diff -Naur kernel.21.3.orig/Documentation/ABI/testing/debugfs-aufs kernel.21.3.new/Documentation/ABI/testing/debugfs-aufs --- kernel.21.3.orig/Documentation/ABI/testing/debugfs-aufs 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/Documentation/ABI/testing/debugfs-aufs 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,50 @@ +What: /debug/aufs/si_/ +Date: March 2009 +Contact: J. R. Okajima +Description: + Under /debug/aufs, a directory named si_ is created + per aufs mount, where is a unique id generated + internally. + +What: /debug/aufs/si_/plink +Date: Apr 2013 +Contact: J. R. Okajima +Description: + It has three lines and shows the information about the + pseudo-link. The first line is a single number + representing a number of buckets. The second line is a + number of pseudo-links per buckets (separated by a + blank). The last line is a single number representing a + total number of psedo-links. + When the aufs mount option 'noplink' is specified, it + will show "1\n0\n0\n". + +What: /debug/aufs/si_/xib +Date: March 2009 +Contact: J. R. Okajima +Description: + It shows the consumed blocks by xib (External Inode Number + Bitmap), its block size and file size. + When the aufs mount option 'noxino' is specified, it + will be empty. About XINO files, see the aufs manual. + +What: /debug/aufs/si_/xino0, xino1 ... xinoN +Date: March 2009 +Contact: J. R. Okajima +Description: + It shows the consumed blocks by xino (External Inode Number + Translation Table), its link count, block size and file + size. + When the aufs mount option 'noxino' is specified, it + will be empty. About XINO files, see the aufs manual. + +What: /debug/aufs/si_/xigen +Date: March 2009 +Contact: J. R. Okajima +Description: + It shows the consumed blocks by xigen (External Inode + Generation Table), its block size and file size. + If CONFIG_AUFS_EXPORT is disabled, this entry will not + be created. + When the aufs mount option 'noxino' is specified, it + will be empty. About XINO files, see the aufs manual. diff -Naur kernel.21.3.orig/Documentation/ABI/testing/sysfs-aufs kernel.21.3.new/Documentation/ABI/testing/sysfs-aufs --- kernel.21.3.orig/Documentation/ABI/testing/sysfs-aufs 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/Documentation/ABI/testing/sysfs-aufs 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,31 @@ +What: /sys/fs/aufs/si_/ +Date: March 2009 +Contact: J. R. Okajima +Description: + Under /sys/fs/aufs, a directory named si_ is created + per aufs mount, where is a unique id generated + internally. + +What: /sys/fs/aufs/si_/br0, br1 ... brN +Date: March 2009 +Contact: J. R. Okajima +Description: + It shows the abolute path of a member directory (which + is called branch) in aufs, and its permission. + +What: /sys/fs/aufs/si_/brid0, brid1 ... bridN +Date: July 2013 +Contact: J. R. Okajima +Description: + It shows the id of a member directory (which is called + branch) in aufs. + +What: /sys/fs/aufs/si_/xi_path +Date: March 2009 +Contact: J. R. Okajima +Description: + It shows the abolute path of XINO (External Inode Number + Bitmap, Translation Table and Generation Table) file + even if it is the default path. + When the aufs mount option 'noxino' is specified, it + will be empty. About XINO files, see the aufs manual. diff -Naur kernel.21.3.orig/Documentation/filesystems/aufs/design/01intro.txt kernel.21.3.new/Documentation/filesystems/aufs/design/01intro.txt --- kernel.21.3.orig/Documentation/filesystems/aufs/design/01intro.txt 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/Documentation/filesystems/aufs/design/01intro.txt 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,161 @@ + +# Copyright (C) 2005-2014 Junjiro R. Okajima +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see . + +Introduction +---------------------------------------- + +aufs [ei ju: ef es] | [a u f s] +1. abbrev. for "advanced multi-layered unification filesystem". +2. abbrev. for "another unionfs". +3. abbrev. for "auf das" in German which means "on the" in English. + Ex. "Butter aufs Brot"(G) means "butter onto bread"(E). + But "Filesystem aufs Filesystem" is hard to understand. + +AUFS is a filesystem with features: +- multi layered stackable unification filesystem, the member directory + is called as a branch. +- branch permission and attribute, 'readonly', 'real-readonly', + 'readwrite', 'whiteout-able', 'link-able whiteout' and their + combination. +- internal "file copy-on-write". +- logical deletion, whiteout. +- dynamic branch manipulation, adding, deleting and changing permission. +- allow bypassing aufs, user's direct branch access. +- external inode number translation table and bitmap which maintains the + persistent aufs inode number. +- seekable directory, including NFS readdir. +- file mapping, mmap and sharing pages. +- pseudo-link, hardlink over branches. +- loopback mounted filesystem as a branch. +- several policies to select one among multiple writable branches. +- revert a single systemcall when an error occurs in aufs. +- and more... + + +Multi Layered Stackable Unification Filesystem +---------------------------------------------------------------------- +Most people already knows what it is. +It is a filesystem which unifies several directories and provides a +merged single directory. When users access a file, the access will be +passed/re-directed/converted (sorry, I am not sure which English word is +correct) to the real file on the member filesystem. The member +filesystem is called 'lower filesystem' or 'branch' and has a mode +'readonly' and 'readwrite.' And the deletion for a file on the lower +readonly branch is handled by creating 'whiteout' on the upper writable +branch. + +On LKML, there have been discussions about UnionMount (Jan Blunck, +Bharata B Rao and Valerie Aurora) and Unionfs (Erez Zadok). They took +different approaches to implement the merged-view. +The former tries putting it into VFS, and the latter implements as a +separate filesystem. +(If I misunderstand about these implementations, please let me know and +I shall correct it. Because it is a long time ago when I read their +source files last time). + +UnionMount's approach will be able to small, but may be hard to share +branches between several UnionMount since the whiteout in it is +implemented in the inode on branch filesystem and always +shared. According to Bharata's post, readdir does not seems to be +finished yet. +There are several missing features known in this implementations such as +- for users, the inode number may change silently. eg. copy-up. +- link(2) may break by copy-up. +- read(2) may get an obsoleted filedata (fstat(2) too). +- fcntl(F_SETLK) may be broken by copy-up. +- unnecessary copy-up may happen, for example mmap(MAP_PRIVATE) after + open(O_RDWR). + +Unionfs has a longer history. When I started implementing a stacking filesystem +(Aug 2005), it already existed. It has virtual super_block, inode, +dentry and file objects and they have an array pointing lower same kind +objects. After contributing many patches for Unionfs, I re-started my +project AUFS (Jun 2006). + +In AUFS, the structure of filesystem resembles to Unionfs, but I +implemented my own ideas, approaches and enhancements and it became +totally different one. + +Comparing DM snapshot and fs based implementation +- the number of bytes to be copied between devices is much smaller. +- the type of filesystem must be one and only. +- the fs must be writable, no readonly fs, even for the lower original + device. so the compression fs will not be usable. but if we use + loopback mount, we may address this issue. + for instance, + mount /cdrom/squashfs.img /sq + losetup /sq/ext2.img + losetup /somewhere/cow + dmsetup "snapshot /dev/loop0 /dev/loop1 ..." +- it will be difficult (or needs more operations) to extract the + difference between the original device and COW. +- DM snapshot-merge may help a lot when users try merging. in the + fs-layer union, users will use rsync(1). + + +Several characters/aspects of aufs +---------------------------------------------------------------------- + +Aufs has several characters or aspects. +1. a filesystem, callee of VFS helper +2. sub-VFS, caller of VFS helper for branches +3. a virtual filesystem which maintains persistent inode number +4. reader/writer of files on branches such like an application + +1. Callee of VFS Helper +As an ordinary linux filesystem, aufs is a callee of VFS. For instance, +unlink(2) from an application reaches sys_unlink() kernel function and +then vfs_unlink() is called. vfs_unlink() is one of VFS helper and it +calls filesystem specific unlink operation. Actually aufs implements the +unlink operation but it behaves like a redirector. + +2. Caller of VFS Helper for Branches +aufs_unlink() passes the unlink request to the branch filesystem as if +it were called from VFS. So the called unlink operation of the branch +filesystem acts as usual. As a caller of VFS helper, aufs should handle +every necessary pre/post operation for the branch filesystem. +- acquire the lock for the parent dir on a branch +- lookup in a branch +- revalidate dentry on a branch +- mnt_want_write() for a branch +- vfs_unlink() for a branch +- mnt_drop_write() for a branch +- release the lock on a branch + +3. Persistent Inode Number +One of the most important issue for a filesystem is to maintain inode +numbers. This is particularly important to support exporting a +filesystem via NFS. Aufs is a virtual filesystem which doesn't have a +backend block device for its own. But some storage is necessary to +maintain inode number. It may be a large space and may not suit to keep +in memory. Aufs rents some space from its first writable branch +filesystem (by default) and creates file(s) on it. These files are +created by aufs internally and removed soon (currently) keeping opened. +Note: Because these files are removed, they are totally gone after + unmounting aufs. It means the inode numbers are not persistent + across unmount or reboot. I have a plan to make them really + persistent which will be important for aufs on NFS server. + +4. Read/Write Files Internally (copy-on-write) +Because a branch can be readonly, when you write a file on it, aufs will +"copy-up" it to the upper writable branch internally. And then write the +originally requested thing to the file. Generally kernel doesn't +open/read/write file actively. In aufs, even a single write may cause a +internal "file copy". This behaviour is very similar to cp(1) command. + +Some people may think it is better to pass such work to user space +helper, instead of doing in kernel space. Actually I am still thinking +about it. But currently I have implemented it in kernel space. diff -Naur kernel.21.3.orig/Documentation/filesystems/aufs/design/02struct.txt kernel.21.3.new/Documentation/filesystems/aufs/design/02struct.txt --- kernel.21.3.orig/Documentation/filesystems/aufs/design/02struct.txt 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/Documentation/filesystems/aufs/design/02struct.txt 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,251 @@ + +# Copyright (C) 2005-2014 Junjiro R. Okajima +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see . + +Basic Aufs Internal Structure + +Superblock/Inode/Dentry/File Objects +---------------------------------------------------------------------- +As like an ordinary filesystem, aufs has its own +superblock/inode/dentry/file objects. All these objects have a +dynamically allocated array and store the same kind of pointers to the +lower filesystem, branch. +For example, when you build a union with one readwrite branch and one +readonly, mounted /au, /rw and /ro respectively. +- /au = /rw + /ro +- /ro/fileA exists but /rw/fileA + +Aufs lookup operation finds /ro/fileA and gets dentry for that. These +pointers are stored in a aufs dentry. The array in aufs dentry will be, +- [0] = NULL +- [1] = /ro/fileA + +This style of an array is essentially same to the aufs +superblock/inode/dentry/file objects. + +Because aufs supports manipulating branches, ie. add/delete/change +dynamically, these objects has its own generation. When branches are +changed, the generation in aufs superblock is incremented. And a +generation in other object are compared when it is accessed. +When a generation in other objects are obsoleted, aufs refreshes the +internal array. + + +Superblock +---------------------------------------------------------------------- +Additionally aufs superblock has some data for policies to select one +among multiple writable branches, XIB files, pseudo-links and kobject. +See below in detail. +About the policies which supports copy-down a directory, see policy.txt +too. + + +Branch and XINO(External Inode Number Translation Table) +---------------------------------------------------------------------- +Every branch has its own xino (external inode number translation table) +file. The xino file is created and unlinked by aufs internally. When two +members of a union exist on the same filesystem, they share the single +xino file. +The struct of a xino file is simple, just a sequence of aufs inode +numbers which is indexed by the lower inode number. +In the above sample, assume the inode number of /ro/fileA is i111 and +aufs assigns the inode number i999 for fileA. Then aufs writes 999 as +4(8) bytes at 111 * 4(8) bytes offset in the xino file. + +When the inode numbers are not contiguous, the xino file will be sparse +which has a hole in it and doesn't consume as much disk space as it +might appear. If your branch filesystem consumes disk space for such +holes, then you should specify 'xino=' option at mounting aufs. + +Also a writable branch has three kinds of "whiteout bases". All these +are existed when the branch is joined to aufs and the names are +whiteout-ed doubly, so that users will never see their names in aufs +hierarchy. +1. a regular file which will be linked to all whiteouts. +2. a directory to store a pseudo-link. +3. a directory to store an "orphan-ed" file temporary. + +1. Whiteout Base + When you remove a file on a readonly branch, aufs handles it as a + logical deletion and creates a whiteout on the upper writable branch + as a hardlink of this file in order not to consume inode on the + writable branch. +2. Pseudo-link Dir + See below, Pseudo-link. +3. Step-Parent Dir + When "fileC" exists on the lower readonly branch only and it is + opened and removed with its parent dir, and then user writes + something into it, then aufs copies-up fileC to this + directory. Because there is no other dir to store fileC. After + creating a file under this dir, the file is unlinked. + +Because aufs supports manipulating branches, ie. add/delete/change +dynamically, a branch has its own id. When the branch order changes, aufs +finds the new index by searching the branch id. + + +Pseudo-link +---------------------------------------------------------------------- +Assume "fileA" exists on the lower readonly branch only and it is +hardlinked to "fileB" on the branch. When you write something to fileA, +aufs copies-up it to the upper writable branch. Additionally aufs +creates a hardlink under the Pseudo-link Directory of the writable +branch. The inode of a pseudo-link is kept in aufs super_block as a +simple list. If fileB is read after unlinking fileA, aufs returns +filedata from the pseudo-link instead of the lower readonly +branch. Because the pseudo-link is based upon the inode, to keep the +inode number by xino (see above) is important. + +All the hardlinks under the Pseudo-link Directory of the writable branch +should be restored in a proper location later. Aufs provides a utility +to do this. The userspace helpers executed at remounting and unmounting +aufs by default. +During this utility is running, it puts aufs into the pseudo-link +maintenance mode. In this mode, only the process which began the +maintenance mode (and its child processes) is allowed to operate in +aufs. Some other processes which are not related to the pseudo-link will +be allowed to run too, but the rest have to return an error or wait +until the maintenance mode ends. If a process already acquires an inode +mutex (in VFS), it has to return an error. + + +XIB(external inode number bitmap) +---------------------------------------------------------------------- +Addition to the xino file per a branch, aufs has an external inode number +bitmap in a superblock object. It is also a file such like a xino file. +It is a simple bitmap to mark whether the aufs inode number is in-use or +not. +To reduce the file I/O, aufs prepares a single memory page to cache xib. + +Aufs implements a feature to truncate/refresh both of xino and xib to +reduce the number of consumed disk blocks for these files. + + +Virtual or Vertical Dir, and Readdir in Userspace +---------------------------------------------------------------------- +In order to support multiple layers (branches), aufs readdir operation +constructs a virtual dir block on memory. For readdir, aufs calls +vfs_readdir() internally for each dir on branches, merges their entries +with eliminating the whiteout-ed ones, and sets it to file (dir) +object. So the file object has its entry list until it is closed. The +entry list will be updated when the file position is zero and becomes +old. This decision is made in aufs automatically. + +The dynamically allocated memory block for the name of entries has a +unit of 512 bytes (by default) and stores the names contiguously (no +padding). Another block for each entry is handled by kmem_cache too. +During building dir blocks, aufs creates hash list and judging whether +the entry is whiteouted by its upper branch or already listed. +The merged result is cached in the corresponding inode object and +maintained by a customizable life-time option. + +Some people may call it can be a security hole or invite DoS attack +since the opened and once readdir-ed dir (file object) holds its entry +list and becomes a pressure for system memory. But I'd say it is similar +to files under /proc or /sys. The virtual files in them also holds a +memory page (generally) while they are opened. When an idea to reduce +memory for them is introduced, it will be applied to aufs too. +For those who really hate this situation, I've developed readdir(3) +library which operates this merging in userspace. You just need to set +LD_PRELOAD environment variable, and aufs will not consume no memory in +kernel space for readdir(3). + + +Workqueue +---------------------------------------------------------------------- +Aufs sometimes requires privilege access to a branch. For instance, +in copy-up/down operation. When a user process is going to make changes +to a file which exists in the lower readonly branch only, and the mode +of one of ancestor directories may not be writable by a user +process. Here aufs copy-up the file with its ancestors and they may +require privilege to set its owner/group/mode/etc. +This is a typical case of a application character of aufs (see +Introduction). + +Aufs uses workqueue synchronously for this case. It creates its own +workqueue. The workqueue is a kernel thread and has privilege. Aufs +passes the request to call mkdir or write (for example), and wait for +its completion. This approach solves a problem of a signal handler +simply. +If aufs didn't adopt the workqueue and changed the privilege of the +process, and if the mkdir/write call arises SIGXFSZ or other signal, +then the user process might gain a privilege or the generated core file +was owned by a superuser. + +Also aufs uses the system global workqueue ("events" kernel thread) too +for asynchronous tasks, such like handling inotify/fsnotify, re-creating a +whiteout base and etc. This is unrelated to a privilege. +Most of aufs operation tries acquiring a rw_semaphore for aufs +superblock at the beginning, at the same time waits for the completion +of all queued asynchronous tasks. + + +Whiteout +---------------------------------------------------------------------- +The whiteout in aufs is very similar to Unionfs's. That is represented +by its filename. UnionMount takes an approach of a file mode, but I am +afraid several utilities (find(1) or something) will have to support it. + +Basically the whiteout represents "logical deletion" which stops aufs to +lookup further, but also it represents "dir is opaque" which also stop +lookup. + +In aufs, rmdir(2) and rename(2) for dir uses whiteout alternatively. +In order to make several functions in a single systemcall to be +revertible, aufs adopts an approach to rename a directory to a temporary +unique whiteouted name. +For example, in rename(2) dir where the target dir already existed, aufs +renames the target dir to a temporary unique whiteouted name before the +actual rename on a branch and then handles other actions (make it opaque, +update the attributes, etc). If an error happens in these actions, aufs +simply renames the whiteouted name back and returns an error. If all are +succeeded, aufs registers a function to remove the whiteouted unique +temporary name completely and asynchronously to the system global +workqueue. + + +Copy-up +---------------------------------------------------------------------- +It is a well-known feature or concept. +When user modifies a file on a readonly branch, aufs operate "copy-up" +internally and makes change to the new file on the upper writable branch. +When the trigger systemcall does not update the timestamps of the parent +dir, aufs reverts it after copy-up. + + +Move-down (aufs3.9 and later) +---------------------------------------------------------------------- +"Copy-up" is one of the essential feature in aufs. It copies a file from +the lower readonly branch to the upper writable branch when a user +changes something about the file. +"Move-down" is an opposite action of copy-up. Basically this action is +ran manually instead of automatically and internally. +For desgin and implementation, aufs has to consider these issues. +- whiteout for the file may exist on the lower branch. +- ancestor directories may not exist on the lower branch. +- diropq for the ancestor directories may exist on the upper branch. +- free space on the lower branch will reduce. +- another access to the file may happen during moving-down, including + UDBA. +- the file should not be hard-linked nor pseudo-linked. they should be + handled by auplink utility later. + +Sometimes users want to move-down a file from the upper writable branch +to the lower readonly or writable branch. For instance, +- the free space of the upper writable branch is going to run out. +- create a new intermediate branch between the upper and lower branch. +- etc. + +For this purpose, use "aumvdown" command in aufs-util.git. diff -Naur kernel.21.3.orig/Documentation/filesystems/aufs/design/03lookup.txt kernel.21.3.new/Documentation/filesystems/aufs/design/03lookup.txt --- kernel.21.3.orig/Documentation/filesystems/aufs/design/03lookup.txt 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/Documentation/filesystems/aufs/design/03lookup.txt 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,133 @@ + +# Copyright (C) 2005-2014 Junjiro R. Okajima +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see . + +Lookup in a Branch +---------------------------------------------------------------------- +Since aufs has a character of sub-VFS (see Introduction), it operates +lookup for branches as VFS does. It may be a heavy work. Generally +speaking struct nameidata is a bigger structure and includes many +information. But almost all lookup operation in aufs is the simplest +case, ie. lookup only an entry directly connected to its parent. Digging +down the directory hierarchy is unnecessary. + +VFS has a function lookup_one_len() for that use, but it is not usable +for a branch filesystem which requires struct nameidata. So aufs +implements a simple lookup wrapper function. When a branch filesystem +allows NULL as nameidata, it calls lookup_one_len(). Otherwise it builds +a simplest nameidata and calls lookup_hash(). +Here aufs applies "a principle in NFSD", ie. if the filesystem supports +NFS-export, then it has to support NULL as a nameidata parameter for +->create(), ->lookup() and ->d_revalidate(). So the lookup wrapper in +aufs tests if ->s_export_op in the branch is NULL or not. + +When a branch is a remote filesystem, aufs basically trusts its +->d_revalidate(), also aufs forces the hardest revalidate tests for +them. +For d_revalidate, aufs implements three levels of revalidate tests. See +"Revalidate Dentry and UDBA" in detail. + + +Test Only the Highest One for the Directory Permission (dirperm1 option) +---------------------------------------------------------------------- +Let's try case study. +- aufs has two branches, upper readwrite and lower readonly. + /au = /rw + /ro +- "dirA" exists under /ro, but /rw. and its mode is 0700. +- user invoked "chmod a+rx /au/dirA" +- the internal copy-up is activated and "/rw/dirA" is created and its + permission bits are set to world readble. +- then "/au/dirA" becomes world readable? + +In this case, /ro/dirA is still 0700 since it exists in readonly branch, +or it may be a natively readonly filesystem. If aufs respects the lower +branch, it should not respond readdir request from other users. But user +allowed it by chmod. Should really aufs rejects showing the entries +under /ro/dirA? + +To be honest, I don't have a best solution for this case. So aufs +implements 'dirperm1' and 'nodirperm1' and leave it to users. +When dirperm1 is specified, aufs checks only the highest one for the +directory permission, and shows the entries. Otherwise, as usual, checks +every dir existing on all branches and rejects the request. + +As a side effect, dirperm1 option improves the performance of aufs +because the number of permission check is reduced when the number of +branch is many. + + +Loopback Mount +---------------------------------------------------------------------- +Basically aufs supports any type of filesystem and block device for a +branch (actually there are some exceptions). But it is prohibited to add +a loopback mounted one whose backend file exists in a filesystem which is +already added to aufs. The reason is to protect aufs from a recursive +lookup. If it was allowed, the aufs lookup operation might re-enter a +lookup for the loopback mounted branch in the same context, and will +cause a deadlock. + + +Revalidate Dentry and UDBA (User's Direct Branch Access) +---------------------------------------------------------------------- +Generally VFS helpers re-validate a dentry as a part of lookup. +0. digging down the directory hierarchy. +1. lock the parent dir by its i_mutex. +2. lookup the final (child) entry. +3. revalidate it. +4. call the actual operation (create, unlink, etc.) +5. unlock the parent dir + +If the filesystem implements its ->d_revalidate() (step 3), then it is +called. Actually aufs implements it and checks the dentry on a branch is +still valid. +But it is not enough. Because aufs has to release the lock for the +parent dir on a branch at the end of ->lookup() (step 2) and +->d_revalidate() (step 3) while the i_mutex of the aufs dir is still +held by VFS. +If the file on a branch is changed directly, eg. bypassing aufs, after +aufs released the lock, then the subsequent operation may cause +something unpleasant result. + +This situation is a result of VFS architecture, ->lookup() and +->d_revalidate() is separated. But I never say it is wrong. It is a good +design from VFS's point of view. It is just not suitable for sub-VFS +character in aufs. + +Aufs supports such case by three level of revalidation which is +selectable by user. +1. Simple Revalidate + Addition to the native flow in VFS's, confirm the child-parent + relationship on the branch just after locking the parent dir on the + branch in the "actual operation" (step 4). When this validation + fails, aufs returns EBUSY. ->d_revalidate() (step 3) in aufs still + checks the validation of the dentry on branches. +2. Monitor Changes Internally by Inotify/Fsnotify + Addition to above, in the "actual operation" (step 4) aufs re-lookup + the dentry on the branch, and returns EBUSY if it finds different + dentry. + Additionally, aufs sets the inotify/fsnotify watch for every dir on branches + during it is in cache. When the event is notified, aufs registers a + function to kernel 'events' thread by schedule_work(). And the + function sets some special status to the cached aufs dentry and inode + private data. If they are not cached, then aufs has nothing to + do. When the same file is accessed through aufs (step 0-3) later, + aufs will detect the status and refresh all necessary data. + In this mode, aufs has to ignore the event which is fired by aufs + itself. +3. No Extra Validation + This is the simplest test and doesn't add any additional revalidation + test, and skip therevalidatin in step 4. It is useful and improves + aufs performance when system surely hide the aufs branches from user, + by over-mounting something (or another method). diff -Naur kernel.21.3.orig/Documentation/filesystems/aufs/design/04branch.txt kernel.21.3.new/Documentation/filesystems/aufs/design/04branch.txt --- kernel.21.3.orig/Documentation/filesystems/aufs/design/04branch.txt 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/Documentation/filesystems/aufs/design/04branch.txt 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,75 @@ + +# Copyright (C) 2005-2014 Junjiro R. Okajima +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see . + +Branch Manipulation + +Since aufs supports dynamic branch manipulation, ie. add/remove a branch +and changing its permission/attribute, there are a lot of works to do. + + +Add a Branch +---------------------------------------------------------------------- +o Confirm the adding dir exists outside of aufs, including loopback + mount. +- and other various attributes... +o Initialize the xino file and whiteout bases if necessary. + See struct.txt. + +o Check the owner/group/mode of the directory + When the owner/group/mode of the adding directory differs from the + existing branch, aufs issues a warning because it may impose a + security risk. + For example, when a upper writable branch has a world writable empty + top directory, a malicious user can create any files on the writable + branch directly, like copy-up and modify manually. If something like + /etc/{passwd,shadow} exists on the lower readonly branch but the upper + writable branch, and the writable branch is world-writable, then a + malicious guy may create /etc/passwd on the writable branch directly + and the infected file will be valid in aufs. + I am afraid it can be a security issue, but nothing to do except + producing a warning. + + +Delete a Branch +---------------------------------------------------------------------- +o Confirm the deleting branch is not busy + To be general, there is one merit to adopt "remount" interface to + manipulate branches. It is to discard caches. At deleting a branch, + aufs checks the still cached (and connected) dentries and inodes. If + there are any, then they are all in-use. An inode without its + corresponding dentry can be alive alone (for example, inotify/fsnotify case). + + For the cached one, aufs checks whether the same named entry exists on + other branches. + If the cached one is a directory, because aufs provides a merged view + to users, as long as one dir is left on any branch aufs can show the + dir to users. In this case, the branch can be removed from aufs. + Otherwise aufs rejects deleting the branch. + + If any file on the deleting branch is opened by aufs, then aufs + rejects deleting. + + +Modify the Permission of a Branch +---------------------------------------------------------------------- +o Re-initialize or remove the xino file and whiteout bases if necessary. + See struct.txt. + +o rw --> ro: Confirm the modifying branch is not busy + Aufs rejects the request if any of these conditions are true. + - a file on the branch is mmap-ed. + - a regular file on the branch is opened for write and there is no + same named entry on the upper branch. diff -Naur kernel.21.3.orig/Documentation/filesystems/aufs/design/05wbr_policy.txt kernel.21.3.new/Documentation/filesystems/aufs/design/05wbr_policy.txt --- kernel.21.3.orig/Documentation/filesystems/aufs/design/05wbr_policy.txt 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/Documentation/filesystems/aufs/design/05wbr_policy.txt 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,64 @@ + +# Copyright (C) 2005-2014 Junjiro R. Okajima +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see . + +Policies to Select One among Multiple Writable Branches +---------------------------------------------------------------------- +When the number of writable branch is more than one, aufs has to decide +the target branch for file creation or copy-up. By default, the highest +writable branch which has the parent (or ancestor) dir of the target +file is chosen (top-down-parent policy). +By user's request, aufs implements some other policies to select the +writable branch, for file creation two policies, round-robin and +most-free-space policies. For copy-up three policies, top-down-parent, +bottom-up-parent and bottom-up policies. + +As expected, the round-robin policy selects the branch in circular. When +you have two writable branches and creates 10 new files, 5 files will be +created for each branch. mkdir(2) systemcall is an exception. When you +create 10 new directories, all will be created on the same branch. +And the most-free-space policy selects the one which has most free +space among the writable branches. The amount of free space will be +checked by aufs internally, and users can specify its time interval. + +The policies for copy-up is more simple, +top-down-parent is equivalent to the same named on in create policy, +bottom-up-parent selects the writable branch where the parent dir +exists and the nearest upper one from the copyup-source, +bottom-up selects the nearest upper writable branch from the +copyup-source, regardless the existence of the parent dir. + +There are some rules or exceptions to apply these policies. +- If there is a readonly branch above the policy-selected branch and + the parent dir is marked as opaque (a variation of whiteout), or the + target (creating) file is whiteout-ed on the upper readonly branch, + then the result of the policy is ignored and the target file will be + created on the nearest upper writable branch than the readonly branch. +- If there is a writable branch above the policy-selected branch and + the parent dir is marked as opaque or the target file is whiteouted + on the branch, then the result of the policy is ignored and the target + file will be created on the highest one among the upper writable + branches who has diropq or whiteout. In case of whiteout, aufs removes + it as usual. +- link(2) and rename(2) systemcalls are exceptions in every policy. + They try selecting the branch where the source exists as possible + since copyup a large file will take long time. If it can't be, + ie. the branch where the source exists is readonly, then they will + follow the copyup policy. +- There is an exception for rename(2) when the target exists. + If the rename target exists, aufs compares the index of the branches + where the source and the target exists and selects the higher + one. If the selected branch is readonly, then aufs follows the + copyup policy. diff -Naur kernel.21.3.orig/Documentation/filesystems/aufs/design/06fhsm.txt kernel.21.3.new/Documentation/filesystems/aufs/design/06fhsm.txt --- kernel.21.3.orig/Documentation/filesystems/aufs/design/06fhsm.txt 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/Documentation/filesystems/aufs/design/06fhsm.txt 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,120 @@ + +# Copyright (C) 2011-2014 Junjiro R. Okajima +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program; if not, write to the Free Software +# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + + +File-based Hierarchical Storage Management (FHSM) +---------------------------------------------------------------------- +Hierarchical Storage Management (or HSM) is a well-known feature in the +storage world. Aufs provides this feature as file-based with multiple +writable branches, based upon the principle of "Colder-Lower". +Here the word "colder" means that the less used files, and "lower" means +that the position in the order of the stacked branches. +These multiple writable branches are prioritized, ie. the topmost one +should be the fastest drive and be used heavily. + +o Characters in aufs FHSM story +- aufs itself and a new branch attribute. +- a new ioctl interface to move-down and to establish a connection with + the daemon ("move-down" is a converse of "copy-up"). +- userspace tool and daemon. + +The userspace daemon establishes a connection with aufs and waits for +the notification. The notified information is very similar to struct +statfs containing the number of consumed blocks and inodes. +When the consumed blocks/inodes of a branch exceeds the user-specified +upper watermark, the daemon activates its move-down process until the +consumed blocks/inodes reaches the user-specified lower watermark. + +The actual move-down is done by aufs based upon the request from +user-space since we need to maintain the inode number and the internal +pointer arrays in aufs. + +Currently aufs FHSM handles the regular files only. Additionally they +must not be hard-linked nor pseudo-linked. + + +o Cowork of aufs and the user-space daemon + During the userspace daemon established the connection, aufs sends a + small notification to it whenever aufs writes something into the + writable branch. But it may cost high since aufs issues statfs(2) + internally. So user can specify a new option to cache the + info. Actually the notification is controlled by these factors. + + the specified cache time. + + classified as "force" by aufs internally. + Until the specified time expires, aufs doesn't send the info + except the forced cases. When aufs decide forcing, the info is always + notified to userspace. + For example, the number of free inodes is generally large enough and + the shortage of it happens rarely. So aufs doesn't force the + notification when creating a new file, directory and others. This is + the typical case which aufs doesn't force. + When aufs writes the actual filedata and the files consumes any of new + blocks, the aufs forces notifying. + + +o Interfaces in aufs +- New branch attribute. + + fhsm + Specifies that the branch is managed by FHSM feature. In other word, + participant in the FHSM. + When nofhsm is set to the branch, it will not be the source/target + branch of the move-down operation. This attribute is set + independently from coo and moo attributes, and if you want full + FHSM, you should specify them as well. +- New mount option. + + fhsm_sec + Specifies a second to suppress many less important info to be + notified. +- New ioctl. + + AUFS_CTL_FHSM_FD + create a new file descriptor which userspace can read the notification + (a subset of struct statfs) from aufs. +- Module parameter 'brs' + It has to be set to 1. Otherwise the new mount option 'fhsm' will not + be set. +- mount helpers /sbin/mount.aufs and /sbin/umount.aufs + When there are two or more branches with fhsm attributes, + /sbin/mount.aufs invokes the user-space daemon and /sbin/umount.aufs + terminates it. As a result of remounting and branch-manipulation, the + number of branches with fhsm attribute can be one. In this case, + /sbin/mount.aufs will terminate the user-space daemon. + + +Finally the operation is done as these steps in kernel-space. +- make sure that, + + no one else is using the file. + + the file is not hard-linked. + + the file is not pseudo-linked. + + the file is a regular file. + + the parent dir is not opaqued. +- find the target writable branch. +- make sure the file is not whiteout-ed by the upper (than the target) + branch. +- make the parent dir on the target branch. +- mutex lock the inode on the branch. +- unlink the whiteout on the target branch (if exists). +- lookup and create the whiteout-ed temporary name on the target branch. +- copy the file as the whiteout-ed temporary name on the target branch. +- rename the whiteout-ed temporary name to the original name. +- unlink the file on the source branch. +- maintain the internal pointer array and the external inode number + table (XINO). +- maintain the timestamps and other attributes of the parent dir and the + file. + +And of course, in every step, an error may happen. So the operation +should restore the original file state after an error happens. diff -Naur kernel.21.3.orig/Documentation/filesystems/aufs/design/06mmap.txt kernel.21.3.new/Documentation/filesystems/aufs/design/06mmap.txt --- kernel.21.3.orig/Documentation/filesystems/aufs/design/06mmap.txt 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/Documentation/filesystems/aufs/design/06mmap.txt 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,46 @@ + +# Copyright (C) 2005-2014 Junjiro R. Okajima +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see . + +mmap(2) -- File Memory Mapping +---------------------------------------------------------------------- +In aufs, the file-mapped pages are handled by a branch fs directly, no +interaction with aufs. It means aufs_mmap() calls the branch fs's +->mmap(). +This approach is simple and good, but there is one problem. +Under /proc, several entries show the mmap-ped files by its path (with +device and inode number), and the printed path will be the path on the +branch fs's instead of virtual aufs's. +This is not a problem in most cases, but some utilities lsof(1) (and its +user) may expect the path on aufs. + +To address this issue, aufs adds a new member called vm_prfile in struct +vm_area_struct (and struct vm_region). The original vm_file points to +the file on the branch fs in order to handle everything correctly as +usual. The new vm_prfile points to a virtual file in aufs, and the +show-functions in procfs refers to vm_prfile if it is set. +Also we need to maintain several other places where touching vm_file +such like +- fork()/clone() copies vma and the reference count of vm_file is + incremented. +- merging vma maintains the ref count too. + +This is not a good approach. It just faking the printed path. But it +leaves all behaviour around f_mapping unchanged. This is surely an +advantage. +Actually aufs had adopted another complicated approach which calls +generic_file_mmap() and handles struct vm_operations_struct. In this +approach, aufs met a hard problem and I could not solve it without +switching the approach. diff -Naur kernel.21.3.orig/Documentation/filesystems/aufs/design/06xattr.txt kernel.21.3.new/Documentation/filesystems/aufs/design/06xattr.txt --- kernel.21.3.orig/Documentation/filesystems/aufs/design/06xattr.txt 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/Documentation/filesystems/aufs/design/06xattr.txt 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,96 @@ + +# Copyright (C) 2014 Junjiro R. Okajima +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program; if not, write to the Free Software +# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + + +Listing XATTR/EA and getting the value +---------------------------------------------------------------------- +For the inode standard attributes (owner, group, timestamps, etc.), aufs +shows the values from the topmost existing file. This behaviour is good +for the non-dir entreis since the bahaviour exactly matches the shown +information. But for the directories, aufs considers all the same named +entries on the lower branches. Which means, if one of the lower entry +rejects readdir call, then aufs returns an error even if the topmost +entry allows it. This behaviour is necessary to respect the branch fs's +security, but can make users confused since the user-visible standard +attributes don't match the behaviour. +To address this issue, aufs has a mount option called dirperm1 which +checks the permission for the topmost entry only, and ignores the lower +entry's permission. + +A similar issue can happen around XATTR. +getxattr(2) and listxattr(2) families behave as if dirperm1 option is +always set. Otherwise these very unpleasant situation can happen. +- listxattr(2) may return the duplicated entires. +- users may not be able to remove or reset the XATTR forever, + + +XATTR/EA support in the internal (copy,move)-(up,down) +---------------------------------------------------------------------- +Generally the extended attributes of inode are categorazied as these. +- "security" for LSM and capability. +- "system" for posix ACL, 'acl' mount option is required for the branch + fs generally. +- "trusted" for userspace, CAP_SYS_ADMIN is required. +- "user" for userspace, 'user_xattr' mount option is required for the + branch fs generally. + +Moreover there are some other categories. Aufs handles these rather +unpopular categories as the ordinary ones, ie. there is no special +condition nor exception. + +In copy-up, the support for XATTR on the dst branch may differ from the +src branch. In this case, the copy-up operation will get an error and +the original user operation which triggered the copy-up fails. It can +happen that even all copy-up will fail. +When both of src and dst branches support XATTR and if an error occurs +during copying XATTR, then the copy-up should fail obviously. That is a +good reason and aufs should return an error to userspace. But when only +the src branch support XATTR, aufs should not return an error. +For example, the src branch supports ACL but the dst branch doesn't +because the dst branch may natively un-support it or temporary +un-support it due to "noacl" mount option. Of course, the dst branch fs +may NOT return an error even if the XATTR is not supported. It is +totally up to the branch fs. + +Anyway when the aufs internal copy-up gets an error from the dst branch +fs, then aufs tries removing the just copied entry and returns the error +to the userspace. The worst case of this situation will be all copy-up +will fail. + +For the copy-up operation, there two basic approaches. +- copy the specified XATTR only (by category above), and return the + error if it happens inconditionally. +- copy all XATTR, and ignore the error on the specified category only. + +In order to support XATTR and to implement the correct behaviour, aufs +chooses the latter approach and introduces some attributes for its +branch, "icexsec", "icexsys", "icextr", "icexusr", and "icexoth". +They correspond to the XATTR namespaces (see above). Additionally, to be +convenient, "icex" is also provided which means all "ix*" attributes are +set. + +The meaning of these attributes is to ignore the error from setting +XATTR on that branch. +Note that aufs tries copying all XATTR unconditionally, and ignores the +error from the dst branch according to the specified attributes. + +Some XATTR may have its default value. The default value may come from +the parent dir or the environment. If the default value is set at the +file creating-time, it will be overwritten by copy-up. +Some contradiction may happen I am afraid. +Do we need another attribute to stop copying XATTR? I am unsure. For +now, aufs implements the branch attributes to ignore the error. diff -Naur kernel.21.3.orig/Documentation/filesystems/aufs/design/07export.txt kernel.21.3.new/Documentation/filesystems/aufs/design/07export.txt --- kernel.21.3.orig/Documentation/filesystems/aufs/design/07export.txt 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/Documentation/filesystems/aufs/design/07export.txt 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,58 @@ + +# Copyright (C) 2005-2014 Junjiro R. Okajima +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see . + +Export Aufs via NFS +---------------------------------------------------------------------- +Here is an approach. +- like xino/xib, add a new file 'xigen' which stores aufs inode + generation. +- iget_locked(): initialize aufs inode generation for a new inode, and + store it in xigen file. +- destroy_inode(): increment aufs inode generation and store it in xigen + file. it is necessary even if it is not unlinked, because any data of + inode may be changed by UDBA. +- encode_fh(): for a root dir, simply return FILEID_ROOT. otherwise + build file handle by + + branch id (4 bytes) + + superblock generation (4 bytes) + + inode number (4 or 8 bytes) + + parent dir inode number (4 or 8 bytes) + + inode generation (4 bytes)) + + return value of exportfs_encode_fh() for the parent on a branch (4 + bytes) + + file handle for a branch (by exportfs_encode_fh()) +- fh_to_dentry(): + + find the index of a branch from its id in handle, and check it is + still exist in aufs. + + 1st level: get the inode number from handle and search it in cache. + + 2nd level: if not found, get the parent inode number from handle and + search it in cache. and then open the parent dir, find the matching + inode number by vfs_readdir() and get its name, and call + lookup_one_len() for the target dentry. + + 3rd level: if the parent dir is not cached, call + exportfs_decode_fh() for a branch and get the parent on a branch, + build a pathname of it, convert it a pathname in aufs, call + path_lookup(). now aufs gets a parent dir dentry, then handle it as + the 2nd level. + + to open the dir, aufs needs struct vfsmount. aufs keeps vfsmount + for every branch, but not itself. to get this, (currently) aufs + searches in current->nsproxy->mnt_ns list. it may not be a good + idea, but I didn't get other approach. + + test the generation of the gotten inode. +- every inode operation: they may get EBUSY due to UDBA. in this case, + convert it into ESTALE for NFSD. +- readdir(): call lockdep_on/off() because filldir in NFSD calls + lookup_one_len(), vfs_getattr(), encode_fh() and others. diff -Naur kernel.21.3.orig/Documentation/filesystems/aufs/design/08shwh.txt kernel.21.3.new/Documentation/filesystems/aufs/design/08shwh.txt --- kernel.21.3.orig/Documentation/filesystems/aufs/design/08shwh.txt 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/Documentation/filesystems/aufs/design/08shwh.txt 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,52 @@ + +# Copyright (C) 2005-2014 Junjiro R. Okajima +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see . + +Show Whiteout Mode (shwh) +---------------------------------------------------------------------- +Generally aufs hides the name of whiteouts. But in some cases, to show +them is very useful for users. For instance, creating a new middle layer +(branch) by merging existing layers. + +(borrowing aufs1 HOW-TO from a user, Michael Towers) +When you have three branches, +- Bottom: 'system', squashfs (underlying base system), read-only +- Middle: 'mods', squashfs, read-only +- Top: 'overlay', ram (tmpfs), read-write + +The top layer is loaded at boot time and saved at shutdown, to preserve +the changes made to the system during the session. +When larger changes have been made, or smaller changes have accumulated, +the size of the saved top layer data grows. At this point, it would be +nice to be able to merge the two overlay branches ('mods' and 'overlay') +and rewrite the 'mods' squashfs, clearing the top layer and thus +restoring save and load speed. + +This merging is simplified by the use of another aufs mount, of just the +two overlay branches using the 'shwh' option. +# mount -t aufs -o ro,shwh,br:/livesys/overlay=ro+wh:/livesys/mods=rr+wh \ + aufs /livesys/merge_union + +A merged view of these two branches is then available at +/livesys/merge_union, and the new feature is that the whiteouts are +visible! +Note that in 'shwh' mode the aufs mount must be 'ro', which will disable +writing to all branches. Also the default mode for all branches is 'ro'. +It is now possible to save the combined contents of the two overlay +branches to a new squashfs, e.g.: +# mksquashfs /livesys/merge_union /path/to/newmods.squash + +This new squashfs archive can be stored on the boot device and the +initramfs will use it to replace the old one at the next boot. diff -Naur kernel.21.3.orig/Documentation/filesystems/aufs/design/10dynop.txt kernel.21.3.new/Documentation/filesystems/aufs/design/10dynop.txt --- kernel.21.3.orig/Documentation/filesystems/aufs/design/10dynop.txt 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/Documentation/filesystems/aufs/design/10dynop.txt 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,46 @@ + +# Copyright (C) 2010-2014 Junjiro R. Okajima +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see . + +Dynamically customizable FS operations +---------------------------------------------------------------------- +Generally FS operations (struct inode_operations, struct +address_space_operations, struct file_operations, etc.) are defined as +"static const", but it never means that FS have only one set of +operation. Some FS have multiple sets of them. For instance, ext2 has +three sets, one for XIP, for NOBH, and for normal. +Since aufs overrides and redirects these operations, sometimes aufs has +to change its behaviour according to the branch FS type. More imporantly +VFS acts differently if a function (member in the struct) is set or +not. It means aufs should have several sets of operations and select one +among them according to the branch FS definition. + +In order to solve this problem and not to affect the behavour of VFS, +aufs defines these operations dynamically. For instance, aufs defines +aio_read function for struct file_operations, but it may not be set to +the file_operations. When the branch FS doesn't have it, aufs doesn't +set it to its file_operations while the function definition itself is +still alive. So the behaviour of io_submit(2) will not change, and it +will return an error when aio_read is not defined. + +The lifetime of these dynamically generated operation object is +maintained by aufs branch object. When the branch is removed from aufs, +the reference counter of the object is decremented. When it reaches +zero, the dynamically generated operation object will be freed. + +This approach is designed to support AIO (io_submit), Direcit I/O and +XIP mainly. +Currently this approach is applied to file_operations and +vm_operations_struct for regular files only. diff -Naur kernel.21.3.orig/Documentation/filesystems/aufs/design/99plan.txt kernel.21.3.new/Documentation/filesystems/aufs/design/99plan.txt --- kernel.21.3.orig/Documentation/filesystems/aufs/design/99plan.txt 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/Documentation/filesystems/aufs/design/99plan.txt 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,58 @@ + +# Copyright (C) 2005-2014 Junjiro R. Okajima +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see . + +Plan + +Restoring some features which was implemented in aufs1. +They were dropped in aufs2 in order to make source files simpler and +easier to be reviewed. + + +Being Another Aufs's Readonly Branch (robr) +---------------------------------------------------------------------- +Aufs1 allows aufs to be another aufs's readonly branch. +This feature was developed by a user's request. But it may not be used +currecnly. + + +Refresh the Opened File (refrof) +---------------------------------------------------------------------- +This option is implemented in aufs1 but incomplete. + +When user reads from a file, he expects to get its latest filedata +generally. If the file is removed and a new same named file is created, +the content he gets is unchanged, ie. the unlinked filedata. + +Let's try case study again. +- aufs has two branches. + /au = /rw + /ro +- "fileA" exists under /ro, but /rw. +- user opened "/au/fileA". +- he or someone else inserts a branch (/new) between /rw and /ro. + /au = /rw + /new + /ro +- the new branch has "fileA". +- user reads from the opened "fileA" +- which filedata should aufs return, from /ro or /new? + +Some people says it has to be "from /ro" and it is a semantics of Unix. +The others say it should be "from /new" because the file is not removed +and it is equivalent to the case of someone else modifies the file. + +Here again I don't have a best and final answer. I got an idea to +implement 'refrof' and 'norefrof' option. When 'refrof' (REFResh the +Opened File) is specified (by default), aufs returns the filedata from +/new. +Otherwise from /new. diff -Naur kernel.21.3.orig/Documentation/filesystems/aufs/README kernel.21.3.new/Documentation/filesystems/aufs/README --- kernel.21.3.orig/Documentation/filesystems/aufs/README 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/Documentation/filesystems/aufs/README 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,370 @@ + +Aufs3 -- advanced multi layered unification filesystem version 3.x +http://aufs.sf.net +Junjiro R. Okajima + + +0. Introduction +---------------------------------------- +In the early days, aufs was entirely re-designed and re-implemented +Unionfs Version 1.x series. After many original ideas, approaches, +improvements and implementations, it becomes totally different from +Unionfs while keeping the basic features. +Recently, Unionfs Version 2.x series begin taking some of the same +approaches to aufs1's. +Unionfs is being developed by Professor Erez Zadok at Stony Brook +University and his team. + +Aufs3 supports linux-3.0 and later. +If you want older kernel version support, try aufs2-2.6.git or +aufs2-standalone.git repository, aufs1 from CVS on SourceForge. + +Note: it becomes clear that "Aufs was rejected. Let's give it up." + According to Christoph Hellwig, linux rejects all union-type + filesystems but UnionMount. + + +PS. Al Viro seems have a plan to merge aufs as well as overlayfs and + UnionMount, and he pointed out an issue around a directory mutex + lock and aufs addressed it. But it is still unsure whether aufs will + be merged (or any other union solution). + + + +1. Features +---------------------------------------- +- unite several directories into a single virtual filesystem. The member + directory is called as a branch. +- you can specify the permission flags to the branch, which are 'readonly', + 'readwrite' and 'whiteout-able.' +- by upper writable branch, internal copyup and whiteout, files/dirs on + readonly branch are modifiable logically. +- dynamic branch manipulation, add, del. +- etc... + +Also there are many enhancements in aufs1, such as: +- readdir(3) in userspace. +- keep inode number by external inode number table +- keep the timestamps of file/dir in internal copyup operation +- seekable directory, supporting NFS readdir. +- whiteout is hardlinked in order to reduce the consumption of inodes + on branch +- do not copyup, nor create a whiteout when it is unnecessary +- revert a single systemcall when an error occurs in aufs +- remount interface instead of ioctl +- maintain /etc/mtab by an external command, /sbin/mount.aufs. +- loopback mounted filesystem as a branch +- kernel thread for removing the dir who has a plenty of whiteouts +- support copyup sparse file (a file which has a 'hole' in it) +- default permission flags for branches +- selectable permission flags for ro branch, whether whiteout can + exist or not +- export via NFS. +- support /fs/aufs and /aufs. +- support multiple writable branches, some policies to select one + among multiple writable branches. +- a new semantics for link(2) and rename(2) to support multiple + writable branches. +- no glibc changes are required. +- pseudo hardlink (hardlink over branches) +- allow a direct access manually to a file on branch, e.g. bypassing aufs. + including NFS or remote filesystem branch. +- userspace wrapper for pathconf(3)/fpathconf(3) with _PC_LINK_MAX. +- and more... + +Currently these features are dropped temporary from aufs3. +See design/08plan.txt in detail. +- test only the highest one for the directory permission (dirperm1) +- copyup on open (coo=) +- nested mount, i.e. aufs as readonly no-whiteout branch of another aufs + (robr) +- statistics of aufs thread (/sys/fs/aufs/stat) +- delegation mode (dlgt) + a delegation of the internal branch access to support task I/O + accounting, which also supports Linux Security Modules (LSM) mainly + for Suse AppArmor. +- intent.open/create (file open in a single lookup) + +Features or just an idea in the future (see also design/*.txt), +- reorder the branch index without del/re-add. +- permanent xino files for NFSD +- an option for refreshing the opened files after add/del branches +- 'move' policy for copy-up between two writable branches, after + checking free space. +- light version, without branch manipulation. (unnecessary?) +- copyup in userspace +- inotify in userspace +- readv/writev +- xattr, acl + + +2. Download +---------------------------------------- +There were three GIT trees for aufs3, aufs3-linux.git, +aufs3-standalone.git, and aufs-util.git. Note that there is no "3" in +"aufs-util.git." +While the aufs-util is always necessary, you need either of aufs3-linux +or aufs3-standalone. + +The aufs3-linux tree includes the whole linux mainline GIT tree, +git://git.kernel.org/.../torvalds/linux.git. +And you cannot select CONFIG_AUFS_FS=m for this version, eg. you cannot +build aufs3 as an external kernel module. + +On the other hand, the aufs3-standalone tree has only aufs source files +and necessary patches, and you can select CONFIG_AUFS_FS=m. + +You will find GIT branches whose name is in form of "aufs3.x" where "x" +represents the linux kernel version, "linux-3.x". For instance, +"aufs3.0" is for linux-3.0. For latest "linux-3.x-rcN", use +"aufs3.x-rcN" branch. + +o aufs3-linux tree +$ git clone --reference /your/linux/git/tree \ + git://git.code.sf.net/p/aufs/aufs3-linux aufs-aufs3-linux \ + aufs3-linux.git +- if you don't have linux GIT tree, then remove "--reference ..." +$ cd aufs3-linux.git +$ git checkout origin/aufs3.0 + +o aufs3-standalone tree +$ git clone git://git.code.sf.net/p/aufs/aufs3-standalone \ + aufs3-standalone.git +$ cd aufs3-standalone.git +$ git checkout origin/aufs3.0 + +o aufs-util tree +$ git clone git://git.code.sf.net/p/aufs/aufs-util \ + aufs-util.git +$ cd aufs-util.git +$ git checkout origin/aufs3.0 + +Note: The 3.x-rcN branch is to be used with `rc' kernel versions ONLY. +The minor version number, 'x' in '3.x', of aufs may not always +follow the minor version number of the kernel. +Because changes in the kernel that cause the use of a new +minor version number do not always require changes to aufs-util. + +Since aufs-util has its own minor version number, you may not be +able to find a GIT branch in aufs-util for your kernel's +exact minor version number. +In this case, you should git-checkout the branch for the +nearest lower number. + +For (an unreleased) example: +If you are using "linux-3.10" and the "aufs3.10" branch +does not exist in aufs-util repository, then "aufs3.9", "aufs3.8" +or something numerically smaller is the branch for your kernel. + +Also you can view all branches by + $ git branch -a + + +3. Configuration and Compilation +---------------------------------------- +Make sure you have git-checkout'ed the correct branch. + +For aufs3-linux tree, +- enable CONFIG_AUFS_FS. +- set other aufs configurations if necessary. + +For aufs3-standalone tree, +There are several ways to build. + +1. +- apply ./aufs3-kbuild.patch to your kernel source files. +- apply ./aufs3-base.patch too. +- apply ./aufs3-mmap.patch too. +- apply ./aufs3-standalone.patch too, if you have a plan to set + CONFIG_AUFS_FS=m. otherwise you don't need ./aufs3-standalone.patch. +- copy ./{Documentation,fs,include/uapi/linux/aufs_type.h} files to your + kernel source tree. Never copy $PWD/include/uapi/linux/Kbuild. +- enable CONFIG_AUFS_FS, you can select either + =m or =y. +- and build your kernel as usual. +- install the built kernel. + Note: Since linux-3.9, every filesystem module requires an alias + "fs-". You should make sure that "fs-aufs" is listed in your + modules.aliases file if you set CONFIG_AUFS_FS=m. +- install the header files too by "make headers_install" to the + directory where you specify. By default, it is $PWD/usr. + "make help" shows a brief note for headers_install. +- and reboot your system. + +2. +- module only (CONFIG_AUFS_FS=m). +- apply ./aufs3-base.patch to your kernel source files. +- apply ./aufs3-mmap.patch too. +- apply ./aufs3-standalone.patch too. +- build your kernel, don't forget "make headers_install", and reboot. +- edit ./config.mk and set other aufs configurations if necessary. + Note: You should read $PWD/fs/aufs/Kconfig carefully which describes + every aufs configurations. +- build the module by simple "make". + Note: Since linux-3.9, every filesystem module requires an alias + "fs-". You should make sure that "fs-aufs" is listed in your + modules.aliases file. +- you can specify ${KDIR} make variable which points to your kernel + source tree. +- install the files + + run "make install" to install the aufs module, or copy the built + $PWD/aufs.ko to /lib/modules/... and run depmod -a (or reboot simply). + + run "make install_headers" (instead of headers_install) to install + the modified aufs header file (you can specify DESTDIR which is + available in aufs standalone version's Makefile only), or copy + $PWD/usr/include/linux/aufs_type.h to /usr/include/linux or wherever + you like manually. By default, the target directory is $PWD/usr. +- no need to apply aufs3-kbuild.patch, nor copying source files to your + kernel source tree. + +Note: The header file aufs_type.h is necessary to build aufs-util + as well as "make headers_install" in the kernel source tree. + headers_install is subject to be forgotten, but it is essentially + necessary, not only for building aufs-util. + You may not meet problems without headers_install in some older + version though. + +And then, +- read README in aufs-util, build and install it +- note that your distribution may contain an obsoleted version of + aufs_type.h in /usr/include/linux or something. When you build aufs + utilities, make sure that your compiler refers the correct aufs header + file which is built by "make headers_install." +- if you want to use readdir(3) in userspace or pathconf(3) wrapper, + then run "make install_ulib" too. And refer to the aufs manual in + detail. + +There several other patches in aufs3-standalone.git. They are all +optional. When you meet some problems, they will help you. +- aufs3-loopback.patch + Supports a nested loopback mount in a branch-fs. This patch is + unnecessary until aufs produces a message like "you may want to try + another patch for loopback file". +- vfs-ino.patch + Modifies a system global kernel internal function get_next_ino() in + order to stop assigning 0 for an inode-number. Not directly related to + aufs, but recommended generally. +- tmpfs-idr.patch + Keeps the tmpfs inode number as the lowest value. Effective to reduce + the size of aufs XINO files for tmpfs branch. Also it prevents the + duplication of inode number, which is important for backup tools and + other utilities. When you find aufs XINO files for tmpfs branch + growing too much, try this patch. + + +4. Usage +---------------------------------------- +At first, make sure aufs-util are installed, and please read the aufs +manual, aufs.5 in aufs-util.git tree. +$ man -l aufs.5 + +And then, +$ mkdir /tmp/rw /tmp/aufs +# mount -t aufs -o br=/tmp/rw:${HOME} none /tmp/aufs + +Here is another example. The result is equivalent. +# mount -t aufs -o br=/tmp/rw=rw:${HOME}=ro none /tmp/aufs + Or +# mount -t aufs -o br:/tmp/rw none /tmp/aufs +# mount -o remount,append:${HOME} /tmp/aufs + +Then, you can see whole tree of your home dir through /tmp/aufs. If +you modify a file under /tmp/aufs, the one on your home directory is +not affected, instead the same named file will be newly created under +/tmp/rw. And all of your modification to a file will be applied to +the one under /tmp/rw. This is called the file based Copy on Write +(COW) method. +Aufs mount options are described in aufs.5. +If you run chroot or something and make your aufs as a root directory, +then you need to customize the shutdown script. See the aufs manual in +detail. + +Additionally, there are some sample usages of aufs which are a +diskless system with network booting, and LiveCD over NFS. +See sample dir in CVS tree on SourceForge. + + +5. Contact +---------------------------------------- +When you have any problems or strange behaviour in aufs, please let me +know with: +- /proc/mounts (instead of the output of mount(8)) +- /sys/module/aufs/* +- /sys/fs/aufs/* (if you have them) +- /debug/aufs/* (if you have them) +- linux kernel version + if your kernel is not plain, for example modified by distributor, + the url where i can download its source is necessary too. +- aufs version which was printed at loading the module or booting the + system, instead of the date you downloaded. +- configuration (define/undefine CONFIG_AUFS_xxx) +- kernel configuration or /proc/config.gz (if you have it) +- behaviour which you think to be incorrect +- actual operation, reproducible one is better +- mailto: aufs-users at lists.sourceforge.net + +Usually, I don't watch the Public Areas(Bugs, Support Requests, Patches, +and Feature Requests) on SourceForge. Please join and write to +aufs-users ML. + + +6. Acknowledgements +---------------------------------------- +Thanks to everyone who have tried and are using aufs, whoever +have reported a bug or any feedback. + +Especially donators: +Tomas Matejicek(slax.org) made a donation (much more than once). + Since Apr 2010, Tomas M (the author of Slax and Linux Live + scripts) is making "doubling" donations. + Unfortunately I cannot list all of the donators, but I really + appreciate. + It ends Aug 2010, but the ordinary donation URL is still available. + +Dai Itasaka made a donation (2007/8). +Chuck Smith made a donation (2008/4, 10 and 12). +Henk Schoneveld made a donation (2008/9). +Chih-Wei Huang, ASUS, CTC donated Eee PC 4G (2008/10). +Francois Dupoux made a donation (2008/11). +Bruno Cesar Ribas and Luis Carlos Erpen de Bona, C3SL serves public + aufs2 GIT tree (2009/2). +William Grant made a donation (2009/3). +Patrick Lane made a donation (2009/4). +The Mail Archive (mail-archive.com) made donations (2009/5). +Nippy Networks (Ed Wildgoose) made a donation (2009/7). +New Dream Network, LLC (www.dreamhost.com) made a donation (2009/11). +Pavel Pronskiy made a donation (2011/2). +Iridium and Inmarsat satellite phone retailer (www.mailasail.com), Nippy + Networks (Ed Wildgoose) made a donation for hardware (2011/3). +Max Lekomcev (DOM-TV project) made a donation (2011/7, 12, 2012/3, 6 and +11). +Sam Liddicott made a donation (2011/9). +Era Scarecrow made a donation (2013/4). +Bor Ratajc made a donation (2013/4). +Alessandro Gorreta made a donation (2013/4). +POIRETTE Marc made a donation (2013/4). +Alessandro Gorreta made a donation (2013/4). +lauri kasvandik made a donation (2013/5). +"pemasu from Finland" made a donation (2013/7). +The Parted Magic Project made a donation (2013/9 and 11). +Pavel Barta made a donation (2013/10). +Nikolay Pertsev made a donation (2014/5). +James B made a donation (2014/7). +Stefano Di Biase made a donation (2014/8). + +Thank you very much. +Donations are always, including future donations, very important and +helpful for me to keep on developing aufs. + + +7. +---------------------------------------- +If you are an experienced user, no explanation is needed. Aufs is +just a linux filesystem. + + +Enjoy! + +# Local variables: ; +# mode: text; +# End: ; diff -Naur kernel.21.3.orig/fs/aufs/aufs.h kernel.21.3.new/fs/aufs/aufs.h --- kernel.21.3.orig/fs/aufs/aufs.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/aufs.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,59 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * all header files + */ + +#ifndef __AUFS_H__ +#define __AUFS_H__ + +#ifdef __KERNEL__ + +#define AuStub(type, name, body, ...) \ + static inline type name(__VA_ARGS__) { body; } + +#define AuStubVoid(name, ...) \ + AuStub(void, name, , __VA_ARGS__) +#define AuStubInt0(name, ...) \ + AuStub(int, name, return 0, __VA_ARGS__) + +#include "debug.h" + +#include "branch.h" +#include "cpup.h" +#include "dcsub.h" +#include "dbgaufs.h" +#include "dentry.h" +#include "dir.h" +#include "dynop.h" +#include "file.h" +#include "fstype.h" +#include "inode.h" +#include "loop.h" +#include "module.h" +#include "opts.h" +#include "rwsem.h" +#include "spl.h" +#include "super.h" +#include "sysaufs.h" +#include "vfsub.h" +#include "whout.h" +#include "wkq.h" + +#endif /* __KERNEL__ */ +#endif /* __AUFS_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/branch.c kernel.21.3.new/fs/aufs/branch.c --- kernel.21.3.orig/fs/aufs/branch.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/branch.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,1436 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * branch management + */ + +#include +#include +#include "aufs.h" + +/* + * free a single branch + */ + +/* prohibit rmdir to the root of the branch */ +/* todo: another new flag? */ +static void au_br_dflags_force(struct au_branch *br) +{ + struct dentry *h_dentry; + + h_dentry = au_br_dentry(br); + spin_lock(&h_dentry->d_lock); + br->br_dflags = h_dentry->d_flags & DCACHE_MOUNTED; + h_dentry->d_flags |= DCACHE_MOUNTED; + spin_unlock(&h_dentry->d_lock); +} + +/* restore its d_flags */ +static void au_br_dflags_restore(struct au_branch *br) +{ + struct dentry *h_dentry; + + if (br->br_dflags) + return; + + h_dentry = au_br_dentry(br); + spin_lock(&h_dentry->d_lock); + h_dentry->d_flags &= ~DCACHE_MOUNTED; + spin_unlock(&h_dentry->d_lock); +} + +static void au_br_do_free(struct au_branch *br) +{ + int i; + struct au_wbr *wbr; + struct au_dykey **key; + + au_hnotify_fin_br(br); + + if (br->br_xino.xi_file) + fput(br->br_xino.xi_file); + mutex_destroy(&br->br_xino.xi_nondir_mtx); + + AuDebugOn(atomic_read(&br->br_count)); + + wbr = br->br_wbr; + if (wbr) { + for (i = 0; i < AuBrWh_Last; i++) + dput(wbr->wbr_wh[i]); + AuDebugOn(atomic_read(&wbr->wbr_wh_running)); + AuRwDestroy(&wbr->wbr_wh_rwsem); + } + + if (br->br_fhsm) { + au_br_fhsm_fin(br->br_fhsm); + kfree(br->br_fhsm); + } + + key = br->br_dykey; + for (i = 0; i < AuBrDynOp; i++, key++) + if (*key) + au_dy_put(*key); + else + break; + + au_br_dflags_restore(br); + + /* recursive lock, s_umount of branch's */ + lockdep_off(); + path_put(&br->br_path); + lockdep_on(); + kfree(wbr); + kfree(br); +} + +/* + * frees all branches + */ +void au_br_free(struct au_sbinfo *sbinfo) +{ + aufs_bindex_t bmax; + struct au_branch **br; + + AuRwMustWriteLock(&sbinfo->si_rwsem); + + bmax = sbinfo->si_bend + 1; + br = sbinfo->si_branch; + while (bmax--) + au_br_do_free(*br++); +} + +/* + * find the index of a branch which is specified by @br_id. + */ +int au_br_index(struct super_block *sb, aufs_bindex_t br_id) +{ + aufs_bindex_t bindex, bend; + + bend = au_sbend(sb); + for (bindex = 0; bindex <= bend; bindex++) + if (au_sbr_id(sb, bindex) == br_id) + return bindex; + return -1; +} + +/* ---------------------------------------------------------------------- */ + +/* + * add a branch + */ + +static int test_overlap(struct super_block *sb, struct dentry *h_adding, + struct dentry *h_root) +{ + if (unlikely(h_adding == h_root + || au_test_loopback_overlap(sb, h_adding))) + return 1; + if (h_adding->d_sb != h_root->d_sb) + return 0; + return au_test_subdir(h_adding, h_root) + || au_test_subdir(h_root, h_adding); +} + +/* + * returns a newly allocated branch. @new_nbranch is a number of branches + * after adding a branch. + */ +static struct au_branch *au_br_alloc(struct super_block *sb, int new_nbranch, + int perm) +{ + struct au_branch *add_branch; + struct dentry *root; + int err; + + err = -ENOMEM; + root = sb->s_root; + add_branch = kmalloc(sizeof(*add_branch), GFP_NOFS); + if (unlikely(!add_branch)) + goto out; + + err = au_hnotify_init_br(add_branch, perm); + if (unlikely(err)) + goto out_br; + + add_branch->br_wbr = NULL; + if (au_br_writable(perm)) { + /* may be freed separately at changing the branch permission */ + add_branch->br_wbr = kmalloc(sizeof(*add_branch->br_wbr), + GFP_NOFS); + if (unlikely(!add_branch->br_wbr)) + goto out_hnotify; + } + + add_branch->br_fhsm = NULL; + if (au_br_fhsm(perm)) { + err = au_fhsm_br_alloc(add_branch); + if (unlikely(err)) + goto out_wbr; + } + + err = au_sbr_realloc(au_sbi(sb), new_nbranch); + if (!err) + err = au_di_realloc(au_di(root), new_nbranch); + if (!err) + err = au_ii_realloc(au_ii(root->d_inode), new_nbranch); + if (!err) + return add_branch; /* success */ + +out_wbr: + kfree(add_branch->br_wbr); +out_hnotify: + au_hnotify_fin_br(add_branch); +out_br: + kfree(add_branch); +out: + return ERR_PTR(err); +} + +/* + * test if the branch permission is legal or not. + */ +static int test_br(struct inode *inode, int brperm, char *path) +{ + int err; + + err = (au_br_writable(brperm) && IS_RDONLY(inode)); + if (!err) + goto out; + + err = -EINVAL; + pr_err("write permission for readonly mount or inode, %s\n", path); + +out: + return err; +} + +/* + * returns: + * 0: success, the caller will add it + * plus: success, it is already unified, the caller should ignore it + * minus: error + */ +static int test_add(struct super_block *sb, struct au_opt_add *add, int remount) +{ + int err; + aufs_bindex_t bend, bindex; + struct dentry *root; + struct inode *inode, *h_inode; + + root = sb->s_root; + bend = au_sbend(sb); + if (unlikely(bend >= 0 + && au_find_dbindex(root, add->path.dentry) >= 0)) { + err = 1; + if (!remount) { + err = -EINVAL; + pr_err("%s duplicated\n", add->pathname); + } + goto out; + } + + err = -ENOSPC; /* -E2BIG; */ + if (unlikely(AUFS_BRANCH_MAX <= add->bindex + || AUFS_BRANCH_MAX - 1 <= bend)) { + pr_err("number of branches exceeded %s\n", add->pathname); + goto out; + } + + err = -EDOM; + if (unlikely(add->bindex < 0 || bend + 1 < add->bindex)) { + pr_err("bad index %d\n", add->bindex); + goto out; + } + + inode = add->path.dentry->d_inode; + err = -ENOENT; + if (unlikely(!inode->i_nlink)) { + pr_err("no existence %s\n", add->pathname); + goto out; + } + + err = -EINVAL; + if (unlikely(inode->i_sb == sb)) { + pr_err("%s must be outside\n", add->pathname); + goto out; + } + + if (unlikely(au_test_fs_unsuppoted(inode->i_sb))) { + pr_err("unsupported filesystem, %s (%s)\n", + add->pathname, au_sbtype(inode->i_sb)); + goto out; + } + + err = test_br(add->path.dentry->d_inode, add->perm, add->pathname); + if (unlikely(err)) + goto out; + + if (bend < 0) + return 0; /* success */ + + err = -EINVAL; + for (bindex = 0; bindex <= bend; bindex++) + if (unlikely(test_overlap(sb, add->path.dentry, + au_h_dptr(root, bindex)))) { + pr_err("%s is overlapped\n", add->pathname); + goto out; + } + + err = 0; + if (au_opt_test(au_mntflags(sb), WARN_PERM)) { + h_inode = au_h_dptr(root, 0)->d_inode; + if ((h_inode->i_mode & S_IALLUGO) != (inode->i_mode & S_IALLUGO) + || !uid_eq(h_inode->i_uid, inode->i_uid) + || !gid_eq(h_inode->i_gid, inode->i_gid)) + pr_warn("uid/gid/perm %s %u/%u/0%o, %u/%u/0%o\n", + add->pathname, + i_uid_read(inode), i_gid_read(inode), + (inode->i_mode & S_IALLUGO), + i_uid_read(h_inode), i_gid_read(h_inode), + (h_inode->i_mode & S_IALLUGO)); + } + +out: + return err; +} + +/* + * initialize or clean the whiteouts for an adding branch + */ +static int au_br_init_wh(struct super_block *sb, struct au_branch *br, + int new_perm) +{ + int err, old_perm; + aufs_bindex_t bindex; + struct mutex *h_mtx; + struct au_wbr *wbr; + struct au_hinode *hdir; + + err = vfsub_mnt_want_write(au_br_mnt(br)); + if (unlikely(err)) + goto out; + + wbr = br->br_wbr; + old_perm = br->br_perm; + br->br_perm = new_perm; + hdir = NULL; + h_mtx = NULL; + bindex = au_br_index(sb, br->br_id); + if (0 <= bindex) { + hdir = au_hi(sb->s_root->d_inode, bindex); + au_hn_imtx_lock_nested(hdir, AuLsc_I_PARENT); + } else { + h_mtx = &au_br_dentry(br)->d_inode->i_mutex; + mutex_lock_nested(h_mtx, AuLsc_I_PARENT); + } + if (!wbr) + err = au_wh_init(br, sb); + else { + wbr_wh_write_lock(wbr); + err = au_wh_init(br, sb); + wbr_wh_write_unlock(wbr); + } + if (hdir) + au_hn_imtx_unlock(hdir); + else + mutex_unlock(h_mtx); + vfsub_mnt_drop_write(au_br_mnt(br)); + br->br_perm = old_perm; + + if (!err && wbr && !au_br_writable(new_perm)) { + kfree(wbr); + br->br_wbr = NULL; + } + +out: + return err; +} + +static int au_wbr_init(struct au_branch *br, struct super_block *sb, + int perm) +{ + int err; + struct kstatfs kst; + struct au_wbr *wbr; + + wbr = br->br_wbr; + au_rw_init(&wbr->wbr_wh_rwsem); + memset(wbr->wbr_wh, 0, sizeof(wbr->wbr_wh)); + atomic_set(&wbr->wbr_wh_running, 0); + wbr->wbr_bytes = 0; + + /* + * a limit for rmdir/rename a dir + * cf. AUFS_MAX_NAMELEN in include/uapi/linux/aufs_type.h + */ + err = vfs_statfs(&br->br_path, &kst); + if (unlikely(err)) + goto out; + err = -EINVAL; + if (kst.f_namelen >= NAME_MAX) + err = au_br_init_wh(sb, br, perm); + else + pr_err("%.*s(%s), unsupported namelen %ld\n", + AuDLNPair(au_br_dentry(br)), + au_sbtype(au_br_dentry(br)->d_sb), kst.f_namelen); + +out: + return err; +} + +/* initialize a new branch */ +static int au_br_init(struct au_branch *br, struct super_block *sb, + struct au_opt_add *add) +{ + int err; + + err = 0; + memset(&br->br_xino, 0, sizeof(br->br_xino)); + mutex_init(&br->br_xino.xi_nondir_mtx); + br->br_perm = add->perm; + BUILD_BUG_ON(sizeof(br->br_dflags) + != sizeof(br->br_path.dentry->d_flags)); + br->br_dflags = DCACHE_MOUNTED; + br->br_path = add->path; /* set first, path_get() later */ + spin_lock_init(&br->br_dykey_lock); + memset(br->br_dykey, 0, sizeof(br->br_dykey)); + atomic_set(&br->br_count, 0); + atomic_set(&br->br_xino_running, 0); + br->br_id = au_new_br_id(sb); + AuDebugOn(br->br_id < 0); + + if (au_br_writable(add->perm)) { + err = au_wbr_init(br, sb, add->perm); + if (unlikely(err)) + goto out_err; + } + + if (au_opt_test(au_mntflags(sb), XINO)) { + err = au_xino_br(sb, br, add->path.dentry->d_inode->i_ino, + au_sbr(sb, 0)->br_xino.xi_file, /*do_test*/1); + if (unlikely(err)) { + AuDebugOn(br->br_xino.xi_file); + goto out_err; + } + } + + sysaufs_br_init(br); + path_get(&br->br_path); + goto out; /* success */ + +out_err: + memset(&br->br_path, 0, sizeof(br->br_path)); +out: + return err; +} + +static void au_br_do_add_brp(struct au_sbinfo *sbinfo, aufs_bindex_t bindex, + struct au_branch *br, aufs_bindex_t bend, + aufs_bindex_t amount) +{ + struct au_branch **brp; + + AuRwMustWriteLock(&sbinfo->si_rwsem); + + brp = sbinfo->si_branch + bindex; + memmove(brp + 1, brp, sizeof(*brp) * amount); + *brp = br; + sbinfo->si_bend++; + if (unlikely(bend < 0)) + sbinfo->si_bend = 0; +} + +static void au_br_do_add_hdp(struct au_dinfo *dinfo, aufs_bindex_t bindex, + aufs_bindex_t bend, aufs_bindex_t amount) +{ + struct au_hdentry *hdp; + + AuRwMustWriteLock(&dinfo->di_rwsem); + + hdp = dinfo->di_hdentry + bindex; + memmove(hdp + 1, hdp, sizeof(*hdp) * amount); + au_h_dentry_init(hdp); + dinfo->di_bend++; + if (unlikely(bend < 0)) + dinfo->di_bstart = 0; +} + +static void au_br_do_add_hip(struct au_iinfo *iinfo, aufs_bindex_t bindex, + aufs_bindex_t bend, aufs_bindex_t amount) +{ + struct au_hinode *hip; + + AuRwMustWriteLock(&iinfo->ii_rwsem); + + hip = iinfo->ii_hinode + bindex; + memmove(hip + 1, hip, sizeof(*hip) * amount); + hip->hi_inode = NULL; + au_hn_init(hip); + iinfo->ii_bend++; + if (unlikely(bend < 0)) + iinfo->ii_bstart = 0; +} + +static void au_br_do_add(struct super_block *sb, struct au_branch *br, + aufs_bindex_t bindex) +{ + struct dentry *root, *h_dentry; + struct inode *root_inode; + aufs_bindex_t bend, amount; + + au_br_dflags_force(br); + + root = sb->s_root; + root_inode = root->d_inode; + bend = au_sbend(sb); + amount = bend + 1 - bindex; + h_dentry = au_br_dentry(br); + au_sbilist_lock(); + au_br_do_add_brp(au_sbi(sb), bindex, br, bend, amount); + au_br_do_add_hdp(au_di(root), bindex, bend, amount); + au_br_do_add_hip(au_ii(root_inode), bindex, bend, amount); + au_set_h_dptr(root, bindex, dget(h_dentry)); + au_set_h_iptr(root_inode, bindex, au_igrab(h_dentry->d_inode), + /*flags*/0); + au_sbilist_unlock(); +} + +int au_br_add(struct super_block *sb, struct au_opt_add *add, int remount) +{ + int err; + aufs_bindex_t bend, add_bindex; + struct dentry *root, *h_dentry; + struct inode *root_inode; + struct au_branch *add_branch; + + root = sb->s_root; + root_inode = root->d_inode; + IMustLock(root_inode); + err = test_add(sb, add, remount); + if (unlikely(err < 0)) + goto out; + if (err) { + err = 0; + goto out; /* success */ + } + + bend = au_sbend(sb); + add_branch = au_br_alloc(sb, bend + 2, add->perm); + err = PTR_ERR(add_branch); + if (IS_ERR(add_branch)) + goto out; + + err = au_br_init(add_branch, sb, add); + if (unlikely(err)) { + au_br_do_free(add_branch); + goto out; + } + + add_bindex = add->bindex; + if (!remount) + au_br_do_add(sb, add_branch, add_bindex); + else { + sysaufs_brs_del(sb, add_bindex); + au_br_do_add(sb, add_branch, add_bindex); + sysaufs_brs_add(sb, add_bindex); + } + + h_dentry = add->path.dentry; + if (!add_bindex) { + au_cpup_attr_all(root_inode, /*force*/1); + sb->s_maxbytes = h_dentry->d_sb->s_maxbytes; + } else + au_add_nlink(root_inode, h_dentry->d_inode); + + /* + * this test/set prevents aufs from handling unnecesary notify events + * of xino files, in case of re-adding a writable branch which was + * once detached from aufs. + */ + if (au_xino_brid(sb) < 0 + && au_br_writable(add_branch->br_perm) + && !au_test_fs_bad_xino(h_dentry->d_sb) + && add_branch->br_xino.xi_file + && add_branch->br_xino.xi_file->f_dentry->d_parent == h_dentry) + au_xino_brid_set(sb, add_branch->br_id); + +out: + return err; +} + +/* ---------------------------------------------------------------------- */ + +static unsigned long long au_farray_cb(void *a, + unsigned long long max __maybe_unused, + void *arg) +{ + unsigned long long n; + struct file **p, *f; + struct super_block *sb = arg; + + n = 0; + p = a; + lg_global_lock(&files_lglock); + do_file_list_for_each_entry(sb, f) { + if (au_fi(f) + && file_count(f) + && !special_file(file_inode(f)->i_mode)) { + get_file(f); + *p++ = f; + n++; + AuDebugOn(n > max); + } + } while_file_list_for_each_entry; + lg_global_unlock(&files_lglock); + + return n; +} + +static struct file **au_farray_alloc(struct super_block *sb, + unsigned long long *max) +{ + *max = atomic_long_read(&au_sbi(sb)->si_nfiles); + return au_array_alloc(max, au_farray_cb, sb); +} + +static void au_farray_free(struct file **a, unsigned long long max) +{ + unsigned long long ull; + + for (ull = 0; ull < max; ull++) + if (a[ull]) + fput(a[ull]); + au_array_free(a); +} + +/* ---------------------------------------------------------------------- */ + +/* + * delete a branch + */ + +/* to show the line number, do not make it inlined function */ +#define AuVerbose(do_info, fmt, ...) do { \ + if (do_info) \ + pr_info(fmt, ##__VA_ARGS__); \ +} while (0) + +static int au_test_ibusy(struct inode *inode, aufs_bindex_t bstart, + aufs_bindex_t bend) +{ + return (inode && !S_ISDIR(inode->i_mode)) || bstart == bend; +} + +static int au_test_dbusy(struct dentry *dentry, aufs_bindex_t bstart, + aufs_bindex_t bend) +{ + return au_test_ibusy(dentry->d_inode, bstart, bend); +} + +/* + * test if the branch is deletable or not. + */ +static int test_dentry_busy(struct dentry *root, aufs_bindex_t bindex, + unsigned int sigen, const unsigned int verbose) +{ + int err, i, j, ndentry; + aufs_bindex_t bstart, bend; + struct au_dcsub_pages dpages; + struct au_dpage *dpage; + struct dentry *d; + + err = au_dpages_init(&dpages, GFP_NOFS); + if (unlikely(err)) + goto out; + err = au_dcsub_pages(&dpages, root, NULL, NULL); + if (unlikely(err)) + goto out_dpages; + + for (i = 0; !err && i < dpages.ndpage; i++) { + dpage = dpages.dpages + i; + ndentry = dpage->ndentry; + for (j = 0; !err && j < ndentry; j++) { + d = dpage->dentries[j]; + AuDebugOn(!d->d_count); + if (!au_digen_test(d, sigen)) { + di_read_lock_child(d, AuLock_IR); + if (unlikely(au_dbrange_test(d))) { + di_read_unlock(d, AuLock_IR); + continue; + } + } else { + di_write_lock_child(d); + if (unlikely(au_dbrange_test(d))) { + di_write_unlock(d); + continue; + } + err = au_reval_dpath(d, sigen); + if (!err) + di_downgrade_lock(d, AuLock_IR); + else { + di_write_unlock(d); + break; + } + } + + /* AuDbgDentry(d); */ + bstart = au_dbstart(d); + bend = au_dbend(d); + if (bstart <= bindex + && bindex <= bend + && au_h_dptr(d, bindex) + && au_test_dbusy(d, bstart, bend)) { + err = -EBUSY; + AuVerbose(verbose, "busy %.*s\n", AuDLNPair(d)); + AuDbgDentry(d); + } + di_read_unlock(d, AuLock_IR); + } + } + +out_dpages: + au_dpages_free(&dpages); +out: + return err; +} + +static int test_inode_busy(struct super_block *sb, aufs_bindex_t bindex, + unsigned int sigen, const unsigned int verbose) +{ + int err; + unsigned long long max, ull; + struct inode *i, **array; + aufs_bindex_t bstart, bend; + + array = au_iarray_alloc(sb, &max); + err = PTR_ERR(array); + if (IS_ERR(array)) + goto out; + + err = 0; + AuDbg("b%d\n", bindex); + for (ull = 0; !err && ull < max; ull++) { + i = array[ull]; + if (unlikely(!i)) + break; + if (i->i_ino == AUFS_ROOT_INO) + continue; + + /* AuDbgInode(i); */ + if (au_iigen(i, NULL) == sigen) + ii_read_lock_child(i); + else { + ii_write_lock_child(i); + err = au_refresh_hinode_self(i); + au_iigen_dec(i); + if (!err) + ii_downgrade_lock(i); + else { + ii_write_unlock(i); + break; + } + } + + bstart = au_ibstart(i); + bend = au_ibend(i); + if (bstart <= bindex + && bindex <= bend + && au_h_iptr(i, bindex) + && au_test_ibusy(i, bstart, bend)) { + err = -EBUSY; + AuVerbose(verbose, "busy i%lu\n", i->i_ino); + AuDbgInode(i); + } + ii_read_unlock(i); + } + au_iarray_free(array, max); + +out: + return err; +} + +static int test_children_busy(struct dentry *root, aufs_bindex_t bindex, + const unsigned int verbose) +{ + int err; + unsigned int sigen; + + sigen = au_sigen(root->d_sb); + DiMustNoWaiters(root); + IiMustNoWaiters(root->d_inode); + di_write_unlock(root); + err = test_dentry_busy(root, bindex, sigen, verbose); + if (!err) + err = test_inode_busy(root->d_sb, bindex, sigen, verbose); + di_write_lock_child(root); /* aufs_write_lock() calls ..._child() */ + + return err; +} + +static int test_dir_busy(struct file *file, aufs_bindex_t br_id, + struct file **to_free, int *idx) +{ + int err; + unsigned char matched, root; + aufs_bindex_t bindex, bend; + struct au_fidir *fidir; + struct au_hfile *hfile; + + err = 0; + root = IS_ROOT(file->f_dentry); + if (root) { + get_file(file); + to_free[*idx] = file; + (*idx)++; + goto out; + } + + matched = 0; + fidir = au_fi(file)->fi_hdir; + AuDebugOn(!fidir); + bend = au_fbend_dir(file); + for (bindex = au_fbstart(file); bindex <= bend; bindex++) { + hfile = fidir->fd_hfile + bindex; + if (!hfile->hf_file) + continue; + + if (hfile->hf_br->br_id == br_id) { + matched = 1; + break; + } + } + if (matched) + err = -EBUSY; + +out: + return err; +} + +static int test_file_busy(struct super_block *sb, aufs_bindex_t br_id, + struct file **to_free, int opened) +{ + int err, idx; + unsigned long long ull, max; + aufs_bindex_t bstart; + struct file *file, **array; + struct inode *inode; + struct dentry *root; + struct au_hfile *hfile; + + array = au_farray_alloc(sb, &max); + err = PTR_ERR(array); + if (IS_ERR(array)) + goto out; + + err = 0; + idx = 0; + root = sb->s_root; + di_write_unlock(root); + for (ull = 0; ull < max; ull++) { + file = array[ull]; + if (unlikely(!file)) + break; + + /* AuDbg("%.*s\n", AuDLNPair(file->f_dentry)); */ + fi_read_lock(file); + bstart = au_fbstart(file); + inode = file_inode(file); + if (!S_ISDIR(inode->i_mode)) { + hfile = &au_fi(file)->fi_htop; + if (hfile->hf_br->br_id == br_id) + err = -EBUSY; + } else + err = test_dir_busy(file, br_id, to_free, &idx); + fi_read_unlock(file); + if (unlikely(err)) + break; + } + di_write_lock_child(root); + au_farray_free(array, max); + AuDebugOn(idx > opened); + +out: + return err; +} + +static void br_del_file(struct file **to_free, unsigned long long opened, + aufs_bindex_t br_id) +{ + unsigned long long ull; + aufs_bindex_t bindex, bstart, bend, bfound; + struct file *file; + struct au_fidir *fidir; + struct au_hfile *hfile; + + for (ull = 0; ull < opened; ull++) { + file = to_free[ull]; + if (unlikely(!file)) + break; + + /* AuDbg("%.*s\n", AuDLNPair(file->f_dentry)); */ + AuDebugOn(!S_ISDIR(file_inode(file)->i_mode)); + bfound = -1; + fidir = au_fi(file)->fi_hdir; + AuDebugOn(!fidir); + fi_write_lock(file); + bstart = au_fbstart(file); + bend = au_fbend_dir(file); + for (bindex = bstart; bindex <= bend; bindex++) { + hfile = fidir->fd_hfile + bindex; + if (!hfile->hf_file) + continue; + + if (hfile->hf_br->br_id == br_id) { + bfound = bindex; + break; + } + } + AuDebugOn(bfound < 0); + au_set_h_fptr(file, bfound, NULL); + if (bfound == bstart) { + for (bstart++; bstart <= bend; bstart++) + if (au_hf_dir(file, bstart)) { + au_set_fbstart(file, bstart); + break; + } + } + fi_write_unlock(file); + } +} + +static void au_br_do_del_brp(struct au_sbinfo *sbinfo, + const aufs_bindex_t bindex, + const aufs_bindex_t bend) +{ + struct au_branch **brp, **p; + + AuRwMustWriteLock(&sbinfo->si_rwsem); + + brp = sbinfo->si_branch + bindex; + if (bindex < bend) + memmove(brp, brp + 1, sizeof(*brp) * (bend - bindex)); + sbinfo->si_branch[0 + bend] = NULL; + sbinfo->si_bend--; + + p = krealloc(sbinfo->si_branch, sizeof(*p) * bend, AuGFP_SBILIST); + if (p) + sbinfo->si_branch = p; + /* harmless error */ +} + +static void au_br_do_del_hdp(struct au_dinfo *dinfo, const aufs_bindex_t bindex, + const aufs_bindex_t bend) +{ + struct au_hdentry *hdp, *p; + + AuRwMustWriteLock(&dinfo->di_rwsem); + + hdp = dinfo->di_hdentry; + if (bindex < bend) + memmove(hdp + bindex, hdp + bindex + 1, + sizeof(*hdp) * (bend - bindex)); + hdp[0 + bend].hd_dentry = NULL; + dinfo->di_bend--; + + p = krealloc(hdp, sizeof(*p) * bend, AuGFP_SBILIST); + if (p) + dinfo->di_hdentry = p; + /* harmless error */ +} + +static void au_br_do_del_hip(struct au_iinfo *iinfo, const aufs_bindex_t bindex, + const aufs_bindex_t bend) +{ + struct au_hinode *hip, *p; + + AuRwMustWriteLock(&iinfo->ii_rwsem); + + hip = iinfo->ii_hinode + bindex; + if (bindex < bend) + memmove(hip, hip + 1, sizeof(*hip) * (bend - bindex)); + iinfo->ii_hinode[0 + bend].hi_inode = NULL; + au_hn_init(iinfo->ii_hinode + bend); + iinfo->ii_bend--; + + p = krealloc(iinfo->ii_hinode, sizeof(*p) * bend, AuGFP_SBILIST); + if (p) + iinfo->ii_hinode = p; + /* harmless error */ +} + +static void au_br_do_del(struct super_block *sb, aufs_bindex_t bindex, + struct au_branch *br) +{ + aufs_bindex_t bend; + struct au_sbinfo *sbinfo; + struct dentry *root, *h_root; + struct inode *inode, *h_inode; + struct au_hinode *hinode; + + SiMustWriteLock(sb); + + root = sb->s_root; + inode = root->d_inode; + sbinfo = au_sbi(sb); + bend = sbinfo->si_bend; + + h_root = au_h_dptr(root, bindex); + hinode = au_hi(inode, bindex); + h_inode = au_igrab(hinode->hi_inode); + au_hiput(hinode); + + au_sbilist_lock(); + au_br_do_del_brp(sbinfo, bindex, bend); + au_br_do_del_hdp(au_di(root), bindex, bend); + au_br_do_del_hip(au_ii(inode), bindex, bend); + au_sbilist_unlock(); + + dput(h_root); + iput(h_inode); + au_br_do_free(br); +} + +static unsigned long long empty_cb(void *array, unsigned long long max, + void *arg) +{ + return max; +} + +int au_br_del(struct super_block *sb, struct au_opt_del *del, int remount) +{ + int err, rerr, i; + unsigned long long opened; + unsigned int mnt_flags; + aufs_bindex_t bindex, bend, br_id; + unsigned char do_wh, verbose; + struct au_branch *br; + struct au_wbr *wbr; + struct dentry *root; + struct file **to_free; + + err = 0; + opened = 0; + to_free = NULL; + root = sb->s_root; + bindex = au_find_dbindex(root, del->h_path.dentry); + if (bindex < 0) { + if (remount) + goto out; /* success */ + err = -ENOENT; + pr_err("%s no such branch\n", del->pathname); + goto out; + } + AuDbg("bindex b%d\n", bindex); + + err = -EBUSY; + mnt_flags = au_mntflags(sb); + verbose = !!au_opt_test(mnt_flags, VERBOSE); + bend = au_sbend(sb); + if (unlikely(!bend)) { + AuVerbose(verbose, "no more branches left\n"); + goto out; + } + br = au_sbr(sb, bindex); + AuDebugOn(!path_equal(&br->br_path, &del->h_path)); + + br_id = br->br_id; + opened = atomic_read(&br->br_count); + if (unlikely(opened)) { + to_free = au_array_alloc(&opened, empty_cb, NULL); + err = PTR_ERR(to_free); + if (IS_ERR(to_free)) + goto out; + + err = test_file_busy(sb, br_id, to_free, opened); + if (unlikely(err)) { + AuVerbose(verbose, "%llu file(s) opened\n", opened); + goto out; + } + } + + wbr = br->br_wbr; + do_wh = wbr && (wbr->wbr_whbase || wbr->wbr_plink || wbr->wbr_orph); + if (do_wh) { + /* instead of WbrWhMustWriteLock(wbr) */ + SiMustWriteLock(sb); + for (i = 0; i < AuBrWh_Last; i++) { + dput(wbr->wbr_wh[i]); + wbr->wbr_wh[i] = NULL; + } + } + + err = test_children_busy(root, bindex, verbose); + if (unlikely(err)) { + if (do_wh) + goto out_wh; + goto out; + } + + err = 0; + if (to_free) { + /* + * now we confirmed the branch is deletable. + * let's free the remaining opened dirs on the branch. + */ + di_write_unlock(root); + br_del_file(to_free, opened, br_id); + di_write_lock_child(root); + } + + if (!remount) + au_br_do_del(sb, bindex, br); + else { + sysaufs_brs_del(sb, bindex); + au_br_do_del(sb, bindex, br); + sysaufs_brs_add(sb, bindex); + } + + if (!bindex) { + au_cpup_attr_all(root->d_inode, /*force*/1); + sb->s_maxbytes = au_sbr_sb(sb, 0)->s_maxbytes; + } else + au_sub_nlink(root->d_inode, del->h_path.dentry->d_inode); + if (au_opt_test(mnt_flags, PLINK)) + au_plink_half_refresh(sb, br_id); + + if (au_xino_brid(sb) == br_id) + au_xino_brid_set(sb, -1); + goto out; /* success */ + +out_wh: + /* revert */ + rerr = au_br_init_wh(sb, br, br->br_perm); + if (rerr) + pr_warn("failed re-creating base whiteout, %s. (%d)\n", + del->pathname, rerr); +out: + if (to_free) + au_farray_free(to_free, opened); + return err; +} + +/* ---------------------------------------------------------------------- */ + +static int au_ibusy(struct super_block *sb, struct aufs_ibusy __user *arg) +{ + int err; + aufs_bindex_t bstart, bend; + struct aufs_ibusy ibusy; + struct inode *inode, *h_inode; + + err = -EPERM; + if (unlikely(!capable(CAP_SYS_ADMIN))) + goto out; + + err = copy_from_user(&ibusy, arg, sizeof(ibusy)); + if (!err) + err = !access_ok(VERIFY_WRITE, &arg->h_ino, sizeof(arg->h_ino)); + if (unlikely(err)) { + err = -EFAULT; + AuTraceErr(err); + goto out; + } + + err = -EINVAL; + si_read_lock(sb, AuLock_FLUSH); + if (unlikely(ibusy.bindex < 0 || ibusy.bindex > au_sbend(sb))) + goto out_unlock; + + err = 0; + ibusy.h_ino = 0; /* invalid */ + inode = ilookup(sb, ibusy.ino); + if (!inode + || inode->i_ino == AUFS_ROOT_INO + || is_bad_inode(inode)) + goto out_unlock; + + ii_read_lock_child(inode); + bstart = au_ibstart(inode); + bend = au_ibend(inode); + if (bstart <= ibusy.bindex && ibusy.bindex <= bend) { + h_inode = au_h_iptr(inode, ibusy.bindex); + if (h_inode && au_test_ibusy(inode, bstart, bend)) + ibusy.h_ino = h_inode->i_ino; + } + ii_read_unlock(inode); + iput(inode); + +out_unlock: + si_read_unlock(sb); + if (!err) { + err = __put_user(ibusy.h_ino, &arg->h_ino); + if (unlikely(err)) { + err = -EFAULT; + AuTraceErr(err); + } + } +out: + return err; +} + +long au_ibusy_ioctl(struct file *file, unsigned long arg) +{ + return au_ibusy(file->f_dentry->d_sb, (void __user *)arg); +} + +#ifdef CONFIG_COMPAT +long au_ibusy_compat_ioctl(struct file *file, unsigned long arg) +{ + return au_ibusy(file->f_dentry->d_sb, compat_ptr(arg)); +} +#endif + +/* ---------------------------------------------------------------------- */ + +/* + * change a branch permission + */ + +static void au_warn_ima(void) +{ +#ifdef CONFIG_IMA + /* since it doesn't support mark_files_ro() */ + AuWarn1("RW -> RO makes IMA to produce wrong message\n"); +#endif +} + +static int do_need_sigen_inc(int a, int b) +{ + return au_br_whable(a) && !au_br_whable(b); +} + +static int need_sigen_inc(int old, int new) +{ + return do_need_sigen_inc(old, new) + || do_need_sigen_inc(new, old); +} + +static int au_br_mod_files_ro(struct super_block *sb, aufs_bindex_t bindex) +{ + int err, do_warn; + unsigned int mnt_flags; + unsigned long long ull, max; + aufs_bindex_t br_id; + unsigned char verbose; + struct file *file, *hf, **array; + struct inode *inode; + struct au_hfile *hfile; + + mnt_flags = au_mntflags(sb); + verbose = !!au_opt_test(mnt_flags, VERBOSE); + + array = au_farray_alloc(sb, &max); + err = PTR_ERR(array); + if (IS_ERR(array)) + goto out; + + do_warn = 0; + br_id = au_sbr_id(sb, bindex); + for (ull = 0; ull < max; ull++) { + file = array[ull]; + if (unlikely(!file)) + break; + + /* AuDbg("%.*s\n", AuDLNPair(file->f_dentry)); */ + fi_read_lock(file); + if (unlikely(au_test_mmapped(file))) { + err = -EBUSY; + AuVerbose(verbose, "mmapped %.*s\n", + AuDLNPair(file->f_dentry)); + AuDbgFile(file); + FiMustNoWaiters(file); + fi_read_unlock(file); + goto out_array; + } + + inode = file_inode(file); + hfile = &au_fi(file)->fi_htop; + hf = hfile->hf_file; + if (!S_ISREG(inode->i_mode) + || !(file->f_mode & FMODE_WRITE) + || hfile->hf_br->br_id != br_id + || !(hf->f_mode & FMODE_WRITE)) + array[ull] = NULL; + else { + do_warn = 1; + get_file(file); + } + + FiMustNoWaiters(file); + fi_read_unlock(file); + fput(file); + } + + err = 0; + if (do_warn) + au_warn_ima(); + + for (ull = 0; ull < max; ull++) { + file = array[ull]; + if (!file) + continue; + + /* todo: already flushed? */ + /* cf. fs/super.c:mark_files_ro() */ + /* fi_read_lock(file); */ + hfile = &au_fi(file)->fi_htop; + hf = hfile->hf_file; + /* fi_read_unlock(file); */ + spin_lock(&hf->f_lock); + hf->f_mode &= ~FMODE_WRITE; + spin_unlock(&hf->f_lock); + if (!file_check_writeable(hf)) { + __mnt_drop_write(hf->f_path.mnt); + file_release_write(hf); + } + } + +out_array: + au_farray_free(array, max); +out: + AuTraceErr(err); + return err; +} + +int au_br_mod(struct super_block *sb, struct au_opt_mod *mod, int remount, + int *do_refresh) +{ + int err, rerr; + aufs_bindex_t bindex; + struct dentry *root; + struct au_branch *br; + struct au_br_fhsm *bf; + + root = sb->s_root; + bindex = au_find_dbindex(root, mod->h_root); + if (bindex < 0) { + if (remount) + return 0; /* success */ + err = -ENOENT; + pr_err("%s no such branch\n", mod->path); + goto out; + } + AuDbg("bindex b%d\n", bindex); + + err = test_br(mod->h_root->d_inode, mod->perm, mod->path); + if (unlikely(err)) + goto out; + + br = au_sbr(sb, bindex); + AuDebugOn(mod->h_root != au_br_dentry(br)); + if (br->br_perm == mod->perm) + return 0; /* success */ + + /* pre-allocate for non-fhsm --> fhsm */ + bf = NULL; + if (!au_br_fhsm(br->br_perm) && au_br_fhsm(mod->perm)) { + err = au_fhsm_br_alloc(br); + if (unlikely(err)) + goto out; + bf = br->br_fhsm; + br->br_fhsm = NULL; + } + + if (au_br_writable(br->br_perm)) { + /* remove whiteout base */ + err = au_br_init_wh(sb, br, mod->perm); + if (unlikely(err)) + goto out_bf; + + if (!au_br_writable(mod->perm)) { + /* rw --> ro, file might be mmapped */ + DiMustNoWaiters(root); + IiMustNoWaiters(root->d_inode); + di_write_unlock(root); + err = au_br_mod_files_ro(sb, bindex); + /* aufs_write_lock() calls ..._child() */ + di_write_lock_child(root); + + if (unlikely(err)) { + rerr = -ENOMEM; + br->br_wbr = kmalloc(sizeof(*br->br_wbr), + GFP_NOFS); + if (br->br_wbr) + rerr = au_wbr_init(br, sb, br->br_perm); + if (unlikely(rerr)) { + AuIOErr("nested error %d (%d)\n", + rerr, err); + br->br_perm = mod->perm; + } + } + } + } else if (au_br_writable(mod->perm)) { + /* ro --> rw */ + err = -ENOMEM; + br->br_wbr = kmalloc(sizeof(*br->br_wbr), GFP_NOFS); + if (br->br_wbr) { + err = au_wbr_init(br, sb, mod->perm); + if (unlikely(err)) { + kfree(br->br_wbr); + br->br_wbr = NULL; + } + } + } + if (unlikely(err)) + goto out_bf; + + if (au_br_fhsm(br->br_perm)) { + if (!au_br_fhsm(mod->perm)) { + /* fhsm --> non-fhsm */ + au_br_fhsm_fin(br->br_fhsm); + kfree(br->br_fhsm); + br->br_fhsm = NULL; + } + } else if (au_br_fhsm(mod->perm)) + /* non-fhsm --> fhsm */ + br->br_fhsm = bf; + + if ((br->br_perm & AuBrAttr_UNPIN) + && !(mod->perm & AuBrAttr_UNPIN)) + au_br_dflags_force(br); + else if (!(br->br_perm & AuBrAttr_UNPIN) + && (mod->perm & AuBrAttr_UNPIN)) + au_br_dflags_restore(br); + *do_refresh |= need_sigen_inc(br->br_perm, mod->perm); + br->br_perm = mod->perm; + goto out; /* success */ + +out_bf: + kfree(bf); +out: + AuTraceErr(err); + return err; +} + +/* ---------------------------------------------------------------------- */ + +int au_br_stfs(struct au_branch *br, struct aufs_stfs *stfs) +{ + int err; + struct kstatfs kstfs; + + err = vfs_statfs(&br->br_path, &kstfs); + if (!err) { + stfs->f_blocks = kstfs.f_blocks; + stfs->f_bavail = kstfs.f_bavail; + stfs->f_files = kstfs.f_files; + stfs->f_ffree = kstfs.f_ffree; + } + + return err; +} diff -Naur kernel.21.3.orig/fs/aufs/branch.h kernel.21.3.new/fs/aufs/branch.h --- kernel.21.3.orig/fs/aufs/branch.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/branch.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,268 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * branch filesystems and xino for them + */ + +#ifndef __AUFS_BRANCH_H__ +#define __AUFS_BRANCH_H__ + +#ifdef __KERNEL__ + +#include +#include "dynop.h" +#include "rwsem.h" +#include "super.h" + +/* ---------------------------------------------------------------------- */ + +/* a xino file */ +struct au_xino_file { + struct file *xi_file; + struct mutex xi_nondir_mtx; + + /* todo: make xino files an array to support huge inode number */ + +#ifdef CONFIG_DEBUG_FS + struct dentry *xi_dbgaufs; +#endif +}; + +/* File-based Hierarchical Storage Management */ +struct au_br_fhsm { +#ifdef CONFIG_AUFS_FHSM + struct mutex bf_lock; + unsigned long bf_jiffy; + struct aufs_stfs bf_stfs; + int bf_readable; +#endif +}; + +/* members for writable branch only */ +enum {AuBrWh_BASE, AuBrWh_PLINK, AuBrWh_ORPH, AuBrWh_Last}; +struct au_wbr { + struct au_rwsem wbr_wh_rwsem; + struct dentry *wbr_wh[AuBrWh_Last]; + atomic_t wbr_wh_running; +#define wbr_whbase wbr_wh[AuBrWh_BASE] /* whiteout base */ +#define wbr_plink wbr_wh[AuBrWh_PLINK] /* pseudo-link dir */ +#define wbr_orph wbr_wh[AuBrWh_ORPH] /* dir for orphans */ + + /* mfs mode */ + unsigned long long wbr_bytes; +}; + +/* ext2 has 3 types of operations at least, ext3 has 4 */ +#define AuBrDynOp (AuDyLast * 4) + +#ifdef CONFIG_AUFS_HFSNOTIFY +/* support for asynchronous destruction */ +struct au_br_hfsnotify { + struct fsnotify_group *hfsn_group; +}; +#endif + +/* sysfs entries */ +struct au_brsysfs { + char name[16]; + struct attribute attr; +}; + +enum { + AuBrSysfs_BR, + AuBrSysfs_BRID, + AuBrSysfs_Last +}; + +/* protected by superblock rwsem */ +struct au_branch { + struct au_xino_file br_xino; + + aufs_bindex_t br_id; + + int br_perm; + unsigned int br_dflags; + struct path br_path; + spinlock_t br_dykey_lock; + struct au_dykey *br_dykey[AuBrDynOp]; + atomic_t br_count; + + struct au_wbr *br_wbr; + struct au_br_fhsm *br_fhsm; + + /* xino truncation */ + atomic_t br_xino_running; + +#ifdef CONFIG_AUFS_HFSNOTIFY + struct au_br_hfsnotify *br_hfsn; +#endif + +#ifdef CONFIG_SYSFS + /* entries under sysfs per mount-point */ + struct au_brsysfs br_sysfs[AuBrSysfs_Last]; +#endif +}; + +/* ---------------------------------------------------------------------- */ + +static inline struct vfsmount *au_br_mnt(struct au_branch *br) +{ + return br->br_path.mnt; +} + +static inline struct dentry *au_br_dentry(struct au_branch *br) +{ + return br->br_path.dentry; +} + +static inline struct super_block *au_br_sb(struct au_branch *br) +{ + return au_br_mnt(br)->mnt_sb; +} + +static inline int au_br_rdonly(struct au_branch *br) +{ + return ((au_br_sb(br)->s_flags & MS_RDONLY) + || !au_br_writable(br->br_perm)) + ? -EROFS : 0; +} + +static inline int au_br_hnotifyable(int brperm __maybe_unused) +{ +#ifdef CONFIG_AUFS_HNOTIFY + return !(brperm & AuBrPerm_RR); +#else + return 0; +#endif +} + +/* ---------------------------------------------------------------------- */ + +/* branch.c */ +struct au_sbinfo; +void au_br_free(struct au_sbinfo *sinfo); +int au_br_index(struct super_block *sb, aufs_bindex_t br_id); +struct au_opt_add; +int au_br_add(struct super_block *sb, struct au_opt_add *add, int remount); +struct au_opt_del; +int au_br_del(struct super_block *sb, struct au_opt_del *del, int remount); +long au_ibusy_ioctl(struct file *file, unsigned long arg); +#ifdef CONFIG_COMPAT +long au_ibusy_compat_ioctl(struct file *file, unsigned long arg); +#endif +struct au_opt_mod; +int au_br_mod(struct super_block *sb, struct au_opt_mod *mod, int remount, + int *do_refresh); +struct aufs_stfs; +int au_br_stfs(struct au_branch *br, struct aufs_stfs *stfs); + +/* xino.c */ +static const loff_t au_loff_max = LLONG_MAX; + +int au_xib_trunc(struct super_block *sb); +ssize_t xino_fread(au_readf_t func, struct file *file, void *buf, size_t size, + loff_t *pos); +ssize_t xino_fwrite(au_writef_t func, struct file *file, void *buf, size_t size, + loff_t *pos); +struct file *au_xino_create2(struct file *base_file, struct file *copy_src); +struct file *au_xino_create(struct super_block *sb, char *fname, int silent); +ino_t au_xino_new_ino(struct super_block *sb); +void au_xino_delete_inode(struct inode *inode, const int unlinked); +int au_xino_write(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino, + ino_t ino); +int au_xino_read(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino, + ino_t *ino); +int au_xino_br(struct super_block *sb, struct au_branch *br, ino_t hino, + struct file *base_file, int do_test); +int au_xino_trunc(struct super_block *sb, aufs_bindex_t bindex); + +struct au_opt_xino; +int au_xino_set(struct super_block *sb, struct au_opt_xino *xino, int remount); +void au_xino_clr(struct super_block *sb); +struct file *au_xino_def(struct super_block *sb); +int au_xino_path(struct seq_file *seq, struct file *file); + +/* ---------------------------------------------------------------------- */ + +/* Superblock to branch */ +static inline +aufs_bindex_t au_sbr_id(struct super_block *sb, aufs_bindex_t bindex) +{ + return au_sbr(sb, bindex)->br_id; +} + +static inline +struct vfsmount *au_sbr_mnt(struct super_block *sb, aufs_bindex_t bindex) +{ + return au_br_mnt(au_sbr(sb, bindex)); +} + +static inline +struct super_block *au_sbr_sb(struct super_block *sb, aufs_bindex_t bindex) +{ + return au_br_sb(au_sbr(sb, bindex)); +} + +static inline void au_sbr_put(struct super_block *sb, aufs_bindex_t bindex) +{ + atomic_dec(&au_sbr(sb, bindex)->br_count); +} + +static inline int au_sbr_perm(struct super_block *sb, aufs_bindex_t bindex) +{ + return au_sbr(sb, bindex)->br_perm; +} + +static inline int au_sbr_whable(struct super_block *sb, aufs_bindex_t bindex) +{ + return au_br_whable(au_sbr_perm(sb, bindex)); +} + +/* ---------------------------------------------------------------------- */ + +/* + * wbr_wh_read_lock, wbr_wh_write_lock + * wbr_wh_read_unlock, wbr_wh_write_unlock, wbr_wh_downgrade_lock + */ +AuSimpleRwsemFuncs(wbr_wh, struct au_wbr *wbr, &wbr->wbr_wh_rwsem); + +#define WbrWhMustNoWaiters(wbr) AuRwMustNoWaiters(&wbr->wbr_wh_rwsem) +#define WbrWhMustAnyLock(wbr) AuRwMustAnyLock(&wbr->wbr_wh_rwsem) +#define WbrWhMustWriteLock(wbr) AuRwMustWriteLock(&wbr->wbr_wh_rwsem) + +/* ---------------------------------------------------------------------- */ + +#ifdef CONFIG_AUFS_FHSM +static inline void au_br_fhsm_init(struct au_br_fhsm *brfhsm) +{ + mutex_init(&brfhsm->bf_lock); + brfhsm->bf_jiffy = 0; + brfhsm->bf_readable = 0; +} + +static inline void au_br_fhsm_fin(struct au_br_fhsm *brfhsm) +{ + mutex_destroy(&brfhsm->bf_lock); +} +#else +AuStubVoid(au_br_fhsm_init, struct au_br_fhsm *brfhsm) +AuStubVoid(au_br_fhsm_fin, struct au_br_fhsm *brfhsm) +#endif + +#endif /* __KERNEL__ */ +#endif /* __AUFS_BRANCH_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/conf.mk kernel.21.3.new/fs/aufs/conf.mk --- kernel.21.3.orig/fs/aufs/conf.mk 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/conf.mk 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,38 @@ + +AuConfStr = CONFIG_AUFS_FS=${CONFIG_AUFS_FS} + +define AuConf +ifdef ${1} +AuConfStr += ${1}=${${1}} +endif +endef + +AuConfAll = BRANCH_MAX_127 BRANCH_MAX_511 BRANCH_MAX_1023 BRANCH_MAX_32767 \ + SBILIST \ + HNOTIFY HFSNOTIFY \ + EXPORT INO_T_64 \ + XATTR \ + FHSM \ + RDU \ + SHWH \ + BR_RAMFS \ + BR_FUSE POLL \ + BR_HFSPLUS \ + BDEV_LOOP \ + DEBUG MAGIC_SYSRQ +$(foreach i, ${AuConfAll}, \ + $(eval $(call AuConf,CONFIG_AUFS_${i}))) + +AuConfName = ${obj}/conf.str +${AuConfName}.tmp: FORCE + @echo ${AuConfStr} | tr ' ' '\n' | sed -e 's/^/"/' -e 's/$$/\\n"/' > $@ +${AuConfName}: ${AuConfName}.tmp + @diff -q $< $@ > /dev/null 2>&1 || { \ + echo ' GEN ' $@; \ + cp -p $< $@; \ + } +FORCE: +clean-files += ${AuConfName} ${AuConfName}.tmp +${obj}/sysfs.o: ${AuConfName} + +-include ${srctree}/${src}/conf_priv.mk diff -Naur kernel.21.3.orig/fs/aufs/cpup.c kernel.21.3.new/fs/aufs/cpup.c --- kernel.21.3.orig/fs/aufs/cpup.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/cpup.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,1285 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * copy-up functions, see wbr_policy.c for copy-down + */ + +#include +#include +#include "aufs.h" + +void au_cpup_attr_flags(struct inode *dst, unsigned int iflags) +{ + const unsigned int mask = S_DEAD | S_SWAPFILE | S_PRIVATE + | S_NOATIME | S_NOCMTIME | S_AUTOMOUNT; + + BUILD_BUG_ON(sizeof(iflags) != sizeof(dst->i_flags)); + + dst->i_flags |= iflags & ~mask; + if (au_test_fs_notime(dst->i_sb)) + dst->i_flags |= S_NOATIME | S_NOCMTIME; +} + +void au_cpup_attr_timesizes(struct inode *inode) +{ + struct inode *h_inode; + + h_inode = au_h_iptr(inode, au_ibstart(inode)); + fsstack_copy_attr_times(inode, h_inode); + fsstack_copy_inode_size(inode, h_inode); +} + +void au_cpup_attr_nlink(struct inode *inode, int force) +{ + struct inode *h_inode; + struct super_block *sb; + aufs_bindex_t bindex, bend; + + sb = inode->i_sb; + bindex = au_ibstart(inode); + h_inode = au_h_iptr(inode, bindex); + if (!force + && !S_ISDIR(h_inode->i_mode) + && au_opt_test(au_mntflags(sb), PLINK) + && au_plink_test(inode)) + return; + + /* + * 0 can happen in revalidating. + * h_inode->i_mutex is not held, but it is harmless since once i_nlink + * reaches 0, it will never become positive. + */ + set_nlink(inode, h_inode->i_nlink); + + /* + * fewer nlink makes find(1) noisy, but larger nlink doesn't. + * it may includes whplink directory. + */ + if (S_ISDIR(h_inode->i_mode)) { + bend = au_ibend(inode); + for (bindex++; bindex <= bend; bindex++) { + h_inode = au_h_iptr(inode, bindex); + if (h_inode) + au_add_nlink(inode, h_inode); + } + } +} + +void au_cpup_attr_changeable(struct inode *inode) +{ + struct inode *h_inode; + + h_inode = au_h_iptr(inode, au_ibstart(inode)); + inode->i_mode = h_inode->i_mode; + inode->i_uid = h_inode->i_uid; + inode->i_gid = h_inode->i_gid; + au_cpup_attr_timesizes(inode); + au_cpup_attr_flags(inode, h_inode->i_flags); +} + +void au_cpup_igen(struct inode *inode, struct inode *h_inode) +{ + struct au_iinfo *iinfo = au_ii(inode); + + IiMustWriteLock(inode); + + iinfo->ii_higen = h_inode->i_generation; + iinfo->ii_hsb1 = h_inode->i_sb; +} + +void au_cpup_attr_all(struct inode *inode, int force) +{ + struct inode *h_inode; + + h_inode = au_h_iptr(inode, au_ibstart(inode)); + au_cpup_attr_changeable(inode); + if (inode->i_nlink > 0) + au_cpup_attr_nlink(inode, force); + inode->i_rdev = h_inode->i_rdev; + inode->i_blkbits = h_inode->i_blkbits; + au_cpup_igen(inode, h_inode); +} + +/* ---------------------------------------------------------------------- */ + +/* Note: dt_dentry and dt_h_dentry are not dget/dput-ed */ + +/* keep the timestamps of the parent dir when cpup */ +void au_dtime_store(struct au_dtime *dt, struct dentry *dentry, + struct path *h_path) +{ + struct inode *h_inode; + + dt->dt_dentry = dentry; + dt->dt_h_path = *h_path; + h_inode = h_path->dentry->d_inode; + dt->dt_atime = h_inode->i_atime; + dt->dt_mtime = h_inode->i_mtime; + /* smp_mb(); */ +} + +void au_dtime_revert(struct au_dtime *dt) +{ + struct iattr attr; + int err; + + attr.ia_atime = dt->dt_atime; + attr.ia_mtime = dt->dt_mtime; + attr.ia_valid = ATTR_FORCE | ATTR_MTIME | ATTR_MTIME_SET + | ATTR_ATIME | ATTR_ATIME_SET; + + err = vfsub_notify_change(&dt->dt_h_path, &attr); + if (unlikely(err)) + pr_warn("restoring timestamps failed(%d). ignored\n", err); +} + +/* ---------------------------------------------------------------------- */ + +/* internal use only */ +struct au_cpup_reg_attr { + int valid; + struct kstat st; + unsigned int iflags; /* inode->i_flags */ +}; + +static noinline_for_stack +int cpup_iattr(struct dentry *dst, aufs_bindex_t bindex, struct dentry *h_src, + struct au_cpup_reg_attr *h_src_attr) +{ + int err, sbits, icex; + struct iattr ia; + struct path h_path; + struct inode *h_isrc, *h_idst; + struct kstat *h_st; + struct au_branch *br; + + h_path.dentry = au_h_dptr(dst, bindex); + h_idst = h_path.dentry->d_inode; + br = au_sbr(dst->d_sb, bindex); + h_path.mnt = au_br_mnt(br); + h_isrc = h_src->d_inode; + ia.ia_valid = ATTR_FORCE | ATTR_UID | ATTR_GID + | ATTR_ATIME | ATTR_MTIME + | ATTR_ATIME_SET | ATTR_MTIME_SET; + if (h_src_attr && h_src_attr->valid) { + h_st = &h_src_attr->st; + ia.ia_uid = h_st->uid; + ia.ia_gid = h_st->gid; + ia.ia_atime = h_st->atime; + ia.ia_mtime = h_st->mtime; + if (h_idst->i_mode != h_st->mode + && !S_ISLNK(h_idst->i_mode)) { + ia.ia_valid |= ATTR_MODE; + ia.ia_mode = h_st->mode; + } + sbits = !!(h_st->mode & (S_ISUID | S_ISGID)); + au_cpup_attr_flags(h_idst, h_src_attr->iflags); + } else { + ia.ia_uid = h_isrc->i_uid; + ia.ia_gid = h_isrc->i_gid; + ia.ia_atime = h_isrc->i_atime; + ia.ia_mtime = h_isrc->i_mtime; + if (h_idst->i_mode != h_isrc->i_mode + && !S_ISLNK(h_idst->i_mode)) { + ia.ia_valid |= ATTR_MODE; + ia.ia_mode = h_isrc->i_mode; + } + sbits = !!(h_isrc->i_mode & (S_ISUID | S_ISGID)); + au_cpup_attr_flags(h_idst, h_isrc->i_flags); + } + err = vfsub_notify_change(&h_path, &ia); + + /* is this nfs only? */ + if (!err && sbits && au_test_nfs(h_path.dentry->d_sb)) { + ia.ia_valid = ATTR_FORCE | ATTR_MODE; + ia.ia_mode = h_isrc->i_mode; + err = vfsub_notify_change(&h_path, &ia); + } + + icex = br->br_perm & AuBrAttr_ICEX; + if (!err) + err = au_cpup_xattr(h_path.dentry, h_src, icex); + + return err; +} + +/* ---------------------------------------------------------------------- */ + +static int au_do_copy_file(struct file *dst, struct file *src, loff_t len, + char *buf, unsigned long blksize) +{ + int err; + size_t sz, rbytes, wbytes; + unsigned char all_zero; + char *p, *zp; + struct mutex *h_mtx; + /* reduce stack usage */ + struct iattr *ia; + + zp = page_address(ZERO_PAGE(0)); + if (unlikely(!zp)) + return -ENOMEM; /* possible? */ + + err = 0; + all_zero = 0; + while (len) { + AuDbg("len %lld\n", len); + sz = blksize; + if (len < blksize) + sz = len; + + rbytes = 0; + /* todo: signal_pending? */ + while (!rbytes || err == -EAGAIN || err == -EINTR) { + rbytes = vfsub_read_k(src, buf, sz, &src->f_pos); + err = rbytes; + } + if (unlikely(err < 0)) + break; + + all_zero = 0; + if (len >= rbytes && rbytes == blksize) + all_zero = !memcmp(buf, zp, rbytes); + if (!all_zero) { + wbytes = rbytes; + p = buf; + while (wbytes) { + size_t b; + + b = vfsub_write_k(dst, p, wbytes, &dst->f_pos); + err = b; + /* todo: signal_pending? */ + if (unlikely(err == -EAGAIN || err == -EINTR)) + continue; + if (unlikely(err < 0)) + break; + wbytes -= b; + p += b; + } + if (unlikely(err < 0)) + break; + } else { + loff_t res; + + AuLabel(hole); + res = vfsub_llseek(dst, rbytes, SEEK_CUR); + err = res; + if (unlikely(res < 0)) + break; + } + len -= rbytes; + err = 0; + } + + /* the last block may be a hole */ + if (!err && all_zero) { + AuLabel(last hole); + + err = 1; + if (au_test_nfs(dst->f_dentry->d_sb)) { + /* nfs requires this step to make last hole */ + /* is this only nfs? */ + do { + /* todo: signal_pending? */ + err = vfsub_write_k(dst, "\0", 1, &dst->f_pos); + } while (err == -EAGAIN || err == -EINTR); + if (err == 1) + dst->f_pos--; + } + + if (err == 1) { + ia = (void *)buf; + ia->ia_size = dst->f_pos; + ia->ia_valid = ATTR_SIZE | ATTR_FILE; + ia->ia_file = dst; + h_mtx = &file_inode(dst)->i_mutex; + mutex_lock_nested(h_mtx, AuLsc_I_CHILD2); + err = vfsub_notify_change(&dst->f_path, ia); + mutex_unlock(h_mtx); + } + } + + return err; +} + +int au_copy_file(struct file *dst, struct file *src, loff_t len) +{ + int err; + unsigned long blksize; + unsigned char do_kfree; + char *buf; + + err = -ENOMEM; + blksize = dst->f_dentry->d_sb->s_blocksize; + if (!blksize || PAGE_SIZE < blksize) + blksize = PAGE_SIZE; + AuDbg("blksize %lu\n", blksize); + do_kfree = (blksize != PAGE_SIZE && blksize >= sizeof(struct iattr *)); + if (do_kfree) + buf = kmalloc(blksize, GFP_NOFS); + else + buf = (void *)__get_free_page(GFP_NOFS); + if (unlikely(!buf)) + goto out; + + if (len > (1 << 22)) + AuDbg("copying a large file %lld\n", (long long)len); + + src->f_pos = 0; + dst->f_pos = 0; + err = au_do_copy_file(dst, src, len, buf, blksize); + if (do_kfree) + kfree(buf); + else + free_page((unsigned long)buf); + +out: + return err; +} + +/* + * to support a sparse file which is opened with O_APPEND, + * we need to close the file. + */ +static int au_cp_regular(struct au_cp_generic *cpg) +{ + int err, i; + enum { SRC, DST }; + struct { + aufs_bindex_t bindex; + unsigned int flags; + struct dentry *dentry; + int force_wr; + struct file *file; + void *label, *label_file; + } *f, file[] = { + { + .bindex = cpg->bsrc, + .flags = O_RDONLY | O_NOATIME | O_LARGEFILE, + .label = &&out, + .label_file = &&out_src + }, + { + .bindex = cpg->bdst, + .flags = O_WRONLY | O_NOATIME | O_LARGEFILE, + .force_wr = !!au_ftest_cpup(cpg->flags, RWDST), + .label = &&out_src, + .label_file = &&out_dst + } + }; + struct super_block *sb; + + /* bsrc branch can be ro/rw. */ + sb = cpg->dentry->d_sb; + f = file; + for (i = 0; i < 2; i++, f++) { + f->dentry = au_h_dptr(cpg->dentry, f->bindex); + f->file = au_h_open(cpg->dentry, f->bindex, f->flags, + /*file*/NULL, f->force_wr); + err = PTR_ERR(f->file); + if (IS_ERR(f->file)) + goto *f->label; + err = -EINVAL; + if (unlikely(!f->file->f_op)) + goto *f->label_file; + } + + /* try stopping to update while we copyup */ + IMustLock(file[SRC].dentry->d_inode); + err = au_copy_file(file[DST].file, file[SRC].file, cpg->len); + +out_dst: + fput(file[DST].file); + au_sbr_put(sb, file[DST].bindex); +out_src: + fput(file[SRC].file); + au_sbr_put(sb, file[SRC].bindex); +out: + return err; +} + +static int au_do_cpup_regular(struct au_cp_generic *cpg, + struct au_cpup_reg_attr *h_src_attr) +{ + int err, rerr; + loff_t l; + struct path h_path; + struct inode *h_src_inode; + + err = 0; + h_src_inode = au_h_iptr(cpg->dentry->d_inode, cpg->bsrc); + l = i_size_read(h_src_inode); + if (cpg->len == -1 || l < cpg->len) + cpg->len = l; + if (cpg->len) { + /* try stopping to update while we are referencing */ + mutex_lock_nested(&h_src_inode->i_mutex, AuLsc_I_CHILD); + au_pin_hdir_unlock(cpg->pin); + + h_path.dentry = au_h_dptr(cpg->dentry, cpg->bsrc); + h_path.mnt = au_sbr_mnt(cpg->dentry->d_sb, cpg->bsrc); + h_src_attr->iflags = h_src_inode->i_flags; + err = vfs_getattr(&h_path, &h_src_attr->st); + if (unlikely(err)) { + mutex_unlock(&h_src_inode->i_mutex); + goto out; + } + h_src_attr->valid = 1; + err = au_cp_regular(cpg); + mutex_unlock(&h_src_inode->i_mutex); + rerr = au_pin_hdir_relock(cpg->pin); + if (!err && rerr) + err = rerr; + } + +out: + return err; +} + +static int au_do_cpup_symlink(struct path *h_path, struct dentry *h_src, + struct inode *h_dir) +{ + int err, symlen; + mm_segment_t old_fs; + union { + char *k; + char __user *u; + } sym; + + err = -ENOSYS; + if (unlikely(!h_src->d_inode->i_op->readlink)) + goto out; + + err = -ENOMEM; + sym.k = (void *)__get_free_page(GFP_NOFS); + if (unlikely(!sym.k)) + goto out; + + /* unnecessary to support mmap_sem since symlink is not mmap-able */ + old_fs = get_fs(); + set_fs(KERNEL_DS); + symlen = h_src->d_inode->i_op->readlink(h_src, sym.u, PATH_MAX); + err = symlen; + set_fs(old_fs); + + if (symlen > 0) { + sym.k[symlen] = 0; + err = vfsub_symlink(h_dir, h_path, sym.k); + } + free_page((unsigned long)sym.k); + +out: + return err; +} + +static noinline_for_stack +int cpup_entry(struct au_cp_generic *cpg, struct dentry *dst_parent, + struct au_cpup_reg_attr *h_src_attr) +{ + int err; + umode_t mode; + unsigned int mnt_flags; + unsigned char isdir, isreg, force; + const unsigned char do_dt = !!au_ftest_cpup(cpg->flags, DTIME); + struct au_dtime dt; + struct path h_path; + struct dentry *h_src, *h_dst, *h_parent; + struct inode *h_inode, *h_dir; + struct super_block *sb; + + /* bsrc branch can be ro/rw. */ + h_src = au_h_dptr(cpg->dentry, cpg->bsrc); + h_inode = h_src->d_inode; + AuDebugOn(h_inode != au_h_iptr(cpg->dentry->d_inode, cpg->bsrc)); + + /* try stopping to be referenced while we are creating */ + h_dst = au_h_dptr(cpg->dentry, cpg->bdst); + if (au_ftest_cpup(cpg->flags, RENAME)) + AuDebugOn(strncmp(h_dst->d_name.name, AUFS_WH_PFX, + AUFS_WH_PFX_LEN)); + h_parent = h_dst->d_parent; /* dir inode is locked */ + h_dir = h_parent->d_inode; + IMustLock(h_dir); + AuDebugOn(h_parent != h_dst->d_parent); + + sb = cpg->dentry->d_sb; + h_path.mnt = au_sbr_mnt(sb, cpg->bdst); + if (do_dt) { + h_path.dentry = h_parent; + au_dtime_store(&dt, dst_parent, &h_path); + } + h_path.dentry = h_dst; + + isreg = 0; + isdir = 0; + mode = h_inode->i_mode; + switch (mode & S_IFMT) { + case S_IFREG: + isreg = 1; + err = vfsub_create(h_dir, &h_path, mode | S_IWUSR, + /*want_excl*/true); + if (!err) + err = au_do_cpup_regular(cpg, h_src_attr); + break; + case S_IFDIR: + isdir = 1; + err = vfsub_mkdir(h_dir, &h_path, mode); + if (!err) { + /* + * strange behaviour from the users view, + * particularry setattr case + */ + if (au_ibstart(dst_parent->d_inode) == cpg->bdst) + au_cpup_attr_nlink(dst_parent->d_inode, + /*force*/1); + au_cpup_attr_nlink(cpg->dentry->d_inode, /*force*/1); + } + break; + case S_IFLNK: + err = au_do_cpup_symlink(&h_path, h_src, h_dir); + break; + case S_IFCHR: + case S_IFBLK: + AuDebugOn(!capable(CAP_MKNOD)); + /*FALLTHROUGH*/ + case S_IFIFO: + case S_IFSOCK: + err = vfsub_mknod(h_dir, &h_path, mode, h_inode->i_rdev); + break; + default: + AuIOErr("Unknown inode type 0%o\n", mode); + err = -EIO; + } + + mnt_flags = au_mntflags(sb); + if (!au_opt_test(mnt_flags, UDBA_NONE) + && !isdir + && au_opt_test(mnt_flags, XINO) + && h_inode->i_nlink == 1 + /* todo: unnecessary? */ + /* && cpg->dentry->d_inode->i_nlink == 1 */ + && cpg->bdst < cpg->bsrc + && !au_ftest_cpup(cpg->flags, KEEPLINO)) + au_xino_write(sb, cpg->bsrc, h_inode->i_ino, /*ino*/0); + /* ignore this error */ + + if (!err) { + force = 0; + if (isreg) { + force = !!cpg->len; + if (cpg->len == -1) + force = !!i_size_read(h_inode); + } + au_fhsm_wrote(sb, cpg->bdst, force); + } + + if (do_dt) + au_dtime_revert(&dt); + return err; +} + +static int au_do_ren_after_cpup(struct au_cp_generic *cpg, struct path *h_path) +{ + int err; + struct dentry *dentry, *h_dentry, *h_parent, *parent; + struct inode *h_dir; + aufs_bindex_t bdst; + + dentry = cpg->dentry; + bdst = cpg->bdst; + h_dentry = au_h_dptr(dentry, bdst); + if (!au_ftest_cpup(cpg->flags, OVERWRITE)) { + dget(h_dentry); + au_set_h_dptr(dentry, bdst, NULL); + err = au_lkup_neg(dentry, bdst, /*wh*/0); + if (!err) + h_path->dentry = dget(au_h_dptr(dentry, bdst)); + au_set_h_dptr(dentry, bdst, h_dentry); + } else { + err = 0; + parent = dget_parent(dentry); + h_parent = au_h_dptr(parent, bdst); + dput(parent); + h_path->dentry = vfsub_lkup_one(&dentry->d_name, h_parent); + if (IS_ERR(h_path->dentry)) + err = PTR_ERR(h_path->dentry); + } + if (unlikely(err)) + goto out; + + h_parent = h_dentry->d_parent; /* dir inode is locked */ + h_dir = h_parent->d_inode; + IMustLock(h_dir); + AuDbg("%.*s %.*s\n", AuDLNPair(h_dentry), AuDLNPair(h_path->dentry)); + err = vfsub_rename(h_dir, h_dentry, h_dir, h_path); + dput(h_path->dentry); + +out: + return err; +} + +/* + * copyup the @dentry from @bsrc to @bdst. + * the caller must set the both of lower dentries. + * @len is for truncating when it is -1 copyup the entire file. + * in link/rename cases, @dst_parent may be different from the real one. + * basic->bsrc can be larger than basic->bdst. + */ +static int au_cpup_single(struct au_cp_generic *cpg, struct dentry *dst_parent) +{ + int err, rerr; + aufs_bindex_t old_ibstart; + unsigned char isdir, plink; + struct dentry *h_src, *h_dst, *h_parent; + struct inode *dst_inode, *h_dir, *inode; + struct super_block *sb; + struct au_branch *br; + /* to reuduce stack size */ + struct { + struct au_dtime dt; + struct path h_path; + struct au_cpup_reg_attr h_src_attr; + } *a; + + err = -ENOMEM; + a = kmalloc(sizeof(*a), GFP_NOFS); + if (unlikely(!a)) + goto out; + a->h_src_attr.valid = 0; + + sb = cpg->dentry->d_sb; + br = au_sbr(sb, cpg->bdst); + a->h_path.mnt = au_br_mnt(br); + h_dst = au_h_dptr(cpg->dentry, cpg->bdst); + h_parent = h_dst->d_parent; /* dir inode is locked */ + h_dir = h_parent->d_inode; + IMustLock(h_dir); + + h_src = au_h_dptr(cpg->dentry, cpg->bsrc); + inode = cpg->dentry->d_inode; + + if (!dst_parent) + dst_parent = dget_parent(cpg->dentry); + else + dget(dst_parent); + + plink = !!au_opt_test(au_mntflags(sb), PLINK); + dst_inode = au_h_iptr(inode, cpg->bdst); + if (dst_inode) { + if (unlikely(!plink)) { + err = -EIO; + AuIOErr("hi%lu(i%lu) exists on b%d " + "but plink is disabled\n", + dst_inode->i_ino, inode->i_ino, cpg->bdst); + goto out_parent; + } + + if (dst_inode->i_nlink) { + const int do_dt = au_ftest_cpup(cpg->flags, DTIME); + + h_src = au_plink_lkup(inode, cpg->bdst); + err = PTR_ERR(h_src); + if (IS_ERR(h_src)) + goto out_parent; + if (unlikely(!h_src->d_inode)) { + err = -EIO; + AuIOErr("i%lu exists on a upper branch " + "but not pseudo-linked\n", + inode->i_ino); + dput(h_src); + goto out_parent; + } + + if (do_dt) { + a->h_path.dentry = h_parent; + au_dtime_store(&a->dt, dst_parent, &a->h_path); + } + + a->h_path.dentry = h_dst; + err = vfsub_link(h_src, h_dir, &a->h_path); + if (!err && au_ftest_cpup(cpg->flags, RENAME)) + err = au_do_ren_after_cpup(cpg, &a->h_path); + if (do_dt) + au_dtime_revert(&a->dt); + dput(h_src); + goto out_parent; + } else + /* todo: cpup_wh_file? */ + /* udba work */ + au_update_ibrange(inode, /*do_put_zero*/1); + } + + isdir = S_ISDIR(inode->i_mode); + old_ibstart = au_ibstart(inode); + err = cpup_entry(cpg, dst_parent, &a->h_src_attr); + if (unlikely(err)) + goto out_rev; + dst_inode = h_dst->d_inode; + mutex_lock_nested(&dst_inode->i_mutex, AuLsc_I_CHILD2); + /* todo: necessary? */ + /* au_pin_hdir_unlock(cpg->pin); */ + + err = cpup_iattr(cpg->dentry, cpg->bdst, h_src, &a->h_src_attr); + if (unlikely(err)) { + /* todo: necessary? */ + /* au_pin_hdir_relock(cpg->pin); */ /* ignore an error */ + mutex_unlock(&dst_inode->i_mutex); + goto out_rev; + } + + if (cpg->bdst < old_ibstart) { + if (S_ISREG(inode->i_mode)) { + err = au_dy_iaop(inode, cpg->bdst, dst_inode); + if (unlikely(err)) { + /* ignore an error */ + /* au_pin_hdir_relock(cpg->pin); */ + mutex_unlock(&dst_inode->i_mutex); + goto out_rev; + } + } + au_set_ibstart(inode, cpg->bdst); + } else + au_set_ibend(inode, cpg->bdst); + au_set_h_iptr(inode, cpg->bdst, au_igrab(dst_inode), + au_hi_flags(inode, isdir)); + + /* todo: necessary? */ + /* err = au_pin_hdir_relock(cpg->pin); */ + mutex_unlock(&dst_inode->i_mutex); + if (unlikely(err)) + goto out_rev; + + if (!isdir + && h_src->d_inode->i_nlink > 1 + && plink) + au_plink_append(inode, cpg->bdst, h_dst); + + if (au_ftest_cpup(cpg->flags, RENAME)) { + a->h_path.dentry = h_dst; + err = au_do_ren_after_cpup(cpg, &a->h_path); + } + if (!err) + goto out_parent; /* success */ + + /* revert */ +out_rev: + a->h_path.dentry = h_parent; + au_dtime_store(&a->dt, dst_parent, &a->h_path); + a->h_path.dentry = h_dst; + rerr = 0; + if (h_dst->d_inode) { + if (!isdir) + rerr = vfsub_unlink(h_dir, &a->h_path, /*force*/0); + else + rerr = vfsub_rmdir(h_dir, &a->h_path); + } + au_dtime_revert(&a->dt); + if (rerr) { + AuIOErr("failed removing broken entry(%d, %d)\n", err, rerr); + err = -EIO; + } +out_parent: + dput(dst_parent); + kfree(a); +out: + return err; +} + +#if 0 /* unused */ +struct au_cpup_single_args { + int *errp; + struct au_cp_generic *cpg; + struct dentry *dst_parent; +}; + +static void au_call_cpup_single(void *args) +{ + struct au_cpup_single_args *a = args; + + au_pin_hdir_acquire_nest(a->cpg->pin); + *a->errp = au_cpup_single(a->cpg, a->dst_parent); + au_pin_hdir_release(a->cpg->pin); +} +#endif + +/* + * prevent SIGXFSZ in copy-up. + * testing CAP_MKNOD is for generic fs, + * but CAP_FSETID is for xfs only, currently. + */ +static int au_cpup_sio_test(struct au_pin *pin, umode_t mode) +{ + int do_sio; + struct super_block *sb; + struct inode *h_dir; + + do_sio = 0; + sb = au_pinned_parent(pin)->d_sb; + if (!au_wkq_test() + && (!au_sbi(sb)->si_plink_maint_pid + || au_plink_maint(sb, AuLock_NOPLM))) { + switch (mode & S_IFMT) { + case S_IFREG: + /* no condition about RLIMIT_FSIZE and the file size */ + do_sio = 1; + break; + case S_IFCHR: + case S_IFBLK: + do_sio = !capable(CAP_MKNOD); + break; + } + if (!do_sio) + do_sio = ((mode & (S_ISUID | S_ISGID)) + && !capable(CAP_FSETID)); + /* this workaround may be removed in the future */ + if (!do_sio) { + h_dir = au_pinned_h_dir(pin); + do_sio = h_dir->i_mode & S_ISVTX; + } + } + + return do_sio; +} + +#if 0 /* unused */ +int au_sio_cpup_single(struct au_cp_generic *cpg, struct dentry *dst_parent) +{ + int err, wkq_err; + struct dentry *h_dentry; + + h_dentry = au_h_dptr(cpg->dentry, cpg->bsrc); + if (!au_cpup_sio_test(pin, h_dentry->d_inode->i_mode)) + err = au_cpup_single(cpg, dst_parent); + else { + struct au_cpup_single_args args = { + .errp = &err, + .cpg = cpg, + .dst_parent = dst_parent + }; + wkq_err = au_wkq_wait(au_call_cpup_single, &args); + if (unlikely(wkq_err)) + err = wkq_err; + } + + return err; +} +#endif + +/* + * copyup the @dentry from the first active lower branch to @bdst, + * using au_cpup_single(). + */ +static int au_cpup_simple(struct au_cp_generic *cpg) +{ + int err; + unsigned int flags_orig; + struct dentry *dentry; + + AuDebugOn(cpg->bsrc < 0); + + dentry = cpg->dentry; + DiMustWriteLock(dentry); + + err = au_lkup_neg(dentry, cpg->bdst, /*wh*/1); + if (!err) { + flags_orig = cpg->flags; + au_fset_cpup(cpg->flags, RENAME); + err = au_cpup_single(cpg, NULL); + cpg->flags = flags_orig; + if (!err) + return 0; /* success */ + + /* revert */ + au_set_h_dptr(dentry, cpg->bdst, NULL); + au_set_dbstart(dentry, cpg->bsrc); + } + + return err; +} + +struct au_cpup_simple_args { + int *errp; + struct au_cp_generic *cpg; +}; + +static void au_call_cpup_simple(void *args) +{ + struct au_cpup_simple_args *a = args; + + au_pin_hdir_acquire_nest(a->cpg->pin); + *a->errp = au_cpup_simple(a->cpg); + au_pin_hdir_release(a->cpg->pin); +} + +static int au_do_sio_cpup_simple(struct au_cp_generic *cpg) +{ + int err, wkq_err; + struct dentry *dentry, *parent; + struct file *h_file; + struct inode *h_dir; + + dentry = cpg->dentry; + h_file = NULL; + if (au_ftest_cpup(cpg->flags, HOPEN)) { + AuDebugOn(cpg->bsrc < 0); + h_file = au_h_open_pre(dentry, cpg->bsrc, /*force_wr*/0); + err = PTR_ERR(h_file); + if (IS_ERR(h_file)) + goto out; + } + + parent = dget_parent(dentry); + h_dir = au_h_iptr(parent->d_inode, cpg->bdst); + if (!au_test_h_perm_sio(h_dir, MAY_EXEC | MAY_WRITE) + && !au_cpup_sio_test(cpg->pin, dentry->d_inode->i_mode)) + err = au_cpup_simple(cpg); + else { + struct au_cpup_simple_args args = { + .errp = &err, + .cpg = cpg + }; + wkq_err = au_wkq_wait(au_call_cpup_simple, &args); + if (unlikely(wkq_err)) + err = wkq_err; + } + + dput(parent); + if (h_file) + au_h_open_post(dentry, cpg->bsrc, h_file); + +out: + return err; +} + +int au_sio_cpup_simple(struct au_cp_generic *cpg) +{ + aufs_bindex_t bsrc, bend; + struct dentry *dentry, *h_dentry; + + if (cpg->bsrc < 0) { + dentry = cpg->dentry; + bend = au_dbend(dentry); + for (bsrc = cpg->bdst + 1; bsrc <= bend; bsrc++) { + h_dentry = au_h_dptr(dentry, bsrc); + if (h_dentry) { + AuDebugOn(!h_dentry->d_inode); + break; + } + } + AuDebugOn(bsrc > bend); + cpg->bsrc = bsrc; + } + AuDebugOn(cpg->bsrc <= cpg->bdst); + return au_do_sio_cpup_simple(cpg); +} + +int au_sio_cpdown_simple(struct au_cp_generic *cpg) +{ + AuDebugOn(cpg->bdst <= cpg->bsrc); + return au_do_sio_cpup_simple(cpg); +} + +/* ---------------------------------------------------------------------- */ + +/* + * copyup the deleted file for writing. + */ +static int au_do_cpup_wh(struct au_cp_generic *cpg, struct dentry *wh_dentry, + struct file *file) +{ + int err; + unsigned int flags_orig; + aufs_bindex_t bsrc_orig; + struct dentry *h_d_dst, *h_d_start; + struct au_dinfo *dinfo; + struct au_hdentry *hdp; + + dinfo = au_di(cpg->dentry); + AuRwMustWriteLock(&dinfo->di_rwsem); + + bsrc_orig = cpg->bsrc; + cpg->bsrc = dinfo->di_bstart; + hdp = dinfo->di_hdentry; + h_d_dst = hdp[0 + cpg->bdst].hd_dentry; + dinfo->di_bstart = cpg->bdst; + hdp[0 + cpg->bdst].hd_dentry = wh_dentry; + h_d_start = NULL; + if (file) { + h_d_start = hdp[0 + cpg->bsrc].hd_dentry; + hdp[0 + cpg->bsrc].hd_dentry = au_hf_top(file)->f_dentry; + } + flags_orig = cpg->flags; + cpg->flags = !AuCpup_DTIME; + err = au_cpup_single(cpg, /*h_parent*/NULL); + cpg->flags = flags_orig; + if (file) { + if (!err) + err = au_reopen_nondir(file); + hdp[0 + cpg->bsrc].hd_dentry = h_d_start; + } + hdp[0 + cpg->bdst].hd_dentry = h_d_dst; + dinfo->di_bstart = cpg->bsrc; + cpg->bsrc = bsrc_orig; + + return err; +} + +static int au_cpup_wh(struct au_cp_generic *cpg, struct file *file) +{ + int err; + aufs_bindex_t bdst; + struct au_dtime dt; + struct dentry *dentry, *parent, *h_parent, *wh_dentry; + struct au_branch *br; + struct path h_path; + + dentry = cpg->dentry; + bdst = cpg->bdst; + br = au_sbr(dentry->d_sb, bdst); + parent = dget_parent(dentry); + h_parent = au_h_dptr(parent, bdst); + wh_dentry = au_whtmp_lkup(h_parent, br, &dentry->d_name); + err = PTR_ERR(wh_dentry); + if (IS_ERR(wh_dentry)) + goto out; + + h_path.dentry = h_parent; + h_path.mnt = au_br_mnt(br); + au_dtime_store(&dt, parent, &h_path); + err = au_do_cpup_wh(cpg, wh_dentry, file); + if (unlikely(err)) + goto out_wh; + + dget(wh_dentry); + h_path.dentry = wh_dentry; + if (!S_ISDIR(wh_dentry->d_inode->i_mode)) + err = vfsub_unlink(h_parent->d_inode, &h_path, /*force*/0); + else + err = vfsub_rmdir(h_parent->d_inode, &h_path); + if (unlikely(err)) { + AuIOErr("failed remove copied-up tmp file %.*s(%d)\n", + AuDLNPair(wh_dentry), err); + err = -EIO; + } + au_dtime_revert(&dt); + au_set_hi_wh(dentry->d_inode, bdst, wh_dentry); + +out_wh: + dput(wh_dentry); +out: + dput(parent); + return err; +} + +struct au_cpup_wh_args { + int *errp; + struct au_cp_generic *cpg; + struct file *file; +}; + +static void au_call_cpup_wh(void *args) +{ + struct au_cpup_wh_args *a = args; + + au_pin_hdir_acquire_nest(a->cpg->pin); + *a->errp = au_cpup_wh(a->cpg, a->file); + au_pin_hdir_release(a->cpg->pin); +} + +int au_sio_cpup_wh(struct au_cp_generic *cpg, struct file *file) +{ + int err, wkq_err; + aufs_bindex_t bdst; + struct dentry *dentry, *parent, *h_orph, *h_parent, *h_dentry; + struct inode *dir, *h_dir, *h_tmpdir; + struct au_wbr *wbr; + struct au_pin wh_pin, *pin_orig; + + dentry = cpg->dentry; + bdst = cpg->bdst; + parent = dget_parent(dentry); + dir = parent->d_inode; + h_orph = NULL; + h_parent = NULL; + h_dir = au_igrab(au_h_iptr(dir, bdst)); + h_tmpdir = h_dir; + pin_orig = NULL; + if (!h_dir->i_nlink) { + wbr = au_sbr(dentry->d_sb, bdst)->br_wbr; + h_orph = wbr->wbr_orph; + + h_parent = dget(au_h_dptr(parent, bdst)); + au_set_h_dptr(parent, bdst, dget(h_orph)); + h_tmpdir = h_orph->d_inode; + au_set_h_iptr(dir, bdst, au_igrab(h_tmpdir), /*flags*/0); + + if (file) + h_dentry = au_hf_top(file)->f_dentry; + else + h_dentry = au_h_dptr(dentry, au_dbstart(dentry)); + mutex_lock_nested(&h_tmpdir->i_mutex, AuLsc_I_PARENT3); + /* todo: au_h_open_pre()? */ + + pin_orig = cpg->pin; + au_pin_init(&wh_pin, dentry, bdst, AuLsc_DI_PARENT, + AuLsc_I_PARENT3, cpg->pin->udba, AuPin_DI_LOCKED); + cpg->pin = &wh_pin; + } + + if (!au_test_h_perm_sio(h_tmpdir, MAY_EXEC | MAY_WRITE) + && !au_cpup_sio_test(cpg->pin, dentry->d_inode->i_mode)) + err = au_cpup_wh(cpg, file); + else { + struct au_cpup_wh_args args = { + .errp = &err, + .cpg = cpg, + .file = file + }; + wkq_err = au_wkq_wait(au_call_cpup_wh, &args); + if (unlikely(wkq_err)) + err = wkq_err; + } + + if (h_orph) { + mutex_unlock(&h_tmpdir->i_mutex); + /* todo: au_h_open_post()? */ + au_set_h_iptr(dir, bdst, au_igrab(h_dir), /*flags*/0); + au_set_h_dptr(parent, bdst, h_parent); + AuDebugOn(!pin_orig); + cpg->pin = pin_orig; + } + iput(h_dir); + dput(parent); + + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* + * generic routine for both of copy-up and copy-down. + */ +/* cf. revalidate function in file.c */ +int au_cp_dirs(struct dentry *dentry, aufs_bindex_t bdst, + int (*cp)(struct dentry *dentry, aufs_bindex_t bdst, + struct au_pin *pin, + struct dentry *h_parent, void *arg), + void *arg) +{ + int err; + struct au_pin pin; + struct dentry *d, *parent, *h_parent, *real_parent; + + err = 0; + parent = dget_parent(dentry); + if (IS_ROOT(parent)) + goto out; + + au_pin_init(&pin, dentry, bdst, AuLsc_DI_PARENT2, AuLsc_I_PARENT2, + au_opt_udba(dentry->d_sb), AuPin_MNT_WRITE); + + /* do not use au_dpage */ + real_parent = parent; + while (1) { + dput(parent); + parent = dget_parent(dentry); + h_parent = au_h_dptr(parent, bdst); + if (h_parent) + goto out; /* success */ + + /* find top dir which is necessary to cpup */ + do { + d = parent; + dput(parent); + parent = dget_parent(d); + di_read_lock_parent3(parent, !AuLock_IR); + h_parent = au_h_dptr(parent, bdst); + di_read_unlock(parent, !AuLock_IR); + } while (!h_parent); + + if (d != real_parent) + di_write_lock_child3(d); + + /* somebody else might create while we were sleeping */ + if (!au_h_dptr(d, bdst) || !au_h_dptr(d, bdst)->d_inode) { + if (au_h_dptr(d, bdst)) + au_update_dbstart(d); + + au_pin_set_dentry(&pin, d); + err = au_do_pin(&pin); + if (!err) { + err = cp(d, bdst, &pin, h_parent, arg); + au_unpin(&pin); + } + } + + if (d != real_parent) + di_write_unlock(d); + if (unlikely(err)) + break; + } + +out: + dput(parent); + return err; +} + +static int au_cpup_dir(struct dentry *dentry, aufs_bindex_t bdst, + struct au_pin *pin, + struct dentry *h_parent __maybe_unused , + void *arg __maybe_unused) +{ + struct au_cp_generic cpg = { + .dentry = dentry, + .bdst = bdst, + .bsrc = -1, + .len = 0, + .pin = pin, + .flags = AuCpup_DTIME + }; + return au_sio_cpup_simple(&cpg); +} + +int au_cpup_dirs(struct dentry *dentry, aufs_bindex_t bdst) +{ + return au_cp_dirs(dentry, bdst, au_cpup_dir, NULL); +} + +int au_test_and_cpup_dirs(struct dentry *dentry, aufs_bindex_t bdst) +{ + int err; + struct dentry *parent; + struct inode *dir; + + parent = dget_parent(dentry); + dir = parent->d_inode; + err = 0; + if (au_h_iptr(dir, bdst)) + goto out; + + di_read_unlock(parent, AuLock_IR); + di_write_lock_parent(parent); + /* someone else might change our inode while we were sleeping */ + if (!au_h_iptr(dir, bdst)) + err = au_cpup_dirs(dentry, bdst); + di_downgrade_lock(parent, AuLock_IR); + +out: + dput(parent); + return err; +} diff -Naur kernel.21.3.orig/fs/aufs/cpup.h kernel.21.3.new/fs/aufs/cpup.h --- kernel.21.3.orig/fs/aufs/cpup.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/cpup.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,94 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * copy-up/down functions + */ + +#ifndef __AUFS_CPUP_H__ +#define __AUFS_CPUP_H__ + +#ifdef __KERNEL__ + +#include + +struct inode; +struct file; +struct au_pin; + +void au_cpup_attr_flags(struct inode *dst, unsigned int iflags); +void au_cpup_attr_timesizes(struct inode *inode); +void au_cpup_attr_nlink(struct inode *inode, int force); +void au_cpup_attr_changeable(struct inode *inode); +void au_cpup_igen(struct inode *inode, struct inode *h_inode); +void au_cpup_attr_all(struct inode *inode, int force); + +/* ---------------------------------------------------------------------- */ + +struct au_cp_generic { + struct dentry *dentry; + aufs_bindex_t bdst, bsrc; + loff_t len; + struct au_pin *pin; + unsigned int flags; +}; + +/* cpup flags */ +#define AuCpup_DTIME 1 /* do dtime_store/revert */ +#define AuCpup_KEEPLINO (1 << 1) /* do not clear the lower xino, + for link(2) */ +#define AuCpup_RENAME (1 << 2) /* rename after cpup */ +#define AuCpup_HOPEN (1 << 3) /* call h_open_pre/post() in + cpup */ +#define AuCpup_OVERWRITE (1 << 4) /* allow overwriting the + existing entry */ +#define AuCpup_RWDST (1 << 5) /* force write target even if + the branch is marked as RO */ + +#define au_ftest_cpup(flags, name) ((flags) & AuCpup_##name) +#define au_fset_cpup(flags, name) \ + do { (flags) |= AuCpup_##name; } while (0) +#define au_fclr_cpup(flags, name) \ + do { (flags) &= ~AuCpup_##name; } while (0) + +int au_copy_file(struct file *dst, struct file *src, loff_t len); +int au_sio_cpup_simple(struct au_cp_generic *cpg); +int au_sio_cpdown_simple(struct au_cp_generic *cpg); +int au_sio_cpup_wh(struct au_cp_generic *cpg, struct file *file); + +int au_cp_dirs(struct dentry *dentry, aufs_bindex_t bdst, + int (*cp)(struct dentry *dentry, aufs_bindex_t bdst, + struct au_pin *pin, + struct dentry *h_parent, void *arg), + void *arg); +int au_cpup_dirs(struct dentry *dentry, aufs_bindex_t bdst); +int au_test_and_cpup_dirs(struct dentry *dentry, aufs_bindex_t bdst); + +/* ---------------------------------------------------------------------- */ + +/* keep timestamps when copyup */ +struct au_dtime { + struct dentry *dt_dentry; + struct path dt_h_path; + struct timespec dt_atime, dt_mtime; +}; +void au_dtime_store(struct au_dtime *dt, struct dentry *dentry, + struct path *h_path); +void au_dtime_revert(struct au_dtime *dt); + +#endif /* __KERNEL__ */ +#endif /* __AUFS_CPUP_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/dbgaufs.c kernel.21.3.new/fs/aufs/dbgaufs.c --- kernel.21.3.orig/fs/aufs/dbgaufs.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/dbgaufs.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,432 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * debugfs interface + */ + +#include +#include "aufs.h" + +#ifndef CONFIG_SYSFS +#error DEBUG_FS depends upon SYSFS +#endif + +static struct dentry *dbgaufs; +static const mode_t dbgaufs_mode = S_IRUSR | S_IRGRP | S_IROTH; + +/* 20 is max digits length of ulong 64 */ +struct dbgaufs_arg { + int n; + char a[20 * 4]; +}; + +/* + * common function for all XINO files + */ +static int dbgaufs_xi_release(struct inode *inode __maybe_unused, + struct file *file) +{ + kfree(file->private_data); + return 0; +} + +static int dbgaufs_xi_open(struct file *xf, struct file *file, int do_fcnt) +{ + int err; + struct kstat st; + struct dbgaufs_arg *p; + + err = -ENOMEM; + p = kmalloc(sizeof(*p), GFP_NOFS); + if (unlikely(!p)) + goto out; + + err = 0; + p->n = 0; + file->private_data = p; + if (!xf) + goto out; + + err = vfs_getattr(&xf->f_path, &st); + if (!err) { + if (do_fcnt) + p->n = snprintf + (p->a, sizeof(p->a), "%ld, %llux%lu %lld\n", + (long)file_count(xf), st.blocks, st.blksize, + (long long)st.size); + else + p->n = snprintf(p->a, sizeof(p->a), "%llux%lu %lld\n", + st.blocks, st.blksize, + (long long)st.size); + AuDebugOn(p->n >= sizeof(p->a)); + } else { + p->n = snprintf(p->a, sizeof(p->a), "err %d\n", err); + err = 0; + } + +out: + return err; + +} + +static ssize_t dbgaufs_xi_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) +{ + struct dbgaufs_arg *p; + + p = file->private_data; + return simple_read_from_buffer(buf, count, ppos, p->a, p->n); +} + +/* ---------------------------------------------------------------------- */ + +struct dbgaufs_plink_arg { + int n; + char a[]; +}; + +static int dbgaufs_plink_release(struct inode *inode __maybe_unused, + struct file *file) +{ + free_page((unsigned long)file->private_data); + return 0; +} + +static int dbgaufs_plink_open(struct inode *inode, struct file *file) +{ + int err, i, limit; + unsigned long n, sum; + struct dbgaufs_plink_arg *p; + struct au_sbinfo *sbinfo; + struct super_block *sb; + struct au_sphlhead *sphl; + + err = -ENOMEM; + p = (void *)get_zeroed_page(GFP_NOFS); + if (unlikely(!p)) + goto out; + + err = -EFBIG; + sbinfo = inode->i_private; + sb = sbinfo->si_sb; + si_noflush_read_lock(sb); + if (au_opt_test(au_mntflags(sb), PLINK)) { + limit = PAGE_SIZE - sizeof(p->n); + + /* the number of buckets */ + n = snprintf(p->a + p->n, limit, "%d\n", AuPlink_NHASH); + p->n += n; + limit -= n; + + sum = 0; + for (i = 0, sphl = sbinfo->si_plink; + i < AuPlink_NHASH; + i++, sphl++) { + n = au_sphl_count(sphl); + sum += n; + + n = snprintf(p->a + p->n, limit, "%lu ", n); + p->n += n; + limit -= n; + if (unlikely(limit <= 0)) + goto out_free; + } + p->a[p->n - 1] = '\n'; + + /* the sum of plinks */ + n = snprintf(p->a + p->n, limit, "%lu\n", sum); + p->n += n; + limit -= n; + if (unlikely(limit <= 0)) + goto out_free; + } else { +#define str "1\n0\n0\n" + p->n = sizeof(str) - 1; + strcpy(p->a, str); +#undef str + } + si_read_unlock(sb); + + err = 0; + file->private_data = p; + goto out; /* success */ + +out_free: + free_page((unsigned long)p); +out: + return err; +} + +static ssize_t dbgaufs_plink_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) +{ + struct dbgaufs_plink_arg *p; + + p = file->private_data; + return simple_read_from_buffer(buf, count, ppos, p->a, p->n); +} + +static const struct file_operations dbgaufs_plink_fop = { + .owner = THIS_MODULE, + .open = dbgaufs_plink_open, + .release = dbgaufs_plink_release, + .read = dbgaufs_plink_read +}; + +/* ---------------------------------------------------------------------- */ + +static int dbgaufs_xib_open(struct inode *inode, struct file *file) +{ + int err; + struct au_sbinfo *sbinfo; + struct super_block *sb; + + sbinfo = inode->i_private; + sb = sbinfo->si_sb; + si_noflush_read_lock(sb); + err = dbgaufs_xi_open(sbinfo->si_xib, file, /*do_fcnt*/0); + si_read_unlock(sb); + return err; +} + +static const struct file_operations dbgaufs_xib_fop = { + .owner = THIS_MODULE, + .open = dbgaufs_xib_open, + .release = dbgaufs_xi_release, + .read = dbgaufs_xi_read +}; + +/* ---------------------------------------------------------------------- */ + +#define DbgaufsXi_PREFIX "xi" + +static int dbgaufs_xino_open(struct inode *inode, struct file *file) +{ + int err; + long l; + struct au_sbinfo *sbinfo; + struct super_block *sb; + struct file *xf; + struct qstr *name; + + err = -ENOENT; + xf = NULL; + name = &file->f_dentry->d_name; + if (unlikely(name->len < sizeof(DbgaufsXi_PREFIX) + || memcmp(name->name, DbgaufsXi_PREFIX, + sizeof(DbgaufsXi_PREFIX) - 1))) + goto out; + err = kstrtol(name->name + sizeof(DbgaufsXi_PREFIX) - 1, 10, &l); + if (unlikely(err)) + goto out; + + sbinfo = inode->i_private; + sb = sbinfo->si_sb; + si_noflush_read_lock(sb); + if (l <= au_sbend(sb)) { + xf = au_sbr(sb, (aufs_bindex_t)l)->br_xino.xi_file; + err = dbgaufs_xi_open(xf, file, /*do_fcnt*/1); + } else + err = -ENOENT; + si_read_unlock(sb); + +out: + return err; +} + +static const struct file_operations dbgaufs_xino_fop = { + .owner = THIS_MODULE, + .open = dbgaufs_xino_open, + .release = dbgaufs_xi_release, + .read = dbgaufs_xi_read +}; + +void dbgaufs_brs_del(struct super_block *sb, aufs_bindex_t bindex) +{ + aufs_bindex_t bend; + struct au_branch *br; + struct au_xino_file *xi; + + if (!au_sbi(sb)->si_dbgaufs) + return; + + bend = au_sbend(sb); + for (; bindex <= bend; bindex++) { + br = au_sbr(sb, bindex); + xi = &br->br_xino; + debugfs_remove(xi->xi_dbgaufs); + xi->xi_dbgaufs = NULL; + } +} + +void dbgaufs_brs_add(struct super_block *sb, aufs_bindex_t bindex) +{ + struct au_sbinfo *sbinfo; + struct dentry *parent; + struct au_branch *br; + struct au_xino_file *xi; + aufs_bindex_t bend; + char name[sizeof(DbgaufsXi_PREFIX) + 5]; /* "xi" bindex NULL */ + + sbinfo = au_sbi(sb); + parent = sbinfo->si_dbgaufs; + if (!parent) + return; + + bend = au_sbend(sb); + for (; bindex <= bend; bindex++) { + snprintf(name, sizeof(name), DbgaufsXi_PREFIX "%d", bindex); + br = au_sbr(sb, bindex); + xi = &br->br_xino; + AuDebugOn(xi->xi_dbgaufs); + xi->xi_dbgaufs = debugfs_create_file(name, dbgaufs_mode, parent, + sbinfo, &dbgaufs_xino_fop); + /* ignore an error */ + if (unlikely(!xi->xi_dbgaufs)) + AuWarn1("failed %s under debugfs\n", name); + } +} + +/* ---------------------------------------------------------------------- */ + +#ifdef CONFIG_AUFS_EXPORT +static int dbgaufs_xigen_open(struct inode *inode, struct file *file) +{ + int err; + struct au_sbinfo *sbinfo; + struct super_block *sb; + + sbinfo = inode->i_private; + sb = sbinfo->si_sb; + si_noflush_read_lock(sb); + err = dbgaufs_xi_open(sbinfo->si_xigen, file, /*do_fcnt*/0); + si_read_unlock(sb); + return err; +} + +static const struct file_operations dbgaufs_xigen_fop = { + .owner = THIS_MODULE, + .open = dbgaufs_xigen_open, + .release = dbgaufs_xi_release, + .read = dbgaufs_xi_read +}; + +static int dbgaufs_xigen_init(struct au_sbinfo *sbinfo) +{ + int err; + + /* + * This function is a dynamic '__init' function actually, + * so the tiny check for si_rwsem is unnecessary. + */ + /* AuRwMustWriteLock(&sbinfo->si_rwsem); */ + + err = -EIO; + sbinfo->si_dbgaufs_xigen = debugfs_create_file + ("xigen", dbgaufs_mode, sbinfo->si_dbgaufs, sbinfo, + &dbgaufs_xigen_fop); + if (sbinfo->si_dbgaufs_xigen) + err = 0; + + return err; +} +#else +static int dbgaufs_xigen_init(struct au_sbinfo *sbinfo) +{ + return 0; +} +#endif /* CONFIG_AUFS_EXPORT */ + +/* ---------------------------------------------------------------------- */ + +void dbgaufs_si_fin(struct au_sbinfo *sbinfo) +{ + /* + * This function is a dynamic '__init' function actually, + * so the tiny check for si_rwsem is unnecessary. + */ + /* AuRwMustWriteLock(&sbinfo->si_rwsem); */ + + debugfs_remove_recursive(sbinfo->si_dbgaufs); + sbinfo->si_dbgaufs = NULL; + kobject_put(&sbinfo->si_kobj); +} + +int dbgaufs_si_init(struct au_sbinfo *sbinfo) +{ + int err; + char name[SysaufsSiNameLen]; + + /* + * This function is a dynamic '__init' function actually, + * so the tiny check for si_rwsem is unnecessary. + */ + /* AuRwMustWriteLock(&sbinfo->si_rwsem); */ + + err = -ENOENT; + if (!dbgaufs) { + AuErr1("/debug/aufs is uninitialized\n"); + goto out; + } + + err = -EIO; + sysaufs_name(sbinfo, name); + sbinfo->si_dbgaufs = debugfs_create_dir(name, dbgaufs); + if (unlikely(!sbinfo->si_dbgaufs)) + goto out; + kobject_get(&sbinfo->si_kobj); + + sbinfo->si_dbgaufs_xib = debugfs_create_file + ("xib", dbgaufs_mode, sbinfo->si_dbgaufs, sbinfo, + &dbgaufs_xib_fop); + if (unlikely(!sbinfo->si_dbgaufs_xib)) + goto out_dir; + + sbinfo->si_dbgaufs_plink = debugfs_create_file + ("plink", dbgaufs_mode, sbinfo->si_dbgaufs, sbinfo, + &dbgaufs_plink_fop); + if (unlikely(!sbinfo->si_dbgaufs_plink)) + goto out_dir; + + err = dbgaufs_xigen_init(sbinfo); + if (!err) + goto out; /* success */ + +out_dir: + dbgaufs_si_fin(sbinfo); +out: + return err; +} + +/* ---------------------------------------------------------------------- */ + +void dbgaufs_fin(void) +{ + debugfs_remove(dbgaufs); +} + +int __init dbgaufs_init(void) +{ + int err; + + err = -EIO; + dbgaufs = debugfs_create_dir(AUFS_NAME, NULL); + if (dbgaufs) + err = 0; + return err; +} diff -Naur kernel.21.3.orig/fs/aufs/dbgaufs.h kernel.21.3.new/fs/aufs/dbgaufs.h --- kernel.21.3.orig/fs/aufs/dbgaufs.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/dbgaufs.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,48 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * debugfs interface + */ + +#ifndef __DBGAUFS_H__ +#define __DBGAUFS_H__ + +#ifdef __KERNEL__ + +struct super_block; +struct au_sbinfo; + +#ifdef CONFIG_DEBUG_FS +/* dbgaufs.c */ +void dbgaufs_brs_del(struct super_block *sb, aufs_bindex_t bindex); +void dbgaufs_brs_add(struct super_block *sb, aufs_bindex_t bindex); +void dbgaufs_si_fin(struct au_sbinfo *sbinfo); +int dbgaufs_si_init(struct au_sbinfo *sbinfo); +void dbgaufs_fin(void); +int __init dbgaufs_init(void); +#else +AuStubVoid(dbgaufs_brs_del, struct super_block *sb, aufs_bindex_t bindex) +AuStubVoid(dbgaufs_brs_add, struct super_block *sb, aufs_bindex_t bindex) +AuStubVoid(dbgaufs_si_fin, struct au_sbinfo *sbinfo) +AuStubInt0(dbgaufs_si_init, struct au_sbinfo *sbinfo) +AuStubVoid(dbgaufs_fin, void) +AuStubInt0(__init dbgaufs_init, void) +#endif /* CONFIG_DEBUG_FS */ + +#endif /* __KERNEL__ */ +#endif /* __DBGAUFS_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/dcsub.c kernel.21.3.new/fs/aufs/dcsub.c --- kernel.21.3.orig/fs/aufs/dcsub.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/dcsub.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,242 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * sub-routines for dentry cache + */ + +#include "aufs.h" + +static void au_dpage_free(struct au_dpage *dpage) +{ + int i; + struct dentry **p; + + p = dpage->dentries; + for (i = 0; i < dpage->ndentry; i++) + dput(*p++); + free_page((unsigned long)dpage->dentries); +} + +int au_dpages_init(struct au_dcsub_pages *dpages, gfp_t gfp) +{ + int err; + void *p; + + err = -ENOMEM; + dpages->dpages = kmalloc(sizeof(*dpages->dpages), gfp); + if (unlikely(!dpages->dpages)) + goto out; + + p = (void *)__get_free_page(gfp); + if (unlikely(!p)) + goto out_dpages; + + dpages->dpages[0].ndentry = 0; + dpages->dpages[0].dentries = p; + dpages->ndpage = 1; + return 0; /* success */ + +out_dpages: + kfree(dpages->dpages); +out: + return err; +} + +void au_dpages_free(struct au_dcsub_pages *dpages) +{ + int i; + struct au_dpage *p; + + p = dpages->dpages; + for (i = 0; i < dpages->ndpage; i++) + au_dpage_free(p++); + kfree(dpages->dpages); +} + +static int au_dpages_append(struct au_dcsub_pages *dpages, + struct dentry *dentry, gfp_t gfp) +{ + int err, sz; + struct au_dpage *dpage; + void *p; + + dpage = dpages->dpages + dpages->ndpage - 1; + sz = PAGE_SIZE / sizeof(dentry); + if (unlikely(dpage->ndentry >= sz)) { + AuLabel(new dpage); + err = -ENOMEM; + sz = dpages->ndpage * sizeof(*dpages->dpages); + p = au_kzrealloc(dpages->dpages, sz, + sz + sizeof(*dpages->dpages), gfp); + if (unlikely(!p)) + goto out; + + dpages->dpages = p; + dpage = dpages->dpages + dpages->ndpage; + p = (void *)__get_free_page(gfp); + if (unlikely(!p)) + goto out; + + dpage->ndentry = 0; + dpage->dentries = p; + dpages->ndpage++; + } + + AuDebugOn(!dentry->d_count); + dpage->dentries[dpage->ndentry++] = dget_dlock(dentry); + return 0; /* success */ + +out: + return err; +} + +int au_dcsub_pages(struct au_dcsub_pages *dpages, struct dentry *root, + au_dpages_test test, void *arg) +{ + int err; + struct dentry *this_parent; + struct list_head *next; + struct super_block *sb = root->d_sb; + + err = 0; + write_seqlock(&rename_lock); + this_parent = root; + spin_lock(&this_parent->d_lock); +repeat: + next = this_parent->d_subdirs.next; +resume: + if (this_parent->d_sb == sb + && !IS_ROOT(this_parent) + && au_di(this_parent) + && this_parent->d_count + && (!test || test(this_parent, arg))) { + err = au_dpages_append(dpages, this_parent, GFP_ATOMIC); + if (unlikely(err)) + goto out; + } + + while (next != &this_parent->d_subdirs) { + struct list_head *tmp = next; + struct dentry *dentry = list_entry(tmp, struct dentry, + d_u.d_child); + + next = tmp->next; + spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); + if (dentry->d_count) { + if (!list_empty(&dentry->d_subdirs)) { + spin_unlock(&this_parent->d_lock); + spin_release(&dentry->d_lock.dep_map, 1, + _RET_IP_); + this_parent = dentry; + spin_acquire(&this_parent->d_lock.dep_map, 0, 1, + _RET_IP_); + goto repeat; + } + if (dentry->d_sb == sb + && au_di(dentry) + && (!test || test(dentry, arg))) + err = au_dpages_append(dpages, dentry, + GFP_ATOMIC); + } + spin_unlock(&dentry->d_lock); + if (unlikely(err)) + goto out; + } + + if (this_parent != root) { + struct dentry *tmp; + struct dentry *child; + + tmp = this_parent->d_parent; + rcu_read_lock(); + spin_unlock(&this_parent->d_lock); + child = this_parent; + this_parent = tmp; + spin_lock(&this_parent->d_lock); + rcu_read_unlock(); + next = child->d_u.d_child.next; + goto resume; + } + +out: + spin_unlock(&this_parent->d_lock); + write_sequnlock(&rename_lock); + return err; +} + +int au_dcsub_pages_rev(struct au_dcsub_pages *dpages, struct dentry *dentry, + int do_include, au_dpages_test test, void *arg) +{ + int err; + + err = 0; + write_seqlock(&rename_lock); + spin_lock(&dentry->d_lock); + if (do_include + && dentry->d_count + && (!test || test(dentry, arg))) + err = au_dpages_append(dpages, dentry, GFP_ATOMIC); + spin_unlock(&dentry->d_lock); + if (unlikely(err)) + goto out; + + /* + * vfsmount_lock is unnecessary since this is a traverse in a single + * mount + */ + while (!IS_ROOT(dentry)) { + dentry = dentry->d_parent; /* rename_lock is locked */ + spin_lock(&dentry->d_lock); + if (dentry->d_count + && (!test || test(dentry, arg))) + err = au_dpages_append(dpages, dentry, GFP_ATOMIC); + spin_unlock(&dentry->d_lock); + if (unlikely(err)) + break; + } + +out: + write_sequnlock(&rename_lock); + return err; +} + +static inline int au_dcsub_dpages_aufs(struct dentry *dentry, void *arg) +{ + return au_di(dentry) && dentry->d_sb == arg; +} + +int au_dcsub_pages_rev_aufs(struct au_dcsub_pages *dpages, + struct dentry *dentry, int do_include) +{ + return au_dcsub_pages_rev(dpages, dentry, do_include, + au_dcsub_dpages_aufs, dentry->d_sb); +} + +int au_test_subdir(struct dentry *d1, struct dentry *d2) +{ + struct path path[2] = { + { + .dentry = d1 + }, + { + .dentry = d2 + } + }; + + return path_is_under(path + 0, path + 1); +} diff -Naur kernel.21.3.orig/fs/aufs/dcsub.h kernel.21.3.new/fs/aufs/dcsub.h --- kernel.21.3.orig/fs/aufs/dcsub.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/dcsub.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,102 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * sub-routines for dentry cache + */ + +#ifndef __AUFS_DCSUB_H__ +#define __AUFS_DCSUB_H__ + +#ifdef __KERNEL__ + +#include +#include + +struct dentry; + +struct au_dpage { + int ndentry; + struct dentry **dentries; +}; + +struct au_dcsub_pages { + int ndpage; + struct au_dpage *dpages; +}; + +/* ---------------------------------------------------------------------- */ + +/* dcsub.c */ +int au_dpages_init(struct au_dcsub_pages *dpages, gfp_t gfp); +void au_dpages_free(struct au_dcsub_pages *dpages); +typedef int (*au_dpages_test)(struct dentry *dentry, void *arg); +int au_dcsub_pages(struct au_dcsub_pages *dpages, struct dentry *root, + au_dpages_test test, void *arg); +int au_dcsub_pages_rev(struct au_dcsub_pages *dpages, struct dentry *dentry, + int do_include, au_dpages_test test, void *arg); +int au_dcsub_pages_rev_aufs(struct au_dcsub_pages *dpages, + struct dentry *dentry, int do_include); +int au_test_subdir(struct dentry *d1, struct dentry *d2); + +/* ---------------------------------------------------------------------- */ + +static inline int au_d_hashed_positive(struct dentry *d) +{ + int err; + struct inode *inode = d->d_inode; + + err = 0; + if (unlikely(d_unhashed(d) || !inode || !inode->i_nlink)) + err = -ENOENT; + return err; +} + +static inline int au_d_alive(struct dentry *d) +{ + int err; + struct inode *inode; + + err = 0; + if (!IS_ROOT(d)) + err = au_d_hashed_positive(d); + else { + inode = d->d_inode; + if (unlikely(d_unlinked(d) || !inode || !inode->i_nlink)) + err = -ENOENT; + } + return err; +} + +static inline int au_alive_dir(struct dentry *d) +{ + int err; + + err = au_d_alive(d); + if (unlikely(err || IS_DEADDIR(d->d_inode))) + err = -ENOENT; + return err; +} + +static inline int au_qstreq(struct qstr *a, struct qstr *b) +{ + return a->len == b->len + && !memcmp(a->name, b->name, a->len); +} + +#endif /* __KERNEL__ */ +#endif /* __AUFS_DCSUB_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/debug.c kernel.21.3.new/fs/aufs/debug.c --- kernel.21.3.orig/fs/aufs/debug.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/debug.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,520 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * debug print functions + */ + +#include +#include "aufs.h" + +/* Returns 0, or -errno. arg is in kp->arg. */ +static int param_atomic_t_set(const char *val, const struct kernel_param *kp) +{ + int err, n; + + err = kstrtoint(val, 0, &n); + if (!err) { + if (n > 0) + au_debug_on(); + else + au_debug_off(); + } + return err; +} + +/* Returns length written or -errno. Buffer is 4k (ie. be short!) */ +static int param_atomic_t_get(char *buffer, const struct kernel_param *kp) +{ + atomic_t *a; + + a = kp->arg; + return sprintf(buffer, "%d", atomic_read(a)); +} + +static struct kernel_param_ops param_ops_atomic_t = { + .set = param_atomic_t_set, + .get = param_atomic_t_get + /* void (*free)(void *arg) */ +}; + +atomic_t aufs_debug = ATOMIC_INIT(0); +MODULE_PARM_DESC(debug, "debug print"); +module_param_named(debug, aufs_debug, atomic_t, S_IRUGO | S_IWUSR | S_IWGRP); + +DEFINE_MUTEX(au_dbg_mtx); /* just to serialize the dbg msgs */ +char *au_plevel = KERN_DEBUG; +#define dpri(fmt, ...) do { \ + if ((au_plevel \ + && strcmp(au_plevel, KERN_DEBUG)) \ + || au_debug_test()) \ + printk("%s" fmt, au_plevel, ##__VA_ARGS__); \ +} while (0) + +/* ---------------------------------------------------------------------- */ + +void au_dpri_whlist(struct au_nhash *whlist) +{ + unsigned long ul, n; + struct hlist_head *head; + struct au_vdir_wh *pos; + + n = whlist->nh_num; + head = whlist->nh_head; + for (ul = 0; ul < n; ul++) { + hlist_for_each_entry(pos, head, wh_hash) + dpri("b%d, %.*s, %d\n", + pos->wh_bindex, + pos->wh_str.len, pos->wh_str.name, + pos->wh_str.len); + head++; + } +} + +void au_dpri_vdir(struct au_vdir *vdir) +{ + unsigned long ul; + union au_vdir_deblk_p p; + unsigned char *o; + + if (!vdir || IS_ERR(vdir)) { + dpri("err %ld\n", PTR_ERR(vdir)); + return; + } + + dpri("deblk %u, nblk %lu, deblk %p, last{%lu, %p}, ver %lu\n", + vdir->vd_deblk_sz, vdir->vd_nblk, vdir->vd_deblk, + vdir->vd_last.ul, vdir->vd_last.p.deblk, vdir->vd_version); + for (ul = 0; ul < vdir->vd_nblk; ul++) { + p.deblk = vdir->vd_deblk[ul]; + o = p.deblk; + dpri("[%lu]: %p\n", ul, o); + } +} + +static int do_pri_inode(aufs_bindex_t bindex, struct inode *inode, int hn, + struct dentry *wh) +{ + char *n = NULL; + int l = 0; + + if (!inode || IS_ERR(inode)) { + dpri("i%d: err %ld\n", bindex, PTR_ERR(inode)); + return -1; + } + + /* the type of i_blocks depends upon CONFIG_LBDAF */ + BUILD_BUG_ON(sizeof(inode->i_blocks) != sizeof(unsigned long) + && sizeof(inode->i_blocks) != sizeof(u64)); + if (wh) { + n = (void *)wh->d_name.name; + l = wh->d_name.len; + } + + dpri("i%d: %p, i%lu, %s, cnt %d, nl %u, 0%o, sz %llu, blk %llu," + " hn %d, ct %lld, np %lu, st 0x%lx, f 0x%x, v %llu, g %x%s%.*s\n", + bindex, inode, + inode->i_ino, inode->i_sb ? au_sbtype(inode->i_sb) : "??", + atomic_read(&inode->i_count), inode->i_nlink, inode->i_mode, + i_size_read(inode), (unsigned long long)inode->i_blocks, + hn, (long long)timespec_to_ns(&inode->i_ctime) & 0x0ffff, + inode->i_mapping ? inode->i_mapping->nrpages : 0, + inode->i_state, inode->i_flags, inode->i_version, + inode->i_generation, + l ? ", wh " : "", l, n); + return 0; +} + +void au_dpri_inode(struct inode *inode) +{ + struct au_iinfo *iinfo; + aufs_bindex_t bindex; + int err, hn; + + err = do_pri_inode(-1, inode, -1, NULL); + if (err || !au_test_aufs(inode->i_sb)) + return; + + iinfo = au_ii(inode); + if (!iinfo) + return; + dpri("i-1: bstart %d, bend %d, gen %d\n", + iinfo->ii_bstart, iinfo->ii_bend, au_iigen(inode, NULL)); + if (iinfo->ii_bstart < 0) + return; + hn = 0; + for (bindex = iinfo->ii_bstart; bindex <= iinfo->ii_bend; bindex++) { + hn = !!au_hn(iinfo->ii_hinode + bindex); + do_pri_inode(bindex, iinfo->ii_hinode[0 + bindex].hi_inode, hn, + iinfo->ii_hinode[0 + bindex].hi_whdentry); + } +} + +void au_dpri_dalias(struct inode *inode) +{ + struct dentry *d; + + spin_lock(&inode->i_lock); + hlist_for_each_entry(d, &inode->i_dentry, d_alias) + au_dpri_dentry(d); + spin_unlock(&inode->i_lock); +} + +static int do_pri_dentry(aufs_bindex_t bindex, struct dentry *dentry) +{ + struct dentry *wh = NULL; + int hn; + struct au_iinfo *iinfo; + + if (!dentry || IS_ERR(dentry)) { + dpri("d%d: err %ld\n", bindex, PTR_ERR(dentry)); + return -1; + } + /* do not call dget_parent() here */ + /* note: access d_xxx without d_lock */ + dpri("d%d: %p, %.*s?/%.*s, %s, cnt %d, flags 0x%x, %shashed\n", + bindex, dentry, + AuDLNPair(dentry->d_parent), AuDLNPair(dentry), + dentry->d_sb ? au_sbtype(dentry->d_sb) : "??", + dentry->d_count, dentry->d_flags, + d_unhashed(dentry) ? "un" : ""); + hn = -1; + if (bindex >= 0 && dentry->d_inode && au_test_aufs(dentry->d_sb)) { + iinfo = au_ii(dentry->d_inode); + if (iinfo) { + hn = !!au_hn(iinfo->ii_hinode + bindex); + wh = iinfo->ii_hinode[0 + bindex].hi_whdentry; + } + } + do_pri_inode(bindex, dentry->d_inode, hn, wh); + return 0; +} + +void au_dpri_dentry(struct dentry *dentry) +{ + struct au_dinfo *dinfo; + aufs_bindex_t bindex; + int err; + struct au_hdentry *hdp; + + err = do_pri_dentry(-1, dentry); + if (err || !au_test_aufs(dentry->d_sb)) + return; + + dinfo = au_di(dentry); + if (!dinfo) + return; + dpri("d-1: bstart %d, bend %d, bwh %d, bdiropq %d, gen %d\n", + dinfo->di_bstart, dinfo->di_bend, + dinfo->di_bwh, dinfo->di_bdiropq, au_digen(dentry)); + if (dinfo->di_bstart < 0) + return; + hdp = dinfo->di_hdentry; + for (bindex = dinfo->di_bstart; bindex <= dinfo->di_bend; bindex++) + do_pri_dentry(bindex, hdp[0 + bindex].hd_dentry); +} + +static int do_pri_file(aufs_bindex_t bindex, struct file *file) +{ + char a[32]; + + if (!file || IS_ERR(file)) { + dpri("f%d: err %ld\n", bindex, PTR_ERR(file)); + return -1; + } + a[0] = 0; + if (bindex < 0 + && file->f_dentry + && au_test_aufs(file->f_dentry->d_sb) + && au_fi(file)) + snprintf(a, sizeof(a), ", gen %d, mmapped %d", + au_figen(file), atomic_read(&au_fi(file)->fi_mmapped)); + dpri("f%d: mode 0x%x, flags 0%o, cnt %ld, v %llu, pos %llu%s\n", + bindex, file->f_mode, file->f_flags, (long)file_count(file), + file->f_version, file->f_pos, a); + if (file->f_dentry) + do_pri_dentry(bindex, file->f_dentry); + return 0; +} + +void au_dpri_file(struct file *file) +{ + struct au_finfo *finfo; + struct au_fidir *fidir; + struct au_hfile *hfile; + aufs_bindex_t bindex; + int err; + + err = do_pri_file(-1, file); + if (err || !file->f_dentry || !au_test_aufs(file->f_dentry->d_sb)) + return; + + finfo = au_fi(file); + if (!finfo) + return; + if (finfo->fi_btop < 0) + return; + fidir = finfo->fi_hdir; + if (!fidir) + do_pri_file(finfo->fi_btop, finfo->fi_htop.hf_file); + else + for (bindex = finfo->fi_btop; + bindex >= 0 && bindex <= fidir->fd_bbot; + bindex++) { + hfile = fidir->fd_hfile + bindex; + do_pri_file(bindex, hfile ? hfile->hf_file : NULL); + } +} + +static int do_pri_br(aufs_bindex_t bindex, struct au_branch *br) +{ + struct vfsmount *mnt; + struct super_block *sb; + + if (!br || IS_ERR(br)) + goto out; + mnt = au_br_mnt(br); + if (!mnt || IS_ERR(mnt)) + goto out; + sb = mnt->mnt_sb; + if (!sb || IS_ERR(sb)) + goto out; + + dpri("s%d: {perm 0x%x, id %d, cnt %d, wbr %p}, " + "%s, dev 0x%02x%02x, flags 0x%lx, cnt %d, active %d, " + "xino %d\n", + bindex, br->br_perm, br->br_id, atomic_read(&br->br_count), + br->br_wbr, au_sbtype(sb), MAJOR(sb->s_dev), MINOR(sb->s_dev), + sb->s_flags, sb->s_count, + atomic_read(&sb->s_active), !!br->br_xino.xi_file); + return 0; + +out: + dpri("s%d: err %ld\n", bindex, PTR_ERR(br)); + return -1; +} + +void au_dpri_sb(struct super_block *sb) +{ + struct au_sbinfo *sbinfo; + aufs_bindex_t bindex; + int err; + /* to reuduce stack size */ + struct { + struct vfsmount mnt; + struct au_branch fake; + } *a; + + /* this function can be called from magic sysrq */ + a = kzalloc(sizeof(*a), GFP_ATOMIC); + if (unlikely(!a)) { + dpri("no memory\n"); + return; + } + + a->mnt.mnt_sb = sb; + a->fake.br_perm = 0; + a->fake.br_path.mnt = &a->mnt; + a->fake.br_xino.xi_file = NULL; + atomic_set(&a->fake.br_count, 0); + smp_mb(); /* atomic_set */ + err = do_pri_br(-1, &a->fake); + kfree(a); + dpri("dev 0x%x\n", sb->s_dev); + if (err || !au_test_aufs(sb)) + return; + + sbinfo = au_sbi(sb); + if (!sbinfo) + return; + dpri("nw %d, gen %u, kobj %d\n", + atomic_read(&sbinfo->si_nowait.nw_len), sbinfo->si_generation, + atomic_read(&sbinfo->si_kobj.kref.refcount)); + for (bindex = 0; bindex <= sbinfo->si_bend; bindex++) + do_pri_br(bindex, sbinfo->si_branch[0 + bindex]); +} + +/* ---------------------------------------------------------------------- */ + +void au_dbg_sleep_jiffy(int jiffy) +{ + while (jiffy) + jiffy = schedule_timeout_uninterruptible(jiffy); +} + +void au_dbg_iattr(struct iattr *ia) +{ +#define AuBit(name) \ + do { \ + if (ia->ia_valid & ATTR_ ## name) \ + dpri(#name "\n"); \ + } while (0) + AuBit(MODE); + AuBit(UID); + AuBit(GID); + AuBit(SIZE); + AuBit(ATIME); + AuBit(MTIME); + AuBit(CTIME); + AuBit(ATIME_SET); + AuBit(MTIME_SET); + AuBit(FORCE); + AuBit(ATTR_FLAG); + AuBit(KILL_SUID); + AuBit(KILL_SGID); + AuBit(FILE); + AuBit(KILL_PRIV); + AuBit(OPEN); + AuBit(TIMES_SET); +#undef AuBit + dpri("ia_file %p\n", ia->ia_file); +} + +/* ---------------------------------------------------------------------- */ + +void __au_dbg_verify_dinode(struct dentry *dentry, const char *func, int line) +{ + struct inode *h_inode, *inode = dentry->d_inode; + struct dentry *h_dentry; + aufs_bindex_t bindex, bend, bi; + + if (!inode /* || au_di(dentry)->di_lsc == AuLsc_DI_TMP */) + return; + + bend = au_dbend(dentry); + bi = au_ibend(inode); + if (bi < bend) + bend = bi; + bindex = au_dbstart(dentry); + bi = au_ibstart(inode); + if (bi > bindex) + bindex = bi; + + for (; bindex <= bend; bindex++) { + h_dentry = au_h_dptr(dentry, bindex); + if (!h_dentry) + continue; + h_inode = au_h_iptr(inode, bindex); + if (unlikely(h_inode != h_dentry->d_inode)) { + au_debug_on(); + AuDbg("b%d, %s:%d\n", bindex, func, line); + AuDbgDentry(dentry); + AuDbgInode(inode); + au_debug_off(); + BUG(); + } + } +} + +void au_dbg_verify_dir_parent(struct dentry *dentry, unsigned int sigen) +{ + struct dentry *parent; + + parent = dget_parent(dentry); + AuDebugOn(!S_ISDIR(dentry->d_inode->i_mode)); + AuDebugOn(IS_ROOT(dentry)); + AuDebugOn(au_digen_test(parent, sigen)); + dput(parent); +} + +void au_dbg_verify_nondir_parent(struct dentry *dentry, unsigned int sigen) +{ + struct dentry *parent; + struct inode *inode; + + parent = dget_parent(dentry); + inode = dentry->d_inode; + AuDebugOn(inode && S_ISDIR(dentry->d_inode->i_mode)); + AuDebugOn(au_digen_test(parent, sigen)); + dput(parent); +} + +void au_dbg_verify_gen(struct dentry *parent, unsigned int sigen) +{ + int err, i, j; + struct au_dcsub_pages dpages; + struct au_dpage *dpage; + struct dentry **dentries; + + err = au_dpages_init(&dpages, GFP_NOFS); + AuDebugOn(err); + err = au_dcsub_pages_rev_aufs(&dpages, parent, /*do_include*/1); + AuDebugOn(err); + for (i = dpages.ndpage - 1; !err && i >= 0; i--) { + dpage = dpages.dpages + i; + dentries = dpage->dentries; + for (j = dpage->ndentry - 1; !err && j >= 0; j--) + AuDebugOn(au_digen_test(dentries[j], sigen)); + } + au_dpages_free(&dpages); +} + +void au_dbg_verify_kthread(void) +{ + if (au_wkq_test()) { + au_dbg_blocked(); + /* + * It may be recursive, but udba=notify between two aufs mounts, + * where a single ro branch is shared, is not a problem. + */ + /* WARN_ON(1); */ + } +} + +/* ---------------------------------------------------------------------- */ + +void au_debug_sbinfo_init(struct au_sbinfo *sbinfo __maybe_unused) +{ +#ifdef AuForceNoPlink + au_opt_clr(sbinfo->si_mntflags, PLINK); +#endif +#ifdef AuForceNoXino + au_opt_clr(sbinfo->si_mntflags, XINO); +#endif +#ifdef AuForceNoRefrof + au_opt_clr(sbinfo->si_mntflags, REFROF); +#endif +#ifdef AuForceHnotify + au_opt_set_udba(sbinfo->si_mntflags, UDBA_HNOTIFY); +#endif +#ifdef AuForceRd0 + sbinfo->si_rdblk = 0; + sbinfo->si_rdhash = 0; +#endif +} + +int __init au_debug_init(void) +{ + aufs_bindex_t bindex; + struct au_vdir_destr destr; + + bindex = -1; + AuDebugOn(bindex >= 0); + + destr.len = -1; + AuDebugOn(destr.len < NAME_MAX); + +#ifdef CONFIG_4KSTACKS + pr_warn("CONFIG_4KSTACKS is defined.\n"); +#endif + +#ifdef AuForceNoBrs + sysaufs_brs = 0; +#endif + + return 0; +} diff -Naur kernel.21.3.orig/fs/aufs/debug.h kernel.21.3.new/fs/aufs/debug.h --- kernel.21.3.orig/fs/aufs/debug.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/debug.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,263 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * debug print functions + */ + +#ifndef __AUFS_DEBUG_H__ +#define __AUFS_DEBUG_H__ + +#ifdef __KERNEL__ + +#include +#include +#include +#include + +#ifdef CONFIG_AUFS_DEBUG +#define AuDebugOn(a) BUG_ON(a) + +/* module parameter */ +extern atomic_t aufs_debug; +static inline void au_debug_on(void) +{ + atomic_inc(&aufs_debug); +} +static inline void au_debug_off(void) +{ + atomic_dec_if_positive(&aufs_debug); +} + +static inline int au_debug_test(void) +{ + return atomic_read(&aufs_debug) > 0; +} +#else +#define AuDebugOn(a) do {} while (0) +AuStubVoid(au_debug_on, void) +AuStubVoid(au_debug_off, void) +AuStubInt0(au_debug_test, void) +#endif /* CONFIG_AUFS_DEBUG */ + +#define param_check_atomic_t(name, p) __param_check(name, p, atomic_t) + +/* ---------------------------------------------------------------------- */ + +/* debug print */ + +#define AuDbg(fmt, ...) do { \ + if (au_debug_test()) \ + pr_debug("DEBUG: " fmt, ##__VA_ARGS__); \ +} while (0) +#define AuLabel(l) AuDbg(#l "\n") +#define AuIOErr(fmt, ...) pr_err("I/O Error, " fmt, ##__VA_ARGS__) +#define AuWarn1(fmt, ...) do { \ + static unsigned char _c; \ + if (!_c++) \ + pr_warn(fmt, ##__VA_ARGS__); \ +} while (0) + +#define AuErr1(fmt, ...) do { \ + static unsigned char _c; \ + if (!_c++) \ + pr_err(fmt, ##__VA_ARGS__); \ +} while (0) + +#define AuIOErr1(fmt, ...) do { \ + static unsigned char _c; \ + if (!_c++) \ + AuIOErr(fmt, ##__VA_ARGS__); \ +} while (0) + +#define AuUnsupportMsg "This operation is not supported." \ + " Please report this application to aufs-users ML." +#define AuUnsupport(fmt, ...) do { \ + pr_err(AuUnsupportMsg "\n" fmt, ##__VA_ARGS__); \ + dump_stack(); \ +} while (0) + +#define AuTraceErr(e) do { \ + if (unlikely((e) < 0)) \ + AuDbg("err %d\n", (int)(e)); \ +} while (0) + +#define AuTraceErrPtr(p) do { \ + if (IS_ERR(p)) \ + AuDbg("err %ld\n", PTR_ERR(p)); \ +} while (0) + +/* dirty macros for debug print, use with "%.*s" and caution */ +#define AuLNPair(qstr) (qstr)->len, (qstr)->name +#define AuDLNPair(d) AuLNPair(&(d)->d_name) + +/* ---------------------------------------------------------------------- */ + +struct au_sbinfo; +struct au_finfo; +struct dentry; +#ifdef CONFIG_AUFS_DEBUG +extern struct mutex au_dbg_mtx; +extern char *au_plevel; +struct au_nhash; +void au_dpri_whlist(struct au_nhash *whlist); +struct au_vdir; +void au_dpri_vdir(struct au_vdir *vdir); +struct inode; +void au_dpri_inode(struct inode *inode); +void au_dpri_dalias(struct inode *inode); +void au_dpri_dentry(struct dentry *dentry); +struct file; +void au_dpri_file(struct file *filp); +struct super_block; +void au_dpri_sb(struct super_block *sb); + +void au_dbg_sleep_jiffy(int jiffy); +struct iattr; +void au_dbg_iattr(struct iattr *ia); + +#define au_dbg_verify_dinode(d) __au_dbg_verify_dinode(d, __func__, __LINE__) +void __au_dbg_verify_dinode(struct dentry *dentry, const char *func, int line); +void au_dbg_verify_dir_parent(struct dentry *dentry, unsigned int sigen); +void au_dbg_verify_nondir_parent(struct dentry *dentry, unsigned int sigen); +void au_dbg_verify_gen(struct dentry *parent, unsigned int sigen); +void au_dbg_verify_kthread(void); + +int __init au_debug_init(void); +void au_debug_sbinfo_init(struct au_sbinfo *sbinfo); +#define AuDbgWhlist(w) do { \ + mutex_lock(&au_dbg_mtx); \ + AuDbg(#w "\n"); \ + au_dpri_whlist(w); \ + mutex_unlock(&au_dbg_mtx); \ +} while (0) + +#define AuDbgVdir(v) do { \ + mutex_lock(&au_dbg_mtx); \ + AuDbg(#v "\n"); \ + au_dpri_vdir(v); \ + mutex_unlock(&au_dbg_mtx); \ +} while (0) + +#define AuDbgInode(i) do { \ + mutex_lock(&au_dbg_mtx); \ + AuDbg(#i "\n"); \ + au_dpri_inode(i); \ + mutex_unlock(&au_dbg_mtx); \ +} while (0) + +#define AuDbgDAlias(i) do { \ + mutex_lock(&au_dbg_mtx); \ + AuDbg(#i "\n"); \ + au_dpri_dalias(i); \ + mutex_unlock(&au_dbg_mtx); \ +} while (0) + +#define AuDbgDentry(d) do { \ + mutex_lock(&au_dbg_mtx); \ + AuDbg(#d "\n"); \ + au_dpri_dentry(d); \ + mutex_unlock(&au_dbg_mtx); \ +} while (0) + +#define AuDbgFile(f) do { \ + mutex_lock(&au_dbg_mtx); \ + AuDbg(#f "\n"); \ + au_dpri_file(f); \ + mutex_unlock(&au_dbg_mtx); \ +} while (0) + +#define AuDbgSb(sb) do { \ + mutex_lock(&au_dbg_mtx); \ + AuDbg(#sb "\n"); \ + au_dpri_sb(sb); \ + mutex_unlock(&au_dbg_mtx); \ +} while (0) + +#define AuDbgSleep(sec) do { \ + AuDbg("sleep %d sec\n", sec); \ + ssleep(sec); \ +} while (0) + +#define AuDbgSleepJiffy(jiffy) do { \ + AuDbg("sleep %d jiffies\n", jiffy); \ + au_dbg_sleep_jiffy(jiffy); \ +} while (0) + +#define AuDbgIAttr(ia) do { \ + AuDbg("ia_valid 0x%x\n", (ia)->ia_valid); \ + au_dbg_iattr(ia); \ +} while (0) + +#define AuDbgSym(addr) do { \ + char sym[KSYM_SYMBOL_LEN]; \ + sprint_symbol(sym, (unsigned long)addr); \ + AuDbg("%s\n", sym); \ +} while (0) + +#define AuInfoSym(addr) do { \ + char sym[KSYM_SYMBOL_LEN]; \ + sprint_symbol(sym, (unsigned long)addr); \ + AuInfo("%s\n", sym); \ +} while (0) +#else +AuStubVoid(au_dbg_verify_dinode, struct dentry *dentry) +AuStubVoid(au_dbg_verify_dir_parent, struct dentry *dentry, unsigned int sigen) +AuStubVoid(au_dbg_verify_nondir_parent, struct dentry *dentry, + unsigned int sigen) +AuStubVoid(au_dbg_verify_gen, struct dentry *parent, unsigned int sigen) +AuStubVoid(au_dbg_verify_kthread, void) +AuStubInt0(__init au_debug_init, void) +AuStubVoid(au_debug_sbinfo_init, struct au_sbinfo *sbinfo) + +#define AuDbgWhlist(w) do {} while (0) +#define AuDbgVdir(v) do {} while (0) +#define AuDbgInode(i) do {} while (0) +#define AuDbgDAlias(i) do {} while (0) +#define AuDbgDentry(d) do {} while (0) +#define AuDbgFile(f) do {} while (0) +#define AuDbgSb(sb) do {} while (0) +#define AuDbgSleep(sec) do {} while (0) +#define AuDbgSleepJiffy(jiffy) do {} while (0) +#define AuDbgIAttr(ia) do {} while (0) +#define AuDbgSym(addr) do {} while (0) +#define AuInfoSym(addr) do {} while (0) +#endif /* CONFIG_AUFS_DEBUG */ + +/* ---------------------------------------------------------------------- */ + +#ifdef CONFIG_AUFS_MAGIC_SYSRQ +int __init au_sysrq_init(void); +void au_sysrq_fin(void); + +#ifdef CONFIG_HW_CONSOLE +#define au_dbg_blocked() do { \ + WARN_ON(1); \ + handle_sysrq('w'); \ +} while (0) +#else +AuStubVoid(au_dbg_blocked, void) +#endif + +#else +AuStubInt0(__init au_sysrq_init, void) +AuStubVoid(au_sysrq_fin, void) +AuStubVoid(au_dbg_blocked, void) +#endif /* CONFIG_AUFS_MAGIC_SYSRQ */ + +#endif /* __KERNEL__ */ +#endif /* __AUFS_DEBUG_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/dentry.c kernel.21.3.new/fs/aufs/dentry.c --- kernel.21.3.orig/fs/aufs/dentry.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/dentry.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,1096 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * lookup and dentry operations + */ + +#include +#include "aufs.h" + +#define AuLkup_ALLOW_NEG 1 +#define AuLkup_IGNORE_PERM (1 << 1) +#define au_ftest_lkup(flags, name) ((flags) & AuLkup_##name) +#define au_fset_lkup(flags, name) \ + do { (flags) |= AuLkup_##name; } while (0) +#define au_fclr_lkup(flags, name) \ + do { (flags) &= ~AuLkup_##name; } while (0) + +struct au_do_lookup_args { + unsigned int flags; + mode_t type; +}; + +/* + * returns positive/negative dentry, NULL or an error. + * NULL means whiteout-ed or not-found. + */ +static struct dentry* +au_do_lookup(struct dentry *h_parent, struct dentry *dentry, + aufs_bindex_t bindex, struct qstr *wh_name, + struct au_do_lookup_args *args) +{ + struct dentry *h_dentry; + struct inode *h_inode, *inode; + struct au_branch *br; + int wh_found, opq; + unsigned char wh_able; + const unsigned char allow_neg = !!au_ftest_lkup(args->flags, ALLOW_NEG); + const unsigned char ignore_perm = !!au_ftest_lkup(args->flags, + IGNORE_PERM); + + wh_found = 0; + br = au_sbr(dentry->d_sb, bindex); + wh_able = !!au_br_whable(br->br_perm); + if (wh_able) + wh_found = au_wh_test(h_parent, wh_name, /*try_sio*/0); + h_dentry = ERR_PTR(wh_found); + if (!wh_found) + goto real_lookup; + if (unlikely(wh_found < 0)) + goto out; + + /* We found a whiteout */ + /* au_set_dbend(dentry, bindex); */ + au_set_dbwh(dentry, bindex); + if (!allow_neg) + return NULL; /* success */ + +real_lookup: + if (!ignore_perm) + h_dentry = vfsub_lkup_one(&dentry->d_name, h_parent); + else + h_dentry = au_sio_lkup_one(&dentry->d_name, h_parent); + if (IS_ERR(h_dentry)) + goto out; + + h_inode = h_dentry->d_inode; + if (!h_inode) { + if (!allow_neg) + goto out_neg; + } else if (wh_found + || (args->type && args->type != (h_inode->i_mode & S_IFMT))) + goto out_neg; + + if (au_dbend(dentry) <= bindex) + au_set_dbend(dentry, bindex); + if (au_dbstart(dentry) < 0 || bindex < au_dbstart(dentry)) + au_set_dbstart(dentry, bindex); + au_set_h_dptr(dentry, bindex, h_dentry); + + inode = dentry->d_inode; + if (!h_inode || !S_ISDIR(h_inode->i_mode) || !wh_able + || (inode && !S_ISDIR(inode->i_mode))) + goto out; /* success */ + + mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD); + opq = au_diropq_test(h_dentry); + mutex_unlock(&h_inode->i_mutex); + if (opq > 0) + au_set_dbdiropq(dentry, bindex); + else if (unlikely(opq < 0)) { + au_set_h_dptr(dentry, bindex, NULL); + h_dentry = ERR_PTR(opq); + } + goto out; + +out_neg: + dput(h_dentry); + h_dentry = NULL; +out: + return h_dentry; +} + +static int au_test_shwh(struct super_block *sb, const struct qstr *name) +{ + if (unlikely(!au_opt_test(au_mntflags(sb), SHWH) + && !strncmp(name->name, AUFS_WH_PFX, AUFS_WH_PFX_LEN))) + return -EPERM; + return 0; +} + +/* + * returns the number of lower positive dentries, + * otherwise an error. + * can be called at unlinking with @type is zero. + */ +int au_lkup_dentry(struct dentry *dentry, aufs_bindex_t bstart, mode_t type) +{ + int npositive, err; + aufs_bindex_t bindex, btail, bdiropq; + unsigned char isdir, dirperm1; + struct qstr whname; + struct au_do_lookup_args args = { + .flags = 0, + .type = type + }; + const struct qstr *name = &dentry->d_name; + struct dentry *parent; + struct inode *inode; + struct super_block *sb; + + sb = dentry->d_sb; + err = au_test_shwh(sb, name); + if (unlikely(err)) + goto out; + + err = au_wh_name_alloc(&whname, name); + if (unlikely(err)) + goto out; + + inode = dentry->d_inode; + isdir = !!(inode && S_ISDIR(inode->i_mode)); + if (!type) + au_fset_lkup(args.flags, ALLOW_NEG); + dirperm1 = !!au_opt_test(au_mntflags(sb), DIRPERM1); + + npositive = 0; + parent = dget_parent(dentry); + btail = au_dbtaildir(parent); + for (bindex = bstart; bindex <= btail; bindex++) { + struct dentry *h_parent, *h_dentry; + struct inode *h_inode, *h_dir; + + h_dentry = au_h_dptr(dentry, bindex); + if (h_dentry) { + if (h_dentry->d_inode) + npositive++; + if (type != S_IFDIR) + break; + continue; + } + h_parent = au_h_dptr(parent, bindex); + if (!h_parent) + continue; + h_dir = h_parent->d_inode; + if (!h_dir || !S_ISDIR(h_dir->i_mode)) + continue; + + mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_PARENT); + h_dentry = au_do_lookup(h_parent, dentry, bindex, &whname, + &args); + mutex_unlock(&h_dir->i_mutex); + err = PTR_ERR(h_dentry); + if (IS_ERR(h_dentry)) + goto out_parent; + au_fclr_lkup(args.flags, ALLOW_NEG); + if (dirperm1) + au_fset_lkup(args.flags, IGNORE_PERM); + + if (au_dbwh(dentry) >= 0) + break; + if (!h_dentry) + continue; + h_inode = h_dentry->d_inode; + if (!h_inode) + continue; + npositive++; + if (!args.type) + args.type = h_inode->i_mode & S_IFMT; + if (args.type != S_IFDIR) + break; + else if (isdir) { + /* the type of lower may be different */ + bdiropq = au_dbdiropq(dentry); + if (bdiropq >= 0 && bdiropq <= bindex) + break; + } + } + + if (npositive) { + AuLabel(positive); + au_update_dbstart(dentry); + } + err = npositive; + if (unlikely(!au_opt_test(au_mntflags(sb), UDBA_NONE) + && au_dbstart(dentry) < 0)) { + err = -EIO; + AuIOErr("both of real entry and whiteout found, %.*s, err %d\n", + AuDLNPair(dentry), err); + } + +out_parent: + dput(parent); + kfree(whname.name); +out: + return err; +} + +struct dentry *au_sio_lkup_one(struct qstr *name, struct dentry *parent) +{ + struct dentry *dentry; + int wkq_err; + + if (!au_test_h_perm_sio(parent->d_inode, MAY_EXEC)) + dentry = vfsub_lkup_one(name, parent); + else { + struct vfsub_lkup_one_args args = { + .errp = &dentry, + .name = name, + .parent = parent + }; + + wkq_err = au_wkq_wait(vfsub_call_lkup_one, &args); + if (unlikely(wkq_err)) + dentry = ERR_PTR(wkq_err); + } + + return dentry; +} + +/* + * lookup @dentry on @bindex which should be negative. + */ +int au_lkup_neg(struct dentry *dentry, aufs_bindex_t bindex, int wh) +{ + int err; + struct dentry *parent, *h_parent, *h_dentry; + struct au_branch *br; + + parent = dget_parent(dentry); + h_parent = au_h_dptr(parent, bindex); + br = au_sbr(dentry->d_sb, bindex); + if (wh) + h_dentry = au_whtmp_lkup(h_parent, br, &dentry->d_name); + else + h_dentry = au_sio_lkup_one(&dentry->d_name, h_parent); + err = PTR_ERR(h_dentry); + if (IS_ERR(h_dentry)) + goto out; + if (unlikely(h_dentry->d_inode)) { + err = -EIO; + AuIOErr("%.*s should be negative on b%d.\n", + AuDLNPair(h_dentry), bindex); + dput(h_dentry); + goto out; + } + + err = 0; + if (bindex < au_dbstart(dentry)) + au_set_dbstart(dentry, bindex); + if (au_dbend(dentry) < bindex) + au_set_dbend(dentry, bindex); + au_set_h_dptr(dentry, bindex, h_dentry); + +out: + dput(parent); + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* subset of struct inode */ +struct au_iattr { + unsigned long i_ino; + /* unsigned int i_nlink; */ + kuid_t i_uid; + kgid_t i_gid; + u64 i_version; +/* + loff_t i_size; + blkcnt_t i_blocks; +*/ + umode_t i_mode; +}; + +static void au_iattr_save(struct au_iattr *ia, struct inode *h_inode) +{ + ia->i_ino = h_inode->i_ino; + /* ia->i_nlink = h_inode->i_nlink; */ + ia->i_uid = h_inode->i_uid; + ia->i_gid = h_inode->i_gid; + ia->i_version = h_inode->i_version; +/* + ia->i_size = h_inode->i_size; + ia->i_blocks = h_inode->i_blocks; +*/ + ia->i_mode = (h_inode->i_mode & S_IFMT); +} + +static int au_iattr_test(struct au_iattr *ia, struct inode *h_inode) +{ + return ia->i_ino != h_inode->i_ino + /* || ia->i_nlink != h_inode->i_nlink */ + || !uid_eq(ia->i_uid, h_inode->i_uid) + || !gid_eq(ia->i_gid, h_inode->i_gid) + || ia->i_version != h_inode->i_version +/* + || ia->i_size != h_inode->i_size + || ia->i_blocks != h_inode->i_blocks +*/ + || ia->i_mode != (h_inode->i_mode & S_IFMT); +} + +static int au_h_verify_dentry(struct dentry *h_dentry, struct dentry *h_parent, + struct au_branch *br) +{ + int err; + struct au_iattr ia; + struct inode *h_inode; + struct dentry *h_d; + struct super_block *h_sb; + + err = 0; + memset(&ia, -1, sizeof(ia)); + h_sb = h_dentry->d_sb; + h_inode = h_dentry->d_inode; + if (h_inode) + au_iattr_save(&ia, h_inode); + else if (au_test_nfs(h_sb) || au_test_fuse(h_sb)) + /* nfs d_revalidate may return 0 for negative dentry */ + /* fuse d_revalidate always return 0 for negative dentry */ + goto out; + + /* main purpose is namei.c:cached_lookup() and d_revalidate */ + h_d = vfsub_lkup_one(&h_dentry->d_name, h_parent); + err = PTR_ERR(h_d); + if (IS_ERR(h_d)) + goto out; + + err = 0; + if (unlikely(h_d != h_dentry + || h_d->d_inode != h_inode + || (h_inode && au_iattr_test(&ia, h_inode)))) + err = au_busy_or_stale(); + dput(h_d); + +out: + AuTraceErr(err); + return err; +} + +int au_h_verify(struct dentry *h_dentry, unsigned int udba, struct inode *h_dir, + struct dentry *h_parent, struct au_branch *br) +{ + int err; + + err = 0; + if (udba == AuOpt_UDBA_REVAL + && !au_test_fs_remote(h_dentry->d_sb)) { + IMustLock(h_dir); + err = (h_dentry->d_parent->d_inode != h_dir); + } else if (udba != AuOpt_UDBA_NONE) + err = au_h_verify_dentry(h_dentry, h_parent, br); + + return err; +} + +/* ---------------------------------------------------------------------- */ + +static int au_do_refresh_hdentry(struct dentry *dentry, struct dentry *parent) +{ + int err; + aufs_bindex_t new_bindex, bindex, bend, bwh, bdiropq; + struct au_hdentry tmp, *p, *q; + struct au_dinfo *dinfo; + struct super_block *sb; + + DiMustWriteLock(dentry); + + sb = dentry->d_sb; + dinfo = au_di(dentry); + bend = dinfo->di_bend; + bwh = dinfo->di_bwh; + bdiropq = dinfo->di_bdiropq; + p = dinfo->di_hdentry + dinfo->di_bstart; + for (bindex = dinfo->di_bstart; bindex <= bend; bindex++, p++) { + if (!p->hd_dentry) + continue; + + new_bindex = au_br_index(sb, p->hd_id); + if (new_bindex == bindex) + continue; + + if (dinfo->di_bwh == bindex) + bwh = new_bindex; + if (dinfo->di_bdiropq == bindex) + bdiropq = new_bindex; + if (new_bindex < 0) { + au_hdput(p); + p->hd_dentry = NULL; + continue; + } + + /* swap two lower dentries, and loop again */ + q = dinfo->di_hdentry + new_bindex; + tmp = *q; + *q = *p; + *p = tmp; + if (tmp.hd_dentry) { + bindex--; + p--; + } + } + + dinfo->di_bwh = -1; + if (bwh >= 0 && bwh <= au_sbend(sb) && au_sbr_whable(sb, bwh)) + dinfo->di_bwh = bwh; + + dinfo->di_bdiropq = -1; + if (bdiropq >= 0 + && bdiropq <= au_sbend(sb) + && au_sbr_whable(sb, bdiropq)) + dinfo->di_bdiropq = bdiropq; + + err = -EIO; + dinfo->di_bstart = -1; + dinfo->di_bend = -1; + bend = au_dbend(parent); + p = dinfo->di_hdentry; + for (bindex = 0; bindex <= bend; bindex++, p++) + if (p->hd_dentry) { + dinfo->di_bstart = bindex; + break; + } + + if (dinfo->di_bstart >= 0) { + p = dinfo->di_hdentry + bend; + for (bindex = bend; bindex >= 0; bindex--, p--) + if (p->hd_dentry) { + dinfo->di_bend = bindex; + err = 0; + break; + } + } + + return err; +} + +static void au_do_hide(struct dentry *dentry) +{ + struct inode *inode; + + inode = dentry->d_inode; + if (inode) { + if (!S_ISDIR(inode->i_mode)) { + if (inode->i_nlink && !d_unhashed(dentry)) + drop_nlink(inode); + } else { + clear_nlink(inode); + /* stop next lookup */ + inode->i_flags |= S_DEAD; + } + smp_mb(); /* necessary? */ + } + d_drop(dentry); +} + +static int au_hide_children(struct dentry *parent) +{ + int err, i, j, ndentry; + struct au_dcsub_pages dpages; + struct au_dpage *dpage; + struct dentry *dentry; + + err = au_dpages_init(&dpages, GFP_NOFS); + if (unlikely(err)) + goto out; + err = au_dcsub_pages(&dpages, parent, NULL, NULL); + if (unlikely(err)) + goto out_dpages; + + /* in reverse order */ + for (i = dpages.ndpage - 1; i >= 0; i--) { + dpage = dpages.dpages + i; + ndentry = dpage->ndentry; + for (j = ndentry - 1; j >= 0; j--) { + dentry = dpage->dentries[j]; + if (dentry != parent) + au_do_hide(dentry); + } + } + +out_dpages: + au_dpages_free(&dpages); +out: + return err; +} + +static void au_hide(struct dentry *dentry) +{ + int err; + struct inode *inode; + + AuDbgDentry(dentry); + inode = dentry->d_inode; + if (inode && S_ISDIR(inode->i_mode)) { + /* shrink_dcache_parent(dentry); */ + err = au_hide_children(dentry); + if (unlikely(err)) + AuIOErr("%.*s, failed hiding children, ignored %d\n", + AuDLNPair(dentry), err); + } + au_do_hide(dentry); +} + +/* + * By adding a dirty branch, a cached dentry may be affected in various ways. + * + * a dirty branch is added + * - on the top of layers + * - in the middle of layers + * - to the bottom of layers + * + * on the added branch there exists + * - a whiteout + * - a diropq + * - a same named entry + * + exist + * * negative --> positive + * * positive --> positive + * - type is unchanged + * - type is changed + * + doesn't exist + * * negative --> negative + * * positive --> negative (rejected by au_br_del() for non-dir case) + * - none + */ +static int au_refresh_by_dinfo(struct dentry *dentry, struct au_dinfo *dinfo, + struct au_dinfo *tmp) +{ + int err; + aufs_bindex_t bindex, bend; + struct { + struct dentry *dentry; + struct inode *inode; + mode_t mode; + } orig_h, tmp_h; + struct au_hdentry *hd; + struct inode *inode, *h_inode; + struct dentry *h_dentry; + + err = 0; + AuDebugOn(dinfo->di_bstart < 0); + orig_h.dentry = dinfo->di_hdentry[dinfo->di_bstart].hd_dentry; + orig_h.inode = orig_h.dentry->d_inode; + orig_h.mode = 0; + if (orig_h.inode) + orig_h.mode = orig_h.inode->i_mode & S_IFMT; + memset(&tmp_h, 0, sizeof(tmp_h)); + if (tmp->di_bstart >= 0) { + tmp_h.dentry = tmp->di_hdentry[tmp->di_bstart].hd_dentry; + tmp_h.inode = tmp_h.dentry->d_inode; + if (tmp_h.inode) + tmp_h.mode = tmp_h.inode->i_mode & S_IFMT; + } + + inode = dentry->d_inode; + if (!orig_h.inode) { + AuDbg("nagative originally\n"); + if (inode) { + au_hide(dentry); + goto out; + } + AuDebugOn(inode); + AuDebugOn(dinfo->di_bstart != dinfo->di_bend); + AuDebugOn(dinfo->di_bdiropq != -1); + + if (!tmp_h.inode) { + AuDbg("negative --> negative\n"); + /* should have only one negative lower */ + if (tmp->di_bstart >= 0 + && tmp->di_bstart < dinfo->di_bstart) { + AuDebugOn(tmp->di_bstart != tmp->di_bend); + AuDebugOn(dinfo->di_bstart != dinfo->di_bend); + au_set_h_dptr(dentry, dinfo->di_bstart, NULL); + au_di_cp(dinfo, tmp); + hd = tmp->di_hdentry + tmp->di_bstart; + au_set_h_dptr(dentry, tmp->di_bstart, + dget(hd->hd_dentry)); + } + au_dbg_verify_dinode(dentry); + } else { + AuDbg("negative --> positive\n"); + /* + * similar to the behaviour of creating with bypassing + * aufs. + * unhash it in order to force an error in the + * succeeding create operation. + * we should not set S_DEAD here. + */ + d_drop(dentry); + /* au_di_swap(tmp, dinfo); */ + au_dbg_verify_dinode(dentry); + } + } else { + AuDbg("positive originally\n"); + /* inode may be NULL */ + AuDebugOn(inode && (inode->i_mode & S_IFMT) != orig_h.mode); + if (!tmp_h.inode) { + AuDbg("positive --> negative\n"); + /* or bypassing aufs */ + au_hide(dentry); + if (tmp->di_bwh >= 0 && tmp->di_bwh <= dinfo->di_bstart) + dinfo->di_bwh = tmp->di_bwh; + if (inode) + err = au_refresh_hinode_self(inode); + au_dbg_verify_dinode(dentry); + } else if (orig_h.mode == tmp_h.mode) { + AuDbg("positive --> positive, same type\n"); + if (!S_ISDIR(orig_h.mode) + && dinfo->di_bstart > tmp->di_bstart) { + /* + * similar to the behaviour of removing and + * creating. + */ + au_hide(dentry); + if (inode) + err = au_refresh_hinode_self(inode); + au_dbg_verify_dinode(dentry); + } else { + /* fill empty slots */ + if (dinfo->di_bstart > tmp->di_bstart) + dinfo->di_bstart = tmp->di_bstart; + if (dinfo->di_bend < tmp->di_bend) + dinfo->di_bend = tmp->di_bend; + dinfo->di_bwh = tmp->di_bwh; + dinfo->di_bdiropq = tmp->di_bdiropq; + hd = tmp->di_hdentry; + bend = dinfo->di_bend; + for (bindex = tmp->di_bstart; bindex <= bend; + bindex++) { + if (au_h_dptr(dentry, bindex)) + continue; + h_dentry = hd[bindex].hd_dentry; + if (!h_dentry) + continue; + h_inode = h_dentry->d_inode; + AuDebugOn(!h_inode); + AuDebugOn(orig_h.mode + != (h_inode->i_mode + & S_IFMT)); + au_set_h_dptr(dentry, bindex, + dget(h_dentry)); + } + err = au_refresh_hinode(inode, dentry); + au_dbg_verify_dinode(dentry); + } + } else { + AuDbg("positive --> positive, different type\n"); + /* similar to the behaviour of removing and creating */ + au_hide(dentry); + if (inode) + err = au_refresh_hinode_self(inode); + au_dbg_verify_dinode(dentry); + } + } + +out: + return err; +} + +int au_refresh_dentry(struct dentry *dentry, struct dentry *parent) +{ + int err, ebrange; + unsigned int sigen; + struct au_dinfo *dinfo, *tmp; + struct super_block *sb; + struct inode *inode; + + DiMustWriteLock(dentry); + AuDebugOn(IS_ROOT(dentry)); + AuDebugOn(!parent->d_inode); + + sb = dentry->d_sb; + inode = dentry->d_inode; + sigen = au_sigen(sb); + err = au_digen_test(parent, sigen); + if (unlikely(err)) + goto out; + + dinfo = au_di(dentry); + err = au_di_realloc(dinfo, au_sbend(sb) + 1); + if (unlikely(err)) + goto out; + ebrange = au_dbrange_test(dentry); + if (!ebrange) + ebrange = au_do_refresh_hdentry(dentry, parent); + + if (d_unhashed(dentry) || ebrange) { + AuDebugOn(au_dbstart(dentry) < 0 && au_dbend(dentry) >= 0); + if (inode) + err = au_refresh_hinode_self(inode); + au_dbg_verify_dinode(dentry); + if (!err) + goto out_dgen; /* success */ + goto out; + } + + /* temporary dinfo */ + AuDbgDentry(dentry); + err = -ENOMEM; + tmp = au_di_alloc(sb, AuLsc_DI_TMP); + if (unlikely(!tmp)) + goto out; + au_di_swap(tmp, dinfo); + /* returns the number of positive dentries */ + /* + * if current working dir is removed, it returns an error. + * but the dentry is legal. + */ + err = au_lkup_dentry(dentry, /*bstart*/0, /*type*/0); + AuDbgDentry(dentry); + au_di_swap(tmp, dinfo); + if (err == -ENOENT) + err = 0; + if (err >= 0) { + /* compare/refresh by dinfo */ + AuDbgDentry(dentry); + err = au_refresh_by_dinfo(dentry, dinfo, tmp); + au_dbg_verify_dinode(dentry); + AuTraceErr(err); + } + au_rw_write_unlock(&tmp->di_rwsem); + au_di_free(tmp); + if (unlikely(err)) + goto out; + +out_dgen: + au_update_digen(dentry); +out: + if (unlikely(err && !(dentry->d_flags & DCACHE_NFSFS_RENAMED))) { + AuIOErr("failed refreshing %.*s, %d\n", + AuDLNPair(dentry), err); + AuDbgDentry(dentry); + } + AuTraceErr(err); + return err; +} + +static int au_do_h_d_reval(struct dentry *h_dentry, unsigned int flags, + struct dentry *dentry, aufs_bindex_t bindex) +{ + int err, valid; + + err = 0; + if (!(h_dentry->d_flags & DCACHE_OP_REVALIDATE)) + goto out; + + AuDbg("b%d\n", bindex); + /* + * gave up supporting LOOKUP_CREATE/OPEN for lower fs, + * due to whiteout and branch permission. + */ + flags &= ~(/*LOOKUP_PARENT |*/ LOOKUP_OPEN | LOOKUP_CREATE + | LOOKUP_FOLLOW | LOOKUP_EXCL); + /* it may return tri-state */ + valid = h_dentry->d_op->d_revalidate(h_dentry, flags); + + if (unlikely(valid < 0)) + err = valid; + else if (!valid) + err = -EINVAL; + +out: + AuTraceErr(err); + return err; +} + +/* todo: remove this */ +static int h_d_revalidate(struct dentry *dentry, struct inode *inode, + unsigned int flags, int do_udba) +{ + int err; + umode_t mode, h_mode; + aufs_bindex_t bindex, btail, bstart, ibs, ibe; + unsigned char plus, unhashed, is_root, h_plus, h_nfs; + struct inode *h_inode, *h_cached_inode; + struct dentry *h_dentry; + struct qstr *name, *h_name; + + err = 0; + plus = 0; + mode = 0; + ibs = -1; + ibe = -1; + unhashed = !!d_unhashed(dentry); + is_root = !!IS_ROOT(dentry); + name = &dentry->d_name; + + /* + * Theoretically, REVAL test should be unnecessary in case of + * {FS,I}NOTIFY. + * But {fs,i}notify doesn't fire some necessary events, + * IN_ATTRIB for atime/nlink/pageio + * IN_DELETE for NFS dentry + * Let's do REVAL test too. + */ + if (do_udba && inode) { + mode = (inode->i_mode & S_IFMT); + plus = (inode->i_nlink > 0); + ibs = au_ibstart(inode); + ibe = au_ibend(inode); + } + + bstart = au_dbstart(dentry); + btail = bstart; + if (inode && S_ISDIR(inode->i_mode)) + btail = au_dbtaildir(dentry); + for (bindex = bstart; bindex <= btail; bindex++) { + h_dentry = au_h_dptr(dentry, bindex); + if (!h_dentry) + continue; + + AuDbg("b%d, %.*s\n", bindex, AuDLNPair(h_dentry)); + h_nfs = !!au_test_nfs(h_dentry->d_sb); + spin_lock(&h_dentry->d_lock); + h_name = &h_dentry->d_name; + if (unlikely(do_udba + && !is_root + && ((!h_nfs + && (unhashed != !!d_unhashed(h_dentry) + || !au_qstreq(name, h_name))) + || (h_nfs + && !(flags & LOOKUP_OPEN) + && (h_dentry->d_flags + & DCACHE_NFSFS_RENAMED))) + )) { + int h_unhashed; + + h_unhashed = d_unhashed(h_dentry); + spin_unlock(&h_dentry->d_lock); + AuDbg("unhash 0x%x 0x%x, %.*s %.*s\n", + unhashed, h_unhashed, + AuDLNPair(dentry), AuDLNPair(h_dentry)); + goto err; + } + spin_unlock(&h_dentry->d_lock); + + err = au_do_h_d_reval(h_dentry, flags, dentry, bindex); + if (unlikely(err)) + /* do not goto err, to keep the errno */ + break; + + /* todo: plink too? */ + if (!do_udba) + continue; + + /* UDBA tests */ + h_inode = h_dentry->d_inode; + if (unlikely(!!inode != !!h_inode)) + goto err; + + h_plus = plus; + h_mode = mode; + h_cached_inode = h_inode; + if (h_inode) { + h_mode = (h_inode->i_mode & S_IFMT); + h_plus = (h_inode->i_nlink > 0); + } + if (inode && ibs <= bindex && bindex <= ibe) + h_cached_inode = au_h_iptr(inode, bindex); + + if (!h_nfs) { + if (unlikely(plus != h_plus)) + goto err; + } else { + if (unlikely(!(h_dentry->d_flags & DCACHE_NFSFS_RENAMED) + && !is_root + && !IS_ROOT(h_dentry) + && unhashed != d_unhashed(h_dentry))) + goto err; + } + if (unlikely(mode != h_mode + || h_cached_inode != h_inode)) + goto err; + continue; + +err: + err = -EINVAL; + break; + } + + AuTraceErr(err); + return err; +} + +/* todo: consolidate with do_refresh() and au_reval_for_attr() */ +static int simple_reval_dpath(struct dentry *dentry, unsigned int sigen) +{ + int err; + struct dentry *parent; + + if (!au_digen_test(dentry, sigen)) + return 0; + + parent = dget_parent(dentry); + di_read_lock_parent(parent, AuLock_IR); + AuDebugOn(au_digen_test(parent, sigen)); + au_dbg_verify_gen(parent, sigen); + err = au_refresh_dentry(dentry, parent); + di_read_unlock(parent, AuLock_IR); + dput(parent); + AuTraceErr(err); + return err; +} + +int au_reval_dpath(struct dentry *dentry, unsigned int sigen) +{ + int err; + struct dentry *d, *parent; + struct inode *inode; + + if (!au_ftest_si(au_sbi(dentry->d_sb), FAILED_REFRESH_DIR)) + return simple_reval_dpath(dentry, sigen); + + /* slow loop, keep it simple and stupid */ + /* cf: au_cpup_dirs() */ + err = 0; + parent = NULL; + while (au_digen_test(dentry, sigen)) { + d = dentry; + while (1) { + dput(parent); + parent = dget_parent(d); + if (!au_digen_test(parent, sigen)) + break; + d = parent; + } + + inode = d->d_inode; + if (d != dentry) + di_write_lock_child2(d); + + /* someone might update our dentry while we were sleeping */ + if (au_digen_test(d, sigen)) { + /* + * todo: consolidate with simple_reval_dpath(), + * do_refresh() and au_reval_for_attr(). + */ + di_read_lock_parent(parent, AuLock_IR); + err = au_refresh_dentry(d, parent); + di_read_unlock(parent, AuLock_IR); + } + + if (d != dentry) + di_write_unlock(d); + dput(parent); + if (unlikely(err)) + break; + } + + return err; +} + +/* + * if valid returns 1, otherwise 0. + */ +static int aufs_d_revalidate(struct dentry *dentry, unsigned int flags) +{ + int valid, err; + unsigned int sigen; + unsigned char do_udba; + struct super_block *sb; + struct inode *inode; + + /* todo: support rcu-walk? */ + if (flags & LOOKUP_RCU) + return -ECHILD; + + valid = 0; + if (unlikely(!au_di(dentry))) + goto out; + + valid = 1; + sb = dentry->d_sb; + /* + * todo: very ugly + * i_mutex of parent dir may be held, + * but we should not return 'invalid' due to busy. + */ + err = aufs_read_lock(dentry, AuLock_FLUSH | AuLock_DW | AuLock_NOPLM); + if (unlikely(err)) { + valid = err; + AuTraceErr(err); + goto out; + } + inode = dentry->d_inode; + if (unlikely(inode && is_bad_inode(inode))) { + err = -EINVAL; + AuTraceErr(err); + goto out_dgrade; + } + if (unlikely(au_dbrange_test(dentry))) { + err = -EINVAL; + AuTraceErr(err); + goto out_dgrade; + } + + sigen = au_sigen(sb); + if (au_digen_test(dentry, sigen)) { + AuDebugOn(IS_ROOT(dentry)); + err = au_reval_dpath(dentry, sigen); + if (unlikely(err)) { + AuTraceErr(err); + goto out_dgrade; + } + } + di_downgrade_lock(dentry, AuLock_IR); + + err = -EINVAL; + if (!(flags & (LOOKUP_OPEN | LOOKUP_EMPTY)) + && inode + && (IS_DEADDIR(inode) || !inode->i_nlink)) + goto out_inval; + + do_udba = !au_opt_test(au_mntflags(sb), UDBA_NONE); + if (do_udba && inode) { + aufs_bindex_t bstart = au_ibstart(inode); + struct inode *h_inode; + + if (bstart >= 0) { + h_inode = au_h_iptr(inode, bstart); + if (h_inode && au_test_higen(inode, h_inode)) + goto out_inval; + } + } + + err = h_d_revalidate(dentry, inode, flags, do_udba); + if (unlikely(!err && do_udba && au_dbstart(dentry) < 0)) { + err = -EIO; + AuDbg("both of real entry and whiteout found, %.*s, err %d\n", + AuDLNPair(dentry), err); + } + goto out_inval; + +out_dgrade: + di_downgrade_lock(dentry, AuLock_IR); +out_inval: + aufs_read_unlock(dentry, AuLock_IR); + AuTraceErr(err); + valid = !err; +out: + if (!valid) { + AuDbg("%.*s invalid, %d\n", AuDLNPair(dentry), valid); + d_drop(dentry); + } + return valid; +} + +static void aufs_d_release(struct dentry *dentry) +{ + if (au_di(dentry)) { + au_di_fin(dentry); + au_hn_di_reinit(dentry); + } +} + +const struct dentry_operations aufs_dop = { + .d_revalidate = aufs_d_revalidate, + .d_weak_revalidate = aufs_d_revalidate, + .d_release = aufs_d_release +}; diff -Naur kernel.21.3.orig/fs/aufs/dentry.h kernel.21.3.new/fs/aufs/dentry.h --- kernel.21.3.orig/fs/aufs/dentry.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/dentry.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,232 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * lookup and dentry operations + */ + +#ifndef __AUFS_DENTRY_H__ +#define __AUFS_DENTRY_H__ + +#ifdef __KERNEL__ + +#include +#include "rwsem.h" + +struct au_hdentry { + struct dentry *hd_dentry; + aufs_bindex_t hd_id; +}; + +struct au_dinfo { + atomic_t di_generation; + + struct au_rwsem di_rwsem; + aufs_bindex_t di_bstart, di_bend, di_bwh, di_bdiropq; + struct au_hdentry *di_hdentry; +} ____cacheline_aligned_in_smp; + +/* ---------------------------------------------------------------------- */ + +/* dentry.c */ +extern const struct dentry_operations aufs_dop; +struct au_branch; +struct dentry *au_sio_lkup_one(struct qstr *name, struct dentry *parent); +int au_h_verify(struct dentry *h_dentry, unsigned int udba, struct inode *h_dir, + struct dentry *h_parent, struct au_branch *br); + +int au_lkup_dentry(struct dentry *dentry, aufs_bindex_t bstart, mode_t type); +int au_lkup_neg(struct dentry *dentry, aufs_bindex_t bindex, int wh); +int au_refresh_dentry(struct dentry *dentry, struct dentry *parent); +int au_reval_dpath(struct dentry *dentry, unsigned int sigen); + +/* dinfo.c */ +void au_di_init_once(void *_di); +struct au_dinfo *au_di_alloc(struct super_block *sb, unsigned int lsc); +void au_di_free(struct au_dinfo *dinfo); +void au_di_swap(struct au_dinfo *a, struct au_dinfo *b); +void au_di_cp(struct au_dinfo *dst, struct au_dinfo *src); +int au_di_init(struct dentry *dentry); +void au_di_fin(struct dentry *dentry); +int au_di_realloc(struct au_dinfo *dinfo, int nbr); + +void di_read_lock(struct dentry *d, int flags, unsigned int lsc); +void di_read_unlock(struct dentry *d, int flags); +void di_downgrade_lock(struct dentry *d, int flags); +void di_write_lock(struct dentry *d, unsigned int lsc); +void di_write_unlock(struct dentry *d); +void di_write_lock2_child(struct dentry *d1, struct dentry *d2, int isdir); +void di_write_lock2_parent(struct dentry *d1, struct dentry *d2, int isdir); +void di_write_unlock2(struct dentry *d1, struct dentry *d2); + +struct dentry *au_h_dptr(struct dentry *dentry, aufs_bindex_t bindex); +struct dentry *au_h_d_alias(struct dentry *dentry, aufs_bindex_t bindex); +aufs_bindex_t au_dbtail(struct dentry *dentry); +aufs_bindex_t au_dbtaildir(struct dentry *dentry); + +void au_set_h_dptr(struct dentry *dentry, aufs_bindex_t bindex, + struct dentry *h_dentry); +int au_digen_test(struct dentry *dentry, unsigned int sigen); +int au_dbrange_test(struct dentry *dentry); +void au_update_digen(struct dentry *dentry); +void au_update_dbrange(struct dentry *dentry, int do_put_zero); +void au_update_dbstart(struct dentry *dentry); +void au_update_dbend(struct dentry *dentry); +int au_find_dbindex(struct dentry *dentry, struct dentry *h_dentry); + +/* ---------------------------------------------------------------------- */ + +static inline struct au_dinfo *au_di(struct dentry *dentry) +{ + return dentry->d_fsdata; +} + +/* ---------------------------------------------------------------------- */ + +/* lock subclass for dinfo */ +enum { + AuLsc_DI_CHILD, /* child first */ + AuLsc_DI_CHILD2, /* rename(2), link(2), and cpup at hnotify */ + AuLsc_DI_CHILD3, /* copyup dirs */ + AuLsc_DI_PARENT, + AuLsc_DI_PARENT2, + AuLsc_DI_PARENT3, + AuLsc_DI_TMP /* temp for replacing dinfo */ +}; + +/* + * di_read_lock_child, di_write_lock_child, + * di_read_lock_child2, di_write_lock_child2, + * di_read_lock_child3, di_write_lock_child3, + * di_read_lock_parent, di_write_lock_parent, + * di_read_lock_parent2, di_write_lock_parent2, + * di_read_lock_parent3, di_write_lock_parent3, + */ +#define AuReadLockFunc(name, lsc) \ +static inline void di_read_lock_##name(struct dentry *d, int flags) \ +{ di_read_lock(d, flags, AuLsc_DI_##lsc); } + +#define AuWriteLockFunc(name, lsc) \ +static inline void di_write_lock_##name(struct dentry *d) \ +{ di_write_lock(d, AuLsc_DI_##lsc); } + +#define AuRWLockFuncs(name, lsc) \ + AuReadLockFunc(name, lsc) \ + AuWriteLockFunc(name, lsc) + +AuRWLockFuncs(child, CHILD); +AuRWLockFuncs(child2, CHILD2); +AuRWLockFuncs(child3, CHILD3); +AuRWLockFuncs(parent, PARENT); +AuRWLockFuncs(parent2, PARENT2); +AuRWLockFuncs(parent3, PARENT3); + +#undef AuReadLockFunc +#undef AuWriteLockFunc +#undef AuRWLockFuncs + +#define DiMustNoWaiters(d) AuRwMustNoWaiters(&au_di(d)->di_rwsem) +#define DiMustAnyLock(d) AuRwMustAnyLock(&au_di(d)->di_rwsem) +#define DiMustWriteLock(d) AuRwMustWriteLock(&au_di(d)->di_rwsem) + +/* ---------------------------------------------------------------------- */ + +/* todo: memory barrier? */ +static inline unsigned int au_digen(struct dentry *d) +{ + return atomic_read(&au_di(d)->di_generation); +} + +static inline void au_h_dentry_init(struct au_hdentry *hdentry) +{ + hdentry->hd_dentry = NULL; +} + +static inline void au_hdput(struct au_hdentry *hd) +{ + if (hd) + dput(hd->hd_dentry); +} + +static inline aufs_bindex_t au_dbstart(struct dentry *dentry) +{ + DiMustAnyLock(dentry); + return au_di(dentry)->di_bstart; +} + +static inline aufs_bindex_t au_dbend(struct dentry *dentry) +{ + DiMustAnyLock(dentry); + return au_di(dentry)->di_bend; +} + +static inline aufs_bindex_t au_dbwh(struct dentry *dentry) +{ + DiMustAnyLock(dentry); + return au_di(dentry)->di_bwh; +} + +static inline aufs_bindex_t au_dbdiropq(struct dentry *dentry) +{ + DiMustAnyLock(dentry); + return au_di(dentry)->di_bdiropq; +} + +/* todo: hard/soft set? */ +static inline void au_set_dbstart(struct dentry *dentry, aufs_bindex_t bindex) +{ + DiMustWriteLock(dentry); + au_di(dentry)->di_bstart = bindex; +} + +static inline void au_set_dbend(struct dentry *dentry, aufs_bindex_t bindex) +{ + DiMustWriteLock(dentry); + au_di(dentry)->di_bend = bindex; +} + +static inline void au_set_dbwh(struct dentry *dentry, aufs_bindex_t bindex) +{ + DiMustWriteLock(dentry); + /* dbwh can be outside of bstart - bend range */ + au_di(dentry)->di_bwh = bindex; +} + +static inline void au_set_dbdiropq(struct dentry *dentry, aufs_bindex_t bindex) +{ + DiMustWriteLock(dentry); + au_di(dentry)->di_bdiropq = bindex; +} + +/* ---------------------------------------------------------------------- */ + +#ifdef CONFIG_AUFS_HNOTIFY +static inline void au_digen_dec(struct dentry *d) +{ + atomic_dec(&au_di(d)->di_generation); +} + +static inline void au_hn_di_reinit(struct dentry *dentry) +{ + dentry->d_fsdata = NULL; +} +#else +AuStubVoid(au_hn_di_reinit, struct dentry *dentry __maybe_unused) +#endif /* CONFIG_AUFS_HNOTIFY */ + +#endif /* __KERNEL__ */ +#endif /* __AUFS_DENTRY_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/dinfo.c kernel.21.3.new/fs/aufs/dinfo.c --- kernel.21.3.orig/fs/aufs/dinfo.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/dinfo.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,542 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * dentry private data + */ + +#include "aufs.h" + +void au_di_init_once(void *_dinfo) +{ + struct au_dinfo *dinfo = _dinfo; + static struct lock_class_key aufs_di; + + au_rw_init(&dinfo->di_rwsem); + au_rw_class(&dinfo->di_rwsem, &aufs_di); +} + +struct au_dinfo *au_di_alloc(struct super_block *sb, unsigned int lsc) +{ + struct au_dinfo *dinfo; + int nbr, i; + + dinfo = au_cache_alloc_dinfo(); + if (unlikely(!dinfo)) + goto out; + + nbr = au_sbend(sb) + 1; + if (nbr <= 0) + nbr = 1; + dinfo->di_hdentry = kcalloc(nbr, sizeof(*dinfo->di_hdentry), GFP_NOFS); + if (dinfo->di_hdentry) { + au_rw_write_lock_nested(&dinfo->di_rwsem, lsc); + dinfo->di_bstart = -1; + dinfo->di_bend = -1; + dinfo->di_bwh = -1; + dinfo->di_bdiropq = -1; + for (i = 0; i < nbr; i++) + dinfo->di_hdentry[i].hd_id = -1; + goto out; + } + + au_cache_free_dinfo(dinfo); + dinfo = NULL; + +out: + return dinfo; +} + +void au_di_free(struct au_dinfo *dinfo) +{ + struct au_hdentry *p; + aufs_bindex_t bend, bindex; + + /* dentry may not be revalidated */ + bindex = dinfo->di_bstart; + if (bindex >= 0) { + bend = dinfo->di_bend; + p = dinfo->di_hdentry + bindex; + while (bindex++ <= bend) + au_hdput(p++); + } + kfree(dinfo->di_hdentry); + au_cache_free_dinfo(dinfo); +} + +void au_di_swap(struct au_dinfo *a, struct au_dinfo *b) +{ + struct au_hdentry *p; + aufs_bindex_t bi; + + AuRwMustWriteLock(&a->di_rwsem); + AuRwMustWriteLock(&b->di_rwsem); + +#define DiSwap(v, name) \ + do { \ + v = a->di_##name; \ + a->di_##name = b->di_##name; \ + b->di_##name = v; \ + } while (0) + + DiSwap(p, hdentry); + DiSwap(bi, bstart); + DiSwap(bi, bend); + DiSwap(bi, bwh); + DiSwap(bi, bdiropq); + /* smp_mb(); */ + +#undef DiSwap +} + +void au_di_cp(struct au_dinfo *dst, struct au_dinfo *src) +{ + AuRwMustWriteLock(&dst->di_rwsem); + AuRwMustWriteLock(&src->di_rwsem); + + dst->di_bstart = src->di_bstart; + dst->di_bend = src->di_bend; + dst->di_bwh = src->di_bwh; + dst->di_bdiropq = src->di_bdiropq; + /* smp_mb(); */ +} + +int au_di_init(struct dentry *dentry) +{ + int err; + struct super_block *sb; + struct au_dinfo *dinfo; + + err = 0; + sb = dentry->d_sb; + dinfo = au_di_alloc(sb, AuLsc_DI_CHILD); + if (dinfo) { + atomic_set(&dinfo->di_generation, au_sigen(sb)); + /* smp_mb(); */ /* atomic_set */ + dentry->d_fsdata = dinfo; + } else + err = -ENOMEM; + + return err; +} + +void au_di_fin(struct dentry *dentry) +{ + struct au_dinfo *dinfo; + + dinfo = au_di(dentry); + AuRwDestroy(&dinfo->di_rwsem); + au_di_free(dinfo); +} + +int au_di_realloc(struct au_dinfo *dinfo, int nbr) +{ + int err, sz; + struct au_hdentry *hdp; + + AuRwMustWriteLock(&dinfo->di_rwsem); + + err = -ENOMEM; + sz = sizeof(*hdp) * (dinfo->di_bend + 1); + if (!sz) + sz = sizeof(*hdp); + hdp = au_kzrealloc(dinfo->di_hdentry, sz, sizeof(*hdp) * nbr, GFP_NOFS); + if (hdp) { + dinfo->di_hdentry = hdp; + err = 0; + } + + return err; +} + +/* ---------------------------------------------------------------------- */ + +static void do_ii_write_lock(struct inode *inode, unsigned int lsc) +{ + switch (lsc) { + case AuLsc_DI_CHILD: + ii_write_lock_child(inode); + break; + case AuLsc_DI_CHILD2: + ii_write_lock_child2(inode); + break; + case AuLsc_DI_CHILD3: + ii_write_lock_child3(inode); + break; + case AuLsc_DI_PARENT: + ii_write_lock_parent(inode); + break; + case AuLsc_DI_PARENT2: + ii_write_lock_parent2(inode); + break; + case AuLsc_DI_PARENT3: + ii_write_lock_parent3(inode); + break; + default: + BUG(); + } +} + +static void do_ii_read_lock(struct inode *inode, unsigned int lsc) +{ + switch (lsc) { + case AuLsc_DI_CHILD: + ii_read_lock_child(inode); + break; + case AuLsc_DI_CHILD2: + ii_read_lock_child2(inode); + break; + case AuLsc_DI_CHILD3: + ii_read_lock_child3(inode); + break; + case AuLsc_DI_PARENT: + ii_read_lock_parent(inode); + break; + case AuLsc_DI_PARENT2: + ii_read_lock_parent2(inode); + break; + case AuLsc_DI_PARENT3: + ii_read_lock_parent3(inode); + break; + default: + BUG(); + } +} + +void di_read_lock(struct dentry *d, int flags, unsigned int lsc) +{ + au_rw_read_lock_nested(&au_di(d)->di_rwsem, lsc); + if (d->d_inode) { + if (au_ftest_lock(flags, IW)) + do_ii_write_lock(d->d_inode, lsc); + else if (au_ftest_lock(flags, IR)) + do_ii_read_lock(d->d_inode, lsc); + } +} + +void di_read_unlock(struct dentry *d, int flags) +{ + if (d->d_inode) { + if (au_ftest_lock(flags, IW)) { + au_dbg_verify_dinode(d); + ii_write_unlock(d->d_inode); + } else if (au_ftest_lock(flags, IR)) { + au_dbg_verify_dinode(d); + ii_read_unlock(d->d_inode); + } + } + au_rw_read_unlock(&au_di(d)->di_rwsem); +} + +void di_downgrade_lock(struct dentry *d, int flags) +{ + if (d->d_inode && au_ftest_lock(flags, IR)) + ii_downgrade_lock(d->d_inode); + au_rw_dgrade_lock(&au_di(d)->di_rwsem); +} + +void di_write_lock(struct dentry *d, unsigned int lsc) +{ + au_rw_write_lock_nested(&au_di(d)->di_rwsem, lsc); + if (d->d_inode) + do_ii_write_lock(d->d_inode, lsc); +} + +void di_write_unlock(struct dentry *d) +{ + au_dbg_verify_dinode(d); + if (d->d_inode) + ii_write_unlock(d->d_inode); + au_rw_write_unlock(&au_di(d)->di_rwsem); +} + +void di_write_lock2_child(struct dentry *d1, struct dentry *d2, int isdir) +{ + AuDebugOn(d1 == d2 + || d1->d_inode == d2->d_inode + || d1->d_sb != d2->d_sb); + + if (isdir && au_test_subdir(d1, d2)) { + di_write_lock_child(d1); + di_write_lock_child2(d2); + } else { + /* there should be no races */ + di_write_lock_child(d2); + di_write_lock_child2(d1); + } +} + +void di_write_lock2_parent(struct dentry *d1, struct dentry *d2, int isdir) +{ + AuDebugOn(d1 == d2 + || d1->d_inode == d2->d_inode + || d1->d_sb != d2->d_sb); + + if (isdir && au_test_subdir(d1, d2)) { + di_write_lock_parent(d1); + di_write_lock_parent2(d2); + } else { + /* there should be no races */ + di_write_lock_parent(d2); + di_write_lock_parent2(d1); + } +} + +void di_write_unlock2(struct dentry *d1, struct dentry *d2) +{ + di_write_unlock(d1); + if (d1->d_inode == d2->d_inode) + au_rw_write_unlock(&au_di(d2)->di_rwsem); + else + di_write_unlock(d2); +} + +/* ---------------------------------------------------------------------- */ + +struct dentry *au_h_dptr(struct dentry *dentry, aufs_bindex_t bindex) +{ + struct dentry *d; + + DiMustAnyLock(dentry); + + if (au_dbstart(dentry) < 0 || bindex < au_dbstart(dentry)) + return NULL; + AuDebugOn(bindex < 0); + d = au_di(dentry)->di_hdentry[0 + bindex].hd_dentry; + AuDebugOn(d && d->d_count <= 0); + return d; +} + +/* + * extended version of au_h_dptr(). + * returns a hashed and positive h_dentry in bindex, NULL, or error. + */ +struct dentry *au_h_d_alias(struct dentry *dentry, aufs_bindex_t bindex) +{ + struct dentry *h_dentry; + struct inode *inode, *h_inode; + + inode = dentry->d_inode; + AuDebugOn(!inode); + + h_dentry = NULL; + if (au_dbstart(dentry) <= bindex + && bindex <= au_dbend(dentry)) + h_dentry = au_h_dptr(dentry, bindex); + if (h_dentry && !au_d_hashed_positive(h_dentry)) { + dget(h_dentry); + goto out; /* success */ + } + + AuDebugOn(bindex < au_ibstart(inode)); + AuDebugOn(au_ibend(inode) < bindex); + h_inode = au_h_iptr(inode, bindex); + h_dentry = d_find_alias(h_inode); + if (h_dentry) { + if (!IS_ERR(h_dentry)) { + if (!au_d_hashed_positive(h_dentry)) + goto out; /* success */ + dput(h_dentry); + } else + goto out; + } + + if (au_opt_test(au_mntflags(dentry->d_sb), PLINK)) { + h_dentry = au_plink_lkup(inode, bindex); + AuDebugOn(!h_dentry); + if (!IS_ERR(h_dentry)) { + if (!au_d_hashed_positive(h_dentry)) + goto out; /* success */ + dput(h_dentry); + h_dentry = NULL; + } + } + +out: + AuDbgDentry(h_dentry); + return h_dentry; +} + +aufs_bindex_t au_dbtail(struct dentry *dentry) +{ + aufs_bindex_t bend, bwh; + + bend = au_dbend(dentry); + if (0 <= bend) { + bwh = au_dbwh(dentry); + if (!bwh) + return bwh; + if (0 < bwh && bwh < bend) + return bwh - 1; + } + return bend; +} + +aufs_bindex_t au_dbtaildir(struct dentry *dentry) +{ + aufs_bindex_t bend, bopq; + + bend = au_dbtail(dentry); + if (0 <= bend) { + bopq = au_dbdiropq(dentry); + if (0 <= bopq && bopq < bend) + bend = bopq; + } + return bend; +} + +/* ---------------------------------------------------------------------- */ + +void au_set_h_dptr(struct dentry *dentry, aufs_bindex_t bindex, + struct dentry *h_dentry) +{ + struct au_hdentry *hd = au_di(dentry)->di_hdentry + bindex; + struct au_branch *br; + + DiMustWriteLock(dentry); + + au_hdput(hd); + hd->hd_dentry = h_dentry; + if (h_dentry) { + br = au_sbr(dentry->d_sb, bindex); + hd->hd_id = br->br_id; + } +} + +int au_dbrange_test(struct dentry *dentry) +{ + int err; + aufs_bindex_t bstart, bend; + + err = 0; + bstart = au_dbstart(dentry); + bend = au_dbend(dentry); + if (bstart >= 0) + AuDebugOn(bend < 0 && bstart > bend); + else { + err = -EIO; + AuDebugOn(bend >= 0); + } + + return err; +} + +int au_digen_test(struct dentry *dentry, unsigned int sigen) +{ + int err; + + err = 0; + if (unlikely(au_digen(dentry) != sigen + || au_iigen_test(dentry->d_inode, sigen))) + err = -EIO; + + return err; +} + +void au_update_digen(struct dentry *dentry) +{ + atomic_set(&au_di(dentry)->di_generation, au_sigen(dentry->d_sb)); + /* smp_mb(); */ /* atomic_set */ +} + +void au_update_dbrange(struct dentry *dentry, int do_put_zero) +{ + struct au_dinfo *dinfo; + struct dentry *h_d; + struct au_hdentry *hdp; + + DiMustWriteLock(dentry); + + dinfo = au_di(dentry); + if (!dinfo || dinfo->di_bstart < 0) + return; + + hdp = dinfo->di_hdentry; + if (do_put_zero) { + aufs_bindex_t bindex, bend; + + bend = dinfo->di_bend; + for (bindex = dinfo->di_bstart; bindex <= bend; bindex++) { + h_d = hdp[0 + bindex].hd_dentry; + if (h_d && !h_d->d_inode) + au_set_h_dptr(dentry, bindex, NULL); + } + } + + dinfo->di_bstart = -1; + while (++dinfo->di_bstart <= dinfo->di_bend) + if (hdp[0 + dinfo->di_bstart].hd_dentry) + break; + if (dinfo->di_bstart > dinfo->di_bend) { + dinfo->di_bstart = -1; + dinfo->di_bend = -1; + return; + } + + dinfo->di_bend++; + while (0 <= --dinfo->di_bend) + if (hdp[0 + dinfo->di_bend].hd_dentry) + break; + AuDebugOn(dinfo->di_bstart > dinfo->di_bend || dinfo->di_bend < 0); +} + +void au_update_dbstart(struct dentry *dentry) +{ + aufs_bindex_t bindex, bend; + struct dentry *h_dentry; + + bend = au_dbend(dentry); + for (bindex = au_dbstart(dentry); bindex <= bend; bindex++) { + h_dentry = au_h_dptr(dentry, bindex); + if (!h_dentry) + continue; + if (h_dentry->d_inode) { + au_set_dbstart(dentry, bindex); + return; + } + au_set_h_dptr(dentry, bindex, NULL); + } +} + +void au_update_dbend(struct dentry *dentry) +{ + aufs_bindex_t bindex, bstart; + struct dentry *h_dentry; + + bstart = au_dbstart(dentry); + for (bindex = au_dbend(dentry); bindex >= bstart; bindex--) { + h_dentry = au_h_dptr(dentry, bindex); + if (!h_dentry) + continue; + if (h_dentry->d_inode) { + au_set_dbend(dentry, bindex); + return; + } + au_set_h_dptr(dentry, bindex, NULL); + } +} + +int au_find_dbindex(struct dentry *dentry, struct dentry *h_dentry) +{ + aufs_bindex_t bindex, bend; + + bend = au_dbend(dentry); + for (bindex = au_dbstart(dentry); bindex <= bend; bindex++) + if (au_h_dptr(dentry, bindex) == h_dentry) + return bindex; + return -1; +} diff -Naur kernel.21.3.orig/fs/aufs/dir.c kernel.21.3.new/fs/aufs/dir.c --- kernel.21.3.orig/fs/aufs/dir.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/dir.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,631 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * directory operations + */ + +#include +#include "aufs.h" + +void au_add_nlink(struct inode *dir, struct inode *h_dir) +{ + unsigned int nlink; + + AuDebugOn(!S_ISDIR(dir->i_mode) || !S_ISDIR(h_dir->i_mode)); + + nlink = dir->i_nlink; + nlink += h_dir->i_nlink - 2; + if (h_dir->i_nlink < 2) + nlink += 2; + smp_mb(); /* for i_nlink */ + /* 0 can happen in revaliding */ + set_nlink(dir, nlink); +} + +void au_sub_nlink(struct inode *dir, struct inode *h_dir) +{ + unsigned int nlink; + + AuDebugOn(!S_ISDIR(dir->i_mode) || !S_ISDIR(h_dir->i_mode)); + + nlink = dir->i_nlink; + nlink -= h_dir->i_nlink - 2; + if (h_dir->i_nlink < 2) + nlink -= 2; + smp_mb(); /* for i_nlink */ + /* nlink == 0 means the branch-fs is broken */ + set_nlink(dir, nlink); +} + +loff_t au_dir_size(struct file *file, struct dentry *dentry) +{ + loff_t sz; + aufs_bindex_t bindex, bend; + struct file *h_file; + struct dentry *h_dentry; + + sz = 0; + if (file) { + AuDebugOn(!file_inode(file)); + AuDebugOn(!S_ISDIR(file_inode(file)->i_mode)); + + bend = au_fbend_dir(file); + for (bindex = au_fbstart(file); + bindex <= bend && sz < KMALLOC_MAX_SIZE; + bindex++) { + h_file = au_hf_dir(file, bindex); + if (h_file && file_inode(h_file)) + sz += vfsub_f_size_read(h_file); + } + } else { + AuDebugOn(!dentry); + AuDebugOn(!dentry->d_inode); + AuDebugOn(!S_ISDIR(dentry->d_inode->i_mode)); + + bend = au_dbtaildir(dentry); + for (bindex = au_dbstart(dentry); + bindex <= bend && sz < KMALLOC_MAX_SIZE; + bindex++) { + h_dentry = au_h_dptr(dentry, bindex); + if (h_dentry && h_dentry->d_inode) + sz += i_size_read(h_dentry->d_inode); + } + } + if (sz < KMALLOC_MAX_SIZE) + sz = roundup_pow_of_two(sz); + if (sz > KMALLOC_MAX_SIZE) + sz = KMALLOC_MAX_SIZE; + else if (sz < NAME_MAX) { + BUILD_BUG_ON(AUFS_RDBLK_DEF < NAME_MAX); + sz = AUFS_RDBLK_DEF; + } + return sz; +} + +/* ---------------------------------------------------------------------- */ + +static int reopen_dir(struct file *file) +{ + int err; + unsigned int flags; + aufs_bindex_t bindex, btail, bstart; + struct dentry *dentry, *h_dentry; + struct file *h_file; + + /* open all lower dirs */ + dentry = file->f_dentry; + bstart = au_dbstart(dentry); + for (bindex = au_fbstart(file); bindex < bstart; bindex++) + au_set_h_fptr(file, bindex, NULL); + au_set_fbstart(file, bstart); + + btail = au_dbtaildir(dentry); + for (bindex = au_fbend_dir(file); btail < bindex; bindex--) + au_set_h_fptr(file, bindex, NULL); + au_set_fbend_dir(file, btail); + + flags = vfsub_file_flags(file); + for (bindex = bstart; bindex <= btail; bindex++) { + h_dentry = au_h_dptr(dentry, bindex); + if (!h_dentry) + continue; + h_file = au_hf_dir(file, bindex); + if (h_file) + continue; + + h_file = au_h_open(dentry, bindex, flags, file, /*force_wr*/0); + err = PTR_ERR(h_file); + if (IS_ERR(h_file)) + goto out; /* close all? */ + au_set_h_fptr(file, bindex, h_file); + } + au_update_figen(file); + /* todo: necessary? */ + /* file->f_ra = h_file->f_ra; */ + err = 0; + +out: + return err; +} + +static int do_open_dir(struct file *file, int flags) +{ + int err; + aufs_bindex_t bindex, btail; + struct dentry *dentry, *h_dentry; + struct file *h_file; + + FiMustWriteLock(file); + + err = 0; + dentry = file->f_dentry; + file->f_version = dentry->d_inode->i_version; + bindex = au_dbstart(dentry); + au_set_fbstart(file, bindex); + btail = au_dbtaildir(dentry); + au_set_fbend_dir(file, btail); + for (; !err && bindex <= btail; bindex++) { + h_dentry = au_h_dptr(dentry, bindex); + if (!h_dentry) + continue; + + h_file = au_h_open(dentry, bindex, flags, file, /*force_wr*/0); + if (IS_ERR(h_file)) { + err = PTR_ERR(h_file); + break; + } + au_set_h_fptr(file, bindex, h_file); + } + au_update_figen(file); + /* todo: necessary? */ + /* file->f_ra = h_file->f_ra; */ + if (!err) + return 0; /* success */ + + /* close all */ + for (bindex = au_fbstart(file); bindex <= btail; bindex++) + au_set_h_fptr(file, bindex, NULL); + au_set_fbstart(file, -1); + au_set_fbend_dir(file, -1); + + return err; +} + +static int aufs_open_dir(struct inode *inode __maybe_unused, + struct file *file) +{ + int err; + struct super_block *sb; + struct au_fidir *fidir; + + err = -ENOMEM; + sb = file->f_dentry->d_sb; + si_read_lock(sb, AuLock_FLUSH); + fidir = au_fidir_alloc(sb); + if (fidir) { + err = au_do_open(file, do_open_dir, fidir); + if (unlikely(err)) + kfree(fidir); + } + si_read_unlock(sb); + return err; +} + +static int aufs_release_dir(struct inode *inode __maybe_unused, + struct file *file) +{ + struct au_vdir *vdir_cache; + struct au_finfo *finfo; + struct au_fidir *fidir; + aufs_bindex_t bindex, bend; + + finfo = au_fi(file); + fidir = finfo->fi_hdir; + if (fidir) { + vdir_cache = fidir->fd_vdir_cache; /* lock-free */ + if (vdir_cache) + au_vdir_free(vdir_cache); + + bindex = finfo->fi_btop; + if (bindex >= 0) { + /* + * calls fput() instead of filp_close(), + * since no dnotify or lock for the lower file. + */ + bend = fidir->fd_bbot; + for (; bindex <= bend; bindex++) + au_set_h_fptr(file, bindex, NULL); + } + kfree(fidir); + finfo->fi_hdir = NULL; + } + au_finfo_fin(file); + return 0; +} + +/* ---------------------------------------------------------------------- */ + +static int au_do_flush_dir(struct file *file, fl_owner_t id) +{ + int err; + aufs_bindex_t bindex, bend; + struct file *h_file; + + err = 0; + bend = au_fbend_dir(file); + for (bindex = au_fbstart(file); !err && bindex <= bend; bindex++) { + h_file = au_hf_dir(file, bindex); + if (h_file) + err = vfsub_flush(h_file, id); + } + return err; +} + +static int aufs_flush_dir(struct file *file, fl_owner_t id) +{ + return au_do_flush(file, id, au_do_flush_dir); +} + +/* ---------------------------------------------------------------------- */ + +static int au_do_fsync_dir_no_file(struct dentry *dentry, int datasync) +{ + int err; + aufs_bindex_t bend, bindex; + struct inode *inode; + struct super_block *sb; + + err = 0; + sb = dentry->d_sb; + inode = dentry->d_inode; + IMustLock(inode); + bend = au_dbend(dentry); + for (bindex = au_dbstart(dentry); !err && bindex <= bend; bindex++) { + struct path h_path; + + if (au_test_ro(sb, bindex, inode)) + continue; + h_path.dentry = au_h_dptr(dentry, bindex); + if (!h_path.dentry) + continue; + + h_path.mnt = au_sbr_mnt(sb, bindex); + err = vfsub_fsync(NULL, &h_path, datasync); + } + + return err; +} + +static int au_do_fsync_dir(struct file *file, int datasync) +{ + int err; + aufs_bindex_t bend, bindex; + struct file *h_file; + struct super_block *sb; + struct inode *inode; + + err = au_reval_and_lock_fdi(file, reopen_dir, /*wlock*/1); + if (unlikely(err)) + goto out; + + sb = file->f_dentry->d_sb; + inode = file_inode(file); + bend = au_fbend_dir(file); + for (bindex = au_fbstart(file); !err && bindex <= bend; bindex++) { + h_file = au_hf_dir(file, bindex); + if (!h_file || au_test_ro(sb, bindex, inode)) + continue; + + err = vfsub_fsync(h_file, &h_file->f_path, datasync); + } + +out: + return err; +} + +/* + * @file may be NULL + */ +static int aufs_fsync_dir(struct file *file, loff_t start, loff_t end, + int datasync) +{ + int err; + struct dentry *dentry; + struct super_block *sb; + struct mutex *mtx; + + err = 0; + dentry = file->f_dentry; + mtx = &dentry->d_inode->i_mutex; + mutex_lock(mtx); + sb = dentry->d_sb; + si_noflush_read_lock(sb); + if (file) + err = au_do_fsync_dir(file, datasync); + else { + di_write_lock_child(dentry); + err = au_do_fsync_dir_no_file(dentry, datasync); + } + au_cpup_attr_timesizes(dentry->d_inode); + di_write_unlock(dentry); + if (file) + fi_write_unlock(file); + + si_read_unlock(sb); + mutex_unlock(mtx); + return err; +} + +/* ---------------------------------------------------------------------- */ + +static int aufs_readdir(struct file *file, void *dirent, filldir_t filldir) +{ + int err; + struct dentry *dentry; + struct inode *inode, *h_inode; + struct super_block *sb; + + dentry = file->f_dentry; + inode = dentry->d_inode; + IMustLock(inode); + + sb = dentry->d_sb; + si_read_lock(sb, AuLock_FLUSH); + err = au_reval_and_lock_fdi(file, reopen_dir, /*wlock*/1); + if (unlikely(err)) + goto out; + err = au_alive_dir(dentry); + if (!err) + err = au_vdir_init(file); + di_downgrade_lock(dentry, AuLock_IR); + if (unlikely(err)) + goto out_unlock; + + h_inode = au_h_iptr(inode, au_ibstart(inode)); + if (!au_test_nfsd()) { + err = au_vdir_fill_de(file, dirent, filldir); + fsstack_copy_attr_atime(inode, h_inode); + } else { + /* + * nfsd filldir may call lookup_one_len(), vfs_getattr(), + * encode_fh() and others. + */ + atomic_inc(&h_inode->i_count); + di_read_unlock(dentry, AuLock_IR); + si_read_unlock(sb); + err = au_vdir_fill_de(file, dirent, filldir); + fsstack_copy_attr_atime(inode, h_inode); + fi_write_unlock(file); + iput(h_inode); + + AuTraceErr(err); + return err; + } + +out_unlock: + di_read_unlock(dentry, AuLock_IR); + fi_write_unlock(file); +out: + si_read_unlock(sb); + return err; +} + +/* ---------------------------------------------------------------------- */ + +#define AuTestEmpty_WHONLY 1 +#define AuTestEmpty_CALLED (1 << 1) +#define AuTestEmpty_SHWH (1 << 2) +#define au_ftest_testempty(flags, name) ((flags) & AuTestEmpty_##name) +#define au_fset_testempty(flags, name) \ + do { (flags) |= AuTestEmpty_##name; } while (0) +#define au_fclr_testempty(flags, name) \ + do { (flags) &= ~AuTestEmpty_##name; } while (0) + +#ifndef CONFIG_AUFS_SHWH +#undef AuTestEmpty_SHWH +#define AuTestEmpty_SHWH 0 +#endif + +struct test_empty_arg { + struct au_nhash *whlist; + unsigned int flags; + int err; + aufs_bindex_t bindex; +}; + +static int test_empty_cb(void *__arg, const char *__name, int namelen, + loff_t offset __maybe_unused, u64 ino, + unsigned int d_type) +{ + struct test_empty_arg *arg = __arg; + char *name = (void *)__name; + + arg->err = 0; + au_fset_testempty(arg->flags, CALLED); + /* smp_mb(); */ + if (name[0] == '.' + && (namelen == 1 || (name[1] == '.' && namelen == 2))) + goto out; /* success */ + + if (namelen <= AUFS_WH_PFX_LEN + || memcmp(name, AUFS_WH_PFX, AUFS_WH_PFX_LEN)) { + if (au_ftest_testempty(arg->flags, WHONLY) + && !au_nhash_test_known_wh(arg->whlist, name, namelen)) + arg->err = -ENOTEMPTY; + goto out; + } + + name += AUFS_WH_PFX_LEN; + namelen -= AUFS_WH_PFX_LEN; + if (!au_nhash_test_known_wh(arg->whlist, name, namelen)) + arg->err = au_nhash_append_wh + (arg->whlist, name, namelen, ino, d_type, arg->bindex, + au_ftest_testempty(arg->flags, SHWH)); + +out: + /* smp_mb(); */ + AuTraceErr(arg->err); + return arg->err; +} + +static int do_test_empty(struct dentry *dentry, struct test_empty_arg *arg) +{ + int err; + struct file *h_file; + + h_file = au_h_open(dentry, arg->bindex, + O_RDONLY | O_NONBLOCK | O_DIRECTORY | O_LARGEFILE, + /*file*/NULL, /*force_wr*/0); + err = PTR_ERR(h_file); + if (IS_ERR(h_file)) + goto out; + + err = 0; + if (!au_opt_test(au_mntflags(dentry->d_sb), UDBA_NONE) + && !file_inode(h_file)->i_nlink) + goto out_put; + + do { + arg->err = 0; + au_fclr_testempty(arg->flags, CALLED); + /* smp_mb(); */ + err = vfsub_readdir(h_file, test_empty_cb, arg); + if (err >= 0) + err = arg->err; + } while (!err && au_ftest_testempty(arg->flags, CALLED)); + +out_put: + fput(h_file); + au_sbr_put(dentry->d_sb, arg->bindex); +out: + return err; +} + +struct do_test_empty_args { + int *errp; + struct dentry *dentry; + struct test_empty_arg *arg; +}; + +static void call_do_test_empty(void *args) +{ + struct do_test_empty_args *a = args; + *a->errp = do_test_empty(a->dentry, a->arg); +} + +static int sio_test_empty(struct dentry *dentry, struct test_empty_arg *arg) +{ + int err, wkq_err; + struct dentry *h_dentry; + struct inode *h_inode; + + h_dentry = au_h_dptr(dentry, arg->bindex); + h_inode = h_dentry->d_inode; + /* todo: i_mode changes anytime? */ + mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD); + err = au_test_h_perm_sio(h_inode, MAY_EXEC | MAY_READ); + mutex_unlock(&h_inode->i_mutex); + if (!err) + err = do_test_empty(dentry, arg); + else { + struct do_test_empty_args args = { + .errp = &err, + .dentry = dentry, + .arg = arg + }; + unsigned int flags = arg->flags; + + wkq_err = au_wkq_wait(call_do_test_empty, &args); + if (unlikely(wkq_err)) + err = wkq_err; + arg->flags = flags; + } + + return err; +} + +int au_test_empty_lower(struct dentry *dentry) +{ + int err; + unsigned int rdhash; + aufs_bindex_t bindex, bstart, btail; + struct au_nhash whlist; + struct test_empty_arg arg; + int (*test_empty)(struct dentry *dentry, struct test_empty_arg *arg); + + SiMustAnyLock(dentry->d_sb); + + rdhash = au_sbi(dentry->d_sb)->si_rdhash; + if (!rdhash) + rdhash = au_rdhash_est(au_dir_size(/*file*/NULL, dentry)); + err = au_nhash_alloc(&whlist, rdhash, GFP_NOFS); + if (unlikely(err)) + goto out; + + arg.flags = 0; + arg.whlist = &whlist; + bstart = au_dbstart(dentry); + if (au_opt_test(au_mntflags(dentry->d_sb), SHWH)) + au_fset_testempty(arg.flags, SHWH); + test_empty = do_test_empty; + if (au_opt_test(au_mntflags(dentry->d_sb), DIRPERM1)) + test_empty = sio_test_empty; + arg.bindex = bstart; + err = test_empty(dentry, &arg); + if (unlikely(err)) + goto out_whlist; + + au_fset_testempty(arg.flags, WHONLY); + btail = au_dbtaildir(dentry); + for (bindex = bstart + 1; !err && bindex <= btail; bindex++) { + struct dentry *h_dentry; + + h_dentry = au_h_dptr(dentry, bindex); + if (h_dentry && h_dentry->d_inode) { + arg.bindex = bindex; + err = test_empty(dentry, &arg); + } + } + +out_whlist: + au_nhash_wh_free(&whlist); +out: + return err; +} + +int au_test_empty(struct dentry *dentry, struct au_nhash *whlist) +{ + int err; + struct test_empty_arg arg; + aufs_bindex_t bindex, btail; + + err = 0; + arg.whlist = whlist; + arg.flags = AuTestEmpty_WHONLY; + if (au_opt_test(au_mntflags(dentry->d_sb), SHWH)) + au_fset_testempty(arg.flags, SHWH); + btail = au_dbtaildir(dentry); + for (bindex = au_dbstart(dentry); !err && bindex <= btail; bindex++) { + struct dentry *h_dentry; + + h_dentry = au_h_dptr(dentry, bindex); + if (h_dentry && h_dentry->d_inode) { + arg.bindex = bindex; + err = sio_test_empty(dentry, &arg); + } + } + + return err; +} + +/* ---------------------------------------------------------------------- */ + +const struct file_operations aufs_dir_fop = { + .owner = THIS_MODULE, + .llseek = default_llseek, + .read = generic_read_dir, + .readdir = aufs_readdir, + .unlocked_ioctl = aufs_ioctl_dir, +#ifdef CONFIG_COMPAT + .compat_ioctl = aufs_compat_ioctl_dir, +#endif + .open = aufs_open_dir, + .release = aufs_release_dir, + .flush = aufs_flush_dir, + .fsync = aufs_fsync_dir +}; diff -Naur kernel.21.3.orig/fs/aufs/dir.h kernel.21.3.new/fs/aufs/dir.h --- kernel.21.3.orig/fs/aufs/dir.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/dir.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,130 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * directory operations + */ + +#ifndef __AUFS_DIR_H__ +#define __AUFS_DIR_H__ + +#ifdef __KERNEL__ + +#include + +/* ---------------------------------------------------------------------- */ + +/* need to be faster and smaller */ + +struct au_nhash { + unsigned int nh_num; + struct hlist_head *nh_head; +}; + +struct au_vdir_destr { + unsigned char len; + unsigned char name[0]; +} __packed; + +struct au_vdir_dehstr { + struct hlist_node hash; + struct au_vdir_destr *str; +} ____cacheline_aligned_in_smp; + +struct au_vdir_de { + ino_t de_ino; + unsigned char de_type; + /* caution: packed */ + struct au_vdir_destr de_str; +} __packed; + +struct au_vdir_wh { + struct hlist_node wh_hash; +#ifdef CONFIG_AUFS_SHWH + ino_t wh_ino; + aufs_bindex_t wh_bindex; + unsigned char wh_type; +#else + aufs_bindex_t wh_bindex; +#endif + /* caution: packed */ + struct au_vdir_destr wh_str; +} __packed; + +union au_vdir_deblk_p { + unsigned char *deblk; + struct au_vdir_de *de; +}; + +struct au_vdir { + unsigned char **vd_deblk; + unsigned long vd_nblk; + struct { + unsigned long ul; + union au_vdir_deblk_p p; + } vd_last; + + unsigned long vd_version; + unsigned int vd_deblk_sz; + unsigned long vd_jiffy; +} ____cacheline_aligned_in_smp; + +/* ---------------------------------------------------------------------- */ + +/* dir.c */ +extern const struct file_operations aufs_dir_fop; +void au_add_nlink(struct inode *dir, struct inode *h_dir); +void au_sub_nlink(struct inode *dir, struct inode *h_dir); +loff_t au_dir_size(struct file *file, struct dentry *dentry); +int au_test_empty_lower(struct dentry *dentry); +int au_test_empty(struct dentry *dentry, struct au_nhash *whlist); + +/* vdir.c */ +unsigned int au_rdhash_est(loff_t sz); +int au_nhash_alloc(struct au_nhash *nhash, unsigned int num_hash, gfp_t gfp); +void au_nhash_wh_free(struct au_nhash *whlist); +int au_nhash_test_longer_wh(struct au_nhash *whlist, aufs_bindex_t btgt, + int limit); +int au_nhash_test_known_wh(struct au_nhash *whlist, char *name, int nlen); +int au_nhash_append_wh(struct au_nhash *whlist, char *name, int nlen, ino_t ino, + unsigned int d_type, aufs_bindex_t bindex, + unsigned char shwh); +void au_vdir_free(struct au_vdir *vdir); +int au_vdir_init(struct file *file); +int au_vdir_fill_de(struct file *file, void *dirent, filldir_t filldir); + +/* ioctl.c */ +long aufs_ioctl_dir(struct file *file, unsigned int cmd, unsigned long arg); + +#ifdef CONFIG_AUFS_RDU +/* rdu.c */ +long au_rdu_ioctl(struct file *file, unsigned int cmd, unsigned long arg); +#ifdef CONFIG_COMPAT +long au_rdu_compat_ioctl(struct file *file, unsigned int cmd, + unsigned long arg); +#endif +#else +AuStub(long, au_rdu_ioctl, return -EINVAL, struct file *file, + unsigned int cmd, unsigned long arg) +#ifdef CONFIG_COMPAT +AuStub(long, au_rdu_compat_ioctl, return -EINVAL, struct file *file, + unsigned int cmd, unsigned long arg) +#endif +#endif + +#endif /* __KERNEL__ */ +#endif /* __AUFS_DIR_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/dynop.c kernel.21.3.new/fs/aufs/dynop.c --- kernel.21.3.orig/fs/aufs/dynop.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/dynop.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,378 @@ +/* + * Copyright (C) 2010-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * dynamically customizable operations for regular files + */ + +#include "aufs.h" + +#define DyPrSym(key) AuDbgSym(key->dk_op.dy_hop) + +/* + * How large will these lists be? + * Usually just a few elements, 20-30 at most for each, I guess. + */ +static struct au_splhead dynop[AuDyLast]; + +static struct au_dykey *dy_gfind_get(struct au_splhead *spl, const void *h_op) +{ + struct au_dykey *key, *tmp; + struct list_head *head; + + key = NULL; + head = &spl->head; + rcu_read_lock(); + list_for_each_entry_rcu(tmp, head, dk_list) + if (tmp->dk_op.dy_hop == h_op) { + key = tmp; + kref_get(&key->dk_kref); + break; + } + rcu_read_unlock(); + + return key; +} + +static struct au_dykey *dy_bradd(struct au_branch *br, struct au_dykey *key) +{ + struct au_dykey **k, *found; + const void *h_op = key->dk_op.dy_hop; + int i; + + found = NULL; + k = br->br_dykey; + for (i = 0; i < AuBrDynOp; i++) + if (k[i]) { + if (k[i]->dk_op.dy_hop == h_op) { + found = k[i]; + break; + } + } else + break; + if (!found) { + spin_lock(&br->br_dykey_lock); + for (; i < AuBrDynOp; i++) + if (k[i]) { + if (k[i]->dk_op.dy_hop == h_op) { + found = k[i]; + break; + } + } else { + k[i] = key; + break; + } + spin_unlock(&br->br_dykey_lock); + BUG_ON(i == AuBrDynOp); /* expand the array */ + } + + return found; +} + +/* kref_get() if @key is already added */ +static struct au_dykey *dy_gadd(struct au_splhead *spl, struct au_dykey *key) +{ + struct au_dykey *tmp, *found; + struct list_head *head; + const void *h_op = key->dk_op.dy_hop; + + found = NULL; + head = &spl->head; + spin_lock(&spl->spin); + list_for_each_entry(tmp, head, dk_list) + if (tmp->dk_op.dy_hop == h_op) { + kref_get(&tmp->dk_kref); + found = tmp; + break; + } + if (!found) + list_add_rcu(&key->dk_list, head); + spin_unlock(&spl->spin); + + if (!found) + DyPrSym(key); + return found; +} + +static void dy_free_rcu(struct rcu_head *rcu) +{ + struct au_dykey *key; + + key = container_of(rcu, struct au_dykey, dk_rcu); + DyPrSym(key); + kfree(key); +} + +static void dy_free(struct kref *kref) +{ + struct au_dykey *key; + struct au_splhead *spl; + + key = container_of(kref, struct au_dykey, dk_kref); + spl = dynop + key->dk_op.dy_type; + au_spl_del_rcu(&key->dk_list, spl); + call_rcu(&key->dk_rcu, dy_free_rcu); +} + +void au_dy_put(struct au_dykey *key) +{ + kref_put(&key->dk_kref, dy_free); +} + +/* ---------------------------------------------------------------------- */ + +#define DyDbgSize(cnt, op) AuDebugOn(cnt != sizeof(op)/sizeof(void *)) + +#ifdef CONFIG_AUFS_DEBUG +#define DyDbgDeclare(cnt) unsigned int cnt = 0 +#define DyDbgInc(cnt) do { cnt++; } while (0) +#else +#define DyDbgDeclare(cnt) do {} while (0) +#define DyDbgInc(cnt) do {} while (0) +#endif + +#define DySet(func, dst, src, h_op, h_sb) do { \ + DyDbgInc(cnt); \ + if (h_op->func) { \ + if (src.func) \ + dst.func = src.func; \ + else \ + AuDbg("%s %s\n", au_sbtype(h_sb), #func); \ + } \ +} while (0) + +#define DySetForce(func, dst, src) do { \ + AuDebugOn(!src.func); \ + DyDbgInc(cnt); \ + dst.func = src.func; \ +} while (0) + +#define DySetAop(func) \ + DySet(func, dyaop->da_op, aufs_aop, h_aop, h_sb) +#define DySetAopForce(func) \ + DySetForce(func, dyaop->da_op, aufs_aop) + +static void dy_aop(struct au_dykey *key, const void *h_op, + struct super_block *h_sb __maybe_unused) +{ + struct au_dyaop *dyaop = (void *)key; + const struct address_space_operations *h_aop = h_op; + DyDbgDeclare(cnt); + + AuDbg("%s\n", au_sbtype(h_sb)); + + DySetAop(writepage); + DySetAopForce(readpage); /* force */ + DySetAop(writepages); + DySetAop(set_page_dirty); + DySetAop(readpages); + DySetAop(write_begin); + DySetAop(write_end); + DySetAop(bmap); + DySetAop(invalidatepage); + DySetAop(releasepage); + DySetAop(freepage); + /* these two will be changed according to an aufs mount option */ + DySetAop(direct_IO); + DySetAop(get_xip_mem); + DySetAop(migratepage); + DySetAop(launder_page); + DySetAop(is_partially_uptodate); + DySetAop(error_remove_page); + DySetAop(swap_activate); + DySetAop(swap_deactivate); + + DyDbgSize(cnt, *h_aop); + dyaop->da_get_xip_mem = h_aop->get_xip_mem; +} + +/* ---------------------------------------------------------------------- */ + +static void dy_bug(struct kref *kref) +{ + BUG(); +} + +static struct au_dykey *dy_get(struct au_dynop *op, struct au_branch *br) +{ + struct au_dykey *key, *old; + struct au_splhead *spl; + struct op { + unsigned int sz; + void (*set)(struct au_dykey *key, const void *h_op, + struct super_block *h_sb __maybe_unused); + }; + static const struct op a[] = { + [AuDy_AOP] = { + .sz = sizeof(struct au_dyaop), + .set = dy_aop + } + }; + const struct op *p; + + spl = dynop + op->dy_type; + key = dy_gfind_get(spl, op->dy_hop); + if (key) + goto out_add; /* success */ + + p = a + op->dy_type; + key = kzalloc(p->sz, GFP_NOFS); + if (unlikely(!key)) { + key = ERR_PTR(-ENOMEM); + goto out; + } + + key->dk_op.dy_hop = op->dy_hop; + kref_init(&key->dk_kref); + p->set(key, op->dy_hop, au_br_sb(br)); + old = dy_gadd(spl, key); + if (old) { + kfree(key); + key = old; + } + +out_add: + old = dy_bradd(br, key); + if (old) + /* its ref-count should never be zero here */ + kref_put(&key->dk_kref, dy_bug); +out: + return key; +} + +/* ---------------------------------------------------------------------- */ +/* + * Aufs prohibits O_DIRECT by defaut even if the branch supports it. + * This behaviour is necessary to return an error from open(O_DIRECT) instead + * of the succeeding I/O. The dio mount option enables O_DIRECT and makes + * open(O_DIRECT) always succeed, but the succeeding I/O may return an error. + * See the aufs manual in detail. + * + * To keep this behaviour, aufs has to set NULL to ->get_xip_mem too, and the + * performance of fadvise() and madvise() may be affected. + */ +static void dy_adx(struct au_dyaop *dyaop, int do_dx) +{ + if (!do_dx) { + dyaop->da_op.direct_IO = NULL; + dyaop->da_op.get_xip_mem = NULL; + } else { + dyaop->da_op.direct_IO = aufs_aop.direct_IO; + dyaop->da_op.get_xip_mem = aufs_aop.get_xip_mem; + if (!dyaop->da_get_xip_mem) + dyaop->da_op.get_xip_mem = NULL; + } +} + +static struct au_dyaop *dy_aget(struct au_branch *br, + const struct address_space_operations *h_aop, + int do_dx) +{ + struct au_dyaop *dyaop; + struct au_dynop op; + + op.dy_type = AuDy_AOP; + op.dy_haop = h_aop; + dyaop = (void *)dy_get(&op, br); + if (IS_ERR(dyaop)) + goto out; + dy_adx(dyaop, do_dx); + +out: + return dyaop; +} + +int au_dy_iaop(struct inode *inode, aufs_bindex_t bindex, + struct inode *h_inode) +{ + int err, do_dx; + struct super_block *sb; + struct au_branch *br; + struct au_dyaop *dyaop; + + AuDebugOn(!S_ISREG(h_inode->i_mode)); + IiMustWriteLock(inode); + + sb = inode->i_sb; + br = au_sbr(sb, bindex); + do_dx = !!au_opt_test(au_mntflags(sb), DIO); + dyaop = dy_aget(br, h_inode->i_mapping->a_ops, do_dx); + err = PTR_ERR(dyaop); + if (IS_ERR(dyaop)) + /* unnecessary to call dy_fput() */ + goto out; + + err = 0; + inode->i_mapping->a_ops = &dyaop->da_op; + +out: + return err; +} + +/* + * Is it safe to replace a_ops during the inode/file is in operation? + * Yes, I hope so. + */ +int au_dy_irefresh(struct inode *inode) +{ + int err; + aufs_bindex_t bstart; + struct inode *h_inode; + + err = 0; + if (S_ISREG(inode->i_mode)) { + bstart = au_ibstart(inode); + h_inode = au_h_iptr(inode, bstart); + err = au_dy_iaop(inode, bstart, h_inode); + } + return err; +} + +void au_dy_arefresh(int do_dx) +{ + struct au_splhead *spl; + struct list_head *head; + struct au_dykey *key; + + spl = dynop + AuDy_AOP; + head = &spl->head; + spin_lock(&spl->spin); + list_for_each_entry(key, head, dk_list) + dy_adx((void *)key, do_dx); + spin_unlock(&spl->spin); +} + +/* ---------------------------------------------------------------------- */ + +void __init au_dy_init(void) +{ + int i; + + /* make sure that 'struct au_dykey *' can be any type */ + BUILD_BUG_ON(offsetof(struct au_dyaop, da_key)); + + for (i = 0; i < AuDyLast; i++) + au_spl_init(dynop + i); +} + +void au_dy_fin(void) +{ + int i; + + for (i = 0; i < AuDyLast; i++) + WARN_ON(!list_empty(&dynop[i].head)); +} diff -Naur kernel.21.3.orig/fs/aufs/dynop.h kernel.21.3.new/fs/aufs/dynop.h --- kernel.21.3.orig/fs/aufs/dynop.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/dynop.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,75 @@ +/* + * Copyright (C) 2010-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * dynamically customizable operations (for regular files only) + */ + +#ifndef __AUFS_DYNOP_H__ +#define __AUFS_DYNOP_H__ + +#ifdef __KERNEL__ + +#include "inode.h" + +enum {AuDy_AOP, AuDyLast}; + +struct au_dynop { + int dy_type; + union { + const void *dy_hop; + const struct address_space_operations *dy_haop; + }; +}; + +struct au_dykey { + union { + struct list_head dk_list; + struct rcu_head dk_rcu; + }; + struct au_dynop dk_op; + + /* + * during I am in the branch local array, kref is gotten. when the + * branch is removed, kref is put. + */ + struct kref dk_kref; +}; + +/* stop unioning since their sizes are very different from each other */ +struct au_dyaop { + struct au_dykey da_key; + struct address_space_operations da_op; /* not const */ + int (*da_get_xip_mem)(struct address_space *, pgoff_t, int, + void **, unsigned long *); +}; + +/* ---------------------------------------------------------------------- */ + +/* dynop.c */ +struct au_branch; +void au_dy_put(struct au_dykey *key); +int au_dy_iaop(struct inode *inode, aufs_bindex_t bindex, + struct inode *h_inode); +int au_dy_irefresh(struct inode *inode); +void au_dy_arefresh(int do_dio); + +void __init au_dy_init(void); +void au_dy_fin(void); + +#endif /* __KERNEL__ */ +#endif /* __AUFS_DYNOP_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/export.c kernel.21.3.new/fs/aufs/export.c --- kernel.21.3.orig/fs/aufs/export.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/export.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,825 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * export via nfs + */ + +#include +#include +#include +#include +#include +#include +#include "../fs/mount.h" +#include "aufs.h" + +union conv { +#ifdef CONFIG_AUFS_INO_T_64 + __u32 a[2]; +#else + __u32 a[1]; +#endif + ino_t ino; +}; + +static ino_t decode_ino(__u32 *a) +{ + union conv u; + + BUILD_BUG_ON(sizeof(u.ino) != sizeof(u.a)); + u.a[0] = a[0]; +#ifdef CONFIG_AUFS_INO_T_64 + u.a[1] = a[1]; +#endif + return u.ino; +} + +static void encode_ino(__u32 *a, ino_t ino) +{ + union conv u; + + u.ino = ino; + a[0] = u.a[0]; +#ifdef CONFIG_AUFS_INO_T_64 + a[1] = u.a[1]; +#endif +} + +/* NFS file handle */ +enum { + Fh_br_id, + Fh_sigen, +#ifdef CONFIG_AUFS_INO_T_64 + /* support 64bit inode number */ + Fh_ino1, + Fh_ino2, + Fh_dir_ino1, + Fh_dir_ino2, +#else + Fh_ino1, + Fh_dir_ino1, +#endif + Fh_igen, + Fh_h_type, + Fh_tail, + + Fh_ino = Fh_ino1, + Fh_dir_ino = Fh_dir_ino1 +}; + +static int au_test_anon(struct dentry *dentry) +{ + /* note: read d_flags without d_lock */ + return !!(dentry->d_flags & DCACHE_DISCONNECTED); +} + +int au_test_nfsd(void) +{ + int ret; + struct task_struct *tsk = current; + char comm[sizeof(tsk->comm)]; + + ret = 0; + if (tsk->flags & PF_KTHREAD) { + get_task_comm(comm, tsk); + ret = !strcmp(comm, "nfsd"); + } + + return ret; +} + +/* ---------------------------------------------------------------------- */ +/* inode generation external table */ + +void au_xigen_inc(struct inode *inode) +{ + loff_t pos; + ssize_t sz; + __u32 igen; + struct super_block *sb; + struct au_sbinfo *sbinfo; + + sb = inode->i_sb; + AuDebugOn(!au_opt_test(au_mntflags(sb), XINO)); + + sbinfo = au_sbi(sb); + pos = inode->i_ino; + pos *= sizeof(igen); + igen = inode->i_generation + 1; + sz = xino_fwrite(sbinfo->si_xwrite, sbinfo->si_xigen, &igen, + sizeof(igen), &pos); + if (sz == sizeof(igen)) + return; /* success */ + + if (unlikely(sz >= 0)) + AuIOErr("xigen error (%zd)\n", sz); +} + +int au_xigen_new(struct inode *inode) +{ + int err; + loff_t pos; + ssize_t sz; + struct super_block *sb; + struct au_sbinfo *sbinfo; + struct file *file; + + err = 0; + /* todo: dirty, at mount time */ + if (inode->i_ino == AUFS_ROOT_INO) + goto out; + sb = inode->i_sb; + SiMustAnyLock(sb); + if (unlikely(!au_opt_test(au_mntflags(sb), XINO))) + goto out; + + err = -EFBIG; + pos = inode->i_ino; + if (unlikely(au_loff_max / sizeof(inode->i_generation) - 1 < pos)) { + AuIOErr1("too large i%lld\n", pos); + goto out; + } + pos *= sizeof(inode->i_generation); + + err = 0; + sbinfo = au_sbi(sb); + file = sbinfo->si_xigen; + BUG_ON(!file); + + if (vfsub_f_size_read(file) + < pos + sizeof(inode->i_generation)) { + inode->i_generation = atomic_inc_return(&sbinfo->si_xigen_next); + sz = xino_fwrite(sbinfo->si_xwrite, file, &inode->i_generation, + sizeof(inode->i_generation), &pos); + } else + sz = xino_fread(sbinfo->si_xread, file, &inode->i_generation, + sizeof(inode->i_generation), &pos); + if (sz == sizeof(inode->i_generation)) + goto out; /* success */ + + err = sz; + if (unlikely(sz >= 0)) { + err = -EIO; + AuIOErr("xigen error (%zd)\n", sz); + } + +out: + return err; +} + +int au_xigen_set(struct super_block *sb, struct file *base) +{ + int err; + struct au_sbinfo *sbinfo; + struct file *file; + + SiMustWriteLock(sb); + + sbinfo = au_sbi(sb); + file = au_xino_create2(base, sbinfo->si_xigen); + err = PTR_ERR(file); + if (IS_ERR(file)) + goto out; + err = 0; + if (sbinfo->si_xigen) + fput(sbinfo->si_xigen); + sbinfo->si_xigen = file; + +out: + return err; +} + +void au_xigen_clr(struct super_block *sb) +{ + struct au_sbinfo *sbinfo; + + SiMustWriteLock(sb); + + sbinfo = au_sbi(sb); + if (sbinfo->si_xigen) { + fput(sbinfo->si_xigen); + sbinfo->si_xigen = NULL; + } +} + +/* ---------------------------------------------------------------------- */ + +static struct dentry *decode_by_ino(struct super_block *sb, ino_t ino, + ino_t dir_ino) +{ + struct dentry *dentry, *d; + struct inode *inode; + unsigned int sigen; + + dentry = NULL; + inode = ilookup(sb, ino); + if (!inode) + goto out; + + dentry = ERR_PTR(-ESTALE); + sigen = au_sigen(sb); + if (unlikely(is_bad_inode(inode) + || IS_DEADDIR(inode) + || sigen != au_iigen(inode, NULL))) + goto out_iput; + + dentry = NULL; + if (!dir_ino || S_ISDIR(inode->i_mode)) + dentry = d_find_alias(inode); + else { + spin_lock(&inode->i_lock); + hlist_for_each_entry(d, &inode->i_dentry, d_alias) { + spin_lock(&d->d_lock); + if (!au_test_anon(d) + && d->d_parent->d_inode->i_ino == dir_ino) { + dentry = dget_dlock(d); + spin_unlock(&d->d_lock); + break; + } + spin_unlock(&d->d_lock); + } + spin_unlock(&inode->i_lock); + } + if (unlikely(dentry && au_digen_test(dentry, sigen))) { + /* need to refresh */ + dput(dentry); + dentry = NULL; + } + +out_iput: + iput(inode); +out: + AuTraceErrPtr(dentry); + return dentry; +} + +/* ---------------------------------------------------------------------- */ + +/* todo: dirty? */ +/* if exportfs_decode_fh() passed vfsmount*, we could be happy */ + +struct au_compare_mnt_args { + /* input */ + struct super_block *sb; + + /* output */ + struct vfsmount *mnt; +}; + +static int au_compare_mnt(struct vfsmount *mnt, void *arg) +{ + struct au_compare_mnt_args *a = arg; + + if (mnt->mnt_sb != a->sb) + return 0; + a->mnt = mntget(mnt); + return 1; +} + +static struct vfsmount *au_mnt_get(struct super_block *sb) +{ + int err; + struct path root; + struct au_compare_mnt_args args = { + .sb = sb + }; + + get_fs_root(current->fs, &root); + br_read_lock(&vfsmount_lock); + err = iterate_mounts(au_compare_mnt, &args, root.mnt); + br_read_unlock(&vfsmount_lock); + path_put(&root); + AuDebugOn(!err); + AuDebugOn(!args.mnt); + return args.mnt; +} + +struct au_nfsd_si_lock { + unsigned int sigen; + aufs_bindex_t bindex, br_id; + unsigned char force_lock; +}; + +static int si_nfsd_read_lock(struct super_block *sb, + struct au_nfsd_si_lock *nsi_lock) +{ + int err; + aufs_bindex_t bindex; + + si_read_lock(sb, AuLock_FLUSH); + + /* branch id may be wrapped around */ + err = 0; + bindex = au_br_index(sb, nsi_lock->br_id); + if (bindex >= 0 && nsi_lock->sigen + AUFS_BRANCH_MAX > au_sigen(sb)) + goto out; /* success */ + + err = -ESTALE; + bindex = -1; + if (!nsi_lock->force_lock) + si_read_unlock(sb); + +out: + nsi_lock->bindex = bindex; + return err; +} + +struct find_name_by_ino { + int called, found; + ino_t ino; + char *name; + int namelen; +}; + +static int +find_name_by_ino(void *arg, const char *name, int namelen, loff_t offset, + u64 ino, unsigned int d_type) +{ + struct find_name_by_ino *a = arg; + + a->called++; + if (a->ino != ino) + return 0; + + memcpy(a->name, name, namelen); + a->namelen = namelen; + a->found = 1; + return 1; +} + +static struct dentry *au_lkup_by_ino(struct path *path, ino_t ino, + struct au_nfsd_si_lock *nsi_lock) +{ + struct dentry *dentry, *parent; + struct file *file; + struct inode *dir; + struct find_name_by_ino arg; + int err; + + parent = path->dentry; + if (nsi_lock) + si_read_unlock(parent->d_sb); + file = vfsub_dentry_open(path, au_dir_roflags); + dentry = (void *)file; + if (IS_ERR(file)) + goto out; + + dentry = ERR_PTR(-ENOMEM); + arg.name = (void *)__get_free_page(GFP_NOFS); + if (unlikely(!arg.name)) + goto out_file; + arg.ino = ino; + arg.found = 0; + do { + arg.called = 0; + /* smp_mb(); */ + err = vfsub_readdir(file, find_name_by_ino, &arg); + } while (!err && !arg.found && arg.called); + dentry = ERR_PTR(err); + if (unlikely(err)) + goto out_name; + /* instead of ENOENT */ + dentry = ERR_PTR(-ESTALE); + if (!arg.found) + goto out_name; + + /* do not call vfsub_lkup_one() */ + dir = parent->d_inode; + mutex_lock(&dir->i_mutex); + dentry = vfsub_lookup_one_len(arg.name, parent, arg.namelen); + mutex_unlock(&dir->i_mutex); + AuTraceErrPtr(dentry); + if (IS_ERR(dentry)) + goto out_name; + AuDebugOn(au_test_anon(dentry)); + if (unlikely(!dentry->d_inode)) { + dput(dentry); + dentry = ERR_PTR(-ENOENT); + } + +out_name: + free_page((unsigned long)arg.name); +out_file: + fput(file); +out: + if (unlikely(nsi_lock + && si_nfsd_read_lock(parent->d_sb, nsi_lock) < 0)) + if (!IS_ERR(dentry)) { + dput(dentry); + dentry = ERR_PTR(-ESTALE); + } + AuTraceErrPtr(dentry); + return dentry; +} + +static struct dentry *decode_by_dir_ino(struct super_block *sb, ino_t ino, + ino_t dir_ino, + struct au_nfsd_si_lock *nsi_lock) +{ + struct dentry *dentry; + struct path path; + + if (dir_ino != AUFS_ROOT_INO) { + path.dentry = decode_by_ino(sb, dir_ino, 0); + dentry = path.dentry; + if (!path.dentry || IS_ERR(path.dentry)) + goto out; + AuDebugOn(au_test_anon(path.dentry)); + } else + path.dentry = dget(sb->s_root); + + path.mnt = au_mnt_get(sb); + dentry = au_lkup_by_ino(&path, ino, nsi_lock); + path_put(&path); + +out: + AuTraceErrPtr(dentry); + return dentry; +} + +/* ---------------------------------------------------------------------- */ + +static int h_acceptable(void *expv, struct dentry *dentry) +{ + return 1; +} + +static char *au_build_path(struct dentry *h_parent, struct path *h_rootpath, + char *buf, int len, struct super_block *sb) +{ + char *p; + int n; + struct path path; + + p = d_path(h_rootpath, buf, len); + if (IS_ERR(p)) + goto out; + n = strlen(p); + + path.mnt = h_rootpath->mnt; + path.dentry = h_parent; + p = d_path(&path, buf, len); + if (IS_ERR(p)) + goto out; + if (n != 1) + p += n; + + path.mnt = au_mnt_get(sb); + path.dentry = sb->s_root; + p = d_path(&path, buf, len - strlen(p)); + mntput(path.mnt); + if (IS_ERR(p)) + goto out; + if (n != 1) + p[strlen(p)] = '/'; + +out: + AuTraceErrPtr(p); + return p; +} + +static +struct dentry *decode_by_path(struct super_block *sb, ino_t ino, __u32 *fh, + int fh_len, struct au_nfsd_si_lock *nsi_lock) +{ + struct dentry *dentry, *h_parent, *root; + struct super_block *h_sb; + char *pathname, *p; + struct vfsmount *h_mnt; + struct au_branch *br; + int err; + struct path path; + + br = au_sbr(sb, nsi_lock->bindex); + h_mnt = au_br_mnt(br); + h_sb = h_mnt->mnt_sb; + /* todo: call lower fh_to_dentry()? fh_to_parent()? */ + h_parent = exportfs_decode_fh(h_mnt, (void *)(fh + Fh_tail), + fh_len - Fh_tail, fh[Fh_h_type], + h_acceptable, /*context*/NULL); + dentry = h_parent; + if (unlikely(!h_parent || IS_ERR(h_parent))) { + AuWarn1("%s decode_fh failed, %ld\n", + au_sbtype(h_sb), PTR_ERR(h_parent)); + goto out; + } + dentry = NULL; + if (unlikely(au_test_anon(h_parent))) { + AuWarn1("%s decode_fh returned a disconnected dentry\n", + au_sbtype(h_sb)); + goto out_h_parent; + } + + dentry = ERR_PTR(-ENOMEM); + pathname = (void *)__get_free_page(GFP_NOFS); + if (unlikely(!pathname)) + goto out_h_parent; + + root = sb->s_root; + path.mnt = h_mnt; + di_read_lock_parent(root, !AuLock_IR); + path.dentry = au_h_dptr(root, nsi_lock->bindex); + di_read_unlock(root, !AuLock_IR); + p = au_build_path(h_parent, &path, pathname, PAGE_SIZE, sb); + dentry = (void *)p; + if (IS_ERR(p)) + goto out_pathname; + + si_read_unlock(sb); + err = vfsub_kern_path(p, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &path); + dentry = ERR_PTR(err); + if (unlikely(err)) + goto out_relock; + + dentry = ERR_PTR(-ENOENT); + AuDebugOn(au_test_anon(path.dentry)); + if (unlikely(!path.dentry->d_inode)) + goto out_path; + + if (ino != path.dentry->d_inode->i_ino) + dentry = au_lkup_by_ino(&path, ino, /*nsi_lock*/NULL); + else + dentry = dget(path.dentry); + +out_path: + path_put(&path); +out_relock: + if (unlikely(si_nfsd_read_lock(sb, nsi_lock) < 0)) + if (!IS_ERR(dentry)) { + dput(dentry); + dentry = ERR_PTR(-ESTALE); + } +out_pathname: + free_page((unsigned long)pathname); +out_h_parent: + dput(h_parent); +out: + AuTraceErrPtr(dentry); + return dentry; +} + +/* ---------------------------------------------------------------------- */ + +static struct dentry * +aufs_fh_to_dentry(struct super_block *sb, struct fid *fid, int fh_len, + int fh_type) +{ + struct dentry *dentry; + __u32 *fh = fid->raw; + struct au_branch *br; + ino_t ino, dir_ino; + struct au_nfsd_si_lock nsi_lock = { + .force_lock = 0 + }; + + dentry = ERR_PTR(-ESTALE); + /* it should never happen, but the file handle is unreliable */ + if (unlikely(fh_len < Fh_tail)) + goto out; + nsi_lock.sigen = fh[Fh_sigen]; + nsi_lock.br_id = fh[Fh_br_id]; + + /* branch id may be wrapped around */ + br = NULL; + if (unlikely(si_nfsd_read_lock(sb, &nsi_lock))) + goto out; + nsi_lock.force_lock = 1; + + /* is this inode still cached? */ + ino = decode_ino(fh + Fh_ino); + /* it should never happen */ + if (unlikely(ino == AUFS_ROOT_INO)) + goto out; + + dir_ino = decode_ino(fh + Fh_dir_ino); + dentry = decode_by_ino(sb, ino, dir_ino); + if (IS_ERR(dentry)) + goto out_unlock; + if (dentry) + goto accept; + + /* is the parent dir cached? */ + br = au_sbr(sb, nsi_lock.bindex); + atomic_inc(&br->br_count); + dentry = decode_by_dir_ino(sb, ino, dir_ino, &nsi_lock); + if (IS_ERR(dentry)) + goto out_unlock; + if (dentry) + goto accept; + + /* lookup path */ + dentry = decode_by_path(sb, ino, fh, fh_len, &nsi_lock); + if (IS_ERR(dentry)) + goto out_unlock; + if (unlikely(!dentry)) + /* todo?: make it ESTALE */ + goto out_unlock; + +accept: + if (!au_digen_test(dentry, au_sigen(sb)) + && dentry->d_inode->i_generation == fh[Fh_igen]) + goto out_unlock; /* success */ + + dput(dentry); + dentry = ERR_PTR(-ESTALE); +out_unlock: + if (br) + atomic_dec(&br->br_count); + si_read_unlock(sb); +out: + AuTraceErrPtr(dentry); + return dentry; +} + +#if 0 /* reserved for future use */ +/* support subtreecheck option */ +static struct dentry *aufs_fh_to_parent(struct super_block *sb, struct fid *fid, + int fh_len, int fh_type) +{ + struct dentry *parent; + __u32 *fh = fid->raw; + ino_t dir_ino; + + dir_ino = decode_ino(fh + Fh_dir_ino); + parent = decode_by_ino(sb, dir_ino, 0); + if (IS_ERR(parent)) + goto out; + if (!parent) + parent = decode_by_path(sb, au_br_index(sb, fh[Fh_br_id]), + dir_ino, fh, fh_len); + +out: + AuTraceErrPtr(parent); + return parent; +} +#endif + +/* ---------------------------------------------------------------------- */ + +static int aufs_encode_fh(struct inode *inode, __u32 *fh, int *max_len, + struct inode *dir) +{ + int err; + aufs_bindex_t bindex; + struct super_block *sb, *h_sb; + struct dentry *dentry, *parent, *h_parent; + struct inode *h_dir; + struct au_branch *br; + + err = -ENOSPC; + if (unlikely(*max_len <= Fh_tail)) { + AuWarn1("NFSv2 client (max_len %d)?\n", *max_len); + goto out; + } + + err = FILEID_ROOT; + if (inode->i_ino == AUFS_ROOT_INO) { + AuDebugOn(inode->i_ino != AUFS_ROOT_INO); + goto out; + } + + h_parent = NULL; + sb = inode->i_sb; + err = si_read_lock(sb, AuLock_FLUSH); + if (unlikely(err)) + goto out; + +#ifdef CONFIG_AUFS_DEBUG + if (unlikely(!au_opt_test(au_mntflags(sb), XINO))) + AuWarn1("NFS-exporting requires xino\n"); +#endif + err = -EIO; + parent = NULL; + ii_read_lock_child(inode); + bindex = au_ibstart(inode); + if (!dir) { + dentry = d_find_any_alias(inode); + if (unlikely(!dentry)) + goto out_unlock; + AuDebugOn(au_test_anon(dentry)); + parent = dget_parent(dentry); + dput(dentry); + if (unlikely(!parent)) + goto out_unlock; + dir = parent->d_inode; + } + + ii_read_lock_parent(dir); + h_dir = au_h_iptr(dir, bindex); + ii_read_unlock(dir); + if (unlikely(!h_dir)) + goto out_parent; + h_parent = d_find_any_alias(h_dir); + if (unlikely(!h_parent)) + goto out_hparent; + + err = -EPERM; + br = au_sbr(sb, bindex); + h_sb = au_br_sb(br); + if (unlikely(!h_sb->s_export_op)) { + AuErr1("%s branch is not exportable\n", au_sbtype(h_sb)); + goto out_hparent; + } + + fh[Fh_br_id] = br->br_id; + fh[Fh_sigen] = au_sigen(sb); + encode_ino(fh + Fh_ino, inode->i_ino); + encode_ino(fh + Fh_dir_ino, dir->i_ino); + fh[Fh_igen] = inode->i_generation; + + *max_len -= Fh_tail; + fh[Fh_h_type] = exportfs_encode_fh(h_parent, (void *)(fh + Fh_tail), + max_len, + /*connectable or subtreecheck*/0); + err = fh[Fh_h_type]; + *max_len += Fh_tail; + /* todo: macros? */ + if (err != FILEID_INVALID) + err = 99; + else + AuWarn1("%s encode_fh failed\n", au_sbtype(h_sb)); + +out_hparent: + dput(h_parent); +out_parent: + dput(parent); +out_unlock: + ii_read_unlock(inode); + si_read_unlock(sb); +out: + if (unlikely(err < 0)) + err = FILEID_INVALID; + return err; +} + +/* ---------------------------------------------------------------------- */ + +static int aufs_commit_metadata(struct inode *inode) +{ + int err; + aufs_bindex_t bindex; + struct super_block *sb; + struct inode *h_inode; + int (*f)(struct inode *inode); + + sb = inode->i_sb; + si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW); + ii_write_lock_child(inode); + bindex = au_ibstart(inode); + AuDebugOn(bindex < 0); + h_inode = au_h_iptr(inode, bindex); + + f = h_inode->i_sb->s_export_op->commit_metadata; + if (f) + err = f(h_inode); + else { + struct writeback_control wbc = { + .sync_mode = WB_SYNC_ALL, + .nr_to_write = 0 /* metadata only */ + }; + + err = sync_inode(h_inode, &wbc); + } + + au_cpup_attr_timesizes(inode); + ii_write_unlock(inode); + si_read_unlock(sb); + return err; +} + +/* ---------------------------------------------------------------------- */ + +static struct export_operations aufs_export_op = { + .fh_to_dentry = aufs_fh_to_dentry, + /* .fh_to_parent = aufs_fh_to_parent, */ + .encode_fh = aufs_encode_fh, + .commit_metadata = aufs_commit_metadata +}; + +void au_export_init(struct super_block *sb) +{ + struct au_sbinfo *sbinfo; + __u32 u; + + sb->s_export_op = &aufs_export_op; + sbinfo = au_sbi(sb); + sbinfo->si_xigen = NULL; + get_random_bytes(&u, sizeof(u)); + BUILD_BUG_ON(sizeof(u) != sizeof(int)); + atomic_set(&sbinfo->si_xigen_next, u); +} diff -Naur kernel.21.3.orig/fs/aufs/fhsm.c kernel.21.3.new/fs/aufs/fhsm.c --- kernel.21.3.orig/fs/aufs/fhsm.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/fhsm.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,426 @@ +/* + * Copyright (C) 2011-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ + +/* + * File-based Hierarchy Storage Management + */ + +#include +#include +#include +#include +#include "aufs.h" + +static aufs_bindex_t au_fhsm_bottom(struct super_block *sb) +{ + struct au_sbinfo *sbinfo; + struct au_fhsm *fhsm; + + SiMustAnyLock(sb); + + sbinfo = au_sbi(sb); + fhsm = &sbinfo->si_fhsm; + AuDebugOn(!fhsm); + return fhsm->fhsm_bottom; +} + +void au_fhsm_set_bottom(struct super_block *sb, aufs_bindex_t bindex) +{ + struct au_sbinfo *sbinfo; + struct au_fhsm *fhsm; + + SiMustWriteLock(sb); + + sbinfo = au_sbi(sb); + fhsm = &sbinfo->si_fhsm; + AuDebugOn(!fhsm); + fhsm->fhsm_bottom = bindex; +} + +/* ---------------------------------------------------------------------- */ + +static int au_fhsm_test_jiffy(struct au_sbinfo *sbinfo, struct au_branch *br) +{ + struct au_br_fhsm *bf; + + bf = br->br_fhsm; + MtxMustLock(&bf->bf_lock); + + return !bf->bf_readable + || time_after(jiffies, + bf->bf_jiffy + sbinfo->si_fhsm.fhsm_expire); +} + +/* ---------------------------------------------------------------------- */ + +static void au_fhsm_notify(struct super_block *sb, int val) +{ + struct au_sbinfo *sbinfo; + struct au_fhsm *fhsm; + + SiMustAnyLock(sb); + + sbinfo = au_sbi(sb); + fhsm = &sbinfo->si_fhsm; + if (au_fhsm_pid(fhsm) + && atomic_read(&fhsm->fhsm_readable) != -1) { + atomic_set(&fhsm->fhsm_readable, val); + if (val) + wake_up(&fhsm->fhsm_wqh); + } +} + +static int au_fhsm_stfs(struct super_block *sb, aufs_bindex_t bindex, + struct aufs_stfs *rstfs, int do_lock, int do_notify) +{ + int err; + struct au_branch *br; + struct au_br_fhsm *bf; + + br = au_sbr(sb, bindex); + AuDebugOn(au_br_rdonly(br)); + bf = br->br_fhsm; + AuDebugOn(!bf); + + if (do_lock) + mutex_lock(&bf->bf_lock); + else + MtxMustLock(&bf->bf_lock); + + /* sb->s_root for NFS is unreliable */ + err = au_br_stfs(br, &bf->bf_stfs); + if (unlikely(err)) { + AuErr1("FHSM failed (%d), b%d, ignored.\n", bindex, err); + goto out; + } + + bf->bf_jiffy = jiffies; + bf->bf_readable = 1; + if (do_notify) + au_fhsm_notify(sb, /*val*/1); + if (rstfs) + *rstfs = bf->bf_stfs; + +out: + if (do_lock) + mutex_unlock(&bf->bf_lock); + au_fhsm_notify(sb, /*val*/1); + + return err; +} + +void au_fhsm_wrote(struct super_block *sb, aufs_bindex_t bindex, int force) +{ + int err; + struct au_sbinfo *sbinfo; + struct au_fhsm *fhsm; + struct au_branch *br; + struct au_br_fhsm *bf; + + AuDbg("b%d, force %d\n", bindex, force); + SiMustAnyLock(sb); + + sbinfo = au_sbi(sb); + fhsm = &sbinfo->si_fhsm; + if (!au_ftest_si(sbinfo, FHSM) + || fhsm->fhsm_bottom == bindex) + return; + + br = au_sbr(sb, bindex); + bf = br->br_fhsm; + AuDebugOn(!bf); + mutex_lock(&bf->bf_lock); + if (force + || au_fhsm_pid(fhsm) + || au_fhsm_test_jiffy(sbinfo, br)) + err = au_fhsm_stfs(sb, bindex, /*rstfs*/NULL, /*do_lock*/0, + /*do_notify*/1); + mutex_unlock(&bf->bf_lock); +} + +void au_fhsm_wrote_all(struct super_block *sb, int force) +{ + aufs_bindex_t bindex, bend; + struct au_branch *br; + + /* exclude the bottom */ + bend = au_fhsm_bottom(sb); + for (bindex = 0; bindex < bend; bindex++) { + br = au_sbr(sb, bindex); + if (au_br_fhsm(br->br_perm)) + au_fhsm_wrote(sb, bindex, force); + } +} + +/* ---------------------------------------------------------------------- */ + +static unsigned int au_fhsm_poll(struct file *file, + struct poll_table_struct *wait) +{ + unsigned int mask; + struct au_sbinfo *sbinfo; + struct au_fhsm *fhsm; + + mask = 0; + sbinfo = file->private_data; + fhsm = &sbinfo->si_fhsm; + poll_wait(file, &fhsm->fhsm_wqh, wait); + if (atomic_read(&fhsm->fhsm_readable)) + mask = POLLIN /* | POLLRDNORM */; + + AuTraceErr((int)mask); + return mask; +} + +static int au_fhsm_do_read_one(struct aufs_stbr __user *stbr, + struct aufs_stfs *stfs, __s16 brid) +{ + int err; + + err = copy_to_user(&stbr->stfs, stfs, sizeof(*stfs)); + if (!err) + err = __put_user(brid, &stbr->brid); + if (unlikely(err)) + err = -EFAULT; + + return err; +} + +static ssize_t au_fhsm_do_read(struct super_block *sb, + struct aufs_stbr __user *stbr, size_t count) +{ + ssize_t err; + int nstbr; + aufs_bindex_t bindex, bend; + struct au_branch *br; + struct au_br_fhsm *bf; + + /* except the bottom branch */ + err = 0; + nstbr = 0; + bend = au_fhsm_bottom(sb); + for (bindex = 0; !err && bindex < bend; bindex++) { + br = au_sbr(sb, bindex); + if (!au_br_fhsm(br->br_perm)) + continue; + + bf = br->br_fhsm; + mutex_lock(&bf->bf_lock); + if (bf->bf_readable) { + err = -EFAULT; + if (count >= sizeof(*stbr)) + err = au_fhsm_do_read_one(stbr++, &bf->bf_stfs, + br->br_id); + if (!err) { + bf->bf_readable = 0; + count -= sizeof(*stbr); + nstbr++; + } + } + mutex_unlock(&bf->bf_lock); + } + if (!err) + err = sizeof(*stbr) * nstbr; + + return err; +} + +static ssize_t au_fhsm_read(struct file *file, char __user *buf, size_t count, + loff_t *pos) +{ + ssize_t err; + int readable; + aufs_bindex_t nfhsm, bindex, bend; + struct au_sbinfo *sbinfo; + struct au_fhsm *fhsm; + struct au_branch *br; + struct super_block *sb; + + err = 0; + sbinfo = file->private_data; + fhsm = &sbinfo->si_fhsm; +need_data: + spin_lock_irq(&fhsm->fhsm_wqh.lock); + if (!atomic_read(&fhsm->fhsm_readable)) { + if (vfsub_file_flags(file) & O_NONBLOCK) + err = -EAGAIN; + else + err = wait_event_interruptible_locked_irq + (fhsm->fhsm_wqh, + atomic_read(&fhsm->fhsm_readable)); + } + spin_unlock_irq(&fhsm->fhsm_wqh.lock); + if (unlikely(err)) + goto out; + + /* sb may already be dead */ + au_rw_read_lock(&sbinfo->si_rwsem); + readable = atomic_read(&fhsm->fhsm_readable); + if (readable > 0) { + sb = sbinfo->si_sb; + AuDebugOn(!sb); + /* exclude the bottom branch */ + nfhsm = 0; + bend = au_fhsm_bottom(sb); + for (bindex = 0; bindex < bend; bindex++) { + br = au_sbr(sb, bindex); + if (au_br_fhsm(br->br_perm)) + nfhsm++; + } + err = -EMSGSIZE; + if (nfhsm * sizeof(struct aufs_stbr) <= count) { + atomic_set(&fhsm->fhsm_readable, 0); + err = au_fhsm_do_read(sbinfo->si_sb, (void __user *)buf, + count); + } + } + au_rw_read_unlock(&sbinfo->si_rwsem); + if (!readable) + goto need_data; + +out: + return err; +} + +static int au_fhsm_release(struct inode *inode, struct file *file) +{ + struct au_sbinfo *sbinfo; + struct au_fhsm *fhsm; + + /* sb may already be dead */ + sbinfo = file->private_data; + fhsm = &sbinfo->si_fhsm; + spin_lock(&fhsm->fhsm_spin); + fhsm->fhsm_pid = 0; + spin_unlock(&fhsm->fhsm_spin); + kobject_put(&sbinfo->si_kobj); + + return 0; +} + +static const struct file_operations au_fhsm_fops = { + .owner = THIS_MODULE, + .llseek = noop_llseek, + .read = au_fhsm_read, + .poll = au_fhsm_poll, + .release = au_fhsm_release +}; + +int au_fhsm_fd(struct super_block *sb, int oflags) +{ + int err, fd; + struct au_sbinfo *sbinfo; + struct au_fhsm *fhsm; + + err = -EPERM; + if (unlikely(!capable(CAP_SYS_ADMIN))) + goto out; + + err = -EINVAL; + if (unlikely(oflags & ~(O_CLOEXEC | O_NONBLOCK))) + goto out; + + err = 0; + sbinfo = au_sbi(sb); + fhsm = &sbinfo->si_fhsm; + spin_lock(&fhsm->fhsm_spin); + if (!fhsm->fhsm_pid) + fhsm->fhsm_pid = current->pid; + else + err = -EBUSY; + spin_unlock(&fhsm->fhsm_spin); + if (unlikely(err)) + goto out; + + oflags |= O_RDONLY; + /* oflags |= FMODE_NONOTIFY; */ + fd = anon_inode_getfd("[aufs_fhsm]", &au_fhsm_fops, sbinfo, oflags); + err = fd; + if (unlikely(fd < 0)) + goto out_pid; + + /* succeed reglardless 'fhsm' status */ + kobject_get(&sbinfo->si_kobj); + si_noflush_read_lock(sb); + if (au_ftest_si(sbinfo, FHSM)) + au_fhsm_wrote_all(sb, /*force*/0); + si_read_unlock(sb); + goto out; /* success */ + +out_pid: + spin_lock(&fhsm->fhsm_spin); + fhsm->fhsm_pid = 0; + spin_unlock(&fhsm->fhsm_spin); +out: + AuTraceErr(err); + return err; +} + +/* ---------------------------------------------------------------------- */ + +int au_fhsm_br_alloc(struct au_branch *br) +{ + int err; + + err = 0; + br->br_fhsm = kmalloc(sizeof(*br->br_fhsm), GFP_NOFS); + if (br->br_fhsm) + au_br_fhsm_init(br->br_fhsm); + else + err = -ENOMEM; + + return err; +} + +/* ---------------------------------------------------------------------- */ + +void au_fhsm_fin(struct super_block *sb) +{ + au_fhsm_notify(sb, /*val*/-1); +} + +void au_fhsm_init(struct au_sbinfo *sbinfo) +{ + struct au_fhsm *fhsm; + + fhsm = &sbinfo->si_fhsm; + spin_lock_init(&fhsm->fhsm_spin); + init_waitqueue_head(&fhsm->fhsm_wqh); + atomic_set(&fhsm->fhsm_readable, 0); + fhsm->fhsm_expire + = msecs_to_jiffies(AUFS_FHSM_CACHE_DEF_SEC * MSEC_PER_SEC); + fhsm->fhsm_bottom = -1; +} + +void au_fhsm_set(struct au_sbinfo *sbinfo, unsigned int sec) +{ + sbinfo->si_fhsm.fhsm_expire + = msecs_to_jiffies(sec * MSEC_PER_SEC); +} + +void au_fhsm_show(struct seq_file *seq, struct au_sbinfo *sbinfo) +{ + unsigned int u; + + if (!au_ftest_si(sbinfo, FHSM)) + return; + + u = jiffies_to_msecs(sbinfo->si_fhsm.fhsm_expire) / MSEC_PER_SEC; + if (u != AUFS_FHSM_CACHE_DEF_SEC) + seq_printf(seq, ",fhsm_sec=%u", u); +} diff -Naur kernel.21.3.orig/fs/aufs/file.c kernel.21.3.new/fs/aufs/file.c --- kernel.21.3.orig/fs/aufs/file.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/file.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,815 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * handling file/dir, and address_space operation + */ + +#ifdef CONFIG_AUFS_DEBUG +#include +#endif +#include +#include "aufs.h" + +/* drop flags for writing */ +unsigned int au_file_roflags(unsigned int flags) +{ + flags &= ~(O_WRONLY | O_RDWR | O_APPEND | O_CREAT | O_TRUNC); + flags |= O_RDONLY | O_NOATIME; + return flags; +} + +/* common functions to regular file and dir */ +struct file *au_h_open(struct dentry *dentry, aufs_bindex_t bindex, int flags, + struct file *file, int force_wr) +{ + struct file *h_file; + struct dentry *h_dentry; + struct inode *h_inode; + struct super_block *sb; + struct au_branch *br; + struct path h_path; + int err, exec_flag; + + /* a race condition can happen between open and unlink/rmdir */ + h_file = ERR_PTR(-ENOENT); + h_dentry = au_h_dptr(dentry, bindex); + if (au_test_nfsd() && !h_dentry) + goto out; + h_inode = h_dentry->d_inode; + if (au_test_nfsd() && !h_inode) + goto out; + spin_lock(&h_dentry->d_lock); + err = (!d_unhashed(dentry) && d_unlinked(h_dentry)) + || !h_inode + /* || !dentry->d_inode->i_nlink */ + ; + spin_unlock(&h_dentry->d_lock); + if (unlikely(err)) + goto out; + + sb = dentry->d_sb; + br = au_sbr(sb, bindex); + h_file = ERR_PTR(-EACCES); + exec_flag = flags & __FMODE_EXEC; + if (exec_flag && (au_br_mnt(br)->mnt_flags & MNT_NOEXEC)) + goto out; + + /* drop flags for writing */ + if (au_test_ro(sb, bindex, dentry->d_inode)) { + if (force_wr && !(flags & O_WRONLY)) + force_wr = 0; + flags = au_file_roflags(flags); + if (force_wr) { + h_file = ERR_PTR(-EROFS); + flags = au_file_roflags(flags); + if (unlikely(vfsub_native_ro(h_inode) + || IS_APPEND(h_inode))) + goto out; + flags &= ~O_ACCMODE; + flags |= O_WRONLY; + } + } + flags &= ~O_CREAT; + atomic_inc(&br->br_count); + h_path.dentry = h_dentry; + h_path.mnt = au_br_mnt(br); + h_file = vfsub_dentry_open(&h_path, flags); + if (IS_ERR(h_file)) + goto out_br; + + if (exec_flag) { + err = deny_write_access(h_file); + if (unlikely(err)) { + fput(h_file); + h_file = ERR_PTR(err); + goto out_br; + } + } + fsnotify_open(h_file); + goto out; /* success */ + +out_br: + atomic_dec(&br->br_count); +out: + return h_file; +} + +static int au_cmoo(struct dentry *dentry) +{ + int err, cmoo; + unsigned int udba; + struct path h_path; + struct au_pin pin; + struct au_cp_generic cpg = { + .dentry = dentry, + .bdst = -1, + .bsrc = -1, + .len = -1, + .pin = &pin, + .flags = AuCpup_DTIME | AuCpup_HOPEN + }; + struct inode *inode; + struct super_block *sb; + struct au_sbinfo *sbinfo; + struct au_fhsm *fhsm; + pid_t pid; + struct au_branch *br; + struct dentry *parent; + struct au_hinode *hdir; + + DiMustWriteLock(dentry); + inode = dentry->d_inode; + IiMustWriteLock(inode); + + err = 0; + if (IS_ROOT(dentry)) + goto out; + cpg.bsrc = au_dbstart(dentry); + if (!cpg.bsrc) + goto out; + + sb = dentry->d_sb; + sbinfo = au_sbi(sb); + fhsm = &sbinfo->si_fhsm; + pid = au_fhsm_pid(fhsm); + if (pid + && (current->pid == pid + || current->real_parent->pid == pid)) + goto out; + + br = au_sbr(sb, cpg.bsrc); + cmoo = au_br_cmoo(br->br_perm); + if (!cmoo) + goto out; + if (!S_ISREG(inode->i_mode)) + cmoo &= AuBrAttr_COO_ALL; + if (!cmoo) + goto out; + + parent = dget_parent(dentry); + di_write_lock_parent(parent); + err = au_wbr_do_copyup_bu(dentry, cpg.bsrc - 1); + cpg.bdst = err; + if (unlikely(err < 0)) { + err = 0; /* there is no upper writable branch */ + goto out_dgrade; + } + AuDbg("bsrc %d, bdst %d\n", cpg.bsrc, cpg.bdst); + + /* do not respect the coo attrib for the target branch */ + err = au_cpup_dirs(dentry, cpg.bdst); + if (unlikely(err)) + goto out_dgrade; + + di_downgrade_lock(parent, AuLock_IR); + udba = au_opt_udba(sb); + err = au_pin(&pin, dentry, cpg.bdst, udba, + AuPin_DI_LOCKED | AuPin_MNT_WRITE); + if (unlikely(err)) + goto out_parent; + + err = au_sio_cpup_simple(&cpg); + au_unpin(&pin); + if (unlikely(err)) + goto out_parent; + if (!(cmoo & AuBrWAttr_MOO)) + goto out_parent; /* success */ + + err = au_pin(&pin, dentry, cpg.bsrc, udba, + AuPin_DI_LOCKED | AuPin_MNT_WRITE); + if (unlikely(err)) + goto out_parent; + + h_path.mnt = au_br_mnt(br); + h_path.dentry = au_h_dptr(dentry, cpg.bsrc); + hdir = au_hi(parent->d_inode, cpg.bsrc); + err = vfsub_unlink(hdir->hi_inode, &h_path, /*force*/1); + au_unpin(&pin); + /* todo: keep h_dentry or not? */ + if (unlikely(err)) { + pr_err("unlink %.*s after coo failed (%d), ignored\n", + AuDLNPair(dentry), err); + err = 0; + } + goto out_parent; /* success */ + +out_dgrade: + di_downgrade_lock(parent, AuLock_IR); +out_parent: + di_read_unlock(parent, AuLock_IR); + dput(parent); +out: + AuTraceErr(err); + return err; +} + +int au_do_open(struct file *file, int (*open)(struct file *file, int flags), + struct au_fidir *fidir) +{ + int err; + struct dentry *dentry; + + err = au_finfo_init(file, fidir); + if (unlikely(err)) + goto out; + + dentry = file->f_dentry; + di_write_lock_child(dentry); + err = au_cmoo(dentry); + di_downgrade_lock(dentry, AuLock_IR); + if (!err) + err = open(file, vfsub_file_flags(file)); + di_read_unlock(dentry, AuLock_IR); + + fi_write_unlock(file); + if (unlikely(err)) { + au_fi(file)->fi_hdir = NULL; + au_finfo_fin(file); + } + +out: + return err; +} + +int au_reopen_nondir(struct file *file) +{ + int err; + aufs_bindex_t bstart; + struct dentry *dentry; + struct file *h_file, *h_file_tmp; + + dentry = file->f_dentry; + bstart = au_dbstart(dentry); + h_file_tmp = NULL; + if (au_fbstart(file) == bstart) { + h_file = au_hf_top(file); + if (file->f_mode == h_file->f_mode) + return 0; /* success */ + h_file_tmp = h_file; + get_file(h_file_tmp); + au_set_h_fptr(file, bstart, NULL); + } + AuDebugOn(au_fi(file)->fi_hdir); + /* + * it can happen + * file exists on both of rw and ro + * open --> dbstart and fbstart are both 0 + * prepend a branch as rw, "rw" become ro + * remove rw/file + * delete the top branch, "rw" becomes rw again + * --> dbstart is 1, fbstart is still 0 + * write --> fbstart is 0 but dbstart is 1 + */ + /* AuDebugOn(au_fbstart(file) < bstart); */ + + h_file = au_h_open(dentry, bstart, vfsub_file_flags(file) & ~O_TRUNC, + file, /*force_wr*/0); + err = PTR_ERR(h_file); + if (IS_ERR(h_file)) { + if (h_file_tmp) { + atomic_inc(&au_sbr(dentry->d_sb, bstart)->br_count); + au_set_h_fptr(file, bstart, h_file_tmp); + h_file_tmp = NULL; + } + goto out; /* todo: close all? */ + } + + err = 0; + au_set_fbstart(file, bstart); + au_set_h_fptr(file, bstart, h_file); + au_update_figen(file); + /* todo: necessary? */ + /* file->f_ra = h_file->f_ra; */ + +out: + if (h_file_tmp) + fput(h_file_tmp); + return err; +} + +/* ---------------------------------------------------------------------- */ + +static int au_reopen_wh(struct file *file, aufs_bindex_t btgt, + struct dentry *hi_wh) +{ + int err; + aufs_bindex_t bstart; + struct au_dinfo *dinfo; + struct dentry *h_dentry; + struct au_hdentry *hdp; + + dinfo = au_di(file->f_dentry); + AuRwMustWriteLock(&dinfo->di_rwsem); + + bstart = dinfo->di_bstart; + dinfo->di_bstart = btgt; + hdp = dinfo->di_hdentry; + h_dentry = hdp[0 + btgt].hd_dentry; + hdp[0 + btgt].hd_dentry = hi_wh; + err = au_reopen_nondir(file); + hdp[0 + btgt].hd_dentry = h_dentry; + dinfo->di_bstart = bstart; + + return err; +} + +static int au_ready_to_write_wh(struct file *file, loff_t len, + aufs_bindex_t bcpup, struct au_pin *pin) +{ + int err; + struct inode *inode, *h_inode; + struct dentry *h_dentry, *hi_wh; + struct au_cp_generic cpg = { + .dentry = file->f_dentry, + .bdst = bcpup, + .bsrc = -1, + .len = len, + .pin = pin + }; + + au_update_dbstart(cpg.dentry); + inode = cpg.dentry->d_inode; + h_inode = NULL; + if (au_dbstart(cpg.dentry) <= bcpup + && au_dbend(cpg.dentry) >= bcpup) { + h_dentry = au_h_dptr(cpg.dentry, bcpup); + if (h_dentry) + h_inode = h_dentry->d_inode; + } + hi_wh = au_hi_wh(inode, bcpup); + if (!hi_wh && !h_inode) + err = au_sio_cpup_wh(&cpg, file); + else + /* already copied-up after unlink */ + err = au_reopen_wh(file, bcpup, hi_wh); + + if (!err + && inode->i_nlink > 1 + && au_opt_test(au_mntflags(cpg.dentry->d_sb), PLINK)) + au_plink_append(inode, bcpup, au_h_dptr(cpg.dentry, bcpup)); + + return err; +} + +/* + * prepare the @file for writing. + */ +int au_ready_to_write(struct file *file, loff_t len, struct au_pin *pin) +{ + int err; + aufs_bindex_t dbstart; + struct dentry *parent, *h_dentry; + struct inode *inode; + struct super_block *sb; + struct file *h_file; + struct au_cp_generic cpg = { + .dentry = file->f_dentry, + .bdst = -1, + .bsrc = -1, + .len = len, + .pin = pin, + .flags = AuCpup_DTIME + }; + + sb = cpg.dentry->d_sb; + inode = cpg.dentry->d_inode; + cpg.bsrc = au_fbstart(file); + err = au_test_ro(sb, cpg.bsrc, inode); + if (!err && (au_hf_top(file)->f_mode & FMODE_WRITE)) { + err = au_pin(pin, cpg.dentry, cpg.bsrc, AuOpt_UDBA_NONE, + /*flags*/0); + goto out; + } + + /* need to cpup or reopen */ + parent = dget_parent(cpg.dentry); + di_write_lock_parent(parent); + err = AuWbrCopyup(au_sbi(sb), cpg.dentry); + cpg.bdst = err; + if (unlikely(err < 0)) + goto out_dgrade; + err = 0; + + if (!d_unhashed(cpg.dentry) && !au_h_dptr(parent, cpg.bdst)) { + err = au_cpup_dirs(cpg.dentry, cpg.bdst); + if (unlikely(err)) + goto out_dgrade; + } + + err = au_pin(pin, cpg.dentry, cpg.bdst, AuOpt_UDBA_NONE, + AuPin_DI_LOCKED | AuPin_MNT_WRITE); + if (unlikely(err)) + goto out_dgrade; + + h_dentry = au_hf_top(file)->f_dentry; + dbstart = au_dbstart(cpg.dentry); + if (dbstart <= cpg.bdst) { + h_dentry = au_h_dptr(cpg.dentry, cpg.bdst); + AuDebugOn(!h_dentry); + cpg.bsrc = cpg.bdst; + } + + if (dbstart <= cpg.bdst /* just reopen */ + || !d_unhashed(cpg.dentry) /* copyup and reopen */ + ) { + h_file = au_h_open_pre(cpg.dentry, cpg.bsrc, /*force_wr*/0); + if (IS_ERR(h_file)) + err = PTR_ERR(h_file); + else { + di_downgrade_lock(parent, AuLock_IR); + if (dbstart > cpg.bdst) + err = au_sio_cpup_simple(&cpg); + if (!err) + err = au_reopen_nondir(file); + au_h_open_post(cpg.dentry, cpg.bsrc, h_file); + } + } else { /* copyup as wh and reopen */ + /* + * since writable hfsplus branch is not supported, + * h_open_pre/post() are unnecessary. + */ + err = au_ready_to_write_wh(file, len, cpg.bdst, pin); + di_downgrade_lock(parent, AuLock_IR); + } + + if (!err) { + au_pin_set_parent_lflag(pin, /*lflag*/0); + goto out_dput; /* success */ + } + au_unpin(pin); + goto out_unlock; + +out_dgrade: + di_downgrade_lock(parent, AuLock_IR); +out_unlock: + di_read_unlock(parent, AuLock_IR); +out_dput: + dput(parent); +out: + return err; +} + +/* ---------------------------------------------------------------------- */ + +int au_do_flush(struct file *file, fl_owner_t id, + int (*flush)(struct file *file, fl_owner_t id)) +{ + int err; + struct super_block *sb; + struct inode *inode; + + inode = file_inode(file); + sb = inode->i_sb; + si_noflush_read_lock(sb); + fi_read_lock(file); + ii_read_lock_child(inode); + + err = flush(file, id); + au_cpup_attr_timesizes(inode); + + ii_read_unlock(inode); + fi_read_unlock(file); + si_read_unlock(sb); + return err; +} + +/* ---------------------------------------------------------------------- */ + +static int au_file_refresh_by_inode(struct file *file, int *need_reopen) +{ + int err; + struct au_pin pin; + struct au_finfo *finfo; + struct dentry *parent, *hi_wh; + struct inode *inode; + struct super_block *sb; + struct au_cp_generic cpg = { + .dentry = file->f_dentry, + .bdst = -1, + .bsrc = -1, + .len = -1, + .pin = &pin, + .flags = AuCpup_DTIME + }; + + FiMustWriteLock(file); + + err = 0; + finfo = au_fi(file); + sb = cpg.dentry->d_sb; + inode = cpg.dentry->d_inode; + cpg.bdst = au_ibstart(inode); + if (cpg.bdst == finfo->fi_btop || IS_ROOT(cpg.dentry)) + goto out; + + parent = dget_parent(cpg.dentry); + if (au_test_ro(sb, cpg.bdst, inode)) { + di_read_lock_parent(parent, !AuLock_IR); + err = AuWbrCopyup(au_sbi(sb), cpg.dentry); + cpg.bdst = err; + di_read_unlock(parent, !AuLock_IR); + if (unlikely(err < 0)) + goto out_parent; + err = 0; + } + + di_read_lock_parent(parent, AuLock_IR); + hi_wh = au_hi_wh(inode, cpg.bdst); + if (!S_ISDIR(inode->i_mode) + && au_opt_test(au_mntflags(sb), PLINK) + && au_plink_test(inode) + && !d_unhashed(cpg.dentry) + && cpg.bdst < au_dbstart(cpg.dentry)) { + err = au_test_and_cpup_dirs(cpg.dentry, cpg.bdst); + if (unlikely(err)) + goto out_unlock; + + /* always superio. */ + err = au_pin(&pin, cpg.dentry, cpg.bdst, AuOpt_UDBA_NONE, + AuPin_DI_LOCKED | AuPin_MNT_WRITE); + if (!err) { + err = au_sio_cpup_simple(&cpg); + au_unpin(&pin); + } + } else if (hi_wh) { + /* already copied-up after unlink */ + err = au_reopen_wh(file, cpg.bdst, hi_wh); + *need_reopen = 0; + } + +out_unlock: + di_read_unlock(parent, AuLock_IR); +out_parent: + dput(parent); +out: + return err; +} + +static void au_do_refresh_dir(struct file *file) +{ + aufs_bindex_t bindex, bend, new_bindex, brid; + struct au_hfile *p, tmp, *q; + struct au_finfo *finfo; + struct super_block *sb; + struct au_fidir *fidir; + + FiMustWriteLock(file); + + sb = file->f_dentry->d_sb; + finfo = au_fi(file); + fidir = finfo->fi_hdir; + AuDebugOn(!fidir); + p = fidir->fd_hfile + finfo->fi_btop; + brid = p->hf_br->br_id; + bend = fidir->fd_bbot; + for (bindex = finfo->fi_btop; bindex <= bend; bindex++, p++) { + if (!p->hf_file) + continue; + + new_bindex = au_br_index(sb, p->hf_br->br_id); + if (new_bindex == bindex) + continue; + if (new_bindex < 0) { + au_set_h_fptr(file, bindex, NULL); + continue; + } + + /* swap two lower inode, and loop again */ + q = fidir->fd_hfile + new_bindex; + tmp = *q; + *q = *p; + *p = tmp; + if (tmp.hf_file) { + bindex--; + p--; + } + } + + p = fidir->fd_hfile; + if (!au_test_mmapped(file) && !d_unlinked(file->f_dentry)) { + bend = au_sbend(sb); + for (finfo->fi_btop = 0; finfo->fi_btop <= bend; + finfo->fi_btop++, p++) + if (p->hf_file) { + if (file_inode(p->hf_file)) + break; + au_hfput(p, file); + } + } else { + bend = au_br_index(sb, brid); + for (finfo->fi_btop = 0; finfo->fi_btop < bend; + finfo->fi_btop++, p++) + if (p->hf_file) + au_hfput(p, file); + bend = au_sbend(sb); + } + + p = fidir->fd_hfile + bend; + for (fidir->fd_bbot = bend; fidir->fd_bbot >= finfo->fi_btop; + fidir->fd_bbot--, p--) + if (p->hf_file) { + if (file_inode(p->hf_file)) + break; + au_hfput(p, file); + } + AuDebugOn(fidir->fd_bbot < finfo->fi_btop); +} + +/* + * after branch manipulating, refresh the file. + */ +static int refresh_file(struct file *file, int (*reopen)(struct file *file)) +{ + int err, need_reopen; + aufs_bindex_t bend, bindex; + struct dentry *dentry; + struct au_finfo *finfo; + struct au_hfile *hfile; + + dentry = file->f_dentry; + finfo = au_fi(file); + if (!finfo->fi_hdir) { + hfile = &finfo->fi_htop; + AuDebugOn(!hfile->hf_file); + bindex = au_br_index(dentry->d_sb, hfile->hf_br->br_id); + AuDebugOn(bindex < 0); + if (bindex != finfo->fi_btop) + au_set_fbstart(file, bindex); + } else { + err = au_fidir_realloc(finfo, au_sbend(dentry->d_sb) + 1); + if (unlikely(err)) + goto out; + au_do_refresh_dir(file); + } + + err = 0; + need_reopen = 1; + if (!au_test_mmapped(file)) + err = au_file_refresh_by_inode(file, &need_reopen); + if (!err && need_reopen && !d_unlinked(dentry)) + err = reopen(file); + if (!err) { + au_update_figen(file); + goto out; /* success */ + } + + /* error, close all lower files */ + if (finfo->fi_hdir) { + bend = au_fbend_dir(file); + for (bindex = au_fbstart(file); bindex <= bend; bindex++) + au_set_h_fptr(file, bindex, NULL); + } + +out: + return err; +} + +/* common function to regular file and dir */ +int au_reval_and_lock_fdi(struct file *file, int (*reopen)(struct file *file), + int wlock) +{ + int err; + unsigned int sigen, figen; + aufs_bindex_t bstart; + unsigned char pseudo_link; + struct dentry *dentry; + struct inode *inode; + + err = 0; + dentry = file->f_dentry; + inode = dentry->d_inode; + sigen = au_sigen(dentry->d_sb); + fi_write_lock(file); + figen = au_figen(file); + di_write_lock_child(dentry); + bstart = au_dbstart(dentry); + pseudo_link = (bstart != au_ibstart(inode)); + if (sigen == figen && !pseudo_link && au_fbstart(file) == bstart) { + if (!wlock) { + di_downgrade_lock(dentry, AuLock_IR); + fi_downgrade_lock(file); + } + goto out; /* success */ + } + + AuDbg("sigen %d, figen %d\n", sigen, figen); + if (au_digen_test(dentry, sigen)) { + err = au_reval_dpath(dentry, sigen); + AuDebugOn(!err && au_digen_test(dentry, sigen)); + } + + if (!err) + err = refresh_file(file, reopen); + if (!err) { + if (!wlock) { + di_downgrade_lock(dentry, AuLock_IR); + fi_downgrade_lock(file); + } + } else { + di_write_unlock(dentry); + fi_write_unlock(file); + } + +out: + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* cf. aufs_nopage() */ +/* for madvise(2) */ +static int aufs_readpage(struct file *file __maybe_unused, struct page *page) +{ + unlock_page(page); + return 0; +} + +/* it will never be called, but necessary to support O_DIRECT */ +static ssize_t aufs_direct_IO(int rw, struct kiocb *iocb, + const struct iovec *iov, loff_t offset, + unsigned long nr_segs) +{ BUG(); return 0; } + +/* + * it will never be called, but madvise and fadvise behaves differently + * when get_xip_mem is defined + */ +static int aufs_get_xip_mem(struct address_space *mapping, pgoff_t pgoff, + int create, void **kmem, unsigned long *pfn) +{ BUG(); return 0; } + +/* they will never be called. */ +#ifdef CONFIG_AUFS_DEBUG +static int aufs_write_begin(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned flags, + struct page **pagep, void **fsdata) +{ AuUnsupport(); return 0; } +static int aufs_write_end(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata) +{ AuUnsupport(); return 0; } +static int aufs_writepage(struct page *page, struct writeback_control *wbc) +{ AuUnsupport(); return 0; } + +static int aufs_set_page_dirty(struct page *page) +{ AuUnsupport(); return 0; } +static void aufs_invalidatepage(struct page *page, unsigned long offset) +{ AuUnsupport(); } +static int aufs_releasepage(struct page *page, gfp_t gfp) +{ AuUnsupport(); return 0; } +static int aufs_migratepage(struct address_space *mapping, struct page *newpage, + struct page *page, enum migrate_mode mode) +{ AuUnsupport(); return 0; } +static int aufs_launder_page(struct page *page) +{ AuUnsupport(); return 0; } +static int aufs_is_partially_uptodate(struct page *page, + read_descriptor_t *desc, + unsigned long from) +{ AuUnsupport(); return 0; } +static int aufs_error_remove_page(struct address_space *mapping, + struct page *page) +{ AuUnsupport(); return 0; } +static int aufs_swap_activate(struct swap_info_struct *sis, struct file *file, + sector_t *span) +{ AuUnsupport(); return 0; } +static void aufs_swap_deactivate(struct file *file) +{ AuUnsupport(); } +#endif /* CONFIG_AUFS_DEBUG */ + +const struct address_space_operations aufs_aop = { + .readpage = aufs_readpage, + .direct_IO = aufs_direct_IO, + .get_xip_mem = aufs_get_xip_mem, +#ifdef CONFIG_AUFS_DEBUG + .writepage = aufs_writepage, + /* no writepages, because of writepage */ + .set_page_dirty = aufs_set_page_dirty, + /* no readpages, because of readpage */ + .write_begin = aufs_write_begin, + .write_end = aufs_write_end, + /* no bmap, no block device */ + .invalidatepage = aufs_invalidatepage, + .releasepage = aufs_releasepage, + .migratepage = aufs_migratepage, + .launder_page = aufs_launder_page, + .is_partially_uptodate = aufs_is_partially_uptodate, + .error_remove_page = aufs_error_remove_page, + .swap_activate = aufs_swap_activate, + .swap_deactivate = aufs_swap_deactivate +#endif /* CONFIG_AUFS_DEBUG */ +}; diff -Naur kernel.21.3.orig/fs/aufs/file.h kernel.21.3.new/fs/aufs/file.h --- kernel.21.3.orig/fs/aufs/file.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/file.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,281 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * file operations + */ + +#ifndef __AUFS_FILE_H__ +#define __AUFS_FILE_H__ + +#ifdef __KERNEL__ + +#include +#include +#include +#include "rwsem.h" + +struct au_branch; +struct au_hfile { + struct file *hf_file; + struct au_branch *hf_br; +}; + +struct au_vdir; +struct au_fidir { + aufs_bindex_t fd_bbot; + aufs_bindex_t fd_nent; + struct au_vdir *fd_vdir_cache; + struct au_hfile fd_hfile[]; +}; + +static inline int au_fidir_sz(int nent) +{ + AuDebugOn(nent < 0); + return sizeof(struct au_fidir) + sizeof(struct au_hfile) * nent; +} + +struct au_finfo { + atomic_t fi_generation; + + struct au_rwsem fi_rwsem; + aufs_bindex_t fi_btop; + + /* do not union them */ + struct { /* for non-dir */ + struct au_hfile fi_htop; + atomic_t fi_mmapped; + }; + struct au_fidir *fi_hdir; /* for dir only */ +} ____cacheline_aligned_in_smp; + +/* ---------------------------------------------------------------------- */ + +/* file.c */ +extern const struct address_space_operations aufs_aop; +unsigned int au_file_roflags(unsigned int flags); +struct file *au_h_open(struct dentry *dentry, aufs_bindex_t bindex, int flags, + struct file *file, int force_wr); +int au_do_open(struct file *file, int (*open)(struct file *file, int flags), + struct au_fidir *fidir); +int au_reopen_nondir(struct file *file); +struct au_pin; +int au_ready_to_write(struct file *file, loff_t len, struct au_pin *pin); +int au_reval_and_lock_fdi(struct file *file, int (*reopen)(struct file *file), + int wlock); +int au_do_flush(struct file *file, fl_owner_t id, + int (*flush)(struct file *file, fl_owner_t id)); + +/* poll.c */ +#ifdef CONFIG_AUFS_POLL +unsigned int aufs_poll(struct file *file, poll_table *wait); +#endif + +#ifdef CONFIG_AUFS_BR_HFSPLUS +/* hfsplus.c */ +struct file *au_h_open_pre(struct dentry *dentry, aufs_bindex_t bindex, + int force_wr); +void au_h_open_post(struct dentry *dentry, aufs_bindex_t bindex, + struct file *h_file); +#else +AuStub(struct file *, au_h_open_pre, return NULL, struct dentry *dentry, + aufs_bindex_t bindex, int force_wr) +AuStubVoid(au_h_open_post, struct dentry *dentry, aufs_bindex_t bindex, + struct file *h_file); +#endif + +/* f_op.c */ +extern const struct file_operations aufs_file_fop; +int au_do_open_nondir(struct file *file, int flags); +int aufs_release_nondir(struct inode *inode __maybe_unused, struct file *file); + +/* finfo.c */ +void au_hfput(struct au_hfile *hf, struct file *file); +void au_set_h_fptr(struct file *file, aufs_bindex_t bindex, + struct file *h_file); + +void au_update_figen(struct file *file); +struct au_fidir *au_fidir_alloc(struct super_block *sb); +int au_fidir_realloc(struct au_finfo *finfo, int nbr); + +void au_fi_init_once(void *_fi); +void au_finfo_fin(struct file *file); +int au_finfo_init(struct file *file, struct au_fidir *fidir); + +/* ioctl.c */ +long aufs_ioctl_nondir(struct file *file, unsigned int cmd, unsigned long arg); +#ifdef CONFIG_COMPAT +long aufs_compat_ioctl_dir(struct file *file, unsigned int cmd, + unsigned long arg); +long aufs_compat_ioctl_nondir(struct file *file, unsigned int cmd, + unsigned long arg); +#endif + +/* ---------------------------------------------------------------------- */ + +static inline struct au_finfo *au_fi(struct file *file) +{ + return file->private_data; +} + +/* ---------------------------------------------------------------------- */ + +/* + * fi_read_lock, fi_write_lock, + * fi_read_unlock, fi_write_unlock, fi_downgrade_lock + */ +AuSimpleRwsemFuncs(fi, struct file *f, &au_fi(f)->fi_rwsem); + +#define FiMustNoWaiters(f) AuRwMustNoWaiters(&au_fi(f)->fi_rwsem) +#define FiMustAnyLock(f) AuRwMustAnyLock(&au_fi(f)->fi_rwsem) +#define FiMustWriteLock(f) AuRwMustWriteLock(&au_fi(f)->fi_rwsem) + +/* ---------------------------------------------------------------------- */ + +/* todo: hard/soft set? */ +static inline aufs_bindex_t au_fbstart(struct file *file) +{ + FiMustAnyLock(file); + return au_fi(file)->fi_btop; +} + +static inline aufs_bindex_t au_fbend_dir(struct file *file) +{ + FiMustAnyLock(file); + AuDebugOn(!au_fi(file)->fi_hdir); + return au_fi(file)->fi_hdir->fd_bbot; +} + +static inline struct au_vdir *au_fvdir_cache(struct file *file) +{ + FiMustAnyLock(file); + AuDebugOn(!au_fi(file)->fi_hdir); + return au_fi(file)->fi_hdir->fd_vdir_cache; +} + +static inline void au_set_fbstart(struct file *file, aufs_bindex_t bindex) +{ + FiMustWriteLock(file); + au_fi(file)->fi_btop = bindex; +} + +static inline void au_set_fbend_dir(struct file *file, aufs_bindex_t bindex) +{ + FiMustWriteLock(file); + AuDebugOn(!au_fi(file)->fi_hdir); + au_fi(file)->fi_hdir->fd_bbot = bindex; +} + +static inline void au_set_fvdir_cache(struct file *file, + struct au_vdir *vdir_cache) +{ + FiMustWriteLock(file); + AuDebugOn(!au_fi(file)->fi_hdir); + au_fi(file)->fi_hdir->fd_vdir_cache = vdir_cache; +} + +static inline struct file *au_hf_top(struct file *file) +{ + FiMustAnyLock(file); + AuDebugOn(au_fi(file)->fi_hdir); + return au_fi(file)->fi_htop.hf_file; +} + +static inline struct file *au_hf_dir(struct file *file, aufs_bindex_t bindex) +{ + FiMustAnyLock(file); + AuDebugOn(!au_fi(file)->fi_hdir); + return au_fi(file)->fi_hdir->fd_hfile[0 + bindex].hf_file; +} + +/* todo: memory barrier? */ +static inline unsigned int au_figen(struct file *f) +{ + return atomic_read(&au_fi(f)->fi_generation); +} + +static inline void au_set_mmapped(struct file *f) +{ + if (atomic_inc_return(&au_fi(f)->fi_mmapped)) + return; + pr_warn("fi_mmapped wrapped around\n"); + while (!atomic_inc_return(&au_fi(f)->fi_mmapped)) + ; +} + +static inline void au_unset_mmapped(struct file *f) +{ + atomic_dec(&au_fi(f)->fi_mmapped); +} + +static inline int au_test_mmapped(struct file *f) +{ + return atomic_read(&au_fi(f)->fi_mmapped); +} + +/* customize vma->vm_file */ + +static inline void au_do_vm_file_reset(struct vm_area_struct *vma, + struct file *file) +{ + struct file *f; + + f = vma->vm_file; + get_file(file); + vma->vm_file = file; + fput(f); +} + +#ifdef CONFIG_MMU +#define AuDbgVmRegion(file, vma) do {} while (0) + +static inline void au_vm_file_reset(struct vm_area_struct *vma, + struct file *file) +{ + au_do_vm_file_reset(vma, file); +} +#else +#define AuDbgVmRegion(file, vma) \ + AuDebugOn((vma)->vm_region && (vma)->vm_region->vm_file != (file)) + +static inline void au_vm_file_reset(struct vm_area_struct *vma, + struct file *file) +{ + struct file *f; + + au_do_vm_file_reset(vma, file); + f = vma->vm_region->vm_file; + get_file(file); + vma->vm_region->vm_file = file; + fput(f); +} +#endif /* CONFIG_MMU */ + +/* handle vma->vm_prfile */ +static inline void au_vm_prfile_set(struct vm_area_struct *vma, + struct file *file) +{ + get_file(file); + vma->vm_prfile = file; +#ifndef CONFIG_MMU + get_file(file); + vma->vm_region->vm_prfile = file; +#endif +} + +#endif /* __KERNEL__ */ +#endif /* __AUFS_FILE_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/finfo.c kernel.21.3.new/fs/aufs/finfo.c --- kernel.21.3.orig/fs/aufs/finfo.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/finfo.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,156 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * file private data + */ + +#include "aufs.h" + +void au_hfput(struct au_hfile *hf, struct file *file) +{ + /* todo: direct access f_flags */ + if (vfsub_file_flags(file) & __FMODE_EXEC) + allow_write_access(hf->hf_file); + fput(hf->hf_file); + hf->hf_file = NULL; + atomic_dec(&hf->hf_br->br_count); + hf->hf_br = NULL; +} + +void au_set_h_fptr(struct file *file, aufs_bindex_t bindex, struct file *val) +{ + struct au_finfo *finfo = au_fi(file); + struct au_hfile *hf; + struct au_fidir *fidir; + + fidir = finfo->fi_hdir; + if (!fidir) { + AuDebugOn(finfo->fi_btop != bindex); + hf = &finfo->fi_htop; + } else + hf = fidir->fd_hfile + bindex; + + if (hf && hf->hf_file) + au_hfput(hf, file); + if (val) { + FiMustWriteLock(file); + hf->hf_file = val; + hf->hf_br = au_sbr(file->f_dentry->d_sb, bindex); + } +} + +void au_update_figen(struct file *file) +{ + atomic_set(&au_fi(file)->fi_generation, au_digen(file->f_dentry)); + /* smp_mb(); */ /* atomic_set */ +} + +/* ---------------------------------------------------------------------- */ + +struct au_fidir *au_fidir_alloc(struct super_block *sb) +{ + struct au_fidir *fidir; + int nbr; + + nbr = au_sbend(sb) + 1; + if (nbr < 2) + nbr = 2; /* initial allocate for 2 branches */ + fidir = kzalloc(au_fidir_sz(nbr), GFP_NOFS); + if (fidir) { + fidir->fd_bbot = -1; + fidir->fd_nent = nbr; + fidir->fd_vdir_cache = NULL; + } + + return fidir; +} + +int au_fidir_realloc(struct au_finfo *finfo, int nbr) +{ + int err; + struct au_fidir *fidir, *p; + + AuRwMustWriteLock(&finfo->fi_rwsem); + fidir = finfo->fi_hdir; + AuDebugOn(!fidir); + + err = -ENOMEM; + p = au_kzrealloc(fidir, au_fidir_sz(fidir->fd_nent), au_fidir_sz(nbr), + GFP_NOFS); + if (p) { + p->fd_nent = nbr; + finfo->fi_hdir = p; + err = 0; + } + + return err; +} + +/* ---------------------------------------------------------------------- */ + +void au_finfo_fin(struct file *file) +{ + struct au_finfo *finfo; + + au_nfiles_dec(file->f_dentry->d_sb); + + finfo = au_fi(file); + AuDebugOn(finfo->fi_hdir); + AuRwDestroy(&finfo->fi_rwsem); + au_cache_free_finfo(finfo); +} + +void au_fi_init_once(void *_finfo) +{ + struct au_finfo *finfo = _finfo; + static struct lock_class_key aufs_fi; + + au_rw_init(&finfo->fi_rwsem); + au_rw_class(&finfo->fi_rwsem, &aufs_fi); +} + +int au_finfo_init(struct file *file, struct au_fidir *fidir) +{ + int err; + struct au_finfo *finfo; + struct dentry *dentry; + + err = -ENOMEM; + dentry = file->f_dentry; + finfo = au_cache_alloc_finfo(); + if (unlikely(!finfo)) + goto out; + + err = 0; + au_nfiles_inc(dentry->d_sb); + /* verbose coding for lock class name */ + if (!fidir) + au_rw_class(&finfo->fi_rwsem, au_lc_key + AuLcNonDir_FIINFO); + else + au_rw_class(&finfo->fi_rwsem, au_lc_key + AuLcDir_FIINFO); + au_rw_write_lock(&finfo->fi_rwsem); + finfo->fi_btop = -1; + finfo->fi_hdir = fidir; + atomic_set(&finfo->fi_generation, au_digen(dentry)); + /* smp_mb(); */ /* atomic_set */ + + file->private_data = finfo; + +out: + return err; +} diff -Naur kernel.21.3.orig/fs/aufs/f_op.c kernel.21.3.new/fs/aufs/f_op.c --- kernel.21.3.orig/fs/aufs/f_op.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/f_op.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,796 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * file and vm operations + */ + +#include +#include +#include +#include +#include "aufs.h" + +int au_do_open_nondir(struct file *file, int flags) +{ + int err; + aufs_bindex_t bindex; + struct file *h_file; + struct dentry *dentry; + struct au_finfo *finfo; + + FiMustWriteLock(file); + + err = 0; + dentry = file->f_dentry; + finfo = au_fi(file); + memset(&finfo->fi_htop, 0, sizeof(finfo->fi_htop)); + atomic_set(&finfo->fi_mmapped, 0); + bindex = au_dbstart(dentry); + h_file = au_h_open(dentry, bindex, flags, file, /*force_wr*/0); + if (IS_ERR(h_file)) + err = PTR_ERR(h_file); + else { + au_set_fbstart(file, bindex); + au_set_h_fptr(file, bindex, h_file); + au_update_figen(file); + /* todo: necessary? */ + /* file->f_ra = h_file->f_ra; */ + } + + return err; +} + +static int aufs_open_nondir(struct inode *inode __maybe_unused, + struct file *file) +{ + int err; + struct super_block *sb; + + AuDbg("%.*s, f_flags 0x%x, f_mode 0x%x\n", + AuDLNPair(file->f_dentry), vfsub_file_flags(file), + file->f_mode); + + sb = file->f_dentry->d_sb; + si_read_lock(sb, AuLock_FLUSH); + err = au_do_open(file, au_do_open_nondir, /*fidir*/NULL); + si_read_unlock(sb); + return err; +} + +int aufs_release_nondir(struct inode *inode __maybe_unused, struct file *file) +{ + struct au_finfo *finfo; + aufs_bindex_t bindex; + + finfo = au_fi(file); + bindex = finfo->fi_btop; + if (bindex >= 0) + au_set_h_fptr(file, bindex, NULL); + + au_finfo_fin(file); + return 0; +} + +/* ---------------------------------------------------------------------- */ + +static int au_do_flush_nondir(struct file *file, fl_owner_t id) +{ + int err; + struct file *h_file; + + err = 0; + h_file = au_hf_top(file); + if (h_file) + err = vfsub_flush(h_file, id); + return err; +} + +static int aufs_flush_nondir(struct file *file, fl_owner_t id) +{ + return au_do_flush(file, id, au_do_flush_nondir); +} + +/* ---------------------------------------------------------------------- */ +/* + * read and write functions acquire [fdi]_rwsem once, but release before + * mmap_sem. This is because to stop a race condition between mmap(2). + * Releasing these aufs-rwsem should be safe, no branch-mamagement (by keeping + * si_rwsem), no harmful copy-up should happen. Actually copy-up may happen in + * read functions after [fdi]_rwsem are released, but it should be harmless. + */ + +static ssize_t aufs_read(struct file *file, char __user *buf, size_t count, + loff_t *ppos) +{ + ssize_t err; + struct dentry *dentry; + struct file *h_file; + struct super_block *sb; + + dentry = file->f_dentry; + sb = dentry->d_sb; + si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW); + err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0); + if (unlikely(err)) + goto out; + + h_file = au_hf_top(file); + get_file(h_file); + di_read_unlock(dentry, AuLock_IR); + fi_read_unlock(file); + + /* filedata may be obsoleted by concurrent copyup, but no problem */ + err = vfsub_read_u(h_file, buf, count, ppos); + /* todo: necessary? */ + /* file->f_ra = h_file->f_ra; */ + /* update without lock, I don't think it a problem */ + fsstack_copy_attr_atime(dentry->d_inode, file_inode(h_file)); + fput(h_file); + +out: + si_read_unlock(sb); + return err; +} + +/* + * todo: very ugly + * it locks both of i_mutex and si_rwsem for read in safe. + * if the plink maintenance mode continues forever (that is the problem), + * may loop forever. + */ +static void au_mtx_and_read_lock(struct inode *inode) +{ + int err; + struct super_block *sb = inode->i_sb; + + while (1) { + mutex_lock(&inode->i_mutex); + err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM); + if (!err) + break; + mutex_unlock(&inode->i_mutex); + si_read_lock(sb, AuLock_NOPLMW); + si_read_unlock(sb); + } +} + +static ssize_t aufs_write(struct file *file, const char __user *ubuf, + size_t count, loff_t *ppos) +{ + ssize_t err; + blkcnt_t blks; + aufs_bindex_t bstart; + struct au_pin pin; + struct dentry *dentry; + struct inode *inode, *h_inode; + struct super_block *sb; + struct file *h_file; + char __user *buf = (char __user *)ubuf; + + dentry = file->f_dentry; + sb = dentry->d_sb; + inode = dentry->d_inode; + au_mtx_and_read_lock(inode); + + err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1); + if (unlikely(err)) + goto out; + + err = au_ready_to_write(file, -1, &pin); + di_downgrade_lock(dentry, AuLock_IR); + if (unlikely(err)) { + di_read_unlock(dentry, AuLock_IR); + fi_write_unlock(file); + goto out; + } + + bstart = au_fbstart(file); + h_file = au_hf_top(file); + get_file(h_file); + h_inode = h_file->f_dentry->d_inode; + blks = h_inode->i_blocks; + au_unpin(&pin); + di_read_unlock(dentry, AuLock_IR); + fi_write_unlock(file); + + err = vfsub_write_u(h_file, buf, count, ppos); + ii_write_lock_child(inode); + au_cpup_attr_timesizes(inode); + inode->i_mode = file_inode(h_file)->i_mode; + AuDbg("blks %llu, %llu\n", (u64)blks, (u64)h_inode->i_blocks); + if (err > 0) + au_fhsm_wrote(sb, bstart, /*force*/h_inode->i_blocks > blks); + ii_write_unlock(inode); + fput(h_file); + +out: + si_read_unlock(sb); + mutex_unlock(&inode->i_mutex); + return err; +} + +static ssize_t au_do_aio(struct file *h_file, int rw, struct kiocb *kio, + const struct iovec *iov, unsigned long nv, loff_t pos) +{ + ssize_t err; + struct file *file; + ssize_t (*func)(struct kiocb *, const struct iovec *, unsigned long, + loff_t); + + err = security_file_permission(h_file, rw); + if (unlikely(err)) + goto out; + + err = -ENOSYS; + func = NULL; + if (rw == MAY_READ) + func = h_file->f_op->aio_read; + else if (rw == MAY_WRITE) + func = h_file->f_op->aio_write; + if (func) { + file = kio->ki_filp; + kio->ki_filp = h_file; + lockdep_off(); + err = func(kio, iov, nv, pos); + lockdep_on(); + kio->ki_filp = file; + } else + /* currently there is no such fs */ + WARN_ON_ONCE(1); + +out: + return err; +} + +static ssize_t aufs_aio_read(struct kiocb *kio, const struct iovec *iov, + unsigned long nv, loff_t pos) +{ + ssize_t err; + struct file *file, *h_file; + struct dentry *dentry; + struct super_block *sb; + + file = kio->ki_filp; + dentry = file->f_dentry; + sb = dentry->d_sb; + si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW); + err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0); + if (unlikely(err)) + goto out; + + h_file = au_hf_top(file); + get_file(h_file); + di_read_unlock(dentry, AuLock_IR); + fi_read_unlock(file); + + err = au_do_aio(h_file, MAY_READ, kio, iov, nv, pos); + /* todo: necessary? */ + /* file->f_ra = h_file->f_ra; */ + /* update without lock, I don't think it a problem */ + fsstack_copy_attr_atime(dentry->d_inode, file_inode(h_file)); + fput(h_file); + +out: + si_read_unlock(sb); + return err; +} + +static ssize_t aufs_aio_write(struct kiocb *kio, const struct iovec *iov, + unsigned long nv, loff_t pos) +{ + ssize_t err; + blkcnt_t blks; + aufs_bindex_t bstart; + struct au_pin pin; + struct dentry *dentry; + struct inode *inode, *h_inode; + struct file *file, *h_file; + struct super_block *sb; + + file = kio->ki_filp; + dentry = file->f_dentry; + sb = dentry->d_sb; + inode = dentry->d_inode; + au_mtx_and_read_lock(inode); + + err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1); + if (unlikely(err)) + goto out; + + err = au_ready_to_write(file, -1, &pin); + di_downgrade_lock(dentry, AuLock_IR); + if (unlikely(err)) { + di_read_unlock(dentry, AuLock_IR); + fi_write_unlock(file); + goto out; + } + + bstart = au_fbstart(file); + h_file = au_hf_top(file); + get_file(h_file); + h_inode = h_file->f_dentry->d_inode; + blks = h_inode->i_blocks; + au_unpin(&pin); + di_read_unlock(dentry, AuLock_IR); + fi_write_unlock(file); + + err = au_do_aio(h_file, MAY_WRITE, kio, iov, nv, pos); + ii_write_lock_child(inode); + au_cpup_attr_timesizes(inode); + inode->i_mode = file_inode(h_file)->i_mode; + AuDbg("blks %llu, %llu\n", (u64)blks, (u64)h_inode->i_blocks); + if (err > 0) + au_fhsm_wrote(sb, bstart, /*force*/h_inode->i_blocks > blks); + ii_write_unlock(inode); + fput(h_file); + +out: + si_read_unlock(sb); + mutex_unlock(&inode->i_mutex); + return err; +} + +static ssize_t aufs_splice_read(struct file *file, loff_t *ppos, + struct pipe_inode_info *pipe, size_t len, + unsigned int flags) +{ + ssize_t err; + struct file *h_file; + struct dentry *dentry; + struct super_block *sb; + + dentry = file->f_dentry; + sb = dentry->d_sb; + si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW); + err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0); + if (unlikely(err)) + goto out; + + err = -EINVAL; + h_file = au_hf_top(file); + get_file(h_file); + if (au_test_loopback_kthread()) { + au_warn_loopback(h_file->f_dentry->d_sb); + if (file->f_mapping != h_file->f_mapping) { + file->f_mapping = h_file->f_mapping; + smp_mb(); /* unnecessary? */ + } + } + di_read_unlock(dentry, AuLock_IR); + fi_read_unlock(file); + + err = vfsub_splice_to(h_file, ppos, pipe, len, flags); + /* todo: necessasry? */ + /* file->f_ra = h_file->f_ra; */ + /* update without lock, I don't think it a problem */ + fsstack_copy_attr_atime(dentry->d_inode, file_inode(h_file)); + fput(h_file); + +out: + si_read_unlock(sb); + return err; +} + +static ssize_t +aufs_splice_write(struct pipe_inode_info *pipe, struct file *file, loff_t *ppos, + size_t len, unsigned int flags) +{ + ssize_t err; + blkcnt_t blks; + aufs_bindex_t bstart; + struct au_pin pin; + struct dentry *dentry; + struct inode *inode, *h_inode; + struct super_block *sb; + struct file *h_file; + + dentry = file->f_dentry; + sb = dentry->d_sb; + inode = dentry->d_inode; + au_mtx_and_read_lock(inode); + + err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1); + if (unlikely(err)) + goto out; + + err = au_ready_to_write(file, -1, &pin); + di_downgrade_lock(dentry, AuLock_IR); + if (unlikely(err)) { + di_read_unlock(dentry, AuLock_IR); + fi_write_unlock(file); + goto out; + } + + bstart = au_fbstart(file); + h_file = au_hf_top(file); + get_file(h_file); + h_inode = h_file->f_dentry->d_inode; + blks = h_inode->i_blocks; + au_unpin(&pin); + di_read_unlock(dentry, AuLock_IR); + fi_write_unlock(file); + + err = vfsub_splice_from(pipe, h_file, ppos, len, flags); + ii_write_lock_child(inode); + au_cpup_attr_timesizes(inode); + inode->i_mode = file_inode(h_file)->i_mode; + AuDbg("blks %llu, %llu\n", (u64)blks, (u64)h_inode->i_blocks); + if (err > 0) + au_fhsm_wrote(sb, bstart, /*force*/h_inode->i_blocks > blks); + ii_write_unlock(inode); + fput(h_file); + +out: + si_read_unlock(sb); + mutex_unlock(&inode->i_mutex); + return err; +} + +static long aufs_fallocate(struct file *file, int mode, loff_t offset, + loff_t len) +{ + long err; + struct au_pin pin; + struct dentry *dentry; + struct super_block *sb; + struct inode *inode; + struct file *h_file; + + dentry = file->f_dentry; + sb = dentry->d_sb; + inode = dentry->d_inode; + au_mtx_and_read_lock(inode); + + err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1); + if (unlikely(err)) + goto out; + + err = au_ready_to_write(file, -1, &pin); + di_downgrade_lock(dentry, AuLock_IR); + if (unlikely(err)) { + di_read_unlock(dentry, AuLock_IR); + fi_write_unlock(file); + goto out; + } + + h_file = au_hf_top(file); + get_file(h_file); + au_unpin(&pin); + di_read_unlock(dentry, AuLock_IR); + fi_write_unlock(file); + + lockdep_off(); + err = do_fallocate(h_file, mode, offset, len); + lockdep_on(); + ii_write_lock_child(inode); + au_cpup_attr_timesizes(inode); + inode->i_mode = file_inode(h_file)->i_mode; + ii_write_unlock(inode); + fput(h_file); + +out: + si_read_unlock(sb); + mutex_unlock(&inode->i_mutex); + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* + * The locking order around current->mmap_sem. + * - in most and regular cases + * file I/O syscall -- aufs_read() or something + * -- si_rwsem for read -- mmap_sem + * (Note that [fdi]i_rwsem are released before mmap_sem). + * - in mmap case + * mmap(2) -- mmap_sem -- aufs_mmap() -- si_rwsem for read -- [fdi]i_rwsem + * This AB-BA order is definitly bad, but is not a problem since "si_rwsem for + * read" allows muliple processes to acquire it and [fdi]i_rwsem are not held in + * file I/O. Aufs needs to stop lockdep in aufs_mmap() though. + * It means that when aufs acquires si_rwsem for write, the process should never + * acquire mmap_sem. + * + * Actually aufs_readdir() holds [fdi]i_rwsem before mmap_sem, but this is not a + * problem either since any directory is not able to be mmap-ed. + * The similar scenario is applied to aufs_readlink() too. + */ + +#if 0 /* stop calling security_file_mmap() */ +/* cf. linux/include/linux/mman.h: calc_vm_prot_bits() */ +#define AuConv_VM_PROT(f, b) _calc_vm_trans(f, VM_##b, PROT_##b) + +static unsigned long au_arch_prot_conv(unsigned long flags) +{ + /* currently ppc64 only */ +#ifdef CONFIG_PPC64 + /* cf. linux/arch/powerpc/include/asm/mman.h */ + AuDebugOn(arch_calc_vm_prot_bits(-1) != VM_SAO); + return AuConv_VM_PROT(flags, SAO); +#else + AuDebugOn(arch_calc_vm_prot_bits(-1)); + return 0; +#endif +} + +static unsigned long au_prot_conv(unsigned long flags) +{ + return AuConv_VM_PROT(flags, READ) + | AuConv_VM_PROT(flags, WRITE) + | AuConv_VM_PROT(flags, EXEC) + | au_arch_prot_conv(flags); +} + +/* cf. linux/include/linux/mman.h: calc_vm_flag_bits() */ +#define AuConv_VM_MAP(f, b) _calc_vm_trans(f, VM_##b, MAP_##b) + +static unsigned long au_flag_conv(unsigned long flags) +{ + return AuConv_VM_MAP(flags, GROWSDOWN) + | AuConv_VM_MAP(flags, DENYWRITE) + | AuConv_VM_MAP(flags, LOCKED); +} +#endif + +static int aufs_mmap(struct file *file, struct vm_area_struct *vma) +{ + int err; + aufs_bindex_t bstart; + const unsigned char wlock + = (file->f_mode & FMODE_WRITE) && (vma->vm_flags & VM_SHARED); + struct dentry *dentry; + struct super_block *sb; + struct file *h_file; + struct au_branch *br; + struct au_pin pin; + + AuDbgVmRegion(file, vma); + + dentry = file->f_dentry; + sb = dentry->d_sb; + lockdep_off(); + si_read_lock(sb, AuLock_NOPLMW); + err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1); + if (unlikely(err)) + goto out; + + if (wlock) { + err = au_ready_to_write(file, -1, &pin); + di_write_unlock(dentry); + if (unlikely(err)) { + fi_write_unlock(file); + goto out; + } + au_unpin(&pin); + } else + di_write_unlock(dentry); + + bstart = au_fbstart(file); + br = au_sbr(sb, bstart); + h_file = au_hf_top(file); + get_file(h_file); + au_set_mmapped(file); + fi_write_unlock(file); + lockdep_on(); + + au_vm_file_reset(vma, h_file); + /* + * we cannot call security_mmap_file() here since it may acquire + * mmap_sem or i_mutex. + * + * err = security_mmap_file(h_file, au_prot_conv(vma->vm_flags), + * au_flag_conv(vma->vm_flags)); + */ + if (!err) + err = h_file->f_op->mmap(h_file, vma); + if (unlikely(err)) + goto out_reset; + + au_vm_prfile_set(vma, file); + /* update without lock, I don't think it a problem */ + fsstack_copy_attr_atime(file_inode(file), file_inode(h_file)); + goto out_fput; /* success */ + +out_reset: + au_unset_mmapped(file); + au_vm_file_reset(vma, file); +out_fput: + fput(h_file); + lockdep_off(); +out: + si_read_unlock(sb); + lockdep_on(); + AuTraceErr(err); + return err; +} + +/* ---------------------------------------------------------------------- */ + +static int aufs_fsync_nondir(struct file *file, loff_t start, loff_t end, + int datasync) +{ + int err; + struct au_pin pin; + struct dentry *dentry; + struct inode *inode; + struct file *h_file; + struct super_block *sb; + + dentry = file->f_dentry; + inode = dentry->d_inode; + sb = dentry->d_sb; + mutex_lock(&inode->i_mutex); + err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM); + if (unlikely(err)) + goto out; + + err = 0; /* -EBADF; */ /* posix? */ + if (unlikely(!(file->f_mode & FMODE_WRITE))) + goto out_si; + err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1); + if (unlikely(err)) + goto out_si; + + err = au_ready_to_write(file, -1, &pin); + di_downgrade_lock(dentry, AuLock_IR); + if (unlikely(err)) + goto out_unlock; + au_unpin(&pin); + + err = -EINVAL; + h_file = au_hf_top(file); + err = vfsub_fsync(h_file, &h_file->f_path, datasync); + au_cpup_attr_timesizes(inode); + +out_unlock: + di_read_unlock(dentry, AuLock_IR); + fi_write_unlock(file); +out_si: + si_read_unlock(sb); +out: + mutex_unlock(&inode->i_mutex); + return err; +} + +/* no one supports this operation, currently */ +#if 0 +static int aufs_aio_fsync_nondir(struct kiocb *kio, int datasync) +{ + int err; + struct au_pin pin; + struct dentry *dentry; + struct inode *inode; + struct file *file, *h_file; + + file = kio->ki_filp; + dentry = file->f_dentry; + inode = dentry->d_inode; + au_mtx_and_read_lock(inode); + + err = 0; /* -EBADF; */ /* posix? */ + if (unlikely(!(file->f_mode & FMODE_WRITE))) + goto out; + err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1); + if (unlikely(err)) + goto out; + + err = au_ready_to_write(file, -1, &pin); + di_downgrade_lock(dentry, AuLock_IR); + if (unlikely(err)) + goto out_unlock; + au_unpin(&pin); + + err = -ENOSYS; + h_file = au_hf_top(file); + if (h_file->f_op && h_file->f_op->aio_fsync) { + struct mutex *h_mtx; + + h_mtx = &file_inode(h_file)->i_mutex; + if (!is_sync_kiocb(kio)) { + get_file(h_file); + fput(file); + } + kio->ki_filp = h_file; + err = h_file->f_op->aio_fsync(kio, datasync); + mutex_lock_nested(h_mtx, AuLsc_I_CHILD); + if (!err) + vfsub_update_h_iattr(&h_file->f_path, /*did*/NULL); + /*ignore*/ + au_cpup_attr_timesizes(inode); + mutex_unlock(h_mtx); + } + +out_unlock: + di_read_unlock(dentry, AuLock_IR); + fi_write_unlock(file); +out: + si_read_unlock(inode->sb); + mutex_unlock(&inode->i_mutex); + return err; +} +#endif + +static int aufs_fasync(int fd, struct file *file, int flag) +{ + int err; + struct file *h_file; + struct dentry *dentry; + struct super_block *sb; + + dentry = file->f_dentry; + sb = dentry->d_sb; + si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW); + err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0); + if (unlikely(err)) + goto out; + + h_file = au_hf_top(file); + if (h_file->f_op && h_file->f_op->fasync) + err = h_file->f_op->fasync(fd, h_file, flag); + + di_read_unlock(dentry, AuLock_IR); + fi_read_unlock(file); + +out: + si_read_unlock(sb); + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* no one supports this operation, currently */ +#if 0 +static ssize_t aufs_sendpage(struct file *file, struct page *page, int offset, + size_t len, loff_t *pos , int more) +{ +} +#endif + +/* ---------------------------------------------------------------------- */ + +const struct file_operations aufs_file_fop = { + .owner = THIS_MODULE, + + .llseek = default_llseek, + + .read = aufs_read, + .write = aufs_write, + .aio_read = aufs_aio_read, + .aio_write = aufs_aio_write, +#ifdef CONFIG_AUFS_POLL + .poll = aufs_poll, +#endif + .unlocked_ioctl = aufs_ioctl_nondir, +#ifdef CONFIG_COMPAT + .compat_ioctl = aufs_compat_ioctl_nondir, +#endif + .mmap = aufs_mmap, + .open = aufs_open_nondir, + .flush = aufs_flush_nondir, + .release = aufs_release_nondir, + .fsync = aufs_fsync_nondir, + /* .aio_fsync = aufs_aio_fsync_nondir, */ + .fasync = aufs_fasync, + /* .sendpage = aufs_sendpage, */ + .splice_write = aufs_splice_write, + .splice_read = aufs_splice_read, +#if 0 + .aio_splice_write = aufs_aio_splice_write, + .aio_splice_read = aufs_aio_splice_read, +#endif + .fallocate = aufs_fallocate +}; diff -Naur kernel.21.3.orig/fs/aufs/fstype.h kernel.21.3.new/fs/aufs/fstype.h --- kernel.21.3.orig/fs/aufs/fstype.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/fstype.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,469 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * judging filesystem type + */ + +#ifndef __AUFS_FSTYPE_H__ +#define __AUFS_FSTYPE_H__ + +#ifdef __KERNEL__ + +#include +#include +#include + +static inline int au_test_aufs(struct super_block *sb) +{ + return sb->s_magic == AUFS_SUPER_MAGIC; +} + +static inline const char *au_sbtype(struct super_block *sb) +{ + return sb->s_type->name; +} + +static inline int au_test_iso9660(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_ROMFS_FS) || defined(CONFIG_ROMFS_FS_MODULE) + return sb->s_magic == ROMFS_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_romfs(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_ISO9660_FS) || defined(CONFIG_ISO9660_FS_MODULE) + return sb->s_magic == ISOFS_SUPER_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_cramfs(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_CRAMFS) || defined(CONFIG_CRAMFS_MODULE) + return sb->s_magic == CRAMFS_MAGIC; +#endif + return 0; +} + +static inline int au_test_nfs(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_NFS_FS) || defined(CONFIG_NFS_FS_MODULE) + return sb->s_magic == NFS_SUPER_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_fuse(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_FUSE_FS) || defined(CONFIG_FUSE_FS_MODULE) + return sb->s_magic == FUSE_SUPER_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_xfs(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_XFS_FS) || defined(CONFIG_XFS_FS_MODULE) + return sb->s_magic == XFS_SB_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_tmpfs(struct super_block *sb __maybe_unused) +{ +#ifdef CONFIG_TMPFS + return sb->s_magic == TMPFS_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_ecryptfs(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_ECRYPT_FS) || defined(CONFIG_ECRYPT_FS_MODULE) + return !strcmp(au_sbtype(sb), "ecryptfs"); +#else + return 0; +#endif +} + +static inline int au_test_ocfs2(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_OCFS2_FS) || defined(CONFIG_OCFS2_FS_MODULE) + return sb->s_magic == OCFS2_SUPER_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_ocfs2_dlmfs(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_OCFS2_FS_O2CB) || defined(CONFIG_OCFS2_FS_O2CB_MODULE) + return sb->s_magic == DLMFS_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_coda(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_CODA_FS) || defined(CONFIG_CODA_FS_MODULE) + return sb->s_magic == CODA_SUPER_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_v9fs(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_9P_FS) || defined(CONFIG_9P_FS_MODULE) + return sb->s_magic == V9FS_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_ext4(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_EXT4_FS) || defined(CONFIG_EXT4_FS_MODULE) + return sb->s_magic == EXT4_SUPER_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_sysv(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_SYSV_FS) || defined(CONFIG_SYSV_FS_MODULE) + return !strcmp(au_sbtype(sb), "sysv"); +#else + return 0; +#endif +} + +static inline int au_test_ramfs(struct super_block *sb) +{ + return sb->s_magic == RAMFS_MAGIC; +} + +static inline int au_test_ubifs(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_UBIFS_FS) || defined(CONFIG_UBIFS_FS_MODULE) + return sb->s_magic == UBIFS_SUPER_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_procfs(struct super_block *sb __maybe_unused) +{ +#ifdef CONFIG_PROC_FS + return sb->s_magic == PROC_SUPER_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_sysfs(struct super_block *sb __maybe_unused) +{ +#ifdef CONFIG_SYSFS + return sb->s_magic == SYSFS_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_configfs(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_CONFIGFS_FS) || defined(CONFIG_CONFIGFS_FS_MODULE) + return sb->s_magic == CONFIGFS_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_minix(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_MINIX_FS) || defined(CONFIG_MINIX_FS_MODULE) + return sb->s_magic == MINIX3_SUPER_MAGIC + || sb->s_magic == MINIX2_SUPER_MAGIC + || sb->s_magic == MINIX2_SUPER_MAGIC2 + || sb->s_magic == MINIX_SUPER_MAGIC + || sb->s_magic == MINIX_SUPER_MAGIC2; +#else + return 0; +#endif +} + +static inline int au_test_cifs(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_CIFS_FS) || defined(CONFIGCIFS_FS_MODULE) + return sb->s_magic == CIFS_MAGIC_NUMBER; +#else + return 0; +#endif +} + +static inline int au_test_fat(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_FAT_FS) || defined(CONFIG_FAT_FS_MODULE) + return sb->s_magic == MSDOS_SUPER_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_msdos(struct super_block *sb) +{ + return au_test_fat(sb); +} + +static inline int au_test_vfat(struct super_block *sb) +{ + return au_test_fat(sb); +} + +static inline int au_test_securityfs(struct super_block *sb __maybe_unused) +{ +#ifdef CONFIG_SECURITYFS + return sb->s_magic == SECURITYFS_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_squashfs(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_SQUASHFS) || defined(CONFIG_SQUASHFS_MODULE) + return sb->s_magic == SQUASHFS_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_btrfs(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_BTRFS_FS) || defined(CONFIG_BTRFS_FS_MODULE) + return sb->s_magic == BTRFS_SUPER_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_xenfs(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_XENFS) || defined(CONFIG_XENFS_MODULE) + return sb->s_magic == XENFS_SUPER_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_debugfs(struct super_block *sb __maybe_unused) +{ +#ifdef CONFIG_DEBUG_FS + return sb->s_magic == DEBUGFS_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_nilfs(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_NILFS) || defined(CONFIG_NILFS_MODULE) + return sb->s_magic == NILFS_SUPER_MAGIC; +#else + return 0; +#endif +} + +static inline int au_test_hfsplus(struct super_block *sb __maybe_unused) +{ +#if defined(CONFIG_HFSPLUS_FS) || defined(CONFIG_HFSPLUS_FS_MODULE) + return sb->s_magic == HFSPLUS_SUPER_MAGIC; +#else + return 0; +#endif +} + +/* ---------------------------------------------------------------------- */ +/* + * they can't be an aufs branch. + */ +static inline int au_test_fs_unsuppoted(struct super_block *sb) +{ + return +#ifndef CONFIG_AUFS_BR_RAMFS + au_test_ramfs(sb) || +#endif + au_test_procfs(sb) + || au_test_sysfs(sb) + || au_test_configfs(sb) + || au_test_debugfs(sb) + || au_test_securityfs(sb) + || au_test_xenfs(sb) + || au_test_ecryptfs(sb) + /* || !strcmp(au_sbtype(sb), "unionfs") */ + || au_test_aufs(sb); /* will be supported in next version */ +} + +static inline int au_test_fs_remote(struct super_block *sb) +{ + return !au_test_tmpfs(sb) +#ifdef CONFIG_AUFS_BR_RAMFS + && !au_test_ramfs(sb) +#endif + && !(sb->s_type->fs_flags & FS_REQUIRES_DEV); +} + +/* ---------------------------------------------------------------------- */ + +/* + * Note: these functions (below) are created after reading ->getattr() in all + * filesystems under linux/fs. it means we have to do so in every update... + */ + +/* + * some filesystems require getattr to refresh the inode attributes before + * referencing. + * in most cases, we can rely on the inode attribute in NFS (or every remote fs) + * and leave the work for d_revalidate() + */ +static inline int au_test_fs_refresh_iattr(struct super_block *sb) +{ + return au_test_nfs(sb) + || au_test_fuse(sb) + /* || au_test_ocfs2(sb) */ /* untested */ + /* || au_test_btrfs(sb) */ /* untested */ + /* || au_test_coda(sb) */ /* untested */ + /* || au_test_v9fs(sb) */ /* untested */ + ; +} + +/* + * filesystems which don't maintain i_size or i_blocks. + */ +static inline int au_test_fs_bad_iattr_size(struct super_block *sb) +{ + return au_test_xfs(sb) + || au_test_btrfs(sb) + || au_test_ubifs(sb) + || au_test_hfsplus(sb) /* maintained, but incorrect */ + /* || au_test_ext4(sb) */ /* untested */ + /* || au_test_ocfs2(sb) */ /* untested */ + /* || au_test_ocfs2_dlmfs(sb) */ /* untested */ + /* || au_test_sysv(sb) */ /* untested */ + /* || au_test_minix(sb) */ /* untested */ + ; +} + +/* + * filesystems which don't store the correct value in some of their inode + * attributes. + */ +static inline int au_test_fs_bad_iattr(struct super_block *sb) +{ + return au_test_fs_bad_iattr_size(sb) + /* || au_test_cifs(sb) */ /* untested */ + || au_test_fat(sb) + || au_test_msdos(sb) + || au_test_vfat(sb); +} + +/* they don't check i_nlink in link(2) */ +static inline int au_test_fs_no_limit_nlink(struct super_block *sb) +{ + return au_test_tmpfs(sb) +#ifdef CONFIG_AUFS_BR_RAMFS + || au_test_ramfs(sb) +#endif + || au_test_ubifs(sb) + || au_test_hfsplus(sb); +} + +/* + * filesystems which sets S_NOATIME and S_NOCMTIME. + */ +static inline int au_test_fs_notime(struct super_block *sb) +{ + return au_test_nfs(sb) + || au_test_fuse(sb) + || au_test_ubifs(sb) + /* || au_test_cifs(sb) */ /* untested */ + ; +} + +/* + * filesystems which requires replacing i_mapping. + */ +static inline int au_test_fs_bad_mapping(struct super_block *sb) +{ + return au_test_fuse(sb) + || au_test_ubifs(sb); +} + +/* temporary support for i#1 in cramfs */ +static inline int au_test_fs_unique_ino(struct inode *inode) +{ + if (au_test_cramfs(inode->i_sb)) + return inode->i_ino != 1; + return 1; +} + +/* ---------------------------------------------------------------------- */ + +/* + * the filesystem where the xino files placed must support i/o after unlink and + * maintain i_size and i_blocks. + */ +static inline int au_test_fs_bad_xino(struct super_block *sb) +{ + return au_test_fs_remote(sb) + || au_test_fs_bad_iattr_size(sb) + /* don't want unnecessary work for xino */ + || au_test_aufs(sb) + || au_test_ecryptfs(sb) + || au_test_nilfs(sb); +} + +static inline int au_test_fs_trunc_xino(struct super_block *sb) +{ + return au_test_tmpfs(sb) + || au_test_ramfs(sb); +} + +/* + * test if the @sb is real-readonly. + */ +static inline int au_test_fs_rr(struct super_block *sb) +{ + return au_test_squashfs(sb) + || au_test_iso9660(sb) + || au_test_cramfs(sb) + || au_test_romfs(sb); +} + +#endif /* __KERNEL__ */ +#endif /* __AUFS_FSTYPE_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/hfsnotify.c kernel.21.3.new/fs/aufs/hfsnotify.c --- kernel.21.3.orig/fs/aufs/hfsnotify.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/hfsnotify.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,302 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * fsnotify for the lower directories + */ + +#include "aufs.h" + +/* FS_IN_IGNORED is unnecessary */ +static const __u32 AuHfsnMask = (FS_MOVED_TO | FS_MOVED_FROM | FS_DELETE + | FS_CREATE | FS_EVENT_ON_CHILD); +static DECLARE_WAIT_QUEUE_HEAD(au_hfsn_wq); +static __cacheline_aligned_in_smp atomic64_t au_hfsn_ifree = ATOMIC64_INIT(0); + +static void au_hfsn_free_mark(struct fsnotify_mark *mark) +{ + struct au_hnotify *hn = container_of(mark, struct au_hnotify, + hn_mark); + AuDbg("here\n"); + au_cache_free_hnotify(hn); + smp_mb__before_atomic_dec(); + if (atomic64_dec_and_test(&au_hfsn_ifree)) + wake_up(&au_hfsn_wq); +} + +static int au_hfsn_alloc(struct au_hinode *hinode) +{ + int err; + struct au_hnotify *hn; + struct super_block *sb; + struct au_branch *br; + struct fsnotify_mark *mark; + aufs_bindex_t bindex; + + hn = hinode->hi_notify; + sb = hn->hn_aufs_inode->i_sb; + bindex = au_br_index(sb, hinode->hi_id); + br = au_sbr(sb, bindex); + AuDebugOn(!br->br_hfsn); + + mark = &hn->hn_mark; + fsnotify_init_mark(mark, au_hfsn_free_mark); + mark->mask = AuHfsnMask; + /* + * by udba rename or rmdir, aufs assign a new inode to the known + * h_inode, so specify 1 to allow dups. + */ + lockdep_off(); + err = fsnotify_add_mark(mark, br->br_hfsn->hfsn_group, hinode->hi_inode, + /*mnt*/NULL, /*allow_dups*/1); + /* even if err */ + fsnotify_put_mark(mark); + lockdep_on(); + + return err; +} + +static int au_hfsn_free(struct au_hinode *hinode, struct au_hnotify *hn) +{ + struct fsnotify_mark *mark; + unsigned long long ull; + struct fsnotify_group *group; + + ull = atomic64_inc_return(&au_hfsn_ifree); + BUG_ON(!ull); + + mark = &hn->hn_mark; + spin_lock(&mark->lock); + group = mark->group; + fsnotify_get_group(group); + spin_unlock(&mark->lock); + lockdep_off(); + fsnotify_destroy_mark(mark, group); + fsnotify_put_group(group); + lockdep_on(); + + /* free hn by myself */ + return 0; +} + +/* ---------------------------------------------------------------------- */ + +static void au_hfsn_ctl(struct au_hinode *hinode, int do_set) +{ + struct fsnotify_mark *mark; + + mark = &hinode->hi_notify->hn_mark; + spin_lock(&mark->lock); + if (do_set) { + AuDebugOn(mark->mask & AuHfsnMask); + mark->mask |= AuHfsnMask; + } else { + AuDebugOn(!(mark->mask & AuHfsnMask)); + mark->mask &= ~AuHfsnMask; + } + spin_unlock(&mark->lock); + /* fsnotify_recalc_inode_mask(hinode->hi_inode); */ +} + +/* ---------------------------------------------------------------------- */ + +/* #define AuDbgHnotify */ +#ifdef AuDbgHnotify +static char *au_hfsn_name(u32 mask) +{ +#ifdef CONFIG_AUFS_DEBUG +#define test_ret(flag) \ + do { \ + if (mask & flag) \ + return #flag; \ + } while (0) + test_ret(FS_ACCESS); + test_ret(FS_MODIFY); + test_ret(FS_ATTRIB); + test_ret(FS_CLOSE_WRITE); + test_ret(FS_CLOSE_NOWRITE); + test_ret(FS_OPEN); + test_ret(FS_MOVED_FROM); + test_ret(FS_MOVED_TO); + test_ret(FS_CREATE); + test_ret(FS_DELETE); + test_ret(FS_DELETE_SELF); + test_ret(FS_MOVE_SELF); + test_ret(FS_UNMOUNT); + test_ret(FS_Q_OVERFLOW); + test_ret(FS_IN_IGNORED); + test_ret(FS_IN_ISDIR); + test_ret(FS_IN_ONESHOT); + test_ret(FS_EVENT_ON_CHILD); + return ""; +#undef test_ret +#else + return "??"; +#endif +} +#endif + +/* ---------------------------------------------------------------------- */ + +static void au_hfsn_free_group(struct fsnotify_group *group) +{ + struct au_br_hfsnotify *hfsn = group->private; + + AuDbg("here\n"); + kfree(hfsn); +} + +static int au_hfsn_handle_event(struct fsnotify_group *group, + struct fsnotify_mark *inode_mark, + struct fsnotify_mark *vfsmount_mark, + struct fsnotify_event *event) +{ + int err; + struct au_hnotify *hnotify; + struct inode *h_dir, *h_inode; + __u32 mask; + struct qstr h_child_qstr = QSTR_INIT(event->file_name, event->name_len); + + AuDebugOn(event->data_type != FSNOTIFY_EVENT_INODE); + + err = 0; + /* if FS_UNMOUNT happens, there must be another bug */ + mask = event->mask; + AuDebugOn(mask & FS_UNMOUNT); + if (mask & (FS_IN_IGNORED | FS_UNMOUNT)) + goto out; + + h_dir = event->to_tell; + h_inode = event->inode; +#ifdef AuDbgHnotify + au_debug_on(); + if (1 || h_child_qstr.len != sizeof(AUFS_XINO_FNAME) - 1 + || strncmp(h_child_qstr.name, AUFS_XINO_FNAME, h_child_qstr.len)) { + AuDbg("i%lu, mask 0x%x %s, hcname %.*s, hi%lu\n", + h_dir->i_ino, mask, au_hfsn_name(mask), + AuLNPair(&h_child_qstr), h_inode ? h_inode->i_ino : 0); + /* WARN_ON(1); */ + } + au_debug_off(); +#endif + + AuDebugOn(!inode_mark); + hnotify = container_of(inode_mark, struct au_hnotify, hn_mark); + err = au_hnotify(h_dir, hnotify, mask, &h_child_qstr, h_inode); + +out: + return err; +} + +/* isn't it waste to ask every registered 'group'? */ +/* copied from linux/fs/notify/inotify/inotify_fsnotiry.c */ +/* it should be exported to modules */ +static bool au_hfsn_should_send_event(struct fsnotify_group *group, + struct inode *h_inode, + struct fsnotify_mark *inode_mark, + struct fsnotify_mark *vfsmount_mark, + __u32 mask, void *data, int data_type) +{ + mask = (mask & ~FS_EVENT_ON_CHILD); + return inode_mark->mask & mask; +} + +static struct fsnotify_ops au_hfsn_ops = { + .should_send_event = au_hfsn_should_send_event, + .handle_event = au_hfsn_handle_event, + .free_group_priv = au_hfsn_free_group +}; + +/* ---------------------------------------------------------------------- */ + +static void au_hfsn_fin_br(struct au_branch *br) +{ + struct au_br_hfsnotify *hfsn; + + hfsn = br->br_hfsn; + if (hfsn) { + lockdep_off(); + fsnotify_put_group(hfsn->hfsn_group); + lockdep_on(); + } +} + +static int au_hfsn_init_br(struct au_branch *br, int perm) +{ + int err; + struct fsnotify_group *group; + struct au_br_hfsnotify *hfsn; + + err = 0; + br->br_hfsn = NULL; + if (!au_br_hnotifyable(perm)) + goto out; + + err = -ENOMEM; + hfsn = kmalloc(sizeof(*hfsn), GFP_NOFS); + if (unlikely(!hfsn)) + goto out; + + err = 0; + group = fsnotify_alloc_group(&au_hfsn_ops); + if (IS_ERR(group)) { + err = PTR_ERR(group); + pr_err("fsnotify_alloc_group() failed, %d\n", err); + goto out_hfsn; + } + + group->private = hfsn; + hfsn->hfsn_group = group; + br->br_hfsn = hfsn; + goto out; /* success */ + +out_hfsn: + kfree(hfsn); +out: + return err; +} + +static int au_hfsn_reset_br(unsigned int udba, struct au_branch *br, int perm) +{ + int err; + + err = 0; + if (!br->br_hfsn) + err = au_hfsn_init_br(br, perm); + + return err; +} + +/* ---------------------------------------------------------------------- */ + +static void au_hfsn_fin(void) +{ + AuDbg("au_hfsn_ifree %lld\n", (long long)atomic64_read(&au_hfsn_ifree)); + wait_event(au_hfsn_wq, !atomic64_read(&au_hfsn_ifree)); +} + +const struct au_hnotify_op au_hnotify_op = { + .ctl = au_hfsn_ctl, + .alloc = au_hfsn_alloc, + .free = au_hfsn_free, + + .fin = au_hfsn_fin, + + .reset_br = au_hfsn_reset_br, + .fin_br = au_hfsn_fin_br, + .init_br = au_hfsn_init_br +}; diff -Naur kernel.21.3.orig/fs/aufs/hfsplus.c kernel.21.3.new/fs/aufs/hfsplus.c --- kernel.21.3.orig/fs/aufs/hfsplus.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/hfsplus.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,56 @@ +/* + * Copyright (C) 2010-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * special support for filesystems which aqucires an inode mutex + * at final closing a file, eg, hfsplus. + * + * This trick is very simple and stupid, just to open the file before really + * neceeary open to tell hfsplus that this is not the final closing. + * The caller should call au_h_open_pre() after acquiring the inode mutex, + * and au_h_open_post() after releasing it. + */ + +#include "aufs.h" + +struct file *au_h_open_pre(struct dentry *dentry, aufs_bindex_t bindex, + int force_wr) +{ + struct file *h_file; + struct dentry *h_dentry; + + h_dentry = au_h_dptr(dentry, bindex); + AuDebugOn(!h_dentry); + AuDebugOn(!h_dentry->d_inode); + + h_file = NULL; + if (au_test_hfsplus(h_dentry->d_sb) + && S_ISREG(h_dentry->d_inode->i_mode)) + h_file = au_h_open(dentry, bindex, + O_RDONLY | O_NOATIME | O_LARGEFILE, + /*file*/NULL, force_wr); + return h_file; +} + +void au_h_open_post(struct dentry *dentry, aufs_bindex_t bindex, + struct file *h_file) +{ + if (h_file) { + fput(h_file); + au_sbr_put(dentry->d_sb, bindex); + } +} diff -Naur kernel.21.3.orig/fs/aufs/hnotify.c kernel.21.3.new/fs/aufs/hnotify.c --- kernel.21.3.orig/fs/aufs/hnotify.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/hnotify.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,712 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * abstraction to notify the direct changes on lower directories + */ + +#include "aufs.h" + +int au_hn_alloc(struct au_hinode *hinode, struct inode *inode) +{ + int err; + struct au_hnotify *hn; + + err = -ENOMEM; + hn = au_cache_alloc_hnotify(); + if (hn) { + hn->hn_aufs_inode = inode; + hinode->hi_notify = hn; + err = au_hnotify_op.alloc(hinode); + AuTraceErr(err); + if (unlikely(err)) { + hinode->hi_notify = NULL; + au_cache_free_hnotify(hn); + /* + * The upper dir was removed by udba, but the same named + * dir left. In this case, aufs assignes a new inode + * number and set the monitor again. + * For the lower dir, the old monitnor is still left. + */ + if (err == -EEXIST) + err = 0; + } + } + + AuTraceErr(err); + return err; +} + +void au_hn_free(struct au_hinode *hinode) +{ + struct au_hnotify *hn; + + hn = hinode->hi_notify; + if (hn) { + hinode->hi_notify = NULL; + if (au_hnotify_op.free(hinode, hn)) + au_cache_free_hnotify(hn); + } +} + +/* ---------------------------------------------------------------------- */ + +void au_hn_ctl(struct au_hinode *hinode, int do_set) +{ + if (hinode->hi_notify) + au_hnotify_op.ctl(hinode, do_set); +} + +void au_hn_reset(struct inode *inode, unsigned int flags) +{ + aufs_bindex_t bindex, bend; + struct inode *hi; + struct dentry *iwhdentry; + + bend = au_ibend(inode); + for (bindex = au_ibstart(inode); bindex <= bend; bindex++) { + hi = au_h_iptr(inode, bindex); + if (!hi) + continue; + + /* mutex_lock_nested(&hi->i_mutex, AuLsc_I_CHILD); */ + iwhdentry = au_hi_wh(inode, bindex); + if (iwhdentry) + dget(iwhdentry); + au_igrab(hi); + au_set_h_iptr(inode, bindex, NULL, 0); + au_set_h_iptr(inode, bindex, au_igrab(hi), + flags & ~AuHi_XINO); + iput(hi); + dput(iwhdentry); + /* mutex_unlock(&hi->i_mutex); */ + } +} + +/* ---------------------------------------------------------------------- */ + +static int hn_xino(struct inode *inode, struct inode *h_inode) +{ + int err; + aufs_bindex_t bindex, bend, bfound, bstart; + struct inode *h_i; + + err = 0; + if (unlikely(inode->i_ino == AUFS_ROOT_INO)) { + pr_warn("branch root dir was changed\n"); + goto out; + } + + bfound = -1; + bend = au_ibend(inode); + bstart = au_ibstart(inode); +#if 0 /* reserved for future use */ + if (bindex == bend) { + /* keep this ino in rename case */ + goto out; + } +#endif + for (bindex = bstart; bindex <= bend; bindex++) + if (au_h_iptr(inode, bindex) == h_inode) { + bfound = bindex; + break; + } + if (bfound < 0) + goto out; + + for (bindex = bstart; bindex <= bend; bindex++) { + h_i = au_h_iptr(inode, bindex); + if (!h_i) + continue; + + err = au_xino_write(inode->i_sb, bindex, h_i->i_ino, /*ino*/0); + /* ignore this error */ + /* bad action? */ + } + + /* children inode number will be broken */ + +out: + AuTraceErr(err); + return err; +} + +static int hn_gen_tree(struct dentry *dentry) +{ + int err, i, j, ndentry; + struct au_dcsub_pages dpages; + struct au_dpage *dpage; + struct dentry **dentries; + + err = au_dpages_init(&dpages, GFP_NOFS); + if (unlikely(err)) + goto out; + err = au_dcsub_pages(&dpages, dentry, NULL, NULL); + if (unlikely(err)) + goto out_dpages; + + for (i = 0; i < dpages.ndpage; i++) { + dpage = dpages.dpages + i; + dentries = dpage->dentries; + ndentry = dpage->ndentry; + for (j = 0; j < ndentry; j++) { + struct dentry *d; + + d = dentries[j]; + if (IS_ROOT(d)) + continue; + + au_digen_dec(d); + if (d->d_inode) + /* todo: reset children xino? + cached children only? */ + au_iigen_dec(d->d_inode); + } + } + +out_dpages: + au_dpages_free(&dpages); + +#if 0 + /* discard children */ + dentry_unhash(dentry); + dput(dentry); +#endif +out: + return err; +} + +/* + * return 0 if processed. + */ +static int hn_gen_by_inode(char *name, unsigned int nlen, struct inode *inode, + const unsigned int isdir) +{ + int err; + struct dentry *d; + struct qstr *dname; + + err = 1; + if (unlikely(inode->i_ino == AUFS_ROOT_INO)) { + pr_warn("branch root dir was changed\n"); + err = 0; + goto out; + } + + if (!isdir) { + AuDebugOn(!name); + au_iigen_dec(inode); + spin_lock(&inode->i_lock); + hlist_for_each_entry(d, &inode->i_dentry, d_alias) { + spin_lock(&d->d_lock); + dname = &d->d_name; + if (dname->len != nlen + && memcmp(dname->name, name, nlen)) { + spin_unlock(&d->d_lock); + continue; + } + err = 0; + au_digen_dec(d); + spin_unlock(&d->d_lock); + break; + } + spin_unlock(&inode->i_lock); + } else { + au_fset_si(au_sbi(inode->i_sb), FAILED_REFRESH_DIR); + d = d_find_any_alias(inode); + if (!d) { + au_iigen_dec(inode); + goto out; + } + + spin_lock(&d->d_lock); + dname = &d->d_name; + if (dname->len == nlen && !memcmp(dname->name, name, nlen)) { + spin_unlock(&d->d_lock); + err = hn_gen_tree(d); + spin_lock(&d->d_lock); + } + spin_unlock(&d->d_lock); + dput(d); + } + +out: + AuTraceErr(err); + return err; +} + +static int hn_gen_by_name(struct dentry *dentry, const unsigned int isdir) +{ + int err; + struct inode *inode; + + inode = dentry->d_inode; + if (IS_ROOT(dentry) + /* || (inode && inode->i_ino == AUFS_ROOT_INO) */ + ) { + pr_warn("branch root dir was changed\n"); + return 0; + } + + err = 0; + if (!isdir) { + au_digen_dec(dentry); + if (inode) + au_iigen_dec(inode); + } else { + au_fset_si(au_sbi(dentry->d_sb), FAILED_REFRESH_DIR); + if (inode) + err = hn_gen_tree(dentry); + } + + AuTraceErr(err); + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* hnotify job flags */ +#define AuHnJob_XINO0 1 +#define AuHnJob_GEN (1 << 1) +#define AuHnJob_DIRENT (1 << 2) +#define AuHnJob_ISDIR (1 << 3) +#define AuHnJob_TRYXINO0 (1 << 4) +#define AuHnJob_MNTPNT (1 << 5) +#define au_ftest_hnjob(flags, name) ((flags) & AuHnJob_##name) +#define au_fset_hnjob(flags, name) \ + do { (flags) |= AuHnJob_##name; } while (0) +#define au_fclr_hnjob(flags, name) \ + do { (flags) &= ~AuHnJob_##name; } while (0) + +enum { + AuHn_CHILD, + AuHn_PARENT, + AuHnLast +}; + +struct au_hnotify_args { + struct inode *h_dir, *dir, *h_child_inode; + u32 mask; + unsigned int flags[AuHnLast]; + unsigned int h_child_nlen; + char h_child_name[]; +}; + +struct hn_job_args { + unsigned int flags; + struct inode *inode, *h_inode, *dir, *h_dir; + struct dentry *dentry; + char *h_name; + int h_nlen; +}; + +static int hn_job(struct hn_job_args *a) +{ + const unsigned int isdir = au_ftest_hnjob(a->flags, ISDIR); + int e; + + /* reset xino */ + if (au_ftest_hnjob(a->flags, XINO0) && a->inode) + hn_xino(a->inode, a->h_inode); /* ignore this error */ + + if (au_ftest_hnjob(a->flags, TRYXINO0) + && a->inode + && a->h_inode) { + mutex_lock_nested(&a->h_inode->i_mutex, AuLsc_I_CHILD); + if (!a->h_inode->i_nlink) + hn_xino(a->inode, a->h_inode); /* ignore this error */ + mutex_unlock(&a->h_inode->i_mutex); + } + + /* make the generation obsolete */ + if (au_ftest_hnjob(a->flags, GEN)) { + e = -1; + if (a->inode) + e = hn_gen_by_inode(a->h_name, a->h_nlen, a->inode, + isdir); + if (e && a->dentry) + hn_gen_by_name(a->dentry, isdir); + /* ignore this error */ + } + + /* make dir entries obsolete */ + if (au_ftest_hnjob(a->flags, DIRENT) && a->inode) { + struct au_vdir *vdir; + + vdir = au_ivdir(a->inode); + if (vdir) + vdir->vd_jiffy = 0; + /* IMustLock(a->inode); */ + /* a->inode->i_version++; */ + } + + /* can do nothing but warn */ + if (au_ftest_hnjob(a->flags, MNTPNT) + && a->dentry + && d_mountpoint(a->dentry)) + pr_warn("mount-point %.*s is removed or renamed\n", + AuDLNPair(a->dentry)); + + return 0; +} + +/* ---------------------------------------------------------------------- */ + +static struct dentry *lookup_wlock_by_name(char *name, unsigned int nlen, + struct inode *dir) +{ + struct dentry *dentry, *d, *parent; + struct qstr *dname; + + parent = d_find_any_alias(dir); + if (!parent) + return NULL; + + dentry = NULL; + spin_lock(&parent->d_lock); + list_for_each_entry(d, &parent->d_subdirs, d_u.d_child) { + /* AuDbg("%.*s\n", AuDLNPair(d)); */ + spin_lock_nested(&d->d_lock, DENTRY_D_LOCK_NESTED); + dname = &d->d_name; + if (dname->len != nlen || memcmp(dname->name, name, nlen)) + goto cont_unlock; + if (au_di(d)) + au_digen_dec(d); + else + goto cont_unlock; + if (d->d_count) { + dentry = dget_dlock(d); + spin_unlock(&d->d_lock); + break; + } + +cont_unlock: + spin_unlock(&d->d_lock); + } + spin_unlock(&parent->d_lock); + dput(parent); + + if (dentry) + di_write_lock_child(dentry); + + return dentry; +} + +static struct inode *lookup_wlock_by_ino(struct super_block *sb, + aufs_bindex_t bindex, ino_t h_ino) +{ + struct inode *inode; + ino_t ino; + int err; + + inode = NULL; + err = au_xino_read(sb, bindex, h_ino, &ino); + if (!err && ino) + inode = ilookup(sb, ino); + if (!inode) + goto out; + + if (unlikely(inode->i_ino == AUFS_ROOT_INO)) { + pr_warn("wrong root branch\n"); + iput(inode); + inode = NULL; + goto out; + } + + ii_write_lock_child(inode); + +out: + return inode; +} + +static void au_hn_bh(void *_args) +{ + struct au_hnotify_args *a = _args; + struct super_block *sb; + aufs_bindex_t bindex, bend, bfound; + unsigned char xino, try_iput; + int err; + struct inode *inode; + ino_t h_ino; + struct hn_job_args args; + struct dentry *dentry; + struct au_sbinfo *sbinfo; + + AuDebugOn(!_args); + AuDebugOn(!a->h_dir); + AuDebugOn(!a->dir); + AuDebugOn(!a->mask); + AuDbg("mask 0x%x, i%lu, hi%lu, hci%lu\n", + a->mask, a->dir->i_ino, a->h_dir->i_ino, + a->h_child_inode ? a->h_child_inode->i_ino : 0); + + inode = NULL; + dentry = NULL; + /* + * do not lock a->dir->i_mutex here + * because of d_revalidate() may cause a deadlock. + */ + sb = a->dir->i_sb; + AuDebugOn(!sb); + sbinfo = au_sbi(sb); + AuDebugOn(!sbinfo); + si_write_lock(sb, AuLock_NOPLMW); + + ii_read_lock_parent(a->dir); + bfound = -1; + bend = au_ibend(a->dir); + for (bindex = au_ibstart(a->dir); bindex <= bend; bindex++) + if (au_h_iptr(a->dir, bindex) == a->h_dir) { + bfound = bindex; + break; + } + ii_read_unlock(a->dir); + if (unlikely(bfound < 0)) + goto out; + + xino = !!au_opt_test(au_mntflags(sb), XINO); + h_ino = 0; + if (a->h_child_inode) + h_ino = a->h_child_inode->i_ino; + + if (a->h_child_nlen + && (au_ftest_hnjob(a->flags[AuHn_CHILD], GEN) + || au_ftest_hnjob(a->flags[AuHn_CHILD], MNTPNT))) + dentry = lookup_wlock_by_name(a->h_child_name, a->h_child_nlen, + a->dir); + try_iput = 0; + if (dentry) + inode = dentry->d_inode; + if (xino && !inode && h_ino + && (au_ftest_hnjob(a->flags[AuHn_CHILD], XINO0) + || au_ftest_hnjob(a->flags[AuHn_CHILD], TRYXINO0) + || au_ftest_hnjob(a->flags[AuHn_CHILD], GEN))) { + inode = lookup_wlock_by_ino(sb, bfound, h_ino); + try_iput = 1; + } + + args.flags = a->flags[AuHn_CHILD]; + args.dentry = dentry; + args.inode = inode; + args.h_inode = a->h_child_inode; + args.dir = a->dir; + args.h_dir = a->h_dir; + args.h_name = a->h_child_name; + args.h_nlen = a->h_child_nlen; + err = hn_job(&args); + if (dentry) { + if (au_di(dentry)) + di_write_unlock(dentry); + dput(dentry); + } + if (inode && try_iput) { + ii_write_unlock(inode); + iput(inode); + } + + ii_write_lock_parent(a->dir); + args.flags = a->flags[AuHn_PARENT]; + args.dentry = NULL; + args.inode = a->dir; + args.h_inode = a->h_dir; + args.dir = NULL; + args.h_dir = NULL; + args.h_name = NULL; + args.h_nlen = 0; + err = hn_job(&args); + ii_write_unlock(a->dir); + +out: + iput(a->h_child_inode); + iput(a->h_dir); + iput(a->dir); + si_write_unlock(sb); + au_nwt_done(&sbinfo->si_nowait); + kfree(a); +} + +/* ---------------------------------------------------------------------- */ + +int au_hnotify(struct inode *h_dir, struct au_hnotify *hnotify, u32 mask, + struct qstr *h_child_qstr, struct inode *h_child_inode) +{ + int err, len; + unsigned int flags[AuHnLast], f; + unsigned char isdir, isroot, wh; + struct inode *dir; + struct au_hnotify_args *args; + char *p, *h_child_name; + + err = 0; + AuDebugOn(!hnotify || !hnotify->hn_aufs_inode); + dir = igrab(hnotify->hn_aufs_inode); + if (!dir) + goto out; + + isroot = (dir->i_ino == AUFS_ROOT_INO); + wh = 0; + h_child_name = (void *)h_child_qstr->name; + len = h_child_qstr->len; + if (h_child_name) { + if (len > AUFS_WH_PFX_LEN + && !memcmp(h_child_name, AUFS_WH_PFX, AUFS_WH_PFX_LEN)) { + h_child_name += AUFS_WH_PFX_LEN; + len -= AUFS_WH_PFX_LEN; + wh = 1; + } + } + + isdir = 0; + if (h_child_inode) + isdir = !!S_ISDIR(h_child_inode->i_mode); + flags[AuHn_PARENT] = AuHnJob_ISDIR; + flags[AuHn_CHILD] = 0; + if (isdir) + flags[AuHn_CHILD] = AuHnJob_ISDIR; + au_fset_hnjob(flags[AuHn_PARENT], DIRENT); + au_fset_hnjob(flags[AuHn_CHILD], GEN); + switch (mask & FS_EVENTS_POSS_ON_CHILD) { + case FS_MOVED_FROM: + case FS_MOVED_TO: + au_fset_hnjob(flags[AuHn_CHILD], XINO0); + au_fset_hnjob(flags[AuHn_CHILD], MNTPNT); + /*FALLTHROUGH*/ + case FS_CREATE: + AuDebugOn(!h_child_name || !h_child_inode); + break; + + case FS_DELETE: + /* + * aufs never be able to get this child inode. + * revalidation should be in d_revalidate() + * by checking i_nlink, i_generation or d_unhashed(). + */ + AuDebugOn(!h_child_name); + au_fset_hnjob(flags[AuHn_CHILD], TRYXINO0); + au_fset_hnjob(flags[AuHn_CHILD], MNTPNT); + break; + + default: + AuDebugOn(1); + } + + if (wh) + h_child_inode = NULL; + + err = -ENOMEM; + /* iput() and kfree() will be called in au_hnotify() */ + args = kmalloc(sizeof(*args) + len + 1, GFP_NOFS); + if (unlikely(!args)) { + AuErr1("no memory\n"); + iput(dir); + goto out; + } + args->flags[AuHn_PARENT] = flags[AuHn_PARENT]; + args->flags[AuHn_CHILD] = flags[AuHn_CHILD]; + args->mask = mask; + args->dir = dir; + args->h_dir = igrab(h_dir); + if (h_child_inode) + h_child_inode = igrab(h_child_inode); /* can be NULL */ + args->h_child_inode = h_child_inode; + args->h_child_nlen = len; + if (len) { + p = (void *)args; + p += sizeof(*args); + memcpy(p, h_child_name, len); + p[len] = 0; + } + + f = 0; + if (!dir->i_nlink) + f = AuWkq_NEST; + err = au_wkq_nowait(au_hn_bh, args, dir->i_sb, f); + if (unlikely(err)) { + pr_err("wkq %d\n", err); + iput(args->h_child_inode); + iput(args->h_dir); + iput(args->dir); + kfree(args); + } + +out: + return err; +} + +/* ---------------------------------------------------------------------- */ + +int au_hnotify_reset_br(unsigned int udba, struct au_branch *br, int perm) +{ + int err; + + AuDebugOn(!(udba & AuOptMask_UDBA)); + + err = 0; + if (au_hnotify_op.reset_br) + err = au_hnotify_op.reset_br(udba, br, perm); + + return err; +} + +int au_hnotify_init_br(struct au_branch *br, int perm) +{ + int err; + + err = 0; + if (au_hnotify_op.init_br) + err = au_hnotify_op.init_br(br, perm); + + return err; +} + +void au_hnotify_fin_br(struct au_branch *br) +{ + if (au_hnotify_op.fin_br) + au_hnotify_op.fin_br(br); +} + +static void au_hn_destroy_cache(void) +{ + kmem_cache_destroy(au_cachep[AuCache_HNOTIFY]); + au_cachep[AuCache_HNOTIFY] = NULL; +} + +int __init au_hnotify_init(void) +{ + int err; + + err = -ENOMEM; + au_cachep[AuCache_HNOTIFY] = AuCache(au_hnotify); + if (au_cachep[AuCache_HNOTIFY]) { + err = 0; + if (au_hnotify_op.init) + err = au_hnotify_op.init(); + if (unlikely(err)) + au_hn_destroy_cache(); + } + AuTraceErr(err); + return err; +} + +void au_hnotify_fin(void) +{ + if (au_hnotify_op.fin) + au_hnotify_op.fin(); + /* cf. au_cache_fin() */ + if (au_cachep[AuCache_HNOTIFY]) + au_hn_destroy_cache(); +} diff -Naur kernel.21.3.orig/fs/aufs/iinfo.c kernel.21.3.new/fs/aufs/iinfo.c --- kernel.21.3.orig/fs/aufs/iinfo.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/iinfo.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,275 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * inode private data + */ + +#include "aufs.h" + +struct inode *au_h_iptr(struct inode *inode, aufs_bindex_t bindex) +{ + struct inode *h_inode; + + IiMustAnyLock(inode); + + h_inode = au_ii(inode)->ii_hinode[0 + bindex].hi_inode; + AuDebugOn(h_inode && atomic_read(&h_inode->i_count) <= 0); + return h_inode; +} + +/* todo: hard/soft set? */ +void au_hiput(struct au_hinode *hinode) +{ + au_hn_free(hinode); + dput(hinode->hi_whdentry); + iput(hinode->hi_inode); +} + +unsigned int au_hi_flags(struct inode *inode, int isdir) +{ + unsigned int flags; + const unsigned int mnt_flags = au_mntflags(inode->i_sb); + + flags = 0; + if (au_opt_test(mnt_flags, XINO)) + au_fset_hi(flags, XINO); + if (isdir && au_opt_test(mnt_flags, UDBA_HNOTIFY)) + au_fset_hi(flags, HNOTIFY); + return flags; +} + +void au_set_h_iptr(struct inode *inode, aufs_bindex_t bindex, + struct inode *h_inode, unsigned int flags) +{ + struct au_hinode *hinode; + struct inode *hi; + struct au_iinfo *iinfo = au_ii(inode); + + IiMustWriteLock(inode); + + hinode = iinfo->ii_hinode + bindex; + hi = hinode->hi_inode; + AuDebugOn(h_inode && atomic_read(&h_inode->i_count) <= 0); + + if (hi) + au_hiput(hinode); + hinode->hi_inode = h_inode; + if (h_inode) { + int err; + struct super_block *sb = inode->i_sb; + struct au_branch *br; + + AuDebugOn(inode->i_mode + && (h_inode->i_mode & S_IFMT) + != (inode->i_mode & S_IFMT)); + if (bindex == iinfo->ii_bstart) + au_cpup_igen(inode, h_inode); + br = au_sbr(sb, bindex); + hinode->hi_id = br->br_id; + if (au_ftest_hi(flags, XINO)) { + err = au_xino_write(sb, bindex, h_inode->i_ino, + inode->i_ino); + if (unlikely(err)) + AuIOErr1("failed au_xino_write() %d\n", err); + } + + if (au_ftest_hi(flags, HNOTIFY) + && au_br_hnotifyable(br->br_perm)) { + err = au_hn_alloc(hinode, inode); + if (unlikely(err)) + AuIOErr1("au_hn_alloc() %d\n", err); + } + } +} + +void au_set_hi_wh(struct inode *inode, aufs_bindex_t bindex, + struct dentry *h_wh) +{ + struct au_hinode *hinode; + + IiMustWriteLock(inode); + + hinode = au_ii(inode)->ii_hinode + bindex; + AuDebugOn(hinode->hi_whdentry); + hinode->hi_whdentry = h_wh; +} + +void au_update_iigen(struct inode *inode, int half) +{ + struct au_iinfo *iinfo; + struct au_iigen *iigen; + unsigned int sigen; + + sigen = au_sigen(inode->i_sb); + iinfo = au_ii(inode); + iigen = &iinfo->ii_generation; + spin_lock(&iinfo->ii_genspin); + iigen->ig_generation = sigen; + if (half) + au_ig_fset(iigen->ig_flags, HALF_REFRESHED); + else + au_ig_fclr(iigen->ig_flags, HALF_REFRESHED); + spin_unlock(&iinfo->ii_genspin); +} + +/* it may be called at remount time, too */ +void au_update_ibrange(struct inode *inode, int do_put_zero) +{ + struct au_iinfo *iinfo; + aufs_bindex_t bindex, bend; + + iinfo = au_ii(inode); + if (!iinfo) + return; + + IiMustWriteLock(inode); + + if (do_put_zero && iinfo->ii_bstart >= 0) { + for (bindex = iinfo->ii_bstart; bindex <= iinfo->ii_bend; + bindex++) { + struct inode *h_i; + + h_i = iinfo->ii_hinode[0 + bindex].hi_inode; + if (h_i && !h_i->i_nlink) + au_set_h_iptr(inode, bindex, NULL, 0); + } + } + + iinfo->ii_bstart = -1; + iinfo->ii_bend = -1; + bend = au_sbend(inode->i_sb); + for (bindex = 0; bindex <= bend; bindex++) + if (iinfo->ii_hinode[0 + bindex].hi_inode) { + iinfo->ii_bstart = bindex; + break; + } + if (iinfo->ii_bstart >= 0) + for (bindex = bend; bindex >= iinfo->ii_bstart; bindex--) + if (iinfo->ii_hinode[0 + bindex].hi_inode) { + iinfo->ii_bend = bindex; + break; + } + AuDebugOn(iinfo->ii_bstart > iinfo->ii_bend); +} + +/* ---------------------------------------------------------------------- */ + +void au_icntnr_init_once(void *_c) +{ + struct au_icntnr *c = _c; + struct au_iinfo *iinfo = &c->iinfo; + static struct lock_class_key aufs_ii; + + spin_lock_init(&iinfo->ii_genspin); + au_rw_init(&iinfo->ii_rwsem); + au_rw_class(&iinfo->ii_rwsem, &aufs_ii); + inode_init_once(&c->vfs_inode); +} + +int au_iinfo_init(struct inode *inode) +{ + struct au_iinfo *iinfo; + struct super_block *sb; + int nbr, i; + + sb = inode->i_sb; + iinfo = &(container_of(inode, struct au_icntnr, vfs_inode)->iinfo); + nbr = au_sbend(sb) + 1; + if (unlikely(nbr <= 0)) + nbr = 1; + iinfo->ii_hinode = kcalloc(nbr, sizeof(*iinfo->ii_hinode), GFP_NOFS); + if (iinfo->ii_hinode) { + au_ninodes_inc(sb); + for (i = 0; i < nbr; i++) + iinfo->ii_hinode[i].hi_id = -1; + + iinfo->ii_generation.ig_generation = au_sigen(sb); + iinfo->ii_bstart = -1; + iinfo->ii_bend = -1; + iinfo->ii_vdir = NULL; + return 0; + } + return -ENOMEM; +} + +int au_ii_realloc(struct au_iinfo *iinfo, int nbr) +{ + int err, sz; + struct au_hinode *hip; + + AuRwMustWriteLock(&iinfo->ii_rwsem); + + err = -ENOMEM; + sz = sizeof(*hip) * (iinfo->ii_bend + 1); + if (!sz) + sz = sizeof(*hip); + hip = au_kzrealloc(iinfo->ii_hinode, sz, sizeof(*hip) * nbr, GFP_NOFS); + if (hip) { + iinfo->ii_hinode = hip; + err = 0; + } + + return err; +} + +void au_iinfo_fin(struct inode *inode) +{ + struct au_iinfo *iinfo; + struct au_hinode *hi; + struct super_block *sb; + aufs_bindex_t bindex, bend; + const unsigned char unlinked = !inode->i_nlink; + + iinfo = au_ii(inode); + /* bad_inode case */ + if (!iinfo) + return; + + sb = inode->i_sb; + au_ninodes_dec(sb); + if (si_pid_test(sb)) + au_xino_delete_inode(inode, unlinked); + else { + /* + * it is safe to hide the dependency between sbinfo and + * sb->s_umount. + */ + lockdep_off(); + si_noflush_read_lock(sb); + au_xino_delete_inode(inode, unlinked); + si_read_unlock(sb); + lockdep_on(); + } + + if (iinfo->ii_vdir) + au_vdir_free(iinfo->ii_vdir); + + bindex = iinfo->ii_bstart; + if (bindex >= 0) { + hi = iinfo->ii_hinode + bindex; + bend = iinfo->ii_bend; + while (bindex++ <= bend) { + if (hi->hi_inode) + au_hiput(hi); + hi++; + } + } + kfree(iinfo->ii_hinode); + iinfo->ii_hinode = NULL; + AuRwDestroy(&iinfo->ii_rwsem); +} diff -Naur kernel.21.3.orig/fs/aufs/inode.c kernel.21.3.new/fs/aufs/inode.c --- kernel.21.3.orig/fs/aufs/inode.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/inode.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,496 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * inode functions + */ + +#include "aufs.h" + +struct inode *au_igrab(struct inode *inode) +{ + if (inode) { + AuDebugOn(!atomic_read(&inode->i_count)); + ihold(inode); + } + return inode; +} + +static void au_refresh_hinode_attr(struct inode *inode, int do_version) +{ + au_cpup_attr_all(inode, /*force*/0); + au_update_iigen(inode, /*half*/1); + if (do_version) + inode->i_version++; +} + +static int au_ii_refresh(struct inode *inode, int *update) +{ + int err, e; + umode_t type; + aufs_bindex_t bindex, new_bindex; + struct super_block *sb; + struct au_iinfo *iinfo; + struct au_hinode *p, *q, tmp; + + IiMustWriteLock(inode); + + *update = 0; + sb = inode->i_sb; + type = inode->i_mode & S_IFMT; + iinfo = au_ii(inode); + err = au_ii_realloc(iinfo, au_sbend(sb) + 1); + if (unlikely(err)) + goto out; + + AuDebugOn(iinfo->ii_bstart < 0); + p = iinfo->ii_hinode + iinfo->ii_bstart; + for (bindex = iinfo->ii_bstart; bindex <= iinfo->ii_bend; + bindex++, p++) { + if (!p->hi_inode) + continue; + + AuDebugOn(type != (p->hi_inode->i_mode & S_IFMT)); + new_bindex = au_br_index(sb, p->hi_id); + if (new_bindex == bindex) + continue; + + if (new_bindex < 0) { + *update = 1; + au_hiput(p); + p->hi_inode = NULL; + continue; + } + + if (new_bindex < iinfo->ii_bstart) + iinfo->ii_bstart = new_bindex; + if (iinfo->ii_bend < new_bindex) + iinfo->ii_bend = new_bindex; + /* swap two lower inode, and loop again */ + q = iinfo->ii_hinode + new_bindex; + tmp = *q; + *q = *p; + *p = tmp; + if (tmp.hi_inode) { + bindex--; + p--; + } + } + au_update_ibrange(inode, /*do_put_zero*/0); + e = au_dy_irefresh(inode); + if (unlikely(e && !err)) + err = e; + +out: + AuTraceErr(err); + return err; +} + +int au_refresh_hinode_self(struct inode *inode) +{ + int err, update; + + err = au_ii_refresh(inode, &update); + if (!err) + au_refresh_hinode_attr(inode, update && S_ISDIR(inode->i_mode)); + + AuTraceErr(err); + return err; +} + +int au_refresh_hinode(struct inode *inode, struct dentry *dentry) +{ + int err, e, update; + unsigned int flags; + umode_t mode; + aufs_bindex_t bindex, bend; + unsigned char isdir; + struct au_hinode *p; + struct au_iinfo *iinfo; + + err = au_ii_refresh(inode, &update); + if (unlikely(err)) + goto out; + + update = 0; + iinfo = au_ii(inode); + p = iinfo->ii_hinode + iinfo->ii_bstart; + mode = (inode->i_mode & S_IFMT); + isdir = S_ISDIR(mode); + flags = au_hi_flags(inode, isdir); + bend = au_dbend(dentry); + for (bindex = au_dbstart(dentry); bindex <= bend; bindex++) { + struct inode *h_i; + struct dentry *h_d; + + h_d = au_h_dptr(dentry, bindex); + if (!h_d || !h_d->d_inode) + continue; + + AuDebugOn(mode != (h_d->d_inode->i_mode & S_IFMT)); + if (iinfo->ii_bstart <= bindex && bindex <= iinfo->ii_bend) { + h_i = au_h_iptr(inode, bindex); + if (h_i) { + if (h_i == h_d->d_inode) + continue; + err = -EIO; + break; + } + } + if (bindex < iinfo->ii_bstart) + iinfo->ii_bstart = bindex; + if (iinfo->ii_bend < bindex) + iinfo->ii_bend = bindex; + au_set_h_iptr(inode, bindex, au_igrab(h_d->d_inode), flags); + update = 1; + } + au_update_ibrange(inode, /*do_put_zero*/0); + e = au_dy_irefresh(inode); + if (unlikely(e && !err)) + err = e; + if (!err) + au_refresh_hinode_attr(inode, update && isdir); + +out: + AuTraceErr(err); + return err; +} + +static int set_inode(struct inode *inode, struct dentry *dentry) +{ + int err; + unsigned int flags; + umode_t mode; + aufs_bindex_t bindex, bstart, btail; + unsigned char isdir; + struct dentry *h_dentry; + struct inode *h_inode; + struct au_iinfo *iinfo; + + IiMustWriteLock(inode); + + err = 0; + isdir = 0; + bstart = au_dbstart(dentry); + h_inode = au_h_dptr(dentry, bstart)->d_inode; + mode = h_inode->i_mode; + switch (mode & S_IFMT) { + case S_IFREG: + btail = au_dbtail(dentry); + inode->i_op = &aufs_iop; + inode->i_fop = &aufs_file_fop; + err = au_dy_iaop(inode, bstart, h_inode); + if (unlikely(err)) + goto out; + break; + case S_IFDIR: + isdir = 1; + btail = au_dbtaildir(dentry); + inode->i_op = &aufs_dir_iop; + inode->i_fop = &aufs_dir_fop; + break; + case S_IFLNK: + btail = au_dbtail(dentry); + inode->i_op = &aufs_symlink_iop; + break; + case S_IFBLK: + case S_IFCHR: + case S_IFIFO: + case S_IFSOCK: + btail = au_dbtail(dentry); + inode->i_op = &aufs_iop; + init_special_inode(inode, mode, h_inode->i_rdev); + break; + default: + AuIOErr("Unknown file type 0%o\n", mode); + err = -EIO; + goto out; + } + + /* do not set hnotify for whiteouted dirs (SHWH mode) */ + flags = au_hi_flags(inode, isdir); + if (au_opt_test(au_mntflags(dentry->d_sb), SHWH) + && au_ftest_hi(flags, HNOTIFY) + && dentry->d_name.len > AUFS_WH_PFX_LEN + && !memcmp(dentry->d_name.name, AUFS_WH_PFX, AUFS_WH_PFX_LEN)) + au_fclr_hi(flags, HNOTIFY); + iinfo = au_ii(inode); + iinfo->ii_bstart = bstart; + iinfo->ii_bend = btail; + for (bindex = bstart; bindex <= btail; bindex++) { + h_dentry = au_h_dptr(dentry, bindex); + if (h_dentry) + au_set_h_iptr(inode, bindex, + au_igrab(h_dentry->d_inode), flags); + } + au_cpup_attr_all(inode, /*force*/1); + /* + * to force calling aufs_get_acl() every time, + * do not call cache_no_acl() for aufs inode. + */ + +out: + return err; +} + +/* + * successful returns with iinfo write_locked + * minus: errno + * zero: success, matched + * plus: no error, but unmatched + */ +static int reval_inode(struct inode *inode, struct dentry *dentry) +{ + int err; + unsigned int gen; + struct au_iigen iigen; + aufs_bindex_t bindex, bend; + struct inode *h_inode, *h_dinode; + + /* + * before this function, if aufs got any iinfo lock, it must be only + * one, the parent dir. + * it can happen by UDBA and the obsoleted inode number. + */ + err = -EIO; + if (unlikely(inode->i_ino == parent_ino(dentry))) + goto out; + + err = 1; + ii_write_lock_new_child(inode); + h_dinode = au_h_dptr(dentry, au_dbstart(dentry))->d_inode; + bend = au_ibend(inode); + for (bindex = au_ibstart(inode); bindex <= bend; bindex++) { + h_inode = au_h_iptr(inode, bindex); + if (!h_inode || h_inode != h_dinode) + continue; + + err = 0; + gen = au_iigen(inode, &iigen); + if (gen == au_digen(dentry) + && !au_ig_ftest(iigen.ig_flags, HALF_REFRESHED)) + break; + + /* fully refresh inode using dentry */ + err = au_refresh_hinode(inode, dentry); + if (!err) + au_update_iigen(inode, /*half*/0); + break; + } + + if (unlikely(err)) + ii_write_unlock(inode); +out: + return err; +} + +int au_ino(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino, + unsigned int d_type, ino_t *ino) +{ + int err; + struct mutex *mtx; + + /* prevent hardlinked inode number from race condition */ + mtx = NULL; + if (d_type != DT_DIR) { + mtx = &au_sbr(sb, bindex)->br_xino.xi_nondir_mtx; + mutex_lock(mtx); + } + err = au_xino_read(sb, bindex, h_ino, ino); + if (unlikely(err)) + goto out; + + if (!*ino) { + err = -EIO; + *ino = au_xino_new_ino(sb); + if (unlikely(!*ino)) + goto out; + err = au_xino_write(sb, bindex, h_ino, *ino); + if (unlikely(err)) + goto out; + } + +out: + if (mtx) + mutex_unlock(mtx); + return err; +} + +/* successful returns with iinfo write_locked */ +/* todo: return with unlocked? */ +struct inode *au_new_inode(struct dentry *dentry, int must_new) +{ + struct inode *inode, *h_inode; + struct dentry *h_dentry; + struct super_block *sb; + struct mutex *mtx; + ino_t h_ino, ino; + int err; + aufs_bindex_t bstart; + + sb = dentry->d_sb; + bstart = au_dbstart(dentry); + h_dentry = au_h_dptr(dentry, bstart); + h_inode = h_dentry->d_inode; + h_ino = h_inode->i_ino; + + /* + * stop 'race'-ing between hardlinks under different + * parents. + */ + mtx = NULL; + if (!S_ISDIR(h_inode->i_mode)) + mtx = &au_sbr(sb, bstart)->br_xino.xi_nondir_mtx; + +new_ino: + if (mtx) + mutex_lock(mtx); + err = au_xino_read(sb, bstart, h_ino, &ino); + inode = ERR_PTR(err); + if (unlikely(err)) + goto out; + + if (!ino) { + ino = au_xino_new_ino(sb); + if (unlikely(!ino)) { + inode = ERR_PTR(-EIO); + goto out; + } + } + + AuDbg("i%lu\n", (unsigned long)ino); + inode = au_iget_locked(sb, ino); + err = PTR_ERR(inode); + if (IS_ERR(inode)) + goto out; + + AuDbg("%lx, new %d\n", inode->i_state, !!(inode->i_state & I_NEW)); + if (inode->i_state & I_NEW) { + /* verbose coding for lock class name */ + if (unlikely(S_ISLNK(h_inode->i_mode))) + au_rw_class(&au_ii(inode)->ii_rwsem, + au_lc_key + AuLcSymlink_IIINFO); + else if (unlikely(S_ISDIR(h_inode->i_mode))) + au_rw_class(&au_ii(inode)->ii_rwsem, + au_lc_key + AuLcDir_IIINFO); + else /* likely */ + au_rw_class(&au_ii(inode)->ii_rwsem, + au_lc_key + AuLcNonDir_IIINFO); + + ii_write_lock_new_child(inode); + err = set_inode(inode, dentry); + if (!err) { + unlock_new_inode(inode); + goto out; /* success */ + } + + /* + * iget_failed() calls iput(), but we need to call + * ii_write_unlock() after iget_failed(). so dirty hack for + * i_count. + */ + atomic_inc(&inode->i_count); + iget_failed(inode); + ii_write_unlock(inode); + au_xino_write(sb, bstart, h_ino, /*ino*/0); + /* ignore this error */ + goto out_iput; + } else if (!must_new && !IS_DEADDIR(inode) && inode->i_nlink) { + /* + * horrible race condition between lookup, readdir and copyup + * (or something). + */ + if (mtx) + mutex_unlock(mtx); + err = reval_inode(inode, dentry); + if (unlikely(err < 0)) { + mtx = NULL; + goto out_iput; + } + + if (!err) { + mtx = NULL; + goto out; /* success */ + } else if (mtx) + mutex_lock(mtx); + } + + if (unlikely(au_test_fs_unique_ino(h_dentry->d_inode))) + AuWarn1("Warning: Un-notified UDBA or repeatedly renamed dir," + " b%d, %s, %.*s, hi%lu, i%lu.\n", + bstart, au_sbtype(h_dentry->d_sb), AuDLNPair(dentry), + (unsigned long)h_ino, (unsigned long)ino); + ino = 0; + err = au_xino_write(sb, bstart, h_ino, /*ino*/0); + if (!err) { + iput(inode); + if (mtx) + mutex_unlock(mtx); + goto new_ino; + } + +out_iput: + iput(inode); + inode = ERR_PTR(err); +out: + if (mtx) + mutex_unlock(mtx); + return inode; +} + +/* ---------------------------------------------------------------------- */ + +int au_test_ro(struct super_block *sb, aufs_bindex_t bindex, + struct inode *inode) +{ + int err; + struct inode *hi; + + err = au_br_rdonly(au_sbr(sb, bindex)); + + /* pseudo-link after flushed may happen out of bounds */ + if (!err + && inode + && au_ibstart(inode) <= bindex + && bindex <= au_ibend(inode)) { + /* + * permission check is unnecessary since vfsub routine + * will be called later + */ + hi = au_h_iptr(inode, bindex); + if (hi) + err = IS_IMMUTABLE(hi) ? -EROFS : 0; + } + + return err; +} + +int au_test_h_perm(struct inode *h_inode, int mask) +{ + if (uid_eq(current_fsuid(), GLOBAL_ROOT_UID)) + return 0; + return inode_permission(h_inode, mask); +} + +int au_test_h_perm_sio(struct inode *h_inode, int mask) +{ + if (au_test_nfs(h_inode->i_sb) + && (mask & MAY_WRITE) + && S_ISDIR(h_inode->i_mode)) + mask |= MAY_READ; /* force permission check */ + return au_test_h_perm(h_inode, mask); +} diff -Naur kernel.21.3.orig/fs/aufs/inode.h kernel.21.3.new/fs/aufs/inode.h --- kernel.21.3.orig/fs/aufs/inode.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/inode.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,633 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * inode operations + */ + +#ifndef __AUFS_INODE_H__ +#define __AUFS_INODE_H__ + +#ifdef __KERNEL__ + +#include +#include "rwsem.h" + +struct vfsmount; + +struct au_hnotify { +#ifdef CONFIG_AUFS_HNOTIFY +#ifdef CONFIG_AUFS_HFSNOTIFY + /* never use fsnotify_add_vfsmount_mark() */ + struct fsnotify_mark hn_mark; +#endif + struct inode *hn_aufs_inode; /* no get/put */ +#endif +} ____cacheline_aligned_in_smp; + +struct au_hinode { + struct inode *hi_inode; + aufs_bindex_t hi_id; +#ifdef CONFIG_AUFS_HNOTIFY + struct au_hnotify *hi_notify; +#endif + + /* reference to the copied-up whiteout with get/put */ + struct dentry *hi_whdentry; +}; + +/* ig_flags */ +#define AuIG_HALF_REFRESHED 1 +#define au_ig_ftest(flags, name) ((flags) & AuIG_##name) +#define au_ig_fset(flags, name) \ + do { (flags) |= AuIG_##name; } while (0) +#define au_ig_fclr(flags, name) \ + do { (flags) &= ~AuIG_##name; } while (0) + +struct au_iigen { + __u32 ig_generation, ig_flags; +}; + +struct au_vdir; +struct au_iinfo { + spinlock_t ii_genspin; + struct au_iigen ii_generation; + struct super_block *ii_hsb1; /* no get/put */ + + struct au_rwsem ii_rwsem; + aufs_bindex_t ii_bstart, ii_bend; + __u32 ii_higen; + struct au_hinode *ii_hinode; + struct au_vdir *ii_vdir; +}; + +struct au_icntnr { + struct au_iinfo iinfo; + struct inode vfs_inode; +} ____cacheline_aligned_in_smp; + +/* au_pin flags */ +#define AuPin_DI_LOCKED 1 +#define AuPin_MNT_WRITE (1 << 1) +#define au_ftest_pin(flags, name) ((flags) & AuPin_##name) +#define au_fset_pin(flags, name) \ + do { (flags) |= AuPin_##name; } while (0) +#define au_fclr_pin(flags, name) \ + do { (flags) &= ~AuPin_##name; } while (0) + +struct au_pin { + /* input */ + struct dentry *dentry; + unsigned int udba; + unsigned char lsc_di, lsc_hi, flags; + aufs_bindex_t bindex; + + /* output */ + struct dentry *parent; + struct au_hinode *hdir; + struct vfsmount *h_mnt; + + /* temporary unlock/relock for copyup */ + struct dentry *h_dentry, *h_parent; + struct au_branch *br; + struct task_struct *task; +}; + +void au_pin_hdir_unlock(struct au_pin *p); +int au_pin_hdir_lock(struct au_pin *p); +int au_pin_hdir_relock(struct au_pin *p); +void au_pin_hdir_set_owner(struct au_pin *p, struct task_struct *task); +void au_pin_hdir_acquire_nest(struct au_pin *p); +void au_pin_hdir_release(struct au_pin *p); + +/* ---------------------------------------------------------------------- */ + +static inline struct au_iinfo *au_ii(struct inode *inode) +{ + struct au_iinfo *iinfo; + + iinfo = &(container_of(inode, struct au_icntnr, vfs_inode)->iinfo); + if (iinfo->ii_hinode) + return iinfo; + return NULL; /* debugging bad_inode case */ +} + +/* ---------------------------------------------------------------------- */ + +/* inode.c */ +struct inode *au_igrab(struct inode *inode); +int au_refresh_hinode_self(struct inode *inode); +int au_refresh_hinode(struct inode *inode, struct dentry *dentry); +int au_ino(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino, + unsigned int d_type, ino_t *ino); +struct inode *au_new_inode(struct dentry *dentry, int must_new); +int au_test_ro(struct super_block *sb, aufs_bindex_t bindex, + struct inode *inode); +int au_test_h_perm(struct inode *h_inode, int mask); +int au_test_h_perm_sio(struct inode *h_inode, int mask); + +static inline int au_wh_ino(struct super_block *sb, aufs_bindex_t bindex, + ino_t h_ino, unsigned int d_type, ino_t *ino) +{ +#ifdef CONFIG_AUFS_SHWH + return au_ino(sb, bindex, h_ino, d_type, ino); +#else + return 0; +#endif +} + +/* i_op.c */ +extern struct inode_operations aufs_iop, aufs_symlink_iop, aufs_dir_iop; + +/* au_wr_dir flags */ +#define AuWrDir_ADD_ENTRY 1 +#define AuWrDir_TMP_WHENTRY (1 << 1) +#define AuWrDir_ISDIR (1 << 2) +#define au_ftest_wrdir(flags, name) ((flags) & AuWrDir_##name) +#define au_fset_wrdir(flags, name) \ + do { (flags) |= AuWrDir_##name; } while (0) +#define au_fclr_wrdir(flags, name) \ + do { (flags) &= ~AuWrDir_##name; } while (0) + +struct au_wr_dir_args { + aufs_bindex_t force_btgt; + unsigned char flags; +}; +int au_wr_dir(struct dentry *dentry, struct dentry *src_dentry, + struct au_wr_dir_args *args); + +struct dentry *au_pinned_h_parent(struct au_pin *pin); +void au_pin_init(struct au_pin *pin, struct dentry *dentry, + aufs_bindex_t bindex, int lsc_di, int lsc_hi, + unsigned int udba, unsigned char flags); +int au_pin(struct au_pin *pin, struct dentry *dentry, aufs_bindex_t bindex, + unsigned int udba, unsigned char flags) __must_check; +int au_do_pin(struct au_pin *pin) __must_check; +void au_unpin(struct au_pin *pin); +int au_reval_for_attr(struct dentry *dentry, unsigned int sigen); + +#define AuIcpup_DID_CPUP 1 +#define au_ftest_icpup(flags, name) ((flags) & AuIcpup_##name) +#define au_fset_icpup(flags, name) \ + do { (flags) |= AuIcpup_##name; } while (0) +#define au_fclr_icpup(flags, name) \ + do { (flags) &= ~AuIcpup_##name; } while (0) + +struct au_icpup_args { + unsigned char flags; + unsigned char pin_flags; + aufs_bindex_t btgt; + unsigned int udba; + struct au_pin pin; + struct path h_path; + struct inode *h_inode; +}; + +int au_pin_and_icpup(struct dentry *dentry, struct iattr *ia, + struct au_icpup_args *a); + +int au_h_path_getattr(struct dentry *dentry, int force, struct path *h_path); + +/* i_op_add.c */ +int au_may_add(struct dentry *dentry, aufs_bindex_t bindex, + struct dentry *h_parent, int isdir); +int aufs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, + dev_t dev); +int aufs_symlink(struct inode *dir, struct dentry *dentry, const char *symname); +int aufs_create(struct inode *dir, struct dentry *dentry, umode_t mode, + bool want_excl); +int aufs_link(struct dentry *src_dentry, struct inode *dir, + struct dentry *dentry); +int aufs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode); + +/* i_op_del.c */ +int au_wr_dir_need_wh(struct dentry *dentry, int isdir, aufs_bindex_t *bcpup); +int au_may_del(struct dentry *dentry, aufs_bindex_t bindex, + struct dentry *h_parent, int isdir); +int aufs_unlink(struct inode *dir, struct dentry *dentry); +int aufs_rmdir(struct inode *dir, struct dentry *dentry); + +/* i_op_ren.c */ +int au_wbr(struct dentry *dentry, aufs_bindex_t btgt); +int aufs_rename(struct inode *src_dir, struct dentry *src_dentry, + struct inode *dir, struct dentry *dentry); + +/* iinfo.c */ +struct inode *au_h_iptr(struct inode *inode, aufs_bindex_t bindex); +void au_hiput(struct au_hinode *hinode); +void au_set_hi_wh(struct inode *inode, aufs_bindex_t bindex, + struct dentry *h_wh); +unsigned int au_hi_flags(struct inode *inode, int isdir); + +/* hinode flags */ +#define AuHi_XINO 1 +#define AuHi_HNOTIFY (1 << 1) +#define au_ftest_hi(flags, name) ((flags) & AuHi_##name) +#define au_fset_hi(flags, name) \ + do { (flags) |= AuHi_##name; } while (0) +#define au_fclr_hi(flags, name) \ + do { (flags) &= ~AuHi_##name; } while (0) + +#ifndef CONFIG_AUFS_HNOTIFY +#undef AuHi_HNOTIFY +#define AuHi_HNOTIFY 0 +#endif + +void au_set_h_iptr(struct inode *inode, aufs_bindex_t bindex, + struct inode *h_inode, unsigned int flags); + +void au_update_iigen(struct inode *inode, int half); +void au_update_ibrange(struct inode *inode, int do_put_zero); + +void au_icntnr_init_once(void *_c); +int au_iinfo_init(struct inode *inode); +void au_iinfo_fin(struct inode *inode); +int au_ii_realloc(struct au_iinfo *iinfo, int nbr); + +#ifdef CONFIG_PROC_FS +/* plink.c */ +int au_plink_maint(struct super_block *sb, int flags); +void au_plink_maint_leave(struct au_sbinfo *sbinfo); +int au_plink_maint_enter(struct super_block *sb); +#ifdef CONFIG_AUFS_DEBUG +void au_plink_list(struct super_block *sb); +#else +AuStubVoid(au_plink_list, struct super_block *sb) +#endif +int au_plink_test(struct inode *inode); +struct dentry *au_plink_lkup(struct inode *inode, aufs_bindex_t bindex); +void au_plink_append(struct inode *inode, aufs_bindex_t bindex, + struct dentry *h_dentry); +void au_plink_put(struct super_block *sb, int verbose); +void au_plink_clean(struct super_block *sb, int verbose); +void au_plink_half_refresh(struct super_block *sb, aufs_bindex_t br_id); +#else +AuStubInt0(au_plink_maint, struct super_block *sb, int flags); +AuStubVoid(au_plink_maint_leave, struct au_sbinfo *sbinfo); +AuStubInt0(au_plink_maint_enter, struct super_block *sb); +AuStubVoid(au_plink_list, struct super_block *sb); +AuStubInt0(au_plink_test, struct inode *inode); +AuStub(struct dentry *, au_plink_lkup, return NULL, + struct inode *inode, aufs_bindex_t bindex); +AuStubVoid(au_plink_append, struct inode *inode, aufs_bindex_t bindex, + struct dentry *h_dentry); +AuStubVoid(au_plink_put, struct super_block *sb, int verbose); +AuStubVoid(au_plink_clean, struct super_block *sb, int verbose); +AuStubVoid(au_plink_half_refresh, struct super_block *sb, aufs_bindex_t br_id); +#endif /* CONFIG_PROC_FS */ + +#ifdef CONFIG_AUFS_XATTR +/* xattr.c */ +int au_cpup_xattr(struct dentry *h_dst, struct dentry *h_src, int ignore_flags); +ssize_t aufs_listxattr(struct dentry *dentry, char *list, size_t size); +ssize_t aufs_getxattr(struct dentry *dentry, const char *name, void *value, + size_t size); +int aufs_setxattr(struct dentry *dentry, const char *name, const void *value, + size_t size, int flags); +int aufs_removexattr(struct dentry *dentry, const char *name); + +/* void au_xattr_init(struct super_block *sb); */ +#else +AuStubInt0(au_cpup_xattr, struct dentry *h_dst, struct dentry *h_src, + int ignore_flags); +/* AuStubVoid(au_xattr_init, struct super_block *sb); */ +#endif + + +/* ---------------------------------------------------------------------- */ + +/* lock subclass for iinfo */ +enum { + AuLsc_II_CHILD, /* child first */ + AuLsc_II_CHILD2, /* rename(2), link(2), and cpup at hnotify */ + AuLsc_II_CHILD3, /* copyup dirs */ + AuLsc_II_PARENT, /* see AuLsc_I_PARENT in vfsub.h */ + AuLsc_II_PARENT2, + AuLsc_II_PARENT3, /* copyup dirs */ + AuLsc_II_NEW_CHILD +}; + +/* + * ii_read_lock_child, ii_write_lock_child, + * ii_read_lock_child2, ii_write_lock_child2, + * ii_read_lock_child3, ii_write_lock_child3, + * ii_read_lock_parent, ii_write_lock_parent, + * ii_read_lock_parent2, ii_write_lock_parent2, + * ii_read_lock_parent3, ii_write_lock_parent3, + * ii_read_lock_new_child, ii_write_lock_new_child, + */ +#define AuReadLockFunc(name, lsc) \ +static inline void ii_read_lock_##name(struct inode *i) \ +{ \ + au_rw_read_lock_nested(&au_ii(i)->ii_rwsem, AuLsc_II_##lsc); \ +} + +#define AuWriteLockFunc(name, lsc) \ +static inline void ii_write_lock_##name(struct inode *i) \ +{ \ + au_rw_write_lock_nested(&au_ii(i)->ii_rwsem, AuLsc_II_##lsc); \ +} + +#define AuRWLockFuncs(name, lsc) \ + AuReadLockFunc(name, lsc) \ + AuWriteLockFunc(name, lsc) + +AuRWLockFuncs(child, CHILD); +AuRWLockFuncs(child2, CHILD2); +AuRWLockFuncs(child3, CHILD3); +AuRWLockFuncs(parent, PARENT); +AuRWLockFuncs(parent2, PARENT2); +AuRWLockFuncs(parent3, PARENT3); +AuRWLockFuncs(new_child, NEW_CHILD); + +#undef AuReadLockFunc +#undef AuWriteLockFunc +#undef AuRWLockFuncs + +/* + * ii_read_unlock, ii_write_unlock, ii_downgrade_lock + */ +AuSimpleUnlockRwsemFuncs(ii, struct inode *i, &au_ii(i)->ii_rwsem); + +#define IiMustNoWaiters(i) AuRwMustNoWaiters(&au_ii(i)->ii_rwsem) +#define IiMustAnyLock(i) AuRwMustAnyLock(&au_ii(i)->ii_rwsem) +#define IiMustWriteLock(i) AuRwMustWriteLock(&au_ii(i)->ii_rwsem) + +/* ---------------------------------------------------------------------- */ + +static inline void au_icntnr_init(struct au_icntnr *c) +{ +#ifdef CONFIG_AUFS_DEBUG + c->vfs_inode.i_mode = 0; +#endif +} + +static inline unsigned int au_iigen(struct inode *inode, struct au_iigen *iigen) +{ + unsigned int gen; + struct au_iinfo *iinfo; + + iinfo = au_ii(inode); + spin_lock(&iinfo->ii_genspin); + if (iigen) + *iigen = iinfo->ii_generation; + gen = iinfo->ii_generation.ig_generation; + spin_unlock(&iinfo->ii_genspin); + + return gen; +} + +/* tiny test for inode number */ +/* tmpfs generation is too rough */ +static inline int au_test_higen(struct inode *inode, struct inode *h_inode) +{ + struct au_iinfo *iinfo; + + iinfo = au_ii(inode); + AuRwMustAnyLock(&iinfo->ii_rwsem); + return !(iinfo->ii_hsb1 == h_inode->i_sb + && iinfo->ii_higen == h_inode->i_generation); +} + +static inline void au_iigen_dec(struct inode *inode) +{ + struct au_iinfo *iinfo; + + iinfo = au_ii(inode); + spin_lock(&iinfo->ii_genspin); + iinfo->ii_generation.ig_generation--; + spin_unlock(&iinfo->ii_genspin); +} + +static inline int au_iigen_test(struct inode *inode, unsigned int sigen) +{ + int err; + + err = 0; + if (unlikely(inode && au_iigen(inode, NULL) != sigen)) + err = -EIO; + + return err; +} + +/* ---------------------------------------------------------------------- */ + +static inline aufs_bindex_t au_ii_br_id(struct inode *inode, + aufs_bindex_t bindex) +{ + IiMustAnyLock(inode); + return au_ii(inode)->ii_hinode[0 + bindex].hi_id; +} + +static inline aufs_bindex_t au_ibstart(struct inode *inode) +{ + IiMustAnyLock(inode); + return au_ii(inode)->ii_bstart; +} + +static inline aufs_bindex_t au_ibend(struct inode *inode) +{ + IiMustAnyLock(inode); + return au_ii(inode)->ii_bend; +} + +static inline struct au_vdir *au_ivdir(struct inode *inode) +{ + IiMustAnyLock(inode); + return au_ii(inode)->ii_vdir; +} + +static inline struct dentry *au_hi_wh(struct inode *inode, aufs_bindex_t bindex) +{ + IiMustAnyLock(inode); + return au_ii(inode)->ii_hinode[0 + bindex].hi_whdentry; +} + +static inline void au_set_ibstart(struct inode *inode, aufs_bindex_t bindex) +{ + IiMustWriteLock(inode); + au_ii(inode)->ii_bstart = bindex; +} + +static inline void au_set_ibend(struct inode *inode, aufs_bindex_t bindex) +{ + IiMustWriteLock(inode); + au_ii(inode)->ii_bend = bindex; +} + +static inline void au_set_ivdir(struct inode *inode, struct au_vdir *vdir) +{ + IiMustWriteLock(inode); + au_ii(inode)->ii_vdir = vdir; +} + +static inline struct au_hinode *au_hi(struct inode *inode, aufs_bindex_t bindex) +{ + IiMustAnyLock(inode); + return au_ii(inode)->ii_hinode + bindex; +} + +/* ---------------------------------------------------------------------- */ + +static inline struct dentry *au_pinned_parent(struct au_pin *pin) +{ + if (pin) + return pin->parent; + return NULL; +} + +static inline struct inode *au_pinned_h_dir(struct au_pin *pin) +{ + if (pin && pin->hdir) + return pin->hdir->hi_inode; + return NULL; +} + +static inline struct au_hinode *au_pinned_hdir(struct au_pin *pin) +{ + if (pin) + return pin->hdir; + return NULL; +} + +static inline void au_pin_set_dentry(struct au_pin *pin, struct dentry *dentry) +{ + if (pin) + pin->dentry = dentry; +} + +static inline void au_pin_set_parent_lflag(struct au_pin *pin, + unsigned char lflag) +{ + if (pin) { + if (lflag) + au_fset_pin(pin->flags, DI_LOCKED); + else + au_fclr_pin(pin->flags, DI_LOCKED); + } +} + +static inline void au_pin_set_parent(struct au_pin *pin, struct dentry *parent) +{ + if (pin) { + dput(pin->parent); + pin->parent = dget(parent); + } +} + +/* ---------------------------------------------------------------------- */ + +struct au_branch; +#ifdef CONFIG_AUFS_HNOTIFY +struct au_hnotify_op { + void (*ctl)(struct au_hinode *hinode, int do_set); + int (*alloc)(struct au_hinode *hinode); + + /* + * if it returns true, the the caller should free hinode->hi_notify, + * otherwise ->free() frees it. + */ + int (*free)(struct au_hinode *hinode, + struct au_hnotify *hn) __must_check; + + void (*fin)(void); + int (*init)(void); + + int (*reset_br)(unsigned int udba, struct au_branch *br, int perm); + void (*fin_br)(struct au_branch *br); + int (*init_br)(struct au_branch *br, int perm); +}; + +/* hnotify.c */ +int au_hn_alloc(struct au_hinode *hinode, struct inode *inode); +void au_hn_free(struct au_hinode *hinode); +void au_hn_ctl(struct au_hinode *hinode, int do_set); +void au_hn_reset(struct inode *inode, unsigned int flags); +int au_hnotify(struct inode *h_dir, struct au_hnotify *hnotify, u32 mask, + struct qstr *h_child_qstr, struct inode *h_child_inode); +int au_hnotify_reset_br(unsigned int udba, struct au_branch *br, int perm); +int au_hnotify_init_br(struct au_branch *br, int perm); +void au_hnotify_fin_br(struct au_branch *br); +int __init au_hnotify_init(void); +void au_hnotify_fin(void); + +/* hfsnotify.c */ +extern const struct au_hnotify_op au_hnotify_op; + +static inline +void au_hn_init(struct au_hinode *hinode) +{ + hinode->hi_notify = NULL; +} + +static inline struct au_hnotify *au_hn(struct au_hinode *hinode) +{ + return hinode->hi_notify; +} + +#else +AuStub(int, au_hn_alloc, return -EOPNOTSUPP, + struct au_hinode *hinode __maybe_unused, + struct inode *inode __maybe_unused) +AuStub(struct au_hnotify *, au_hn, return NULL, struct au_hinode *hinode) +AuStubVoid(au_hn_free, struct au_hinode *hinode __maybe_unused) +AuStubVoid(au_hn_ctl, struct au_hinode *hinode __maybe_unused, + int do_set __maybe_unused) +AuStubVoid(au_hn_reset, struct inode *inode __maybe_unused, + unsigned int flags __maybe_unused) +AuStubInt0(au_hnotify_reset_br, unsigned int udba __maybe_unused, + struct au_branch *br __maybe_unused, + int perm __maybe_unused) +AuStubInt0(au_hnotify_init_br, struct au_branch *br __maybe_unused, + int perm __maybe_unused) +AuStubVoid(au_hnotify_fin_br, struct au_branch *br __maybe_unused) +AuStubInt0(__init au_hnotify_init, void) +AuStubVoid(au_hnotify_fin, void) +AuStubVoid(au_hn_init, struct au_hinode *hinode __maybe_unused) +#endif /* CONFIG_AUFS_HNOTIFY */ + +static inline void au_hn_suspend(struct au_hinode *hdir) +{ + au_hn_ctl(hdir, /*do_set*/0); +} + +static inline void au_hn_resume(struct au_hinode *hdir) +{ + au_hn_ctl(hdir, /*do_set*/1); +} + +static inline void au_hn_imtx_lock(struct au_hinode *hdir) +{ + mutex_lock(&hdir->hi_inode->i_mutex); + au_hn_suspend(hdir); +} + +static inline void au_hn_imtx_lock_nested(struct au_hinode *hdir, + unsigned int sc __maybe_unused) +{ + mutex_lock_nested(&hdir->hi_inode->i_mutex, sc); + au_hn_suspend(hdir); +} + +static inline void au_hn_imtx_unlock(struct au_hinode *hdir) +{ + au_hn_resume(hdir); + mutex_unlock(&hdir->hi_inode->i_mutex); +} + +#endif /* __KERNEL__ */ +#endif /* __AUFS_INODE_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/ioctl.c kernel.21.3.new/fs/aufs/ioctl.c --- kernel.21.3.orig/fs/aufs/ioctl.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/ioctl.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,219 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * ioctl + * plink-management and readdir in userspace. + * assist the pathconf(3) wrapper library. + * move-down + * File-based Hierarchical Storage Management. + */ + +#include +#include +#include "aufs.h" + +static int au_wbr_fd(struct path *path, struct aufs_wbr_fd __user *arg) +{ + int err, fd; + aufs_bindex_t wbi, bindex, bend; + struct file *h_file; + struct super_block *sb; + struct dentry *root; + struct au_branch *br; + struct aufs_wbr_fd wbrfd = { + .oflags = au_dir_roflags, + .brid = -1 + }; + const int valid = O_RDONLY | O_NONBLOCK | O_LARGEFILE | O_DIRECTORY + | O_NOATIME | O_CLOEXEC; + + AuDebugOn(wbrfd.oflags & ~valid); + + if (arg) { + err = copy_from_user(&wbrfd, arg, sizeof(wbrfd)); + if (unlikely(err)) { + err = -EFAULT; + goto out; + } + + err = -EINVAL; + AuDbg("wbrfd{0%o, %d}\n", wbrfd.oflags, wbrfd.brid); + wbrfd.oflags |= au_dir_roflags; + AuDbg("0%o\n", wbrfd.oflags); + if (unlikely(wbrfd.oflags & ~valid)) + goto out; + } + + fd = get_unused_fd(); + err = fd; + if (unlikely(fd < 0)) + goto out; + + h_file = ERR_PTR(-EINVAL); + wbi = 0; + br = NULL; + sb = path->dentry->d_sb; + root = sb->s_root; + aufs_read_lock(root, AuLock_IR); + bend = au_sbend(sb); + if (wbrfd.brid >= 0) { + wbi = au_br_index(sb, wbrfd.brid); + if (unlikely(wbi < 0 || wbi > bend)) + goto out_unlock; + } + + h_file = ERR_PTR(-ENOENT); + br = au_sbr(sb, wbi); + if (!au_br_writable(br->br_perm)) { + if (arg) + goto out_unlock; + + bindex = wbi + 1; + wbi = -1; + for (; bindex <= bend; bindex++) { + br = au_sbr(sb, bindex); + if (au_br_writable(br->br_perm)) { + wbi = bindex; + br = au_sbr(sb, wbi); + break; + } + } + } + AuDbg("wbi %d\n", wbi); + if (wbi >= 0) + h_file = au_h_open(root, wbi, wbrfd.oflags, NULL, + /*force_wr*/0); + +out_unlock: + aufs_read_unlock(root, AuLock_IR); + err = PTR_ERR(h_file); + if (IS_ERR(h_file)) + goto out_fd; + + atomic_dec(&br->br_count); /* cf. au_h_open() */ + fd_install(fd, h_file); + err = fd; + goto out; /* success */ + +out_fd: + put_unused_fd(fd); +out: + AuTraceErr(err); + return err; +} + +/* ---------------------------------------------------------------------- */ + +long aufs_ioctl_dir(struct file *file, unsigned int cmd, unsigned long arg) +{ + long err; + struct dentry *dentry; + + switch (cmd) { + case AUFS_CTL_RDU: + case AUFS_CTL_RDU_INO: + err = au_rdu_ioctl(file, cmd, arg); + break; + + case AUFS_CTL_WBR_FD: + err = au_wbr_fd(&file->f_path, (void __user *)arg); + break; + + case AUFS_CTL_IBUSY: + err = au_ibusy_ioctl(file, arg); + break; + + case AUFS_CTL_BRINFO: + err = au_brinfo_ioctl(file, arg); + break; + + case AUFS_CTL_FHSM_FD: + dentry = file->f_dentry; + if (IS_ROOT(dentry)) + err = au_fhsm_fd(dentry->d_sb, arg); + else + err = -ENOTTY; + break; + + default: + /* do not call the lower */ + AuDbg("0x%x\n", cmd); + err = -ENOTTY; + } + + AuTraceErr(err); + return err; +} + +long aufs_ioctl_nondir(struct file *file, unsigned int cmd, unsigned long arg) +{ + long err; + + switch (cmd) { + case AUFS_CTL_MVDOWN: + err = au_mvdown(file->f_dentry, (void __user *)arg); + break; + + case AUFS_CTL_WBR_FD: + err = au_wbr_fd(&file->f_path, (void __user *)arg); + break; + + default: + /* do not call the lower */ + AuDbg("0x%x\n", cmd); + err = -ENOTTY; + } + + AuTraceErr(err); + return err; +} + +#ifdef CONFIG_COMPAT +long aufs_compat_ioctl_dir(struct file *file, unsigned int cmd, + unsigned long arg) +{ + long err; + + switch (cmd) { + case AUFS_CTL_RDU: + case AUFS_CTL_RDU_INO: + err = au_rdu_compat_ioctl(file, cmd, arg); + break; + + case AUFS_CTL_IBUSY: + err = au_ibusy_compat_ioctl(file, arg); + break; + + case AUFS_CTL_BRINFO: + err = au_brinfo_compat_ioctl(file, arg); + break; + + default: + err = aufs_ioctl_dir(file, cmd, arg); + } + + AuTraceErr(err); + return err; +} + +long aufs_compat_ioctl_nondir(struct file *file, unsigned int cmd, + unsigned long arg) +{ + return aufs_ioctl_nondir(file, cmd, (unsigned long)compat_ptr(arg)); +} +#endif diff -Naur kernel.21.3.orig/fs/aufs/i_op_add.c kernel.21.3.new/fs/aufs/i_op_add.c --- kernel.21.3.orig/fs/aufs/i_op_add.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/i_op_add.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,750 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * inode operations (add entry) + */ + +#include "aufs.h" + +/* + * final procedure of adding a new entry, except link(2). + * remove whiteout, instantiate, copyup the parent dir's times and size + * and update version. + * if it failed, re-create the removed whiteout. + */ +static int epilog(struct inode *dir, aufs_bindex_t bindex, + struct dentry *wh_dentry, struct dentry *dentry) +{ + int err, rerr; + aufs_bindex_t bwh; + struct path h_path; + struct super_block *sb; + struct inode *inode, *h_dir; + struct dentry *wh; + + bwh = -1; + sb = dir->i_sb; + if (wh_dentry) { + h_dir = wh_dentry->d_parent->d_inode; /* dir inode is locked */ + IMustLock(h_dir); + AuDebugOn(au_h_iptr(dir, bindex) != h_dir); + bwh = au_dbwh(dentry); + h_path.dentry = wh_dentry; + h_path.mnt = au_sbr_mnt(sb, bindex); + err = au_wh_unlink_dentry(au_h_iptr(dir, bindex), &h_path, + dentry); + if (unlikely(err)) + goto out; + } + + inode = au_new_inode(dentry, /*must_new*/1); + if (!IS_ERR(inode)) { + d_instantiate(dentry, inode); + dir = dentry->d_parent->d_inode; /* dir inode is locked */ + IMustLock(dir); + if (au_ibstart(dir) == au_dbstart(dentry)) + au_cpup_attr_timesizes(dir); + dir->i_version++; + au_fhsm_wrote(sb, bindex, /*force*/0); + return 0; /* success */ + } + + err = PTR_ERR(inode); + if (!wh_dentry) + goto out; + + /* revert */ + /* dir inode is locked */ + wh = au_wh_create(dentry, bwh, wh_dentry->d_parent); + rerr = PTR_ERR(wh); + if (IS_ERR(wh)) { + AuIOErr("%.*s reverting whiteout failed(%d, %d)\n", + AuDLNPair(dentry), err, rerr); + err = -EIO; + } else + dput(wh); + +out: + return err; +} + +static int au_d_may_add(struct dentry *dentry) +{ + int err; + + err = 0; + if (unlikely(d_unhashed(dentry))) + err = -ENOENT; + if (unlikely(dentry->d_inode)) + err = -EEXIST; + return err; +} + +/* + * simple tests for the adding inode operations. + * following the checks in vfs, plus the parent-child relationship. + */ +int au_may_add(struct dentry *dentry, aufs_bindex_t bindex, + struct dentry *h_parent, int isdir) +{ + int err; + umode_t h_mode; + struct dentry *h_dentry; + struct inode *h_inode; + + err = -ENAMETOOLONG; + if (unlikely(dentry->d_name.len > AUFS_MAX_NAMELEN)) + goto out; + + h_dentry = au_h_dptr(dentry, bindex); + h_inode = h_dentry->d_inode; + if (!dentry->d_inode) { + err = -EEXIST; + if (unlikely(h_inode)) + goto out; + } else { + /* rename(2) case */ + err = -EIO; + if (unlikely(!h_inode || !h_inode->i_nlink)) + goto out; + + h_mode = h_inode->i_mode; + if (!isdir) { + err = -EISDIR; + if (unlikely(S_ISDIR(h_mode))) + goto out; + } else if (unlikely(!S_ISDIR(h_mode))) { + err = -ENOTDIR; + goto out; + } + } + + err = 0; + /* expected parent dir is locked */ + if (unlikely(h_parent != h_dentry->d_parent)) + err = -EIO; + +out: + AuTraceErr(err); + return err; +} + +/* + * initial procedure of adding a new entry. + * prepare writable branch and the parent dir, lock it, + * and lookup whiteout for the new entry. + */ +static struct dentry* +lock_hdir_lkup_wh(struct dentry *dentry, struct au_dtime *dt, + struct dentry *src_dentry, struct au_pin *pin, + struct au_wr_dir_args *wr_dir_args) +{ + struct dentry *wh_dentry, *h_parent; + struct super_block *sb; + struct au_branch *br; + int err; + unsigned int udba; + aufs_bindex_t bcpup; + + AuDbg("%.*s\n", AuDLNPair(dentry)); + + err = au_wr_dir(dentry, src_dentry, wr_dir_args); + bcpup = err; + wh_dentry = ERR_PTR(err); + if (unlikely(err < 0)) + goto out; + + sb = dentry->d_sb; + udba = au_opt_udba(sb); + err = au_pin(pin, dentry, bcpup, udba, + AuPin_DI_LOCKED | AuPin_MNT_WRITE); + wh_dentry = ERR_PTR(err); + if (unlikely(err)) + goto out; + + h_parent = au_pinned_h_parent(pin); + if (udba != AuOpt_UDBA_NONE + && au_dbstart(dentry) == bcpup) + err = au_may_add(dentry, bcpup, h_parent, + au_ftest_wrdir(wr_dir_args->flags, ISDIR)); + else if (unlikely(dentry->d_name.len > AUFS_MAX_NAMELEN)) + err = -ENAMETOOLONG; + wh_dentry = ERR_PTR(err); + if (unlikely(err)) + goto out_unpin; + + br = au_sbr(sb, bcpup); + if (dt) { + struct path tmp = { + .dentry = h_parent, + .mnt = au_br_mnt(br) + }; + au_dtime_store(dt, au_pinned_parent(pin), &tmp); + } + + wh_dentry = NULL; + if (bcpup != au_dbwh(dentry)) + goto out; /* success */ + + wh_dentry = au_wh_lkup(h_parent, &dentry->d_name, br); + +out_unpin: + if (IS_ERR(wh_dentry)) + au_unpin(pin); +out: + return wh_dentry; +} + +/* ---------------------------------------------------------------------- */ + +enum { Mknod, Symlink, Creat }; +struct simple_arg { + int type; + union { + struct { + umode_t mode; + bool want_excl; + } c; + struct { + const char *symname; + } s; + struct { + umode_t mode; + dev_t dev; + } m; + } u; +}; + +static int add_simple(struct inode *dir, struct dentry *dentry, + struct simple_arg *arg) +{ + int err, rerr; + aufs_bindex_t bstart; + unsigned char created; + struct dentry *wh_dentry, *parent; + struct inode *h_dir; + /* to reuduce stack size */ + struct { + struct au_dtime dt; + struct au_pin pin; + struct path h_path; + struct au_wr_dir_args wr_dir_args; + } *a; + + AuDbg("%.*s\n", AuDLNPair(dentry)); + IMustLock(dir); + + err = -ENOMEM; + a = kmalloc(sizeof(*a), GFP_NOFS); + if (unlikely(!a)) + goto out; + a->wr_dir_args.force_btgt = -1; + a->wr_dir_args.flags = AuWrDir_ADD_ENTRY; + + parent = dentry->d_parent; /* dir inode is locked */ + err = aufs_read_lock(dentry, AuLock_DW | AuLock_GEN); + if (unlikely(err)) + goto out_free; + err = au_d_may_add(dentry); + if (unlikely(err)) + goto out_unlock; + di_write_lock_parent(parent); + wh_dentry = lock_hdir_lkup_wh(dentry, &a->dt, /*src_dentry*/NULL, + &a->pin, &a->wr_dir_args); + err = PTR_ERR(wh_dentry); + if (IS_ERR(wh_dentry)) + goto out_parent; + + bstart = au_dbstart(dentry); + a->h_path.dentry = au_h_dptr(dentry, bstart); + a->h_path.mnt = au_sbr_mnt(dentry->d_sb, bstart); + h_dir = au_pinned_h_dir(&a->pin); + switch (arg->type) { + case Creat: + err = vfsub_create(h_dir, &a->h_path, arg->u.c.mode, + arg->u.c.want_excl); + break; + case Symlink: + err = vfsub_symlink(h_dir, &a->h_path, arg->u.s.symname); + break; + case Mknod: + err = vfsub_mknod(h_dir, &a->h_path, arg->u.m.mode, + arg->u.m.dev); + break; + default: + BUG(); + } + created = !err; + if (!err) + err = epilog(dir, bstart, wh_dentry, dentry); + + /* revert */ + if (unlikely(created && err && a->h_path.dentry->d_inode)) { + rerr = vfsub_unlink(h_dir, &a->h_path, /*force*/0); + if (rerr) { + AuIOErr("%.*s revert failure(%d, %d)\n", + AuDLNPair(dentry), err, rerr); + err = -EIO; + } + au_dtime_revert(&a->dt); + } + + au_unpin(&a->pin); + dput(wh_dentry); + +out_parent: + di_write_unlock(parent); +out_unlock: + if (unlikely(err)) { + au_update_dbstart(dentry); + d_drop(dentry); + } + aufs_read_unlock(dentry, AuLock_DW); +out_free: + kfree(a); +out: + return err; +} + +int aufs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, + dev_t dev) +{ + struct simple_arg arg = { + .type = Mknod, + .u.m = { + .mode = mode, + .dev = dev + } + }; + return add_simple(dir, dentry, &arg); +} + +int aufs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) +{ + struct simple_arg arg = { + .type = Symlink, + .u.s.symname = symname + }; + return add_simple(dir, dentry, &arg); +} + +int aufs_create(struct inode *dir, struct dentry *dentry, umode_t mode, + bool want_excl) +{ + struct simple_arg arg = { + .type = Creat, + .u.c = { + .mode = mode, + .want_excl = want_excl + } + }; + return add_simple(dir, dentry, &arg); +} + +/* ---------------------------------------------------------------------- */ + +struct au_link_args { + aufs_bindex_t bdst, bsrc; + struct au_pin pin; + struct path h_path; + struct dentry *src_parent, *parent; +}; + +static int au_cpup_before_link(struct dentry *src_dentry, + struct au_link_args *a) +{ + int err; + struct dentry *h_src_dentry; + struct au_cp_generic cpg = { + .dentry = src_dentry, + .bdst = a->bdst, + .bsrc = a->bsrc, + .len = -1, + .pin = &a->pin, + .flags = AuCpup_DTIME | AuCpup_HOPEN /* | AuCpup_KEEPLINO */ + }; + + di_read_lock_parent(a->src_parent, AuLock_IR); + err = au_test_and_cpup_dirs(src_dentry, a->bdst); + if (unlikely(err)) + goto out; + + h_src_dentry = au_h_dptr(src_dentry, a->bsrc); + err = au_pin(&a->pin, src_dentry, a->bdst, + au_opt_udba(src_dentry->d_sb), + AuPin_DI_LOCKED | AuPin_MNT_WRITE); + if (unlikely(err)) + goto out; + + err = au_sio_cpup_simple(&cpg); + au_unpin(&a->pin); + +out: + di_read_unlock(a->src_parent, AuLock_IR); + return err; +} + +static int au_cpup_or_link(struct dentry *src_dentry, struct dentry *dentry, + struct au_link_args *a) +{ + int err; + unsigned char plink; + aufs_bindex_t bend; + struct dentry *h_src_dentry; + struct inode *h_inode, *inode; + struct super_block *sb; + struct file *h_file; + + plink = 0; + h_inode = NULL; + sb = src_dentry->d_sb; + inode = src_dentry->d_inode; + if (au_ibstart(inode) <= a->bdst) + h_inode = au_h_iptr(inode, a->bdst); + if (!h_inode || !h_inode->i_nlink) { + /* copyup src_dentry as the name of dentry. */ + bend = au_dbend(dentry); + if (bend < a->bsrc) + au_set_dbend(dentry, a->bsrc); + au_set_h_dptr(dentry, a->bsrc, + dget(au_h_dptr(src_dentry, a->bsrc))); + dget(a->h_path.dentry); + au_set_h_dptr(dentry, a->bdst, NULL); + AuDbg("temporary d_inode...\n"); + spin_lock(&dentry->d_lock); + dentry->d_inode = src_dentry->d_inode; /* tmp */ + spin_unlock(&dentry->d_lock); + h_file = au_h_open_pre(dentry, a->bsrc, /*force_wr*/0); + if (IS_ERR(h_file)) + err = PTR_ERR(h_file); + else { + struct au_cp_generic cpg = { + .dentry = dentry, + .bdst = a->bdst, + .bsrc = -1, + .len = -1, + .pin = &a->pin, + .flags = AuCpup_KEEPLINO + }; + err = au_sio_cpup_simple(&cpg); + au_h_open_post(dentry, a->bsrc, h_file); + if (!err) { + dput(a->h_path.dentry); + a->h_path.dentry = au_h_dptr(dentry, a->bdst); + } else + au_set_h_dptr(dentry, a->bdst, + a->h_path.dentry); + } + spin_lock(&dentry->d_lock); + dentry->d_inode = NULL; /* restore */ + spin_unlock(&dentry->d_lock); + AuDbg("temporary d_inode...done\n"); + au_set_h_dptr(dentry, a->bsrc, NULL); + au_set_dbend(dentry, bend); + } else { + /* the inode of src_dentry already exists on a.bdst branch */ + h_src_dentry = d_find_alias(h_inode); + if (!h_src_dentry && au_plink_test(inode)) { + plink = 1; + h_src_dentry = au_plink_lkup(inode, a->bdst); + err = PTR_ERR(h_src_dentry); + if (IS_ERR(h_src_dentry)) + goto out; + + if (unlikely(!h_src_dentry->d_inode)) { + dput(h_src_dentry); + h_src_dentry = NULL; + } + + } + if (h_src_dentry) { + err = vfsub_link(h_src_dentry, au_pinned_h_dir(&a->pin), + &a->h_path); + dput(h_src_dentry); + } else { + AuIOErr("no dentry found for hi%lu on b%d\n", + h_inode->i_ino, a->bdst); + err = -EIO; + } + } + + if (!err && !plink) + au_plink_append(inode, a->bdst, a->h_path.dentry); + +out: + AuTraceErr(err); + return err; +} + +int aufs_link(struct dentry *src_dentry, struct inode *dir, + struct dentry *dentry) +{ + int err, rerr; + struct au_dtime dt; + struct au_link_args *a; + struct dentry *wh_dentry, *h_src_dentry; + struct inode *inode; + struct super_block *sb; + struct au_wr_dir_args wr_dir_args = { + /* .force_btgt = -1, */ + .flags = AuWrDir_ADD_ENTRY + }; + + IMustLock(dir); + inode = src_dentry->d_inode; + IMustLock(inode); + + err = -ENOMEM; + a = kzalloc(sizeof(*a), GFP_NOFS); + if (unlikely(!a)) + goto out; + + a->parent = dentry->d_parent; /* dir inode is locked */ + err = aufs_read_and_write_lock2(dentry, src_dentry, + AuLock_NOPLM | AuLock_GEN); + if (unlikely(err)) + goto out_kfree; + err = au_d_hashed_positive(src_dentry); + if (unlikely(err)) + goto out_unlock; + err = au_d_may_add(dentry); + if (unlikely(err)) + goto out_unlock; + + a->src_parent = dget_parent(src_dentry); + wr_dir_args.force_btgt = au_ibstart(inode); + + di_write_lock_parent(a->parent); + wr_dir_args.force_btgt = au_wbr(dentry, wr_dir_args.force_btgt); + wh_dentry = lock_hdir_lkup_wh(dentry, &dt, src_dentry, &a->pin, + &wr_dir_args); + err = PTR_ERR(wh_dentry); + if (IS_ERR(wh_dentry)) + goto out_parent; + + err = 0; + sb = dentry->d_sb; + a->bdst = au_dbstart(dentry); + a->h_path.dentry = au_h_dptr(dentry, a->bdst); + a->h_path.mnt = au_sbr_mnt(sb, a->bdst); + a->bsrc = au_ibstart(inode); + h_src_dentry = au_h_d_alias(src_dentry, a->bsrc); + if (!h_src_dentry) { + a->bsrc = au_dbstart(src_dentry); + h_src_dentry = au_h_d_alias(src_dentry, a->bsrc); + AuDebugOn(!h_src_dentry); + } else if (IS_ERR(h_src_dentry)) { + err = PTR_ERR(h_src_dentry); + goto out_parent; + } + + if (au_opt_test(au_mntflags(sb), PLINK)) { + if (a->bdst < a->bsrc + /* && h_src_dentry->d_sb != a->h_path.dentry->d_sb */) + err = au_cpup_or_link(src_dentry, dentry, a); + else + err = vfsub_link(h_src_dentry, au_pinned_h_dir(&a->pin), + &a->h_path); + dput(h_src_dentry); + } else { + /* + * copyup src_dentry to the branch we process, + * and then link(2) to it. + */ + dput(h_src_dentry); + if (a->bdst < a->bsrc + /* && h_src_dentry->d_sb != a->h_path.dentry->d_sb */) { + au_unpin(&a->pin); + di_write_unlock(a->parent); + err = au_cpup_before_link(src_dentry, a); + di_write_lock_parent(a->parent); + if (!err) + err = au_pin(&a->pin, dentry, a->bdst, + au_opt_udba(sb), + AuPin_DI_LOCKED | AuPin_MNT_WRITE); + if (unlikely(err)) + goto out_wh; + } + if (!err) { + h_src_dentry = au_h_dptr(src_dentry, a->bdst); + err = -ENOENT; + if (h_src_dentry && h_src_dentry->d_inode) + err = vfsub_link(h_src_dentry, + au_pinned_h_dir(&a->pin), + &a->h_path); + } + } + if (unlikely(err)) + goto out_unpin; + + if (wh_dentry) { + a->h_path.dentry = wh_dentry; + err = au_wh_unlink_dentry(au_pinned_h_dir(&a->pin), &a->h_path, + dentry); + if (unlikely(err)) + goto out_revert; + } + + dir->i_version++; + if (au_ibstart(dir) == au_dbstart(dentry)) + au_cpup_attr_timesizes(dir); + inc_nlink(inode); + inode->i_ctime = dir->i_ctime; + d_instantiate(dentry, au_igrab(inode)); + if (d_unhashed(a->h_path.dentry)) + /* some filesystem calls d_drop() */ + d_drop(dentry); + /* some filesystems consume an inode even hardlink */ + au_fhsm_wrote(sb, a->bdst, /*force*/0); + goto out_unpin; /* success */ + +out_revert: + rerr = vfsub_unlink(au_pinned_h_dir(&a->pin), &a->h_path, /*force*/0); + if (unlikely(rerr)) { + AuIOErr("%.*s reverting failed(%d, %d)\n", + AuDLNPair(dentry), err, rerr); + err = -EIO; + } + au_dtime_revert(&dt); +out_unpin: + au_unpin(&a->pin); +out_wh: + dput(wh_dentry); +out_parent: + di_write_unlock(a->parent); + dput(a->src_parent); +out_unlock: + if (unlikely(err)) { + au_update_dbstart(dentry); + d_drop(dentry); + } + aufs_read_and_write_unlock2(dentry, src_dentry); +out_kfree: + kfree(a); +out: + AuTraceErr(err); + return err; +} + +int aufs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +{ + int err, rerr; + aufs_bindex_t bindex; + unsigned char diropq; + struct path h_path; + struct dentry *wh_dentry, *parent, *opq_dentry; + struct mutex *h_mtx; + struct super_block *sb; + struct { + struct au_pin pin; + struct au_dtime dt; + } *a; /* reduce the stack usage */ + struct au_wr_dir_args wr_dir_args = { + .force_btgt = -1, + .flags = AuWrDir_ADD_ENTRY | AuWrDir_ISDIR + }; + + IMustLock(dir); + + err = -ENOMEM; + a = kmalloc(sizeof(*a), GFP_NOFS); + if (unlikely(!a)) + goto out; + + err = aufs_read_lock(dentry, AuLock_DW | AuLock_GEN); + if (unlikely(err)) + goto out_free; + err = au_d_may_add(dentry); + if (unlikely(err)) + goto out_unlock; + + parent = dentry->d_parent; /* dir inode is locked */ + di_write_lock_parent(parent); + wh_dentry = lock_hdir_lkup_wh(dentry, &a->dt, /*src_dentry*/NULL, + &a->pin, &wr_dir_args); + err = PTR_ERR(wh_dentry); + if (IS_ERR(wh_dentry)) + goto out_parent; + + sb = dentry->d_sb; + bindex = au_dbstart(dentry); + h_path.dentry = au_h_dptr(dentry, bindex); + h_path.mnt = au_sbr_mnt(sb, bindex); + err = vfsub_mkdir(au_pinned_h_dir(&a->pin), &h_path, mode); + if (unlikely(err)) + goto out_unpin; + + /* make the dir opaque */ + diropq = 0; + h_mtx = &h_path.dentry->d_inode->i_mutex; + if (wh_dentry + || au_opt_test(au_mntflags(sb), ALWAYS_DIROPQ)) { + mutex_lock_nested(h_mtx, AuLsc_I_CHILD); + opq_dentry = au_diropq_create(dentry, bindex); + mutex_unlock(h_mtx); + err = PTR_ERR(opq_dentry); + if (IS_ERR(opq_dentry)) + goto out_dir; + dput(opq_dentry); + diropq = 1; + } + + err = epilog(dir, bindex, wh_dentry, dentry); + if (!err) { + inc_nlink(dir); + goto out_unpin; /* success */ + } + + /* revert */ + if (diropq) { + AuLabel(revert opq); + mutex_lock_nested(h_mtx, AuLsc_I_CHILD); + rerr = au_diropq_remove(dentry, bindex); + mutex_unlock(h_mtx); + if (rerr) { + AuIOErr("%.*s reverting diropq failed(%d, %d)\n", + AuDLNPair(dentry), err, rerr); + err = -EIO; + } + } + +out_dir: + AuLabel(revert dir); + rerr = vfsub_rmdir(au_pinned_h_dir(&a->pin), &h_path); + if (rerr) { + AuIOErr("%.*s reverting dir failed(%d, %d)\n", + AuDLNPair(dentry), err, rerr); + err = -EIO; + } + au_dtime_revert(&a->dt); +out_unpin: + au_unpin(&a->pin); + dput(wh_dentry); +out_parent: + di_write_unlock(parent); +out_unlock: + if (unlikely(err)) { + au_update_dbstart(dentry); + d_drop(dentry); + } + aufs_read_unlock(dentry, AuLock_DW); +out_free: + kfree(a); +out: + return err; +} diff -Naur kernel.21.3.orig/fs/aufs/i_op.c kernel.21.3.new/fs/aufs/i_op.c --- kernel.21.3.orig/fs/aufs/i_op.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/i_op.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,1255 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * inode operations (except add/del/rename) + */ + +#include +#include +#include +#include +#include +#include "aufs.h" + +static int h_permission(struct inode *h_inode, int mask, + struct vfsmount *h_mnt, int brperm) +{ + int err; + const unsigned char write_mask = !!(mask & (MAY_WRITE | MAY_APPEND)); + + err = -EACCES; + if ((write_mask && IS_IMMUTABLE(h_inode)) + || ((mask & MAY_EXEC) + && S_ISREG(h_inode->i_mode) + && ((h_mnt->mnt_flags & MNT_NOEXEC) + || !(h_inode->i_mode & S_IXUGO)))) + goto out; + + /* + * - skip the lower fs test in the case of write to ro branch. + * - nfs dir permission write check is optimized, but a policy for + * link/rename requires a real check. + */ + if ((write_mask && !au_br_writable(brperm)) + || (au_test_nfs(h_inode->i_sb) && S_ISDIR(h_inode->i_mode) + && write_mask && !(mask & MAY_READ)) + || !h_inode->i_op->permission) { + /* AuLabel(generic_permission); */ + err = generic_permission(h_inode, mask); + } else { + /* AuLabel(h_inode->permission); */ + err = h_inode->i_op->permission(h_inode, mask); + AuTraceErr(err); + } + + if (!err) + err = devcgroup_inode_permission(h_inode, mask); + if (!err) + err = security_inode_permission(h_inode, mask); + +#if 0 + if (!err) { + /* todo: do we need to call ima_path_check()? */ + struct path h_path = { + .dentry = + .mnt = h_mnt + }; + err = ima_path_check(&h_path, + mask & (MAY_READ | MAY_WRITE | MAY_EXEC), + IMA_COUNT_LEAVE); + } +#endif + +out: + return err; +} + +static int aufs_permission(struct inode *inode, int mask) +{ + int err; + aufs_bindex_t bindex, bend; + const unsigned char isdir = !!S_ISDIR(inode->i_mode), + write_mask = !!(mask & (MAY_WRITE | MAY_APPEND)); + struct inode *h_inode; + struct super_block *sb; + struct au_branch *br; + + /* todo: support rcu-walk? */ + if (mask & MAY_NOT_BLOCK) + return -ECHILD; + + sb = inode->i_sb; + si_read_lock(sb, AuLock_FLUSH); + ii_read_lock_child(inode); +#if 0 + err = au_iigen_test(inode, au_sigen(sb)); + if (unlikely(err)) + goto out; +#endif + + if (!isdir + || write_mask + || au_opt_test(au_mntflags(sb), DIRPERM1)) { + err = au_busy_or_stale(); + h_inode = au_h_iptr(inode, au_ibstart(inode)); + if (unlikely(!h_inode + || (h_inode->i_mode & S_IFMT) + != (inode->i_mode & S_IFMT))) + goto out; + + err = 0; + bindex = au_ibstart(inode); + br = au_sbr(sb, bindex); + err = h_permission(h_inode, mask, au_br_mnt(br), br->br_perm); + if (write_mask + && !err + && !special_file(h_inode->i_mode)) { + /* test whether the upper writable branch exists */ + err = -EROFS; + for (; bindex >= 0; bindex--) + if (!au_br_rdonly(au_sbr(sb, bindex))) { + err = 0; + break; + } + } + goto out; + } + + /* non-write to dir */ + err = 0; + bend = au_ibend(inode); + for (bindex = au_ibstart(inode); !err && bindex <= bend; bindex++) { + h_inode = au_h_iptr(inode, bindex); + if (h_inode) { + err = au_busy_or_stale(); + if (unlikely(!S_ISDIR(h_inode->i_mode))) + break; + + br = au_sbr(sb, bindex); + err = h_permission(h_inode, mask, au_br_mnt(br), + br->br_perm); + } + } + +out: + ii_read_unlock(inode); + si_read_unlock(sb); + return err; +} + +#ifdef CONFIG_FS_POSIX_ACL +static struct posix_acl *aufs_get_acl(struct inode *inode, int mask) +{ + struct posix_acl *acl; + int err; + aufs_bindex_t bindex, bend; + const unsigned char isdir = !!S_ISDIR(inode->i_mode), + write_mask = !!(mask & (MAY_WRITE | MAY_APPEND)); + struct inode *h_inode; + struct super_block *sb; + + WARN_ON(mask & MAY_NOT_BLOCK); + + err = 0; + sb = inode->i_sb; + si_read_lock(sb, AuLock_FLUSH); + ii_read_lock_child(inode); + if (!(sb->s_flags & MS_POSIXACL)) + goto out_ii; + + err = au_busy_or_stale(); + bindex = au_ibstart(inode); + h_inode = au_h_iptr(inode, bindex); + if (unlikely(!h_inode + || (h_inode->i_mode & S_IFMT) != (inode->i_mode & S_IFMT))) + goto out_ii; + + /* cf: fs/namei.c:acl_permission_check() */ + err = -EAGAIN; + if (!IS_POSIXACL(h_inode)) + goto out_ii; + + if (!isdir + || write_mask + || au_opt_test(au_mntflags(sb), DIRPERM1)) { + err = check_acl(h_inode, mask); + if (unlikely(err && err != -EAGAIN)) + goto out_ii; + + if (write_mask + && !special_file(h_inode->i_mode)) { + /* test whether the upper writable branch exists */ + err = -EROFS; + for (; bindex >= 0; bindex--) + if (!au_br_rdonly(au_sbr(sb, bindex))) { + err = 0; + break; + } + } + goto out_ii; + } + + /* non-write to dir */ + err = 0; + bend = au_ibend(inode); + for (; (!err || err == -EAGAIN) && bindex <= bend; bindex++) { + h_inode = au_h_iptr(inode, bindex); + if (h_inode) { + err = au_busy_or_stale(); + if (unlikely(!S_ISDIR(h_inode->i_mode))) + break; + + err = check_acl(h_inode, mask); + } + } + +out_ii: + ii_read_unlock(inode); + si_read_unlock(sb); + acl = ERR_PTR(err); + + return acl; +} +#endif + +/* ---------------------------------------------------------------------- */ + +static struct dentry *aufs_lookup(struct inode *dir, struct dentry *dentry, + unsigned int flags) +{ + struct dentry *ret, *parent; + struct inode *inode; + struct super_block *sb; + int err, npositive; + + IMustLock(dir); + + /* todo: support rcu-walk? */ + ret = ERR_PTR(-ECHILD); + if (flags & LOOKUP_RCU) + goto out; + + ret = ERR_PTR(-ENAMETOOLONG); + if (unlikely(dentry->d_name.len > AUFS_MAX_NAMELEN)) + goto out; + + sb = dir->i_sb; + err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM); + ret = ERR_PTR(err); + if (unlikely(err)) + goto out; + + err = au_di_init(dentry); + ret = ERR_PTR(err); + if (unlikely(err)) + goto out_si; + + inode = NULL; + npositive = 0; /* suppress a warning */ + parent = dentry->d_parent; /* dir inode is locked */ + di_read_lock_parent(parent, AuLock_IR); + err = au_alive_dir(parent); + if (!err) + err = au_digen_test(parent, au_sigen(sb)); + if (!err) { + npositive = au_lkup_dentry(dentry, au_dbstart(parent), + /*type*/0); + err = npositive; + } + di_read_unlock(parent, AuLock_IR); + ret = ERR_PTR(err); + if (unlikely(err < 0)) + goto out_unlock; + + if (npositive) { + inode = au_new_inode(dentry, /*must_new*/0); + if (IS_ERR(inode)) { + ret = (void *)inode; + inode = NULL; + goto out_unlock; + } + } + + ret = d_splice_alias(inode, dentry); +#if 0 + if (unlikely(d_need_lookup(dentry))) { + spin_lock(&dentry->d_lock); + dentry->d_flags &= ~DCACHE_NEED_LOOKUP; + spin_unlock(&dentry->d_lock); + } else +#endif + if (unlikely(IS_ERR(ret) && inode)) { + ii_write_unlock(inode); + iput(inode); + inode = NULL; + } + +out_unlock: + di_write_unlock(dentry); + if (inode) { + /* verbose coding for lock class name */ + if (unlikely(S_ISLNK(inode->i_mode))) + au_rw_class(&au_di(dentry)->di_rwsem, + au_lc_key + AuLcSymlink_DIINFO); + else if (unlikely(S_ISDIR(inode->i_mode))) + au_rw_class(&au_di(dentry)->di_rwsem, + au_lc_key + AuLcDir_DIINFO); + else /* likely */ + au_rw_class(&au_di(dentry)->di_rwsem, + au_lc_key + AuLcNonDir_DIINFO); + } +out_si: + si_read_unlock(sb); +out: + return ret; +} + +/* ---------------------------------------------------------------------- */ + +static int au_wr_dir_cpup(struct dentry *dentry, struct dentry *parent, + const unsigned char add_entry, aufs_bindex_t bcpup, + aufs_bindex_t bstart) +{ + int err; + struct dentry *h_parent; + struct inode *h_dir; + + if (add_entry) + IMustLock(parent->d_inode); + else + di_write_lock_parent(parent); + + err = 0; + if (!au_h_dptr(parent, bcpup)) { + if (bstart > bcpup) + err = au_cpup_dirs(dentry, bcpup); + else if (bstart < bcpup) + err = au_cpdown_dirs(dentry, bcpup); + else + BUG(); + } + if (!err && add_entry) { + h_parent = au_h_dptr(parent, bcpup); + h_dir = h_parent->d_inode; + mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_PARENT); + err = au_lkup_neg(dentry, bcpup, + au_ftest_wrdir(add_entry, TMP_WHENTRY)); + /* todo: no unlock here */ + mutex_unlock(&h_dir->i_mutex); + + AuDbg("bcpup %d\n", bcpup); + if (!err) { + if (!dentry->d_inode) + au_set_h_dptr(dentry, bstart, NULL); + au_update_dbrange(dentry, /*do_put_zero*/0); + } + } + + if (!add_entry) + di_write_unlock(parent); + if (!err) + err = bcpup; /* success */ + + AuTraceErr(err); + return err; +} + +/* + * decide the branch and the parent dir where we will create a new entry. + * returns new bindex or an error. + * copyup the parent dir if needed. + */ +int au_wr_dir(struct dentry *dentry, struct dentry *src_dentry, + struct au_wr_dir_args *args) +{ + int err; + unsigned int flags; + aufs_bindex_t bcpup, bstart, src_bstart; + const unsigned char add_entry + = au_ftest_wrdir(args->flags, ADD_ENTRY) + | au_ftest_wrdir(args->flags, TMP_WHENTRY); + struct super_block *sb; + struct dentry *parent; + struct au_sbinfo *sbinfo; + + sb = dentry->d_sb; + sbinfo = au_sbi(sb); + parent = dget_parent(dentry); + bstart = au_dbstart(dentry); + bcpup = bstart; + if (args->force_btgt < 0) { + if (src_dentry) { + src_bstart = au_dbstart(src_dentry); + if (src_bstart < bstart) + bcpup = src_bstart; + } else if (add_entry) { + flags = 0; + if (au_ftest_wrdir(args->flags, ISDIR)) + au_fset_wbr(flags, DIR); + err = AuWbrCreate(sbinfo, dentry, flags); + bcpup = err; + } + + if (bcpup < 0 || au_test_ro(sb, bcpup, dentry->d_inode)) { + if (add_entry) + err = AuWbrCopyup(sbinfo, dentry); + else { + if (!IS_ROOT(dentry)) { + di_read_lock_parent(parent, !AuLock_IR); + err = AuWbrCopyup(sbinfo, dentry); + di_read_unlock(parent, !AuLock_IR); + } else + err = AuWbrCopyup(sbinfo, dentry); + } + bcpup = err; + if (unlikely(err < 0)) + goto out; + } + } else { + bcpup = args->force_btgt; + AuDebugOn(au_test_ro(sb, bcpup, dentry->d_inode)); + } + + AuDbg("bstart %d, bcpup %d\n", bstart, bcpup); + err = bcpup; + if (bcpup == bstart) + goto out; /* success */ + + /* copyup the new parent into the branch we process */ + err = au_wr_dir_cpup(dentry, parent, add_entry, bcpup, bstart); + if (err >= 0) { + if (!dentry->d_inode) { + au_set_h_dptr(dentry, bstart, NULL); + au_set_dbstart(dentry, bcpup); + au_set_dbend(dentry, bcpup); + } + AuDebugOn(add_entry && !au_h_dptr(dentry, bcpup)); + } + +out: + dput(parent); + return err; +} + +/* ---------------------------------------------------------------------- */ + +void au_pin_hdir_unlock(struct au_pin *p) +{ + if (p->hdir) + au_hn_imtx_unlock(p->hdir); +} + +int au_pin_hdir_lock(struct au_pin *p) +{ + int err; + + err = 0; + if (!p->hdir) + goto out; + + /* even if an error happens later, keep this lock */ + au_hn_imtx_lock_nested(p->hdir, p->lsc_hi); + + err = -EBUSY; + if (unlikely(p->hdir->hi_inode != p->h_parent->d_inode)) + goto out; + + err = 0; + if (p->h_dentry) + err = au_h_verify(p->h_dentry, p->udba, p->hdir->hi_inode, + p->h_parent, p->br); + +out: + return err; +} + +int au_pin_hdir_relock(struct au_pin *p) +{ + int err, i; + struct inode *h_i; + struct dentry *h_d[] = { + p->h_dentry, + p->h_parent + }; + + err = au_pin_hdir_lock(p); + if (unlikely(err)) + goto out; + + for (i = 0; !err && i < sizeof(h_d)/sizeof(*h_d); i++) { + if (!h_d[i]) + continue; + h_i = h_d[i]->d_inode; + if (h_i) + err = !h_i->i_nlink; + } + +out: + return err; +} + +void au_pin_hdir_set_owner(struct au_pin *p, struct task_struct *task) +{ +#if defined(CONFIG_DEBUG_MUTEXES) || defined(CONFIG_SMP) + p->hdir->hi_inode->i_mutex.owner = task; +#endif +} + +void au_pin_hdir_acquire_nest(struct au_pin *p) +{ + if (p->hdir) { + mutex_acquire_nest(&p->hdir->hi_inode->i_mutex.dep_map, + p->lsc_hi, 0, NULL, _RET_IP_); + au_pin_hdir_set_owner(p, current); + } +} + +void au_pin_hdir_release(struct au_pin *p) +{ + if (p->hdir) { + au_pin_hdir_set_owner(p, p->task); + mutex_release(&p->hdir->hi_inode->i_mutex.dep_map, 1, _RET_IP_); + } +} + +struct dentry *au_pinned_h_parent(struct au_pin *pin) +{ + if (pin && pin->parent) + return au_h_dptr(pin->parent, pin->bindex); + return NULL; +} + +void au_unpin(struct au_pin *p) +{ + if (p->hdir) + au_pin_hdir_unlock(p); + if (p->h_mnt && au_ftest_pin(p->flags, MNT_WRITE)) + vfsub_mnt_drop_write(p->h_mnt); + if (!p->hdir) + return; + + if (!au_ftest_pin(p->flags, DI_LOCKED)) + di_read_unlock(p->parent, AuLock_IR); + iput(p->hdir->hi_inode); + dput(p->parent); + p->parent = NULL; + p->hdir = NULL; + p->h_mnt = NULL; + /* do not clear p->task */ +} + +int au_do_pin(struct au_pin *p) +{ + int err; + struct super_block *sb; + struct inode *h_dir; + + err = 0; + sb = p->dentry->d_sb; + p->br = au_sbr(sb, p->bindex); + if (IS_ROOT(p->dentry)) { + if (au_ftest_pin(p->flags, MNT_WRITE)) { + p->h_mnt = au_br_mnt(p->br); + err = vfsub_mnt_want_write(p->h_mnt); + if (unlikely(err)) { + au_fclr_pin(p->flags, MNT_WRITE); + goto out_err; + } + } + goto out; + } + + p->h_dentry = NULL; + if (p->bindex <= au_dbend(p->dentry)) + p->h_dentry = au_h_dptr(p->dentry, p->bindex); + + p->parent = dget_parent(p->dentry); + if (!au_ftest_pin(p->flags, DI_LOCKED)) + di_read_lock(p->parent, AuLock_IR, p->lsc_di); + + h_dir = NULL; + p->h_parent = au_h_dptr(p->parent, p->bindex); + p->hdir = au_hi(p->parent->d_inode, p->bindex); + if (p->hdir) + h_dir = p->hdir->hi_inode; + + /* + * udba case, or + * if DI_LOCKED is not set, then p->parent may be different + * and h_parent can be NULL. + */ + if (unlikely(!p->hdir || !h_dir || !p->h_parent)) { + err = -EBUSY; + if (!au_ftest_pin(p->flags, DI_LOCKED)) + di_read_unlock(p->parent, AuLock_IR); + dput(p->parent); + p->parent = NULL; + goto out_err; + } + + if (au_ftest_pin(p->flags, MNT_WRITE)) { + p->h_mnt = au_br_mnt(p->br); + err = vfsub_mnt_want_write(p->h_mnt); + if (unlikely(err)) { + au_fclr_pin(p->flags, MNT_WRITE); + if (!au_ftest_pin(p->flags, DI_LOCKED)) + di_read_unlock(p->parent, AuLock_IR); + dput(p->parent); + p->parent = NULL; + goto out_err; + } + } + + au_igrab(h_dir); + err = au_pin_hdir_lock(p); + if (!err) + goto out; /* success */ + + au_unpin(p); + +out_err: + pr_err("err %d\n", err); + err = au_busy_or_stale(); +out: + return err; +} + +void au_pin_init(struct au_pin *p, struct dentry *dentry, + aufs_bindex_t bindex, int lsc_di, int lsc_hi, + unsigned int udba, unsigned char flags) +{ + p->dentry = dentry; + p->udba = udba; + p->lsc_di = lsc_di; + p->lsc_hi = lsc_hi; + p->flags = flags; + p->bindex = bindex; + + p->parent = NULL; + p->hdir = NULL; + p->h_mnt = NULL; + + p->h_dentry = NULL; + p->h_parent = NULL; + p->br = NULL; + p->task = current; +} + +int au_pin(struct au_pin *pin, struct dentry *dentry, aufs_bindex_t bindex, + unsigned int udba, unsigned char flags) +{ + au_pin_init(pin, dentry, bindex, AuLsc_DI_PARENT, AuLsc_I_PARENT2, + udba, flags); + return au_do_pin(pin); +} + +/* ---------------------------------------------------------------------- */ + +/* + * ->setattr() and ->getattr() are called in various cases. + * chmod, stat: dentry is revalidated. + * fchmod, fstat: file and dentry are not revalidated, additionally they may be + * unhashed. + * for ->setattr(), ia->ia_file is passed from ftruncate only. + */ +/* todo: consolidate with do_refresh() and simple_reval_dpath() */ +int au_reval_for_attr(struct dentry *dentry, unsigned int sigen) +{ + int err; + struct inode *inode; + struct dentry *parent; + + err = 0; + inode = dentry->d_inode; + if (au_digen_test(dentry, sigen)) { + parent = dget_parent(dentry); + di_read_lock_parent(parent, AuLock_IR); + err = au_refresh_dentry(dentry, parent); + di_read_unlock(parent, AuLock_IR); + dput(parent); + } + + AuTraceErr(err); + return err; +} + +int au_pin_and_icpup(struct dentry *dentry, struct iattr *ia, + struct au_icpup_args *a) +{ + int err; + loff_t sz; + aufs_bindex_t bstart, ibstart; + struct dentry *hi_wh, *parent; + struct inode *inode; + struct au_wr_dir_args wr_dir_args = { + .force_btgt = -1, + .flags = 0 + }; + + bstart = au_dbstart(dentry); + inode = dentry->d_inode; + if (S_ISDIR(inode->i_mode)) + au_fset_wrdir(wr_dir_args.flags, ISDIR); + /* plink or hi_wh() case */ + ibstart = au_ibstart(inode); + if (bstart != ibstart && !au_test_ro(inode->i_sb, ibstart, inode)) + wr_dir_args.force_btgt = ibstart; + err = au_wr_dir(dentry, /*src_dentry*/NULL, &wr_dir_args); + if (unlikely(err < 0)) + goto out; + a->btgt = err; + if (err != bstart) + au_fset_icpup(a->flags, DID_CPUP); + + err = 0; + a->pin_flags = AuPin_MNT_WRITE; + parent = NULL; + if (!IS_ROOT(dentry)) { + au_fset_pin(a->pin_flags, DI_LOCKED); + parent = dget_parent(dentry); + di_write_lock_parent(parent); + } + + err = au_pin(&a->pin, dentry, a->btgt, a->udba, a->pin_flags); + if (unlikely(err)) + goto out_parent; + + a->h_path.dentry = au_h_dptr(dentry, bstart); + a->h_inode = a->h_path.dentry->d_inode; + sz = -1; + if (ia && (ia->ia_valid & ATTR_SIZE)) { + mutex_lock_nested(&a->h_inode->i_mutex, AuLsc_I_CHILD); + if (ia->ia_size < i_size_read(a->h_inode)) + sz = ia->ia_size; + mutex_unlock(&a->h_inode->i_mutex); + } + + hi_wh = NULL; + if (au_ftest_icpup(a->flags, DID_CPUP) && d_unlinked(dentry)) { + hi_wh = au_hi_wh(inode, a->btgt); + if (!hi_wh) { + struct au_cp_generic cpg = { + .dentry = dentry, + .bdst = a->btgt, + .bsrc = -1, + .len = sz, + .pin = &a->pin + }; + err = au_sio_cpup_wh(&cpg, /*file*/NULL); + if (unlikely(err)) + goto out_unlock; + hi_wh = au_hi_wh(inode, a->btgt); + /* todo: revalidate hi_wh? */ + } + } + + if (parent) { + au_pin_set_parent_lflag(&a->pin, /*lflag*/0); + di_downgrade_lock(parent, AuLock_IR); + dput(parent); + parent = NULL; + } + if (!au_ftest_icpup(a->flags, DID_CPUP)) + goto out; /* success */ + + if (!d_unhashed(dentry)) { + struct au_cp_generic cpg = { + .dentry = dentry, + .bdst = a->btgt, + .bsrc = bstart, + .len = sz, + .pin = &a->pin, + .flags = AuCpup_DTIME | AuCpup_HOPEN + }; + err = au_sio_cpup_simple(&cpg); + if (!err) + a->h_path.dentry = au_h_dptr(dentry, a->btgt); + } else if (!hi_wh) + a->h_path.dentry = au_h_dptr(dentry, a->btgt); + else + a->h_path.dentry = hi_wh; /* do not dget here */ + +out_unlock: + a->h_inode = a->h_path.dentry->d_inode; + if (!err) + goto out; /* success */ + au_unpin(&a->pin); +out_parent: + if (parent) { + di_write_unlock(parent); + dput(parent); + } +out: + if (!err) + mutex_lock_nested(&a->h_inode->i_mutex, AuLsc_I_CHILD); + return err; +} + +static int aufs_setattr(struct dentry *dentry, struct iattr *ia) +{ + int err; + struct inode *inode; + struct super_block *sb; + struct file *file; + struct au_icpup_args *a; + + inode = dentry->d_inode; + IMustLock(inode); + + err = -ENOMEM; + a = kzalloc(sizeof(*a), GFP_NOFS); + if (unlikely(!a)) + goto out; + + if (ia->ia_valid & (ATTR_KILL_SUID | ATTR_KILL_SGID)) + ia->ia_valid &= ~ATTR_MODE; + + file = NULL; + sb = dentry->d_sb; + err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM); + if (unlikely(err)) + goto out_kfree; + + if (ia->ia_valid & ATTR_FILE) { + /* currently ftruncate(2) only */ + AuDebugOn(!S_ISREG(inode->i_mode)); + file = ia->ia_file; + err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1); + if (unlikely(err)) + goto out_si; + ia->ia_file = au_hf_top(file); + a->udba = AuOpt_UDBA_NONE; + } else { + /* fchmod() doesn't pass ia_file */ + a->udba = au_opt_udba(sb); + di_write_lock_child(dentry); + /* no d_unlinked(), to set UDBA_NONE for root */ + if (d_unhashed(dentry)) + a->udba = AuOpt_UDBA_NONE; + if (a->udba != AuOpt_UDBA_NONE) { + AuDebugOn(IS_ROOT(dentry)); + err = au_reval_for_attr(dentry, au_sigen(sb)); + if (unlikely(err)) + goto out_dentry; + } + } + + err = au_pin_and_icpup(dentry, ia, a); + if (unlikely(err < 0)) + goto out_dentry; + if (au_ftest_icpup(a->flags, DID_CPUP)) { + ia->ia_file = NULL; + ia->ia_valid &= ~ATTR_FILE; + } + + a->h_path.mnt = au_sbr_mnt(sb, a->btgt); + if ((ia->ia_valid & (ATTR_MODE | ATTR_CTIME)) + == (ATTR_MODE | ATTR_CTIME)) { + err = security_path_chmod(&a->h_path, ia->ia_mode); + if (unlikely(err)) + goto out_unlock; + } else if ((ia->ia_valid & (ATTR_UID | ATTR_GID)) + && (ia->ia_valid & ATTR_CTIME)) { + err = security_path_chown(&a->h_path, ia->ia_uid, ia->ia_gid); + if (unlikely(err)) + goto out_unlock; + } + + if (ia->ia_valid & ATTR_SIZE) { + struct file *f; + + if (ia->ia_size < i_size_read(inode)) + /* unmap only */ + truncate_setsize(inode, ia->ia_size); + + f = NULL; + if (ia->ia_valid & ATTR_FILE) + f = ia->ia_file; + mutex_unlock(&a->h_inode->i_mutex); + err = vfsub_trunc(&a->h_path, ia->ia_size, ia->ia_valid, f); + mutex_lock_nested(&a->h_inode->i_mutex, AuLsc_I_CHILD); + } else + err = vfsub_notify_change(&a->h_path, ia); + if (!err) + au_cpup_attr_changeable(inode); + +out_unlock: + mutex_unlock(&a->h_inode->i_mutex); + au_unpin(&a->pin); + if (unlikely(err)) + au_update_dbstart(dentry); +out_dentry: + di_write_unlock(dentry); + if (file) { + fi_write_unlock(file); + ia->ia_file = file; + ia->ia_valid |= ATTR_FILE; + } +out_si: + si_read_unlock(sb); +out_kfree: + kfree(a); +out: + AuTraceErr(err); + return err; +} + +static void au_refresh_iattr(struct inode *inode, struct kstat *st, + unsigned int nlink) +{ + unsigned int n; + + inode->i_mode = st->mode; + /* don't i_[ug]id_write() here */ + inode->i_uid = st->uid; + inode->i_gid = st->gid; + inode->i_atime = st->atime; + inode->i_mtime = st->mtime; + inode->i_ctime = st->ctime; + + au_cpup_attr_nlink(inode, /*force*/0); + if (S_ISDIR(inode->i_mode)) { + n = inode->i_nlink; + n -= nlink; + n += st->nlink; + smp_mb(); /* for i_nlink */ + /* 0 can happen */ + set_nlink(inode, n); + } + + spin_lock(&inode->i_lock); + inode->i_blocks = st->blocks; + i_size_write(inode, st->size); + spin_unlock(&inode->i_lock); +} + +/* + * common routine for aufs_getattr() and aufs_getxattr(). + * returns zero or negative (an error). + * @dentry will be read-locked in success. + */ +int au_h_path_getattr(struct dentry *dentry, int force, struct path *h_path) +{ + int err; + unsigned int mnt_flags, sigen; + unsigned char udba_none; + aufs_bindex_t bindex; + struct super_block *sb, *h_sb; + struct inode *inode; + + h_path->mnt = NULL; + h_path->dentry = NULL; + + err = 0; + sb = dentry->d_sb; + mnt_flags = au_mntflags(sb); + udba_none = !!au_opt_test(mnt_flags, UDBA_NONE); + + /* support fstat(2) */ + if (!d_unlinked(dentry) && !udba_none) { + sigen = au_sigen(sb); + err = au_digen_test(dentry, sigen); + if (!err) { + di_read_lock_child(dentry, AuLock_IR); + err = au_dbrange_test(dentry); + if (unlikely(err)) { + di_read_unlock(dentry, AuLock_IR); + goto out; + } + } else { + AuDebugOn(IS_ROOT(dentry)); + di_write_lock_child(dentry); + err = au_dbrange_test(dentry); + if (!err) + err = au_reval_for_attr(dentry, sigen); + if (!err) + di_downgrade_lock(dentry, AuLock_IR); + else { + di_write_unlock(dentry); + goto out; + } + } + } else + di_read_lock_child(dentry, AuLock_IR); + + inode = dentry->d_inode; + bindex = au_ibstart(inode); + h_path->mnt = au_sbr_mnt(sb, bindex); + h_sb = h_path->mnt->mnt_sb; + if (!force + && !au_test_fs_bad_iattr(h_sb) + && udba_none) + goto out; /* success */ + + if (au_dbstart(dentry) == bindex) + h_path->dentry = au_h_dptr(dentry, bindex); + else if (au_opt_test(mnt_flags, PLINK) && au_plink_test(inode)) { + h_path->dentry = au_plink_lkup(inode, bindex); + if (IS_ERR(h_path->dentry)) + /* pretending success */ + h_path->dentry = NULL; + else + dput(h_path->dentry); + } + +out: + return err; +} + +static int aufs_getattr(struct vfsmount *mnt __maybe_unused, + struct dentry *dentry, struct kstat *st) +{ + int err; + unsigned char positive; + struct path h_path; + struct inode *inode; + struct super_block *sb; + + inode = dentry->d_inode; + sb = dentry->d_sb; + err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM); + if (unlikely(err)) + goto out; + err = au_h_path_getattr(dentry, /*force*/0, &h_path); + if (unlikely(err)) + goto out_si; + if (unlikely(!h_path.dentry)) + /* illegally overlapped or something */ + goto out_fill; /* pretending success */ + + positive = !!h_path.dentry->d_inode; + if (positive) + err = vfs_getattr(&h_path, st); + if (!err) { + if (positive) + au_refresh_iattr(inode, st, + h_path.dentry->d_inode->i_nlink); + goto out_fill; /* success */ + } + AuTraceErr(err); + goto out_di; + +out_fill: + generic_fillattr(inode, st); +out_di: + di_read_unlock(dentry, AuLock_IR); +out_si: + si_read_unlock(sb); +out: + AuTraceErr(err); + return err; +} + +/* ---------------------------------------------------------------------- */ + +static int h_readlink(struct dentry *dentry, int bindex, char __user *buf, + int bufsiz) +{ + int err; + struct super_block *sb; + struct dentry *h_dentry; + + err = -EINVAL; + h_dentry = au_h_dptr(dentry, bindex); + if (unlikely(!h_dentry->d_inode->i_op->readlink)) + goto out; + + err = security_inode_readlink(h_dentry); + if (unlikely(err)) + goto out; + + sb = dentry->d_sb; + if (!au_test_ro(sb, bindex, dentry->d_inode)) { + vfsub_touch_atime(au_sbr_mnt(sb, bindex), h_dentry); + fsstack_copy_attr_atime(dentry->d_inode, h_dentry->d_inode); + } + err = h_dentry->d_inode->i_op->readlink(h_dentry, buf, bufsiz); + +out: + return err; +} + +static int aufs_readlink(struct dentry *dentry, char __user *buf, int bufsiz) +{ + int err; + + err = aufs_read_lock(dentry, AuLock_IR | AuLock_GEN); + if (unlikely(err)) + goto out; + err = au_d_hashed_positive(dentry); + if (!err) + err = h_readlink(dentry, au_dbstart(dentry), buf, bufsiz); + aufs_read_unlock(dentry, AuLock_IR); + +out: + return err; +} + +static void *aufs_follow_link(struct dentry *dentry, struct nameidata *nd) +{ + int err; + mm_segment_t old_fs; + union { + char *k; + char __user *u; + } buf; + + err = -ENOMEM; + buf.k = (void *)__get_free_page(GFP_NOFS); + if (unlikely(!buf.k)) + goto out; + + err = aufs_read_lock(dentry, AuLock_IR | AuLock_GEN); + if (unlikely(err)) + goto out_name; + + err = au_d_hashed_positive(dentry); + if (!err) { + old_fs = get_fs(); + set_fs(KERNEL_DS); + err = h_readlink(dentry, au_dbstart(dentry), buf.u, PATH_MAX); + set_fs(old_fs); + } + aufs_read_unlock(dentry, AuLock_IR); + + if (err >= 0) { + buf.k[err] = 0; + /* will be freed by put_link */ + nd_set_link(nd, buf.k); + return NULL; /* success */ + } + +out_name: + free_page((unsigned long)buf.k); +out: + AuTraceErr(err); + return ERR_PTR(err); +} + +static void aufs_put_link(struct dentry *dentry __maybe_unused, + struct nameidata *nd, void *cookie __maybe_unused) +{ + char *p; + + p = nd_get_link(nd); + if (!IS_ERR_OR_NULL(p)) + free_page((unsigned long)p); +} + +/* ---------------------------------------------------------------------- */ + +static int aufs_update_time(struct inode *inode, struct timespec *ts, int flags) +{ + int err; + struct super_block *sb; + struct inode *h_inode; + + sb = inode->i_sb; + /* mmap_sem might be acquired already, cf. aufs_mmap() */ + lockdep_off(); + si_read_lock(sb, AuLock_FLUSH); + ii_write_lock_child(inode); + lockdep_on(); + h_inode = au_h_iptr(inode, au_ibstart(inode)); + err = vfsub_update_time(h_inode, ts, flags); + lockdep_off(); + if (!err) + au_cpup_attr_timesizes(inode); + ii_write_unlock(inode); + si_read_unlock(sb); + lockdep_on(); + + if (!err && (flags & S_VERSION)) + inode_inc_iversion(inode); + + return err; +} + +/* ---------------------------------------------------------------------- */ + +struct inode_operations aufs_symlink_iop = { + .permission = aufs_permission, +#ifdef CONFIG_FS_POSIX_ACL + .get_acl = aufs_get_acl, +#endif + + .setattr = aufs_setattr, + .getattr = aufs_getattr, + +#ifdef CONFIG_AUFS_XATTR + .setxattr = aufs_setxattr, + .getxattr = aufs_getxattr, + .listxattr = aufs_listxattr, + .removexattr = aufs_removexattr, +#endif + + .readlink = aufs_readlink, + .follow_link = aufs_follow_link, + .put_link = aufs_put_link, + + /* .update_time = aufs_update_time */ +}; + +struct inode_operations aufs_dir_iop = { + .create = aufs_create, + .lookup = aufs_lookup, + .link = aufs_link, + .unlink = aufs_unlink, + .symlink = aufs_symlink, + .mkdir = aufs_mkdir, + .rmdir = aufs_rmdir, + .mknod = aufs_mknod, + .rename = aufs_rename, + + .permission = aufs_permission, +#ifdef CONFIG_FS_POSIX_ACL + .get_acl = aufs_get_acl, +#endif + + .setattr = aufs_setattr, + .getattr = aufs_getattr, + +#ifdef CONFIG_AUFS_XATTR + .setxattr = aufs_setxattr, + .getxattr = aufs_getxattr, + .listxattr = aufs_listxattr, + .removexattr = aufs_removexattr, +#endif + + .update_time = aufs_update_time + /* no support for atomic_open() */ +}; + +struct inode_operations aufs_iop = { + .permission = aufs_permission, +#ifdef CONFIG_FS_POSIX_ACL + .get_acl = aufs_get_acl, +#endif + + .setattr = aufs_setattr, + .getattr = aufs_getattr, + +#ifdef CONFIG_AUFS_XATTR + .setxattr = aufs_setxattr, + .getxattr = aufs_getxattr, + .listxattr = aufs_listxattr, + .removexattr = aufs_removexattr, +#endif + + .update_time = aufs_update_time +}; diff -Naur kernel.21.3.orig/fs/aufs/i_op_del.c kernel.21.3.new/fs/aufs/i_op_del.c --- kernel.21.3.orig/fs/aufs/i_op_del.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/i_op_del.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,502 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * inode operations (del entry) + */ + +#include "aufs.h" + +/* + * decide if a new whiteout for @dentry is necessary or not. + * when it is necessary, prepare the parent dir for the upper branch whose + * branch index is @bcpup for creation. the actual creation of the whiteout will + * be done by caller. + * return value: + * 0: wh is unnecessary + * plus: wh is necessary + * minus: error + */ +int au_wr_dir_need_wh(struct dentry *dentry, int isdir, aufs_bindex_t *bcpup) +{ + int need_wh, err; + aufs_bindex_t bstart; + struct super_block *sb; + + sb = dentry->d_sb; + bstart = au_dbstart(dentry); + if (*bcpup < 0) { + *bcpup = bstart; + if (au_test_ro(sb, bstart, dentry->d_inode)) { + err = AuWbrCopyup(au_sbi(sb), dentry); + *bcpup = err; + if (unlikely(err < 0)) + goto out; + } + } else + AuDebugOn(bstart < *bcpup + || au_test_ro(sb, *bcpup, dentry->d_inode)); + AuDbg("bcpup %d, bstart %d\n", *bcpup, bstart); + + if (*bcpup != bstart) { + err = au_cpup_dirs(dentry, *bcpup); + if (unlikely(err)) + goto out; + need_wh = 1; + } else { + struct au_dinfo *dinfo, *tmp; + + need_wh = -ENOMEM; + dinfo = au_di(dentry); + tmp = au_di_alloc(sb, AuLsc_DI_TMP); + if (tmp) { + au_di_cp(tmp, dinfo); + au_di_swap(tmp, dinfo); + /* returns the number of positive dentries */ + need_wh = au_lkup_dentry(dentry, bstart + 1, /*type*/0); + au_di_swap(tmp, dinfo); + au_rw_write_unlock(&tmp->di_rwsem); + au_di_free(tmp); + } + } + AuDbg("need_wh %d\n", need_wh); + err = need_wh; + +out: + return err; +} + +/* + * simple tests for the del-entry operations. + * following the checks in vfs, plus the parent-child relationship. + */ +int au_may_del(struct dentry *dentry, aufs_bindex_t bindex, + struct dentry *h_parent, int isdir) +{ + int err; + umode_t h_mode; + struct dentry *h_dentry, *h_latest; + struct inode *h_inode; + + h_dentry = au_h_dptr(dentry, bindex); + h_inode = h_dentry->d_inode; + if (dentry->d_inode) { + err = -ENOENT; + if (unlikely(!h_inode || !h_inode->i_nlink)) + goto out; + + h_mode = h_inode->i_mode; + if (!isdir) { + err = -EISDIR; + if (unlikely(S_ISDIR(h_mode))) + goto out; + } else if (unlikely(!S_ISDIR(h_mode))) { + err = -ENOTDIR; + goto out; + } + } else { + /* rename(2) case */ + err = -EIO; + if (unlikely(h_inode)) + goto out; + } + + err = -ENOENT; + /* expected parent dir is locked */ + if (unlikely(h_parent != h_dentry->d_parent)) + goto out; + err = 0; + + /* + * rmdir a dir may break the consistency on some filesystem. + * let's try heavy test. + */ + err = -EACCES; + if (unlikely(!au_opt_test(au_mntflags(dentry->d_sb), DIRPERM1) + && au_test_h_perm(h_parent->d_inode, + MAY_EXEC | MAY_WRITE))) + goto out; + + h_latest = au_sio_lkup_one(&dentry->d_name, h_parent); + err = -EIO; + if (IS_ERR(h_latest)) + goto out; + if (h_latest == h_dentry) + err = 0; + dput(h_latest); + +out: + return err; +} + +/* + * decide the branch where we operate for @dentry. the branch index will be set + * @rbcpup. after diciding it, 'pin' it and store the timestamps of the parent + * dir for reverting. + * when a new whiteout is necessary, create it. + */ +static struct dentry* +lock_hdir_create_wh(struct dentry *dentry, int isdir, aufs_bindex_t *rbcpup, + struct au_dtime *dt, struct au_pin *pin) +{ + struct dentry *wh_dentry; + struct super_block *sb; + struct path h_path; + int err, need_wh; + unsigned int udba; + aufs_bindex_t bcpup; + + need_wh = au_wr_dir_need_wh(dentry, isdir, rbcpup); + wh_dentry = ERR_PTR(need_wh); + if (unlikely(need_wh < 0)) + goto out; + + sb = dentry->d_sb; + udba = au_opt_udba(sb); + bcpup = *rbcpup; + err = au_pin(pin, dentry, bcpup, udba, + AuPin_DI_LOCKED | AuPin_MNT_WRITE); + wh_dentry = ERR_PTR(err); + if (unlikely(err)) + goto out; + + h_path.dentry = au_pinned_h_parent(pin); + if (udba != AuOpt_UDBA_NONE + && au_dbstart(dentry) == bcpup) { + err = au_may_del(dentry, bcpup, h_path.dentry, isdir); + wh_dentry = ERR_PTR(err); + if (unlikely(err)) + goto out_unpin; + } + + h_path.mnt = au_sbr_mnt(sb, bcpup); + au_dtime_store(dt, au_pinned_parent(pin), &h_path); + wh_dentry = NULL; + if (!need_wh) + goto out; /* success, no need to create whiteout */ + + wh_dentry = au_wh_create(dentry, bcpup, h_path.dentry); + if (IS_ERR(wh_dentry)) + goto out_unpin; + + /* returns with the parent is locked and wh_dentry is dget-ed */ + goto out; /* success */ + +out_unpin: + au_unpin(pin); +out: + return wh_dentry; +} + +/* + * when removing a dir, rename it to a unique temporary whiteout-ed name first + * in order to be revertible and save time for removing many child whiteouts + * under the dir. + * returns 1 when there are too many child whiteout and caller should remove + * them asynchronously. returns 0 when the number of children is enough small to + * remove now or the branch fs is a remote fs. + * otherwise return an error. + */ +static int renwh_and_rmdir(struct dentry *dentry, aufs_bindex_t bindex, + struct au_nhash *whlist, struct inode *dir) +{ + int rmdir_later, err, dirwh; + struct dentry *h_dentry; + struct super_block *sb; + + sb = dentry->d_sb; + SiMustAnyLock(sb); + h_dentry = au_h_dptr(dentry, bindex); + err = au_whtmp_ren(h_dentry, au_sbr(sb, bindex)); + if (unlikely(err)) + goto out; + + /* stop monitoring */ + au_hn_free(au_hi(dentry->d_inode, bindex)); + + if (!au_test_fs_remote(h_dentry->d_sb)) { + dirwh = au_sbi(sb)->si_dirwh; + rmdir_later = (dirwh <= 1); + if (!rmdir_later) + rmdir_later = au_nhash_test_longer_wh(whlist, bindex, + dirwh); + if (rmdir_later) + return rmdir_later; + } + + err = au_whtmp_rmdir(dir, bindex, h_dentry, whlist); + if (unlikely(err)) { + AuIOErr("rmdir %.*s, b%d failed, %d. ignored\n", + AuDLNPair(h_dentry), bindex, err); + err = 0; + } + +out: + AuTraceErr(err); + return err; +} + +/* + * final procedure for deleting a entry. + * maintain dentry and iattr. + */ +static void epilog(struct inode *dir, struct dentry *dentry, + aufs_bindex_t bindex) +{ + struct inode *inode; + + inode = dentry->d_inode; + d_drop(dentry); + inode->i_ctime = dir->i_ctime; + + if (au_ibstart(dir) == bindex) + au_cpup_attr_timesizes(dir); + dir->i_version++; +} + +/* + * when an error happened, remove the created whiteout and revert everything. + */ +static int do_revert(int err, struct inode *dir, aufs_bindex_t bindex, + aufs_bindex_t bwh, struct dentry *wh_dentry, + struct dentry *dentry, struct au_dtime *dt) +{ + int rerr; + struct path h_path = { + .dentry = wh_dentry, + .mnt = au_sbr_mnt(dir->i_sb, bindex) + }; + + rerr = au_wh_unlink_dentry(au_h_iptr(dir, bindex), &h_path, dentry); + if (!rerr) { + au_set_dbwh(dentry, bwh); + au_dtime_revert(dt); + return 0; + } + + AuIOErr("%.*s reverting whiteout failed(%d, %d)\n", + AuDLNPair(dentry), err, rerr); + return -EIO; +} + +/* ---------------------------------------------------------------------- */ + +int aufs_unlink(struct inode *dir, struct dentry *dentry) +{ + int err; + aufs_bindex_t bwh, bindex, bstart; + struct inode *inode, *h_dir; + struct dentry *parent, *wh_dentry; + /* to reuduce stack size */ + struct { + struct au_dtime dt; + struct au_pin pin; + struct path h_path; + } *a; + + IMustLock(dir); + + err = -ENOMEM; + a = kmalloc(sizeof(*a), GFP_NOFS); + if (unlikely(!a)) + goto out; + + err = aufs_read_lock(dentry, AuLock_DW | AuLock_GEN); + if (unlikely(err)) + goto out_free; + err = au_d_hashed_positive(dentry); + if (unlikely(err)) + goto out_unlock; + inode = dentry->d_inode; + IMustLock(inode); + err = -EISDIR; + if (unlikely(S_ISDIR(inode->i_mode))) + goto out_unlock; /* possible? */ + + bstart = au_dbstart(dentry); + bwh = au_dbwh(dentry); + bindex = -1; + parent = dentry->d_parent; /* dir inode is locked */ + di_write_lock_parent(parent); + wh_dentry = lock_hdir_create_wh(dentry, /*isdir*/0, &bindex, &a->dt, + &a->pin); + err = PTR_ERR(wh_dentry); + if (IS_ERR(wh_dentry)) + goto out_parent; + + a->h_path.mnt = au_sbr_mnt(dentry->d_sb, bstart); + a->h_path.dentry = au_h_dptr(dentry, bstart); + dget(a->h_path.dentry); + if (bindex == bstart) { + h_dir = au_pinned_h_dir(&a->pin); + err = vfsub_unlink(h_dir, &a->h_path, /*force*/0); + } else { + /* dir inode is locked */ + h_dir = wh_dentry->d_parent->d_inode; + IMustLock(h_dir); + err = 0; + } + + if (!err) { + vfsub_drop_nlink(inode); + epilog(dir, dentry, bindex); + + /* update target timestamps */ + if (bindex == bstart) { + vfsub_update_h_iattr(&a->h_path, /*did*/NULL); + /*ignore*/ + inode->i_ctime = a->h_path.dentry->d_inode->i_ctime; + } else + /* todo: this timestamp may be reverted later */ + inode->i_ctime = h_dir->i_ctime; + goto out_unpin; /* success */ + } + + /* revert */ + if (wh_dentry) { + int rerr; + + rerr = do_revert(err, dir, bindex, bwh, wh_dentry, dentry, + &a->dt); + if (rerr) + err = rerr; + } + +out_unpin: + au_unpin(&a->pin); + dput(wh_dentry); + dput(a->h_path.dentry); +out_parent: + di_write_unlock(parent); +out_unlock: + aufs_read_unlock(dentry, AuLock_DW); +out_free: + kfree(a); +out: + return err; +} + +int aufs_rmdir(struct inode *dir, struct dentry *dentry) +{ + int err, rmdir_later; + aufs_bindex_t bwh, bindex, bstart; + struct inode *inode; + struct dentry *parent, *wh_dentry, *h_dentry; + struct au_whtmp_rmdir *args; + /* to reuduce stack size */ + struct { + struct au_dtime dt; + struct au_pin pin; + } *a; + + IMustLock(dir); + + err = -ENOMEM; + a = kmalloc(sizeof(*a), GFP_NOFS); + if (unlikely(!a)) + goto out; + + err = aufs_read_lock(dentry, AuLock_DW | AuLock_FLUSH | AuLock_GEN); + if (unlikely(err)) + goto out_free; + err = au_alive_dir(dentry); + if (unlikely(err)) + goto out_unlock; + inode = dentry->d_inode; + IMustLock(inode); + err = -ENOTDIR; + if (unlikely(!S_ISDIR(inode->i_mode))) + goto out_unlock; /* possible? */ + + err = -ENOMEM; + args = au_whtmp_rmdir_alloc(dir->i_sb, GFP_NOFS); + if (unlikely(!args)) + goto out_unlock; + + parent = dentry->d_parent; /* dir inode is locked */ + di_write_lock_parent(parent); + err = au_test_empty(dentry, &args->whlist); + if (unlikely(err)) + goto out_parent; + + bstart = au_dbstart(dentry); + bwh = au_dbwh(dentry); + bindex = -1; + wh_dentry = lock_hdir_create_wh(dentry, /*isdir*/1, &bindex, &a->dt, + &a->pin); + err = PTR_ERR(wh_dentry); + if (IS_ERR(wh_dentry)) + goto out_parent; + + h_dentry = au_h_dptr(dentry, bstart); + dget(h_dentry); + rmdir_later = 0; + if (bindex == bstart) { + err = renwh_and_rmdir(dentry, bstart, &args->whlist, dir); + if (err > 0) { + rmdir_later = err; + err = 0; + } + } else { + /* stop monitoring */ + au_hn_free(au_hi(inode, bstart)); + + /* dir inode is locked */ + IMustLock(wh_dentry->d_parent->d_inode); + err = 0; + } + + if (!err) { + vfsub_dead_dir(inode); + au_set_dbdiropq(dentry, -1); + epilog(dir, dentry, bindex); + + if (rmdir_later) { + au_whtmp_kick_rmdir(dir, bstart, h_dentry, args); + args = NULL; + } + + goto out_unpin; /* success */ + } + + /* revert */ + AuLabel(revert); + if (wh_dentry) { + int rerr; + + rerr = do_revert(err, dir, bindex, bwh, wh_dentry, dentry, + &a->dt); + if (rerr) + err = rerr; + } + +out_unpin: + au_unpin(&a->pin); + dput(wh_dentry); + dput(h_dentry); +out_parent: + di_write_unlock(parent); + if (args) + au_whtmp_rmdir_free(args); +out_unlock: + aufs_read_unlock(dentry, AuLock_DW); +out_free: + kfree(a); +out: + AuTraceErr(err); + return err; +} diff -Naur kernel.21.3.orig/fs/aufs/i_op_ren.c kernel.21.3.new/fs/aufs/i_op_ren.c --- kernel.21.3.orig/fs/aufs/i_op_ren.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/i_op_ren.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,1010 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * inode operation (rename entry) + * todo: this is crazy monster + */ + +#include "aufs.h" + +enum { AuSRC, AuDST, AuSrcDst }; +enum { AuPARENT, AuCHILD, AuParentChild }; + +#define AuRen_ISDIR 1 +#define AuRen_ISSAMEDIR (1 << 1) +#define AuRen_WHSRC (1 << 2) +#define AuRen_WHDST (1 << 3) +#define AuRen_MNT_WRITE (1 << 4) +#define AuRen_DT_DSTDIR (1 << 5) +#define AuRen_DIROPQ (1 << 6) +#define AuRen_CPUP (1 << 7) +#define au_ftest_ren(flags, name) ((flags) & AuRen_##name) +#define au_fset_ren(flags, name) \ + do { (flags) |= AuRen_##name; } while (0) +#define au_fclr_ren(flags, name) \ + do { (flags) &= ~AuRen_##name; } while (0) + +struct au_ren_args { + struct { + struct dentry *dentry, *h_dentry, *parent, *h_parent, + *wh_dentry; + struct inode *dir, *inode; + struct au_hinode *hdir; + struct au_dtime dt[AuParentChild]; + aufs_bindex_t bstart; + } sd[AuSrcDst]; + +#define src_dentry sd[AuSRC].dentry +#define src_dir sd[AuSRC].dir +#define src_inode sd[AuSRC].inode +#define src_h_dentry sd[AuSRC].h_dentry +#define src_parent sd[AuSRC].parent +#define src_h_parent sd[AuSRC].h_parent +#define src_wh_dentry sd[AuSRC].wh_dentry +#define src_hdir sd[AuSRC].hdir +#define src_h_dir sd[AuSRC].hdir->hi_inode +#define src_dt sd[AuSRC].dt +#define src_bstart sd[AuSRC].bstart + +#define dst_dentry sd[AuDST].dentry +#define dst_dir sd[AuDST].dir +#define dst_inode sd[AuDST].inode +#define dst_h_dentry sd[AuDST].h_dentry +#define dst_parent sd[AuDST].parent +#define dst_h_parent sd[AuDST].h_parent +#define dst_wh_dentry sd[AuDST].wh_dentry +#define dst_hdir sd[AuDST].hdir +#define dst_h_dir sd[AuDST].hdir->hi_inode +#define dst_dt sd[AuDST].dt +#define dst_bstart sd[AuDST].bstart + + struct dentry *h_trap; + struct au_branch *br; + struct au_hinode *src_hinode; + struct path h_path; + struct au_nhash whlist; + aufs_bindex_t btgt, src_bwh, src_bdiropq; + + unsigned int flags; + + struct au_whtmp_rmdir *thargs; + struct dentry *h_dst; +}; + +/* ---------------------------------------------------------------------- */ + +/* + * functions for reverting. + * when an error happened in a single rename systemcall, we should revert + * everything as if nothing happend. + * we don't need to revert the copied-up/down the parent dir since they are + * harmless. + */ + +#define RevertFailure(fmt, ...) do { \ + AuIOErr("revert failure: " fmt " (%d, %d)\n", \ + ##__VA_ARGS__, err, rerr); \ + err = -EIO; \ +} while (0) + +static void au_ren_rev_diropq(int err, struct au_ren_args *a) +{ + int rerr; + + au_hn_imtx_lock_nested(a->src_hinode, AuLsc_I_CHILD); + rerr = au_diropq_remove(a->src_dentry, a->btgt); + au_hn_imtx_unlock(a->src_hinode); + au_set_dbdiropq(a->src_dentry, a->src_bdiropq); + if (rerr) + RevertFailure("remove diropq %.*s", AuDLNPair(a->src_dentry)); +} + +static void au_ren_rev_rename(int err, struct au_ren_args *a) +{ + int rerr; + + a->h_path.dentry = vfsub_lkup_one(&a->src_dentry->d_name, + a->src_h_parent); + rerr = PTR_ERR(a->h_path.dentry); + if (IS_ERR(a->h_path.dentry)) { + RevertFailure("lkup one %.*s", AuDLNPair(a->src_dentry)); + return; + } + + rerr = vfsub_rename(a->dst_h_dir, + au_h_dptr(a->src_dentry, a->btgt), + a->src_h_dir, &a->h_path); + d_drop(a->h_path.dentry); + dput(a->h_path.dentry); + /* au_set_h_dptr(a->src_dentry, a->btgt, NULL); */ + if (rerr) + RevertFailure("rename %.*s", AuDLNPair(a->src_dentry)); +} + +static void au_ren_rev_cpup(int err, struct au_ren_args *a) +{ + int rerr; + + a->h_path.dentry = a->dst_h_dentry; + rerr = vfsub_unlink(a->dst_h_dir, &a->h_path, /*force*/0); + au_set_h_dptr(a->src_dentry, a->btgt, NULL); + au_set_dbstart(a->src_dentry, a->src_bstart); + if (rerr) + RevertFailure("unlink %.*s", AuDLNPair(a->dst_h_dentry)); +} + +static void au_ren_rev_whtmp(int err, struct au_ren_args *a) +{ + int rerr; + + a->h_path.dentry = vfsub_lkup_one(&a->dst_dentry->d_name, + a->dst_h_parent); + rerr = PTR_ERR(a->h_path.dentry); + if (IS_ERR(a->h_path.dentry)) { + RevertFailure("lkup one %.*s", AuDLNPair(a->dst_dentry)); + return; + } + if (a->h_path.dentry->d_inode) { + d_drop(a->h_path.dentry); + dput(a->h_path.dentry); + return; + } + + rerr = vfsub_rename(a->dst_h_dir, a->h_dst, a->dst_h_dir, &a->h_path); + d_drop(a->h_path.dentry); + dput(a->h_path.dentry); + if (!rerr) + au_set_h_dptr(a->dst_dentry, a->btgt, dget(a->h_dst)); + else + RevertFailure("rename %.*s", AuDLNPair(a->h_dst)); +} + +static void au_ren_rev_whsrc(int err, struct au_ren_args *a) +{ + int rerr; + + a->h_path.dentry = a->src_wh_dentry; + rerr = au_wh_unlink_dentry(a->src_h_dir, &a->h_path, a->src_dentry); + au_set_dbwh(a->src_dentry, a->src_bwh); + if (rerr) + RevertFailure("unlink %.*s", AuDLNPair(a->src_wh_dentry)); +} +#undef RevertFailure + +/* ---------------------------------------------------------------------- */ + +/* + * when we have to copyup the renaming entry, do it with the rename-target name + * in order to minimize the cost (the later actual rename is unnecessary). + * otherwise rename it on the target branch. + */ +static int au_ren_or_cpup(struct au_ren_args *a) +{ + int err; + struct dentry *d; + + d = a->src_dentry; + if (au_dbstart(d) == a->btgt) { + a->h_path.dentry = a->dst_h_dentry; + if (au_ftest_ren(a->flags, DIROPQ) + && au_dbdiropq(d) == a->btgt) + au_fclr_ren(a->flags, DIROPQ); + AuDebugOn(au_dbstart(d) != a->btgt); + err = vfsub_rename(a->src_h_dir, au_h_dptr(d, a->btgt), + a->dst_h_dir, &a->h_path); + } else + BUG(); + + if (!err && a->h_dst) + /* it will be set to dinfo later */ + dget(a->h_dst); + + return err; +} + +/* cf. aufs_rmdir() */ +static int au_ren_del_whtmp(struct au_ren_args *a) +{ + int err; + struct inode *dir; + + dir = a->dst_dir; + SiMustAnyLock(dir->i_sb); + if (!au_nhash_test_longer_wh(&a->whlist, a->btgt, + au_sbi(dir->i_sb)->si_dirwh) + || au_test_fs_remote(a->h_dst->d_sb)) { + err = au_whtmp_rmdir(dir, a->btgt, a->h_dst, &a->whlist); + if (unlikely(err)) + pr_warn("failed removing whtmp dir %.*s (%d), " + "ignored.\n", AuDLNPair(a->h_dst), err); + } else { + au_nhash_wh_free(&a->thargs->whlist); + a->thargs->whlist = a->whlist; + a->whlist.nh_num = 0; + au_whtmp_kick_rmdir(dir, a->btgt, a->h_dst, a->thargs); + dput(a->h_dst); + a->thargs = NULL; + } + + return 0; +} + +/* make it 'opaque' dir. */ +static int au_ren_diropq(struct au_ren_args *a) +{ + int err; + struct dentry *diropq; + + err = 0; + a->src_bdiropq = au_dbdiropq(a->src_dentry); + a->src_hinode = au_hi(a->src_inode, a->btgt); + au_hn_imtx_lock_nested(a->src_hinode, AuLsc_I_CHILD); + diropq = au_diropq_create(a->src_dentry, a->btgt); + au_hn_imtx_unlock(a->src_hinode); + if (IS_ERR(diropq)) + err = PTR_ERR(diropq); + else + dput(diropq); + + return err; +} + +static int do_rename(struct au_ren_args *a) +{ + int err; + struct dentry *d, *h_d; + + /* prepare workqueue args for asynchronous rmdir */ + h_d = a->dst_h_dentry; + if (au_ftest_ren(a->flags, ISDIR) && h_d->d_inode) { + err = -ENOMEM; + a->thargs = au_whtmp_rmdir_alloc(a->src_dentry->d_sb, GFP_NOFS); + if (unlikely(!a->thargs)) + goto out; + a->h_dst = dget(h_d); + } + + /* create whiteout for src_dentry */ + if (au_ftest_ren(a->flags, WHSRC)) { + a->src_bwh = au_dbwh(a->src_dentry); + AuDebugOn(a->src_bwh >= 0); + a->src_wh_dentry + = au_wh_create(a->src_dentry, a->btgt, a->src_h_parent); + err = PTR_ERR(a->src_wh_dentry); + if (IS_ERR(a->src_wh_dentry)) + goto out_thargs; + } + + /* lookup whiteout for dentry */ + if (au_ftest_ren(a->flags, WHDST)) { + h_d = au_wh_lkup(a->dst_h_parent, &a->dst_dentry->d_name, + a->br); + err = PTR_ERR(h_d); + if (IS_ERR(h_d)) + goto out_whsrc; + if (!h_d->d_inode) + dput(h_d); + else + a->dst_wh_dentry = h_d; + } + + /* rename dentry to tmpwh */ + if (a->thargs) { + err = au_whtmp_ren(a->dst_h_dentry, a->br); + if (unlikely(err)) + goto out_whdst; + + d = a->dst_dentry; + au_set_h_dptr(d, a->btgt, NULL); + err = au_lkup_neg(d, a->btgt, /*wh*/0); + if (unlikely(err)) + goto out_whtmp; + a->dst_h_dentry = au_h_dptr(d, a->btgt); + } + + BUG_ON(a->dst_h_dentry->d_inode && a->src_bstart != a->btgt); + + /* rename by vfs_rename or cpup */ + d = a->dst_dentry; + if (au_ftest_ren(a->flags, ISDIR) + && (a->dst_wh_dentry + || au_dbdiropq(d) == a->btgt + /* hide the lower to keep xino */ + || a->btgt < au_dbend(d) + || au_opt_test(au_mntflags(d->d_sb), ALWAYS_DIROPQ))) + au_fset_ren(a->flags, DIROPQ); + err = au_ren_or_cpup(a); + if (unlikely(err)) + /* leave the copied-up one */ + goto out_whtmp; + + /* make dir opaque */ + if (au_ftest_ren(a->flags, DIROPQ)) { + err = au_ren_diropq(a); + if (unlikely(err)) + goto out_rename; + } + + /* update target timestamps */ + AuDebugOn(au_dbstart(a->src_dentry) != a->btgt); + a->h_path.dentry = au_h_dptr(a->src_dentry, a->btgt); + vfsub_update_h_iattr(&a->h_path, /*did*/NULL); /*ignore*/ + a->src_inode->i_ctime = a->h_path.dentry->d_inode->i_ctime; + + /* remove whiteout for dentry */ + if (a->dst_wh_dentry) { + a->h_path.dentry = a->dst_wh_dentry; + err = au_wh_unlink_dentry(a->dst_h_dir, &a->h_path, + a->dst_dentry); + if (unlikely(err)) + goto out_diropq; + } + + /* remove whtmp */ + if (a->thargs) + au_ren_del_whtmp(a); /* ignore this error */ + + au_fhsm_wrote(a->src_dentry->d_sb, a->btgt, /*force*/0); + err = 0; + goto out_success; + +out_diropq: + if (au_ftest_ren(a->flags, DIROPQ)) + au_ren_rev_diropq(err, a); +out_rename: + if (!au_ftest_ren(a->flags, CPUP)) + au_ren_rev_rename(err, a); + else + au_ren_rev_cpup(err, a); + dput(a->h_dst); +out_whtmp: + if (a->thargs) + au_ren_rev_whtmp(err, a); +out_whdst: + dput(a->dst_wh_dentry); + a->dst_wh_dentry = NULL; +out_whsrc: + if (a->src_wh_dentry) + au_ren_rev_whsrc(err, a); +out_success: + dput(a->src_wh_dentry); + dput(a->dst_wh_dentry); +out_thargs: + if (a->thargs) { + dput(a->h_dst); + au_whtmp_rmdir_free(a->thargs); + a->thargs = NULL; + } +out: + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* + * test if @dentry dir can be rename destination or not. + * success means, it is a logically empty dir. + */ +static int may_rename_dstdir(struct dentry *dentry, struct au_nhash *whlist) +{ + return au_test_empty(dentry, whlist); +} + +/* + * test if @dentry dir can be rename source or not. + * if it can, return 0 and @children is filled. + * success means, + * - it is a logically empty dir. + * - or, it exists on writable branch and has no children including whiteouts + * on the lower branch. + */ +static int may_rename_srcdir(struct dentry *dentry, aufs_bindex_t btgt) +{ + int err; + unsigned int rdhash; + aufs_bindex_t bstart; + + bstart = au_dbstart(dentry); + if (bstart != btgt) { + struct au_nhash whlist; + + SiMustAnyLock(dentry->d_sb); + rdhash = au_sbi(dentry->d_sb)->si_rdhash; + if (!rdhash) + rdhash = au_rdhash_est(au_dir_size(/*file*/NULL, + dentry)); + err = au_nhash_alloc(&whlist, rdhash, GFP_NOFS); + if (unlikely(err)) + goto out; + err = au_test_empty(dentry, &whlist); + au_nhash_wh_free(&whlist); + goto out; + } + + if (bstart == au_dbtaildir(dentry)) + return 0; /* success */ + + err = au_test_empty_lower(dentry); + +out: + if (err == -ENOTEMPTY) { + AuWarn1("renaming dir who has child(ren) on multiple branches," + " is not supported\n"); + err = -EXDEV; + } + return err; +} + +/* side effect: sets whlist and h_dentry */ +static int au_ren_may_dir(struct au_ren_args *a) +{ + int err; + unsigned int rdhash; + struct dentry *d; + + d = a->dst_dentry; + SiMustAnyLock(d->d_sb); + + err = 0; + if (au_ftest_ren(a->flags, ISDIR) && a->dst_inode) { + rdhash = au_sbi(d->d_sb)->si_rdhash; + if (!rdhash) + rdhash = au_rdhash_est(au_dir_size(/*file*/NULL, d)); + err = au_nhash_alloc(&a->whlist, rdhash, GFP_NOFS); + if (unlikely(err)) + goto out; + + au_set_dbstart(d, a->dst_bstart); + err = may_rename_dstdir(d, &a->whlist); + au_set_dbstart(d, a->btgt); + } + a->dst_h_dentry = au_h_dptr(d, au_dbstart(d)); + if (unlikely(err)) + goto out; + + d = a->src_dentry; + a->src_h_dentry = au_h_dptr(d, au_dbstart(d)); + if (au_ftest_ren(a->flags, ISDIR)) { + err = may_rename_srcdir(d, a->btgt); + if (unlikely(err)) { + au_nhash_wh_free(&a->whlist); + a->whlist.nh_num = 0; + } + } +out: + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* + * simple tests for rename. + * following the checks in vfs, plus the parent-child relationship. + */ +static int au_may_ren(struct au_ren_args *a) +{ + int err, isdir; + struct inode *h_inode; + + if (a->src_bstart == a->btgt) { + err = au_may_del(a->src_dentry, a->btgt, a->src_h_parent, + au_ftest_ren(a->flags, ISDIR)); + if (unlikely(err)) + goto out; + err = -EINVAL; + if (unlikely(a->src_h_dentry == a->h_trap)) + goto out; + } + + err = 0; + if (a->dst_bstart != a->btgt) + goto out; + + err = -ENOTEMPTY; + if (unlikely(a->dst_h_dentry == a->h_trap)) + goto out; + + err = -EIO; + h_inode = a->dst_h_dentry->d_inode; + isdir = !!au_ftest_ren(a->flags, ISDIR); + if (!a->dst_dentry->d_inode) { + if (unlikely(h_inode)) + goto out; + err = au_may_add(a->dst_dentry, a->btgt, a->dst_h_parent, + isdir); + } else { + if (unlikely(!h_inode || !h_inode->i_nlink)) + goto out; + err = au_may_del(a->dst_dentry, a->btgt, a->dst_h_parent, + isdir); + if (unlikely(err)) + goto out; + } + +out: + if (unlikely(err == -ENOENT || err == -EEXIST)) + err = -EIO; + AuTraceErr(err); + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* + * locking order + * (VFS) + * - src_dir and dir by lock_rename() + * - inode if exitsts + * (aufs) + * - lock all + * + src_dentry and dentry by aufs_read_and_write_lock2() which calls, + * + si_read_lock + * + di_write_lock2_child() + * + di_write_lock_child() + * + ii_write_lock_child() + * + di_write_lock_child2() + * + ii_write_lock_child2() + * + src_parent and parent + * + di_write_lock_parent() + * + ii_write_lock_parent() + * + di_write_lock_parent2() + * + ii_write_lock_parent2() + * + lower src_dir and dir by vfsub_lock_rename() + * + verify the every relationships between child and parent. if any + * of them failed, unlock all and return -EBUSY. + */ +static void au_ren_unlock(struct au_ren_args *a) +{ + vfsub_unlock_rename(a->src_h_parent, a->src_hdir, + a->dst_h_parent, a->dst_hdir); + if (au_ftest_ren(a->flags, MNT_WRITE)) + vfsub_mnt_drop_write(au_br_mnt(a->br)); +} + +static int au_ren_lock(struct au_ren_args *a) +{ + int err; + unsigned int udba; + + err = 0; + a->src_h_parent = au_h_dptr(a->src_parent, a->btgt); + a->src_hdir = au_hi(a->src_dir, a->btgt); + a->dst_h_parent = au_h_dptr(a->dst_parent, a->btgt); + a->dst_hdir = au_hi(a->dst_dir, a->btgt); + + err = vfsub_mnt_want_write(au_br_mnt(a->br)); + if (unlikely(err)) + goto out; + au_fset_ren(a->flags, MNT_WRITE); + a->h_trap = vfsub_lock_rename(a->src_h_parent, a->src_hdir, + a->dst_h_parent, a->dst_hdir); + udba = au_opt_udba(a->src_dentry->d_sb); + if (unlikely(a->src_hdir->hi_inode != a->src_h_parent->d_inode + || a->dst_hdir->hi_inode != a->dst_h_parent->d_inode)) + err = au_busy_or_stale(); + if (!err && au_dbstart(a->src_dentry) == a->btgt) + err = au_h_verify(a->src_h_dentry, udba, + a->src_h_parent->d_inode, a->src_h_parent, + a->br); + if (!err && au_dbstart(a->dst_dentry) == a->btgt) + err = au_h_verify(a->dst_h_dentry, udba, + a->dst_h_parent->d_inode, a->dst_h_parent, + a->br); + if (!err) + goto out; /* success */ + + err = au_busy_or_stale(); + au_ren_unlock(a); + +out: + return err; +} + +/* ---------------------------------------------------------------------- */ + +static void au_ren_refresh_dir(struct au_ren_args *a) +{ + struct inode *dir; + + dir = a->dst_dir; + dir->i_version++; + if (au_ftest_ren(a->flags, ISDIR)) { + /* is this updating defined in POSIX? */ + au_cpup_attr_timesizes(a->src_inode); + au_cpup_attr_nlink(dir, /*force*/1); + } + + if (au_ibstart(dir) == a->btgt) + au_cpup_attr_timesizes(dir); + + if (au_ftest_ren(a->flags, ISSAMEDIR)) + return; + + dir = a->src_dir; + dir->i_version++; + if (au_ftest_ren(a->flags, ISDIR)) + au_cpup_attr_nlink(dir, /*force*/1); + if (au_ibstart(dir) == a->btgt) + au_cpup_attr_timesizes(dir); +} + +static void au_ren_refresh(struct au_ren_args *a) +{ + aufs_bindex_t bend, bindex; + struct dentry *d, *h_d; + struct inode *i, *h_i; + struct super_block *sb; + + d = a->dst_dentry; + d_drop(d); + if (a->h_dst) + /* already dget-ed by au_ren_or_cpup() */ + au_set_h_dptr(d, a->btgt, a->h_dst); + + i = a->dst_inode; + if (i) { + if (!au_ftest_ren(a->flags, ISDIR)) + vfsub_drop_nlink(i); + else { + vfsub_dead_dir(i); + au_cpup_attr_timesizes(i); + } + au_update_dbrange(d, /*do_put_zero*/1); + } else { + bend = a->btgt; + for (bindex = au_dbstart(d); bindex < bend; bindex++) + au_set_h_dptr(d, bindex, NULL); + bend = au_dbend(d); + for (bindex = a->btgt + 1; bindex <= bend; bindex++) + au_set_h_dptr(d, bindex, NULL); + au_update_dbrange(d, /*do_put_zero*/0); + } + + d = a->src_dentry; + au_set_dbwh(d, -1); + bend = au_dbend(d); + for (bindex = a->btgt + 1; bindex <= bend; bindex++) { + h_d = au_h_dptr(d, bindex); + if (h_d) + au_set_h_dptr(d, bindex, NULL); + } + au_set_dbend(d, a->btgt); + + sb = d->d_sb; + i = a->src_inode; + if (au_opt_test(au_mntflags(sb), PLINK) && au_plink_test(i)) + return; /* success */ + + bend = au_ibend(i); + for (bindex = a->btgt + 1; bindex <= bend; bindex++) { + h_i = au_h_iptr(i, bindex); + if (h_i) { + au_xino_write(sb, bindex, h_i->i_ino, /*ino*/0); + /* ignore this error */ + au_set_h_iptr(i, bindex, NULL, 0); + } + } + au_set_ibend(i, a->btgt); +} + +/* ---------------------------------------------------------------------- */ + +/* mainly for link(2) and rename(2) */ +int au_wbr(struct dentry *dentry, aufs_bindex_t btgt) +{ + aufs_bindex_t bdiropq, bwh; + struct dentry *parent; + struct au_branch *br; + + parent = dentry->d_parent; + IMustLock(parent->d_inode); /* dir is locked */ + + bdiropq = au_dbdiropq(parent); + bwh = au_dbwh(dentry); + br = au_sbr(dentry->d_sb, btgt); + if (au_br_rdonly(br) + || (0 <= bdiropq && bdiropq < btgt) + || (0 <= bwh && bwh < btgt)) + btgt = -1; + + AuDbg("btgt %d\n", btgt); + return btgt; +} + +/* sets src_bstart, dst_bstart and btgt */ +static int au_ren_wbr(struct au_ren_args *a) +{ + int err; + struct au_wr_dir_args wr_dir_args = { + /* .force_btgt = -1, */ + .flags = AuWrDir_ADD_ENTRY + }; + + a->src_bstart = au_dbstart(a->src_dentry); + a->dst_bstart = au_dbstart(a->dst_dentry); + if (au_ftest_ren(a->flags, ISDIR)) + au_fset_wrdir(wr_dir_args.flags, ISDIR); + wr_dir_args.force_btgt = a->src_bstart; + if (a->dst_inode && a->dst_bstart < a->src_bstart) + wr_dir_args.force_btgt = a->dst_bstart; + wr_dir_args.force_btgt = au_wbr(a->dst_dentry, wr_dir_args.force_btgt); + err = au_wr_dir(a->dst_dentry, a->src_dentry, &wr_dir_args); + a->btgt = err; + + return err; +} + +static void au_ren_dt(struct au_ren_args *a) +{ + a->h_path.dentry = a->src_h_parent; + au_dtime_store(a->src_dt + AuPARENT, a->src_parent, &a->h_path); + if (!au_ftest_ren(a->flags, ISSAMEDIR)) { + a->h_path.dentry = a->dst_h_parent; + au_dtime_store(a->dst_dt + AuPARENT, a->dst_parent, &a->h_path); + } + + au_fclr_ren(a->flags, DT_DSTDIR); + if (!au_ftest_ren(a->flags, ISDIR)) + return; + + a->h_path.dentry = a->src_h_dentry; + au_dtime_store(a->src_dt + AuCHILD, a->src_dentry, &a->h_path); + if (a->dst_h_dentry->d_inode) { + au_fset_ren(a->flags, DT_DSTDIR); + a->h_path.dentry = a->dst_h_dentry; + au_dtime_store(a->dst_dt + AuCHILD, a->dst_dentry, &a->h_path); + } +} + +static void au_ren_rev_dt(int err, struct au_ren_args *a) +{ + struct dentry *h_d; + struct mutex *h_mtx; + + au_dtime_revert(a->src_dt + AuPARENT); + if (!au_ftest_ren(a->flags, ISSAMEDIR)) + au_dtime_revert(a->dst_dt + AuPARENT); + + if (au_ftest_ren(a->flags, ISDIR) && err != -EIO) { + h_d = a->src_dt[AuCHILD].dt_h_path.dentry; + h_mtx = &h_d->d_inode->i_mutex; + mutex_lock_nested(h_mtx, AuLsc_I_CHILD); + au_dtime_revert(a->src_dt + AuCHILD); + mutex_unlock(h_mtx); + + if (au_ftest_ren(a->flags, DT_DSTDIR)) { + h_d = a->dst_dt[AuCHILD].dt_h_path.dentry; + h_mtx = &h_d->d_inode->i_mutex; + mutex_lock_nested(h_mtx, AuLsc_I_CHILD); + au_dtime_revert(a->dst_dt + AuCHILD); + mutex_unlock(h_mtx); + } + } +} + +/* ---------------------------------------------------------------------- */ + +int aufs_rename(struct inode *_src_dir, struct dentry *_src_dentry, + struct inode *_dst_dir, struct dentry *_dst_dentry) +{ + int err, flags; + /* reduce stack space */ + struct au_ren_args *a; + + AuDbg("%.*s, %.*s\n", AuDLNPair(_src_dentry), AuDLNPair(_dst_dentry)); + IMustLock(_src_dir); + IMustLock(_dst_dir); + + err = -ENOMEM; + BUILD_BUG_ON(sizeof(*a) > PAGE_SIZE); + a = kzalloc(sizeof(*a), GFP_NOFS); + if (unlikely(!a)) + goto out; + + a->src_dir = _src_dir; + a->src_dentry = _src_dentry; + a->src_inode = a->src_dentry->d_inode; + a->src_parent = a->src_dentry->d_parent; /* dir inode is locked */ + a->dst_dir = _dst_dir; + a->dst_dentry = _dst_dentry; + a->dst_inode = a->dst_dentry->d_inode; + a->dst_parent = a->dst_dentry->d_parent; /* dir inode is locked */ + if (a->dst_inode) { + IMustLock(a->dst_inode); + au_igrab(a->dst_inode); + } + + err = -ENOTDIR; + flags = AuLock_FLUSH | AuLock_NOPLM | AuLock_GEN; + if (S_ISDIR(a->src_inode->i_mode)) { + au_fset_ren(a->flags, ISDIR); + if (unlikely(a->dst_inode && !S_ISDIR(a->dst_inode->i_mode))) + goto out_free; + err = aufs_read_and_write_lock2(a->dst_dentry, a->src_dentry, + AuLock_DIR | flags); + } else + err = aufs_read_and_write_lock2(a->dst_dentry, a->src_dentry, + flags); + if (unlikely(err)) + goto out_free; + + err = au_d_hashed_positive(a->src_dentry); + if (unlikely(err)) + goto out_unlock; + err = -ENOENT; + if (a->dst_inode) { + /* + * If it is a dir, VFS unhash dst_dentry before this + * function. It means we cannot rely upon d_unhashed(). + */ + if (unlikely(!a->dst_inode->i_nlink)) + goto out_unlock; + if (!S_ISDIR(a->dst_inode->i_mode)) { + err = au_d_hashed_positive(a->dst_dentry); + if (unlikely(err)) + goto out_unlock; + } else if (unlikely(IS_DEADDIR(a->dst_inode))) + goto out_unlock; + } else if (unlikely(d_unhashed(a->dst_dentry))) + goto out_unlock; + + /* + * is it possible? + * yes, it happend (in linux-3.3-rcN) but I don't know why. + * there may exist a problem somewhere else. + */ + err = -EINVAL; + if (unlikely(a->dst_parent->d_inode == a->src_dentry->d_inode)) + goto out_unlock; + + au_fset_ren(a->flags, ISSAMEDIR); /* temporary */ + di_write_lock_parent(a->dst_parent); + + /* which branch we process */ + err = au_ren_wbr(a); + if (unlikely(err < 0)) + goto out_parent; + a->br = au_sbr(a->dst_dentry->d_sb, a->btgt); + a->h_path.mnt = au_br_mnt(a->br); + + /* are they available to be renamed */ + err = au_ren_may_dir(a); + if (unlikely(err)) + goto out_children; + + /* prepare the writable parent dir on the same branch */ + if (a->dst_bstart == a->btgt) { + au_fset_ren(a->flags, WHDST); + } else { + err = au_cpup_dirs(a->dst_dentry, a->btgt); + if (unlikely(err)) + goto out_children; + } + + if (a->src_dir != a->dst_dir) { + /* + * this temporary unlock is safe, + * because both dir->i_mutex are locked. + */ + di_write_unlock(a->dst_parent); + di_write_lock_parent(a->src_parent); + err = au_wr_dir_need_wh(a->src_dentry, + au_ftest_ren(a->flags, ISDIR), + &a->btgt); + di_write_unlock(a->src_parent); + di_write_lock2_parent(a->src_parent, a->dst_parent, /*isdir*/1); + au_fclr_ren(a->flags, ISSAMEDIR); + } else + err = au_wr_dir_need_wh(a->src_dentry, + au_ftest_ren(a->flags, ISDIR), + &a->btgt); + if (unlikely(err < 0)) + goto out_children; + if (err) + au_fset_ren(a->flags, WHSRC); + + /* cpup src */ + if (a->src_bstart != a->btgt) { + struct au_pin pin; + + err = au_pin(&pin, a->src_dentry, a->btgt, + au_opt_udba(a->src_dentry->d_sb), + AuPin_DI_LOCKED | AuPin_MNT_WRITE); + if (!err) { + struct au_cp_generic cpg = { + .dentry = a->src_dentry, + .bdst = a->btgt, + .bsrc = a->src_bstart, + .len = -1, + .pin = &pin, + .flags = AuCpup_DTIME | AuCpup_HOPEN + }; + AuDebugOn(au_dbstart(a->src_dentry) != a->src_bstart); + err = au_sio_cpup_simple(&cpg); + au_unpin(&pin); + } + if (unlikely(err)) + goto out_children; + a->src_bstart = a->btgt; + a->src_h_dentry = au_h_dptr(a->src_dentry, a->btgt); + au_fset_ren(a->flags, WHSRC); + } + + /* lock them all */ + err = au_ren_lock(a); + if (unlikely(err)) + /* leave the copied-up one */ + goto out_children; + + if (!au_opt_test(au_mntflags(a->dst_dir->i_sb), UDBA_NONE)) + err = au_may_ren(a); + else if (unlikely(a->dst_dentry->d_name.len > AUFS_MAX_NAMELEN)) + err = -ENAMETOOLONG; + if (unlikely(err)) + goto out_hdir; + + /* store timestamps to be revertible */ + au_ren_dt(a); + + /* here we go */ + err = do_rename(a); + if (unlikely(err)) + goto out_dt; + + /* update dir attributes */ + au_ren_refresh_dir(a); + + /* dput/iput all lower dentries */ + au_ren_refresh(a); + + goto out_hdir; /* success */ + +out_dt: + au_ren_rev_dt(err, a); +out_hdir: + au_ren_unlock(a); +out_children: + au_nhash_wh_free(&a->whlist); + if (err && a->dst_inode && a->dst_bstart != a->btgt) { + AuDbg("bstart %d, btgt %d\n", a->dst_bstart, a->btgt); + au_set_h_dptr(a->dst_dentry, a->btgt, NULL); + au_set_dbstart(a->dst_dentry, a->dst_bstart); + } +out_parent: + if (!err) + d_move(a->src_dentry, a->dst_dentry); + else { + au_update_dbstart(a->dst_dentry); + if (!a->dst_inode) + d_drop(a->dst_dentry); + } + if (au_ftest_ren(a->flags, ISSAMEDIR)) + di_write_unlock(a->dst_parent); + else + di_write_unlock2(a->src_parent, a->dst_parent); +out_unlock: + aufs_read_and_write_unlock2(a->dst_dentry, a->src_dentry); +out_free: + iput(a->dst_inode); + if (a->thargs) + au_whtmp_rmdir_free(a->thargs); + kfree(a); +out: + AuTraceErr(err); + return err; +} diff -Naur kernel.21.3.orig/fs/aufs/Kconfig kernel.21.3.new/fs/aufs/Kconfig --- kernel.21.3.orig/fs/aufs/Kconfig 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/Kconfig 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,185 @@ +config AUFS_FS + tristate "Aufs (Advanced multi layered unification filesystem) support" + help + Aufs is a stackable unification filesystem such as Unionfs, + which unifies several directories and provides a merged single + directory. + In the early days, aufs was entirely re-designed and + re-implemented Unionfs Version 1.x series. Introducing many + original ideas, approaches and improvements, it becomes totally + different from Unionfs while keeping the basic features. + +if AUFS_FS +choice + prompt "Maximum number of branches" + default AUFS_BRANCH_MAX_127 + help + Specifies the maximum number of branches (or member directories) + in a single aufs. The larger value consumes more system + resources and has a minor impact to performance. +config AUFS_BRANCH_MAX_127 + bool "127" + help + Specifies the maximum number of branches (or member directories) + in a single aufs. The larger value consumes more system + resources and has a minor impact to performance. +config AUFS_BRANCH_MAX_511 + bool "511" + help + Specifies the maximum number of branches (or member directories) + in a single aufs. The larger value consumes more system + resources and has a minor impact to performance. +config AUFS_BRANCH_MAX_1023 + bool "1023" + help + Specifies the maximum number of branches (or member directories) + in a single aufs. The larger value consumes more system + resources and has a minor impact to performance. +config AUFS_BRANCH_MAX_32767 + bool "32767" + help + Specifies the maximum number of branches (or member directories) + in a single aufs. The larger value consumes more system + resources and has a minor impact to performance. +endchoice + +config AUFS_SBILIST + bool + depends on AUFS_MAGIC_SYSRQ || PROC_FS + default y + help + Automatic configuration for internal use. + When aufs supports Magic SysRq or /proc, enabled automatically. + +config AUFS_HNOTIFY + bool "Detect direct branch access (bypassing aufs)" + help + If you want to modify files on branches directly, eg. bypassing aufs, + and want aufs to detect the changes of them fully, then enable this + option and use 'udba=notify' mount option. + Currently there is only one available configuration, "fsnotify". + It will have a negative impact to the performance. + See detail in aufs.5. + +choice + prompt "method" if AUFS_HNOTIFY + default AUFS_HFSNOTIFY +config AUFS_HFSNOTIFY + bool "fsnotify" + select FSNOTIFY +endchoice + +config AUFS_EXPORT + bool "NFS-exportable aufs" + depends on EXPORTFS + help + If you want to export your mounted aufs via NFS, then enable this + option. There are several requirements for this configuration. + See detail in aufs.5. + +config AUFS_INO_T_64 + bool + depends on AUFS_EXPORT + depends on 64BIT && !(ALPHA || S390) + default y + help + Automatic configuration for internal use. + /* typedef unsigned long/int __kernel_ino_t */ + /* alpha and s390x are int */ + +config AUFS_XATTR + bool "support for XATTR/EA (including Security Labels)" + help + If your branch fs supports XATTR/EA and you want to make them + available in aufs too, then enable this opsion and specify the + branch attributes for EA. + See detail in aufs.5. + +config AUFS_FHSM + bool "File-based Hierarchical Storage Management" + help + Hierarchical Storage Management (or HSM) is a well-known feature + in the storage world. Aufs provides this feature as file-based. + with multiple branches. + These multiple branches are prioritized, ie. the topmost one + should be the fastest drive and be used heavily. + +config AUFS_RDU + bool "Readdir in userspace" + help + Aufs has two methods to provide a merged view for a directory, + by a user-space library and by kernel-space natively. The latter + is always enabled but sometimes large and slow. + If you enable this option, install the library in aufs2-util + package, and set some environment variables for your readdir(3), + then the work will be handled in user-space which generally + shows better performance in most cases. + See detail in aufs.5. + +config AUFS_SHWH + bool "Show whiteouts" + help + If you want to make the whiteouts in aufs visible, then enable + this option and specify 'shwh' mount option. Although it may + sounds like philosophy or something, but in technically it + simply shows the name of whiteout with keeping its behaviour. + +config AUFS_BR_RAMFS + bool "Ramfs (initramfs/rootfs) as an aufs branch" + help + If you want to use ramfs as an aufs branch fs, then enable this + option. Generally tmpfs is recommended. + Aufs prohibited them to be a branch fs by default, because + initramfs becomes unusable after switch_root or something + generally. If you sets initramfs as an aufs branch and boot your + system by switch_root, you will meet a problem easily since the + files in initramfs may be inaccessible. + Unless you are going to use ramfs as an aufs branch fs without + switch_root or something, leave it N. + +config AUFS_BR_FUSE + bool "Fuse fs as an aufs branch" + depends on FUSE_FS + select AUFS_POLL + help + If you want to use fuse-based userspace filesystem as an aufs + branch fs, then enable this option. + It implements the internal poll(2) operation which is + implemented by fuse only (curretnly). + +config AUFS_POLL + bool + help + Automatic configuration for internal use. + +config AUFS_BR_HFSPLUS + bool "Hfsplus as an aufs branch" + depends on HFSPLUS_FS + default y + help + If you want to use hfsplus fs as an aufs branch fs, then enable + this option. This option introduces a small overhead at + copying-up a file on hfsplus. + +config AUFS_BDEV_LOOP + bool + depends on BLK_DEV_LOOP + default y + help + Automatic configuration for internal use. + Convert =[ym] into =y. + +config AUFS_DEBUG + bool "Debug aufs" + help + Enable this to compile aufs internal debug code. + It will have a negative impact to the performance. + +config AUFS_MAGIC_SYSRQ + bool + depends on AUFS_DEBUG && MAGIC_SYSRQ + default y + help + Automatic configuration for internal use. + When aufs supports Magic SysRq, enabled automatically. +endif diff -Naur kernel.21.3.orig/fs/aufs/loop.c kernel.21.3.new/fs/aufs/loop.c --- kernel.21.3.orig/fs/aufs/loop.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/loop.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,134 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * support for loopback block device as a branch + */ + +#include +#include "aufs.h" + +/* + * test if two lower dentries have overlapping branches. + */ +int au_test_loopback_overlap(struct super_block *sb, struct dentry *h_adding) +{ + struct super_block *h_sb; + struct loop_device *l; + + h_sb = h_adding->d_sb; + if (MAJOR(h_sb->s_dev) != LOOP_MAJOR) + return 0; + + l = h_sb->s_bdev->bd_disk->private_data; + h_adding = l->lo_backing_file->f_dentry; + /* + * h_adding can be local NFS. + * in this case aufs cannot detect the loop. + */ + if (unlikely(h_adding->d_sb == sb)) + return 1; + return !!au_test_subdir(h_adding, sb->s_root); +} + +/* true if a kernel thread named 'loop[0-9].*' accesses a file */ +int au_test_loopback_kthread(void) +{ + int ret; + struct task_struct *tsk = current; + char c, comm[sizeof(tsk->comm)]; + + ret = 0; + if (tsk->flags & PF_KTHREAD) { + get_task_comm(comm, tsk); + c = comm[4]; + ret = ('0' <= c && c <= '9' + && !strncmp(comm, "loop", 4)); + } + + return ret; +} + +/* ---------------------------------------------------------------------- */ + +#define au_warn_loopback_step 16 +static int au_warn_loopback_nelem = au_warn_loopback_step; +static unsigned long *au_warn_loopback_array; + +void au_warn_loopback(struct super_block *h_sb) +{ + int i, new_nelem; + unsigned long *a, magic; + static DEFINE_SPINLOCK(spin); + + magic = h_sb->s_magic; + spin_lock(&spin); + a = au_warn_loopback_array; + for (i = 0; i < au_warn_loopback_nelem && *a; i++) + if (a[i] == magic) { + spin_unlock(&spin); + return; + } + + /* h_sb is new to us, print it */ + if (i < au_warn_loopback_nelem) { + a[i] = magic; + goto pr; + } + + /* expand the array */ + new_nelem = au_warn_loopback_nelem + au_warn_loopback_step; + a = au_kzrealloc(au_warn_loopback_array, + au_warn_loopback_nelem * sizeof(unsigned long), + new_nelem * sizeof(unsigned long), GFP_ATOMIC); + if (a) { + au_warn_loopback_nelem = new_nelem; + au_warn_loopback_array = a; + a[i] = magic; + goto pr; + } + + spin_unlock(&spin); + AuWarn1("realloc failed, ignored\n"); + return; + +pr: + spin_unlock(&spin); + pr_warn("you may want to try another patch for loopback file " + "on %s(0x%lx) branch\n", au_sbtype(h_sb), magic); +} + +int au_loopback_init(void) +{ + int err; + struct super_block *sb __maybe_unused; + + AuDebugOn(sizeof(sb->s_magic) != sizeof(unsigned long)); + + err = 0; + au_warn_loopback_array = kcalloc(au_warn_loopback_step, + sizeof(unsigned long), GFP_NOFS); + if (unlikely(!au_warn_loopback_array)) + err = -ENOMEM; + + return err; +} + +void au_loopback_fin(void) +{ + kfree(au_warn_loopback_array); +} diff -Naur kernel.21.3.orig/fs/aufs/loop.h kernel.21.3.new/fs/aufs/loop.h --- kernel.21.3.orig/fs/aufs/loop.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/loop.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,49 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * support for loopback mount as a branch + */ + +#ifndef __AUFS_LOOP_H__ +#define __AUFS_LOOP_H__ + +#ifdef __KERNEL__ + +struct dentry; +struct super_block; + +#ifdef CONFIG_AUFS_BDEV_LOOP +/* loop.c */ +int au_test_loopback_overlap(struct super_block *sb, struct dentry *h_adding); +int au_test_loopback_kthread(void); +void au_warn_loopback(struct super_block *h_sb); + +int au_loopback_init(void); +void au_loopback_fin(void); +#else +AuStubInt0(au_test_loopback_overlap, struct super_block *sb, + struct dentry *h_adding) +AuStubInt0(au_test_loopback_kthread, void) +AuStubVoid(au_warn_loopback, struct super_block *h_sb) + +AuStubInt0(au_loopback_init, void) +AuStubVoid(au_loopback_fin, void) +#endif /* BLK_DEV_LOOP */ + +#endif /* __KERNEL__ */ +#endif /* __AUFS_LOOP_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/magic.mk kernel.21.3.new/fs/aufs/magic.mk --- kernel.21.3.orig/fs/aufs/magic.mk 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/magic.mk 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,54 @@ + +# defined in ${srctree}/fs/fuse/inode.c +# tristate +ifdef CONFIG_FUSE_FS +ccflags-y += -DFUSE_SUPER_MAGIC=0x65735546 +endif + +# defined in ${srctree}/fs/ocfs2/ocfs2_fs.h +# tristate +ifdef CONFIG_OCFS2_FS +ccflags-y += -DOCFS2_SUPER_MAGIC=0x7461636f +endif + +# defined in ${srctree}/fs/ocfs2/dlm/userdlm.h +# tristate +ifdef CONFIG_OCFS2_FS_O2CB +ccflags-y += -DDLMFS_MAGIC=0x76a9f425 +endif + +# defined in ${srctree}/fs/cifs/cifsfs.c +# tristate +ifdef CONFIG_CIFS_FS +ccflags-y += -DCIFS_MAGIC_NUMBER=0xFF534D42 +endif + +# defined in ${srctree}/fs/xfs/xfs_sb.h +# tristate +ifdef CONFIG_XFS_FS +ccflags-y += -DXFS_SB_MAGIC=0x58465342 +endif + +# defined in ${srctree}/fs/configfs/mount.c +# tristate +ifdef CONFIG_CONFIGFS_FS +ccflags-y += -DCONFIGFS_MAGIC=0x62656570 +endif + +# defined in ${srctree}/fs/9p/v9fs.h +# tristate +ifdef CONFIG_9P_FS +ccflags-y += -DV9FS_MAGIC=0x01021997 +endif + +# defined in ${srctree}/fs/ubifs/ubifs.h +# tristate +ifdef CONFIG_UBIFS_FS +ccflags-y += -DUBIFS_SUPER_MAGIC=0x24051905 +endif + +# defined in ${srctree}/fs/hfsplus/hfsplus_raw.h +# tristate +ifdef CONFIG_HFSPLUS_FS +ccflags-y += -DHFSPLUS_SUPER_MAGIC=0x482b +endif diff -Naur kernel.21.3.orig/fs/aufs/Makefile kernel.21.3.new/fs/aufs/Makefile --- kernel.21.3.orig/fs/aufs/Makefile 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/Makefile 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,43 @@ + +include ${src}/magic.mk +ifeq (${CONFIG_AUFS_FS},m) +include ${src}/conf.mk +endif +-include ${src}/priv_def.mk + +# cf. include/linux/kernel.h +# enable pr_debug +ccflags-y += -DDEBUG +# sparse requires the full pathname +ifdef M +ccflags-y += -include ${M}/../../include/uapi/linux/aufs_type.h +else +ccflags-y += -include ${srctree}/include/uapi/linux/aufs_type.h +endif + +obj-$(CONFIG_AUFS_FS) += aufs.o +aufs-y := module.o sbinfo.o super.o branch.o xino.o sysaufs.o opts.o \ + wkq.o vfsub.o dcsub.o \ + cpup.o whout.o wbr_policy.o \ + dinfo.o dentry.o \ + dynop.o \ + finfo.o file.o f_op.o \ + dir.o vdir.o \ + iinfo.o inode.o i_op.o i_op_add.o i_op_del.o i_op_ren.o \ + mvdown.o ioctl.o + +# all are boolean +aufs-$(CONFIG_PROC_FS) += procfs.o plink.o +aufs-$(CONFIG_SYSFS) += sysfs.o +aufs-$(CONFIG_DEBUG_FS) += dbgaufs.o +aufs-$(CONFIG_AUFS_BDEV_LOOP) += loop.o +aufs-$(CONFIG_AUFS_HNOTIFY) += hnotify.o +aufs-$(CONFIG_AUFS_HFSNOTIFY) += hfsnotify.o +aufs-$(CONFIG_AUFS_EXPORT) += export.o +aufs-$(CONFIG_AUFS_XATTR) += xattr.o +aufs-$(CONFIG_AUFS_FHSM) += fhsm.o +aufs-$(CONFIG_AUFS_POLL) += poll.o +aufs-$(CONFIG_AUFS_RDU) += rdu.o +aufs-$(CONFIG_AUFS_BR_HFSPLUS) += hfsplus.o +aufs-$(CONFIG_AUFS_DEBUG) += debug.o +aufs-$(CONFIG_AUFS_MAGIC_SYSRQ) += sysrq.o diff -Naur kernel.21.3.orig/fs/aufs/module.c kernel.21.3.new/fs/aufs/module.c --- kernel.21.3.orig/fs/aufs/module.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/module.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,210 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * module global variables and operations + */ + +#include +#include +#include "aufs.h" + +void *au_kzrealloc(void *p, unsigned int nused, unsigned int new_sz, gfp_t gfp) +{ + if (new_sz <= nused) + return p; + + p = krealloc(p, new_sz, gfp); + if (p) + memset(p + nused, 0, new_sz - nused); + return p; +} + +/* ---------------------------------------------------------------------- */ + +/* + * aufs caches + */ +struct kmem_cache *au_cachep[AuCache_Last]; +static int __init au_cache_init(void) +{ + au_cachep[AuCache_DINFO] = AuCacheCtor(au_dinfo, au_di_init_once); + if (au_cachep[AuCache_DINFO]) + /* SLAB_DESTROY_BY_RCU */ + au_cachep[AuCache_ICNTNR] = AuCacheCtor(au_icntnr, + au_icntnr_init_once); + if (au_cachep[AuCache_ICNTNR]) + au_cachep[AuCache_FINFO] = AuCacheCtor(au_finfo, + au_fi_init_once); + if (au_cachep[AuCache_FINFO]) + au_cachep[AuCache_VDIR] = AuCache(au_vdir); + if (au_cachep[AuCache_VDIR]) + au_cachep[AuCache_DEHSTR] = AuCache(au_vdir_dehstr); + if (au_cachep[AuCache_DEHSTR]) + return 0; + + return -ENOMEM; +} + +static void au_cache_fin(void) +{ + int i; + + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); + + /* excluding AuCache_HNOTIFY */ + BUILD_BUG_ON(AuCache_HNOTIFY + 1 != AuCache_Last); + for (i = 0; i < AuCache_HNOTIFY; i++) + if (au_cachep[i]) { + kmem_cache_destroy(au_cachep[i]); + au_cachep[i] = NULL; + } +} + +/* ---------------------------------------------------------------------- */ + +int au_dir_roflags; + +#ifdef CONFIG_AUFS_SBILIST +/* + * iterate_supers_type() doesn't protect us from + * remounting (branch management) + */ +struct au_splhead au_sbilist; +#endif + +struct lock_class_key au_lc_key[AuLcKey_Last]; + +/* + * functions for module interface. + */ +MODULE_LICENSE("GPL"); +/* MODULE_LICENSE("GPL v2"); */ +MODULE_AUTHOR("Junjiro R. Okajima "); +MODULE_DESCRIPTION(AUFS_NAME + " -- Advanced multi layered unification filesystem"); +MODULE_VERSION(AUFS_VERSION); +MODULE_ALIAS_FS(AUFS_NAME); + +/* this module parameter has no meaning when SYSFS is disabled */ +int sysaufs_brs = 1; +MODULE_PARM_DESC(brs, "use /fs/aufs/si_*/brN"); +module_param_named(brs, sysaufs_brs, int, S_IRUGO); + +/* this module parameter has no meaning when USER_NS is disabled */ +static bool au_userns; +MODULE_PARM_DESC(allow_userns, "allow unprivileged to mount under userns"); +module_param_named(allow_userns, au_userns, bool, S_IRUGO); + +/* ---------------------------------------------------------------------- */ + +static char au_esc_chars[0x20 + 3]; /* 0x01-0x20, backslash, del, and NULL */ + +int au_seq_path(struct seq_file *seq, struct path *path) +{ + return seq_path(seq, path, au_esc_chars); +} + +/* ---------------------------------------------------------------------- */ + +static int __init aufs_init(void) +{ + int err, i; + char *p; + + p = au_esc_chars; + for (i = 1; i <= ' '; i++) + *p++ = i; + *p++ = '\\'; + *p++ = '\x7f'; + *p = 0; + + au_dir_roflags = au_file_roflags(O_DIRECTORY | O_LARGEFILE); + + au_sbilist_init(); + sysaufs_brs_init(); + au_debug_init(); + au_dy_init(); + err = sysaufs_init(); + if (unlikely(err)) + goto out; + err = au_procfs_init(); + if (unlikely(err)) + goto out_sysaufs; + err = au_wkq_init(); + if (unlikely(err)) + goto out_procfs; + err = au_loopback_init(); + if (unlikely(err)) + goto out_wkq; + err = au_hnotify_init(); + if (unlikely(err)) + goto out_loopback; + err = au_sysrq_init(); + if (unlikely(err)) + goto out_hin; + err = au_cache_init(); + if (unlikely(err)) + goto out_sysrq; + + aufs_fs_type.fs_flags |= au_userns ? FS_USERNS_MOUNT : 0; + err = register_filesystem(&aufs_fs_type); + if (unlikely(err)) + goto out_cache; + + /* since we define pr_fmt, call printk directly */ + printk(KERN_INFO AUFS_NAME " " AUFS_VERSION "\n"); + goto out; /* success */ + +out_cache: + au_cache_fin(); +out_sysrq: + au_sysrq_fin(); +out_hin: + au_hnotify_fin(); +out_loopback: + au_loopback_fin(); +out_wkq: + au_wkq_fin(); +out_procfs: + au_procfs_fin(); +out_sysaufs: + sysaufs_fin(); + au_dy_fin(); +out: + return err; +} + +static void __exit aufs_exit(void) +{ + unregister_filesystem(&aufs_fs_type); + au_cache_fin(); + au_sysrq_fin(); + au_hnotify_fin(); + au_loopback_fin(); + au_wkq_fin(); + au_procfs_fin(); + sysaufs_fin(); + au_dy_fin(); +} + +module_init(aufs_init); +module_exit(aufs_exit); diff -Naur kernel.21.3.orig/fs/aufs/module.h kernel.21.3.new/fs/aufs/module.h --- kernel.21.3.orig/fs/aufs/module.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/module.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,104 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * module initialization and module-global + */ + +#ifndef __AUFS_MODULE_H__ +#define __AUFS_MODULE_H__ + +#ifdef __KERNEL__ + +#include + +struct path; +struct seq_file; + +/* module parameters */ +extern int sysaufs_brs; + +/* ---------------------------------------------------------------------- */ + +extern int au_dir_roflags; + +enum { + AuLcNonDir_FIINFO, + AuLcNonDir_DIINFO, + AuLcNonDir_IIINFO, + + AuLcDir_FIINFO, + AuLcDir_DIINFO, + AuLcDir_IIINFO, + + AuLcSymlink_DIINFO, + AuLcSymlink_IIINFO, + + AuLcKey_Last +}; +extern struct lock_class_key au_lc_key[AuLcKey_Last]; + +void *au_kzrealloc(void *p, unsigned int nused, unsigned int new_sz, gfp_t gfp); +int au_seq_path(struct seq_file *seq, struct path *path); + +#ifdef CONFIG_PROC_FS +/* procfs.c */ +int __init au_procfs_init(void); +void au_procfs_fin(void); +#else +AuStubInt0(au_procfs_init, void); +AuStubVoid(au_procfs_fin, void); +#endif + +/* ---------------------------------------------------------------------- */ + +/* kmem cache */ +enum { + AuCache_DINFO, + AuCache_ICNTNR, + AuCache_FINFO, + AuCache_VDIR, + AuCache_DEHSTR, + AuCache_HNOTIFY, /* must be last */ + AuCache_Last +}; + +#define AuCacheFlags (SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD) +#define AuCache(type) KMEM_CACHE(type, AuCacheFlags) +#define AuCacheCtor(type, ctor) \ + kmem_cache_create(#type, sizeof(struct type), \ + __alignof__(struct type), AuCacheFlags, ctor) + +extern struct kmem_cache *au_cachep[]; + +#define AuCacheFuncs(name, index) \ +static inline struct au_##name *au_cache_alloc_##name(void) \ +{ return kmem_cache_alloc(au_cachep[AuCache_##index], GFP_NOFS); } \ +static inline void au_cache_free_##name(struct au_##name *p) \ +{ kmem_cache_free(au_cachep[AuCache_##index], p); } + +AuCacheFuncs(dinfo, DINFO); +AuCacheFuncs(icntnr, ICNTNR); +AuCacheFuncs(finfo, FINFO); +AuCacheFuncs(vdir, VDIR); +AuCacheFuncs(vdir_dehstr, DEHSTR); +#ifdef CONFIG_AUFS_HNOTIFY +AuCacheFuncs(hnotify, HNOTIFY); +#endif + +#endif /* __KERNEL__ */ +#endif /* __AUFS_MODULE_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/mvdown.c kernel.21.3.new/fs/aufs/mvdown.c --- kernel.21.3.orig/fs/aufs/mvdown.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/mvdown.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,680 @@ +/* + * Copyright (C) 2011-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * move-down, opposite of copy-up + */ + +#include "aufs.h" + +struct au_mvd_args { + struct { + struct super_block *h_sb; + struct dentry *h_parent; + struct au_hinode *hdir; + struct inode *h_dir, *h_inode; + struct au_pin pin; + } info[AUFS_MVDOWN_NARRAY]; + + struct aufs_mvdown mvdown; + struct dentry *dentry, *parent; + struct inode *inode, *dir; + struct super_block *sb; + aufs_bindex_t bopq, bwh, bfound; + unsigned char rename_lock; +}; + +#define mvd_errno mvdown.au_errno +#define mvd_bsrc mvdown.stbr[AUFS_MVDOWN_UPPER].bindex +#define mvd_src_brid mvdown.stbr[AUFS_MVDOWN_UPPER].brid +#define mvd_bdst mvdown.stbr[AUFS_MVDOWN_LOWER].bindex +#define mvd_dst_brid mvdown.stbr[AUFS_MVDOWN_LOWER].brid + +#define mvd_h_src_sb info[AUFS_MVDOWN_UPPER].h_sb +#define mvd_h_src_parent info[AUFS_MVDOWN_UPPER].h_parent +#define mvd_hdir_src info[AUFS_MVDOWN_UPPER].hdir +#define mvd_h_src_dir info[AUFS_MVDOWN_UPPER].h_dir +#define mvd_h_src_inode info[AUFS_MVDOWN_UPPER].h_inode +#define mvd_pin_src info[AUFS_MVDOWN_UPPER].pin + +#define mvd_h_dst_sb info[AUFS_MVDOWN_LOWER].h_sb +#define mvd_h_dst_parent info[AUFS_MVDOWN_LOWER].h_parent +#define mvd_hdir_dst info[AUFS_MVDOWN_LOWER].hdir +#define mvd_h_dst_dir info[AUFS_MVDOWN_LOWER].h_dir +#define mvd_h_dst_inode info[AUFS_MVDOWN_LOWER].h_inode +#define mvd_pin_dst info[AUFS_MVDOWN_LOWER].pin + +#define AU_MVD_PR(flag, ...) do { \ + if (flag) \ + pr_err(__VA_ARGS__); \ + } while (0) + +static int find_lower_writable(struct au_mvd_args *a) +{ + struct super_block *sb; + aufs_bindex_t bindex, bend; + struct au_branch *br; + + sb = a->sb; + bindex = a->mvd_bsrc; + bend = au_sbend(sb); + if (a->mvdown.flags & AUFS_MVDOWN_FHSM_LOWER) + for (bindex++; bindex <= bend; bindex++) { + br = au_sbr(sb, bindex); + if (au_br_fhsm(br->br_perm) + && (!(au_br_sb(br)->s_flags & MS_RDONLY))) + return bindex; + } + else if (!(a->mvdown.flags & AUFS_MVDOWN_ROLOWER)) + for (bindex++; bindex <= bend; bindex++) { + br = au_sbr(sb, bindex); + if (!au_br_rdonly(br)) + return bindex; + } + else + for (bindex++; bindex <= bend; bindex++) { + br = au_sbr(sb, bindex); + if (!(au_br_sb(br)->s_flags & MS_RDONLY)) { + if (au_br_rdonly(br)) + a->mvdown.flags + |= AUFS_MVDOWN_ROLOWER_R; + return bindex; + } + } + + return -1; +} + +/* make the parent dir on bdst */ +static int au_do_mkdir(const unsigned char dmsg, struct au_mvd_args *a) +{ + int err; + + err = 0; + a->mvd_hdir_src = au_hi(a->dir, a->mvd_bsrc); + a->mvd_hdir_dst = au_hi(a->dir, a->mvd_bdst); + a->mvd_h_src_parent = au_h_dptr(a->parent, a->mvd_bsrc); + a->mvd_h_dst_parent = NULL; + if (au_dbend(a->parent) >= a->mvd_bdst) + a->mvd_h_dst_parent = au_h_dptr(a->parent, a->mvd_bdst); + if (!a->mvd_h_dst_parent) { + err = au_cpdown_dirs(a->dentry, a->mvd_bdst); + if (unlikely(err)) { + AU_MVD_PR(dmsg, "cpdown_dirs failed\n"); + goto out; + } + a->mvd_h_dst_parent = au_h_dptr(a->parent, a->mvd_bdst); + } + +out: + AuTraceErr(err); + return err; +} + +/* lock them all */ +static int au_do_lock(const unsigned char dmsg, struct au_mvd_args *a) +{ + int err; + struct dentry *h_trap; + + a->mvd_h_src_sb = au_sbr_sb(a->sb, a->mvd_bsrc); + a->mvd_h_dst_sb = au_sbr_sb(a->sb, a->mvd_bdst); + err = au_pin(&a->mvd_pin_dst, a->dentry, a->mvd_bdst, + au_opt_udba(a->sb), + AuPin_MNT_WRITE | AuPin_DI_LOCKED); + AuTraceErr(err); + if (unlikely(err)) { + AU_MVD_PR(dmsg, "pin_dst failed\n"); + goto out; + } + + if (a->mvd_h_src_sb != a->mvd_h_dst_sb) { + a->rename_lock = 0; + au_pin_init(&a->mvd_pin_src, a->dentry, a->mvd_bsrc, + AuLsc_DI_PARENT, AuLsc_I_PARENT3, + au_opt_udba(a->sb), + AuPin_MNT_WRITE | AuPin_DI_LOCKED); + err = au_do_pin(&a->mvd_pin_src); + AuTraceErr(err); + a->mvd_h_src_dir = a->mvd_h_src_parent->d_inode; + if (unlikely(err)) { + AU_MVD_PR(dmsg, "pin_src failed\n"); + goto out_dst; + } + goto out; /* success */ + } + + a->rename_lock = 1; + au_pin_hdir_unlock(&a->mvd_pin_dst); + err = au_pin(&a->mvd_pin_src, a->dentry, a->mvd_bsrc, + au_opt_udba(a->sb), + AuPin_MNT_WRITE | AuPin_DI_LOCKED); + AuTraceErr(err); + a->mvd_h_src_dir = a->mvd_h_src_parent->d_inode; + if (unlikely(err)) { + AU_MVD_PR(dmsg, "pin_src failed\n"); + au_pin_hdir_lock(&a->mvd_pin_dst); + goto out_dst; + } + au_pin_hdir_unlock(&a->mvd_pin_src); + h_trap = vfsub_lock_rename(a->mvd_h_src_parent, a->mvd_hdir_src, + a->mvd_h_dst_parent, a->mvd_hdir_dst); + if (h_trap) { + err = (h_trap != a->mvd_h_src_parent); + if (err) + err = (h_trap != a->mvd_h_dst_parent); + } + BUG_ON(err); /* it should never happen */ + if (unlikely(a->mvd_h_src_dir != au_pinned_h_dir(&a->mvd_pin_src))) { + err = -EBUSY; + AuTraceErr(err); + vfsub_unlock_rename(a->mvd_h_src_parent, a->mvd_hdir_src, + a->mvd_h_dst_parent, a->mvd_hdir_dst); + au_pin_hdir_lock(&a->mvd_pin_src); + au_unpin(&a->mvd_pin_src); + au_pin_hdir_lock(&a->mvd_pin_dst); + goto out_dst; + } + goto out; /* success */ + +out_dst: + au_unpin(&a->mvd_pin_dst); +out: + AuTraceErr(err); + return err; +} + +static void au_do_unlock(const unsigned char dmsg, struct au_mvd_args *a) +{ + if (!a->rename_lock) + au_unpin(&a->mvd_pin_src); + else { + vfsub_unlock_rename(a->mvd_h_src_parent, a->mvd_hdir_src, + a->mvd_h_dst_parent, a->mvd_hdir_dst); + au_pin_hdir_lock(&a->mvd_pin_src); + au_unpin(&a->mvd_pin_src); + au_pin_hdir_lock(&a->mvd_pin_dst); + } + au_unpin(&a->mvd_pin_dst); +} + +/* copy-down the file */ +static int au_do_cpdown(const unsigned char dmsg, struct au_mvd_args *a) +{ + int err; + struct au_cp_generic cpg = { + .dentry = a->dentry, + .bdst = a->mvd_bdst, + .bsrc = a->mvd_bsrc, + .len = -1, + .pin = &a->mvd_pin_dst, + .flags = AuCpup_DTIME | AuCpup_HOPEN + }; + + AuDbg("b%d, b%d\n", cpg.bsrc, cpg.bdst); + if (a->mvdown.flags & AUFS_MVDOWN_OWLOWER) + au_fset_cpup(cpg.flags, OVERWRITE); + if (a->mvdown.flags & AUFS_MVDOWN_ROLOWER) + au_fset_cpup(cpg.flags, RWDST); + err = au_sio_cpdown_simple(&cpg); + if (unlikely(err)) + AU_MVD_PR(dmsg, "cpdown failed\n"); + + AuTraceErr(err); + return err; +} + +/* + * unlink the whiteout on bdst if exist which may be created by UDBA while we + * were sleeping + */ +static int au_do_unlink_wh(const unsigned char dmsg, struct au_mvd_args *a) +{ + int err; + struct path h_path; + struct au_branch *br; + + br = au_sbr(a->sb, a->mvd_bdst); + h_path.dentry = au_wh_lkup(a->mvd_h_dst_parent, &a->dentry->d_name, br); + err = PTR_ERR(h_path.dentry); + if (IS_ERR(h_path.dentry)) { + AU_MVD_PR(dmsg, "wh_lkup failed\n"); + goto out; + } + + err = 0; + if (h_path.dentry->d_inode) { + h_path.mnt = au_br_mnt(br); + err = vfsub_unlink(a->mvd_h_dst_parent->d_inode, &h_path, + /*force*/0); + if (unlikely(err)) + AU_MVD_PR(dmsg, "wh_unlink failed\n"); + } + dput(h_path.dentry); + +out: + AuTraceErr(err); + return err; +} + +/* + * unlink the topmost h_dentry + */ +static int au_do_unlink(const unsigned char dmsg, struct au_mvd_args *a) +{ + int err; + struct path h_path; + + h_path.mnt = au_sbr_mnt(a->sb, a->mvd_bsrc); + h_path.dentry = au_h_dptr(a->dentry, a->mvd_bsrc); + err = vfsub_unlink(a->mvd_h_src_dir, &h_path, /*force*/0); + if (unlikely(err)) + AU_MVD_PR(dmsg, "unlink failed\n"); + + AuTraceErr(err); + return err; +} + +/* Since mvdown succeeded, we ignore an error of this function */ +static void au_do_stfs(const unsigned char dmsg, struct au_mvd_args *a) +{ + int err; + struct au_branch *br; + + a->mvdown.flags |= AUFS_MVDOWN_STFS_FAILED; + br = au_sbr(a->sb, a->mvd_bsrc); + err = au_br_stfs(br, &a->mvdown.stbr[AUFS_MVDOWN_UPPER].stfs); + if (!err) { + br = au_sbr(a->sb, a->mvd_bdst); + a->mvdown.stbr[AUFS_MVDOWN_LOWER].brid = br->br_id; + err = au_br_stfs(br, &a->mvdown.stbr[AUFS_MVDOWN_LOWER].stfs); + } + if (!err) + a->mvdown.flags &= ~AUFS_MVDOWN_STFS_FAILED; + else + AU_MVD_PR(dmsg, "statfs failed (%d), ignored\n", err); +} + +/* + * copy-down the file and unlink the bsrc file. + * - unlink the bdst whout if exist + * - copy-down the file (with whtmp name and rename) + * - unlink the bsrc file + */ +static int au_do_mvdown(const unsigned char dmsg, struct au_mvd_args *a) +{ + int err; + + err = au_do_mkdir(dmsg, a); + if (!err) + err = au_do_lock(dmsg, a); + if (unlikely(err)) + goto out; + + /* + * do not revert the activities we made on bdst since they should be + * harmless in aufs. + */ + + err = au_do_cpdown(dmsg, a); + if (!err) + err = au_do_unlink_wh(dmsg, a); + if (!err && !(a->mvdown.flags & AUFS_MVDOWN_KUPPER)) + err = au_do_unlink(dmsg, a); + if (unlikely(err)) + goto out_unlock; + + AuDbg("%pd2, 0x%x, %d --> %d\n", + a->dentry, a->mvdown.flags, a->mvd_bsrc, a->mvd_bdst); + if (find_lower_writable(a) < 0) + a->mvdown.flags |= AUFS_MVDOWN_BOTTOM; + + if (a->mvdown.flags & AUFS_MVDOWN_STFS) + au_do_stfs(dmsg, a); + + /* maintain internal array */ + if (!(a->mvdown.flags & AUFS_MVDOWN_KUPPER)) { + au_set_h_dptr(a->dentry, a->mvd_bsrc, NULL); + au_set_dbstart(a->dentry, a->mvd_bdst); + au_set_h_iptr(a->inode, a->mvd_bsrc, NULL, /*flags*/0); + au_set_ibstart(a->inode, a->mvd_bdst); + } + if (au_dbend(a->dentry) < a->mvd_bdst) + au_set_dbend(a->dentry, a->mvd_bdst); + if (au_ibend(a->inode) < a->mvd_bdst) + au_set_ibend(a->inode, a->mvd_bdst); + +out_unlock: + au_do_unlock(dmsg, a); +out: + AuTraceErr(err); + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* make sure the file is idle */ +static int au_mvd_args_busy(const unsigned char dmsg, struct au_mvd_args *a) +{ + int err, plinked; + + err = 0; + plinked = !!au_opt_test(au_mntflags(a->sb), PLINK); + if (au_dbstart(a->dentry) == a->mvd_bsrc + && a->dentry->d_count == 1 + && atomic_read(&a->inode->i_count) == 1 + /* && a->mvd_h_src_inode->i_nlink == 1 */ + && (!plinked || !au_plink_test(a->inode)) + && a->inode->i_nlink == 1) + goto out; + + err = -EBUSY; + AU_MVD_PR(dmsg, + "b%d, d{b%d, c%u?}, i{c%d?, l%u}, hi{l%u}, p{%d, %d}\n", + a->mvd_bsrc, au_dbstart(a->dentry), a->dentry->d_count, + atomic_read(&a->inode->i_count), a->inode->i_nlink, + a->mvd_h_src_inode->i_nlink, + plinked, plinked ? au_plink_test(a->inode) : 0); + +out: + AuTraceErr(err); + return err; +} + +/* make sure the parent dir is fine */ +static int au_mvd_args_parent(const unsigned char dmsg, + struct au_mvd_args *a) +{ + int err; + aufs_bindex_t bindex; + + err = 0; + if (unlikely(au_alive_dir(a->parent))) { + err = -ENOENT; + AU_MVD_PR(dmsg, "parent dir is dead\n"); + goto out; + } + + a->bopq = au_dbdiropq(a->parent); + bindex = au_wbr_nonopq(a->dentry, a->mvd_bdst); + AuDbg("b%d\n", bindex); + if (unlikely((bindex >= 0 && bindex < a->mvd_bdst) + || (a->bopq != -1 && a->bopq < a->mvd_bdst))) { + err = -EINVAL; + a->mvd_errno = EAU_MVDOWN_OPAQUE; + AU_MVD_PR(dmsg, "ancestor is opaque b%d, b%d\n", + a->bopq, a->mvd_bdst); + } + +out: + AuTraceErr(err); + return err; +} + +static int au_mvd_args_intermediate(const unsigned char dmsg, + struct au_mvd_args *a) +{ + int err; + struct au_dinfo *dinfo, *tmp; + + /* lookup the next lower positive entry */ + err = -ENOMEM; + tmp = au_di_alloc(a->sb, AuLsc_DI_TMP); + if (unlikely(!tmp)) + goto out; + + a->bfound = -1; + a->bwh = -1; + dinfo = au_di(a->dentry); + au_di_cp(tmp, dinfo); + au_di_swap(tmp, dinfo); + + /* returns the number of positive dentries */ + err = au_lkup_dentry(a->dentry, a->mvd_bsrc + 1, /*type*/0); + if (!err) + a->bwh = au_dbwh(a->dentry); + else if (err > 0) + a->bfound = au_dbstart(a->dentry); + + au_di_swap(tmp, dinfo); + au_rw_write_unlock(&tmp->di_rwsem); + au_di_free(tmp); + if (unlikely(err < 0)) + AU_MVD_PR(dmsg, "failed look-up lower\n"); + + /* + * here, we have these cases. + * bfound == -1 + * no positive dentry under bsrc. there are more sub-cases. + * bwh < 0 + * there no whiteout, we can safely move-down. + * bwh <= bsrc + * impossible + * bsrc < bwh && bwh < bdst + * there is a whiteout on RO branch. cannot proceed. + * bwh == bdst + * there is a whiteout on the RW target branch. it should + * be removed. + * bdst < bwh + * there is a whiteout somewhere unrelated branch. + * -1 < bfound && bfound <= bsrc + * impossible. + * bfound < bdst + * found, but it is on RO branch between bsrc and bdst. cannot + * proceed. + * bfound == bdst + * found, replace it if AUFS_MVDOWN_FORCE is set. otherwise return + * error. + * bdst < bfound + * found, after we create the file on bdst, it will be hidden. + */ + + AuDebugOn(a->bfound == -1 + && a->bwh != -1 + && a->bwh <= a->mvd_bsrc); + AuDebugOn(-1 < a->bfound + && a->bfound <= a->mvd_bsrc); + + err = -EINVAL; + if (a->bfound == -1 + && a->mvd_bsrc < a->bwh + && a->bwh != -1 + && a->bwh < a->mvd_bdst) { + a->mvd_errno = EAU_MVDOWN_WHITEOUT; + AU_MVD_PR(dmsg, "bsrc %d, bdst %d, bfound %d, bwh %d\n", + a->mvd_bsrc, a->mvd_bdst, a->bfound, a->bwh); + goto out; + } else if (a->bfound != -1 && a->bfound < a->mvd_bdst) { + a->mvd_errno = EAU_MVDOWN_UPPER; + AU_MVD_PR(dmsg, "bdst %d, bfound %d\n", + a->mvd_bdst, a->bfound); + goto out; + } + + err = 0; /* success */ + +out: + AuTraceErr(err); + return err; +} + +static int au_mvd_args_exist(const unsigned char dmsg, struct au_mvd_args *a) +{ + int err; + + err = 0; + if (!(a->mvdown.flags & AUFS_MVDOWN_OWLOWER) + && a->bfound == a->mvd_bdst) + err = -EEXIST; + AuTraceErr(err); + return err; +} + +static int au_mvd_args(const unsigned char dmsg, struct au_mvd_args *a) +{ + int err; + struct au_branch *br; + + err = -EISDIR; + if (unlikely(S_ISDIR(a->inode->i_mode))) + goto out; + + err = -EINVAL; + if (!(a->mvdown.flags & AUFS_MVDOWN_BRID_UPPER)) + a->mvd_bsrc = au_ibstart(a->inode); + else { + a->mvd_bsrc = au_br_index(a->sb, a->mvd_src_brid); + if (unlikely(a->mvd_bsrc < 0 + || (a->mvd_bsrc < au_dbstart(a->dentry) + || au_dbend(a->dentry) < a->mvd_bsrc + || !au_h_dptr(a->dentry, a->mvd_bsrc)) + || (a->mvd_bsrc < au_ibstart(a->inode) + || au_ibend(a->inode) < a->mvd_bsrc + || !au_h_iptr(a->inode, a->mvd_bsrc)))) { + a->mvd_errno = EAU_MVDOWN_NOUPPER; + AU_MVD_PR(dmsg, "no upper\n"); + goto out; + } + } + if (unlikely(a->mvd_bsrc == au_sbend(a->sb))) { + a->mvd_errno = EAU_MVDOWN_BOTTOM; + AU_MVD_PR(dmsg, "on the bottom\n"); + goto out; + } + a->mvd_h_src_inode = au_h_iptr(a->inode, a->mvd_bsrc); + br = au_sbr(a->sb, a->mvd_bsrc); + err = au_br_rdonly(br); + if (!(a->mvdown.flags & AUFS_MVDOWN_ROUPPER)) { + if (unlikely(err)) + goto out; + } else if (!(vfsub_native_ro(a->mvd_h_src_inode) + || IS_APPEND(a->mvd_h_src_inode))) { + if (err) + a->mvdown.flags |= AUFS_MVDOWN_ROUPPER_R; + /* go on */ + } else + goto out; + + err = -EINVAL; + if (!(a->mvdown.flags & AUFS_MVDOWN_BRID_LOWER)) { + a->mvd_bdst = find_lower_writable(a); + if (unlikely(a->mvd_bdst < 0)) { + a->mvd_errno = EAU_MVDOWN_BOTTOM; + AU_MVD_PR(dmsg, "no writable lower branch\n"); + goto out; + } + } else { + a->mvd_bdst = au_br_index(a->sb, a->mvd_dst_brid); + if (unlikely(a->mvd_bdst < 0 + || au_sbend(a->sb) < a->mvd_bdst)) { + a->mvd_errno = EAU_MVDOWN_NOLOWERBR; + AU_MVD_PR(dmsg, "no lower brid\n"); + goto out; + } + } + + err = au_mvd_args_busy(dmsg, a); + if (!err) + err = au_mvd_args_parent(dmsg, a); + if (!err) + err = au_mvd_args_intermediate(dmsg, a); + if (!err) + err = au_mvd_args_exist(dmsg, a); + if (!err) + AuDbg("b%d, b%d\n", a->mvd_bsrc, a->mvd_bdst); + +out: + AuTraceErr(err); + return err; +} + +int au_mvdown(struct dentry *dentry, struct aufs_mvdown __user *uarg) +{ + int err, e; + unsigned char dmsg; + struct au_mvd_args *args; + + err = -EPERM; + if (unlikely(!capable(CAP_SYS_ADMIN))) + goto out; + + err = -ENOMEM; + args = kmalloc(sizeof(*args), GFP_NOFS); + if (unlikely(!args)) + goto out; + + err = copy_from_user(&args->mvdown, uarg, sizeof(args->mvdown)); + if (!err) + err = !access_ok(VERIFY_WRITE, uarg, sizeof(*uarg)); + if (unlikely(err)) { + err = -EFAULT; + AuTraceErr(err); + goto out_free; + } + AuDbg("flags 0x%x\n", args->mvdown.flags); + args->mvdown.flags &= ~(AUFS_MVDOWN_ROLOWER_R | AUFS_MVDOWN_ROUPPER_R); + args->mvdown.au_errno = 0; + args->dentry = dentry; + args->inode = dentry->d_inode; + args->sb = dentry->d_sb; + + err = -ENOENT; + dmsg = !!(args->mvdown.flags & AUFS_MVDOWN_DMSG); + args->parent = dget_parent(dentry); + args->dir = args->parent->d_inode; + mutex_lock_nested(&args->dir->i_mutex, I_MUTEX_PARENT); + dput(args->parent); + if (unlikely(args->parent != dentry->d_parent)) { + AU_MVD_PR(dmsg, "parent dir is moved\n"); + goto out_dir; + } + + mutex_lock_nested(&args->inode->i_mutex, I_MUTEX_CHILD); + err = aufs_read_lock(dentry, AuLock_DW | AuLock_FLUSH); + if (unlikely(err)) + goto out_inode; + + di_write_lock_parent(args->parent); + err = au_mvd_args(dmsg, args); + if (unlikely(err)) + goto out_parent; + + err = au_do_mvdown(dmsg, args); + if (unlikely(err)) + goto out_parent; + + au_cpup_attr_timesizes(args->dir); + au_cpup_attr_timesizes(args->inode); + au_cpup_igen(args->inode, au_h_iptr(args->inode, args->mvd_bdst)); + /* au_digen_dec(dentry); */ + +out_parent: + di_write_unlock(args->parent); + aufs_read_unlock(dentry, AuLock_DW); +out_inode: + mutex_unlock(&args->inode->i_mutex); +out_dir: + mutex_unlock(&args->dir->i_mutex); +out_free: + e = copy_to_user(uarg, &args->mvdown, sizeof(args->mvdown)); + if (unlikely(e)) + err = -EFAULT; + kfree(args); +out: + AuTraceErr(err); + return err; +} diff -Naur kernel.21.3.orig/fs/aufs/opts.c kernel.21.3.new/fs/aufs/opts.c --- kernel.21.3.orig/fs/aufs/opts.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/opts.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,1840 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * mount options/flags + */ + +#include +#include /* a distribution requires */ +#include +#include "aufs.h" + +/* ---------------------------------------------------------------------- */ + +enum { + Opt_br, + Opt_add, Opt_del, Opt_mod, Opt_reorder, Opt_append, Opt_prepend, + Opt_idel, Opt_imod, Opt_ireorder, + Opt_dirwh, Opt_rdcache, Opt_rdblk, Opt_rdhash, Opt_rendir, + Opt_rdblk_def, Opt_rdhash_def, + Opt_xino, Opt_zxino, Opt_noxino, + Opt_trunc_xino, Opt_trunc_xino_v, Opt_notrunc_xino, + Opt_trunc_xino_path, Opt_itrunc_xino, + Opt_trunc_xib, Opt_notrunc_xib, + Opt_shwh, Opt_noshwh, + Opt_plink, Opt_noplink, Opt_list_plink, + Opt_udba, + Opt_dio, Opt_nodio, + /* Opt_lock, Opt_unlock, */ + Opt_cmd, Opt_cmd_args, + Opt_diropq_a, Opt_diropq_w, + Opt_warn_perm, Opt_nowarn_perm, + Opt_wbr_copyup, Opt_wbr_create, + Opt_fhsm_sec, + Opt_refrof, Opt_norefrof, + Opt_verbose, Opt_noverbose, + Opt_sum, Opt_nosum, Opt_wsum, + Opt_dirperm1, Opt_nodirperm1, + Opt_acl, Opt_noacl, + Opt_tail, Opt_ignore, Opt_ignore_silent, Opt_err +}; + +static match_table_t options = { + {Opt_br, "br=%s"}, + {Opt_br, "br:%s"}, + + {Opt_add, "add=%d:%s"}, + {Opt_add, "add:%d:%s"}, + {Opt_add, "ins=%d:%s"}, + {Opt_add, "ins:%d:%s"}, + {Opt_append, "append=%s"}, + {Opt_append, "append:%s"}, + {Opt_prepend, "prepend=%s"}, + {Opt_prepend, "prepend:%s"}, + + {Opt_del, "del=%s"}, + {Opt_del, "del:%s"}, + /* {Opt_idel, "idel:%d"}, */ + {Opt_mod, "mod=%s"}, + {Opt_mod, "mod:%s"}, + /* {Opt_imod, "imod:%d:%s"}, */ + + {Opt_dirwh, "dirwh=%d"}, + + {Opt_xino, "xino=%s"}, + {Opt_noxino, "noxino"}, + {Opt_trunc_xino, "trunc_xino"}, + {Opt_trunc_xino_v, "trunc_xino_v=%d:%d"}, + {Opt_notrunc_xino, "notrunc_xino"}, + {Opt_trunc_xino_path, "trunc_xino=%s"}, + {Opt_itrunc_xino, "itrunc_xino=%d"}, + /* {Opt_zxino, "zxino=%s"}, */ + {Opt_trunc_xib, "trunc_xib"}, + {Opt_notrunc_xib, "notrunc_xib"}, + +#ifdef CONFIG_PROC_FS + {Opt_plink, "plink"}, +#else + {Opt_ignore_silent, "plink"}, +#endif + + {Opt_noplink, "noplink"}, + +#ifdef CONFIG_AUFS_DEBUG + {Opt_list_plink, "list_plink"}, +#endif + + {Opt_udba, "udba=%s"}, + + {Opt_dio, "dio"}, + {Opt_nodio, "nodio"}, + +#ifdef CONFIG_AUFS_FHSM + {Opt_fhsm_sec, "fhsm_sec=%d"}, +#else + {Opt_ignore_silent, "fhsm_sec=%d"}, +#endif + + {Opt_diropq_a, "diropq=always"}, + {Opt_diropq_a, "diropq=a"}, + {Opt_diropq_w, "diropq=whiteouted"}, + {Opt_diropq_w, "diropq=w"}, + + {Opt_warn_perm, "warn_perm"}, + {Opt_nowarn_perm, "nowarn_perm"}, + + /* keep them temporary */ + {Opt_ignore_silent, "nodlgt"}, + {Opt_ignore_silent, "clean_plink"}, + +#ifdef CONFIG_AUFS_SHWH + {Opt_shwh, "shwh"}, +#endif + {Opt_noshwh, "noshwh"}, + + {Opt_dirperm1, "dirperm1"}, + {Opt_nodirperm1, "nodirperm1"}, + + {Opt_rendir, "rendir=%d"}, + + {Opt_refrof, "refrof"}, + {Opt_norefrof, "norefrof"}, + + {Opt_verbose, "verbose"}, + {Opt_verbose, "v"}, + {Opt_noverbose, "noverbose"}, + {Opt_noverbose, "quiet"}, + {Opt_noverbose, "q"}, + {Opt_noverbose, "silent"}, + + {Opt_sum, "sum"}, + {Opt_nosum, "nosum"}, + {Opt_wsum, "wsum"}, + + {Opt_rdcache, "rdcache=%d"}, + {Opt_rdblk, "rdblk=%d"}, + {Opt_rdblk_def, "rdblk=def"}, + {Opt_rdhash, "rdhash=%d"}, + {Opt_rdhash_def, "rdhash=def"}, + + {Opt_wbr_create, "create=%s"}, + {Opt_wbr_create, "create_policy=%s"}, + {Opt_wbr_copyup, "cpup=%s"}, + {Opt_wbr_copyup, "copyup=%s"}, + {Opt_wbr_copyup, "copyup_policy=%s"}, + + /* generic VFS flag */ +#ifdef CONFIG_FS_POSIX_ACL + {Opt_acl, "acl"}, + {Opt_noacl, "noacl"}, +#else + {Opt_ignore_silent, "acl"}, + {Opt_ignore_silent, "noacl"}, +#endif + + /* internal use for the scripts */ + {Opt_ignore_silent, "si=%s"}, + + {Opt_br, "dirs=%s"}, + {Opt_ignore, "debug=%d"}, + {Opt_ignore, "delete=whiteout"}, + {Opt_ignore, "delete=all"}, + {Opt_ignore, "imap=%s"}, + + /* temporary workaround, due to old mount(8)? */ + {Opt_ignore_silent, "relatime"}, + + {Opt_err, NULL} +}; + +/* ---------------------------------------------------------------------- */ + +static const char *au_parser_pattern(int val, match_table_t tbl) +{ + struct match_token *p; + + p = tbl; + while (p->pattern) { + if (p->token == val) + return p->pattern; + p++; + } + BUG(); + return "??"; +} + +static const char *au_optstr(int *val, match_table_t tbl) +{ + struct match_token *p; + int v; + + v = *val; + p = tbl; + while (p->token) { + if ((v & p->token) == p->token) { + *val &= ~p->token; + return p->pattern; + } + p++; + } + return NULL; +} + +/* ---------------------------------------------------------------------- */ + +static match_table_t brperm = { + {AuBrPerm_RO, AUFS_BRPERM_RO}, + {AuBrPerm_RR, AUFS_BRPERM_RR}, + {AuBrPerm_RW, AUFS_BRPERM_RW}, + {0, NULL} +}; + +static match_table_t brattr = { + /* general */ + {AuBrAttr_COO_REG, AUFS_BRATTR_COO_REG}, + {AuBrAttr_COO_ALL, AUFS_BRATTR_COO_ALL}, + {AuBrAttr_UNPIN, AUFS_BRATTR_UNPIN}, + {AuBrAttr_FHSM, AUFS_BRATTR_FHSM}, + {AuBrAttr_ICEX, AUFS_BRATTR_ICEX}, + {AuBrAttr_ICEX_SEC, AUFS_BRATTR_ICEX_SEC}, + {AuBrAttr_ICEX_SYS, AUFS_BRATTR_ICEX_SYS}, + {AuBrAttr_ICEX_TR, AUFS_BRATTR_ICEX_TR}, + {AuBrAttr_ICEX_USR, AUFS_BRATTR_ICEX_USR}, + {AuBrAttr_ICEX_OTH, AUFS_BRATTR_ICEX_OTH}, + + /* ro/rr branch */ + {AuBrRAttr_WH, AUFS_BRRATTR_WH}, + + /* rw branch */ + {AuBrWAttr_MOO, AUFS_BRWATTR_MOO}, + {AuBrWAttr_NoLinkWH, AUFS_BRWATTR_NLWH}, + + {0, NULL} +}; + +static int br_attr_val(char *str, match_table_t table, substring_t args[]) +{ + int attr, v; + char *p; + + attr = 0; + do { + p = strchr(str, '+'); + if (p) + *p = 0; + v = match_token(str, table, args); + if (v) { + if (v & AuBrAttr_CMOO_Mask) + attr &= ~AuBrAttr_CMOO_Mask; + attr |= v; + } else { + if (p) + *p = '+'; + pr_warn("ignored branch attribute %s\n", str); + break; + } + if (p) + str = p + 1; + } while (p); + + return attr; +} + +static int au_do_optstr_br_attr(au_br_perm_str_t *str, int perm) +{ + int sz; + const char *p; + char *q; + + q = str->a; + *q = 0; + p = au_optstr(&perm, brattr); + if (p) { + sz = strlen(p); + memcpy(q, p, sz + 1); + q += sz; + } else + goto out; + + do { + p = au_optstr(&perm, brattr); + if (p) { + *q++ = '+'; + sz = strlen(p); + memcpy(q, p, sz + 1); + q += sz; + } + } while (p); + +out: + return q - str->a; +} + +static int noinline_for_stack br_perm_val(char *perm) +{ + int val, bad, sz; + char *p; + substring_t args[MAX_OPT_ARGS]; + au_br_perm_str_t attr; + + p = strchr(perm, '+'); + if (p) + *p = 0; + val = match_token(perm, brperm, args); + if (!val) { + if (p) + *p = '+'; + pr_warn("ignored branch permission %s\n", perm); + val = AuBrPerm_RO; + goto out; + } + if (!p) + goto out; + + val |= br_attr_val(p + 1, brattr, args); + + bad = 0; + switch (val & AuBrPerm_Mask) { + case AuBrPerm_RO: + case AuBrPerm_RR: + bad = val & AuBrWAttr_Mask; + val &= ~AuBrWAttr_Mask; + break; + case AuBrPerm_RW: + bad = val & AuBrRAttr_Mask; + val &= ~AuBrRAttr_Mask; + break; + } + if (unlikely(bad)) { + sz = au_do_optstr_br_attr(&attr, bad); + AuDebugOn(!sz); + pr_warn("ignored branch attribute %s\n", attr.a); + } + +out: + return val; +} + +void au_optstr_br_perm(au_br_perm_str_t *str, int perm) +{ + au_br_perm_str_t attr; + const char *p; + char *q; + int sz; + + q = str->a; + p = au_optstr(&perm, brperm); + AuDebugOn(!p || !*p); + sz = strlen(p); + memcpy(q, p, sz + 1); + q += sz; + + sz = au_do_optstr_br_attr(&attr, perm); + if (sz) { + *q++ = '+'; + memcpy(q, attr.a, sz + 1); + } + + AuDebugOn(strlen(str->a) >= sizeof(str->a)); +} + +/* ---------------------------------------------------------------------- */ + +static match_table_t udbalevel = { + {AuOpt_UDBA_REVAL, "reval"}, + {AuOpt_UDBA_NONE, "none"}, +#ifdef CONFIG_AUFS_HNOTIFY + {AuOpt_UDBA_HNOTIFY, "notify"}, /* abstraction */ +#ifdef CONFIG_AUFS_HFSNOTIFY + {AuOpt_UDBA_HNOTIFY, "fsnotify"}, +#endif +#endif + {-1, NULL} +}; + +static int noinline_for_stack udba_val(char *str) +{ + substring_t args[MAX_OPT_ARGS]; + + return match_token(str, udbalevel, args); +} + +const char *au_optstr_udba(int udba) +{ + return au_parser_pattern(udba, udbalevel); +} + +/* ---------------------------------------------------------------------- */ + +static match_table_t au_wbr_create_policy = { + {AuWbrCreate_TDP, "tdp"}, + {AuWbrCreate_TDP, "top-down-parent"}, + {AuWbrCreate_RR, "rr"}, + {AuWbrCreate_RR, "round-robin"}, + {AuWbrCreate_MFS, "mfs"}, + {AuWbrCreate_MFS, "most-free-space"}, + {AuWbrCreate_MFSV, "mfs:%d"}, + {AuWbrCreate_MFSV, "most-free-space:%d"}, + + {AuWbrCreate_MFSRR, "mfsrr:%d"}, + {AuWbrCreate_MFSRRV, "mfsrr:%d:%d"}, + {AuWbrCreate_PMFS, "pmfs"}, + {AuWbrCreate_PMFSV, "pmfs:%d"}, + {AuWbrCreate_PMFSRR, "pmfsrr:%d"}, + {AuWbrCreate_PMFSRRV, "pmfsrr:%d:%d"}, + + {-1, NULL} +}; + +/* + * cf. linux/lib/parser.c and cmdline.c + * gave up calling memparse() since it uses simple_strtoull() instead of + * kstrto...(). + */ +static int noinline_for_stack +au_match_ull(substring_t *s, unsigned long long *result) +{ + int err; + unsigned int len; + char a[32]; + + err = -ERANGE; + len = s->to - s->from; + if (len + 1 <= sizeof(a)) { + memcpy(a, s->from, len); + a[len] = '\0'; + err = kstrtoull(a, 0, result); + } + return err; +} + +static int au_wbr_mfs_wmark(substring_t *arg, char *str, + struct au_opt_wbr_create *create) +{ + int err; + unsigned long long ull; + + err = 0; + if (!au_match_ull(arg, &ull)) + create->mfsrr_watermark = ull; + else { + pr_err("bad integer in %s\n", str); + err = -EINVAL; + } + + return err; +} + +static int au_wbr_mfs_sec(substring_t *arg, char *str, + struct au_opt_wbr_create *create) +{ + int n, err; + + err = 0; + if (!match_int(arg, &n) && 0 <= n && n <= AUFS_MFS_MAX_SEC) + create->mfs_second = n; + else { + pr_err("bad integer in %s\n", str); + err = -EINVAL; + } + + return err; +} + +static int noinline_for_stack +au_wbr_create_val(char *str, struct au_opt_wbr_create *create) +{ + int err, e; + substring_t args[MAX_OPT_ARGS]; + + err = match_token(str, au_wbr_create_policy, args); + create->wbr_create = err; + switch (err) { + case AuWbrCreate_MFSRRV: + case AuWbrCreate_PMFSRRV: + e = au_wbr_mfs_wmark(&args[0], str, create); + if (!e) + e = au_wbr_mfs_sec(&args[1], str, create); + if (unlikely(e)) + err = e; + break; + case AuWbrCreate_MFSRR: + case AuWbrCreate_PMFSRR: + e = au_wbr_mfs_wmark(&args[0], str, create); + if (unlikely(e)) { + err = e; + break; + } + /*FALLTHROUGH*/ + case AuWbrCreate_MFS: + case AuWbrCreate_PMFS: + create->mfs_second = AUFS_MFS_DEF_SEC; + break; + case AuWbrCreate_MFSV: + case AuWbrCreate_PMFSV: + e = au_wbr_mfs_sec(&args[0], str, create); + if (unlikely(e)) + err = e; + break; + } + + return err; +} + +const char *au_optstr_wbr_create(int wbr_create) +{ + return au_parser_pattern(wbr_create, au_wbr_create_policy); +} + +static match_table_t au_wbr_copyup_policy = { + {AuWbrCopyup_TDP, "tdp"}, + {AuWbrCopyup_TDP, "top-down-parent"}, + {AuWbrCopyup_BUP, "bup"}, + {AuWbrCopyup_BUP, "bottom-up-parent"}, + {AuWbrCopyup_BU, "bu"}, + {AuWbrCopyup_BU, "bottom-up"}, + {-1, NULL} +}; + +static int noinline_for_stack au_wbr_copyup_val(char *str) +{ + substring_t args[MAX_OPT_ARGS]; + + return match_token(str, au_wbr_copyup_policy, args); +} + +const char *au_optstr_wbr_copyup(int wbr_copyup) +{ + return au_parser_pattern(wbr_copyup, au_wbr_copyup_policy); +} + +/* ---------------------------------------------------------------------- */ + +static const int lkup_dirflags = LOOKUP_FOLLOW | LOOKUP_DIRECTORY; + +static void dump_opts(struct au_opts *opts) +{ +#ifdef CONFIG_AUFS_DEBUG + /* reduce stack space */ + union { + struct au_opt_add *add; + struct au_opt_del *del; + struct au_opt_mod *mod; + struct au_opt_xino *xino; + struct au_opt_xino_itrunc *xino_itrunc; + struct au_opt_wbr_create *create; + } u; + struct au_opt *opt; + + opt = opts->opt; + while (opt->type != Opt_tail) { + switch (opt->type) { + case Opt_add: + u.add = &opt->add; + AuDbg("add {b%d, %s, 0x%x, %p}\n", + u.add->bindex, u.add->pathname, u.add->perm, + u.add->path.dentry); + break; + case Opt_del: + case Opt_idel: + u.del = &opt->del; + AuDbg("del {%s, %p}\n", + u.del->pathname, u.del->h_path.dentry); + break; + case Opt_mod: + case Opt_imod: + u.mod = &opt->mod; + AuDbg("mod {%s, 0x%x, %p}\n", + u.mod->path, u.mod->perm, u.mod->h_root); + break; + case Opt_append: + u.add = &opt->add; + AuDbg("append {b%d, %s, 0x%x, %p}\n", + u.add->bindex, u.add->pathname, u.add->perm, + u.add->path.dentry); + break; + case Opt_prepend: + u.add = &opt->add; + AuDbg("prepend {b%d, %s, 0x%x, %p}\n", + u.add->bindex, u.add->pathname, u.add->perm, + u.add->path.dentry); + break; + case Opt_dirwh: + AuDbg("dirwh %d\n", opt->dirwh); + break; + case Opt_rdcache: + AuDbg("rdcache %d\n", opt->rdcache); + break; + case Opt_rdblk: + AuDbg("rdblk %u\n", opt->rdblk); + break; + case Opt_rdblk_def: + AuDbg("rdblk_def\n"); + break; + case Opt_rdhash: + AuDbg("rdhash %u\n", opt->rdhash); + break; + case Opt_rdhash_def: + AuDbg("rdhash_def\n"); + break; + case Opt_xino: + u.xino = &opt->xino; + AuDbg("xino {%s %.*s}\n", + u.xino->path, + AuDLNPair(u.xino->file->f_dentry)); + break; + case Opt_trunc_xino: + AuLabel(trunc_xino); + break; + case Opt_notrunc_xino: + AuLabel(notrunc_xino); + break; + case Opt_trunc_xino_path: + case Opt_itrunc_xino: + u.xino_itrunc = &opt->xino_itrunc; + AuDbg("trunc_xino %d\n", u.xino_itrunc->bindex); + break; + + case Opt_noxino: + AuLabel(noxino); + break; + case Opt_trunc_xib: + AuLabel(trunc_xib); + break; + case Opt_notrunc_xib: + AuLabel(notrunc_xib); + break; + case Opt_shwh: + AuLabel(shwh); + break; + case Opt_noshwh: + AuLabel(noshwh); + break; + case Opt_dirperm1: + AuLabel(dirperm1); + break; + case Opt_nodirperm1: + AuLabel(nodirperm1); + break; + case Opt_plink: + AuLabel(plink); + break; + case Opt_noplink: + AuLabel(noplink); + break; + case Opt_list_plink: + AuLabel(list_plink); + break; + case Opt_udba: + AuDbg("udba %d, %s\n", + opt->udba, au_optstr_udba(opt->udba)); + break; + case Opt_dio: + AuLabel(dio); + break; + case Opt_nodio: + AuLabel(nodio); + break; + case Opt_diropq_a: + AuLabel(diropq_a); + break; + case Opt_diropq_w: + AuLabel(diropq_w); + break; + case Opt_warn_perm: + AuLabel(warn_perm); + break; + case Opt_nowarn_perm: + AuLabel(nowarn_perm); + break; + case Opt_refrof: + AuLabel(refrof); + break; + case Opt_norefrof: + AuLabel(norefrof); + break; + case Opt_verbose: + AuLabel(verbose); + break; + case Opt_noverbose: + AuLabel(noverbose); + break; + case Opt_sum: + AuLabel(sum); + break; + case Opt_nosum: + AuLabel(nosum); + break; + case Opt_wsum: + AuLabel(wsum); + break; + case Opt_wbr_create: + u.create = &opt->wbr_create; + AuDbg("create %d, %s\n", u.create->wbr_create, + au_optstr_wbr_create(u.create->wbr_create)); + switch (u.create->wbr_create) { + case AuWbrCreate_MFSV: + case AuWbrCreate_PMFSV: + AuDbg("%d sec\n", u.create->mfs_second); + break; + case AuWbrCreate_MFSRR: + AuDbg("%llu watermark\n", + u.create->mfsrr_watermark); + break; + case AuWbrCreate_MFSRRV: + case AuWbrCreate_PMFSRRV: + AuDbg("%llu watermark, %d sec\n", + u.create->mfsrr_watermark, + u.create->mfs_second); + break; + } + break; + case Opt_wbr_copyup: + AuDbg("copyup %d, %s\n", opt->wbr_copyup, + au_optstr_wbr_copyup(opt->wbr_copyup)); + break; + case Opt_fhsm_sec: + AuDbg("fhsm_sec %u\n", opt->fhsm_second); + break; + case Opt_acl: + AuLabel(acl); + break; + case Opt_noacl: + AuLabel(noacl); + break; + default: + BUG(); + } + opt++; + } +#endif +} + +void au_opts_free(struct au_opts *opts) +{ + struct au_opt *opt; + + opt = opts->opt; + while (opt->type != Opt_tail) { + switch (opt->type) { + case Opt_add: + case Opt_append: + case Opt_prepend: + path_put(&opt->add.path); + break; + case Opt_del: + case Opt_idel: + path_put(&opt->del.h_path); + break; + case Opt_mod: + case Opt_imod: + dput(opt->mod.h_root); + break; + case Opt_xino: + fput(opt->xino.file); + break; + } + opt++; + } +} + +static int opt_add(struct au_opt *opt, char *opt_str, unsigned long sb_flags, + aufs_bindex_t bindex) +{ + int err; + struct au_opt_add *add = &opt->add; + char *p; + + add->bindex = bindex; + add->perm = AuBrPerm_RO; + add->pathname = opt_str; + p = strchr(opt_str, '='); + if (p) { + *p++ = 0; + if (*p) + add->perm = br_perm_val(p); + } + + err = vfsub_kern_path(add->pathname, lkup_dirflags, &add->path); + if (!err) { + if (!p) { + add->perm = AuBrPerm_RO; + if (au_test_fs_rr(add->path.dentry->d_sb)) + add->perm = AuBrPerm_RR; + else if (!bindex && !(sb_flags & MS_RDONLY)) + add->perm = AuBrPerm_RW; + } + opt->type = Opt_add; + goto out; + } + pr_err("lookup failed %s (%d)\n", add->pathname, err); + err = -EINVAL; + +out: + return err; +} + +static int au_opts_parse_del(struct au_opt_del *del, substring_t args[]) +{ + int err; + + del->pathname = args[0].from; + AuDbg("del path %s\n", del->pathname); + + err = vfsub_kern_path(del->pathname, lkup_dirflags, &del->h_path); + if (unlikely(err)) + pr_err("lookup failed %s (%d)\n", del->pathname, err); + + return err; +} + +#if 0 /* reserved for future use */ +static int au_opts_parse_idel(struct super_block *sb, aufs_bindex_t bindex, + struct au_opt_del *del, substring_t args[]) +{ + int err; + struct dentry *root; + + err = -EINVAL; + root = sb->s_root; + aufs_read_lock(root, AuLock_FLUSH); + if (bindex < 0 || au_sbend(sb) < bindex) { + pr_err("out of bounds, %d\n", bindex); + goto out; + } + + err = 0; + del->h_path.dentry = dget(au_h_dptr(root, bindex)); + del->h_path.mnt = mntget(au_sbr_mnt(sb, bindex)); + +out: + aufs_read_unlock(root, !AuLock_IR); + return err; +} +#endif + +static int noinline_for_stack +au_opts_parse_mod(struct au_opt_mod *mod, substring_t args[]) +{ + int err; + struct path path; + char *p; + + err = -EINVAL; + mod->path = args[0].from; + p = strchr(mod->path, '='); + if (unlikely(!p)) { + pr_err("no permssion %s\n", args[0].from); + goto out; + } + + *p++ = 0; + err = vfsub_kern_path(mod->path, lkup_dirflags, &path); + if (unlikely(err)) { + pr_err("lookup failed %s (%d)\n", mod->path, err); + goto out; + } + + mod->perm = br_perm_val(p); + AuDbg("mod path %s, perm 0x%x, %s\n", mod->path, mod->perm, p); + mod->h_root = dget(path.dentry); + path_put(&path); + +out: + return err; +} + +#if 0 /* reserved for future use */ +static int au_opts_parse_imod(struct super_block *sb, aufs_bindex_t bindex, + struct au_opt_mod *mod, substring_t args[]) +{ + int err; + struct dentry *root; + + err = -EINVAL; + root = sb->s_root; + aufs_read_lock(root, AuLock_FLUSH); + if (bindex < 0 || au_sbend(sb) < bindex) { + pr_err("out of bounds, %d\n", bindex); + goto out; + } + + err = 0; + mod->perm = br_perm_val(args[1].from); + AuDbg("mod path %s, perm 0x%x, %s\n", + mod->path, mod->perm, args[1].from); + mod->h_root = dget(au_h_dptr(root, bindex)); + +out: + aufs_read_unlock(root, !AuLock_IR); + return err; +} +#endif + +static int au_opts_parse_xino(struct super_block *sb, struct au_opt_xino *xino, + substring_t args[]) +{ + int err; + struct file *file; + + file = au_xino_create(sb, args[0].from, /*silent*/0); + err = PTR_ERR(file); + if (IS_ERR(file)) + goto out; + + err = -EINVAL; + if (unlikely(file->f_dentry->d_sb == sb)) { + fput(file); + pr_err("%s must be outside\n", args[0].from); + goto out; + } + + err = 0; + xino->file = file; + xino->path = args[0].from; + +out: + return err; +} + +static int noinline_for_stack +au_opts_parse_xino_itrunc_path(struct super_block *sb, + struct au_opt_xino_itrunc *xino_itrunc, + substring_t args[]) +{ + int err; + aufs_bindex_t bend, bindex; + struct path path; + struct dentry *root; + + err = vfsub_kern_path(args[0].from, lkup_dirflags, &path); + if (unlikely(err)) { + pr_err("lookup failed %s (%d)\n", args[0].from, err); + goto out; + } + + xino_itrunc->bindex = -1; + root = sb->s_root; + aufs_read_lock(root, AuLock_FLUSH); + bend = au_sbend(sb); + for (bindex = 0; bindex <= bend; bindex++) { + if (au_h_dptr(root, bindex) == path.dentry) { + xino_itrunc->bindex = bindex; + break; + } + } + aufs_read_unlock(root, !AuLock_IR); + path_put(&path); + + if (unlikely(xino_itrunc->bindex < 0)) { + pr_err("no such branch %s\n", args[0].from); + err = -EINVAL; + } + +out: + return err; +} + +/* called without aufs lock */ +int au_opts_parse(struct super_block *sb, char *str, struct au_opts *opts) +{ + int err, n, token; + aufs_bindex_t bindex; + unsigned char skipped; + struct dentry *root; + struct au_opt *opt, *opt_tail; + char *opt_str; + /* reduce the stack space */ + union { + struct au_opt_xino_itrunc *xino_itrunc; + struct au_opt_wbr_create *create; + } u; + struct { + substring_t args[MAX_OPT_ARGS]; + } *a; + + err = -ENOMEM; + a = kmalloc(sizeof(*a), GFP_NOFS); + if (unlikely(!a)) + goto out; + + root = sb->s_root; + err = 0; + bindex = 0; + opt = opts->opt; + opt_tail = opt + opts->max_opt - 1; + opt->type = Opt_tail; + while (!err && (opt_str = strsep(&str, ",")) && *opt_str) { + err = -EINVAL; + skipped = 0; + token = match_token(opt_str, options, a->args); + switch (token) { + case Opt_br: + err = 0; + while (!err && (opt_str = strsep(&a->args[0].from, ":")) + && *opt_str) { + err = opt_add(opt, opt_str, opts->sb_flags, + bindex++); + if (unlikely(!err && ++opt > opt_tail)) { + err = -E2BIG; + break; + } + opt->type = Opt_tail; + skipped = 1; + } + break; + case Opt_add: + if (unlikely(match_int(&a->args[0], &n))) { + pr_err("bad integer in %s\n", opt_str); + break; + } + bindex = n; + err = opt_add(opt, a->args[1].from, opts->sb_flags, + bindex); + if (!err) + opt->type = token; + break; + case Opt_append: + err = opt_add(opt, a->args[0].from, opts->sb_flags, + /*dummy bindex*/1); + if (!err) + opt->type = token; + break; + case Opt_prepend: + err = opt_add(opt, a->args[0].from, opts->sb_flags, + /*bindex*/0); + if (!err) + opt->type = token; + break; + case Opt_del: + err = au_opts_parse_del(&opt->del, a->args); + if (!err) + opt->type = token; + break; +#if 0 /* reserved for future use */ + case Opt_idel: + del->pathname = "(indexed)"; + if (unlikely(match_int(&args[0], &n))) { + pr_err("bad integer in %s\n", opt_str); + break; + } + err = au_opts_parse_idel(sb, n, &opt->del, a->args); + if (!err) + opt->type = token; + break; +#endif + case Opt_mod: + err = au_opts_parse_mod(&opt->mod, a->args); + if (!err) + opt->type = token; + break; +#ifdef IMOD /* reserved for future use */ + case Opt_imod: + u.mod->path = "(indexed)"; + if (unlikely(match_int(&a->args[0], &n))) { + pr_err("bad integer in %s\n", opt_str); + break; + } + err = au_opts_parse_imod(sb, n, &opt->mod, a->args); + if (!err) + opt->type = token; + break; +#endif + case Opt_xino: + err = au_opts_parse_xino(sb, &opt->xino, a->args); + if (!err) + opt->type = token; + break; + + case Opt_trunc_xino_path: + err = au_opts_parse_xino_itrunc_path + (sb, &opt->xino_itrunc, a->args); + if (!err) + opt->type = token; + break; + + case Opt_itrunc_xino: + u.xino_itrunc = &opt->xino_itrunc; + if (unlikely(match_int(&a->args[0], &n))) { + pr_err("bad integer in %s\n", opt_str); + break; + } + u.xino_itrunc->bindex = n; + aufs_read_lock(root, AuLock_FLUSH); + if (n < 0 || au_sbend(sb) < n) { + pr_err("out of bounds, %d\n", n); + aufs_read_unlock(root, !AuLock_IR); + break; + } + aufs_read_unlock(root, !AuLock_IR); + err = 0; + opt->type = token; + break; + + case Opt_dirwh: + if (unlikely(match_int(&a->args[0], &opt->dirwh))) + break; + err = 0; + opt->type = token; + break; + + case Opt_rdcache: + if (unlikely(match_int(&a->args[0], &n))) { + pr_err("bad integer in %s\n", opt_str); + break; + } + if (unlikely(n > AUFS_RDCACHE_MAX)) { + pr_err("rdcache must be smaller than %d\n", + AUFS_RDCACHE_MAX); + break; + } + opt->rdcache = n; + err = 0; + opt->type = token; + break; + case Opt_rdblk: + if (unlikely(match_int(&a->args[0], &n) + || n < 0 + || n > KMALLOC_MAX_SIZE)) { + pr_err("bad integer in %s\n", opt_str); + break; + } + if (unlikely(n && n < NAME_MAX)) { + pr_err("rdblk must be larger than %d\n", + NAME_MAX); + break; + } + opt->rdblk = n; + err = 0; + opt->type = token; + break; + case Opt_rdhash: + if (unlikely(match_int(&a->args[0], &n) + || n < 0 + || n * sizeof(struct hlist_head) + > KMALLOC_MAX_SIZE)) { + pr_err("bad integer in %s\n", opt_str); + break; + } + opt->rdhash = n; + err = 0; + opt->type = token; + break; + + case Opt_trunc_xino: + case Opt_notrunc_xino: + case Opt_noxino: + case Opt_trunc_xib: + case Opt_notrunc_xib: + case Opt_shwh: + case Opt_noshwh: + case Opt_dirperm1: + case Opt_nodirperm1: + case Opt_plink: + case Opt_noplink: + case Opt_list_plink: + case Opt_dio: + case Opt_nodio: + case Opt_diropq_a: + case Opt_diropq_w: + case Opt_warn_perm: + case Opt_nowarn_perm: + case Opt_refrof: + case Opt_norefrof: + case Opt_verbose: + case Opt_noverbose: + case Opt_sum: + case Opt_nosum: + case Opt_wsum: + case Opt_rdblk_def: + case Opt_rdhash_def: + case Opt_acl: + case Opt_noacl: + err = 0; + opt->type = token; + break; + + case Opt_udba: + opt->udba = udba_val(a->args[0].from); + if (opt->udba >= 0) { + err = 0; + opt->type = token; + } else + pr_err("wrong value, %s\n", opt_str); + break; + + case Opt_wbr_create: + u.create = &opt->wbr_create; + u.create->wbr_create + = au_wbr_create_val(a->args[0].from, u.create); + if (u.create->wbr_create >= 0) { + err = 0; + opt->type = token; + } else + pr_err("wrong value, %s\n", opt_str); + break; + case Opt_wbr_copyup: + opt->wbr_copyup = au_wbr_copyup_val(a->args[0].from); + if (opt->wbr_copyup >= 0) { + err = 0; + opt->type = token; + } else + pr_err("wrong value, %s\n", opt_str); + break; + + case Opt_fhsm_sec: + if (unlikely(match_int(&a->args[0], &n) + || n < 0)) { + pr_err("bad integer in %s\n", opt_str); + break; + } + if (sysaufs_brs) { + opt->fhsm_second = n; + opt->type = token; + } else + pr_warn("ignored %s\n", opt_str); + err = 0; + break; + + case Opt_ignore: + pr_warn("ignored %s\n", opt_str); + /*FALLTHROUGH*/ + case Opt_ignore_silent: + skipped = 1; + err = 0; + break; + case Opt_err: + pr_err("unknown option %s\n", opt_str); + break; + } + + if (!err && !skipped) { + if (unlikely(++opt > opt_tail)) { + err = -E2BIG; + opt--; + opt->type = Opt_tail; + break; + } + opt->type = Opt_tail; + } + } + + kfree(a); + dump_opts(opts); + if (unlikely(err)) + au_opts_free(opts); + +out: + return err; +} + +static int au_opt_wbr_create(struct super_block *sb, + struct au_opt_wbr_create *create) +{ + int err; + struct au_sbinfo *sbinfo; + + SiMustWriteLock(sb); + + err = 1; /* handled */ + sbinfo = au_sbi(sb); + if (sbinfo->si_wbr_create_ops->fin) { + err = sbinfo->si_wbr_create_ops->fin(sb); + if (!err) + err = 1; + } + + sbinfo->si_wbr_create = create->wbr_create; + sbinfo->si_wbr_create_ops = au_wbr_create_ops + create->wbr_create; + switch (create->wbr_create) { + case AuWbrCreate_MFSRRV: + case AuWbrCreate_MFSRR: + case AuWbrCreate_PMFSRR: + case AuWbrCreate_PMFSRRV: + sbinfo->si_wbr_mfs.mfsrr_watermark = create->mfsrr_watermark; + /*FALLTHROUGH*/ + case AuWbrCreate_MFS: + case AuWbrCreate_MFSV: + case AuWbrCreate_PMFS: + case AuWbrCreate_PMFSV: + sbinfo->si_wbr_mfs.mfs_expire + = msecs_to_jiffies(create->mfs_second * MSEC_PER_SEC); + break; + } + + if (sbinfo->si_wbr_create_ops->init) + sbinfo->si_wbr_create_ops->init(sb); /* ignore */ + + return err; +} + +/* + * returns, + * plus: processed without an error + * zero: unprocessed + */ +static int au_opt_simple(struct super_block *sb, struct au_opt *opt, + struct au_opts *opts) +{ + int err; + struct au_sbinfo *sbinfo; + + SiMustWriteLock(sb); + + err = 1; /* handled */ + sbinfo = au_sbi(sb); + switch (opt->type) { + case Opt_udba: + sbinfo->si_mntflags &= ~AuOptMask_UDBA; + sbinfo->si_mntflags |= opt->udba; + opts->given_udba |= opt->udba; + break; + + case Opt_plink: + au_opt_set(sbinfo->si_mntflags, PLINK); + break; + case Opt_noplink: + if (au_opt_test(sbinfo->si_mntflags, PLINK)) + au_plink_put(sb, /*verbose*/1); + au_opt_clr(sbinfo->si_mntflags, PLINK); + break; + case Opt_list_plink: + if (au_opt_test(sbinfo->si_mntflags, PLINK)) + au_plink_list(sb); + break; + + case Opt_dio: + au_opt_set(sbinfo->si_mntflags, DIO); + au_fset_opts(opts->flags, REFRESH_DYAOP); + break; + case Opt_nodio: + au_opt_clr(sbinfo->si_mntflags, DIO); + au_fset_opts(opts->flags, REFRESH_DYAOP); + break; + + case Opt_fhsm_sec: + au_fhsm_set(sbinfo, opt->fhsm_second); + break; + + case Opt_diropq_a: + au_opt_set(sbinfo->si_mntflags, ALWAYS_DIROPQ); + break; + case Opt_diropq_w: + au_opt_clr(sbinfo->si_mntflags, ALWAYS_DIROPQ); + break; + + case Opt_warn_perm: + au_opt_set(sbinfo->si_mntflags, WARN_PERM); + break; + case Opt_nowarn_perm: + au_opt_clr(sbinfo->si_mntflags, WARN_PERM); + break; + + case Opt_refrof: + au_opt_set(sbinfo->si_mntflags, REFROF); + break; + case Opt_norefrof: + au_opt_clr(sbinfo->si_mntflags, REFROF); + break; + + case Opt_verbose: + au_opt_set(sbinfo->si_mntflags, VERBOSE); + break; + case Opt_noverbose: + au_opt_clr(sbinfo->si_mntflags, VERBOSE); + break; + + case Opt_sum: + au_opt_set(sbinfo->si_mntflags, SUM); + break; + case Opt_wsum: + au_opt_clr(sbinfo->si_mntflags, SUM); + au_opt_set(sbinfo->si_mntflags, SUM_W); + case Opt_nosum: + au_opt_clr(sbinfo->si_mntflags, SUM); + au_opt_clr(sbinfo->si_mntflags, SUM_W); + break; + + case Opt_wbr_create: + err = au_opt_wbr_create(sb, &opt->wbr_create); + break; + case Opt_wbr_copyup: + sbinfo->si_wbr_copyup = opt->wbr_copyup; + sbinfo->si_wbr_copyup_ops = au_wbr_copyup_ops + opt->wbr_copyup; + break; + + case Opt_dirwh: + sbinfo->si_dirwh = opt->dirwh; + break; + + case Opt_rdcache: + sbinfo->si_rdcache + = msecs_to_jiffies(opt->rdcache * MSEC_PER_SEC); + break; + case Opt_rdblk: + sbinfo->si_rdblk = opt->rdblk; + break; + case Opt_rdblk_def: + sbinfo->si_rdblk = AUFS_RDBLK_DEF; + break; + case Opt_rdhash: + sbinfo->si_rdhash = opt->rdhash; + break; + case Opt_rdhash_def: + sbinfo->si_rdhash = AUFS_RDHASH_DEF; + break; + + case Opt_shwh: + au_opt_set(sbinfo->si_mntflags, SHWH); + break; + case Opt_noshwh: + au_opt_clr(sbinfo->si_mntflags, SHWH); + break; + + case Opt_dirperm1: + au_opt_set(sbinfo->si_mntflags, DIRPERM1); + break; + case Opt_nodirperm1: + au_opt_clr(sbinfo->si_mntflags, DIRPERM1); + break; + + case Opt_trunc_xino: + au_opt_set(sbinfo->si_mntflags, TRUNC_XINO); + break; + case Opt_notrunc_xino: + au_opt_clr(sbinfo->si_mntflags, TRUNC_XINO); + break; + + case Opt_trunc_xino_path: + case Opt_itrunc_xino: + err = au_xino_trunc(sb, opt->xino_itrunc.bindex); + if (!err) + err = 1; + break; + + case Opt_trunc_xib: + au_fset_opts(opts->flags, TRUNC_XIB); + break; + case Opt_notrunc_xib: + au_fclr_opts(opts->flags, TRUNC_XIB); + break; + + case Opt_acl: + sb->s_flags |= MS_POSIXACL; + break; + case Opt_noacl: + sb->s_flags &= ~MS_POSIXACL; + break; + + default: + err = 0; + break; + } + + return err; +} + +/* + * returns tri-state. + * plus: processed without an error + * zero: unprocessed + * minus: error + */ +static int au_opt_br(struct super_block *sb, struct au_opt *opt, + struct au_opts *opts) +{ + int err, do_refresh; + + err = 0; + switch (opt->type) { + case Opt_append: + opt->add.bindex = au_sbend(sb) + 1; + if (opt->add.bindex < 0) + opt->add.bindex = 0; + goto add; + case Opt_prepend: + opt->add.bindex = 0; + add: /* indented label */ + case Opt_add: + err = au_br_add(sb, &opt->add, + au_ftest_opts(opts->flags, REMOUNT)); + if (!err) { + err = 1; + au_fset_opts(opts->flags, REFRESH); + } + break; + + case Opt_del: + case Opt_idel: + err = au_br_del(sb, &opt->del, + au_ftest_opts(opts->flags, REMOUNT)); + if (!err) { + err = 1; + au_fset_opts(opts->flags, TRUNC_XIB); + au_fset_opts(opts->flags, REFRESH); + } + break; + + case Opt_mod: + case Opt_imod: + err = au_br_mod(sb, &opt->mod, + au_ftest_opts(opts->flags, REMOUNT), + &do_refresh); + if (!err) { + err = 1; + if (do_refresh) + au_fset_opts(opts->flags, REFRESH); + } + break; + } + + return err; +} + +static int au_opt_xino(struct super_block *sb, struct au_opt *opt, + struct au_opt_xino **opt_xino, + struct au_opts *opts) +{ + int err; + aufs_bindex_t bend, bindex; + struct dentry *root, *parent, *h_root; + + err = 0; + switch (opt->type) { + case Opt_xino: + err = au_xino_set(sb, &opt->xino, + !!au_ftest_opts(opts->flags, REMOUNT)); + if (unlikely(err)) + break; + + *opt_xino = &opt->xino; + au_xino_brid_set(sb, -1); + + /* safe d_parent access */ + parent = opt->xino.file->f_dentry->d_parent; + root = sb->s_root; + bend = au_sbend(sb); + for (bindex = 0; bindex <= bend; bindex++) { + h_root = au_h_dptr(root, bindex); + if (h_root == parent) { + au_xino_brid_set(sb, au_sbr_id(sb, bindex)); + break; + } + } + break; + + case Opt_noxino: + au_xino_clr(sb); + au_xino_brid_set(sb, -1); + *opt_xino = (void *)-1; + break; + } + + return err; +} + +int au_opts_verify(struct super_block *sb, unsigned long sb_flags, + unsigned int pending) +{ + int err, fhsm; + aufs_bindex_t bindex, bend; + unsigned char do_plink, skip, do_free; + struct au_branch *br; + struct au_wbr *wbr; + struct dentry *root; + struct inode *dir, *h_dir; + struct au_sbinfo *sbinfo; + struct au_hinode *hdir; + + SiMustAnyLock(sb); + + sbinfo = au_sbi(sb); + AuDebugOn(!(sbinfo->si_mntflags & AuOptMask_UDBA)); + + if (!(sb_flags & MS_RDONLY)) { + if (unlikely(!au_br_writable(au_sbr_perm(sb, 0)))) + pr_warn("first branch should be rw\n"); + if (unlikely(au_opt_test(sbinfo->si_mntflags, SHWH))) + pr_warn("shwh should be used with ro\n"); + } + + if (au_opt_test((sbinfo->si_mntflags | pending), UDBA_HNOTIFY) + && !au_opt_test(sbinfo->si_mntflags, XINO)) + pr_warn("udba=*notify requires xino\n"); + + if (au_opt_test(sbinfo->si_mntflags, DIRPERM1)) + pr_warn("dirperm1 breaks the protection" + " by the permission bits on the lower branch\n"); + + err = 0; + fhsm = 0; + root = sb->s_root; + dir = root->d_inode; + do_plink = !!au_opt_test(sbinfo->si_mntflags, PLINK); + bend = au_sbend(sb); + for (bindex = 0; !err && bindex <= bend; bindex++) { + skip = 0; + h_dir = au_h_iptr(dir, bindex); + br = au_sbr(sb, bindex); + + if ((br->br_perm & AuBrAttr_ICEX) + && !h_dir->i_op->listxattr) + br->br_perm &= ~AuBrAttr_ICEX; +#if 0 + if ((br->br_perm & AuBrAttr_ICEX_SEC) + && (au_br_sb(br)->s_flags & MS_NOSEC)) + br->br_perm &= ~AuBrAttr_ICEX_SEC; +#endif + + do_free = 0; + wbr = br->br_wbr; + if (wbr) + wbr_wh_read_lock(wbr); + + if (!au_br_writable(br->br_perm)) { + do_free = !!wbr; + skip = (!wbr + || (!wbr->wbr_whbase + && !wbr->wbr_plink + && !wbr->wbr_orph)); + } else if (!au_br_wh_linkable(br->br_perm)) { + /* skip = (!br->br_whbase && !br->br_orph); */ + skip = (!wbr || !wbr->wbr_whbase); + if (skip && wbr) { + if (do_plink) + skip = !!wbr->wbr_plink; + else + skip = !wbr->wbr_plink; + } + } else { + /* skip = (br->br_whbase && br->br_ohph); */ + skip = (wbr && wbr->wbr_whbase); + if (skip) { + if (do_plink) + skip = !!wbr->wbr_plink; + else + skip = !wbr->wbr_plink; + } + } + if (wbr) + wbr_wh_read_unlock(wbr); + + if (au_br_fhsm(br->br_perm)) { + fhsm++; + AuDebugOn(!br->br_fhsm); + } + + if (skip) + continue; + + hdir = au_hi(dir, bindex); + au_hn_imtx_lock_nested(hdir, AuLsc_I_PARENT); + if (wbr) + wbr_wh_write_lock(wbr); + err = au_wh_init(br, sb); + if (wbr) + wbr_wh_write_unlock(wbr); + au_hn_imtx_unlock(hdir); + + if (!err && do_free) { + kfree(wbr); + br->br_wbr = NULL; + } + } + + if (fhsm >= 2) { + au_fset_si(sbinfo, FHSM); + for (bindex = bend; bindex >= 0; bindex--) { + br = au_sbr(sb, bindex); + if (au_br_fhsm(br->br_perm)) { + au_fhsm_set_bottom(sb, bindex); + break; + } + } + } else { + au_fclr_si(sbinfo, FHSM); + au_fhsm_set_bottom(sb, -1); + } + + return err; +} + +int au_opts_mount(struct super_block *sb, struct au_opts *opts) +{ + int err; + unsigned int tmp; + aufs_bindex_t bindex, bend; + struct au_opt *opt; + struct au_opt_xino *opt_xino, xino; + struct au_sbinfo *sbinfo; + struct au_branch *br; + struct inode *dir; + + SiMustWriteLock(sb); + + err = 0; + opt_xino = NULL; + opt = opts->opt; + while (err >= 0 && opt->type != Opt_tail) + err = au_opt_simple(sb, opt++, opts); + if (err > 0) + err = 0; + else if (unlikely(err < 0)) + goto out; + + /* disable xino and udba temporary */ + sbinfo = au_sbi(sb); + tmp = sbinfo->si_mntflags; + au_opt_clr(sbinfo->si_mntflags, XINO); + au_opt_set_udba(sbinfo->si_mntflags, UDBA_REVAL); + + opt = opts->opt; + while (err >= 0 && opt->type != Opt_tail) + err = au_opt_br(sb, opt++, opts); + if (err > 0) + err = 0; + else if (unlikely(err < 0)) + goto out; + + bend = au_sbend(sb); + if (unlikely(bend < 0)) { + err = -EINVAL; + pr_err("no branches\n"); + goto out; + } + + if (au_opt_test(tmp, XINO)) + au_opt_set(sbinfo->si_mntflags, XINO); + opt = opts->opt; + while (!err && opt->type != Opt_tail) + err = au_opt_xino(sb, opt++, &opt_xino, opts); + if (unlikely(err)) + goto out; + + err = au_opts_verify(sb, sb->s_flags, tmp); + if (unlikely(err)) + goto out; + + /* restore xino */ + if (au_opt_test(tmp, XINO) && !opt_xino) { + xino.file = au_xino_def(sb); + err = PTR_ERR(xino.file); + if (IS_ERR(xino.file)) + goto out; + + err = au_xino_set(sb, &xino, /*remount*/0); + fput(xino.file); + if (unlikely(err)) + goto out; + } + + /* restore udba */ + tmp &= AuOptMask_UDBA; + sbinfo->si_mntflags &= ~AuOptMask_UDBA; + sbinfo->si_mntflags |= tmp; + bend = au_sbend(sb); + for (bindex = 0; bindex <= bend; bindex++) { + br = au_sbr(sb, bindex); + err = au_hnotify_reset_br(tmp, br, br->br_perm); + if (unlikely(err)) + AuIOErr("hnotify failed on br %d, %d, ignored\n", + bindex, err); + /* go on even if err */ + } + if (au_opt_test(tmp, UDBA_HNOTIFY)) { + dir = sb->s_root->d_inode; + au_hn_reset(dir, au_hi_flags(dir, /*isdir*/1) & ~AuHi_XINO); + } + +out: + return err; +} + +int au_opts_remount(struct super_block *sb, struct au_opts *opts) +{ + int err, rerr; + struct inode *dir; + struct au_opt_xino *opt_xino; + struct au_opt *opt; + struct au_sbinfo *sbinfo; + + SiMustWriteLock(sb); + + dir = sb->s_root->d_inode; + sbinfo = au_sbi(sb); + err = 0; + opt_xino = NULL; + opt = opts->opt; + while (err >= 0 && opt->type != Opt_tail) { + err = au_opt_simple(sb, opt, opts); + if (!err) + err = au_opt_br(sb, opt, opts); + if (!err) + err = au_opt_xino(sb, opt, &opt_xino, opts); + opt++; + } + if (err > 0) + err = 0; + AuTraceErr(err); + /* go on even err */ + + rerr = au_opts_verify(sb, opts->sb_flags, /*pending*/0); + if (unlikely(rerr && !err)) + err = rerr; + + if (au_ftest_opts(opts->flags, TRUNC_XIB)) { + rerr = au_xib_trunc(sb); + if (unlikely(rerr && !err)) + err = rerr; + } + + /* will be handled by the caller */ + if (!au_ftest_opts(opts->flags, REFRESH) + && (opts->given_udba || au_opt_test(sbinfo->si_mntflags, XINO))) + au_fset_opts(opts->flags, REFRESH); + + AuDbg("status 0x%x\n", opts->flags); + return err; +} + +/* ---------------------------------------------------------------------- */ + +unsigned int au_opt_udba(struct super_block *sb) +{ + return au_mntflags(sb) & AuOptMask_UDBA; +} diff -Naur kernel.21.3.orig/fs/aufs/opts.h kernel.21.3.new/fs/aufs/opts.h --- kernel.21.3.orig/fs/aufs/opts.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/opts.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,213 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * mount options/flags + */ + +#ifndef __AUFS_OPTS_H__ +#define __AUFS_OPTS_H__ + +#ifdef __KERNEL__ + +#include +#include "branch.h" + +struct file; +struct super_block; + +/* ---------------------------------------------------------------------- */ + +/* mount flags */ +#define AuOpt_XINO 1 /* external inode number bitmap + and translation table */ +#define AuOpt_TRUNC_XINO (1 << 1) /* truncate xino files */ +#define AuOpt_UDBA_NONE (1 << 2) /* users direct branch access */ +#define AuOpt_UDBA_REVAL (1 << 3) +#define AuOpt_UDBA_HNOTIFY (1 << 4) +#define AuOpt_SHWH (1 << 5) /* show whiteout */ +#define AuOpt_PLINK (1 << 6) /* pseudo-link */ +#define AuOpt_DIRPERM1 (1 << 7) /* ignore the lower dir's perm + bits */ +#define AuOpt_REFROF (1 << 8) /* unimplemented */ +#define AuOpt_ALWAYS_DIROPQ (1 << 9) /* policy to creating diropq */ +#define AuOpt_SUM (1 << 10) /* summation for statfs(2) */ +#define AuOpt_SUM_W (1 << 11) /* unimplemented */ +#define AuOpt_WARN_PERM (1 << 12) /* warn when add-branch */ +#define AuOpt_VERBOSE (1 << 13) /* busy inode when del-branch */ +#define AuOpt_DIO (1 << 14) /* direct io */ + +#ifndef CONFIG_AUFS_HNOTIFY +#undef AuOpt_UDBA_HNOTIFY +#define AuOpt_UDBA_HNOTIFY 0 +#endif +#ifndef CONFIG_AUFS_SHWH +#undef AuOpt_SHWH +#define AuOpt_SHWH 0 +#endif + +#define AuOpt_Def (AuOpt_XINO \ + | AuOpt_UDBA_REVAL \ + | AuOpt_PLINK \ + /* | AuOpt_DIRPERM1 */ \ + | AuOpt_WARN_PERM) +#define AuOptMask_UDBA (AuOpt_UDBA_NONE \ + | AuOpt_UDBA_REVAL \ + | AuOpt_UDBA_HNOTIFY) + +#define au_opt_test(flags, name) (flags & AuOpt_##name) +#define au_opt_set(flags, name) do { \ + BUILD_BUG_ON(AuOpt_##name & AuOptMask_UDBA); \ + ((flags) |= AuOpt_##name); \ +} while (0) +#define au_opt_set_udba(flags, name) do { \ + (flags) &= ~AuOptMask_UDBA; \ + ((flags) |= AuOpt_##name); \ +} while (0) +#define au_opt_clr(flags, name) do { \ + ((flags) &= ~AuOpt_##name); \ +} while (0) + +static inline unsigned int au_opts_plink(unsigned int mntflags) +{ +#ifdef CONFIG_PROC_FS + return mntflags; +#else + return mntflags & ~AuOpt_PLINK; +#endif +} + +/* ---------------------------------------------------------------------- */ + +/* policies to select one among multiple writable branches */ +enum { + AuWbrCreate_TDP, /* top down parent */ + AuWbrCreate_RR, /* round robin */ + AuWbrCreate_MFS, /* most free space */ + AuWbrCreate_MFSV, /* mfs with seconds */ + AuWbrCreate_MFSRR, /* mfs then rr */ + AuWbrCreate_MFSRRV, /* mfs then rr with seconds */ + AuWbrCreate_PMFS, /* parent and mfs */ + AuWbrCreate_PMFSV, /* parent and mfs with seconds */ + AuWbrCreate_PMFSRR, /* parent, mfs and round-robin */ + AuWbrCreate_PMFSRRV, /* plus seconds */ + + AuWbrCreate_Def = AuWbrCreate_TDP +}; + +enum { + AuWbrCopyup_TDP, /* top down parent */ + AuWbrCopyup_BUP, /* bottom up parent */ + AuWbrCopyup_BU, /* bottom up */ + + AuWbrCopyup_Def = AuWbrCopyup_TDP +}; + +/* ---------------------------------------------------------------------- */ + +struct au_opt_add { + aufs_bindex_t bindex; + char *pathname; + int perm; + struct path path; +}; + +struct au_opt_del { + char *pathname; + struct path h_path; +}; + +struct au_opt_mod { + char *path; + int perm; + struct dentry *h_root; +}; + +struct au_opt_xino { + char *path; + struct file *file; +}; + +struct au_opt_xino_itrunc { + aufs_bindex_t bindex; +}; + +struct au_opt_wbr_create { + int wbr_create; + int mfs_second; + unsigned long long mfsrr_watermark; +}; + +struct au_opt { + int type; + union { + struct au_opt_xino xino; + struct au_opt_xino_itrunc xino_itrunc; + struct au_opt_add add; + struct au_opt_del del; + struct au_opt_mod mod; + int dirwh; + int rdcache; + unsigned int rdblk; + unsigned int rdhash; + int udba; + struct au_opt_wbr_create wbr_create; + int wbr_copyup; + unsigned int fhsm_second; + }; +}; + +/* opts flags */ +#define AuOpts_REMOUNT 1 +#define AuOpts_REFRESH (1 << 1) +#define AuOpts_TRUNC_XIB (1 << 2) +#define AuOpts_REFRESH_DYAOP (1 << 3) +#define au_ftest_opts(flags, name) ((flags) & AuOpts_##name) +#define au_fset_opts(flags, name) \ + do { (flags) |= AuOpts_##name; } while (0) +#define au_fclr_opts(flags, name) \ + do { (flags) &= ~AuOpts_##name; } while (0) + +struct au_opts { + struct au_opt *opt; + int max_opt; + + unsigned int given_udba; + unsigned int flags; + unsigned long sb_flags; +}; + +/* ---------------------------------------------------------------------- */ + +void au_optstr_br_perm(au_br_perm_str_t *str, int perm); +const char *au_optstr_udba(int udba); +const char *au_optstr_wbr_copyup(int wbr_copyup); +const char *au_optstr_wbr_create(int wbr_create); + +void au_opts_free(struct au_opts *opts); +int au_opts_parse(struct super_block *sb, char *str, struct au_opts *opts); +int au_opts_verify(struct super_block *sb, unsigned long sb_flags, + unsigned int pending); +int au_opts_mount(struct super_block *sb, struct au_opts *opts); +int au_opts_remount(struct super_block *sb, struct au_opts *opts); + +unsigned int au_opt_udba(struct super_block *sb); + +/* ---------------------------------------------------------------------- */ + +#endif /* __KERNEL__ */ +#endif /* __AUFS_OPTS_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/plink.c kernel.21.3.new/fs/aufs/plink.c --- kernel.21.3.orig/fs/aufs/plink.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/plink.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,519 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * pseudo-link + */ + +#include "aufs.h" + +/* + * the pseudo-link maintenance mode. + * during a user process maintains the pseudo-links, + * prohibit adding a new plink and branch manipulation. + * + * Flags + * NOPLM: + * For entry functions which will handle plink, and i_mutex is already held + * in VFS. + * They cannot wait and should return an error at once. + * Callers has to check the error. + * NOPLMW: + * For entry functions which will handle plink, but i_mutex is not held + * in VFS. + * They can wait the plink maintenance mode to finish. + * + * They behave like F_SETLK and F_SETLKW. + * If the caller never handle plink, then both flags are unnecessary. + */ + +int au_plink_maint(struct super_block *sb, int flags) +{ + int err; + pid_t pid, ppid; + struct au_sbinfo *sbi; + + SiMustAnyLock(sb); + + err = 0; + if (!au_opt_test(au_mntflags(sb), PLINK)) + goto out; + + sbi = au_sbi(sb); + pid = sbi->si_plink_maint_pid; + if (!pid || pid == current->pid) + goto out; + + /* todo: it highly depends upon /sbin/mount.aufs */ + rcu_read_lock(); + ppid = task_pid_vnr(rcu_dereference(current->real_parent)); + rcu_read_unlock(); + if (pid == ppid) + goto out; + + if (au_ftest_lock(flags, NOPLMW)) { + /* if there is no i_mutex lock in VFS, we don't need to wait */ + /* AuDebugOn(!lockdep_depth(current)); */ + while (sbi->si_plink_maint_pid) { + si_read_unlock(sb); + /* gave up wake_up_bit() */ + wait_event(sbi->si_plink_wq, !sbi->si_plink_maint_pid); + + if (au_ftest_lock(flags, FLUSH)) + au_nwt_flush(&sbi->si_nowait); + si_noflush_read_lock(sb); + } + } else if (au_ftest_lock(flags, NOPLM)) { + AuDbg("ppid %d, pid %d\n", ppid, pid); + err = -EAGAIN; + } + +out: + return err; +} + +void au_plink_maint_leave(struct au_sbinfo *sbinfo) +{ + spin_lock(&sbinfo->si_plink_maint_lock); + sbinfo->si_plink_maint_pid = 0; + spin_unlock(&sbinfo->si_plink_maint_lock); + wake_up_all(&sbinfo->si_plink_wq); +} + +int au_plink_maint_enter(struct super_block *sb) +{ + int err; + struct au_sbinfo *sbinfo; + + err = 0; + sbinfo = au_sbi(sb); + /* make sure i am the only one in this fs */ + si_write_lock(sb, AuLock_FLUSH); + if (au_opt_test(au_mntflags(sb), PLINK)) { + spin_lock(&sbinfo->si_plink_maint_lock); + if (!sbinfo->si_plink_maint_pid) + sbinfo->si_plink_maint_pid = current->pid; + else + err = -EBUSY; + spin_unlock(&sbinfo->si_plink_maint_lock); + } + si_write_unlock(sb); + + return err; +} + +/* ---------------------------------------------------------------------- */ + +#ifdef CONFIG_AUFS_DEBUG +void au_plink_list(struct super_block *sb) +{ + int i; + struct au_sbinfo *sbinfo; + struct hlist_head *plink_hlist; + struct pseudo_link *plink; + + SiMustAnyLock(sb); + + sbinfo = au_sbi(sb); + AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK)); + AuDebugOn(au_plink_maint(sb, AuLock_NOPLM)); + + for (i = 0; i < AuPlink_NHASH; i++) { + plink_hlist = &sbinfo->si_plink[i].head; + rcu_read_lock(); + hlist_for_each_entry_rcu(plink, plink_hlist, hlist) + AuDbg("%lu\n", plink->inode->i_ino); + rcu_read_unlock(); + } +} +#endif + +/* is the inode pseudo-linked? */ +int au_plink_test(struct inode *inode) +{ + int found, i; + struct au_sbinfo *sbinfo; + struct hlist_head *plink_hlist; + struct pseudo_link *plink; + + sbinfo = au_sbi(inode->i_sb); + AuRwMustAnyLock(&sbinfo->si_rwsem); + AuDebugOn(!au_opt_test(au_mntflags(inode->i_sb), PLINK)); + AuDebugOn(au_plink_maint(inode->i_sb, AuLock_NOPLM)); + + found = 0; + i = au_plink_hash(inode->i_ino); + plink_hlist = &sbinfo->si_plink[i].head; + rcu_read_lock(); + hlist_for_each_entry_rcu(plink, plink_hlist, hlist) + if (plink->inode == inode) { + found = 1; + break; + } + rcu_read_unlock(); + return found; +} + +/* ---------------------------------------------------------------------- */ + +/* + * generate a name for plink. + * the file will be stored under AUFS_WH_PLINKDIR. + */ +/* 20 is max digits length of ulong 64 */ +#define PLINK_NAME_LEN ((20 + 1) * 2) + +static int plink_name(char *name, int len, struct inode *inode, + aufs_bindex_t bindex) +{ + int rlen; + struct inode *h_inode; + + h_inode = au_h_iptr(inode, bindex); + rlen = snprintf(name, len, "%lu.%lu", inode->i_ino, h_inode->i_ino); + return rlen; +} + +struct au_do_plink_lkup_args { + struct dentry **errp; + struct qstr *tgtname; + struct dentry *h_parent; + struct au_branch *br; +}; + +static struct dentry *au_do_plink_lkup(struct qstr *tgtname, + struct dentry *h_parent, + struct au_branch *br) +{ + struct dentry *h_dentry; + struct mutex *h_mtx; + + h_mtx = &h_parent->d_inode->i_mutex; + mutex_lock_nested(h_mtx, AuLsc_I_CHILD2); + h_dentry = vfsub_lkup_one(tgtname, h_parent); + mutex_unlock(h_mtx); + return h_dentry; +} + +static void au_call_do_plink_lkup(void *args) +{ + struct au_do_plink_lkup_args *a = args; + *a->errp = au_do_plink_lkup(a->tgtname, a->h_parent, a->br); +} + +/* lookup the plink-ed @inode under the branch at @bindex */ +struct dentry *au_plink_lkup(struct inode *inode, aufs_bindex_t bindex) +{ + struct dentry *h_dentry, *h_parent; + struct au_branch *br; + struct inode *h_dir; + int wkq_err; + char a[PLINK_NAME_LEN]; + struct qstr tgtname = QSTR_INIT(a, 0); + + AuDebugOn(au_plink_maint(inode->i_sb, AuLock_NOPLM)); + + br = au_sbr(inode->i_sb, bindex); + h_parent = br->br_wbr->wbr_plink; + h_dir = h_parent->d_inode; + tgtname.len = plink_name(a, sizeof(a), inode, bindex); + + if (!uid_eq(current_fsuid(), GLOBAL_ROOT_UID)) { + struct au_do_plink_lkup_args args = { + .errp = &h_dentry, + .tgtname = &tgtname, + .h_parent = h_parent, + .br = br + }; + + wkq_err = au_wkq_wait(au_call_do_plink_lkup, &args); + if (unlikely(wkq_err)) + h_dentry = ERR_PTR(wkq_err); + } else + h_dentry = au_do_plink_lkup(&tgtname, h_parent, br); + + return h_dentry; +} + +/* create a pseudo-link */ +static int do_whplink(struct qstr *tgt, struct dentry *h_parent, + struct dentry *h_dentry, struct au_branch *br) +{ + int err; + struct path h_path = { + .mnt = au_br_mnt(br) + }; + struct inode *h_dir; + + h_dir = h_parent->d_inode; + mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_CHILD2); +again: + h_path.dentry = vfsub_lkup_one(tgt, h_parent); + err = PTR_ERR(h_path.dentry); + if (IS_ERR(h_path.dentry)) + goto out; + + err = 0; + /* wh.plink dir is not monitored */ + /* todo: is it really safe? */ + if (h_path.dentry->d_inode + && h_path.dentry->d_inode != h_dentry->d_inode) { + err = vfsub_unlink(h_dir, &h_path, /*force*/0); + dput(h_path.dentry); + h_path.dentry = NULL; + if (!err) + goto again; + } + if (!err && !h_path.dentry->d_inode) + err = vfsub_link(h_dentry, h_dir, &h_path); + dput(h_path.dentry); + +out: + mutex_unlock(&h_dir->i_mutex); + return err; +} + +struct do_whplink_args { + int *errp; + struct qstr *tgt; + struct dentry *h_parent; + struct dentry *h_dentry; + struct au_branch *br; +}; + +static void call_do_whplink(void *args) +{ + struct do_whplink_args *a = args; + *a->errp = do_whplink(a->tgt, a->h_parent, a->h_dentry, a->br); +} + +static int whplink(struct dentry *h_dentry, struct inode *inode, + aufs_bindex_t bindex, struct au_branch *br) +{ + int err, wkq_err; + struct au_wbr *wbr; + struct dentry *h_parent; + struct inode *h_dir; + char a[PLINK_NAME_LEN]; + struct qstr tgtname = QSTR_INIT(a, 0); + + wbr = au_sbr(inode->i_sb, bindex)->br_wbr; + h_parent = wbr->wbr_plink; + h_dir = h_parent->d_inode; + tgtname.len = plink_name(a, sizeof(a), inode, bindex); + + /* always superio. */ + if (!uid_eq(current_fsuid(), GLOBAL_ROOT_UID)) { + struct do_whplink_args args = { + .errp = &err, + .tgt = &tgtname, + .h_parent = h_parent, + .h_dentry = h_dentry, + .br = br + }; + wkq_err = au_wkq_wait(call_do_whplink, &args); + if (unlikely(wkq_err)) + err = wkq_err; + } else + err = do_whplink(&tgtname, h_parent, h_dentry, br); + + return err; +} + +/* free a single plink */ +static void do_put_plink(struct pseudo_link *plink, int do_del) +{ + if (do_del) + hlist_del(&plink->hlist); + iput(plink->inode); + kfree(plink); +} + +static void do_put_plink_rcu(struct rcu_head *rcu) +{ + struct pseudo_link *plink; + + plink = container_of(rcu, struct pseudo_link, rcu); + iput(plink->inode); + kfree(plink); +} + +/* + * create a new pseudo-link for @h_dentry on @bindex. + * the linked inode is held in aufs @inode. + */ +void au_plink_append(struct inode *inode, aufs_bindex_t bindex, + struct dentry *h_dentry) +{ + struct super_block *sb; + struct au_sbinfo *sbinfo; + struct hlist_head *plink_hlist; + struct pseudo_link *plink, *tmp; + struct au_sphlhead *sphl; + int found, err, cnt, i; + + sb = inode->i_sb; + sbinfo = au_sbi(sb); + AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK)); + AuDebugOn(au_plink_maint(sb, AuLock_NOPLM)); + + found = au_plink_test(inode); + if (found) + return; + + i = au_plink_hash(inode->i_ino); + sphl = sbinfo->si_plink + i; + plink_hlist = &sphl->head; + tmp = kmalloc(sizeof(*plink), GFP_NOFS); + if (tmp) + tmp->inode = au_igrab(inode); + else { + err = -ENOMEM; + goto out; + } + + spin_lock(&sphl->spin); + hlist_for_each_entry(plink, plink_hlist, hlist) { + if (plink->inode == inode) { + found = 1; + break; + } + } + if (!found) + hlist_add_head_rcu(&tmp->hlist, plink_hlist); + spin_unlock(&sphl->spin); + if (!found) { + cnt = au_sphl_count(sphl); +#define msg "unexpectedly unblanced or too many pseudo-links" + if (cnt > AUFS_PLINK_WARN) + AuWarn1(msg ", %d\n", cnt); +#undef msg + err = whplink(h_dentry, inode, bindex, au_sbr(sb, bindex)); + } else { + do_put_plink(tmp, 0); + return; + } + +out: + if (unlikely(err)) { + pr_warn("err %d, damaged pseudo link.\n", err); + if (tmp) { + au_sphl_del_rcu(&tmp->hlist, sphl); + call_rcu(&tmp->rcu, do_put_plink_rcu); + } + } +} + +/* free all plinks */ +void au_plink_put(struct super_block *sb, int verbose) +{ + int i, warned; + struct au_sbinfo *sbinfo; + struct hlist_head *plink_hlist; + struct hlist_node *tmp; + struct pseudo_link *plink; + + SiMustWriteLock(sb); + + sbinfo = au_sbi(sb); + AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK)); + AuDebugOn(au_plink_maint(sb, AuLock_NOPLM)); + + /* no spin_lock since sbinfo is write-locked */ + warned = 0; + for (i = 0; i < AuPlink_NHASH; i++) { + plink_hlist = &sbinfo->si_plink[i].head; + if (!warned && verbose && !hlist_empty(plink_hlist)) { + pr_warn("pseudo-link is not flushed"); + warned = 1; + } + hlist_for_each_entry_safe(plink, tmp, plink_hlist, hlist) + do_put_plink(plink, 0); + INIT_HLIST_HEAD(plink_hlist); + } +} + +void au_plink_clean(struct super_block *sb, int verbose) +{ + struct dentry *root; + + root = sb->s_root; + aufs_write_lock(root); + if (au_opt_test(au_mntflags(sb), PLINK)) + au_plink_put(sb, verbose); + aufs_write_unlock(root); +} + +static int au_plink_do_half_refresh(struct inode *inode, aufs_bindex_t br_id) +{ + int do_put; + aufs_bindex_t bstart, bend, bindex; + + do_put = 0; + bstart = au_ibstart(inode); + bend = au_ibend(inode); + if (bstart >= 0) { + for (bindex = bstart; bindex <= bend; bindex++) { + if (!au_h_iptr(inode, bindex) + || au_ii_br_id(inode, bindex) != br_id) + continue; + au_set_h_iptr(inode, bindex, NULL, 0); + do_put = 1; + break; + } + if (do_put) + for (bindex = bstart; bindex <= bend; bindex++) + if (au_h_iptr(inode, bindex)) { + do_put = 0; + break; + } + } else + do_put = 1; + + return do_put; +} + +/* free the plinks on a branch specified by @br_id */ +void au_plink_half_refresh(struct super_block *sb, aufs_bindex_t br_id) +{ + struct au_sbinfo *sbinfo; + struct hlist_head *plink_hlist; + struct hlist_node *tmp; + struct pseudo_link *plink; + struct inode *inode; + int i, do_put; + + SiMustWriteLock(sb); + + sbinfo = au_sbi(sb); + AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK)); + AuDebugOn(au_plink_maint(sb, AuLock_NOPLM)); + + /* no spin_lock since sbinfo is write-locked */ + for (i = 0; i < AuPlink_NHASH; i++) { + plink_hlist = &sbinfo->si_plink[i].head; + hlist_for_each_entry_safe(plink, tmp, plink_hlist, hlist) { + inode = au_igrab(plink->inode); + ii_write_lock_child(inode); + do_put = au_plink_do_half_refresh(inode, br_id); + if (do_put) + do_put_plink(plink, 1); + ii_write_unlock(inode); + iput(inode); + } + } +} diff -Naur kernel.21.3.orig/fs/aufs/poll.c kernel.21.3.new/fs/aufs/poll.c --- kernel.21.3.orig/fs/aufs/poll.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/poll.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,55 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * poll operation + * There is only one filesystem which implements ->poll operation, currently. + */ + +#include "aufs.h" + +unsigned int aufs_poll(struct file *file, poll_table *wait) +{ + unsigned int mask; + int err; + struct file *h_file; + struct dentry *dentry; + struct super_block *sb; + + /* We should pretend an error happened. */ + mask = POLLERR /* | POLLIN | POLLOUT */; + dentry = file->f_dentry; + sb = dentry->d_sb; + si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW); + err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0); + if (unlikely(err)) + goto out; + + /* it is not an error if h_file has no operation */ + mask = DEFAULT_POLLMASK; + h_file = au_hf_top(file); + if (h_file->f_op && h_file->f_op->poll) + mask = h_file->f_op->poll(h_file, wait); + + di_read_unlock(dentry, AuLock_IR); + fi_read_unlock(file); + +out: + si_read_unlock(sb); + AuTraceErr((int)mask); + return mask; +} diff -Naur kernel.21.3.orig/fs/aufs/procfs.c kernel.21.3.new/fs/aufs/procfs.c --- kernel.21.3.orig/fs/aufs/procfs.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/procfs.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,169 @@ +/* + * Copyright (C) 2010-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * procfs interfaces + */ + +#include +#include "aufs.h" + +static int au_procfs_plm_release(struct inode *inode, struct file *file) +{ + struct au_sbinfo *sbinfo; + + sbinfo = file->private_data; + if (sbinfo) { + au_plink_maint_leave(sbinfo); + kobject_put(&sbinfo->si_kobj); + } + + return 0; +} + +static void au_procfs_plm_write_clean(struct file *file) +{ + struct au_sbinfo *sbinfo; + + sbinfo = file->private_data; + if (sbinfo) + au_plink_clean(sbinfo->si_sb, /*verbose*/0); +} + +static int au_procfs_plm_write_si(struct file *file, unsigned long id) +{ + int err; + struct super_block *sb; + struct au_sbinfo *sbinfo; + + err = -EBUSY; + if (unlikely(file->private_data)) + goto out; + + sb = NULL; + /* don't use au_sbilist_lock() here */ + spin_lock(&au_sbilist.spin); + list_for_each_entry(sbinfo, &au_sbilist.head, si_list) + if (id == sysaufs_si_id(sbinfo)) { + kobject_get(&sbinfo->si_kobj); + sb = sbinfo->si_sb; + break; + } + spin_unlock(&au_sbilist.spin); + + err = -EINVAL; + if (unlikely(!sb)) + goto out; + + err = au_plink_maint_enter(sb); + if (!err) + /* keep kobject_get() */ + file->private_data = sbinfo; + else + kobject_put(&sbinfo->si_kobj); +out: + return err; +} + +/* + * Accept a valid "si=xxxx" only. + * Once it is accepted successfully, accept "clean" too. + */ +static ssize_t au_procfs_plm_write(struct file *file, const char __user *ubuf, + size_t count, loff_t *ppos) +{ + ssize_t err; + unsigned long id; + /* last newline is allowed */ + char buf[3 + sizeof(unsigned long) * 2 + 1]; + + err = -EACCES; + if (unlikely(!capable(CAP_SYS_ADMIN))) + goto out; + + err = -EINVAL; + if (unlikely(count > sizeof(buf))) + goto out; + + err = copy_from_user(buf, ubuf, count); + if (unlikely(err)) { + err = -EFAULT; + goto out; + } + buf[count] = 0; + + err = -EINVAL; + if (!strcmp("clean", buf)) { + au_procfs_plm_write_clean(file); + goto out_success; + } else if (unlikely(strncmp("si=", buf, 3))) + goto out; + + err = kstrtoul(buf + 3, 16, &id); + if (unlikely(err)) + goto out; + + err = au_procfs_plm_write_si(file, id); + if (unlikely(err)) + goto out; + +out_success: + err = count; /* success */ +out: + return err; +} + +static const struct file_operations au_procfs_plm_fop = { + .write = au_procfs_plm_write, + .release = au_procfs_plm_release, + .owner = THIS_MODULE +}; + +/* ---------------------------------------------------------------------- */ + +static struct proc_dir_entry *au_procfs_dir; + +void au_procfs_fin(void) +{ + remove_proc_entry(AUFS_PLINK_MAINT_NAME, au_procfs_dir); + remove_proc_entry(AUFS_PLINK_MAINT_DIR, NULL); +} + +int __init au_procfs_init(void) +{ + int err; + struct proc_dir_entry *entry; + + err = -ENOMEM; + au_procfs_dir = proc_mkdir(AUFS_PLINK_MAINT_DIR, NULL); + if (unlikely(!au_procfs_dir)) + goto out; + + entry = proc_create(AUFS_PLINK_MAINT_NAME, S_IFREG | S_IWUSR, + au_procfs_dir, &au_procfs_plm_fop); + if (unlikely(!entry)) + goto out_dir; + + err = 0; + goto out; /* success */ + + +out_dir: + remove_proc_entry(AUFS_PLINK_MAINT_DIR, NULL); +out: + return err; +} diff -Naur kernel.21.3.orig/fs/aufs/rdu.c kernel.21.3.new/fs/aufs/rdu.c --- kernel.21.3.orig/fs/aufs/rdu.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/rdu.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,383 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * readdir in userspace. + */ + +#include +#include +#include +#include "aufs.h" + +/* bits for struct aufs_rdu.flags */ +#define AuRdu_CALLED 1 +#define AuRdu_CONT (1 << 1) +#define AuRdu_FULL (1 << 2) +#define au_ftest_rdu(flags, name) ((flags) & AuRdu_##name) +#define au_fset_rdu(flags, name) \ + do { (flags) |= AuRdu_##name; } while (0) +#define au_fclr_rdu(flags, name) \ + do { (flags) &= ~AuRdu_##name; } while (0) + +struct au_rdu_arg { + struct aufs_rdu *rdu; + union au_rdu_ent_ul ent; + unsigned long end; + + struct super_block *sb; + int err; +}; + +static int au_rdu_fill(void *__arg, const char *name, int nlen, + loff_t offset, u64 h_ino, unsigned int d_type) +{ + int err, len; + struct au_rdu_arg *arg = __arg; + struct aufs_rdu *rdu = arg->rdu; + struct au_rdu_ent ent; + + err = 0; + arg->err = 0; + au_fset_rdu(rdu->cookie.flags, CALLED); + len = au_rdu_len(nlen); + if (arg->ent.ul + len < arg->end) { + ent.ino = h_ino; + ent.bindex = rdu->cookie.bindex; + ent.type = d_type; + ent.nlen = nlen; + if (unlikely(nlen > AUFS_MAX_NAMELEN)) + ent.type = DT_UNKNOWN; + + /* unnecessary to support mmap_sem since this is a dir */ + err = -EFAULT; + if (copy_to_user(arg->ent.e, &ent, sizeof(ent))) + goto out; + if (copy_to_user(arg->ent.e->name, name, nlen)) + goto out; + /* the terminating NULL */ + if (__put_user(0, arg->ent.e->name + nlen)) + goto out; + err = 0; + /* AuDbg("%p, %.*s\n", arg->ent.p, nlen, name); */ + arg->ent.ul += len; + rdu->rent++; + } else { + err = -EFAULT; + au_fset_rdu(rdu->cookie.flags, FULL); + rdu->full = 1; + rdu->tail = arg->ent; + } + +out: + /* AuTraceErr(err); */ + return err; +} + +static int au_rdu_do(struct file *h_file, struct au_rdu_arg *arg) +{ + int err; + loff_t offset; + struct au_rdu_cookie *cookie = &arg->rdu->cookie; + + /* we don't have to care (FMODE_32BITHASH | FMODE_64BITHASH) for ext4 */ + offset = vfsub_llseek(h_file, cookie->h_pos, SEEK_SET); + err = offset; + if (unlikely(offset != cookie->h_pos)) + goto out; + + err = 0; + do { + arg->err = 0; + au_fclr_rdu(cookie->flags, CALLED); + /* smp_mb(); */ + err = vfsub_readdir(h_file, au_rdu_fill, arg); + if (err >= 0) + err = arg->err; + } while (!err + && au_ftest_rdu(cookie->flags, CALLED) + && !au_ftest_rdu(cookie->flags, FULL)); + cookie->h_pos = h_file->f_pos; + +out: + AuTraceErr(err); + return err; +} + +static int au_rdu(struct file *file, struct aufs_rdu *rdu) +{ + int err; + aufs_bindex_t bend; + struct au_rdu_arg arg; + struct dentry *dentry; + struct inode *inode; + struct file *h_file; + struct au_rdu_cookie *cookie = &rdu->cookie; + + err = !access_ok(VERIFY_WRITE, rdu->ent.e, rdu->sz); + if (unlikely(err)) { + err = -EFAULT; + AuTraceErr(err); + goto out; + } + rdu->rent = 0; + rdu->tail = rdu->ent; + rdu->full = 0; + arg.rdu = rdu; + arg.ent = rdu->ent; + arg.end = arg.ent.ul; + arg.end += rdu->sz; + + err = -ENOTDIR; + if (unlikely(!file->f_op || !file->f_op->readdir)) + goto out; + + err = security_file_permission(file, MAY_READ); + AuTraceErr(err); + if (unlikely(err)) + goto out; + + dentry = file->f_dentry; + inode = dentry->d_inode; +#if 1 + mutex_lock(&inode->i_mutex); +#else + err = mutex_lock_killable(&inode->i_mutex); + AuTraceErr(err); + if (unlikely(err)) + goto out; +#endif + + arg.sb = inode->i_sb; + err = si_read_lock(arg.sb, AuLock_FLUSH | AuLock_NOPLM); + if (unlikely(err)) + goto out_mtx; + err = au_alive_dir(dentry); + if (unlikely(err)) + goto out_si; + /* todo: reval? */ + fi_read_lock(file); + + err = -EAGAIN; + if (unlikely(au_ftest_rdu(cookie->flags, CONT) + && cookie->generation != au_figen(file))) + goto out_unlock; + + err = 0; + if (!rdu->blk) { + rdu->blk = au_sbi(arg.sb)->si_rdblk; + if (!rdu->blk) + rdu->blk = au_dir_size(file, /*dentry*/NULL); + } + bend = au_fbstart(file); + if (cookie->bindex < bend) + cookie->bindex = bend; + bend = au_fbend_dir(file); + /* AuDbg("b%d, b%d\n", cookie->bindex, bend); */ + for (; !err && cookie->bindex <= bend; + cookie->bindex++, cookie->h_pos = 0) { + h_file = au_hf_dir(file, cookie->bindex); + if (!h_file) + continue; + + au_fclr_rdu(cookie->flags, FULL); + err = au_rdu_do(h_file, &arg); + AuTraceErr(err); + if (unlikely(au_ftest_rdu(cookie->flags, FULL) || err)) + break; + } + AuDbg("rent %llu\n", rdu->rent); + + if (!err && !au_ftest_rdu(cookie->flags, CONT)) { + rdu->shwh = !!au_opt_test(au_sbi(arg.sb)->si_mntflags, SHWH); + au_fset_rdu(cookie->flags, CONT); + cookie->generation = au_figen(file); + } + + ii_read_lock_child(inode); + fsstack_copy_attr_atime(inode, au_h_iptr(inode, au_ibstart(inode))); + ii_read_unlock(inode); + +out_unlock: + fi_read_unlock(file); +out_si: + si_read_unlock(arg.sb); +out_mtx: + mutex_unlock(&inode->i_mutex); +out: + AuTraceErr(err); + return err; +} + +static int au_rdu_ino(struct file *file, struct aufs_rdu *rdu) +{ + int err; + ino_t ino; + unsigned long long nent; + union au_rdu_ent_ul *u; + struct au_rdu_ent ent; + struct super_block *sb; + + err = 0; + nent = rdu->nent; + u = &rdu->ent; + sb = file->f_dentry->d_sb; + si_read_lock(sb, AuLock_FLUSH); + while (nent-- > 0) { + /* unnecessary to support mmap_sem since this is a dir */ + err = copy_from_user(&ent, u->e, sizeof(ent)); + if (!err) + err = !access_ok(VERIFY_WRITE, &u->e->ino, sizeof(ino)); + if (unlikely(err)) { + err = -EFAULT; + AuTraceErr(err); + break; + } + + /* AuDbg("b%d, i%llu\n", ent.bindex, ent.ino); */ + if (!ent.wh) + err = au_ino(sb, ent.bindex, ent.ino, ent.type, &ino); + else + err = au_wh_ino(sb, ent.bindex, ent.ino, ent.type, + &ino); + if (unlikely(err)) { + AuTraceErr(err); + break; + } + + err = __put_user(ino, &u->e->ino); + if (unlikely(err)) { + err = -EFAULT; + AuTraceErr(err); + break; + } + u->ul += au_rdu_len(ent.nlen); + } + si_read_unlock(sb); + + return err; +} + +/* ---------------------------------------------------------------------- */ + +static int au_rdu_verify(struct aufs_rdu *rdu) +{ + AuDbg("rdu{%llu, %p, %u | %u | %llu, %u, %u | " + "%llu, b%d, 0x%x, g%u}\n", + rdu->sz, rdu->ent.e, rdu->verify[AufsCtlRduV_SZ], + rdu->blk, + rdu->rent, rdu->shwh, rdu->full, + rdu->cookie.h_pos, rdu->cookie.bindex, rdu->cookie.flags, + rdu->cookie.generation); + + if (rdu->verify[AufsCtlRduV_SZ] == sizeof(*rdu)) + return 0; + + AuDbg("%u:%u\n", + rdu->verify[AufsCtlRduV_SZ], (unsigned int)sizeof(*rdu)); + return -EINVAL; +} + +long au_rdu_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + long err, e; + struct aufs_rdu rdu; + void __user *p = (void __user *)arg; + + err = copy_from_user(&rdu, p, sizeof(rdu)); + if (unlikely(err)) { + err = -EFAULT; + AuTraceErr(err); + goto out; + } + err = au_rdu_verify(&rdu); + if (unlikely(err)) + goto out; + + switch (cmd) { + case AUFS_CTL_RDU: + err = au_rdu(file, &rdu); + if (unlikely(err)) + break; + + e = copy_to_user(p, &rdu, sizeof(rdu)); + if (unlikely(e)) { + err = -EFAULT; + AuTraceErr(err); + } + break; + case AUFS_CTL_RDU_INO: + err = au_rdu_ino(file, &rdu); + break; + + default: + /* err = -ENOTTY; */ + err = -EINVAL; + } + +out: + AuTraceErr(err); + return err; +} + +#ifdef CONFIG_COMPAT +long au_rdu_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + long err, e; + struct aufs_rdu rdu; + void __user *p = compat_ptr(arg); + + /* todo: get_user()? */ + err = copy_from_user(&rdu, p, sizeof(rdu)); + if (unlikely(err)) { + err = -EFAULT; + AuTraceErr(err); + goto out; + } + rdu.ent.e = compat_ptr(rdu.ent.ul); + err = au_rdu_verify(&rdu); + if (unlikely(err)) + goto out; + + switch (cmd) { + case AUFS_CTL_RDU: + err = au_rdu(file, &rdu); + if (unlikely(err)) + break; + + rdu.ent.ul = ptr_to_compat(rdu.ent.e); + rdu.tail.ul = ptr_to_compat(rdu.tail.e); + e = copy_to_user(p, &rdu, sizeof(rdu)); + if (unlikely(e)) { + err = -EFAULT; + AuTraceErr(err); + } + break; + case AUFS_CTL_RDU_INO: + err = au_rdu_ino(file, &rdu); + break; + + default: + /* err = -ENOTTY; */ + err = -EINVAL; + } + +out: + AuTraceErr(err); + return err; +} +#endif diff -Naur kernel.21.3.orig/fs/aufs/rwsem.h kernel.21.3.new/fs/aufs/rwsem.h --- kernel.21.3.orig/fs/aufs/rwsem.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/rwsem.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,191 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * simple read-write semaphore wrappers + */ + +#ifndef __AUFS_RWSEM_H__ +#define __AUFS_RWSEM_H__ + +#ifdef __KERNEL__ + +#include "debug.h" + +struct au_rwsem { + struct rw_semaphore rwsem; +#ifdef CONFIG_AUFS_DEBUG + /* just for debugging, not almighty counter */ + atomic_t rcnt, wcnt; +#endif +}; + +#ifdef CONFIG_AUFS_DEBUG +#define AuDbgCntInit(rw) do { \ + atomic_set(&(rw)->rcnt, 0); \ + atomic_set(&(rw)->wcnt, 0); \ + smp_mb(); /* atomic set */ \ +} while (0) + +#define AuDbgRcntInc(rw) atomic_inc(&(rw)->rcnt) +#define AuDbgRcntDec(rw) WARN_ON(atomic_dec_return(&(rw)->rcnt) < 0) +#define AuDbgWcntInc(rw) atomic_inc(&(rw)->wcnt) +#define AuDbgWcntDec(rw) WARN_ON(atomic_dec_return(&(rw)->wcnt) < 0) +#else +#define AuDbgCntInit(rw) do {} while (0) +#define AuDbgRcntInc(rw) do {} while (0) +#define AuDbgRcntDec(rw) do {} while (0) +#define AuDbgWcntInc(rw) do {} while (0) +#define AuDbgWcntDec(rw) do {} while (0) +#endif /* CONFIG_AUFS_DEBUG */ + +/* to debug easier, do not make them inlined functions */ +#define AuRwMustNoWaiters(rw) AuDebugOn(!list_empty(&(rw)->rwsem.wait_list)) +/* rwsem_is_locked() is unusable */ +#define AuRwMustReadLock(rw) AuDebugOn(atomic_read(&(rw)->rcnt) <= 0) +#define AuRwMustWriteLock(rw) AuDebugOn(atomic_read(&(rw)->wcnt) <= 0) +#define AuRwMustAnyLock(rw) AuDebugOn(atomic_read(&(rw)->rcnt) <= 0 \ + && atomic_read(&(rw)->wcnt) <= 0) +#define AuRwDestroy(rw) AuDebugOn(atomic_read(&(rw)->rcnt) \ + || atomic_read(&(rw)->wcnt)) + +#define au_rw_class(rw, key) lockdep_set_class(&(rw)->rwsem, key) + +static inline void au_rw_init(struct au_rwsem *rw) +{ + AuDbgCntInit(rw); + init_rwsem(&rw->rwsem); +} + +static inline void au_rw_init_wlock(struct au_rwsem *rw) +{ + au_rw_init(rw); + down_write(&rw->rwsem); + AuDbgWcntInc(rw); +} + +static inline void au_rw_init_wlock_nested(struct au_rwsem *rw, + unsigned int lsc) +{ + au_rw_init(rw); + down_write_nested(&rw->rwsem, lsc); + AuDbgWcntInc(rw); +} + +static inline void au_rw_read_lock(struct au_rwsem *rw) +{ + down_read(&rw->rwsem); + AuDbgRcntInc(rw); +} + +static inline void au_rw_read_lock_nested(struct au_rwsem *rw, unsigned int lsc) +{ + down_read_nested(&rw->rwsem, lsc); + AuDbgRcntInc(rw); +} + +static inline void au_rw_read_unlock(struct au_rwsem *rw) +{ + AuRwMustReadLock(rw); + AuDbgRcntDec(rw); + up_read(&rw->rwsem); +} + +static inline void au_rw_dgrade_lock(struct au_rwsem *rw) +{ + AuRwMustWriteLock(rw); + AuDbgRcntInc(rw); + AuDbgWcntDec(rw); + downgrade_write(&rw->rwsem); +} + +static inline void au_rw_write_lock(struct au_rwsem *rw) +{ + down_write(&rw->rwsem); + AuDbgWcntInc(rw); +} + +static inline void au_rw_write_lock_nested(struct au_rwsem *rw, + unsigned int lsc) +{ + down_write_nested(&rw->rwsem, lsc); + AuDbgWcntInc(rw); +} + +static inline void au_rw_write_unlock(struct au_rwsem *rw) +{ + AuRwMustWriteLock(rw); + AuDbgWcntDec(rw); + up_write(&rw->rwsem); +} + +/* why is not _nested version defined */ +static inline int au_rw_read_trylock(struct au_rwsem *rw) +{ + int ret; + + ret = down_read_trylock(&rw->rwsem); + if (ret) + AuDbgRcntInc(rw); + return ret; +} + +static inline int au_rw_write_trylock(struct au_rwsem *rw) +{ + int ret; + + ret = down_write_trylock(&rw->rwsem); + if (ret) + AuDbgWcntInc(rw); + return ret; +} + +#undef AuDbgCntInit +#undef AuDbgRcntInc +#undef AuDbgRcntDec +#undef AuDbgWcntInc +#undef AuDbgWcntDec + +#define AuSimpleLockRwsemFuncs(prefix, param, rwsem) \ +static inline void prefix##_read_lock(param) \ +{ au_rw_read_lock(rwsem); } \ +static inline void prefix##_write_lock(param) \ +{ au_rw_write_lock(rwsem); } \ +static inline int prefix##_read_trylock(param) \ +{ return au_rw_read_trylock(rwsem); } \ +static inline int prefix##_write_trylock(param) \ +{ return au_rw_write_trylock(rwsem); } +/* why is not _nested version defined */ +/* static inline void prefix##_read_trylock_nested(param, lsc) +{ au_rw_read_trylock_nested(rwsem, lsc)); } +static inline void prefix##_write_trylock_nestd(param, lsc) +{ au_rw_write_trylock_nested(rwsem, lsc); } */ + +#define AuSimpleUnlockRwsemFuncs(prefix, param, rwsem) \ +static inline void prefix##_read_unlock(param) \ +{ au_rw_read_unlock(rwsem); } \ +static inline void prefix##_write_unlock(param) \ +{ au_rw_write_unlock(rwsem); } \ +static inline void prefix##_downgrade_lock(param) \ +{ au_rw_dgrade_lock(rwsem); } + +#define AuSimpleRwsemFuncs(prefix, param, rwsem) \ + AuSimpleLockRwsemFuncs(prefix, param, rwsem) \ + AuSimpleUnlockRwsemFuncs(prefix, param, rwsem) + +#endif /* __KERNEL__ */ +#endif /* __AUFS_RWSEM_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/sbinfo.c kernel.21.3.new/fs/aufs/sbinfo.c --- kernel.21.3.orig/fs/aufs/sbinfo.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/sbinfo.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,351 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * superblock private data + */ + +#include "aufs.h" + +/* + * they are necessary regardless sysfs is disabled. + */ +void au_si_free(struct kobject *kobj) +{ + int i; + struct au_sbinfo *sbinfo; + char *locked __maybe_unused; /* debug only */ + + sbinfo = container_of(kobj, struct au_sbinfo, si_kobj); + for (i = 0; i < AuPlink_NHASH; i++) + AuDebugOn(!hlist_empty(&sbinfo->si_plink[i].head)); + AuDebugOn(atomic_read(&sbinfo->si_nowait.nw_len)); + + au_rw_write_lock(&sbinfo->si_rwsem); + au_br_free(sbinfo); + au_rw_write_unlock(&sbinfo->si_rwsem); + + AuDebugOn(radix_tree_gang_lookup + (&sbinfo->au_si_pid.tree, (void **)&locked, + /*first_index*/PID_MAX_DEFAULT - 1, + /*max_items*/sizeof(locked)/sizeof(*locked))); + + kfree(sbinfo->si_branch); + kfree(sbinfo->au_si_pid.bitmap); + mutex_destroy(&sbinfo->si_xib_mtx); + AuRwDestroy(&sbinfo->si_rwsem); + + kfree(sbinfo); +} + +int au_si_alloc(struct super_block *sb) +{ + int err, i; + struct au_sbinfo *sbinfo; + static struct lock_class_key aufs_si; + + err = -ENOMEM; + sbinfo = kzalloc(sizeof(*sbinfo), GFP_NOFS); + if (unlikely(!sbinfo)) + goto out; + + BUILD_BUG_ON(sizeof(unsigned long) != + sizeof(*sbinfo->au_si_pid.bitmap)); + sbinfo->au_si_pid.bitmap = kcalloc(BITS_TO_LONGS(PID_MAX_DEFAULT), + sizeof(*sbinfo->au_si_pid.bitmap), + GFP_NOFS); + if (unlikely(!sbinfo->au_si_pid.bitmap)) + goto out_sbinfo; + + /* will be reallocated separately */ + sbinfo->si_branch = kzalloc(sizeof(*sbinfo->si_branch), GFP_NOFS); + if (unlikely(!sbinfo->si_branch)) + goto out_pidmap; + + err = sysaufs_si_init(sbinfo); + if (unlikely(err)) + goto out_br; + + au_nwt_init(&sbinfo->si_nowait); + au_rw_init_wlock(&sbinfo->si_rwsem); + au_rw_class(&sbinfo->si_rwsem, &aufs_si); + spin_lock_init(&sbinfo->au_si_pid.tree_lock); + INIT_RADIX_TREE(&sbinfo->au_si_pid.tree, GFP_ATOMIC | __GFP_NOFAIL); + + atomic_long_set(&sbinfo->si_ninodes, 0); + atomic_long_set(&sbinfo->si_nfiles, 0); + + sbinfo->si_bend = -1; + sbinfo->si_last_br_id = AUFS_BRANCH_MAX / 2; + + sbinfo->si_wbr_copyup = AuWbrCopyup_Def; + sbinfo->si_wbr_create = AuWbrCreate_Def; + sbinfo->si_wbr_copyup_ops = au_wbr_copyup_ops + sbinfo->si_wbr_copyup; + sbinfo->si_wbr_create_ops = au_wbr_create_ops + sbinfo->si_wbr_create; + + au_fhsm_init(sbinfo); + + sbinfo->si_mntflags = au_opts_plink(AuOpt_Def); + + sbinfo->si_xino_jiffy = jiffies; + sbinfo->si_xino_expire + = msecs_to_jiffies(AUFS_XINO_DEF_SEC * MSEC_PER_SEC); + mutex_init(&sbinfo->si_xib_mtx); + sbinfo->si_xino_brid = -1; + /* leave si_xib_last_pindex and si_xib_next_bit */ + + sbinfo->si_rdcache = msecs_to_jiffies(AUFS_RDCACHE_DEF * MSEC_PER_SEC); + sbinfo->si_rdblk = AUFS_RDBLK_DEF; + sbinfo->si_rdhash = AUFS_RDHASH_DEF; + sbinfo->si_dirwh = AUFS_DIRWH_DEF; + + for (i = 0; i < AuPlink_NHASH; i++) + au_sphl_init(sbinfo->si_plink + i); + init_waitqueue_head(&sbinfo->si_plink_wq); + spin_lock_init(&sbinfo->si_plink_maint_lock); + + /* leave other members for sysaufs and si_mnt. */ + sbinfo->si_sb = sb; + sb->s_fs_info = sbinfo; + si_pid_set(sb); + au_debug_sbinfo_init(sbinfo); + return 0; /* success */ + +out_br: + kfree(sbinfo->si_branch); +out_pidmap: + kfree(sbinfo->au_si_pid.bitmap); +out_sbinfo: + kfree(sbinfo); +out: + return err; +} + +int au_sbr_realloc(struct au_sbinfo *sbinfo, int nbr) +{ + int err, sz; + struct au_branch **brp; + + AuRwMustWriteLock(&sbinfo->si_rwsem); + + err = -ENOMEM; + sz = sizeof(*brp) * (sbinfo->si_bend + 1); + if (unlikely(!sz)) + sz = sizeof(*brp); + brp = au_kzrealloc(sbinfo->si_branch, sz, sizeof(*brp) * nbr, GFP_NOFS); + if (brp) { + sbinfo->si_branch = brp; + err = 0; + } + + return err; +} + +/* ---------------------------------------------------------------------- */ + +unsigned int au_sigen_inc(struct super_block *sb) +{ + unsigned int gen; + + SiMustWriteLock(sb); + + gen = ++au_sbi(sb)->si_generation; + au_update_digen(sb->s_root); + au_update_iigen(sb->s_root->d_inode, /*half*/0); + sb->s_root->d_inode->i_version++; + return gen; +} + +aufs_bindex_t au_new_br_id(struct super_block *sb) +{ + aufs_bindex_t br_id; + int i; + struct au_sbinfo *sbinfo; + + SiMustWriteLock(sb); + + sbinfo = au_sbi(sb); + for (i = 0; i <= AUFS_BRANCH_MAX; i++) { + br_id = ++sbinfo->si_last_br_id; + AuDebugOn(br_id < 0); + if (br_id && au_br_index(sb, br_id) < 0) + return br_id; + } + + return -1; +} + +/* ---------------------------------------------------------------------- */ + +/* it is ok that new 'nwt' tasks are appended while we are sleeping */ +int si_read_lock(struct super_block *sb, int flags) +{ + int err; + + err = 0; + if (au_ftest_lock(flags, FLUSH)) + au_nwt_flush(&au_sbi(sb)->si_nowait); + + si_noflush_read_lock(sb); + err = au_plink_maint(sb, flags); + if (unlikely(err)) + si_read_unlock(sb); + + return err; +} + +int si_write_lock(struct super_block *sb, int flags) +{ + int err; + + if (au_ftest_lock(flags, FLUSH)) + au_nwt_flush(&au_sbi(sb)->si_nowait); + + si_noflush_write_lock(sb); + err = au_plink_maint(sb, flags); + if (unlikely(err)) + si_write_unlock(sb); + + return err; +} + +/* dentry and super_block lock. call at entry point */ +int aufs_read_lock(struct dentry *dentry, int flags) +{ + int err; + struct super_block *sb; + + sb = dentry->d_sb; + err = si_read_lock(sb, flags); + if (unlikely(err)) + goto out; + + if (au_ftest_lock(flags, DW)) + di_write_lock_child(dentry); + else + di_read_lock_child(dentry, flags); + + if (au_ftest_lock(flags, GEN)) { + err = au_digen_test(dentry, au_sigen(sb)); + AuDebugOn(!err && au_dbrange_test(dentry)); + if (unlikely(err)) + aufs_read_unlock(dentry, flags); + } + +out: + return err; +} + +void aufs_read_unlock(struct dentry *dentry, int flags) +{ + if (au_ftest_lock(flags, DW)) + di_write_unlock(dentry); + else + di_read_unlock(dentry, flags); + si_read_unlock(dentry->d_sb); +} + +void aufs_write_lock(struct dentry *dentry) +{ + si_write_lock(dentry->d_sb, AuLock_FLUSH | AuLock_NOPLMW); + di_write_lock_child(dentry); +} + +void aufs_write_unlock(struct dentry *dentry) +{ + di_write_unlock(dentry); + si_write_unlock(dentry->d_sb); +} + +int aufs_read_and_write_lock2(struct dentry *d1, struct dentry *d2, int flags) +{ + int err; + unsigned int sigen; + struct super_block *sb; + + sb = d1->d_sb; + err = si_read_lock(sb, flags); + if (unlikely(err)) + goto out; + + di_write_lock2_child(d1, d2, au_ftest_lock(flags, DIR)); + + if (au_ftest_lock(flags, GEN)) { + sigen = au_sigen(sb); + err = au_digen_test(d1, sigen); + AuDebugOn(!err && au_dbrange_test(d1)); + if (!err) { + err = au_digen_test(d2, sigen); + AuDebugOn(!err && au_dbrange_test(d2)); + } + if (unlikely(err)) + aufs_read_and_write_unlock2(d1, d2); + } + +out: + return err; +} + +void aufs_read_and_write_unlock2(struct dentry *d1, struct dentry *d2) +{ + di_write_unlock2(d1, d2); + si_read_unlock(d1->d_sb); +} + +/* ---------------------------------------------------------------------- */ + +int si_pid_test_slow(struct super_block *sb) +{ + void *p; + + rcu_read_lock(); + p = radix_tree_lookup(&au_sbi(sb)->au_si_pid.tree, current->pid); + rcu_read_unlock(); + + return (long)!!p; +} + +void si_pid_set_slow(struct super_block *sb) +{ + int err; + struct au_sbinfo *sbinfo; + + AuDebugOn(si_pid_test_slow(sb)); + + sbinfo = au_sbi(sb); + err = radix_tree_preload(GFP_NOFS | __GFP_NOFAIL); + AuDebugOn(err); + spin_lock(&sbinfo->au_si_pid.tree_lock); + err = radix_tree_insert(&sbinfo->au_si_pid.tree, current->pid, + /*any valid ptr*/sb); + spin_unlock(&sbinfo->au_si_pid.tree_lock); + AuDebugOn(err); + radix_tree_preload_end(); +} + +void si_pid_clr_slow(struct super_block *sb) +{ + void *p; + struct au_sbinfo *sbinfo; + + AuDebugOn(!si_pid_test_slow(sb)); + + sbinfo = au_sbi(sb); + spin_lock(&sbinfo->au_si_pid.tree_lock); + p = radix_tree_delete(&sbinfo->au_si_pid.tree, current->pid); + spin_unlock(&sbinfo->au_si_pid.tree_lock); +} diff -Naur kernel.21.3.orig/fs/aufs/spl.h kernel.21.3.new/fs/aufs/spl.h --- kernel.21.3.orig/fs/aufs/spl.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/spl.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,111 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * simple list protected by a spinlock + */ + +#ifndef __AUFS_SPL_H__ +#define __AUFS_SPL_H__ + +#ifdef __KERNEL__ + +struct au_splhead { + spinlock_t spin; + struct list_head head; +}; + +static inline void au_spl_init(struct au_splhead *spl) +{ + spin_lock_init(&spl->spin); + INIT_LIST_HEAD(&spl->head); +} + +static inline void au_spl_add(struct list_head *list, struct au_splhead *spl) +{ + spin_lock(&spl->spin); + list_add(list, &spl->head); + spin_unlock(&spl->spin); +} + +static inline void au_spl_del(struct list_head *list, struct au_splhead *spl) +{ + spin_lock(&spl->spin); + list_del(list); + spin_unlock(&spl->spin); +} + +static inline void au_spl_del_rcu(struct list_head *list, + struct au_splhead *spl) +{ + spin_lock(&spl->spin); + list_del_rcu(list); + spin_unlock(&spl->spin); +} + +/* ---------------------------------------------------------------------- */ + +struct au_sphlhead { + spinlock_t spin; + struct hlist_head head; +}; + +static inline void au_sphl_init(struct au_sphlhead *sphl) +{ + spin_lock_init(&sphl->spin); + INIT_HLIST_HEAD(&sphl->head); +} + +static inline void au_sphl_add(struct hlist_node *hlist, + struct au_sphlhead *sphl) +{ + spin_lock(&sphl->spin); + hlist_add_head(hlist, &sphl->head); + spin_unlock(&sphl->spin); +} + +static inline void au_sphl_del(struct hlist_node *hlist, + struct au_sphlhead *sphl) +{ + spin_lock(&sphl->spin); + hlist_del(hlist); + spin_unlock(&sphl->spin); +} + +static inline void au_sphl_del_rcu(struct hlist_node *hlist, + struct au_sphlhead *sphl) +{ + spin_lock(&sphl->spin); + hlist_del_rcu(hlist); + spin_unlock(&sphl->spin); +} + +static inline unsigned long au_sphl_count(struct au_sphlhead *sphl) +{ + unsigned long cnt; + struct hlist_node *pos; + + cnt = 0; + spin_lock(&sphl->spin); + hlist_for_each(pos, &sphl->head) + cnt++; + spin_unlock(&sphl->spin); + return cnt; +} + +#endif /* __KERNEL__ */ +#endif /* __AUFS_SPL_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/super.c kernel.21.3.new/fs/aufs/super.c --- kernel.21.3.orig/fs/aufs/super.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/super.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,1008 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * mount and super_block operations + */ + +#include +#include +#include +#include +#include +#include +#include "aufs.h" + +/* + * super_operations + */ +static struct inode *aufs_alloc_inode(struct super_block *sb __maybe_unused) +{ + struct au_icntnr *c; + + c = au_cache_alloc_icntnr(); + if (c) { + au_icntnr_init(c); + c->vfs_inode.i_version = 1; /* sigen(sb); */ + c->iinfo.ii_hinode = NULL; + return &c->vfs_inode; + } + return NULL; +} + +static void aufs_destroy_inode_cb(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + + INIT_HLIST_HEAD(&inode->i_dentry); + au_cache_free_icntnr(container_of(inode, struct au_icntnr, vfs_inode)); +} + +static void aufs_destroy_inode(struct inode *inode) +{ + au_iinfo_fin(inode); + call_rcu(&inode->i_rcu, aufs_destroy_inode_cb); +} + +struct inode *au_iget_locked(struct super_block *sb, ino_t ino) +{ + struct inode *inode; + int err; + + inode = iget_locked(sb, ino); + if (unlikely(!inode)) { + inode = ERR_PTR(-ENOMEM); + goto out; + } + if (!(inode->i_state & I_NEW)) + goto out; + + err = au_xigen_new(inode); + if (!err) + err = au_iinfo_init(inode); + if (!err) + inode->i_version++; + else { + iget_failed(inode); + inode = ERR_PTR(err); + } + +out: + /* never return NULL */ + AuDebugOn(!inode); + AuTraceErrPtr(inode); + return inode; +} + +/* lock free root dinfo */ +static int au_show_brs(struct seq_file *seq, struct super_block *sb) +{ + int err; + aufs_bindex_t bindex, bend; + struct path path; + struct au_hdentry *hdp; + struct au_branch *br; + au_br_perm_str_t perm; + + err = 0; + bend = au_sbend(sb); + hdp = au_di(sb->s_root)->di_hdentry; + for (bindex = 0; !err && bindex <= bend; bindex++) { + br = au_sbr(sb, bindex); + path.mnt = au_br_mnt(br); + path.dentry = hdp[bindex].hd_dentry; + err = au_seq_path(seq, &path); + if (err > 0) { + au_optstr_br_perm(&perm, br->br_perm); + err = seq_printf(seq, "=%s", perm.a); + if (err == -1) + err = -E2BIG; + } + if (!err && bindex != bend) + err = seq_putc(seq, ':'); + } + + return err; +} + +static void au_show_wbr_create(struct seq_file *m, int v, + struct au_sbinfo *sbinfo) +{ + const char *pat; + + AuRwMustAnyLock(&sbinfo->si_rwsem); + + seq_puts(m, ",create="); + pat = au_optstr_wbr_create(v); + switch (v) { + case AuWbrCreate_TDP: + case AuWbrCreate_RR: + case AuWbrCreate_MFS: + case AuWbrCreate_PMFS: + seq_puts(m, pat); + break; + case AuWbrCreate_MFSV: + seq_printf(m, /*pat*/"mfs:%lu", + jiffies_to_msecs(sbinfo->si_wbr_mfs.mfs_expire) + / MSEC_PER_SEC); + break; + case AuWbrCreate_PMFSV: + seq_printf(m, /*pat*/"pmfs:%lu", + jiffies_to_msecs(sbinfo->si_wbr_mfs.mfs_expire) + / MSEC_PER_SEC); + break; + case AuWbrCreate_MFSRR: + seq_printf(m, /*pat*/"mfsrr:%llu", + sbinfo->si_wbr_mfs.mfsrr_watermark); + break; + case AuWbrCreate_MFSRRV: + seq_printf(m, /*pat*/"mfsrr:%llu:%lu", + sbinfo->si_wbr_mfs.mfsrr_watermark, + jiffies_to_msecs(sbinfo->si_wbr_mfs.mfs_expire) + / MSEC_PER_SEC); + break; + case AuWbrCreate_PMFSRR: + seq_printf(m, /*pat*/"pmfsrr:%llu", + sbinfo->si_wbr_mfs.mfsrr_watermark); + break; + case AuWbrCreate_PMFSRRV: + seq_printf(m, /*pat*/"pmfsrr:%llu:%lu", + sbinfo->si_wbr_mfs.mfsrr_watermark, + jiffies_to_msecs(sbinfo->si_wbr_mfs.mfs_expire) + / MSEC_PER_SEC); + break; + } +} + +static int au_show_xino(struct seq_file *seq, struct super_block *sb) +{ +#ifdef CONFIG_SYSFS + return 0; +#else + int err; + const int len = sizeof(AUFS_XINO_FNAME) - 1; + aufs_bindex_t bindex, brid; + struct qstr *name; + struct file *f; + struct dentry *d, *h_root; + struct au_hdentry *hdp; + + AuRwMustAnyLock(&sbinfo->si_rwsem); + + err = 0; + f = au_sbi(sb)->si_xib; + if (!f) + goto out; + + /* stop printing the default xino path on the first writable branch */ + h_root = NULL; + brid = au_xino_brid(sb); + if (brid >= 0) { + bindex = au_br_index(sb, brid); + hdp = au_di(sb->s_root)->di_hdentry; + h_root = hdp[0 + bindex].hd_dentry; + } + d = f->f_dentry; + name = &d->d_name; + /* safe ->d_parent because the file is unlinked */ + if (d->d_parent == h_root + && name->len == len + && !memcmp(name->name, AUFS_XINO_FNAME, len)) + goto out; + + seq_puts(seq, ",xino="); + err = au_xino_path(seq, f); + +out: + return err; +#endif +} + +/* seq_file will re-call me in case of too long string */ +static int aufs_show_options(struct seq_file *m, struct dentry *dentry) +{ + int err; + unsigned int mnt_flags, v; + struct super_block *sb; + struct au_sbinfo *sbinfo; + +#define AuBool(name, str) do { \ + v = au_opt_test(mnt_flags, name); \ + if (v != au_opt_test(AuOpt_Def, name)) \ + seq_printf(m, ",%s" #str, v ? "" : "no"); \ +} while (0) + +#define AuStr(name, str) do { \ + v = mnt_flags & AuOptMask_##name; \ + if (v != (AuOpt_Def & AuOptMask_##name)) \ + seq_printf(m, "," #str "=%s", au_optstr_##str(v)); \ +} while (0) + +#define AuUInt(name, str, val) do { \ + if (val != AUFS_##name##_DEF) \ + seq_printf(m, "," #str "=%u", val); \ +} while (0) + + sb = dentry->d_sb; + if (sb->s_flags & MS_POSIXACL) + seq_puts(m, ",acl"); + + /* lock free root dinfo */ + si_noflush_read_lock(sb); + sbinfo = au_sbi(sb); + seq_printf(m, ",si=%lx", sysaufs_si_id(sbinfo)); + + mnt_flags = au_mntflags(sb); + if (au_opt_test(mnt_flags, XINO)) { + err = au_show_xino(m, sb); + if (unlikely(err)) + goto out; + } else + seq_puts(m, ",noxino"); + + AuBool(TRUNC_XINO, trunc_xino); + AuStr(UDBA, udba); + AuBool(SHWH, shwh); + AuBool(PLINK, plink); + AuBool(DIO, dio); + AuBool(DIRPERM1, dirperm1); + /* AuBool(REFROF, refrof); */ + + v = sbinfo->si_wbr_create; + if (v != AuWbrCreate_Def) + au_show_wbr_create(m, v, sbinfo); + + v = sbinfo->si_wbr_copyup; + if (v != AuWbrCopyup_Def) + seq_printf(m, ",cpup=%s", au_optstr_wbr_copyup(v)); + + v = au_opt_test(mnt_flags, ALWAYS_DIROPQ); + if (v != au_opt_test(AuOpt_Def, ALWAYS_DIROPQ)) + seq_printf(m, ",diropq=%c", v ? 'a' : 'w'); + + AuUInt(DIRWH, dirwh, sbinfo->si_dirwh); + + v = jiffies_to_msecs(sbinfo->si_rdcache) / MSEC_PER_SEC; + AuUInt(RDCACHE, rdcache, v); + + AuUInt(RDBLK, rdblk, sbinfo->si_rdblk); + AuUInt(RDHASH, rdhash, sbinfo->si_rdhash); + + au_fhsm_show(m, sbinfo); + + AuBool(SUM, sum); + /* AuBool(SUM_W, wsum); */ + AuBool(WARN_PERM, warn_perm); + AuBool(VERBOSE, verbose); + +out: + /* be sure to print "br:" last */ + if (!sysaufs_brs) { + seq_puts(m, ",br:"); + au_show_brs(m, sb); + } + si_read_unlock(sb); + return 0; + +#undef AuBool +#undef AuStr +#undef AuUInt +} + +/* ---------------------------------------------------------------------- */ + +/* sum mode which returns the summation for statfs(2) */ + +static u64 au_add_till_max(u64 a, u64 b) +{ + u64 old; + + old = a; + a += b; + if (old <= a) + return a; + return ULLONG_MAX; +} + +static u64 au_mul_till_max(u64 a, long mul) +{ + u64 old; + + old = a; + a *= mul; + if (old <= a) + return a; + return ULLONG_MAX; +} + +static int au_statfs_sum(struct super_block *sb, struct kstatfs *buf) +{ + int err; + long bsize, factor; + u64 blocks, bfree, bavail, files, ffree; + aufs_bindex_t bend, bindex, i; + unsigned char shared; + struct path h_path; + struct super_block *h_sb; + + err = 0; + bsize = LONG_MAX; + files = 0; + ffree = 0; + blocks = 0; + bfree = 0; + bavail = 0; + bend = au_sbend(sb); + for (bindex = 0; bindex <= bend; bindex++) { + h_path.mnt = au_sbr_mnt(sb, bindex); + h_sb = h_path.mnt->mnt_sb; + shared = 0; + for (i = 0; !shared && i < bindex; i++) + shared = (au_sbr_sb(sb, i) == h_sb); + if (shared) + continue; + + /* sb->s_root for NFS is unreliable */ + h_path.dentry = h_path.mnt->mnt_root; + err = vfs_statfs(&h_path, buf); + if (unlikely(err)) + goto out; + + if (bsize > buf->f_bsize) { + /* + * we will reduce bsize, so we have to expand blocks + * etc. to match them again + */ + factor = (bsize / buf->f_bsize); + blocks = au_mul_till_max(blocks, factor); + bfree = au_mul_till_max(bfree, factor); + bavail = au_mul_till_max(bavail, factor); + bsize = buf->f_bsize; + } + + factor = (buf->f_bsize / bsize); + blocks = au_add_till_max(blocks, + au_mul_till_max(buf->f_blocks, factor)); + bfree = au_add_till_max(bfree, + au_mul_till_max(buf->f_bfree, factor)); + bavail = au_add_till_max(bavail, + au_mul_till_max(buf->f_bavail, factor)); + files = au_add_till_max(files, buf->f_files); + ffree = au_add_till_max(ffree, buf->f_ffree); + } + + buf->f_bsize = bsize; + buf->f_blocks = blocks; + buf->f_bfree = bfree; + buf->f_bavail = bavail; + buf->f_files = files; + buf->f_ffree = ffree; + buf->f_frsize = 0; + +out: + return err; +} + +static int aufs_statfs(struct dentry *dentry, struct kstatfs *buf) +{ + int err; + struct path h_path; + struct super_block *sb; + + /* lock free root dinfo */ + sb = dentry->d_sb; + si_noflush_read_lock(sb); + if (!au_opt_test(au_mntflags(sb), SUM)) { + /* sb->s_root for NFS is unreliable */ + h_path.mnt = au_sbr_mnt(sb, 0); + h_path.dentry = h_path.mnt->mnt_root; + err = vfs_statfs(&h_path, buf); + } else + err = au_statfs_sum(sb, buf); + si_read_unlock(sb); + + if (!err) { + buf->f_type = AUFS_SUPER_MAGIC; + buf->f_namelen = AUFS_MAX_NAMELEN; + memset(&buf->f_fsid, 0, sizeof(buf->f_fsid)); + } + /* buf->f_bsize = buf->f_blocks = buf->f_bfree = buf->f_bavail = -1; */ + + return err; +} + +/* ---------------------------------------------------------------------- */ + +static int aufs_sync_fs(struct super_block *sb, int wait) +{ + int err, e; + aufs_bindex_t bend, bindex; + struct au_branch *br; + struct super_block *h_sb; + + err = 0; + si_noflush_read_lock(sb); + bend = au_sbend(sb); + for (bindex = 0; bindex <= bend; bindex++) { + br = au_sbr(sb, bindex); + if (!au_br_writable(br->br_perm)) + continue; + + h_sb = au_sbr_sb(sb, bindex); + if (h_sb->s_op->sync_fs) { + e = h_sb->s_op->sync_fs(h_sb, wait); + if (unlikely(e && !err)) + err = e; + /* go on even if an error happens */ + } + } + si_read_unlock(sb); + + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* final actions when unmounting a file system */ +static void aufs_put_super(struct super_block *sb) +{ + struct au_sbinfo *sbinfo; + + sbinfo = au_sbi(sb); + if (!sbinfo) + return; + + dbgaufs_si_fin(sbinfo); + kobject_put(&sbinfo->si_kobj); +} + +/* ---------------------------------------------------------------------- */ + +void au_array_free(void *array) +{ + if (array) { + if (!is_vmalloc_addr(array)) + kfree(array); + else + vfree(array); + } +} + +void *au_array_alloc(unsigned long long *hint, au_arraycb_t cb, void *arg) +{ + void *array; + unsigned long long n, sz; + + array = NULL; + n = 0; + if (!*hint) + goto out; + + if (*hint > ULLONG_MAX / sizeof(array)) { + array = ERR_PTR(-EMFILE); + pr_err("hint %llu\n", *hint); + goto out; + } + + sz = sizeof(array) * *hint; + array = kzalloc(sz, GFP_NOFS); + if (unlikely(!array)) + array = vzalloc(sz); + if (unlikely(!array)) { + array = ERR_PTR(-ENOMEM); + goto out; + } + + n = cb(array, *hint, arg); + AuDebugOn(n > *hint); + +out: + *hint = n; + return array; +} + +static unsigned long long au_iarray_cb(void *a, + unsigned long long max __maybe_unused, + void *arg) +{ + unsigned long long n; + struct inode **p, *inode; + struct list_head *head; + + n = 0; + p = a; + head = arg; + spin_lock(&inode_sb_list_lock); + list_for_each_entry(inode, head, i_sb_list) { + if (!is_bad_inode(inode) + && au_ii(inode)->ii_bstart >= 0) { + spin_lock(&inode->i_lock); + if (atomic_read(&inode->i_count)) { + au_igrab(inode); + *p++ = inode; + n++; + AuDebugOn(n > max); + } + spin_unlock(&inode->i_lock); + } + } + spin_unlock(&inode_sb_list_lock); + + return n; +} + +struct inode **au_iarray_alloc(struct super_block *sb, unsigned long long *max) +{ + *max = atomic_long_read(&au_sbi(sb)->si_ninodes); + return au_array_alloc(max, au_iarray_cb, &sb->s_inodes); +} + +void au_iarray_free(struct inode **a, unsigned long long max) +{ + unsigned long long ull; + + for (ull = 0; ull < max; ull++) + iput(a[ull]); + au_array_free(a); +} + +/* ---------------------------------------------------------------------- */ + +/* + * refresh dentry and inode at remount time. + */ +/* todo: consolidate with simple_reval_dpath() and au_reval_for_attr() */ +static int au_do_refresh(struct dentry *dentry, unsigned int dir_flags, + struct dentry *parent) +{ + int err; + + di_write_lock_child(dentry); + di_read_lock_parent(parent, AuLock_IR); + err = au_refresh_dentry(dentry, parent); + if (!err && dir_flags) + au_hn_reset(dentry->d_inode, dir_flags); + di_read_unlock(parent, AuLock_IR); + di_write_unlock(dentry); + + return err; +} + +static int au_do_refresh_d(struct dentry *dentry, unsigned int sigen, + struct au_sbinfo *sbinfo, + const unsigned int dir_flags) +{ + int err; + struct dentry *parent; + struct inode *inode; + + err = 0; + parent = dget_parent(dentry); + if (!au_digen_test(parent, sigen) && au_digen_test(dentry, sigen)) { + inode = dentry->d_inode; + if (inode) { + if (!S_ISDIR(inode->i_mode)) + err = au_do_refresh(dentry, /*dir_flags*/0, + parent); + else { + err = au_do_refresh(dentry, dir_flags, parent); + if (unlikely(err)) + au_fset_si(sbinfo, FAILED_REFRESH_DIR); + } + } else + err = au_do_refresh(dentry, /*dir_flags*/0, parent); + AuDbgDentry(dentry); + } + dput(parent); + + AuTraceErr(err); + return err; +} + +static int au_refresh_d(struct super_block *sb) +{ + int err, i, j, ndentry, e; + unsigned int sigen; + struct au_dcsub_pages dpages; + struct au_dpage *dpage; + struct dentry **dentries, *d; + struct au_sbinfo *sbinfo; + struct dentry *root = sb->s_root; + const unsigned int dir_flags = au_hi_flags(root->d_inode, /*isdir*/1); + + err = au_dpages_init(&dpages, GFP_NOFS); + if (unlikely(err)) + goto out; + err = au_dcsub_pages(&dpages, root, NULL, NULL); + if (unlikely(err)) + goto out_dpages; + + sigen = au_sigen(sb); + sbinfo = au_sbi(sb); + for (i = 0; i < dpages.ndpage; i++) { + dpage = dpages.dpages + i; + dentries = dpage->dentries; + ndentry = dpage->ndentry; + for (j = 0; j < ndentry; j++) { + d = dentries[j]; + e = au_do_refresh_d(d, sigen, sbinfo, dir_flags); + if (unlikely(e && !err)) + err = e; + /* go on even err */ + } + } + +out_dpages: + au_dpages_free(&dpages); +out: + return err; +} + +static int au_refresh_i(struct super_block *sb) +{ + int err, e; + unsigned int sigen; + unsigned long long max, ull; + struct inode *inode, **array; + + array = au_iarray_alloc(sb, &max); + err = PTR_ERR(array); + if (IS_ERR(array)) + goto out; + + err = 0; + sigen = au_sigen(sb); + for (ull = 0; ull < max; ull++) { + inode = array[ull]; + if (unlikely(!inode)) + break; + if (au_iigen(inode, NULL) != sigen) { + ii_write_lock_child(inode); + e = au_refresh_hinode_self(inode); + ii_write_unlock(inode); + if (unlikely(e)) { + pr_err("error %d, i%lu\n", e, inode->i_ino); + if (!err) + err = e; + /* go on even if err */ + } + } + } + + au_iarray_free(array, max); + +out: + return err; +} + +static void au_remount_refresh(struct super_block *sb) +{ + int err, e; + unsigned int udba; + aufs_bindex_t bindex, bend; + struct dentry *root; + struct inode *inode; + struct au_branch *br; + + au_sigen_inc(sb); + au_fclr_si(au_sbi(sb), FAILED_REFRESH_DIR); + + root = sb->s_root; + DiMustNoWaiters(root); + inode = root->d_inode; + IiMustNoWaiters(inode); + + udba = au_opt_udba(sb); + bend = au_sbend(sb); + for (bindex = 0; bindex <= bend; bindex++) { + br = au_sbr(sb, bindex); + err = au_hnotify_reset_br(udba, br, br->br_perm); + if (unlikely(err)) + AuIOErr("hnotify failed on br %d, %d, ignored\n", + bindex, err); + /* go on even if err */ + } + au_hn_reset(inode, au_hi_flags(inode, /*isdir*/1)); + + di_write_unlock(root); + err = au_refresh_d(sb); + e = au_refresh_i(sb); + if (unlikely(e && !err)) + err = e; + /* aufs_write_lock() calls ..._child() */ + di_write_lock_child(root); + + au_cpup_attr_all(inode, /*force*/1); + + if (unlikely(err)) + AuIOErr("refresh failed, ignored, %d\n", err); +} + +/* stop extra interpretation of errno in mount(8), and strange error messages */ +static int cvt_err(int err) +{ + AuTraceErr(err); + + switch (err) { + case -ENOENT: + case -ENOTDIR: + case -EEXIST: + case -EIO: + err = -EINVAL; + } + return err; +} + +static int aufs_remount_fs(struct super_block *sb, int *flags, char *data) +{ + int err, do_dx; + unsigned int mntflags; + struct au_opts opts; + struct dentry *root; + struct inode *inode; + struct au_sbinfo *sbinfo; + + err = 0; + root = sb->s_root; + if (!data || !*data) { + err = si_write_lock(sb, AuLock_FLUSH | AuLock_NOPLM); + if (!err) { + di_write_lock_child(root); + err = au_opts_verify(sb, *flags, /*pending*/0); + aufs_write_unlock(root); + } + goto out; + } + + err = -ENOMEM; + memset(&opts, 0, sizeof(opts)); + opts.opt = (void *)__get_free_page(GFP_NOFS); + if (unlikely(!opts.opt)) + goto out; + opts.max_opt = PAGE_SIZE / sizeof(*opts.opt); + opts.flags = AuOpts_REMOUNT; + opts.sb_flags = *flags; + + /* parse it before aufs lock */ + err = au_opts_parse(sb, data, &opts); + if (unlikely(err)) + goto out_opts; + + sbinfo = au_sbi(sb); + inode = root->d_inode; + mutex_lock(&inode->i_mutex); + err = si_write_lock(sb, AuLock_FLUSH | AuLock_NOPLM); + if (unlikely(err)) + goto out_mtx; + di_write_lock_child(root); + + /* au_opts_remount() may return an error */ + err = au_opts_remount(sb, &opts); + au_opts_free(&opts); + + if (au_ftest_opts(opts.flags, REFRESH)) + au_remount_refresh(sb); + + if (au_ftest_opts(opts.flags, REFRESH_DYAOP)) { + mntflags = au_mntflags(sb); + do_dx = !!au_opt_test(mntflags, DIO); + au_dy_arefresh(do_dx); + } + + au_fhsm_wrote_all(sb, /*force*/1); /* ?? */ + aufs_write_unlock(root); + +out_mtx: + mutex_unlock(&inode->i_mutex); +out_opts: + free_page((unsigned long)opts.opt); +out: + err = cvt_err(err); + AuTraceErr(err); + return err; +} + +static const struct super_operations aufs_sop = { + .alloc_inode = aufs_alloc_inode, + .destroy_inode = aufs_destroy_inode, + /* always deleting, no clearing */ + .drop_inode = generic_delete_inode, + .show_options = aufs_show_options, + .statfs = aufs_statfs, + .put_super = aufs_put_super, + .sync_fs = aufs_sync_fs, + .remount_fs = aufs_remount_fs +}; + +/* ---------------------------------------------------------------------- */ + +static int alloc_root(struct super_block *sb) +{ + int err; + struct inode *inode; + struct dentry *root; + + err = -ENOMEM; + inode = au_iget_locked(sb, AUFS_ROOT_INO); + err = PTR_ERR(inode); + if (IS_ERR(inode)) + goto out; + + inode->i_op = &aufs_dir_iop; + inode->i_fop = &aufs_dir_fop; + inode->i_mode = S_IFDIR; + set_nlink(inode, 2); + unlock_new_inode(inode); + + root = d_make_root(inode); + if (unlikely(!root)) + goto out; + err = PTR_ERR(root); + if (IS_ERR(root)) + goto out; + + err = au_di_init(root); + if (!err) { + sb->s_root = root; + return 0; /* success */ + } + dput(root); + +out: + return err; +} + +static int aufs_fill_super(struct super_block *sb, void *raw_data, + int silent __maybe_unused) +{ + int err; + struct au_opts opts; + struct dentry *root; + struct inode *inode; + char *arg = raw_data; + + if (unlikely(!arg || !*arg)) { + err = -EINVAL; + pr_err("no arg\n"); + goto out; + } + + err = -ENOMEM; + memset(&opts, 0, sizeof(opts)); + opts.opt = (void *)__get_free_page(GFP_NOFS); + if (unlikely(!opts.opt)) + goto out; + opts.max_opt = PAGE_SIZE / sizeof(*opts.opt); + opts.sb_flags = sb->s_flags; + + err = au_si_alloc(sb); + if (unlikely(err)) + goto out_opts; + + /* all timestamps always follow the ones on the branch */ + sb->s_flags |= MS_NOATIME | MS_NODIRATIME; + sb->s_op = &aufs_sop; + sb->s_d_op = &aufs_dop; + sb->s_magic = AUFS_SUPER_MAGIC; + sb->s_maxbytes = 0; + au_export_init(sb); + /* au_xattr_init(sb); */ + + err = alloc_root(sb); + if (unlikely(err)) { + si_write_unlock(sb); + goto out_info; + } + root = sb->s_root; + inode = root->d_inode; + + /* + * actually we can parse options regardless aufs lock here. + * but at remount time, parsing must be done before aufs lock. + * so we follow the same rule. + */ + ii_write_lock_parent(inode); + aufs_write_unlock(root); + err = au_opts_parse(sb, arg, &opts); + if (unlikely(err)) + goto out_root; + + /* lock vfs_inode first, then aufs. */ + mutex_lock(&inode->i_mutex); + aufs_write_lock(root); + err = au_opts_mount(sb, &opts); + au_opts_free(&opts); + aufs_write_unlock(root); + mutex_unlock(&inode->i_mutex); + if (!err) + goto out_opts; /* success */ + +out_root: + dput(root); + sb->s_root = NULL; +out_info: + dbgaufs_si_fin(au_sbi(sb)); + kobject_put(&au_sbi(sb)->si_kobj); + sb->s_fs_info = NULL; +out_opts: + free_page((unsigned long)opts.opt); +out: + AuTraceErr(err); + err = cvt_err(err); + AuTraceErr(err); + return err; +} + +/* ---------------------------------------------------------------------- */ + +static struct dentry *aufs_mount(struct file_system_type *fs_type, int flags, + const char *dev_name __maybe_unused, + void *raw_data) +{ + struct dentry *root; + struct super_block *sb; + + /* all timestamps always follow the ones on the branch */ + /* mnt->mnt_flags |= MNT_NOATIME | MNT_NODIRATIME; */ + root = mount_nodev(fs_type, flags, raw_data, aufs_fill_super); + if (IS_ERR(root)) + goto out; + + sb = root->d_sb; + si_write_lock(sb, !AuLock_FLUSH); + sysaufs_brs_add(sb, 0); + si_write_unlock(sb); + au_sbilist_add(sb); + +out: + return root; +} + +static void aufs_kill_sb(struct super_block *sb) +{ + struct au_sbinfo *sbinfo; + + sbinfo = au_sbi(sb); + if (sbinfo) { + au_sbilist_del(sb); + aufs_write_lock(sb->s_root); + au_fhsm_fin(sb); + if (sbinfo->si_wbr_create_ops->fin) + sbinfo->si_wbr_create_ops->fin(sb); + if (au_opt_test(sbinfo->si_mntflags, UDBA_HNOTIFY)) { + au_opt_set_udba(sbinfo->si_mntflags, UDBA_NONE); + au_remount_refresh(sb); + } + if (au_opt_test(sbinfo->si_mntflags, PLINK)) + au_plink_put(sb, /*verbose*/1); + au_xino_clr(sb); + sbinfo->si_sb = NULL; + aufs_write_unlock(sb->s_root); + au_nwt_flush(&sbinfo->si_nowait); + } + kill_anon_super(sb); +} + +struct file_system_type aufs_fs_type = { + .name = AUFS_FSTYPE, + /* a race between rename and others */ + .fs_flags = FS_RENAME_DOES_D_MOVE, + .mount = aufs_mount, + .kill_sb = aufs_kill_sb, + /* no need to __module_get() and module_put(). */ + .owner = THIS_MODULE, +}; diff -Naur kernel.21.3.orig/fs/aufs/super.h kernel.21.3.new/fs/aufs/super.h --- kernel.21.3.orig/fs/aufs/super.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/super.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,637 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * super_block operations + */ + +#ifndef __AUFS_SUPER_H__ +#define __AUFS_SUPER_H__ + +#ifdef __KERNEL__ + +#include +#include "rwsem.h" +#include "spl.h" +#include "wkq.h" + +typedef ssize_t (*au_readf_t)(struct file *, char __user *, size_t, loff_t *); +typedef ssize_t (*au_writef_t)(struct file *, const char __user *, size_t, + loff_t *); + +/* policies to select one among multiple writable branches */ +struct au_wbr_copyup_operations { + int (*copyup)(struct dentry *dentry); +}; + +#define AuWbr_DIR 1 /* target is a dir */ +#define AuWbr_PARENT (1 << 1) /* always require a parent */ + +#define au_ftest_wbr(flags, name) ((flags) & AuWbr_##name) +#define au_fset_wbr(flags, name) { (flags) |= AuWbr_##name; } +#define au_fclr_wbr(flags, name) { (flags) &= ~AuWbr_##name; } + +struct au_wbr_create_operations { + int (*create)(struct dentry *dentry, unsigned int flags); + int (*init)(struct super_block *sb); + int (*fin)(struct super_block *sb); +}; + +struct au_wbr_mfs { + struct mutex mfs_lock; /* protect this structure */ + unsigned long mfs_jiffy; + unsigned long mfs_expire; + aufs_bindex_t mfs_bindex; + + unsigned long long mfsrr_bytes; + unsigned long long mfsrr_watermark; +}; + +struct pseudo_link { + union { + struct hlist_node hlist; + struct rcu_head rcu; + }; + struct inode *inode; +}; + +#define AuPlink_NHASH 100 +static inline int au_plink_hash(ino_t ino) +{ + return ino % AuPlink_NHASH; +} + +/* File-based Hierarchical Storage Management */ +struct au_fhsm { +#ifdef CONFIG_AUFS_FHSM + /* allow only one process who can receive the notification */ + spinlock_t fhsm_spin; + pid_t fhsm_pid; + wait_queue_head_t fhsm_wqh; + atomic_t fhsm_readable; + + /* these are protected by si_rwsem */ + unsigned long fhsm_expire; + aufs_bindex_t fhsm_bottom; +#endif +}; + +struct au_branch; +struct au_sbinfo { + /* nowait tasks in the system-wide workqueue */ + struct au_nowait_tasks si_nowait; + + /* + * tried sb->s_umount, but failed due to the dependecy between i_mutex. + * rwsem for au_sbinfo is necessary. + */ + struct au_rwsem si_rwsem; + + /* prevent recursive locking in deleting inode */ + struct { + unsigned long *bitmap; + spinlock_t tree_lock; + struct radix_tree_root tree; + } au_si_pid; + + /* + * dirty approach to protect sb->sb_inodes and ->s_files from remount. + */ + atomic_long_t si_ninodes, si_nfiles; + + /* branch management */ + unsigned int si_generation; + + /* see above flags */ + unsigned char au_si_status; + + aufs_bindex_t si_bend; + + /* dirty trick to keep br_id plus */ + unsigned int si_last_br_id : + sizeof(aufs_bindex_t) * BITS_PER_BYTE - 1; + struct au_branch **si_branch; + + /* policy to select a writable branch */ + unsigned char si_wbr_copyup; + unsigned char si_wbr_create; + struct au_wbr_copyup_operations *si_wbr_copyup_ops; + struct au_wbr_create_operations *si_wbr_create_ops; + + /* round robin */ + atomic_t si_wbr_rr_next; + + /* most free space */ + struct au_wbr_mfs si_wbr_mfs; + + /* File-based Hierarchical Storage Management */ + struct au_fhsm si_fhsm; + + /* mount flags */ + /* include/asm-ia64/siginfo.h defines a macro named si_flags */ + unsigned int si_mntflags; + + /* external inode number (bitmap and translation table) */ + au_readf_t si_xread; + au_writef_t si_xwrite; + struct file *si_xib; + struct mutex si_xib_mtx; /* protect xib members */ + unsigned long *si_xib_buf; + unsigned long si_xib_last_pindex; + int si_xib_next_bit; + aufs_bindex_t si_xino_brid; + unsigned long si_xino_jiffy; + unsigned long si_xino_expire; + /* reserved for future use */ + /* unsigned long long si_xib_limit; */ /* Max xib file size */ + +#ifdef CONFIG_AUFS_EXPORT + /* i_generation */ + struct file *si_xigen; + atomic_t si_xigen_next; +#endif + + /* vdir parameters */ + unsigned long si_rdcache; /* max cache time in jiffies */ + unsigned int si_rdblk; /* deblk size */ + unsigned int si_rdhash; /* hash size */ + + /* + * If the number of whiteouts are larger than si_dirwh, leave all of + * them after au_whtmp_ren to reduce the cost of rmdir(2). + * future fsck.aufs or kernel thread will remove them later. + * Otherwise, remove all whiteouts and the dir in rmdir(2). + */ + unsigned int si_dirwh; + + /* + * rename(2) a directory with all children. + */ + /* reserved for future use */ + /* int si_rendir; */ + + /* pseudo_link list */ + struct au_sphlhead si_plink[AuPlink_NHASH]; + wait_queue_head_t si_plink_wq; + spinlock_t si_plink_maint_lock; + pid_t si_plink_maint_pid; + + /* + * sysfs and lifetime management. + * this is not a small structure and it may be a waste of memory in case + * of sysfs is disabled, particulary when many aufs-es are mounted. + * but using sysfs is majority. + */ + struct kobject si_kobj; +#ifdef CONFIG_DEBUG_FS + struct dentry *si_dbgaufs; + struct dentry *si_dbgaufs_plink; + struct dentry *si_dbgaufs_xib; +#ifdef CONFIG_AUFS_EXPORT + struct dentry *si_dbgaufs_xigen; +#endif +#endif + +#ifdef CONFIG_AUFS_SBILIST + struct list_head si_list; +#endif + + /* dirty, necessary for unmounting, sysfs and sysrq */ + struct super_block *si_sb; +}; + +/* sbinfo status flags */ +/* + * set true when refresh_dirs() failed at remount time. + * then try refreshing dirs at access time again. + * if it is false, refreshing dirs at access time is unnecesary + */ +#define AuSi_FAILED_REFRESH_DIR 1 + +#define AuSi_FHSM (1 << 1) /* fhsm is active now */ + +#ifndef CONFIG_AUFS_FHSM +#undef AuSi_FHSM +#define AuSi_FHSM 0 +#endif + +static inline unsigned char au_do_ftest_si(struct au_sbinfo *sbi, + unsigned int flag) +{ + AuRwMustAnyLock(&sbi->si_rwsem); + return sbi->au_si_status & flag; +} +#define au_ftest_si(sbinfo, name) au_do_ftest_si(sbinfo, AuSi_##name) +#define au_fset_si(sbinfo, name) do { \ + AuRwMustWriteLock(&(sbinfo)->si_rwsem); \ + (sbinfo)->au_si_status |= AuSi_##name; \ +} while (0) +#define au_fclr_si(sbinfo, name) do { \ + AuRwMustWriteLock(&(sbinfo)->si_rwsem); \ + (sbinfo)->au_si_status &= ~AuSi_##name; \ +} while (0) + +/* ---------------------------------------------------------------------- */ + +/* policy to select one among writable branches */ +#define AuWbrCopyup(sbinfo, ...) \ + ((sbinfo)->si_wbr_copyup_ops->copyup(__VA_ARGS__)) +#define AuWbrCreate(sbinfo, ...) \ + ((sbinfo)->si_wbr_create_ops->create(__VA_ARGS__)) + +/* flags for si_read_lock()/aufs_read_lock()/di_read_lock() */ +#define AuLock_DW 1 /* write-lock dentry */ +#define AuLock_IR (1 << 1) /* read-lock inode */ +#define AuLock_IW (1 << 2) /* write-lock inode */ +#define AuLock_FLUSH (1 << 3) /* wait for 'nowait' tasks */ +#define AuLock_DIR (1 << 4) /* target is a dir */ +#define AuLock_NOPLM (1 << 5) /* return err in plm mode */ +#define AuLock_NOPLMW (1 << 6) /* wait for plm mode ends */ +#define AuLock_GEN (1 << 7) /* test digen/iigen */ +#define au_ftest_lock(flags, name) ((flags) & AuLock_##name) +#define au_fset_lock(flags, name) \ + do { (flags) |= AuLock_##name; } while (0) +#define au_fclr_lock(flags, name) \ + do { (flags) &= ~AuLock_##name; } while (0) + +/* ---------------------------------------------------------------------- */ + +/* super.c */ +extern struct file_system_type aufs_fs_type; +struct inode *au_iget_locked(struct super_block *sb, ino_t ino); +typedef unsigned long long (*au_arraycb_t)(void *array, unsigned long long max, + void *arg); +void au_array_free(void *array); +void *au_array_alloc(unsigned long long *hint, au_arraycb_t cb, void *arg); +struct inode **au_iarray_alloc(struct super_block *sb, unsigned long long *max); +void au_iarray_free(struct inode **a, unsigned long long max); + +/* sbinfo.c */ +void au_si_free(struct kobject *kobj); +int au_si_alloc(struct super_block *sb); +int au_sbr_realloc(struct au_sbinfo *sbinfo, int nbr); + +unsigned int au_sigen_inc(struct super_block *sb); +aufs_bindex_t au_new_br_id(struct super_block *sb); + +int si_read_lock(struct super_block *sb, int flags); +int si_write_lock(struct super_block *sb, int flags); +int aufs_read_lock(struct dentry *dentry, int flags); +void aufs_read_unlock(struct dentry *dentry, int flags); +void aufs_write_lock(struct dentry *dentry); +void aufs_write_unlock(struct dentry *dentry); +int aufs_read_and_write_lock2(struct dentry *d1, struct dentry *d2, int flags); +void aufs_read_and_write_unlock2(struct dentry *d1, struct dentry *d2); + +int si_pid_test_slow(struct super_block *sb); +void si_pid_set_slow(struct super_block *sb); +void si_pid_clr_slow(struct super_block *sb); + +/* wbr_policy.c */ +extern struct au_wbr_copyup_operations au_wbr_copyup_ops[]; +extern struct au_wbr_create_operations au_wbr_create_ops[]; +int au_cpdown_dirs(struct dentry *dentry, aufs_bindex_t bdst); +int au_wbr_nonopq(struct dentry *dentry, aufs_bindex_t bindex); +int au_wbr_do_copyup_bu(struct dentry *dentry, aufs_bindex_t bstart); + +/* mvdown.c */ +int au_mvdown(struct dentry *dentry, struct aufs_mvdown __user *arg); + +#ifdef CONFIG_AUFS_FHSM +/* fhsm.c */ + +static inline pid_t au_fhsm_pid(struct au_fhsm *fhsm) +{ + pid_t pid; + + spin_lock(&fhsm->fhsm_spin); + pid = fhsm->fhsm_pid; + spin_unlock(&fhsm->fhsm_spin); + + return pid; +} + +void au_fhsm_wrote(struct super_block *sb, aufs_bindex_t bindex, int force); +void au_fhsm_wrote_all(struct super_block *sb, int force); +int au_fhsm_fd(struct super_block *sb, int oflags); +int au_fhsm_br_alloc(struct au_branch *br); +void au_fhsm_set_bottom(struct super_block *sb, aufs_bindex_t bindex); +void au_fhsm_fin(struct super_block *sb); +void au_fhsm_init(struct au_sbinfo *sbinfo); +void au_fhsm_set(struct au_sbinfo *sbinfo, unsigned int sec); +void au_fhsm_show(struct seq_file *seq, struct au_sbinfo *sbinfo); +#else +AuStubVoid(au_fhsm_wrote, struct super_block *sb, aufs_bindex_t bindex, + int force) +AuStubVoid(au_fhsm_wrote_all, struct super_block *sb, int force) +AuStub(int, au_fhsm_fd, return -EOPNOTSUPP, struct super_block *sb, int oflags) +AuStub(pid_t, au_fhsm_pid, return 0, struct au_fhsm *fhsm) +AuStubInt0(au_fhsm_br_alloc, struct au_branch *br) +AuStubVoid(au_fhsm_set_bottom, struct super_block *sb, aufs_bindex_t bindex) +AuStubVoid(au_fhsm_fin, struct super_block *sb) +AuStubVoid(au_fhsm_init, struct au_sbinfo *sbinfo) +AuStubVoid(au_fhsm_set, struct au_sbinfo *sbinfo, unsigned int sec) +AuStubVoid(au_fhsm_show, struct seq_file *seq, struct au_sbinfo *sbinfo) +#endif + +/* ---------------------------------------------------------------------- */ + +static inline struct au_sbinfo *au_sbi(struct super_block *sb) +{ + return sb->s_fs_info; +} + +/* ---------------------------------------------------------------------- */ + +#ifdef CONFIG_AUFS_EXPORT +int au_test_nfsd(void); +void au_export_init(struct super_block *sb); +void au_xigen_inc(struct inode *inode); +int au_xigen_new(struct inode *inode); +int au_xigen_set(struct super_block *sb, struct file *base); +void au_xigen_clr(struct super_block *sb); + +static inline int au_busy_or_stale(void) +{ + if (!au_test_nfsd()) + return -EBUSY; + return -ESTALE; +} +#else +AuStubInt0(au_test_nfsd, void) +AuStubVoid(au_export_init, struct super_block *sb) +AuStubVoid(au_xigen_inc, struct inode *inode) +AuStubInt0(au_xigen_new, struct inode *inode) +AuStubInt0(au_xigen_set, struct super_block *sb, struct file *base) +AuStubVoid(au_xigen_clr, struct super_block *sb) +AuStub(int, au_busy_or_stale, return -EBUSY, void) +#endif /* CONFIG_AUFS_EXPORT */ + +/* ---------------------------------------------------------------------- */ + +#ifdef CONFIG_AUFS_SBILIST +/* module.c */ +extern struct au_splhead au_sbilist; + +static inline void au_sbilist_init(void) +{ + au_spl_init(&au_sbilist); +} + +static inline void au_sbilist_add(struct super_block *sb) +{ + au_spl_add(&au_sbi(sb)->si_list, &au_sbilist); +} + +static inline void au_sbilist_del(struct super_block *sb) +{ + au_spl_del(&au_sbi(sb)->si_list, &au_sbilist); +} + +#ifdef CONFIG_AUFS_MAGIC_SYSRQ +static inline void au_sbilist_lock(void) +{ + spin_lock(&au_sbilist.spin); +} + +static inline void au_sbilist_unlock(void) +{ + spin_unlock(&au_sbilist.spin); +} +#define AuGFP_SBILIST GFP_ATOMIC +#else +AuStubVoid(au_sbilist_lock, void) +AuStubVoid(au_sbilist_unlock, void) +#define AuGFP_SBILIST GFP_NOFS +#endif /* CONFIG_AUFS_MAGIC_SYSRQ */ +#else +AuStubVoid(au_sbilist_init, void) +AuStubVoid(au_sbilist_add, struct super_block *sb) +AuStubVoid(au_sbilist_del, struct super_block *sb) +AuStubVoid(au_sbilist_lock, void) +AuStubVoid(au_sbilist_unlock, void) +#define AuGFP_SBILIST GFP_NOFS +#endif + +/* ---------------------------------------------------------------------- */ + +static inline void dbgaufs_si_null(struct au_sbinfo *sbinfo) +{ + /* + * This function is a dynamic '__init' function actually, + * so the tiny check for si_rwsem is unnecessary. + */ + /* AuRwMustWriteLock(&sbinfo->si_rwsem); */ +#ifdef CONFIG_DEBUG_FS + sbinfo->si_dbgaufs = NULL; + sbinfo->si_dbgaufs_plink = NULL; + sbinfo->si_dbgaufs_xib = NULL; +#ifdef CONFIG_AUFS_EXPORT + sbinfo->si_dbgaufs_xigen = NULL; +#endif +#endif +} + +/* ---------------------------------------------------------------------- */ + +static inline pid_t si_pid_bit(void) +{ + /* the origin of pid is 1, but the bitmap's is 0 */ + return current->pid - 1; +} + +static inline int si_pid_test(struct super_block *sb) +{ + pid_t bit; + + bit = si_pid_bit(); + if (bit < PID_MAX_DEFAULT) + return test_bit(bit, au_sbi(sb)->au_si_pid.bitmap); + return si_pid_test_slow(sb); +} + +static inline void si_pid_set(struct super_block *sb) +{ + pid_t bit; + + bit = si_pid_bit(); + if (bit < PID_MAX_DEFAULT) { + AuDebugOn(test_bit(bit, au_sbi(sb)->au_si_pid.bitmap)); + set_bit(bit, au_sbi(sb)->au_si_pid.bitmap); + /* smp_mb(); */ + } else + si_pid_set_slow(sb); +} + +static inline void si_pid_clr(struct super_block *sb) +{ + pid_t bit; + + bit = si_pid_bit(); + if (bit < PID_MAX_DEFAULT) { + AuDebugOn(!test_bit(bit, au_sbi(sb)->au_si_pid.bitmap)); + clear_bit(bit, au_sbi(sb)->au_si_pid.bitmap); + /* smp_mb(); */ + } else + si_pid_clr_slow(sb); +} + +/* ---------------------------------------------------------------------- */ + +/* lock superblock. mainly for entry point functions */ +/* + * __si_read_lock, __si_write_lock, + * __si_read_unlock, __si_write_unlock, __si_downgrade_lock + */ +AuSimpleRwsemFuncs(__si, struct super_block *sb, &au_sbi(sb)->si_rwsem); + +#define SiMustNoWaiters(sb) AuRwMustNoWaiters(&au_sbi(sb)->si_rwsem) +#define SiMustAnyLock(sb) AuRwMustAnyLock(&au_sbi(sb)->si_rwsem) +#define SiMustWriteLock(sb) AuRwMustWriteLock(&au_sbi(sb)->si_rwsem) + +static inline void si_noflush_read_lock(struct super_block *sb) +{ + __si_read_lock(sb); + si_pid_set(sb); +} + +static inline int si_noflush_read_trylock(struct super_block *sb) +{ + int locked; + + locked = __si_read_trylock(sb); + if (locked) + si_pid_set(sb); + return locked; +} + +static inline void si_noflush_write_lock(struct super_block *sb) +{ + __si_write_lock(sb); + si_pid_set(sb); +} + +static inline int si_noflush_write_trylock(struct super_block *sb) +{ + int locked; + + locked = __si_write_trylock(sb); + if (locked) + si_pid_set(sb); + return locked; +} + +#if 0 /* unused */ +static inline int si_read_trylock(struct super_block *sb, int flags) +{ + if (au_ftest_lock(flags, FLUSH)) + au_nwt_flush(&au_sbi(sb)->si_nowait); + return si_noflush_read_trylock(sb); +} +#endif + +static inline void si_read_unlock(struct super_block *sb) +{ + si_pid_clr(sb); + __si_read_unlock(sb); +} + +#if 0 /* unused */ +static inline int si_write_trylock(struct super_block *sb, int flags) +{ + if (au_ftest_lock(flags, FLUSH)) + au_nwt_flush(&au_sbi(sb)->si_nowait); + return si_noflush_write_trylock(sb); +} +#endif + +static inline void si_write_unlock(struct super_block *sb) +{ + si_pid_clr(sb); + __si_write_unlock(sb); +} + +#if 0 /* unused */ +static inline void si_downgrade_lock(struct super_block *sb) +{ + __si_downgrade_lock(sb); +} +#endif + +/* ---------------------------------------------------------------------- */ + +static inline aufs_bindex_t au_sbend(struct super_block *sb) +{ + SiMustAnyLock(sb); + return au_sbi(sb)->si_bend; +} + +static inline unsigned int au_mntflags(struct super_block *sb) +{ + SiMustAnyLock(sb); + return au_sbi(sb)->si_mntflags; +} + +static inline unsigned int au_sigen(struct super_block *sb) +{ + SiMustAnyLock(sb); + return au_sbi(sb)->si_generation; +} + +static inline void au_ninodes_inc(struct super_block *sb) +{ + atomic_long_inc(&au_sbi(sb)->si_ninodes); +} + +static inline void au_ninodes_dec(struct super_block *sb) +{ + AuDebugOn(!atomic_long_read(&au_sbi(sb)->si_ninodes)); + atomic_long_dec(&au_sbi(sb)->si_ninodes); +} + +static inline void au_nfiles_inc(struct super_block *sb) +{ + atomic_long_inc(&au_sbi(sb)->si_nfiles); +} + +static inline void au_nfiles_dec(struct super_block *sb) +{ + AuDebugOn(!atomic_long_read(&au_sbi(sb)->si_nfiles)); + atomic_long_dec(&au_sbi(sb)->si_nfiles); +} + +static inline struct au_branch *au_sbr(struct super_block *sb, + aufs_bindex_t bindex) +{ + SiMustAnyLock(sb); + return au_sbi(sb)->si_branch[0 + bindex]; +} + +static inline void au_xino_brid_set(struct super_block *sb, aufs_bindex_t brid) +{ + SiMustWriteLock(sb); + au_sbi(sb)->si_xino_brid = brid; +} + +static inline aufs_bindex_t au_xino_brid(struct super_block *sb) +{ + SiMustAnyLock(sb); + return au_sbi(sb)->si_xino_brid; +} + +#endif /* __KERNEL__ */ +#endif /* __AUFS_SUPER_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/sysaufs.c kernel.21.3.new/fs/aufs/sysaufs.c --- kernel.21.3.orig/fs/aufs/sysaufs.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/sysaufs.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,104 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * sysfs interface and lifetime management + * they are necessary regardless sysfs is disabled. + */ + +#include +#include "aufs.h" + +unsigned long sysaufs_si_mask; +struct kset *sysaufs_kset; + +#define AuSiAttr(_name) { \ + .attr = { .name = __stringify(_name), .mode = 0444 }, \ + .show = sysaufs_si_##_name, \ +} + +static struct sysaufs_si_attr sysaufs_si_attr_xi_path = AuSiAttr(xi_path); +struct attribute *sysaufs_si_attrs[] = { + &sysaufs_si_attr_xi_path.attr, + NULL, +}; + +static const struct sysfs_ops au_sbi_ops = { + .show = sysaufs_si_show +}; + +static struct kobj_type au_sbi_ktype = { + .release = au_si_free, + .sysfs_ops = &au_sbi_ops, + .default_attrs = sysaufs_si_attrs +}; + +/* ---------------------------------------------------------------------- */ + +int sysaufs_si_init(struct au_sbinfo *sbinfo) +{ + int err; + + sbinfo->si_kobj.kset = sysaufs_kset; + /* cf. sysaufs_name() */ + err = kobject_init_and_add + (&sbinfo->si_kobj, &au_sbi_ktype, /*&sysaufs_kset->kobj*/NULL, + SysaufsSiNamePrefix "%lx", sysaufs_si_id(sbinfo)); + + dbgaufs_si_null(sbinfo); + if (!err) { + err = dbgaufs_si_init(sbinfo); + if (unlikely(err)) + kobject_put(&sbinfo->si_kobj); + } + return err; +} + +void sysaufs_fin(void) +{ + dbgaufs_fin(); + sysfs_remove_group(&sysaufs_kset->kobj, sysaufs_attr_group); + kset_unregister(sysaufs_kset); +} + +int __init sysaufs_init(void) +{ + int err; + + do { + get_random_bytes(&sysaufs_si_mask, sizeof(sysaufs_si_mask)); + } while (!sysaufs_si_mask); + + err = -EINVAL; + sysaufs_kset = kset_create_and_add(AUFS_NAME, NULL, fs_kobj); + if (unlikely(!sysaufs_kset)) + goto out; + err = PTR_ERR(sysaufs_kset); + if (IS_ERR(sysaufs_kset)) + goto out; + err = sysfs_create_group(&sysaufs_kset->kobj, sysaufs_attr_group); + if (unlikely(err)) { + kset_unregister(sysaufs_kset); + goto out; + } + + err = dbgaufs_init(); + if (unlikely(err)) + sysaufs_fin(); +out: + return err; +} diff -Naur kernel.21.3.orig/fs/aufs/sysaufs.h kernel.21.3.new/fs/aufs/sysaufs.h --- kernel.21.3.orig/fs/aufs/sysaufs.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/sysaufs.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,101 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * sysfs interface and mount lifetime management + */ + +#ifndef __SYSAUFS_H__ +#define __SYSAUFS_H__ + +#ifdef __KERNEL__ + +#include +#include "module.h" + +struct super_block; +struct au_sbinfo; + +struct sysaufs_si_attr { + struct attribute attr; + int (*show)(struct seq_file *seq, struct super_block *sb); +}; + +/* ---------------------------------------------------------------------- */ + +/* sysaufs.c */ +extern unsigned long sysaufs_si_mask; +extern struct kset *sysaufs_kset; +extern struct attribute *sysaufs_si_attrs[]; +int sysaufs_si_init(struct au_sbinfo *sbinfo); +int __init sysaufs_init(void); +void sysaufs_fin(void); + +/* ---------------------------------------------------------------------- */ + +/* some people doesn't like to show a pointer in kernel */ +static inline unsigned long sysaufs_si_id(struct au_sbinfo *sbinfo) +{ + return sysaufs_si_mask ^ (unsigned long)sbinfo; +} + +#define SysaufsSiNamePrefix "si_" +#define SysaufsSiNameLen (sizeof(SysaufsSiNamePrefix) + 16) +static inline void sysaufs_name(struct au_sbinfo *sbinfo, char *name) +{ + snprintf(name, SysaufsSiNameLen, SysaufsSiNamePrefix "%lx", + sysaufs_si_id(sbinfo)); +} + +struct au_branch; +#ifdef CONFIG_SYSFS +/* sysfs.c */ +extern struct attribute_group *sysaufs_attr_group; + +int sysaufs_si_xi_path(struct seq_file *seq, struct super_block *sb); +ssize_t sysaufs_si_show(struct kobject *kobj, struct attribute *attr, + char *buf); +long au_brinfo_ioctl(struct file *file, unsigned long arg); +#ifdef CONFIG_COMPAT +long au_brinfo_compat_ioctl(struct file *file, unsigned long arg); +#endif + +void sysaufs_br_init(struct au_branch *br); +void sysaufs_brs_add(struct super_block *sb, aufs_bindex_t bindex); +void sysaufs_brs_del(struct super_block *sb, aufs_bindex_t bindex); + +#define sysaufs_brs_init() do {} while (0) + +#else +#define sysaufs_attr_group NULL + +AuStubInt0(sysaufs_si_xi_path, struct seq_file *seq, struct super_block *sb) +AuStub(ssize_t, sysaufs_si_show, return 0, struct kobject *kobj, + struct attribute *attr, char *buf) +AuStubVoid(sysaufs_br_init, struct au_branch *br) +AuStubVoid(sysaufs_brs_add, struct super_block *sb, aufs_bindex_t bindex) +AuStubVoid(sysaufs_brs_del, struct super_block *sb, aufs_bindex_t bindex) + +static inline void sysaufs_brs_init(void) +{ + sysaufs_brs = 0; +} + +#endif /* CONFIG_SYSFS */ + +#endif /* __KERNEL__ */ +#endif /* __SYSAUFS_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/sysfs.c kernel.21.3.new/fs/aufs/sysfs.c --- kernel.21.3.orig/fs/aufs/sysfs.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/sysfs.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,372 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * sysfs interface + */ + +#include +#include +#include "aufs.h" + +#ifdef CONFIG_AUFS_FS_MODULE +/* this entry violates the "one line per file" policy of sysfs */ +static ssize_t config_show(struct kobject *kobj, struct kobj_attribute *attr, + char *buf) +{ + ssize_t err; + static char *conf = +/* this file is generated at compiling */ +#include "conf.str" + ; + + err = snprintf(buf, PAGE_SIZE, conf); + if (unlikely(err >= PAGE_SIZE)) + err = -EFBIG; + return err; +} + +static struct kobj_attribute au_config_attr = __ATTR_RO(config); +#endif + +static struct attribute *au_attr[] = { +#ifdef CONFIG_AUFS_FS_MODULE + &au_config_attr.attr, +#endif + NULL, /* need to NULL terminate the list of attributes */ +}; + +static struct attribute_group sysaufs_attr_group_body = { + .attrs = au_attr +}; + +struct attribute_group *sysaufs_attr_group = &sysaufs_attr_group_body; + +/* ---------------------------------------------------------------------- */ + +int sysaufs_si_xi_path(struct seq_file *seq, struct super_block *sb) +{ + int err; + + SiMustAnyLock(sb); + + err = 0; + if (au_opt_test(au_mntflags(sb), XINO)) { + err = au_xino_path(seq, au_sbi(sb)->si_xib); + seq_putc(seq, '\n'); + } + return err; +} + +/* + * the lifetime of branch is independent from the entry under sysfs. + * sysfs handles the lifetime of the entry, and never call ->show() after it is + * unlinked. + */ +static int sysaufs_si_br(struct seq_file *seq, struct super_block *sb, + aufs_bindex_t bindex, int idx) +{ + int err; + struct path path; + struct dentry *root; + struct au_branch *br; + au_br_perm_str_t perm; + + AuDbg("b%d\n", bindex); + + err = 0; + root = sb->s_root; + di_read_lock_parent(root, !AuLock_IR); + br = au_sbr(sb, bindex); + + switch (idx) { + case AuBrSysfs_BR: + path.mnt = au_br_mnt(br); + path.dentry = au_h_dptr(root, bindex); + au_seq_path(seq, &path); + au_optstr_br_perm(&perm, br->br_perm); + err = seq_printf(seq, "=%s\n", perm.a); + break; + case AuBrSysfs_BRID: + err = seq_printf(seq, "%d\n", br->br_id); + break; + } + di_read_unlock(root, !AuLock_IR); + if (err == -1) + err = -E2BIG; + + return err; +} + +/* ---------------------------------------------------------------------- */ + +static struct seq_file *au_seq(char *p, ssize_t len) +{ + struct seq_file *seq; + + seq = kzalloc(sizeof(*seq), GFP_NOFS); + if (seq) { + /* mutex_init(&seq.lock); */ + seq->buf = p; + seq->size = len; + return seq; /* success */ + } + + seq = ERR_PTR(-ENOMEM); + return seq; +} + +#define SysaufsBr_PREFIX "br" +#define SysaufsBrid_PREFIX "brid" + +/* todo: file size may exceed PAGE_SIZE */ +ssize_t sysaufs_si_show(struct kobject *kobj, struct attribute *attr, + char *buf) +{ + ssize_t err; + int idx; + long l; + aufs_bindex_t bend; + struct au_sbinfo *sbinfo; + struct super_block *sb; + struct seq_file *seq; + char *name; + struct attribute **cattr; + + sbinfo = container_of(kobj, struct au_sbinfo, si_kobj); + sb = sbinfo->si_sb; + + /* + * prevent a race condition between sysfs and aufs. + * for instance, sysfs_file_read() calls sysfs_get_active_two() which + * prohibits maintaining the sysfs entries. + * hew we acquire read lock after sysfs_get_active_two(). + * on the other hand, the remount process may maintain the sysfs/aufs + * entries after acquiring write lock. + * it can cause a deadlock. + * simply we gave up processing read here. + */ + err = -EBUSY; + if (unlikely(!si_noflush_read_trylock(sb))) + goto out; + + seq = au_seq(buf, PAGE_SIZE); + err = PTR_ERR(seq); + if (IS_ERR(seq)) + goto out_unlock; + + name = (void *)attr->name; + cattr = sysaufs_si_attrs; + while (*cattr) { + if (!strcmp(name, (*cattr)->name)) { + err = container_of(*cattr, struct sysaufs_si_attr, attr) + ->show(seq, sb); + goto out_seq; + } + cattr++; + } + + if (!strncmp(name, SysaufsBrid_PREFIX, + sizeof(SysaufsBrid_PREFIX) - 1)) { + idx = AuBrSysfs_BRID; + name += sizeof(SysaufsBrid_PREFIX) - 1; + } else if (!strncmp(name, SysaufsBr_PREFIX, + sizeof(SysaufsBr_PREFIX) - 1)) { + idx = AuBrSysfs_BR; + name += sizeof(SysaufsBr_PREFIX) - 1; + } else + BUG(); + + err = kstrtol(name, 10, &l); + if (!err) { + bend = au_sbend(sb); + if (l <= bend) + err = sysaufs_si_br(seq, sb, (aufs_bindex_t)l, idx); + else + err = -ENOENT; + } + +out_seq: + if (!err) { + err = seq->count; + /* sysfs limit */ + if (unlikely(err == PAGE_SIZE)) + err = -EFBIG; + } + kfree(seq); +out_unlock: + si_read_unlock(sb); +out: + return err; +} + +/* ---------------------------------------------------------------------- */ + +static int au_brinfo(struct super_block *sb, union aufs_brinfo __user *arg) +{ + int err; + int16_t brid; + aufs_bindex_t bindex, bend; + size_t sz; + char *buf; + struct seq_file *seq; + struct au_branch *br; + + si_read_lock(sb, AuLock_FLUSH); + bend = au_sbend(sb); + err = bend + 1; + if (!arg) + goto out; + + err = -ENOMEM; + buf = (void *)__get_free_page(GFP_NOFS); + if (unlikely(!buf)) + goto out; + + seq = au_seq(buf, PAGE_SIZE); + err = PTR_ERR(seq); + if (IS_ERR(seq)) + goto out_buf; + + sz = sizeof(*arg) - offsetof(union aufs_brinfo, path); + for (bindex = 0; bindex <= bend; bindex++, arg++) { + err = !access_ok(VERIFY_WRITE, arg, sizeof(*arg)); + if (unlikely(err)) + break; + + br = au_sbr(sb, bindex); + brid = br->br_id; + BUILD_BUG_ON(sizeof(brid) != sizeof(arg->id)); + err = __put_user(brid, &arg->id); + if (unlikely(err)) + break; + + BUILD_BUG_ON(sizeof(br->br_perm) != sizeof(arg->perm)); + err = __put_user(br->br_perm, &arg->perm); + if (unlikely(err)) + break; + + au_seq_path(seq, &br->br_path); + err = seq_putc(seq, '\0'); + if (!err && seq->count <= sz) { + err = copy_to_user(arg->path, seq->buf, seq->count); + seq->count = 0; + if (unlikely(err)) + break; + } else { + err = -E2BIG; + goto out_seq; + } + } + if (unlikely(err)) + err = -EFAULT; + +out_seq: + kfree(seq); +out_buf: + free_page((unsigned long)buf); +out: + si_read_unlock(sb); + return err; +} + +long au_brinfo_ioctl(struct file *file, unsigned long arg) +{ + return au_brinfo(file->f_dentry->d_sb, (void __user *)arg); +} + +#ifdef CONFIG_COMPAT +long au_brinfo_compat_ioctl(struct file *file, unsigned long arg) +{ + return au_brinfo(file->f_dentry->d_sb, compat_ptr(arg)); +} +#endif + +/* ---------------------------------------------------------------------- */ + +void sysaufs_br_init(struct au_branch *br) +{ + int i; + struct au_brsysfs *br_sysfs; + struct attribute *attr; + + br_sysfs = br->br_sysfs; + for (i = 0; i < ARRAY_SIZE(br->br_sysfs); i++) { + attr = &br_sysfs->attr; + sysfs_attr_init(attr); + attr->name = br_sysfs->name; + attr->mode = S_IRUGO; + br_sysfs++; + } +} + +void sysaufs_brs_del(struct super_block *sb, aufs_bindex_t bindex) +{ + struct au_branch *br; + struct kobject *kobj; + struct au_brsysfs *br_sysfs; + int i; + aufs_bindex_t bend; + + dbgaufs_brs_del(sb, bindex); + + if (!sysaufs_brs) + return; + + kobj = &au_sbi(sb)->si_kobj; + bend = au_sbend(sb); + for (; bindex <= bend; bindex++) { + br = au_sbr(sb, bindex); + br_sysfs = br->br_sysfs; + for (i = 0; i < ARRAY_SIZE(br->br_sysfs); i++) { + sysfs_remove_file(kobj, &br_sysfs->attr); + br_sysfs++; + } + } +} + +void sysaufs_brs_add(struct super_block *sb, aufs_bindex_t bindex) +{ + int err, i; + aufs_bindex_t bend; + struct kobject *kobj; + struct au_branch *br; + struct au_brsysfs *br_sysfs; + + dbgaufs_brs_add(sb, bindex); + + if (!sysaufs_brs) + return; + + kobj = &au_sbi(sb)->si_kobj; + bend = au_sbend(sb); + for (; bindex <= bend; bindex++) { + br = au_sbr(sb, bindex); + br_sysfs = br->br_sysfs; + snprintf(br_sysfs[AuBrSysfs_BR].name, sizeof(br_sysfs->name), + SysaufsBr_PREFIX "%d", bindex); + snprintf(br_sysfs[AuBrSysfs_BRID].name, sizeof(br_sysfs->name), + SysaufsBrid_PREFIX "%d", bindex); + for (i = 0; i < ARRAY_SIZE(br->br_sysfs); i++) { + err = sysfs_create_file(kobj, &br_sysfs->attr); + if (unlikely(err)) + pr_warn("failed %s under sysfs(%d)\n", + br_sysfs->name, err); + br_sysfs++; + } + } +} diff -Naur kernel.21.3.orig/fs/aufs/sysrq.c kernel.21.3.new/fs/aufs/sysrq.c --- kernel.21.3.orig/fs/aufs/sysrq.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/sysrq.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,153 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * magic sysrq hanlder + */ + +/* #include */ +#include +#include "aufs.h" + +/* ---------------------------------------------------------------------- */ + +static void sysrq_sb(struct super_block *sb) +{ + char *plevel; + struct au_sbinfo *sbinfo; + struct file *file; + + plevel = au_plevel; + au_plevel = KERN_WARNING; + + /* since we define pr_fmt, call printk directly */ +#define pr(str) printk(KERN_WARNING AUFS_NAME ": " str) + + sbinfo = au_sbi(sb); + printk(KERN_WARNING "si=%lx\n", sysaufs_si_id(sbinfo)); + pr("superblock\n"); + au_dpri_sb(sb); + +#if 0 + pr("root dentry\n"); + au_dpri_dentry(sb->s_root); + pr("root inode\n"); + au_dpri_inode(sb->s_root->d_inode); +#endif + +#if 0 + do { + int err, i, j, ndentry; + struct au_dcsub_pages dpages; + struct au_dpage *dpage; + + err = au_dpages_init(&dpages, GFP_ATOMIC); + if (unlikely(err)) + break; + err = au_dcsub_pages(&dpages, sb->s_root, NULL, NULL); + if (!err) + for (i = 0; i < dpages.ndpage; i++) { + dpage = dpages.dpages + i; + ndentry = dpage->ndentry; + for (j = 0; j < ndentry; j++) + au_dpri_dentry(dpage->dentries[j]); + } + au_dpages_free(&dpages); + } while (0); +#endif + +#if 1 + { + struct inode *i; + + pr("isolated inode\n"); + spin_lock(&inode_sb_list_lock); + list_for_each_entry(i, &sb->s_inodes, i_sb_list) { + spin_lock(&i->i_lock); + if (1 || hlist_empty(&i->i_dentry)) + au_dpri_inode(i); + spin_unlock(&i->i_lock); + } + spin_unlock(&inode_sb_list_lock); + } +#endif + pr("files\n"); + lg_global_lock(&files_lglock); + do_file_list_for_each_entry(sb, file) { + umode_t mode; + + mode = file_inode(file)->i_mode; + if (!special_file(mode)) + au_dpri_file(file); + } while_file_list_for_each_entry; + lg_global_unlock(&files_lglock); + pr("done\n"); + +#undef pr + au_plevel = plevel; +} + +/* ---------------------------------------------------------------------- */ + +/* module parameter */ +static char *aufs_sysrq_key = "a"; +module_param_named(sysrq, aufs_sysrq_key, charp, S_IRUGO); +MODULE_PARM_DESC(sysrq, "MagicSysRq key for " AUFS_NAME); + +static void au_sysrq(int key __maybe_unused) +{ + struct au_sbinfo *sbinfo; + + lockdep_off(); + au_sbilist_lock(); + list_for_each_entry(sbinfo, &au_sbilist.head, si_list) + sysrq_sb(sbinfo->si_sb); + au_sbilist_unlock(); + lockdep_on(); +} + +static struct sysrq_key_op au_sysrq_op = { + .handler = au_sysrq, + .help_msg = "Aufs", + .action_msg = "Aufs", + .enable_mask = SYSRQ_ENABLE_DUMP +}; + +/* ---------------------------------------------------------------------- */ + +int __init au_sysrq_init(void) +{ + int err; + char key; + + err = -1; + key = *aufs_sysrq_key; + if ('a' <= key && key <= 'z') + err = register_sysrq_key(key, &au_sysrq_op); + if (unlikely(err)) + pr_err("err %d, sysrq=%c\n", err, key); + return err; +} + +void au_sysrq_fin(void) +{ + int err; + + err = unregister_sysrq_key(*aufs_sysrq_key, &au_sysrq_op); + if (unlikely(err)) + pr_err("err %d (ignored)\n", err); +} diff -Naur kernel.21.3.orig/fs/aufs/vdir.c kernel.21.3.new/fs/aufs/vdir.c --- kernel.21.3.orig/fs/aufs/vdir.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/vdir.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,879 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * virtual or vertical directory + */ + +#include "aufs.h" + +static unsigned int calc_size(int nlen) +{ + return ALIGN(sizeof(struct au_vdir_de) + nlen, sizeof(ino_t)); +} + +static int set_deblk_end(union au_vdir_deblk_p *p, + union au_vdir_deblk_p *deblk_end) +{ + if (calc_size(0) <= deblk_end->deblk - p->deblk) { + p->de->de_str.len = 0; + /* smp_mb(); */ + return 0; + } + return -1; /* error */ +} + +/* returns true or false */ +static int is_deblk_end(union au_vdir_deblk_p *p, + union au_vdir_deblk_p *deblk_end) +{ + if (calc_size(0) <= deblk_end->deblk - p->deblk) + return !p->de->de_str.len; + return 1; +} + +static unsigned char *last_deblk(struct au_vdir *vdir) +{ + return vdir->vd_deblk[vdir->vd_nblk - 1]; +} + +/* ---------------------------------------------------------------------- */ + +/* estimate the apropriate size for name hash table */ +unsigned int au_rdhash_est(loff_t sz) +{ + unsigned int n; + + n = UINT_MAX; + sz >>= 10; + if (sz < n) + n = sz; + if (sz < AUFS_RDHASH_DEF) + n = AUFS_RDHASH_DEF; + /* pr_info("n %u\n", n); */ + return n; +} + +/* + * the allocated memory has to be freed by + * au_nhash_wh_free() or au_nhash_de_free(). + */ +int au_nhash_alloc(struct au_nhash *nhash, unsigned int num_hash, gfp_t gfp) +{ + struct hlist_head *head; + unsigned int u; + size_t sz; + + sz = sizeof(*nhash->nh_head) * num_hash; + head = kmalloc(sz, gfp); + if (head) { + nhash->nh_num = num_hash; + nhash->nh_head = head; + for (u = 0; u < num_hash; u++) + INIT_HLIST_HEAD(head++); + return 0; /* success */ + } + + return -ENOMEM; +} + +static void nhash_count(struct hlist_head *head) +{ +#if 0 + unsigned long n; + struct hlist_node *pos; + + n = 0; + hlist_for_each(pos, head) + n++; + pr_info("%lu\n", n); +#endif +} + +static void au_nhash_wh_do_free(struct hlist_head *head) +{ + struct au_vdir_wh *pos; + struct hlist_node *node; + + hlist_for_each_entry_safe(pos, node, head, wh_hash) + kfree(pos); +} + +static void au_nhash_de_do_free(struct hlist_head *head) +{ + struct au_vdir_dehstr *pos; + struct hlist_node *node; + + hlist_for_each_entry_safe(pos, node, head, hash) + au_cache_free_vdir_dehstr(pos); +} + +static void au_nhash_do_free(struct au_nhash *nhash, + void (*free)(struct hlist_head *head)) +{ + unsigned int n; + struct hlist_head *head; + + n = nhash->nh_num; + if (!n) + return; + + head = nhash->nh_head; + while (n-- > 0) { + nhash_count(head); + free(head++); + } + kfree(nhash->nh_head); +} + +void au_nhash_wh_free(struct au_nhash *whlist) +{ + au_nhash_do_free(whlist, au_nhash_wh_do_free); +} + +static void au_nhash_de_free(struct au_nhash *delist) +{ + au_nhash_do_free(delist, au_nhash_de_do_free); +} + +/* ---------------------------------------------------------------------- */ + +int au_nhash_test_longer_wh(struct au_nhash *whlist, aufs_bindex_t btgt, + int limit) +{ + int num; + unsigned int u, n; + struct hlist_head *head; + struct au_vdir_wh *pos; + + num = 0; + n = whlist->nh_num; + head = whlist->nh_head; + for (u = 0; u < n; u++, head++) + hlist_for_each_entry(pos, head, wh_hash) + if (pos->wh_bindex == btgt && ++num > limit) + return 1; + return 0; +} + +static struct hlist_head *au_name_hash(struct au_nhash *nhash, + unsigned char *name, + unsigned int len) +{ + unsigned int v; + /* const unsigned int magic_bit = 12; */ + + AuDebugOn(!nhash->nh_num || !nhash->nh_head); + + v = 0; + while (len--) + v += *name++; + /* v = hash_long(v, magic_bit); */ + v %= nhash->nh_num; + return nhash->nh_head + v; +} + +static int au_nhash_test_name(struct au_vdir_destr *str, const char *name, + int nlen) +{ + return str->len == nlen && !memcmp(str->name, name, nlen); +} + +/* returns found or not */ +int au_nhash_test_known_wh(struct au_nhash *whlist, char *name, int nlen) +{ + struct hlist_head *head; + struct au_vdir_wh *pos; + struct au_vdir_destr *str; + + head = au_name_hash(whlist, name, nlen); + hlist_for_each_entry(pos, head, wh_hash) { + str = &pos->wh_str; + AuDbg("%.*s\n", str->len, str->name); + if (au_nhash_test_name(str, name, nlen)) + return 1; + } + return 0; +} + +/* returns found(true) or not */ +static int test_known(struct au_nhash *delist, char *name, int nlen) +{ + struct hlist_head *head; + struct au_vdir_dehstr *pos; + struct au_vdir_destr *str; + + head = au_name_hash(delist, name, nlen); + hlist_for_each_entry(pos, head, hash) { + str = pos->str; + AuDbg("%.*s\n", str->len, str->name); + if (au_nhash_test_name(str, name, nlen)) + return 1; + } + return 0; +} + +static void au_shwh_init_wh(struct au_vdir_wh *wh, ino_t ino, + unsigned char d_type) +{ +#ifdef CONFIG_AUFS_SHWH + wh->wh_ino = ino; + wh->wh_type = d_type; +#endif +} + +/* ---------------------------------------------------------------------- */ + +int au_nhash_append_wh(struct au_nhash *whlist, char *name, int nlen, ino_t ino, + unsigned int d_type, aufs_bindex_t bindex, + unsigned char shwh) +{ + int err; + struct au_vdir_destr *str; + struct au_vdir_wh *wh; + + AuDbg("%.*s\n", nlen, name); + AuDebugOn(!whlist->nh_num || !whlist->nh_head); + + err = -ENOMEM; + wh = kmalloc(sizeof(*wh) + nlen, GFP_NOFS); + if (unlikely(!wh)) + goto out; + + err = 0; + wh->wh_bindex = bindex; + if (shwh) + au_shwh_init_wh(wh, ino, d_type); + str = &wh->wh_str; + str->len = nlen; + memcpy(str->name, name, nlen); + hlist_add_head(&wh->wh_hash, au_name_hash(whlist, name, nlen)); + /* smp_mb(); */ + +out: + return err; +} + +static int append_deblk(struct au_vdir *vdir) +{ + int err; + unsigned long ul; + const unsigned int deblk_sz = vdir->vd_deblk_sz; + union au_vdir_deblk_p p, deblk_end; + unsigned char **o; + + err = -ENOMEM; + o = krealloc(vdir->vd_deblk, sizeof(*o) * (vdir->vd_nblk + 1), + GFP_NOFS); + if (unlikely(!o)) + goto out; + + vdir->vd_deblk = o; + p.deblk = kmalloc(deblk_sz, GFP_NOFS); + if (p.deblk) { + ul = vdir->vd_nblk++; + vdir->vd_deblk[ul] = p.deblk; + vdir->vd_last.ul = ul; + vdir->vd_last.p.deblk = p.deblk; + deblk_end.deblk = p.deblk + deblk_sz; + err = set_deblk_end(&p, &deblk_end); + } + +out: + return err; +} + +static int append_de(struct au_vdir *vdir, char *name, int nlen, ino_t ino, + unsigned int d_type, struct au_nhash *delist) +{ + int err; + unsigned int sz; + const unsigned int deblk_sz = vdir->vd_deblk_sz; + union au_vdir_deblk_p p, *room, deblk_end; + struct au_vdir_dehstr *dehstr; + + p.deblk = last_deblk(vdir); + deblk_end.deblk = p.deblk + deblk_sz; + room = &vdir->vd_last.p; + AuDebugOn(room->deblk < p.deblk || deblk_end.deblk <= room->deblk + || !is_deblk_end(room, &deblk_end)); + + sz = calc_size(nlen); + if (unlikely(sz > deblk_end.deblk - room->deblk)) { + err = append_deblk(vdir); + if (unlikely(err)) + goto out; + + p.deblk = last_deblk(vdir); + deblk_end.deblk = p.deblk + deblk_sz; + /* smp_mb(); */ + AuDebugOn(room->deblk != p.deblk); + } + + err = -ENOMEM; + dehstr = au_cache_alloc_vdir_dehstr(); + if (unlikely(!dehstr)) + goto out; + + dehstr->str = &room->de->de_str; + hlist_add_head(&dehstr->hash, au_name_hash(delist, name, nlen)); + room->de->de_ino = ino; + room->de->de_type = d_type; + room->de->de_str.len = nlen; + memcpy(room->de->de_str.name, name, nlen); + + err = 0; + room->deblk += sz; + if (unlikely(set_deblk_end(room, &deblk_end))) + err = append_deblk(vdir); + /* smp_mb(); */ + +out: + return err; +} + +/* ---------------------------------------------------------------------- */ + +void au_vdir_free(struct au_vdir *vdir) +{ + unsigned char **deblk; + + deblk = vdir->vd_deblk; + while (vdir->vd_nblk--) + kfree(*deblk++); + kfree(vdir->vd_deblk); + au_cache_free_vdir(vdir); +} + +static struct au_vdir *alloc_vdir(struct file *file) +{ + struct au_vdir *vdir; + struct super_block *sb; + int err; + + sb = file->f_dentry->d_sb; + SiMustAnyLock(sb); + + err = -ENOMEM; + vdir = au_cache_alloc_vdir(); + if (unlikely(!vdir)) + goto out; + + vdir->vd_deblk = kzalloc(sizeof(*vdir->vd_deblk), GFP_NOFS); + if (unlikely(!vdir->vd_deblk)) + goto out_free; + + vdir->vd_deblk_sz = au_sbi(sb)->si_rdblk; + if (!vdir->vd_deblk_sz) { + /* estimate the apropriate size for deblk */ + vdir->vd_deblk_sz = au_dir_size(file, /*dentry*/NULL); + /* pr_info("vd_deblk_sz %u\n", vdir->vd_deblk_sz); */ + } + vdir->vd_nblk = 0; + vdir->vd_version = 0; + vdir->vd_jiffy = 0; + err = append_deblk(vdir); + if (!err) + return vdir; /* success */ + + kfree(vdir->vd_deblk); + +out_free: + au_cache_free_vdir(vdir); +out: + vdir = ERR_PTR(err); + return vdir; +} + +static int reinit_vdir(struct au_vdir *vdir) +{ + int err; + union au_vdir_deblk_p p, deblk_end; + + while (vdir->vd_nblk > 1) { + kfree(vdir->vd_deblk[vdir->vd_nblk - 1]); + /* vdir->vd_deblk[vdir->vd_nblk - 1] = NULL; */ + vdir->vd_nblk--; + } + p.deblk = vdir->vd_deblk[0]; + deblk_end.deblk = p.deblk + vdir->vd_deblk_sz; + err = set_deblk_end(&p, &deblk_end); + /* keep vd_dblk_sz */ + vdir->vd_last.ul = 0; + vdir->vd_last.p.deblk = vdir->vd_deblk[0]; + vdir->vd_version = 0; + vdir->vd_jiffy = 0; + /* smp_mb(); */ + return err; +} + +/* ---------------------------------------------------------------------- */ + +#define AuFillVdir_CALLED 1 +#define AuFillVdir_WHABLE (1 << 1) +#define AuFillVdir_SHWH (1 << 2) +#define au_ftest_fillvdir(flags, name) ((flags) & AuFillVdir_##name) +#define au_fset_fillvdir(flags, name) \ + do { (flags) |= AuFillVdir_##name; } while (0) +#define au_fclr_fillvdir(flags, name) \ + do { (flags) &= ~AuFillVdir_##name; } while (0) + +#ifndef CONFIG_AUFS_SHWH +#undef AuFillVdir_SHWH +#define AuFillVdir_SHWH 0 +#endif + +struct fillvdir_arg { + struct file *file; + struct au_vdir *vdir; + struct au_nhash delist; + struct au_nhash whlist; + aufs_bindex_t bindex; + unsigned int flags; + int err; +}; + +static int fillvdir(void *__arg, const char *__name, int nlen, + loff_t offset __maybe_unused, u64 h_ino, + unsigned int d_type) +{ + struct fillvdir_arg *arg = __arg; + char *name = (void *)__name; + struct super_block *sb; + ino_t ino; + const unsigned char shwh = !!au_ftest_fillvdir(arg->flags, SHWH); + + arg->err = 0; + sb = arg->file->f_dentry->d_sb; + au_fset_fillvdir(arg->flags, CALLED); + /* smp_mb(); */ + if (nlen <= AUFS_WH_PFX_LEN + || memcmp(name, AUFS_WH_PFX, AUFS_WH_PFX_LEN)) { + if (test_known(&arg->delist, name, nlen) + || au_nhash_test_known_wh(&arg->whlist, name, nlen)) + goto out; /* already exists or whiteouted */ + + sb = arg->file->f_dentry->d_sb; + arg->err = au_ino(sb, arg->bindex, h_ino, d_type, &ino); + if (!arg->err) { + if (unlikely(nlen > AUFS_MAX_NAMELEN)) + d_type = DT_UNKNOWN; + arg->err = append_de(arg->vdir, name, nlen, ino, + d_type, &arg->delist); + } + } else if (au_ftest_fillvdir(arg->flags, WHABLE)) { + name += AUFS_WH_PFX_LEN; + nlen -= AUFS_WH_PFX_LEN; + if (au_nhash_test_known_wh(&arg->whlist, name, nlen)) + goto out; /* already whiteouted */ + + if (shwh) + arg->err = au_wh_ino(sb, arg->bindex, h_ino, d_type, + &ino); + if (!arg->err) { + if (nlen <= AUFS_MAX_NAMELEN + AUFS_WH_PFX_LEN) + d_type = DT_UNKNOWN; + arg->err = au_nhash_append_wh + (&arg->whlist, name, nlen, ino, d_type, + arg->bindex, shwh); + } + } + +out: + if (!arg->err) + arg->vdir->vd_jiffy = jiffies; + /* smp_mb(); */ + AuTraceErr(arg->err); + return arg->err; +} + +static int au_handle_shwh(struct super_block *sb, struct au_vdir *vdir, + struct au_nhash *whlist, struct au_nhash *delist) +{ +#ifdef CONFIG_AUFS_SHWH + int err; + unsigned int nh, u; + struct hlist_head *head; + struct au_vdir_wh *pos; + struct hlist_node *n; + char *p, *o; + struct au_vdir_destr *destr; + + AuDebugOn(!au_opt_test(au_mntflags(sb), SHWH)); + + err = -ENOMEM; + o = p = (void *)__get_free_page(GFP_NOFS); + if (unlikely(!p)) + goto out; + + err = 0; + nh = whlist->nh_num; + memcpy(p, AUFS_WH_PFX, AUFS_WH_PFX_LEN); + p += AUFS_WH_PFX_LEN; + for (u = 0; u < nh; u++) { + head = whlist->nh_head + u; + hlist_for_each_entry_safe(pos, n, head, wh_hash) { + destr = &pos->wh_str; + memcpy(p, destr->name, destr->len); + err = append_de(vdir, o, destr->len + AUFS_WH_PFX_LEN, + pos->wh_ino, pos->wh_type, delist); + if (unlikely(err)) + break; + } + } + + free_page((unsigned long)o); + +out: + AuTraceErr(err); + return err; +#else + return 0; +#endif +} + +static int au_do_read_vdir(struct fillvdir_arg *arg) +{ + int err; + unsigned int rdhash; + loff_t offset; + aufs_bindex_t bend, bindex, bstart; + unsigned char shwh; + struct file *hf, *file; + struct super_block *sb; + + file = arg->file; + sb = file->f_dentry->d_sb; + SiMustAnyLock(sb); + + rdhash = au_sbi(sb)->si_rdhash; + if (!rdhash) + rdhash = au_rdhash_est(au_dir_size(file, /*dentry*/NULL)); + err = au_nhash_alloc(&arg->delist, rdhash, GFP_NOFS); + if (unlikely(err)) + goto out; + err = au_nhash_alloc(&arg->whlist, rdhash, GFP_NOFS); + if (unlikely(err)) + goto out_delist; + + err = 0; + arg->flags = 0; + shwh = 0; + if (au_opt_test(au_mntflags(sb), SHWH)) { + shwh = 1; + au_fset_fillvdir(arg->flags, SHWH); + } + bstart = au_fbstart(file); + bend = au_fbend_dir(file); + for (bindex = bstart; !err && bindex <= bend; bindex++) { + hf = au_hf_dir(file, bindex); + if (!hf) + continue; + + offset = vfsub_llseek(hf, 0, SEEK_SET); + err = offset; + if (unlikely(offset)) + break; + + arg->bindex = bindex; + au_fclr_fillvdir(arg->flags, WHABLE); + if (shwh + || (bindex != bend + && au_br_whable(au_sbr_perm(sb, bindex)))) + au_fset_fillvdir(arg->flags, WHABLE); + do { + arg->err = 0; + au_fclr_fillvdir(arg->flags, CALLED); + /* smp_mb(); */ + err = vfsub_readdir(hf, fillvdir, arg); + if (err >= 0) + err = arg->err; + } while (!err && au_ftest_fillvdir(arg->flags, CALLED)); + } + + if (!err && shwh) + err = au_handle_shwh(sb, arg->vdir, &arg->whlist, &arg->delist); + + au_nhash_wh_free(&arg->whlist); + +out_delist: + au_nhash_de_free(&arg->delist); +out: + return err; +} + +static int read_vdir(struct file *file, int may_read) +{ + int err; + unsigned long expire; + unsigned char do_read; + struct fillvdir_arg arg; + struct inode *inode; + struct au_vdir *vdir, *allocated; + + err = 0; + inode = file_inode(file); + IMustLock(inode); + SiMustAnyLock(inode->i_sb); + + allocated = NULL; + do_read = 0; + expire = au_sbi(inode->i_sb)->si_rdcache; + vdir = au_ivdir(inode); + if (!vdir) { + do_read = 1; + vdir = alloc_vdir(file); + err = PTR_ERR(vdir); + if (IS_ERR(vdir)) + goto out; + err = 0; + allocated = vdir; + } else if (may_read + && (inode->i_version != vdir->vd_version + || time_after(jiffies, vdir->vd_jiffy + expire))) { + do_read = 1; + err = reinit_vdir(vdir); + if (unlikely(err)) + goto out; + } + + if (!do_read) + return 0; /* success */ + + arg.file = file; + arg.vdir = vdir; + err = au_do_read_vdir(&arg); + if (!err) { + /* file->f_pos = 0; */ + vdir->vd_version = inode->i_version; + vdir->vd_last.ul = 0; + vdir->vd_last.p.deblk = vdir->vd_deblk[0]; + if (allocated) + au_set_ivdir(inode, allocated); + } else if (allocated) + au_vdir_free(allocated); + +out: + return err; +} + +static int copy_vdir(struct au_vdir *tgt, struct au_vdir *src) +{ + int err, rerr; + unsigned long ul, n; + const unsigned int deblk_sz = src->vd_deblk_sz; + + AuDebugOn(tgt->vd_nblk != 1); + + err = -ENOMEM; + if (tgt->vd_nblk < src->vd_nblk) { + unsigned char **p; + + p = krealloc(tgt->vd_deblk, sizeof(*p) * src->vd_nblk, + GFP_NOFS); + if (unlikely(!p)) + goto out; + tgt->vd_deblk = p; + } + + if (tgt->vd_deblk_sz != deblk_sz) { + unsigned char *p; + + tgt->vd_deblk_sz = deblk_sz; + p = krealloc(tgt->vd_deblk[0], deblk_sz, GFP_NOFS); + if (unlikely(!p)) + goto out; + tgt->vd_deblk[0] = p; + } + memcpy(tgt->vd_deblk[0], src->vd_deblk[0], deblk_sz); + tgt->vd_version = src->vd_version; + tgt->vd_jiffy = src->vd_jiffy; + + n = src->vd_nblk; + for (ul = 1; ul < n; ul++) { + tgt->vd_deblk[ul] = kmemdup(src->vd_deblk[ul], deblk_sz, + GFP_NOFS); + if (unlikely(!tgt->vd_deblk[ul])) + goto out; + tgt->vd_nblk++; + } + tgt->vd_nblk = n; + tgt->vd_last.ul = tgt->vd_last.ul; + tgt->vd_last.p.deblk = tgt->vd_deblk[tgt->vd_last.ul]; + tgt->vd_last.p.deblk += src->vd_last.p.deblk + - src->vd_deblk[src->vd_last.ul]; + /* smp_mb(); */ + return 0; /* success */ + +out: + rerr = reinit_vdir(tgt); + BUG_ON(rerr); + return err; +} + +int au_vdir_init(struct file *file) +{ + int err; + struct inode *inode; + struct au_vdir *vdir_cache, *allocated; + + err = read_vdir(file, !file->f_pos); + if (unlikely(err)) + goto out; + + allocated = NULL; + vdir_cache = au_fvdir_cache(file); + if (!vdir_cache) { + vdir_cache = alloc_vdir(file); + err = PTR_ERR(vdir_cache); + if (IS_ERR(vdir_cache)) + goto out; + allocated = vdir_cache; + } else if (!file->f_pos && vdir_cache->vd_version != file->f_version) { + err = reinit_vdir(vdir_cache); + if (unlikely(err)) + goto out; + } else + return 0; /* success */ + + inode = file_inode(file); + err = copy_vdir(vdir_cache, au_ivdir(inode)); + if (!err) { + file->f_version = inode->i_version; + if (allocated) + au_set_fvdir_cache(file, allocated); + } else if (allocated) + au_vdir_free(allocated); + +out: + return err; +} + +static loff_t calc_offset(struct au_vdir *vdir) +{ + loff_t offset; + union au_vdir_deblk_p p; + + p.deblk = vdir->vd_deblk[vdir->vd_last.ul]; + offset = vdir->vd_last.p.deblk - p.deblk; + offset += vdir->vd_deblk_sz * vdir->vd_last.ul; + return offset; +} + +/* returns true or false */ +static int seek_vdir(struct file *file) +{ + int valid; + unsigned int deblk_sz; + unsigned long ul, n; + loff_t offset; + union au_vdir_deblk_p p, deblk_end; + struct au_vdir *vdir_cache; + + valid = 1; + vdir_cache = au_fvdir_cache(file); + offset = calc_offset(vdir_cache); + AuDbg("offset %lld\n", offset); + if (file->f_pos == offset) + goto out; + + vdir_cache->vd_last.ul = 0; + vdir_cache->vd_last.p.deblk = vdir_cache->vd_deblk[0]; + if (!file->f_pos) + goto out; + + valid = 0; + deblk_sz = vdir_cache->vd_deblk_sz; + ul = div64_u64(file->f_pos, deblk_sz); + AuDbg("ul %lu\n", ul); + if (ul >= vdir_cache->vd_nblk) + goto out; + + n = vdir_cache->vd_nblk; + for (; ul < n; ul++) { + p.deblk = vdir_cache->vd_deblk[ul]; + deblk_end.deblk = p.deblk + deblk_sz; + offset = ul; + offset *= deblk_sz; + while (!is_deblk_end(&p, &deblk_end) && offset < file->f_pos) { + unsigned int l; + + l = calc_size(p.de->de_str.len); + offset += l; + p.deblk += l; + } + if (!is_deblk_end(&p, &deblk_end)) { + valid = 1; + vdir_cache->vd_last.ul = ul; + vdir_cache->vd_last.p = p; + break; + } + } + +out: + /* smp_mb(); */ + AuTraceErr(!valid); + return valid; +} + +int au_vdir_fill_de(struct file *file, void *dirent, filldir_t filldir) +{ + int err; + unsigned int l, deblk_sz; + union au_vdir_deblk_p deblk_end; + struct au_vdir *vdir_cache; + struct au_vdir_de *de; + + vdir_cache = au_fvdir_cache(file); + if (!seek_vdir(file)) + return 0; + + deblk_sz = vdir_cache->vd_deblk_sz; + while (1) { + deblk_end.deblk = vdir_cache->vd_deblk[vdir_cache->vd_last.ul]; + deblk_end.deblk += deblk_sz; + while (!is_deblk_end(&vdir_cache->vd_last.p, &deblk_end)) { + de = vdir_cache->vd_last.p.de; + AuDbg("%.*s, off%lld, i%lu, dt%d\n", + de->de_str.len, de->de_str.name, file->f_pos, + (unsigned long)de->de_ino, de->de_type); + err = filldir(dirent, de->de_str.name, de->de_str.len, + file->f_pos, de->de_ino, de->de_type); + if (unlikely(err)) { + AuTraceErr(err); + /* todo: ignore the error caused by udba? */ + /* return err; */ + return 0; + } + + l = calc_size(de->de_str.len); + vdir_cache->vd_last.p.deblk += l; + file->f_pos += l; + } + if (vdir_cache->vd_last.ul < vdir_cache->vd_nblk - 1) { + vdir_cache->vd_last.ul++; + vdir_cache->vd_last.p.deblk + = vdir_cache->vd_deblk[vdir_cache->vd_last.ul]; + file->f_pos = deblk_sz * vdir_cache->vd_last.ul; + continue; + } + break; + } + + /* smp_mb(); */ + return 0; +} diff -Naur kernel.21.3.orig/fs/aufs/vfsub.c kernel.21.3.new/fs/aufs/vfsub.c --- kernel.21.3.orig/fs/aufs/vfsub.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/vfsub.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,782 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * sub-routines for VFS + */ + +#include +#include +#include +#include +#include "aufs.h" + +int vfsub_update_h_iattr(struct path *h_path, int *did) +{ + int err; + struct kstat st; + struct super_block *h_sb; + + /* for remote fs, leave work for its getattr or d_revalidate */ + /* for bad i_attr fs, handle them in aufs_getattr() */ + /* still some fs may acquire i_mutex. we need to skip them */ + err = 0; + if (!did) + did = &err; + h_sb = h_path->dentry->d_sb; + *did = (!au_test_fs_remote(h_sb) && au_test_fs_refresh_iattr(h_sb)); + if (*did) + err = vfs_getattr(h_path, &st); + + return err; +} + +/* ---------------------------------------------------------------------- */ + +struct file *vfsub_dentry_open(struct path *path, int flags) +{ + struct file *file; + + file = dentry_open(path, flags /* | __FMODE_NONOTIFY */, + current_cred()); + if (!IS_ERR_OR_NULL(file) + && (file->f_mode & (FMODE_READ | FMODE_WRITE)) == FMODE_READ) + i_readcount_inc(path->dentry->d_inode); + + return file; +} + +struct file *vfsub_filp_open(const char *path, int oflags, int mode) +{ + struct file *file; + + lockdep_off(); + file = filp_open(path, + oflags /* | __FMODE_NONOTIFY */, + mode); + lockdep_on(); + if (IS_ERR(file)) + goto out; + vfsub_update_h_iattr(&file->f_path, /*did*/NULL); /*ignore*/ + +out: + return file; +} + +int vfsub_kern_path(const char *name, unsigned int flags, struct path *path) +{ + int err; + + err = kern_path(name, flags, path); + if (!err && path->dentry->d_inode) + vfsub_update_h_iattr(path, /*did*/NULL); /*ignore*/ + return err; +} + +struct dentry *vfsub_lookup_one_len(const char *name, struct dentry *parent, + int len) +{ + struct path path = { + .mnt = NULL + }; + + /* VFS checks it too, but by WARN_ON_ONCE() */ + IMustLock(parent->d_inode); + + path.dentry = lookup_one_len(name, parent, len); + if (IS_ERR(path.dentry)) + goto out; + if (path.dentry->d_inode) + vfsub_update_h_iattr(&path, /*did*/NULL); /*ignore*/ + +out: + AuTraceErrPtr(path.dentry); + return path.dentry; +} + +void vfsub_call_lkup_one(void *args) +{ + struct vfsub_lkup_one_args *a = args; + *a->errp = vfsub_lkup_one(a->name, a->parent); +} + +/* ---------------------------------------------------------------------- */ + +struct dentry *vfsub_lock_rename(struct dentry *d1, struct au_hinode *hdir1, + struct dentry *d2, struct au_hinode *hdir2) +{ + struct dentry *d; + + lockdep_off(); + d = lock_rename(d1, d2); + lockdep_on(); + au_hn_suspend(hdir1); + if (hdir1 != hdir2) + au_hn_suspend(hdir2); + + return d; +} + +void vfsub_unlock_rename(struct dentry *d1, struct au_hinode *hdir1, + struct dentry *d2, struct au_hinode *hdir2) +{ + au_hn_resume(hdir1); + if (hdir1 != hdir2) + au_hn_resume(hdir2); + lockdep_off(); + unlock_rename(d1, d2); + lockdep_on(); +} + +/* ---------------------------------------------------------------------- */ + +int vfsub_create(struct inode *dir, struct path *path, int mode, bool want_excl) +{ + int err; + struct dentry *d; + + IMustLock(dir); + + d = path->dentry; + path->dentry = d->d_parent; + err = security_path_mknod(path, d, mode, 0); + path->dentry = d; + if (unlikely(err)) + goto out; + + lockdep_off(); + err = vfs_create(dir, path->dentry, mode, want_excl); + lockdep_on(); + if (!err) { + struct path tmp = *path; + int did; + + vfsub_update_h_iattr(&tmp, &did); + if (did) { + tmp.dentry = path->dentry->d_parent; + vfsub_update_h_iattr(&tmp, /*did*/NULL); + } + /*ignore*/ + } + +out: + return err; +} + +int vfsub_symlink(struct inode *dir, struct path *path, const char *symname) +{ + int err; + struct dentry *d; + + IMustLock(dir); + + d = path->dentry; + path->dentry = d->d_parent; + err = security_path_symlink(path, d, symname); + path->dentry = d; + if (unlikely(err)) + goto out; + + lockdep_off(); + err = vfs_symlink(dir, path->dentry, symname); + lockdep_on(); + if (!err) { + struct path tmp = *path; + int did; + + vfsub_update_h_iattr(&tmp, &did); + if (did) { + tmp.dentry = path->dentry->d_parent; + vfsub_update_h_iattr(&tmp, /*did*/NULL); + } + /*ignore*/ + } + +out: + return err; +} + +int vfsub_mknod(struct inode *dir, struct path *path, int mode, dev_t dev) +{ + int err; + struct dentry *d; + + IMustLock(dir); + + d = path->dentry; + path->dentry = d->d_parent; + err = security_path_mknod(path, d, mode, new_encode_dev(dev)); + path->dentry = d; + if (unlikely(err)) + goto out; + + lockdep_off(); + err = vfs_mknod(dir, path->dentry, mode, dev); + lockdep_on(); + if (!err) { + struct path tmp = *path; + int did; + + vfsub_update_h_iattr(&tmp, &did); + if (did) { + tmp.dentry = path->dentry->d_parent; + vfsub_update_h_iattr(&tmp, /*did*/NULL); + } + /*ignore*/ + } + +out: + return err; +} + +static int au_test_nlink(struct inode *inode) +{ + const unsigned int link_max = UINT_MAX >> 1; /* rough margin */ + + if (!au_test_fs_no_limit_nlink(inode->i_sb) + || inode->i_nlink < link_max) + return 0; + return -EMLINK; +} + +int vfsub_link(struct dentry *src_dentry, struct inode *dir, struct path *path) +{ + int err; + struct dentry *d; + + IMustLock(dir); + + err = au_test_nlink(src_dentry->d_inode); + if (unlikely(err)) + return err; + + /* we don't call may_linkat() */ + d = path->dentry; + path->dentry = d->d_parent; + err = security_path_link(src_dentry, path, d); + path->dentry = d; + if (unlikely(err)) + goto out; + + lockdep_off(); + err = vfs_link(src_dentry, dir, path->dentry); + lockdep_on(); + if (!err) { + struct path tmp = *path; + int did; + + /* fuse has different memory inode for the same inumber */ + vfsub_update_h_iattr(&tmp, &did); + if (did) { + tmp.dentry = path->dentry->d_parent; + vfsub_update_h_iattr(&tmp, /*did*/NULL); + tmp.dentry = src_dentry; + vfsub_update_h_iattr(&tmp, /*did*/NULL); + } + /*ignore*/ + } + +out: + return err; +} + +int vfsub_rename(struct inode *src_dir, struct dentry *src_dentry, + struct inode *dir, struct path *path) +{ + int err; + struct path tmp = { + .mnt = path->mnt + }; + struct dentry *d; + + IMustLock(dir); + IMustLock(src_dir); + + d = path->dentry; + path->dentry = d->d_parent; + tmp.dentry = src_dentry->d_parent; + err = security_path_rename(&tmp, src_dentry, path, d); + path->dentry = d; + if (unlikely(err)) + goto out; + + lockdep_off(); + err = vfs_rename(src_dir, src_dentry, dir, path->dentry); + lockdep_on(); + if (!err) { + int did; + + tmp.dentry = d->d_parent; + vfsub_update_h_iattr(&tmp, &did); + if (did) { + tmp.dentry = src_dentry; + vfsub_update_h_iattr(&tmp, /*did*/NULL); + tmp.dentry = src_dentry->d_parent; + vfsub_update_h_iattr(&tmp, /*did*/NULL); + } + /*ignore*/ + } + +out: + return err; +} + +int vfsub_mkdir(struct inode *dir, struct path *path, int mode) +{ + int err; + struct dentry *d; + + IMustLock(dir); + + d = path->dentry; + path->dentry = d->d_parent; + err = security_path_mkdir(path, d, mode); + path->dentry = d; + if (unlikely(err)) + goto out; + + lockdep_off(); + err = vfs_mkdir(dir, path->dentry, mode); + lockdep_on(); + if (!err) { + struct path tmp = *path; + int did; + + vfsub_update_h_iattr(&tmp, &did); + if (did) { + tmp.dentry = path->dentry->d_parent; + vfsub_update_h_iattr(&tmp, /*did*/NULL); + } + /*ignore*/ + } + +out: + return err; +} + +int vfsub_rmdir(struct inode *dir, struct path *path) +{ + int err; + struct dentry *d; + + IMustLock(dir); + + d = path->dentry; + path->dentry = d->d_parent; + err = security_path_rmdir(path, d); + path->dentry = d; + if (unlikely(err)) + goto out; + + lockdep_off(); + err = vfs_rmdir(dir, path->dentry); + lockdep_on(); + if (!err) { + struct path tmp = { + .dentry = path->dentry->d_parent, + .mnt = path->mnt + }; + + vfsub_update_h_iattr(&tmp, /*did*/NULL); /*ignore*/ + } + +out: + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* todo: support mmap_sem? */ +ssize_t vfsub_read_u(struct file *file, char __user *ubuf, size_t count, + loff_t *ppos) +{ + ssize_t err; + + lockdep_off(); + err = vfs_read(file, ubuf, count, ppos); + lockdep_on(); + if (err >= 0) + vfsub_update_h_iattr(&file->f_path, /*did*/NULL); /*ignore*/ + return err; +} + +/* todo: kernel_read()? */ +ssize_t vfsub_read_k(struct file *file, void *kbuf, size_t count, + loff_t *ppos) +{ + ssize_t err; + mm_segment_t oldfs; + union { + void *k; + char __user *u; + } buf; + + buf.k = kbuf; + oldfs = get_fs(); + set_fs(KERNEL_DS); + err = vfsub_read_u(file, buf.u, count, ppos); + set_fs(oldfs); + return err; +} + +ssize_t vfsub_write_u(struct file *file, const char __user *ubuf, size_t count, + loff_t *ppos) +{ + ssize_t err; + + lockdep_off(); + err = vfs_write(file, ubuf, count, ppos); + lockdep_on(); + if (err >= 0) + vfsub_update_h_iattr(&file->f_path, /*did*/NULL); /*ignore*/ + return err; +} + +ssize_t vfsub_write_k(struct file *file, void *kbuf, size_t count, loff_t *ppos) +{ + ssize_t err; + mm_segment_t oldfs; + union { + void *k; + const char __user *u; + } buf; + + buf.k = kbuf; + oldfs = get_fs(); + set_fs(KERNEL_DS); + err = vfsub_write_u(file, buf.u, count, ppos); + set_fs(oldfs); + return err; +} + +int vfsub_flush(struct file *file, fl_owner_t id) +{ + int err; + + err = 0; + if (file->f_op && file->f_op->flush) { + if (!au_test_nfs(file->f_dentry->d_sb)) + err = file->f_op->flush(file, id); + else { + lockdep_off(); + err = file->f_op->flush(file, id); + lockdep_on(); + } + if (!err) + vfsub_update_h_iattr(&file->f_path, /*did*/NULL); + /*ignore*/ + } + return err; +} + +int vfsub_readdir(struct file *file, filldir_t filldir, void *arg) +{ + int err; + + lockdep_off(); + err = vfs_readdir(file, filldir, arg); + lockdep_on(); + if (err >= 0) + vfsub_update_h_iattr(&file->f_path, /*did*/NULL); /*ignore*/ + return err; +} + +long vfsub_splice_to(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, size_t len, + unsigned int flags) +{ + long err; + + lockdep_off(); + err = do_splice_to(in, ppos, pipe, len, flags); + lockdep_on(); + file_accessed(in); + if (err >= 0) + vfsub_update_h_iattr(&in->f_path, /*did*/NULL); /*ignore*/ + return err; +} + +long vfsub_splice_from(struct pipe_inode_info *pipe, struct file *out, + loff_t *ppos, size_t len, unsigned int flags) +{ + long err; + + lockdep_off(); + err = do_splice_from(pipe, out, ppos, len, flags); + lockdep_on(); + if (err >= 0) + vfsub_update_h_iattr(&out->f_path, /*did*/NULL); /*ignore*/ + return err; +} + +int vfsub_fsync(struct file *file, struct path *path, int datasync) +{ + int err; + + /* file can be NULL */ + lockdep_off(); + err = vfs_fsync(file, datasync); + lockdep_on(); + if (!err) { + if (!path) { + AuDebugOn(!file); + path = &file->f_path; + } + vfsub_update_h_iattr(path, /*did*/NULL); /*ignore*/ + } + return err; +} + +/* cf. open.c:do_sys_truncate() and do_sys_ftruncate() */ +int vfsub_trunc(struct path *h_path, loff_t length, unsigned int attr, + struct file *h_file) +{ + int err; + struct inode *h_inode; + struct super_block *h_sb; + + if (!h_file) { + err = vfsub_truncate(h_path, length); + goto out; + } + + h_inode = h_path->dentry->d_inode; + h_sb = h_inode->i_sb; + lockdep_off(); + sb_start_write(h_sb); + lockdep_on(); + err = locks_verify_truncate(h_inode, h_file, length); + if (!err) + err = security_path_truncate(h_path); + if (!err) { + lockdep_off(); + err = do_truncate(h_path->dentry, length, attr, h_file); + lockdep_on(); + } + lockdep_off(); + sb_end_write(h_sb); + lockdep_on(); + +out: + return err; +} + +/* ---------------------------------------------------------------------- */ + +struct au_vfsub_mkdir_args { + int *errp; + struct inode *dir; + struct path *path; + int mode; +}; + +static void au_call_vfsub_mkdir(void *args) +{ + struct au_vfsub_mkdir_args *a = args; + *a->errp = vfsub_mkdir(a->dir, a->path, a->mode); +} + +int vfsub_sio_mkdir(struct inode *dir, struct path *path, int mode) +{ + int err, do_sio, wkq_err; + + do_sio = au_test_h_perm_sio(dir, MAY_EXEC | MAY_WRITE); + if (!do_sio) { + lockdep_off(); + err = vfsub_mkdir(dir, path, mode); + lockdep_on(); + } else { + struct au_vfsub_mkdir_args args = { + .errp = &err, + .dir = dir, + .path = path, + .mode = mode + }; + wkq_err = au_wkq_wait(au_call_vfsub_mkdir, &args); + if (unlikely(wkq_err)) + err = wkq_err; + } + + return err; +} + +struct au_vfsub_rmdir_args { + int *errp; + struct inode *dir; + struct path *path; +}; + +static void au_call_vfsub_rmdir(void *args) +{ + struct au_vfsub_rmdir_args *a = args; + *a->errp = vfsub_rmdir(a->dir, a->path); +} + +int vfsub_sio_rmdir(struct inode *dir, struct path *path) +{ + int err, do_sio, wkq_err; + + do_sio = au_test_h_perm_sio(dir, MAY_EXEC | MAY_WRITE); + if (!do_sio) { + lockdep_off(); + err = vfsub_rmdir(dir, path); + lockdep_on(); + } else { + struct au_vfsub_rmdir_args args = { + .errp = &err, + .dir = dir, + .path = path + }; + wkq_err = au_wkq_wait(au_call_vfsub_rmdir, &args); + if (unlikely(wkq_err)) + err = wkq_err; + } + + return err; +} + +/* ---------------------------------------------------------------------- */ + +struct notify_change_args { + int *errp; + struct path *path; + struct iattr *ia; +}; + +static void call_notify_change(void *args) +{ + struct notify_change_args *a = args; + struct inode *h_inode; + + h_inode = a->path->dentry->d_inode; + IMustLock(h_inode); + + *a->errp = -EPERM; + if (!IS_IMMUTABLE(h_inode) && !IS_APPEND(h_inode)) { + lockdep_off(); + *a->errp = notify_change(a->path->dentry, a->ia); + lockdep_on(); + if (!*a->errp) + vfsub_update_h_iattr(a->path, /*did*/NULL); /*ignore*/ + } + AuTraceErr(*a->errp); +} + +int vfsub_notify_change(struct path *path, struct iattr *ia) +{ + int err; + struct notify_change_args args = { + .errp = &err, + .path = path, + .ia = ia + }; + + call_notify_change(&args); + + return err; +} + +int vfsub_sio_notify_change(struct path *path, struct iattr *ia) +{ + int err, wkq_err; + struct notify_change_args args = { + .errp = &err, + .path = path, + .ia = ia + }; + + wkq_err = au_wkq_wait(call_notify_change, &args); + if (unlikely(wkq_err)) + err = wkq_err; + + return err; +} + +/* ---------------------------------------------------------------------- */ + +struct unlink_args { + int *errp; + struct inode *dir; + struct path *path; +}; + +static void call_unlink(void *args) +{ + struct unlink_args *a = args; + struct dentry *d = a->path->dentry; + struct inode *h_inode; + const int stop_sillyrename = (au_test_nfs(d->d_sb) + && d->d_count == 1); + + IMustLock(a->dir); + + a->path->dentry = d->d_parent; + *a->errp = security_path_unlink(a->path, d); + a->path->dentry = d; + if (unlikely(*a->errp)) + return; + + if (!stop_sillyrename) + dget(d); + h_inode = d->d_inode; + if (h_inode) + ihold(h_inode); + + lockdep_off(); + *a->errp = vfs_unlink(a->dir, d); + lockdep_on(); + if (!*a->errp) { + struct path tmp = { + .dentry = d->d_parent, + .mnt = a->path->mnt + }; + vfsub_update_h_iattr(&tmp, /*did*/NULL); /*ignore*/ + } + + if (!stop_sillyrename) + dput(d); + if (h_inode) + iput(h_inode); + + AuTraceErr(*a->errp); +} + +/* + * @dir: must be locked. + * @dentry: target dentry. + */ +int vfsub_unlink(struct inode *dir, struct path *path, int force) +{ + int err; + struct unlink_args args = { + .errp = &err, + .dir = dir, + .path = path + }; + + if (!force) + call_unlink(&args); + else { + int wkq_err; + + wkq_err = au_wkq_wait(call_unlink, &args); + if (unlikely(wkq_err)) + err = wkq_err; + } + + return err; +} diff -Naur kernel.21.3.orig/fs/aufs/vfsub.h kernel.21.3.new/fs/aufs/vfsub.h --- kernel.21.3.orig/fs/aufs/vfsub.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/vfsub.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,329 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * sub-routines for VFS + */ + +#ifndef __AUFS_VFSUB_H__ +#define __AUFS_VFSUB_H__ + +#ifdef __KERNEL__ + +#include +#include +#include +#include +#include "debug.h" + +/* copied from linux/fs/internal.h */ +/* todo: BAD approach!! */ +extern struct lglock vfsmount_lock; +extern void __mnt_drop_write(struct vfsmount *); +extern spinlock_t inode_sb_list_lock; + +/* copied from linux/fs/file_table.c */ +extern struct lglock files_lglock; +#ifdef CONFIG_SMP +/* + * These macros iterate all files on all CPUs for a given superblock. + * files_lglock must be held globally. + */ +#define do_file_list_for_each_entry(__sb, __file) \ +{ \ + int i; \ + for_each_possible_cpu(i) { \ + struct list_head *list; \ + list = per_cpu_ptr((__sb)->s_files, i); \ + list_for_each_entry((__file), list, f_u.fu_list) + +#define while_file_list_for_each_entry \ + } \ +} + +#else + +#define do_file_list_for_each_entry(__sb, __file) \ +{ \ + struct list_head *list; \ + list = &(sb)->s_files; \ + list_for_each_entry((__file), list, f_u.fu_list) + +#define while_file_list_for_each_entry \ +} +#endif + +/* ---------------------------------------------------------------------- */ + +/* lock subclass for lower inode */ +/* default MAX_LOCKDEP_SUBCLASSES(8) is not enough */ +/* reduce? gave up. */ +enum { + AuLsc_I_Begin = I_MUTEX_QUOTA, /* 4 */ + AuLsc_I_PARENT, /* lower inode, parent first */ + AuLsc_I_PARENT2, /* copyup dirs */ + AuLsc_I_PARENT3, /* copyup wh */ + AuLsc_I_CHILD, + AuLsc_I_CHILD2, + AuLsc_I_End +}; + +/* to debug easier, do not make them inlined functions */ +#define MtxMustLock(mtx) AuDebugOn(!mutex_is_locked(mtx)) +#define IMustLock(i) MtxMustLock(&(i)->i_mutex) + +/* ---------------------------------------------------------------------- */ + +static inline void vfsub_drop_nlink(struct inode *inode) +{ + AuDebugOn(!inode->i_nlink); + drop_nlink(inode); +} + +static inline void vfsub_dead_dir(struct inode *inode) +{ + AuDebugOn(!S_ISDIR(inode->i_mode)); + inode->i_flags |= S_DEAD; + clear_nlink(inode); +} + +static inline int vfsub_native_ro(struct inode *inode) +{ + return (inode->i_sb->s_flags & MS_RDONLY) + || IS_RDONLY(inode) + /* || IS_APPEND(inode) */ + || IS_IMMUTABLE(inode); +} + +/* ---------------------------------------------------------------------- */ + +int vfsub_update_h_iattr(struct path *h_path, int *did); +struct file *vfsub_dentry_open(struct path *path, int flags); +struct file *vfsub_filp_open(const char *path, int oflags, int mode); +int vfsub_kern_path(const char *name, unsigned int flags, struct path *path); + +struct dentry *vfsub_lookup_one_len(const char *name, struct dentry *parent, + int len); + +struct vfsub_lkup_one_args { + struct dentry **errp; + struct qstr *name; + struct dentry *parent; +}; + +static inline struct dentry *vfsub_lkup_one(struct qstr *name, + struct dentry *parent) +{ + return vfsub_lookup_one_len(name->name, parent, name->len); +} + +void vfsub_call_lkup_one(void *args); + +/* ---------------------------------------------------------------------- */ + +static inline int vfsub_mnt_want_write(struct vfsmount *mnt) +{ + int err; + + lockdep_off(); + err = mnt_want_write(mnt); + lockdep_on(); + return err; +} + +static inline void vfsub_mnt_drop_write(struct vfsmount *mnt) +{ + lockdep_off(); + mnt_drop_write(mnt); + lockdep_on(); +} + +static inline void vfsub_mnt_drop_write_file(struct file *file) +{ + lockdep_off(); + mnt_drop_write_file(file); + lockdep_on(); +} + +/* ---------------------------------------------------------------------- */ + +struct au_hinode; +struct dentry *vfsub_lock_rename(struct dentry *d1, struct au_hinode *hdir1, + struct dentry *d2, struct au_hinode *hdir2); +void vfsub_unlock_rename(struct dentry *d1, struct au_hinode *hdir1, + struct dentry *d2, struct au_hinode *hdir2); + +int vfsub_create(struct inode *dir, struct path *path, int mode, + bool want_excl); +int vfsub_symlink(struct inode *dir, struct path *path, + const char *symname); +int vfsub_mknod(struct inode *dir, struct path *path, int mode, dev_t dev); +int vfsub_link(struct dentry *src_dentry, struct inode *dir, + struct path *path); +int vfsub_rename(struct inode *src_hdir, struct dentry *src_dentry, + struct inode *hdir, struct path *path); +int vfsub_mkdir(struct inode *dir, struct path *path, int mode); +int vfsub_rmdir(struct inode *dir, struct path *path); + +/* ---------------------------------------------------------------------- */ + +ssize_t vfsub_read_u(struct file *file, char __user *ubuf, size_t count, + loff_t *ppos); +ssize_t vfsub_read_k(struct file *file, void *kbuf, size_t count, + loff_t *ppos); +ssize_t vfsub_write_u(struct file *file, const char __user *ubuf, size_t count, + loff_t *ppos); +ssize_t vfsub_write_k(struct file *file, void *kbuf, size_t count, + loff_t *ppos); +int vfsub_flush(struct file *file, fl_owner_t id); +int vfsub_readdir(struct file *file, filldir_t filldir, void *arg); + +static inline loff_t vfsub_f_size_read(struct file *file) +{ + return i_size_read(file_inode(file)); +} + +static inline unsigned int vfsub_file_flags(struct file *file) +{ + unsigned int flags; + + spin_lock(&file->f_lock); + flags = file->f_flags; + spin_unlock(&file->f_lock); + + return flags; +} + +static inline void vfsub_file_accessed(struct file *h_file) +{ + file_accessed(h_file); + vfsub_update_h_iattr(&h_file->f_path, /*did*/NULL); /*ignore*/ +} + +static inline void vfsub_touch_atime(struct vfsmount *h_mnt, + struct dentry *h_dentry) +{ + struct path h_path = { + .dentry = h_dentry, + .mnt = h_mnt + }; + touch_atime(&h_path); + vfsub_update_h_iattr(&h_path, /*did*/NULL); /*ignore*/ +} + +static inline int vfsub_update_time(struct inode *h_inode, struct timespec *ts, + int flags) +{ + return update_time(h_inode, ts, flags); + /* no vfsub_update_h_iattr() since we don't have struct path */ +} + +long vfsub_splice_to(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, size_t len, + unsigned int flags); +long vfsub_splice_from(struct pipe_inode_info *pipe, struct file *out, + loff_t *ppos, size_t len, unsigned int flags); + +static inline long vfsub_truncate(struct path *path, loff_t length) +{ + long err; + + lockdep_off(); + err = vfs_truncate(path, length); + lockdep_on(); + return err; +} + +int vfsub_trunc(struct path *h_path, loff_t length, unsigned int attr, + struct file *h_file); +int vfsub_fsync(struct file *file, struct path *path, int datasync); + +/* ---------------------------------------------------------------------- */ + +static inline loff_t vfsub_llseek(struct file *file, loff_t offset, int origin) +{ + loff_t err; + + lockdep_off(); + err = vfs_llseek(file, offset, origin); + lockdep_on(); + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* dirty workaround for strict type of fmode_t */ +union vfsub_fmu { + fmode_t fm; + unsigned int ui; +}; + +static inline unsigned int vfsub_fmode_to_uint(fmode_t fm) +{ + union vfsub_fmu u = { + .fm = fm + }; + + BUILD_BUG_ON(sizeof(u.fm) != sizeof(u.ui)); + + return u.ui; +} + +static inline fmode_t vfsub_uint_to_fmode(unsigned int ui) +{ + union vfsub_fmu u = { + .ui = ui + }; + + return u.fm; +} + +/* ---------------------------------------------------------------------- */ + +int vfsub_sio_mkdir(struct inode *dir, struct path *path, int mode); +int vfsub_sio_rmdir(struct inode *dir, struct path *path); +int vfsub_sio_notify_change(struct path *path, struct iattr *ia); +int vfsub_notify_change(struct path *path, struct iattr *ia); +int vfsub_unlink(struct inode *dir, struct path *path, int force); + +/* ---------------------------------------------------------------------- */ + +static inline int vfsub_setxattr(struct dentry *dentry, const char *name, + const void *value, size_t size, int flags) +{ + int err; + + lockdep_off(); + err = vfs_setxattr(dentry, name, value, size, flags); + lockdep_on(); + + return err; +} + +static inline int vfsub_removexattr(struct dentry *dentry, const char *name) +{ + int err; + + lockdep_off(); + err = vfs_removexattr(dentry, name); + lockdep_on(); + + return err; +} + +#endif /* __KERNEL__ */ +#endif /* __AUFS_VFSUB_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/wbr_policy.c kernel.21.3.new/fs/aufs/wbr_policy.c --- kernel.21.3.orig/fs/aufs/wbr_policy.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/wbr_policy.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,764 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * policies for selecting one among multiple writable branches + */ + +#include +#include "aufs.h" + +/* subset of cpup_attr() */ +static noinline_for_stack +int au_cpdown_attr(struct path *h_path, struct dentry *h_src) +{ + int err, sbits; + struct iattr ia; + struct inode *h_isrc; + + h_isrc = h_src->d_inode; + ia.ia_valid = ATTR_FORCE | ATTR_MODE | ATTR_UID | ATTR_GID; + ia.ia_mode = h_isrc->i_mode; + ia.ia_uid = h_isrc->i_uid; + ia.ia_gid = h_isrc->i_gid; + sbits = !!(ia.ia_mode & (S_ISUID | S_ISGID)); + au_cpup_attr_flags(h_path->dentry->d_inode, h_isrc->i_flags); + err = vfsub_sio_notify_change(h_path, &ia); + + /* is this nfs only? */ + if (!err && sbits && au_test_nfs(h_path->dentry->d_sb)) { + ia.ia_valid = ATTR_FORCE | ATTR_MODE; + ia.ia_mode = h_isrc->i_mode; + err = vfsub_sio_notify_change(h_path, &ia); + } + + return err; +} + +#define AuCpdown_PARENT_OPQ 1 +#define AuCpdown_WHED (1 << 1) +#define AuCpdown_MADE_DIR (1 << 2) +#define AuCpdown_DIROPQ (1 << 3) +#define au_ftest_cpdown(flags, name) ((flags) & AuCpdown_##name) +#define au_fset_cpdown(flags, name) \ + do { (flags) |= AuCpdown_##name; } while (0) +#define au_fclr_cpdown(flags, name) \ + do { (flags) &= ~AuCpdown_##name; } while (0) + +static int au_cpdown_dir_opq(struct dentry *dentry, aufs_bindex_t bdst, + unsigned int *flags) +{ + int err; + struct dentry *opq_dentry; + + opq_dentry = au_diropq_create(dentry, bdst); + err = PTR_ERR(opq_dentry); + if (IS_ERR(opq_dentry)) + goto out; + dput(opq_dentry); + au_fset_cpdown(*flags, DIROPQ); + +out: + return err; +} + +static int au_cpdown_dir_wh(struct dentry *dentry, struct dentry *h_parent, + struct inode *dir, aufs_bindex_t bdst) +{ + int err; + struct path h_path; + struct au_branch *br; + + br = au_sbr(dentry->d_sb, bdst); + h_path.dentry = au_wh_lkup(h_parent, &dentry->d_name, br); + err = PTR_ERR(h_path.dentry); + if (IS_ERR(h_path.dentry)) + goto out; + + err = 0; + if (h_path.dentry->d_inode) { + h_path.mnt = au_br_mnt(br); + err = au_wh_unlink_dentry(au_h_iptr(dir, bdst), &h_path, + dentry); + } + dput(h_path.dentry); + +out: + return err; +} + +static int au_cpdown_dir(struct dentry *dentry, aufs_bindex_t bdst, + struct au_pin *pin, + struct dentry *h_parent, void *arg) +{ + int err, rerr; + aufs_bindex_t bopq, bstart; + struct path h_path; + struct dentry *parent; + struct inode *h_dir, *h_inode, *inode, *dir; + unsigned int *flags = arg; + + bstart = au_dbstart(dentry); + /* dentry is di-locked */ + parent = dget_parent(dentry); + dir = parent->d_inode; + h_dir = h_parent->d_inode; + AuDebugOn(h_dir != au_h_iptr(dir, bdst)); + IMustLock(h_dir); + + err = au_lkup_neg(dentry, bdst, /*wh*/0); + if (unlikely(err < 0)) + goto out; + h_path.dentry = au_h_dptr(dentry, bdst); + h_path.mnt = au_sbr_mnt(dentry->d_sb, bdst); + err = vfsub_sio_mkdir(au_h_iptr(dir, bdst), &h_path, + S_IRWXU | S_IRUGO | S_IXUGO); + if (unlikely(err)) + goto out_put; + au_fset_cpdown(*flags, MADE_DIR); + + bopq = au_dbdiropq(dentry); + au_fclr_cpdown(*flags, WHED); + au_fclr_cpdown(*flags, DIROPQ); + if (au_dbwh(dentry) == bdst) + au_fset_cpdown(*flags, WHED); + if (!au_ftest_cpdown(*flags, PARENT_OPQ) && bopq <= bdst) + au_fset_cpdown(*flags, PARENT_OPQ); + h_inode = h_path.dentry->d_inode; + mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD); + if (au_ftest_cpdown(*flags, WHED)) { + err = au_cpdown_dir_opq(dentry, bdst, flags); + if (unlikely(err)) { + mutex_unlock(&h_inode->i_mutex); + goto out_dir; + } + } + + err = au_cpdown_attr(&h_path, au_h_dptr(dentry, bstart)); + mutex_unlock(&h_inode->i_mutex); + if (unlikely(err)) + goto out_opq; + + if (au_ftest_cpdown(*flags, WHED)) { + err = au_cpdown_dir_wh(dentry, h_parent, dir, bdst); + if (unlikely(err)) + goto out_opq; + } + + inode = dentry->d_inode; + if (au_ibend(inode) < bdst) + au_set_ibend(inode, bdst); + au_set_h_iptr(inode, bdst, au_igrab(h_inode), + au_hi_flags(inode, /*isdir*/1)); + au_fhsm_wrote(dentry->d_sb, bdst, /*force*/0); + goto out; /* success */ + + /* revert */ +out_opq: + if (au_ftest_cpdown(*flags, DIROPQ)) { + mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD); + rerr = au_diropq_remove(dentry, bdst); + mutex_unlock(&h_inode->i_mutex); + if (unlikely(rerr)) { + AuIOErr("failed removing diropq for %.*s b%d (%d)\n", + AuDLNPair(dentry), bdst, rerr); + err = -EIO; + goto out; + } + } +out_dir: + if (au_ftest_cpdown(*flags, MADE_DIR)) { + rerr = vfsub_sio_rmdir(au_h_iptr(dir, bdst), &h_path); + if (unlikely(rerr)) { + AuIOErr("failed removing %.*s b%d (%d)\n", + AuDLNPair(dentry), bdst, rerr); + err = -EIO; + } + } +out_put: + au_set_h_dptr(dentry, bdst, NULL); + if (au_dbend(dentry) == bdst) + au_update_dbend(dentry); +out: + dput(parent); + return err; +} + +int au_cpdown_dirs(struct dentry *dentry, aufs_bindex_t bdst) +{ + int err; + unsigned int flags; + + flags = 0; + err = au_cp_dirs(dentry, bdst, au_cpdown_dir, &flags); + + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* policies for create */ + +int au_wbr_nonopq(struct dentry *dentry, aufs_bindex_t bindex) +{ + int err, i, j, ndentry; + aufs_bindex_t bopq; + struct au_dcsub_pages dpages; + struct au_dpage *dpage; + struct dentry **dentries, *parent, *d; + + err = au_dpages_init(&dpages, GFP_NOFS); + if (unlikely(err)) + goto out; + parent = dget_parent(dentry); + err = au_dcsub_pages_rev_aufs(&dpages, parent, /*do_include*/0); + if (unlikely(err)) + goto out_free; + + err = bindex; + for (i = 0; i < dpages.ndpage; i++) { + dpage = dpages.dpages + i; + dentries = dpage->dentries; + ndentry = dpage->ndentry; + for (j = 0; j < ndentry; j++) { + d = dentries[j]; + di_read_lock_parent2(d, !AuLock_IR); + bopq = au_dbdiropq(d); + di_read_unlock(d, !AuLock_IR); + if (bopq >= 0 && bopq < err) + err = bopq; + } + } + +out_free: + dput(parent); + au_dpages_free(&dpages); +out: + return err; +} + +static int au_wbr_bu(struct super_block *sb, aufs_bindex_t bindex) +{ + for (; bindex >= 0; bindex--) + if (!au_br_rdonly(au_sbr(sb, bindex))) + return bindex; + return -EROFS; +} + +/* top down parent */ +static int au_wbr_create_tdp(struct dentry *dentry, + unsigned int flags __maybe_unused) +{ + int err; + aufs_bindex_t bstart, bindex; + struct super_block *sb; + struct dentry *parent, *h_parent; + + sb = dentry->d_sb; + bstart = au_dbstart(dentry); + err = bstart; + if (!au_br_rdonly(au_sbr(sb, bstart))) + goto out; + + err = -EROFS; + parent = dget_parent(dentry); + for (bindex = au_dbstart(parent); bindex < bstart; bindex++) { + h_parent = au_h_dptr(parent, bindex); + if (!h_parent || !h_parent->d_inode) + continue; + + if (!au_br_rdonly(au_sbr(sb, bindex))) { + err = bindex; + break; + } + } + dput(parent); + + /* bottom up here */ + if (unlikely(err < 0)) { + err = au_wbr_bu(sb, bstart - 1); + if (err >= 0) + err = au_wbr_nonopq(dentry, err); + } + +out: + AuDbg("b%d\n", err); + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* an exception for the policy other than tdp */ +static int au_wbr_create_exp(struct dentry *dentry) +{ + int err; + aufs_bindex_t bwh, bdiropq; + struct dentry *parent; + + err = -1; + bwh = au_dbwh(dentry); + parent = dget_parent(dentry); + bdiropq = au_dbdiropq(parent); + if (bwh >= 0) { + if (bdiropq >= 0) + err = min(bdiropq, bwh); + else + err = bwh; + AuDbg("%d\n", err); + } else if (bdiropq >= 0) { + err = bdiropq; + AuDbg("%d\n", err); + } + dput(parent); + + if (err >= 0) + err = au_wbr_nonopq(dentry, err); + + if (err >= 0 && au_br_rdonly(au_sbr(dentry->d_sb, err))) + err = -1; + + AuDbg("%d\n", err); + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* round robin */ +static int au_wbr_create_init_rr(struct super_block *sb) +{ + int err; + + err = au_wbr_bu(sb, au_sbend(sb)); + atomic_set(&au_sbi(sb)->si_wbr_rr_next, -err); /* less important */ + /* smp_mb(); */ + + AuDbg("b%d\n", err); + return err; +} + +static int au_wbr_create_rr(struct dentry *dentry, unsigned int flags) +{ + int err, nbr; + unsigned int u; + aufs_bindex_t bindex, bend; + struct super_block *sb; + atomic_t *next; + + err = au_wbr_create_exp(dentry); + if (err >= 0) + goto out; + + sb = dentry->d_sb; + next = &au_sbi(sb)->si_wbr_rr_next; + bend = au_sbend(sb); + nbr = bend + 1; + for (bindex = 0; bindex <= bend; bindex++) { + if (!au_ftest_wbr(flags, DIR)) { + err = atomic_dec_return(next) + 1; + /* modulo for 0 is meaningless */ + if (unlikely(!err)) + err = atomic_dec_return(next) + 1; + } else + err = atomic_read(next); + AuDbg("%d\n", err); + u = err; + err = u % nbr; + AuDbg("%d\n", err); + if (!au_br_rdonly(au_sbr(sb, err))) + break; + err = -EROFS; + } + + if (err >= 0) + err = au_wbr_nonopq(dentry, err); + +out: + AuDbg("%d\n", err); + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* most free space */ +static void au_mfs(struct dentry *dentry, struct dentry *parent) +{ + struct super_block *sb; + struct au_branch *br; + struct au_wbr_mfs *mfs; + struct dentry *h_parent; + aufs_bindex_t bindex, bend; + int err; + unsigned long long b, bavail; + struct path h_path; + /* reduce the stack usage */ + struct kstatfs *st; + + st = kmalloc(sizeof(*st), GFP_NOFS); + if (unlikely(!st)) { + AuWarn1("failed updating mfs(%d), ignored\n", -ENOMEM); + return; + } + + bavail = 0; + sb = dentry->d_sb; + mfs = &au_sbi(sb)->si_wbr_mfs; + MtxMustLock(&mfs->mfs_lock); + mfs->mfs_bindex = -EROFS; + mfs->mfsrr_bytes = 0; + if (!parent) { + bindex = 0; + bend = au_sbend(sb); + } else { + bindex = au_dbstart(parent); + bend = au_dbtaildir(parent); + } + + for (; bindex <= bend; bindex++) { + if (parent) { + h_parent = au_h_dptr(parent, bindex); + if (!h_parent || !h_parent->d_inode) + continue; + } + br = au_sbr(sb, bindex); + if (au_br_rdonly(br)) + continue; + + /* sb->s_root for NFS is unreliable */ + h_path.mnt = au_br_mnt(br); + h_path.dentry = h_path.mnt->mnt_root; + err = vfs_statfs(&h_path, st); + if (unlikely(err)) { + AuWarn1("failed statfs, b%d, %d\n", bindex, err); + continue; + } + + /* when the available size is equal, select the lower one */ + BUILD_BUG_ON(sizeof(b) < sizeof(st->f_bavail) + || sizeof(b) < sizeof(st->f_bsize)); + b = st->f_bavail * st->f_bsize; + br->br_wbr->wbr_bytes = b; + if (b >= bavail) { + bavail = b; + mfs->mfs_bindex = bindex; + mfs->mfs_jiffy = jiffies; + } + } + + mfs->mfsrr_bytes = bavail; + AuDbg("b%d\n", mfs->mfs_bindex); + kfree(st); +} + +static int au_wbr_create_mfs(struct dentry *dentry, unsigned int flags) +{ + int err; + struct dentry *parent; + struct super_block *sb; + struct au_wbr_mfs *mfs; + + err = au_wbr_create_exp(dentry); + if (err >= 0) + goto out; + + sb = dentry->d_sb; + parent = NULL; + if (au_ftest_wbr(flags, PARENT)) + parent = dget_parent(dentry); + mfs = &au_sbi(sb)->si_wbr_mfs; + mutex_lock(&mfs->mfs_lock); + if (time_after(jiffies, mfs->mfs_jiffy + mfs->mfs_expire) + || mfs->mfs_bindex < 0 + || au_br_rdonly(au_sbr(sb, mfs->mfs_bindex))) + au_mfs(dentry, parent); + mutex_unlock(&mfs->mfs_lock); + err = mfs->mfs_bindex; + dput(parent); + + if (err >= 0) + err = au_wbr_nonopq(dentry, err); + +out: + AuDbg("b%d\n", err); + return err; +} + +static int au_wbr_create_init_mfs(struct super_block *sb) +{ + struct au_wbr_mfs *mfs; + + mfs = &au_sbi(sb)->si_wbr_mfs; + mutex_init(&mfs->mfs_lock); + mfs->mfs_jiffy = 0; + mfs->mfs_bindex = -EROFS; + + return 0; +} + +static int au_wbr_create_fin_mfs(struct super_block *sb __maybe_unused) +{ + mutex_destroy(&au_sbi(sb)->si_wbr_mfs.mfs_lock); + return 0; +} + +/* ---------------------------------------------------------------------- */ + +/* most free space and then round robin */ +static int au_wbr_create_mfsrr(struct dentry *dentry, unsigned int flags) +{ + int err; + struct au_wbr_mfs *mfs; + + err = au_wbr_create_mfs(dentry, flags); + if (err >= 0) { + mfs = &au_sbi(dentry->d_sb)->si_wbr_mfs; + mutex_lock(&mfs->mfs_lock); + if (mfs->mfsrr_bytes < mfs->mfsrr_watermark) + err = au_wbr_create_rr(dentry, flags); + mutex_unlock(&mfs->mfs_lock); + } + + AuDbg("b%d\n", err); + return err; +} + +static int au_wbr_create_init_mfsrr(struct super_block *sb) +{ + int err; + + au_wbr_create_init_mfs(sb); /* ignore */ + err = au_wbr_create_init_rr(sb); + + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* top down parent and most free space */ +static int au_wbr_create_pmfs(struct dentry *dentry, unsigned int flags) +{ + int err, e2; + unsigned long long b; + aufs_bindex_t bindex, bstart, bend; + struct super_block *sb; + struct dentry *parent, *h_parent; + struct au_branch *br; + + err = au_wbr_create_tdp(dentry, flags); + if (unlikely(err < 0)) + goto out; + parent = dget_parent(dentry); + bstart = au_dbstart(parent); + bend = au_dbtaildir(parent); + if (bstart == bend) + goto out_parent; /* success */ + + e2 = au_wbr_create_mfs(dentry, flags); + if (e2 < 0) + goto out_parent; /* success */ + + /* when the available size is equal, select upper one */ + sb = dentry->d_sb; + br = au_sbr(sb, err); + b = br->br_wbr->wbr_bytes; + AuDbg("b%d, %llu\n", err, b); + + for (bindex = bstart; bindex <= bend; bindex++) { + h_parent = au_h_dptr(parent, bindex); + if (!h_parent || !h_parent->d_inode) + continue; + + br = au_sbr(sb, bindex); + if (!au_br_rdonly(br) && br->br_wbr->wbr_bytes > b) { + b = br->br_wbr->wbr_bytes; + err = bindex; + AuDbg("b%d, %llu\n", err, b); + } + } + + if (err >= 0) + err = au_wbr_nonopq(dentry, err); + +out_parent: + dput(parent); +out: + AuDbg("b%d\n", err); + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* + * - top down parent + * - most free space with parent + * - most free space round-robin regardless parent + */ +static int au_wbr_create_pmfsrr(struct dentry *dentry, unsigned int flags) +{ + int err; + unsigned long long watermark; + struct super_block *sb; + struct au_branch *br; + struct au_wbr_mfs *mfs; + + err = au_wbr_create_pmfs(dentry, flags | AuWbr_PARENT); + if (unlikely(err < 0)) + goto out; + + sb = dentry->d_sb; + br = au_sbr(sb, err); + mfs = &au_sbi(sb)->si_wbr_mfs; + mutex_lock(&mfs->mfs_lock); + watermark = mfs->mfsrr_watermark; + mutex_unlock(&mfs->mfs_lock); + if (br->br_wbr->wbr_bytes < watermark) + /* regardless the parent dir */ + err = au_wbr_create_mfsrr(dentry, flags); + +out: + AuDbg("b%d\n", err); + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* policies for copyup */ + +/* top down parent */ +static int au_wbr_copyup_tdp(struct dentry *dentry) +{ + return au_wbr_create_tdp(dentry, /*flags, anything is ok*/0); +} + +/* bottom up parent */ +static int au_wbr_copyup_bup(struct dentry *dentry) +{ + int err; + aufs_bindex_t bindex, bstart; + struct dentry *parent, *h_parent; + struct super_block *sb; + + err = -EROFS; + sb = dentry->d_sb; + parent = dget_parent(dentry); + bstart = au_dbstart(parent); + for (bindex = au_dbstart(dentry); bindex >= bstart; bindex--) { + h_parent = au_h_dptr(parent, bindex); + if (!h_parent || !h_parent->d_inode) + continue; + + if (!au_br_rdonly(au_sbr(sb, bindex))) { + err = bindex; + break; + } + } + dput(parent); + + /* bottom up here */ + if (unlikely(err < 0)) + err = au_wbr_bu(sb, bstart - 1); + + AuDbg("b%d\n", err); + return err; +} + +/* bottom up */ +int au_wbr_do_copyup_bu(struct dentry *dentry, aufs_bindex_t bstart) +{ + int err; + + err = au_wbr_bu(dentry->d_sb, bstart); + AuDbg("b%d\n", err); + if (err > bstart) + err = au_wbr_nonopq(dentry, err); + + AuDbg("b%d\n", err); + return err; +} + +static int au_wbr_copyup_bu(struct dentry *dentry) +{ + int err; + aufs_bindex_t bstart; + + bstart = au_dbstart(dentry); + err = au_wbr_do_copyup_bu(dentry, bstart); + return err; +} + +/* ---------------------------------------------------------------------- */ + +struct au_wbr_copyup_operations au_wbr_copyup_ops[] = { + [AuWbrCopyup_TDP] = { + .copyup = au_wbr_copyup_tdp + }, + [AuWbrCopyup_BUP] = { + .copyup = au_wbr_copyup_bup + }, + [AuWbrCopyup_BU] = { + .copyup = au_wbr_copyup_bu + } +}; + +struct au_wbr_create_operations au_wbr_create_ops[] = { + [AuWbrCreate_TDP] = { + .create = au_wbr_create_tdp + }, + [AuWbrCreate_RR] = { + .create = au_wbr_create_rr, + .init = au_wbr_create_init_rr + }, + [AuWbrCreate_MFS] = { + .create = au_wbr_create_mfs, + .init = au_wbr_create_init_mfs, + .fin = au_wbr_create_fin_mfs + }, + [AuWbrCreate_MFSV] = { + .create = au_wbr_create_mfs, + .init = au_wbr_create_init_mfs, + .fin = au_wbr_create_fin_mfs + }, + [AuWbrCreate_MFSRR] = { + .create = au_wbr_create_mfsrr, + .init = au_wbr_create_init_mfsrr, + .fin = au_wbr_create_fin_mfs + }, + [AuWbrCreate_MFSRRV] = { + .create = au_wbr_create_mfsrr, + .init = au_wbr_create_init_mfsrr, + .fin = au_wbr_create_fin_mfs + }, + [AuWbrCreate_PMFS] = { + .create = au_wbr_create_pmfs, + .init = au_wbr_create_init_mfs, + .fin = au_wbr_create_fin_mfs + }, + [AuWbrCreate_PMFSV] = { + .create = au_wbr_create_pmfs, + .init = au_wbr_create_init_mfs, + .fin = au_wbr_create_fin_mfs + }, + [AuWbrCreate_PMFSRR] = { + .create = au_wbr_create_pmfsrr, + .init = au_wbr_create_init_mfsrr, + .fin = au_wbr_create_fin_mfs + }, + [AuWbrCreate_PMFSRRV] = { + .create = au_wbr_create_pmfsrr, + .init = au_wbr_create_init_mfsrr, + .fin = au_wbr_create_fin_mfs + } +}; diff -Naur kernel.21.3.orig/fs/aufs/whout.c kernel.21.3.new/fs/aufs/whout.c --- kernel.21.3.orig/fs/aufs/whout.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/whout.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,1025 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * whiteout for logical deletion and opaque directory + */ + +#include "aufs.h" + +#define WH_MASK S_IRUGO + +/* + * If a directory contains this file, then it is opaque. We start with the + * .wh. flag so that it is blocked by lookup. + */ +static struct qstr diropq_name = QSTR_INIT(AUFS_WH_DIROPQ, + sizeof(AUFS_WH_DIROPQ) - 1); + +/* + * generate whiteout name, which is NOT terminated by NULL. + * @name: original d_name.name + * @len: original d_name.len + * @wh: whiteout qstr + * returns zero when succeeds, otherwise error. + * succeeded value as wh->name should be freed by kfree(). + */ +int au_wh_name_alloc(struct qstr *wh, const struct qstr *name) +{ + char *p; + + if (unlikely(name->len > PATH_MAX - AUFS_WH_PFX_LEN)) + return -ENAMETOOLONG; + + wh->len = name->len + AUFS_WH_PFX_LEN; + p = kmalloc(wh->len, GFP_NOFS); + wh->name = p; + if (p) { + memcpy(p, AUFS_WH_PFX, AUFS_WH_PFX_LEN); + memcpy(p + AUFS_WH_PFX_LEN, name->name, name->len); + /* smp_mb(); */ + return 0; + } + return -ENOMEM; +} + +/* ---------------------------------------------------------------------- */ + +/* + * test if the @wh_name exists under @h_parent. + * @try_sio specifies the necessary of super-io. + */ +int au_wh_test(struct dentry *h_parent, struct qstr *wh_name, int try_sio) +{ + int err; + struct dentry *wh_dentry; + + if (!try_sio) + wh_dentry = vfsub_lkup_one(wh_name, h_parent); + else + wh_dentry = au_sio_lkup_one(wh_name, h_parent); + err = PTR_ERR(wh_dentry); + if (IS_ERR(wh_dentry)) + goto out; + + err = 0; + if (!wh_dentry->d_inode) + goto out_wh; /* success */ + + err = 1; + if (S_ISREG(wh_dentry->d_inode->i_mode)) + goto out_wh; /* success */ + + err = -EIO; + AuIOErr("%.*s Invalid whiteout entry type 0%o.\n", + AuDLNPair(wh_dentry), wh_dentry->d_inode->i_mode); + +out_wh: + dput(wh_dentry); +out: + return err; +} + +/* + * test if the @h_dentry sets opaque or not. + */ +int au_diropq_test(struct dentry *h_dentry) +{ + int err; + struct inode *h_dir; + + h_dir = h_dentry->d_inode; + err = au_wh_test(h_dentry, &diropq_name, + au_test_h_perm_sio(h_dir, MAY_EXEC)); + return err; +} + +/* + * returns a negative dentry whose name is unique and temporary. + */ +struct dentry *au_whtmp_lkup(struct dentry *h_parent, struct au_branch *br, + struct qstr *prefix) +{ + struct dentry *dentry; + int i; + char defname[NAME_MAX - AUFS_MAX_NAMELEN + DNAME_INLINE_LEN + 1], + *name, *p; + /* strict atomic_t is unnecessary here */ + static unsigned short cnt; + struct qstr qs; + + BUILD_BUG_ON(sizeof(cnt) * 2 > AUFS_WH_TMP_LEN); + + name = defname; + qs.len = sizeof(defname) - DNAME_INLINE_LEN + prefix->len - 1; + if (unlikely(prefix->len > DNAME_INLINE_LEN)) { + dentry = ERR_PTR(-ENAMETOOLONG); + if (unlikely(qs.len > NAME_MAX)) + goto out; + dentry = ERR_PTR(-ENOMEM); + name = kmalloc(qs.len + 1, GFP_NOFS); + if (unlikely(!name)) + goto out; + } + + /* doubly whiteout-ed */ + memcpy(name, AUFS_WH_PFX AUFS_WH_PFX, AUFS_WH_PFX_LEN * 2); + p = name + AUFS_WH_PFX_LEN * 2; + memcpy(p, prefix->name, prefix->len); + p += prefix->len; + *p++ = '.'; + AuDebugOn(name + qs.len + 1 - p <= AUFS_WH_TMP_LEN); + + qs.name = name; + for (i = 0; i < 3; i++) { + sprintf(p, "%.*x", AUFS_WH_TMP_LEN, cnt++); + dentry = au_sio_lkup_one(&qs, h_parent); + if (IS_ERR(dentry) || !dentry->d_inode) + goto out_name; + dput(dentry); + } + /* pr_warn("could not get random name\n"); */ + dentry = ERR_PTR(-EEXIST); + AuDbg("%.*s\n", AuLNPair(&qs)); + BUG(); + +out_name: + if (name != defname) + kfree(name); +out: + AuTraceErrPtr(dentry); + return dentry; +} + +/* + * rename the @h_dentry on @br to the whiteouted temporary name. + */ +int au_whtmp_ren(struct dentry *h_dentry, struct au_branch *br) +{ + int err; + struct path h_path = { + .mnt = au_br_mnt(br) + }; + struct inode *h_dir; + struct dentry *h_parent; + + h_parent = h_dentry->d_parent; /* dir inode is locked */ + h_dir = h_parent->d_inode; + IMustLock(h_dir); + + h_path.dentry = au_whtmp_lkup(h_parent, br, &h_dentry->d_name); + err = PTR_ERR(h_path.dentry); + if (IS_ERR(h_path.dentry)) + goto out; + + /* under the same dir, no need to lock_rename() */ + err = vfsub_rename(h_dir, h_dentry, h_dir, &h_path); + AuTraceErr(err); + dput(h_path.dentry); + +out: + AuTraceErr(err); + return err; +} + +/* ---------------------------------------------------------------------- */ +/* + * functions for removing a whiteout + */ + +static int do_unlink_wh(struct inode *h_dir, struct path *h_path) +{ + int force; + + /* + * forces superio when the dir has a sticky bit. + * this may be a violation of unix fs semantics. + */ + force = (h_dir->i_mode & S_ISVTX) + && !uid_eq(current_fsuid(), h_path->dentry->d_inode->i_uid); + return vfsub_unlink(h_dir, h_path, force); +} + +int au_wh_unlink_dentry(struct inode *h_dir, struct path *h_path, + struct dentry *dentry) +{ + int err; + + err = do_unlink_wh(h_dir, h_path); + if (!err && dentry) + au_set_dbwh(dentry, -1); + + return err; +} + +static int unlink_wh_name(struct dentry *h_parent, struct qstr *wh, + struct au_branch *br) +{ + int err; + struct path h_path = { + .mnt = au_br_mnt(br) + }; + + err = 0; + h_path.dentry = vfsub_lkup_one(wh, h_parent); + if (IS_ERR(h_path.dentry)) + err = PTR_ERR(h_path.dentry); + else { + if (h_path.dentry->d_inode + && S_ISREG(h_path.dentry->d_inode->i_mode)) + err = do_unlink_wh(h_parent->d_inode, &h_path); + dput(h_path.dentry); + } + + return err; +} + +/* ---------------------------------------------------------------------- */ +/* + * initialize/clean whiteout for a branch + */ + +static void au_wh_clean(struct inode *h_dir, struct path *whpath, + const int isdir) +{ + int err; + + if (!whpath->dentry->d_inode) + return; + + if (isdir) + err = vfsub_rmdir(h_dir, whpath); + else + err = vfsub_unlink(h_dir, whpath, /*force*/0); + if (unlikely(err)) + pr_warn("failed removing %.*s (%d), ignored.\n", + AuDLNPair(whpath->dentry), err); +} + +static int test_linkable(struct dentry *h_root) +{ + struct inode *h_dir = h_root->d_inode; + + if (h_dir->i_op->link) + return 0; + + pr_err("%.*s (%s) doesn't support link(2), use noplink and rw+nolwh\n", + AuDLNPair(h_root), au_sbtype(h_root->d_sb)); + return -ENOSYS; +} + +/* todo: should this mkdir be done in /sbin/mount.aufs helper? */ +static int au_whdir(struct inode *h_dir, struct path *path) +{ + int err; + + err = -EEXIST; + if (!path->dentry->d_inode) { + int mode = S_IRWXU; + + if (au_test_nfs(path->dentry->d_sb)) + mode |= S_IXUGO; + err = vfsub_mkdir(h_dir, path, mode); + } else if (S_ISDIR(path->dentry->d_inode->i_mode)) + err = 0; + else + pr_err("unknown %.*s exists\n", AuDLNPair(path->dentry)); + + return err; +} + +struct au_wh_base { + const struct qstr *name; + struct dentry *dentry; +}; + +static void au_wh_init_ro(struct inode *h_dir, struct au_wh_base base[], + struct path *h_path) +{ + h_path->dentry = base[AuBrWh_BASE].dentry; + au_wh_clean(h_dir, h_path, /*isdir*/0); + h_path->dentry = base[AuBrWh_PLINK].dentry; + au_wh_clean(h_dir, h_path, /*isdir*/1); + h_path->dentry = base[AuBrWh_ORPH].dentry; + au_wh_clean(h_dir, h_path, /*isdir*/1); +} + +/* + * returns tri-state, + * minus: error, caller should print the message + * zero: succuess + * plus: error, caller should NOT print the message + */ +static int au_wh_init_rw_nolink(struct dentry *h_root, struct au_wbr *wbr, + int do_plink, struct au_wh_base base[], + struct path *h_path) +{ + int err; + struct inode *h_dir; + + h_dir = h_root->d_inode; + h_path->dentry = base[AuBrWh_BASE].dentry; + au_wh_clean(h_dir, h_path, /*isdir*/0); + h_path->dentry = base[AuBrWh_PLINK].dentry; + if (do_plink) { + err = test_linkable(h_root); + if (unlikely(err)) { + err = 1; + goto out; + } + + err = au_whdir(h_dir, h_path); + if (unlikely(err)) + goto out; + wbr->wbr_plink = dget(base[AuBrWh_PLINK].dentry); + } else + au_wh_clean(h_dir, h_path, /*isdir*/1); + h_path->dentry = base[AuBrWh_ORPH].dentry; + err = au_whdir(h_dir, h_path); + if (unlikely(err)) + goto out; + wbr->wbr_orph = dget(base[AuBrWh_ORPH].dentry); + +out: + return err; +} + +/* + * for the moment, aufs supports the branch filesystem which does not support + * link(2). testing on FAT which does not support i_op->setattr() fully either, + * copyup failed. finally, such filesystem will not be used as the writable + * branch. + * + * returns tri-state, see above. + */ +static int au_wh_init_rw(struct dentry *h_root, struct au_wbr *wbr, + int do_plink, struct au_wh_base base[], + struct path *h_path) +{ + int err; + struct inode *h_dir; + + WbrWhMustWriteLock(wbr); + + err = test_linkable(h_root); + if (unlikely(err)) { + err = 1; + goto out; + } + + /* + * todo: should this create be done in /sbin/mount.aufs helper? + */ + err = -EEXIST; + h_dir = h_root->d_inode; + if (!base[AuBrWh_BASE].dentry->d_inode) { + h_path->dentry = base[AuBrWh_BASE].dentry; + err = vfsub_create(h_dir, h_path, WH_MASK, /*want_excl*/true); + } else if (S_ISREG(base[AuBrWh_BASE].dentry->d_inode->i_mode)) + err = 0; + else + pr_err("unknown %.*s/%.*s exists\n", + AuDLNPair(h_root), AuDLNPair(base[AuBrWh_BASE].dentry)); + if (unlikely(err)) + goto out; + + h_path->dentry = base[AuBrWh_PLINK].dentry; + if (do_plink) { + err = au_whdir(h_dir, h_path); + if (unlikely(err)) + goto out; + wbr->wbr_plink = dget(base[AuBrWh_PLINK].dentry); + } else + au_wh_clean(h_dir, h_path, /*isdir*/1); + wbr->wbr_whbase = dget(base[AuBrWh_BASE].dentry); + + h_path->dentry = base[AuBrWh_ORPH].dentry; + err = au_whdir(h_dir, h_path); + if (unlikely(err)) + goto out; + wbr->wbr_orph = dget(base[AuBrWh_ORPH].dentry); + +out: + return err; +} + +/* + * initialize the whiteout base file/dir for @br. + */ +int au_wh_init(struct au_branch *br, struct super_block *sb) +{ + int err, i; + const unsigned char do_plink + = !!au_opt_test(au_mntflags(sb), PLINK); + struct inode *h_dir; + struct path path = br->br_path; + struct dentry *h_root = path.dentry; + struct au_wbr *wbr = br->br_wbr; + static const struct qstr base_name[] = { + [AuBrWh_BASE] = QSTR_INIT(AUFS_BASE_NAME, + sizeof(AUFS_BASE_NAME) - 1), + [AuBrWh_PLINK] = QSTR_INIT(AUFS_PLINKDIR_NAME, + sizeof(AUFS_PLINKDIR_NAME) - 1), + [AuBrWh_ORPH] = QSTR_INIT(AUFS_ORPHDIR_NAME, + sizeof(AUFS_ORPHDIR_NAME) - 1) + }; + struct au_wh_base base[] = { + [AuBrWh_BASE] = { + .name = base_name + AuBrWh_BASE, + .dentry = NULL + }, + [AuBrWh_PLINK] = { + .name = base_name + AuBrWh_PLINK, + .dentry = NULL + }, + [AuBrWh_ORPH] = { + .name = base_name + AuBrWh_ORPH, + .dentry = NULL + } + }; + + if (wbr) + WbrWhMustWriteLock(wbr); + + for (i = 0; i < AuBrWh_Last; i++) { + /* doubly whiteouted */ + struct dentry *d; + + d = au_wh_lkup(h_root, (void *)base[i].name, br); + err = PTR_ERR(d); + if (IS_ERR(d)) + goto out; + + base[i].dentry = d; + AuDebugOn(wbr + && wbr->wbr_wh[i] + && wbr->wbr_wh[i] != base[i].dentry); + } + + if (wbr) + for (i = 0; i < AuBrWh_Last; i++) { + dput(wbr->wbr_wh[i]); + wbr->wbr_wh[i] = NULL; + } + + err = 0; + if (!au_br_writable(br->br_perm)) { + h_dir = h_root->d_inode; + au_wh_init_ro(h_dir, base, &path); + } else if (!au_br_wh_linkable(br->br_perm)) { + err = au_wh_init_rw_nolink(h_root, wbr, do_plink, base, &path); + if (err > 0) + goto out; + else if (err) + goto out_err; + } else { + err = au_wh_init_rw(h_root, wbr, do_plink, base, &path); + if (err > 0) + goto out; + else if (err) + goto out_err; + } + goto out; /* success */ + +out_err: + pr_err("an error(%d) on the writable branch %.*s(%s)\n", + err, AuDLNPair(h_root), au_sbtype(h_root->d_sb)); +out: + for (i = 0; i < AuBrWh_Last; i++) + dput(base[i].dentry); + return err; +} + +/* ---------------------------------------------------------------------- */ +/* + * whiteouts are all hard-linked usually. + * when its link count reaches a ceiling, we create a new whiteout base + * asynchronously. + */ + +struct reinit_br_wh { + struct super_block *sb; + struct au_branch *br; +}; + +static void reinit_br_wh(void *arg) +{ + int err; + aufs_bindex_t bindex; + struct path h_path; + struct reinit_br_wh *a = arg; + struct au_wbr *wbr; + struct inode *dir; + struct dentry *h_root; + struct au_hinode *hdir; + + err = 0; + wbr = a->br->br_wbr; + /* big aufs lock */ + si_noflush_write_lock(a->sb); + if (!au_br_writable(a->br->br_perm)) + goto out; + bindex = au_br_index(a->sb, a->br->br_id); + if (unlikely(bindex < 0)) + goto out; + + di_read_lock_parent(a->sb->s_root, AuLock_IR); + dir = a->sb->s_root->d_inode; + hdir = au_hi(dir, bindex); + h_root = au_h_dptr(a->sb->s_root, bindex); + AuDebugOn(h_root != au_br_dentry(a->br)); + + au_hn_imtx_lock_nested(hdir, AuLsc_I_PARENT); + wbr_wh_write_lock(wbr); + err = au_h_verify(wbr->wbr_whbase, au_opt_udba(a->sb), hdir->hi_inode, + h_root, a->br); + if (!err) { + h_path.dentry = wbr->wbr_whbase; + h_path.mnt = au_br_mnt(a->br); + err = vfsub_unlink(hdir->hi_inode, &h_path, /*force*/0); + } else { + pr_warn("%.*s is moved, ignored\n", + AuDLNPair(wbr->wbr_whbase)); + err = 0; + } + dput(wbr->wbr_whbase); + wbr->wbr_whbase = NULL; + if (!err) + err = au_wh_init(a->br, a->sb); + wbr_wh_write_unlock(wbr); + au_hn_imtx_unlock(hdir); + di_read_unlock(a->sb->s_root, AuLock_IR); + if (!err) + au_fhsm_wrote(a->sb, bindex, /*force*/0); + +out: + if (wbr) + atomic_dec(&wbr->wbr_wh_running); + atomic_dec(&a->br->br_count); + si_write_unlock(a->sb); + au_nwt_done(&au_sbi(a->sb)->si_nowait); + kfree(arg); + if (unlikely(err)) + AuIOErr("err %d\n", err); +} + +static void kick_reinit_br_wh(struct super_block *sb, struct au_branch *br) +{ + int do_dec, wkq_err; + struct reinit_br_wh *arg; + + do_dec = 1; + if (atomic_inc_return(&br->br_wbr->wbr_wh_running) != 1) + goto out; + + /* ignore ENOMEM */ + arg = kmalloc(sizeof(*arg), GFP_NOFS); + if (arg) { + /* + * dec(wh_running), kfree(arg) and dec(br_count) + * in reinit function + */ + arg->sb = sb; + arg->br = br; + atomic_inc(&br->br_count); + wkq_err = au_wkq_nowait(reinit_br_wh, arg, sb, /*flags*/0); + if (unlikely(wkq_err)) { + atomic_dec(&br->br_wbr->wbr_wh_running); + atomic_dec(&br->br_count); + kfree(arg); + } + do_dec = 0; + } + +out: + if (do_dec) + atomic_dec(&br->br_wbr->wbr_wh_running); +} + +/* ---------------------------------------------------------------------- */ + +/* + * create the whiteout @wh. + */ +static int link_or_create_wh(struct super_block *sb, aufs_bindex_t bindex, + struct dentry *wh) +{ + int err; + struct path h_path = { + .dentry = wh + }; + struct au_branch *br; + struct au_wbr *wbr; + struct dentry *h_parent; + struct inode *h_dir; + + h_parent = wh->d_parent; /* dir inode is locked */ + h_dir = h_parent->d_inode; + IMustLock(h_dir); + + br = au_sbr(sb, bindex); + h_path.mnt = au_br_mnt(br); + wbr = br->br_wbr; + wbr_wh_read_lock(wbr); + if (wbr->wbr_whbase) { + err = vfsub_link(wbr->wbr_whbase, h_dir, &h_path); + if (!err || err != -EMLINK) + goto out; + + /* link count full. re-initialize br_whbase. */ + kick_reinit_br_wh(sb, br); + } + + /* return this error in this context */ + err = vfsub_create(h_dir, &h_path, WH_MASK, /*want_excl*/true); + if (!err) + au_fhsm_wrote(sb, bindex, /*force*/0); + +out: + wbr_wh_read_unlock(wbr); + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* + * create or remove the diropq. + */ +static struct dentry *do_diropq(struct dentry *dentry, aufs_bindex_t bindex, + unsigned int flags) +{ + struct dentry *opq_dentry, *h_dentry; + struct super_block *sb; + struct au_branch *br; + int err; + + sb = dentry->d_sb; + br = au_sbr(sb, bindex); + h_dentry = au_h_dptr(dentry, bindex); + opq_dentry = vfsub_lkup_one(&diropq_name, h_dentry); + if (IS_ERR(opq_dentry)) + goto out; + + if (au_ftest_diropq(flags, CREATE)) { + err = link_or_create_wh(sb, bindex, opq_dentry); + if (!err) { + au_set_dbdiropq(dentry, bindex); + goto out; /* success */ + } + } else { + struct path tmp = { + .dentry = opq_dentry, + .mnt = au_br_mnt(br) + }; + err = do_unlink_wh(au_h_iptr(dentry->d_inode, bindex), &tmp); + if (!err) + au_set_dbdiropq(dentry, -1); + } + dput(opq_dentry); + opq_dentry = ERR_PTR(err); + +out: + return opq_dentry; +} + +struct do_diropq_args { + struct dentry **errp; + struct dentry *dentry; + aufs_bindex_t bindex; + unsigned int flags; +}; + +static void call_do_diropq(void *args) +{ + struct do_diropq_args *a = args; + *a->errp = do_diropq(a->dentry, a->bindex, a->flags); +} + +struct dentry *au_diropq_sio(struct dentry *dentry, aufs_bindex_t bindex, + unsigned int flags) +{ + struct dentry *diropq, *h_dentry; + + h_dentry = au_h_dptr(dentry, bindex); + if (!au_test_h_perm_sio(h_dentry->d_inode, MAY_EXEC | MAY_WRITE)) + diropq = do_diropq(dentry, bindex, flags); + else { + int wkq_err; + struct do_diropq_args args = { + .errp = &diropq, + .dentry = dentry, + .bindex = bindex, + .flags = flags + }; + + wkq_err = au_wkq_wait(call_do_diropq, &args); + if (unlikely(wkq_err)) + diropq = ERR_PTR(wkq_err); + } + + return diropq; +} + +/* ---------------------------------------------------------------------- */ + +/* + * lookup whiteout dentry. + * @h_parent: lower parent dentry which must exist and be locked + * @base_name: name of dentry which will be whiteouted + * returns dentry for whiteout. + */ +struct dentry *au_wh_lkup(struct dentry *h_parent, struct qstr *base_name, + struct au_branch *br) +{ + int err; + struct qstr wh_name; + struct dentry *wh_dentry; + + err = au_wh_name_alloc(&wh_name, base_name); + wh_dentry = ERR_PTR(err); + if (!err) { + wh_dentry = vfsub_lkup_one(&wh_name, h_parent); + kfree(wh_name.name); + } + return wh_dentry; +} + +/* + * link/create a whiteout for @dentry on @bindex. + */ +struct dentry *au_wh_create(struct dentry *dentry, aufs_bindex_t bindex, + struct dentry *h_parent) +{ + struct dentry *wh_dentry; + struct super_block *sb; + int err; + + sb = dentry->d_sb; + wh_dentry = au_wh_lkup(h_parent, &dentry->d_name, au_sbr(sb, bindex)); + if (!IS_ERR(wh_dentry) && !wh_dentry->d_inode) { + err = link_or_create_wh(sb, bindex, wh_dentry); + if (!err) { + au_set_dbwh(dentry, bindex); + au_fhsm_wrote(sb, bindex, /*force*/0); + } else { + dput(wh_dentry); + wh_dentry = ERR_PTR(err); + } + } + + return wh_dentry; +} + +/* ---------------------------------------------------------------------- */ + +/* Delete all whiteouts in this directory on branch bindex. */ +static int del_wh_children(struct dentry *h_dentry, struct au_nhash *whlist, + aufs_bindex_t bindex, struct au_branch *br) +{ + int err; + unsigned long ul, n; + struct qstr wh_name; + char *p; + struct hlist_head *head; + struct au_vdir_wh *pos; + struct au_vdir_destr *str; + + err = -ENOMEM; + p = (void *)__get_free_page(GFP_NOFS); + wh_name.name = p; + if (unlikely(!wh_name.name)) + goto out; + + err = 0; + memcpy(p, AUFS_WH_PFX, AUFS_WH_PFX_LEN); + p += AUFS_WH_PFX_LEN; + n = whlist->nh_num; + head = whlist->nh_head; + for (ul = 0; !err && ul < n; ul++, head++) { + hlist_for_each_entry(pos, head, wh_hash) { + if (pos->wh_bindex != bindex) + continue; + + str = &pos->wh_str; + if (str->len + AUFS_WH_PFX_LEN <= PATH_MAX) { + memcpy(p, str->name, str->len); + wh_name.len = AUFS_WH_PFX_LEN + str->len; + err = unlink_wh_name(h_dentry, &wh_name, br); + if (!err) + continue; + break; + } + AuIOErr("whiteout name too long %.*s\n", + str->len, str->name); + err = -EIO; + break; + } + } + free_page((unsigned long)wh_name.name); + +out: + return err; +} + +struct del_wh_children_args { + int *errp; + struct dentry *h_dentry; + struct au_nhash *whlist; + aufs_bindex_t bindex; + struct au_branch *br; +}; + +static void call_del_wh_children(void *args) +{ + struct del_wh_children_args *a = args; + *a->errp = del_wh_children(a->h_dentry, a->whlist, a->bindex, a->br); +} + +/* ---------------------------------------------------------------------- */ + +struct au_whtmp_rmdir *au_whtmp_rmdir_alloc(struct super_block *sb, gfp_t gfp) +{ + struct au_whtmp_rmdir *whtmp; + int err; + unsigned int rdhash; + + SiMustAnyLock(sb); + + whtmp = kmalloc(sizeof(*whtmp), gfp); + if (unlikely(!whtmp)) { + whtmp = ERR_PTR(-ENOMEM); + goto out; + } + + whtmp->dir = NULL; + whtmp->br = NULL; + whtmp->wh_dentry = NULL; + /* no estimation for dir size */ + rdhash = au_sbi(sb)->si_rdhash; + if (!rdhash) + rdhash = AUFS_RDHASH_DEF; + err = au_nhash_alloc(&whtmp->whlist, rdhash, gfp); + if (unlikely(err)) { + kfree(whtmp); + whtmp = ERR_PTR(err); + } + +out: + return whtmp; +} + +void au_whtmp_rmdir_free(struct au_whtmp_rmdir *whtmp) +{ + if (whtmp->br) + atomic_dec(&whtmp->br->br_count); + dput(whtmp->wh_dentry); + iput(whtmp->dir); + au_nhash_wh_free(&whtmp->whlist); + kfree(whtmp); +} + +/* + * rmdir the whiteouted temporary named dir @h_dentry. + * @whlist: whiteouted children. + */ +int au_whtmp_rmdir(struct inode *dir, aufs_bindex_t bindex, + struct dentry *wh_dentry, struct au_nhash *whlist) +{ + int err; + struct path h_tmp; + struct inode *wh_inode, *h_dir; + struct au_branch *br; + + h_dir = wh_dentry->d_parent->d_inode; /* dir inode is locked */ + IMustLock(h_dir); + + br = au_sbr(dir->i_sb, bindex); + wh_inode = wh_dentry->d_inode; + mutex_lock_nested(&wh_inode->i_mutex, AuLsc_I_CHILD); + + /* + * someone else might change some whiteouts while we were sleeping. + * it means this whlist may have an obsoleted entry. + */ + if (!au_test_h_perm_sio(wh_inode, MAY_EXEC | MAY_WRITE)) + err = del_wh_children(wh_dentry, whlist, bindex, br); + else { + int wkq_err; + struct del_wh_children_args args = { + .errp = &err, + .h_dentry = wh_dentry, + .whlist = whlist, + .bindex = bindex, + .br = br + }; + + wkq_err = au_wkq_wait(call_del_wh_children, &args); + if (unlikely(wkq_err)) + err = wkq_err; + } + mutex_unlock(&wh_inode->i_mutex); + + if (!err) { + h_tmp.dentry = wh_dentry; + h_tmp.mnt = au_br_mnt(br); + err = vfsub_rmdir(h_dir, &h_tmp); + } + + if (!err) { + if (au_ibstart(dir) == bindex) { + /* todo: dir->i_mutex is necessary */ + au_cpup_attr_timesizes(dir); + vfsub_drop_nlink(dir); + } + return 0; /* success */ + } + + pr_warn("failed removing %.*s(%d), ignored\n", + AuDLNPair(wh_dentry), err); + return err; +} + +static void call_rmdir_whtmp(void *args) +{ + int err; + aufs_bindex_t bindex; + struct au_whtmp_rmdir *a = args; + struct super_block *sb; + struct dentry *h_parent; + struct inode *h_dir; + struct au_hinode *hdir; + + /* rmdir by nfsd may cause deadlock with this i_mutex */ + /* mutex_lock(&a->dir->i_mutex); */ + err = -EROFS; + sb = a->dir->i_sb; + si_read_lock(sb, !AuLock_FLUSH); + if (!au_br_writable(a->br->br_perm)) + goto out; + bindex = au_br_index(sb, a->br->br_id); + if (unlikely(bindex < 0)) + goto out; + + err = -EIO; + ii_write_lock_parent(a->dir); + h_parent = dget_parent(a->wh_dentry); + h_dir = h_parent->d_inode; + hdir = au_hi(a->dir, bindex); + err = vfsub_mnt_want_write(au_br_mnt(a->br)); + if (unlikely(err)) + goto out_mnt; + au_hn_imtx_lock_nested(hdir, AuLsc_I_PARENT); + err = au_h_verify(a->wh_dentry, au_opt_udba(sb), h_dir, h_parent, + a->br); + if (!err) + err = au_whtmp_rmdir(a->dir, bindex, a->wh_dentry, &a->whlist); + au_hn_imtx_unlock(hdir); + vfsub_mnt_drop_write(au_br_mnt(a->br)); + +out_mnt: + dput(h_parent); + ii_write_unlock(a->dir); +out: + /* mutex_unlock(&a->dir->i_mutex); */ + au_whtmp_rmdir_free(a); + si_read_unlock(sb); + au_nwt_done(&au_sbi(sb)->si_nowait); + if (unlikely(err)) + AuIOErr("err %d\n", err); +} + +void au_whtmp_kick_rmdir(struct inode *dir, aufs_bindex_t bindex, + struct dentry *wh_dentry, struct au_whtmp_rmdir *args) +{ + int wkq_err; + struct super_block *sb; + + IMustLock(dir); + + /* all post-process will be done in do_rmdir_whtmp(). */ + sb = dir->i_sb; + args->dir = au_igrab(dir); + args->br = au_sbr(sb, bindex); + atomic_inc(&args->br->br_count); + args->wh_dentry = dget(wh_dentry); + wkq_err = au_wkq_nowait(call_rmdir_whtmp, args, sb, /*flags*/0); + if (unlikely(wkq_err)) { + pr_warn("rmdir error %.*s (%d), ignored\n", + AuDLNPair(wh_dentry), wkq_err); + au_whtmp_rmdir_free(args); + } +} diff -Naur kernel.21.3.orig/fs/aufs/whout.h kernel.21.3.new/fs/aufs/whout.h --- kernel.21.3.orig/fs/aufs/whout.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/whout.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,85 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * whiteout for logical deletion and opaque directory + */ + +#ifndef __AUFS_WHOUT_H__ +#define __AUFS_WHOUT_H__ + +#ifdef __KERNEL__ + +#include "dir.h" + +/* whout.c */ +int au_wh_name_alloc(struct qstr *wh, const struct qstr *name); +struct au_branch; +int au_wh_test(struct dentry *h_parent, struct qstr *wh_name, int try_sio); +int au_diropq_test(struct dentry *h_dentry); +struct dentry *au_whtmp_lkup(struct dentry *h_parent, struct au_branch *br, + struct qstr *prefix); +int au_whtmp_ren(struct dentry *h_dentry, struct au_branch *br); +int au_wh_unlink_dentry(struct inode *h_dir, struct path *h_path, + struct dentry *dentry); +int au_wh_init(struct au_branch *br, struct super_block *sb); + +/* diropq flags */ +#define AuDiropq_CREATE 1 +#define au_ftest_diropq(flags, name) ((flags) & AuDiropq_##name) +#define au_fset_diropq(flags, name) \ + do { (flags) |= AuDiropq_##name; } while (0) +#define au_fclr_diropq(flags, name) \ + do { (flags) &= ~AuDiropq_##name; } while (0) + +struct dentry *au_diropq_sio(struct dentry *dentry, aufs_bindex_t bindex, + unsigned int flags); +struct dentry *au_wh_lkup(struct dentry *h_parent, struct qstr *base_name, + struct au_branch *br); +struct dentry *au_wh_create(struct dentry *dentry, aufs_bindex_t bindex, + struct dentry *h_parent); + +/* real rmdir for the whiteout-ed dir */ +struct au_whtmp_rmdir { + struct inode *dir; + struct au_branch *br; + struct dentry *wh_dentry; + struct au_nhash whlist; +}; + +struct au_whtmp_rmdir *au_whtmp_rmdir_alloc(struct super_block *sb, gfp_t gfp); +void au_whtmp_rmdir_free(struct au_whtmp_rmdir *whtmp); +int au_whtmp_rmdir(struct inode *dir, aufs_bindex_t bindex, + struct dentry *wh_dentry, struct au_nhash *whlist); +void au_whtmp_kick_rmdir(struct inode *dir, aufs_bindex_t bindex, + struct dentry *wh_dentry, struct au_whtmp_rmdir *args); + +/* ---------------------------------------------------------------------- */ + +static inline struct dentry *au_diropq_create(struct dentry *dentry, + aufs_bindex_t bindex) +{ + return au_diropq_sio(dentry, bindex, AuDiropq_CREATE); +} + +static inline int au_diropq_remove(struct dentry *dentry, aufs_bindex_t bindex) +{ + return PTR_ERR(au_diropq_sio(dentry, bindex, !AuDiropq_CREATE)); +} + +#endif /* __KERNEL__ */ +#endif /* __AUFS_WHOUT_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/wkq.c kernel.21.3.new/fs/aufs/wkq.c --- kernel.21.3.orig/fs/aufs/wkq.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/wkq.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,212 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * workqueue for asynchronous/super-io operations + * todo: try new dredential scheme + */ + +#include +#include "aufs.h" + +/* internal workqueue named AUFS_WKQ_NAME */ + +static struct workqueue_struct *au_wkq; + +struct au_wkinfo { + struct work_struct wk; + struct kobject *kobj; + + unsigned int flags; /* see wkq.h */ + + au_wkq_func_t func; + void *args; + + struct completion *comp; +}; + +/* ---------------------------------------------------------------------- */ + +static void wkq_func(struct work_struct *wk) +{ + struct au_wkinfo *wkinfo = container_of(wk, struct au_wkinfo, wk); + + AuDebugOn(!uid_eq(current_fsuid(), GLOBAL_ROOT_UID)); + AuDebugOn(rlimit(RLIMIT_FSIZE) != RLIM_INFINITY); + + wkinfo->func(wkinfo->args); + if (au_ftest_wkq(wkinfo->flags, WAIT)) + complete(wkinfo->comp); + else { + kobject_put(wkinfo->kobj); + module_put(THIS_MODULE); /* todo: ?? */ + kfree(wkinfo); + } +} + +/* + * Since struct completion is large, try allocating it dynamically. + */ +#if 1 /* defined(CONFIG_4KSTACKS) || defined(AuTest4KSTACKS) */ +#define AuWkqCompDeclare(name) struct completion *comp = NULL + +static int au_wkq_comp_alloc(struct au_wkinfo *wkinfo, struct completion **comp) +{ + *comp = kmalloc(sizeof(**comp), GFP_NOFS); + if (*comp) { + init_completion(*comp); + wkinfo->comp = *comp; + return 0; + } + return -ENOMEM; +} + +static void au_wkq_comp_free(struct completion *comp) +{ + kfree(comp); +} + +#else + +/* no braces */ +#define AuWkqCompDeclare(name) \ + DECLARE_COMPLETION_ONSTACK(_ ## name); \ + struct completion *comp = &_ ## name + +static int au_wkq_comp_alloc(struct au_wkinfo *wkinfo, struct completion **comp) +{ + wkinfo->comp = *comp; + return 0; +} + +static void au_wkq_comp_free(struct completion *comp __maybe_unused) +{ + /* empty */ +} +#endif /* 4KSTACKS */ + +static void au_wkq_run(struct au_wkinfo *wkinfo) +{ + if (au_ftest_wkq(wkinfo->flags, NEST)) { + if (au_wkq_test()) { + AuWarn1("wkq from wkq, due to a dead dir by UDBA?\n"); + AuDebugOn(au_ftest_wkq(wkinfo->flags, WAIT)); + } + } else + au_dbg_verify_kthread(); + + if (au_ftest_wkq(wkinfo->flags, WAIT)) { + INIT_WORK_ONSTACK(&wkinfo->wk, wkq_func); + queue_work(au_wkq, &wkinfo->wk); + } else { + INIT_WORK(&wkinfo->wk, wkq_func); + schedule_work(&wkinfo->wk); + } +} + +/* + * Be careful. It is easy to make deadlock happen. + * processA: lock, wkq and wait + * processB: wkq and wait, lock in wkq + * --> deadlock + */ +int au_wkq_do_wait(unsigned int flags, au_wkq_func_t func, void *args) +{ + int err; + AuWkqCompDeclare(comp); + struct au_wkinfo wkinfo = { + .flags = flags, + .func = func, + .args = args + }; + + err = au_wkq_comp_alloc(&wkinfo, &comp); + if (!err) { + au_wkq_run(&wkinfo); + /* no timeout, no interrupt */ + wait_for_completion(wkinfo.comp); + au_wkq_comp_free(comp); + destroy_work_on_stack(&wkinfo.wk); + } + + return err; + +} + +/* + * Note: dget/dput() in func for aufs dentries are not supported. It will be a + * problem in a concurrent umounting. + */ +int au_wkq_nowait(au_wkq_func_t func, void *args, struct super_block *sb, + unsigned int flags) +{ + int err; + struct au_wkinfo *wkinfo; + + atomic_inc(&au_sbi(sb)->si_nowait.nw_len); + + /* + * wkq_func() must free this wkinfo. + * it highly depends upon the implementation of workqueue. + */ + err = 0; + wkinfo = kmalloc(sizeof(*wkinfo), GFP_NOFS); + if (wkinfo) { + wkinfo->kobj = &au_sbi(sb)->si_kobj; + wkinfo->flags = flags & ~AuWkq_WAIT; + wkinfo->func = func; + wkinfo->args = args; + wkinfo->comp = NULL; + kobject_get(wkinfo->kobj); + __module_get(THIS_MODULE); /* todo: ?? */ + + au_wkq_run(wkinfo); + } else { + err = -ENOMEM; + au_nwt_done(&au_sbi(sb)->si_nowait); + } + + return err; +} + +/* ---------------------------------------------------------------------- */ + +void au_nwt_init(struct au_nowait_tasks *nwt) +{ + atomic_set(&nwt->nw_len, 0); + /* smp_mb(); */ /* atomic_set */ + init_waitqueue_head(&nwt->nw_wq); +} + +void au_wkq_fin(void) +{ + destroy_workqueue(au_wkq); +} + +int __init au_wkq_init(void) +{ + int err; + + err = 0; + au_wkq = alloc_workqueue(AUFS_WKQ_NAME, 0, WQ_DFL_ACTIVE); + if (IS_ERR(au_wkq)) + err = PTR_ERR(au_wkq); + else if (!au_wkq) + err = -ENOMEM; + + return err; +} diff -Naur kernel.21.3.orig/fs/aufs/wkq.h kernel.21.3.new/fs/aufs/wkq.h --- kernel.21.3.orig/fs/aufs/wkq.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/wkq.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,91 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * workqueue for asynchronous/super-io operations + * todo: try new credentials management scheme + */ + +#ifndef __AUFS_WKQ_H__ +#define __AUFS_WKQ_H__ + +#ifdef __KERNEL__ + +struct super_block; + +/* ---------------------------------------------------------------------- */ + +/* + * in the next operation, wait for the 'nowait' tasks in system-wide workqueue + */ +struct au_nowait_tasks { + atomic_t nw_len; + wait_queue_head_t nw_wq; +}; + +/* ---------------------------------------------------------------------- */ + +typedef void (*au_wkq_func_t)(void *args); + +/* wkq flags */ +#define AuWkq_WAIT 1 +#define AuWkq_NEST (1 << 1) +#define au_ftest_wkq(flags, name) ((flags) & AuWkq_##name) +#define au_fset_wkq(flags, name) \ + do { (flags) |= AuWkq_##name; } while (0) +#define au_fclr_wkq(flags, name) \ + do { (flags) &= ~AuWkq_##name; } while (0) + +#ifndef CONFIG_AUFS_HNOTIFY +#undef AuWkq_NEST +#define AuWkq_NEST 0 +#endif + +/* wkq.c */ +int au_wkq_do_wait(unsigned int flags, au_wkq_func_t func, void *args); +int au_wkq_nowait(au_wkq_func_t func, void *args, struct super_block *sb, + unsigned int flags); +void au_nwt_init(struct au_nowait_tasks *nwt); +int __init au_wkq_init(void); +void au_wkq_fin(void); + +/* ---------------------------------------------------------------------- */ + +static inline int au_wkq_test(void) +{ + return current->flags & PF_WQ_WORKER; +} + +static inline int au_wkq_wait(au_wkq_func_t func, void *args) +{ + return au_wkq_do_wait(AuWkq_WAIT, func, args); +} + +static inline void au_nwt_done(struct au_nowait_tasks *nwt) +{ + if (atomic_dec_and_test(&nwt->nw_len)) + wake_up_all(&nwt->nw_wq); +} + +static inline int au_nwt_flush(struct au_nowait_tasks *nwt) +{ + wait_event(nwt->nw_wq, !atomic_read(&nwt->nw_len)); + return 0; +} + +#endif /* __KERNEL__ */ +#endif /* __AUFS_WKQ_H__ */ diff -Naur kernel.21.3.orig/fs/aufs/xattr.c kernel.21.3.new/fs/aufs/xattr.c --- kernel.21.3.orig/fs/aufs/xattr.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/xattr.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,437 @@ +/* + * Copyright (C) 2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * handling xattr functions + */ + +#include +#include "aufs.h" + +static int au_xattr_ignore(int err, char *name, unsigned int ignore_flags) +{ + if (!ignore_flags) + goto out; + switch (err) { + case -ENOMEM: + case -EDQUOT: + goto out; + } + + if ((ignore_flags & AuBrAttr_ICEX) == AuBrAttr_ICEX) { + err = 0; + goto out; + } + +#define cmp(brattr, prefix) do { \ + if (!strncmp(name, XATTR_##prefix##_PREFIX, \ + XATTR_##prefix##_PREFIX_LEN)) { \ + if (ignore_flags & AuBrAttr_ICEX_##brattr) \ + err = 0; \ + goto out; \ + } \ + } while (0) + + cmp(SEC, SECURITY); + cmp(SYS, SYSTEM); + cmp(TR, TRUSTED); + cmp(USR, USER); +#undef cmp + + if (ignore_flags & AuBrAttr_ICEX_OTH) + err = 0; + +out: + return err; +} + +static const int au_xattr_out_of_list = AuBrAttr_ICEX_OTH << 1; + +static int au_do_cpup_xattr(struct dentry *h_dst, struct dentry *h_src, + char *name, char **buf, unsigned int ignore_flags) +{ + int err; + ssize_t ssz; + struct inode *h_idst; + + ssz = vfs_getxattr_alloc(h_src, name, buf, 0, GFP_NOFS); + err = ssz; + if (unlikely(err <= 0)) { + AuTraceErr(err); + if (err == -ENODATA + || (err == -EOPNOTSUPP + && (ignore_flags & au_xattr_out_of_list))) + err = 0; + goto out; + } + + /* unlock it temporary */ + h_idst = h_dst->d_inode; + mutex_unlock(&h_idst->i_mutex); + err = vfsub_setxattr(h_dst, name, *buf, ssz, /*flags*/0); + mutex_lock_nested(&h_idst->i_mutex, AuLsc_I_CHILD2); + if (unlikely(err)) { + AuDbg("%s, err %d\n", name, err); + err = au_xattr_ignore(err, name, ignore_flags); + } + +out: + return err; +} + +int au_cpup_xattr(struct dentry *h_dst, struct dentry *h_src, int ignore_flags) +{ + int err, unlocked, acl_access, acl_default; + ssize_t ssz; + struct inode *h_isrc, *h_idst; + char *value, *p, *o, *e; + + /* try stopping to update the source inode while we are referencing */ + /* there should not be the parent-child relation ship between them */ + h_isrc = h_src->d_inode; + h_idst = h_dst->d_inode; + mutex_unlock(&h_idst->i_mutex); + mutex_lock_nested(&h_isrc->i_mutex, AuLsc_I_CHILD); + mutex_lock_nested(&h_idst->i_mutex, AuLsc_I_CHILD2); + unlocked = 0; + + /* some filesystems don't list POSIX ACL, for example tmpfs */ + ssz = vfs_listxattr(h_src, NULL, 0); + err = ssz; + if (unlikely(err < 0)) { + if (err == -ENODATA + || err == -EOPNOTSUPP) + err = 0; /* ignore */ + goto out; + } + + err = 0; + p = NULL; + o = NULL; + if (ssz) { + err = -ENOMEM; + p = kmalloc(ssz, GFP_NOFS); + o = p; + if (unlikely(!p)) + goto out; + err = vfs_listxattr(h_src, p, ssz); + } + mutex_unlock(&h_isrc->i_mutex); + unlocked = 1; + AuDbg("err %d, ssz %zd\n", err, ssz); + if (unlikely(err < 0)) + goto out_free; + + err = 0; + e = p + ssz; + value = NULL; + acl_access = 0; + acl_default = 0; + while (!err && p < e) { + acl_access |= !strncmp(p, XATTR_NAME_POSIX_ACL_ACCESS, + sizeof(XATTR_NAME_POSIX_ACL_ACCESS) - 1); + acl_default |= !strncmp(p, XATTR_NAME_POSIX_ACL_DEFAULT, + sizeof(XATTR_NAME_POSIX_ACL_DEFAULT) + - 1); + err = au_do_cpup_xattr(h_dst, h_src, p, &value, ignore_flags); + p += strlen(p) + 1; + } + AuTraceErr(err); + ignore_flags |= au_xattr_out_of_list; + if (!err && !acl_access) { + err = au_do_cpup_xattr(h_dst, h_src, + XATTR_NAME_POSIX_ACL_ACCESS, &value, + ignore_flags); + AuTraceErr(err); + } + if (!err && !acl_default) { + err = au_do_cpup_xattr(h_dst, h_src, + XATTR_NAME_POSIX_ACL_DEFAULT, &value, + ignore_flags); + AuTraceErr(err); + } + + kfree(value); + +out_free: + kfree(o); +out: + if (!unlocked) + mutex_unlock(&h_isrc->i_mutex); + AuTraceErr(err); + return err; +} + +/* ---------------------------------------------------------------------- */ + +enum { + AU_XATTR_LIST, + AU_XATTR_GET +}; + +struct au_lgxattr { + int type; + union { + struct { + char *list; + size_t size; + } list; + struct { + const char *name; + void *value; + size_t size; + } get; + } u; +}; + +static ssize_t au_lgxattr(struct dentry *dentry, struct au_lgxattr *arg) +{ + ssize_t err; + struct path h_path; + struct super_block *sb; + + sb = dentry->d_sb; + err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM); + if (unlikely(err)) + goto out; + err = au_h_path_getattr(dentry, /*force*/1, &h_path); + if (unlikely(err)) + goto out_si; + if (unlikely(!h_path.dentry)) + /* illegally overlapped or something */ + goto out_di; /* pretending success */ + + /* always topmost entry only */ + switch (arg->type) { + case AU_XATTR_LIST: + err = vfs_listxattr(h_path.dentry, + arg->u.list.list, arg->u.list.size); + break; + case AU_XATTR_GET: + err = vfs_getxattr(h_path.dentry, + arg->u.get.name, arg->u.get.value, + arg->u.get.size); + break; + } + +out_di: + di_read_unlock(dentry, AuLock_IR); +out_si: + si_read_unlock(sb); +out: + AuTraceErr(err); + return err; +} + +ssize_t aufs_listxattr(struct dentry *dentry, char *list, size_t size) +{ + struct au_lgxattr arg = { + .type = AU_XATTR_LIST, + .u.list = { + .list = list, + .size = size + }, + }; + + return au_lgxattr(dentry, &arg); +} + +ssize_t aufs_getxattr(struct dentry *dentry, const char *name, void *value, + size_t size) +{ + struct au_lgxattr arg = { + .type = AU_XATTR_GET, + .u.get = { + .name = name, + .value = value, + .size = size + }, + }; + + return au_lgxattr(dentry, &arg); +} + +/* cf fs/aufs/i_op.c:aufs_setattr() */ +static int au_h_path_to_set_attr(struct dentry *dentry, + struct au_icpup_args *a, struct path *h_path) +{ + int err; + struct super_block *sb; + + sb = dentry->d_sb; + a->udba = au_opt_udba(sb); + /* no d_unlinked(), to set UDBA_NONE for root */ + if (d_unhashed(dentry)) + a->udba = AuOpt_UDBA_NONE; + if (a->udba != AuOpt_UDBA_NONE) { + AuDebugOn(IS_ROOT(dentry)); + err = au_reval_for_attr(dentry, au_sigen(sb)); + if (unlikely(err)) + goto out; + } + err = au_pin_and_icpup(dentry, /*ia*/NULL, a); + if (unlikely(err < 0)) + goto out; + + h_path->dentry = a->h_path.dentry; + h_path->mnt = au_sbr_mnt(sb, a->btgt); + +out: + return err; +} + +enum { + AU_XATTR_SET, + AU_XATTR_REMOVE +}; + +struct au_srxattr { + int type; + union { + struct { + const char *name; + const void *value; + size_t size; + int flags; + } set; + struct { + const char *name; + } remove; + } u; +}; + +static ssize_t au_srxattr(struct dentry *dentry, struct au_srxattr *arg) +{ + int err; + struct path h_path; + struct super_block *sb; + struct au_icpup_args *a; + struct inode *inode; + + inode = dentry->d_inode; + IMustLock(inode); + + err = -ENOMEM; + a = kzalloc(sizeof(*a), GFP_NOFS); + if (unlikely(!a)) + goto out; + + sb = dentry->d_sb; + err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM); + if (unlikely(err)) + goto out_kfree; + + h_path.dentry = NULL; /* silence gcc */ + di_write_lock_child(dentry); + err = au_h_path_to_set_attr(dentry, a, &h_path); + if (unlikely(err)) + goto out_di; + + mutex_unlock(&a->h_inode->i_mutex); + switch (arg->type) { + case AU_XATTR_SET: + err = vfsub_setxattr(h_path.dentry, + arg->u.set.name, arg->u.set.value, + arg->u.set.size, arg->u.set.flags); + break; + case AU_XATTR_REMOVE: + err = vfsub_removexattr(h_path.dentry, arg->u.remove.name); + break; + } + if (!err) + au_cpup_attr_timesizes(inode); + + au_unpin(&a->pin); + if (unlikely(err)) + au_update_dbstart(dentry); + +out_di: + di_write_unlock(dentry); + si_read_unlock(sb); +out_kfree: + kfree(a); +out: + AuTraceErr(err); + return err; +} + +int aufs_setxattr(struct dentry *dentry, const char *name, const void *value, + size_t size, int flags) +{ + struct au_srxattr arg = { + .type = AU_XATTR_SET, + .u.set = { + .name = name, + .value = value, + .size = size, + .flags = flags + }, + }; + + return au_srxattr(dentry, &arg); +} + +int aufs_removexattr(struct dentry *dentry, const char *name) +{ + struct au_srxattr arg = { + .type = AU_XATTR_REMOVE, + .u.remove = { + .name = name + }, + }; + + return au_srxattr(dentry, &arg); +} + +/* ---------------------------------------------------------------------- */ + +#if 0 +static size_t au_xattr_list(struct dentry *dentry, char *list, size_t list_size, + const char *name, size_t name_len, int type) +{ + return aufs_listxattr(dentry, list, list_size); +} + +static int au_xattr_get(struct dentry *dentry, const char *name, void *buffer, + size_t size, int type) +{ + return aufs_getxattr(dentry, name, buffer, size); +} + +static int au_xattr_set(struct dentry *dentry, const char *name, + const void *value, size_t size, int flags, int type) +{ + return aufs_setxattr(dentry, name, value, size, flags); +} + +static const struct xattr_handler au_xattr_handler = { + /* no prefix, no flags */ + .list = au_xattr_list, + .get = au_xattr_get, + .set = au_xattr_set + /* why no remove? */ +}; + +static const struct xattr_handler *au_xattr_handlers[] = { + &au_xattr_handler +}; + +void au_xattr_init(struct super_block *sb) +{ + /* sb->s_xattr = au_xattr_handlers; */ +} +#endif diff -Naur kernel.21.3.orig/fs/aufs/xino.c kernel.21.3.new/fs/aufs/xino.c --- kernel.21.3.orig/fs/aufs/xino.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/aufs/xino.c 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,1309 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +/* + * external inode number translation table and bitmap + */ + +#include +#include +#include "aufs.h" + +/* todo: unnecessary to support mmap_sem since kernel-space? */ +ssize_t xino_fread(au_readf_t func, struct file *file, void *kbuf, size_t size, + loff_t *pos) +{ + ssize_t err; + mm_segment_t oldfs; + union { + void *k; + char __user *u; + } buf; + + buf.k = kbuf; + oldfs = get_fs(); + set_fs(KERNEL_DS); + do { + /* todo: signal_pending? */ + err = func(file, buf.u, size, pos); + } while (err == -EAGAIN || err == -EINTR); + set_fs(oldfs); + +#if 0 /* reserved for future use */ + if (err > 0) + fsnotify_access(file->f_dentry); +#endif + + return err; +} + +/* ---------------------------------------------------------------------- */ + +static ssize_t do_xino_fwrite(au_writef_t func, struct file *file, void *kbuf, + size_t size, loff_t *pos) +{ + ssize_t err; + mm_segment_t oldfs; + union { + void *k; + const char __user *u; + } buf; + + buf.k = kbuf; + oldfs = get_fs(); + set_fs(KERNEL_DS); + do { + /* todo: signal_pending? */ + err = func(file, buf.u, size, pos); + } while (err == -EAGAIN || err == -EINTR); + set_fs(oldfs); + +#if 0 /* reserved for future use */ + if (err > 0) + fsnotify_modify(file->f_dentry); +#endif + + return err; +} + +struct do_xino_fwrite_args { + ssize_t *errp; + au_writef_t func; + struct file *file; + void *buf; + size_t size; + loff_t *pos; +}; + +static void call_do_xino_fwrite(void *args) +{ + struct do_xino_fwrite_args *a = args; + *a->errp = do_xino_fwrite(a->func, a->file, a->buf, a->size, a->pos); +} + +ssize_t xino_fwrite(au_writef_t func, struct file *file, void *buf, size_t size, + loff_t *pos) +{ + ssize_t err; + + /* todo: signal block and no wkq? */ + if (rlimit(RLIMIT_FSIZE) == RLIM_INFINITY) { + lockdep_off(); + err = do_xino_fwrite(func, file, buf, size, pos); + lockdep_on(); + } else { + /* + * it breaks RLIMIT_FSIZE and normal user's limit, + * users should care about quota and real 'filesystem full.' + */ + int wkq_err; + struct do_xino_fwrite_args args = { + .errp = &err, + .func = func, + .file = file, + .buf = buf, + .size = size, + .pos = pos + }; + + wkq_err = au_wkq_wait(call_do_xino_fwrite, &args); + if (unlikely(wkq_err)) + err = wkq_err; + } + + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* + * create a new xinofile at the same place/path as @base_file. + */ +struct file *au_xino_create2(struct file *base_file, struct file *copy_src) +{ + struct file *file; + struct dentry *base, *parent; + struct inode *dir; + struct qstr *name; + struct path path; + int err; + + base = base_file->f_dentry; + parent = base->d_parent; /* dir inode is locked */ + dir = parent->d_inode; + IMustLock(dir); + + file = ERR_PTR(-EINVAL); + name = &base->d_name; + path.dentry = vfsub_lookup_one_len(name->name, parent, name->len); + if (IS_ERR(path.dentry)) { + file = (void *)path.dentry; + pr_err("%.*s lookup err %ld\n", + AuLNPair(name), PTR_ERR(path.dentry)); + goto out; + } + + /* no need to mnt_want_write() since we call dentry_open() later */ + err = vfs_create(dir, path.dentry, S_IRUGO | S_IWUGO, NULL); + if (unlikely(err)) { + file = ERR_PTR(err); + pr_err("%.*s create err %d\n", AuLNPair(name), err); + goto out_dput; + } + + path.mnt = base_file->f_path.mnt; + file = vfsub_dentry_open(&path, + O_RDWR | O_CREAT | O_EXCL | O_LARGEFILE + /* | __FMODE_NONOTIFY */); + if (IS_ERR(file)) { + pr_err("%.*s open err %ld\n", AuLNPair(name), PTR_ERR(file)); + goto out_dput; + } + + err = vfsub_unlink(dir, &file->f_path, /*force*/0); + if (unlikely(err)) { + pr_err("%.*s unlink err %d\n", AuLNPair(name), err); + goto out_fput; + } + + if (copy_src) { + /* no one can touch copy_src xino */ + err = au_copy_file(file, copy_src, vfsub_f_size_read(copy_src)); + if (unlikely(err)) { + pr_err("%.*s copy err %d\n", AuLNPair(name), err); + goto out_fput; + } + } + goto out_dput; /* success */ + +out_fput: + fput(file); + file = ERR_PTR(err); +out_dput: + dput(path.dentry); +out: + return file; +} + +struct au_xino_lock_dir { + struct au_hinode *hdir; + struct dentry *parent; + struct mutex *mtx; +}; + +static void au_xino_lock_dir(struct super_block *sb, struct file *xino, + struct au_xino_lock_dir *ldir) +{ + aufs_bindex_t brid, bindex; + + ldir->hdir = NULL; + bindex = -1; + brid = au_xino_brid(sb); + if (brid >= 0) + bindex = au_br_index(sb, brid); + if (bindex >= 0) { + ldir->hdir = au_hi(sb->s_root->d_inode, bindex); + au_hn_imtx_lock_nested(ldir->hdir, AuLsc_I_PARENT); + } else { + ldir->parent = dget_parent(xino->f_dentry); + ldir->mtx = &ldir->parent->d_inode->i_mutex; + mutex_lock_nested(ldir->mtx, AuLsc_I_PARENT); + } +} + +static void au_xino_unlock_dir(struct au_xino_lock_dir *ldir) +{ + if (ldir->hdir) + au_hn_imtx_unlock(ldir->hdir); + else { + mutex_unlock(ldir->mtx); + dput(ldir->parent); + } +} + +/* ---------------------------------------------------------------------- */ + +/* trucate xino files asynchronously */ + +int au_xino_trunc(struct super_block *sb, aufs_bindex_t bindex) +{ + int err; + unsigned long jiffy; + blkcnt_t blocks; + aufs_bindex_t bi, bend; + struct kstatfs *st; + struct au_branch *br; + struct file *new_xino, *file; + struct super_block *h_sb; + struct au_xino_lock_dir ldir; + + err = -ENOMEM; + st = kzalloc(sizeof(*st), GFP_NOFS); + if (unlikely(!st)) + goto out; + + err = -EINVAL; + bend = au_sbend(sb); + if (unlikely(bindex < 0 || bend < bindex)) + goto out_st; + br = au_sbr(sb, bindex); + file = br->br_xino.xi_file; + if (!file) + goto out_st; + + err = vfs_statfs(&file->f_path, st); + if (unlikely(err)) + AuErr1("statfs err %d, ignored\n", err); + jiffy = jiffies; + blocks = file_inode(file)->i_blocks; + pr_info("begin truncating xino(b%d), ib%llu, %llu/%llu free blks\n", + bindex, (u64)blocks, st->f_bfree, st->f_blocks); + + au_xino_lock_dir(sb, file, &ldir); + /* mnt_want_write() is unnecessary here */ + new_xino = au_xino_create2(file, file); + au_xino_unlock_dir(&ldir); + err = PTR_ERR(new_xino); + if (IS_ERR(new_xino)) { + pr_err("err %d, ignored\n", err); + goto out_st; + } + err = 0; + fput(file); + br->br_xino.xi_file = new_xino; + + h_sb = au_br_sb(br); + for (bi = 0; bi <= bend; bi++) { + if (unlikely(bi == bindex)) + continue; + br = au_sbr(sb, bi); + if (au_br_sb(br) != h_sb) + continue; + + fput(br->br_xino.xi_file); + br->br_xino.xi_file = new_xino; + get_file(new_xino); + } + + err = vfs_statfs(&new_xino->f_path, st); + if (!err) { + pr_info("end truncating xino(b%d), ib%llu, %llu/%llu free blks\n", + bindex, (u64)file_inode(new_xino)->i_blocks, + st->f_bfree, st->f_blocks); + if (file_inode(new_xino)->i_blocks < blocks) + au_sbi(sb)->si_xino_jiffy = jiffy; + } else + AuErr1("statfs err %d, ignored\n", err); + +out_st: + kfree(st); +out: + return err; +} + +struct xino_do_trunc_args { + struct super_block *sb; + struct au_branch *br; +}; + +static void xino_do_trunc(void *_args) +{ + struct xino_do_trunc_args *args = _args; + struct super_block *sb; + struct au_branch *br; + struct inode *dir; + int err; + aufs_bindex_t bindex; + + err = 0; + sb = args->sb; + dir = sb->s_root->d_inode; + br = args->br; + + si_noflush_write_lock(sb); + ii_read_lock_parent(dir); + bindex = au_br_index(sb, br->br_id); + err = au_xino_trunc(sb, bindex); + ii_read_unlock(dir); + if (unlikely(err)) + pr_warn("err b%d, (%d)\n", bindex, err); + atomic_dec(&br->br_xino_running); + atomic_dec(&br->br_count); + si_write_unlock(sb); + au_nwt_done(&au_sbi(sb)->si_nowait); + kfree(args); +} + +static int xino_trunc_test(struct super_block *sb, struct au_branch *br) +{ + int err; + struct kstatfs st; + struct au_sbinfo *sbinfo; + + /* todo: si_xino_expire and the ratio should be customizable */ + sbinfo = au_sbi(sb); + if (time_before(jiffies, + sbinfo->si_xino_jiffy + sbinfo->si_xino_expire)) + return 0; + + /* truncation border */ + err = vfs_statfs(&br->br_xino.xi_file->f_path, &st); + if (unlikely(err)) { + AuErr1("statfs err %d, ignored\n", err); + return 0; + } + if (div64_u64(st.f_bfree * 100, st.f_blocks) >= AUFS_XINO_DEF_TRUNC) + return 0; + + return 1; +} + +static void xino_try_trunc(struct super_block *sb, struct au_branch *br) +{ + struct xino_do_trunc_args *args; + int wkq_err; + + if (!xino_trunc_test(sb, br)) + return; + + if (atomic_inc_return(&br->br_xino_running) > 1) + goto out; + + /* lock and kfree() will be called in trunc_xino() */ + args = kmalloc(sizeof(*args), GFP_NOFS); + if (unlikely(!args)) { + AuErr1("no memory\n"); + goto out_args; + } + + atomic_inc(&br->br_count); + args->sb = sb; + args->br = br; + wkq_err = au_wkq_nowait(xino_do_trunc, args, sb, /*flags*/0); + if (!wkq_err) + return; /* success */ + + pr_err("wkq %d\n", wkq_err); + atomic_dec(&br->br_count); + +out_args: + kfree(args); +out: + atomic_dec(&br->br_xino_running); +} + +/* ---------------------------------------------------------------------- */ + +static int au_xino_do_write(au_writef_t write, struct file *file, + ino_t h_ino, ino_t ino) +{ + loff_t pos; + ssize_t sz; + + pos = h_ino; + if (unlikely(au_loff_max / sizeof(ino) - 1 < pos)) { + AuIOErr1("too large hi%lu\n", (unsigned long)h_ino); + return -EFBIG; + } + pos *= sizeof(ino); + sz = xino_fwrite(write, file, &ino, sizeof(ino), &pos); + if (sz == sizeof(ino)) + return 0; /* success */ + + AuIOErr("write failed (%zd)\n", sz); + return -EIO; +} + +/* + * write @ino to the xinofile for the specified branch{@sb, @bindex} + * at the position of @h_ino. + * even if @ino is zero, it is written to the xinofile and means no entry. + * if the size of the xino file on a specific filesystem exceeds the watermark, + * try truncating it. + */ +int au_xino_write(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino, + ino_t ino) +{ + int err; + unsigned int mnt_flags; + struct au_branch *br; + + BUILD_BUG_ON(sizeof(long long) != sizeof(au_loff_max) + || ((loff_t)-1) > 0); + SiMustAnyLock(sb); + + mnt_flags = au_mntflags(sb); + if (!au_opt_test(mnt_flags, XINO)) + return 0; + + br = au_sbr(sb, bindex); + err = au_xino_do_write(au_sbi(sb)->si_xwrite, br->br_xino.xi_file, + h_ino, ino); + if (!err) { + if (au_opt_test(mnt_flags, TRUNC_XINO) + && au_test_fs_trunc_xino(au_br_sb(br))) + xino_try_trunc(sb, br); + return 0; /* success */ + } + + AuIOErr("write failed (%d)\n", err); + return -EIO; +} + +/* ---------------------------------------------------------------------- */ + +/* aufs inode number bitmap */ + +static const int page_bits = (int)PAGE_SIZE * BITS_PER_BYTE; +static ino_t xib_calc_ino(unsigned long pindex, int bit) +{ + ino_t ino; + + AuDebugOn(bit < 0 || page_bits <= bit); + ino = AUFS_FIRST_INO + pindex * page_bits + bit; + return ino; +} + +static void xib_calc_bit(ino_t ino, unsigned long *pindex, int *bit) +{ + AuDebugOn(ino < AUFS_FIRST_INO); + ino -= AUFS_FIRST_INO; + *pindex = ino / page_bits; + *bit = ino % page_bits; +} + +static int xib_pindex(struct super_block *sb, unsigned long pindex) +{ + int err; + loff_t pos; + ssize_t sz; + struct au_sbinfo *sbinfo; + struct file *xib; + unsigned long *p; + + sbinfo = au_sbi(sb); + MtxMustLock(&sbinfo->si_xib_mtx); + AuDebugOn(pindex > ULONG_MAX / PAGE_SIZE + || !au_opt_test(sbinfo->si_mntflags, XINO)); + + if (pindex == sbinfo->si_xib_last_pindex) + return 0; + + xib = sbinfo->si_xib; + p = sbinfo->si_xib_buf; + pos = sbinfo->si_xib_last_pindex; + pos *= PAGE_SIZE; + sz = xino_fwrite(sbinfo->si_xwrite, xib, p, PAGE_SIZE, &pos); + if (unlikely(sz != PAGE_SIZE)) + goto out; + + pos = pindex; + pos *= PAGE_SIZE; + if (vfsub_f_size_read(xib) >= pos + PAGE_SIZE) + sz = xino_fread(sbinfo->si_xread, xib, p, PAGE_SIZE, &pos); + else { + memset(p, 0, PAGE_SIZE); + sz = xino_fwrite(sbinfo->si_xwrite, xib, p, PAGE_SIZE, &pos); + } + if (sz == PAGE_SIZE) { + sbinfo->si_xib_last_pindex = pindex; + return 0; /* success */ + } + +out: + AuIOErr1("write failed (%zd)\n", sz); + err = sz; + if (sz >= 0) + err = -EIO; + return err; +} + +/* ---------------------------------------------------------------------- */ + +static void au_xib_clear_bit(struct inode *inode) +{ + int err, bit; + unsigned long pindex; + struct super_block *sb; + struct au_sbinfo *sbinfo; + + AuDebugOn(inode->i_nlink); + + sb = inode->i_sb; + xib_calc_bit(inode->i_ino, &pindex, &bit); + AuDebugOn(page_bits <= bit); + sbinfo = au_sbi(sb); + mutex_lock(&sbinfo->si_xib_mtx); + err = xib_pindex(sb, pindex); + if (!err) { + clear_bit(bit, sbinfo->si_xib_buf); + sbinfo->si_xib_next_bit = bit; + } + mutex_unlock(&sbinfo->si_xib_mtx); +} + +/* for s_op->delete_inode() */ +void au_xino_delete_inode(struct inode *inode, const int unlinked) +{ + int err; + unsigned int mnt_flags; + aufs_bindex_t bindex, bend, bi; + unsigned char try_trunc; + struct au_iinfo *iinfo; + struct super_block *sb; + struct au_hinode *hi; + struct inode *h_inode; + struct au_branch *br; + au_writef_t xwrite; + + sb = inode->i_sb; + mnt_flags = au_mntflags(sb); + if (!au_opt_test(mnt_flags, XINO) + || inode->i_ino == AUFS_ROOT_INO) + return; + + if (unlinked) { + au_xigen_inc(inode); + au_xib_clear_bit(inode); + } + + iinfo = au_ii(inode); + if (!iinfo) + return; + + bindex = iinfo->ii_bstart; + if (bindex < 0) + return; + + xwrite = au_sbi(sb)->si_xwrite; + try_trunc = !!au_opt_test(mnt_flags, TRUNC_XINO); + hi = iinfo->ii_hinode + bindex; + bend = iinfo->ii_bend; + for (; bindex <= bend; bindex++, hi++) { + h_inode = hi->hi_inode; + if (!h_inode + || (!unlinked && h_inode->i_nlink)) + continue; + + /* inode may not be revalidated */ + bi = au_br_index(sb, hi->hi_id); + if (bi < 0) + continue; + + br = au_sbr(sb, bi); + err = au_xino_do_write(xwrite, br->br_xino.xi_file, + h_inode->i_ino, /*ino*/0); + if (!err && try_trunc + && au_test_fs_trunc_xino(au_br_sb(br))) + xino_try_trunc(sb, br); + } +} + +/* get an unused inode number from bitmap */ +ino_t au_xino_new_ino(struct super_block *sb) +{ + ino_t ino; + unsigned long *p, pindex, ul, pend; + struct au_sbinfo *sbinfo; + struct file *file; + int free_bit, err; + + if (!au_opt_test(au_mntflags(sb), XINO)) + return iunique(sb, AUFS_FIRST_INO); + + sbinfo = au_sbi(sb); + mutex_lock(&sbinfo->si_xib_mtx); + p = sbinfo->si_xib_buf; + free_bit = sbinfo->si_xib_next_bit; + if (free_bit < page_bits && !test_bit(free_bit, p)) + goto out; /* success */ + free_bit = find_first_zero_bit(p, page_bits); + if (free_bit < page_bits) + goto out; /* success */ + + pindex = sbinfo->si_xib_last_pindex; + for (ul = pindex - 1; ul < ULONG_MAX; ul--) { + err = xib_pindex(sb, ul); + if (unlikely(err)) + goto out_err; + free_bit = find_first_zero_bit(p, page_bits); + if (free_bit < page_bits) + goto out; /* success */ + } + + file = sbinfo->si_xib; + pend = vfsub_f_size_read(file) / PAGE_SIZE; + for (ul = pindex + 1; ul <= pend; ul++) { + err = xib_pindex(sb, ul); + if (unlikely(err)) + goto out_err; + free_bit = find_first_zero_bit(p, page_bits); + if (free_bit < page_bits) + goto out; /* success */ + } + BUG(); + +out: + set_bit(free_bit, p); + sbinfo->si_xib_next_bit = free_bit + 1; + pindex = sbinfo->si_xib_last_pindex; + mutex_unlock(&sbinfo->si_xib_mtx); + ino = xib_calc_ino(pindex, free_bit); + AuDbg("i%lu\n", (unsigned long)ino); + return ino; +out_err: + mutex_unlock(&sbinfo->si_xib_mtx); + AuDbg("i0\n"); + return 0; +} + +/* + * read @ino from xinofile for the specified branch{@sb, @bindex} + * at the position of @h_ino. + * if @ino does not exist and @do_new is true, get new one. + */ +int au_xino_read(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino, + ino_t *ino) +{ + int err; + ssize_t sz; + loff_t pos; + struct file *file; + struct au_sbinfo *sbinfo; + + *ino = 0; + if (!au_opt_test(au_mntflags(sb), XINO)) + return 0; /* no xino */ + + err = 0; + sbinfo = au_sbi(sb); + pos = h_ino; + if (unlikely(au_loff_max / sizeof(*ino) - 1 < pos)) { + AuIOErr1("too large hi%lu\n", (unsigned long)h_ino); + return -EFBIG; + } + pos *= sizeof(*ino); + + file = au_sbr(sb, bindex)->br_xino.xi_file; + if (vfsub_f_size_read(file) < pos + sizeof(*ino)) + return 0; /* no ino */ + + sz = xino_fread(sbinfo->si_xread, file, ino, sizeof(*ino), &pos); + if (sz == sizeof(*ino)) + return 0; /* success */ + + err = sz; + if (unlikely(sz >= 0)) { + err = -EIO; + AuIOErr("xino read error (%zd)\n", sz); + } + + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* create and set a new xino file */ + +struct file *au_xino_create(struct super_block *sb, char *fname, int silent) +{ + struct file *file; + struct dentry *h_parent, *d; + struct inode *h_dir; + int err; + + /* + * at mount-time, and the xino file is the default path, + * hnotify is disabled so we have no notify events to ignore. + * when a user specified the xino, we cannot get au_hdir to be ignored. + */ + file = vfsub_filp_open(fname, O_RDWR | O_CREAT | O_EXCL | O_LARGEFILE + /* | __FMODE_NONOTIFY */, + S_IRUGO | S_IWUGO); + if (IS_ERR(file)) { + if (!silent) + pr_err("open %s(%ld)\n", fname, PTR_ERR(file)); + return file; + } + + /* keep file count */ + h_parent = dget_parent(file->f_dentry); + h_dir = h_parent->d_inode; + mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_PARENT); + /* mnt_want_write() is unnecessary here */ + err = vfsub_unlink(h_dir, &file->f_path, /*force*/0); + mutex_unlock(&h_dir->i_mutex); + dput(h_parent); + if (unlikely(err)) { + if (!silent) + pr_err("unlink %s(%d)\n", fname, err); + goto out; + } + + err = -EINVAL; + d = file->f_dentry; + if (unlikely(sb == d->d_sb)) { + if (!silent) + pr_err("%s must be outside\n", fname); + goto out; + } + if (unlikely(au_test_fs_bad_xino(d->d_sb))) { + if (!silent) + pr_err("xino doesn't support %s(%s)\n", + fname, au_sbtype(d->d_sb)); + goto out; + } + return file; /* success */ + +out: + fput(file); + file = ERR_PTR(err); + return file; +} + +/* + * find another branch who is on the same filesystem of the specified + * branch{@btgt}. search until @bend. + */ +static int is_sb_shared(struct super_block *sb, aufs_bindex_t btgt, + aufs_bindex_t bend) +{ + aufs_bindex_t bindex; + struct super_block *tgt_sb = au_sbr_sb(sb, btgt); + + for (bindex = 0; bindex < btgt; bindex++) + if (unlikely(tgt_sb == au_sbr_sb(sb, bindex))) + return bindex; + for (bindex++; bindex <= bend; bindex++) + if (unlikely(tgt_sb == au_sbr_sb(sb, bindex))) + return bindex; + return -1; +} + +/* ---------------------------------------------------------------------- */ + +/* + * initialize the xinofile for the specified branch @br + * at the place/path where @base_file indicates. + * test whether another branch is on the same filesystem or not, + * if @do_test is true. + */ +int au_xino_br(struct super_block *sb, struct au_branch *br, ino_t h_ino, + struct file *base_file, int do_test) +{ + int err; + ino_t ino; + aufs_bindex_t bend, bindex; + struct au_branch *shared_br, *b; + struct file *file; + struct super_block *tgt_sb; + + shared_br = NULL; + bend = au_sbend(sb); + if (do_test) { + tgt_sb = au_br_sb(br); + for (bindex = 0; bindex <= bend; bindex++) { + b = au_sbr(sb, bindex); + if (tgt_sb == au_br_sb(b)) { + shared_br = b; + break; + } + } + } + + if (!shared_br || !shared_br->br_xino.xi_file) { + struct au_xino_lock_dir ldir; + + au_xino_lock_dir(sb, base_file, &ldir); + /* mnt_want_write() is unnecessary here */ + file = au_xino_create2(base_file, NULL); + au_xino_unlock_dir(&ldir); + err = PTR_ERR(file); + if (IS_ERR(file)) + goto out; + br->br_xino.xi_file = file; + } else { + br->br_xino.xi_file = shared_br->br_xino.xi_file; + get_file(br->br_xino.xi_file); + } + + ino = AUFS_ROOT_INO; + err = au_xino_do_write(au_sbi(sb)->si_xwrite, br->br_xino.xi_file, + h_ino, ino); + if (unlikely(err)) { + fput(br->br_xino.xi_file); + br->br_xino.xi_file = NULL; + } + +out: + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* trucate a xino bitmap file */ + +/* todo: slow */ +static int do_xib_restore(struct super_block *sb, struct file *file, void *page) +{ + int err, bit; + ssize_t sz; + unsigned long pindex; + loff_t pos, pend; + struct au_sbinfo *sbinfo; + au_readf_t func; + ino_t *ino; + unsigned long *p; + + err = 0; + sbinfo = au_sbi(sb); + MtxMustLock(&sbinfo->si_xib_mtx); + p = sbinfo->si_xib_buf; + func = sbinfo->si_xread; + pend = vfsub_f_size_read(file); + pos = 0; + while (pos < pend) { + sz = xino_fread(func, file, page, PAGE_SIZE, &pos); + err = sz; + if (unlikely(sz <= 0)) + goto out; + + err = 0; + for (ino = page; sz > 0; ino++, sz -= sizeof(ino)) { + if (unlikely(*ino < AUFS_FIRST_INO)) + continue; + + xib_calc_bit(*ino, &pindex, &bit); + AuDebugOn(page_bits <= bit); + err = xib_pindex(sb, pindex); + if (!err) + set_bit(bit, p); + else + goto out; + } + } + +out: + return err; +} + +static int xib_restore(struct super_block *sb) +{ + int err; + aufs_bindex_t bindex, bend; + void *page; + + err = -ENOMEM; + page = (void *)__get_free_page(GFP_NOFS); + if (unlikely(!page)) + goto out; + + err = 0; + bend = au_sbend(sb); + for (bindex = 0; !err && bindex <= bend; bindex++) + if (!bindex || is_sb_shared(sb, bindex, bindex - 1) < 0) + err = do_xib_restore + (sb, au_sbr(sb, bindex)->br_xino.xi_file, page); + else + AuDbg("b%d\n", bindex); + free_page((unsigned long)page); + +out: + return err; +} + +int au_xib_trunc(struct super_block *sb) +{ + int err; + ssize_t sz; + loff_t pos; + struct au_xino_lock_dir ldir; + struct au_sbinfo *sbinfo; + unsigned long *p; + struct file *file; + + SiMustWriteLock(sb); + + err = 0; + sbinfo = au_sbi(sb); + if (!au_opt_test(sbinfo->si_mntflags, XINO)) + goto out; + + file = sbinfo->si_xib; + if (vfsub_f_size_read(file) <= PAGE_SIZE) + goto out; + + au_xino_lock_dir(sb, file, &ldir); + /* mnt_want_write() is unnecessary here */ + file = au_xino_create2(sbinfo->si_xib, NULL); + au_xino_unlock_dir(&ldir); + err = PTR_ERR(file); + if (IS_ERR(file)) + goto out; + fput(sbinfo->si_xib); + sbinfo->si_xib = file; + + p = sbinfo->si_xib_buf; + memset(p, 0, PAGE_SIZE); + pos = 0; + sz = xino_fwrite(sbinfo->si_xwrite, sbinfo->si_xib, p, PAGE_SIZE, &pos); + if (unlikely(sz != PAGE_SIZE)) { + err = sz; + AuIOErr("err %d\n", err); + if (sz >= 0) + err = -EIO; + goto out; + } + + mutex_lock(&sbinfo->si_xib_mtx); + /* mnt_want_write() is unnecessary here */ + err = xib_restore(sb); + mutex_unlock(&sbinfo->si_xib_mtx); + +out: + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* + * xino mount option handlers + */ +static au_readf_t find_readf(struct file *h_file) +{ + const struct file_operations *fop = h_file->f_op; + + if (fop) { + if (fop->read) + return fop->read; + if (fop->aio_read) + return do_sync_read; + } + return ERR_PTR(-ENOSYS); +} + +static au_writef_t find_writef(struct file *h_file) +{ + const struct file_operations *fop = h_file->f_op; + + if (fop) { + if (fop->write) + return fop->write; + if (fop->aio_write) + return do_sync_write; + } + return ERR_PTR(-ENOSYS); +} + +/* xino bitmap */ +static void xino_clear_xib(struct super_block *sb) +{ + struct au_sbinfo *sbinfo; + + SiMustWriteLock(sb); + + sbinfo = au_sbi(sb); + sbinfo->si_xread = NULL; + sbinfo->si_xwrite = NULL; + if (sbinfo->si_xib) + fput(sbinfo->si_xib); + sbinfo->si_xib = NULL; + free_page((unsigned long)sbinfo->si_xib_buf); + sbinfo->si_xib_buf = NULL; +} + +static int au_xino_set_xib(struct super_block *sb, struct file *base) +{ + int err; + loff_t pos; + struct au_sbinfo *sbinfo; + struct file *file; + + SiMustWriteLock(sb); + + sbinfo = au_sbi(sb); + file = au_xino_create2(base, sbinfo->si_xib); + err = PTR_ERR(file); + if (IS_ERR(file)) + goto out; + if (sbinfo->si_xib) + fput(sbinfo->si_xib); + sbinfo->si_xib = file; + sbinfo->si_xread = find_readf(file); + sbinfo->si_xwrite = find_writef(file); + + err = -ENOMEM; + if (!sbinfo->si_xib_buf) + sbinfo->si_xib_buf = (void *)get_zeroed_page(GFP_NOFS); + if (unlikely(!sbinfo->si_xib_buf)) + goto out_unset; + + sbinfo->si_xib_last_pindex = 0; + sbinfo->si_xib_next_bit = 0; + if (vfsub_f_size_read(file) < PAGE_SIZE) { + pos = 0; + err = xino_fwrite(sbinfo->si_xwrite, file, sbinfo->si_xib_buf, + PAGE_SIZE, &pos); + if (unlikely(err != PAGE_SIZE)) + goto out_free; + } + err = 0; + goto out; /* success */ + +out_free: + free_page((unsigned long)sbinfo->si_xib_buf); + sbinfo->si_xib_buf = NULL; + if (err >= 0) + err = -EIO; +out_unset: + fput(sbinfo->si_xib); + sbinfo->si_xib = NULL; + sbinfo->si_xread = NULL; + sbinfo->si_xwrite = NULL; +out: + return err; +} + +/* xino for each branch */ +static void xino_clear_br(struct super_block *sb) +{ + aufs_bindex_t bindex, bend; + struct au_branch *br; + + bend = au_sbend(sb); + for (bindex = 0; bindex <= bend; bindex++) { + br = au_sbr(sb, bindex); + if (!br || !br->br_xino.xi_file) + continue; + + fput(br->br_xino.xi_file); + br->br_xino.xi_file = NULL; + } +} + +static int au_xino_set_br(struct super_block *sb, struct file *base) +{ + int err; + ino_t ino; + aufs_bindex_t bindex, bend, bshared; + struct { + struct file *old, *new; + } *fpair, *p; + struct au_branch *br; + struct inode *inode; + au_writef_t writef; + + SiMustWriteLock(sb); + + err = -ENOMEM; + bend = au_sbend(sb); + fpair = kcalloc(bend + 1, sizeof(*fpair), GFP_NOFS); + if (unlikely(!fpair)) + goto out; + + inode = sb->s_root->d_inode; + ino = AUFS_ROOT_INO; + writef = au_sbi(sb)->si_xwrite; + for (bindex = 0, p = fpair; bindex <= bend; bindex++, p++) { + br = au_sbr(sb, bindex); + bshared = is_sb_shared(sb, bindex, bindex - 1); + if (bshared >= 0) { + /* shared xino */ + *p = fpair[bshared]; + get_file(p->new); + } + + if (!p->new) { + /* new xino */ + p->old = br->br_xino.xi_file; + p->new = au_xino_create2(base, br->br_xino.xi_file); + err = PTR_ERR(p->new); + if (IS_ERR(p->new)) { + p->new = NULL; + goto out_pair; + } + } + + err = au_xino_do_write(writef, p->new, + au_h_iptr(inode, bindex)->i_ino, ino); + if (unlikely(err)) + goto out_pair; + } + + for (bindex = 0, p = fpair; bindex <= bend; bindex++, p++) { + br = au_sbr(sb, bindex); + if (br->br_xino.xi_file) + fput(br->br_xino.xi_file); + get_file(p->new); + br->br_xino.xi_file = p->new; + } + +out_pair: + for (bindex = 0, p = fpair; bindex <= bend; bindex++, p++) + if (p->new) + fput(p->new); + else + break; + kfree(fpair); +out: + return err; +} + +void au_xino_clr(struct super_block *sb) +{ + struct au_sbinfo *sbinfo; + + au_xigen_clr(sb); + xino_clear_xib(sb); + xino_clear_br(sb); + sbinfo = au_sbi(sb); + /* lvalue, do not call au_mntflags() */ + au_opt_clr(sbinfo->si_mntflags, XINO); +} + +int au_xino_set(struct super_block *sb, struct au_opt_xino *xino, int remount) +{ + int err, skip; + struct dentry *parent, *cur_parent; + struct qstr *dname, *cur_name; + struct file *cur_xino; + struct inode *dir; + struct au_sbinfo *sbinfo; + + SiMustWriteLock(sb); + + err = 0; + sbinfo = au_sbi(sb); + parent = dget_parent(xino->file->f_dentry); + if (remount) { + skip = 0; + dname = &xino->file->f_dentry->d_name; + cur_xino = sbinfo->si_xib; + if (cur_xino) { + cur_parent = dget_parent(cur_xino->f_dentry); + cur_name = &cur_xino->f_dentry->d_name; + skip = (cur_parent == parent + && au_qstreq(dname, cur_name)); + dput(cur_parent); + } + if (skip) + goto out; + } + + au_opt_set(sbinfo->si_mntflags, XINO); + dir = parent->d_inode; + mutex_lock_nested(&dir->i_mutex, AuLsc_I_PARENT); + /* mnt_want_write() is unnecessary here */ + err = au_xino_set_xib(sb, xino->file); + if (!err) + err = au_xigen_set(sb, xino->file); + if (!err) + err = au_xino_set_br(sb, xino->file); + mutex_unlock(&dir->i_mutex); + if (!err) + goto out; /* success */ + + /* reset all */ + AuIOErr("failed creating xino(%d).\n", err); + +out: + dput(parent); + return err; +} + +/* ---------------------------------------------------------------------- */ + +/* + * create a xinofile at the default place/path. + */ +struct file *au_xino_def(struct super_block *sb) +{ + struct file *file; + char *page, *p; + struct au_branch *br; + struct super_block *h_sb; + struct path path; + aufs_bindex_t bend, bindex, bwr; + + br = NULL; + bend = au_sbend(sb); + bwr = -1; + for (bindex = 0; bindex <= bend; bindex++) { + br = au_sbr(sb, bindex); + if (au_br_writable(br->br_perm) + && !au_test_fs_bad_xino(au_br_sb(br))) { + bwr = bindex; + break; + } + } + + if (bwr >= 0) { + file = ERR_PTR(-ENOMEM); + page = (void *)__get_free_page(GFP_NOFS); + if (unlikely(!page)) + goto out; + path.mnt = au_br_mnt(br); + path.dentry = au_h_dptr(sb->s_root, bwr); + p = d_path(&path, page, PATH_MAX - sizeof(AUFS_XINO_FNAME)); + file = (void *)p; + if (!IS_ERR(p)) { + strcat(p, "/" AUFS_XINO_FNAME); + AuDbg("%s\n", p); + file = au_xino_create(sb, p, /*silent*/0); + if (!IS_ERR(file)) + au_xino_brid_set(sb, br->br_id); + } + free_page((unsigned long)page); + } else { + file = au_xino_create(sb, AUFS_XINO_DEFPATH, /*silent*/0); + if (IS_ERR(file)) + goto out; + h_sb = file->f_dentry->d_sb; + if (unlikely(au_test_fs_bad_xino(h_sb))) { + pr_err("xino doesn't support %s(%s)\n", + AUFS_XINO_DEFPATH, au_sbtype(h_sb)); + fput(file); + file = ERR_PTR(-EINVAL); + } + if (!IS_ERR(file)) + au_xino_brid_set(sb, -1); + } + +out: + return file; +} + +/* ---------------------------------------------------------------------- */ + +int au_xino_path(struct seq_file *seq, struct file *file) +{ + int err; + + err = au_seq_path(seq, &file->f_path); + if (unlikely(err < 0)) + goto out; + + err = 0; +#define Deleted "\\040(deleted)" + seq->count -= sizeof(Deleted) - 1; + AuDebugOn(memcmp(seq->buf + seq->count, Deleted, + sizeof(Deleted) - 1)); +#undef Deleted + +out: + return err; +} diff -Naur kernel.21.3.orig/fs/buffer.c kernel.21.3.new/fs/buffer.c --- kernel.21.3.orig/fs/buffer.c 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/fs/buffer.c 2015-04-25 23:28:45.284481882 +0000 @@ -2389,7 +2389,7 @@ * Update file times before taking page lock. We may end up failing the * fault so this update may be superfluous but who really cares... */ - file_update_time(vma->vm_file); + vma_file_update_time(vma); ret = __block_page_mkwrite(vma, vmf, get_block); sb_end_pagefault(sb); diff -Naur kernel.21.3.orig/fs/buffer.c.orig kernel.21.3.new/fs/buffer.c.orig --- kernel.21.3.orig/fs/buffer.c.orig 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/buffer.c.orig 2015-02-03 01:44:32.000000000 +0000 @@ -0,0 +1,3381 @@ +/* + * linux/fs/buffer.c + * + * Copyright (C) 1991, 1992, 2002 Linus Torvalds + */ + +/* + * Start bdflush() with kernel_thread not syscall - Paul Gortmaker, 12/95 + * + * Removed a lot of unnecessary code and simplified things now that + * the buffer cache isn't our primary cache - Andrew Tridgell 12/96 + * + * Speed up hash, lru, and free list operations. Use gfp() for allocating + * hash table, use SLAB cache for buffer heads. SMP threading. -DaveM + * + * Added 32k buffer block sizes - these are required older ARM systems. - RMK + * + * async buffer flushing, 1999 Andrea Arcangeli + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +static int fsync_buffers_list(spinlock_t *lock, struct list_head *list); + +#define BH_ENTRY(list) list_entry((list), struct buffer_head, b_assoc_buffers) + +void init_buffer(struct buffer_head *bh, bh_end_io_t *handler, void *private) +{ + bh->b_end_io = handler; + bh->b_private = private; +} +EXPORT_SYMBOL(init_buffer); + +inline void touch_buffer(struct buffer_head *bh) +{ + trace_block_touch_buffer(bh); + mark_page_accessed(bh->b_page); +} +EXPORT_SYMBOL(touch_buffer); + +static int sleep_on_buffer(void *word) +{ + io_schedule(); + return 0; +} + +void __lock_buffer(struct buffer_head *bh) +{ + wait_on_bit_lock(&bh->b_state, BH_Lock, sleep_on_buffer, + TASK_UNINTERRUPTIBLE); +} +EXPORT_SYMBOL(__lock_buffer); + +void unlock_buffer(struct buffer_head *bh) +{ + clear_bit_unlock(BH_Lock, &bh->b_state); + smp_mb__after_clear_bit(); + wake_up_bit(&bh->b_state, BH_Lock); +} +EXPORT_SYMBOL(unlock_buffer); + +/* + * Block until a buffer comes unlocked. This doesn't stop it + * from becoming locked again - you have to lock it yourself + * if you want to preserve its state. + */ +void __wait_on_buffer(struct buffer_head * bh) +{ + wait_on_bit(&bh->b_state, BH_Lock, sleep_on_buffer, TASK_UNINTERRUPTIBLE); +} +EXPORT_SYMBOL(__wait_on_buffer); + +static void +__clear_page_buffers(struct page *page) +{ + ClearPagePrivate(page); + set_page_private(page, 0); + page_cache_release(page); +} + + +static int quiet_error(struct buffer_head *bh) +{ + if (!test_bit(BH_Quiet, &bh->b_state) && printk_ratelimit()) + return 0; + return 1; +} + + +static void buffer_io_error(struct buffer_head *bh) +{ + char b[BDEVNAME_SIZE]; + printk(KERN_ERR "Buffer I/O error on device %s, logical block %Lu\n", + bdevname(bh->b_bdev, b), + (unsigned long long)bh->b_blocknr); +} + +/* + * End-of-IO handler helper function which does not touch the bh after + * unlocking it. + * Note: unlock_buffer() sort-of does touch the bh after unlocking it, but + * a race there is benign: unlock_buffer() only use the bh's address for + * hashing after unlocking the buffer, so it doesn't actually touch the bh + * itself. + */ +static void __end_buffer_read_notouch(struct buffer_head *bh, int uptodate) +{ + if (uptodate) { + set_buffer_uptodate(bh); + } else { + /* This happens, due to failed READA attempts. */ + clear_buffer_uptodate(bh); + } + unlock_buffer(bh); +} + +/* + * Default synchronous end-of-IO handler.. Just mark it up-to-date and + * unlock the buffer. This is what ll_rw_block uses too. + */ +void end_buffer_read_sync(struct buffer_head *bh, int uptodate) +{ + __end_buffer_read_notouch(bh, uptodate); + put_bh(bh); +} +EXPORT_SYMBOL(end_buffer_read_sync); + +void end_buffer_write_sync(struct buffer_head *bh, int uptodate) +{ + char b[BDEVNAME_SIZE]; + + if (uptodate) { + set_buffer_uptodate(bh); + } else { + if (!quiet_error(bh)) { + buffer_io_error(bh); + printk(KERN_WARNING "lost page write due to " + "I/O error on %s\n", + bdevname(bh->b_bdev, b)); + } + set_buffer_write_io_error(bh); + clear_buffer_uptodate(bh); + } + unlock_buffer(bh); + put_bh(bh); +} +EXPORT_SYMBOL(end_buffer_write_sync); + +/* + * Various filesystems appear to want __find_get_block to be non-blocking. + * But it's the page lock which protects the buffers. To get around this, + * we get exclusion from try_to_free_buffers with the blockdev mapping's + * private_lock. + * + * Hack idea: for the blockdev mapping, i_bufferlist_lock contention + * may be quite high. This code could TryLock the page, and if that + * succeeds, there is no need to take private_lock. (But if + * private_lock is contended then so is mapping->tree_lock). + */ +static struct buffer_head * +__find_get_block_slow(struct block_device *bdev, sector_t block) +{ + struct inode *bd_inode = bdev->bd_inode; + struct address_space *bd_mapping = bd_inode->i_mapping; + struct buffer_head *ret = NULL; + pgoff_t index; + struct buffer_head *bh; + struct buffer_head *head; + struct page *page; + int all_mapped = 1; + + index = block >> (PAGE_CACHE_SHIFT - bd_inode->i_blkbits); + page = find_get_page(bd_mapping, index); + if (!page) + goto out; + + spin_lock(&bd_mapping->private_lock); + if (!page_has_buffers(page)) + goto out_unlock; + head = page_buffers(page); + bh = head; + do { + if (!buffer_mapped(bh)) + all_mapped = 0; + else if (bh->b_blocknr == block) { + ret = bh; + get_bh(bh); + goto out_unlock; + } + bh = bh->b_this_page; + } while (bh != head); + + /* we might be here because some of the buffers on this page are + * not mapped. This is due to various races between + * file io on the block device and getblk. It gets dealt with + * elsewhere, don't buffer_error if we had some unmapped buffers + */ + if (all_mapped) { + char b[BDEVNAME_SIZE]; + + printk("__find_get_block_slow() failed. " + "block=%llu, b_blocknr=%llu\n", + (unsigned long long)block, + (unsigned long long)bh->b_blocknr); + printk("b_state=0x%08lx, b_size=%zu\n", + bh->b_state, bh->b_size); + printk("device %s blocksize: %d\n", bdevname(bdev, b), + 1 << bd_inode->i_blkbits); + } +out_unlock: + spin_unlock(&bd_mapping->private_lock); + page_cache_release(page); +out: + return ret; +} + +/* + * Kick the writeback threads then try to free up some ZONE_NORMAL memory. + */ +static void free_more_memory(void) +{ + struct zone *zone; + int nid; + + wakeup_flusher_threads(1024, WB_REASON_FREE_MORE_MEM); + yield(); + + for_each_online_node(nid) { + (void)first_zones_zonelist(node_zonelist(nid, GFP_NOFS), + gfp_zone(GFP_NOFS), NULL, + &zone); + if (zone) + try_to_free_pages(node_zonelist(nid, GFP_NOFS), 0, + GFP_NOFS, NULL); + } +} + +/* + * I/O completion handler for block_read_full_page() - pages + * which come unlocked at the end of I/O. + */ +static void end_buffer_async_read(struct buffer_head *bh, int uptodate) +{ + unsigned long flags; + struct buffer_head *first; + struct buffer_head *tmp; + struct page *page; + int page_uptodate = 1; + + BUG_ON(!buffer_async_read(bh)); + + page = bh->b_page; + if (uptodate) { + set_buffer_uptodate(bh); + } else { + clear_buffer_uptodate(bh); + if (!quiet_error(bh)) + buffer_io_error(bh); + SetPageError(page); + } + + /* + * Be _very_ careful from here on. Bad things can happen if + * two buffer heads end IO at almost the same time and both + * decide that the page is now completely done. + */ + first = page_buffers(page); + local_irq_save(flags); + bit_spin_lock(BH_Uptodate_Lock, &first->b_state); + clear_buffer_async_read(bh); + unlock_buffer(bh); + tmp = bh; + do { + if (!buffer_uptodate(tmp)) + page_uptodate = 0; + if (buffer_async_read(tmp)) { + BUG_ON(!buffer_locked(tmp)); + goto still_busy; + } + tmp = tmp->b_this_page; + } while (tmp != bh); + bit_spin_unlock(BH_Uptodate_Lock, &first->b_state); + local_irq_restore(flags); + + /* + * If none of the buffers had errors and they are all + * uptodate then we can set the page uptodate. + */ + if (page_uptodate && !PageError(page)) + SetPageUptodate(page); + unlock_page(page); + return; + +still_busy: + bit_spin_unlock(BH_Uptodate_Lock, &first->b_state); + local_irq_restore(flags); + return; +} + +/* + * Completion handler for block_write_full_page() - pages which are unlocked + * during I/O, and which have PageWriteback cleared upon I/O completion. + */ +void end_buffer_async_write(struct buffer_head *bh, int uptodate) +{ + char b[BDEVNAME_SIZE]; + unsigned long flags; + struct buffer_head *first; + struct buffer_head *tmp; + struct page *page; + + BUG_ON(!buffer_async_write(bh)); + + page = bh->b_page; + if (uptodate) { + set_buffer_uptodate(bh); + } else { + if (!quiet_error(bh)) { + buffer_io_error(bh); + printk(KERN_WARNING "lost page write due to " + "I/O error on %s\n", + bdevname(bh->b_bdev, b)); + } + set_bit(AS_EIO, &page->mapping->flags); + set_buffer_write_io_error(bh); + clear_buffer_uptodate(bh); + SetPageError(page); + } + + first = page_buffers(page); + local_irq_save(flags); + bit_spin_lock(BH_Uptodate_Lock, &first->b_state); + + clear_buffer_async_write(bh); + unlock_buffer(bh); + tmp = bh->b_this_page; + while (tmp != bh) { + if (buffer_async_write(tmp)) { + BUG_ON(!buffer_locked(tmp)); + goto still_busy; + } + tmp = tmp->b_this_page; + } + bit_spin_unlock(BH_Uptodate_Lock, &first->b_state); + local_irq_restore(flags); + end_page_writeback(page); + return; + +still_busy: + bit_spin_unlock(BH_Uptodate_Lock, &first->b_state); + local_irq_restore(flags); + return; +} +EXPORT_SYMBOL(end_buffer_async_write); + +/* + * If a page's buffers are under async readin (end_buffer_async_read + * completion) then there is a possibility that another thread of + * control could lock one of the buffers after it has completed + * but while some of the other buffers have not completed. This + * locked buffer would confuse end_buffer_async_read() into not unlocking + * the page. So the absence of BH_Async_Read tells end_buffer_async_read() + * that this buffer is not under async I/O. + * + * The page comes unlocked when it has no locked buffer_async buffers + * left. + * + * PageLocked prevents anyone starting new async I/O reads any of + * the buffers. + * + * PageWriteback is used to prevent simultaneous writeout of the same + * page. + * + * PageLocked prevents anyone from starting writeback of a page which is + * under read I/O (PageWriteback is only ever set against a locked page). + */ +static void mark_buffer_async_read(struct buffer_head *bh) +{ + bh->b_end_io = end_buffer_async_read; + set_buffer_async_read(bh); +} + +static void mark_buffer_async_write_endio(struct buffer_head *bh, + bh_end_io_t *handler) +{ + bh->b_end_io = handler; + set_buffer_async_write(bh); +} + +void mark_buffer_async_write(struct buffer_head *bh) +{ + mark_buffer_async_write_endio(bh, end_buffer_async_write); +} +EXPORT_SYMBOL(mark_buffer_async_write); + + +/* + * fs/buffer.c contains helper functions for buffer-backed address space's + * fsync functions. A common requirement for buffer-based filesystems is + * that certain data from the backing blockdev needs to be written out for + * a successful fsync(). For example, ext2 indirect blocks need to be + * written back and waited upon before fsync() returns. + * + * The functions mark_buffer_inode_dirty(), fsync_inode_buffers(), + * inode_has_buffers() and invalidate_inode_buffers() are provided for the + * management of a list of dependent buffers at ->i_mapping->private_list. + * + * Locking is a little subtle: try_to_free_buffers() will remove buffers + * from their controlling inode's queue when they are being freed. But + * try_to_free_buffers() will be operating against the *blockdev* mapping + * at the time, not against the S_ISREG file which depends on those buffers. + * So the locking for private_list is via the private_lock in the address_space + * which backs the buffers. Which is different from the address_space + * against which the buffers are listed. So for a particular address_space, + * mapping->private_lock does *not* protect mapping->private_list! In fact, + * mapping->private_list will always be protected by the backing blockdev's + * ->private_lock. + * + * Which introduces a requirement: all buffers on an address_space's + * ->private_list must be from the same address_space: the blockdev's. + * + * address_spaces which do not place buffers at ->private_list via these + * utility functions are free to use private_lock and private_list for + * whatever they want. The only requirement is that list_empty(private_list) + * be true at clear_inode() time. + * + * FIXME: clear_inode should not call invalidate_inode_buffers(). The + * filesystems should do that. invalidate_inode_buffers() should just go + * BUG_ON(!list_empty). + * + * FIXME: mark_buffer_dirty_inode() is a data-plane operation. It should + * take an address_space, not an inode. And it should be called + * mark_buffer_dirty_fsync() to clearly define why those buffers are being + * queued up. + * + * FIXME: mark_buffer_dirty_inode() doesn't need to add the buffer to the + * list if it is already on a list. Because if the buffer is on a list, + * it *must* already be on the right one. If not, the filesystem is being + * silly. This will save a ton of locking. But first we have to ensure + * that buffers are taken *off* the old inode's list when they are freed + * (presumably in truncate). That requires careful auditing of all + * filesystems (do it inside bforget()). It could also be done by bringing + * b_inode back. + */ + +/* + * The buffer's backing address_space's private_lock must be held + */ +static void __remove_assoc_queue(struct buffer_head *bh) +{ + list_del_init(&bh->b_assoc_buffers); + WARN_ON(!bh->b_assoc_map); + if (buffer_write_io_error(bh)) + set_bit(AS_EIO, &bh->b_assoc_map->flags); + bh->b_assoc_map = NULL; +} + +int inode_has_buffers(struct inode *inode) +{ + return !list_empty(&inode->i_data.private_list); +} + +/* + * osync is designed to support O_SYNC io. It waits synchronously for + * all already-submitted IO to complete, but does not queue any new + * writes to the disk. + * + * To do O_SYNC writes, just queue the buffer writes with ll_rw_block as + * you dirty the buffers, and then use osync_inode_buffers to wait for + * completion. Any other dirty buffers which are not yet queued for + * write will not be flushed to disk by the osync. + */ +static int osync_buffers_list(spinlock_t *lock, struct list_head *list) +{ + struct buffer_head *bh; + struct list_head *p; + int err = 0; + + spin_lock(lock); +repeat: + list_for_each_prev(p, list) { + bh = BH_ENTRY(p); + if (buffer_locked(bh)) { + get_bh(bh); + spin_unlock(lock); + wait_on_buffer(bh); + if (!buffer_uptodate(bh)) + err = -EIO; + brelse(bh); + spin_lock(lock); + goto repeat; + } + } + spin_unlock(lock); + return err; +} + +static void do_thaw_one(struct super_block *sb, void *unused) +{ + char b[BDEVNAME_SIZE]; + while (sb->s_bdev && !thaw_bdev(sb->s_bdev, sb)) + printk(KERN_WARNING "Emergency Thaw on %s\n", + bdevname(sb->s_bdev, b)); +} + +static void do_thaw_all(struct work_struct *work) +{ + iterate_supers(do_thaw_one, NULL); + kfree(work); + printk(KERN_WARNING "Emergency Thaw complete\n"); +} + +/** + * emergency_thaw_all -- forcibly thaw every frozen filesystem + * + * Used for emergency unfreeze of all filesystems via SysRq + */ +void emergency_thaw_all(void) +{ + struct work_struct *work; + + work = kmalloc(sizeof(*work), GFP_ATOMIC); + if (work) { + INIT_WORK(work, do_thaw_all); + schedule_work(work); + } +} + +/** + * sync_mapping_buffers - write out & wait upon a mapping's "associated" buffers + * @mapping: the mapping which wants those buffers written + * + * Starts I/O against the buffers at mapping->private_list, and waits upon + * that I/O. + * + * Basically, this is a convenience function for fsync(). + * @mapping is a file or directory which needs those buffers to be written for + * a successful fsync(). + */ +int sync_mapping_buffers(struct address_space *mapping) +{ + struct address_space *buffer_mapping = mapping->private_data; + + if (buffer_mapping == NULL || list_empty(&mapping->private_list)) + return 0; + + return fsync_buffers_list(&buffer_mapping->private_lock, + &mapping->private_list); +} +EXPORT_SYMBOL(sync_mapping_buffers); + +/* + * Called when we've recently written block `bblock', and it is known that + * `bblock' was for a buffer_boundary() buffer. This means that the block at + * `bblock + 1' is probably a dirty indirect block. Hunt it down and, if it's + * dirty, schedule it for IO. So that indirects merge nicely with their data. + */ +void write_boundary_block(struct block_device *bdev, + sector_t bblock, unsigned blocksize) +{ + struct buffer_head *bh = __find_get_block(bdev, bblock + 1, blocksize); + if (bh) { + if (buffer_dirty(bh)) + ll_rw_block(WRITE, 1, &bh); + put_bh(bh); + } +} + +void mark_buffer_dirty_inode(struct buffer_head *bh, struct inode *inode) +{ + struct address_space *mapping = inode->i_mapping; + struct address_space *buffer_mapping = bh->b_page->mapping; + + mark_buffer_dirty(bh); + if (!mapping->private_data) { + mapping->private_data = buffer_mapping; + } else { + BUG_ON(mapping->private_data != buffer_mapping); + } + if (!bh->b_assoc_map) { + spin_lock(&buffer_mapping->private_lock); + list_move_tail(&bh->b_assoc_buffers, + &mapping->private_list); + bh->b_assoc_map = mapping; + spin_unlock(&buffer_mapping->private_lock); + } +} +EXPORT_SYMBOL(mark_buffer_dirty_inode); + +/* + * Mark the page dirty, and set it dirty in the radix tree, and mark the inode + * dirty. + * + * If warn is true, then emit a warning if the page is not uptodate and has + * not been truncated. + */ +static void __set_page_dirty(struct page *page, + struct address_space *mapping, int warn) +{ + unsigned long flags; + + spin_lock_irqsave(&mapping->tree_lock, flags); + if (page->mapping) { /* Race with truncate? */ + WARN_ON_ONCE(warn && !PageUptodate(page)); + account_page_dirtied(page, mapping); + radix_tree_tag_set(&mapping->page_tree, + page_index(page), PAGECACHE_TAG_DIRTY); + } + spin_unlock_irqrestore(&mapping->tree_lock, flags); + __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); +} + +/* + * Add a page to the dirty page list. + * + * It is a sad fact of life that this function is called from several places + * deeply under spinlocking. It may not sleep. + * + * If the page has buffers, the uptodate buffers are set dirty, to preserve + * dirty-state coherency between the page and the buffers. It the page does + * not have buffers then when they are later attached they will all be set + * dirty. + * + * The buffers are dirtied before the page is dirtied. There's a small race + * window in which a writepage caller may see the page cleanness but not the + * buffer dirtiness. That's fine. If this code were to set the page dirty + * before the buffers, a concurrent writepage caller could clear the page dirty + * bit, see a bunch of clean buffers and we'd end up with dirty buffers/clean + * page on the dirty page list. + * + * We use private_lock to lock against try_to_free_buffers while using the + * page's buffer list. Also use this to protect against clean buffers being + * added to the page after it was set dirty. + * + * FIXME: may need to call ->reservepage here as well. That's rather up to the + * address_space though. + */ +int __set_page_dirty_buffers(struct page *page) +{ + int newly_dirty; + struct address_space *mapping = page_mapping(page); + + if (unlikely(!mapping)) + return !TestSetPageDirty(page); + + spin_lock(&mapping->private_lock); + if (page_has_buffers(page)) { + struct buffer_head *head = page_buffers(page); + struct buffer_head *bh = head; + + do { + set_buffer_dirty(bh); + bh = bh->b_this_page; + } while (bh != head); + } + newly_dirty = !TestSetPageDirty(page); + spin_unlock(&mapping->private_lock); + + if (newly_dirty) + __set_page_dirty(page, mapping, 1); + return newly_dirty; +} +EXPORT_SYMBOL(__set_page_dirty_buffers); + +/* + * Write out and wait upon a list of buffers. + * + * We have conflicting pressures: we want to make sure that all + * initially dirty buffers get waited on, but that any subsequently + * dirtied buffers don't. After all, we don't want fsync to last + * forever if somebody is actively writing to the file. + * + * Do this in two main stages: first we copy dirty buffers to a + * temporary inode list, queueing the writes as we go. Then we clean + * up, waiting for those writes to complete. + * + * During this second stage, any subsequent updates to the file may end + * up refiling the buffer on the original inode's dirty list again, so + * there is a chance we will end up with a buffer queued for write but + * not yet completed on that list. So, as a final cleanup we go through + * the osync code to catch these locked, dirty buffers without requeuing + * any newly dirty buffers for write. + */ +static int fsync_buffers_list(spinlock_t *lock, struct list_head *list) +{ + struct buffer_head *bh; + struct list_head tmp; + struct address_space *mapping; + int err = 0, err2; + struct blk_plug plug; + + INIT_LIST_HEAD(&tmp); + blk_start_plug(&plug); + + spin_lock(lock); + while (!list_empty(list)) { + bh = BH_ENTRY(list->next); + mapping = bh->b_assoc_map; + __remove_assoc_queue(bh); + /* Avoid race with mark_buffer_dirty_inode() which does + * a lockless check and we rely on seeing the dirty bit */ + smp_mb(); + if (buffer_dirty(bh) || buffer_locked(bh)) { + list_add(&bh->b_assoc_buffers, &tmp); + bh->b_assoc_map = mapping; + if (buffer_dirty(bh)) { + get_bh(bh); + spin_unlock(lock); + /* + * Ensure any pending I/O completes so that + * write_dirty_buffer() actually writes the + * current contents - it is a noop if I/O is + * still in flight on potentially older + * contents. + */ + write_dirty_buffer(bh, WRITE_SYNC); + + /* + * Kick off IO for the previous mapping. Note + * that we will not run the very last mapping, + * wait_on_buffer() will do that for us + * through sync_buffer(). + */ + brelse(bh); + spin_lock(lock); + } + } + } + + spin_unlock(lock); + blk_finish_plug(&plug); + spin_lock(lock); + + while (!list_empty(&tmp)) { + bh = BH_ENTRY(tmp.prev); + get_bh(bh); + mapping = bh->b_assoc_map; + __remove_assoc_queue(bh); + /* Avoid race with mark_buffer_dirty_inode() which does + * a lockless check and we rely on seeing the dirty bit */ + smp_mb(); + if (buffer_dirty(bh)) { + list_add(&bh->b_assoc_buffers, + &mapping->private_list); + bh->b_assoc_map = mapping; + } + spin_unlock(lock); + wait_on_buffer(bh); + if (!buffer_uptodate(bh)) + err = -EIO; + brelse(bh); + spin_lock(lock); + } + + spin_unlock(lock); + err2 = osync_buffers_list(lock, list); + if (err) + return err; + else + return err2; +} + +/* + * Invalidate any and all dirty buffers on a given inode. We are + * probably unmounting the fs, but that doesn't mean we have already + * done a sync(). Just drop the buffers from the inode list. + * + * NOTE: we take the inode's blockdev's mapping's private_lock. Which + * assumes that all the buffers are against the blockdev. Not true + * for reiserfs. + */ +void invalidate_inode_buffers(struct inode *inode) +{ + if (inode_has_buffers(inode)) { + struct address_space *mapping = &inode->i_data; + struct list_head *list = &mapping->private_list; + struct address_space *buffer_mapping = mapping->private_data; + + spin_lock(&buffer_mapping->private_lock); + while (!list_empty(list)) + __remove_assoc_queue(BH_ENTRY(list->next)); + spin_unlock(&buffer_mapping->private_lock); + } +} +EXPORT_SYMBOL(invalidate_inode_buffers); + +/* + * Remove any clean buffers from the inode's buffer list. This is called + * when we're trying to free the inode itself. Those buffers can pin it. + * + * Returns true if all buffers were removed. + */ +int remove_inode_buffers(struct inode *inode) +{ + int ret = 1; + + if (inode_has_buffers(inode)) { + struct address_space *mapping = &inode->i_data; + struct list_head *list = &mapping->private_list; + struct address_space *buffer_mapping = mapping->private_data; + + spin_lock(&buffer_mapping->private_lock); + while (!list_empty(list)) { + struct buffer_head *bh = BH_ENTRY(list->next); + if (buffer_dirty(bh)) { + ret = 0; + break; + } + __remove_assoc_queue(bh); + } + spin_unlock(&buffer_mapping->private_lock); + } + return ret; +} + +/* + * Create the appropriate buffers when given a page for data area and + * the size of each buffer.. Use the bh->b_this_page linked list to + * follow the buffers created. Return NULL if unable to create more + * buffers. + * + * The retry flag is used to differentiate async IO (paging, swapping) + * which may not fail from ordinary buffer allocations. + */ +struct buffer_head *alloc_page_buffers(struct page *page, unsigned long size, + int retry) +{ + struct buffer_head *bh, *head; + long offset; + +try_again: + head = NULL; + offset = PAGE_SIZE; + while ((offset -= size) >= 0) { + bh = alloc_buffer_head(GFP_NOFS); + if (!bh) + goto no_grow; + + bh->b_this_page = head; + bh->b_blocknr = -1; + head = bh; + + bh->b_size = size; + + /* Link the buffer to its page */ + set_bh_page(bh, page, offset); + } + return head; +/* + * In case anything failed, we just free everything we got. + */ +no_grow: + if (head) { + do { + bh = head; + head = head->b_this_page; + free_buffer_head(bh); + } while (head); + } + + /* + * Return failure for non-async IO requests. Async IO requests + * are not allowed to fail, so we have to wait until buffer heads + * become available. But we don't want tasks sleeping with + * partially complete buffers, so all were released above. + */ + if (!retry) + return NULL; + + /* We're _really_ low on memory. Now we just + * wait for old buffer heads to become free due to + * finishing IO. Since this is an async request and + * the reserve list is empty, we're sure there are + * async buffer heads in use. + */ + free_more_memory(); + goto try_again; +} +EXPORT_SYMBOL_GPL(alloc_page_buffers); + +static inline void +link_dev_buffers(struct page *page, struct buffer_head *head) +{ + struct buffer_head *bh, *tail; + + bh = head; + do { + tail = bh; + bh = bh->b_this_page; + } while (bh); + tail->b_this_page = head; + attach_page_buffers(page, head); +} + +static sector_t blkdev_max_block(struct block_device *bdev, unsigned int size) +{ + sector_t retval = ~((sector_t)0); + loff_t sz = i_size_read(bdev->bd_inode); + + if (sz) { + unsigned int sizebits = blksize_bits(size); + retval = (sz >> sizebits); + } + return retval; +} + +/* + * Initialise the state of a blockdev page's buffers. + */ +static sector_t +init_page_buffers(struct page *page, struct block_device *bdev, + sector_t block, int size) +{ + struct buffer_head *head = page_buffers(page); + struct buffer_head *bh = head; + int uptodate = PageUptodate(page); + sector_t end_block = blkdev_max_block(I_BDEV(bdev->bd_inode), size); + + do { + if (!buffer_mapped(bh)) { + init_buffer(bh, NULL, NULL); + bh->b_bdev = bdev; + bh->b_blocknr = block; + if (uptodate) + set_buffer_uptodate(bh); + if (block < end_block) + set_buffer_mapped(bh); + } + block++; + bh = bh->b_this_page; + } while (bh != head); + + /* + * Caller needs to validate requested block against end of device. + */ + return end_block; +} + +/* + * Create the page-cache page that contains the requested block. + * + * This is used purely for blockdev mappings. + */ +static int +grow_dev_page(struct block_device *bdev, sector_t block, + pgoff_t index, int size, int sizebits) +{ + struct inode *inode = bdev->bd_inode; + struct page *page; + struct buffer_head *bh; + sector_t end_block; + int ret = 0; /* Will call free_more_memory() */ + + page = find_or_create_page(inode->i_mapping, index, + (mapping_gfp_mask(inode->i_mapping) & ~__GFP_FS)); + if (!page) + return ret; + + BUG_ON(!PageLocked(page)); + + if (page_has_buffers(page)) { + bh = page_buffers(page); + if (bh->b_size == size) { + end_block = init_page_buffers(page, bdev, + index << sizebits, size); + goto done; + } + if (!try_to_free_buffers(page)) + goto failed; + } + + /* + * Allocate some buffers for this page + */ + bh = alloc_page_buffers(page, size, 0); + if (!bh) + goto failed; + + /* + * Link the page to the buffers and initialise them. Take the + * lock to be atomic wrt __find_get_block(), which does not + * run under the page lock. + */ + spin_lock(&inode->i_mapping->private_lock); + link_dev_buffers(page, bh); + end_block = init_page_buffers(page, bdev, index << sizebits, size); + spin_unlock(&inode->i_mapping->private_lock); +done: + ret = (block < end_block) ? 1 : -ENXIO; +failed: + unlock_page(page); + page_cache_release(page); + return ret; +} + +/* + * Create buffers for the specified block device block's page. If + * that page was dirty, the buffers are set dirty also. + */ +static int +grow_buffers(struct block_device *bdev, sector_t block, int size) +{ + pgoff_t index; + int sizebits; + + sizebits = -1; + do { + sizebits++; + } while ((size << sizebits) < PAGE_SIZE); + + index = block >> sizebits; + + /* + * Check for a block which wants to lie outside our maximum possible + * pagecache index. (this comparison is done using sector_t types). + */ + if (unlikely(index != block >> sizebits)) { + char b[BDEVNAME_SIZE]; + + printk(KERN_ERR "%s: requested out-of-range block %llu for " + "device %s\n", + __func__, (unsigned long long)block, + bdevname(bdev, b)); + return -EIO; + } + + /* Create a page with the proper size buffers.. */ + return grow_dev_page(bdev, block, index, size, sizebits); +} + +static struct buffer_head * +__getblk_slow(struct block_device *bdev, sector_t block, int size) +{ + /* Size must be multiple of hard sectorsize */ + if (unlikely(size & (bdev_logical_block_size(bdev)-1) || + (size < 512 || size > PAGE_SIZE))) { + printk(KERN_ERR "getblk(): invalid block size %d requested\n", + size); + printk(KERN_ERR "logical block size: %d\n", + bdev_logical_block_size(bdev)); + + dump_stack(); + return NULL; + } + + for (;;) { + struct buffer_head *bh; + int ret; + + bh = __find_get_block(bdev, block, size); + if (bh) + return bh; + + ret = grow_buffers(bdev, block, size); + if (ret < 0) + return NULL; + if (ret == 0) + free_more_memory(); + } +} + +/* + * The relationship between dirty buffers and dirty pages: + * + * Whenever a page has any dirty buffers, the page's dirty bit is set, and + * the page is tagged dirty in its radix tree. + * + * At all times, the dirtiness of the buffers represents the dirtiness of + * subsections of the page. If the page has buffers, the page dirty bit is + * merely a hint about the true dirty state. + * + * When a page is set dirty in its entirety, all its buffers are marked dirty + * (if the page has buffers). + * + * When a buffer is marked dirty, its page is dirtied, but the page's other + * buffers are not. + * + * Also. When blockdev buffers are explicitly read with bread(), they + * individually become uptodate. But their backing page remains not + * uptodate - even if all of its buffers are uptodate. A subsequent + * block_read_full_page() against that page will discover all the uptodate + * buffers, will set the page uptodate and will perform no I/O. + */ + +/** + * mark_buffer_dirty - mark a buffer_head as needing writeout + * @bh: the buffer_head to mark dirty + * + * mark_buffer_dirty() will set the dirty bit against the buffer, then set its + * backing page dirty, then tag the page as dirty in its address_space's radix + * tree and then attach the address_space's inode to its superblock's dirty + * inode list. + * + * mark_buffer_dirty() is atomic. It takes bh->b_page->mapping->private_lock, + * mapping->tree_lock and mapping->host->i_lock. + */ +void mark_buffer_dirty(struct buffer_head *bh) +{ + WARN_ON_ONCE(!buffer_uptodate(bh)); + + trace_block_dirty_buffer(bh); + + /* + * Very *carefully* optimize the it-is-already-dirty case. + * + * Don't let the final "is it dirty" escape to before we + * perhaps modified the buffer. + */ + if (buffer_dirty(bh)) { + smp_mb(); + if (buffer_dirty(bh)) + return; + } + + if (!test_set_buffer_dirty(bh)) { + struct page *page = bh->b_page; + if (!TestSetPageDirty(page)) { + struct address_space *mapping = page_mapping(page); + if (mapping) + __set_page_dirty(page, mapping, 0); + } + } +} +EXPORT_SYMBOL(mark_buffer_dirty); + +/* + * Decrement a buffer_head's reference count. If all buffers against a page + * have zero reference count, are clean and unlocked, and if the page is clean + * and unlocked then try_to_free_buffers() may strip the buffers from the page + * in preparation for freeing it (sometimes, rarely, buffers are removed from + * a page but it ends up not being freed, and buffers may later be reattached). + */ +void __brelse(struct buffer_head * buf) +{ + if (atomic_read(&buf->b_count)) { + put_bh(buf); + return; + } + WARN(1, KERN_ERR "VFS: brelse: Trying to free free buffer\n"); +} +EXPORT_SYMBOL(__brelse); + +/* + * bforget() is like brelse(), except it discards any + * potentially dirty data. + */ +void __bforget(struct buffer_head *bh) +{ + clear_buffer_dirty(bh); + if (bh->b_assoc_map) { + struct address_space *buffer_mapping = bh->b_page->mapping; + + spin_lock(&buffer_mapping->private_lock); + list_del_init(&bh->b_assoc_buffers); + bh->b_assoc_map = NULL; + spin_unlock(&buffer_mapping->private_lock); + } + __brelse(bh); +} +EXPORT_SYMBOL(__bforget); + +static struct buffer_head *__bread_slow(struct buffer_head *bh) +{ + lock_buffer(bh); + if (buffer_uptodate(bh)) { + unlock_buffer(bh); + return bh; + } else { + get_bh(bh); + bh->b_end_io = end_buffer_read_sync; + submit_bh(READ, bh); + wait_on_buffer(bh); + if (buffer_uptodate(bh)) + return bh; + } + brelse(bh); + return NULL; +} + +/* + * Per-cpu buffer LRU implementation. To reduce the cost of __find_get_block(). + * The bhs[] array is sorted - newest buffer is at bhs[0]. Buffers have their + * refcount elevated by one when they're in an LRU. A buffer can only appear + * once in a particular CPU's LRU. A single buffer can be present in multiple + * CPU's LRUs at the same time. + * + * This is a transparent caching front-end to sb_bread(), sb_getblk() and + * sb_find_get_block(). + * + * The LRUs themselves only need locking against invalidate_bh_lrus. We use + * a local interrupt disable for that. + */ + +#define BH_LRU_SIZE 8 + +struct bh_lru { + struct buffer_head *bhs[BH_LRU_SIZE]; +}; + +static DEFINE_PER_CPU(struct bh_lru, bh_lrus) = {{ NULL }}; + +#ifdef CONFIG_SMP +#define bh_lru_lock() local_irq_disable() +#define bh_lru_unlock() local_irq_enable() +#else +#define bh_lru_lock() preempt_disable() +#define bh_lru_unlock() preempt_enable() +#endif + +static inline void check_irqs_on(void) +{ +#ifdef irqs_disabled + BUG_ON(irqs_disabled()); +#endif +} + +/* + * The LRU management algorithm is dopey-but-simple. Sorry. + */ +static void bh_lru_install(struct buffer_head *bh) +{ + struct buffer_head *evictee = NULL; + + check_irqs_on(); + bh_lru_lock(); + if (__this_cpu_read(bh_lrus.bhs[0]) != bh) { + struct buffer_head *bhs[BH_LRU_SIZE]; + int in; + int out = 0; + + get_bh(bh); + bhs[out++] = bh; + for (in = 0; in < BH_LRU_SIZE; in++) { + struct buffer_head *bh2 = + __this_cpu_read(bh_lrus.bhs[in]); + + if (bh2 == bh) { + __brelse(bh2); + } else { + if (out >= BH_LRU_SIZE) { + BUG_ON(evictee != NULL); + evictee = bh2; + } else { + bhs[out++] = bh2; + } + } + } + while (out < BH_LRU_SIZE) + bhs[out++] = NULL; + memcpy(__this_cpu_ptr(&bh_lrus.bhs), bhs, sizeof(bhs)); + } + bh_lru_unlock(); + + if (evictee) + __brelse(evictee); +} + +/* + * Look up the bh in this cpu's LRU. If it's there, move it to the head. + */ +static struct buffer_head * +lookup_bh_lru(struct block_device *bdev, sector_t block, unsigned size) +{ + struct buffer_head *ret = NULL; + unsigned int i; + + check_irqs_on(); + bh_lru_lock(); + for (i = 0; i < BH_LRU_SIZE; i++) { + struct buffer_head *bh = __this_cpu_read(bh_lrus.bhs[i]); + + if (bh && bh->b_bdev == bdev && + bh->b_blocknr == block && bh->b_size == size) { + if (i) { + while (i) { + __this_cpu_write(bh_lrus.bhs[i], + __this_cpu_read(bh_lrus.bhs[i - 1])); + i--; + } + __this_cpu_write(bh_lrus.bhs[0], bh); + } + get_bh(bh); + ret = bh; + break; + } + } + bh_lru_unlock(); + return ret; +} + +/* + * Perform a pagecache lookup for the matching buffer. If it's there, refresh + * it in the LRU and mark it as accessed. If it is not present then return + * NULL + */ +struct buffer_head * +__find_get_block(struct block_device *bdev, sector_t block, unsigned size) +{ + struct buffer_head *bh = lookup_bh_lru(bdev, block, size); + + if (bh == NULL) { + bh = __find_get_block_slow(bdev, block); + if (bh) + bh_lru_install(bh); + } + if (bh) + touch_buffer(bh); + return bh; +} +EXPORT_SYMBOL(__find_get_block); + +/* + * __getblk will locate (and, if necessary, create) the buffer_head + * which corresponds to the passed block_device, block and size. The + * returned buffer has its reference count incremented. + * + * __getblk() will lock up the machine if grow_dev_page's try_to_free_buffers() + * attempt is failing. FIXME, perhaps? + */ +struct buffer_head * +__getblk(struct block_device *bdev, sector_t block, unsigned size) +{ + struct buffer_head *bh = __find_get_block(bdev, block, size); + + might_sleep(); + if (bh == NULL) + bh = __getblk_slow(bdev, block, size); + return bh; +} +EXPORT_SYMBOL(__getblk); + +/* + * Do async read-ahead on a buffer.. + */ +void __breadahead(struct block_device *bdev, sector_t block, unsigned size) +{ + struct buffer_head *bh = __getblk(bdev, block, size); + if (likely(bh)) { + ll_rw_block(READA, 1, &bh); + brelse(bh); + } +} +EXPORT_SYMBOL(__breadahead); + +/** + * __bread() - reads a specified block and returns the bh + * @bdev: the block_device to read from + * @block: number of block + * @size: size (in bytes) to read + * + * Reads a specified block, and returns buffer head that contains it. + * It returns NULL if the block was unreadable. + */ +struct buffer_head * +__bread(struct block_device *bdev, sector_t block, unsigned size) +{ + struct buffer_head *bh = __getblk(bdev, block, size); + + if (likely(bh) && !buffer_uptodate(bh)) + bh = __bread_slow(bh); + return bh; +} +EXPORT_SYMBOL(__bread); + +/* + * invalidate_bh_lrus() is called rarely - but not only at unmount. + * This doesn't race because it runs in each cpu either in irq + * or with preempt disabled. + */ +static void invalidate_bh_lru(void *arg) +{ + struct bh_lru *b = &get_cpu_var(bh_lrus); + int i; + + for (i = 0; i < BH_LRU_SIZE; i++) { + brelse(b->bhs[i]); + b->bhs[i] = NULL; + } + put_cpu_var(bh_lrus); +} + +static bool has_bh_in_lru(int cpu, void *dummy) +{ + struct bh_lru *b = per_cpu_ptr(&bh_lrus, cpu); + int i; + + for (i = 0; i < BH_LRU_SIZE; i++) { + if (b->bhs[i]) + return 1; + } + + return 0; +} + +void invalidate_bh_lrus(void) +{ + on_each_cpu_cond(has_bh_in_lru, invalidate_bh_lru, NULL, 1, GFP_KERNEL); +} +EXPORT_SYMBOL_GPL(invalidate_bh_lrus); + +void set_bh_page(struct buffer_head *bh, + struct page *page, unsigned long offset) +{ + bh->b_page = page; + BUG_ON(offset >= PAGE_SIZE); + if (PageHighMem(page)) + /* + * This catches illegal uses and preserves the offset: + */ + bh->b_data = (char *)(0 + offset); + else + bh->b_data = page_address(page) + offset; +} +EXPORT_SYMBOL(set_bh_page); + +/* + * Called when truncating a buffer on a page completely. + */ +static void discard_buffer(struct buffer_head * bh) +{ + lock_buffer(bh); + clear_buffer_dirty(bh); + bh->b_bdev = NULL; + clear_buffer_mapped(bh); + clear_buffer_req(bh); + clear_buffer_new(bh); + clear_buffer_delay(bh); + clear_buffer_unwritten(bh); + unlock_buffer(bh); +} + +/** + * block_invalidatepage - invalidate part or all of a buffer-backed page + * + * @page: the page which is affected + * @offset: the index of the truncation point + * + * block_invalidatepage() is called when all or part of the page has become + * invalidated by a truncate operation. + * + * block_invalidatepage() does not have to release all buffers, but it must + * ensure that no dirty buffer is left outside @offset and that no I/O + * is underway against any of the blocks which are outside the truncation + * point. Because the caller is about to free (and possibly reuse) those + * blocks on-disk. + */ +void block_invalidatepage(struct page *page, unsigned long offset) +{ + struct buffer_head *head, *bh, *next; + unsigned int curr_off = 0; + + BUG_ON(!PageLocked(page)); + if (!page_has_buffers(page)) + goto out; + + head = page_buffers(page); + bh = head; + do { + unsigned int next_off = curr_off + bh->b_size; + next = bh->b_this_page; + + /* + * is this block fully invalidated? + */ + if (offset <= curr_off) + discard_buffer(bh); + curr_off = next_off; + bh = next; + } while (bh != head); + + /* + * We release buffers only if the entire page is being invalidated. + * The get_block cached value has been unconditionally invalidated, + * so real IO is not possible anymore. + */ + if (offset == 0) + try_to_release_page(page, 0); +out: + return; +} +EXPORT_SYMBOL(block_invalidatepage); + +/* + * We attach and possibly dirty the buffers atomically wrt + * __set_page_dirty_buffers() via private_lock. try_to_free_buffers + * is already excluded via the page lock. + */ +void create_empty_buffers(struct page *page, + unsigned long blocksize, unsigned long b_state) +{ + struct buffer_head *bh, *head, *tail; + + head = alloc_page_buffers(page, blocksize, 1); + bh = head; + do { + bh->b_state |= b_state; + tail = bh; + bh = bh->b_this_page; + } while (bh); + tail->b_this_page = head; + + spin_lock(&page->mapping->private_lock); + if (PageUptodate(page) || PageDirty(page)) { + bh = head; + do { + if (PageDirty(page)) + set_buffer_dirty(bh); + if (PageUptodate(page)) + set_buffer_uptodate(bh); + bh = bh->b_this_page; + } while (bh != head); + } + attach_page_buffers(page, head); + spin_unlock(&page->mapping->private_lock); +} +EXPORT_SYMBOL(create_empty_buffers); + +/* + * We are taking a block for data and we don't want any output from any + * buffer-cache aliases starting from return from that function and + * until the moment when something will explicitly mark the buffer + * dirty (hopefully that will not happen until we will free that block ;-) + * We don't even need to mark it not-uptodate - nobody can expect + * anything from a newly allocated buffer anyway. We used to used + * unmap_buffer() for such invalidation, but that was wrong. We definitely + * don't want to mark the alias unmapped, for example - it would confuse + * anyone who might pick it with bread() afterwards... + * + * Also.. Note that bforget() doesn't lock the buffer. So there can + * be writeout I/O going on against recently-freed buffers. We don't + * wait on that I/O in bforget() - it's more efficient to wait on the I/O + * only if we really need to. That happens here. + */ +void unmap_underlying_metadata(struct block_device *bdev, sector_t block) +{ + struct buffer_head *old_bh; + + might_sleep(); + + old_bh = __find_get_block_slow(bdev, block); + if (old_bh) { + clear_buffer_dirty(old_bh); + wait_on_buffer(old_bh); + clear_buffer_req(old_bh); + __brelse(old_bh); + } +} +EXPORT_SYMBOL(unmap_underlying_metadata); + +/* + * Size is a power-of-two in the range 512..PAGE_SIZE, + * and the case we care about most is PAGE_SIZE. + * + * So this *could* possibly be written with those + * constraints in mind (relevant mostly if some + * architecture has a slow bit-scan instruction) + */ +static inline int block_size_bits(unsigned int blocksize) +{ + return ilog2(blocksize); +} + +static struct buffer_head *create_page_buffers(struct page *page, struct inode *inode, unsigned int b_state) +{ + BUG_ON(!PageLocked(page)); + + if (!page_has_buffers(page)) + create_empty_buffers(page, 1 << ACCESS_ONCE(inode->i_blkbits), b_state); + return page_buffers(page); +} + +/* + * NOTE! All mapped/uptodate combinations are valid: + * + * Mapped Uptodate Meaning + * + * No No "unknown" - must do get_block() + * No Yes "hole" - zero-filled + * Yes No "allocated" - allocated on disk, not read in + * Yes Yes "valid" - allocated and up-to-date in memory. + * + * "Dirty" is valid only with the last case (mapped+uptodate). + */ + +/* + * While block_write_full_page is writing back the dirty buffers under + * the page lock, whoever dirtied the buffers may decide to clean them + * again at any time. We handle that by only looking at the buffer + * state inside lock_buffer(). + * + * If block_write_full_page() is called for regular writeback + * (wbc->sync_mode == WB_SYNC_NONE) then it will redirty a page which has a + * locked buffer. This only can happen if someone has written the buffer + * directly, with submit_bh(). At the address_space level PageWriteback + * prevents this contention from occurring. + * + * If block_write_full_page() is called with wbc->sync_mode == + * WB_SYNC_ALL, the writes are posted using WRITE_SYNC; this + * causes the writes to be flagged as synchronous writes. + */ +static int __block_write_full_page(struct inode *inode, struct page *page, + get_block_t *get_block, struct writeback_control *wbc, + bh_end_io_t *handler) +{ + int err; + sector_t block; + sector_t last_block; + struct buffer_head *bh, *head; + unsigned int blocksize, bbits; + int nr_underway = 0; + int write_op = (wbc->sync_mode == WB_SYNC_ALL ? + WRITE_SYNC : WRITE); + + head = create_page_buffers(page, inode, + (1 << BH_Dirty)|(1 << BH_Uptodate)); + + /* + * Be very careful. We have no exclusion from __set_page_dirty_buffers + * here, and the (potentially unmapped) buffers may become dirty at + * any time. If a buffer becomes dirty here after we've inspected it + * then we just miss that fact, and the page stays dirty. + * + * Buffers outside i_size may be dirtied by __set_page_dirty_buffers; + * handle that here by just cleaning them. + */ + + bh = head; + blocksize = bh->b_size; + bbits = block_size_bits(blocksize); + + block = (sector_t)page->index << (PAGE_CACHE_SHIFT - bbits); + last_block = (i_size_read(inode) - 1) >> bbits; + + /* + * Get all the dirty buffers mapped to disk addresses and + * handle any aliases from the underlying blockdev's mapping. + */ + do { + if (block > last_block) { + /* + * mapped buffers outside i_size will occur, because + * this page can be outside i_size when there is a + * truncate in progress. + */ + /* + * The buffer was zeroed by block_write_full_page() + */ + clear_buffer_dirty(bh); + set_buffer_uptodate(bh); + } else if ((!buffer_mapped(bh) || buffer_delay(bh)) && + buffer_dirty(bh)) { + WARN_ON(bh->b_size != blocksize); + err = get_block(inode, block, bh, 1); + if (err) + goto recover; + clear_buffer_delay(bh); + if (buffer_new(bh)) { + /* blockdev mappings never come here */ + clear_buffer_new(bh); + unmap_underlying_metadata(bh->b_bdev, + bh->b_blocknr); + } + } + bh = bh->b_this_page; + block++; + } while (bh != head); + + do { + if (!buffer_mapped(bh)) + continue; + /* + * If it's a fully non-blocking write attempt and we cannot + * lock the buffer then redirty the page. Note that this can + * potentially cause a busy-wait loop from writeback threads + * and kswapd activity, but those code paths have their own + * higher-level throttling. + */ + if (wbc->sync_mode != WB_SYNC_NONE) { + lock_buffer(bh); + } else if (!trylock_buffer(bh)) { + redirty_page_for_writepage(wbc, page); + continue; + } + if (test_clear_buffer_dirty(bh)) { + mark_buffer_async_write_endio(bh, handler); + } else { + unlock_buffer(bh); + } + } while ((bh = bh->b_this_page) != head); + + /* + * The page and its buffers are protected by PageWriteback(), so we can + * drop the bh refcounts early. + */ + BUG_ON(PageWriteback(page)); + set_page_writeback(page); + + do { + struct buffer_head *next = bh->b_this_page; + if (buffer_async_write(bh)) { + submit_bh(write_op, bh); + nr_underway++; + } + bh = next; + } while (bh != head); + unlock_page(page); + + err = 0; +done: + if (nr_underway == 0) { + /* + * The page was marked dirty, but the buffers were + * clean. Someone wrote them back by hand with + * ll_rw_block/submit_bh. A rare case. + */ + end_page_writeback(page); + + /* + * The page and buffer_heads can be released at any time from + * here on. + */ + } + return err; + +recover: + /* + * ENOSPC, or some other error. We may already have added some + * blocks to the file, so we need to write these out to avoid + * exposing stale data. + * The page is currently locked and not marked for writeback + */ + bh = head; + /* Recovery: lock and submit the mapped buffers */ + do { + if (buffer_mapped(bh) && buffer_dirty(bh) && + !buffer_delay(bh)) { + lock_buffer(bh); + mark_buffer_async_write_endio(bh, handler); + } else { + /* + * The buffer may have been set dirty during + * attachment to a dirty page. + */ + clear_buffer_dirty(bh); + } + } while ((bh = bh->b_this_page) != head); + SetPageError(page); + BUG_ON(PageWriteback(page)); + mapping_set_error(page->mapping, err); + set_page_writeback(page); + do { + struct buffer_head *next = bh->b_this_page; + if (buffer_async_write(bh)) { + clear_buffer_dirty(bh); + submit_bh(write_op, bh); + nr_underway++; + } + bh = next; + } while (bh != head); + unlock_page(page); + goto done; +} + +/* + * If a page has any new buffers, zero them out here, and mark them uptodate + * and dirty so they'll be written out (in order to prevent uninitialised + * block data from leaking). And clear the new bit. + */ +void page_zero_new_buffers(struct page *page, unsigned from, unsigned to) +{ + unsigned int block_start, block_end; + struct buffer_head *head, *bh; + + BUG_ON(!PageLocked(page)); + if (!page_has_buffers(page)) + return; + + bh = head = page_buffers(page); + block_start = 0; + do { + block_end = block_start + bh->b_size; + + if (buffer_new(bh)) { + if (block_end > from && block_start < to) { + if (!PageUptodate(page)) { + unsigned start, size; + + start = max(from, block_start); + size = min(to, block_end) - start; + + zero_user(page, start, size); + set_buffer_uptodate(bh); + } + + clear_buffer_new(bh); + mark_buffer_dirty(bh); + } + } + + block_start = block_end; + bh = bh->b_this_page; + } while (bh != head); +} +EXPORT_SYMBOL(page_zero_new_buffers); + +int __block_write_begin(struct page *page, loff_t pos, unsigned len, + get_block_t *get_block) +{ + unsigned from = pos & (PAGE_CACHE_SIZE - 1); + unsigned to = from + len; + struct inode *inode = page->mapping->host; + unsigned block_start, block_end; + sector_t block; + int err = 0; + unsigned blocksize, bbits; + struct buffer_head *bh, *head, *wait[2], **wait_bh=wait; + + BUG_ON(!PageLocked(page)); + BUG_ON(from > PAGE_CACHE_SIZE); + BUG_ON(to > PAGE_CACHE_SIZE); + BUG_ON(from > to); + + head = create_page_buffers(page, inode, 0); + blocksize = head->b_size; + bbits = block_size_bits(blocksize); + + block = (sector_t)page->index << (PAGE_CACHE_SHIFT - bbits); + + for(bh = head, block_start = 0; bh != head || !block_start; + block++, block_start=block_end, bh = bh->b_this_page) { + block_end = block_start + blocksize; + if (block_end <= from || block_start >= to) { + if (PageUptodate(page)) { + if (!buffer_uptodate(bh)) + set_buffer_uptodate(bh); + } + continue; + } + if (buffer_new(bh)) + clear_buffer_new(bh); + if (!buffer_mapped(bh)) { + WARN_ON(bh->b_size != blocksize); + err = get_block(inode, block, bh, 1); + if (err) + break; + if (buffer_new(bh)) { + unmap_underlying_metadata(bh->b_bdev, + bh->b_blocknr); + if (PageUptodate(page)) { + clear_buffer_new(bh); + set_buffer_uptodate(bh); + mark_buffer_dirty(bh); + continue; + } + if (block_end > to || block_start < from) + zero_user_segments(page, + to, block_end, + block_start, from); + continue; + } + } + if (PageUptodate(page)) { + if (!buffer_uptodate(bh)) + set_buffer_uptodate(bh); + continue; + } + if (!buffer_uptodate(bh) && !buffer_delay(bh) && + !buffer_unwritten(bh) && + (block_start < from || block_end > to)) { + ll_rw_block(READ, 1, &bh); + *wait_bh++=bh; + } + } + /* + * If we issued read requests - let them complete. + */ + while(wait_bh > wait) { + wait_on_buffer(*--wait_bh); + if (!buffer_uptodate(*wait_bh)) + err = -EIO; + } + if (unlikely(err)) + page_zero_new_buffers(page, from, to); + return err; +} +EXPORT_SYMBOL(__block_write_begin); + +static int __block_commit_write(struct inode *inode, struct page *page, + unsigned from, unsigned to) +{ + unsigned block_start, block_end; + int partial = 0; + unsigned blocksize; + struct buffer_head *bh, *head; + + bh = head = page_buffers(page); + blocksize = bh->b_size; + + block_start = 0; + do { + block_end = block_start + blocksize; + if (block_end <= from || block_start >= to) { + if (!buffer_uptodate(bh)) + partial = 1; + } else { + set_buffer_uptodate(bh); + mark_buffer_dirty(bh); + } + clear_buffer_new(bh); + + block_start = block_end; + bh = bh->b_this_page; + } while (bh != head); + + /* + * If this is a partial write which happened to make all buffers + * uptodate then we can optimize away a bogus readpage() for + * the next read(). Here we 'discover' whether the page went + * uptodate as a result of this (potentially partial) write. + */ + if (!partial) + SetPageUptodate(page); + return 0; +} + +/* + * block_write_begin takes care of the basic task of block allocation and + * bringing partial write blocks uptodate first. + * + * The filesystem needs to handle block truncation upon failure. + */ +int block_write_begin(struct address_space *mapping, loff_t pos, unsigned len, + unsigned flags, struct page **pagep, get_block_t *get_block) +{ + pgoff_t index = pos >> PAGE_CACHE_SHIFT; + struct page *page; + int status; + + page = grab_cache_page_write_begin(mapping, index, flags); + if (!page) + return -ENOMEM; + + status = __block_write_begin(page, pos, len, get_block); + if (unlikely(status)) { + unlock_page(page); + page_cache_release(page); + page = NULL; + } + + *pagep = page; + return status; +} +EXPORT_SYMBOL(block_write_begin); + +int block_write_end(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata) +{ + struct inode *inode = mapping->host; + unsigned start; + + start = pos & (PAGE_CACHE_SIZE - 1); + + if (unlikely(copied < len)) { + /* + * The buffers that were written will now be uptodate, so we + * don't have to worry about a readpage reading them and + * overwriting a partial write. However if we have encountered + * a short write and only partially written into a buffer, it + * will not be marked uptodate, so a readpage might come in and + * destroy our partial write. + * + * Do the simplest thing, and just treat any short write to a + * non uptodate page as a zero-length write, and force the + * caller to redo the whole thing. + */ + if (!PageUptodate(page)) + copied = 0; + + page_zero_new_buffers(page, start+copied, start+len); + } + flush_dcache_page(page); + + /* This could be a short (even 0-length) commit */ + __block_commit_write(inode, page, start, start+copied); + + return copied; +} +EXPORT_SYMBOL(block_write_end); + +int generic_write_end(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata) +{ + struct inode *inode = mapping->host; + int i_size_changed = 0; + + copied = block_write_end(file, mapping, pos, len, copied, page, fsdata); + + /* + * No need to use i_size_read() here, the i_size + * cannot change under us because we hold i_mutex. + * + * But it's important to update i_size while still holding page lock: + * page writeout could otherwise come in and zero beyond i_size. + */ + if (pos+copied > inode->i_size) { + i_size_write(inode, pos+copied); + i_size_changed = 1; + } + + unlock_page(page); + page_cache_release(page); + + /* + * Don't mark the inode dirty under page lock. First, it unnecessarily + * makes the holding time of page lock longer. Second, it forces lock + * ordering of page lock and transaction start for journaling + * filesystems. + */ + if (i_size_changed) + mark_inode_dirty(inode); + + return copied; +} +EXPORT_SYMBOL(generic_write_end); + +/* + * block_is_partially_uptodate checks whether buffers within a page are + * uptodate or not. + * + * Returns true if all buffers which correspond to a file portion + * we want to read are uptodate. + */ +int block_is_partially_uptodate(struct page *page, read_descriptor_t *desc, + unsigned long from) +{ + unsigned block_start, block_end, blocksize; + unsigned to; + struct buffer_head *bh, *head; + int ret = 1; + + if (!page_has_buffers(page)) + return 0; + + head = page_buffers(page); + blocksize = head->b_size; + to = min_t(unsigned, PAGE_CACHE_SIZE - from, desc->count); + to = from + to; + if (from < blocksize && to > PAGE_CACHE_SIZE - blocksize) + return 0; + + bh = head; + block_start = 0; + do { + block_end = block_start + blocksize; + if (block_end > from && block_start < to) { + if (!buffer_uptodate(bh)) { + ret = 0; + break; + } + if (block_end >= to) + break; + } + block_start = block_end; + bh = bh->b_this_page; + } while (bh != head); + + return ret; +} +EXPORT_SYMBOL(block_is_partially_uptodate); + +/* + * Generic "read page" function for block devices that have the normal + * get_block functionality. This is most of the block device filesystems. + * Reads the page asynchronously --- the unlock_buffer() and + * set/clear_buffer_uptodate() functions propagate buffer state into the + * page struct once IO has completed. + */ +int block_read_full_page(struct page *page, get_block_t *get_block) +{ + struct inode *inode = page->mapping->host; + sector_t iblock, lblock; + struct buffer_head *bh, *head, *arr[MAX_BUF_PER_PAGE]; + unsigned int blocksize, bbits; + int nr, i; + int fully_mapped = 1; + + head = create_page_buffers(page, inode, 0); + blocksize = head->b_size; + bbits = block_size_bits(blocksize); + + iblock = (sector_t)page->index << (PAGE_CACHE_SHIFT - bbits); + lblock = (i_size_read(inode)+blocksize-1) >> bbits; + bh = head; + nr = 0; + i = 0; + + do { + if (buffer_uptodate(bh)) + continue; + + if (!buffer_mapped(bh)) { + int err = 0; + + fully_mapped = 0; + if (iblock < lblock) { + WARN_ON(bh->b_size != blocksize); + err = get_block(inode, iblock, bh, 0); + if (err) + SetPageError(page); + } + if (!buffer_mapped(bh)) { + zero_user(page, i * blocksize, blocksize); + if (!err) + set_buffer_uptodate(bh); + continue; + } + /* + * get_block() might have updated the buffer + * synchronously + */ + if (buffer_uptodate(bh)) + continue; + } + arr[nr++] = bh; + } while (i++, iblock++, (bh = bh->b_this_page) != head); + + if (fully_mapped) + SetPageMappedToDisk(page); + + if (!nr) { + /* + * All buffers are uptodate - we can set the page uptodate + * as well. But not if get_block() returned an error. + */ + if (!PageError(page)) + SetPageUptodate(page); + unlock_page(page); + return 0; + } + + /* Stage two: lock the buffers */ + for (i = 0; i < nr; i++) { + bh = arr[i]; + lock_buffer(bh); + mark_buffer_async_read(bh); + } + + /* + * Stage 3: start the IO. Check for uptodateness + * inside the buffer lock in case another process reading + * the underlying blockdev brought it uptodate (the sct fix). + */ + for (i = 0; i < nr; i++) { + bh = arr[i]; + if (buffer_uptodate(bh)) + end_buffer_async_read(bh, 1); + else + submit_bh(READ, bh); + } + return 0; +} +EXPORT_SYMBOL(block_read_full_page); + +/* utility function for filesystems that need to do work on expanding + * truncates. Uses filesystem pagecache writes to allow the filesystem to + * deal with the hole. + */ +int generic_cont_expand_simple(struct inode *inode, loff_t size) +{ + struct address_space *mapping = inode->i_mapping; + struct page *page; + void *fsdata; + int err; + + err = inode_newsize_ok(inode, size); + if (err) + goto out; + + err = pagecache_write_begin(NULL, mapping, size, 0, + AOP_FLAG_UNINTERRUPTIBLE|AOP_FLAG_CONT_EXPAND, + &page, &fsdata); + if (err) + goto out; + + err = pagecache_write_end(NULL, mapping, size, 0, 0, page, fsdata); + BUG_ON(err > 0); + +out: + return err; +} +EXPORT_SYMBOL(generic_cont_expand_simple); + +static int cont_expand_zero(struct file *file, struct address_space *mapping, + loff_t pos, loff_t *bytes) +{ + struct inode *inode = mapping->host; + unsigned blocksize = 1 << inode->i_blkbits; + struct page *page; + void *fsdata; + pgoff_t index, curidx; + loff_t curpos; + unsigned zerofrom, offset, len; + int err = 0; + + index = pos >> PAGE_CACHE_SHIFT; + offset = pos & ~PAGE_CACHE_MASK; + + while (index > (curidx = (curpos = *bytes)>>PAGE_CACHE_SHIFT)) { + zerofrom = curpos & ~PAGE_CACHE_MASK; + if (zerofrom & (blocksize-1)) { + *bytes |= (blocksize-1); + (*bytes)++; + } + len = PAGE_CACHE_SIZE - zerofrom; + + err = pagecache_write_begin(file, mapping, curpos, len, + AOP_FLAG_UNINTERRUPTIBLE, + &page, &fsdata); + if (err) + goto out; + zero_user(page, zerofrom, len); + err = pagecache_write_end(file, mapping, curpos, len, len, + page, fsdata); + if (err < 0) + goto out; + BUG_ON(err != len); + err = 0; + + balance_dirty_pages_ratelimited(mapping); + } + + /* page covers the boundary, find the boundary offset */ + if (index == curidx) { + zerofrom = curpos & ~PAGE_CACHE_MASK; + /* if we will expand the thing last block will be filled */ + if (offset <= zerofrom) { + goto out; + } + if (zerofrom & (blocksize-1)) { + *bytes |= (blocksize-1); + (*bytes)++; + } + len = offset - zerofrom; + + err = pagecache_write_begin(file, mapping, curpos, len, + AOP_FLAG_UNINTERRUPTIBLE, + &page, &fsdata); + if (err) + goto out; + zero_user(page, zerofrom, len); + err = pagecache_write_end(file, mapping, curpos, len, len, + page, fsdata); + if (err < 0) + goto out; + BUG_ON(err != len); + err = 0; + } +out: + return err; +} + +/* + * For moronic filesystems that do not allow holes in file. + * We may have to extend the file. + */ +int cont_write_begin(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned flags, + struct page **pagep, void **fsdata, + get_block_t *get_block, loff_t *bytes) +{ + struct inode *inode = mapping->host; + unsigned blocksize = 1 << inode->i_blkbits; + unsigned zerofrom; + int err; + + err = cont_expand_zero(file, mapping, pos, bytes); + if (err) + return err; + + zerofrom = *bytes & ~PAGE_CACHE_MASK; + if (pos+len > *bytes && zerofrom & (blocksize-1)) { + *bytes |= (blocksize-1); + (*bytes)++; + } + + return block_write_begin(mapping, pos, len, flags, pagep, get_block); +} +EXPORT_SYMBOL(cont_write_begin); + +int block_commit_write(struct page *page, unsigned from, unsigned to) +{ + struct inode *inode = page->mapping->host; + __block_commit_write(inode,page,from,to); + return 0; +} +EXPORT_SYMBOL(block_commit_write); + +/* + * block_page_mkwrite() is not allowed to change the file size as it gets + * called from a page fault handler when a page is first dirtied. Hence we must + * be careful to check for EOF conditions here. We set the page up correctly + * for a written page which means we get ENOSPC checking when writing into + * holes and correct delalloc and unwritten extent mapping on filesystems that + * support these features. + * + * We are not allowed to take the i_mutex here so we have to play games to + * protect against truncate races as the page could now be beyond EOF. Because + * truncate writes the inode size before removing pages, once we have the + * page lock we can determine safely if the page is beyond EOF. If it is not + * beyond EOF, then the page is guaranteed safe against truncation until we + * unlock the page. + * + * Direct callers of this function should protect against filesystem freezing + * using sb_start_write() - sb_end_write() functions. + */ +int __block_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf, + get_block_t get_block) +{ + struct page *page = vmf->page; + struct inode *inode = file_inode(vma->vm_file); + unsigned long end; + loff_t size; + int ret; + + lock_page(page); + size = i_size_read(inode); + if ((page->mapping != inode->i_mapping) || + (page_offset(page) > size)) { + /* We overload EFAULT to mean page got truncated */ + ret = -EFAULT; + goto out_unlock; + } + + /* page is wholly or partially inside EOF */ + if (((page->index + 1) << PAGE_CACHE_SHIFT) > size) + end = size & ~PAGE_CACHE_MASK; + else + end = PAGE_CACHE_SIZE; + + ret = __block_write_begin(page, 0, end, get_block); + if (!ret) + ret = block_commit_write(page, 0, end); + + if (unlikely(ret < 0)) + goto out_unlock; + set_page_dirty(page); + wait_for_stable_page(page); + return 0; +out_unlock: + unlock_page(page); + return ret; +} +EXPORT_SYMBOL(__block_page_mkwrite); + +int block_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf, + get_block_t get_block) +{ + int ret; + struct super_block *sb = file_inode(vma->vm_file)->i_sb; + + sb_start_pagefault(sb); + + /* + * Update file times before taking page lock. We may end up failing the + * fault so this update may be superfluous but who really cares... + */ + file_update_time(vma->vm_file); + + ret = __block_page_mkwrite(vma, vmf, get_block); + sb_end_pagefault(sb); + return block_page_mkwrite_return(ret); +} +EXPORT_SYMBOL(block_page_mkwrite); + +/* + * nobh_write_begin()'s prereads are special: the buffer_heads are freed + * immediately, while under the page lock. So it needs a special end_io + * handler which does not touch the bh after unlocking it. + */ +static void end_buffer_read_nobh(struct buffer_head *bh, int uptodate) +{ + __end_buffer_read_notouch(bh, uptodate); +} + +/* + * Attach the singly-linked list of buffers created by nobh_write_begin, to + * the page (converting it to circular linked list and taking care of page + * dirty races). + */ +static void attach_nobh_buffers(struct page *page, struct buffer_head *head) +{ + struct buffer_head *bh; + + BUG_ON(!PageLocked(page)); + + spin_lock(&page->mapping->private_lock); + bh = head; + do { + if (PageDirty(page)) + set_buffer_dirty(bh); + if (!bh->b_this_page) + bh->b_this_page = head; + bh = bh->b_this_page; + } while (bh != head); + attach_page_buffers(page, head); + spin_unlock(&page->mapping->private_lock); +} + +/* + * On entry, the page is fully not uptodate. + * On exit the page is fully uptodate in the areas outside (from,to) + * The filesystem needs to handle block truncation upon failure. + */ +int nobh_write_begin(struct address_space *mapping, + loff_t pos, unsigned len, unsigned flags, + struct page **pagep, void **fsdata, + get_block_t *get_block) +{ + struct inode *inode = mapping->host; + const unsigned blkbits = inode->i_blkbits; + const unsigned blocksize = 1 << blkbits; + struct buffer_head *head, *bh; + struct page *page; + pgoff_t index; + unsigned from, to; + unsigned block_in_page; + unsigned block_start, block_end; + sector_t block_in_file; + int nr_reads = 0; + int ret = 0; + int is_mapped_to_disk = 1; + + index = pos >> PAGE_CACHE_SHIFT; + from = pos & (PAGE_CACHE_SIZE - 1); + to = from + len; + + page = grab_cache_page_write_begin(mapping, index, flags); + if (!page) + return -ENOMEM; + *pagep = page; + *fsdata = NULL; + + if (page_has_buffers(page)) { + ret = __block_write_begin(page, pos, len, get_block); + if (unlikely(ret)) + goto out_release; + return ret; + } + + if (PageMappedToDisk(page)) + return 0; + + /* + * Allocate buffers so that we can keep track of state, and potentially + * attach them to the page if an error occurs. In the common case of + * no error, they will just be freed again without ever being attached + * to the page (which is all OK, because we're under the page lock). + * + * Be careful: the buffer linked list is a NULL terminated one, rather + * than the circular one we're used to. + */ + head = alloc_page_buffers(page, blocksize, 0); + if (!head) { + ret = -ENOMEM; + goto out_release; + } + + block_in_file = (sector_t)page->index << (PAGE_CACHE_SHIFT - blkbits); + + /* + * We loop across all blocks in the page, whether or not they are + * part of the affected region. This is so we can discover if the + * page is fully mapped-to-disk. + */ + for (block_start = 0, block_in_page = 0, bh = head; + block_start < PAGE_CACHE_SIZE; + block_in_page++, block_start += blocksize, bh = bh->b_this_page) { + int create; + + block_end = block_start + blocksize; + bh->b_state = 0; + create = 1; + if (block_start >= to) + create = 0; + ret = get_block(inode, block_in_file + block_in_page, + bh, create); + if (ret) + goto failed; + if (!buffer_mapped(bh)) + is_mapped_to_disk = 0; + if (buffer_new(bh)) + unmap_underlying_metadata(bh->b_bdev, bh->b_blocknr); + if (PageUptodate(page)) { + set_buffer_uptodate(bh); + continue; + } + if (buffer_new(bh) || !buffer_mapped(bh)) { + zero_user_segments(page, block_start, from, + to, block_end); + continue; + } + if (buffer_uptodate(bh)) + continue; /* reiserfs does this */ + if (block_start < from || block_end > to) { + lock_buffer(bh); + bh->b_end_io = end_buffer_read_nobh; + submit_bh(READ, bh); + nr_reads++; + } + } + + if (nr_reads) { + /* + * The page is locked, so these buffers are protected from + * any VM or truncate activity. Hence we don't need to care + * for the buffer_head refcounts. + */ + for (bh = head; bh; bh = bh->b_this_page) { + wait_on_buffer(bh); + if (!buffer_uptodate(bh)) + ret = -EIO; + } + if (ret) + goto failed; + } + + if (is_mapped_to_disk) + SetPageMappedToDisk(page); + + *fsdata = head; /* to be released by nobh_write_end */ + + return 0; + +failed: + BUG_ON(!ret); + /* + * Error recovery is a bit difficult. We need to zero out blocks that + * were newly allocated, and dirty them to ensure they get written out. + * Buffers need to be attached to the page at this point, otherwise + * the handling of potential IO errors during writeout would be hard + * (could try doing synchronous writeout, but what if that fails too?) + */ + attach_nobh_buffers(page, head); + page_zero_new_buffers(page, from, to); + +out_release: + unlock_page(page); + page_cache_release(page); + *pagep = NULL; + + return ret; +} +EXPORT_SYMBOL(nobh_write_begin); + +int nobh_write_end(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata) +{ + struct inode *inode = page->mapping->host; + struct buffer_head *head = fsdata; + struct buffer_head *bh; + BUG_ON(fsdata != NULL && page_has_buffers(page)); + + if (unlikely(copied < len) && head) + attach_nobh_buffers(page, head); + if (page_has_buffers(page)) + return generic_write_end(file, mapping, pos, len, + copied, page, fsdata); + + SetPageUptodate(page); + set_page_dirty(page); + if (pos+copied > inode->i_size) { + i_size_write(inode, pos+copied); + mark_inode_dirty(inode); + } + + unlock_page(page); + page_cache_release(page); + + while (head) { + bh = head; + head = head->b_this_page; + free_buffer_head(bh); + } + + return copied; +} +EXPORT_SYMBOL(nobh_write_end); + +/* + * nobh_writepage() - based on block_full_write_page() except + * that it tries to operate without attaching bufferheads to + * the page. + */ +int nobh_writepage(struct page *page, get_block_t *get_block, + struct writeback_control *wbc) +{ + struct inode * const inode = page->mapping->host; + loff_t i_size = i_size_read(inode); + const pgoff_t end_index = i_size >> PAGE_CACHE_SHIFT; + unsigned offset; + int ret; + + /* Is the page fully inside i_size? */ + if (page->index < end_index) + goto out; + + /* Is the page fully outside i_size? (truncate in progress) */ + offset = i_size & (PAGE_CACHE_SIZE-1); + if (page->index >= end_index+1 || !offset) { + /* + * The page may have dirty, unmapped buffers. For example, + * they may have been added in ext3_writepage(). Make them + * freeable here, so the page does not leak. + */ +#if 0 + /* Not really sure about this - do we need this ? */ + if (page->mapping->a_ops->invalidatepage) + page->mapping->a_ops->invalidatepage(page, offset); +#endif + unlock_page(page); + return 0; /* don't care */ + } + + /* + * The page straddles i_size. It must be zeroed out on each and every + * writepage invocation because it may be mmapped. "A file is mapped + * in multiples of the page size. For a file that is not a multiple of + * the page size, the remaining memory is zeroed when mapped, and + * writes to that region are not written out to the file." + */ + zero_user_segment(page, offset, PAGE_CACHE_SIZE); +out: + ret = mpage_writepage(page, get_block, wbc); + if (ret == -EAGAIN) + ret = __block_write_full_page(inode, page, get_block, wbc, + end_buffer_async_write); + return ret; +} +EXPORT_SYMBOL(nobh_writepage); + +int nobh_truncate_page(struct address_space *mapping, + loff_t from, get_block_t *get_block) +{ + pgoff_t index = from >> PAGE_CACHE_SHIFT; + unsigned offset = from & (PAGE_CACHE_SIZE-1); + unsigned blocksize; + sector_t iblock; + unsigned length, pos; + struct inode *inode = mapping->host; + struct page *page; + struct buffer_head map_bh; + int err; + + blocksize = 1 << inode->i_blkbits; + length = offset & (blocksize - 1); + + /* Block boundary? Nothing to do */ + if (!length) + return 0; + + length = blocksize - length; + iblock = (sector_t)index << (PAGE_CACHE_SHIFT - inode->i_blkbits); + + page = grab_cache_page(mapping, index); + err = -ENOMEM; + if (!page) + goto out; + + if (page_has_buffers(page)) { +has_buffers: + unlock_page(page); + page_cache_release(page); + return block_truncate_page(mapping, from, get_block); + } + + /* Find the buffer that contains "offset" */ + pos = blocksize; + while (offset >= pos) { + iblock++; + pos += blocksize; + } + + map_bh.b_size = blocksize; + map_bh.b_state = 0; + err = get_block(inode, iblock, &map_bh, 0); + if (err) + goto unlock; + /* unmapped? It's a hole - nothing to do */ + if (!buffer_mapped(&map_bh)) + goto unlock; + + /* Ok, it's mapped. Make sure it's up-to-date */ + if (!PageUptodate(page)) { + err = mapping->a_ops->readpage(NULL, page); + if (err) { + page_cache_release(page); + goto out; + } + lock_page(page); + if (!PageUptodate(page)) { + err = -EIO; + goto unlock; + } + if (page_has_buffers(page)) + goto has_buffers; + } + zero_user(page, offset, length); + set_page_dirty(page); + err = 0; + +unlock: + unlock_page(page); + page_cache_release(page); +out: + return err; +} +EXPORT_SYMBOL(nobh_truncate_page); + +int block_truncate_page(struct address_space *mapping, + loff_t from, get_block_t *get_block) +{ + pgoff_t index = from >> PAGE_CACHE_SHIFT; + unsigned offset = from & (PAGE_CACHE_SIZE-1); + unsigned blocksize; + sector_t iblock; + unsigned length, pos; + struct inode *inode = mapping->host; + struct page *page; + struct buffer_head *bh; + int err; + + blocksize = 1 << inode->i_blkbits; + length = offset & (blocksize - 1); + + /* Block boundary? Nothing to do */ + if (!length) + return 0; + + length = blocksize - length; + iblock = (sector_t)index << (PAGE_CACHE_SHIFT - inode->i_blkbits); + + page = grab_cache_page(mapping, index); + err = -ENOMEM; + if (!page) + goto out; + + if (!page_has_buffers(page)) + create_empty_buffers(page, blocksize, 0); + + /* Find the buffer that contains "offset" */ + bh = page_buffers(page); + pos = blocksize; + while (offset >= pos) { + bh = bh->b_this_page; + iblock++; + pos += blocksize; + } + + err = 0; + if (!buffer_mapped(bh)) { + WARN_ON(bh->b_size != blocksize); + err = get_block(inode, iblock, bh, 0); + if (err) + goto unlock; + /* unmapped? It's a hole - nothing to do */ + if (!buffer_mapped(bh)) + goto unlock; + } + + /* Ok, it's mapped. Make sure it's up-to-date */ + if (PageUptodate(page)) + set_buffer_uptodate(bh); + + if (!buffer_uptodate(bh) && !buffer_delay(bh) && !buffer_unwritten(bh)) { + err = -EIO; + ll_rw_block(READ, 1, &bh); + wait_on_buffer(bh); + /* Uhhuh. Read error. Complain and punt. */ + if (!buffer_uptodate(bh)) + goto unlock; + } + + zero_user(page, offset, length); + mark_buffer_dirty(bh); + err = 0; + +unlock: + unlock_page(page); + page_cache_release(page); +out: + return err; +} +EXPORT_SYMBOL(block_truncate_page); + +/* + * The generic ->writepage function for buffer-backed address_spaces + * this form passes in the end_io handler used to finish the IO. + */ +int block_write_full_page_endio(struct page *page, get_block_t *get_block, + struct writeback_control *wbc, bh_end_io_t *handler) +{ + struct inode * const inode = page->mapping->host; + loff_t i_size = i_size_read(inode); + const pgoff_t end_index = i_size >> PAGE_CACHE_SHIFT; + unsigned offset; + + /* Is the page fully inside i_size? */ + if (page->index < end_index) + return __block_write_full_page(inode, page, get_block, wbc, + handler); + + /* Is the page fully outside i_size? (truncate in progress) */ + offset = i_size & (PAGE_CACHE_SIZE-1); + if (page->index >= end_index+1 || !offset) { + /* + * The page may have dirty, unmapped buffers. For example, + * they may have been added in ext3_writepage(). Make them + * freeable here, so the page does not leak. + */ + do_invalidatepage(page, 0); + unlock_page(page); + return 0; /* don't care */ + } + + /* + * The page straddles i_size. It must be zeroed out on each and every + * writepage invocation because it may be mmapped. "A file is mapped + * in multiples of the page size. For a file that is not a multiple of + * the page size, the remaining memory is zeroed when mapped, and + * writes to that region are not written out to the file." + */ + zero_user_segment(page, offset, PAGE_CACHE_SIZE); + return __block_write_full_page(inode, page, get_block, wbc, handler); +} +EXPORT_SYMBOL(block_write_full_page_endio); + +/* + * The generic ->writepage function for buffer-backed address_spaces + */ +int block_write_full_page(struct page *page, get_block_t *get_block, + struct writeback_control *wbc) +{ + return block_write_full_page_endio(page, get_block, wbc, + end_buffer_async_write); +} +EXPORT_SYMBOL(block_write_full_page); + +sector_t generic_block_bmap(struct address_space *mapping, sector_t block, + get_block_t *get_block) +{ + struct buffer_head tmp; + struct inode *inode = mapping->host; + tmp.b_state = 0; + tmp.b_blocknr = 0; + tmp.b_size = 1 << inode->i_blkbits; + get_block(inode, block, &tmp, 0); + return tmp.b_blocknr; +} +EXPORT_SYMBOL(generic_block_bmap); + +static void end_bio_bh_io_sync(struct bio *bio, int err) +{ + struct buffer_head *bh = bio->bi_private; + + if (err == -EOPNOTSUPP) { + set_bit(BIO_EOPNOTSUPP, &bio->bi_flags); + } + + if (unlikely (test_bit(BIO_QUIET,&bio->bi_flags))) + set_bit(BH_Quiet, &bh->b_state); + + bh->b_end_io(bh, test_bit(BIO_UPTODATE, &bio->bi_flags)); + bio_put(bio); +} + +/* + * This allows us to do IO even on the odd last sectors + * of a device, even if the bh block size is some multiple + * of the physical sector size. + * + * We'll just truncate the bio to the size of the device, + * and clear the end of the buffer head manually. + * + * Truly out-of-range accesses will turn into actual IO + * errors, this only handles the "we need to be able to + * do IO at the final sector" case. + */ +static void guard_bh_eod(int rw, struct bio *bio, struct buffer_head *bh) +{ + sector_t maxsector; + unsigned bytes; + + maxsector = i_size_read(bio->bi_bdev->bd_inode) >> 9; + if (!maxsector) + return; + + /* + * If the *whole* IO is past the end of the device, + * let it through, and the IO layer will turn it into + * an EIO. + */ + if (unlikely(bio->bi_sector >= maxsector)) + return; + + maxsector -= bio->bi_sector; + bytes = bio->bi_size; + if (likely((bytes >> 9) <= maxsector)) + return; + + /* Uhhuh. We've got a bh that straddles the device size! */ + bytes = maxsector << 9; + + /* Truncate the bio.. */ + bio->bi_size = bytes; + bio->bi_io_vec[0].bv_len = bytes; + + /* ..and clear the end of the buffer for reads */ + if ((rw & RW_MASK) == READ) { + void *kaddr = kmap_atomic(bh->b_page); + memset(kaddr + bh_offset(bh) + bytes, 0, bh->b_size - bytes); + kunmap_atomic(kaddr); + flush_dcache_page(bh->b_page); + } +} + +int _submit_bh(int rw, struct buffer_head *bh, unsigned long bio_flags) +{ + struct bio *bio; + int ret = 0; + + BUG_ON(!buffer_locked(bh)); + BUG_ON(!buffer_mapped(bh)); + BUG_ON(!bh->b_end_io); + BUG_ON(buffer_delay(bh)); + BUG_ON(buffer_unwritten(bh)); + + /* + * Only clear out a write error when rewriting + */ + if (test_set_buffer_req(bh) && (rw & WRITE)) + clear_buffer_write_io_error(bh); + + /* + * from here on down, it's all bio -- do the initial mapping, + * submit_bio -> generic_make_request may further map this bio around + */ + bio = bio_alloc(GFP_NOIO, 1); + + bio->bi_sector = bh->b_blocknr * (bh->b_size >> 9); + bio->bi_bdev = bh->b_bdev; + bio->bi_io_vec[0].bv_page = bh->b_page; + bio->bi_io_vec[0].bv_len = bh->b_size; + bio->bi_io_vec[0].bv_offset = bh_offset(bh); + + bio->bi_vcnt = 1; + bio->bi_size = bh->b_size; + + bio->bi_end_io = end_bio_bh_io_sync; + bio->bi_private = bh; + bio->bi_flags |= bio_flags; + + /* Take care of bh's that straddle the end of the device */ + guard_bh_eod(rw, bio, bh); + + if (buffer_meta(bh)) + rw |= REQ_META; + if (buffer_prio(bh)) + rw |= REQ_PRIO; + + bio_get(bio); + submit_bio(rw, bio); + + if (bio_flagged(bio, BIO_EOPNOTSUPP)) + ret = -EOPNOTSUPP; + + bio_put(bio); + return ret; +} +EXPORT_SYMBOL_GPL(_submit_bh); + +int submit_bh(int rw, struct buffer_head *bh) +{ + return _submit_bh(rw, bh, 0); +} +EXPORT_SYMBOL(submit_bh); + +/** + * ll_rw_block: low-level access to block devices (DEPRECATED) + * @rw: whether to %READ or %WRITE or maybe %READA (readahead) + * @nr: number of &struct buffer_heads in the array + * @bhs: array of pointers to &struct buffer_head + * + * ll_rw_block() takes an array of pointers to &struct buffer_heads, and + * requests an I/O operation on them, either a %READ or a %WRITE. The third + * %READA option is described in the documentation for generic_make_request() + * which ll_rw_block() calls. + * + * This function drops any buffer that it cannot get a lock on (with the + * BH_Lock state bit), any buffer that appears to be clean when doing a write + * request, and any buffer that appears to be up-to-date when doing read + * request. Further it marks as clean buffers that are processed for + * writing (the buffer cache won't assume that they are actually clean + * until the buffer gets unlocked). + * + * ll_rw_block sets b_end_io to simple completion handler that marks + * the buffer up-to-date (if approriate), unlocks the buffer and wakes + * any waiters. + * + * All of the buffers must be for the same device, and must also be a + * multiple of the current approved size for the device. + */ +void ll_rw_block(int rw, int nr, struct buffer_head *bhs[]) +{ + int i; + + for (i = 0; i < nr; i++) { + struct buffer_head *bh = bhs[i]; + + if (!trylock_buffer(bh)) + continue; + if (rw == WRITE) { + if (test_clear_buffer_dirty(bh)) { + bh->b_end_io = end_buffer_write_sync; + get_bh(bh); + submit_bh(WRITE, bh); + continue; + } + } else { + if (!buffer_uptodate(bh)) { + bh->b_end_io = end_buffer_read_sync; + get_bh(bh); + submit_bh(rw, bh); + continue; + } + } + unlock_buffer(bh); + } +} +EXPORT_SYMBOL(ll_rw_block); + +void write_dirty_buffer(struct buffer_head *bh, int rw) +{ + lock_buffer(bh); + if (!test_clear_buffer_dirty(bh)) { + unlock_buffer(bh); + return; + } + bh->b_end_io = end_buffer_write_sync; + get_bh(bh); + submit_bh(rw, bh); +} +EXPORT_SYMBOL(write_dirty_buffer); + +/* + * For a data-integrity writeout, we need to wait upon any in-progress I/O + * and then start new I/O and then wait upon it. The caller must have a ref on + * the buffer_head. + */ +int __sync_dirty_buffer(struct buffer_head *bh, int rw) +{ + int ret = 0; + + WARN_ON(atomic_read(&bh->b_count) < 1); + lock_buffer(bh); + if (test_clear_buffer_dirty(bh)) { + get_bh(bh); + bh->b_end_io = end_buffer_write_sync; + ret = submit_bh(rw, bh); + wait_on_buffer(bh); + if (!ret && !buffer_uptodate(bh)) + ret = -EIO; + } else { + unlock_buffer(bh); + } + return ret; +} +EXPORT_SYMBOL(__sync_dirty_buffer); + +int sync_dirty_buffer(struct buffer_head *bh) +{ + return __sync_dirty_buffer(bh, WRITE_SYNC); +} +EXPORT_SYMBOL(sync_dirty_buffer); + +/* + * try_to_free_buffers() checks if all the buffers on this particular page + * are unused, and releases them if so. + * + * Exclusion against try_to_free_buffers may be obtained by either + * locking the page or by holding its mapping's private_lock. + * + * If the page is dirty but all the buffers are clean then we need to + * be sure to mark the page clean as well. This is because the page + * may be against a block device, and a later reattachment of buffers + * to a dirty page will set *all* buffers dirty. Which would corrupt + * filesystem data on the same device. + * + * The same applies to regular filesystem pages: if all the buffers are + * clean then we set the page clean and proceed. To do that, we require + * total exclusion from __set_page_dirty_buffers(). That is obtained with + * private_lock. + * + * try_to_free_buffers() is non-blocking. + */ +static inline int buffer_busy(struct buffer_head *bh) +{ + return atomic_read(&bh->b_count) | + (bh->b_state & ((1 << BH_Dirty) | (1 << BH_Lock))); +} + +static int +drop_buffers(struct page *page, struct buffer_head **buffers_to_free) +{ + struct buffer_head *head = page_buffers(page); + struct buffer_head *bh; + + bh = head; + do { + if (buffer_write_io_error(bh) && page->mapping) + set_bit(AS_EIO, &page->mapping->flags); + if (buffer_busy(bh)) + goto failed; + bh = bh->b_this_page; + } while (bh != head); + + do { + struct buffer_head *next = bh->b_this_page; + + if (bh->b_assoc_map) + __remove_assoc_queue(bh); + bh = next; + } while (bh != head); + *buffers_to_free = head; + __clear_page_buffers(page); + return 1; +failed: + return 0; +} + +int try_to_free_buffers(struct page *page) +{ + struct address_space * const mapping = page->mapping; + struct buffer_head *buffers_to_free = NULL; + int ret = 0; + + BUG_ON(!PageLocked(page)); + if (PageWriteback(page)) + return 0; + + if (mapping == NULL) { /* can this still happen? */ + ret = drop_buffers(page, &buffers_to_free); + goto out; + } + + spin_lock(&mapping->private_lock); + ret = drop_buffers(page, &buffers_to_free); + + /* + * If the filesystem writes its buffers by hand (eg ext3) + * then we can have clean buffers against a dirty page. We + * clean the page here; otherwise the VM will never notice + * that the filesystem did any IO at all. + * + * Also, during truncate, discard_buffer will have marked all + * the page's buffers clean. We discover that here and clean + * the page also. + * + * private_lock must be held over this entire operation in order + * to synchronise against __set_page_dirty_buffers and prevent the + * dirty bit from being lost. + */ + if (ret) + cancel_dirty_page(page, PAGE_CACHE_SIZE); + spin_unlock(&mapping->private_lock); +out: + if (buffers_to_free) { + struct buffer_head *bh = buffers_to_free; + + do { + struct buffer_head *next = bh->b_this_page; + free_buffer_head(bh); + bh = next; + } while (bh != buffers_to_free); + } + return ret; +} +EXPORT_SYMBOL(try_to_free_buffers); + +/* + * There are no bdflush tunables left. But distributions are + * still running obsolete flush daemons, so we terminate them here. + * + * Use of bdflush() is deprecated and will be removed in a future kernel. + * The `flush-X' kernel threads fully replace bdflush daemons and this call. + */ +SYSCALL_DEFINE2(bdflush, int, func, long, data) +{ + static int msg_count; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (msg_count < 5) { + msg_count++; + printk(KERN_INFO + "warning: process `%s' used the obsolete bdflush" + " system call\n", current->comm); + printk(KERN_INFO "Fix your initscripts?\n"); + } + + if (func == 1) + do_exit(0); + return 0; +} + +/* + * Buffer-head allocation + */ +static struct kmem_cache *bh_cachep __read_mostly; + +/* + * Once the number of bh's in the machine exceeds this level, we start + * stripping them in writeback. + */ +static unsigned long max_buffer_heads; + +int buffer_heads_over_limit; + +struct bh_accounting { + int nr; /* Number of live bh's */ + int ratelimit; /* Limit cacheline bouncing */ +}; + +static DEFINE_PER_CPU(struct bh_accounting, bh_accounting) = {0, 0}; + +static void recalc_bh_state(void) +{ + int i; + int tot = 0; + + if (__this_cpu_inc_return(bh_accounting.ratelimit) - 1 < 4096) + return; + __this_cpu_write(bh_accounting.ratelimit, 0); + for_each_online_cpu(i) + tot += per_cpu(bh_accounting, i).nr; + buffer_heads_over_limit = (tot > max_buffer_heads); +} + +struct buffer_head *alloc_buffer_head(gfp_t gfp_flags) +{ + struct buffer_head *ret = kmem_cache_zalloc(bh_cachep, gfp_flags); + if (ret) { + INIT_LIST_HEAD(&ret->b_assoc_buffers); + preempt_disable(); + __this_cpu_inc(bh_accounting.nr); + recalc_bh_state(); + preempt_enable(); + } + return ret; +} +EXPORT_SYMBOL(alloc_buffer_head); + +void free_buffer_head(struct buffer_head *bh) +{ + BUG_ON(!list_empty(&bh->b_assoc_buffers)); + kmem_cache_free(bh_cachep, bh); + preempt_disable(); + __this_cpu_dec(bh_accounting.nr); + recalc_bh_state(); + preempt_enable(); +} +EXPORT_SYMBOL(free_buffer_head); + +static void buffer_exit_cpu(int cpu) +{ + int i; + struct bh_lru *b = &per_cpu(bh_lrus, cpu); + + for (i = 0; i < BH_LRU_SIZE; i++) { + brelse(b->bhs[i]); + b->bhs[i] = NULL; + } + this_cpu_add(bh_accounting.nr, per_cpu(bh_accounting, cpu).nr); + per_cpu(bh_accounting, cpu).nr = 0; +} + +static int buffer_cpu_notify(struct notifier_block *self, + unsigned long action, void *hcpu) +{ + if (action == CPU_DEAD || action == CPU_DEAD_FROZEN) + buffer_exit_cpu((unsigned long)hcpu); + return NOTIFY_OK; +} + +/** + * bh_uptodate_or_lock - Test whether the buffer is uptodate + * @bh: struct buffer_head + * + * Return true if the buffer is up-to-date and false, + * with the buffer locked, if not. + */ +int bh_uptodate_or_lock(struct buffer_head *bh) +{ + if (!buffer_uptodate(bh)) { + lock_buffer(bh); + if (!buffer_uptodate(bh)) + return 0; + unlock_buffer(bh); + } + return 1; +} +EXPORT_SYMBOL(bh_uptodate_or_lock); + +/** + * bh_submit_read - Submit a locked buffer for reading + * @bh: struct buffer_head + * + * Returns zero on success and -EIO on error. + */ +int bh_submit_read(struct buffer_head *bh) +{ + BUG_ON(!buffer_locked(bh)); + + if (buffer_uptodate(bh)) { + unlock_buffer(bh); + return 0; + } + + get_bh(bh); + bh->b_end_io = end_buffer_read_sync; + submit_bh(READ, bh); + wait_on_buffer(bh); + if (buffer_uptodate(bh)) + return 0; + return -EIO; +} +EXPORT_SYMBOL(bh_submit_read); + +void __init buffer_init(void) +{ + unsigned long nrpages; + + bh_cachep = kmem_cache_create("buffer_head", + sizeof(struct buffer_head), 0, + (SLAB_RECLAIM_ACCOUNT|SLAB_PANIC| + SLAB_MEM_SPREAD), + NULL); + + /* + * Limit the bh occupancy to 10% of ZONE_NORMAL + */ + nrpages = (nr_free_buffer_pages() * 10) / 100; + max_buffer_heads = nrpages * (PAGE_SIZE / sizeof(struct buffer_head)); + hotcpu_notifier(buffer_cpu_notify, 0); +} diff -Naur kernel.21.3.orig/fs/file_table.c kernel.21.3.new/fs/file_table.c --- kernel.21.3.orig/fs/file_table.c 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/fs/file_table.c 2015-04-25 23:28:53.669398047 +0000 @@ -36,7 +36,8 @@ .max_files = NR_FILE }; -DEFINE_STATIC_LGLOCK(files_lglock); +DEFINE_LGLOCK(files_lglock); +EXPORT_SYMBOL(files_lglock); /* SLAB cache for file structures */ static struct kmem_cache *filp_cachep __read_mostly; @@ -405,6 +406,8 @@ } } +EXPORT_SYMBOL(file_sb_list_del); + #ifdef CONFIG_SMP /* diff -Naur kernel.21.3.orig/fs/inode.c kernel.21.3.new/fs/inode.c --- kernel.21.3.orig/fs/inode.c 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/fs/inode.c 2015-04-25 23:28:53.678397957 +0000 @@ -56,6 +56,7 @@ static __cacheline_aligned_in_smp DEFINE_SPINLOCK(inode_hash_lock); __cacheline_aligned_in_smp DEFINE_SPINLOCK(inode_sb_list_lock); +EXPORT_SYMBOL(inode_sb_list_lock); /* * Empty aops. Can be used for the cases where the user does not @@ -1498,7 +1499,7 @@ * This does the actual work of updating an inodes time or version. Must have * had called mnt_want_write() before calling this. */ -static int update_time(struct inode *inode, struct timespec *time, int flags) +int update_time(struct inode *inode, struct timespec *time, int flags) { if (inode->i_op->update_time) return inode->i_op->update_time(inode, time, flags); @@ -1514,6 +1515,7 @@ mark_inode_dirty_sync(inode); return 0; } +EXPORT_SYMBOL(update_time); /** * touch_atime - update the access time diff -Naur kernel.21.3.orig/fs/Kconfig kernel.21.3.new/fs/Kconfig --- kernel.21.3.orig/fs/Kconfig 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/fs/Kconfig 2015-04-25 23:28:39.211542285 +0000 @@ -212,6 +212,7 @@ source "fs/exofs/Kconfig" source "fs/f2fs/Kconfig" source "fs/efivarfs/Kconfig" +source "fs/aufs/Kconfig" endif # MISC_FILESYSTEMS diff -Naur kernel.21.3.orig/fs/Makefile kernel.21.3.new/fs/Makefile --- kernel.21.3.orig/fs/Makefile 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/fs/Makefile 2015-04-25 23:28:39.213542266 +0000 @@ -126,3 +126,4 @@ obj-$(CONFIG_CEPH_FS) += ceph/ obj-$(CONFIG_PSTORE) += pstore/ obj-$(CONFIG_EFIVAR_FS) += efivarfs/ +obj-$(CONFIG_AUFS_FS) += aufs/ diff -Naur kernel.21.3.orig/fs/namei.c kernel.21.3.new/fs/namei.c --- kernel.21.3.orig/fs/namei.c 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/fs/namei.c 2015-04-25 23:28:53.690397836 +0000 @@ -220,7 +220,7 @@ } #endif -static int check_acl(struct inode *inode, int mask) +int check_acl(struct inode *inode, int mask) { #ifdef CONFIG_FS_POSIX_ACL struct posix_acl *acl; @@ -265,6 +265,7 @@ return -EAGAIN; } +EXPORT_SYMBOL(check_acl); /* * This does the basic permission checking diff -Naur kernel.21.3.orig/fs/namespace.c kernel.21.3.new/fs/namespace.c --- kernel.21.3.orig/fs/namespace.c 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/fs/namespace.c 2015-04-25 23:28:53.693397806 +0000 @@ -54,6 +54,7 @@ * tree or hash is modified or when a vfsmount structure is modified. */ DEFINE_BRLOCK(vfsmount_lock); +EXPORT_SYMBOL(vfsmount_lock); static inline unsigned long hash(struct vfsmount *mnt, struct dentry *dentry) { @@ -427,6 +428,7 @@ mnt_dec_writers(real_mount(mnt)); preempt_enable(); } +EXPORT_SYMBOL_GPL(__mnt_drop_write); /** * mnt_drop_write - give up write access to a mount @@ -1456,6 +1458,7 @@ } return 0; } +EXPORT_SYMBOL(iterate_mounts); static void cleanup_group_ids(struct mount *mnt, struct mount *end) { diff -Naur kernel.21.3.orig/fs/notify/group.c kernel.21.3.new/fs/notify/group.c --- kernel.21.3.orig/fs/notify/group.c 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/fs/notify/group.c 2015-04-25 23:28:53.699397746 +0000 @@ -22,6 +22,7 @@ #include #include #include +#include #include #include "fsnotify.h" @@ -65,6 +66,7 @@ { atomic_inc(&group->refcnt); } +EXPORT_SYMBOL(fsnotify_get_group); /* * Drop a reference to a group. Free it if it's through. @@ -74,6 +76,7 @@ if (atomic_dec_and_test(&group->refcnt)) fsnotify_final_destroy_group(group); } +EXPORT_SYMBOL(fsnotify_put_group); /* * Create a new fsnotify_group and hold a reference for the group returned. @@ -102,6 +105,7 @@ return group; } +EXPORT_SYMBOL(fsnotify_alloc_group); int fsnotify_fasync(int fd, struct file *file, int on) { diff -Naur kernel.21.3.orig/fs/notify/mark.c kernel.21.3.new/fs/notify/mark.c --- kernel.21.3.orig/fs/notify/mark.c 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/fs/notify/mark.c 2015-04-25 23:28:53.715397586 +0000 @@ -115,6 +115,7 @@ mark->free_mark(mark); } } +EXPORT_SYMBOL(fsnotify_put_mark); /* * Any time a mark is getting freed we end up here. @@ -197,6 +198,7 @@ fsnotify_destroy_mark_locked(mark, group); mutex_unlock(&group->mark_mutex); } +EXPORT_SYMBOL(fsnotify_destroy_mark); void fsnotify_set_mark_mask_locked(struct fsnotify_mark *mark, __u32 mask) { @@ -281,6 +283,7 @@ return ret; } +EXPORT_SYMBOL(fsnotify_add_mark); int fsnotify_add_mark(struct fsnotify_mark *mark, struct fsnotify_group *group, struct inode *inode, struct vfsmount *mnt, int allow_dups) @@ -342,6 +345,7 @@ atomic_set(&mark->refcnt, 1); mark->free_mark = free_mark; } +EXPORT_SYMBOL(fsnotify_init_mark); static int fsnotify_mark_destroy(void *ignored) { diff -Naur kernel.21.3.orig/fs/open.c kernel.21.3.new/fs/open.c --- kernel.21.3.orig/fs/open.c 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/fs/open.c 2015-04-25 23:28:53.715397586 +0000 @@ -61,6 +61,7 @@ mutex_unlock(&dentry->d_inode->i_mutex); return ret; } +EXPORT_SYMBOL(do_truncate); long vfs_truncate(struct path *path, loff_t length) { @@ -278,6 +279,7 @@ sb_end_write(inode->i_sb); return ret; } +EXPORT_SYMBOL(do_fallocate); SYSCALL_DEFINE4(fallocate, int, fd, int, mode, loff_t, offset, loff_t, len) { diff -Naur kernel.21.3.orig/fs/proc/base.c kernel.21.3.new/fs/proc/base.c --- kernel.21.3.orig/fs/proc/base.c 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/fs/proc/base.c 2015-04-25 23:28:45.300481722 +0000 @@ -1871,7 +1871,7 @@ down_read(&mm->mmap_sem); vma = find_exact_vma(mm, vm_start, vm_end); if (vma && vma->vm_file) { - *path = vma->vm_file->f_path; + *path = vma_pr_or_file(vma)->f_path; path_get(path); rc = 0; } diff -Naur kernel.21.3.orig/fs/proc/base.c.orig kernel.21.3.new/fs/proc/base.c.orig --- kernel.21.3.orig/fs/proc/base.c.orig 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/proc/base.c.orig 2015-02-03 01:44:32.000000000 +0000 @@ -0,0 +1,3376 @@ +/* + * linux/fs/proc/base.c + * + * Copyright (C) 1991, 1992 Linus Torvalds + * + * proc base directory handling functions + * + * 1999, Al Viro. Rewritten. Now it covers the whole per-process part. + * Instead of using magical inumbers to determine the kind of object + * we allocate and fill in-core inodes upon lookup. They don't even + * go into icache. We cache the reference to task_struct upon lookup too. + * Eventually it should become a filesystem in its own. We don't use the + * rest of procfs anymore. + * + * + * Changelog: + * 17-Jan-2005 + * Allan Bezerra + * Bruna Moreira + * Edjard Mota + * Ilias Biris + * Mauricio Lin + * + * Embedded Linux Lab - 10LE Instituto Nokia de Tecnologia - INdT + * + * A new process specific entry (smaps) included in /proc. It shows the + * size of rss for each memory area. The maps entry lacks information + * about physical memory size (rss) for each mapped file, i.e., + * rss information for executables and library files. + * This additional information is useful for any tools that need to know + * about physical memory consumption for a process specific library. + * + * Changelog: + * 21-Feb-2005 + * Embedded Linux Lab - 10LE Instituto Nokia de Tecnologia - INdT + * Pud inclusion in the page table walking. + * + * ChangeLog: + * 10-Mar-2005 + * 10LE Instituto Nokia de Tecnologia - INdT: + * A better way to walks through the page table as suggested by Hugh Dickins. + * + * Simo Piiroinen : + * Smaps information related to shared, private, clean and dirty pages. + * + * Paul Mundt : + * Overall revision about smaps. + * + * Copyright (c) 2013, NVIDIA CORPORATION. All rights reserved. + */ + +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#ifdef CONFIG_HARDWALL +#include +#endif +#include +#include "internal.h" +#include "fd.h" + +/* NOTE: + * Implementing inode permission operations in /proc is almost + * certainly an error. Permission checks need to happen during + * each system call not at open time. The reason is that most of + * what we wish to check for permissions in /proc varies at runtime. + * + * The classic example of a problem is opening file descriptors + * in /proc for a task before it execs a suid executable. + */ + +struct pid_entry { + char *name; + int len; + umode_t mode; + const struct inode_operations *iop; + const struct file_operations *fop; + union proc_op op; +}; + +#define NOD(NAME, MODE, IOP, FOP, OP) { \ + .name = (NAME), \ + .len = sizeof(NAME) - 1, \ + .mode = MODE, \ + .iop = IOP, \ + .fop = FOP, \ + .op = OP, \ +} + +#define DIR(NAME, MODE, iops, fops) \ + NOD(NAME, (S_IFDIR|(MODE)), &iops, &fops, {} ) +#define LNK(NAME, get_link) \ + NOD(NAME, (S_IFLNK|S_IRWXUGO), \ + &proc_pid_link_inode_operations, NULL, \ + { .proc_get_link = get_link } ) +#define REG(NAME, MODE, fops) \ + NOD(NAME, (S_IFREG|(MODE)), NULL, &fops, {}) +#define INF(NAME, MODE, read) \ + NOD(NAME, (S_IFREG|(MODE)), \ + NULL, &proc_info_file_operations, \ + { .proc_read = read } ) +#define ONE(NAME, MODE, show) \ + NOD(NAME, (S_IFREG|(MODE)), \ + NULL, &proc_single_file_operations, \ + { .proc_show = show } ) + +/* ANDROID is for special files in /proc. */ +#define ANDROID(NAME, MODE, OTYPE) \ + NOD(NAME, (S_IFREG|(MODE)), \ + &proc_##OTYPE##_inode_operations, \ + &proc_##OTYPE##_operations, {}) + +/* + * Count the number of hardlinks for the pid_entry table, excluding the . + * and .. links. + */ +static unsigned int pid_entry_count_dirs(const struct pid_entry *entries, + unsigned int n) +{ + unsigned int i; + unsigned int count; + + count = 0; + for (i = 0; i < n; ++i) { + if (S_ISDIR(entries[i].mode)) + ++count; + } + + return count; +} + +static int get_task_root(struct task_struct *task, struct path *root) +{ + int result = -ENOENT; + + task_lock(task); + if (task->fs) { + get_fs_root(task->fs, root); + result = 0; + } + task_unlock(task); + return result; +} + +static int proc_cwd_link(struct dentry *dentry, struct path *path) +{ + struct task_struct *task = get_proc_task(dentry->d_inode); + int result = -ENOENT; + + if (task) { + task_lock(task); + if (task->fs) { + get_fs_pwd(task->fs, path); + result = 0; + } + task_unlock(task); + put_task_struct(task); + } + return result; +} + +static int proc_root_link(struct dentry *dentry, struct path *path) +{ + struct task_struct *task = get_proc_task(dentry->d_inode); + int result = -ENOENT; + + if (task) { + result = get_task_root(task, path); + put_task_struct(task); + } + return result; +} + +struct mm_struct *mm_for_maps(struct task_struct *task) +{ + return mm_access(task, PTRACE_MODE_READ); +} + +static int proc_pid_cmdline(struct task_struct *task, char *buffer) +{ + int res = 0; + unsigned int len; + struct mm_struct *mm = get_task_mm(task); + if (!mm) + goto out; + if (!mm->arg_end) + goto out_mm; /* Shh! No looking before we're done */ + + len = mm->arg_end - mm->arg_start; + + if (len > PAGE_SIZE) + len = PAGE_SIZE; + + res = access_process_vm(task, mm->arg_start, buffer, len, 0); + + // If the nul at the end of args has been overwritten, then + // assume application is using setproctitle(3). + if (res > 0 && buffer[res-1] != '\0' && len < PAGE_SIZE) { + len = strnlen(buffer, res); + if (len < res) { + res = len; + } else { + len = mm->env_end - mm->env_start; + if (len > PAGE_SIZE - res) + len = PAGE_SIZE - res; + res += access_process_vm(task, mm->env_start, buffer+res, len, 0); + res = strnlen(buffer, res); + } + } +out_mm: + mmput(mm); +out: + return res; +} + +static int proc_pid_auxv(struct task_struct *task, char *buffer) +{ + struct mm_struct *mm = mm_access(task, PTRACE_MODE_READ); + int res = PTR_ERR(mm); + if (mm && !IS_ERR(mm)) { + unsigned int nwords = 0; + do { + nwords += 2; + } while (mm->saved_auxv[nwords - 2] != 0); /* AT_NULL */ + res = nwords * sizeof(mm->saved_auxv[0]); + if (res > PAGE_SIZE) + res = PAGE_SIZE; + memcpy(buffer, mm->saved_auxv, res); + mmput(mm); + } + return res; +} + + +#ifdef CONFIG_KALLSYMS +/* + * Provides a wchan file via kallsyms in a proper one-value-per-file format. + * Returns the resolved symbol. If that fails, simply return the address. + */ +static int proc_pid_wchan(struct task_struct *task, char *buffer) +{ + unsigned long wchan; + char symname[KSYM_NAME_LEN]; + + wchan = get_wchan(task); + + if (lookup_symbol_name(wchan, symname) < 0) + if (!ptrace_may_access(task, PTRACE_MODE_READ)) + return 0; + else + return sprintf(buffer, "%lu", wchan); + else + return sprintf(buffer, "%s", symname); +} +#endif /* CONFIG_KALLSYMS */ + +static int lock_trace(struct task_struct *task) +{ + int err = mutex_lock_killable(&task->signal->cred_guard_mutex); + if (err) + return err; + if (!ptrace_may_access(task, PTRACE_MODE_ATTACH)) { + mutex_unlock(&task->signal->cred_guard_mutex); + return -EPERM; + } + return 0; +} + +static void unlock_trace(struct task_struct *task) +{ + mutex_unlock(&task->signal->cred_guard_mutex); +} + +#ifdef CONFIG_STACKTRACE + +#define MAX_STACK_TRACE_DEPTH 64 + +static int proc_pid_stack(struct seq_file *m, struct pid_namespace *ns, + struct pid *pid, struct task_struct *task) +{ + struct stack_trace trace; + unsigned long *entries; + int err; + int i; + + entries = kmalloc(MAX_STACK_TRACE_DEPTH * sizeof(*entries), GFP_KERNEL); + if (!entries) + return -ENOMEM; + + trace.nr_entries = 0; + trace.max_entries = MAX_STACK_TRACE_DEPTH; + trace.entries = entries; + trace.skip = 0; + + err = lock_trace(task); + if (!err) { + save_stack_trace_tsk(task, &trace); + + for (i = 0; i < trace.nr_entries; i++) { + seq_printf(m, "[<%pK>] %pS\n", + (void *)entries[i], (void *)entries[i]); + } + unlock_trace(task); + } + kfree(entries); + + return err; +} +#endif + +#ifdef CONFIG_SCHEDSTATS +/* + * Provides /proc/PID/schedstat + */ +static int proc_pid_schedstat(struct task_struct *task, char *buffer) +{ + return sprintf(buffer, "%llu %llu %lu\n", + (unsigned long long)task->se.sum_exec_runtime, + (unsigned long long)task->sched_info.run_delay, + task->sched_info.pcount); +} +#endif + +#ifdef CONFIG_LATENCYTOP +static int lstats_show_proc(struct seq_file *m, void *v) +{ + int i; + struct inode *inode = m->private; + struct task_struct *task = get_proc_task(inode); + + if (!task) + return -ESRCH; + seq_puts(m, "Latency Top version : v0.1\n"); + for (i = 0; i < 32; i++) { + struct latency_record *lr = &task->latency_record[i]; + if (lr->backtrace[0]) { + int q; + seq_printf(m, "%i %li %li", + lr->count, lr->time, lr->max); + for (q = 0; q < LT_BACKTRACEDEPTH; q++) { + unsigned long bt = lr->backtrace[q]; + if (!bt) + break; + if (bt == ULONG_MAX) + break; + seq_printf(m, " %ps", (void *)bt); + } + seq_putc(m, '\n'); + } + + } + put_task_struct(task); + return 0; +} + +static int lstats_open(struct inode *inode, struct file *file) +{ + return single_open(file, lstats_show_proc, inode); +} + +static ssize_t lstats_write(struct file *file, const char __user *buf, + size_t count, loff_t *offs) +{ + struct task_struct *task = get_proc_task(file_inode(file)); + + if (!task) + return -ESRCH; + clear_all_latency_tracing(task); + put_task_struct(task); + + return count; +} + +static const struct file_operations proc_lstats_operations = { + .open = lstats_open, + .read = seq_read, + .write = lstats_write, + .llseek = seq_lseek, + .release = single_release, +}; + +#endif + +#ifdef CONFIG_CGROUPS +static int cgroup_open(struct inode *inode, struct file *file) +{ + struct pid *pid = PROC_I(inode)->pid; + return single_open(file, proc_cgroup_show, pid); +} + +static const struct file_operations proc_cgroup_operations = { + .open = cgroup_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; +#endif + +#ifdef CONFIG_PROC_PID_CPUSET + +static int cpuset_open(struct inode *inode, struct file *file) +{ + struct pid *pid = PROC_I(inode)->pid; + return single_open(file, proc_cpuset_show, pid); +} + +static const struct file_operations proc_cpuset_operations = { + .open = cpuset_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; +#endif + +static int proc_oom_score(struct task_struct *task, char *buffer) +{ + unsigned long totalpages = totalram_pages + total_swap_pages; + unsigned long points = 0; + + read_lock(&tasklist_lock); + if (pid_alive(task)) + points = oom_badness(task, NULL, NULL, totalpages) * + 1000 / totalpages; + read_unlock(&tasklist_lock); + return sprintf(buffer, "%lu\n", points); +} + +struct limit_names { + char *name; + char *unit; +}; + +static const struct limit_names lnames[RLIM_NLIMITS] = { + [RLIMIT_CPU] = {"Max cpu time", "seconds"}, + [RLIMIT_FSIZE] = {"Max file size", "bytes"}, + [RLIMIT_DATA] = {"Max data size", "bytes"}, + [RLIMIT_STACK] = {"Max stack size", "bytes"}, + [RLIMIT_CORE] = {"Max core file size", "bytes"}, + [RLIMIT_RSS] = {"Max resident set", "bytes"}, + [RLIMIT_NPROC] = {"Max processes", "processes"}, + [RLIMIT_NOFILE] = {"Max open files", "files"}, + [RLIMIT_MEMLOCK] = {"Max locked memory", "bytes"}, + [RLIMIT_AS] = {"Max address space", "bytes"}, + [RLIMIT_LOCKS] = {"Max file locks", "locks"}, + [RLIMIT_SIGPENDING] = {"Max pending signals", "signals"}, + [RLIMIT_MSGQUEUE] = {"Max msgqueue size", "bytes"}, + [RLIMIT_NICE] = {"Max nice priority", NULL}, + [RLIMIT_RTPRIO] = {"Max realtime priority", NULL}, + [RLIMIT_RTTIME] = {"Max realtime timeout", "us"}, +}; + +/* Display limits for a process */ +static int proc_pid_limits(struct task_struct *task, char *buffer) +{ + unsigned int i; + int count = 0; + unsigned long flags; + char *bufptr = buffer; + + struct rlimit rlim[RLIM_NLIMITS]; + + if (!lock_task_sighand(task, &flags)) + return 0; + memcpy(rlim, task->signal->rlim, sizeof(struct rlimit) * RLIM_NLIMITS); + unlock_task_sighand(task, &flags); + + /* + * print the file header + */ + count += sprintf(&bufptr[count], "%-25s %-20s %-20s %-10s\n", + "Limit", "Soft Limit", "Hard Limit", "Units"); + + for (i = 0; i < RLIM_NLIMITS; i++) { + if (rlim[i].rlim_cur == RLIM_INFINITY) + count += sprintf(&bufptr[count], "%-25s %-20s ", + lnames[i].name, "unlimited"); + else + count += sprintf(&bufptr[count], "%-25s %-20lu ", + lnames[i].name, rlim[i].rlim_cur); + + if (rlim[i].rlim_max == RLIM_INFINITY) + count += sprintf(&bufptr[count], "%-20s ", "unlimited"); + else + count += sprintf(&bufptr[count], "%-20lu ", + rlim[i].rlim_max); + + if (lnames[i].unit) + count += sprintf(&bufptr[count], "%-10s\n", + lnames[i].unit); + else + count += sprintf(&bufptr[count], "\n"); + } + + return count; +} + +#ifdef CONFIG_HAVE_ARCH_TRACEHOOK +static int proc_pid_syscall(struct task_struct *task, char *buffer) +{ + long nr; + unsigned long args[6], sp, pc; + int res = lock_trace(task); + if (res) + return res; + + if (task_current_syscall(task, &nr, args, 6, &sp, &pc)) + res = sprintf(buffer, "running\n"); + else if (nr < 0) + res = sprintf(buffer, "%ld 0x%lx 0x%lx\n", nr, sp, pc); + else + res = sprintf(buffer, + "%ld 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx\n", + nr, + args[0], args[1], args[2], args[3], args[4], args[5], + sp, pc); + unlock_trace(task); + return res; +} +#endif /* CONFIG_HAVE_ARCH_TRACEHOOK */ + +/************************************************************************/ +/* Here the fs part begins */ +/************************************************************************/ + +/* permission checks */ +static int proc_fd_access_allowed(struct inode *inode) +{ + struct task_struct *task; + int allowed = 0; + /* Allow access to a task's file descriptors if it is us or we + * may use ptrace attach to the process and find out that + * information. + */ + task = get_proc_task(inode); + if (task) { + allowed = ptrace_may_access(task, PTRACE_MODE_READ); + put_task_struct(task); + } + return allowed; +} + +int proc_setattr(struct dentry *dentry, struct iattr *attr) +{ + int error; + struct inode *inode = dentry->d_inode; + + if (attr->ia_valid & ATTR_MODE) + return -EPERM; + + error = inode_change_ok(inode, attr); + if (error) + return error; + + setattr_copy(inode, attr); + mark_inode_dirty(inode); + return 0; +} + +/* + * May current process learn task's sched/cmdline info (for hide_pid_min=1) + * or euid/egid (for hide_pid_min=2)? + */ +static bool has_pid_permissions(struct pid_namespace *pid, + struct task_struct *task, + int hide_pid_min) +{ + if (pid->hide_pid < hide_pid_min) + return true; + if (in_group_p(pid->pid_gid)) + return true; + return ptrace_may_access(task, PTRACE_MODE_READ); +} + + +static int proc_pid_permission(struct inode *inode, int mask) +{ + struct pid_namespace *pid = inode->i_sb->s_fs_info; + struct task_struct *task; + bool has_perms; + + task = get_proc_task(inode); + if (!task) + return -ESRCH; + has_perms = has_pid_permissions(pid, task, 1); + put_task_struct(task); + + if (!has_perms) { + if (pid->hide_pid == 2) { + /* + * Let's make getdents(), stat(), and open() + * consistent with each other. If a process + * may not stat() a file, it shouldn't be seen + * in procfs at all. + */ + return -ENOENT; + } + + return -EPERM; + } + return generic_permission(inode, mask); +} + + + +static const struct inode_operations proc_def_inode_operations = { + .setattr = proc_setattr, +}; + +#define PROC_BLOCK_SIZE (3*1024) /* 4K page size but our output routines use some slack for overruns */ + +static ssize_t proc_info_read(struct file * file, char __user * buf, + size_t count, loff_t *ppos) +{ + struct inode * inode = file_inode(file); + unsigned long page; + ssize_t length; + struct task_struct *task = get_proc_task(inode); + + length = -ESRCH; + if (!task) + goto out_no_task; + + if (count > PROC_BLOCK_SIZE) + count = PROC_BLOCK_SIZE; + + length = -ENOMEM; + if (!(page = __get_free_page(GFP_TEMPORARY))) + goto out; + + length = PROC_I(inode)->op.proc_read(task, (char*)page); + + if (length >= 0) + length = simple_read_from_buffer(buf, count, ppos, (char *)page, length); + free_page(page); +out: + put_task_struct(task); +out_no_task: + return length; +} + +static const struct file_operations proc_info_file_operations = { + .read = proc_info_read, + .llseek = generic_file_llseek, +}; + +static int proc_single_show(struct seq_file *m, void *v) +{ + struct inode *inode = m->private; + struct pid_namespace *ns; + struct pid *pid; + struct task_struct *task; + int ret; + + ns = inode->i_sb->s_fs_info; + pid = proc_pid(inode); + task = get_pid_task(pid, PIDTYPE_PID); + if (!task) + return -ESRCH; + + ret = PROC_I(inode)->op.proc_show(m, ns, pid, task); + + put_task_struct(task); + return ret; +} + +static int proc_single_open(struct inode *inode, struct file *filp) +{ + return single_open(filp, proc_single_show, inode); +} + +static const struct file_operations proc_single_file_operations = { + .open = proc_single_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static int __mem_open(struct inode *inode, struct file *file, unsigned int mode) +{ + struct task_struct *task = get_proc_task(file_inode(file)); + struct mm_struct *mm; + + if (!task) + return -ESRCH; + + mm = mm_access(task, mode); + put_task_struct(task); + + if (IS_ERR(mm)) + return PTR_ERR(mm); + + if (mm) { + /* ensure this mm_struct can't be freed */ + atomic_inc(&mm->mm_count); + /* but do not pin its memory */ + mmput(mm); + } + + file->private_data = mm; + + return 0; +} + +static int mem_open(struct inode *inode, struct file *file) +{ + int ret = __mem_open(inode, file, PTRACE_MODE_ATTACH); + + /* OK to pass negative loff_t, we can catch out-of-range */ + file->f_mode |= FMODE_UNSIGNED_OFFSET; + + return ret; +} + +static ssize_t mem_rw(struct file *file, char __user *buf, + size_t count, loff_t *ppos, int write) +{ + struct mm_struct *mm = file->private_data; + unsigned long addr = *ppos; + ssize_t copied; + char *page; + + if (!mm) + return 0; + + page = (char *)__get_free_page(GFP_TEMPORARY); + if (!page) + return -ENOMEM; + + copied = 0; + if (!atomic_inc_not_zero(&mm->mm_users)) + goto free; + + while (count > 0) { + int this_len = min_t(int, count, PAGE_SIZE); + + if (write && copy_from_user(page, buf, this_len)) { + copied = -EFAULT; + break; + } + + this_len = access_remote_vm(mm, addr, page, this_len, write); + if (!this_len) { + if (!copied) + copied = -EIO; + break; + } + + if (!write && copy_to_user(buf, page, this_len)) { + copied = -EFAULT; + break; + } + + buf += this_len; + addr += this_len; + copied += this_len; + count -= this_len; + } + *ppos = addr; + + mmput(mm); +free: + free_page((unsigned long) page); + return copied; +} + +static ssize_t mem_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) +{ + return mem_rw(file, buf, count, ppos, 0); +} + +static ssize_t mem_write(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) +{ + return mem_rw(file, (char __user*)buf, count, ppos, 1); +} + +loff_t mem_lseek(struct file *file, loff_t offset, int orig) +{ + switch (orig) { + case 0: + file->f_pos = offset; + break; + case 1: + file->f_pos += offset; + break; + default: + return -EINVAL; + } + force_successful_syscall_return(); + return file->f_pos; +} + +static int mem_release(struct inode *inode, struct file *file) +{ + struct mm_struct *mm = file->private_data; + if (mm) + mmdrop(mm); + return 0; +} + +static const struct file_operations proc_mem_operations = { + .llseek = mem_lseek, + .read = mem_read, + .write = mem_write, + .open = mem_open, + .release = mem_release, +}; + +static int environ_open(struct inode *inode, struct file *file) +{ + return __mem_open(inode, file, PTRACE_MODE_READ); +} + +static ssize_t environ_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) +{ + char *page; + unsigned long src = *ppos; + int ret = 0; + struct mm_struct *mm = file->private_data; + + if (!mm) + return 0; + + page = (char *)__get_free_page(GFP_TEMPORARY); + if (!page) + return -ENOMEM; + + ret = 0; + if (!atomic_inc_not_zero(&mm->mm_users)) + goto free; + while (count > 0) { + size_t this_len, max_len; + int retval; + + if (src >= (mm->env_end - mm->env_start)) + break; + + this_len = mm->env_end - (mm->env_start + src); + + max_len = min_t(size_t, PAGE_SIZE, count); + this_len = min(max_len, this_len); + + retval = access_remote_vm(mm, (mm->env_start + src), + page, this_len, 0); + + if (retval <= 0) { + ret = retval; + break; + } + + if (copy_to_user(buf, page, retval)) { + ret = -EFAULT; + break; + } + + ret += retval; + src += retval; + buf += retval; + count -= retval; + } + *ppos = src; + mmput(mm); + +free: + free_page((unsigned long) page); + return ret; +} + +static const struct file_operations proc_environ_operations = { + .open = environ_open, + .read = environ_read, + .llseek = generic_file_llseek, + .release = mem_release, +}; + +static ssize_t oom_adj_read(struct file *file, char __user *buf, size_t count, + loff_t *ppos) +{ + struct task_struct *task = get_proc_task(file_inode(file)); + char buffer[PROC_NUMBUF]; + int oom_adj = OOM_ADJUST_MIN; + size_t len; + unsigned long flags; + + if (!task) + return -ESRCH; + if (lock_task_sighand(task, &flags)) { + if (task->signal->oom_score_adj == OOM_SCORE_ADJ_MAX) + oom_adj = OOM_ADJUST_MAX; + else + oom_adj = (task->signal->oom_score_adj * -OOM_DISABLE) / + OOM_SCORE_ADJ_MAX; + unlock_task_sighand(task, &flags); + } + put_task_struct(task); + len = snprintf(buffer, sizeof(buffer), "%d\n", oom_adj); + return simple_read_from_buffer(buf, count, ppos, buffer, len); +} + +static ssize_t oom_adj_write(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) +{ + struct task_struct *task; + char buffer[PROC_NUMBUF]; + int oom_adj; + unsigned long flags; + int err; + + memset(buffer, 0, sizeof(buffer)); + if (count > sizeof(buffer) - 1) + count = sizeof(buffer) - 1; + if (copy_from_user(buffer, buf, count)) { + err = -EFAULT; + goto out; + } + + err = kstrtoint(strstrip(buffer), 0, &oom_adj); + if (err) + goto out; + if ((oom_adj < OOM_ADJUST_MIN || oom_adj > OOM_ADJUST_MAX) && + oom_adj != OOM_DISABLE) { + err = -EINVAL; + goto out; + } + + task = get_proc_task(file_inode(file)); + if (!task) { + err = -ESRCH; + goto out; + } + + task_lock(task); + if (!task->mm) { + err = -EINVAL; + goto err_task_lock; + } + + if (!lock_task_sighand(task, &flags)) { + err = -ESRCH; + goto err_task_lock; + } + + /* + * Scale /proc/pid/oom_score_adj appropriately ensuring that a maximum + * value is always attainable. + */ + if (oom_adj == OOM_ADJUST_MAX) + oom_adj = OOM_SCORE_ADJ_MAX; + else + oom_adj = (oom_adj * OOM_SCORE_ADJ_MAX) / -OOM_DISABLE; + + if (oom_adj < task->signal->oom_score_adj && + !capable(CAP_SYS_RESOURCE)) { + err = -EACCES; + goto err_sighand; + } + + /* + * /proc/pid/oom_adj is provided for legacy purposes, ask users to use + * /proc/pid/oom_score_adj instead. + */ + pr_warn_once("%s (%d): /proc/%d/oom_adj is deprecated, please use /proc/%d/oom_score_adj instead.\n", + current->comm, task_pid_nr(current), task_pid_nr(task), + task_pid_nr(task)); + + task->signal->oom_score_adj = oom_adj; + trace_oom_score_adj_update(task); +err_sighand: + unlock_task_sighand(task, &flags); +err_task_lock: + task_unlock(task); + put_task_struct(task); +out: + return err < 0 ? err : count; +} + +static int oom_adjust_permission(struct inode *inode, int mask) +{ + uid_t uid; + struct task_struct *p; + + p = get_proc_task(inode); + if(p) { + uid = task_uid(p); + put_task_struct(p); + } + + /* + * System Server (uid == 1000) is granted access to oom_adj of all + * android applications (uid > 10000) as and services (uid >= 1000) + */ + if (p && (current_fsuid() == 1000) && (uid >= 1000)) { + if (inode->i_mode >> 6 & mask) { + return 0; + } + } + + /* Fall back to default. */ + return generic_permission(inode, mask); +} + +static const struct inode_operations proc_oom_adj_inode_operations = { + .permission = oom_adjust_permission, +}; + +static const struct file_operations proc_oom_adj_operations = { + .read = oom_adj_read, + .write = oom_adj_write, + .llseek = generic_file_llseek, +}; + +static ssize_t oom_score_adj_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) +{ + struct task_struct *task = get_proc_task(file_inode(file)); + char buffer[PROC_NUMBUF]; + short oom_score_adj = OOM_SCORE_ADJ_MIN; + unsigned long flags; + size_t len; + + if (!task) + return -ESRCH; + if (lock_task_sighand(task, &flags)) { + oom_score_adj = task->signal->oom_score_adj; + unlock_task_sighand(task, &flags); + } + put_task_struct(task); + len = snprintf(buffer, sizeof(buffer), "%hd\n", oom_score_adj); + return simple_read_from_buffer(buf, count, ppos, buffer, len); +} + +static ssize_t oom_score_adj_write(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) +{ + struct task_struct *task; + char buffer[PROC_NUMBUF]; + unsigned long flags; + int oom_score_adj; + int err; + + memset(buffer, 0, sizeof(buffer)); + if (count > sizeof(buffer) - 1) + count = sizeof(buffer) - 1; + if (copy_from_user(buffer, buf, count)) { + err = -EFAULT; + goto out; + } + + err = kstrtoint(strstrip(buffer), 0, &oom_score_adj); + if (err) + goto out; + if (oom_score_adj < OOM_SCORE_ADJ_MIN || + oom_score_adj > OOM_SCORE_ADJ_MAX) { + err = -EINVAL; + goto out; + } + + task = get_proc_task(file_inode(file)); + if (!task) { + err = -ESRCH; + goto out; + } + + task_lock(task); + if (!task->mm) { + err = -EINVAL; + goto err_task_lock; + } + + if (!lock_task_sighand(task, &flags)) { + err = -ESRCH; + goto err_task_lock; + } + + if ((short)oom_score_adj < task->signal->oom_score_adj_min && + !capable(CAP_SYS_RESOURCE)) { + err = -EACCES; + goto err_sighand; + } + + task->signal->oom_score_adj = (short)oom_score_adj; + if (has_capability_noaudit(current, CAP_SYS_RESOURCE)) + task->signal->oom_score_adj_min = (short)oom_score_adj; + trace_oom_score_adj_update(task); + +err_sighand: + unlock_task_sighand(task, &flags); +err_task_lock: + task_unlock(task); + put_task_struct(task); +out: + return err < 0 ? err : count; +} + +static const struct file_operations proc_oom_score_adj_operations = { + .read = oom_score_adj_read, + .write = oom_score_adj_write, + .llseek = default_llseek, +}; + +#ifdef CONFIG_AUDITSYSCALL +#define TMPBUFLEN 21 +static ssize_t proc_loginuid_read(struct file * file, char __user * buf, + size_t count, loff_t *ppos) +{ + struct inode * inode = file_inode(file); + struct task_struct *task = get_proc_task(inode); + ssize_t length; + char tmpbuf[TMPBUFLEN]; + + if (!task) + return -ESRCH; + length = scnprintf(tmpbuf, TMPBUFLEN, "%u", + from_kuid(file->f_cred->user_ns, + audit_get_loginuid(task))); + put_task_struct(task); + return simple_read_from_buffer(buf, count, ppos, tmpbuf, length); +} + +static ssize_t proc_loginuid_write(struct file * file, const char __user * buf, + size_t count, loff_t *ppos) +{ + struct inode * inode = file_inode(file); + char *page, *tmp; + ssize_t length; + uid_t loginuid; + kuid_t kloginuid; + + rcu_read_lock(); + if (current != pid_task(proc_pid(inode), PIDTYPE_PID)) { + rcu_read_unlock(); + return -EPERM; + } + rcu_read_unlock(); + + if (count >= PAGE_SIZE) + count = PAGE_SIZE - 1; + + if (*ppos != 0) { + /* No partial writes. */ + return -EINVAL; + } + page = (char*)__get_free_page(GFP_TEMPORARY); + if (!page) + return -ENOMEM; + length = -EFAULT; + if (copy_from_user(page, buf, count)) + goto out_free_page; + + page[count] = '\0'; + loginuid = simple_strtoul(page, &tmp, 10); + if (tmp == page) { + length = -EINVAL; + goto out_free_page; + + } + kloginuid = make_kuid(file->f_cred->user_ns, loginuid); + if (!uid_valid(kloginuid)) { + length = -EINVAL; + goto out_free_page; + } + + length = audit_set_loginuid(kloginuid); + if (likely(length == 0)) + length = count; + +out_free_page: + free_page((unsigned long) page); + return length; +} + +static const struct file_operations proc_loginuid_operations = { + .read = proc_loginuid_read, + .write = proc_loginuid_write, + .llseek = generic_file_llseek, +}; + +static ssize_t proc_sessionid_read(struct file * file, char __user * buf, + size_t count, loff_t *ppos) +{ + struct inode * inode = file_inode(file); + struct task_struct *task = get_proc_task(inode); + ssize_t length; + char tmpbuf[TMPBUFLEN]; + + if (!task) + return -ESRCH; + length = scnprintf(tmpbuf, TMPBUFLEN, "%u", + audit_get_sessionid(task)); + put_task_struct(task); + return simple_read_from_buffer(buf, count, ppos, tmpbuf, length); +} + +static const struct file_operations proc_sessionid_operations = { + .read = proc_sessionid_read, + .llseek = generic_file_llseek, +}; +#endif + +#ifdef CONFIG_FAULT_INJECTION +static ssize_t proc_fault_inject_read(struct file * file, char __user * buf, + size_t count, loff_t *ppos) +{ + struct task_struct *task = get_proc_task(file_inode(file)); + char buffer[PROC_NUMBUF]; + size_t len; + int make_it_fail; + + if (!task) + return -ESRCH; + make_it_fail = task->make_it_fail; + put_task_struct(task); + + len = snprintf(buffer, sizeof(buffer), "%i\n", make_it_fail); + + return simple_read_from_buffer(buf, count, ppos, buffer, len); +} + +static ssize_t proc_fault_inject_write(struct file * file, + const char __user * buf, size_t count, loff_t *ppos) +{ + struct task_struct *task; + char buffer[PROC_NUMBUF], *end; + int make_it_fail; + + if (!capable(CAP_SYS_RESOURCE)) + return -EPERM; + memset(buffer, 0, sizeof(buffer)); + if (count > sizeof(buffer) - 1) + count = sizeof(buffer) - 1; + if (copy_from_user(buffer, buf, count)) + return -EFAULT; + make_it_fail = simple_strtol(strstrip(buffer), &end, 0); + if (*end) + return -EINVAL; + task = get_proc_task(file_inode(file)); + if (!task) + return -ESRCH; + task->make_it_fail = make_it_fail; + put_task_struct(task); + + return count; +} + +static const struct file_operations proc_fault_inject_operations = { + .read = proc_fault_inject_read, + .write = proc_fault_inject_write, + .llseek = generic_file_llseek, +}; +#endif + + +#ifdef CONFIG_SCHED_DEBUG +/* + * Print out various scheduling related per-task fields: + */ +static int sched_show(struct seq_file *m, void *v) +{ + struct inode *inode = m->private; + struct task_struct *p; + + p = get_proc_task(inode); + if (!p) + return -ESRCH; + proc_sched_show_task(p, m); + + put_task_struct(p); + + return 0; +} + +static ssize_t +sched_write(struct file *file, const char __user *buf, + size_t count, loff_t *offset) +{ + struct inode *inode = file_inode(file); + struct task_struct *p; + + p = get_proc_task(inode); + if (!p) + return -ESRCH; + proc_sched_set_task(p); + + put_task_struct(p); + + return count; +} + +static int sched_open(struct inode *inode, struct file *filp) +{ + return single_open(filp, sched_show, inode); +} + +static const struct file_operations proc_pid_sched_operations = { + .open = sched_open, + .read = seq_read, + .write = sched_write, + .llseek = seq_lseek, + .release = single_release, +}; + +#endif + +#ifdef CONFIG_SCHED_AUTOGROUP +/* + * Print out autogroup related information: + */ +static int sched_autogroup_show(struct seq_file *m, void *v) +{ + struct inode *inode = m->private; + struct task_struct *p; + + p = get_proc_task(inode); + if (!p) + return -ESRCH; + proc_sched_autogroup_show_task(p, m); + + put_task_struct(p); + + return 0; +} + +static ssize_t +sched_autogroup_write(struct file *file, const char __user *buf, + size_t count, loff_t *offset) +{ + struct inode *inode = file_inode(file); + struct task_struct *p; + char buffer[PROC_NUMBUF]; + int nice; + int err; + + memset(buffer, 0, sizeof(buffer)); + if (count > sizeof(buffer) - 1) + count = sizeof(buffer) - 1; + if (copy_from_user(buffer, buf, count)) + return -EFAULT; + + err = kstrtoint(strstrip(buffer), 0, &nice); + if (err < 0) + return err; + + p = get_proc_task(inode); + if (!p) + return -ESRCH; + + err = proc_sched_autogroup_set_nice(p, nice); + if (err) + count = err; + + put_task_struct(p); + + return count; +} + +static int sched_autogroup_open(struct inode *inode, struct file *filp) +{ + int ret; + + ret = single_open(filp, sched_autogroup_show, NULL); + if (!ret) { + struct seq_file *m = filp->private_data; + + m->private = inode; + } + return ret; +} + +static const struct file_operations proc_pid_sched_autogroup_operations = { + .open = sched_autogroup_open, + .read = seq_read, + .write = sched_autogroup_write, + .llseek = seq_lseek, + .release = single_release, +}; + +#endif /* CONFIG_SCHED_AUTOGROUP */ + +static ssize_t comm_write(struct file *file, const char __user *buf, + size_t count, loff_t *offset) +{ + struct inode *inode = file_inode(file); + struct task_struct *p; + char buffer[TASK_COMM_LEN]; + const size_t maxlen = sizeof(buffer) - 1; + + memset(buffer, 0, sizeof(buffer)); + if (copy_from_user(buffer, buf, count > maxlen ? maxlen : count)) + return -EFAULT; + + p = get_proc_task(inode); + if (!p) + return -ESRCH; + + if (same_thread_group(current, p)) + set_task_comm(p, buffer); + else + count = -EINVAL; + + put_task_struct(p); + + return count; +} + +static int comm_show(struct seq_file *m, void *v) +{ + struct inode *inode = m->private; + struct task_struct *p; + + p = get_proc_task(inode); + if (!p) + return -ESRCH; + + task_lock(p); + seq_printf(m, "%s\n", p->comm); + task_unlock(p); + + put_task_struct(p); + + return 0; +} + +static int comm_open(struct inode *inode, struct file *filp) +{ + return single_open(filp, comm_show, inode); +} + +static const struct file_operations proc_pid_set_comm_operations = { + .open = comm_open, + .read = seq_read, + .write = comm_write, + .llseek = seq_lseek, + .release = single_release, +}; + +static int proc_exe_link(struct dentry *dentry, struct path *exe_path) +{ + struct task_struct *task; + struct mm_struct *mm; + struct file *exe_file; + + task = get_proc_task(dentry->d_inode); + if (!task) + return -ENOENT; + mm = get_task_mm(task); + put_task_struct(task); + if (!mm) + return -ENOENT; + exe_file = get_mm_exe_file(mm); + mmput(mm); + if (exe_file) { + *exe_path = exe_file->f_path; + path_get(&exe_file->f_path); + fput(exe_file); + return 0; + } else + return -ENOENT; +} + +static void *proc_pid_follow_link(struct dentry *dentry, struct nameidata *nd) +{ + struct inode *inode = dentry->d_inode; + struct path path; + int error = -EACCES; + + /* Are we allowed to snoop on the tasks file descriptors? */ + if (!proc_fd_access_allowed(inode)) + goto out; + + error = PROC_I(inode)->op.proc_get_link(dentry, &path); + if (error) + goto out; + + nd_jump_link(nd, &path); + return NULL; +out: + return ERR_PTR(error); +} + +static int do_proc_readlink(struct path *path, char __user *buffer, int buflen) +{ + char *tmp = (char*)__get_free_page(GFP_TEMPORARY); + char *pathname; + int len; + + if (!tmp) + return -ENOMEM; + + pathname = d_path(path, tmp, PAGE_SIZE); + len = PTR_ERR(pathname); + if (IS_ERR(pathname)) + goto out; + len = tmp + PAGE_SIZE - 1 - pathname; + + if (len > buflen) + len = buflen; + if (copy_to_user(buffer, pathname, len)) + len = -EFAULT; + out: + free_page((unsigned long)tmp); + return len; +} + +static int proc_pid_readlink(struct dentry * dentry, char __user * buffer, int buflen) +{ + int error = -EACCES; + struct inode *inode = dentry->d_inode; + struct path path; + + /* Are we allowed to snoop on the tasks file descriptors? */ + if (!proc_fd_access_allowed(inode)) + goto out; + + error = PROC_I(inode)->op.proc_get_link(dentry, &path); + if (error) + goto out; + + error = do_proc_readlink(&path, buffer, buflen); + path_put(&path); +out: + return error; +} + +const struct inode_operations proc_pid_link_inode_operations = { + .readlink = proc_pid_readlink, + .follow_link = proc_pid_follow_link, + .setattr = proc_setattr, +}; + + +/* building an inode */ + +struct inode *proc_pid_make_inode(struct super_block * sb, struct task_struct *task) +{ + struct inode * inode; + struct proc_inode *ei; + const struct cred *cred; + + /* We need a new inode */ + + inode = new_inode(sb); + if (!inode) + goto out; + + /* Common stuff */ + ei = PROC_I(inode); + inode->i_ino = get_next_ino(); + inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; + inode->i_op = &proc_def_inode_operations; + + /* + * grab the reference to task. + */ + ei->pid = get_task_pid(task, PIDTYPE_PID); + if (!ei->pid) + goto out_unlock; + + if (task_dumpable(task)) { + rcu_read_lock(); + cred = __task_cred(task); + inode->i_uid = cred->euid; + inode->i_gid = cred->egid; + rcu_read_unlock(); + } + security_task_to_inode(task, inode); + +out: + return inode; + +out_unlock: + iput(inode); + return NULL; +} + +int pid_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) +{ + struct inode *inode = dentry->d_inode; + struct task_struct *task; + const struct cred *cred; + struct pid_namespace *pid = dentry->d_sb->s_fs_info; + + generic_fillattr(inode, stat); + + rcu_read_lock(); + stat->uid = GLOBAL_ROOT_UID; + stat->gid = GLOBAL_ROOT_GID; + task = pid_task(proc_pid(inode), PIDTYPE_PID); + if (task) { + if (!has_pid_permissions(pid, task, 2)) { + rcu_read_unlock(); + /* + * This doesn't prevent learning whether PID exists, + * it only makes getattr() consistent with readdir(). + */ + return -ENOENT; + } + if ((inode->i_mode == (S_IFDIR|S_IRUGO|S_IXUGO)) || + task_dumpable(task)) { + cred = __task_cred(task); + stat->uid = cred->euid; + stat->gid = cred->egid; + } + } + rcu_read_unlock(); + return 0; +} + +/* dentry stuff */ + +/* + * Exceptional case: normally we are not allowed to unhash a busy + * directory. In this case, however, we can do it - no aliasing problems + * due to the way we treat inodes. + * + * Rewrite the inode's ownerships here because the owning task may have + * performed a setuid(), etc. + * + * Before the /proc/pid/status file was created the only way to read + * the effective uid of a /process was to stat /proc/pid. Reading + * /proc/pid/status is slow enough that procps and other packages + * kept stating /proc/pid. To keep the rules in /proc simple I have + * made this apply to all per process world readable and executable + * directories. + */ +int pid_revalidate(struct dentry *dentry, unsigned int flags) +{ + struct inode *inode; + struct task_struct *task; + const struct cred *cred; + + if (flags & LOOKUP_RCU) + return -ECHILD; + + inode = dentry->d_inode; + task = get_proc_task(inode); + + if (task) { + if ((inode->i_mode == (S_IFDIR|S_IRUGO|S_IXUGO)) || + task_dumpable(task)) { + rcu_read_lock(); + cred = __task_cred(task); + inode->i_uid = cred->euid; + inode->i_gid = cred->egid; + rcu_read_unlock(); + } else { + inode->i_uid = GLOBAL_ROOT_UID; + inode->i_gid = GLOBAL_ROOT_GID; + } + inode->i_mode &= ~(S_ISUID | S_ISGID); + security_task_to_inode(task, inode); + put_task_struct(task); + return 1; + } + d_drop(dentry); + return 0; +} + +int pid_delete_dentry(const struct dentry *dentry) +{ + /* Is the task we represent dead? + * If so, then don't put the dentry on the lru list, + * kill it immediately. + */ + return !proc_pid(dentry->d_inode)->tasks[PIDTYPE_PID].first; +} + +const struct dentry_operations pid_dentry_operations = +{ + .d_revalidate = pid_revalidate, + .d_delete = pid_delete_dentry, +}; + +/* Lookups */ + +/* + * Fill a directory entry. + * + * If possible create the dcache entry and derive our inode number and + * file type from dcache entry. + * + * Since all of the proc inode numbers are dynamically generated, the inode + * numbers do not exist until the inode is cache. This means creating the + * the dcache entry in readdir is necessary to keep the inode numbers + * reported by readdir in sync with the inode numbers reported + * by stat. + */ +int proc_fill_cache(struct file *filp, void *dirent, filldir_t filldir, + const char *name, int len, + instantiate_t instantiate, struct task_struct *task, const void *ptr) +{ + struct dentry *child, *dir = filp->f_path.dentry; + struct inode *inode; + struct qstr qname; + ino_t ino = 0; + unsigned type = DT_UNKNOWN; + + qname.name = name; + qname.len = len; + qname.hash = full_name_hash(name, len); + + child = d_lookup(dir, &qname); + if (!child) { + struct dentry *new; + new = d_alloc(dir, &qname); + if (new) { + child = instantiate(dir->d_inode, new, task, ptr); + if (child) + dput(new); + else + child = new; + } + } + if (!child || IS_ERR(child) || !child->d_inode) + goto end_instantiate; + inode = child->d_inode; + if (inode) { + ino = inode->i_ino; + type = inode->i_mode >> 12; + } + dput(child); +end_instantiate: + if (!ino) + ino = find_inode_number(dir, &qname); + if (!ino) + ino = 1; + return filldir(dirent, name, len, filp->f_pos, ino, type); +} + +#ifdef CONFIG_CHECKPOINT_RESTORE + +/* + * dname_to_vma_addr - maps a dentry name into two unsigned longs + * which represent vma start and end addresses. + */ +static int dname_to_vma_addr(struct dentry *dentry, + unsigned long *start, unsigned long *end) +{ + if (sscanf(dentry->d_name.name, "%lx-%lx", start, end) != 2) + return -EINVAL; + + return 0; +} + +static int map_files_d_revalidate(struct dentry *dentry, unsigned int flags) +{ + unsigned long vm_start, vm_end; + bool exact_vma_exists = false; + struct mm_struct *mm = NULL; + struct task_struct *task; + const struct cred *cred; + struct inode *inode; + int status = 0; + + if (flags & LOOKUP_RCU) + return -ECHILD; + + if (!capable(CAP_SYS_ADMIN)) { + status = -EPERM; + goto out_notask; + } + + inode = dentry->d_inode; + task = get_proc_task(inode); + if (!task) + goto out_notask; + + mm = mm_access(task, PTRACE_MODE_READ); + if (IS_ERR_OR_NULL(mm)) + goto out; + + if (!dname_to_vma_addr(dentry, &vm_start, &vm_end)) { + down_read(&mm->mmap_sem); + exact_vma_exists = !!find_exact_vma(mm, vm_start, vm_end); + up_read(&mm->mmap_sem); + } + + mmput(mm); + + if (exact_vma_exists) { + if (task_dumpable(task)) { + rcu_read_lock(); + cred = __task_cred(task); + inode->i_uid = cred->euid; + inode->i_gid = cred->egid; + rcu_read_unlock(); + } else { + inode->i_uid = GLOBAL_ROOT_UID; + inode->i_gid = GLOBAL_ROOT_GID; + } + security_task_to_inode(task, inode); + status = 1; + } + +out: + put_task_struct(task); + +out_notask: + if (status <= 0) + d_drop(dentry); + + return status; +} + +static const struct dentry_operations tid_map_files_dentry_operations = { + .d_revalidate = map_files_d_revalidate, + .d_delete = pid_delete_dentry, +}; + +static int proc_map_files_get_link(struct dentry *dentry, struct path *path) +{ + unsigned long vm_start, vm_end; + struct vm_area_struct *vma; + struct task_struct *task; + struct mm_struct *mm; + int rc; + + rc = -ENOENT; + task = get_proc_task(dentry->d_inode); + if (!task) + goto out; + + mm = get_task_mm(task); + put_task_struct(task); + if (!mm) + goto out; + + rc = dname_to_vma_addr(dentry, &vm_start, &vm_end); + if (rc) + goto out_mmput; + + rc = -ENOENT; + down_read(&mm->mmap_sem); + vma = find_exact_vma(mm, vm_start, vm_end); + if (vma && vma->vm_file) { + *path = vma->vm_file->f_path; + path_get(path); + rc = 0; + } + up_read(&mm->mmap_sem); + +out_mmput: + mmput(mm); +out: + return rc; +} + +struct map_files_info { + fmode_t mode; + unsigned long len; + unsigned char name[4*sizeof(long)+2]; /* max: %lx-%lx\0 */ +}; + +static struct dentry * +proc_map_files_instantiate(struct inode *dir, struct dentry *dentry, + struct task_struct *task, const void *ptr) +{ + fmode_t mode = (fmode_t)(unsigned long)ptr; + struct proc_inode *ei; + struct inode *inode; + + inode = proc_pid_make_inode(dir->i_sb, task); + if (!inode) + return ERR_PTR(-ENOENT); + + ei = PROC_I(inode); + ei->op.proc_get_link = proc_map_files_get_link; + + inode->i_op = &proc_pid_link_inode_operations; + inode->i_size = 64; + inode->i_mode = S_IFLNK; + + if (mode & FMODE_READ) + inode->i_mode |= S_IRUSR; + if (mode & FMODE_WRITE) + inode->i_mode |= S_IWUSR; + + d_set_d_op(dentry, &tid_map_files_dentry_operations); + d_add(dentry, inode); + + return NULL; +} + +static struct dentry *proc_map_files_lookup(struct inode *dir, + struct dentry *dentry, unsigned int flags) +{ + unsigned long vm_start, vm_end; + struct vm_area_struct *vma; + struct task_struct *task; + struct dentry *result; + struct mm_struct *mm; + + result = ERR_PTR(-EPERM); + if (!capable(CAP_SYS_ADMIN)) + goto out; + + result = ERR_PTR(-ENOENT); + task = get_proc_task(dir); + if (!task) + goto out; + + result = ERR_PTR(-EACCES); + if (!ptrace_may_access(task, PTRACE_MODE_READ)) + goto out_put_task; + + result = ERR_PTR(-ENOENT); + if (dname_to_vma_addr(dentry, &vm_start, &vm_end)) + goto out_put_task; + + mm = get_task_mm(task); + if (!mm) + goto out_put_task; + + down_read(&mm->mmap_sem); + vma = find_exact_vma(mm, vm_start, vm_end); + if (!vma) + goto out_no_vma; + + if (vma->vm_file) + result = proc_map_files_instantiate(dir, dentry, task, + (void *)(unsigned long)vma->vm_file->f_mode); + +out_no_vma: + up_read(&mm->mmap_sem); + mmput(mm); +out_put_task: + put_task_struct(task); +out: + return result; +} + +static const struct inode_operations proc_map_files_inode_operations = { + .lookup = proc_map_files_lookup, + .permission = proc_fd_permission, + .setattr = proc_setattr, +}; + +static int +proc_map_files_readdir(struct file *filp, void *dirent, filldir_t filldir) +{ + struct dentry *dentry = filp->f_path.dentry; + struct inode *inode = dentry->d_inode; + struct vm_area_struct *vma; + struct task_struct *task; + struct mm_struct *mm; + ino_t ino; + int ret; + + ret = -EPERM; + if (!capable(CAP_SYS_ADMIN)) + goto out; + + ret = -ENOENT; + task = get_proc_task(inode); + if (!task) + goto out; + + ret = -EACCES; + if (!ptrace_may_access(task, PTRACE_MODE_READ)) + goto out_put_task; + + ret = 0; + switch (filp->f_pos) { + case 0: + ino = inode->i_ino; + if (filldir(dirent, ".", 1, 0, ino, DT_DIR) < 0) + goto out_put_task; + filp->f_pos++; + case 1: + ino = parent_ino(dentry); + if (filldir(dirent, "..", 2, 1, ino, DT_DIR) < 0) + goto out_put_task; + filp->f_pos++; + default: + { + unsigned long nr_files, pos, i; + struct flex_array *fa = NULL; + struct map_files_info info; + struct map_files_info *p; + + mm = get_task_mm(task); + if (!mm) + goto out_put_task; + down_read(&mm->mmap_sem); + + nr_files = 0; + + /* + * We need two passes here: + * + * 1) Collect vmas of mapped files with mmap_sem taken + * 2) Release mmap_sem and instantiate entries + * + * otherwise we get lockdep complained, since filldir() + * routine might require mmap_sem taken in might_fault(). + */ + + for (vma = mm->mmap, pos = 2; vma; vma = vma->vm_next) { + if (vma->vm_file && ++pos > filp->f_pos) + nr_files++; + } + + if (nr_files) { + fa = flex_array_alloc(sizeof(info), nr_files, + GFP_KERNEL); + if (!fa || flex_array_prealloc(fa, 0, nr_files, + GFP_KERNEL)) { + ret = -ENOMEM; + if (fa) + flex_array_free(fa); + up_read(&mm->mmap_sem); + mmput(mm); + goto out_put_task; + } + for (i = 0, vma = mm->mmap, pos = 2; vma; + vma = vma->vm_next) { + if (!vma->vm_file) + continue; + if (++pos <= filp->f_pos) + continue; + + info.mode = vma->vm_file->f_mode; + info.len = snprintf(info.name, + sizeof(info.name), "%lx-%lx", + vma->vm_start, vma->vm_end); + if (flex_array_put(fa, i++, &info, GFP_KERNEL)) + BUG(); + } + } + up_read(&mm->mmap_sem); + + for (i = 0; i < nr_files; i++) { + p = flex_array_get(fa, i); + ret = proc_fill_cache(filp, dirent, filldir, + p->name, p->len, + proc_map_files_instantiate, + task, + (void *)(unsigned long)p->mode); + if (ret) + break; + filp->f_pos++; + } + if (fa) + flex_array_free(fa); + mmput(mm); + } + } + +out_put_task: + put_task_struct(task); +out: + return ret; +} + +static const struct file_operations proc_map_files_operations = { + .read = generic_read_dir, + .readdir = proc_map_files_readdir, + .llseek = default_llseek, +}; + +struct timers_private { + struct pid *pid; + struct task_struct *task; + struct sighand_struct *sighand; + struct pid_namespace *ns; + unsigned long flags; +}; + +static void *timers_start(struct seq_file *m, loff_t *pos) +{ + struct timers_private *tp = m->private; + + tp->task = get_pid_task(tp->pid, PIDTYPE_PID); + if (!tp->task) + return ERR_PTR(-ESRCH); + + tp->sighand = lock_task_sighand(tp->task, &tp->flags); + if (!tp->sighand) + return ERR_PTR(-ESRCH); + + return seq_list_start(&tp->task->signal->posix_timers, *pos); +} + +static void *timers_next(struct seq_file *m, void *v, loff_t *pos) +{ + struct timers_private *tp = m->private; + return seq_list_next(v, &tp->task->signal->posix_timers, pos); +} + +static void timers_stop(struct seq_file *m, void *v) +{ + struct timers_private *tp = m->private; + + if (tp->sighand) { + unlock_task_sighand(tp->task, &tp->flags); + tp->sighand = NULL; + } + + if (tp->task) { + put_task_struct(tp->task); + tp->task = NULL; + } +} + +static int show_timer(struct seq_file *m, void *v) +{ + struct k_itimer *timer; + struct timers_private *tp = m->private; + int notify; + static char *nstr[] = { + [SIGEV_SIGNAL] = "signal", + [SIGEV_NONE] = "none", + [SIGEV_THREAD] = "thread", + }; + + timer = list_entry((struct list_head *)v, struct k_itimer, list); + notify = timer->it_sigev_notify; + + seq_printf(m, "ID: %d\n", timer->it_id); + seq_printf(m, "signal: %d/%p\n", timer->sigq->info.si_signo, + timer->sigq->info.si_value.sival_ptr); + seq_printf(m, "notify: %s/%s.%d\n", + nstr[notify & ~SIGEV_THREAD_ID], + (notify & SIGEV_THREAD_ID) ? "tid" : "pid", + pid_nr_ns(timer->it_pid, tp->ns)); + seq_printf(m, "ClockID: %d\n", timer->it_clock); + + return 0; +} + +static const struct seq_operations proc_timers_seq_ops = { + .start = timers_start, + .next = timers_next, + .stop = timers_stop, + .show = show_timer, +}; + +static int proc_timers_open(struct inode *inode, struct file *file) +{ + struct timers_private *tp; + + tp = __seq_open_private(file, &proc_timers_seq_ops, + sizeof(struct timers_private)); + if (!tp) + return -ENOMEM; + + tp->pid = proc_pid(inode); + tp->ns = inode->i_sb->s_fs_info; + return 0; +} + +static const struct file_operations proc_timers_operations = { + .open = proc_timers_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; +#endif /* CONFIG_CHECKPOINT_RESTORE */ + +static struct dentry *proc_pident_instantiate(struct inode *dir, + struct dentry *dentry, struct task_struct *task, const void *ptr) +{ + const struct pid_entry *p = ptr; + struct inode *inode; + struct proc_inode *ei; + struct dentry *error = ERR_PTR(-ENOENT); + + inode = proc_pid_make_inode(dir->i_sb, task); + if (!inode) + goto out; + + ei = PROC_I(inode); + inode->i_mode = p->mode; + if (S_ISDIR(inode->i_mode)) + set_nlink(inode, 2); /* Use getattr to fix if necessary */ + if (p->iop) + inode->i_op = p->iop; + if (p->fop) + inode->i_fop = p->fop; + ei->op = p->op; + d_set_d_op(dentry, &pid_dentry_operations); + d_add(dentry, inode); + /* Close the race of the process dying before we return the dentry */ + if (pid_revalidate(dentry, 0)) + error = NULL; +out: + return error; +} + +static struct dentry *proc_pident_lookup(struct inode *dir, + struct dentry *dentry, + const struct pid_entry *ents, + unsigned int nents) +{ + struct dentry *error; + struct task_struct *task = get_proc_task(dir); + const struct pid_entry *p, *last; + + error = ERR_PTR(-ENOENT); + + if (!task) + goto out_no_task; + + /* + * Yes, it does not scale. And it should not. Don't add + * new entries into /proc// without very good reasons. + */ + last = &ents[nents - 1]; + for (p = ents; p <= last; p++) { + if (p->len != dentry->d_name.len) + continue; + if (!memcmp(dentry->d_name.name, p->name, p->len)) + break; + } + if (p > last) + goto out; + + error = proc_pident_instantiate(dir, dentry, task, p); +out: + put_task_struct(task); +out_no_task: + return error; +} + +static int proc_pident_fill_cache(struct file *filp, void *dirent, + filldir_t filldir, struct task_struct *task, const struct pid_entry *p) +{ + return proc_fill_cache(filp, dirent, filldir, p->name, p->len, + proc_pident_instantiate, task, p); +} + +static int proc_pident_readdir(struct file *filp, + void *dirent, filldir_t filldir, + const struct pid_entry *ents, unsigned int nents) +{ + int i; + struct dentry *dentry = filp->f_path.dentry; + struct inode *inode = dentry->d_inode; + struct task_struct *task = get_proc_task(inode); + const struct pid_entry *p, *last; + ino_t ino; + int ret; + + ret = -ENOENT; + if (!task) + goto out_no_task; + + ret = 0; + i = filp->f_pos; + switch (i) { + case 0: + ino = inode->i_ino; + if (filldir(dirent, ".", 1, i, ino, DT_DIR) < 0) + goto out; + i++; + filp->f_pos++; + /* fall through */ + case 1: + ino = parent_ino(dentry); + if (filldir(dirent, "..", 2, i, ino, DT_DIR) < 0) + goto out; + i++; + filp->f_pos++; + /* fall through */ + default: + i -= 2; + if (i >= nents) { + ret = 1; + goto out; + } + p = ents + i; + last = &ents[nents - 1]; + while (p <= last) { + if (proc_pident_fill_cache(filp, dirent, filldir, task, p) < 0) + goto out; + filp->f_pos++; + p++; + } + } + + ret = 1; +out: + put_task_struct(task); +out_no_task: + return ret; +} + +#ifdef CONFIG_SECURITY +static ssize_t proc_pid_attr_read(struct file * file, char __user * buf, + size_t count, loff_t *ppos) +{ + struct inode * inode = file_inode(file); + char *p = NULL; + ssize_t length; + struct task_struct *task = get_proc_task(inode); + + if (!task) + return -ESRCH; + + length = security_getprocattr(task, + (char*)file->f_path.dentry->d_name.name, + &p); + put_task_struct(task); + if (length > 0) + length = simple_read_from_buffer(buf, count, ppos, p, length); + kfree(p); + return length; +} + +static ssize_t proc_pid_attr_write(struct file * file, const char __user * buf, + size_t count, loff_t *ppos) +{ + struct inode * inode = file_inode(file); + char *page; + ssize_t length; + struct task_struct *task = get_proc_task(inode); + + length = -ESRCH; + if (!task) + goto out_no_task; + if (count > PAGE_SIZE) + count = PAGE_SIZE; + + /* No partial writes. */ + length = -EINVAL; + if (*ppos != 0) + goto out; + + length = -ENOMEM; + page = (char*)__get_free_page(GFP_TEMPORARY); + if (!page) + goto out; + + length = -EFAULT; + if (copy_from_user(page, buf, count)) + goto out_free; + + /* Guard against adverse ptrace interaction */ + length = mutex_lock_interruptible(&task->signal->cred_guard_mutex); + if (length < 0) + goto out_free; + + length = security_setprocattr(task, + (char*)file->f_path.dentry->d_name.name, + (void*)page, count); + mutex_unlock(&task->signal->cred_guard_mutex); +out_free: + free_page((unsigned long) page); +out: + put_task_struct(task); +out_no_task: + return length; +} + +static const struct file_operations proc_pid_attr_operations = { + .read = proc_pid_attr_read, + .write = proc_pid_attr_write, + .llseek = generic_file_llseek, +}; + +static const struct pid_entry attr_dir_stuff[] = { + REG("current", S_IRUGO|S_IWUGO, proc_pid_attr_operations), + REG("prev", S_IRUGO, proc_pid_attr_operations), + REG("exec", S_IRUGO|S_IWUGO, proc_pid_attr_operations), + REG("fscreate", S_IRUGO|S_IWUGO, proc_pid_attr_operations), + REG("keycreate", S_IRUGO|S_IWUGO, proc_pid_attr_operations), + REG("sockcreate", S_IRUGO|S_IWUGO, proc_pid_attr_operations), +}; + +static int proc_attr_dir_readdir(struct file * filp, + void * dirent, filldir_t filldir) +{ + return proc_pident_readdir(filp,dirent,filldir, + attr_dir_stuff,ARRAY_SIZE(attr_dir_stuff)); +} + +static const struct file_operations proc_attr_dir_operations = { + .read = generic_read_dir, + .readdir = proc_attr_dir_readdir, + .llseek = default_llseek, +}; + +static struct dentry *proc_attr_dir_lookup(struct inode *dir, + struct dentry *dentry, unsigned int flags) +{ + return proc_pident_lookup(dir, dentry, + attr_dir_stuff, ARRAY_SIZE(attr_dir_stuff)); +} + +static const struct inode_operations proc_attr_dir_inode_operations = { + .lookup = proc_attr_dir_lookup, + .getattr = pid_getattr, + .setattr = proc_setattr, +}; + +#endif + +#ifdef CONFIG_ELF_CORE +static ssize_t proc_coredump_filter_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) +{ + struct task_struct *task = get_proc_task(file_inode(file)); + struct mm_struct *mm; + char buffer[PROC_NUMBUF]; + size_t len; + int ret; + + if (!task) + return -ESRCH; + + ret = 0; + mm = get_task_mm(task); + if (mm) { + len = snprintf(buffer, sizeof(buffer), "%08lx\n", + ((mm->flags & MMF_DUMP_FILTER_MASK) >> + MMF_DUMP_FILTER_SHIFT)); + mmput(mm); + ret = simple_read_from_buffer(buf, count, ppos, buffer, len); + } + + put_task_struct(task); + + return ret; +} + +static ssize_t proc_coredump_filter_write(struct file *file, + const char __user *buf, + size_t count, + loff_t *ppos) +{ + struct task_struct *task; + struct mm_struct *mm; + char buffer[PROC_NUMBUF], *end; + unsigned int val; + int ret; + int i; + unsigned long mask; + + ret = -EFAULT; + memset(buffer, 0, sizeof(buffer)); + if (count > sizeof(buffer) - 1) + count = sizeof(buffer) - 1; + if (copy_from_user(buffer, buf, count)) + goto out_no_task; + + ret = -EINVAL; + val = (unsigned int)simple_strtoul(buffer, &end, 0); + if (*end == '\n') + end++; + if (end - buffer == 0) + goto out_no_task; + + ret = -ESRCH; + task = get_proc_task(file_inode(file)); + if (!task) + goto out_no_task; + + ret = end - buffer; + mm = get_task_mm(task); + if (!mm) + goto out_no_mm; + + for (i = 0, mask = 1; i < MMF_DUMP_FILTER_BITS; i++, mask <<= 1) { + if (val & mask) + set_bit(i + MMF_DUMP_FILTER_SHIFT, &mm->flags); + else + clear_bit(i + MMF_DUMP_FILTER_SHIFT, &mm->flags); + } + + mmput(mm); + out_no_mm: + put_task_struct(task); + out_no_task: + return ret; +} + +static const struct file_operations proc_coredump_filter_operations = { + .read = proc_coredump_filter_read, + .write = proc_coredump_filter_write, + .llseek = generic_file_llseek, +}; +#endif + +#ifdef CONFIG_TASK_IO_ACCOUNTING +static int do_io_accounting(struct task_struct *task, char *buffer, int whole) +{ + struct task_io_accounting acct = task->ioac; + unsigned long flags; + int result; + + result = mutex_lock_killable(&task->signal->cred_guard_mutex); + if (result) + return result; + + if (!ptrace_may_access(task, PTRACE_MODE_READ)) { + result = -EACCES; + goto out_unlock; + } + + if (whole && lock_task_sighand(task, &flags)) { + struct task_struct *t = task; + + task_io_accounting_add(&acct, &task->signal->ioac); + while_each_thread(task, t) + task_io_accounting_add(&acct, &t->ioac); + + unlock_task_sighand(task, &flags); + } + result = sprintf(buffer, + "rchar: %llu\n" + "wchar: %llu\n" + "syscr: %llu\n" + "syscw: %llu\n" + "read_bytes: %llu\n" + "write_bytes: %llu\n" + "cancelled_write_bytes: %llu\n", + (unsigned long long)acct.rchar, + (unsigned long long)acct.wchar, + (unsigned long long)acct.syscr, + (unsigned long long)acct.syscw, + (unsigned long long)acct.read_bytes, + (unsigned long long)acct.write_bytes, + (unsigned long long)acct.cancelled_write_bytes); +out_unlock: + mutex_unlock(&task->signal->cred_guard_mutex); + return result; +} + +static int proc_tid_io_accounting(struct task_struct *task, char *buffer) +{ + return do_io_accounting(task, buffer, 0); +} + +static int proc_tgid_io_accounting(struct task_struct *task, char *buffer) +{ + return do_io_accounting(task, buffer, 1); +} +#endif /* CONFIG_TASK_IO_ACCOUNTING */ + +#ifdef CONFIG_USER_NS +static int proc_id_map_open(struct inode *inode, struct file *file, + struct seq_operations *seq_ops) +{ + struct user_namespace *ns = NULL; + struct task_struct *task; + struct seq_file *seq; + int ret = -EINVAL; + + task = get_proc_task(inode); + if (task) { + rcu_read_lock(); + ns = get_user_ns(task_cred_xxx(task, user_ns)); + rcu_read_unlock(); + put_task_struct(task); + } + if (!ns) + goto err; + + ret = seq_open(file, seq_ops); + if (ret) + goto err_put_ns; + + seq = file->private_data; + seq->private = ns; + + return 0; +err_put_ns: + put_user_ns(ns); +err: + return ret; +} + +static int proc_id_map_release(struct inode *inode, struct file *file) +{ + struct seq_file *seq = file->private_data; + struct user_namespace *ns = seq->private; + put_user_ns(ns); + return seq_release(inode, file); +} + +static int proc_uid_map_open(struct inode *inode, struct file *file) +{ + return proc_id_map_open(inode, file, &proc_uid_seq_operations); +} + +static int proc_gid_map_open(struct inode *inode, struct file *file) +{ + return proc_id_map_open(inode, file, &proc_gid_seq_operations); +} + +static int proc_projid_map_open(struct inode *inode, struct file *file) +{ + return proc_id_map_open(inode, file, &proc_projid_seq_operations); +} + +static const struct file_operations proc_uid_map_operations = { + .open = proc_uid_map_open, + .write = proc_uid_map_write, + .read = seq_read, + .llseek = seq_lseek, + .release = proc_id_map_release, +}; + +static const struct file_operations proc_gid_map_operations = { + .open = proc_gid_map_open, + .write = proc_gid_map_write, + .read = seq_read, + .llseek = seq_lseek, + .release = proc_id_map_release, +}; + +static const struct file_operations proc_projid_map_operations = { + .open = proc_projid_map_open, + .write = proc_projid_map_write, + .read = seq_read, + .llseek = seq_lseek, + .release = proc_id_map_release, +}; +#endif /* CONFIG_USER_NS */ + +static int proc_pid_personality(struct seq_file *m, struct pid_namespace *ns, + struct pid *pid, struct task_struct *task) +{ + int err = lock_trace(task); + if (!err) { + seq_printf(m, "%08x\n", task->personality); + unlock_trace(task); + } + return err; +} + +/* + * Thread groups + */ +static const struct file_operations proc_task_operations; +static const struct inode_operations proc_task_inode_operations; + +static const struct pid_entry tgid_base_stuff[] = { + DIR("task", S_IRUGO|S_IXUGO, proc_task_inode_operations, proc_task_operations), + DIR("fd", S_IRUSR|S_IXUSR, proc_fd_inode_operations, proc_fd_operations), +#ifdef CONFIG_CHECKPOINT_RESTORE + DIR("map_files", S_IRUSR|S_IXUSR, proc_map_files_inode_operations, proc_map_files_operations), +#endif + DIR("fdinfo", S_IRUSR|S_IXUSR, proc_fdinfo_inode_operations, proc_fdinfo_operations), + DIR("ns", S_IRUSR|S_IXUGO, proc_ns_dir_inode_operations, proc_ns_dir_operations), +#ifdef CONFIG_NET + DIR("net", S_IRUGO|S_IXUGO, proc_net_inode_operations, proc_net_operations), +#endif + REG("environ", S_IRUSR, proc_environ_operations), + INF("auxv", S_IRUSR, proc_pid_auxv), + ONE("status", S_IRUGO, proc_pid_status), + ONE("personality", S_IRUGO, proc_pid_personality), + INF("limits", S_IRUGO, proc_pid_limits), +#ifdef CONFIG_SCHED_DEBUG + REG("sched", S_IRUGO|S_IWUSR, proc_pid_sched_operations), +#endif +#ifdef CONFIG_SCHED_AUTOGROUP + REG("autogroup", S_IRUGO|S_IWUSR, proc_pid_sched_autogroup_operations), +#endif + REG("comm", S_IRUGO|S_IWUSR, proc_pid_set_comm_operations), +#ifdef CONFIG_HAVE_ARCH_TRACEHOOK + INF("syscall", S_IRUGO, proc_pid_syscall), +#endif + INF("cmdline", S_IRUGO, proc_pid_cmdline), + ONE("stat", S_IRUGO, proc_tgid_stat), + ONE("statm", S_IRUGO, proc_pid_statm), + REG("maps", S_IRUGO, proc_pid_maps_operations), +#ifdef CONFIG_NUMA + REG("numa_maps", S_IRUGO, proc_pid_numa_maps_operations), +#endif + REG("mem", S_IRUSR|S_IWUSR, proc_mem_operations), + LNK("cwd", proc_cwd_link), + LNK("root", proc_root_link), + LNK("exe", proc_exe_link), + REG("mounts", S_IRUGO, proc_mounts_operations), + REG("mountinfo", S_IRUGO, proc_mountinfo_operations), + REG("mountstats", S_IRUSR, proc_mountstats_operations), +#ifdef CONFIG_PROC_PAGE_MONITOR + REG("clear_refs", S_IWUSR, proc_clear_refs_operations), + REG("smaps", S_IRUGO, proc_pid_smaps_operations), + REG("pagemap", S_IRUGO, proc_pagemap_operations), +#endif +#ifdef CONFIG_SECURITY + DIR("attr", S_IRUGO|S_IXUGO, proc_attr_dir_inode_operations, proc_attr_dir_operations), +#endif +#ifdef CONFIG_KALLSYMS + INF("wchan", S_IRUGO, proc_pid_wchan), +#endif +#ifdef CONFIG_STACKTRACE + ONE("stack", S_IRUGO, proc_pid_stack), +#endif +#ifdef CONFIG_SCHEDSTATS + INF("schedstat", S_IRUGO, proc_pid_schedstat), +#endif +#ifdef CONFIG_LATENCYTOP + REG("latency", S_IRUGO, proc_lstats_operations), +#endif +#ifdef CONFIG_PROC_PID_CPUSET + REG("cpuset", S_IRUGO, proc_cpuset_operations), +#endif +#ifdef CONFIG_CGROUPS + REG("cgroup", S_IRUGO, proc_cgroup_operations), +#endif + INF("oom_score", S_IRUGO, proc_oom_score), + ANDROID("oom_adj", S_IRUGO|S_IWUSR, oom_adj), + REG("oom_score_adj", S_IRUGO|S_IWUSR, proc_oom_score_adj_operations), +#ifdef CONFIG_AUDITSYSCALL + REG("loginuid", S_IWUSR|S_IRUGO, proc_loginuid_operations), + REG("sessionid", S_IRUGO, proc_sessionid_operations), +#endif +#ifdef CONFIG_FAULT_INJECTION + REG("make-it-fail", S_IRUGO|S_IWUSR, proc_fault_inject_operations), +#endif +#ifdef CONFIG_ELF_CORE + REG("coredump_filter", S_IRUGO|S_IWUSR, proc_coredump_filter_operations), +#endif +#ifdef CONFIG_TASK_IO_ACCOUNTING + INF("io", S_IRUSR, proc_tgid_io_accounting), +#endif +#ifdef CONFIG_HARDWALL + INF("hardwall", S_IRUGO, proc_pid_hardwall), +#endif +#ifdef CONFIG_USER_NS + REG("uid_map", S_IRUGO|S_IWUSR, proc_uid_map_operations), + REG("gid_map", S_IRUGO|S_IWUSR, proc_gid_map_operations), + REG("projid_map", S_IRUGO|S_IWUSR, proc_projid_map_operations), +#endif +#ifdef CONFIG_CHECKPOINT_RESTORE + REG("timers", S_IRUGO, proc_timers_operations), +#endif +}; + +static int proc_tgid_base_readdir(struct file * filp, + void * dirent, filldir_t filldir) +{ + return proc_pident_readdir(filp,dirent,filldir, + tgid_base_stuff,ARRAY_SIZE(tgid_base_stuff)); +} + +static const struct file_operations proc_tgid_base_operations = { + .read = generic_read_dir, + .readdir = proc_tgid_base_readdir, + .llseek = default_llseek, +}; + +static struct dentry *proc_tgid_base_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) +{ + return proc_pident_lookup(dir, dentry, + tgid_base_stuff, ARRAY_SIZE(tgid_base_stuff)); +} + +static const struct inode_operations proc_tgid_base_inode_operations = { + .lookup = proc_tgid_base_lookup, + .getattr = pid_getattr, + .setattr = proc_setattr, + .permission = proc_pid_permission, +}; + +static void proc_flush_task_mnt(struct vfsmount *mnt, pid_t pid, pid_t tgid) +{ + struct dentry *dentry, *leader, *dir; + char buf[PROC_NUMBUF]; + struct qstr name; + + name.name = buf; + name.len = snprintf(buf, sizeof(buf), "%d", pid); + /* no ->d_hash() rejects on procfs */ + dentry = d_hash_and_lookup(mnt->mnt_root, &name); + if (dentry) { + shrink_dcache_parent(dentry); + d_drop(dentry); + dput(dentry); + } + + name.name = buf; + name.len = snprintf(buf, sizeof(buf), "%d", tgid); + leader = d_hash_and_lookup(mnt->mnt_root, &name); + if (!leader) + goto out; + + name.name = "task"; + name.len = strlen(name.name); + dir = d_hash_and_lookup(leader, &name); + if (!dir) + goto out_put_leader; + + name.name = buf; + name.len = snprintf(buf, sizeof(buf), "%d", pid); + dentry = d_hash_and_lookup(dir, &name); + if (dentry) { + shrink_dcache_parent(dentry); + d_drop(dentry); + dput(dentry); + } + + dput(dir); +out_put_leader: + dput(leader); +out: + return; +} + +/** + * proc_flush_task - Remove dcache entries for @task from the /proc dcache. + * @task: task that should be flushed. + * + * When flushing dentries from proc, one needs to flush them from global + * proc (proc_mnt) and from all the namespaces' procs this task was seen + * in. This call is supposed to do all of this job. + * + * Looks in the dcache for + * /proc/@pid + * /proc/@tgid/task/@pid + * if either directory is present flushes it and all of it'ts children + * from the dcache. + * + * It is safe and reasonable to cache /proc entries for a task until + * that task exits. After that they just clog up the dcache with + * useless entries, possibly causing useful dcache entries to be + * flushed instead. This routine is proved to flush those useless + * dcache entries at process exit time. + * + * NOTE: This routine is just an optimization so it does not guarantee + * that no dcache entries will exist at process exit time it + * just makes it very unlikely that any will persist. + */ + +void proc_flush_task(struct task_struct *task) +{ + int i; + struct pid *pid, *tgid; + struct upid *upid; + + pid = task_pid(task); + tgid = task_tgid(task); + + for (i = 0; i <= pid->level; i++) { + upid = &pid->numbers[i]; + proc_flush_task_mnt(upid->ns->proc_mnt, upid->nr, + tgid->numbers[i].nr); + } +} + +static struct dentry *proc_pid_instantiate(struct inode *dir, + struct dentry * dentry, + struct task_struct *task, const void *ptr) +{ + struct dentry *error = ERR_PTR(-ENOENT); + struct inode *inode; + + inode = proc_pid_make_inode(dir->i_sb, task); + if (!inode) + goto out; + + inode->i_mode = S_IFDIR|S_IRUGO|S_IXUGO; + inode->i_op = &proc_tgid_base_inode_operations; + inode->i_fop = &proc_tgid_base_operations; + inode->i_flags|=S_IMMUTABLE; + + set_nlink(inode, 2 + pid_entry_count_dirs(tgid_base_stuff, + ARRAY_SIZE(tgid_base_stuff))); + + d_set_d_op(dentry, &pid_dentry_operations); + + d_add(dentry, inode); + /* Close the race of the process dying before we return the dentry */ + if (pid_revalidate(dentry, 0)) + error = NULL; +out: + return error; +} + +struct dentry *proc_pid_lookup(struct inode *dir, struct dentry * dentry, unsigned int flags) +{ + struct dentry *result = NULL; + struct task_struct *task; + unsigned tgid; + struct pid_namespace *ns; + + tgid = name_to_int(dentry); + if (tgid == ~0U) + goto out; + + ns = dentry->d_sb->s_fs_info; + rcu_read_lock(); + task = find_task_by_pid_ns(tgid, ns); + if (task) + get_task_struct(task); + rcu_read_unlock(); + if (!task) + goto out; + + result = proc_pid_instantiate(dir, dentry, task, NULL); + put_task_struct(task); +out: + return result; +} + +/* + * Find the first task with tgid >= tgid + * + */ +struct tgid_iter { + unsigned int tgid; + struct task_struct *task; +}; +static struct tgid_iter next_tgid(struct pid_namespace *ns, struct tgid_iter iter) +{ + struct pid *pid; + + if (iter.task) + put_task_struct(iter.task); + rcu_read_lock(); +retry: + iter.task = NULL; + pid = find_ge_pid(iter.tgid, ns); + if (pid) { + iter.tgid = pid_nr_ns(pid, ns); + iter.task = pid_task(pid, PIDTYPE_PID); + /* What we to know is if the pid we have find is the + * pid of a thread_group_leader. Testing for task + * being a thread_group_leader is the obvious thing + * todo but there is a window when it fails, due to + * the pid transfer logic in de_thread. + * + * So we perform the straight forward test of seeing + * if the pid we have found is the pid of a thread + * group leader, and don't worry if the task we have + * found doesn't happen to be a thread group leader. + * As we don't care in the case of readdir. + */ + if (!iter.task || !has_group_leader_pid(iter.task)) { + iter.tgid += 1; + goto retry; + } + get_task_struct(iter.task); + } + rcu_read_unlock(); + return iter; +} + +#define TGID_OFFSET (FIRST_PROCESS_ENTRY + 1) + +static int proc_pid_fill_cache(struct file *filp, void *dirent, filldir_t filldir, + struct tgid_iter iter) +{ + char name[PROC_NUMBUF]; + int len = snprintf(name, sizeof(name), "%d", iter.tgid); + return proc_fill_cache(filp, dirent, filldir, name, len, + proc_pid_instantiate, iter.task, NULL); +} + +static int fake_filldir(void *buf, const char *name, int namelen, + loff_t offset, u64 ino, unsigned d_type) +{ + return 0; +} + +/* for the /proc/ directory itself, after non-process stuff has been done */ +int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir) +{ + struct tgid_iter iter; + struct pid_namespace *ns; + filldir_t __filldir; + loff_t pos = filp->f_pos; + + if (pos >= PID_MAX_LIMIT + TGID_OFFSET) + goto out; + + if (pos == TGID_OFFSET - 1) { + if (proc_fill_cache(filp, dirent, filldir, "self", 4, + NULL, NULL, NULL) < 0) + goto out; + iter.tgid = 0; + } else { + iter.tgid = pos - TGID_OFFSET; + } + iter.task = NULL; + ns = filp->f_dentry->d_sb->s_fs_info; + for (iter = next_tgid(ns, iter); + iter.task; + iter.tgid += 1, iter = next_tgid(ns, iter)) { + if (has_pid_permissions(ns, iter.task, 2)) + __filldir = filldir; + else + __filldir = fake_filldir; + + filp->f_pos = iter.tgid + TGID_OFFSET; + if (proc_pid_fill_cache(filp, dirent, __filldir, iter) < 0) { + put_task_struct(iter.task); + goto out; + } + } + filp->f_pos = PID_MAX_LIMIT + TGID_OFFSET; +out: + return 0; +} + +/* + * Tasks + */ +static const struct pid_entry tid_base_stuff[] = { + DIR("fd", S_IRUSR|S_IXUSR, proc_fd_inode_operations, proc_fd_operations), + DIR("fdinfo", S_IRUSR|S_IXUSR, proc_fdinfo_inode_operations, proc_fdinfo_operations), + DIR("ns", S_IRUSR|S_IXUGO, proc_ns_dir_inode_operations, proc_ns_dir_operations), + REG("environ", S_IRUSR, proc_environ_operations), + INF("auxv", S_IRUSR, proc_pid_auxv), + ONE("status", S_IRUGO, proc_pid_status), + ONE("personality", S_IRUGO, proc_pid_personality), + INF("limits", S_IRUGO, proc_pid_limits), +#ifdef CONFIG_SCHED_DEBUG + REG("sched", S_IRUGO|S_IWUSR, proc_pid_sched_operations), +#endif + REG("comm", S_IRUGO|S_IWUSR, proc_pid_set_comm_operations), +#ifdef CONFIG_HAVE_ARCH_TRACEHOOK + INF("syscall", S_IRUGO, proc_pid_syscall), +#endif + INF("cmdline", S_IRUGO, proc_pid_cmdline), + ONE("stat", S_IRUGO, proc_tid_stat), + ONE("statm", S_IRUGO, proc_pid_statm), + REG("maps", S_IRUGO, proc_tid_maps_operations), +#ifdef CONFIG_CHECKPOINT_RESTORE + REG("children", S_IRUGO, proc_tid_children_operations), +#endif +#ifdef CONFIG_NUMA + REG("numa_maps", S_IRUGO, proc_tid_numa_maps_operations), +#endif + REG("mem", S_IRUSR|S_IWUSR, proc_mem_operations), + LNK("cwd", proc_cwd_link), + LNK("root", proc_root_link), + LNK("exe", proc_exe_link), + REG("mounts", S_IRUGO, proc_mounts_operations), + REG("mountinfo", S_IRUGO, proc_mountinfo_operations), +#ifdef CONFIG_PROC_PAGE_MONITOR + REG("clear_refs", S_IWUSR, proc_clear_refs_operations), + REG("smaps", S_IRUGO, proc_tid_smaps_operations), + REG("pagemap", S_IRUGO, proc_pagemap_operations), +#endif +#ifdef CONFIG_SECURITY + DIR("attr", S_IRUGO|S_IXUGO, proc_attr_dir_inode_operations, proc_attr_dir_operations), +#endif +#ifdef CONFIG_KALLSYMS + INF("wchan", S_IRUGO, proc_pid_wchan), +#endif +#ifdef CONFIG_STACKTRACE + ONE("stack", S_IRUGO, proc_pid_stack), +#endif +#ifdef CONFIG_SCHEDSTATS + INF("schedstat", S_IRUGO, proc_pid_schedstat), +#endif +#ifdef CONFIG_LATENCYTOP + REG("latency", S_IRUGO, proc_lstats_operations), +#endif +#ifdef CONFIG_PROC_PID_CPUSET + REG("cpuset", S_IRUGO, proc_cpuset_operations), +#endif +#ifdef CONFIG_CGROUPS + REG("cgroup", S_IRUGO, proc_cgroup_operations), +#endif + INF("oom_score", S_IRUGO, proc_oom_score), + REG("oom_adj", S_IRUGO|S_IWUSR, proc_oom_adj_operations), + REG("oom_score_adj", S_IRUGO|S_IWUSR, proc_oom_score_adj_operations), +#ifdef CONFIG_AUDITSYSCALL + REG("loginuid", S_IWUSR|S_IRUGO, proc_loginuid_operations), + REG("sessionid", S_IRUGO, proc_sessionid_operations), +#endif +#ifdef CONFIG_FAULT_INJECTION + REG("make-it-fail", S_IRUGO|S_IWUSR, proc_fault_inject_operations), +#endif +#ifdef CONFIG_TASK_IO_ACCOUNTING + INF("io", S_IRUSR, proc_tid_io_accounting), +#endif +#ifdef CONFIG_HARDWALL + INF("hardwall", S_IRUGO, proc_pid_hardwall), +#endif +#ifdef CONFIG_USER_NS + REG("uid_map", S_IRUGO|S_IWUSR, proc_uid_map_operations), + REG("gid_map", S_IRUGO|S_IWUSR, proc_gid_map_operations), + REG("projid_map", S_IRUGO|S_IWUSR, proc_projid_map_operations), +#endif +}; + +static int proc_tid_base_readdir(struct file * filp, + void * dirent, filldir_t filldir) +{ + return proc_pident_readdir(filp,dirent,filldir, + tid_base_stuff,ARRAY_SIZE(tid_base_stuff)); +} + +static struct dentry *proc_tid_base_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) +{ + return proc_pident_lookup(dir, dentry, + tid_base_stuff, ARRAY_SIZE(tid_base_stuff)); +} + +static const struct file_operations proc_tid_base_operations = { + .read = generic_read_dir, + .readdir = proc_tid_base_readdir, + .llseek = default_llseek, +}; + +static const struct inode_operations proc_tid_base_inode_operations = { + .lookup = proc_tid_base_lookup, + .getattr = pid_getattr, + .setattr = proc_setattr, +}; + +static struct dentry *proc_task_instantiate(struct inode *dir, + struct dentry *dentry, struct task_struct *task, const void *ptr) +{ + struct dentry *error = ERR_PTR(-ENOENT); + struct inode *inode; + inode = proc_pid_make_inode(dir->i_sb, task); + + if (!inode) + goto out; + inode->i_mode = S_IFDIR|S_IRUGO|S_IXUGO; + inode->i_op = &proc_tid_base_inode_operations; + inode->i_fop = &proc_tid_base_operations; + inode->i_flags|=S_IMMUTABLE; + + set_nlink(inode, 2 + pid_entry_count_dirs(tid_base_stuff, + ARRAY_SIZE(tid_base_stuff))); + + d_set_d_op(dentry, &pid_dentry_operations); + + d_add(dentry, inode); + /* Close the race of the process dying before we return the dentry */ + if (pid_revalidate(dentry, 0)) + error = NULL; +out: + return error; +} + +static struct dentry *proc_task_lookup(struct inode *dir, struct dentry * dentry, unsigned int flags) +{ + struct dentry *result = ERR_PTR(-ENOENT); + struct task_struct *task; + struct task_struct *leader = get_proc_task(dir); + unsigned tid; + struct pid_namespace *ns; + + if (!leader) + goto out_no_task; + + tid = name_to_int(dentry); + if (tid == ~0U) + goto out; + + ns = dentry->d_sb->s_fs_info; + rcu_read_lock(); + task = find_task_by_pid_ns(tid, ns); + if (task) + get_task_struct(task); + rcu_read_unlock(); + if (!task) + goto out; + if (!same_thread_group(leader, task)) + goto out_drop_task; + + result = proc_task_instantiate(dir, dentry, task, NULL); +out_drop_task: + put_task_struct(task); +out: + put_task_struct(leader); +out_no_task: + return result; +} + +/* + * Find the first tid of a thread group to return to user space. + * + * Usually this is just the thread group leader, but if the users + * buffer was too small or there was a seek into the middle of the + * directory we have more work todo. + * + * In the case of a short read we start with find_task_by_pid. + * + * In the case of a seek we start with the leader and walk nr + * threads past it. + */ +static struct task_struct *first_tid(struct task_struct *leader, + int tid, int nr, struct pid_namespace *ns) +{ + struct task_struct *pos; + + rcu_read_lock(); + /* Attempt to start with the pid of a thread */ + if (tid && (nr > 0)) { + pos = find_task_by_pid_ns(tid, ns); + if (pos && (pos->group_leader == leader)) + goto found; + } + + /* If nr exceeds the number of threads there is nothing todo */ + pos = NULL; + if (nr && nr >= get_nr_threads(leader)) + goto out; + + /* If we haven't found our starting place yet start + * with the leader and walk nr threads forward. + */ + for (pos = leader; nr > 0; --nr) { + pos = next_thread(pos); + if (pos == leader) { + pos = NULL; + goto out; + } + } +found: + get_task_struct(pos); +out: + rcu_read_unlock(); + return pos; +} + +/* + * Find the next thread in the thread list. + * Return NULL if there is an error or no next thread. + * + * The reference to the input task_struct is released. + */ +static struct task_struct *next_tid(struct task_struct *start) +{ + struct task_struct *pos = NULL; + rcu_read_lock(); + if (pid_alive(start)) { + pos = next_thread(start); + if (thread_group_leader(pos)) + pos = NULL; + else + get_task_struct(pos); + } + rcu_read_unlock(); + put_task_struct(start); + return pos; +} + +static int proc_task_fill_cache(struct file *filp, void *dirent, filldir_t filldir, + struct task_struct *task, int tid) +{ + char name[PROC_NUMBUF]; + int len = snprintf(name, sizeof(name), "%d", tid); + return proc_fill_cache(filp, dirent, filldir, name, len, + proc_task_instantiate, task, NULL); +} + +/* for the /proc/TGID/task/ directories */ +static int proc_task_readdir(struct file * filp, void * dirent, filldir_t filldir) +{ + struct dentry *dentry = filp->f_path.dentry; + struct inode *inode = dentry->d_inode; + struct task_struct *leader = NULL; + struct task_struct *task; + int retval = -ENOENT; + ino_t ino; + int tid; + struct pid_namespace *ns; + + task = get_proc_task(inode); + if (!task) + goto out_no_task; + rcu_read_lock(); + if (pid_alive(task)) { + leader = task->group_leader; + get_task_struct(leader); + } + rcu_read_unlock(); + put_task_struct(task); + if (!leader) + goto out_no_task; + retval = 0; + + switch ((unsigned long)filp->f_pos) { + case 0: + ino = inode->i_ino; + if (filldir(dirent, ".", 1, filp->f_pos, ino, DT_DIR) < 0) + goto out; + filp->f_pos++; + /* fall through */ + case 1: + ino = parent_ino(dentry); + if (filldir(dirent, "..", 2, filp->f_pos, ino, DT_DIR) < 0) + goto out; + filp->f_pos++; + /* fall through */ + } + + /* f_version caches the tgid value that the last readdir call couldn't + * return. lseek aka telldir automagically resets f_version to 0. + */ + ns = filp->f_dentry->d_sb->s_fs_info; + tid = (int)filp->f_version; + filp->f_version = 0; + for (task = first_tid(leader, tid, filp->f_pos - 2, ns); + task; + task = next_tid(task), filp->f_pos++) { + tid = task_pid_nr_ns(task, ns); + if (proc_task_fill_cache(filp, dirent, filldir, task, tid) < 0) { + /* returning this tgid failed, save it as the first + * pid for the next readir call */ + filp->f_version = (u64)tid; + put_task_struct(task); + break; + } + } +out: + put_task_struct(leader); +out_no_task: + return retval; +} + +static int proc_task_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) +{ + struct inode *inode = dentry->d_inode; + struct task_struct *p = get_proc_task(inode); + generic_fillattr(inode, stat); + + if (p) { + stat->nlink += get_nr_threads(p); + put_task_struct(p); + } + + return 0; +} + +static const struct inode_operations proc_task_inode_operations = { + .lookup = proc_task_lookup, + .getattr = proc_task_getattr, + .setattr = proc_setattr, + .permission = proc_pid_permission, +}; + +static const struct file_operations proc_task_operations = { + .read = generic_read_dir, + .readdir = proc_task_readdir, + .llseek = default_llseek, +}; diff -Naur kernel.21.3.orig/fs/proc/nommu.c kernel.21.3.new/fs/proc/nommu.c --- kernel.21.3.orig/fs/proc/nommu.c 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/fs/proc/nommu.c 2015-04-25 23:28:45.302481702 +0000 @@ -45,7 +45,10 @@ file = region->vm_file; if (file) { - struct inode *inode = file_inode(region->vm_file); + struct inode *inode; + + file = vmr_pr_or_file(region); + inode = file_inode(file); dev = inode->i_sb->s_dev; ino = inode->i_ino; } diff -Naur kernel.21.3.orig/fs/proc/task_mmu.c kernel.21.3.new/fs/proc/task_mmu.c --- kernel.21.3.orig/fs/proc/task_mmu.c 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/fs/proc/task_mmu.c 2015-04-25 23:28:45.311481613 +0000 @@ -321,7 +321,10 @@ const char *name = NULL; if (file) { - struct inode *inode = file_inode(vma->vm_file); + struct inode *inode; + + file = vma_pr_or_file(vma); + inode = file_inode(file); dev = inode->i_sb->s_dev; ino = inode->i_ino; pgoff = ((loff_t)vma->vm_pgoff) << PAGE_SHIFT; @@ -1346,6 +1349,7 @@ seq_printf(m, "%08lx %s", vma->vm_start, buffer); if (file) { + file = vma_pr_or_file(vma); seq_printf(m, " file="); seq_path(m, &file->f_path, "\n\t= "); } else if (vma->vm_start <= mm->brk && vma->vm_end >= mm->start_brk) { diff -Naur kernel.21.3.orig/fs/proc/task_mmu.c.orig kernel.21.3.new/fs/proc/task_mmu.c.orig --- kernel.21.3.orig/fs/proc/task_mmu.c.orig 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/proc/task_mmu.c.orig 2015-02-03 01:44:32.000000000 +0000 @@ -0,0 +1,1474 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include "internal.h" + +void task_mem(struct seq_file *m, struct mm_struct *mm) +{ + unsigned long data, text, lib, swap; + unsigned long hiwater_vm, total_vm, hiwater_rss, total_rss; + + /* + * Note: to minimize their overhead, mm maintains hiwater_vm and + * hiwater_rss only when about to *lower* total_vm or rss. Any + * collector of these hiwater stats must therefore get total_vm + * and rss too, which will usually be the higher. Barriers? not + * worth the effort, such snapshots can always be inconsistent. + */ + hiwater_vm = total_vm = mm->total_vm; + if (hiwater_vm < mm->hiwater_vm) + hiwater_vm = mm->hiwater_vm; + hiwater_rss = total_rss = get_mm_rss(mm); + if (hiwater_rss < mm->hiwater_rss) + hiwater_rss = mm->hiwater_rss; + + data = mm->total_vm - mm->shared_vm - mm->stack_vm; + text = (PAGE_ALIGN(mm->end_code) - (mm->start_code & PAGE_MASK)) >> 10; + lib = (mm->exec_vm << (PAGE_SHIFT-10)) - text; + swap = get_mm_counter(mm, MM_SWAPENTS); + seq_printf(m, + "VmPeak:\t%8lu kB\n" + "VmSize:\t%8lu kB\n" + "VmLck:\t%8lu kB\n" + "VmPin:\t%8lu kB\n" + "VmHWM:\t%8lu kB\n" + "VmRSS:\t%8lu kB\n" + "VmData:\t%8lu kB\n" + "VmStk:\t%8lu kB\n" + "VmExe:\t%8lu kB\n" + "VmLib:\t%8lu kB\n" + "VmPTE:\t%8lu kB\n" + "VmSwap:\t%8lu kB\n", + hiwater_vm << (PAGE_SHIFT-10), + total_vm << (PAGE_SHIFT-10), + mm->locked_vm << (PAGE_SHIFT-10), + mm->pinned_vm << (PAGE_SHIFT-10), + hiwater_rss << (PAGE_SHIFT-10), + total_rss << (PAGE_SHIFT-10), + data << (PAGE_SHIFT-10), + mm->stack_vm << (PAGE_SHIFT-10), text, lib, + (PTRS_PER_PTE*sizeof(pte_t)*mm->nr_ptes) >> 10, + swap << (PAGE_SHIFT-10)); +} + +unsigned long task_vsize(struct mm_struct *mm) +{ + return PAGE_SIZE * mm->total_vm; +} + +unsigned long task_statm(struct mm_struct *mm, + unsigned long *shared, unsigned long *text, + unsigned long *data, unsigned long *resident) +{ + *shared = get_mm_counter(mm, MM_FILEPAGES); + *text = (PAGE_ALIGN(mm->end_code) - (mm->start_code & PAGE_MASK)) + >> PAGE_SHIFT; + *data = mm->total_vm - mm->shared_vm; + *resident = *shared + get_mm_counter(mm, MM_ANONPAGES); + return mm->total_vm; +} + +static void pad_len_spaces(struct seq_file *m, int len) +{ + len = 25 + sizeof(void*) * 6 - len; + if (len < 1) + len = 1; + seq_printf(m, "%*c", len, ' '); +} + +#ifdef CONFIG_NUMA +/* + * These functions are for numa_maps but called in generic **maps seq_file + * ->start(), ->stop() ops. + * + * numa_maps scans all vmas under mmap_sem and checks their mempolicy. + * Each mempolicy object is controlled by reference counting. The problem here + * is how to avoid accessing dead mempolicy object. + * + * Because we're holding mmap_sem while reading seq_file, it's safe to access + * each vma's mempolicy, no vma objects will never drop refs to mempolicy. + * + * A task's mempolicy (task->mempolicy) has different behavior. task->mempolicy + * is set and replaced under mmap_sem but unrefed and cleared under task_lock(). + * So, without task_lock(), we cannot trust get_vma_policy() because we cannot + * gurantee the task never exits under us. But taking task_lock() around + * get_vma_plicy() causes lock order problem. + * + * To access task->mempolicy without lock, we hold a reference count of an + * object pointed by task->mempolicy and remember it. This will guarantee + * that task->mempolicy points to an alive object or NULL in numa_maps accesses. + */ +static void hold_task_mempolicy(struct proc_maps_private *priv) +{ + struct task_struct *task = priv->task; + + task_lock(task); + priv->task_mempolicy = task->mempolicy; + mpol_get(priv->task_mempolicy); + task_unlock(task); +} +static void release_task_mempolicy(struct proc_maps_private *priv) +{ + mpol_put(priv->task_mempolicy); +} +#else +static void hold_task_mempolicy(struct proc_maps_private *priv) +{ +} +static void release_task_mempolicy(struct proc_maps_private *priv) +{ +} +#endif + +static void seq_print_vma_name(struct seq_file *m, struct vm_area_struct *vma) +{ + const char __user *name = vma_get_anon_name(vma); + struct mm_struct *mm = vma->vm_mm; + + unsigned long page_start_vaddr; + unsigned long page_offset; + unsigned long num_pages; + unsigned long max_len = NAME_MAX; + int i; + + page_start_vaddr = (unsigned long)name & PAGE_MASK; + page_offset = (unsigned long)name - page_start_vaddr; + num_pages = DIV_ROUND_UP(page_offset + max_len, PAGE_SIZE); + + seq_puts(m, "[anon:"); + + for (i = 0; i < num_pages; i++) { + int len; + int write_len; + const char *kaddr; + long pages_pinned; + struct page *page; + + pages_pinned = get_user_pages(current, mm, page_start_vaddr, + 1, 0, 0, &page, NULL); + if (pages_pinned < 1) { + seq_puts(m, "]"); + return; + } + + kaddr = (const char *)kmap(page); + len = min(max_len, PAGE_SIZE - page_offset); + write_len = strnlen(kaddr + page_offset, len); + seq_write(m, kaddr + page_offset, write_len); + kunmap(page); + put_page(page); + + /* if strnlen hit a null terminator then we're done */ + if (write_len != len) + break; + + max_len -= len; + page_offset = 0; + page_start_vaddr += PAGE_SIZE; + } + + seq_putc(m, ']'); +} + +static void vma_stop(struct proc_maps_private *priv, struct vm_area_struct *vma) +{ + if (vma && vma != priv->tail_vma) { + struct mm_struct *mm = vma->vm_mm; + release_task_mempolicy(priv); + up_read(&mm->mmap_sem); + mmput(mm); + } +} + +static void *m_start(struct seq_file *m, loff_t *pos) +{ + struct proc_maps_private *priv = m->private; + unsigned long last_addr = m->version; + struct mm_struct *mm; + struct vm_area_struct *vma, *tail_vma = NULL; + loff_t l = *pos; + + /* Clear the per syscall fields in priv */ + priv->task = NULL; + priv->tail_vma = NULL; + + /* + * We remember last_addr rather than next_addr to hit with + * mmap_cache most of the time. We have zero last_addr at + * the beginning and also after lseek. We will have -1 last_addr + * after the end of the vmas. + */ + + if (last_addr == -1UL) + return NULL; + + priv->task = get_pid_task(priv->pid, PIDTYPE_PID); + if (!priv->task) + return ERR_PTR(-ESRCH); + + mm = mm_access(priv->task, PTRACE_MODE_READ); + if (!mm || IS_ERR(mm)) + return mm; + down_read(&mm->mmap_sem); + + tail_vma = get_gate_vma(priv->task->mm); + priv->tail_vma = tail_vma; + hold_task_mempolicy(priv); + /* Start with last addr hint */ + vma = find_vma(mm, last_addr); + if (last_addr && vma) { + vma = vma->vm_next; + goto out; + } + + /* + * Check the vma index is within the range and do + * sequential scan until m_index. + */ + vma = NULL; + if ((unsigned long)l < mm->map_count) { + vma = mm->mmap; + while (l-- && vma) + vma = vma->vm_next; + goto out; + } + + if (l != mm->map_count) + tail_vma = NULL; /* After gate vma */ + +out: + if (vma) + return vma; + + release_task_mempolicy(priv); + /* End of vmas has been reached */ + m->version = (tail_vma != NULL)? 0: -1UL; + up_read(&mm->mmap_sem); + mmput(mm); + return tail_vma; +} + +static void *m_next(struct seq_file *m, void *v, loff_t *pos) +{ + struct proc_maps_private *priv = m->private; + struct vm_area_struct *vma = v; + struct vm_area_struct *tail_vma = priv->tail_vma; + + (*pos)++; + if (vma && (vma != tail_vma) && vma->vm_next) + return vma->vm_next; + vma_stop(priv, vma); + return (vma != tail_vma)? tail_vma: NULL; +} + +static void m_stop(struct seq_file *m, void *v) +{ + struct proc_maps_private *priv = m->private; + struct vm_area_struct *vma = v; + + if (!IS_ERR(vma)) + vma_stop(priv, vma); + if (priv->task) + put_task_struct(priv->task); +} + +static int do_maps_open(struct inode *inode, struct file *file, + const struct seq_operations *ops) +{ + struct proc_maps_private *priv; + int ret = -ENOMEM; + priv = kzalloc(sizeof(*priv), GFP_KERNEL); + if (priv) { + priv->pid = proc_pid(inode); + ret = seq_open(file, ops); + if (!ret) { + struct seq_file *m = file->private_data; + m->private = priv; + } else { + kfree(priv); + } + } + return ret; +} + +static void +show_map_vma(struct seq_file *m, struct vm_area_struct *vma, int is_pid) +{ + struct mm_struct *mm = vma->vm_mm; + struct file *file = vma->vm_file; + struct proc_maps_private *priv = m->private; + struct task_struct *task = priv->task; + vm_flags_t flags = vma->vm_flags; + unsigned long ino = 0; + unsigned long long pgoff = 0; + unsigned long start, end; + dev_t dev = 0; + int len; + const char *name = NULL; + + if (file) { + struct inode *inode = file_inode(vma->vm_file); + dev = inode->i_sb->s_dev; + ino = inode->i_ino; + pgoff = ((loff_t)vma->vm_pgoff) << PAGE_SHIFT; + } + + /* We don't show the stack guard page in /proc/maps */ + start = vma->vm_start; + if (stack_guard_page_start(vma, start)) + start += PAGE_SIZE; + end = vma->vm_end; + if (stack_guard_page_end(vma, end)) + end -= PAGE_SIZE; + + seq_printf(m, "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu %n", + start, + end, + flags & VM_READ ? 'r' : '-', + flags & VM_WRITE ? 'w' : '-', + flags & VM_EXEC ? 'x' : '-', + flags & VM_MAYSHARE ? 's' : 'p', + pgoff, + MAJOR(dev), MINOR(dev), ino, &len); + + /* + * Print the dentry name for named mappings, and a + * special [heap] marker for the heap: + */ + if (file) { + pad_len_spaces(m, len); + seq_path(m, &file->f_path, "\n"); + goto done; + } + + name = arch_vma_name(vma); + if (!name) { + pid_t tid; + + if (!mm) { + name = "[vdso]"; + goto done; + } + + if (vma->vm_start <= mm->brk && + vma->vm_end >= mm->start_brk) { + name = "[heap]"; + goto done; + } + + tid = vm_is_stack(task, vma, is_pid); + + if (tid != 0) { + /* + * Thread stack in /proc/PID/task/TID/maps or + * the main process stack. + */ + if (!is_pid || (vma->vm_start <= mm->start_stack && + vma->vm_end >= mm->start_stack)) { + name = "[stack]"; + } else { + /* Thread stack in /proc/PID/maps */ + pad_len_spaces(m, len); + seq_printf(m, "[stack:%d]", tid); + } + goto done; + } + + if (vma_get_anon_name(vma)) { + pad_len_spaces(m, len); + seq_print_vma_name(m, vma); + } + } + +done: + if (name) { + pad_len_spaces(m, len); + seq_puts(m, name); + } + seq_putc(m, '\n'); +} + +static int show_map(struct seq_file *m, void *v, int is_pid) +{ + struct vm_area_struct *vma = v; + struct proc_maps_private *priv = m->private; + struct task_struct *task = priv->task; + + show_map_vma(m, vma, is_pid); + + if (m->count < m->size) /* vma is copied successfully */ + m->version = (vma != get_gate_vma(task->mm)) + ? vma->vm_start : 0; + return 0; +} + +static int show_pid_map(struct seq_file *m, void *v) +{ + return show_map(m, v, 1); +} + +static int show_tid_map(struct seq_file *m, void *v) +{ + return show_map(m, v, 0); +} + +static const struct seq_operations proc_pid_maps_op = { + .start = m_start, + .next = m_next, + .stop = m_stop, + .show = show_pid_map +}; + +static const struct seq_operations proc_tid_maps_op = { + .start = m_start, + .next = m_next, + .stop = m_stop, + .show = show_tid_map +}; + +static int pid_maps_open(struct inode *inode, struct file *file) +{ + return do_maps_open(inode, file, &proc_pid_maps_op); +} + +static int tid_maps_open(struct inode *inode, struct file *file) +{ + return do_maps_open(inode, file, &proc_tid_maps_op); +} + +const struct file_operations proc_pid_maps_operations = { + .open = pid_maps_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; + +const struct file_operations proc_tid_maps_operations = { + .open = tid_maps_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; + +/* + * Proportional Set Size(PSS): my share of RSS. + * + * PSS of a process is the count of pages it has in memory, where each + * page is divided by the number of processes sharing it. So if a + * process has 1000 pages all to itself, and 1000 shared with one other + * process, its PSS will be 1500. + * + * To keep (accumulated) division errors low, we adopt a 64bit + * fixed-point pss counter to minimize division errors. So (pss >> + * PSS_SHIFT) would be the real byte count. + * + * A shift of 12 before division means (assuming 4K page size): + * - 1M 3-user-pages add up to 8KB errors; + * - supports mapcount up to 2^24, or 16M; + * - supports PSS up to 2^52 bytes, or 4PB. + */ +#define PSS_SHIFT 12 + +#ifdef CONFIG_PROC_PAGE_MONITOR +struct mem_size_stats { + struct vm_area_struct *vma; + unsigned long resident; + unsigned long shared_clean; + unsigned long shared_dirty; + unsigned long private_clean; + unsigned long private_dirty; + unsigned long referenced; + unsigned long anonymous; + unsigned long anonymous_thp; + unsigned long swap; + unsigned long nonlinear; + u64 pss; +}; + + +static void smaps_pte_entry(pte_t ptent, unsigned long addr, + unsigned long ptent_size, struct mm_walk *walk) +{ + struct mem_size_stats *mss = walk->private; + struct vm_area_struct *vma = mss->vma; + pgoff_t pgoff = linear_page_index(vma, addr); + struct page *page = NULL; + int mapcount; + + if (pte_present(ptent)) { + page = vm_normal_page(vma, addr, ptent); + } else if (is_swap_pte(ptent)) { + swp_entry_t swpent = pte_to_swp_entry(ptent); + + if (!non_swap_entry(swpent)) + mss->swap += ptent_size; + else if (is_migration_entry(swpent)) + page = migration_entry_to_page(swpent); + } else if (pte_file(ptent)) { + if (pte_to_pgoff(ptent) != pgoff) + mss->nonlinear += ptent_size; + } + + if (!page) + return; + + if (PageAnon(page)) + mss->anonymous += ptent_size; + + if (page->index != pgoff) + mss->nonlinear += ptent_size; + + mss->resident += ptent_size; + /* Accumulate the size in pages that have been accessed. */ + if (pte_young(ptent) || PageReferenced(page)) + mss->referenced += ptent_size; + mapcount = page_mapcount(page); + if (mapcount >= 2) { + if (pte_dirty(ptent) || PageDirty(page)) + mss->shared_dirty += ptent_size; + else + mss->shared_clean += ptent_size; + mss->pss += (ptent_size << PSS_SHIFT) / mapcount; + } else { + if (pte_dirty(ptent) || PageDirty(page)) + mss->private_dirty += ptent_size; + else + mss->private_clean += ptent_size; + mss->pss += (ptent_size << PSS_SHIFT); + } +} + +static int smaps_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end, + struct mm_walk *walk) +{ + struct mem_size_stats *mss = walk->private; + struct vm_area_struct *vma = mss->vma; + pte_t *pte; + spinlock_t *ptl; + + if (pmd_trans_huge_lock(pmd, vma) == 1) { + smaps_pte_entry(*(pte_t *)pmd, addr, HPAGE_PMD_SIZE, walk); + spin_unlock(&walk->mm->page_table_lock); + mss->anonymous_thp += HPAGE_PMD_SIZE; + return 0; + } + + if (pmd_trans_unstable(pmd)) + return 0; + /* + * The mmap_sem held all the way back in m_start() is what + * keeps khugepaged out of here and from collapsing things + * in here. + */ + pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl); + for (; addr != end; pte++, addr += PAGE_SIZE) + smaps_pte_entry(*pte, addr, PAGE_SIZE, walk); + pte_unmap_unlock(pte - 1, ptl); + cond_resched(); + return 0; +} + +static void show_smap_vma_flags(struct seq_file *m, struct vm_area_struct *vma) +{ + /* + * Don't forget to update Documentation/ on changes. + */ + static const char mnemonics[BITS_PER_LONG][2] = { + /* + * In case if we meet a flag we don't know about. + */ + [0 ... (BITS_PER_LONG-1)] = "??", + + [ilog2(VM_READ)] = "rd", + [ilog2(VM_WRITE)] = "wr", + [ilog2(VM_EXEC)] = "ex", + [ilog2(VM_SHARED)] = "sh", + [ilog2(VM_MAYREAD)] = "mr", + [ilog2(VM_MAYWRITE)] = "mw", + [ilog2(VM_MAYEXEC)] = "me", + [ilog2(VM_MAYSHARE)] = "ms", + [ilog2(VM_GROWSDOWN)] = "gd", + [ilog2(VM_PFNMAP)] = "pf", + [ilog2(VM_DENYWRITE)] = "dw", + [ilog2(VM_LOCKED)] = "lo", + [ilog2(VM_IO)] = "io", + [ilog2(VM_SEQ_READ)] = "sr", + [ilog2(VM_RAND_READ)] = "rr", + [ilog2(VM_DONTCOPY)] = "dc", + [ilog2(VM_DONTEXPAND)] = "de", + [ilog2(VM_ACCOUNT)] = "ac", + [ilog2(VM_NORESERVE)] = "nr", + [ilog2(VM_HUGETLB)] = "ht", + [ilog2(VM_NONLINEAR)] = "nl", + [ilog2(VM_ARCH_1)] = "ar", + [ilog2(VM_DONTDUMP)] = "dd", + [ilog2(VM_MIXEDMAP)] = "mm", + [ilog2(VM_HUGEPAGE)] = "hg", + [ilog2(VM_NOHUGEPAGE)] = "nh", + [ilog2(VM_MERGEABLE)] = "mg", + }; + size_t i; + + seq_puts(m, "VmFlags: "); + for (i = 0; i < BITS_PER_LONG; i++) { + if (vma->vm_flags & (1UL << i)) { + seq_printf(m, "%c%c ", + mnemonics[i][0], mnemonics[i][1]); + } + } + seq_putc(m, '\n'); +} + +static int show_smap(struct seq_file *m, void *v, int is_pid) +{ + struct proc_maps_private *priv = m->private; + struct task_struct *task = priv->task; + struct vm_area_struct *vma = v; + struct mem_size_stats mss; + struct mm_walk smaps_walk = { + .pmd_entry = smaps_pte_range, + .mm = vma->vm_mm, + .private = &mss, + }; + + memset(&mss, 0, sizeof mss); + mss.vma = vma; + /* mmap_sem is held in m_start */ + if (vma->vm_mm && !is_vm_hugetlb_page(vma)) + walk_page_range(vma->vm_start, vma->vm_end, &smaps_walk); + + show_map_vma(m, vma, is_pid); + + seq_printf(m, + "Size: %8lu kB\n" + "Rss: %8lu kB\n" + "Pss: %8lu kB\n" + "Shared_Clean: %8lu kB\n" + "Shared_Dirty: %8lu kB\n" + "Private_Clean: %8lu kB\n" + "Private_Dirty: %8lu kB\n" + "Referenced: %8lu kB\n" + "Anonymous: %8lu kB\n" + "AnonHugePages: %8lu kB\n" + "Swap: %8lu kB\n" + "KernelPageSize: %8lu kB\n" + "MMUPageSize: %8lu kB\n" + "Locked: %8lu kB\n", + (vma->vm_end - vma->vm_start) >> 10, + mss.resident >> 10, + (unsigned long)(mss.pss >> (10 + PSS_SHIFT)), + mss.shared_clean >> 10, + mss.shared_dirty >> 10, + mss.private_clean >> 10, + mss.private_dirty >> 10, + mss.referenced >> 10, + mss.anonymous >> 10, + mss.anonymous_thp >> 10, + mss.swap >> 10, + vma_kernel_pagesize(vma) >> 10, + vma_mmu_pagesize(vma) >> 10, + (vma->vm_flags & VM_LOCKED) ? + (unsigned long)(mss.pss >> (10 + PSS_SHIFT)) : 0); + + if (vma->vm_flags & VM_NONLINEAR) + seq_printf(m, "Nonlinear: %8lu kB\n", + mss.nonlinear >> 10); + + show_smap_vma_flags(m, vma); + + if (vma_get_anon_name(vma)) { + seq_puts(m, "Name: "); + seq_print_vma_name(m, vma); + seq_putc(m, '\n'); + } + + if (m->count < m->size) /* vma is copied successfully */ + m->version = (vma != get_gate_vma(task->mm)) + ? vma->vm_start : 0; + return 0; +} + +static int show_pid_smap(struct seq_file *m, void *v) +{ + return show_smap(m, v, 1); +} + +static int show_tid_smap(struct seq_file *m, void *v) +{ + return show_smap(m, v, 0); +} + +static const struct seq_operations proc_pid_smaps_op = { + .start = m_start, + .next = m_next, + .stop = m_stop, + .show = show_pid_smap +}; + +static const struct seq_operations proc_tid_smaps_op = { + .start = m_start, + .next = m_next, + .stop = m_stop, + .show = show_tid_smap +}; + +static int pid_smaps_open(struct inode *inode, struct file *file) +{ + return do_maps_open(inode, file, &proc_pid_smaps_op); +} + +static int tid_smaps_open(struct inode *inode, struct file *file) +{ + return do_maps_open(inode, file, &proc_tid_smaps_op); +} + +const struct file_operations proc_pid_smaps_operations = { + .open = pid_smaps_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; + +const struct file_operations proc_tid_smaps_operations = { + .open = tid_smaps_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; + +static int clear_refs_pte_range(pmd_t *pmd, unsigned long addr, + unsigned long end, struct mm_walk *walk) +{ + struct vm_area_struct *vma = walk->private; + pte_t *pte, ptent; + spinlock_t *ptl; + struct page *page; + + split_huge_page_pmd(vma, addr, pmd); + if (pmd_trans_unstable(pmd)) + return 0; + + pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl); + for (; addr != end; pte++, addr += PAGE_SIZE) { + ptent = *pte; + if (!pte_present(ptent)) + continue; + + page = vm_normal_page(vma, addr, ptent); + if (!page) + continue; + + /* Clear accessed and referenced bits. */ + ptep_test_and_clear_young(vma, addr, pte); + ClearPageReferenced(page); + } + pte_unmap_unlock(pte - 1, ptl); + cond_resched(); + return 0; +} + +#define CLEAR_REFS_ALL 1 +#define CLEAR_REFS_ANON 2 +#define CLEAR_REFS_MAPPED 3 + +static ssize_t clear_refs_write(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) +{ + struct task_struct *task; + char buffer[PROC_NUMBUF]; + struct mm_struct *mm; + struct vm_area_struct *vma; + int type; + int rv; + + memset(buffer, 0, sizeof(buffer)); + if (count > sizeof(buffer) - 1) + count = sizeof(buffer) - 1; + if (copy_from_user(buffer, buf, count)) + return -EFAULT; + rv = kstrtoint(strstrip(buffer), 10, &type); + if (rv < 0) + return rv; + if (type < CLEAR_REFS_ALL || type > CLEAR_REFS_MAPPED) + return -EINVAL; + task = get_proc_task(file_inode(file)); + if (!task) + return -ESRCH; + mm = get_task_mm(task); + if (mm) { + struct mm_walk clear_refs_walk = { + .pmd_entry = clear_refs_pte_range, + .mm = mm, + }; + down_read(&mm->mmap_sem); + for (vma = mm->mmap; vma; vma = vma->vm_next) { + clear_refs_walk.private = vma; + if (is_vm_hugetlb_page(vma)) + continue; + /* + * Writing 1 to /proc/pid/clear_refs affects all pages. + * + * Writing 2 to /proc/pid/clear_refs only affects + * Anonymous pages. + * + * Writing 3 to /proc/pid/clear_refs only affects file + * mapped pages. + */ + if (type == CLEAR_REFS_ANON && vma->vm_file) + continue; + if (type == CLEAR_REFS_MAPPED && !vma->vm_file) + continue; + walk_page_range(vma->vm_start, vma->vm_end, + &clear_refs_walk); + } + flush_tlb_mm(mm); + up_read(&mm->mmap_sem); + mmput(mm); + } + put_task_struct(task); + + return count; +} + +const struct file_operations proc_clear_refs_operations = { + .write = clear_refs_write, + .llseek = noop_llseek, +}; + +typedef struct { + u64 pme; +} pagemap_entry_t; + +struct pagemapread { + int pos, len; /* units: PM_ENTRY_BYTES, not bytes */ + pagemap_entry_t *buffer; +}; + +#define PAGEMAP_WALK_SIZE (PMD_SIZE) +#define PAGEMAP_WALK_MASK (PMD_MASK) + +#define PM_ENTRY_BYTES sizeof(pagemap_entry_t) +#define PM_STATUS_BITS 3 +#define PM_STATUS_OFFSET (64 - PM_STATUS_BITS) +#define PM_STATUS_MASK (((1LL << PM_STATUS_BITS) - 1) << PM_STATUS_OFFSET) +#define PM_STATUS(nr) (((nr) << PM_STATUS_OFFSET) & PM_STATUS_MASK) +#define PM_PSHIFT_BITS 6 +#define PM_PSHIFT_OFFSET (PM_STATUS_OFFSET - PM_PSHIFT_BITS) +#define PM_PSHIFT_MASK (((1LL << PM_PSHIFT_BITS) - 1) << PM_PSHIFT_OFFSET) +#define PM_PSHIFT(x) (((u64) (x) << PM_PSHIFT_OFFSET) & PM_PSHIFT_MASK) +#define PM_PFRAME_MASK ((1LL << PM_PSHIFT_OFFSET) - 1) +#define PM_PFRAME(x) ((x) & PM_PFRAME_MASK) + +#define PM_PRESENT PM_STATUS(4LL) +#define PM_SWAP PM_STATUS(2LL) +#define PM_FILE PM_STATUS(1LL) +#define PM_NOT_PRESENT PM_PSHIFT(PAGE_SHIFT) +#define PM_END_OF_BUFFER 1 + +static inline pagemap_entry_t make_pme(u64 val) +{ + return (pagemap_entry_t) { .pme = val }; +} + +static int add_to_pagemap(unsigned long addr, pagemap_entry_t *pme, + struct pagemapread *pm) +{ + pm->buffer[pm->pos++] = *pme; + if (pm->pos >= pm->len) + return PM_END_OF_BUFFER; + return 0; +} + +static int pagemap_pte_hole(unsigned long start, unsigned long end, + struct mm_walk *walk) +{ + struct pagemapread *pm = walk->private; + unsigned long addr; + int err = 0; + pagemap_entry_t pme = make_pme(PM_NOT_PRESENT); + + for (addr = start; addr < end; addr += PAGE_SIZE) { + err = add_to_pagemap(addr, &pme, pm); + if (err) + break; + } + return err; +} + +static void pte_to_pagemap_entry(pagemap_entry_t *pme, + struct vm_area_struct *vma, unsigned long addr, pte_t pte) +{ + u64 frame, flags; + struct page *page = NULL; + + if (pte_present(pte)) { + frame = pte_pfn(pte); + flags = PM_PRESENT; + page = vm_normal_page(vma, addr, pte); + } else if (is_swap_pte(pte)) { + swp_entry_t entry = pte_to_swp_entry(pte); + + frame = swp_type(entry) | + (swp_offset(entry) << MAX_SWAPFILES_SHIFT); + flags = PM_SWAP; + if (is_migration_entry(entry)) + page = migration_entry_to_page(entry); + } else { + *pme = make_pme(PM_NOT_PRESENT); + return; + } + + if (page && !PageAnon(page)) + flags |= PM_FILE; + + *pme = make_pme(PM_PFRAME(frame) | PM_PSHIFT(PAGE_SHIFT) | flags); +} + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +static void thp_pmd_to_pagemap_entry(pagemap_entry_t *pme, + pmd_t pmd, int offset) +{ + /* + * Currently pmd for thp is always present because thp can not be + * swapped-out, migrated, or HWPOISONed (split in such cases instead.) + * This if-check is just to prepare for future implementation. + */ + if (pmd_present(pmd)) + *pme = make_pme(PM_PFRAME(pmd_pfn(pmd) + offset) + | PM_PSHIFT(PAGE_SHIFT) | PM_PRESENT); + else + *pme = make_pme(PM_NOT_PRESENT); +} +#else +static inline void thp_pmd_to_pagemap_entry(pagemap_entry_t *pme, + pmd_t pmd, int offset) +{ +} +#endif + +static int pagemap_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end, + struct mm_walk *walk) +{ + struct vm_area_struct *vma; + struct pagemapread *pm = walk->private; + pte_t *pte; + int err = 0; + pagemap_entry_t pme = make_pme(PM_NOT_PRESENT); + + /* find the first VMA at or above 'addr' */ + vma = find_vma(walk->mm, addr); + if (vma && pmd_trans_huge_lock(pmd, vma) == 1) { + for (; addr != end; addr += PAGE_SIZE) { + unsigned long offset; + + offset = (addr & ~PAGEMAP_WALK_MASK) >> + PAGE_SHIFT; + thp_pmd_to_pagemap_entry(&pme, *pmd, offset); + err = add_to_pagemap(addr, &pme, pm); + if (err) + break; + } + spin_unlock(&walk->mm->page_table_lock); + return err; + } + + if (pmd_trans_unstable(pmd)) + return 0; + for (; addr != end; addr += PAGE_SIZE) { + + /* check to see if we've left 'vma' behind + * and need a new, higher one */ + if (vma && (addr >= vma->vm_end)) { + vma = find_vma(walk->mm, addr); + pme = make_pme(PM_NOT_PRESENT); + } + + /* check that 'vma' actually covers this address, + * and that it isn't a huge page vma */ + if (vma && (vma->vm_start <= addr) && + !is_vm_hugetlb_page(vma)) { + pte = pte_offset_map(pmd, addr); + pte_to_pagemap_entry(&pme, vma, addr, *pte); + /* unmap before userspace copy */ + pte_unmap(pte); + } + err = add_to_pagemap(addr, &pme, pm); + if (err) + return err; + } + + cond_resched(); + + return err; +} + +#ifdef CONFIG_HUGETLB_PAGE +static void huge_pte_to_pagemap_entry(pagemap_entry_t *pme, + pte_t pte, int offset) +{ + if (pte_present(pte)) + *pme = make_pme(PM_PFRAME(pte_pfn(pte) + offset) + | PM_PSHIFT(PAGE_SHIFT) | PM_PRESENT); + else + *pme = make_pme(PM_NOT_PRESENT); +} + +/* This function walks within one hugetlb entry in the single call */ +static int pagemap_hugetlb_range(pte_t *pte, unsigned long hmask, + unsigned long addr, unsigned long end, + struct mm_walk *walk) +{ + struct pagemapread *pm = walk->private; + int err = 0; + pagemap_entry_t pme; + + for (; addr != end; addr += PAGE_SIZE) { + int offset = (addr & ~hmask) >> PAGE_SHIFT; + huge_pte_to_pagemap_entry(&pme, *pte, offset); + err = add_to_pagemap(addr, &pme, pm); + if (err) + return err; + } + + cond_resched(); + + return err; +} +#endif /* HUGETLB_PAGE */ + +/* + * /proc/pid/pagemap - an array mapping virtual pages to pfns + * + * For each page in the address space, this file contains one 64-bit entry + * consisting of the following: + * + * Bits 0-54 page frame number (PFN) if present + * Bits 0-4 swap type if swapped + * Bits 5-54 swap offset if swapped + * Bits 55-60 page shift (page size = 1<> PAGE_SHIFT); + pm.buffer = kmalloc(pm.len * PM_ENTRY_BYTES, GFP_TEMPORARY); + ret = -ENOMEM; + if (!pm.buffer) + goto out_task; + + mm = mm_access(task, PTRACE_MODE_READ); + ret = PTR_ERR(mm); + if (!mm || IS_ERR(mm)) + goto out_free; + + pagemap_walk.pmd_entry = pagemap_pte_range; + pagemap_walk.pte_hole = pagemap_pte_hole; +#ifdef CONFIG_HUGETLB_PAGE + pagemap_walk.hugetlb_entry = pagemap_hugetlb_range; +#endif + pagemap_walk.mm = mm; + pagemap_walk.private = ± + + src = *ppos; + svpfn = src / PM_ENTRY_BYTES; + start_vaddr = svpfn << PAGE_SHIFT; + end_vaddr = TASK_SIZE_OF(task); + + /* watch out for wraparound */ + if (svpfn > TASK_SIZE_OF(task) >> PAGE_SHIFT) + start_vaddr = end_vaddr; + + /* + * The odds are that this will stop walking way + * before end_vaddr, because the length of the + * user buffer is tracked in "pm", and the walk + * will stop when we hit the end of the buffer. + */ + ret = 0; + while (count && (start_vaddr < end_vaddr)) { + int len; + unsigned long end; + + pm.pos = 0; + end = (start_vaddr + PAGEMAP_WALK_SIZE) & PAGEMAP_WALK_MASK; + /* overflow ? */ + if (end < start_vaddr || end > end_vaddr) + end = end_vaddr; + down_read(&mm->mmap_sem); + ret = walk_page_range(start_vaddr, end, &pagemap_walk); + up_read(&mm->mmap_sem); + start_vaddr = end; + + len = min(count, PM_ENTRY_BYTES * pm.pos); + if (copy_to_user(buf, pm.buffer, len)) { + ret = -EFAULT; + goto out_mm; + } + copied += len; + buf += len; + count -= len; + } + *ppos += copied; + if (!ret || ret == PM_END_OF_BUFFER) + ret = copied; + +out_mm: + mmput(mm); +out_free: + kfree(pm.buffer); +out_task: + put_task_struct(task); +out: + return ret; +} + +const struct file_operations proc_pagemap_operations = { + .llseek = mem_lseek, /* borrow this */ + .read = pagemap_read, +}; +#endif /* CONFIG_PROC_PAGE_MONITOR */ + +#ifdef CONFIG_NUMA + +struct numa_maps { + struct vm_area_struct *vma; + unsigned long pages; + unsigned long anon; + unsigned long active; + unsigned long writeback; + unsigned long mapcount_max; + unsigned long dirty; + unsigned long swapcache; + unsigned long node[MAX_NUMNODES]; +}; + +struct numa_maps_private { + struct proc_maps_private proc_maps; + struct numa_maps md; +}; + +static void gather_stats(struct page *page, struct numa_maps *md, int pte_dirty, + unsigned long nr_pages) +{ + int count = page_mapcount(page); + + md->pages += nr_pages; + if (pte_dirty || PageDirty(page)) + md->dirty += nr_pages; + + if (PageSwapCache(page)) + md->swapcache += nr_pages; + + if (PageActive(page) || PageUnevictable(page)) + md->active += nr_pages; + + if (PageWriteback(page)) + md->writeback += nr_pages; + + if (PageAnon(page)) + md->anon += nr_pages; + + if (count > md->mapcount_max) + md->mapcount_max = count; + + md->node[page_to_nid(page)] += nr_pages; +} + +static struct page *can_gather_numa_stats(pte_t pte, struct vm_area_struct *vma, + unsigned long addr) +{ + struct page *page; + int nid; + + if (!pte_present(pte)) + return NULL; + + page = vm_normal_page(vma, addr, pte); + if (!page) + return NULL; + + if (PageReserved(page)) + return NULL; + + nid = page_to_nid(page); + if (!node_isset(nid, node_states[N_MEMORY])) + return NULL; + + return page; +} + +static int gather_pte_stats(pmd_t *pmd, unsigned long addr, + unsigned long end, struct mm_walk *walk) +{ + struct numa_maps *md; + spinlock_t *ptl; + pte_t *orig_pte; + pte_t *pte; + + md = walk->private; + + if (pmd_trans_huge_lock(pmd, md->vma) == 1) { + pte_t huge_pte = *(pte_t *)pmd; + struct page *page; + + page = can_gather_numa_stats(huge_pte, md->vma, addr); + if (page) + gather_stats(page, md, pte_dirty(huge_pte), + HPAGE_PMD_SIZE/PAGE_SIZE); + spin_unlock(&walk->mm->page_table_lock); + return 0; + } + + if (pmd_trans_unstable(pmd)) + return 0; + orig_pte = pte = pte_offset_map_lock(walk->mm, pmd, addr, &ptl); + do { + struct page *page = can_gather_numa_stats(*pte, md->vma, addr); + if (!page) + continue; + gather_stats(page, md, pte_dirty(*pte), 1); + + } while (pte++, addr += PAGE_SIZE, addr != end); + pte_unmap_unlock(orig_pte, ptl); + return 0; +} +#ifdef CONFIG_HUGETLB_PAGE +static int gather_hugetbl_stats(pte_t *pte, unsigned long hmask, + unsigned long addr, unsigned long end, struct mm_walk *walk) +{ + struct numa_maps *md; + struct page *page; + + if (pte_none(*pte)) + return 0; + + page = pte_page(*pte); + if (!page) + return 0; + + md = walk->private; + gather_stats(page, md, pte_dirty(*pte), 1); + return 0; +} + +#else +static int gather_hugetbl_stats(pte_t *pte, unsigned long hmask, + unsigned long addr, unsigned long end, struct mm_walk *walk) +{ + return 0; +} +#endif + +/* + * Display pages allocated per node and memory policy via /proc. + */ +static int show_numa_map(struct seq_file *m, void *v, int is_pid) +{ + struct numa_maps_private *numa_priv = m->private; + struct proc_maps_private *proc_priv = &numa_priv->proc_maps; + struct vm_area_struct *vma = v; + struct numa_maps *md = &numa_priv->md; + struct file *file = vma->vm_file; + struct task_struct *task = proc_priv->task; + struct mm_struct *mm = vma->vm_mm; + struct mm_walk walk = {}; + struct mempolicy *pol; + int n; + char buffer[50]; + + if (!mm) + return 0; + + /* Ensure we start with an empty set of numa_maps statistics. */ + memset(md, 0, sizeof(*md)); + + md->vma = vma; + + walk.hugetlb_entry = gather_hugetbl_stats; + walk.pmd_entry = gather_pte_stats; + walk.private = md; + walk.mm = mm; + + pol = get_vma_policy(task, vma, vma->vm_start); + mpol_to_str(buffer, sizeof(buffer), pol); + mpol_cond_put(pol); + + seq_printf(m, "%08lx %s", vma->vm_start, buffer); + + if (file) { + seq_printf(m, " file="); + seq_path(m, &file->f_path, "\n\t= "); + } else if (vma->vm_start <= mm->brk && vma->vm_end >= mm->start_brk) { + seq_printf(m, " heap"); + } else { + pid_t tid = vm_is_stack(task, vma, is_pid); + if (tid != 0) { + /* + * Thread stack in /proc/PID/task/TID/maps or + * the main process stack. + */ + if (!is_pid || (vma->vm_start <= mm->start_stack && + vma->vm_end >= mm->start_stack)) + seq_printf(m, " stack"); + else + seq_printf(m, " stack:%d", tid); + } + } + + if (is_vm_hugetlb_page(vma)) + seq_printf(m, " huge"); + + walk_page_range(vma->vm_start, vma->vm_end, &walk); + + if (!md->pages) + goto out; + + if (md->anon) + seq_printf(m, " anon=%lu", md->anon); + + if (md->dirty) + seq_printf(m, " dirty=%lu", md->dirty); + + if (md->pages != md->anon && md->pages != md->dirty) + seq_printf(m, " mapped=%lu", md->pages); + + if (md->mapcount_max > 1) + seq_printf(m, " mapmax=%lu", md->mapcount_max); + + if (md->swapcache) + seq_printf(m, " swapcache=%lu", md->swapcache); + + if (md->active < md->pages && !is_vm_hugetlb_page(vma)) + seq_printf(m, " active=%lu", md->active); + + if (md->writeback) + seq_printf(m, " writeback=%lu", md->writeback); + + for_each_node_state(n, N_MEMORY) + if (md->node[n]) + seq_printf(m, " N%d=%lu", n, md->node[n]); +out: + seq_putc(m, '\n'); + + if (m->count < m->size) + m->version = (vma != proc_priv->tail_vma) ? vma->vm_start : 0; + return 0; +} + +static int show_pid_numa_map(struct seq_file *m, void *v) +{ + return show_numa_map(m, v, 1); +} + +static int show_tid_numa_map(struct seq_file *m, void *v) +{ + return show_numa_map(m, v, 0); +} + +static const struct seq_operations proc_pid_numa_maps_op = { + .start = m_start, + .next = m_next, + .stop = m_stop, + .show = show_pid_numa_map, +}; + +static const struct seq_operations proc_tid_numa_maps_op = { + .start = m_start, + .next = m_next, + .stop = m_stop, + .show = show_tid_numa_map, +}; + +static int numa_maps_open(struct inode *inode, struct file *file, + const struct seq_operations *ops) +{ + struct numa_maps_private *priv; + int ret = -ENOMEM; + priv = kzalloc(sizeof(*priv), GFP_KERNEL); + if (priv) { + priv->proc_maps.pid = proc_pid(inode); + ret = seq_open(file, ops); + if (!ret) { + struct seq_file *m = file->private_data; + m->private = priv; + } else { + kfree(priv); + } + } + return ret; +} + +static int pid_numa_maps_open(struct inode *inode, struct file *file) +{ + return numa_maps_open(inode, file, &proc_pid_numa_maps_op); +} + +static int tid_numa_maps_open(struct inode *inode, struct file *file) +{ + return numa_maps_open(inode, file, &proc_tid_numa_maps_op); +} + +const struct file_operations proc_pid_numa_maps_operations = { + .open = pid_numa_maps_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; + +const struct file_operations proc_tid_numa_maps_operations = { + .open = tid_numa_maps_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; +#endif /* CONFIG_NUMA */ diff -Naur kernel.21.3.orig/fs/proc/task_nommu.c kernel.21.3.new/fs/proc/task_nommu.c --- kernel.21.3.orig/fs/proc/task_nommu.c 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/fs/proc/task_nommu.c 2015-04-25 23:28:45.312481603 +0000 @@ -149,7 +149,10 @@ file = vma->vm_file; if (file) { - struct inode *inode = file_inode(vma->vm_file); + struct inode *inode; + + file = vma_pr_or_file(file); + inode = file_inode(file); dev = inode->i_sb->s_dev; ino = inode->i_ino; pgoff = (loff_t)vma->vm_pgoff << PAGE_SHIFT; diff -Naur kernel.21.3.orig/fs/splice.c kernel.21.3.new/fs/splice.c --- kernel.21.3.orig/fs/splice.c 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/fs/splice.c 2015-04-25 23:28:53.719397545 +0000 @@ -1111,8 +1111,8 @@ /* * Attempt to initiate a splice from pipe to file. */ -static long do_splice_from(struct pipe_inode_info *pipe, struct file *out, - loff_t *ppos, size_t len, unsigned int flags) +long do_splice_from(struct pipe_inode_info *pipe, struct file *out, + loff_t *ppos, size_t len, unsigned int flags) { ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, loff_t *, size_t, unsigned int); @@ -1138,13 +1138,14 @@ file_end_write(out); return ret; } +EXPORT_SYMBOL(do_splice_from); /* * Attempt to initiate a splice from a file to a pipe. */ -static long do_splice_to(struct file *in, loff_t *ppos, - struct pipe_inode_info *pipe, size_t len, - unsigned int flags) +long do_splice_to(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, size_t len, + unsigned int flags) { ssize_t (*splice_read)(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int); @@ -1164,6 +1165,7 @@ return splice_read(in, ppos, pipe, len, flags); } +EXPORT_SYMBOL(do_splice_to); /** * splice_direct_to_actor - splices data directly between two non-pipes diff -Naur kernel.21.3.orig/fs/splice.c.orig kernel.21.3.new/fs/splice.c.orig --- kernel.21.3.orig/fs/splice.c.orig 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/fs/splice.c.orig 2015-04-25 23:28:26.644666407 +0000 @@ -0,0 +1,2096 @@ +/* + * "splice": joining two ropes together by interweaving their strands. + * + * This is the "extended pipe" functionality, where a pipe is used as + * an arbitrary in-memory buffer. Think of a pipe as a small kernel + * buffer that you can use to transfer data from one end to the other. + * + * The traditional unix read/write is extended with a "splice()" operation + * that transfers data buffers to or from a pipe buffer. + * + * Named by Larry McVoy, original implementation from Linus, extended by + * Jens to support splicing to files, network, direct splicing, etc and + * fixing lots of bugs. + * + * Copyright (C) 2005-2006 Jens Axboe + * Copyright (C) 2005-2006 Linus Torvalds + * Copyright (C) 2006 Ingo Molnar + * + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "internal.h" + +/* + * Attempt to steal a page from a pipe buffer. This should perhaps go into + * a vm helper function, it's already simplified quite a bit by the + * addition of remove_mapping(). If success is returned, the caller may + * attempt to reuse this page for another destination. + */ +static int page_cache_pipe_buf_steal(struct pipe_inode_info *pipe, + struct pipe_buffer *buf) +{ + struct page *page = buf->page; + struct address_space *mapping; + + lock_page(page); + + mapping = page_mapping(page); + if (mapping) { + WARN_ON(!PageUptodate(page)); + + /* + * At least for ext2 with nobh option, we need to wait on + * writeback completing on this page, since we'll remove it + * from the pagecache. Otherwise truncate wont wait on the + * page, allowing the disk blocks to be reused by someone else + * before we actually wrote our data to them. fs corruption + * ensues. + */ + wait_on_page_writeback(page); + + if (page_has_private(page) && + !try_to_release_page(page, GFP_KERNEL)) + goto out_unlock; + + /* + * If we succeeded in removing the mapping, set LRU flag + * and return good. + */ + if (remove_mapping(mapping, page)) { + buf->flags |= PIPE_BUF_FLAG_LRU; + return 0; + } + } + + /* + * Raced with truncate or failed to remove page from current + * address space, unlock and return failure. + */ +out_unlock: + unlock_page(page); + return 1; +} + +static void page_cache_pipe_buf_release(struct pipe_inode_info *pipe, + struct pipe_buffer *buf) +{ + page_cache_release(buf->page); + buf->flags &= ~PIPE_BUF_FLAG_LRU; +} + +/* + * Check whether the contents of buf is OK to access. Since the content + * is a page cache page, IO may be in flight. + */ +static int page_cache_pipe_buf_confirm(struct pipe_inode_info *pipe, + struct pipe_buffer *buf) +{ + struct page *page = buf->page; + int err; + + if (!PageUptodate(page)) { + lock_page(page); + + /* + * Page got truncated/unhashed. This will cause a 0-byte + * splice, if this is the first page. + */ + if (!page->mapping) { + err = -ENODATA; + goto error; + } + + /* + * Uh oh, read-error from disk. + */ + if (!PageUptodate(page)) { + err = -EIO; + goto error; + } + + /* + * Page is ok afterall, we are done. + */ + unlock_page(page); + } + + return 0; +error: + unlock_page(page); + return err; +} + +const struct pipe_buf_operations page_cache_pipe_buf_ops = { + .can_merge = 0, + .map = generic_pipe_buf_map, + .unmap = generic_pipe_buf_unmap, + .confirm = page_cache_pipe_buf_confirm, + .release = page_cache_pipe_buf_release, + .steal = page_cache_pipe_buf_steal, + .get = generic_pipe_buf_get, +}; + +static int user_page_pipe_buf_steal(struct pipe_inode_info *pipe, + struct pipe_buffer *buf) +{ + if (!(buf->flags & PIPE_BUF_FLAG_GIFT)) + return 1; + + buf->flags |= PIPE_BUF_FLAG_LRU; + return generic_pipe_buf_steal(pipe, buf); +} + +static const struct pipe_buf_operations user_page_pipe_buf_ops = { + .can_merge = 0, + .map = generic_pipe_buf_map, + .unmap = generic_pipe_buf_unmap, + .confirm = generic_pipe_buf_confirm, + .release = page_cache_pipe_buf_release, + .steal = user_page_pipe_buf_steal, + .get = generic_pipe_buf_get, +}; + +static void wakeup_pipe_readers(struct pipe_inode_info *pipe) +{ + smp_mb(); + if (waitqueue_active(&pipe->wait)) + wake_up_interruptible(&pipe->wait); + kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN); +} + +/** + * splice_to_pipe - fill passed data into a pipe + * @pipe: pipe to fill + * @spd: data to fill + * + * Description: + * @spd contains a map of pages and len/offset tuples, along with + * the struct pipe_buf_operations associated with these pages. This + * function will link that data to the pipe. + * + */ +ssize_t splice_to_pipe(struct pipe_inode_info *pipe, + struct splice_pipe_desc *spd) +{ + unsigned int spd_pages = spd->nr_pages; + int ret, do_wakeup, page_nr; + + ret = 0; + do_wakeup = 0; + page_nr = 0; + + pipe_lock(pipe); + + for (;;) { + if (!pipe->readers) { + send_sig(SIGPIPE, current, 0); + if (!ret) + ret = -EPIPE; + break; + } + + if (pipe->nrbufs < pipe->buffers) { + int newbuf = (pipe->curbuf + pipe->nrbufs) & (pipe->buffers - 1); + struct pipe_buffer *buf = pipe->bufs + newbuf; + + buf->page = spd->pages[page_nr]; + buf->offset = spd->partial[page_nr].offset; + buf->len = spd->partial[page_nr].len; + buf->private = spd->partial[page_nr].private; + buf->ops = spd->ops; + if (spd->flags & SPLICE_F_GIFT) + buf->flags |= PIPE_BUF_FLAG_GIFT; + + pipe->nrbufs++; + page_nr++; + ret += buf->len; + + if (pipe->files) + do_wakeup = 1; + + if (!--spd->nr_pages) + break; + if (pipe->nrbufs < pipe->buffers) + continue; + + break; + } + + if (spd->flags & SPLICE_F_NONBLOCK) { + if (!ret) + ret = -EAGAIN; + break; + } + + if (signal_pending(current)) { + if (!ret) + ret = -ERESTARTSYS; + break; + } + + if (do_wakeup) { + smp_mb(); + if (waitqueue_active(&pipe->wait)) + wake_up_interruptible_sync(&pipe->wait); + kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN); + do_wakeup = 0; + } + + pipe->waiting_writers++; + pipe_wait(pipe); + pipe->waiting_writers--; + } + + pipe_unlock(pipe); + + if (do_wakeup) + wakeup_pipe_readers(pipe); + + while (page_nr < spd_pages) + spd->spd_release(spd, page_nr++); + + return ret; +} + +void spd_release_page(struct splice_pipe_desc *spd, unsigned int i) +{ + page_cache_release(spd->pages[i]); +} + +/* + * Check if we need to grow the arrays holding pages and partial page + * descriptions. + */ +int splice_grow_spd(const struct pipe_inode_info *pipe, struct splice_pipe_desc *spd) +{ + unsigned int buffers = ACCESS_ONCE(pipe->buffers); + + spd->nr_pages_max = buffers; + if (buffers <= PIPE_DEF_BUFFERS) + return 0; + + spd->pages = kmalloc(buffers * sizeof(struct page *), GFP_KERNEL); + spd->partial = kmalloc(buffers * sizeof(struct partial_page), GFP_KERNEL); + + if (spd->pages && spd->partial) + return 0; + + kfree(spd->pages); + kfree(spd->partial); + return -ENOMEM; +} + +void splice_shrink_spd(struct splice_pipe_desc *spd) +{ + if (spd->nr_pages_max <= PIPE_DEF_BUFFERS) + return; + + kfree(spd->pages); + kfree(spd->partial); +} + +static int +__generic_file_splice_read(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, size_t len, + unsigned int flags) +{ + struct address_space *mapping = in->f_mapping; + unsigned int loff, nr_pages, req_pages; + struct page *pages[PIPE_DEF_BUFFERS]; + struct partial_page partial[PIPE_DEF_BUFFERS]; + struct page *page; + pgoff_t index, end_index; + loff_t isize; + int error, page_nr; + struct splice_pipe_desc spd = { + .pages = pages, + .partial = partial, + .nr_pages_max = PIPE_DEF_BUFFERS, + .flags = flags, + .ops = &page_cache_pipe_buf_ops, + .spd_release = spd_release_page, + }; + + if (splice_grow_spd(pipe, &spd)) + return -ENOMEM; + + index = *ppos >> PAGE_CACHE_SHIFT; + loff = *ppos & ~PAGE_CACHE_MASK; + req_pages = (len + loff + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + nr_pages = min(req_pages, spd.nr_pages_max); + + /* + * Lookup the (hopefully) full range of pages we need. + */ + spd.nr_pages = find_get_pages_contig(mapping, index, nr_pages, spd.pages); + index += spd.nr_pages; + + /* + * If find_get_pages_contig() returned fewer pages than we needed, + * readahead/allocate the rest and fill in the holes. + */ + if (spd.nr_pages < nr_pages) + page_cache_sync_readahead(mapping, &in->f_ra, in, + index, req_pages - spd.nr_pages); + + error = 0; + while (spd.nr_pages < nr_pages) { + /* + * Page could be there, find_get_pages_contig() breaks on + * the first hole. + */ + page = find_get_page(mapping, index); + if (!page) { + /* + * page didn't exist, allocate one. + */ + page = page_cache_alloc_cold(mapping); + if (!page) + break; + + error = add_to_page_cache_lru(page, mapping, index, + GFP_KERNEL); + if (unlikely(error)) { + page_cache_release(page); + if (error == -EEXIST) + continue; + break; + } + /* + * add_to_page_cache() locks the page, unlock it + * to avoid convoluting the logic below even more. + */ + unlock_page(page); + } + + spd.pages[spd.nr_pages++] = page; + index++; + } + + /* + * Now loop over the map and see if we need to start IO on any + * pages, fill in the partial map, etc. + */ + index = *ppos >> PAGE_CACHE_SHIFT; + nr_pages = spd.nr_pages; + spd.nr_pages = 0; + for (page_nr = 0; page_nr < nr_pages; page_nr++) { + unsigned int this_len; + + if (!len) + break; + + /* + * this_len is the max we'll use from this page + */ + this_len = min_t(unsigned long, len, PAGE_CACHE_SIZE - loff); + page = spd.pages[page_nr]; + + if (PageReadahead(page)) + page_cache_async_readahead(mapping, &in->f_ra, in, + page, index, req_pages - page_nr); + + /* + * If the page isn't uptodate, we may need to start io on it + */ + if (!PageUptodate(page)) { + lock_page(page); + + /* + * Page was truncated, or invalidated by the + * filesystem. Redo the find/create, but this time the + * page is kept locked, so there's no chance of another + * race with truncate/invalidate. + */ + if (!page->mapping) { + unlock_page(page); + page = find_or_create_page(mapping, index, + mapping_gfp_mask(mapping)); + + if (!page) { + error = -ENOMEM; + break; + } + page_cache_release(spd.pages[page_nr]); + spd.pages[page_nr] = page; + } + /* + * page was already under io and is now done, great + */ + if (PageUptodate(page)) { + unlock_page(page); + goto fill_it; + } + + /* + * need to read in the page + */ + error = mapping->a_ops->readpage(in, page); + if (unlikely(error)) { + /* + * We really should re-lookup the page here, + * but it complicates things a lot. Instead + * lets just do what we already stored, and + * we'll get it the next time we are called. + */ + if (error == AOP_TRUNCATED_PAGE) + error = 0; + + break; + } + } +fill_it: + /* + * i_size must be checked after PageUptodate. + */ + isize = i_size_read(mapping->host); + end_index = (isize - 1) >> PAGE_CACHE_SHIFT; + if (unlikely(!isize || index > end_index)) + break; + + /* + * if this is the last page, see if we need to shrink + * the length and stop + */ + if (end_index == index) { + unsigned int plen; + + /* + * max good bytes in this page + */ + plen = ((isize - 1) & ~PAGE_CACHE_MASK) + 1; + if (plen <= loff) + break; + + /* + * force quit after adding this page + */ + this_len = min(this_len, plen - loff); + len = this_len; + } + + spd.partial[page_nr].offset = loff; + spd.partial[page_nr].len = this_len; + len -= this_len; + loff = 0; + spd.nr_pages++; + index++; + } + + /* + * Release any pages at the end, if we quit early. 'page_nr' is how far + * we got, 'nr_pages' is how many pages are in the map. + */ + while (page_nr < nr_pages) + page_cache_release(spd.pages[page_nr++]); + in->f_ra.prev_pos = (loff_t)index << PAGE_CACHE_SHIFT; + + if (spd.nr_pages) + error = splice_to_pipe(pipe, &spd); + + splice_shrink_spd(&spd); + return error; +} + +/** + * generic_file_splice_read - splice data from file to a pipe + * @in: file to splice from + * @ppos: position in @in + * @pipe: pipe to splice to + * @len: number of bytes to splice + * @flags: splice modifier flags + * + * Description: + * Will read pages from given file and fill them into a pipe. Can be + * used as long as the address_space operations for the source implements + * a readpage() hook. + * + */ +ssize_t generic_file_splice_read(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, size_t len, + unsigned int flags) +{ + loff_t isize, left; + int ret; + + isize = i_size_read(in->f_mapping->host); + if (unlikely(*ppos >= isize)) + return 0; + + left = isize - *ppos; + if (unlikely(left < len)) + len = left; + + ret = __generic_file_splice_read(in, ppos, pipe, len, flags); + if (ret > 0) { + *ppos += ret; + file_accessed(in); + } + + return ret; +} +EXPORT_SYMBOL(generic_file_splice_read); + +static const struct pipe_buf_operations default_pipe_buf_ops = { + .can_merge = 0, + .map = generic_pipe_buf_map, + .unmap = generic_pipe_buf_unmap, + .confirm = generic_pipe_buf_confirm, + .release = generic_pipe_buf_release, + .steal = generic_pipe_buf_steal, + .get = generic_pipe_buf_get, +}; + +static int generic_pipe_buf_nosteal(struct pipe_inode_info *pipe, + struct pipe_buffer *buf) +{ + return 1; +} + +/* Pipe buffer operations for a socket and similar. */ +const struct pipe_buf_operations nosteal_pipe_buf_ops = { + .can_merge = 0, + .map = generic_pipe_buf_map, + .unmap = generic_pipe_buf_unmap, + .confirm = generic_pipe_buf_confirm, + .release = generic_pipe_buf_release, + .steal = generic_pipe_buf_nosteal, + .get = generic_pipe_buf_get, +}; +EXPORT_SYMBOL(nosteal_pipe_buf_ops); + +static ssize_t kernel_readv(struct file *file, const struct iovec *vec, + unsigned long vlen, loff_t offset) +{ + mm_segment_t old_fs; + loff_t pos = offset; + ssize_t res; + + old_fs = get_fs(); + set_fs(get_ds()); + /* The cast to a user pointer is valid due to the set_fs() */ + res = vfs_readv(file, (const struct iovec __user *)vec, vlen, &pos); + set_fs(old_fs); + + return res; +} + +ssize_t kernel_write(struct file *file, const char *buf, size_t count, + loff_t pos) +{ + mm_segment_t old_fs; + ssize_t res; + + old_fs = get_fs(); + set_fs(get_ds()); + /* The cast to a user pointer is valid due to the set_fs() */ + res = vfs_write(file, (__force const char __user *)buf, count, &pos); + set_fs(old_fs); + + return res; +} +EXPORT_SYMBOL(kernel_write); + +ssize_t default_file_splice_read(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, size_t len, + unsigned int flags) +{ + unsigned int nr_pages; + unsigned int nr_freed; + size_t offset; + struct page *pages[PIPE_DEF_BUFFERS]; + struct partial_page partial[PIPE_DEF_BUFFERS]; + struct iovec *vec, __vec[PIPE_DEF_BUFFERS]; + ssize_t res; + size_t this_len; + int error; + int i; + struct splice_pipe_desc spd = { + .pages = pages, + .partial = partial, + .nr_pages_max = PIPE_DEF_BUFFERS, + .flags = flags, + .ops = &default_pipe_buf_ops, + .spd_release = spd_release_page, + }; + + if (splice_grow_spd(pipe, &spd)) + return -ENOMEM; + + res = -ENOMEM; + vec = __vec; + if (spd.nr_pages_max > PIPE_DEF_BUFFERS) { + vec = kmalloc(spd.nr_pages_max * sizeof(struct iovec), GFP_KERNEL); + if (!vec) + goto shrink_ret; + } + + offset = *ppos & ~PAGE_CACHE_MASK; + nr_pages = (len + offset + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + + for (i = 0; i < nr_pages && i < spd.nr_pages_max && len; i++) { + struct page *page; + + page = alloc_page(GFP_USER); + error = -ENOMEM; + if (!page) + goto err; + + this_len = min_t(size_t, len, PAGE_CACHE_SIZE - offset); + vec[i].iov_base = (void __user *) page_address(page); + vec[i].iov_len = this_len; + spd.pages[i] = page; + spd.nr_pages++; + len -= this_len; + offset = 0; + } + + res = kernel_readv(in, vec, spd.nr_pages, *ppos); + if (res < 0) { + error = res; + goto err; + } + + error = 0; + if (!res) + goto err; + + nr_freed = 0; + for (i = 0; i < spd.nr_pages; i++) { + this_len = min_t(size_t, vec[i].iov_len, res); + spd.partial[i].offset = 0; + spd.partial[i].len = this_len; + if (!this_len) { + __free_page(spd.pages[i]); + spd.pages[i] = NULL; + nr_freed++; + } + res -= this_len; + } + spd.nr_pages -= nr_freed; + + res = splice_to_pipe(pipe, &spd); + if (res > 0) + *ppos += res; + +shrink_ret: + if (vec != __vec) + kfree(vec); + splice_shrink_spd(&spd); + return res; + +err: + for (i = 0; i < spd.nr_pages; i++) + __free_page(spd.pages[i]); + + res = error; + goto shrink_ret; +} +EXPORT_SYMBOL(default_file_splice_read); + +/* + * Send 'sd->len' bytes to socket from 'sd->file' at position 'sd->pos' + * using sendpage(). Return the number of bytes sent. + */ +static int pipe_to_sendpage(struct pipe_inode_info *pipe, + struct pipe_buffer *buf, struct splice_desc *sd) +{ + struct file *file = sd->u.file; + loff_t pos = sd->pos; + int more; + + if (!likely(file->f_op && file->f_op->sendpage)) + return -EINVAL; + + more = (sd->flags & SPLICE_F_MORE) ? MSG_MORE : 0; + + if (sd->len < sd->total_len && pipe->nrbufs > 1) + more |= MSG_SENDPAGE_NOTLAST; + + return file->f_op->sendpage(file, buf->page, buf->offset, + sd->len, &pos, more); +} + +/* + * This is a little more tricky than the file -> pipe splicing. There are + * basically three cases: + * + * - Destination page already exists in the address space and there + * are users of it. For that case we have no other option that + * copying the data. Tough luck. + * - Destination page already exists in the address space, but there + * are no users of it. Make sure it's uptodate, then drop it. Fall + * through to last case. + * - Destination page does not exist, we can add the pipe page to + * the page cache and avoid the copy. + * + * If asked to move pages to the output file (SPLICE_F_MOVE is set in + * sd->flags), we attempt to migrate pages from the pipe to the output + * file address space page cache. This is possible if no one else has + * the pipe page referenced outside of the pipe and page cache. If + * SPLICE_F_MOVE isn't set, or we cannot move the page, we simply create + * a new page in the output file page cache and fill/dirty that. + */ +int pipe_to_file(struct pipe_inode_info *pipe, struct pipe_buffer *buf, + struct splice_desc *sd) +{ + struct file *file = sd->u.file; + struct address_space *mapping = file->f_mapping; + unsigned int offset, this_len; + struct page *page; + void *fsdata; + int ret; + + offset = sd->pos & ~PAGE_CACHE_MASK; + + this_len = sd->len; + if (this_len + offset > PAGE_CACHE_SIZE) + this_len = PAGE_CACHE_SIZE - offset; + + ret = pagecache_write_begin(file, mapping, sd->pos, this_len, + AOP_FLAG_UNINTERRUPTIBLE, &page, &fsdata); + if (unlikely(ret)) + goto out; + + if (buf->page != page) { + char *src = buf->ops->map(pipe, buf, 1); + char *dst = kmap_atomic(page); + + memcpy(dst + offset, src + buf->offset, this_len); + flush_dcache_page(page); + kunmap_atomic(dst); + buf->ops->unmap(pipe, buf, src); + } + ret = pagecache_write_end(file, mapping, sd->pos, this_len, this_len, + page, fsdata); +out: + return ret; +} +EXPORT_SYMBOL(pipe_to_file); + +static void wakeup_pipe_writers(struct pipe_inode_info *pipe) +{ + smp_mb(); + if (waitqueue_active(&pipe->wait)) + wake_up_interruptible(&pipe->wait); + kill_fasync(&pipe->fasync_writers, SIGIO, POLL_OUT); +} + +/** + * splice_from_pipe_feed - feed available data from a pipe to a file + * @pipe: pipe to splice from + * @sd: information to @actor + * @actor: handler that splices the data + * + * Description: + * This function loops over the pipe and calls @actor to do the + * actual moving of a single struct pipe_buffer to the desired + * destination. It returns when there's no more buffers left in + * the pipe or if the requested number of bytes (@sd->total_len) + * have been copied. It returns a positive number (one) if the + * pipe needs to be filled with more data, zero if the required + * number of bytes have been copied and -errno on error. + * + * This, together with splice_from_pipe_{begin,end,next}, may be + * used to implement the functionality of __splice_from_pipe() when + * locking is required around copying the pipe buffers to the + * destination. + */ +int splice_from_pipe_feed(struct pipe_inode_info *pipe, struct splice_desc *sd, + splice_actor *actor) +{ + int ret; + + while (pipe->nrbufs) { + struct pipe_buffer *buf = pipe->bufs + pipe->curbuf; + const struct pipe_buf_operations *ops = buf->ops; + + sd->len = buf->len; + if (sd->len > sd->total_len) + sd->len = sd->total_len; + + ret = buf->ops->confirm(pipe, buf); + if (unlikely(ret)) { + if (ret == -ENODATA) + ret = 0; + return ret; + } + + ret = actor(pipe, buf, sd); + if (ret <= 0) + return ret; + + buf->offset += ret; + buf->len -= ret; + + sd->num_spliced += ret; + sd->len -= ret; + sd->pos += ret; + sd->total_len -= ret; + + if (!buf->len) { + buf->ops = NULL; + ops->release(pipe, buf); + pipe->curbuf = (pipe->curbuf + 1) & (pipe->buffers - 1); + pipe->nrbufs--; + if (pipe->files) + sd->need_wakeup = true; + } + + if (!sd->total_len) + return 0; + } + + return 1; +} +EXPORT_SYMBOL(splice_from_pipe_feed); + +/** + * splice_from_pipe_next - wait for some data to splice from + * @pipe: pipe to splice from + * @sd: information about the splice operation + * + * Description: + * This function will wait for some data and return a positive + * value (one) if pipe buffers are available. It will return zero + * or -errno if no more data needs to be spliced. + */ +int splice_from_pipe_next(struct pipe_inode_info *pipe, struct splice_desc *sd) +{ + while (!pipe->nrbufs) { + if (!pipe->writers) + return 0; + + if (!pipe->waiting_writers && sd->num_spliced) + return 0; + + if (sd->flags & SPLICE_F_NONBLOCK) + return -EAGAIN; + + if (signal_pending(current)) + return -ERESTARTSYS; + + if (sd->need_wakeup) { + wakeup_pipe_writers(pipe); + sd->need_wakeup = false; + } + + pipe_wait(pipe); + } + + return 1; +} +EXPORT_SYMBOL(splice_from_pipe_next); + +/** + * splice_from_pipe_begin - start splicing from pipe + * @sd: information about the splice operation + * + * Description: + * This function should be called before a loop containing + * splice_from_pipe_next() and splice_from_pipe_feed() to + * initialize the necessary fields of @sd. + */ +void splice_from_pipe_begin(struct splice_desc *sd) +{ + sd->num_spliced = 0; + sd->need_wakeup = false; +} +EXPORT_SYMBOL(splice_from_pipe_begin); + +/** + * splice_from_pipe_end - finish splicing from pipe + * @pipe: pipe to splice from + * @sd: information about the splice operation + * + * Description: + * This function will wake up pipe writers if necessary. It should + * be called after a loop containing splice_from_pipe_next() and + * splice_from_pipe_feed(). + */ +void splice_from_pipe_end(struct pipe_inode_info *pipe, struct splice_desc *sd) +{ + if (sd->need_wakeup) + wakeup_pipe_writers(pipe); +} +EXPORT_SYMBOL(splice_from_pipe_end); + +/** + * __splice_from_pipe - splice data from a pipe to given actor + * @pipe: pipe to splice from + * @sd: information to @actor + * @actor: handler that splices the data + * + * Description: + * This function does little more than loop over the pipe and call + * @actor to do the actual moving of a single struct pipe_buffer to + * the desired destination. See pipe_to_file, pipe_to_sendpage, or + * pipe_to_user. + * + */ +ssize_t __splice_from_pipe(struct pipe_inode_info *pipe, struct splice_desc *sd, + splice_actor *actor) +{ + int ret; + + splice_from_pipe_begin(sd); + do { + ret = splice_from_pipe_next(pipe, sd); + if (ret > 0) + ret = splice_from_pipe_feed(pipe, sd, actor); + } while (ret > 0); + splice_from_pipe_end(pipe, sd); + + return sd->num_spliced ? sd->num_spliced : ret; +} +EXPORT_SYMBOL(__splice_from_pipe); + +/** + * splice_from_pipe - splice data from a pipe to a file + * @pipe: pipe to splice from + * @out: file to splice to + * @ppos: position in @out + * @len: how many bytes to splice + * @flags: splice modifier flags + * @actor: handler that splices the data + * + * Description: + * See __splice_from_pipe. This function locks the pipe inode, + * otherwise it's identical to __splice_from_pipe(). + * + */ +ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, + loff_t *ppos, size_t len, unsigned int flags, + splice_actor *actor) +{ + ssize_t ret; + struct splice_desc sd = { + .total_len = len, + .flags = flags, + .pos = *ppos, + .u.file = out, + }; + + pipe_lock(pipe); + ret = __splice_from_pipe(pipe, &sd, actor); + pipe_unlock(pipe); + + return ret; +} + +/** + * generic_file_splice_write - splice data from a pipe to a file + * @pipe: pipe info + * @out: file to write to + * @ppos: position in @out + * @len: number of bytes to splice + * @flags: splice modifier flags + * + * Description: + * Will either move or copy pages (determined by @flags options) from + * the given pipe inode to the given file. + * + */ +ssize_t +generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out, + loff_t *ppos, size_t len, unsigned int flags) +{ + struct address_space *mapping = out->f_mapping; + struct inode *inode = mapping->host; + struct splice_desc sd = { + .total_len = len, + .flags = flags, + .pos = *ppos, + .u.file = out, + }; + ssize_t ret; + + pipe_lock(pipe); + + splice_from_pipe_begin(&sd); + do { + ret = splice_from_pipe_next(pipe, &sd); + if (ret <= 0) + break; + + mutex_lock_nested(&inode->i_mutex, I_MUTEX_CHILD); + ret = file_remove_suid(out); + if (!ret) { + ret = file_update_time(out); + if (!ret) + ret = splice_from_pipe_feed(pipe, &sd, + pipe_to_file); + } + mutex_unlock(&inode->i_mutex); + } while (ret > 0); + splice_from_pipe_end(pipe, &sd); + + pipe_unlock(pipe); + + if (sd.num_spliced) + ret = sd.num_spliced; + + if (ret > 0) { + int err; + + err = generic_write_sync(out, *ppos, ret); + if (err) + ret = err; + else + *ppos += ret; + balance_dirty_pages_ratelimited(mapping); + } + + return ret; +} + +EXPORT_SYMBOL(generic_file_splice_write); + +static int write_pipe_buf(struct pipe_inode_info *pipe, struct pipe_buffer *buf, + struct splice_desc *sd) +{ + int ret; + void *data; + loff_t tmp = sd->pos; + + data = buf->ops->map(pipe, buf, 0); + ret = __kernel_write(sd->u.file, data + buf->offset, sd->len, &tmp); + buf->ops->unmap(pipe, buf, data); + + return ret; +} + +static ssize_t default_file_splice_write(struct pipe_inode_info *pipe, + struct file *out, loff_t *ppos, + size_t len, unsigned int flags) +{ + ssize_t ret; + + ret = splice_from_pipe(pipe, out, ppos, len, flags, write_pipe_buf); + if (ret > 0) + *ppos += ret; + + return ret; +} + +/** + * generic_splice_sendpage - splice data from a pipe to a socket + * @pipe: pipe to splice from + * @out: socket to write to + * @ppos: position in @out + * @len: number of bytes to splice + * @flags: splice modifier flags + * + * Description: + * Will send @len bytes from the pipe to a network socket. No data copying + * is involved. + * + */ +ssize_t generic_splice_sendpage(struct pipe_inode_info *pipe, struct file *out, + loff_t *ppos, size_t len, unsigned int flags) +{ + return splice_from_pipe(pipe, out, ppos, len, flags, pipe_to_sendpage); +} + +EXPORT_SYMBOL(generic_splice_sendpage); + +/* + * Attempt to initiate a splice from pipe to file. + */ +long do_splice_from(struct pipe_inode_info *pipe, struct file *out, + loff_t *ppos, size_t len, unsigned int flags) +{ + ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, + loff_t *, size_t, unsigned int); + int ret; + + if (unlikely(!(out->f_mode & FMODE_WRITE))) + return -EBADF; + + if (unlikely(out->f_flags & O_APPEND)) + return -EINVAL; + + ret = rw_verify_area(WRITE, out, ppos, len); + if (unlikely(ret < 0)) + return ret; + + if (out->f_op && out->f_op->splice_write) + splice_write = out->f_op->splice_write; + else + splice_write = default_file_splice_write; + + file_start_write(out); + ret = splice_write(pipe, out, ppos, len, flags); + file_end_write(out); + return ret; +} + +/* + * Attempt to initiate a splice from a file to a pipe. + */ +long do_splice_to(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, size_t len, + unsigned int flags) +{ + ssize_t (*splice_read)(struct file *, loff_t *, + struct pipe_inode_info *, size_t, unsigned int); + int ret; + + if (unlikely(!(in->f_mode & FMODE_READ))) + return -EBADF; + + ret = rw_verify_area(READ, in, ppos, len); + if (unlikely(ret < 0)) + return ret; + + if (in->f_op && in->f_op->splice_read) + splice_read = in->f_op->splice_read; + else + splice_read = default_file_splice_read; + + return splice_read(in, ppos, pipe, len, flags); +} + +/** + * splice_direct_to_actor - splices data directly between two non-pipes + * @in: file to splice from + * @sd: actor information on where to splice to + * @actor: handles the data splicing + * + * Description: + * This is a special case helper to splice directly between two + * points, without requiring an explicit pipe. Internally an allocated + * pipe is cached in the process, and reused during the lifetime of + * that process. + * + */ +ssize_t splice_direct_to_actor(struct file *in, struct splice_desc *sd, + splice_direct_actor *actor) +{ + struct pipe_inode_info *pipe; + long ret, bytes; + umode_t i_mode; + size_t len; + int i, flags; + + /* + * We require the input being a regular file, as we don't want to + * randomly drop data for eg socket -> socket splicing. Use the + * piped splicing for that! + */ + i_mode = file_inode(in)->i_mode; + if (unlikely(!S_ISREG(i_mode) && !S_ISBLK(i_mode))) + return -EINVAL; + + /* + * neither in nor out is a pipe, setup an internal pipe attached to + * 'out' and transfer the wanted data from 'in' to 'out' through that + */ + pipe = current->splice_pipe; + if (unlikely(!pipe)) { + pipe = alloc_pipe_info(); + if (!pipe) + return -ENOMEM; + + /* + * We don't have an immediate reader, but we'll read the stuff + * out of the pipe right after the splice_to_pipe(). So set + * PIPE_READERS appropriately. + */ + pipe->readers = 1; + + current->splice_pipe = pipe; + } + + /* + * Do the splice. + */ + ret = 0; + bytes = 0; + len = sd->total_len; + flags = sd->flags; + + /* + * Don't block on output, we have to drain the direct pipe. + */ + sd->flags &= ~SPLICE_F_NONBLOCK; + + while (len) { + size_t read_len; + loff_t pos = sd->pos, prev_pos = pos; + + ret = do_splice_to(in, &pos, pipe, len, flags); + if (unlikely(ret <= 0)) + goto out_release; + + read_len = ret; + sd->total_len = read_len; + + /* + * NOTE: nonblocking mode only applies to the input. We + * must not do the output in nonblocking mode as then we + * could get stuck data in the internal pipe: + */ + ret = actor(pipe, sd); + if (unlikely(ret <= 0)) { + sd->pos = prev_pos; + goto out_release; + } + + bytes += ret; + len -= ret; + sd->pos = pos; + + if (ret < read_len) { + sd->pos = prev_pos + ret; + goto out_release; + } + } + +done: + pipe->nrbufs = pipe->curbuf = 0; + file_accessed(in); + return bytes; + +out_release: + /* + * If we did an incomplete transfer we must release + * the pipe buffers in question: + */ + for (i = 0; i < pipe->buffers; i++) { + struct pipe_buffer *buf = pipe->bufs + i; + + if (buf->ops) { + buf->ops->release(pipe, buf); + buf->ops = NULL; + } + } + + if (!bytes) + bytes = ret; + + goto done; +} +EXPORT_SYMBOL(splice_direct_to_actor); + +static int direct_splice_actor(struct pipe_inode_info *pipe, + struct splice_desc *sd) +{ + struct file *file = sd->u.file; + + return do_splice_from(pipe, file, sd->opos, sd->total_len, + sd->flags); +} + +/** + * do_splice_direct - splices data directly between two files + * @in: file to splice from + * @ppos: input file offset + * @out: file to splice to + * @opos: output file offset + * @len: number of bytes to splice + * @flags: splice modifier flags + * + * Description: + * For use by do_sendfile(). splice can easily emulate sendfile, but + * doing it in the application would incur an extra system call + * (splice in + splice out, as compared to just sendfile()). So this helper + * can splice directly through a process-private pipe. + * + */ +long do_splice_direct(struct file *in, loff_t *ppos, struct file *out, + loff_t *opos, size_t len, unsigned int flags) +{ + struct splice_desc sd = { + .len = len, + .total_len = len, + .flags = flags, + .pos = *ppos, + .u.file = out, + .opos = opos, + }; + long ret; + + ret = splice_direct_to_actor(in, &sd, direct_splice_actor); + if (ret > 0) + *ppos = sd.pos; + + return ret; +} + +static int splice_pipe_to_pipe(struct pipe_inode_info *ipipe, + struct pipe_inode_info *opipe, + size_t len, unsigned int flags); + +/* + * Determine where to splice to/from. + */ +static long do_splice(struct file *in, loff_t __user *off_in, + struct file *out, loff_t __user *off_out, + size_t len, unsigned int flags) +{ + struct pipe_inode_info *ipipe; + struct pipe_inode_info *opipe; + loff_t offset; + long ret; + + ipipe = get_pipe_info(in); + opipe = get_pipe_info(out); + + if (ipipe && opipe) { + if (off_in || off_out) + return -ESPIPE; + + if (!(in->f_mode & FMODE_READ)) + return -EBADF; + + if (!(out->f_mode & FMODE_WRITE)) + return -EBADF; + + /* Splicing to self would be fun, but... */ + if (ipipe == opipe) + return -EINVAL; + + return splice_pipe_to_pipe(ipipe, opipe, len, flags); + } + + if (ipipe) { + if (off_in) + return -ESPIPE; + if (off_out) { + if (!(out->f_mode & FMODE_PWRITE)) + return -EINVAL; + if (copy_from_user(&offset, off_out, sizeof(loff_t))) + return -EFAULT; + } else { + offset = out->f_pos; + } + + ret = do_splice_from(ipipe, out, &offset, len, flags); + + if (!off_out) + out->f_pos = offset; + else if (copy_to_user(off_out, &offset, sizeof(loff_t))) + ret = -EFAULT; + + return ret; + } + + if (opipe) { + if (off_out) + return -ESPIPE; + if (off_in) { + if (!(in->f_mode & FMODE_PREAD)) + return -EINVAL; + if (copy_from_user(&offset, off_in, sizeof(loff_t))) + return -EFAULT; + } else { + offset = in->f_pos; + } + + ret = do_splice_to(in, &offset, opipe, len, flags); + + if (!off_in) + in->f_pos = offset; + else if (copy_to_user(off_in, &offset, sizeof(loff_t))) + ret = -EFAULT; + + return ret; + } + + return -EINVAL; +} + +/* + * Map an iov into an array of pages and offset/length tupples. With the + * partial_page structure, we can map several non-contiguous ranges into + * our ones pages[] map instead of splitting that operation into pieces. + * Could easily be exported as a generic helper for other users, in which + * case one would probably want to add a 'max_nr_pages' parameter as well. + */ +static int get_iovec_page_array(const struct iovec __user *iov, + unsigned int nr_vecs, struct page **pages, + struct partial_page *partial, bool aligned, + unsigned int pipe_buffers) +{ + int buffers = 0, error = 0; + + while (nr_vecs) { + unsigned long off, npages; + struct iovec entry; + void __user *base; + size_t len; + int i; + + error = -EFAULT; + if (copy_from_user(&entry, iov, sizeof(entry))) + break; + + base = entry.iov_base; + len = entry.iov_len; + + /* + * Sanity check this iovec. 0 read succeeds. + */ + error = 0; + if (unlikely(!len)) + break; + error = -EFAULT; + if (!access_ok(VERIFY_READ, base, len)) + break; + + /* + * Get this base offset and number of pages, then map + * in the user pages. + */ + off = (unsigned long) base & ~PAGE_MASK; + + /* + * If asked for alignment, the offset must be zero and the + * length a multiple of the PAGE_SIZE. + */ + error = -EINVAL; + if (aligned && (off || len & ~PAGE_MASK)) + break; + + npages = (off + len + PAGE_SIZE - 1) >> PAGE_SHIFT; + if (npages > pipe_buffers - buffers) + npages = pipe_buffers - buffers; + + error = get_user_pages_fast((unsigned long)base, npages, + 0, &pages[buffers]); + + if (unlikely(error <= 0)) + break; + + /* + * Fill this contiguous range into the partial page map. + */ + for (i = 0; i < error; i++) { + const int plen = min_t(size_t, len, PAGE_SIZE - off); + + partial[buffers].offset = off; + partial[buffers].len = plen; + + off = 0; + len -= plen; + buffers++; + } + + /* + * We didn't complete this iov, stop here since it probably + * means we have to move some of this into a pipe to + * be able to continue. + */ + if (len) + break; + + /* + * Don't continue if we mapped fewer pages than we asked for, + * or if we mapped the max number of pages that we have + * room for. + */ + if (error < npages || buffers == pipe_buffers) + break; + + nr_vecs--; + iov++; + } + + if (buffers) + return buffers; + + return error; +} + +static int pipe_to_user(struct pipe_inode_info *pipe, struct pipe_buffer *buf, + struct splice_desc *sd) +{ + char *src; + int ret; + + /* + * See if we can use the atomic maps, by prefaulting in the + * pages and doing an atomic copy + */ + if (!fault_in_pages_writeable(sd->u.userptr, sd->len)) { + src = buf->ops->map(pipe, buf, 1); + ret = __copy_to_user_inatomic(sd->u.userptr, src + buf->offset, + sd->len); + buf->ops->unmap(pipe, buf, src); + if (!ret) { + ret = sd->len; + goto out; + } + } + + /* + * No dice, use slow non-atomic map and copy + */ + src = buf->ops->map(pipe, buf, 0); + + ret = sd->len; + if (copy_to_user(sd->u.userptr, src + buf->offset, sd->len)) + ret = -EFAULT; + + buf->ops->unmap(pipe, buf, src); +out: + if (ret > 0) + sd->u.userptr += ret; + return ret; +} + +/* + * For lack of a better implementation, implement vmsplice() to userspace + * as a simple copy of the pipes pages to the user iov. + */ +static long vmsplice_to_user(struct file *file, const struct iovec __user *iov, + unsigned long nr_segs, unsigned int flags) +{ + struct pipe_inode_info *pipe; + struct splice_desc sd; + ssize_t size; + int error; + long ret; + + pipe = get_pipe_info(file); + if (!pipe) + return -EBADF; + + pipe_lock(pipe); + + error = ret = 0; + while (nr_segs) { + void __user *base; + size_t len; + + /* + * Get user address base and length for this iovec. + */ + error = get_user(base, &iov->iov_base); + if (unlikely(error)) + break; + error = get_user(len, &iov->iov_len); + if (unlikely(error)) + break; + + /* + * Sanity check this iovec. 0 read succeeds. + */ + if (unlikely(!len)) + break; + if (unlikely(!base)) { + error = -EFAULT; + break; + } + + if (unlikely(!access_ok(VERIFY_WRITE, base, len))) { + error = -EFAULT; + break; + } + + sd.len = 0; + sd.total_len = len; + sd.flags = flags; + sd.u.userptr = base; + sd.pos = 0; + + size = __splice_from_pipe(pipe, &sd, pipe_to_user); + if (size < 0) { + if (!ret) + ret = size; + + break; + } + + ret += size; + + if (size < len) + break; + + nr_segs--; + iov++; + } + + pipe_unlock(pipe); + + if (!ret) + ret = error; + + return ret; +} + +/* + * vmsplice splices a user address range into a pipe. It can be thought of + * as splice-from-memory, where the regular splice is splice-from-file (or + * to file). In both cases the output is a pipe, naturally. + */ +static long vmsplice_to_pipe(struct file *file, const struct iovec __user *iov, + unsigned long nr_segs, unsigned int flags) +{ + struct pipe_inode_info *pipe; + struct page *pages[PIPE_DEF_BUFFERS]; + struct partial_page partial[PIPE_DEF_BUFFERS]; + struct splice_pipe_desc spd = { + .pages = pages, + .partial = partial, + .nr_pages_max = PIPE_DEF_BUFFERS, + .flags = flags, + .ops = &user_page_pipe_buf_ops, + .spd_release = spd_release_page, + }; + long ret; + + pipe = get_pipe_info(file); + if (!pipe) + return -EBADF; + + if (splice_grow_spd(pipe, &spd)) + return -ENOMEM; + + spd.nr_pages = get_iovec_page_array(iov, nr_segs, spd.pages, + spd.partial, false, + spd.nr_pages_max); + if (spd.nr_pages <= 0) + ret = spd.nr_pages; + else + ret = splice_to_pipe(pipe, &spd); + + splice_shrink_spd(&spd); + return ret; +} + +/* + * Note that vmsplice only really supports true splicing _from_ user memory + * to a pipe, not the other way around. Splicing from user memory is a simple + * operation that can be supported without any funky alignment restrictions + * or nasty vm tricks. We simply map in the user memory and fill them into + * a pipe. The reverse isn't quite as easy, though. There are two possible + * solutions for that: + * + * - memcpy() the data internally, at which point we might as well just + * do a regular read() on the buffer anyway. + * - Lots of nasty vm tricks, that are neither fast nor flexible (it + * has restriction limitations on both ends of the pipe). + * + * Currently we punt and implement it as a normal copy, see pipe_to_user(). + * + */ +SYSCALL_DEFINE4(vmsplice, int, fd, const struct iovec __user *, iov, + unsigned long, nr_segs, unsigned int, flags) +{ + struct fd f; + long error; + + if (unlikely(nr_segs > UIO_MAXIOV)) + return -EINVAL; + else if (unlikely(!nr_segs)) + return 0; + + error = -EBADF; + f = fdget(fd); + if (f.file) { + if (f.file->f_mode & FMODE_WRITE) + error = vmsplice_to_pipe(f.file, iov, nr_segs, flags); + else if (f.file->f_mode & FMODE_READ) + error = vmsplice_to_user(f.file, iov, nr_segs, flags); + + fdput(f); + } + + return error; +} + +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE4(vmsplice, int, fd, const struct compat_iovec __user *, iov32, + unsigned int, nr_segs, unsigned int, flags) +{ + unsigned i; + struct iovec __user *iov; + if (nr_segs > UIO_MAXIOV) + return -EINVAL; + iov = compat_alloc_user_space(nr_segs * sizeof(struct iovec)); + for (i = 0; i < nr_segs; i++) { + struct compat_iovec v; + if (get_user(v.iov_base, &iov32[i].iov_base) || + get_user(v.iov_len, &iov32[i].iov_len) || + put_user(compat_ptr(v.iov_base), &iov[i].iov_base) || + put_user(v.iov_len, &iov[i].iov_len)) + return -EFAULT; + } + return sys_vmsplice(fd, iov, nr_segs, flags); +} +#endif + +SYSCALL_DEFINE6(splice, int, fd_in, loff_t __user *, off_in, + int, fd_out, loff_t __user *, off_out, + size_t, len, unsigned int, flags) +{ + struct fd in, out; + long error; + + if (unlikely(!len)) + return 0; + + error = -EBADF; + in = fdget(fd_in); + if (in.file) { + if (in.file->f_mode & FMODE_READ) { + out = fdget(fd_out); + if (out.file) { + if (out.file->f_mode & FMODE_WRITE) + error = do_splice(in.file, off_in, + out.file, off_out, + len, flags); + fdput(out); + } + } + fdput(in); + } + return error; +} + +/* + * Make sure there's data to read. Wait for input if we can, otherwise + * return an appropriate error. + */ +static int ipipe_prep(struct pipe_inode_info *pipe, unsigned int flags) +{ + int ret; + + /* + * Check ->nrbufs without the inode lock first. This function + * is speculative anyways, so missing one is ok. + */ + if (pipe->nrbufs) + return 0; + + ret = 0; + pipe_lock(pipe); + + while (!pipe->nrbufs) { + if (signal_pending(current)) { + ret = -ERESTARTSYS; + break; + } + if (!pipe->writers) + break; + if (!pipe->waiting_writers) { + if (flags & SPLICE_F_NONBLOCK) { + ret = -EAGAIN; + break; + } + } + pipe_wait(pipe); + } + + pipe_unlock(pipe); + return ret; +} + +/* + * Make sure there's writeable room. Wait for room if we can, otherwise + * return an appropriate error. + */ +static int opipe_prep(struct pipe_inode_info *pipe, unsigned int flags) +{ + int ret; + + /* + * Check ->nrbufs without the inode lock first. This function + * is speculative anyways, so missing one is ok. + */ + if (pipe->nrbufs < pipe->buffers) + return 0; + + ret = 0; + pipe_lock(pipe); + + while (pipe->nrbufs >= pipe->buffers) { + if (!pipe->readers) { + send_sig(SIGPIPE, current, 0); + ret = -EPIPE; + break; + } + if (flags & SPLICE_F_NONBLOCK) { + ret = -EAGAIN; + break; + } + if (signal_pending(current)) { + ret = -ERESTARTSYS; + break; + } + pipe->waiting_writers++; + pipe_wait(pipe); + pipe->waiting_writers--; + } + + pipe_unlock(pipe); + return ret; +} + +/* + * Splice contents of ipipe to opipe. + */ +static int splice_pipe_to_pipe(struct pipe_inode_info *ipipe, + struct pipe_inode_info *opipe, + size_t len, unsigned int flags) +{ + struct pipe_buffer *ibuf, *obuf; + int ret = 0, nbuf; + bool input_wakeup = false; + + +retry: + ret = ipipe_prep(ipipe, flags); + if (ret) + return ret; + + ret = opipe_prep(opipe, flags); + if (ret) + return ret; + + /* + * Potential ABBA deadlock, work around it by ordering lock + * grabbing by pipe info address. Otherwise two different processes + * could deadlock (one doing tee from A -> B, the other from B -> A). + */ + pipe_double_lock(ipipe, opipe); + + do { + if (!opipe->readers) { + send_sig(SIGPIPE, current, 0); + if (!ret) + ret = -EPIPE; + break; + } + + if (!ipipe->nrbufs && !ipipe->writers) + break; + + /* + * Cannot make any progress, because either the input + * pipe is empty or the output pipe is full. + */ + if (!ipipe->nrbufs || opipe->nrbufs >= opipe->buffers) { + /* Already processed some buffers, break */ + if (ret) + break; + + if (flags & SPLICE_F_NONBLOCK) { + ret = -EAGAIN; + break; + } + + /* + * We raced with another reader/writer and haven't + * managed to process any buffers. A zero return + * value means EOF, so retry instead. + */ + pipe_unlock(ipipe); + pipe_unlock(opipe); + goto retry; + } + + ibuf = ipipe->bufs + ipipe->curbuf; + nbuf = (opipe->curbuf + opipe->nrbufs) & (opipe->buffers - 1); + obuf = opipe->bufs + nbuf; + + if (len >= ibuf->len) { + /* + * Simply move the whole buffer from ipipe to opipe + */ + *obuf = *ibuf; + ibuf->ops = NULL; + opipe->nrbufs++; + ipipe->curbuf = (ipipe->curbuf + 1) & (ipipe->buffers - 1); + ipipe->nrbufs--; + input_wakeup = true; + } else { + /* + * Get a reference to this pipe buffer, + * so we can copy the contents over. + */ + ibuf->ops->get(ipipe, ibuf); + *obuf = *ibuf; + + /* + * Don't inherit the gift flag, we need to + * prevent multiple steals of this page. + */ + obuf->flags &= ~PIPE_BUF_FLAG_GIFT; + + obuf->len = len; + opipe->nrbufs++; + ibuf->offset += obuf->len; + ibuf->len -= obuf->len; + } + ret += obuf->len; + len -= obuf->len; + } while (len); + + pipe_unlock(ipipe); + pipe_unlock(opipe); + + /* + * If we put data in the output pipe, wakeup any potential readers. + */ + if (ret > 0) + wakeup_pipe_readers(opipe); + + if (input_wakeup) + wakeup_pipe_writers(ipipe); + + return ret; +} + +/* + * Link contents of ipipe to opipe. + */ +static int link_pipe(struct pipe_inode_info *ipipe, + struct pipe_inode_info *opipe, + size_t len, unsigned int flags) +{ + struct pipe_buffer *ibuf, *obuf; + int ret = 0, i = 0, nbuf; + + /* + * Potential ABBA deadlock, work around it by ordering lock + * grabbing by pipe info address. Otherwise two different processes + * could deadlock (one doing tee from A -> B, the other from B -> A). + */ + pipe_double_lock(ipipe, opipe); + + do { + if (!opipe->readers) { + send_sig(SIGPIPE, current, 0); + if (!ret) + ret = -EPIPE; + break; + } + + /* + * If we have iterated all input buffers or ran out of + * output room, break. + */ + if (i >= ipipe->nrbufs || opipe->nrbufs >= opipe->buffers) + break; + + ibuf = ipipe->bufs + ((ipipe->curbuf + i) & (ipipe->buffers-1)); + nbuf = (opipe->curbuf + opipe->nrbufs) & (opipe->buffers - 1); + + /* + * Get a reference to this pipe buffer, + * so we can copy the contents over. + */ + ibuf->ops->get(ipipe, ibuf); + + obuf = opipe->bufs + nbuf; + *obuf = *ibuf; + + /* + * Don't inherit the gift flag, we need to + * prevent multiple steals of this page. + */ + obuf->flags &= ~PIPE_BUF_FLAG_GIFT; + + if (obuf->len > len) + obuf->len = len; + + opipe->nrbufs++; + ret += obuf->len; + len -= obuf->len; + i++; + } while (len); + + /* + * return EAGAIN if we have the potential of some data in the + * future, otherwise just return 0 + */ + if (!ret && ipipe->waiting_writers && (flags & SPLICE_F_NONBLOCK)) + ret = -EAGAIN; + + pipe_unlock(ipipe); + pipe_unlock(opipe); + + /* + * If we put data in the output pipe, wakeup any potential readers. + */ + if (ret > 0) + wakeup_pipe_readers(opipe); + + return ret; +} + +/* + * This is a tee(1) implementation that works on pipes. It doesn't copy + * any data, it simply references the 'in' pages on the 'out' pipe. + * The 'flags' used are the SPLICE_F_* variants, currently the only + * applicable one is SPLICE_F_NONBLOCK. + */ +static long do_tee(struct file *in, struct file *out, size_t len, + unsigned int flags) +{ + struct pipe_inode_info *ipipe = get_pipe_info(in); + struct pipe_inode_info *opipe = get_pipe_info(out); + int ret = -EINVAL; + + /* + * Duplicate the contents of ipipe to opipe without actually + * copying the data. + */ + if (ipipe && opipe && ipipe != opipe) { + /* + * Keep going, unless we encounter an error. The ipipe/opipe + * ordering doesn't really matter. + */ + ret = ipipe_prep(ipipe, flags); + if (!ret) { + ret = opipe_prep(opipe, flags); + if (!ret) + ret = link_pipe(ipipe, opipe, len, flags); + } + } + + return ret; +} + +SYSCALL_DEFINE4(tee, int, fdin, int, fdout, size_t, len, unsigned int, flags) +{ + struct fd in; + int error; + + if (unlikely(!len)) + return 0; + + error = -EBADF; + in = fdget(fdin); + if (in.file) { + if (in.file->f_mode & FMODE_READ) { + struct fd out = fdget(fdout); + if (out.file) { + if (out.file->f_mode & FMODE_WRITE) + error = do_tee(in.file, out.file, + len, flags); + fdput(out); + } + } + fdput(in); + } + + return error; +} diff -Naur kernel.21.3.orig/fs/xattr.c kernel.21.3.new/fs/xattr.c --- kernel.21.3.orig/fs/xattr.c 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/fs/xattr.c 2015-04-25 23:28:53.720397535 +0000 @@ -207,6 +207,7 @@ *xattr_value = value; return error; } +EXPORT_SYMBOL(vfs_getxattr_alloc); /* Compare an extended attribute value with the given value */ int vfs_xattr_cmp(struct dentry *dentry, const char *xattr_name, diff -Naur kernel.21.3.orig/include/linux/fs.h kernel.21.3.new/include/linux/fs.h --- kernel.21.3.orig/include/linux/fs.h 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/include/linux/fs.h 2015-04-25 23:28:26.656666289 +0000 @@ -2012,6 +2012,7 @@ extern int filp_close(struct file *, fl_owner_t id); extern struct filename *getname(const char __user *); +int check_acl(struct inode *, int); enum { FILE_CREATED = 1, @@ -2574,6 +2575,7 @@ extern int inode_newsize_ok(const struct inode *, loff_t offset); extern void setattr_copy(struct inode *inode, const struct iattr *attr); +extern int update_time(struct inode *, struct timespec *, int); extern int file_update_time(struct file *file); extern int generic_show_options(struct seq_file *m, struct dentry *root); diff -Naur kernel.21.3.orig/include/linux/mm.h kernel.21.3.new/include/linux/mm.h --- kernel.21.3.orig/include/linux/mm.h 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/include/linux/mm.h 2015-04-25 23:28:45.318481543 +0000 @@ -1036,6 +1036,28 @@ } #endif +#ifdef CONFIG_MMU +extern void vma_do_file_update_time(struct vm_area_struct *, const char[], int); +extern struct file *vma_do_pr_or_file(struct vm_area_struct *, const char[], + int); +extern void vma_do_get_file(struct vm_area_struct *, const char[], int); +extern void vma_do_fput(struct vm_area_struct *, const char[], int); + +#define vma_file_update_time(vma) vma_do_file_update_time(vma, __func__, \ + __LINE__) +#define vma_pr_or_file(vma) vma_do_pr_or_file(vma, __func__, \ + __LINE__) +#define vma_get_file(vma) vma_do_get_file(vma, __func__, __LINE__) +#define vma_fput(vma) vma_do_fput(vma, __func__, __LINE__) +#else +extern struct file *vmr_do_pr_or_file(struct vm_region *, const char[], int); +extern void vmr_do_fput(struct vm_region *, const char[], int); + +#define vmr_pr_or_file(region) vmr_do_pr_or_file(region, __func__, \ + __LINE__) +#define vmr_fput(region) vmr_do_fput(region, __func__, __LINE__) +#endif /* CONFIG_MMU */ + extern int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write); extern int access_remote_vm(struct mm_struct *mm, unsigned long addr, void *buf, int len, int write); diff -Naur kernel.21.3.orig/include/linux/mm.h.orig kernel.21.3.new/include/linux/mm.h.orig --- kernel.21.3.orig/include/linux/mm.h.orig 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/include/linux/mm.h.orig 2015-02-03 01:44:32.000000000 +0000 @@ -0,0 +1,1916 @@ +#ifndef _LINUX_MM_H +#define _LINUX_MM_H + +#include + +#ifdef __KERNEL__ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +struct mempolicy; +struct anon_vma; +struct anon_vma_chain; +struct file_ra_state; +struct user_struct; +struct writeback_control; + +#ifndef CONFIG_DISCONTIGMEM /* Don't use mapnrs, do it properly */ +extern unsigned long max_mapnr; +#endif + +extern unsigned long num_physpages; +extern unsigned long totalram_pages; +extern void * high_memory; +extern int page_cluster; + +#ifdef CONFIG_SYSCTL +extern int sysctl_legacy_va_layout; +#else +#define sysctl_legacy_va_layout 0 +#endif + +#include +#include +#include + +extern unsigned long sysctl_user_reserve_kbytes; +extern unsigned long sysctl_admin_reserve_kbytes; + +#define nth_page(page,n) pfn_to_page(page_to_pfn((page)) + (n)) + +/* to align the pointer to the (next) page boundary */ +#define PAGE_ALIGN(addr) ALIGN(addr, PAGE_SIZE) + +/* + * Linux kernel virtual memory manager primitives. + * The idea being to have a "virtual" mm in the same way + * we have a virtual fs - giving a cleaner interface to the + * mm details, and allowing different kinds of memory mappings + * (from shared memory to executable loading to arbitrary + * mmap() functions). + */ + +extern struct kmem_cache *vm_area_cachep; + +#ifndef CONFIG_MMU +extern struct rb_root nommu_region_tree; +extern struct rw_semaphore nommu_region_sem; + +extern unsigned int kobjsize(const void *objp); +#endif + +/* + * vm_flags in vm_area_struct, see mm_types.h. + */ +#define VM_NONE 0x00000000 + +#define VM_READ 0x00000001 /* currently active flags */ +#define VM_WRITE 0x00000002 +#define VM_EXEC 0x00000004 +#define VM_SHARED 0x00000008 + +/* mprotect() hardcodes VM_MAYREAD >> 4 == VM_READ, and so for r/w/x bits. */ +#define VM_MAYREAD 0x00000010 /* limits for mprotect() etc */ +#define VM_MAYWRITE 0x00000020 +#define VM_MAYEXEC 0x00000040 +#define VM_MAYSHARE 0x00000080 + +#define VM_GROWSDOWN 0x00000100 /* general info on the segment */ +#define VM_PFNMAP 0x00000400 /* Page-ranges managed without "struct page", just pure PFN */ +#define VM_DENYWRITE 0x00000800 /* ETXTBSY on write attempts.. */ + +#define VM_LOCKED 0x00002000 +#define VM_IO 0x00004000 /* Memory mapped I/O or similar */ + + /* Used by sys_madvise() */ +#define VM_SEQ_READ 0x00008000 /* App will access data sequentially */ +#define VM_RAND_READ 0x00010000 /* App will not benefit from clustered reads */ + +#define VM_DONTCOPY 0x00020000 /* Do not copy this vma on fork */ +#define VM_DONTEXPAND 0x00040000 /* Cannot expand with mremap() */ +#define VM_ACCOUNT 0x00100000 /* Is a VM accounted object */ +#define VM_NORESERVE 0x00200000 /* should the VM suppress accounting */ +#define VM_HUGETLB 0x00400000 /* Huge TLB Page VM */ +#define VM_NONLINEAR 0x00800000 /* Is non-linear (remap_file_pages) */ +#define VM_ARCH_1 0x01000000 /* Architecture-specific flag */ +#define VM_DONTDUMP 0x04000000 /* Do not include in the core dump */ + +#define VM_MIXEDMAP 0x10000000 /* Can contain "struct page" and pure PFN pages */ +#define VM_HUGEPAGE 0x20000000 /* MADV_HUGEPAGE marked this vma */ +#define VM_NOHUGEPAGE 0x40000000 /* MADV_NOHUGEPAGE marked this vma */ +#define VM_MERGEABLE 0x80000000 /* KSM may merge identical pages */ + +#if defined(CONFIG_X86) +# define VM_PAT VM_ARCH_1 /* PAT reserves whole VMA at once (x86) */ +#elif defined(CONFIG_PPC) +# define VM_SAO VM_ARCH_1 /* Strong Access Ordering (powerpc) */ +#elif defined(CONFIG_PARISC) +# define VM_GROWSUP VM_ARCH_1 +#elif defined(CONFIG_METAG) +# define VM_GROWSUP VM_ARCH_1 +#elif defined(CONFIG_IA64) +# define VM_GROWSUP VM_ARCH_1 +#elif !defined(CONFIG_MMU) +# define VM_MAPPED_COPY VM_ARCH_1 /* T if mapped copy of data (nommu mmap) */ +#endif + +#ifndef VM_GROWSUP +# define VM_GROWSUP VM_NONE +#endif + +/* Bits set in the VMA until the stack is in its final location */ +#define VM_STACK_INCOMPLETE_SETUP (VM_RAND_READ | VM_SEQ_READ) + +#ifndef VM_STACK_DEFAULT_FLAGS /* arch can override this */ +#define VM_STACK_DEFAULT_FLAGS VM_DATA_DEFAULT_FLAGS +#endif + +#ifdef CONFIG_STACK_GROWSUP +#define VM_STACK_FLAGS (VM_GROWSUP | VM_STACK_DEFAULT_FLAGS | VM_ACCOUNT) +#else +#define VM_STACK_FLAGS (VM_GROWSDOWN | VM_STACK_DEFAULT_FLAGS | VM_ACCOUNT) +#endif + +#define VM_READHINTMASK (VM_SEQ_READ | VM_RAND_READ) +#define VM_ClearReadHint(v) (v)->vm_flags &= ~VM_READHINTMASK +#define VM_NormalReadHint(v) (!((v)->vm_flags & VM_READHINTMASK)) +#define VM_SequentialReadHint(v) ((v)->vm_flags & VM_SEQ_READ) +#define VM_RandomReadHint(v) ((v)->vm_flags & VM_RAND_READ) + +/* + * Special vmas that are non-mergable, non-mlock()able. + * Note: mm/huge_memory.c VM_NO_THP depends on this definition. + */ +#define VM_SPECIAL (VM_IO | VM_DONTEXPAND | VM_PFNMAP) + +/* + * mapping from the currently active vm_flags protection bits (the + * low four bits) to a page protection mask.. + */ +extern pgprot_t protection_map[16]; + +#define FAULT_FLAG_WRITE 0x01 /* Fault was a write access */ +#define FAULT_FLAG_NONLINEAR 0x02 /* Fault was via a nonlinear mapping */ +#define FAULT_FLAG_MKWRITE 0x04 /* Fault was mkwrite of existing pte */ +#define FAULT_FLAG_ALLOW_RETRY 0x08 /* Retry fault if blocking */ +#define FAULT_FLAG_RETRY_NOWAIT 0x10 /* Don't drop mmap_sem and wait when retrying */ +#define FAULT_FLAG_KILLABLE 0x20 /* The fault task is in SIGKILL killable region */ +#define FAULT_FLAG_TRIED 0x40 /* second try */ +#define FAULT_FLAG_USER 0x80 /* The fault originated in userspace */ +#define FAULT_FLAG_NO_CMA 0x100 /* don't use CMA pages */ + +/* + * vm_fault is filled by the the pagefault handler and passed to the vma's + * ->fault function. The vma's ->fault is responsible for returning a bitmask + * of VM_FAULT_xxx flags that give details about how the fault was handled. + * + * pgoff should be used in favour of virtual_address, if possible. If pgoff + * is used, one may implement ->remap_pages to get nonlinear mapping support. + */ +struct vm_fault { + unsigned int flags; /* FAULT_FLAG_xxx flags */ + pgoff_t pgoff; /* Logical page offset based on vma */ + void __user *virtual_address; /* Faulting virtual address */ + + struct page *page; /* ->fault handlers should return a + * page here, unless VM_FAULT_NOPAGE + * is set (which is also implied by + * VM_FAULT_ERROR). + */ +}; + +/* + * These are the virtual MM functions - opening of an area, closing and + * unmapping it (needed to keep files on disk up-to-date etc), pointer + * to the functions called when a no-page or a wp-page exception occurs. + */ +struct vm_operations_struct { + void (*open)(struct vm_area_struct * area); + void (*close)(struct vm_area_struct * area); + int (*fault)(struct vm_area_struct *vma, struct vm_fault *vmf); + + /* notification that a previously read-only page is about to become + * writable, if an error is returned it will cause a SIGBUS */ + int (*page_mkwrite)(struct vm_area_struct *vma, struct vm_fault *vmf); + + /* called by access_process_vm when get_user_pages() fails, typically + * for use by special VMAs that can switch between memory and hardware + */ + int (*access)(struct vm_area_struct *vma, unsigned long addr, + void *buf, int len, int write); +#ifdef CONFIG_NUMA + /* + * set_policy() op must add a reference to any non-NULL @new mempolicy + * to hold the policy upon return. Caller should pass NULL @new to + * remove a policy and fall back to surrounding context--i.e. do not + * install a MPOL_DEFAULT policy, nor the task or system default + * mempolicy. + */ + int (*set_policy)(struct vm_area_struct *vma, struct mempolicy *new); + + /* + * get_policy() op must add reference [mpol_get()] to any policy at + * (vma,addr) marked as MPOL_SHARED. The shared policy infrastructure + * in mm/mempolicy.c will do this automatically. + * get_policy() must NOT add a ref if the policy at (vma,addr) is not + * marked as MPOL_SHARED. vma policies are protected by the mmap_sem. + * If no [shared/vma] mempolicy exists at the addr, get_policy() op + * must return NULL--i.e., do not "fallback" to task or system default + * policy. + */ + struct mempolicy *(*get_policy)(struct vm_area_struct *vma, + unsigned long addr); + int (*migrate)(struct vm_area_struct *vma, const nodemask_t *from, + const nodemask_t *to, unsigned long flags); +#endif + /* called by sys_remap_file_pages() to populate non-linear mapping */ + int (*remap_pages)(struct vm_area_struct *vma, unsigned long addr, + unsigned long size, pgoff_t pgoff); +}; + +struct mmu_gather; +struct inode; + +#define page_private(page) ((page)->private) +#define set_page_private(page, v) ((page)->private = (v)) + +/* It's valid only if the page is free path or free_list */ +static inline void set_freepage_migratetype(struct page *page, int migratetype) +{ + page->index = migratetype; +} + +/* It's valid only if the page is free path or free_list */ +static inline int get_freepage_migratetype(struct page *page) +{ + return page->index; +} + +/* + * FIXME: take this include out, include page-flags.h in + * files which need it (119 of them) + */ +#include +#include + +/* + * Methods to modify the page usage count. + * + * What counts for a page usage: + * - cache mapping (page->mapping) + * - private data (page->private) + * - page mapped in a task's page tables, each mapping + * is counted separately + * + * Also, many kernel routines increase the page count before a critical + * routine so they can be sure the page doesn't go away from under them. + */ + +/* + * Drop a ref, return true if the refcount fell to zero (the page has no users) + */ +static inline int put_page_testzero(struct page *page) +{ + VM_BUG_ON(atomic_read(&page->_count) == 0); + return atomic_dec_and_test(&page->_count); +} + +/* + * Try to grab a ref unless the page has a refcount of zero, return false if + * that is the case. + */ +static inline int get_page_unless_zero(struct page *page) +{ + return atomic_inc_not_zero(&page->_count); +} + +extern int page_is_ram(unsigned long pfn); + +/* Support for virtually mapped pages */ +struct page *vmalloc_to_page(const void *addr); +unsigned long vmalloc_to_pfn(const void *addr); + +/* + * Determine if an address is within the vmalloc range + * + * On nommu, vmalloc/vfree wrap through kmalloc/kfree directly, so there + * is no special casing required. + */ +static inline int is_vmalloc_addr(const void *x) +{ +#ifdef CONFIG_MMU + unsigned long addr = (unsigned long)x; + + return addr >= VMALLOC_START && addr < VMALLOC_END; +#else + return 0; +#endif +} +#ifdef CONFIG_MMU +extern int is_vmalloc_or_module_addr(const void *x); +#else +static inline int is_vmalloc_or_module_addr(const void *x) +{ + return 0; +} +#endif + +static inline void compound_lock(struct page *page) +{ +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + VM_BUG_ON(PageSlab(page)); + bit_spin_lock(PG_compound_lock, &page->flags); +#endif +} + +static inline void compound_unlock(struct page *page) +{ +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + VM_BUG_ON(PageSlab(page)); + bit_spin_unlock(PG_compound_lock, &page->flags); +#endif +} + +static inline unsigned long compound_lock_irqsave(struct page *page) +{ + unsigned long uninitialized_var(flags); +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + local_irq_save(flags); + compound_lock(page); +#endif + return flags; +} + +static inline void compound_unlock_irqrestore(struct page *page, + unsigned long flags) +{ +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + compound_unlock(page); + local_irq_restore(flags); +#endif +} + +static inline struct page *compound_head(struct page *page) +{ + if (unlikely(PageTail(page))) { + struct page *head = page->first_page; + + /* + * page->first_page may be a dangling pointer to an old + * compound page, so recheck that it is still a tail + * page before returning. + */ + smp_rmb(); + if (likely(PageTail(page))) + return head; + } + return page; +} + +/* + * The atomic page->_mapcount, starts from -1: so that transitions + * both from it and to it can be tracked, using atomic_inc_and_test + * and atomic_add_negative(-1). + */ +static inline void page_mapcount_reset(struct page *page) +{ + atomic_set(&(page)->_mapcount, -1); +} + +static inline int page_mapcount(struct page *page) +{ + return atomic_read(&(page)->_mapcount) + 1; +} + +static inline int page_count(struct page *page) +{ + return atomic_read(&compound_head(page)->_count); +} + +static inline void get_huge_page_tail(struct page *page) +{ + /* + * __split_huge_page_refcount() cannot run + * from under us. + */ + VM_BUG_ON(page_mapcount(page) < 0); + VM_BUG_ON(atomic_read(&page->_count) != 0); + atomic_inc(&page->_mapcount); +} + +extern bool __get_page_tail(struct page *page); + +static inline void get_page(struct page *page) +{ + if (unlikely(PageTail(page))) + if (likely(__get_page_tail(page))) + return; + /* + * Getting a normal page or the head of a compound page + * requires to already have an elevated page->_count. + */ + VM_BUG_ON(atomic_read(&page->_count) <= 0); + atomic_inc(&page->_count); +} + +static inline struct page *virt_to_head_page(const void *x) +{ + struct page *page = virt_to_page(x); + return compound_head(page); +} + +/* + * Setup the page count before being freed into the page allocator for + * the first time (boot or memory hotplug) + */ +static inline void init_page_count(struct page *page) +{ + atomic_set(&page->_count, 1); +} + +/* + * PageBuddy() indicate that the page is free and in the buddy system + * (see mm/page_alloc.c). + * + * PAGE_BUDDY_MAPCOUNT_VALUE must be <= -2 but better not too close to + * -2 so that an underflow of the page_mapcount() won't be mistaken + * for a genuine PAGE_BUDDY_MAPCOUNT_VALUE. -128 can be created very + * efficiently by most CPU architectures. + */ +#define PAGE_BUDDY_MAPCOUNT_VALUE (-128) + +static inline int PageBuddy(struct page *page) +{ + return atomic_read(&page->_mapcount) == PAGE_BUDDY_MAPCOUNT_VALUE; +} + +static inline void __SetPageBuddy(struct page *page) +{ + VM_BUG_ON(atomic_read(&page->_mapcount) != -1); + atomic_set(&page->_mapcount, PAGE_BUDDY_MAPCOUNT_VALUE); +} + +static inline void __ClearPageBuddy(struct page *page) +{ + VM_BUG_ON(!PageBuddy(page)); + atomic_set(&page->_mapcount, -1); +} + +void put_page(struct page *page); +void put_pages_list(struct list_head *pages); + +void split_page(struct page *page, unsigned int order); +int split_free_page(struct page *page); + +/* + * Compound pages have a destructor function. Provide a + * prototype for that function and accessor functions. + * These are _only_ valid on the head of a PG_compound page. + */ +typedef void compound_page_dtor(struct page *); + +static inline void set_compound_page_dtor(struct page *page, + compound_page_dtor *dtor) +{ + page[1].lru.next = (void *)dtor; +} + +static inline compound_page_dtor *get_compound_page_dtor(struct page *page) +{ + return (compound_page_dtor *)page[1].lru.next; +} + +static inline int compound_order(struct page *page) +{ + if (!PageHead(page)) + return 0; + return (unsigned long)page[1].lru.prev; +} + +static inline int compound_trans_order(struct page *page) +{ + int order; + unsigned long flags; + + if (!PageHead(page)) + return 0; + + flags = compound_lock_irqsave(page); + order = compound_order(page); + compound_unlock_irqrestore(page, flags); + return order; +} + +static inline void set_compound_order(struct page *page, unsigned long order) +{ + page[1].lru.prev = (void *)order; +} + +#ifdef CONFIG_MMU +/* + * Do pte_mkwrite, but only if the vma says VM_WRITE. We do this when + * servicing faults for write access. In the normal case, do always want + * pte_mkwrite. But get_user_pages can cause write faults for mappings + * that do not have writing enabled, when used by access_process_vm. + */ +static inline pte_t maybe_mkwrite(pte_t pte, struct vm_area_struct *vma) +{ + if (likely(vma->vm_flags & VM_WRITE)) + pte = pte_mkwrite(pte); + return pte; +} +#endif + +/* + * Multiple processes may "see" the same page. E.g. for untouched + * mappings of /dev/null, all processes see the same page full of + * zeroes, and text pages of executables and shared libraries have + * only one copy in memory, at most, normally. + * + * For the non-reserved pages, page_count(page) denotes a reference count. + * page_count() == 0 means the page is free. page->lru is then used for + * freelist management in the buddy allocator. + * page_count() > 0 means the page has been allocated. + * + * Pages are allocated by the slab allocator in order to provide memory + * to kmalloc and kmem_cache_alloc. In this case, the management of the + * page, and the fields in 'struct page' are the responsibility of mm/slab.c + * unless a particular usage is carefully commented. (the responsibility of + * freeing the kmalloc memory is the caller's, of course). + * + * A page may be used by anyone else who does a __get_free_page(). + * In this case, page_count still tracks the references, and should only + * be used through the normal accessor functions. The top bits of page->flags + * and page->virtual store page management information, but all other fields + * are unused and could be used privately, carefully. The management of this + * page is the responsibility of the one who allocated it, and those who have + * subsequently been given references to it. + * + * The other pages (we may call them "pagecache pages") are completely + * managed by the Linux memory manager: I/O, buffers, swapping etc. + * The following discussion applies only to them. + * + * A pagecache page contains an opaque `private' member, which belongs to the + * page's address_space. Usually, this is the address of a circular list of + * the page's disk buffers. PG_private must be set to tell the VM to call + * into the filesystem to release these pages. + * + * A page may belong to an inode's memory mapping. In this case, page->mapping + * is the pointer to the inode, and page->index is the file offset of the page, + * in units of PAGE_CACHE_SIZE. + * + * If pagecache pages are not associated with an inode, they are said to be + * anonymous pages. These may become associated with the swapcache, and in that + * case PG_swapcache is set, and page->private is an offset into the swapcache. + * + * In either case (swapcache or inode backed), the pagecache itself holds one + * reference to the page. Setting PG_private should also increment the + * refcount. The each user mapping also has a reference to the page. + * + * The pagecache pages are stored in a per-mapping radix tree, which is + * rooted at mapping->page_tree, and indexed by offset. + * Where 2.4 and early 2.6 kernels kept dirty/clean pages in per-address_space + * lists, we instead now tag pages as dirty/writeback in the radix tree. + * + * All pagecache pages may be subject to I/O: + * - inode pages may need to be read from disk, + * - inode pages which have been modified and are MAP_SHARED may need + * to be written back to the inode on disk, + * - anonymous pages (including MAP_PRIVATE file mappings) which have been + * modified may need to be swapped out to swap space and (later) to be read + * back into memory. + */ + +/* + * The zone field is never updated after free_area_init_core() + * sets it, so none of the operations on it need to be atomic. + */ + +/* Page flags: | [SECTION] | [NODE] | ZONE | [LAST_NID] | ... | FLAGS | */ +#define SECTIONS_PGOFF ((sizeof(unsigned long)*8) - SECTIONS_WIDTH) +#define NODES_PGOFF (SECTIONS_PGOFF - NODES_WIDTH) +#define ZONES_PGOFF (NODES_PGOFF - ZONES_WIDTH) +#define LAST_NID_PGOFF (ZONES_PGOFF - LAST_NID_WIDTH) + +/* + * Define the bit shifts to access each section. For non-existent + * sections we define the shift as 0; that plus a 0 mask ensures + * the compiler will optimise away reference to them. + */ +#define SECTIONS_PGSHIFT (SECTIONS_PGOFF * (SECTIONS_WIDTH != 0)) +#define NODES_PGSHIFT (NODES_PGOFF * (NODES_WIDTH != 0)) +#define ZONES_PGSHIFT (ZONES_PGOFF * (ZONES_WIDTH != 0)) +#define LAST_NID_PGSHIFT (LAST_NID_PGOFF * (LAST_NID_WIDTH != 0)) + +/* NODE:ZONE or SECTION:ZONE is used to ID a zone for the buddy allocator */ +#ifdef NODE_NOT_IN_PAGE_FLAGS +#define ZONEID_SHIFT (SECTIONS_SHIFT + ZONES_SHIFT) +#define ZONEID_PGOFF ((SECTIONS_PGOFF < ZONES_PGOFF)? \ + SECTIONS_PGOFF : ZONES_PGOFF) +#else +#define ZONEID_SHIFT (NODES_SHIFT + ZONES_SHIFT) +#define ZONEID_PGOFF ((NODES_PGOFF < ZONES_PGOFF)? \ + NODES_PGOFF : ZONES_PGOFF) +#endif + +#define ZONEID_PGSHIFT (ZONEID_PGOFF * (ZONEID_SHIFT != 0)) + +#if SECTIONS_WIDTH+NODES_WIDTH+ZONES_WIDTH > BITS_PER_LONG - NR_PAGEFLAGS +#error SECTIONS_WIDTH+NODES_WIDTH+ZONES_WIDTH > BITS_PER_LONG - NR_PAGEFLAGS +#endif + +#define ZONES_MASK ((1UL << ZONES_WIDTH) - 1) +#define NODES_MASK ((1UL << NODES_WIDTH) - 1) +#define SECTIONS_MASK ((1UL << SECTIONS_WIDTH) - 1) +#define LAST_NID_MASK ((1UL << LAST_NID_WIDTH) - 1) +#define ZONEID_MASK ((1UL << ZONEID_SHIFT) - 1) + +static inline enum zone_type page_zonenum(const struct page *page) +{ + return (page->flags >> ZONES_PGSHIFT) & ZONES_MASK; +} + +#if defined(CONFIG_SPARSEMEM) && !defined(CONFIG_SPARSEMEM_VMEMMAP) +#define SECTION_IN_PAGE_FLAGS +#endif + +/* + * The identification function is only used by the buddy allocator for + * determining if two pages could be buddies. We are not really + * identifying a zone since we could be using a the section number + * id if we have not node id available in page flags. + * We guarantee only that it will return the same value for two + * combinable pages in a zone. + */ +static inline int page_zone_id(struct page *page) +{ + return (page->flags >> ZONEID_PGSHIFT) & ZONEID_MASK; +} + +static inline int zone_to_nid(struct zone *zone) +{ +#ifdef CONFIG_NUMA + return zone->node; +#else + return 0; +#endif +} + +#ifdef NODE_NOT_IN_PAGE_FLAGS +extern int page_to_nid(const struct page *page); +#else +static inline int page_to_nid(const struct page *page) +{ + return (page->flags >> NODES_PGSHIFT) & NODES_MASK; +} +#endif + +#ifdef CONFIG_NUMA_BALANCING +#ifdef LAST_NID_NOT_IN_PAGE_FLAGS +static inline int page_nid_xchg_last(struct page *page, int nid) +{ + return xchg(&page->_last_nid, nid); +} + +static inline int page_nid_last(struct page *page) +{ + return page->_last_nid; +} +static inline void page_nid_reset_last(struct page *page) +{ + page->_last_nid = -1; +} +#else +static inline int page_nid_last(struct page *page) +{ + return (page->flags >> LAST_NID_PGSHIFT) & LAST_NID_MASK; +} + +extern int page_nid_xchg_last(struct page *page, int nid); + +static inline void page_nid_reset_last(struct page *page) +{ + int nid = (1 << LAST_NID_SHIFT) - 1; + + page->flags &= ~(LAST_NID_MASK << LAST_NID_PGSHIFT); + page->flags |= (nid & LAST_NID_MASK) << LAST_NID_PGSHIFT; +} +#endif /* LAST_NID_NOT_IN_PAGE_FLAGS */ +#else +static inline int page_nid_xchg_last(struct page *page, int nid) +{ + return page_to_nid(page); +} + +static inline int page_nid_last(struct page *page) +{ + return page_to_nid(page); +} + +static inline void page_nid_reset_last(struct page *page) +{ +} +#endif + +static inline struct zone *page_zone(const struct page *page) +{ + return &NODE_DATA(page_to_nid(page))->node_zones[page_zonenum(page)]; +} + +#ifdef SECTION_IN_PAGE_FLAGS +static inline void set_page_section(struct page *page, unsigned long section) +{ + page->flags &= ~(SECTIONS_MASK << SECTIONS_PGSHIFT); + page->flags |= (section & SECTIONS_MASK) << SECTIONS_PGSHIFT; +} + +static inline unsigned long page_to_section(const struct page *page) +{ + return (page->flags >> SECTIONS_PGSHIFT) & SECTIONS_MASK; +} +#endif + +static inline void set_page_zone(struct page *page, enum zone_type zone) +{ + page->flags &= ~(ZONES_MASK << ZONES_PGSHIFT); + page->flags |= (zone & ZONES_MASK) << ZONES_PGSHIFT; +} + +static inline void set_page_node(struct page *page, unsigned long node) +{ + page->flags &= ~(NODES_MASK << NODES_PGSHIFT); + page->flags |= (node & NODES_MASK) << NODES_PGSHIFT; +} + +static inline void set_page_links(struct page *page, enum zone_type zone, + unsigned long node, unsigned long pfn) +{ + set_page_zone(page, zone); + set_page_node(page, node); +#ifdef SECTION_IN_PAGE_FLAGS + set_page_section(page, pfn_to_section_nr(pfn)); +#endif +} + +/* + * Some inline functions in vmstat.h depend on page_zone() + */ +#include + +static __always_inline void *lowmem_page_address(const struct page *page) +{ + return __va(PFN_PHYS(page_to_pfn(page))); +} + +#if defined(CONFIG_HIGHMEM) && !defined(WANT_PAGE_VIRTUAL) +#define HASHED_PAGE_VIRTUAL +#endif + +#if defined(WANT_PAGE_VIRTUAL) +static inline void *page_address(const struct page *page) +{ + return page->virtual; +} +static inline void set_page_address(struct page *page, void *address) +{ + page->virtual = address; +} +#define page_address_init() do { } while(0) +#endif + +#if defined(HASHED_PAGE_VIRTUAL) +void *page_address(const struct page *page); +void set_page_address(struct page *page, void *virtual); +void page_address_init(void); +#endif + +#if !defined(HASHED_PAGE_VIRTUAL) && !defined(WANT_PAGE_VIRTUAL) +#define page_address(page) lowmem_page_address(page) +#define set_page_address(page, address) do { } while(0) +#define page_address_init() do { } while(0) +#endif + +/* + * On an anonymous page mapped into a user virtual memory area, + * page->mapping points to its anon_vma, not to a struct address_space; + * with the PAGE_MAPPING_ANON bit set to distinguish it. See rmap.h. + * + * On an anonymous page in a VM_MERGEABLE area, if CONFIG_KSM is enabled, + * the PAGE_MAPPING_KSM bit may be set along with the PAGE_MAPPING_ANON bit; + * and then page->mapping points, not to an anon_vma, but to a private + * structure which KSM associates with that merged page. See ksm.h. + * + * PAGE_MAPPING_KSM without PAGE_MAPPING_ANON is currently never used. + * + * Please note that, confusingly, "page_mapping" refers to the inode + * address_space which maps the page from disk; whereas "page_mapped" + * refers to user virtual address space into which the page is mapped. + */ +#define PAGE_MAPPING_ANON 1 +#define PAGE_MAPPING_KSM 2 +#define PAGE_MAPPING_FLAGS (PAGE_MAPPING_ANON | PAGE_MAPPING_KSM) + +extern struct address_space *page_mapping(struct page *page); + +/* Neutral page->mapping pointer to address_space or anon_vma or other */ +static inline void *page_rmapping(struct page *page) +{ + return (void *)((unsigned long)page->mapping & ~PAGE_MAPPING_FLAGS); +} + +extern struct address_space *__page_file_mapping(struct page *); + +static inline +struct address_space *page_file_mapping(struct page *page) +{ + if (unlikely(PageSwapCache(page))) + return __page_file_mapping(page); + + return page->mapping; +} + +static inline int PageAnon(struct page *page) +{ + return ((unsigned long)page->mapping & PAGE_MAPPING_ANON) != 0; +} + +/* + * Return the pagecache index of the passed page. Regular pagecache pages + * use ->index whereas swapcache pages use ->private + */ +static inline pgoff_t page_index(struct page *page) +{ + if (unlikely(PageSwapCache(page))) + return page_private(page); + return page->index; +} + +extern pgoff_t __page_file_index(struct page *page); + +/* + * Return the file index of the page. Regular pagecache pages use ->index + * whereas swapcache pages use swp_offset(->private) + */ +static inline pgoff_t page_file_index(struct page *page) +{ + if (unlikely(PageSwapCache(page))) + return __page_file_index(page); + + return page->index; +} + +/* + * Return true if this page is mapped into pagetables. + */ +static inline int page_mapped(struct page *page) +{ + return atomic_read(&(page)->_mapcount) >= 0; +} + +/* + * Different kinds of faults, as returned by handle_mm_fault(). + * Used to decide whether a process gets delivered SIGBUS or + * just gets major/minor fault counters bumped up. + */ + +#define VM_FAULT_MINOR 0 /* For backwards compat. Remove me quickly. */ + +#define VM_FAULT_OOM 0x0001 +#define VM_FAULT_SIGBUS 0x0002 +#define VM_FAULT_MAJOR 0x0004 +#define VM_FAULT_WRITE 0x0008 /* Special case for get_user_pages */ +#define VM_FAULT_HWPOISON 0x0010 /* Hit poisoned small page */ +#define VM_FAULT_HWPOISON_LARGE 0x0020 /* Hit poisoned large page. Index encoded in upper bits */ + +#define VM_FAULT_NOPAGE 0x0100 /* ->fault installed the pte, not return page */ +#define VM_FAULT_LOCKED 0x0200 /* ->fault locked the returned page */ +#define VM_FAULT_RETRY 0x0400 /* ->fault blocked, must retry */ + +#define VM_FAULT_HWPOISON_LARGE_MASK 0xf000 /* encodes hpage index for large hwpoison */ + +#define VM_FAULT_ERROR (VM_FAULT_OOM | VM_FAULT_SIGBUS | VM_FAULT_HWPOISON | \ + VM_FAULT_HWPOISON_LARGE) + +/* Encode hstate index for a hwpoisoned large page */ +#define VM_FAULT_SET_HINDEX(x) ((x) << 12) +#define VM_FAULT_GET_HINDEX(x) (((x) >> 12) & 0xf) + +/* + * Can be called by the pagefault handler when it gets a VM_FAULT_OOM. + */ +extern void pagefault_out_of_memory(void); + +#define offset_in_page(p) ((unsigned long)(p) & ~PAGE_MASK) + +/* + * Flags passed to show_mem() and show_free_areas() to suppress output in + * various contexts. + */ +#define SHOW_MEM_FILTER_NODES (0x0001u) /* disallowed nodes */ +#define SHOW_MEM_FILTER_PAGE_COUNT (0x0002u) /* page type count */ + +extern void show_free_areas(unsigned int flags); +extern bool skip_free_areas_node(unsigned int flags, int nid); + +void shmem_set_file(struct vm_area_struct *vma, struct file *file); +int shmem_zero_setup(struct vm_area_struct *); + +extern int can_do_mlock(void); +extern int user_shm_lock(size_t, struct user_struct *); +extern void user_shm_unlock(size_t, struct user_struct *); + +/* + * Parameter block passed down to zap_pte_range in exceptional cases. + */ +struct zap_details { + struct vm_area_struct *nonlinear_vma; /* Check page->index if set */ + struct address_space *check_mapping; /* Check page->mapping if set */ + pgoff_t first_index; /* Lowest page->index to unmap */ + pgoff_t last_index; /* Highest page->index to unmap */ +}; + +struct page *vm_normal_page(struct vm_area_struct *vma, unsigned long addr, + pte_t pte); + +int zap_vma_ptes(struct vm_area_struct *vma, unsigned long address, + unsigned long size); +void zap_page_range(struct vm_area_struct *vma, unsigned long address, + unsigned long size, struct zap_details *); +void unmap_vmas(struct mmu_gather *tlb, struct vm_area_struct *start_vma, + unsigned long start, unsigned long end); + +/** + * mm_walk - callbacks for walk_page_range + * @pgd_entry: if set, called for each non-empty PGD (top-level) entry + * @pud_entry: if set, called for each non-empty PUD (2nd-level) entry + * @pmd_entry: if set, called for each non-empty PMD (3rd-level) entry + * this handler is required to be able to handle + * pmd_trans_huge() pmds. They may simply choose to + * split_huge_page() instead of handling it explicitly. + * @pte_entry: if set, called for each non-empty PTE (4th-level) entry + * @pte_hole: if set, called for each hole at all levels + * @hugetlb_entry: if set, called for each hugetlb entry + * *Caution*: The caller must hold mmap_sem() if @hugetlb_entry + * is used. + * + * (see walk_page_range for more details) + */ +struct mm_walk { + int (*pgd_entry)(pgd_t *pgd, unsigned long addr, + unsigned long next, struct mm_walk *walk); + int (*pud_entry)(pud_t *pud, unsigned long addr, + unsigned long next, struct mm_walk *walk); + int (*pmd_entry)(pmd_t *pmd, unsigned long addr, + unsigned long next, struct mm_walk *walk); + int (*pte_entry)(pte_t *pte, unsigned long addr, + unsigned long next, struct mm_walk *walk); + int (*pte_hole)(unsigned long addr, unsigned long next, + struct mm_walk *walk); + int (*hugetlb_entry)(pte_t *pte, unsigned long hmask, + unsigned long addr, unsigned long next, + struct mm_walk *walk); + struct mm_struct *mm; + void *private; +}; + +int walk_page_range(unsigned long addr, unsigned long end, + struct mm_walk *walk); +void free_pgd_range(struct mmu_gather *tlb, unsigned long addr, + unsigned long end, unsigned long floor, unsigned long ceiling); +int copy_page_range(struct mm_struct *dst, struct mm_struct *src, + struct vm_area_struct *vma); +void unmap_mapping_range(struct address_space *mapping, + loff_t const holebegin, loff_t const holelen, int even_cows); +int follow_pfn(struct vm_area_struct *vma, unsigned long address, + unsigned long *pfn); +int follow_phys(struct vm_area_struct *vma, unsigned long address, + unsigned int flags, unsigned long *prot, resource_size_t *phys); +int generic_access_phys(struct vm_area_struct *vma, unsigned long addr, + void *buf, int len, int write); + +static inline void unmap_shared_mapping_range(struct address_space *mapping, + loff_t const holebegin, loff_t const holelen) +{ + unmap_mapping_range(mapping, holebegin, holelen, 0); +} + +extern void truncate_pagecache(struct inode *inode, loff_t old, loff_t new); +extern void truncate_setsize(struct inode *inode, loff_t newsize); +void truncate_pagecache_range(struct inode *inode, loff_t offset, loff_t end); +int truncate_inode_page(struct address_space *mapping, struct page *page); +int generic_error_remove_page(struct address_space *mapping, struct page *page); +int invalidate_inode_page(struct page *page); + +#ifdef CONFIG_MMU +extern int handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long address, unsigned int flags); +extern int fixup_user_fault(struct task_struct *tsk, struct mm_struct *mm, + unsigned long address, unsigned int fault_flags); +#else +static inline int handle_mm_fault(struct mm_struct *mm, + struct vm_area_struct *vma, unsigned long address, + unsigned int flags) +{ + /* should never happen if there's no MMU */ + BUG(); + return VM_FAULT_SIGBUS; +} +static inline int fixup_user_fault(struct task_struct *tsk, + struct mm_struct *mm, unsigned long address, + unsigned int fault_flags) +{ + /* should never happen if there's no MMU */ + BUG(); + return -EFAULT; +} +#endif + +extern int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write); +extern int access_remote_vm(struct mm_struct *mm, unsigned long addr, + void *buf, int len, int write); + +long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm, + unsigned long start, unsigned long nr_pages, + unsigned int foll_flags, struct page **pages, + struct vm_area_struct **vmas, int *nonblocking); + +int get_user_pages_fast(unsigned long start, int nr_pages, int write, + struct page **pages); +struct kvec; +int get_kernel_pages(const struct kvec *iov, int nr_pages, int write, + struct page **pages); +int get_kernel_page(unsigned long start, int write, struct page **pages); +struct page *get_dump_page(unsigned long addr); + +extern int try_to_release_page(struct page * page, gfp_t gfp_mask); +extern void do_invalidatepage(struct page *page, unsigned long offset); + +int __set_page_dirty_nobuffers(struct page *page); +int __set_page_dirty_no_writeback(struct page *page); +int redirty_page_for_writepage(struct writeback_control *wbc, + struct page *page); +void account_page_dirtied(struct page *page, struct address_space *mapping); +void account_page_writeback(struct page *page); +int set_page_dirty(struct page *page); +int set_page_dirty_lock(struct page *page); +int clear_page_dirty_for_io(struct page *page); + +/* Is the vma a continuation of the stack vma above it? */ +static inline int vma_growsdown(struct vm_area_struct *vma, unsigned long addr) +{ + return vma && (vma->vm_end == addr) && (vma->vm_flags & VM_GROWSDOWN); +} + +static inline int stack_guard_page_start(struct vm_area_struct *vma, + unsigned long addr) +{ + return (vma->vm_flags & VM_GROWSDOWN) && + (vma->vm_start == addr) && + !vma_growsdown(vma->vm_prev, addr); +} + +/* Is the vma a continuation of the stack vma below it? */ +static inline int vma_growsup(struct vm_area_struct *vma, unsigned long addr) +{ + return vma && (vma->vm_start == addr) && (vma->vm_flags & VM_GROWSUP); +} + +static inline int stack_guard_page_end(struct vm_area_struct *vma, + unsigned long addr) +{ + return (vma->vm_flags & VM_GROWSUP) && + (vma->vm_end == addr) && + !vma_growsup(vma->vm_next, addr); +} + +extern pid_t +vm_is_stack(struct task_struct *task, struct vm_area_struct *vma, int in_group); + +extern unsigned long move_page_tables(struct vm_area_struct *vma, + unsigned long old_addr, struct vm_area_struct *new_vma, + unsigned long new_addr, unsigned long len, + bool need_rmap_locks); +extern unsigned long change_protection(struct vm_area_struct *vma, unsigned long start, + unsigned long end, pgprot_t newprot, + int dirty_accountable, int prot_numa); +extern int mprotect_fixup(struct vm_area_struct *vma, + struct vm_area_struct **pprev, unsigned long start, + unsigned long end, unsigned long newflags); + +/* + * doesn't attempt to fault and will return short. + */ +int __get_user_pages_fast(unsigned long start, int nr_pages, int write, + struct page **pages); +/* + * per-process(per-mm_struct) statistics. + */ +static inline unsigned long get_mm_counter(struct mm_struct *mm, int member) +{ + long val = atomic_long_read(&mm->rss_stat.count[member]); + +#ifdef SPLIT_RSS_COUNTING + /* + * counter is updated in asynchronous manner and may go to minus. + * But it's never be expected number for users. + */ + if (val < 0) + val = 0; +#endif + return (unsigned long)val; +} + +static inline void add_mm_counter(struct mm_struct *mm, int member, long value) +{ + atomic_long_add(value, &mm->rss_stat.count[member]); +} + +static inline void inc_mm_counter(struct mm_struct *mm, int member) +{ + atomic_long_inc(&mm->rss_stat.count[member]); +} + +static inline void dec_mm_counter(struct mm_struct *mm, int member) +{ + atomic_long_dec(&mm->rss_stat.count[member]); +} + +static inline unsigned long get_mm_rss(struct mm_struct *mm) +{ + return get_mm_counter(mm, MM_FILEPAGES) + + get_mm_counter(mm, MM_ANONPAGES); +} + +static inline unsigned long get_mm_hiwater_rss(struct mm_struct *mm) +{ + return max(mm->hiwater_rss, get_mm_rss(mm)); +} + +static inline unsigned long get_mm_hiwater_vm(struct mm_struct *mm) +{ + return max(mm->hiwater_vm, mm->total_vm); +} + +static inline void update_hiwater_rss(struct mm_struct *mm) +{ + unsigned long _rss = get_mm_rss(mm); + + if ((mm)->hiwater_rss < _rss) + (mm)->hiwater_rss = _rss; +} + +static inline void update_hiwater_vm(struct mm_struct *mm) +{ + if (mm->hiwater_vm < mm->total_vm) + mm->hiwater_vm = mm->total_vm; +} + +static inline void setmax_mm_hiwater_rss(unsigned long *maxrss, + struct mm_struct *mm) +{ + unsigned long hiwater_rss = get_mm_hiwater_rss(mm); + + if (*maxrss < hiwater_rss) + *maxrss = hiwater_rss; +} + +#if defined(SPLIT_RSS_COUNTING) +void sync_mm_rss(struct mm_struct *mm); +#else +static inline void sync_mm_rss(struct mm_struct *mm) +{ +} +#endif + +int vma_wants_writenotify(struct vm_area_struct *vma); + +extern pte_t *__get_locked_pte(struct mm_struct *mm, unsigned long addr, + spinlock_t **ptl); +static inline pte_t *get_locked_pte(struct mm_struct *mm, unsigned long addr, + spinlock_t **ptl) +{ + pte_t *ptep; + __cond_lock(*ptl, ptep = __get_locked_pte(mm, addr, ptl)); + return ptep; +} + +#ifdef __PAGETABLE_PUD_FOLDED +static inline int __pud_alloc(struct mm_struct *mm, pgd_t *pgd, + unsigned long address) +{ + return 0; +} +#else +int __pud_alloc(struct mm_struct *mm, pgd_t *pgd, unsigned long address); +#endif + +#ifdef __PAGETABLE_PMD_FOLDED +static inline int __pmd_alloc(struct mm_struct *mm, pud_t *pud, + unsigned long address) +{ + return 0; +} +#else +int __pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address); +#endif + +int __pte_alloc(struct mm_struct *mm, struct vm_area_struct *vma, + pmd_t *pmd, unsigned long address); +int __pte_alloc_kernel(pmd_t *pmd, unsigned long address); + +/* + * The following ifdef needed to get the 4level-fixup.h header to work. + * Remove it when 4level-fixup.h has been removed. + */ +#if defined(CONFIG_MMU) && !defined(__ARCH_HAS_4LEVEL_HACK) +static inline pud_t *pud_alloc(struct mm_struct *mm, pgd_t *pgd, unsigned long address) +{ + return (unlikely(pgd_none(*pgd)) && __pud_alloc(mm, pgd, address))? + NULL: pud_offset(pgd, address); +} + +static inline pmd_t *pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address) +{ + return (unlikely(pud_none(*pud)) && __pmd_alloc(mm, pud, address))? + NULL: pmd_offset(pud, address); +} +#endif /* CONFIG_MMU && !__ARCH_HAS_4LEVEL_HACK */ + +#if USE_SPLIT_PTLOCKS +/* + * We tuck a spinlock to guard each pagetable page into its struct page, + * at page->private, with BUILD_BUG_ON to make sure that this will not + * overflow into the next struct page (as it might with DEBUG_SPINLOCK). + * When freeing, reset page->mapping so free_pages_check won't complain. + */ +#define __pte_lockptr(page) &((page)->ptl) +#define pte_lock_init(_page) do { \ + spin_lock_init(__pte_lockptr(_page)); \ +} while (0) +#define pte_lock_deinit(page) ((page)->mapping = NULL) +#define pte_lockptr(mm, pmd) ({(void)(mm); __pte_lockptr(pmd_page(*(pmd)));}) +#else /* !USE_SPLIT_PTLOCKS */ +/* + * We use mm->page_table_lock to guard all pagetable pages of the mm. + */ +#define pte_lock_init(page) do {} while (0) +#define pte_lock_deinit(page) do {} while (0) +#define pte_lockptr(mm, pmd) ({(void)(pmd); &(mm)->page_table_lock;}) +#endif /* USE_SPLIT_PTLOCKS */ + +static inline bool pgtable_page_ctor(struct page *page) +{ + pte_lock_init(page); + inc_zone_page_state(page, NR_PAGETABLE); + return true; +} + +static inline void pgtable_page_dtor(struct page *page) +{ + pte_lock_deinit(page); + dec_zone_page_state(page, NR_PAGETABLE); +} + +#define pte_offset_map_lock(mm, pmd, address, ptlp) \ +({ \ + spinlock_t *__ptl = pte_lockptr(mm, pmd); \ + pte_t *__pte = pte_offset_map(pmd, address); \ + *(ptlp) = __ptl; \ + spin_lock(__ptl); \ + __pte; \ +}) + +#define pte_unmap_unlock(pte, ptl) do { \ + spin_unlock(ptl); \ + pte_unmap(pte); \ +} while (0) + +#define pte_alloc_map(mm, vma, pmd, address) \ + ((unlikely(pmd_none(*(pmd))) && __pte_alloc(mm, vma, \ + pmd, address))? \ + NULL: pte_offset_map(pmd, address)) + +#define pte_alloc_map_lock(mm, pmd, address, ptlp) \ + ((unlikely(pmd_none(*(pmd))) && __pte_alloc(mm, NULL, \ + pmd, address))? \ + NULL: pte_offset_map_lock(mm, pmd, address, ptlp)) + +#define pte_alloc_kernel(pmd, address) \ + ((unlikely(pmd_none(*(pmd))) && __pte_alloc_kernel(pmd, address))? \ + NULL: pte_offset_kernel(pmd, address)) + +extern void free_area_init(unsigned long * zones_size); +extern void free_area_init_node(int nid, unsigned long * zones_size, + unsigned long zone_start_pfn, unsigned long *zholes_size); +extern void free_initmem(void); + +/* + * Free reserved pages within range [PAGE_ALIGN(start), end & PAGE_MASK) + * into the buddy system. The freed pages will be poisoned with pattern + * "poison" if it's non-zero. + * Return pages freed into the buddy system. + */ +extern unsigned long free_reserved_area(unsigned long start, unsigned long end, + int poison, char *s); +#ifdef CONFIG_HIGHMEM +/* + * Free a highmem page into the buddy system, adjusting totalhigh_pages + * and totalram_pages. + */ +extern void free_highmem_page(struct page *page); +#endif + +static inline void adjust_managed_page_count(struct page *page, long count) +{ + totalram_pages += count; +} + +/* Free the reserved page into the buddy system, so it gets managed. */ +static inline void __free_reserved_page(struct page *page) +{ + ClearPageReserved(page); + init_page_count(page); + __free_page(page); +} + +static inline void free_reserved_page(struct page *page) +{ + __free_reserved_page(page); + adjust_managed_page_count(page, 1); +} + +static inline void mark_page_reserved(struct page *page) +{ + SetPageReserved(page); + adjust_managed_page_count(page, -1); +} + +/* + * Default method to free all the __init memory into the buddy system. + * The freed pages will be poisoned with pattern "poison" if it is + * non-zero. Return pages freed into the buddy system. + */ +static inline unsigned long free_initmem_default(int poison) +{ + extern char __init_begin[], __init_end[]; + + return free_reserved_area(PAGE_ALIGN((unsigned long)&__init_begin) , + ((unsigned long)&__init_end) & PAGE_MASK, + poison, "unused kernel"); +} + +#ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP +/* + * With CONFIG_HAVE_MEMBLOCK_NODE_MAP set, an architecture may initialise its + * zones, allocate the backing mem_map and account for memory holes in a more + * architecture independent manner. This is a substitute for creating the + * zone_sizes[] and zholes_size[] arrays and passing them to + * free_area_init_node() + * + * An architecture is expected to register range of page frames backed by + * physical memory with memblock_add[_node]() before calling + * free_area_init_nodes() passing in the PFN each zone ends at. At a basic + * usage, an architecture is expected to do something like + * + * unsigned long max_zone_pfns[MAX_NR_ZONES] = {max_dma, max_normal_pfn, + * max_highmem_pfn}; + * for_each_valid_physical_page_range() + * memblock_add_node(base, size, nid) + * free_area_init_nodes(max_zone_pfns); + * + * free_bootmem_with_active_regions() calls free_bootmem_node() for each + * registered physical page range. Similarly + * sparse_memory_present_with_active_regions() calls memory_present() for + * each range when SPARSEMEM is enabled. + * + * See mm/page_alloc.c for more information on each function exposed by + * CONFIG_HAVE_MEMBLOCK_NODE_MAP. + */ +extern void free_area_init_nodes(unsigned long *max_zone_pfn); +unsigned long node_map_pfn_alignment(void); +unsigned long __absent_pages_in_range(int nid, unsigned long start_pfn, + unsigned long end_pfn); +extern unsigned long absent_pages_in_range(unsigned long start_pfn, + unsigned long end_pfn); +extern void get_pfn_range_for_nid(unsigned int nid, + unsigned long *start_pfn, unsigned long *end_pfn); +extern unsigned long find_min_pfn_with_active_regions(void); +extern void free_bootmem_with_active_regions(int nid, + unsigned long max_low_pfn); +extern void sparse_memory_present_with_active_regions(int nid); + +#endif /* CONFIG_HAVE_MEMBLOCK_NODE_MAP */ + +#if !defined(CONFIG_HAVE_MEMBLOCK_NODE_MAP) && \ + !defined(CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID) +static inline int __early_pfn_to_nid(unsigned long pfn) +{ + return 0; +} +#else +/* please see mm/page_alloc.c */ +extern int __meminit early_pfn_to_nid(unsigned long pfn); +#ifdef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID +/* there is a per-arch backend function. */ +extern int __meminit __early_pfn_to_nid(unsigned long pfn); +#endif /* CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID */ +#endif + +extern void set_dma_reserve(unsigned long new_dma_reserve); +extern void memmap_init_zone(unsigned long, int, unsigned long, + unsigned long, enum memmap_context); +extern void setup_per_zone_wmarks(void); +extern int __meminit init_per_zone_wmark_min(void); +extern void mem_init(void); +extern void __init mmap_init(void); +extern void show_mem(unsigned int flags); +extern void si_meminfo(struct sysinfo * val); +extern void si_meminfo_node(struct sysinfo *val, int nid); + +extern __printf(3, 4) +void warn_alloc_failed(gfp_t gfp_mask, int order, const char *fmt, ...); + +extern void setup_per_cpu_pageset(void); + +extern void zone_pcp_update(struct zone *zone); +extern void zone_pcp_reset(struct zone *zone); + +/* page_alloc.c */ +extern int min_free_kbytes; + +/* nommu.c */ +extern atomic_long_t mmap_pages_allocated; +extern int nommu_shrink_inode_mappings(struct inode *, size_t, size_t); + +/* interval_tree.c */ +void vma_interval_tree_insert(struct vm_area_struct *node, + struct rb_root *root); +void vma_interval_tree_insert_after(struct vm_area_struct *node, + struct vm_area_struct *prev, + struct rb_root *root); +void vma_interval_tree_remove(struct vm_area_struct *node, + struct rb_root *root); +struct vm_area_struct *vma_interval_tree_iter_first(struct rb_root *root, + unsigned long start, unsigned long last); +struct vm_area_struct *vma_interval_tree_iter_next(struct vm_area_struct *node, + unsigned long start, unsigned long last); + +#define vma_interval_tree_foreach(vma, root, start, last) \ + for (vma = vma_interval_tree_iter_first(root, start, last); \ + vma; vma = vma_interval_tree_iter_next(vma, start, last)) + +static inline void vma_nonlinear_insert(struct vm_area_struct *vma, + struct list_head *list) +{ + list_add_tail(&vma->shared.nonlinear, list); +} + +void anon_vma_interval_tree_insert(struct anon_vma_chain *node, + struct rb_root *root); +void anon_vma_interval_tree_remove(struct anon_vma_chain *node, + struct rb_root *root); +struct anon_vma_chain *anon_vma_interval_tree_iter_first( + struct rb_root *root, unsigned long start, unsigned long last); +struct anon_vma_chain *anon_vma_interval_tree_iter_next( + struct anon_vma_chain *node, unsigned long start, unsigned long last); +#ifdef CONFIG_DEBUG_VM_RB +void anon_vma_interval_tree_verify(struct anon_vma_chain *node); +#endif + +#define anon_vma_interval_tree_foreach(avc, root, start, last) \ + for (avc = anon_vma_interval_tree_iter_first(root, start, last); \ + avc; avc = anon_vma_interval_tree_iter_next(avc, start, last)) + +/* mmap.c */ +extern int __vm_enough_memory(struct mm_struct *mm, long pages, int cap_sys_admin); +extern int vma_adjust(struct vm_area_struct *vma, unsigned long start, + unsigned long end, pgoff_t pgoff, struct vm_area_struct *insert); +extern struct vm_area_struct *vma_merge(struct mm_struct *, + struct vm_area_struct *prev, unsigned long addr, unsigned long end, + unsigned long vm_flags, struct anon_vma *, struct file *, pgoff_t, + struct mempolicy *, const char __user *); +extern struct anon_vma *find_mergeable_anon_vma(struct vm_area_struct *); +extern int split_vma(struct mm_struct *, + struct vm_area_struct *, unsigned long addr, int new_below); +extern int insert_vm_struct(struct mm_struct *, struct vm_area_struct *); +extern void __vma_link_rb(struct mm_struct *, struct vm_area_struct *, + struct rb_node **, struct rb_node *); +extern void unlink_file_vma(struct vm_area_struct *); +extern struct vm_area_struct *copy_vma(struct vm_area_struct **, + unsigned long addr, unsigned long len, pgoff_t pgoff, + bool *need_rmap_locks); +extern void exit_mmap(struct mm_struct *); + +extern int mm_take_all_locks(struct mm_struct *mm); +extern void mm_drop_all_locks(struct mm_struct *mm); + +extern void set_mm_exe_file(struct mm_struct *mm, struct file *new_exe_file); +extern struct file *get_mm_exe_file(struct mm_struct *mm); + +extern int may_expand_vm(struct mm_struct *mm, unsigned long npages); +extern int install_special_mapping(struct mm_struct *mm, + unsigned long addr, unsigned long len, + unsigned long flags, struct page **pages); + +extern unsigned long get_unmapped_area(struct file *, unsigned long, unsigned long, unsigned long, unsigned long); + +extern unsigned long mmap_region(struct file *file, unsigned long addr, + unsigned long len, vm_flags_t vm_flags, unsigned long pgoff); +extern unsigned long do_mmap_pgoff(struct file *file, unsigned long addr, + unsigned long len, unsigned long prot, unsigned long flags, + unsigned long pgoff, unsigned long *populate); +extern int do_munmap(struct mm_struct *, unsigned long, size_t); + +#ifdef CONFIG_MMU +extern int __mm_populate(unsigned long addr, unsigned long len, + int ignore_errors); +static inline void mm_populate(unsigned long addr, unsigned long len) +{ + /* Ignore errors */ + (void) __mm_populate(addr, len, 1); +} +#else +static inline void mm_populate(unsigned long addr, unsigned long len) {} +#endif + +/* These take the mm semaphore themselves */ +extern unsigned long vm_brk(unsigned long, unsigned long); +extern int vm_munmap(unsigned long, size_t); +extern unsigned long vm_mmap(struct file *, unsigned long, + unsigned long, unsigned long, + unsigned long, unsigned long); + +struct vm_unmapped_area_info { +#define VM_UNMAPPED_AREA_TOPDOWN 1 + unsigned long flags; + unsigned long length; + unsigned long low_limit; + unsigned long high_limit; + unsigned long align_mask; + unsigned long align_offset; +}; + +extern unsigned long unmapped_area(struct vm_unmapped_area_info *info); +extern unsigned long unmapped_area_topdown(struct vm_unmapped_area_info *info); + +/* + * Search for an unmapped address range. + * + * We are looking for a range that: + * - does not intersect with any VMA; + * - is contained within the [low_limit, high_limit) interval; + * - is at least the desired size. + * - satisfies (begin_addr & align_mask) == (align_offset & align_mask) + */ +static inline unsigned long +vm_unmapped_area(struct vm_unmapped_area_info *info) +{ + if (!(info->flags & VM_UNMAPPED_AREA_TOPDOWN)) + return unmapped_area(info); + else + return unmapped_area_topdown(info); +} + +/* truncate.c */ +extern void truncate_inode_pages(struct address_space *, loff_t); +extern void truncate_inode_pages_range(struct address_space *, + loff_t lstart, loff_t lend); + +/* generic vm_area_ops exported for stackable file systems */ +extern int filemap_fault(struct vm_area_struct *, struct vm_fault *); +extern int filemap_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf); + +/* mm/page-writeback.c */ +int write_one_page(struct page *page, int wait); +void task_dirty_inc(struct task_struct *tsk); + +/* readahead.c */ +#define VM_MAX_READAHEAD 128 /* kbytes */ +#define VM_MIN_READAHEAD 16 /* kbytes (includes current page) */ + +int force_page_cache_readahead(struct address_space *mapping, struct file *filp, + pgoff_t offset, unsigned long nr_to_read); + +void page_cache_sync_readahead(struct address_space *mapping, + struct file_ra_state *ra, + struct file *filp, + pgoff_t offset, + unsigned long size); + +void page_cache_async_readahead(struct address_space *mapping, + struct file_ra_state *ra, + struct file *filp, + struct page *pg, + pgoff_t offset, + unsigned long size); + +unsigned long max_sane_readahead(unsigned long nr); +unsigned long ra_submit(struct file_ra_state *ra, + struct address_space *mapping, + struct file *filp); + +/* Generic expand stack which grows the stack according to GROWS{UP,DOWN} */ +extern int expand_stack(struct vm_area_struct *vma, unsigned long address); + +/* CONFIG_STACK_GROWSUP still needs to to grow downwards at some places */ +extern int expand_downwards(struct vm_area_struct *vma, + unsigned long address); +#if VM_GROWSUP +extern int expand_upwards(struct vm_area_struct *vma, unsigned long address); +#else + #define expand_upwards(vma, address) do { } while (0) +#endif + +/* Look up the first VMA which satisfies addr < vm_end, NULL if none. */ +extern struct vm_area_struct * find_vma(struct mm_struct * mm, unsigned long addr); +extern struct vm_area_struct * find_vma_prev(struct mm_struct * mm, unsigned long addr, + struct vm_area_struct **pprev); + +/* Look up the first VMA which intersects the interval start_addr..end_addr-1, + NULL if none. Assume start_addr < end_addr. */ +static inline struct vm_area_struct * find_vma_intersection(struct mm_struct * mm, unsigned long start_addr, unsigned long end_addr) +{ + struct vm_area_struct * vma = find_vma(mm,start_addr); + + if (vma && end_addr <= vma->vm_start) + vma = NULL; + return vma; +} + +static inline unsigned long vma_pages(struct vm_area_struct *vma) +{ + return (vma->vm_end - vma->vm_start) >> PAGE_SHIFT; +} + +/* Look up the first VMA which exactly match the interval vm_start ... vm_end */ +static inline struct vm_area_struct *find_exact_vma(struct mm_struct *mm, + unsigned long vm_start, unsigned long vm_end) +{ + struct vm_area_struct *vma = find_vma(mm, vm_start); + + if (vma && (vma->vm_start != vm_start || vma->vm_end != vm_end)) + vma = NULL; + + return vma; +} + +#ifdef CONFIG_MMU +pgprot_t vm_get_page_prot(unsigned long vm_flags); +#else +static inline pgprot_t vm_get_page_prot(unsigned long vm_flags) +{ + return __pgprot(0); +} +#endif + +#ifdef CONFIG_ARCH_USES_NUMA_PROT_NONE +unsigned long change_prot_numa(struct vm_area_struct *vma, + unsigned long start, unsigned long end); +#endif + +struct vm_area_struct *find_extend_vma(struct mm_struct *, unsigned long addr); +int remap_pfn_range(struct vm_area_struct *, unsigned long addr, + unsigned long pfn, unsigned long size, pgprot_t); +int vm_insert_page(struct vm_area_struct *, unsigned long addr, struct page *); +int vm_insert_pfn(struct vm_area_struct *vma, unsigned long addr, + unsigned long pfn); +int vm_insert_mixed(struct vm_area_struct *vma, unsigned long addr, + unsigned long pfn); +int vm_iomap_memory(struct vm_area_struct *vma, phys_addr_t start, unsigned long len); + + +struct page *follow_page_mask(struct vm_area_struct *vma, + unsigned long address, unsigned int foll_flags, + unsigned int *page_mask); + +static inline struct page *follow_page(struct vm_area_struct *vma, + unsigned long address, unsigned int foll_flags) +{ + unsigned int unused_page_mask; + return follow_page_mask(vma, address, foll_flags, &unused_page_mask); +} + +#define FOLL_WRITE 0x01 /* check pte is writable */ +#define FOLL_TOUCH 0x02 /* mark page accessed */ +#define FOLL_GET 0x04 /* do get_page on page */ +#define FOLL_DUMP 0x08 /* give error on hole if it would be zero */ +#define FOLL_FORCE 0x10 /* get_user_pages read/write w/o permission */ +#define FOLL_NOWAIT 0x20 /* if a disk transfer is needed, start the IO + * and return without waiting upon it */ +#define FOLL_MLOCK 0x40 /* mark page as mlocked */ +#define FOLL_SPLIT 0x80 /* don't return transhuge pages, split them */ +#define FOLL_HWPOISON 0x100 /* check page is hwpoisoned */ +#define FOLL_NUMA 0x200 /* force NUMA hinting page fault */ +#define FOLL_MIGRATION 0x400 /* wait for page to replace migration entry */ +#define FOLL_DURABLE 0x800 /* get the page reference for a long time */ + +typedef int (*pte_fn_t)(pte_t *pte, pgtable_t token, unsigned long addr, + void *data); +extern int apply_to_page_range(struct mm_struct *mm, unsigned long address, + unsigned long size, pte_fn_t fn, void *data); + +/* + * get_user_pages() - pin user pages in memory + * @tsk: the task_struct to use for page fault accounting, or + * NULL if faults are not to be recorded. + * @mm: mm_struct of target mm + * @start: starting user address + * @nr_pages: number of pages from start to pin + * @write: whether pages will be written to by the caller + * @force: whether to force write access even if user mapping is + * readonly. This will result in the page being COWed even + * in MAP_SHARED mappings. You do not want this. + * @pages: array that receives pointers to the pages pinned. + * Should be at least nr_pages long. Or NULL, if caller + * only intends to ensure the pages are faulted in. + * @vmas: array of pointers to vmas corresponding to each page. + * Or NULL if the caller does not require them. + * + * Returns number of pages pinned. This may be fewer than the number + * requested. If nr_pages is 0 or negative, returns 0. If no pages + * were pinned, returns -errno. Each page returned must be released + * with a put_page() call when it is finished with. vmas will only + * remain valid while mmap_sem is held. + * + * Must be called with mmap_sem held for read or write. + * + * get_user_pages walks a process's page tables and takes a reference to + * each struct page that each user address corresponds to at a given + * instant. That is, it takes the page that would be accessed if a user + * thread accesses the given user virtual address at that instant. + * + * This does not guarantee that the page exists in the user mappings when + * get_user_pages returns, and there may even be a completely different + * page there in some cases (eg. if mmapped pagecache has been invalidated + * and subsequently re faulted). However it does guarantee that the page + * won't be freed completely. And mostly callers simply care that the page + * contains data that was valid *at some point in time*. Typically, an IO + * or similar operation cannot guarantee anything stronger anyway because + * locks can't be held over the syscall boundary. + * + * If write=0, the page must not be written to. If the page is written to, + * set_page_dirty (or set_page_dirty_lock, as appropriate) must be called + * after the page is finished with, and before put_page is called. + * + * get_user_pages is typically used for fewer-copy IO operations, to get a + * handle on the memory by some means other than accesses via the user virtual + * addresses. The pages may be submitted for DMA to devices or accessed via + * their kernel linear mapping (via the kmap APIs). Care should be taken to + * use the correct cache flushing APIs. + * + * See also get_user_pages_fast, for performance critical applications. + */ +static inline long get_user_pages(struct task_struct *tsk, struct mm_struct *mm, + unsigned long start, unsigned long nr_pages, int write, + int force, struct page **pages, + struct vm_area_struct **vmas) +{ + int flags = FOLL_TOUCH; + + if (pages) + flags |= FOLL_GET; + if (write) + flags |= FOLL_WRITE; + if (force) + flags |= FOLL_FORCE; + + return __get_user_pages(tsk, mm, start, nr_pages, flags, pages, vmas, + NULL); +} + +#ifdef CONFIG_PROC_FS +void vm_stat_account(struct mm_struct *, unsigned long, struct file *, long); +#else +static inline void vm_stat_account(struct mm_struct *mm, + unsigned long flags, struct file *file, long pages) +{ + mm->total_vm += pages; +} +#endif /* CONFIG_PROC_FS */ + +#ifdef CONFIG_DEBUG_PAGEALLOC +extern void kernel_map_pages(struct page *page, int numpages, int enable); +#ifdef CONFIG_HIBERNATION +extern bool kernel_page_present(struct page *page); +#endif /* CONFIG_HIBERNATION */ +#else +static inline void +kernel_map_pages(struct page *page, int numpages, int enable) {} +#ifdef CONFIG_HIBERNATION +static inline bool kernel_page_present(struct page *page) { return true; } +#endif /* CONFIG_HIBERNATION */ +#endif + +extern struct vm_area_struct *get_gate_vma(struct mm_struct *mm); +#ifdef __HAVE_ARCH_GATE_AREA +int in_gate_area_no_mm(unsigned long addr); +int in_gate_area(struct mm_struct *mm, unsigned long addr); +#else +int in_gate_area_no_mm(unsigned long addr); +#define in_gate_area(mm, addr) ({(void)mm; in_gate_area_no_mm(addr);}) +#endif /* __HAVE_ARCH_GATE_AREA */ + +#ifdef CONFIG_SYSCTL +extern int sysctl_drop_caches; +int drop_caches_sysctl_handler(struct ctl_table *, int, + void __user *, size_t *, loff_t *); +#endif + +unsigned long shrink_slab(struct shrink_control *shrink, + unsigned long nr_pages_scanned, + unsigned long lru_pages); + +#ifndef CONFIG_MMU +#define randomize_va_space 0 +#else +extern int randomize_va_space; +#endif + +const char * arch_vma_name(struct vm_area_struct *vma); +void print_vma_addr(char *prefix, unsigned long rip); + +void sparse_mem_maps_populate_node(struct page **map_map, + unsigned long pnum_begin, + unsigned long pnum_end, + unsigned long map_count, + int nodeid); + +struct page *sparse_mem_map_populate(unsigned long pnum, int nid); +pgd_t *vmemmap_pgd_populate(unsigned long addr, int node); +pud_t *vmemmap_pud_populate(pgd_t *pgd, unsigned long addr, int node); +pmd_t *vmemmap_pmd_populate(pud_t *pud, unsigned long addr, int node); +pte_t *vmemmap_pte_populate(pmd_t *pmd, unsigned long addr, int node); +void *vmemmap_alloc_block(unsigned long size, int node); +void *vmemmap_alloc_block_buf(unsigned long size, int node); +void vmemmap_verify(pte_t *, int, unsigned long, unsigned long); +int vmemmap_populate_basepages(unsigned long start, unsigned long end, + int node); +int vmemmap_populate(unsigned long start, unsigned long end, int node); +void vmemmap_populate_print_last(void); +#ifdef CONFIG_MEMORY_HOTPLUG +void vmemmap_free(unsigned long start, unsigned long end); +#endif +void register_page_bootmem_memmap(unsigned long section_nr, struct page *map, + unsigned long size); + +enum mf_flags { + MF_COUNT_INCREASED = 1 << 0, + MF_ACTION_REQUIRED = 1 << 1, + MF_MUST_KILL = 1 << 2, +}; +extern int memory_failure(unsigned long pfn, int trapno, int flags); +extern void memory_failure_queue(unsigned long pfn, int trapno, int flags); +extern int unpoison_memory(unsigned long pfn); +extern int sysctl_memory_failure_early_kill; +extern int sysctl_memory_failure_recovery; +extern void shake_page(struct page *p, int access); +extern atomic_long_t num_poisoned_pages; +extern int soft_offline_page(struct page *page, int flags); + +extern void dump_page(struct page *page); + +#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLBFS) +extern void clear_huge_page(struct page *page, + unsigned long addr, + unsigned int pages_per_huge_page); +extern void copy_user_huge_page(struct page *dst, struct page *src, + unsigned long addr, struct vm_area_struct *vma, + unsigned int pages_per_huge_page); +#endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLBFS */ + +#ifdef CONFIG_DEBUG_PAGEALLOC +extern unsigned int _debug_guardpage_minorder; + +static inline unsigned int debug_guardpage_minorder(void) +{ + return _debug_guardpage_minorder; +} + +static inline bool page_is_guard(struct page *page) +{ + return test_bit(PAGE_DEBUG_FLAG_GUARD, &page->debug_flags); +} +#else +static inline unsigned int debug_guardpage_minorder(void) { return 0; } +static inline bool page_is_guard(struct page *page) { return false; } +#endif /* CONFIG_DEBUG_PAGEALLOC */ + +#if MAX_NUMNODES > 1 +void __init setup_nr_node_ids(void); +#else +static inline void setup_nr_node_ids(void) {} +#endif + +#endif /* __KERNEL__ */ +#endif /* _LINUX_MM_H */ diff -Naur kernel.21.3.orig/include/linux/mm_types.h kernel.21.3.new/include/linux/mm_types.h --- kernel.21.3.orig/include/linux/mm_types.h 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/include/linux/mm_types.h 2015-04-25 23:28:45.321481513 +0000 @@ -213,6 +213,7 @@ unsigned long vm_top; /* region allocated to here */ unsigned long vm_pgoff; /* the offset in vm_file corresponding to vm_start */ struct file *vm_file; /* the backing file or NULL */ + struct file *vm_prfile; /* the virtual backing file or NULL */ int vm_usage; /* region usage count (access under nommu_region_sem) */ bool vm_icache_flushed : 1; /* true if the icache has been flushed for @@ -286,6 +287,7 @@ unsigned long vm_pgoff; /* Offset (within vm_file) in PAGE_SIZE units, *not* PAGE_CACHE_SIZE */ struct file * vm_file; /* File we map to (can be NULL). */ + struct file *vm_prfile; /* shadow of vm_file */ void * vm_private_data; /* was vm_pte (shared mem) */ #ifndef CONFIG_MMU diff -Naur kernel.21.3.orig/include/linux/mm_types.h.orig kernel.21.3.new/include/linux/mm_types.h.orig --- kernel.21.3.orig/include/linux/mm_types.h.orig 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/include/linux/mm_types.h.orig 2015-02-03 01:44:32.000000000 +0000 @@ -0,0 +1,524 @@ +#ifndef _LINUX_MM_TYPES_H +#define _LINUX_MM_TYPES_H + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#ifndef AT_VECTOR_SIZE_ARCH +#define AT_VECTOR_SIZE_ARCH 0 +#endif +#define AT_VECTOR_SIZE (2*(AT_VECTOR_SIZE_ARCH + AT_VECTOR_SIZE_BASE + 1)) + +struct address_space; + +#define USE_SPLIT_PTLOCKS (NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS) + +/* + * Each physical page in the system has a struct page associated with + * it to keep track of whatever it is we are using the page for at the + * moment. Note that we have no way to track which tasks are using + * a page, though if it is a pagecache page, rmap structures can tell us + * who is mapping it. + * + * The objects in struct page are organized in double word blocks in + * order to allows us to use atomic double word operations on portions + * of struct page. That is currently only used by slub but the arrangement + * allows the use of atomic double word operations on the flags/mapping + * and lru list pointers also. + */ +struct page { + /* First double word block */ + unsigned long flags; /* Atomic flags, some possibly + * updated asynchronously */ + struct address_space *mapping; /* If low bit clear, points to + * inode address_space, or NULL. + * If page mapped as anonymous + * memory, low bit is set, and + * it points to anon_vma object: + * see PAGE_MAPPING_ANON below. + */ + /* Second double word */ + struct { + union { + pgoff_t index; /* Our offset within mapping. */ + void *freelist; /* slub/slob first free object */ + bool pfmemalloc; /* If set by the page allocator, + * ALLOC_NO_WATERMARKS was set + * and the low watermark was not + * met implying that the system + * is under some pressure. The + * caller should try ensure + * this page is only used to + * free other pages. + */ + }; + + union { +#if defined(CONFIG_HAVE_CMPXCHG_DOUBLE) && \ + defined(CONFIG_HAVE_ALIGNED_STRUCT_PAGE) + /* Used for cmpxchg_double in slub */ + unsigned long counters; +#else + /* + * Keep _count separate from slub cmpxchg_double data. + * As the rest of the double word is protected by + * slab_lock but _count is not. + */ + unsigned counters; +#endif + + struct { + + union { + /* + * Count of ptes mapped in + * mms, to show when page is + * mapped & limit reverse map + * searches. + * + * Used also for tail pages + * refcounting instead of + * _count. Tail pages cannot + * be mapped and keeping the + * tail page _count zero at + * all times guarantees + * get_page_unless_zero() will + * never succeed on tail + * pages. + */ + atomic_t _mapcount; + + struct { /* SLUB */ + unsigned inuse:16; + unsigned objects:15; + unsigned frozen:1; + }; + int units; /* SLOB */ + }; + atomic_t _count; /* Usage count, see below. */ + }; + }; + }; + + /* Third double word block */ + union { + struct list_head lru; /* Pageout list, eg. active_list + * protected by zone->lru_lock ! + */ + struct { /* slub per cpu partial pages */ + struct page *next; /* Next partial slab */ +#ifdef CONFIG_64BIT + int pages; /* Nr of partial slabs left */ + int pobjects; /* Approximate # of objects */ +#else + short int pages; + short int pobjects; +#endif + }; + + struct list_head list; /* slobs list of pages */ + struct slab *slab_page; /* slab fields */ + }; + + /* Remainder is not double word aligned */ + union { + unsigned long private; /* Mapping-private opaque data: + * usually used for buffer_heads + * if PagePrivate set; used for + * swp_entry_t if PageSwapCache; + * indicates order in the buddy + * system if PG_buddy is set. + */ +#if USE_SPLIT_PTLOCKS + spinlock_t ptl; +#endif + struct kmem_cache *slab_cache; /* SL[AU]B: Pointer to slab */ + struct page *first_page; /* Compound tail pages */ + }; + + /* + * On machines where all RAM is mapped into kernel address space, + * we can simply calculate the virtual address. On machines with + * highmem some memory is mapped into kernel virtual memory + * dynamically, so we need a place to store that address. + * Note that this field could be 16 bits on x86 ... ;) + * + * Architectures with slow multiplication can define + * WANT_PAGE_VIRTUAL in asm/page.h + */ +#if defined(WANT_PAGE_VIRTUAL) + void *virtual; /* Kernel virtual address (NULL if + not kmapped, ie. highmem) */ +#endif /* WANT_PAGE_VIRTUAL */ +#ifdef CONFIG_WANT_PAGE_DEBUG_FLAGS + unsigned long debug_flags; /* Use atomic bitops on this */ +#endif + +#ifdef CONFIG_KMEMCHECK + /* + * kmemcheck wants to track the status of each byte in a page; this + * is a pointer to such a status block. NULL if not tracked. + */ + void *shadow; +#endif + +#ifdef LAST_NID_NOT_IN_PAGE_FLAGS + int _last_nid; +#endif +} +/* + * The struct page can be forced to be double word aligned so that atomic ops + * on double words work. The SLUB allocator can make use of such a feature. + */ +#ifdef CONFIG_HAVE_ALIGNED_STRUCT_PAGE + __aligned(2 * sizeof(unsigned long)) +#endif +; + +struct page_frag { + struct page *page; +#if (BITS_PER_LONG > 32) || (PAGE_SIZE >= 65536) + __u32 offset; + __u32 size; +#else + __u16 offset; + __u16 size; +#endif +}; + +typedef unsigned long __nocast vm_flags_t; + +/* + * A region containing a mapping of a non-memory backed file under NOMMU + * conditions. These are held in a global tree and are pinned by the VMAs that + * map parts of them. + */ +struct vm_region { + struct rb_node vm_rb; /* link in global region tree */ + vm_flags_t vm_flags; /* VMA vm_flags */ + unsigned long vm_start; /* start address of region */ + unsigned long vm_end; /* region initialised to here */ + unsigned long vm_top; /* region allocated to here */ + unsigned long vm_pgoff; /* the offset in vm_file corresponding to vm_start */ + struct file *vm_file; /* the backing file or NULL */ + + int vm_usage; /* region usage count (access under nommu_region_sem) */ + bool vm_icache_flushed : 1; /* true if the icache has been flushed for + * this region */ +}; + +/* + * This struct defines a memory VMM memory area. There is one of these + * per VM-area/task. A VM area is any part of the process virtual memory + * space that has a special rule for the page-fault handlers (ie a shared + * library, the executable area etc). + */ +struct vm_area_struct { + /* The first cache line has the info for VMA tree walking. */ + + unsigned long vm_start; /* Our start address within vm_mm. */ + unsigned long vm_end; /* The first byte after our end address + within vm_mm. */ + + /* linked list of VM areas per task, sorted by address */ + struct vm_area_struct *vm_next, *vm_prev; + + struct rb_node vm_rb; + + /* + * Largest free memory gap in bytes to the left of this VMA. + * Either between this VMA and vma->vm_prev, or between one of the + * VMAs below us in the VMA rbtree and its ->vm_prev. This helps + * get_unmapped_area find a free area of the right size. + */ + unsigned long rb_subtree_gap; + + /* Second cache line starts here. */ + + struct mm_struct *vm_mm; /* The address space we belong to. */ + pgprot_t vm_page_prot; /* Access permissions of this VMA. */ + unsigned long vm_flags; /* Flags, see mm.h. */ + + /* + * For areas with an address space and backing store, + * linkage into the address_space->i_mmap interval tree, or + * linkage of vma in the address_space->i_mmap_nonlinear list. + * + * For private anonymous mappings, a pointer to a null terminated string + * in the user process containing the name given to the vma, or NULL + * if unnamed. + */ + union { + struct { + struct rb_node rb; + unsigned long rb_subtree_last; + } linear; + struct list_head nonlinear; + const char __user *anon_name; + } shared; + + /* + * A file's MAP_PRIVATE vma can be in both i_mmap tree and anon_vma + * list, after a COW of one of the file pages. A MAP_SHARED vma + * can only be in the i_mmap tree. An anonymous MAP_PRIVATE, stack + * or brk vma (with NULL file) can only be in an anon_vma list. + */ + struct list_head anon_vma_chain; /* Serialized by mmap_sem & + * page_table_lock */ + struct anon_vma *anon_vma; /* Serialized by page_table_lock */ + + /* Function pointers to deal with this struct. */ + const struct vm_operations_struct *vm_ops; + + /* Information about our backing store: */ + unsigned long vm_pgoff; /* Offset (within vm_file) in PAGE_SIZE + units, *not* PAGE_CACHE_SIZE */ + struct file * vm_file; /* File we map to (can be NULL). */ + void * vm_private_data; /* was vm_pte (shared mem) */ + +#ifndef CONFIG_MMU + struct vm_region *vm_region; /* NOMMU mapping region */ +#endif +#ifdef CONFIG_NUMA + struct mempolicy *vm_policy; /* NUMA policy for the VMA */ +#endif +}; + +struct core_thread { + struct task_struct *task; + struct core_thread *next; +}; + +struct core_state { + atomic_t nr_threads; + struct core_thread dumper; + struct completion startup; +}; + +enum { + MM_FILEPAGES, + MM_ANONPAGES, + MM_SWAPENTS, + NR_MM_COUNTERS +}; + +#if USE_SPLIT_PTLOCKS && defined(CONFIG_MMU) +#define SPLIT_RSS_COUNTING +/* per-thread cached information, */ +struct task_rss_stat { + int events; /* for synchronization threshold */ + int count[NR_MM_COUNTERS]; +}; +#endif /* USE_SPLIT_PTLOCKS */ + +struct mm_rss_stat { + atomic_long_t count[NR_MM_COUNTERS]; +}; + +struct mm_struct { + struct vm_area_struct * mmap; /* list of VMAs */ + struct rb_root mm_rb; + struct vm_area_struct * mmap_cache; /* last find_vma result */ +#ifdef CONFIG_MMU + unsigned long (*get_unmapped_area) (struct file *filp, + unsigned long addr, unsigned long len, + unsigned long pgoff, unsigned long flags); + void (*unmap_area) (struct mm_struct *mm, unsigned long addr); +#endif + unsigned long mmap_base; /* base of mmap area */ + unsigned long mmap_legacy_base; /* base of mmap area in bottom-up allocations */ + unsigned long task_size; /* size of task vm space */ + unsigned long cached_hole_size; /* if non-zero, the largest hole below free_area_cache */ + unsigned long free_area_cache; /* first hole of size cached_hole_size or larger */ + unsigned long highest_vm_end; /* highest vma end address */ + pgd_t * pgd; + atomic_t mm_users; /* How many users with user space? */ + atomic_t mm_count; /* How many references to "struct mm_struct" (users count as 1) */ + int map_count; /* number of VMAs */ + + spinlock_t page_table_lock; /* Protects page tables and some counters */ + struct rw_semaphore mmap_sem; + + struct list_head mmlist; /* List of maybe swapped mm's. These are globally strung + * together off init_mm.mmlist, and are protected + * by mmlist_lock + */ + + + unsigned long hiwater_rss; /* High-watermark of RSS usage */ + unsigned long hiwater_vm; /* High-water virtual memory usage */ + + unsigned long total_vm; /* Total pages mapped */ + unsigned long locked_vm; /* Pages that have PG_mlocked set */ + unsigned long pinned_vm; /* Refcount permanently increased */ + unsigned long shared_vm; /* Shared pages (files) */ + unsigned long exec_vm; /* VM_EXEC & ~VM_WRITE */ + unsigned long stack_vm; /* VM_GROWSUP/DOWN */ + unsigned long def_flags; + unsigned long nr_ptes; /* Page table pages */ + unsigned long start_code, end_code, start_data, end_data; + unsigned long start_brk, brk, start_stack; + unsigned long arg_start, arg_end, env_start, env_end; + + unsigned long saved_auxv[AT_VECTOR_SIZE]; /* for /proc/PID/auxv */ + + /* + * Special counters, in some configurations protected by the + * page_table_lock, in other configurations by being atomic. + */ + struct mm_rss_stat rss_stat; + + struct linux_binfmt *binfmt; + + cpumask_var_t cpu_vm_mask_var; + + /* Architecture-specific MM context */ + mm_context_t context; + + unsigned long flags; /* Must use atomic bitops to access the bits */ + + struct core_state *core_state; /* coredumping support */ +#ifdef CONFIG_AIO + spinlock_t ioctx_lock; + struct hlist_head ioctx_list; +#endif +#ifdef CONFIG_MM_OWNER + /* + * "owner" points to a task that is regarded as the canonical + * user/owner of this mm. All of the following must be true in + * order for it to be changed: + * + * current == mm->owner + * current->mm != mm + * new_owner->mm == mm + * new_owner->alloc_lock is held + */ + struct task_struct __rcu *owner; +#endif + + /* store ref to file /proc//exe symlink points to */ + struct file *exe_file; +#ifdef CONFIG_MMU_NOTIFIER + struct mmu_notifier_mm *mmu_notifier_mm; +#endif +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + pgtable_t pmd_huge_pte; /* protected by page_table_lock */ +#endif +#ifdef CONFIG_CPUMASK_OFFSTACK + struct cpumask cpumask_allocation; +#endif +#ifdef CONFIG_NUMA_BALANCING + /* + * numa_next_scan is the next time that the PTEs will be marked + * pte_numa. NUMA hinting faults will gather statistics and migrate + * pages to new nodes if necessary. + */ + unsigned long numa_next_scan; + + /* numa_next_reset is when the PTE scanner period will be reset */ + unsigned long numa_next_reset; + + /* Restart point for scanning and setting pte_numa */ + unsigned long numa_scan_offset; + + /* numa_scan_seq prevents two threads setting pte_numa */ + int numa_scan_seq; + + /* + * The first node a task was scheduled on. If a task runs on + * a different node than Make PTE Scan Go Now. + */ + int first_nid; +#endif +#if defined(CONFIG_NUMA_BALANCING) || defined(CONFIG_COMPACTION) + /* + * An operation with batched TLB flushing is going on. Anything that + * can move process memory needs to flush the TLB when moving a + * PROT_NONE or PROT_NUMA mapped page. + */ + bool tlb_flush_pending; +#endif + struct uprobes_state uprobes_state; +}; + +/* first nid will either be a valid NID or one of these values */ +#define NUMA_PTE_SCAN_INIT -1 +#define NUMA_PTE_SCAN_ACTIVE -2 + +static inline void mm_init_cpumask(struct mm_struct *mm) +{ +#ifdef CONFIG_CPUMASK_OFFSTACK + mm->cpu_vm_mask_var = &mm->cpumask_allocation; +#endif +} + +/* Future-safe accessor for struct mm_struct's cpu_vm_mask. */ +static inline cpumask_t *mm_cpumask(struct mm_struct *mm) +{ + return mm->cpu_vm_mask_var; +} + + +/* Return the name for an anonymous mapping or NULL for a file-backed mapping */ +static inline const char __user *vma_get_anon_name(struct vm_area_struct *vma) +{ + if (vma->vm_file) + return NULL; + + return vma->shared.anon_name; +} + +#if defined(CONFIG_NUMA_BALANCING) || defined(CONFIG_COMPACTION) +/* + * Memory barriers to keep this state in sync are graciously provided by + * the page table locks, outside of which no page table modifications happen. + * The barriers below prevent the compiler from re-ordering the instructions + * around the memory barriers that are already present in the code. + */ +static inline bool mm_tlb_flush_pending(struct mm_struct *mm) +{ + barrier(); + return mm->tlb_flush_pending; +} +static inline void set_tlb_flush_pending(struct mm_struct *mm) +{ + mm->tlb_flush_pending = true; + + /* + * Guarantee that the tlb_flush_pending store does not leak into the + * critical section updating the page tables + */ + smp_mb__before_spinlock(); +} +/* Clearing is done after a TLB flush, which also provides a barrier. */ +static inline void clear_tlb_flush_pending(struct mm_struct *mm) +{ + barrier(); + mm->tlb_flush_pending = false; +} +#else +static inline bool mm_tlb_flush_pending(struct mm_struct *mm) +{ + return false; +} +static inline void set_tlb_flush_pending(struct mm_struct *mm) +{ +} +static inline void clear_tlb_flush_pending(struct mm_struct *mm) +{ +} +#endif + +#endif /* _LINUX_MM_TYPES_H */ diff -Naur kernel.21.3.orig/include/linux/splice.h kernel.21.3.new/include/linux/splice.h --- kernel.21.3.orig/include/linux/splice.h 2015-02-03 01:44:32.000000000 +0000 +++ kernel.21.3.new/include/linux/splice.h 2015-04-25 23:28:26.656666289 +0000 @@ -92,4 +92,10 @@ extern void spd_release_page(struct splice_pipe_desc *, unsigned int); extern const struct pipe_buf_operations page_cache_pipe_buf_ops; + +extern long do_splice_from(struct pipe_inode_info *pipe, struct file *out, + loff_t *ppos, size_t len, unsigned int flags); +extern long do_splice_to(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, size_t len, + unsigned int flags); #endif diff -Naur kernel.21.3.orig/include/uapi/linux/aufs_type.h kernel.21.3.new/include/uapi/linux/aufs_type.h --- kernel.21.3.orig/include/uapi/linux/aufs_type.h 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/include/uapi/linux/aufs_type.h 2015-04-25 22:45:35.000000000 +0000 @@ -0,0 +1,418 @@ +/* + * Copyright (C) 2005-2014 Junjiro R. Okajima + * + * This program, aufs is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +#ifndef __AUFS_TYPE_H__ +#define __AUFS_TYPE_H__ + +#define AUFS_NAME "aufs" + +#ifdef __KERNEL__ +/* + * define it before including all other headers. + * sched.h may use pr_* macros before defining "current", so define the + * no-current version first, and re-define later. + */ +#define pr_fmt(fmt) AUFS_NAME " %s:%d: " fmt, __func__, __LINE__ +#include +#undef pr_fmt +#define pr_fmt(fmt) \ + AUFS_NAME " %s:%d:%.*s[%d]: " fmt, __func__, __LINE__, \ + (int)sizeof(current->comm), current->comm, current->pid +#else +#include +#include +#endif /* __KERNEL__ */ + +#include + +#define AUFS_VERSION "3.10.x-20141215" + +/* todo? move this to linux-2.6.19/include/magic.h */ +#define AUFS_SUPER_MAGIC ('a' << 24 | 'u' << 16 | 'f' << 8 | 's') + +/* ---------------------------------------------------------------------- */ + +#ifdef CONFIG_AUFS_BRANCH_MAX_127 +typedef int8_t aufs_bindex_t; +#define AUFS_BRANCH_MAX 127 +#else +typedef int16_t aufs_bindex_t; +#ifdef CONFIG_AUFS_BRANCH_MAX_511 +#define AUFS_BRANCH_MAX 511 +#elif defined(CONFIG_AUFS_BRANCH_MAX_1023) +#define AUFS_BRANCH_MAX 1023 +#elif defined(CONFIG_AUFS_BRANCH_MAX_32767) +#define AUFS_BRANCH_MAX 32767 +#endif +#endif + +#ifdef __KERNEL__ +#ifndef AUFS_BRANCH_MAX +#error unknown CONFIG_AUFS_BRANCH_MAX value +#endif +#endif /* __KERNEL__ */ + +/* ---------------------------------------------------------------------- */ + +#define AUFS_FSTYPE AUFS_NAME + +#define AUFS_ROOT_INO 2 +#define AUFS_FIRST_INO 11 + +#define AUFS_WH_PFX ".wh." +#define AUFS_WH_PFX_LEN ((int)sizeof(AUFS_WH_PFX) - 1) +#define AUFS_WH_TMP_LEN 4 +/* a limit for rmdir/rename a dir and copyup */ +#define AUFS_MAX_NAMELEN (NAME_MAX \ + - AUFS_WH_PFX_LEN * 2 /* doubly whiteouted */\ + - 1 /* dot */\ + - AUFS_WH_TMP_LEN) /* hex */ +#define AUFS_XINO_FNAME "." AUFS_NAME ".xino" +#define AUFS_XINO_DEFPATH "/tmp/" AUFS_XINO_FNAME +#define AUFS_XINO_DEF_SEC 30 /* seconds */ +#define AUFS_XINO_DEF_TRUNC 45 /* percentage */ +#define AUFS_DIRWH_DEF 3 +#define AUFS_RDCACHE_DEF 10 /* seconds */ +#define AUFS_RDCACHE_MAX 3600 /* seconds */ +#define AUFS_RDBLK_DEF 512 /* bytes */ +#define AUFS_RDHASH_DEF 32 +#define AUFS_WKQ_NAME AUFS_NAME "d" +#define AUFS_MFS_DEF_SEC 30 /* seconds */ +#define AUFS_MFS_MAX_SEC 3600 /* seconds */ +#define AUFS_FHSM_CACHE_DEF_SEC 30 /* seconds */ +#define AUFS_PLINK_WARN 50 /* number of plinks in a single bucket */ + +/* pseudo-link maintenace under /proc */ +#define AUFS_PLINK_MAINT_NAME "plink_maint" +#define AUFS_PLINK_MAINT_DIR "fs/" AUFS_NAME +#define AUFS_PLINK_MAINT_PATH AUFS_PLINK_MAINT_DIR "/" AUFS_PLINK_MAINT_NAME + +#define AUFS_DIROPQ_NAME AUFS_WH_PFX ".opq" /* whiteouted doubly */ +#define AUFS_WH_DIROPQ AUFS_WH_PFX AUFS_DIROPQ_NAME + +#define AUFS_BASE_NAME AUFS_WH_PFX AUFS_NAME +#define AUFS_PLINKDIR_NAME AUFS_WH_PFX "plnk" +#define AUFS_ORPHDIR_NAME AUFS_WH_PFX "orph" + +/* doubly whiteouted */ +#define AUFS_WH_BASE AUFS_WH_PFX AUFS_BASE_NAME +#define AUFS_WH_PLINKDIR AUFS_WH_PFX AUFS_PLINKDIR_NAME +#define AUFS_WH_ORPHDIR AUFS_WH_PFX AUFS_ORPHDIR_NAME + +/* branch permissions and attributes */ +#define AUFS_BRPERM_RW "rw" +#define AUFS_BRPERM_RO "ro" +#define AUFS_BRPERM_RR "rr" +#define AUFS_BRATTR_COO_REG "coo_reg" +#define AUFS_BRATTR_COO_ALL "coo_all" +#define AUFS_BRATTR_FHSM "fhsm" +#define AUFS_BRATTR_UNPIN "unpin" +#define AUFS_BRATTR_ICEX "icex" +#define AUFS_BRATTR_ICEX_SEC "icexsec" +#define AUFS_BRATTR_ICEX_SYS "icexsys" +#define AUFS_BRATTR_ICEX_TR "icextr" +#define AUFS_BRATTR_ICEX_USR "icexusr" +#define AUFS_BRATTR_ICEX_OTH "icexoth" +#define AUFS_BRRATTR_WH "wh" +#define AUFS_BRWATTR_NLWH "nolwh" +#define AUFS_BRWATTR_MOO "moo" + +#define AuBrPerm_RW 1 /* writable, hardlinkable wh */ +#define AuBrPerm_RO (1 << 1) /* readonly */ +#define AuBrPerm_RR (1 << 2) /* natively readonly */ +#define AuBrPerm_Mask (AuBrPerm_RW | AuBrPerm_RO | AuBrPerm_RR) + +#define AuBrAttr_COO_REG (1 << 3) /* copy-up on open */ +#define AuBrAttr_COO_ALL (1 << 4) +#define AuBrAttr_COO_Mask (AuBrAttr_COO_REG | AuBrAttr_COO_ALL) + +#define AuBrAttr_FHSM (1 << 5) /* file-based hsm */ +#define AuBrAttr_UNPIN (1 << 6) /* rename-able top dir of + branch */ + +/* ignore error in copying XATTR */ +#define AuBrAttr_ICEX_SEC (1 << 7) +#define AuBrAttr_ICEX_SYS (1 << 8) +#define AuBrAttr_ICEX_TR (1 << 9) +#define AuBrAttr_ICEX_USR (1 << 10) +#define AuBrAttr_ICEX_OTH (1 << 11) +#define AuBrAttr_ICEX (AuBrAttr_ICEX_SEC \ + | AuBrAttr_ICEX_SYS \ + | AuBrAttr_ICEX_TR \ + | AuBrAttr_ICEX_USR \ + | AuBrAttr_ICEX_OTH) + +#define AuBrRAttr_WH (1 << 12) /* whiteout-able */ +#define AuBrRAttr_Mask AuBrRAttr_WH + +#define AuBrWAttr_NoLinkWH (1 << 13) /* un-hardlinkable whiteouts */ +#define AuBrWAttr_MOO (1 << 14) /* move-up on open */ +#define AuBrWAttr_Mask (AuBrWAttr_NoLinkWH | AuBrWAttr_MOO) + +#define AuBrAttr_CMOO_Mask (AuBrAttr_COO_Mask | AuBrWAttr_MOO) + +/* #warning test userspace */ +#ifdef __KERNEL__ +#ifndef CONFIG_AUFS_FHSM +#undef AuBrAttr_FHSM +#define AuBrAttr_FHSM 0 +#endif +#ifndef CONFIG_AUFS_XATTR +#undef AuBrAttr_ICEX +#define AuBrAttr_ICEX 0 +#undef AuBrAttr_ICEX_SEC +#define AuBrAttr_ICEX_SEC 0 +#undef AuBrAttr_ICEX_SYS +#define AuBrAttr_ICEX_SYS 0 +#undef AuBrAttr_ICEX_TR +#define AuBrAttr_ICEX_TR 0 +#undef AuBrAttr_ICEX_USR +#define AuBrAttr_ICEX_USR 0 +#undef AuBrAttr_ICEX_OTH +#define AuBrAttr_ICEX_OTH 0 +#endif +#endif + +/* the longest combination */ +/* AUFS_BRATTR_ICEX and AUFS_BRATTR_ICEX_TR don't affect here */ +#define AuBrPermStrSz sizeof(AUFS_BRPERM_RW \ + "+" AUFS_BRATTR_COO_REG \ + "+" AUFS_BRATTR_FHSM \ + "+" AUFS_BRATTR_UNPIN \ + "+" AUFS_BRATTR_ICEX_SEC \ + "+" AUFS_BRATTR_ICEX_SYS \ + "+" AUFS_BRATTR_ICEX_USR \ + "+" AUFS_BRATTR_ICEX_OTH \ + "+" AUFS_BRWATTR_NLWH) + +typedef struct { + char a[AuBrPermStrSz]; +} au_br_perm_str_t; + +static inline int au_br_writable(int brperm) +{ + return brperm & AuBrPerm_RW; +} + +static inline int au_br_whable(int brperm) +{ + return brperm & (AuBrPerm_RW | AuBrRAttr_WH); +} + +static inline int au_br_wh_linkable(int brperm) +{ + return !(brperm & AuBrWAttr_NoLinkWH); +} + +static inline int au_br_cmoo(int brperm) +{ + return brperm & AuBrAttr_CMOO_Mask; +} + +static inline int au_br_fhsm(int brperm) +{ + return brperm & AuBrAttr_FHSM; +} + +/* ---------------------------------------------------------------------- */ + +/* ioctl */ +enum { + /* readdir in userspace */ + AuCtl_RDU, + AuCtl_RDU_INO, + + AuCtl_WBR_FD, /* pathconf wrapper */ + AuCtl_IBUSY, /* busy inode */ + AuCtl_MVDOWN, /* move-down */ + AuCtl_BR, /* info about branches */ + AuCtl_FHSM_FD /* connection for fhsm */ +}; + +/* borrowed from linux/include/linux/kernel.h */ +#ifndef ALIGN +#define ALIGN(x, a) __ALIGN_MASK(x, (typeof(x))(a)-1) +#define __ALIGN_MASK(x, mask) (((x)+(mask))&~(mask)) +#endif + +/* borrowed from linux/include/linux/compiler-gcc3.h */ +#ifndef __aligned +#define __aligned(x) __attribute__((aligned(x))) +#endif + +#ifdef __KERNEL__ +#ifndef __packed +#define __packed __attribute__((packed)) +#endif +#endif + +struct au_rdu_cookie { + uint64_t h_pos; + int16_t bindex; + uint8_t flags; + uint8_t pad; + uint32_t generation; +} __aligned(8); + +struct au_rdu_ent { + uint64_t ino; + int16_t bindex; + uint8_t type; + uint8_t nlen; + uint8_t wh; + char name[0]; +} __aligned(8); + +static inline int au_rdu_len(int nlen) +{ + /* include the terminating NULL */ + return ALIGN(sizeof(struct au_rdu_ent) + nlen + 1, + sizeof(uint64_t)); +} + +union au_rdu_ent_ul { + struct au_rdu_ent __user *e; + uint64_t ul; +}; + +enum { + AufsCtlRduV_SZ, + AufsCtlRduV_End +}; + +struct aufs_rdu { + /* input */ + union { + uint64_t sz; /* AuCtl_RDU */ + uint64_t nent; /* AuCtl_RDU_INO */ + }; + union au_rdu_ent_ul ent; + uint16_t verify[AufsCtlRduV_End]; + + /* input/output */ + uint32_t blk; + + /* output */ + union au_rdu_ent_ul tail; + /* number of entries which were added in a single call */ + uint64_t rent; + uint8_t full; + uint8_t shwh; + + struct au_rdu_cookie cookie; +} __aligned(8); + +/* ---------------------------------------------------------------------- */ + +struct aufs_wbr_fd { + uint32_t oflags; + int16_t brid; +} __aligned(8); + +/* ---------------------------------------------------------------------- */ + +struct aufs_ibusy { + uint64_t ino, h_ino; + int16_t bindex; +} __aligned(8); + +/* ---------------------------------------------------------------------- */ + +/* error code for move-down */ +/* the actual message strings are implemented in aufs-util.git */ +enum { + EAU_MVDOWN_OPAQUE = 1, + EAU_MVDOWN_WHITEOUT, + EAU_MVDOWN_UPPER, + EAU_MVDOWN_BOTTOM, + EAU_MVDOWN_NOUPPER, + EAU_MVDOWN_NOLOWERBR, + EAU_Last +}; + +/* flags for move-down */ +#define AUFS_MVDOWN_DMSG 1 +#define AUFS_MVDOWN_OWLOWER (1 << 1) /* overwrite lower */ +#define AUFS_MVDOWN_KUPPER (1 << 2) /* keep upper */ +#define AUFS_MVDOWN_ROLOWER (1 << 3) /* do even if lower is RO */ +#define AUFS_MVDOWN_ROLOWER_R (1 << 4) /* did on lower RO */ +#define AUFS_MVDOWN_ROUPPER (1 << 5) /* do even if upper is RO */ +#define AUFS_MVDOWN_ROUPPER_R (1 << 6) /* did on upper RO */ +#define AUFS_MVDOWN_BRID_UPPER (1 << 7) /* upper brid */ +#define AUFS_MVDOWN_BRID_LOWER (1 << 8) /* lower brid */ +#define AUFS_MVDOWN_FHSM_LOWER (1 << 9) /* find fhsm attr for lower */ +#define AUFS_MVDOWN_STFS (1 << 10) /* req. stfs */ +#define AUFS_MVDOWN_STFS_FAILED (1 << 11) /* output: stfs is unusable */ +#define AUFS_MVDOWN_BOTTOM (1 << 12) /* output: no more lowers */ + +/* index for move-down */ +enum { + AUFS_MVDOWN_UPPER, + AUFS_MVDOWN_LOWER, + AUFS_MVDOWN_NARRAY +}; + +/* + * additional info of move-down + * number of free blocks and inodes. + * subset of struct kstatfs, but smaller and always 64bit. + */ +struct aufs_stfs { + uint64_t f_blocks; + uint64_t f_bavail; + uint64_t f_files; + uint64_t f_ffree; +}; + +struct aufs_stbr { + int16_t brid; /* optional input */ + int16_t bindex; /* output */ + struct aufs_stfs stfs; /* output when AUFS_MVDOWN_STFS set */ +} __aligned(8); + +struct aufs_mvdown { + uint32_t flags; /* input/output */ + struct aufs_stbr stbr[AUFS_MVDOWN_NARRAY]; /* input/output */ + int8_t au_errno; /* output */ +} __aligned(8); + +/* ---------------------------------------------------------------------- */ + +union aufs_brinfo { + /* PATH_MAX may differ between kernel-space and user-space */ + char _spacer[4096]; + struct { + int16_t id; + int perm; + char path[0]; + }; +} __aligned(8); + +/* ---------------------------------------------------------------------- */ + +#define AuCtlType 'A' +#define AUFS_CTL_RDU _IOWR(AuCtlType, AuCtl_RDU, struct aufs_rdu) +#define AUFS_CTL_RDU_INO _IOWR(AuCtlType, AuCtl_RDU_INO, struct aufs_rdu) +#define AUFS_CTL_WBR_FD _IOW(AuCtlType, AuCtl_WBR_FD, \ + struct aufs_wbr_fd) +#define AUFS_CTL_IBUSY _IOWR(AuCtlType, AuCtl_IBUSY, struct aufs_ibusy) +#define AUFS_CTL_MVDOWN _IOWR(AuCtlType, AuCtl_MVDOWN, \ + struct aufs_mvdown) +#define AUFS_CTL_BRINFO _IOW(AuCtlType, AuCtl_BR, union aufs_brinfo) +#define AUFS_CTL_FHSM_FD _IOW(AuCtlType, AuCtl_FHSM_FD, int) + +#endif /* __AUFS_TYPE_H__ */ diff -Naur kernel.21.3.orig/include/uapi/linux/Kbuild kernel.21.3.new/include/uapi/linux/Kbuild --- kernel.21.3.orig/include/uapi/linux/Kbuild 2015-02-03 01:44:33.000000000 +0000 +++ kernel.21.3.new/include/uapi/linux/Kbuild 2015-04-25 23:28:39.215542246 +0000 @@ -56,6 +56,7 @@ header-y += atmsap.h header-y += atmsvc.h header-y += audit.h +header-y += aufs_type.h header-y += auto_fs.h header-y += auto_fs4.h header-y += auxvec.h diff -Naur kernel.21.3.orig/kernel/fork.c kernel.21.3.new/kernel/fork.c --- kernel.21.3.orig/kernel/fork.c 2015-02-03 01:44:33.000000000 +0000 +++ kernel.21.3.new/kernel/fork.c 2015-04-25 23:28:45.330481423 +0000 @@ -434,7 +434,7 @@ struct inode *inode = file_inode(file); struct address_space *mapping = file->f_mapping; - get_file(file); + vma_get_file(tmp); if (tmp->vm_flags & VM_DENYWRITE) atomic_dec(&inode->i_writecount); mutex_lock(&mapping->i_mmap_mutex); diff -Naur kernel.21.3.orig/kernel/fork.c.orig kernel.21.3.new/kernel/fork.c.orig --- kernel.21.3.orig/kernel/fork.c.orig 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/kernel/fork.c.orig 2015-02-03 01:44:33.000000000 +0000 @@ -0,0 +1,1971 @@ +/* + * linux/kernel/fork.c + * + * Copyright (C) 1991, 1992 Linus Torvalds + */ + +/* + * 'fork.c' contains the help-routines for the 'fork' system call + * (see also entry.S and others). + * Fork is rather simple, once you get the hang of it, but the memory + * management can be a bitch. See 'mm/memory.c': 'copy_page_range()' + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +#include +#include + +#define CREATE_TRACE_POINTS +#include + +/* + * Protected counters by write_lock_irq(&tasklist_lock) + */ +unsigned long total_forks; /* Handle normal Linux uptimes. */ +int nr_threads; /* The idle threads do not count.. */ + +int max_threads; /* tunable limit on nr_threads */ + +DEFINE_PER_CPU(unsigned long, process_counts) = 0; + +__cacheline_aligned DEFINE_RWLOCK(tasklist_lock); /* outer */ + +#ifdef CONFIG_PROVE_RCU +int lockdep_tasklist_lock_is_held(void) +{ + return lockdep_is_held(&tasklist_lock); +} +EXPORT_SYMBOL_GPL(lockdep_tasklist_lock_is_held); +#endif /* #ifdef CONFIG_PROVE_RCU */ + +int nr_processes(void) +{ + int cpu; + int total = 0; + + for_each_possible_cpu(cpu) + total += per_cpu(process_counts, cpu); + + return total; +} + +void __weak arch_release_task_struct(struct task_struct *tsk) +{ +} + +#ifndef CONFIG_ARCH_TASK_STRUCT_ALLOCATOR +static struct kmem_cache *task_struct_cachep; + +static inline struct task_struct *alloc_task_struct_node(int node) +{ + return kmem_cache_alloc_node(task_struct_cachep, GFP_KERNEL, node); +} + +static inline void free_task_struct(struct task_struct *tsk) +{ + kmem_cache_free(task_struct_cachep, tsk); +} +#endif + +void __weak arch_release_thread_info(struct thread_info *ti) +{ +} + +#ifndef CONFIG_ARCH_THREAD_INFO_ALLOCATOR + +/* + * Allocate pages if THREAD_SIZE is >= PAGE_SIZE, otherwise use a + * kmemcache based allocator. + */ +# if THREAD_SIZE >= PAGE_SIZE +static struct thread_info *alloc_thread_info_node(struct task_struct *tsk, + int node) +{ + struct page *page = alloc_pages_node(node, THREADINFO_GFP_ACCOUNTED, + THREAD_SIZE_ORDER); + + return page ? page_address(page) : NULL; +} + +static inline void free_thread_info(struct thread_info *ti) +{ + free_memcg_kmem_pages((unsigned long)ti, THREAD_SIZE_ORDER); +} +# else +static struct kmem_cache *thread_info_cache; + +static struct thread_info *alloc_thread_info_node(struct task_struct *tsk, + int node) +{ + return kmem_cache_alloc_node(thread_info_cache, THREADINFO_GFP, node); +} + +static void free_thread_info(struct thread_info *ti) +{ + kmem_cache_free(thread_info_cache, ti); +} + +void thread_info_cache_init(void) +{ + thread_info_cache = kmem_cache_create("thread_info", THREAD_SIZE, + THREAD_SIZE, 0, NULL); + BUG_ON(thread_info_cache == NULL); +} +# endif +#endif + +/* SLAB cache for signal_struct structures (tsk->signal) */ +static struct kmem_cache *signal_cachep; + +/* SLAB cache for sighand_struct structures (tsk->sighand) */ +struct kmem_cache *sighand_cachep; + +/* SLAB cache for files_struct structures (tsk->files) */ +struct kmem_cache *files_cachep; + +/* SLAB cache for fs_struct structures (tsk->fs) */ +struct kmem_cache *fs_cachep; + +/* SLAB cache for vm_area_struct structures */ +struct kmem_cache *vm_area_cachep; + +/* SLAB cache for mm_struct structures (tsk->mm) */ +static struct kmem_cache *mm_cachep; + +/* Notifier list called when a task struct is freed */ +static ATOMIC_NOTIFIER_HEAD(task_free_notifier); + +static void account_kernel_stack(struct thread_info *ti, int account) +{ + struct zone *zone = page_zone(virt_to_page(ti)); + + mod_zone_page_state(zone, NR_KERNEL_STACK, account); +} + +void free_task(struct task_struct *tsk) +{ + account_kernel_stack(tsk->stack, -1); + arch_release_thread_info(tsk->stack); + free_thread_info(tsk->stack); + rt_mutex_debug_task_free(tsk); + ftrace_graph_exit_task(tsk); + put_seccomp_filter(tsk); + arch_release_task_struct(tsk); + free_task_struct(tsk); +} +EXPORT_SYMBOL(free_task); + +static inline void free_signal_struct(struct signal_struct *sig) +{ + taskstats_tgid_free(sig); + sched_autogroup_exit(sig); + kmem_cache_free(signal_cachep, sig); +} + +static inline void put_signal_struct(struct signal_struct *sig) +{ + if (atomic_dec_and_test(&sig->sigcnt)) + free_signal_struct(sig); +} + +int task_free_register(struct notifier_block *n) +{ + return atomic_notifier_chain_register(&task_free_notifier, n); +} +EXPORT_SYMBOL(task_free_register); + +int task_free_unregister(struct notifier_block *n) +{ + return atomic_notifier_chain_unregister(&task_free_notifier, n); +} +EXPORT_SYMBOL(task_free_unregister); + +void __put_task_struct(struct task_struct *tsk) +{ + WARN_ON(!tsk->exit_state); + WARN_ON(atomic_read(&tsk->usage)); + WARN_ON(tsk == current); + + security_task_free(tsk); + exit_creds(tsk); + delayacct_tsk_free(tsk); + put_signal_struct(tsk->signal); + + atomic_notifier_call_chain(&task_free_notifier, 0, tsk); + if (!profile_handoff_task(tsk)) + free_task(tsk); +} +EXPORT_SYMBOL_GPL(__put_task_struct); + +void __init __weak arch_task_cache_init(void) { } + +void __init fork_init(unsigned long mempages) +{ +#ifndef CONFIG_ARCH_TASK_STRUCT_ALLOCATOR +#ifndef ARCH_MIN_TASKALIGN +#define ARCH_MIN_TASKALIGN L1_CACHE_BYTES +#endif + /* create a slab on which task_structs can be allocated */ + task_struct_cachep = + kmem_cache_create("task_struct", sizeof(struct task_struct), + ARCH_MIN_TASKALIGN, SLAB_PANIC | SLAB_NOTRACK, NULL); +#endif + + /* do the arch specific task caches init */ + arch_task_cache_init(); + + /* + * The default maximum number of threads is set to a safe + * value: the thread structures can take up at most half + * of memory. + */ + max_threads = mempages / (8 * THREAD_SIZE / PAGE_SIZE); + + /* + * we need to allow at least 20 threads to boot a system + */ + if (max_threads < 20) + max_threads = 20; + + init_task.signal->rlim[RLIMIT_NPROC].rlim_cur = max_threads/2; + init_task.signal->rlim[RLIMIT_NPROC].rlim_max = max_threads/2; + init_task.signal->rlim[RLIMIT_SIGPENDING] = + init_task.signal->rlim[RLIMIT_NPROC]; +} + +int __attribute__((weak)) arch_dup_task_struct(struct task_struct *dst, + struct task_struct *src) +{ + *dst = *src; + return 0; +} + +static struct task_struct *dup_task_struct(struct task_struct *orig) +{ + struct task_struct *tsk; + struct thread_info *ti; + unsigned long *stackend; + int node = tsk_fork_get_node(orig); + int err; + + tsk = alloc_task_struct_node(node); + if (!tsk) + return NULL; + + ti = alloc_thread_info_node(tsk, node); + if (!ti) + goto free_tsk; + + err = arch_dup_task_struct(tsk, orig); + if (err) + goto free_ti; + + tsk->stack = ti; + + setup_thread_stack(tsk, orig); + clear_user_return_notifier(tsk); + clear_tsk_need_resched(tsk); + stackend = end_of_stack(tsk); + *stackend = STACK_END_MAGIC; /* for overflow detection */ + +#ifdef CONFIG_CC_STACKPROTECTOR + tsk->stack_canary = get_random_int(); +#endif + + /* + * One for us, one for whoever does the "release_task()" (usually + * parent) + */ + atomic_set(&tsk->usage, 2); +#ifdef CONFIG_BLK_DEV_IO_TRACE + tsk->btrace_seq = 0; +#endif + tsk->splice_pipe = NULL; + tsk->task_frag.page = NULL; + + account_kernel_stack(ti, 1); + + return tsk; + +free_ti: + free_thread_info(ti); +free_tsk: + free_task_struct(tsk); + return NULL; +} + +#ifdef CONFIG_MMU +static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) +{ + struct vm_area_struct *mpnt, *tmp, *prev, **pprev; + struct rb_node **rb_link, *rb_parent; + int retval; + unsigned long charge; + struct mempolicy *pol; + + uprobe_start_dup_mmap(); + down_write(&oldmm->mmap_sem); + flush_cache_dup_mm(oldmm); + uprobe_dup_mmap(oldmm, mm); + /* + * Not linked in yet - no deadlock potential: + */ + down_write_nested(&mm->mmap_sem, SINGLE_DEPTH_NESTING); + + mm->locked_vm = 0; + mm->mmap = NULL; + mm->mmap_cache = NULL; + mm->free_area_cache = oldmm->mmap_base; + mm->cached_hole_size = ~0UL; + mm->map_count = 0; + cpumask_clear(mm_cpumask(mm)); + mm->mm_rb = RB_ROOT; + rb_link = &mm->mm_rb.rb_node; + rb_parent = NULL; + pprev = &mm->mmap; + retval = ksm_fork(mm, oldmm); + if (retval) + goto out; + retval = khugepaged_fork(mm, oldmm); + if (retval) + goto out; + + prev = NULL; + for (mpnt = oldmm->mmap; mpnt; mpnt = mpnt->vm_next) { + struct file *file; + + if (mpnt->vm_flags & VM_DONTCOPY) { + vm_stat_account(mm, mpnt->vm_flags, mpnt->vm_file, + -vma_pages(mpnt)); + continue; + } + charge = 0; + if (mpnt->vm_flags & VM_ACCOUNT) { + unsigned long len = vma_pages(mpnt); + + if (security_vm_enough_memory_mm(oldmm, len)) /* sic */ + goto fail_nomem; + charge = len; + } + tmp = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); + if (!tmp) + goto fail_nomem; + *tmp = *mpnt; + INIT_LIST_HEAD(&tmp->anon_vma_chain); + pol = mpol_dup(vma_policy(mpnt)); + retval = PTR_ERR(pol); + if (IS_ERR(pol)) + goto fail_nomem_policy; + vma_set_policy(tmp, pol); + tmp->vm_mm = mm; + if (anon_vma_fork(tmp, mpnt)) + goto fail_nomem_anon_vma_fork; + tmp->vm_flags &= ~VM_LOCKED; + tmp->vm_next = tmp->vm_prev = NULL; + file = tmp->vm_file; + if (file) { + struct inode *inode = file_inode(file); + struct address_space *mapping = file->f_mapping; + + get_file(file); + if (tmp->vm_flags & VM_DENYWRITE) + atomic_dec(&inode->i_writecount); + mutex_lock(&mapping->i_mmap_mutex); + if (tmp->vm_flags & VM_SHARED) + mapping->i_mmap_writable++; + flush_dcache_mmap_lock(mapping); + /* insert tmp into the share list, just after mpnt */ + if (unlikely(tmp->vm_flags & VM_NONLINEAR)) + vma_nonlinear_insert(tmp, + &mapping->i_mmap_nonlinear); + else + vma_interval_tree_insert_after(tmp, mpnt, + &mapping->i_mmap); + flush_dcache_mmap_unlock(mapping); + mutex_unlock(&mapping->i_mmap_mutex); + } + + /* + * Clear hugetlb-related page reserves for children. This only + * affects MAP_PRIVATE mappings. Faults generated by the child + * are not guaranteed to succeed, even if read-only + */ + if (is_vm_hugetlb_page(tmp)) + reset_vma_resv_huge_pages(tmp); + + /* + * Link in the new vma and copy the page table entries. + */ + *pprev = tmp; + pprev = &tmp->vm_next; + tmp->vm_prev = prev; + prev = tmp; + + __vma_link_rb(mm, tmp, rb_link, rb_parent); + rb_link = &tmp->vm_rb.rb_right; + rb_parent = &tmp->vm_rb; + + mm->map_count++; + retval = copy_page_range(mm, oldmm, mpnt); + + if (tmp->vm_ops && tmp->vm_ops->open) + tmp->vm_ops->open(tmp); + + if (retval) + goto out; + } + /* a new mm has just been created */ + arch_dup_mmap(oldmm, mm); + retval = 0; +out: + up_write(&mm->mmap_sem); + flush_tlb_mm(oldmm); + up_write(&oldmm->mmap_sem); + uprobe_end_dup_mmap(); + return retval; +fail_nomem_anon_vma_fork: + mpol_put(pol); +fail_nomem_policy: + kmem_cache_free(vm_area_cachep, tmp); +fail_nomem: + retval = -ENOMEM; + vm_unacct_memory(charge); + goto out; +} + +static inline int mm_alloc_pgd(struct mm_struct *mm) +{ + mm->pgd = pgd_alloc(mm); + if (unlikely(!mm->pgd)) + return -ENOMEM; + return 0; +} + +static inline void mm_free_pgd(struct mm_struct *mm) +{ + pgd_free(mm, mm->pgd); +} +#else +#define dup_mmap(mm, oldmm) (0) +#define mm_alloc_pgd(mm) (0) +#define mm_free_pgd(mm) +#endif /* CONFIG_MMU */ + +__cacheline_aligned_in_smp DEFINE_SPINLOCK(mmlist_lock); + +#define allocate_mm() (kmem_cache_alloc(mm_cachep, GFP_KERNEL)) +#define free_mm(mm) (kmem_cache_free(mm_cachep, (mm))) + +static unsigned long default_dump_filter = MMF_DUMP_FILTER_DEFAULT; + +static int __init coredump_filter_setup(char *s) +{ + default_dump_filter = + (simple_strtoul(s, NULL, 0) << MMF_DUMP_FILTER_SHIFT) & + MMF_DUMP_FILTER_MASK; + return 1; +} + +__setup("coredump_filter=", coredump_filter_setup); + +#include + +static void mm_init_aio(struct mm_struct *mm) +{ +#ifdef CONFIG_AIO + spin_lock_init(&mm->ioctx_lock); + INIT_HLIST_HEAD(&mm->ioctx_list); +#endif +} + +static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p) +{ + atomic_set(&mm->mm_users, 1); + atomic_set(&mm->mm_count, 1); + init_rwsem(&mm->mmap_sem); + INIT_LIST_HEAD(&mm->mmlist); + mm->flags = (current->mm) ? + (current->mm->flags & MMF_INIT_MASK) : default_dump_filter; + mm->core_state = NULL; + mm->nr_ptes = 0; + memset(&mm->rss_stat, 0, sizeof(mm->rss_stat)); + spin_lock_init(&mm->page_table_lock); + mm->free_area_cache = TASK_UNMAPPED_BASE; + mm->cached_hole_size = ~0UL; + mm_init_aio(mm); + mm_init_owner(mm, p); + clear_tlb_flush_pending(mm); + + if (likely(!mm_alloc_pgd(mm))) { + mm->def_flags = 0; + mmu_notifier_mm_init(mm); + return mm; + } + + free_mm(mm); + return NULL; +} + +static void check_mm(struct mm_struct *mm) +{ + int i; + + for (i = 0; i < NR_MM_COUNTERS; i++) { + long x = atomic_long_read(&mm->rss_stat.count[i]); + + if (unlikely(x)) + printk(KERN_ALERT "BUG: Bad rss-counter state " + "mm:%p idx:%d val:%ld\n", mm, i, x); + } + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + VM_BUG_ON(mm->pmd_huge_pte); +#endif +} + +/* + * Allocate and initialize an mm_struct. + */ +struct mm_struct *mm_alloc(void) +{ + struct mm_struct *mm; + + mm = allocate_mm(); + if (!mm) + return NULL; + + memset(mm, 0, sizeof(*mm)); + mm_init_cpumask(mm); + return mm_init(mm, current); +} + +/* + * Called when the last reference to the mm + * is dropped: either by a lazy thread or by + * mmput. Free the page directory and the mm. + */ +void __mmdrop(struct mm_struct *mm) +{ + BUG_ON(mm == &init_mm); + mm_free_pgd(mm); + destroy_context(mm); + mmu_notifier_mm_destroy(mm); + check_mm(mm); + free_mm(mm); +} +EXPORT_SYMBOL_GPL(__mmdrop); + +/* + * Decrement the use count and release all resources for an mm. + */ +void mmput(struct mm_struct *mm) +{ + might_sleep(); + + if (atomic_dec_and_test(&mm->mm_users)) { + uprobe_clear_state(mm); + exit_aio(mm); + ksm_exit(mm); + khugepaged_exit(mm); /* must run before exit_mmap */ + exit_mmap(mm); + set_mm_exe_file(mm, NULL); + if (!list_empty(&mm->mmlist)) { + spin_lock(&mmlist_lock); + list_del(&mm->mmlist); + spin_unlock(&mmlist_lock); + } + if (mm->binfmt) + module_put(mm->binfmt->module); + mmdrop(mm); + } +} +EXPORT_SYMBOL_GPL(mmput); + +void set_mm_exe_file(struct mm_struct *mm, struct file *new_exe_file) +{ + if (new_exe_file) + get_file(new_exe_file); + if (mm->exe_file) + fput(mm->exe_file); + mm->exe_file = new_exe_file; +} + +struct file *get_mm_exe_file(struct mm_struct *mm) +{ + struct file *exe_file; + + /* We need mmap_sem to protect against races with removal of exe_file */ + down_read(&mm->mmap_sem); + exe_file = mm->exe_file; + if (exe_file) + get_file(exe_file); + up_read(&mm->mmap_sem); + return exe_file; +} + +static void dup_mm_exe_file(struct mm_struct *oldmm, struct mm_struct *newmm) +{ + /* It's safe to write the exe_file pointer without exe_file_lock because + * this is called during fork when the task is not yet in /proc */ + newmm->exe_file = get_mm_exe_file(oldmm); +} + +/** + * get_task_mm - acquire a reference to the task's mm + * + * Returns %NULL if the task has no mm. Checks PF_KTHREAD (meaning + * this kernel workthread has transiently adopted a user mm with use_mm, + * to do its AIO) is not set and if so returns a reference to it, after + * bumping up the use count. User must release the mm via mmput() + * after use. Typically used by /proc and ptrace. + */ +struct mm_struct *get_task_mm(struct task_struct *task) +{ + struct mm_struct *mm; + + task_lock(task); + mm = task->mm; + if (mm) { + if (task->flags & PF_KTHREAD) + mm = NULL; + else + atomic_inc(&mm->mm_users); + } + task_unlock(task); + return mm; +} +EXPORT_SYMBOL_GPL(get_task_mm); + +struct mm_struct *mm_access(struct task_struct *task, unsigned int mode) +{ + struct mm_struct *mm; + int err; + + err = mutex_lock_killable(&task->signal->cred_guard_mutex); + if (err) + return ERR_PTR(err); + + mm = get_task_mm(task); + if (mm && mm != current->mm && + !ptrace_may_access(task, mode) && + !capable(CAP_SYS_RESOURCE)) { + mmput(mm); + mm = ERR_PTR(-EACCES); + } + mutex_unlock(&task->signal->cred_guard_mutex); + + return mm; +} + +static void complete_vfork_done(struct task_struct *tsk) +{ + struct completion *vfork; + + task_lock(tsk); + vfork = tsk->vfork_done; + if (likely(vfork)) { + tsk->vfork_done = NULL; + complete(vfork); + } + task_unlock(tsk); +} + +static int wait_for_vfork_done(struct task_struct *child, + struct completion *vfork) +{ + int killed; + + freezer_do_not_count(); + killed = wait_for_completion_killable(vfork); + freezer_count(); + + if (killed) { + task_lock(child); + child->vfork_done = NULL; + task_unlock(child); + } + + put_task_struct(child); + return killed; +} + +/* Please note the differences between mmput and mm_release. + * mmput is called whenever we stop holding onto a mm_struct, + * error success whatever. + * + * mm_release is called after a mm_struct has been removed + * from the current process. + * + * This difference is important for error handling, when we + * only half set up a mm_struct for a new process and need to restore + * the old one. Because we mmput the new mm_struct before + * restoring the old one. . . + * Eric Biederman 10 January 1998 + */ +void mm_release(struct task_struct *tsk, struct mm_struct *mm) +{ + /* Get rid of any futexes when releasing the mm */ +#ifdef CONFIG_FUTEX + if (unlikely(tsk->robust_list)) { + exit_robust_list(tsk); + tsk->robust_list = NULL; + } +#ifdef CONFIG_COMPAT + if (unlikely(tsk->compat_robust_list)) { + compat_exit_robust_list(tsk); + tsk->compat_robust_list = NULL; + } +#endif + if (unlikely(!list_empty(&tsk->pi_state_list))) + exit_pi_state_list(tsk); +#endif + + uprobe_free_utask(tsk); + + /* Get rid of any cached register state */ + deactivate_mm(tsk, mm); + + /* + * If we're exiting normally, clear a user-space tid field if + * requested. We leave this alone when dying by signal, to leave + * the value intact in a core dump, and to save the unnecessary + * trouble, say, a killed vfork parent shouldn't touch this mm. + * Userland only wants this done for a sys_exit. + */ + if (tsk->clear_child_tid) { + if (!(tsk->flags & PF_SIGNALED) && + atomic_read(&mm->mm_users) > 1) { + /* + * We don't check the error code - if userspace has + * not set up a proper pointer then tough luck. + */ + put_user(0, tsk->clear_child_tid); + sys_futex(tsk->clear_child_tid, FUTEX_WAKE, + 1, NULL, NULL, 0); + } + tsk->clear_child_tid = NULL; + } + + /* + * All done, finally we can wake up parent and return this mm to him. + * Also kthread_stop() uses this completion for synchronization. + */ + if (tsk->vfork_done) + complete_vfork_done(tsk); +} + +/* + * Allocate a new mm structure and copy contents from the + * mm structure of the passed in task structure. + */ +struct mm_struct *dup_mm(struct task_struct *tsk) +{ + struct mm_struct *mm, *oldmm = current->mm; + int err; + + if (!oldmm) + return NULL; + + mm = allocate_mm(); + if (!mm) + goto fail_nomem; + + memcpy(mm, oldmm, sizeof(*mm)); + mm_init_cpumask(mm); + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + mm->pmd_huge_pte = NULL; +#endif +#ifdef CONFIG_NUMA_BALANCING + mm->first_nid = NUMA_PTE_SCAN_INIT; +#endif + if (!mm_init(mm, tsk)) + goto fail_nomem; + + if (init_new_context(tsk, mm)) + goto fail_nocontext; + + dup_mm_exe_file(oldmm, mm); + + err = dup_mmap(mm, oldmm); + if (err) + goto free_pt; + + mm->hiwater_rss = get_mm_rss(mm); + mm->hiwater_vm = mm->total_vm; + + if (mm->binfmt && !try_module_get(mm->binfmt->module)) + goto free_pt; + + return mm; + +free_pt: + /* don't put binfmt in mmput, we haven't got module yet */ + mm->binfmt = NULL; + mmput(mm); + +fail_nomem: + return NULL; + +fail_nocontext: + /* + * If init_new_context() failed, we cannot use mmput() to free the mm + * because it calls destroy_context() + */ + mm_free_pgd(mm); + free_mm(mm); + return NULL; +} + +static int copy_mm(unsigned long clone_flags, struct task_struct *tsk) +{ + struct mm_struct *mm, *oldmm; + int retval; + + tsk->min_flt = tsk->maj_flt = 0; + tsk->nvcsw = tsk->nivcsw = 0; +#ifdef CONFIG_DETECT_HUNG_TASK + tsk->last_switch_count = tsk->nvcsw + tsk->nivcsw; +#endif + + tsk->mm = NULL; + tsk->active_mm = NULL; + + /* + * Are we cloning a kernel thread? + * + * We need to steal a active VM for that.. + */ + oldmm = current->mm; + if (!oldmm) + return 0; + + if (clone_flags & CLONE_VM) { + atomic_inc(&oldmm->mm_users); + mm = oldmm; + goto good_mm; + } + + retval = -ENOMEM; + mm = dup_mm(tsk); + if (!mm) + goto fail_nomem; + +good_mm: + tsk->mm = mm; + tsk->active_mm = mm; + return 0; + +fail_nomem: + return retval; +} + +static int copy_fs(unsigned long clone_flags, struct task_struct *tsk) +{ + struct fs_struct *fs = current->fs; + if (clone_flags & CLONE_FS) { + /* tsk->fs is already what we want */ + spin_lock(&fs->lock); + if (fs->in_exec) { + spin_unlock(&fs->lock); + return -EAGAIN; + } + fs->users++; + spin_unlock(&fs->lock); + return 0; + } + tsk->fs = copy_fs_struct(fs); + if (!tsk->fs) + return -ENOMEM; + return 0; +} + +static int copy_files(unsigned long clone_flags, struct task_struct *tsk) +{ + struct files_struct *oldf, *newf; + int error = 0; + + /* + * A background process may not have any files ... + */ + oldf = current->files; + if (!oldf) + goto out; + + if (clone_flags & CLONE_FILES) { + atomic_inc(&oldf->count); + goto out; + } + + newf = dup_fd(oldf, &error); + if (!newf) + goto out; + + tsk->files = newf; + error = 0; +out: + return error; +} + +static int copy_io(unsigned long clone_flags, struct task_struct *tsk) +{ +#ifdef CONFIG_BLOCK + struct io_context *ioc = current->io_context; + struct io_context *new_ioc; + + if (!ioc) + return 0; + /* + * Share io context with parent, if CLONE_IO is set + */ + if (clone_flags & CLONE_IO) { + ioc_task_link(ioc); + tsk->io_context = ioc; + } else if (ioprio_valid(ioc->ioprio)) { + new_ioc = get_task_io_context(tsk, GFP_KERNEL, NUMA_NO_NODE); + if (unlikely(!new_ioc)) + return -ENOMEM; + + new_ioc->ioprio = ioc->ioprio; + put_io_context(new_ioc); + } +#endif + return 0; +} + +static int copy_sighand(unsigned long clone_flags, struct task_struct *tsk) +{ + struct sighand_struct *sig; + + if (clone_flags & CLONE_SIGHAND) { + atomic_inc(¤t->sighand->count); + return 0; + } + sig = kmem_cache_alloc(sighand_cachep, GFP_KERNEL); + rcu_assign_pointer(tsk->sighand, sig); + if (!sig) + return -ENOMEM; + atomic_set(&sig->count, 1); + memcpy(sig->action, current->sighand->action, sizeof(sig->action)); + return 0; +} + +void __cleanup_sighand(struct sighand_struct *sighand) +{ + if (atomic_dec_and_test(&sighand->count)) { + signalfd_cleanup(sighand); + kmem_cache_free(sighand_cachep, sighand); + } +} + + +/* + * Initialize POSIX timer handling for a thread group. + */ +static void posix_cpu_timers_init_group(struct signal_struct *sig) +{ + unsigned long cpu_limit; + + /* Thread group counters. */ + thread_group_cputime_init(sig); + + cpu_limit = ACCESS_ONCE(sig->rlim[RLIMIT_CPU].rlim_cur); + if (cpu_limit != RLIM_INFINITY) { + sig->cputime_expires.prof_exp = secs_to_cputime(cpu_limit); + sig->cputimer.running = 1; + } + + /* The timer lists. */ + INIT_LIST_HEAD(&sig->cpu_timers[0]); + INIT_LIST_HEAD(&sig->cpu_timers[1]); + INIT_LIST_HEAD(&sig->cpu_timers[2]); +} + +static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) +{ + struct signal_struct *sig; + + if (clone_flags & CLONE_THREAD) + return 0; + + sig = kmem_cache_zalloc(signal_cachep, GFP_KERNEL); + tsk->signal = sig; + if (!sig) + return -ENOMEM; + + sig->nr_threads = 1; + atomic_set(&sig->live, 1); + atomic_set(&sig->sigcnt, 1); + + /* list_add(thread_node, thread_head) without INIT_LIST_HEAD() */ + sig->thread_head = (struct list_head)LIST_HEAD_INIT(tsk->thread_node); + tsk->thread_node = (struct list_head)LIST_HEAD_INIT(sig->thread_head); + + init_waitqueue_head(&sig->wait_chldexit); + sig->curr_target = tsk; + init_sigpending(&sig->shared_pending); + INIT_LIST_HEAD(&sig->posix_timers); + + hrtimer_init(&sig->real_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + sig->real_timer.function = it_real_fn; + + task_lock(current->group_leader); + memcpy(sig->rlim, current->signal->rlim, sizeof sig->rlim); + task_unlock(current->group_leader); + + posix_cpu_timers_init_group(sig); + + tty_audit_fork(sig); + sched_autogroup_fork(sig); + +#ifdef CONFIG_CGROUPS + init_rwsem(&sig->group_rwsem); +#endif + + sig->oom_score_adj = current->signal->oom_score_adj; + sig->oom_score_adj_min = current->signal->oom_score_adj_min; + + sig->has_child_subreaper = current->signal->has_child_subreaper || + current->signal->is_child_subreaper; + + mutex_init(&sig->cred_guard_mutex); + + return 0; +} + +static void copy_flags(unsigned long clone_flags, struct task_struct *p) +{ + unsigned long new_flags = p->flags; + + new_flags &= ~(PF_SUPERPRIV | PF_WQ_WORKER); + new_flags |= PF_FORKNOEXEC; + p->flags = new_flags; +} + +SYSCALL_DEFINE1(set_tid_address, int __user *, tidptr) +{ + current->clear_child_tid = tidptr; + + return task_pid_vnr(current); +} + +static void rt_mutex_init_task(struct task_struct *p) +{ + raw_spin_lock_init(&p->pi_lock); +#ifdef CONFIG_RT_MUTEXES + plist_head_init(&p->pi_waiters); + p->pi_blocked_on = NULL; +#endif +} + +#ifdef CONFIG_MM_OWNER +void mm_init_owner(struct mm_struct *mm, struct task_struct *p) +{ + mm->owner = p; +} +#endif /* CONFIG_MM_OWNER */ + +/* + * Initialize POSIX timer handling for a single task. + */ +static void posix_cpu_timers_init(struct task_struct *tsk) +{ + tsk->cputime_expires.prof_exp = 0; + tsk->cputime_expires.virt_exp = 0; + tsk->cputime_expires.sched_exp = 0; + INIT_LIST_HEAD(&tsk->cpu_timers[0]); + INIT_LIST_HEAD(&tsk->cpu_timers[1]); + INIT_LIST_HEAD(&tsk->cpu_timers[2]); +} + +/* + * This creates a new process as a copy of the old one, + * but does not actually start it yet. + * + * It copies the registers, and all the appropriate + * parts of the process environment (as per the clone + * flags). The actual kick-off is left to the caller. + */ +static struct task_struct *copy_process(unsigned long clone_flags, + unsigned long stack_start, + unsigned long stack_size, + int __user *child_tidptr, + struct pid *pid, + int trace) +{ + int retval; + struct task_struct *p; + + if ((clone_flags & (CLONE_NEWNS|CLONE_FS)) == (CLONE_NEWNS|CLONE_FS)) + return ERR_PTR(-EINVAL); + + if ((clone_flags & (CLONE_NEWUSER|CLONE_FS)) == (CLONE_NEWUSER|CLONE_FS)) + return ERR_PTR(-EINVAL); + + /* + * Thread groups must share signals as well, and detached threads + * can only be started up within the thread group. + */ + if ((clone_flags & CLONE_THREAD) && !(clone_flags & CLONE_SIGHAND)) + return ERR_PTR(-EINVAL); + + /* + * Shared signal handlers imply shared VM. By way of the above, + * thread groups also imply shared VM. Blocking this case allows + * for various simplifications in other code. + */ + if ((clone_flags & CLONE_SIGHAND) && !(clone_flags & CLONE_VM)) + return ERR_PTR(-EINVAL); + + /* + * Siblings of global init remain as zombies on exit since they are + * not reaped by their parent (swapper). To solve this and to avoid + * multi-rooted process trees, prevent global and container-inits + * from creating siblings. + */ + if ((clone_flags & CLONE_PARENT) && + current->signal->flags & SIGNAL_UNKILLABLE) + return ERR_PTR(-EINVAL); + + /* + * If the new process will be in a different pid namespace don't + * allow it to share a thread group or signal handlers with the + * forking task. + */ + if ((clone_flags & (CLONE_SIGHAND | CLONE_NEWPID)) && + (task_active_pid_ns(current) != current->nsproxy->pid_ns)) + return ERR_PTR(-EINVAL); + + retval = security_task_create(clone_flags); + if (retval) + goto fork_out; + + retval = -ENOMEM; + p = dup_task_struct(current); + if (!p) + goto fork_out; + + ftrace_graph_init_task(p); + get_seccomp_filter(p); + + rt_mutex_init_task(p); + +#ifdef CONFIG_PROVE_LOCKING + DEBUG_LOCKS_WARN_ON(!p->hardirqs_enabled); + DEBUG_LOCKS_WARN_ON(!p->softirqs_enabled); +#endif + retval = -EAGAIN; + if (atomic_read(&p->real_cred->user->processes) >= + task_rlimit(p, RLIMIT_NPROC)) { + if (!capable(CAP_SYS_ADMIN) && !capable(CAP_SYS_RESOURCE) && + p->real_cred->user != INIT_USER) + goto bad_fork_free; + } + current->flags &= ~PF_NPROC_EXCEEDED; + + retval = copy_creds(p, clone_flags); + if (retval < 0) + goto bad_fork_free; + + /* + * If multiple threads are within copy_process(), then this check + * triggers too late. This doesn't hurt, the check is only there + * to stop root fork bombs. + */ + retval = -EAGAIN; + if (nr_threads >= max_threads) + goto bad_fork_cleanup_count; + + if (!try_module_get(task_thread_info(p)->exec_domain->module)) + goto bad_fork_cleanup_count; + + p->did_exec = 0; + delayacct_tsk_init(p); /* Must remain after dup_task_struct() */ + copy_flags(clone_flags, p); + INIT_LIST_HEAD(&p->children); + INIT_LIST_HEAD(&p->sibling); + rcu_copy_process(p); + p->vfork_done = NULL; + spin_lock_init(&p->alloc_lock); + + init_sigpending(&p->pending); + + p->utime = p->stime = p->gtime = 0; + p->utimescaled = p->stimescaled = 0; +#ifndef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE + p->prev_cputime.utime = p->prev_cputime.stime = 0; +#endif +#ifdef CONFIG_VIRT_CPU_ACCOUNTING_GEN + seqlock_init(&p->vtime_seqlock); + p->vtime_snap = 0; + p->vtime_snap_whence = VTIME_SLEEPING; +#endif + +#if defined(SPLIT_RSS_COUNTING) + memset(&p->rss_stat, 0, sizeof(p->rss_stat)); +#endif + + p->default_timer_slack_ns = current->timer_slack_ns; + + task_io_accounting_init(&p->ioac); + acct_clear_integrals(p); + + posix_cpu_timers_init(p); + + do_posix_clock_monotonic_gettime(&p->start_time); + p->real_start_time = p->start_time; + monotonic_to_bootbased(&p->real_start_time); + p->io_context = NULL; + p->audit_context = NULL; + if (clone_flags & CLONE_THREAD) + threadgroup_change_begin(current); + cgroup_fork(p); +#ifdef CONFIG_NUMA + p->mempolicy = mpol_dup(p->mempolicy); + if (IS_ERR(p->mempolicy)) { + retval = PTR_ERR(p->mempolicy); + p->mempolicy = NULL; + goto bad_fork_cleanup_cgroup; + } + mpol_fix_fork_child_flag(p); +#endif +#ifdef CONFIG_CPUSETS + p->cpuset_mem_spread_rotor = NUMA_NO_NODE; + p->cpuset_slab_spread_rotor = NUMA_NO_NODE; + seqcount_init(&p->mems_allowed_seq); +#endif +#ifdef CONFIG_TRACE_IRQFLAGS + p->irq_events = 0; + p->hardirqs_enabled = 0; + p->hardirq_enable_ip = 0; + p->hardirq_enable_event = 0; + p->hardirq_disable_ip = _THIS_IP_; + p->hardirq_disable_event = 0; + p->softirqs_enabled = 1; + p->softirq_enable_ip = _THIS_IP_; + p->softirq_enable_event = 0; + p->softirq_disable_ip = 0; + p->softirq_disable_event = 0; + p->hardirq_context = 0; + p->softirq_context = 0; +#endif +#ifdef CONFIG_LOCKDEP + p->lockdep_depth = 0; /* no locks held yet */ + p->curr_chain_key = 0; + p->lockdep_recursion = 0; +#endif + +#ifdef CONFIG_DEBUG_MUTEXES + p->blocked_on = NULL; /* not blocked yet */ +#endif +#ifdef CONFIG_MEMCG + p->memcg_batch.do_batch = 0; + p->memcg_batch.memcg = NULL; +#endif +#ifdef CONFIG_BCACHE + p->sequential_io = 0; + p->sequential_io_avg = 0; +#endif + + /* Perform scheduler related setup. Assign this task to a CPU. */ + sched_fork(p); + + retval = perf_event_init_task(p); + if (retval) + goto bad_fork_cleanup_policy; + retval = audit_alloc(p); + if (retval) + goto bad_fork_cleanup_policy; + /* copy all the process information */ + retval = copy_semundo(clone_flags, p); + if (retval) + goto bad_fork_cleanup_audit; + retval = copy_files(clone_flags, p); + if (retval) + goto bad_fork_cleanup_semundo; + retval = copy_fs(clone_flags, p); + if (retval) + goto bad_fork_cleanup_files; + retval = copy_sighand(clone_flags, p); + if (retval) + goto bad_fork_cleanup_fs; + retval = copy_signal(clone_flags, p); + if (retval) + goto bad_fork_cleanup_sighand; + retval = copy_mm(clone_flags, p); + if (retval) + goto bad_fork_cleanup_signal; + retval = copy_namespaces(clone_flags, p); + if (retval) + goto bad_fork_cleanup_mm; + retval = copy_io(clone_flags, p); + if (retval) + goto bad_fork_cleanup_namespaces; + retval = copy_thread(clone_flags, stack_start, stack_size, p); + if (retval) + goto bad_fork_cleanup_io; + + if (pid != &init_struct_pid) { + retval = -ENOMEM; + pid = alloc_pid(p->nsproxy->pid_ns); + if (!pid) + goto bad_fork_cleanup_io; + } + + p->pid = pid_nr(pid); + p->tgid = p->pid; + if (clone_flags & CLONE_THREAD) + p->tgid = current->tgid; + + p->set_child_tid = (clone_flags & CLONE_CHILD_SETTID) ? child_tidptr : NULL; + /* + * Clear TID on mm_release()? + */ + p->clear_child_tid = (clone_flags & CLONE_CHILD_CLEARTID) ? child_tidptr : NULL; +#ifdef CONFIG_BLOCK + p->plug = NULL; +#endif +#ifdef CONFIG_FUTEX + p->robust_list = NULL; +#ifdef CONFIG_COMPAT + p->compat_robust_list = NULL; +#endif + INIT_LIST_HEAD(&p->pi_state_list); + p->pi_state_cache = NULL; +#endif + uprobe_copy_process(p); + /* + * sigaltstack should be cleared when sharing the same VM + */ + if ((clone_flags & (CLONE_VM|CLONE_VFORK)) == CLONE_VM) + p->sas_ss_sp = p->sas_ss_size = 0; + + /* + * Syscall tracing and stepping should be turned off in the + * child regardless of CLONE_PTRACE. + */ + user_disable_single_step(p); + clear_tsk_thread_flag(p, TIF_SYSCALL_TRACE); +#ifdef TIF_SYSCALL_EMU + clear_tsk_thread_flag(p, TIF_SYSCALL_EMU); +#endif + clear_all_latency_tracing(p); + + /* ok, now we should be set up.. */ + if (clone_flags & CLONE_THREAD) + p->exit_signal = -1; + else if (clone_flags & CLONE_PARENT) + p->exit_signal = current->group_leader->exit_signal; + else + p->exit_signal = (clone_flags & CSIGNAL); + + p->pdeath_signal = 0; + p->exit_state = 0; + + p->nr_dirtied = 0; + p->nr_dirtied_pause = 128 >> (PAGE_SHIFT - 10); + p->dirty_paused_when = 0; + + /* + * Ok, make it visible to the rest of the system. + * We dont wake it up yet. + */ + p->group_leader = p; + INIT_LIST_HEAD(&p->thread_group); + p->task_works = NULL; + + /* Need tasklist lock for parent etc handling! */ + write_lock_irq(&tasklist_lock); + + /* CLONE_PARENT re-uses the old parent */ + if (clone_flags & (CLONE_PARENT|CLONE_THREAD)) { + p->real_parent = current->real_parent; + p->parent_exec_id = current->parent_exec_id; + } else { + p->real_parent = current; + p->parent_exec_id = current->self_exec_id; + } + + spin_lock(¤t->sighand->siglock); + + /* + * Process group and session signals need to be delivered to just the + * parent before the fork or both the parent and the child after the + * fork. Restart if a signal comes in before we add the new process to + * it's process group. + * A fatal signal pending means that current will exit, so the new + * thread can't slip out of an OOM kill (or normal SIGKILL). + */ + recalc_sigpending(); + if (signal_pending(current)) { + spin_unlock(¤t->sighand->siglock); + write_unlock_irq(&tasklist_lock); + retval = -ERESTARTNOINTR; + goto bad_fork_free_pid; + } + + if (clone_flags & CLONE_THREAD) { + current->signal->nr_threads++; + atomic_inc(¤t->signal->live); + atomic_inc(¤t->signal->sigcnt); + p->group_leader = current->group_leader; + list_add_tail_rcu(&p->thread_group, &p->group_leader->thread_group); + } + + if (likely(p->pid)) { + ptrace_init_task(p, (clone_flags & CLONE_PTRACE) || trace); + + if (thread_group_leader(p)) { + if (is_child_reaper(pid)) { + ns_of_pid(pid)->child_reaper = p; + p->signal->flags |= SIGNAL_UNKILLABLE; + } + + p->signal->leader_pid = pid; + p->signal->tty = tty_kref_get(current->signal->tty); + attach_pid(p, PIDTYPE_PGID, task_pgrp(current)); + attach_pid(p, PIDTYPE_SID, task_session(current)); + list_add_tail(&p->sibling, &p->real_parent->children); + list_add_tail_rcu(&p->tasks, &init_task.tasks); + __this_cpu_inc(process_counts); + } else { + list_add_tail_rcu(&p->thread_node, + &p->signal->thread_head); + } + attach_pid(p, PIDTYPE_PID, pid); + nr_threads++; + } + + total_forks++; + spin_unlock(¤t->sighand->siglock); + write_unlock_irq(&tasklist_lock); + proc_fork_connector(p); + cgroup_post_fork(p); + if (clone_flags & CLONE_THREAD) + threadgroup_change_end(current); + perf_event_fork(p); + + trace_task_newtask(p, clone_flags); + + return p; + +bad_fork_free_pid: + if (pid != &init_struct_pid) + free_pid(pid); +bad_fork_cleanup_io: + if (p->io_context) + exit_io_context(p); +bad_fork_cleanup_namespaces: + exit_task_namespaces(p); +bad_fork_cleanup_mm: + if (p->mm) + mmput(p->mm); +bad_fork_cleanup_signal: + if (!(clone_flags & CLONE_THREAD)) + free_signal_struct(p->signal); +bad_fork_cleanup_sighand: + __cleanup_sighand(p->sighand); +bad_fork_cleanup_fs: + exit_fs(p); /* blocking */ +bad_fork_cleanup_files: + exit_files(p); /* blocking */ +bad_fork_cleanup_semundo: + exit_sem(p); +bad_fork_cleanup_audit: + audit_free(p); +bad_fork_cleanup_policy: + perf_event_free_task(p); +#ifdef CONFIG_NUMA + mpol_put(p->mempolicy); +bad_fork_cleanup_cgroup: +#endif + if (clone_flags & CLONE_THREAD) + threadgroup_change_end(current); + cgroup_exit(p, 0); + delayacct_tsk_free(p); + module_put(task_thread_info(p)->exec_domain->module); +bad_fork_cleanup_count: + atomic_dec(&p->cred->user->processes); + exit_creds(p); +bad_fork_free: + free_task(p); +fork_out: + return ERR_PTR(retval); +} + +static inline void init_idle_pids(struct pid_link *links) +{ + enum pid_type type; + + for (type = PIDTYPE_PID; type < PIDTYPE_MAX; ++type) { + INIT_HLIST_NODE(&links[type].node); /* not really needed */ + links[type].pid = &init_struct_pid; + } +} + +struct task_struct * __cpuinit fork_idle(int cpu) +{ + struct task_struct *task; + task = copy_process(CLONE_VM, 0, 0, NULL, &init_struct_pid, 0); + if (!IS_ERR(task)) { + init_idle_pids(task->pids); + init_idle(task, cpu); + } + + return task; +} + +/* + * Ok, this is the main fork-routine. + * + * It copies the process, and if successful kick-starts + * it and waits for it to finish using the VM if required. + */ +long do_fork(unsigned long clone_flags, + unsigned long stack_start, + unsigned long stack_size, + int __user *parent_tidptr, + int __user *child_tidptr) +{ + struct task_struct *p; + int trace = 0; + long nr; + + /* + * Do some preliminary argument and permissions checking before we + * actually start allocating stuff + */ + if (clone_flags & (CLONE_NEWUSER | CLONE_NEWPID)) { + if (clone_flags & (CLONE_THREAD|CLONE_PARENT)) + return -EINVAL; + } + + /* + * Determine whether and which event to report to ptracer. When + * called from kernel_thread or CLONE_UNTRACED is explicitly + * requested, no event is reported; otherwise, report if the event + * for the type of forking is enabled. + */ + if (!(clone_flags & CLONE_UNTRACED)) { + if (clone_flags & CLONE_VFORK) + trace = PTRACE_EVENT_VFORK; + else if ((clone_flags & CSIGNAL) != SIGCHLD) + trace = PTRACE_EVENT_CLONE; + else + trace = PTRACE_EVENT_FORK; + + if (likely(!ptrace_event_enabled(current, trace))) + trace = 0; + } + + p = copy_process(clone_flags, stack_start, stack_size, + child_tidptr, NULL, trace); + /* + * Do this prior waking up the new thread - the thread pointer + * might get invalid after that point, if the thread exits quickly. + */ + if (!IS_ERR(p)) { + struct completion vfork; + + trace_sched_process_fork(current, p); + + nr = task_pid_vnr(p); + + if (clone_flags & CLONE_PARENT_SETTID) + put_user(nr, parent_tidptr); + + if (clone_flags & CLONE_VFORK) { + p->vfork_done = &vfork; + init_completion(&vfork); + get_task_struct(p); + } + + wake_up_new_task(p); + + /* forking complete and child started to run, tell ptracer */ + if (unlikely(trace)) + ptrace_event(trace, nr); + + if (clone_flags & CLONE_VFORK) { + if (!wait_for_vfork_done(p, &vfork)) + ptrace_event(PTRACE_EVENT_VFORK_DONE, nr); + } + } else { + nr = PTR_ERR(p); + } + return nr; +} + +/* + * Create a kernel thread. + */ +pid_t kernel_thread(int (*fn)(void *), void *arg, unsigned long flags) +{ + return do_fork(flags|CLONE_VM|CLONE_UNTRACED, (unsigned long)fn, + (unsigned long)arg, NULL, NULL); +} + +#ifdef __ARCH_WANT_SYS_FORK +SYSCALL_DEFINE0(fork) +{ +#ifdef CONFIG_MMU + return do_fork(SIGCHLD, 0, 0, NULL, NULL); +#else + /* can not support in nommu mode */ + return(-EINVAL); +#endif +} +#endif + +#ifdef __ARCH_WANT_SYS_VFORK +SYSCALL_DEFINE0(vfork) +{ + return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, 0, + 0, NULL, NULL); +} +#endif + +#ifdef __ARCH_WANT_SYS_CLONE +#ifdef CONFIG_CLONE_BACKWARDS +SYSCALL_DEFINE5(clone, unsigned long, clone_flags, unsigned long, newsp, + int __user *, parent_tidptr, + int, tls_val, + int __user *, child_tidptr) +#elif defined(CONFIG_CLONE_BACKWARDS2) +SYSCALL_DEFINE5(clone, unsigned long, newsp, unsigned long, clone_flags, + int __user *, parent_tidptr, + int __user *, child_tidptr, + int, tls_val) +#elif defined(CONFIG_CLONE_BACKWARDS3) +SYSCALL_DEFINE6(clone, unsigned long, clone_flags, unsigned long, newsp, + int, stack_size, + int __user *, parent_tidptr, + int __user *, child_tidptr, + int, tls_val) +#else +SYSCALL_DEFINE5(clone, unsigned long, clone_flags, unsigned long, newsp, + int __user *, parent_tidptr, + int __user *, child_tidptr, + int, tls_val) +#endif +{ + return do_fork(clone_flags, newsp, 0, parent_tidptr, child_tidptr); +} +#endif + +#ifndef ARCH_MIN_MMSTRUCT_ALIGN +#define ARCH_MIN_MMSTRUCT_ALIGN 0 +#endif + +static void sighand_ctor(void *data) +{ + struct sighand_struct *sighand = data; + + spin_lock_init(&sighand->siglock); + init_waitqueue_head(&sighand->signalfd_wqh); +} + +void __init proc_caches_init(void) +{ + sighand_cachep = kmem_cache_create("sighand_cache", + sizeof(struct sighand_struct), 0, + SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_DESTROY_BY_RCU| + SLAB_NOTRACK, sighand_ctor); + signal_cachep = kmem_cache_create("signal_cache", + sizeof(struct signal_struct), 0, + SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_NOTRACK, NULL); + files_cachep = kmem_cache_create("files_cache", + sizeof(struct files_struct), 0, + SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_NOTRACK, NULL); + fs_cachep = kmem_cache_create("fs_cache", + sizeof(struct fs_struct), 0, + SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_NOTRACK, NULL); + /* + * FIXME! The "sizeof(struct mm_struct)" currently includes the + * whole struct cpumask for the OFFSTACK case. We could change + * this to *only* allocate as much of it as required by the + * maximum number of CPU's we can ever have. The cpumask_allocation + * is at the end of the structure, exactly for that reason. + */ + mm_cachep = kmem_cache_create("mm_struct", + sizeof(struct mm_struct), ARCH_MIN_MMSTRUCT_ALIGN, + SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_NOTRACK, NULL); + vm_area_cachep = KMEM_CACHE(vm_area_struct, SLAB_PANIC); + mmap_init(); + nsproxy_cache_init(); +} + +/* + * Check constraints on flags passed to the unshare system call. + */ +static int check_unshare_flags(unsigned long unshare_flags) +{ + if (unshare_flags & ~(CLONE_THREAD|CLONE_FS|CLONE_NEWNS|CLONE_SIGHAND| + CLONE_VM|CLONE_FILES|CLONE_SYSVSEM| + CLONE_NEWUTS|CLONE_NEWIPC|CLONE_NEWNET| + CLONE_NEWUSER|CLONE_NEWPID)) + return -EINVAL; + /* + * Not implemented, but pretend it works if there is nothing to + * unshare. Note that unsharing CLONE_THREAD or CLONE_SIGHAND + * needs to unshare vm. + */ + if (unshare_flags & (CLONE_THREAD | CLONE_SIGHAND | CLONE_VM)) { + /* FIXME: get_task_mm() increments ->mm_users */ + if (atomic_read(¤t->mm->mm_users) > 1) + return -EINVAL; + } + + return 0; +} + +/* + * Unshare the filesystem structure if it is being shared + */ +static int unshare_fs(unsigned long unshare_flags, struct fs_struct **new_fsp) +{ + struct fs_struct *fs = current->fs; + + if (!(unshare_flags & CLONE_FS) || !fs) + return 0; + + /* don't need lock here; in the worst case we'll do useless copy */ + if (fs->users == 1) + return 0; + + *new_fsp = copy_fs_struct(fs); + if (!*new_fsp) + return -ENOMEM; + + return 0; +} + +/* + * Unshare file descriptor table if it is being shared + */ +static int unshare_fd(unsigned long unshare_flags, struct files_struct **new_fdp) +{ + struct files_struct *fd = current->files; + int error = 0; + + if ((unshare_flags & CLONE_FILES) && + (fd && atomic_read(&fd->count) > 1)) { + *new_fdp = dup_fd(fd, &error); + if (!*new_fdp) + return error; + } + + return 0; +} + +/* + * unshare allows a process to 'unshare' part of the process + * context which was originally shared using clone. copy_* + * functions used by do_fork() cannot be used here directly + * because they modify an inactive task_struct that is being + * constructed. Here we are modifying the current, active, + * task_struct. + */ +SYSCALL_DEFINE1(unshare, unsigned long, unshare_flags) +{ + struct fs_struct *fs, *new_fs = NULL; + struct files_struct *fd, *new_fd = NULL; + struct cred *new_cred = NULL; + struct nsproxy *new_nsproxy = NULL; + int do_sysvsem = 0; + int err; + + /* + * If unsharing a user namespace must also unshare the thread. + */ + if (unshare_flags & CLONE_NEWUSER) + unshare_flags |= CLONE_THREAD | CLONE_FS; + /* + * If unsharing a pid namespace must also unshare the thread. + */ + if (unshare_flags & CLONE_NEWPID) + unshare_flags |= CLONE_THREAD; + /* + * If unsharing a thread from a thread group, must also unshare vm. + */ + if (unshare_flags & CLONE_THREAD) + unshare_flags |= CLONE_VM; + /* + * If unsharing vm, must also unshare signal handlers. + */ + if (unshare_flags & CLONE_VM) + unshare_flags |= CLONE_SIGHAND; + /* + * If unsharing namespace, must also unshare filesystem information. + */ + if (unshare_flags & CLONE_NEWNS) + unshare_flags |= CLONE_FS; + + err = check_unshare_flags(unshare_flags); + if (err) + goto bad_unshare_out; + /* + * CLONE_NEWIPC must also detach from the undolist: after switching + * to a new ipc namespace, the semaphore arrays from the old + * namespace are unreachable. + */ + if (unshare_flags & (CLONE_NEWIPC|CLONE_SYSVSEM)) + do_sysvsem = 1; + err = unshare_fs(unshare_flags, &new_fs); + if (err) + goto bad_unshare_out; + err = unshare_fd(unshare_flags, &new_fd); + if (err) + goto bad_unshare_cleanup_fs; + err = unshare_userns(unshare_flags, &new_cred); + if (err) + goto bad_unshare_cleanup_fd; + err = unshare_nsproxy_namespaces(unshare_flags, &new_nsproxy, + new_cred, new_fs); + if (err) + goto bad_unshare_cleanup_cred; + + if (new_fs || new_fd || do_sysvsem || new_cred || new_nsproxy) { + if (do_sysvsem) { + /* + * CLONE_SYSVSEM is equivalent to sys_exit(). + */ + exit_sem(current); + } + + if (new_nsproxy) + switch_task_namespaces(current, new_nsproxy); + + task_lock(current); + + if (new_fs) { + fs = current->fs; + spin_lock(&fs->lock); + current->fs = new_fs; + if (--fs->users) + new_fs = NULL; + else + new_fs = fs; + spin_unlock(&fs->lock); + } + + if (new_fd) { + fd = current->files; + current->files = new_fd; + new_fd = fd; + } + + task_unlock(current); + + if (new_cred) { + /* Install the new user namespace */ + commit_creds(new_cred); + new_cred = NULL; + } + } + +bad_unshare_cleanup_cred: + if (new_cred) + put_cred(new_cred); +bad_unshare_cleanup_fd: + if (new_fd) + put_files_struct(new_fd); + +bad_unshare_cleanup_fs: + if (new_fs) + free_fs_struct(new_fs); + +bad_unshare_out: + return err; +} + +/* + * Helper to unshare the files of the current task. + * We don't want to expose copy_files internals to + * the exec layer of the kernel. + */ + +int unshare_files(struct files_struct **displaced) +{ + struct task_struct *task = current; + struct files_struct *copy = NULL; + int error; + + error = unshare_fd(CLONE_FILES, ©); + if (error || !copy) { + *displaced = NULL; + return error; + } + *displaced = task->files; + task_lock(task); + task->files = copy; + task_unlock(task); + return 0; +} diff -Naur kernel.21.3.orig/MAINTAINERS kernel.21.3.new/MAINTAINERS --- kernel.21.3.orig/MAINTAINERS 2015-02-03 01:44:24.000000000 +0000 +++ kernel.21.3.new/MAINTAINERS 2015-04-25 23:28:26.599666849 +0000 @@ -1533,6 +1533,20 @@ F: include/uapi/linux/audit.h F: kernel/audit* +AUFS (advanced multi layered unification filesystem) FILESYSTEM +M: "J. R. Okajima" +L: linux-unionfs@vger.kernel.org +L: aufs-users@lists.sourceforge.net (members only) +W: http://aufs.sourceforge.net +T: git://git.code.sf.net/p/aufs/aufs3-linux +T: git://github.com/sfjro/aufs3-linux.git +S: Supported +F: Documentation/filesystems/aufs/ +F: Documentation/ABI/testing/debugfs-aufs +F: Documentation/ABI/testing/sysfs-aufs +F: fs/aufs/ +F: include/uapi/linux/aufs_type.h + AUXILIARY DISPLAY DRIVERS M: Miguel Ojeda Sandonis W: http://miguelojeda.es/auxdisplay.htm diff -Naur kernel.21.3.orig/mm/filemap.c kernel.21.3.new/mm/filemap.c --- kernel.21.3.orig/mm/filemap.c 2015-02-03 01:44:33.000000000 +0000 +++ kernel.21.3.new/mm/filemap.c 2015-04-25 23:28:45.348481244 +0000 @@ -1742,7 +1742,7 @@ int ret = VM_FAULT_LOCKED; sb_start_pagefault(inode->i_sb); - file_update_time(vma->vm_file); + vma_file_update_time(vma); lock_page(page); if (page->mapping != inode->i_mapping) { unlock_page(page); diff -Naur kernel.21.3.orig/mm/filemap.c.orig kernel.21.3.new/mm/filemap.c.orig --- kernel.21.3.orig/mm/filemap.c.orig 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/mm/filemap.c.orig 2015-02-03 01:44:33.000000000 +0000 @@ -0,0 +1,2603 @@ +/* + * linux/mm/filemap.c + * + * Copyright (C) 1994-1999 Linus Torvalds + */ + +/* + * This file handles the generic file mmap semantics used by + * most "normal" filesystems (but you don't /have/ to use this: + * the NFS filesystem used to do this differently, for example) + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include /* for BUG_ON(!in_atomic()) only */ +#include +#include +#include "internal.h" + +#define CREATE_TRACE_POINTS +#include + +/* + * FIXME: remove all knowledge of the buffer layer from the core VM + */ +#include /* for try_to_free_buffers */ + +#include + +/* + * Shared mappings implemented 30.11.1994. It's not fully working yet, + * though. + * + * Shared mappings now work. 15.8.1995 Bruno. + * + * finished 'unifying' the page and buffer cache and SMP-threaded the + * page-cache, 21.05.1999, Ingo Molnar + * + * SMP-threaded pagemap-LRU 1999, Andrea Arcangeli + */ + +/* + * Lock ordering: + * + * ->i_mmap_mutex (truncate_pagecache) + * ->private_lock (__free_pte->__set_page_dirty_buffers) + * ->swap_lock (exclusive_swap_page, others) + * ->mapping->tree_lock + * + * ->i_mutex + * ->i_mmap_mutex (truncate->unmap_mapping_range) + * + * ->mmap_sem + * ->i_mmap_mutex + * ->page_table_lock or pte_lock (various, mainly in memory.c) + * ->mapping->tree_lock (arch-dependent flush_dcache_mmap_lock) + * + * ->mmap_sem + * ->lock_page (access_process_vm) + * + * ->i_mutex (generic_file_buffered_write) + * ->mmap_sem (fault_in_pages_readable->do_page_fault) + * + * bdi->wb.list_lock + * sb_lock (fs/fs-writeback.c) + * ->mapping->tree_lock (__sync_single_inode) + * + * ->i_mmap_mutex + * ->anon_vma.lock (vma_adjust) + * + * ->anon_vma.lock + * ->page_table_lock or pte_lock (anon_vma_prepare and various) + * + * ->page_table_lock or pte_lock + * ->swap_lock (try_to_unmap_one) + * ->private_lock (try_to_unmap_one) + * ->tree_lock (try_to_unmap_one) + * ->zone.lru_lock (follow_page->mark_page_accessed) + * ->zone.lru_lock (check_pte_range->isolate_lru_page) + * ->private_lock (page_remove_rmap->set_page_dirty) + * ->tree_lock (page_remove_rmap->set_page_dirty) + * bdi.wb->list_lock (page_remove_rmap->set_page_dirty) + * ->inode->i_lock (page_remove_rmap->set_page_dirty) + * bdi.wb->list_lock (zap_pte_range->set_page_dirty) + * ->inode->i_lock (zap_pte_range->set_page_dirty) + * ->private_lock (zap_pte_range->__set_page_dirty_buffers) + * + * ->i_mmap_mutex + * ->tasklist_lock (memory_failure, collect_procs_ao) + */ + +/* + * Delete a page from the page cache and free it. Caller has to make + * sure the page is locked and that nobody else uses it - or that usage + * is safe. The caller must hold the mapping's tree_lock. + */ +void __delete_from_page_cache(struct page *page) +{ + struct address_space *mapping = page->mapping; + + trace_mm_filemap_delete_from_page_cache(page); + /* + * if we're uptodate, flush out into the cleancache, otherwise + * invalidate any existing cleancache entries. We can't leave + * stale data around in the cleancache once our page is gone + */ + if (PageUptodate(page) && PageMappedToDisk(page)) + cleancache_put_page(page); + else + cleancache_invalidate_page(mapping, page); + + radix_tree_delete(&mapping->page_tree, page->index); + page->mapping = NULL; + /* Leave page->index set: truncation lookup relies upon it */ + mapping->nrpages--; + __dec_zone_page_state(page, NR_FILE_PAGES); + if (PageSwapBacked(page)) + __dec_zone_page_state(page, NR_SHMEM); + BUG_ON(page_mapped(page)); + + /* + * Some filesystems seem to re-dirty the page even after + * the VM has canceled the dirty bit (eg ext3 journaling). + * + * Fix it up by doing a final dirty accounting check after + * having removed the page entirely. + */ + if (PageDirty(page) && mapping_cap_account_dirty(mapping)) { + dec_zone_page_state(page, NR_FILE_DIRTY); + dec_bdi_stat(mapping->backing_dev_info, BDI_RECLAIMABLE); + } +} + +/** + * delete_from_page_cache - delete page from page cache + * @page: the page which the kernel is trying to remove from page cache + * + * This must be called only on pages that have been verified to be in the page + * cache and locked. It will never put the page into the free list, the caller + * has a reference on the page. + */ +void delete_from_page_cache(struct page *page) +{ + struct address_space *mapping = page->mapping; + void (*freepage)(struct page *); + + BUG_ON(!PageLocked(page)); + + freepage = mapping->a_ops->freepage; + spin_lock_irq(&mapping->tree_lock); + __delete_from_page_cache(page); + spin_unlock_irq(&mapping->tree_lock); + mem_cgroup_uncharge_cache_page(page); + + if (freepage) + freepage(page); + page_cache_release(page); +} +EXPORT_SYMBOL(delete_from_page_cache); + +static int sleep_on_page(void *word) +{ + io_schedule(); + return 0; +} + +static int sleep_on_page_killable(void *word) +{ + sleep_on_page(word); + return fatal_signal_pending(current) ? -EINTR : 0; +} + +static int filemap_check_errors(struct address_space *mapping) +{ + int ret = 0; + /* Check for outstanding write errors */ + if (test_and_clear_bit(AS_ENOSPC, &mapping->flags)) + ret = -ENOSPC; + if (test_and_clear_bit(AS_EIO, &mapping->flags)) + ret = -EIO; + return ret; +} + +/** + * __filemap_fdatawrite_range - start writeback on mapping dirty pages in range + * @mapping: address space structure to write + * @start: offset in bytes where the range starts + * @end: offset in bytes where the range ends (inclusive) + * @sync_mode: enable synchronous operation + * + * Start writeback against all of a mapping's dirty pages that lie + * within the byte offsets inclusive. + * + * If sync_mode is WB_SYNC_ALL then this is a "data integrity" operation, as + * opposed to a regular memory cleansing writeback. The difference between + * these two operations is that if a dirty page/buffer is encountered, it must + * be waited upon, and not just skipped over. + */ +int __filemap_fdatawrite_range(struct address_space *mapping, loff_t start, + loff_t end, int sync_mode) +{ + int ret; + struct writeback_control wbc = { + .sync_mode = sync_mode, + .nr_to_write = LONG_MAX, + .range_start = start, + .range_end = end, + }; + + if (!mapping_cap_writeback_dirty(mapping)) + return 0; + + ret = do_writepages(mapping, &wbc); + return ret; +} + +static inline int __filemap_fdatawrite(struct address_space *mapping, + int sync_mode) +{ + return __filemap_fdatawrite_range(mapping, 0, LLONG_MAX, sync_mode); +} + +int filemap_fdatawrite(struct address_space *mapping) +{ + return __filemap_fdatawrite(mapping, WB_SYNC_ALL); +} +EXPORT_SYMBOL(filemap_fdatawrite); + +int filemap_fdatawrite_range(struct address_space *mapping, loff_t start, + loff_t end) +{ + return __filemap_fdatawrite_range(mapping, start, end, WB_SYNC_ALL); +} +EXPORT_SYMBOL(filemap_fdatawrite_range); + +/** + * filemap_flush - mostly a non-blocking flush + * @mapping: target address_space + * + * This is a mostly non-blocking flush. Not suitable for data-integrity + * purposes - I/O may not be started against all dirty pages. + */ +int filemap_flush(struct address_space *mapping) +{ + return __filemap_fdatawrite(mapping, WB_SYNC_NONE); +} +EXPORT_SYMBOL(filemap_flush); + +/** + * filemap_fdatawait_range - wait for writeback to complete + * @mapping: address space structure to wait for + * @start_byte: offset in bytes where the range starts + * @end_byte: offset in bytes where the range ends (inclusive) + * + * Walk the list of under-writeback pages of the given address space + * in the given range and wait for all of them. + */ +int filemap_fdatawait_range(struct address_space *mapping, loff_t start_byte, + loff_t end_byte) +{ + pgoff_t index = start_byte >> PAGE_CACHE_SHIFT; + pgoff_t end = end_byte >> PAGE_CACHE_SHIFT; + struct pagevec pvec; + int nr_pages; + int ret2, ret = 0; + + if (end_byte < start_byte) + goto out; + + pagevec_init(&pvec, 0); + while ((index <= end) && + (nr_pages = pagevec_lookup_tag(&pvec, mapping, &index, + PAGECACHE_TAG_WRITEBACK, + min(end - index, (pgoff_t)PAGEVEC_SIZE-1) + 1)) != 0) { + unsigned i; + + for (i = 0; i < nr_pages; i++) { + struct page *page = pvec.pages[i]; + + /* until radix tree lookup accepts end_index */ + if (page->index > end) + continue; + + wait_on_page_writeback(page); + if (TestClearPageError(page)) + ret = -EIO; + } + pagevec_release(&pvec); + cond_resched(); + } +out: + ret2 = filemap_check_errors(mapping); + if (!ret) + ret = ret2; + + return ret; +} +EXPORT_SYMBOL(filemap_fdatawait_range); + +/** + * filemap_fdatawait - wait for all under-writeback pages to complete + * @mapping: address space structure to wait for + * + * Walk the list of under-writeback pages of the given address space + * and wait for all of them. + */ +int filemap_fdatawait(struct address_space *mapping) +{ + loff_t i_size = i_size_read(mapping->host); + + if (i_size == 0) + return 0; + + return filemap_fdatawait_range(mapping, 0, i_size - 1); +} +EXPORT_SYMBOL(filemap_fdatawait); + +int filemap_write_and_wait(struct address_space *mapping) +{ + int err = 0; + + if (mapping->nrpages) { + err = filemap_fdatawrite(mapping); + /* + * Even if the above returned error, the pages may be + * written partially (e.g. -ENOSPC), so we wait for it. + * But the -EIO is special case, it may indicate the worst + * thing (e.g. bug) happened, so we avoid waiting for it. + */ + if (err != -EIO) { + int err2 = filemap_fdatawait(mapping); + if (!err) + err = err2; + } + } else { + err = filemap_check_errors(mapping); + } + return err; +} +EXPORT_SYMBOL(filemap_write_and_wait); + +/** + * filemap_write_and_wait_range - write out & wait on a file range + * @mapping: the address_space for the pages + * @lstart: offset in bytes where the range starts + * @lend: offset in bytes where the range ends (inclusive) + * + * Write out and wait upon file offsets lstart->lend, inclusive. + * + * Note that `lend' is inclusive (describes the last byte to be written) so + * that this function can be used to write to the very end-of-file (end = -1). + */ +int filemap_write_and_wait_range(struct address_space *mapping, + loff_t lstart, loff_t lend) +{ + int err = 0; + + if (mapping->nrpages) { + err = __filemap_fdatawrite_range(mapping, lstart, lend, + WB_SYNC_ALL); + /* See comment of filemap_write_and_wait() */ + if (err != -EIO) { + int err2 = filemap_fdatawait_range(mapping, + lstart, lend); + if (!err) + err = err2; + } + } else { + err = filemap_check_errors(mapping); + } + return err; +} +EXPORT_SYMBOL(filemap_write_and_wait_range); + +/** + * replace_page_cache_page - replace a pagecache page with a new one + * @old: page to be replaced + * @new: page to replace with + * @gfp_mask: allocation mode + * + * This function replaces a page in the pagecache with a new one. On + * success it acquires the pagecache reference for the new page and + * drops it for the old page. Both the old and new pages must be + * locked. This function does not add the new page to the LRU, the + * caller must do that. + * + * The remove + add is atomic. The only way this function can fail is + * memory allocation failure. + */ +int replace_page_cache_page(struct page *old, struct page *new, gfp_t gfp_mask) +{ + int error; + + VM_BUG_ON(!PageLocked(old)); + VM_BUG_ON(!PageLocked(new)); + VM_BUG_ON(new->mapping); + + error = radix_tree_preload(gfp_mask & ~__GFP_HIGHMEM); + if (!error) { + struct address_space *mapping = old->mapping; + void (*freepage)(struct page *); + + pgoff_t offset = old->index; + freepage = mapping->a_ops->freepage; + + page_cache_get(new); + new->mapping = mapping; + new->index = offset; + + spin_lock_irq(&mapping->tree_lock); + __delete_from_page_cache(old); + error = radix_tree_insert(&mapping->page_tree, offset, new); + BUG_ON(error); + mapping->nrpages++; + __inc_zone_page_state(new, NR_FILE_PAGES); + if (PageSwapBacked(new)) + __inc_zone_page_state(new, NR_SHMEM); + spin_unlock_irq(&mapping->tree_lock); + /* mem_cgroup codes must not be called under tree_lock */ + mem_cgroup_replace_page_cache(old, new); + radix_tree_preload_end(); + if (freepage) + freepage(old); + page_cache_release(old); + } + + return error; +} +EXPORT_SYMBOL_GPL(replace_page_cache_page); + +/** + * add_to_page_cache_locked - add a locked page to the pagecache + * @page: page to add + * @mapping: the page's address_space + * @offset: page index + * @gfp_mask: page allocation mode + * + * This function is used to add a page to the pagecache. It must be locked. + * This function does not add the page to the LRU. The caller must do that. + */ +int add_to_page_cache_locked(struct page *page, struct address_space *mapping, + pgoff_t offset, gfp_t gfp_mask) +{ + int error; + + VM_BUG_ON(!PageLocked(page)); + VM_BUG_ON(PageSwapBacked(page)); + + error = mem_cgroup_cache_charge(page, current->mm, + gfp_mask & GFP_RECLAIM_MASK); + if (error) + goto out; + + error = radix_tree_preload(gfp_mask & ~__GFP_HIGHMEM); + if (error == 0) { + page_cache_get(page); + page->mapping = mapping; + page->index = offset; + + spin_lock_irq(&mapping->tree_lock); + error = radix_tree_insert(&mapping->page_tree, offset, page); + if (likely(!error)) { + mapping->nrpages++; + __inc_zone_page_state(page, NR_FILE_PAGES); + spin_unlock_irq(&mapping->tree_lock); + trace_mm_filemap_add_to_page_cache(page); + } else { + page->mapping = NULL; + /* Leave page->index set: truncation relies upon it */ + spin_unlock_irq(&mapping->tree_lock); + mem_cgroup_uncharge_cache_page(page); + page_cache_release(page); + } + radix_tree_preload_end(); + } else + mem_cgroup_uncharge_cache_page(page); +out: + return error; +} +EXPORT_SYMBOL(add_to_page_cache_locked); + +int add_to_page_cache_lru(struct page *page, struct address_space *mapping, + pgoff_t offset, gfp_t gfp_mask) +{ + int ret; + + ret = add_to_page_cache(page, mapping, offset, gfp_mask); + if (ret == 0) + lru_cache_add_file(page); + return ret; +} +EXPORT_SYMBOL_GPL(add_to_page_cache_lru); + +#ifdef CONFIG_NUMA +struct page *__page_cache_alloc(gfp_t gfp) +{ + int n; + struct page *page; + + if (cpuset_do_page_mem_spread()) { + unsigned int cpuset_mems_cookie; + do { + cpuset_mems_cookie = get_mems_allowed(); + n = cpuset_mem_spread_node(); + page = alloc_pages_exact_node(n, gfp, 0); + } while (!put_mems_allowed(cpuset_mems_cookie) && !page); + + return page; + } + return alloc_pages(gfp, 0); +} +EXPORT_SYMBOL(__page_cache_alloc); +#endif + +/* + * In order to wait for pages to become available there must be + * waitqueues associated with pages. By using a hash table of + * waitqueues where the bucket discipline is to maintain all + * waiters on the same queue and wake all when any of the pages + * become available, and for the woken contexts to check to be + * sure the appropriate page became available, this saves space + * at a cost of "thundering herd" phenomena during rare hash + * collisions. + */ +static wait_queue_head_t *page_waitqueue(struct page *page) +{ + const struct zone *zone = page_zone(page); + + return &zone->wait_table[hash_ptr(page, zone->wait_table_bits)]; +} + +static inline void wake_up_page(struct page *page, int bit) +{ + __wake_up_bit(page_waitqueue(page), &page->flags, bit); +} + +void wait_on_page_bit(struct page *page, int bit_nr) +{ + DEFINE_WAIT_BIT(wait, &page->flags, bit_nr); + + if (test_bit(bit_nr, &page->flags)) + __wait_on_bit(page_waitqueue(page), &wait, sleep_on_page, + TASK_UNINTERRUPTIBLE); +} +EXPORT_SYMBOL(wait_on_page_bit); + +int wait_on_page_bit_killable(struct page *page, int bit_nr) +{ + DEFINE_WAIT_BIT(wait, &page->flags, bit_nr); + + if (!test_bit(bit_nr, &page->flags)) + return 0; + + return __wait_on_bit(page_waitqueue(page), &wait, + sleep_on_page_killable, TASK_KILLABLE); +} + +/** + * add_page_wait_queue - Add an arbitrary waiter to a page's wait queue + * @page: Page defining the wait queue of interest + * @waiter: Waiter to add to the queue + * + * Add an arbitrary @waiter to the wait queue for the nominated @page. + */ +void add_page_wait_queue(struct page *page, wait_queue_t *waiter) +{ + wait_queue_head_t *q = page_waitqueue(page); + unsigned long flags; + + spin_lock_irqsave(&q->lock, flags); + __add_wait_queue(q, waiter); + spin_unlock_irqrestore(&q->lock, flags); +} +EXPORT_SYMBOL_GPL(add_page_wait_queue); + +/** + * unlock_page - unlock a locked page + * @page: the page + * + * Unlocks the page and wakes up sleepers in ___wait_on_page_locked(). + * Also wakes sleepers in wait_on_page_writeback() because the wakeup + * mechananism between PageLocked pages and PageWriteback pages is shared. + * But that's OK - sleepers in wait_on_page_writeback() just go back to sleep. + * + * The mb is necessary to enforce ordering between the clear_bit and the read + * of the waitqueue (to avoid SMP races with a parallel wait_on_page_locked()). + */ +void unlock_page(struct page *page) +{ + VM_BUG_ON(!PageLocked(page)); + clear_bit_unlock(PG_locked, &page->flags); + smp_mb__after_clear_bit(); + wake_up_page(page, PG_locked); +} +EXPORT_SYMBOL(unlock_page); + +/** + * end_page_writeback - end writeback against a page + * @page: the page + */ +void end_page_writeback(struct page *page) +{ + if (TestClearPageReclaim(page)) + rotate_reclaimable_page(page); + + if (!test_clear_page_writeback(page)) + BUG(); + + smp_mb__after_clear_bit(); + wake_up_page(page, PG_writeback); +} +EXPORT_SYMBOL(end_page_writeback); + +/** + * __lock_page - get a lock on the page, assuming we need to sleep to get it + * @page: the page to lock + */ +void __lock_page(struct page *page) +{ + DEFINE_WAIT_BIT(wait, &page->flags, PG_locked); + + __wait_on_bit_lock(page_waitqueue(page), &wait, sleep_on_page, + TASK_UNINTERRUPTIBLE); +} +EXPORT_SYMBOL(__lock_page); + +int __lock_page_killable(struct page *page) +{ + DEFINE_WAIT_BIT(wait, &page->flags, PG_locked); + + return __wait_on_bit_lock(page_waitqueue(page), &wait, + sleep_on_page_killable, TASK_KILLABLE); +} +EXPORT_SYMBOL_GPL(__lock_page_killable); + +int __lock_page_or_retry(struct page *page, struct mm_struct *mm, + unsigned int flags) +{ + if (flags & FAULT_FLAG_ALLOW_RETRY) { + /* + * CAUTION! In this case, mmap_sem is not released + * even though return 0. + */ + if (flags & FAULT_FLAG_RETRY_NOWAIT) + return 0; + + up_read(&mm->mmap_sem); + if (flags & FAULT_FLAG_KILLABLE) + wait_on_page_locked_killable(page); + else + wait_on_page_locked(page); + return 0; + } else { + if (flags & FAULT_FLAG_KILLABLE) { + int ret; + + ret = __lock_page_killable(page); + if (ret) { + up_read(&mm->mmap_sem); + return 0; + } + } else + __lock_page(page); + return 1; + } +} + +/** + * find_get_page - find and get a page reference + * @mapping: the address_space to search + * @offset: the page index + * + * Is there a pagecache struct page at the given (mapping, offset) tuple? + * If yes, increment its refcount and return it; if no, return NULL. + */ +struct page *find_get_page(struct address_space *mapping, pgoff_t offset) +{ + void **pagep; + struct page *page; + + rcu_read_lock(); +repeat: + page = NULL; + pagep = radix_tree_lookup_slot(&mapping->page_tree, offset); + if (pagep) { + page = radix_tree_deref_slot(pagep); + if (unlikely(!page)) + goto out; + if (radix_tree_exception(page)) { + if (radix_tree_deref_retry(page)) + goto repeat; + /* + * Otherwise, shmem/tmpfs must be storing a swap entry + * here as an exceptional entry: so return it without + * attempting to raise page count. + */ + goto out; + } + if (!page_cache_get_speculative(page)) + goto repeat; + + /* + * Has the page moved? + * This is part of the lockless pagecache protocol. See + * include/linux/pagemap.h for details. + */ + if (unlikely(page != *pagep)) { + page_cache_release(page); + goto repeat; + } + } +out: + rcu_read_unlock(); + + return page; +} +EXPORT_SYMBOL(find_get_page); + +/** + * find_lock_page - locate, pin and lock a pagecache page + * @mapping: the address_space to search + * @offset: the page index + * + * Locates the desired pagecache page, locks it, increments its reference + * count and returns its address. + * + * Returns zero if the page was not present. find_lock_page() may sleep. + */ +struct page *find_lock_page(struct address_space *mapping, pgoff_t offset) +{ + struct page *page; + +repeat: + page = find_get_page(mapping, offset); + if (page && !radix_tree_exception(page)) { + lock_page(page); + /* Has the page been truncated? */ + if (unlikely(page->mapping != mapping)) { + unlock_page(page); + page_cache_release(page); + goto repeat; + } + VM_BUG_ON(page->index != offset); + } + return page; +} +EXPORT_SYMBOL(find_lock_page); + +/** + * find_or_create_page - locate or add a pagecache page + * @mapping: the page's address_space + * @index: the page's index into the mapping + * @gfp_mask: page allocation mode + * + * Locates a page in the pagecache. If the page is not present, a new page + * is allocated using @gfp_mask and is added to the pagecache and to the VM's + * LRU list. The returned page is locked and has its reference count + * incremented. + * + * find_or_create_page() may sleep, even if @gfp_flags specifies an atomic + * allocation! + * + * find_or_create_page() returns the desired page's address, or zero on + * memory exhaustion. + */ +struct page *find_or_create_page(struct address_space *mapping, + pgoff_t index, gfp_t gfp_mask) +{ + struct page *page; + int err; +repeat: + page = find_lock_page(mapping, index); + if (!page) { + page = __page_cache_alloc(gfp_mask); + if (!page) + return NULL; + /* + * We want a regular kernel memory (not highmem or DMA etc) + * allocation for the radix tree nodes, but we need to honour + * the context-specific requirements the caller has asked for. + * GFP_RECLAIM_MASK collects those requirements. + */ + err = add_to_page_cache_lru(page, mapping, index, + (gfp_mask & GFP_RECLAIM_MASK)); + if (unlikely(err)) { + page_cache_release(page); + page = NULL; + if (err == -EEXIST) + goto repeat; + } + } + return page; +} +EXPORT_SYMBOL(find_or_create_page); + +/** + * find_get_pages - gang pagecache lookup + * @mapping: The address_space to search + * @start: The starting page index + * @nr_pages: The maximum number of pages + * @pages: Where the resulting pages are placed + * + * find_get_pages() will search for and return a group of up to + * @nr_pages pages in the mapping. The pages are placed at @pages. + * find_get_pages() takes a reference against the returned pages. + * + * The search returns a group of mapping-contiguous pages with ascending + * indexes. There may be holes in the indices due to not-present pages. + * + * find_get_pages() returns the number of pages which were found. + */ +unsigned find_get_pages(struct address_space *mapping, pgoff_t start, + unsigned int nr_pages, struct page **pages) +{ + struct radix_tree_iter iter; + void **slot; + unsigned ret = 0; + + if (unlikely(!nr_pages)) + return 0; + + rcu_read_lock(); +restart: + radix_tree_for_each_slot(slot, &mapping->page_tree, &iter, start) { + struct page *page; +repeat: + page = radix_tree_deref_slot(slot); + if (unlikely(!page)) + continue; + + if (radix_tree_exception(page)) { + if (radix_tree_deref_retry(page)) { + /* + * Transient condition which can only trigger + * when entry at index 0 moves out of or back + * to root: none yet gotten, safe to restart. + */ + WARN_ON(iter.index); + goto restart; + } + /* + * Otherwise, shmem/tmpfs must be storing a swap entry + * here as an exceptional entry: so skip over it - + * we only reach this from invalidate_mapping_pages(). + */ + continue; + } + + if (!page_cache_get_speculative(page)) + goto repeat; + + /* Has the page moved? */ + if (unlikely(page != *slot)) { + page_cache_release(page); + goto repeat; + } + + pages[ret] = page; + if (++ret == nr_pages) + break; + } + + rcu_read_unlock(); + return ret; +} + +/** + * find_get_pages_contig - gang contiguous pagecache lookup + * @mapping: The address_space to search + * @index: The starting page index + * @nr_pages: The maximum number of pages + * @pages: Where the resulting pages are placed + * + * find_get_pages_contig() works exactly like find_get_pages(), except + * that the returned number of pages are guaranteed to be contiguous. + * + * find_get_pages_contig() returns the number of pages which were found. + */ +unsigned find_get_pages_contig(struct address_space *mapping, pgoff_t index, + unsigned int nr_pages, struct page **pages) +{ + struct radix_tree_iter iter; + void **slot; + unsigned int ret = 0; + + if (unlikely(!nr_pages)) + return 0; + + rcu_read_lock(); +restart: + radix_tree_for_each_contig(slot, &mapping->page_tree, &iter, index) { + struct page *page; +repeat: + page = radix_tree_deref_slot(slot); + /* The hole, there no reason to continue */ + if (unlikely(!page)) + break; + + if (radix_tree_exception(page)) { + if (radix_tree_deref_retry(page)) { + /* + * Transient condition which can only trigger + * when entry at index 0 moves out of or back + * to root: none yet gotten, safe to restart. + */ + goto restart; + } + /* + * Otherwise, shmem/tmpfs must be storing a swap entry + * here as an exceptional entry: so stop looking for + * contiguous pages. + */ + break; + } + + if (!page_cache_get_speculative(page)) + goto repeat; + + /* Has the page moved? */ + if (unlikely(page != *slot)) { + page_cache_release(page); + goto repeat; + } + + /* + * must check mapping and index after taking the ref. + * otherwise we can get both false positives and false + * negatives, which is just confusing to the caller. + */ + if (page->mapping == NULL || page->index != iter.index) { + page_cache_release(page); + break; + } + + pages[ret] = page; + if (++ret == nr_pages) + break; + } + rcu_read_unlock(); + return ret; +} +EXPORT_SYMBOL(find_get_pages_contig); + +/** + * find_get_pages_tag - find and return pages that match @tag + * @mapping: the address_space to search + * @index: the starting page index + * @tag: the tag index + * @nr_pages: the maximum number of pages + * @pages: where the resulting pages are placed + * + * Like find_get_pages, except we only return pages which are tagged with + * @tag. We update @index to index the next page for the traversal. + */ +unsigned find_get_pages_tag(struct address_space *mapping, pgoff_t *index, + int tag, unsigned int nr_pages, struct page **pages) +{ + struct radix_tree_iter iter; + void **slot; + unsigned ret = 0; + + if (unlikely(!nr_pages)) + return 0; + + rcu_read_lock(); +restart: + radix_tree_for_each_tagged(slot, &mapping->page_tree, + &iter, *index, tag) { + struct page *page; +repeat: + page = radix_tree_deref_slot(slot); + if (unlikely(!page)) + continue; + + if (radix_tree_exception(page)) { + if (radix_tree_deref_retry(page)) { + /* + * Transient condition which can only trigger + * when entry at index 0 moves out of or back + * to root: none yet gotten, safe to restart. + */ + goto restart; + } + /* + * This function is never used on a shmem/tmpfs + * mapping, so a swap entry won't be found here. + */ + BUG(); + } + + if (!page_cache_get_speculative(page)) + goto repeat; + + /* Has the page moved? */ + if (unlikely(page != *slot)) { + page_cache_release(page); + goto repeat; + } + + pages[ret] = page; + if (++ret == nr_pages) + break; + } + + rcu_read_unlock(); + + if (ret) + *index = pages[ret - 1]->index + 1; + + return ret; +} +EXPORT_SYMBOL(find_get_pages_tag); + +/** + * grab_cache_page_nowait - returns locked page at given index in given cache + * @mapping: target address_space + * @index: the page index + * + * Same as grab_cache_page(), but do not wait if the page is unavailable. + * This is intended for speculative data generators, where the data can + * be regenerated if the page couldn't be grabbed. This routine should + * be safe to call while holding the lock for another page. + * + * Clear __GFP_FS when allocating the page to avoid recursion into the fs + * and deadlock against the caller's locked page. + */ +struct page * +grab_cache_page_nowait(struct address_space *mapping, pgoff_t index) +{ + struct page *page = find_get_page(mapping, index); + + if (page) { + if (trylock_page(page)) + return page; + page_cache_release(page); + return NULL; + } + page = __page_cache_alloc(mapping_gfp_mask(mapping) & ~__GFP_FS); + if (page && add_to_page_cache_lru(page, mapping, index, GFP_NOFS)) { + page_cache_release(page); + page = NULL; + } + return page; +} +EXPORT_SYMBOL(grab_cache_page_nowait); + +/* + * CD/DVDs are error prone. When a medium error occurs, the driver may fail + * a _large_ part of the i/o request. Imagine the worst scenario: + * + * ---R__________________________________________B__________ + * ^ reading here ^ bad block(assume 4k) + * + * read(R) => miss => readahead(R...B) => media error => frustrating retries + * => failing the whole request => read(R) => read(R+1) => + * readahead(R+1...B+1) => bang => read(R+2) => read(R+3) => + * readahead(R+3...B+2) => bang => read(R+3) => read(R+4) => + * readahead(R+4...B+3) => bang => read(R+4) => read(R+5) => ...... + * + * It is going insane. Fix it by quickly scaling down the readahead size. + */ +static void shrink_readahead_size_eio(struct file *filp, + struct file_ra_state *ra) +{ + ra->ra_pages /= 4; +} + +/** + * do_generic_file_read - generic file read routine + * @filp: the file to read + * @ppos: current file position + * @desc: read_descriptor + * @actor: read method + * + * This is a generic file read routine, and uses the + * mapping->a_ops->readpage() function for the actual low-level stuff. + * + * This is really ugly. But the goto's actually try to clarify some + * of the logic when it comes to error handling etc. + */ +static void do_generic_file_read(struct file *filp, loff_t *ppos, + read_descriptor_t *desc, read_actor_t actor) +{ + struct address_space *mapping = filp->f_mapping; + struct inode *inode = mapping->host; + struct file_ra_state *ra = &filp->f_ra; + pgoff_t index; + pgoff_t last_index; + pgoff_t prev_index; + unsigned long offset; /* offset into pagecache page */ + unsigned int prev_offset; + int error; + + index = *ppos >> PAGE_CACHE_SHIFT; + prev_index = ra->prev_pos >> PAGE_CACHE_SHIFT; + prev_offset = ra->prev_pos & (PAGE_CACHE_SIZE-1); + last_index = (*ppos + desc->count + PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT; + offset = *ppos & ~PAGE_CACHE_MASK; + + for (;;) { + struct page *page; + pgoff_t end_index; + loff_t isize; + unsigned long nr, ret; + + cond_resched(); +find_page: + page = find_get_page(mapping, index); + if (!page) { + page_cache_sync_readahead(mapping, + ra, filp, + index, last_index - index); + page = find_get_page(mapping, index); + if (unlikely(page == NULL)) + goto no_cached_page; + } + if (PageReadahead(page)) { + page_cache_async_readahead(mapping, + ra, filp, page, + index, last_index - index); + } + if (!PageUptodate(page)) { + if (inode->i_blkbits == PAGE_CACHE_SHIFT || + !mapping->a_ops->is_partially_uptodate) + goto page_not_up_to_date; + if (!trylock_page(page)) + goto page_not_up_to_date; + /* Did it get truncated before we got the lock? */ + if (!page->mapping) + goto page_not_up_to_date_locked; + if (!mapping->a_ops->is_partially_uptodate(page, + desc, offset)) + goto page_not_up_to_date_locked; + unlock_page(page); + } +page_ok: + /* + * i_size must be checked after we know the page is Uptodate. + * + * Checking i_size after the check allows us to calculate + * the correct value for "nr", which means the zero-filled + * part of the page is not copied back to userspace (unless + * another truncate extends the file - this is desired though). + */ + + isize = i_size_read(inode); + end_index = (isize - 1) >> PAGE_CACHE_SHIFT; + if (unlikely(!isize || index > end_index)) { + page_cache_release(page); + goto out; + } + + /* nr is the maximum number of bytes to copy from this page */ + nr = PAGE_CACHE_SIZE; + if (index == end_index) { + nr = ((isize - 1) & ~PAGE_CACHE_MASK) + 1; + if (nr <= offset) { + page_cache_release(page); + goto out; + } + } + nr = nr - offset; + + /* If users can be writing to this page using arbitrary + * virtual addresses, take care about potential aliasing + * before reading the page on the kernel side. + */ + if (mapping_writably_mapped(mapping)) + flush_dcache_page(page); + + /* + * When a sequential read accesses a page several times, + * only mark it as accessed the first time. + */ + if (prev_index != index || offset != prev_offset) + mark_page_accessed(page); + prev_index = index; + + /* + * Ok, we have the page, and it's up-to-date, so + * now we can copy it to user space... + * + * The actor routine returns how many bytes were actually used.. + * NOTE! This may not be the same as how much of a user buffer + * we filled up (we may be padding etc), so we can only update + * "pos" here (the actor routine has to update the user buffer + * pointers and the remaining count). + */ + ret = actor(desc, page, offset, nr); + offset += ret; + index += offset >> PAGE_CACHE_SHIFT; + offset &= ~PAGE_CACHE_MASK; + prev_offset = offset; + + page_cache_release(page); + if (ret == nr && desc->count) + continue; + goto out; + +page_not_up_to_date: + /* Get exclusive access to the page ... */ + error = lock_page_killable(page); + if (unlikely(error)) + goto readpage_error; + +page_not_up_to_date_locked: + /* Did it get truncated before we got the lock? */ + if (!page->mapping) { + unlock_page(page); + page_cache_release(page); + continue; + } + + /* Did somebody else fill it already? */ + if (PageUptodate(page)) { + unlock_page(page); + goto page_ok; + } + +readpage: + /* + * A previous I/O error may have been due to temporary + * failures, eg. multipath errors. + * PG_error will be set again if readpage fails. + */ + ClearPageError(page); + /* Start the actual read. The read will unlock the page. */ + error = mapping->a_ops->readpage(filp, page); + + if (unlikely(error)) { + if (error == AOP_TRUNCATED_PAGE) { + page_cache_release(page); + goto find_page; + } + goto readpage_error; + } + + if (!PageUptodate(page)) { + error = lock_page_killable(page); + if (unlikely(error)) + goto readpage_error; + if (!PageUptodate(page)) { + if (page->mapping == NULL) { + /* + * invalidate_mapping_pages got it + */ + unlock_page(page); + page_cache_release(page); + goto find_page; + } + unlock_page(page); + shrink_readahead_size_eio(filp, ra); + error = -EIO; + goto readpage_error; + } + unlock_page(page); + } + + goto page_ok; + +readpage_error: + /* UHHUH! A synchronous read error occurred. Report it */ + desc->error = error; + page_cache_release(page); + goto out; + +no_cached_page: + /* + * Ok, it wasn't cached, so we need to create a new + * page.. + */ + page = page_cache_alloc_cold(mapping); + if (!page) { + desc->error = -ENOMEM; + goto out; + } + error = add_to_page_cache_lru(page, mapping, + index, GFP_KERNEL); + if (error) { + page_cache_release(page); + if (error == -EEXIST) + goto find_page; + desc->error = error; + goto out; + } + goto readpage; + } + +out: + ra->prev_pos = prev_index; + ra->prev_pos <<= PAGE_CACHE_SHIFT; + ra->prev_pos |= prev_offset; + + *ppos = ((loff_t)index << PAGE_CACHE_SHIFT) + offset; + file_accessed(filp); +} + +int file_read_actor(read_descriptor_t *desc, struct page *page, + unsigned long offset, unsigned long size) +{ + char *kaddr; + unsigned long left, count = desc->count; + + if (size > count) + size = count; + + /* + * Faults on the destination of a read are common, so do it before + * taking the kmap. + */ + if (!fault_in_pages_writeable(desc->arg.buf, size)) { + kaddr = kmap_atomic(page); + left = __copy_to_user_inatomic(desc->arg.buf, + kaddr + offset, size); + kunmap_atomic(kaddr); + if (left == 0) + goto success; + } + + /* Do it the slow way */ + kaddr = kmap(page); + left = __copy_to_user(desc->arg.buf, kaddr + offset, size); + kunmap(page); + + if (left) { + size -= left; + desc->error = -EFAULT; + } +success: + desc->count = count - size; + desc->written += size; + desc->arg.buf += size; + return size; +} + +/* + * Performs necessary checks before doing a write + * @iov: io vector request + * @nr_segs: number of segments in the iovec + * @count: number of bytes to write + * @access_flags: type of access: %VERIFY_READ or %VERIFY_WRITE + * + * Adjust number of segments and amount of bytes to write (nr_segs should be + * properly initialized first). Returns appropriate error code that caller + * should return or zero in case that write should be allowed. + */ +int generic_segment_checks(const struct iovec *iov, + unsigned long *nr_segs, size_t *count, int access_flags) +{ + unsigned long seg; + size_t cnt = 0; + for (seg = 0; seg < *nr_segs; seg++) { + const struct iovec *iv = &iov[seg]; + + /* + * If any segment has a negative length, or the cumulative + * length ever wraps negative then return -EINVAL. + */ + cnt += iv->iov_len; + if (unlikely((ssize_t)(cnt|iv->iov_len) < 0)) + return -EINVAL; + if (access_ok(access_flags, iv->iov_base, iv->iov_len)) + continue; + if (seg == 0) + return -EFAULT; + *nr_segs = seg; + cnt -= iv->iov_len; /* This segment is no good */ + break; + } + *count = cnt; + return 0; +} +EXPORT_SYMBOL(generic_segment_checks); + +/** + * generic_file_aio_read - generic filesystem read routine + * @iocb: kernel I/O control block + * @iov: io vector request + * @nr_segs: number of segments in the iovec + * @pos: current file position + * + * This is the "read()" routine for all filesystems + * that can use the page cache directly. + */ +ssize_t +generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov, + unsigned long nr_segs, loff_t pos) +{ + struct file *filp = iocb->ki_filp; + ssize_t retval; + unsigned long seg = 0; + size_t count; + loff_t *ppos = &iocb->ki_pos; + + count = 0; + retval = generic_segment_checks(iov, &nr_segs, &count, VERIFY_WRITE); + if (retval) + return retval; + + /* coalesce the iovecs and go direct-to-BIO for O_DIRECT */ + if (filp->f_flags & O_DIRECT) { + loff_t size; + struct address_space *mapping; + struct inode *inode; + + mapping = filp->f_mapping; + inode = mapping->host; + if (!count) + goto out; /* skip atime */ + size = i_size_read(inode); + if (pos < size) { + retval = filemap_write_and_wait_range(mapping, pos, + pos + iov_length(iov, nr_segs) - 1); + if (!retval) { + retval = mapping->a_ops->direct_IO(READ, iocb, + iov, pos, nr_segs); + } + if (retval > 0) { + *ppos = pos + retval; + count -= retval; + } + + /* + * Btrfs can have a short DIO read if we encounter + * compressed extents, so if there was an error, or if + * we've already read everything we wanted to, or if + * there was a short read because we hit EOF, go ahead + * and return. Otherwise fallthrough to buffered io for + * the rest of the read. + */ + if (retval < 0 || !count || *ppos >= size) { + file_accessed(filp); + goto out; + } + } + } + + count = retval; + for (seg = 0; seg < nr_segs; seg++) { + read_descriptor_t desc; + loff_t offset = 0; + + /* + * If we did a short DIO read we need to skip the section of the + * iov that we've already read data into. + */ + if (count) { + if (count > iov[seg].iov_len) { + count -= iov[seg].iov_len; + continue; + } + offset = count; + count = 0; + } + + desc.written = 0; + desc.arg.buf = iov[seg].iov_base + offset; + desc.count = iov[seg].iov_len - offset; + if (desc.count == 0) + continue; + desc.error = 0; + do_generic_file_read(filp, ppos, &desc, file_read_actor); + retval += desc.written; + if (desc.error) { + retval = retval ?: desc.error; + break; + } + if (desc.count > 0) + break; + } +out: + return retval; +} +EXPORT_SYMBOL(generic_file_aio_read); + +#ifdef CONFIG_MMU +/** + * page_cache_read - adds requested page to the page cache if not already there + * @file: file to read + * @offset: page index + * + * This adds the requested page to the page cache if it isn't already there, + * and schedules an I/O to read in its contents from disk. + */ +static int page_cache_read(struct file *file, pgoff_t offset) +{ + struct address_space *mapping = file->f_mapping; + struct page *page; + int ret; + + do { + page = page_cache_alloc_cold(mapping); + if (!page) + return -ENOMEM; + + ret = add_to_page_cache_lru(page, mapping, offset, GFP_KERNEL); + if (ret == 0) + ret = mapping->a_ops->readpage(file, page); + else if (ret == -EEXIST) + ret = 0; /* losing race to add is OK */ + + page_cache_release(page); + + } while (ret == AOP_TRUNCATED_PAGE); + + return ret; +} + +#define MMAP_LOTSAMISS (100) + +/* + * Synchronous readahead happens when we don't even find + * a page in the page cache at all. + */ +static void do_sync_mmap_readahead(struct vm_area_struct *vma, + struct file_ra_state *ra, + struct file *file, + pgoff_t offset) +{ + unsigned long ra_pages; + struct address_space *mapping = file->f_mapping; + + /* If we don't want any read-ahead, don't bother */ + if (VM_RandomReadHint(vma)) + return; + if (!ra->ra_pages) + return; + + if (VM_SequentialReadHint(vma)) { + page_cache_sync_readahead(mapping, ra, file, offset, + ra->ra_pages); + return; + } + + /* Avoid banging the cache line if not needed */ + if (ra->mmap_miss < MMAP_LOTSAMISS * 10) + ra->mmap_miss++; + + /* + * Do we miss much more than hit in this file? If so, + * stop bothering with read-ahead. It will only hurt. + */ + if (ra->mmap_miss > MMAP_LOTSAMISS) + return; + + /* + * mmap read-around + */ + ra_pages = max_sane_readahead(ra->ra_pages); + ra->start = max_t(long, 0, offset - ra_pages / 2); + ra->size = ra_pages; + ra->async_size = ra_pages / 4; + ra_submit(ra, mapping, file); +} + +/* + * Asynchronous readahead happens when we find the page and PG_readahead, + * so we want to possibly extend the readahead further.. + */ +static void do_async_mmap_readahead(struct vm_area_struct *vma, + struct file_ra_state *ra, + struct file *file, + struct page *page, + pgoff_t offset) +{ + struct address_space *mapping = file->f_mapping; + + /* If we don't want any read-ahead, don't bother */ + if (VM_RandomReadHint(vma)) + return; + if (ra->mmap_miss > 0) + ra->mmap_miss--; + if (PageReadahead(page)) + page_cache_async_readahead(mapping, ra, file, + page, offset, ra->ra_pages); +} + +/** + * filemap_fault - read in file data for page fault handling + * @vma: vma in which the fault was taken + * @vmf: struct vm_fault containing details of the fault + * + * filemap_fault() is invoked via the vma operations vector for a + * mapped memory region to read in file data during a page fault. + * + * The goto's are kind of ugly, but this streamlines the normal case of having + * it in the page cache, and handles the special cases reasonably without + * having a lot of duplicated code. + */ +int filemap_fault(struct vm_area_struct *vma, struct vm_fault *vmf) +{ + int error; + struct file *file = vma->vm_file; + struct address_space *mapping = file->f_mapping; + struct file_ra_state *ra = &file->f_ra; + struct inode *inode = mapping->host; + pgoff_t offset = vmf->pgoff; + struct page *page; + bool memcg_oom; + pgoff_t size; + int ret = 0; + + size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + if (offset >= size) + return VM_FAULT_SIGBUS; + + /* + * Do we have something in the page cache already? Either + * way, try readahead, but disable the memcg OOM killer for it + * as readahead is optional and no errors are propagated up + * the fault stack. The OOM killer is enabled while trying to + * instantiate the faulting page individually below. + */ + page = find_get_page(mapping, offset); + if (likely(page) && !(vmf->flags & FAULT_FLAG_TRIED)) { + /* + * We found the page, so try async readahead before + * waiting for the lock. + */ + memcg_oom = mem_cgroup_toggle_oom(false); + do_async_mmap_readahead(vma, ra, file, page, offset); + mem_cgroup_toggle_oom(memcg_oom); + } else if (!page) { + /* No page in the page cache at all */ + memcg_oom = mem_cgroup_toggle_oom(false); + do_sync_mmap_readahead(vma, ra, file, offset); + mem_cgroup_toggle_oom(memcg_oom); + count_vm_event(PGMAJFAULT); + mem_cgroup_count_vm_event(vma->vm_mm, PGMAJFAULT); + ret = VM_FAULT_MAJOR; +retry_find: + page = find_get_page(mapping, offset); + if (!page) + goto no_cached_page; + } + + if (!lock_page_or_retry(page, vma->vm_mm, vmf->flags)) { + page_cache_release(page); + return ret | VM_FAULT_RETRY; + } + + /* Did it get truncated? */ + if (unlikely(page->mapping != mapping)) { + unlock_page(page); + put_page(page); + goto retry_find; + } + VM_BUG_ON(page->index != offset); + + /* + * We have a locked page in the page cache, now we need to check + * that it's up-to-date. If not, it is going to be due to an error. + */ + if (unlikely(!PageUptodate(page))) + goto page_not_uptodate; + + /* + * Found the page and have a reference on it. + * We must recheck i_size under page lock. + */ + size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + if (unlikely(offset >= size)) { + unlock_page(page); + page_cache_release(page); + return VM_FAULT_SIGBUS; + } + + vmf->page = page; + return ret | VM_FAULT_LOCKED; + +no_cached_page: + /* + * We're only likely to ever get here if MADV_RANDOM is in + * effect. + */ + error = page_cache_read(file, offset); + + /* + * The page we want has now been added to the page cache. + * In the unlikely event that someone removed it in the + * meantime, we'll just come back here and read it again. + */ + if (error >= 0) + goto retry_find; + + /* + * An error return from page_cache_read can result if the + * system is low on memory, or a problem occurs while trying + * to schedule I/O. + */ + if (error == -ENOMEM) + return VM_FAULT_OOM; + return VM_FAULT_SIGBUS; + +page_not_uptodate: + /* + * Umm, take care of errors if the page isn't up-to-date. + * Try to re-read it _once_. We do this synchronously, + * because there really aren't any performance issues here + * and we need to check for errors. + */ + ClearPageError(page); + error = mapping->a_ops->readpage(file, page); + if (!error) { + wait_on_page_locked(page); + if (!PageUptodate(page)) + error = -EIO; + } + page_cache_release(page); + + if (!error || error == AOP_TRUNCATED_PAGE) + goto retry_find; + + /* Things didn't work out. Return zero to tell the mm layer so. */ + shrink_readahead_size_eio(file, ra); + return VM_FAULT_SIGBUS; +} +EXPORT_SYMBOL(filemap_fault); + +int filemap_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) +{ + struct page *page = vmf->page; + struct inode *inode = file_inode(vma->vm_file); + int ret = VM_FAULT_LOCKED; + + sb_start_pagefault(inode->i_sb); + file_update_time(vma->vm_file); + lock_page(page); + if (page->mapping != inode->i_mapping) { + unlock_page(page); + ret = VM_FAULT_NOPAGE; + goto out; + } + /* + * We mark the page dirty already here so that when freeze is in + * progress, we are guaranteed that writeback during freezing will + * see the dirty page and writeprotect it again. + */ + set_page_dirty(page); + wait_for_stable_page(page); +out: + sb_end_pagefault(inode->i_sb); + return ret; +} +EXPORT_SYMBOL(filemap_page_mkwrite); + +const struct vm_operations_struct generic_file_vm_ops = { + .fault = filemap_fault, + .page_mkwrite = filemap_page_mkwrite, + .remap_pages = generic_file_remap_pages, +}; + +/* This is used for a general mmap of a disk file */ + +int generic_file_mmap(struct file * file, struct vm_area_struct * vma) +{ + struct address_space *mapping = file->f_mapping; + + if (!mapping->a_ops->readpage) + return -ENOEXEC; + file_accessed(file); + vma->vm_ops = &generic_file_vm_ops; + return 0; +} + +/* + * This is for filesystems which do not implement ->writepage. + */ +int generic_file_readonly_mmap(struct file *file, struct vm_area_struct *vma) +{ + if ((vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_MAYWRITE)) + return -EINVAL; + return generic_file_mmap(file, vma); +} +#else +int generic_file_mmap(struct file * file, struct vm_area_struct * vma) +{ + return -ENOSYS; +} +int generic_file_readonly_mmap(struct file * file, struct vm_area_struct * vma) +{ + return -ENOSYS; +} +#endif /* CONFIG_MMU */ + +EXPORT_SYMBOL(generic_file_mmap); +EXPORT_SYMBOL(generic_file_readonly_mmap); + +static struct page *__read_cache_page(struct address_space *mapping, + pgoff_t index, + int (*filler)(void *, struct page *), + void *data, + gfp_t gfp) +{ + struct page *page; + int err; +repeat: + page = find_get_page(mapping, index); + if (!page) { + page = __page_cache_alloc(gfp | __GFP_COLD); + if (!page) + return ERR_PTR(-ENOMEM); + err = add_to_page_cache_lru(page, mapping, index, gfp); + if (unlikely(err)) { + page_cache_release(page); + if (err == -EEXIST) + goto repeat; + /* Presumably ENOMEM for radix tree node */ + return ERR_PTR(err); + } + err = filler(data, page); + if (err < 0) { + page_cache_release(page); + page = ERR_PTR(err); + } + } + return page; +} + +static struct page *do_read_cache_page(struct address_space *mapping, + pgoff_t index, + int (*filler)(void *, struct page *), + void *data, + gfp_t gfp) + +{ + struct page *page; + int err; + +retry: + page = __read_cache_page(mapping, index, filler, data, gfp); + if (IS_ERR(page)) + return page; + if (PageUptodate(page)) + goto out; + + lock_page(page); + if (!page->mapping) { + unlock_page(page); + page_cache_release(page); + goto retry; + } + if (PageUptodate(page)) { + unlock_page(page); + goto out; + } + err = filler(data, page); + if (err < 0) { + page_cache_release(page); + return ERR_PTR(err); + } +out: + mark_page_accessed(page); + return page; +} + +/** + * read_cache_page_async - read into page cache, fill it if needed + * @mapping: the page's address_space + * @index: the page index + * @filler: function to perform the read + * @data: first arg to filler(data, page) function, often left as NULL + * + * Same as read_cache_page, but don't wait for page to become unlocked + * after submitting it to the filler. + * + * Read into the page cache. If a page already exists, and PageUptodate() is + * not set, try to fill the page but don't wait for it to become unlocked. + * + * If the page does not get brought uptodate, return -EIO. + */ +struct page *read_cache_page_async(struct address_space *mapping, + pgoff_t index, + int (*filler)(void *, struct page *), + void *data) +{ + return do_read_cache_page(mapping, index, filler, data, mapping_gfp_mask(mapping)); +} +EXPORT_SYMBOL(read_cache_page_async); + +static struct page *wait_on_page_read(struct page *page) +{ + if (!IS_ERR(page)) { + wait_on_page_locked(page); + if (!PageUptodate(page)) { + page_cache_release(page); + page = ERR_PTR(-EIO); + } + } + return page; +} + +/** + * read_cache_page_gfp - read into page cache, using specified page allocation flags. + * @mapping: the page's address_space + * @index: the page index + * @gfp: the page allocator flags to use if allocating + * + * This is the same as "read_mapping_page(mapping, index, NULL)", but with + * any new page allocations done using the specified allocation flags. + * + * If the page does not get brought uptodate, return -EIO. + */ +struct page *read_cache_page_gfp(struct address_space *mapping, + pgoff_t index, + gfp_t gfp) +{ + filler_t *filler = (filler_t *)mapping->a_ops->readpage; + + return wait_on_page_read(do_read_cache_page(mapping, index, filler, NULL, gfp)); +} +EXPORT_SYMBOL(read_cache_page_gfp); + +/** + * read_cache_page - read into page cache, fill it if needed + * @mapping: the page's address_space + * @index: the page index + * @filler: function to perform the read + * @data: first arg to filler(data, page) function, often left as NULL + * + * Read into the page cache. If a page already exists, and PageUptodate() is + * not set, try to fill the page then wait for it to become unlocked. + * + * If the page does not get brought uptodate, return -EIO. + */ +struct page *read_cache_page(struct address_space *mapping, + pgoff_t index, + int (*filler)(void *, struct page *), + void *data) +{ + return wait_on_page_read(read_cache_page_async(mapping, index, filler, data)); +} +EXPORT_SYMBOL(read_cache_page); + +static size_t __iovec_copy_from_user_inatomic(char *vaddr, + const struct iovec *iov, size_t base, size_t bytes) +{ + size_t copied = 0, left = 0; + + while (bytes) { + char __user *buf = iov->iov_base + base; + int copy = min(bytes, iov->iov_len - base); + + base = 0; + left = __copy_from_user_inatomic(vaddr, buf, copy); + copied += copy; + bytes -= copy; + vaddr += copy; + iov++; + + if (unlikely(left)) + break; + } + return copied - left; +} + +/* + * Copy as much as we can into the page and return the number of bytes which + * were successfully copied. If a fault is encountered then return the number of + * bytes which were copied. + */ +size_t iov_iter_copy_from_user_atomic(struct page *page, + struct iov_iter *i, unsigned long offset, size_t bytes) +{ + char *kaddr; + size_t copied; + + BUG_ON(!in_atomic()); + kaddr = kmap_atomic(page); + if (likely(i->nr_segs == 1)) { + int left; + char __user *buf = i->iov->iov_base + i->iov_offset; + left = __copy_from_user_inatomic(kaddr + offset, buf, bytes); + copied = bytes - left; + } else { + copied = __iovec_copy_from_user_inatomic(kaddr + offset, + i->iov, i->iov_offset, bytes); + } + kunmap_atomic(kaddr); + + return copied; +} +EXPORT_SYMBOL(iov_iter_copy_from_user_atomic); + +/* + * This has the same sideeffects and return value as + * iov_iter_copy_from_user_atomic(). + * The difference is that it attempts to resolve faults. + * Page must not be locked. + */ +size_t iov_iter_copy_from_user(struct page *page, + struct iov_iter *i, unsigned long offset, size_t bytes) +{ + char *kaddr; + size_t copied; + + kaddr = kmap(page); + if (likely(i->nr_segs == 1)) { + int left; + char __user *buf = i->iov->iov_base + i->iov_offset; + left = __copy_from_user(kaddr + offset, buf, bytes); + copied = bytes - left; + } else { + copied = __iovec_copy_from_user_inatomic(kaddr + offset, + i->iov, i->iov_offset, bytes); + } + kunmap(page); + return copied; +} +EXPORT_SYMBOL(iov_iter_copy_from_user); + +void iov_iter_advance(struct iov_iter *i, size_t bytes) +{ + BUG_ON(i->count < bytes); + + if (likely(i->nr_segs == 1)) { + i->iov_offset += bytes; + i->count -= bytes; + } else { + const struct iovec *iov = i->iov; + size_t base = i->iov_offset; + unsigned long nr_segs = i->nr_segs; + + /* + * The !iov->iov_len check ensures we skip over unlikely + * zero-length segments (without overruning the iovec). + */ + while (bytes || unlikely(i->count && !iov->iov_len)) { + int copy; + + copy = min(bytes, iov->iov_len - base); + BUG_ON(!i->count || i->count < copy); + i->count -= copy; + bytes -= copy; + base += copy; + if (iov->iov_len == base) { + iov++; + nr_segs--; + base = 0; + } + } + i->iov = iov; + i->iov_offset = base; + i->nr_segs = nr_segs; + } +} +EXPORT_SYMBOL(iov_iter_advance); + +/* + * Fault in the first iovec of the given iov_iter, to a maximum length + * of bytes. Returns 0 on success, or non-zero if the memory could not be + * accessed (ie. because it is an invalid address). + * + * writev-intensive code may want this to prefault several iovecs -- that + * would be possible (callers must not rely on the fact that _only_ the + * first iovec will be faulted with the current implementation). + */ +int iov_iter_fault_in_readable(struct iov_iter *i, size_t bytes) +{ + char __user *buf = i->iov->iov_base + i->iov_offset; + bytes = min(bytes, i->iov->iov_len - i->iov_offset); + return fault_in_pages_readable(buf, bytes); +} +EXPORT_SYMBOL(iov_iter_fault_in_readable); + +/* + * Return the count of just the current iov_iter segment. + */ +size_t iov_iter_single_seg_count(const struct iov_iter *i) +{ + const struct iovec *iov = i->iov; + if (i->nr_segs == 1) + return i->count; + else + return min(i->count, iov->iov_len - i->iov_offset); +} +EXPORT_SYMBOL(iov_iter_single_seg_count); + +/* + * Performs necessary checks before doing a write + * + * Can adjust writing position or amount of bytes to write. + * Returns appropriate error code that caller should return or + * zero in case that write should be allowed. + */ +inline int generic_write_checks(struct file *file, loff_t *pos, size_t *count, int isblk) +{ + struct inode *inode = file->f_mapping->host; + unsigned long limit = rlimit(RLIMIT_FSIZE); + + if (unlikely(*pos < 0)) + return -EINVAL; + + if (!isblk) { + /* FIXME: this is for backwards compatibility with 2.4 */ + if (file->f_flags & O_APPEND) + *pos = i_size_read(inode); + + if (limit != RLIM_INFINITY) { + if (*pos >= limit) { + send_sig(SIGXFSZ, current, 0); + return -EFBIG; + } + if (*count > limit - (typeof(limit))*pos) { + *count = limit - (typeof(limit))*pos; + } + } + } + + /* + * LFS rule + */ + if (unlikely(*pos + *count > MAX_NON_LFS && + !(file->f_flags & O_LARGEFILE))) { + if (*pos >= MAX_NON_LFS) { + return -EFBIG; + } + if (*count > MAX_NON_LFS - (unsigned long)*pos) { + *count = MAX_NON_LFS - (unsigned long)*pos; + } + } + + /* + * Are we about to exceed the fs block limit ? + * + * If we have written data it becomes a short write. If we have + * exceeded without writing data we send a signal and return EFBIG. + * Linus frestrict idea will clean these up nicely.. + */ + if (likely(!isblk)) { + if (unlikely(*pos >= inode->i_sb->s_maxbytes)) { + if (*count || *pos > inode->i_sb->s_maxbytes) { + return -EFBIG; + } + /* zero-length writes at ->s_maxbytes are OK */ + } + + if (unlikely(*pos + *count > inode->i_sb->s_maxbytes)) + *count = inode->i_sb->s_maxbytes - *pos; + } else { +#ifdef CONFIG_BLOCK + loff_t isize; + if (bdev_read_only(I_BDEV(inode))) + return -EPERM; + isize = i_size_read(inode); + if (*pos >= isize) { + if (*count || *pos > isize) + return -ENOSPC; + } + + if (*pos + *count > isize) + *count = isize - *pos; +#else + return -EPERM; +#endif + } + return 0; +} +EXPORT_SYMBOL(generic_write_checks); + +int pagecache_write_begin(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned flags, + struct page **pagep, void **fsdata) +{ + const struct address_space_operations *aops = mapping->a_ops; + + return aops->write_begin(file, mapping, pos, len, flags, + pagep, fsdata); +} +EXPORT_SYMBOL(pagecache_write_begin); + +int pagecache_write_end(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata) +{ + const struct address_space_operations *aops = mapping->a_ops; + + mark_page_accessed(page); + return aops->write_end(file, mapping, pos, len, copied, page, fsdata); +} +EXPORT_SYMBOL(pagecache_write_end); + +ssize_t +generic_file_direct_write(struct kiocb *iocb, const struct iovec *iov, + unsigned long *nr_segs, loff_t pos, loff_t *ppos, + size_t count, size_t ocount) +{ + struct file *file = iocb->ki_filp; + struct address_space *mapping = file->f_mapping; + struct inode *inode = mapping->host; + ssize_t written; + size_t write_len; + pgoff_t end; + + if (count != ocount) + *nr_segs = iov_shorten((struct iovec *)iov, *nr_segs, count); + + write_len = iov_length(iov, *nr_segs); + end = (pos + write_len - 1) >> PAGE_CACHE_SHIFT; + + written = filemap_write_and_wait_range(mapping, pos, pos + write_len - 1); + if (written) + goto out; + + /* + * After a write we want buffered reads to be sure to go to disk to get + * the new data. We invalidate clean cached page from the region we're + * about to write. We do this *before* the write so that we can return + * without clobbering -EIOCBQUEUED from ->direct_IO(). + */ + if (mapping->nrpages) { + written = invalidate_inode_pages2_range(mapping, + pos >> PAGE_CACHE_SHIFT, end); + /* + * If a page can not be invalidated, return 0 to fall back + * to buffered write. + */ + if (written) { + if (written == -EBUSY) + return 0; + goto out; + } + } + + written = mapping->a_ops->direct_IO(WRITE, iocb, iov, pos, *nr_segs); + + /* + * Finally, try again to invalidate clean pages which might have been + * cached by non-direct readahead, or faulted in by get_user_pages() + * if the source of the write was an mmap'ed region of the file + * we're writing. Either one is a pretty crazy thing to do, + * so we don't support it 100%. If this invalidation + * fails, tough, the write still worked... + */ + if (mapping->nrpages) { + invalidate_inode_pages2_range(mapping, + pos >> PAGE_CACHE_SHIFT, end); + } + + if (written > 0) { + pos += written; + if (pos > i_size_read(inode) && !S_ISBLK(inode->i_mode)) { + i_size_write(inode, pos); + mark_inode_dirty(inode); + } + *ppos = pos; + } +out: + return written; +} +EXPORT_SYMBOL(generic_file_direct_write); + +/* + * Find or create a page at the given pagecache position. Return the locked + * page. This function is specifically for buffered writes. + */ +struct page *grab_cache_page_write_begin(struct address_space *mapping, + pgoff_t index, unsigned flags) +{ + int status; + gfp_t gfp_mask; + struct page *page; + gfp_t gfp_notmask = 0; + + gfp_mask = mapping_gfp_mask(mapping); + if (mapping_cap_account_dirty(mapping)) + gfp_mask |= __GFP_WRITE; + if (flags & AOP_FLAG_NOFS) + gfp_notmask = __GFP_FS; +repeat: + page = find_lock_page(mapping, index); + if (page) + goto found; + + page = __page_cache_alloc(gfp_mask & ~gfp_notmask); + if (!page) + return NULL; + status = add_to_page_cache_lru(page, mapping, index, + GFP_KERNEL & ~gfp_notmask); + if (unlikely(status)) { + page_cache_release(page); + if (status == -EEXIST) + goto repeat; + return NULL; + } +found: + wait_for_stable_page(page); + return page; +} +EXPORT_SYMBOL(grab_cache_page_write_begin); + +static ssize_t generic_perform_write(struct file *file, + struct iov_iter *i, loff_t pos) +{ + struct address_space *mapping = file->f_mapping; + const struct address_space_operations *a_ops = mapping->a_ops; + long status = 0; + ssize_t written = 0; + unsigned int flags = 0; + + /* + * Copies from kernel address space cannot fail (NFSD is a big user). + */ + if (segment_eq(get_fs(), KERNEL_DS)) + flags |= AOP_FLAG_UNINTERRUPTIBLE; + + do { + struct page *page; + unsigned long offset; /* Offset into pagecache page */ + unsigned long bytes; /* Bytes to write to page */ + size_t copied; /* Bytes copied from user */ + void *fsdata; + + offset = (pos & (PAGE_CACHE_SIZE - 1)); + bytes = min_t(unsigned long, PAGE_CACHE_SIZE - offset, + iov_iter_count(i)); + +again: + /* + * Bring in the user page that we will copy from _first_. + * Otherwise there's a nasty deadlock on copying from the + * same page as we're writing to, without it being marked + * up-to-date. + * + * Not only is this an optimisation, but it is also required + * to check that the address is actually valid, when atomic + * usercopies are used, below. + */ + if (unlikely(iov_iter_fault_in_readable(i, bytes))) { + status = -EFAULT; + break; + } + + status = a_ops->write_begin(file, mapping, pos, bytes, flags, + &page, &fsdata); + if (unlikely(status)) + break; + + if (mapping_writably_mapped(mapping)) + flush_dcache_page(page); + + pagefault_disable(); + copied = iov_iter_copy_from_user_atomic(page, i, offset, bytes); + pagefault_enable(); + flush_dcache_page(page); + + mark_page_accessed(page); + status = a_ops->write_end(file, mapping, pos, bytes, copied, + page, fsdata); + if (unlikely(status < 0)) + break; + copied = status; + + cond_resched(); + + iov_iter_advance(i, copied); + if (unlikely(copied == 0)) { + /* + * If we were unable to copy any data at all, we must + * fall back to a single segment length write. + * + * If we didn't fallback here, we could livelock + * because not all segments in the iov can be copied at + * once without a pagefault. + */ + bytes = min_t(unsigned long, PAGE_CACHE_SIZE - offset, + iov_iter_single_seg_count(i)); + goto again; + } + pos += copied; + written += copied; + + balance_dirty_pages_ratelimited(mapping); + if (fatal_signal_pending(current)) { + status = -EINTR; + break; + } + } while (iov_iter_count(i)); + + return written ? written : status; +} + +ssize_t +generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov, + unsigned long nr_segs, loff_t pos, loff_t *ppos, + size_t count, ssize_t written) +{ + struct file *file = iocb->ki_filp; + ssize_t status; + struct iov_iter i; + + iov_iter_init(&i, iov, nr_segs, count, written); + status = generic_perform_write(file, &i, pos); + + if (likely(status >= 0)) { + written += status; + *ppos = pos + status; + } + + return written ? written : status; +} +EXPORT_SYMBOL(generic_file_buffered_write); + +/** + * __generic_file_aio_write - write data to a file + * @iocb: IO state structure (file, offset, etc.) + * @iov: vector with data to write + * @nr_segs: number of segments in the vector + * @ppos: position where to write + * + * This function does all the work needed for actually writing data to a + * file. It does all basic checks, removes SUID from the file, updates + * modification times and calls proper subroutines depending on whether we + * do direct IO or a standard buffered write. + * + * It expects i_mutex to be grabbed unless we work on a block device or similar + * object which does not need locking at all. + * + * This function does *not* take care of syncing data in case of O_SYNC write. + * A caller has to handle it. This is mainly due to the fact that we want to + * avoid syncing under i_mutex. + */ +ssize_t __generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, + unsigned long nr_segs, loff_t *ppos) +{ + struct file *file = iocb->ki_filp; + struct address_space * mapping = file->f_mapping; + size_t ocount; /* original count */ + size_t count; /* after file limit checks */ + struct inode *inode = mapping->host; + loff_t pos; + ssize_t written; + ssize_t err; + + ocount = 0; + err = generic_segment_checks(iov, &nr_segs, &ocount, VERIFY_READ); + if (err) + return err; + + count = ocount; + pos = *ppos; + + /* We can write back this queue in page reclaim */ + current->backing_dev_info = mapping->backing_dev_info; + written = 0; + + err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode)); + if (err) + goto out; + + if (count == 0) + goto out; + + err = file_remove_suid(file); + if (err) + goto out; + + err = file_update_time(file); + if (err) + goto out; + + /* coalesce the iovecs and go direct-to-BIO for O_DIRECT */ + if (unlikely(file->f_flags & O_DIRECT)) { + loff_t endbyte; + ssize_t written_buffered; + + written = generic_file_direct_write(iocb, iov, &nr_segs, pos, + ppos, count, ocount); + if (written < 0 || written == count) + goto out; + /* + * direct-io write to a hole: fall through to buffered I/O + * for completing the rest of the request. + */ + pos += written; + count -= written; + written_buffered = generic_file_buffered_write(iocb, iov, + nr_segs, pos, ppos, count, + written); + /* + * If generic_file_buffered_write() retuned a synchronous error + * then we want to return the number of bytes which were + * direct-written, or the error code if that was zero. Note + * that this differs from normal direct-io semantics, which + * will return -EFOO even if some bytes were written. + */ + if (written_buffered < 0) { + err = written_buffered; + goto out; + } + + /* + * We need to ensure that the page cache pages are written to + * disk and invalidated to preserve the expected O_DIRECT + * semantics. + */ + endbyte = pos + written_buffered - written - 1; + err = filemap_write_and_wait_range(file->f_mapping, pos, endbyte); + if (err == 0) { + written = written_buffered; + invalidate_mapping_pages(mapping, + pos >> PAGE_CACHE_SHIFT, + endbyte >> PAGE_CACHE_SHIFT); + } else { + /* + * We don't know how much we wrote, so just return + * the number of bytes which were direct-written + */ + } + } else { + written = generic_file_buffered_write(iocb, iov, nr_segs, + pos, ppos, count, written); + } +out: + current->backing_dev_info = NULL; + return written ? written : err; +} +EXPORT_SYMBOL(__generic_file_aio_write); + +/** + * generic_file_aio_write - write data to a file + * @iocb: IO state structure + * @iov: vector with data to write + * @nr_segs: number of segments in the vector + * @pos: position in file where to write + * + * This is a wrapper around __generic_file_aio_write() to be used by most + * filesystems. It takes care of syncing the file in case of O_SYNC file + * and acquires i_mutex as needed. + */ +ssize_t generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, + unsigned long nr_segs, loff_t pos) +{ + struct file *file = iocb->ki_filp; + struct inode *inode = file->f_mapping->host; + ssize_t ret; + + BUG_ON(iocb->ki_pos != pos); + + mutex_lock(&inode->i_mutex); + ret = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos); + mutex_unlock(&inode->i_mutex); + + if (ret > 0 || ret == -EIOCBQUEUED) { + ssize_t err; + + err = generic_write_sync(file, pos, ret); + if (err < 0 && ret > 0) + ret = err; + } + return ret; +} +EXPORT_SYMBOL(generic_file_aio_write); + +/** + * try_to_release_page() - release old fs-specific metadata on a page + * + * @page: the page which the kernel is trying to free + * @gfp_mask: memory allocation flags (and I/O mode) + * + * The address_space is to try to release any data against the page + * (presumably at page->private). If the release was successful, return `1'. + * Otherwise return zero. + * + * This may also be called if PG_fscache is set on a page, indicating that the + * page is known to the local caching routines. + * + * The @gfp_mask argument specifies whether I/O may be performed to release + * this page (__GFP_IO), and whether the call may block (__GFP_WAIT & __GFP_FS). + * + */ +int try_to_release_page(struct page *page, gfp_t gfp_mask) +{ + struct address_space * const mapping = page->mapping; + + BUG_ON(!PageLocked(page)); + if (PageWriteback(page)) + return 0; + + if (mapping && mapping->a_ops->releasepage) + return mapping->a_ops->releasepage(page, gfp_mask); + return try_to_free_buffers(page); +} + +EXPORT_SYMBOL(try_to_release_page); diff -Naur kernel.21.3.orig/mm/fremap.c kernel.21.3.new/mm/fremap.c --- kernel.21.3.orig/mm/fremap.c 2015-02-03 01:44:33.000000000 +0000 +++ kernel.21.3.new/mm/fremap.c 2015-04-25 23:28:45.348481244 +0000 @@ -202,16 +202,28 @@ */ if (mapping_cap_account_dirty(mapping)) { unsigned long addr; - struct file *file = get_file(vma->vm_file); + struct file *file = vma->vm_file, + *prfile = vma->vm_prfile; + /* mmap_region may free vma; grab the info now */ vm_flags = vma->vm_flags; + vma_get_file(vma); addr = mmap_region(file, start, size, vm_flags, pgoff); - fput(file); + vma_fput(vma); if (IS_ERR_VALUE(addr)) { err = addr; } else { BUG_ON(addr != start); + if (prfile) { + struct vm_area_struct *new_vma; + + new_vma = find_vma(mm, addr); + if (!new_vma->vm_prfile) + new_vma->vm_prfile = prfile; + if (new_vma != vma) + get_file(prfile); + } err = 0; } goto out_freed; diff -Naur kernel.21.3.orig/mm/madvise.c kernel.21.3.new/mm/madvise.c --- kernel.21.3.orig/mm/madvise.c 2015-02-03 01:44:33.000000000 +0000 +++ kernel.21.3.new/mm/madvise.c 2015-04-25 23:28:45.350481224 +0000 @@ -328,12 +328,12 @@ * vma's reference to the file) can go away as soon as we drop * mmap_sem. */ - get_file(f); + vma_get_file(vma); up_read(¤t->mm->mmap_sem); error = do_fallocate(f, FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE, offset, end - start); - fput(f); + vma_fput(vma); down_read(¤t->mm->mmap_sem); return error; } diff -Naur kernel.21.3.orig/mm/madvise.c.orig kernel.21.3.new/mm/madvise.c.orig --- kernel.21.3.orig/mm/madvise.c.orig 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/mm/madvise.c.orig 2015-02-03 01:44:33.000000000 +0000 @@ -0,0 +1,551 @@ +/* + * linux/mm/madvise.c + * + * Copyright (C) 1999 Linus Torvalds + * Copyright (C) 2002 Christoph Hellwig + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* + * Any behaviour which results in changes to the vma->vm_flags needs to + * take mmap_sem for writing. Others, which simply traverse vmas, need + * to only take it for reading. + */ +static int madvise_need_mmap_write(int behavior) +{ + switch (behavior) { + case MADV_REMOVE: + case MADV_WILLNEED: + case MADV_DONTNEED: + return 0; + default: + /* be safe, default to 1. list exceptions explicitly */ + return 1; + } +} + +/* + * We can potentially split a vm area into separate + * areas, each area with its own behavior. + */ +static long madvise_behavior(struct vm_area_struct * vma, + struct vm_area_struct **prev, + unsigned long start, unsigned long end, int behavior) +{ + struct mm_struct * mm = vma->vm_mm; + int error = 0; + pgoff_t pgoff; + unsigned long new_flags = vma->vm_flags; + + switch (behavior) { + case MADV_NORMAL: + new_flags = new_flags & ~VM_RAND_READ & ~VM_SEQ_READ; + break; + case MADV_SEQUENTIAL: + new_flags = (new_flags & ~VM_RAND_READ) | VM_SEQ_READ; + break; + case MADV_RANDOM: + new_flags = (new_flags & ~VM_SEQ_READ) | VM_RAND_READ; + break; + case MADV_DONTFORK: + new_flags |= VM_DONTCOPY; + break; + case MADV_DOFORK: + if (vma->vm_flags & VM_IO) { + error = -EINVAL; + goto out; + } + new_flags &= ~VM_DONTCOPY; + break; + case MADV_DONTDUMP: + new_flags |= VM_DONTDUMP; + break; + case MADV_DODUMP: + if (new_flags & VM_SPECIAL) { + error = -EINVAL; + goto out; + } + new_flags &= ~VM_DONTDUMP; + break; + case MADV_MERGEABLE: + case MADV_UNMERGEABLE: + error = ksm_madvise(vma, start, end, behavior, &new_flags); + if (error) + goto out; + break; + case MADV_HUGEPAGE: + case MADV_NOHUGEPAGE: + error = hugepage_madvise(vma, &new_flags, behavior); + if (error) + goto out; + break; + } + + if (new_flags == vma->vm_flags) { + *prev = vma; + goto out; + } + + pgoff = vma->vm_pgoff + ((start - vma->vm_start) >> PAGE_SHIFT); + *prev = vma_merge(mm, *prev, start, end, new_flags, vma->anon_vma, + vma->vm_file, pgoff, vma_policy(vma), + vma_get_anon_name(vma)); + if (*prev) { + vma = *prev; + goto success; + } + + *prev = vma; + + if (start != vma->vm_start) { + error = split_vma(mm, vma, start, 1); + if (error) + goto out; + } + + if (end != vma->vm_end) { + error = split_vma(mm, vma, end, 0); + if (error) + goto out; + } + +success: + /* + * vm_flags is protected by the mmap_sem held in write mode. + */ + vma->vm_flags = new_flags; + +out: + if (error == -ENOMEM) + error = -EAGAIN; + return error; +} + +#ifdef CONFIG_SWAP +static int swapin_walk_pmd_entry(pmd_t *pmd, unsigned long start, + unsigned long end, struct mm_walk *walk) +{ + pte_t *orig_pte; + struct vm_area_struct *vma = walk->private; + unsigned long index; + + if (pmd_none_or_trans_huge_or_clear_bad(pmd)) + return 0; + + for (index = start; index != end; index += PAGE_SIZE) { + pte_t pte; + swp_entry_t entry; + struct page *page; + spinlock_t *ptl; + + orig_pte = pte_offset_map_lock(vma->vm_mm, pmd, start, &ptl); + pte = *(orig_pte + ((index - start) / PAGE_SIZE)); + pte_unmap_unlock(orig_pte, ptl); + + if (pte_present(pte) || pte_none(pte) || pte_file(pte)) + continue; + entry = pte_to_swp_entry(pte); + if (unlikely(non_swap_entry(entry))) + continue; + + page = read_swap_cache_async(entry, GFP_HIGHUSER_MOVABLE, + vma, index); + if (page) + page_cache_release(page); + } + + return 0; +} + +static void force_swapin_readahead(struct vm_area_struct *vma, + unsigned long start, unsigned long end) +{ + struct mm_walk walk = { + .mm = vma->vm_mm, + .pmd_entry = swapin_walk_pmd_entry, + .private = vma, + }; + + walk_page_range(start, end, &walk); + + lru_add_drain(); /* Push any new pages onto the LRU now */ +} + +static void force_shm_swapin_readahead(struct vm_area_struct *vma, + unsigned long start, unsigned long end, + struct address_space *mapping) +{ + pgoff_t index; + struct page *page; + swp_entry_t swap; + + for (; start < end; start += PAGE_SIZE) { + index = ((start - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff; + + page = find_get_page(mapping, index); + if (!radix_tree_exceptional_entry(page)) { + if (page) + page_cache_release(page); + continue; + } + swap = radix_to_swp_entry(page); + page = read_swap_cache_async(swap, GFP_HIGHUSER_MOVABLE, + NULL, 0); + if (page) + page_cache_release(page); + } + + lru_add_drain(); /* Push any new pages onto the LRU now */ +} +#endif /* CONFIG_SWAP */ + +/* + * Schedule all required I/O operations. Do not wait for completion. + */ +static long madvise_willneed(struct vm_area_struct * vma, + struct vm_area_struct ** prev, + unsigned long start, unsigned long end) +{ + struct file *file = vma->vm_file; + +#ifdef CONFIG_SWAP + if (!file || mapping_cap_swap_backed(file->f_mapping)) { + *prev = vma; + if (!file) + force_swapin_readahead(vma, start, end); + else + force_shm_swapin_readahead(vma, start, end, + file->f_mapping); + return 0; + } +#endif + + if (!file) + return -EBADF; + + if (file->f_mapping->a_ops->get_xip_mem) { + /* no bad return value, but ignore advice */ + return 0; + } + + *prev = vma; + start = ((start - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff; + if (end > vma->vm_end) + end = vma->vm_end; + end = ((end - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff; + + force_page_cache_readahead(file->f_mapping, file, start, end - start); + return 0; +} + +/* + * Application no longer needs these pages. If the pages are dirty, + * it's OK to just throw them away. The app will be more careful about + * data it wants to keep. Be sure to free swap resources too. The + * zap_page_range call sets things up for shrink_active_list to actually free + * these pages later if no one else has touched them in the meantime, + * although we could add these pages to a global reuse list for + * shrink_active_list to pick up before reclaiming other pages. + * + * NB: This interface discards data rather than pushes it out to swap, + * as some implementations do. This has performance implications for + * applications like large transactional databases which want to discard + * pages in anonymous maps after committing to backing store the data + * that was kept in them. There is no reason to write this data out to + * the swap area if the application is discarding it. + * + * An interface that causes the system to free clean pages and flush + * dirty pages is already available as msync(MS_INVALIDATE). + */ +static long madvise_dontneed(struct vm_area_struct * vma, + struct vm_area_struct ** prev, + unsigned long start, unsigned long end) +{ + *prev = vma; + if (vma->vm_flags & (VM_LOCKED|VM_HUGETLB|VM_PFNMAP)) + return -EINVAL; + + if (unlikely(vma->vm_flags & VM_NONLINEAR)) { + struct zap_details details = { + .nonlinear_vma = vma, + .last_index = ULONG_MAX, + }; + zap_page_range(vma, start, end - start, &details); + } else + zap_page_range(vma, start, end - start, NULL); + return 0; +} + +/* + * Application wants to free up the pages and associated backing store. + * This is effectively punching a hole into the middle of a file. + * + * NOTE: Currently, only shmfs/tmpfs is supported for this operation. + * Other filesystems return -ENOSYS. + */ +static long madvise_remove(struct vm_area_struct *vma, + struct vm_area_struct **prev, + unsigned long start, unsigned long end) +{ + loff_t offset; + int error; + struct file *f; + + *prev = NULL; /* tell sys_madvise we drop mmap_sem */ + + if (vma->vm_flags & (VM_LOCKED|VM_NONLINEAR|VM_HUGETLB)) + return -EINVAL; + + f = vma->vm_file; + + if (!f || !f->f_mapping || !f->f_mapping->host) { + return -EINVAL; + } + + if ((vma->vm_flags & (VM_SHARED|VM_WRITE)) != (VM_SHARED|VM_WRITE)) + return -EACCES; + + offset = (loff_t)(start - vma->vm_start) + + ((loff_t)vma->vm_pgoff << PAGE_SHIFT); + + /* + * Filesystem's fallocate may need to take i_mutex. We need to + * explicitly grab a reference because the vma (and hence the + * vma's reference to the file) can go away as soon as we drop + * mmap_sem. + */ + get_file(f); + up_read(¤t->mm->mmap_sem); + error = do_fallocate(f, + FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE, + offset, end - start); + fput(f); + down_read(¤t->mm->mmap_sem); + return error; +} + +#ifdef CONFIG_MEMORY_FAILURE +/* + * Error injection support for memory error handling. + */ +static int madvise_hwpoison(int bhv, unsigned long start, unsigned long end) +{ + int ret = 0; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + for (; start < end; start += PAGE_SIZE) { + struct page *p; + int ret = get_user_pages_fast(start, 1, 0, &p); + if (ret != 1) + return ret; + if (bhv == MADV_SOFT_OFFLINE) { + printk(KERN_INFO "Soft offlining page %lx at %lx\n", + page_to_pfn(p), start); + ret = soft_offline_page(p, MF_COUNT_INCREASED); + if (ret) + break; + continue; + } + printk(KERN_INFO "Injecting memory failure for page %lx at %lx\n", + page_to_pfn(p), start); + /* Ignore return value for now */ + memory_failure(page_to_pfn(p), 0, MF_COUNT_INCREASED); + } + return ret; +} +#endif + +static long +madvise_vma(struct vm_area_struct *vma, struct vm_area_struct **prev, + unsigned long start, unsigned long end, int behavior) +{ + switch (behavior) { + case MADV_REMOVE: + return madvise_remove(vma, prev, start, end); + case MADV_WILLNEED: + return madvise_willneed(vma, prev, start, end); + case MADV_DONTNEED: + return madvise_dontneed(vma, prev, start, end); + default: + return madvise_behavior(vma, prev, start, end, behavior); + } +} + +static int +madvise_behavior_valid(int behavior) +{ + switch (behavior) { + case MADV_DOFORK: + case MADV_DONTFORK: + case MADV_NORMAL: + case MADV_SEQUENTIAL: + case MADV_RANDOM: + case MADV_REMOVE: + case MADV_WILLNEED: + case MADV_DONTNEED: +#ifdef CONFIG_KSM + case MADV_MERGEABLE: + case MADV_UNMERGEABLE: +#endif +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + case MADV_HUGEPAGE: + case MADV_NOHUGEPAGE: +#endif + case MADV_DONTDUMP: + case MADV_DODUMP: + return 1; + + default: + return 0; + } +} + +/* + * The madvise(2) system call. + * + * Applications can use madvise() to advise the kernel how it should + * handle paging I/O in this VM area. The idea is to help the kernel + * use appropriate read-ahead and caching techniques. The information + * provided is advisory only, and can be safely disregarded by the + * kernel without affecting the correct operation of the application. + * + * behavior values: + * MADV_NORMAL - the default behavior is to read clusters. This + * results in some read-ahead and read-behind. + * MADV_RANDOM - the system should read the minimum amount of data + * on any access, since it is unlikely that the appli- + * cation will need more than what it asks for. + * MADV_SEQUENTIAL - pages in the given range will probably be accessed + * once, so they can be aggressively read ahead, and + * can be freed soon after they are accessed. + * MADV_WILLNEED - the application is notifying the system to read + * some pages ahead. + * MADV_DONTNEED - the application is finished with the given range, + * so the kernel can free resources associated with it. + * MADV_REMOVE - the application wants to free up the given range of + * pages and associated backing store. + * MADV_DONTFORK - omit this area from child's address space when forking: + * typically, to avoid COWing pages pinned by get_user_pages(). + * MADV_DOFORK - cancel MADV_DONTFORK: no longer omit this area when forking. + * MADV_MERGEABLE - the application recommends that KSM try to merge pages in + * this area with pages of identical content from other such areas. + * MADV_UNMERGEABLE- cancel MADV_MERGEABLE: no longer merge pages with others. + * + * return values: + * zero - success + * -EINVAL - start + len < 0, start is not page-aligned, + * "behavior" is not a valid value, or application + * is attempting to release locked or shared pages. + * -ENOMEM - addresses in the specified range are not currently + * mapped, or are outside the AS of the process. + * -EIO - an I/O error occurred while paging in data. + * -EBADF - map exists, but area maps something that isn't a file. + * -EAGAIN - a kernel resource was temporarily unavailable. + */ +SYSCALL_DEFINE3(madvise, unsigned long, start, size_t, len_in, int, behavior) +{ + unsigned long end, tmp; + struct vm_area_struct * vma, *prev; + int unmapped_error = 0; + int error = -EINVAL; + int write; + size_t len; + struct blk_plug plug; + +#ifdef CONFIG_MEMORY_FAILURE + if (behavior == MADV_HWPOISON || behavior == MADV_SOFT_OFFLINE) + return madvise_hwpoison(behavior, start, start+len_in); +#endif + if (!madvise_behavior_valid(behavior)) + return error; + + if (start & ~PAGE_MASK) + return error; + len = (len_in + ~PAGE_MASK) & PAGE_MASK; + + /* Check to see whether len was rounded up from small -ve to zero */ + if (len_in && !len) + return error; + + end = start + len; + if (end < start) + return error; + + error = 0; + if (end == start) + return error; + + write = madvise_need_mmap_write(behavior); + if (write) + down_write(¤t->mm->mmap_sem); + else + down_read(¤t->mm->mmap_sem); + + /* + * If the interval [start,end) covers some unmapped address + * ranges, just ignore them, but return -ENOMEM at the end. + * - different from the way of handling in mlock etc. + */ + vma = find_vma_prev(current->mm, start, &prev); + if (vma && start > vma->vm_start) + prev = vma; + + blk_start_plug(&plug); + for (;;) { + /* Still start < end. */ + error = -ENOMEM; + if (!vma) + goto out; + + /* Here start < (end|vma->vm_end). */ + if (start < vma->vm_start) { + unmapped_error = -ENOMEM; + start = vma->vm_start; + if (start >= end) + goto out; + } + + /* Here vma->vm_start <= start < (end|vma->vm_end) */ + tmp = vma->vm_end; + if (end < tmp) + tmp = end; + + /* Here vma->vm_start <= start < tmp <= (end|vma->vm_end). */ + error = madvise_vma(vma, &prev, start, tmp, behavior); + if (error) + goto out; + start = tmp; + if (prev && start < prev->vm_end) + start = prev->vm_end; + error = unmapped_error; + if (start >= end) + goto out; + if (prev) + vma = prev->vm_next; + else /* madvise_remove dropped mmap_sem */ + vma = find_vma(current->mm, start); + } +out: + blk_finish_plug(&plug); + if (write) + up_write(¤t->mm->mmap_sem); + else + up_read(¤t->mm->mmap_sem); + + return error; +} diff -Naur kernel.21.3.orig/mm/Makefile kernel.21.3.new/mm/Makefile --- kernel.21.3.orig/mm/Makefile 2015-02-03 01:44:33.000000000 +0000 +++ kernel.21.3.new/mm/Makefile 2015-04-25 23:28:45.331481413 +0000 @@ -17,7 +17,7 @@ util.o mmzone.o vmstat.o backing-dev.o \ mm_init.o mmu_context.o percpu.o slab_common.o \ compaction.o balloon_compaction.o \ - interval_tree.o $(mmu-y) + interval_tree.o prfile.o $(mmu-y) obj-y += init-mm.o diff -Naur kernel.21.3.orig/mm/memory.c kernel.21.3.new/mm/memory.c --- kernel.21.3.orig/mm/memory.c 2015-02-03 01:44:33.000000000 +0000 +++ kernel.21.3.new/mm/memory.c 2015-04-25 23:28:45.378480945 +0000 @@ -2739,7 +2739,7 @@ set_page_dirty_balance(dirty_page, page_mkwrite); /* file_update_time outside page_lock */ if (vma->vm_file) - file_update_time(vma->vm_file); + vma_file_update_time(vma); } put_page(dirty_page); if (page_mkwrite) { @@ -3457,7 +3457,7 @@ /* file_update_time outside page_lock */ if (vma->vm_file && !page_mkwrite) - file_update_time(vma->vm_file); + vma_file_update_time(vma); } else { unlock_page(vmf.page); if (anon) diff -Naur kernel.21.3.orig/mm/memory.c.orig kernel.21.3.new/mm/memory.c.orig --- kernel.21.3.orig/mm/memory.c.orig 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/mm/memory.c.orig 2015-02-03 01:44:33.000000000 +0000 @@ -0,0 +1,4316 @@ +/* + * linux/mm/memory.c + * + * Copyright (C) 1991, 1992, 1993, 1994 Linus Torvalds + */ + +/* + * demand-loading started 01.12.91 - seems it is high on the list of + * things wanted, and it should be easy to implement. - Linus + */ + +/* + * Ok, demand-loading was easy, shared pages a little bit tricker. Shared + * pages started 02.12.91, seems to work. - Linus. + * + * Tested sharing by executing about 30 /bin/sh: under the old kernel it + * would have taken more than the 6M I have free, but it worked well as + * far as I could see. + * + * Also corrected some "invalidate()"s - I wasn't doing enough of them. + */ + +/* + * Real VM (paging to/from disk) started 18.12.91. Much more work and + * thought has to go into this. Oh, well.. + * 19.12.91 - works, somewhat. Sometimes I get faults, don't know why. + * Found it. Everything seems to work now. + * 20.12.91 - Ok, making the swap-device changeable like the root. + */ + +/* + * 05.04.94 - Multi-page memory management added for v1.1. + * Idea by Alex Bligh (alex@cconcepts.co.uk) + * + * 16.07.99 - Support of BIGMEM added by Gerhard Wichert, Siemens AG + * (Gerhard.Wichert@pdb.siemens.de) + * + * Aug/Sep 2004 Changed to four level page tables (Andi Kleen) + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +#include "internal.h" + +#ifdef LAST_NID_NOT_IN_PAGE_FLAGS +#warning Unfortunate NUMA and NUMA Balancing config, growing page-frame for last_nid. +#endif + +#ifndef CONFIG_NEED_MULTIPLE_NODES +/* use the per-pgdat data instead for discontigmem - mbligh */ +unsigned long max_mapnr; +struct page *mem_map; + +EXPORT_SYMBOL(max_mapnr); +EXPORT_SYMBOL(mem_map); +#endif + +unsigned long num_physpages; +/* + * A number of key systems in x86 including ioremap() rely on the assumption + * that high_memory defines the upper bound on direct map memory, then end + * of ZONE_NORMAL. Under CONFIG_DISCONTIG this means that max_low_pfn and + * highstart_pfn must be the same; there must be no gap between ZONE_NORMAL + * and ZONE_HIGHMEM. + */ +void * high_memory; + +EXPORT_SYMBOL(num_physpages); +EXPORT_SYMBOL(high_memory); + +/* + * Randomize the address space (stacks, mmaps, brk, etc.). + * + * ( When CONFIG_COMPAT_BRK=y we exclude brk from randomization, + * as ancient (libc5 based) binaries can segfault. ) + */ +int randomize_va_space __read_mostly = +#ifdef CONFIG_COMPAT_BRK + 1; +#else + 2; +#endif + +static int __init disable_randmaps(char *s) +{ + randomize_va_space = 0; + return 1; +} +__setup("norandmaps", disable_randmaps); + +unsigned long zero_pfn __read_mostly; +unsigned long highest_memmap_pfn __read_mostly; + +/* + * CONFIG_MMU architectures set up ZERO_PAGE in their paging_init() + */ +static int __init init_zero_pfn(void) +{ + zero_pfn = page_to_pfn(ZERO_PAGE(0)); + return 0; +} +core_initcall(init_zero_pfn); + + +#if defined(SPLIT_RSS_COUNTING) + +void sync_mm_rss(struct mm_struct *mm) +{ + int i; + + for (i = 0; i < NR_MM_COUNTERS; i++) { + if (current->rss_stat.count[i]) { + add_mm_counter(mm, i, current->rss_stat.count[i]); + current->rss_stat.count[i] = 0; + } + } + current->rss_stat.events = 0; +} + +static void add_mm_counter_fast(struct mm_struct *mm, int member, int val) +{ + struct task_struct *task = current; + + if (likely(task->mm == mm)) + task->rss_stat.count[member] += val; + else + add_mm_counter(mm, member, val); +} +#define inc_mm_counter_fast(mm, member) add_mm_counter_fast(mm, member, 1) +#define dec_mm_counter_fast(mm, member) add_mm_counter_fast(mm, member, -1) + +/* sync counter once per 64 page faults */ +#define TASK_RSS_EVENTS_THRESH (64) +static void check_sync_rss_stat(struct task_struct *task) +{ + if (unlikely(task != current)) + return; + if (unlikely(task->rss_stat.events++ > TASK_RSS_EVENTS_THRESH)) + sync_mm_rss(task->mm); +} +#else /* SPLIT_RSS_COUNTING */ + +#define inc_mm_counter_fast(mm, member) inc_mm_counter(mm, member) +#define dec_mm_counter_fast(mm, member) dec_mm_counter(mm, member) + +static void check_sync_rss_stat(struct task_struct *task) +{ +} + +#endif /* SPLIT_RSS_COUNTING */ + +#ifdef HAVE_GENERIC_MMU_GATHER + +static int tlb_next_batch(struct mmu_gather *tlb) +{ + struct mmu_gather_batch *batch; + + batch = tlb->active; + if (batch->next) { + tlb->active = batch->next; + return 1; + } + + if (tlb->batch_count == MAX_GATHER_BATCH_COUNT) + return 0; + + batch = (void *)__get_free_pages(GFP_NOWAIT | __GFP_NOWARN, 0); + if (!batch) + return 0; + + tlb->batch_count++; + batch->next = NULL; + batch->nr = 0; + batch->max = MAX_GATHER_BATCH; + + tlb->active->next = batch; + tlb->active = batch; + + return 1; +} + +/* tlb_gather_mmu + * Called to initialize an (on-stack) mmu_gather structure for page-table + * tear-down from @mm. The @fullmm argument is used when @mm is without + * users and we're going to destroy the full address space (exit/execve). + */ +void tlb_gather_mmu(struct mmu_gather *tlb, struct mm_struct *mm, unsigned long start, unsigned long end) +{ + tlb->mm = mm; + + /* Is it from 0 to ~0? */ + tlb->fullmm = !(start | (end+1)); + tlb->need_flush_all = 0; + tlb->start = start; + tlb->end = end; + tlb->need_flush = 0; + tlb->local.next = NULL; + tlb->local.nr = 0; + tlb->local.max = ARRAY_SIZE(tlb->__pages); + tlb->active = &tlb->local; + tlb->batch_count = 0; + +#ifdef CONFIG_HAVE_RCU_TABLE_FREE + tlb->batch = NULL; +#endif +} + +void tlb_flush_mmu(struct mmu_gather *tlb) +{ + struct mmu_gather_batch *batch; + + if (!tlb->need_flush) + return; + tlb->need_flush = 0; + tlb_flush(tlb); +#ifdef CONFIG_HAVE_RCU_TABLE_FREE + tlb_table_flush(tlb); +#endif + + for (batch = &tlb->local; batch; batch = batch->next) { + free_pages_and_swap_cache(batch->pages, batch->nr); + batch->nr = 0; + } + tlb->active = &tlb->local; +} + +/* tlb_finish_mmu + * Called at the end of the shootdown operation to free up any resources + * that were required. + */ +void tlb_finish_mmu(struct mmu_gather *tlb, unsigned long start, unsigned long end) +{ + struct mmu_gather_batch *batch, *next; + + tlb_flush_mmu(tlb); + + /* keep the page table cache within bounds */ + check_pgt_cache(); + + for (batch = tlb->local.next; batch; batch = next) { + next = batch->next; + free_pages((unsigned long)batch, 0); + } + tlb->local.next = NULL; +} + +/* __tlb_remove_page + * Must perform the equivalent to __free_pte(pte_get_and_clear(ptep)), while + * handling the additional races in SMP caused by other CPUs caching valid + * mappings in their TLBs. Returns the number of free page slots left. + * When out of page slots we must call tlb_flush_mmu(). + */ +int __tlb_remove_page(struct mmu_gather *tlb, struct page *page) +{ + struct mmu_gather_batch *batch; + + VM_BUG_ON(!tlb->need_flush); + + batch = tlb->active; + batch->pages[batch->nr++] = page; + if (batch->nr == batch->max) { + if (!tlb_next_batch(tlb)) + return 0; + batch = tlb->active; + } + VM_BUG_ON(batch->nr > batch->max); + + return batch->max - batch->nr; +} + +#endif /* HAVE_GENERIC_MMU_GATHER */ + +#ifdef CONFIG_HAVE_RCU_TABLE_FREE + +/* + * See the comment near struct mmu_table_batch. + */ + +static void tlb_remove_table_smp_sync(void *arg) +{ + /* Simply deliver the interrupt */ +} + +static void tlb_remove_table_one(void *table) +{ + /* + * This isn't an RCU grace period and hence the page-tables cannot be + * assumed to be actually RCU-freed. + * + * It is however sufficient for software page-table walkers that rely on + * IRQ disabling. See the comment near struct mmu_table_batch. + */ + smp_call_function(tlb_remove_table_smp_sync, NULL, 1); + __tlb_remove_table(table); +} + +static void tlb_remove_table_rcu(struct rcu_head *head) +{ + struct mmu_table_batch *batch; + int i; + + batch = container_of(head, struct mmu_table_batch, rcu); + + for (i = 0; i < batch->nr; i++) + __tlb_remove_table(batch->tables[i]); + + free_page((unsigned long)batch); +} + +void tlb_table_flush(struct mmu_gather *tlb) +{ + struct mmu_table_batch **batch = &tlb->batch; + + if (*batch) { + call_rcu_sched(&(*batch)->rcu, tlb_remove_table_rcu); + *batch = NULL; + } +} + +void tlb_remove_table(struct mmu_gather *tlb, void *table) +{ + struct mmu_table_batch **batch = &tlb->batch; + + tlb->need_flush = 1; + + /* + * When there's less then two users of this mm there cannot be a + * concurrent page-table walk. + */ + if (atomic_read(&tlb->mm->mm_users) < 2) { + __tlb_remove_table(table); + return; + } + + if (*batch == NULL) { + *batch = (struct mmu_table_batch *)__get_free_page(GFP_NOWAIT | __GFP_NOWARN); + if (*batch == NULL) { + tlb_remove_table_one(table); + return; + } + (*batch)->nr = 0; + } + (*batch)->tables[(*batch)->nr++] = table; + if ((*batch)->nr == MAX_TABLE_BATCH) + tlb_table_flush(tlb); +} + +#endif /* CONFIG_HAVE_RCU_TABLE_FREE */ + +/* + * If a p?d_bad entry is found while walking page tables, report + * the error, before resetting entry to p?d_none. Usually (but + * very seldom) called out from the p?d_none_or_clear_bad macros. + */ + +void pgd_clear_bad(pgd_t *pgd) +{ + pgd_ERROR(*pgd); + pgd_clear(pgd); +} + +void pud_clear_bad(pud_t *pud) +{ + pud_ERROR(*pud); + pud_clear(pud); +} + +void pmd_clear_bad(pmd_t *pmd) +{ + pmd_ERROR(*pmd); + pmd_clear(pmd); +} + +/* + * Note: this doesn't free the actual pages themselves. That + * has been handled earlier when unmapping all the memory regions. + */ +static void free_pte_range(struct mmu_gather *tlb, pmd_t *pmd, + unsigned long addr) +{ + pgtable_t token = pmd_pgtable(*pmd); + pmd_clear(pmd); + pte_free_tlb(tlb, token, addr); + tlb->mm->nr_ptes--; +} + +static inline void free_pmd_range(struct mmu_gather *tlb, pud_t *pud, + unsigned long addr, unsigned long end, + unsigned long floor, unsigned long ceiling) +{ + pmd_t *pmd; + unsigned long next; + unsigned long start; + + start = addr; + pmd = pmd_offset(pud, addr); + do { + next = pmd_addr_end(addr, end); + if (pmd_none_or_clear_bad(pmd)) + continue; + free_pte_range(tlb, pmd, addr); + } while (pmd++, addr = next, addr != end); + + start &= PUD_MASK; + if (start < floor) + return; + if (ceiling) { + ceiling &= PUD_MASK; + if (!ceiling) + return; + } + if (end - 1 > ceiling - 1) + return; + + pmd = pmd_offset(pud, start); + pud_clear(pud); + pmd_free_tlb(tlb, pmd, start); +} + +static inline void free_pud_range(struct mmu_gather *tlb, pgd_t *pgd, + unsigned long addr, unsigned long end, + unsigned long floor, unsigned long ceiling) +{ + pud_t *pud; + unsigned long next; + unsigned long start; + + start = addr; + pud = pud_offset(pgd, addr); + do { + next = pud_addr_end(addr, end); + if (pud_none_or_clear_bad(pud)) + continue; + free_pmd_range(tlb, pud, addr, next, floor, ceiling); + } while (pud++, addr = next, addr != end); + + start &= PGDIR_MASK; + if (start < floor) + return; + if (ceiling) { + ceiling &= PGDIR_MASK; + if (!ceiling) + return; + } + if (end - 1 > ceiling - 1) + return; + + pud = pud_offset(pgd, start); + pgd_clear(pgd); + pud_free_tlb(tlb, pud, start); +} + +/* + * This function frees user-level page tables of a process. + * + * Must be called with pagetable lock held. + */ +void free_pgd_range(struct mmu_gather *tlb, + unsigned long addr, unsigned long end, + unsigned long floor, unsigned long ceiling) +{ + pgd_t *pgd; + unsigned long next; + + /* + * The next few lines have given us lots of grief... + * + * Why are we testing PMD* at this top level? Because often + * there will be no work to do at all, and we'd prefer not to + * go all the way down to the bottom just to discover that. + * + * Why all these "- 1"s? Because 0 represents both the bottom + * of the address space and the top of it (using -1 for the + * top wouldn't help much: the masks would do the wrong thing). + * The rule is that addr 0 and floor 0 refer to the bottom of + * the address space, but end 0 and ceiling 0 refer to the top + * Comparisons need to use "end - 1" and "ceiling - 1" (though + * that end 0 case should be mythical). + * + * Wherever addr is brought up or ceiling brought down, we must + * be careful to reject "the opposite 0" before it confuses the + * subsequent tests. But what about where end is brought down + * by PMD_SIZE below? no, end can't go down to 0 there. + * + * Whereas we round start (addr) and ceiling down, by different + * masks at different levels, in order to test whether a table + * now has no other vmas using it, so can be freed, we don't + * bother to round floor or end up - the tests don't need that. + */ + + addr &= PMD_MASK; + if (addr < floor) { + addr += PMD_SIZE; + if (!addr) + return; + } + if (ceiling) { + ceiling &= PMD_MASK; + if (!ceiling) + return; + } + if (end - 1 > ceiling - 1) + end -= PMD_SIZE; + if (addr > end - 1) + return; + + pgd = pgd_offset(tlb->mm, addr); + do { + next = pgd_addr_end(addr, end); + if (pgd_none_or_clear_bad(pgd)) + continue; + free_pud_range(tlb, pgd, addr, next, floor, ceiling); + } while (pgd++, addr = next, addr != end); +} + +void free_pgtables(struct mmu_gather *tlb, struct vm_area_struct *vma, + unsigned long floor, unsigned long ceiling) +{ + while (vma) { + struct vm_area_struct *next = vma->vm_next; + unsigned long addr = vma->vm_start; + + /* + * Hide vma from rmap and truncate_pagecache before freeing + * pgtables + */ + unlink_anon_vmas(vma); + unlink_file_vma(vma); + + if (is_vm_hugetlb_page(vma)) { + hugetlb_free_pgd_range(tlb, addr, vma->vm_end, + floor, next? next->vm_start: ceiling); + } else { + /* + * Optimization: gather nearby vmas into one call down + */ + while (next && next->vm_start <= vma->vm_end + PMD_SIZE + && !is_vm_hugetlb_page(next)) { + vma = next; + next = vma->vm_next; + unlink_anon_vmas(vma); + unlink_file_vma(vma); + } + free_pgd_range(tlb, addr, vma->vm_end, + floor, next? next->vm_start: ceiling); + } + vma = next; + } +} + +int __pte_alloc(struct mm_struct *mm, struct vm_area_struct *vma, + pmd_t *pmd, unsigned long address) +{ + pgtable_t new = pte_alloc_one(mm, address); + int wait_split_huge_page; + if (!new) + return -ENOMEM; + + /* + * Ensure all pte setup (eg. pte page lock and page clearing) are + * visible before the pte is made visible to other CPUs by being + * put into page tables. + * + * The other side of the story is the pointer chasing in the page + * table walking code (when walking the page table without locking; + * ie. most of the time). Fortunately, these data accesses consist + * of a chain of data-dependent loads, meaning most CPUs (alpha + * being the notable exception) will already guarantee loads are + * seen in-order. See the alpha page table accessors for the + * smp_read_barrier_depends() barriers in page table walking code. + */ + smp_wmb(); /* Could be smp_wmb__xxx(before|after)_spin_lock */ + + spin_lock(&mm->page_table_lock); + wait_split_huge_page = 0; + if (likely(pmd_none(*pmd))) { /* Has another populated it ? */ + mm->nr_ptes++; + pmd_populate(mm, pmd, new); + new = NULL; + } else if (unlikely(pmd_trans_splitting(*pmd))) + wait_split_huge_page = 1; + spin_unlock(&mm->page_table_lock); + if (new) + pte_free(mm, new); + if (wait_split_huge_page) + wait_split_huge_page(vma->anon_vma, pmd); + return 0; +} + +int __pte_alloc_kernel(pmd_t *pmd, unsigned long address) +{ + pte_t *new = pte_alloc_one_kernel(&init_mm, address); + if (!new) + return -ENOMEM; + + smp_wmb(); /* See comment in __pte_alloc */ + + spin_lock(&init_mm.page_table_lock); + if (likely(pmd_none(*pmd))) { /* Has another populated it ? */ + pmd_populate_kernel(&init_mm, pmd, new); + new = NULL; + } else + VM_BUG_ON(pmd_trans_splitting(*pmd)); + spin_unlock(&init_mm.page_table_lock); + if (new) + pte_free_kernel(&init_mm, new); + return 0; +} + +static inline void init_rss_vec(int *rss) +{ + memset(rss, 0, sizeof(int) * NR_MM_COUNTERS); +} + +static inline void add_mm_rss_vec(struct mm_struct *mm, int *rss) +{ + int i; + + if (current->mm == mm) + sync_mm_rss(mm); + for (i = 0; i < NR_MM_COUNTERS; i++) + if (rss[i]) + add_mm_counter(mm, i, rss[i]); +} + +/* + * This function is called to print an error when a bad pte + * is found. For example, we might have a PFN-mapped pte in + * a region that doesn't allow it. + * + * The calling function must still handle the error. + */ +static void print_bad_pte(struct vm_area_struct *vma, unsigned long addr, + pte_t pte, struct page *page) +{ + pgd_t *pgd = pgd_offset(vma->vm_mm, addr); + pud_t *pud = pud_offset(pgd, addr); + pmd_t *pmd = pmd_offset(pud, addr); + struct address_space *mapping; + pgoff_t index; + static unsigned long resume; + static unsigned long nr_shown; + static unsigned long nr_unshown; + + /* + * Allow a burst of 60 reports, then keep quiet for that minute; + * or allow a steady drip of one report per second. + */ + if (nr_shown == 60) { + if (time_before(jiffies, resume)) { + nr_unshown++; + return; + } + if (nr_unshown) { + printk(KERN_ALERT + "BUG: Bad page map: %lu messages suppressed\n", + nr_unshown); + nr_unshown = 0; + } + nr_shown = 0; + } + if (nr_shown++ == 0) + resume = jiffies + 60 * HZ; + + mapping = vma->vm_file ? vma->vm_file->f_mapping : NULL; + index = linear_page_index(vma, addr); + + printk(KERN_ALERT + "BUG: Bad page map in process %s pte:%08llx pmd:%08llx\n", + current->comm, + (long long)pte_val(pte), (long long)pmd_val(*pmd)); + if (page) + dump_page(page); + printk(KERN_ALERT + "addr:%p vm_flags:%08lx anon_vma:%p mapping:%p index:%lx\n", + (void *)addr, vma->vm_flags, vma->anon_vma, mapping, index); + /* + * Choose text because data symbols depend on CONFIG_KALLSYMS_ALL=y + */ + if (vma->vm_ops) + printk(KERN_ALERT "vma->vm_ops->fault: %pSR\n", + vma->vm_ops->fault); + if (vma->vm_file && vma->vm_file->f_op) + printk(KERN_ALERT "vma->vm_file->f_op->mmap: %pSR\n", + vma->vm_file->f_op->mmap); + dump_stack(); + add_taint(TAINT_BAD_PAGE, LOCKDEP_NOW_UNRELIABLE); +} + +static inline bool is_cow_mapping(vm_flags_t flags) +{ + return (flags & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE; +} + +/* + * vm_normal_page -- This function gets the "struct page" associated with a pte. + * + * "Special" mappings do not wish to be associated with a "struct page" (either + * it doesn't exist, or it exists but they don't want to touch it). In this + * case, NULL is returned here. "Normal" mappings do have a struct page. + * + * There are 2 broad cases. Firstly, an architecture may define a pte_special() + * pte bit, in which case this function is trivial. Secondly, an architecture + * may not have a spare pte bit, which requires a more complicated scheme, + * described below. + * + * A raw VM_PFNMAP mapping (ie. one that is not COWed) is always considered a + * special mapping (even if there are underlying and valid "struct pages"). + * COWed pages of a VM_PFNMAP are always normal. + * + * The way we recognize COWed pages within VM_PFNMAP mappings is through the + * rules set up by "remap_pfn_range()": the vma will have the VM_PFNMAP bit + * set, and the vm_pgoff will point to the first PFN mapped: thus every special + * mapping will always honor the rule + * + * pfn_of_page == vma->vm_pgoff + ((addr - vma->vm_start) >> PAGE_SHIFT) + * + * And for normal mappings this is false. + * + * This restricts such mappings to be a linear translation from virtual address + * to pfn. To get around this restriction, we allow arbitrary mappings so long + * as the vma is not a COW mapping; in that case, we know that all ptes are + * special (because none can have been COWed). + * + * + * In order to support COW of arbitrary special mappings, we have VM_MIXEDMAP. + * + * VM_MIXEDMAP mappings can likewise contain memory with or without "struct + * page" backing, however the difference is that _all_ pages with a struct + * page (that is, those where pfn_valid is true) are refcounted and considered + * normal pages by the VM. The disadvantage is that pages are refcounted + * (which can be slower and simply not an option for some PFNMAP users). The + * advantage is that we don't have to follow the strict linearity rule of + * PFNMAP mappings in order to support COWable mappings. + * + */ +#ifdef __HAVE_ARCH_PTE_SPECIAL +# define HAVE_PTE_SPECIAL 1 +#else +# define HAVE_PTE_SPECIAL 0 +#endif +struct page *vm_normal_page(struct vm_area_struct *vma, unsigned long addr, + pte_t pte) +{ + unsigned long pfn = pte_pfn(pte); + + if (HAVE_PTE_SPECIAL) { + if (likely(!pte_special(pte))) + goto check_pfn; + if (vma->vm_flags & (VM_PFNMAP | VM_MIXEDMAP)) + return NULL; + if (!is_zero_pfn(pfn)) + print_bad_pte(vma, addr, pte, NULL); + return NULL; + } + + /* !HAVE_PTE_SPECIAL case follows: */ + + if (unlikely(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP))) { + if (vma->vm_flags & VM_MIXEDMAP) { + if (!pfn_valid(pfn)) + return NULL; + goto out; + } else { + unsigned long off; + off = (addr - vma->vm_start) >> PAGE_SHIFT; + if (pfn == vma->vm_pgoff + off) + return NULL; + if (!is_cow_mapping(vma->vm_flags)) + return NULL; + } + } + + if (is_zero_pfn(pfn)) + return NULL; +check_pfn: + if (unlikely(pfn > highest_memmap_pfn)) { + print_bad_pte(vma, addr, pte, NULL); + return NULL; + } + + /* + * NOTE! We still have PageReserved() pages in the page tables. + * eg. VDSO mappings can cause them to exist. + */ +out: + return pfn_to_page(pfn); +} + +/* + * copy one vm_area from one task to the other. Assumes the page tables + * already present in the new task to be cleared in the whole range + * covered by this vma. + */ + +static inline unsigned long +copy_one_pte(struct mm_struct *dst_mm, struct mm_struct *src_mm, + pte_t *dst_pte, pte_t *src_pte, struct vm_area_struct *vma, + unsigned long addr, int *rss) +{ + unsigned long vm_flags = vma->vm_flags; + pte_t pte = *src_pte; + struct page *page; + + /* pte contains position in swap or file, so copy. */ + if (unlikely(!pte_present(pte))) { + if (!pte_file(pte)) { + swp_entry_t entry = pte_to_swp_entry(pte); + + if (swap_duplicate(entry) < 0) + return entry.val; + + /* make sure dst_mm is on swapoff's mmlist. */ + if (unlikely(list_empty(&dst_mm->mmlist))) { + spin_lock(&mmlist_lock); + if (list_empty(&dst_mm->mmlist)) + list_add(&dst_mm->mmlist, + &src_mm->mmlist); + spin_unlock(&mmlist_lock); + } + if (likely(!non_swap_entry(entry))) + rss[MM_SWAPENTS]++; + else if (is_migration_entry(entry)) { + page = migration_entry_to_page(entry); + + if (PageAnon(page)) + rss[MM_ANONPAGES]++; + else + rss[MM_FILEPAGES]++; + + if (is_write_migration_entry(entry) && + is_cow_mapping(vm_flags)) { + /* + * COW mappings require pages in both + * parent and child to be set to read. + */ + make_migration_entry_read(&entry); + pte = swp_entry_to_pte(entry); + set_pte_at(src_mm, addr, src_pte, pte); + } + } + } + goto out_set_pte; + } + + /* + * If it's a COW mapping, write protect it both + * in the parent and the child + */ + if (is_cow_mapping(vm_flags)) { + ptep_set_wrprotect(src_mm, addr, src_pte); + pte = pte_wrprotect(pte); + } + + /* + * If it's a shared mapping, mark it clean in + * the child + */ + if (vm_flags & VM_SHARED) + pte = pte_mkclean(pte); + pte = pte_mkold(pte); + + page = vm_normal_page(vma, addr, pte); + if (page) { + get_page(page); + page_dup_rmap(page); + if (PageAnon(page)) + rss[MM_ANONPAGES]++; + else + rss[MM_FILEPAGES]++; + } + +out_set_pte: + set_pte_at(dst_mm, addr, dst_pte, pte); + return 0; +} + +int copy_pte_range(struct mm_struct *dst_mm, struct mm_struct *src_mm, + pmd_t *dst_pmd, pmd_t *src_pmd, struct vm_area_struct *vma, + unsigned long addr, unsigned long end) +{ + pte_t *orig_src_pte, *orig_dst_pte; + pte_t *src_pte, *dst_pte; + spinlock_t *src_ptl, *dst_ptl; + int progress = 0; + int rss[NR_MM_COUNTERS]; + swp_entry_t entry = (swp_entry_t){0}; + +again: + init_rss_vec(rss); + + dst_pte = pte_alloc_map_lock(dst_mm, dst_pmd, addr, &dst_ptl); + if (!dst_pte) + return -ENOMEM; + src_pte = pte_offset_map(src_pmd, addr); + src_ptl = pte_lockptr(src_mm, src_pmd); + spin_lock_nested(src_ptl, SINGLE_DEPTH_NESTING); + orig_src_pte = src_pte; + orig_dst_pte = dst_pte; + arch_enter_lazy_mmu_mode(); + + do { + /* + * We are holding two locks at this point - either of them + * could generate latencies in another task on another CPU. + */ + if (progress >= 32) { + progress = 0; + if (need_resched() || + spin_needbreak(src_ptl) || spin_needbreak(dst_ptl)) + break; + } + if (pte_none(*src_pte)) { + progress++; + continue; + } + entry.val = copy_one_pte(dst_mm, src_mm, dst_pte, src_pte, + vma, addr, rss); + if (entry.val) + break; + progress += 8; + } while (dst_pte++, src_pte++, addr += PAGE_SIZE, addr != end); + + arch_leave_lazy_mmu_mode(); + spin_unlock(src_ptl); + pte_unmap(orig_src_pte); + add_mm_rss_vec(dst_mm, rss); + pte_unmap_unlock(orig_dst_pte, dst_ptl); + cond_resched(); + + if (entry.val) { + if (add_swap_count_continuation(entry, GFP_KERNEL) < 0) + return -ENOMEM; + progress = 0; + } + if (addr != end) + goto again; + return 0; +} + +static inline int copy_pmd_range(struct mm_struct *dst_mm, struct mm_struct *src_mm, + pud_t *dst_pud, pud_t *src_pud, struct vm_area_struct *vma, + unsigned long addr, unsigned long end) +{ + pmd_t *src_pmd, *dst_pmd; + unsigned long next; + + dst_pmd = pmd_alloc(dst_mm, dst_pud, addr); + if (!dst_pmd) + return -ENOMEM; + src_pmd = pmd_offset(src_pud, addr); + do { + next = pmd_addr_end(addr, end); + if (pmd_trans_huge(*src_pmd)) { + int err; + VM_BUG_ON(next-addr != HPAGE_PMD_SIZE); + err = copy_huge_pmd(dst_mm, src_mm, + dst_pmd, src_pmd, addr, vma); + if (err == -ENOMEM) + return -ENOMEM; + if (!err) + continue; + /* fall through */ + } + if (pmd_none_or_clear_bad(src_pmd)) + continue; + if (copy_pte_range(dst_mm, src_mm, dst_pmd, src_pmd, + vma, addr, next)) + return -ENOMEM; + } while (dst_pmd++, src_pmd++, addr = next, addr != end); + return 0; +} + +static inline int copy_pud_range(struct mm_struct *dst_mm, struct mm_struct *src_mm, + pgd_t *dst_pgd, pgd_t *src_pgd, struct vm_area_struct *vma, + unsigned long addr, unsigned long end) +{ + pud_t *src_pud, *dst_pud; + unsigned long next; + + dst_pud = pud_alloc(dst_mm, dst_pgd, addr); + if (!dst_pud) + return -ENOMEM; + src_pud = pud_offset(src_pgd, addr); + do { + next = pud_addr_end(addr, end); + if (pud_none_or_clear_bad(src_pud)) + continue; + if (copy_pmd_range(dst_mm, src_mm, dst_pud, src_pud, + vma, addr, next)) + return -ENOMEM; + } while (dst_pud++, src_pud++, addr = next, addr != end); + return 0; +} + +int copy_page_range(struct mm_struct *dst_mm, struct mm_struct *src_mm, + struct vm_area_struct *vma) +{ + pgd_t *src_pgd, *dst_pgd; + unsigned long next; + unsigned long addr = vma->vm_start; + unsigned long end = vma->vm_end; + unsigned long mmun_start; /* For mmu_notifiers */ + unsigned long mmun_end; /* For mmu_notifiers */ + bool is_cow; + int ret; + + /* + * Don't copy ptes where a page fault will fill them correctly. + * Fork becomes much lighter when there are big shared or private + * readonly mappings. The tradeoff is that copy_page_range is more + * efficient than faulting. + */ + if (!(vma->vm_flags & (VM_HUGETLB | VM_NONLINEAR | + VM_PFNMAP | VM_MIXEDMAP))) { + if (!vma->anon_vma) + return 0; + } + + if (is_vm_hugetlb_page(vma)) + return copy_hugetlb_page_range(dst_mm, src_mm, vma); + + if (unlikely(vma->vm_flags & VM_PFNMAP)) { + /* + * We do not free on error cases below as remove_vma + * gets called on error from higher level routine + */ + ret = track_pfn_copy(vma); + if (ret) + return ret; + } + + /* + * We need to invalidate the secondary MMU mappings only when + * there could be a permission downgrade on the ptes of the + * parent mm. And a permission downgrade will only happen if + * is_cow_mapping() returns true. + */ + is_cow = is_cow_mapping(vma->vm_flags); + mmun_start = addr; + mmun_end = end; + if (is_cow) + mmu_notifier_invalidate_range_start(src_mm, mmun_start, + mmun_end); + + ret = 0; + dst_pgd = pgd_offset(dst_mm, addr); + src_pgd = pgd_offset(src_mm, addr); + do { + next = pgd_addr_end(addr, end); + if (pgd_none_or_clear_bad(src_pgd)) + continue; + if (unlikely(copy_pud_range(dst_mm, src_mm, dst_pgd, src_pgd, + vma, addr, next))) { + ret = -ENOMEM; + break; + } + } while (dst_pgd++, src_pgd++, addr = next, addr != end); + + if (is_cow) + mmu_notifier_invalidate_range_end(src_mm, mmun_start, mmun_end); + return ret; +} + +static unsigned long zap_pte_range(struct mmu_gather *tlb, + struct vm_area_struct *vma, pmd_t *pmd, + unsigned long addr, unsigned long end, + struct zap_details *details) +{ + struct mm_struct *mm = tlb->mm; + int force_flush = 0; + int rss[NR_MM_COUNTERS]; + spinlock_t *ptl; + pte_t *start_pte; + pte_t *pte; + +again: + init_rss_vec(rss); + start_pte = pte_offset_map_lock(mm, pmd, addr, &ptl); + pte = start_pte; + arch_enter_lazy_mmu_mode(); + do { + pte_t ptent = *pte; + if (pte_none(ptent)) { + continue; + } + + if (pte_present(ptent)) { + struct page *page; + + page = vm_normal_page(vma, addr, ptent); + if (unlikely(details) && page) { + /* + * unmap_shared_mapping_pages() wants to + * invalidate cache without truncating: + * unmap shared but keep private pages. + */ + if (details->check_mapping && + details->check_mapping != page->mapping) + continue; + /* + * Each page->index must be checked when + * invalidating or truncating nonlinear. + */ + if (details->nonlinear_vma && + (page->index < details->first_index || + page->index > details->last_index)) + continue; + } + ptent = ptep_get_and_clear_full(mm, addr, pte, + tlb->fullmm); + tlb_remove_tlb_entry(tlb, pte, addr); + if (unlikely(!page)) + continue; + if (unlikely(details) && details->nonlinear_vma + && linear_page_index(details->nonlinear_vma, + addr) != page->index) + set_pte_at(mm, addr, pte, + pgoff_to_pte(page->index)); + if (PageAnon(page)) + rss[MM_ANONPAGES]--; + else { + if (pte_dirty(ptent)) + set_page_dirty(page); + if (pte_young(ptent) && + likely(!VM_SequentialReadHint(vma))) + mark_page_accessed(page); + rss[MM_FILEPAGES]--; + } + page_remove_rmap(page); + if (unlikely(page_mapcount(page) < 0)) + print_bad_pte(vma, addr, ptent, page); + force_flush = !__tlb_remove_page(tlb, page); + if (force_flush) + break; + continue; + } + /* + * If details->check_mapping, we leave swap entries; + * if details->nonlinear_vma, we leave file entries. + */ + if (unlikely(details)) + continue; + if (pte_file(ptent)) { + if (unlikely(!(vma->vm_flags & VM_NONLINEAR))) + print_bad_pte(vma, addr, ptent, NULL); + } else { + swp_entry_t entry = pte_to_swp_entry(ptent); + + if (!non_swap_entry(entry)) + rss[MM_SWAPENTS]--; + else if (is_migration_entry(entry)) { + struct page *page; + + page = migration_entry_to_page(entry); + + if (PageAnon(page)) + rss[MM_ANONPAGES]--; + else + rss[MM_FILEPAGES]--; + } + if (unlikely(!free_swap_and_cache(entry))) + print_bad_pte(vma, addr, ptent, NULL); + } + pte_clear_not_present_full(mm, addr, pte, tlb->fullmm); + } while (pte++, addr += PAGE_SIZE, addr != end); + + add_mm_rss_vec(mm, rss); + arch_leave_lazy_mmu_mode(); + pte_unmap_unlock(start_pte, ptl); + + /* + * mmu_gather ran out of room to batch pages, we break out of + * the PTE lock to avoid doing the potential expensive TLB invalidate + * and page-free while holding it. + */ + if (force_flush) { + unsigned long old_end; + + force_flush = 0; + + /* + * Flush the TLB just for the previous segment, + * then update the range to be the remaining + * TLB range. + */ + old_end = tlb->end; + tlb->end = addr; + + tlb_flush_mmu(tlb); + + tlb->start = addr; + tlb->end = old_end; + + if (addr != end) + goto again; + } + + return addr; +} + +static inline unsigned long zap_pmd_range(struct mmu_gather *tlb, + struct vm_area_struct *vma, pud_t *pud, + unsigned long addr, unsigned long end, + struct zap_details *details) +{ + pmd_t *pmd; + unsigned long next; + + pmd = pmd_offset(pud, addr); + do { + next = pmd_addr_end(addr, end); + if (pmd_trans_huge(*pmd)) { + if (next - addr != HPAGE_PMD_SIZE) { +#ifdef CONFIG_DEBUG_VM + if (!rwsem_is_locked(&tlb->mm->mmap_sem)) { + pr_err("%s: mmap_sem is unlocked! addr=0x%lx end=0x%lx vma->vm_start=0x%lx vma->vm_end=0x%lx\n", + __func__, addr, end, + vma->vm_start, + vma->vm_end); + BUG(); + } +#endif + split_huge_page_pmd(vma, addr, pmd); + } else if (zap_huge_pmd(tlb, vma, pmd, addr)) + goto next; + /* fall through */ + } + /* + * Here there can be other concurrent MADV_DONTNEED or + * trans huge page faults running, and if the pmd is + * none or trans huge it can change under us. This is + * because MADV_DONTNEED holds the mmap_sem in read + * mode. + */ + if (pmd_none_or_trans_huge_or_clear_bad(pmd)) + goto next; + next = zap_pte_range(tlb, vma, pmd, addr, next, details); +next: + cond_resched(); + } while (pmd++, addr = next, addr != end); + + return addr; +} + +static inline unsigned long zap_pud_range(struct mmu_gather *tlb, + struct vm_area_struct *vma, pgd_t *pgd, + unsigned long addr, unsigned long end, + struct zap_details *details) +{ + pud_t *pud; + unsigned long next; + + pud = pud_offset(pgd, addr); + do { + next = pud_addr_end(addr, end); + if (pud_none_or_clear_bad(pud)) + continue; + next = zap_pmd_range(tlb, vma, pud, addr, next, details); + } while (pud++, addr = next, addr != end); + + return addr; +} + +static void unmap_page_range(struct mmu_gather *tlb, + struct vm_area_struct *vma, + unsigned long addr, unsigned long end, + struct zap_details *details) +{ + pgd_t *pgd; + unsigned long next; + + if (details && !details->check_mapping && !details->nonlinear_vma) + details = NULL; + + BUG_ON(addr >= end); + mem_cgroup_uncharge_start(); + tlb_start_vma(tlb, vma); + pgd = pgd_offset(vma->vm_mm, addr); + do { + next = pgd_addr_end(addr, end); + if (pgd_none_or_clear_bad(pgd)) + continue; + next = zap_pud_range(tlb, vma, pgd, addr, next, details); + } while (pgd++, addr = next, addr != end); + tlb_end_vma(tlb, vma); + mem_cgroup_uncharge_end(); +} + + +static void unmap_single_vma(struct mmu_gather *tlb, + struct vm_area_struct *vma, unsigned long start_addr, + unsigned long end_addr, + struct zap_details *details) +{ + unsigned long start = max(vma->vm_start, start_addr); + unsigned long end; + + if (start >= vma->vm_end) + return; + end = min(vma->vm_end, end_addr); + if (end <= vma->vm_start) + return; + + if (vma->vm_file) + uprobe_munmap(vma, start, end); + + if (unlikely(vma->vm_flags & VM_PFNMAP)) + untrack_pfn(vma, 0, 0); + + if (start != end) { + if (unlikely(is_vm_hugetlb_page(vma))) { + /* + * It is undesirable to test vma->vm_file as it + * should be non-null for valid hugetlb area. + * However, vm_file will be NULL in the error + * cleanup path of do_mmap_pgoff. When + * hugetlbfs ->mmap method fails, + * do_mmap_pgoff() nullifies vma->vm_file + * before calling this function to clean up. + * Since no pte has actually been setup, it is + * safe to do nothing in this case. + */ + if (vma->vm_file) { + mutex_lock(&vma->vm_file->f_mapping->i_mmap_mutex); + __unmap_hugepage_range_final(tlb, vma, start, end, NULL); + mutex_unlock(&vma->vm_file->f_mapping->i_mmap_mutex); + } + } else + unmap_page_range(tlb, vma, start, end, details); + } +} + +/** + * unmap_vmas - unmap a range of memory covered by a list of vma's + * @tlb: address of the caller's struct mmu_gather + * @vma: the starting vma + * @start_addr: virtual address at which to start unmapping + * @end_addr: virtual address at which to end unmapping + * + * Unmap all pages in the vma list. + * + * Only addresses between `start' and `end' will be unmapped. + * + * The VMA list must be sorted in ascending virtual address order. + * + * unmap_vmas() assumes that the caller will flush the whole unmapped address + * range after unmap_vmas() returns. So the only responsibility here is to + * ensure that any thus-far unmapped pages are flushed before unmap_vmas() + * drops the lock and schedules. + */ +void unmap_vmas(struct mmu_gather *tlb, + struct vm_area_struct *vma, unsigned long start_addr, + unsigned long end_addr) +{ + struct mm_struct *mm = vma->vm_mm; + + mmu_notifier_invalidate_range_start(mm, start_addr, end_addr); + for ( ; vma && vma->vm_start < end_addr; vma = vma->vm_next) + unmap_single_vma(tlb, vma, start_addr, end_addr, NULL); + mmu_notifier_invalidate_range_end(mm, start_addr, end_addr); +} + +/** + * zap_page_range - remove user pages in a given range + * @vma: vm_area_struct holding the applicable pages + * @start: starting address of pages to zap + * @size: number of bytes to zap + * @details: details of nonlinear truncation or shared cache invalidation + * + * Caller must protect the VMA list + */ +void zap_page_range(struct vm_area_struct *vma, unsigned long start, + unsigned long size, struct zap_details *details) +{ + struct mm_struct *mm = vma->vm_mm; + struct mmu_gather tlb; + unsigned long end = start + size; + + lru_add_drain(); + tlb_gather_mmu(&tlb, mm, start, end); + update_hiwater_rss(mm); + mmu_notifier_invalidate_range_start(mm, start, end); + for ( ; vma && vma->vm_start < end; vma = vma->vm_next) + unmap_single_vma(&tlb, vma, start, end, details); + mmu_notifier_invalidate_range_end(mm, start, end); + tlb_finish_mmu(&tlb, start, end); +} + +/** + * zap_page_range_single - remove user pages in a given range + * @vma: vm_area_struct holding the applicable pages + * @address: starting address of pages to zap + * @size: number of bytes to zap + * @details: details of nonlinear truncation or shared cache invalidation + * + * The range must fit into one VMA. + */ +static void zap_page_range_single(struct vm_area_struct *vma, unsigned long address, + unsigned long size, struct zap_details *details) +{ + struct mm_struct *mm = vma->vm_mm; + struct mmu_gather tlb; + unsigned long end = address + size; + + lru_add_drain(); + tlb_gather_mmu(&tlb, mm, address, end); + update_hiwater_rss(mm); + mmu_notifier_invalidate_range_start(mm, address, end); + unmap_single_vma(&tlb, vma, address, end, details); + mmu_notifier_invalidate_range_end(mm, address, end); + tlb_finish_mmu(&tlb, address, end); +} + +/** + * zap_vma_ptes - remove ptes mapping the vma + * @vma: vm_area_struct holding ptes to be zapped + * @address: starting address of pages to zap + * @size: number of bytes to zap + * + * This function only unmaps ptes assigned to VM_PFNMAP vmas. + * + * The entire address range must be fully contained within the vma. + * + * Returns 0 if successful. + */ +int zap_vma_ptes(struct vm_area_struct *vma, unsigned long address, + unsigned long size) +{ + if (address < vma->vm_start || address + size > vma->vm_end || + !(vma->vm_flags & VM_PFNMAP)) + return -1; + zap_page_range_single(vma, address, size, NULL); + return 0; +} +EXPORT_SYMBOL_GPL(zap_vma_ptes); + +/** + * follow_page_mask - look up a page descriptor from a user-virtual address + * @vma: vm_area_struct mapping @address + * @address: virtual address to look up + * @flags: flags modifying lookup behaviour + * @page_mask: on output, *page_mask is set according to the size of the page + * + * @flags can have FOLL_ flags set, defined in + * + * Returns the mapped (struct page *), %NULL if no mapping exists, or + * an error pointer if there is a mapping to something not represented + * by a page descriptor (see also vm_normal_page()). + */ +struct page *follow_page_mask(struct vm_area_struct *vma, + unsigned long address, unsigned int flags, + unsigned int *page_mask) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *ptep, pte; + spinlock_t *ptl; + struct page *page; + struct mm_struct *mm; + + if (WARN_ON(!vma)) + return NULL; + + mm = vma->vm_mm; + + *page_mask = 0; + + page = follow_huge_addr(mm, address, flags & FOLL_WRITE); + if (!IS_ERR(page)) { + BUG_ON(flags & FOLL_GET); + goto out; + } + + page = NULL; + pgd = pgd_offset(mm, address); + if (pgd_none(*pgd) || unlikely(pgd_bad(*pgd))) + goto no_page_table; + + pud = pud_offset(pgd, address); + if (pud_none(*pud)) + goto no_page_table; + if (pud_huge(*pud) && vma->vm_flags & VM_HUGETLB) { + BUG_ON(flags & FOLL_GET); + page = follow_huge_pud(mm, address, pud, flags & FOLL_WRITE); + goto out; + } + if (unlikely(pud_bad(*pud))) + goto no_page_table; + + pmd = pmd_offset(pud, address); + if (pmd_none(*pmd)) + goto no_page_table; + if (pmd_huge(*pmd) && vma->vm_flags & VM_HUGETLB) { + BUG_ON(flags & FOLL_GET); + page = follow_huge_pmd(mm, address, pmd, flags & FOLL_WRITE); + goto out; + } + if ((flags & FOLL_NUMA) && pmd_numa(*pmd)) + goto no_page_table; + if (pmd_trans_huge(*pmd)) { + if (flags & FOLL_SPLIT) { + split_huge_page_pmd(vma, address, pmd); + goto split_fallthrough; + } + spin_lock(&mm->page_table_lock); + if (likely(pmd_trans_huge(*pmd))) { + if (unlikely(pmd_trans_splitting(*pmd))) { + spin_unlock(&mm->page_table_lock); + wait_split_huge_page(vma->anon_vma, pmd); + } else { + page = follow_trans_huge_pmd(vma, address, + pmd, flags); + spin_unlock(&mm->page_table_lock); + *page_mask = HPAGE_PMD_NR - 1; + goto out; + } + } else + spin_unlock(&mm->page_table_lock); + /* fall through */ + } +split_fallthrough: + if (unlikely(pmd_bad(*pmd))) + goto no_page_table; + + ptep = pte_offset_map_lock(mm, pmd, address, &ptl); + + pte = *ptep; + if (!pte_present(pte)) { + swp_entry_t entry; + /* + * KSM's break_ksm() relies upon recognizing a ksm page + * even while it is being migrated, so for that case we + * need migration_entry_wait(). + */ + if (likely(!(flags & FOLL_MIGRATION))) + goto no_page; + if (pte_none(pte) || pte_file(pte)) + goto no_page; + entry = pte_to_swp_entry(pte); + if (!is_migration_entry(entry)) + goto no_page; + pte_unmap_unlock(ptep, ptl); + migration_entry_wait(mm, pmd, address); + goto split_fallthrough; + } + if ((flags & FOLL_NUMA) && pte_numa(pte)) + goto no_page; + if ((flags & FOLL_WRITE) && !pte_write(pte)) + goto unlock; + + page = vm_normal_page(vma, address, pte); + if (unlikely(!page)) { + if ((flags & FOLL_DUMP) || + !is_zero_pfn(pte_pfn(pte))) + goto bad_page; + page = pte_page(pte); + } + + if (flags & FOLL_GET) + get_page_foll(page); + if (flags & FOLL_TOUCH) { + if ((flags & FOLL_WRITE) && + !pte_dirty(pte) && !PageDirty(page)) + set_page_dirty(page); + /* + * pte_mkyoung() would be more correct here, but atomic care + * is needed to avoid losing the dirty bit: it is easier to use + * mark_page_accessed(). + */ + mark_page_accessed(page); + } + if ((flags & FOLL_MLOCK) && (vma->vm_flags & VM_LOCKED)) { + /* + * The preliminary mapping check is mainly to avoid the + * pointless overhead of lock_page on the ZERO_PAGE + * which might bounce very badly if there is contention. + * + * If the page is already locked, we don't need to + * handle it now - vmscan will handle it later if and + * when it attempts to reclaim the page. + */ + if (page->mapping && trylock_page(page)) { + lru_add_drain(); /* push cached pages to LRU */ + /* + * Because we lock page here, and migration is + * blocked by the pte's page reference, and we + * know the page is still mapped, we don't even + * need to check for file-cache page truncation. + */ + mlock_vma_page(page); + unlock_page(page); + } + } +unlock: + pte_unmap_unlock(ptep, ptl); +out: + return page; + +bad_page: + pte_unmap_unlock(ptep, ptl); + return ERR_PTR(-EFAULT); + +no_page: + pte_unmap_unlock(ptep, ptl); + if (!pte_none(pte)) + return page; + +no_page_table: + /* + * When core dumping an enormous anonymous area that nobody + * has touched so far, we don't want to allocate unnecessary pages or + * page tables. Return error instead of NULL to skip handle_mm_fault, + * then get_dump_page() will return NULL to leave a hole in the dump. + * But we can only make this optimization where a hole would surely + * be zero-filled if handle_mm_fault() actually did handle it. + */ + if ((flags & FOLL_DUMP) && + (!vma->vm_ops || !vma->vm_ops->fault)) + return ERR_PTR(-EFAULT); + return page; +} + +static inline int stack_guard_page(struct vm_area_struct *vma, unsigned long addr) +{ + return stack_guard_page_start(vma, addr) || + stack_guard_page_end(vma, addr+PAGE_SIZE); +} + +/** + * replace_cma_page() - migrate page out of CMA page blocks + * @page: source page to be migrated + * + * Returns either the old page (if migration was not possible) or the pointer + * to the newly allocated page (with additional reference taken). + * + * get_user_pages() might take a reference to a page for a long period of time, + * what prevent such page from migration. This is fatal to the preffered usage + * pattern of CMA pageblocks. This function replaces the given user page with + * a new one allocated from NON-MOVABLE pageblock, so locking CMA page can be + * avoided. + */ +static inline struct page *migrate_replace_cma_page(struct page *page) +{ + struct page *newpage = alloc_page(GFP_HIGHUSER); + + if (!newpage) + goto out; + + /* + * Take additional reference to the new page to ensure it won't get + * freed after migration procedure end. + */ + get_page_foll(newpage); + + if (migrate_replace_page(page, newpage) == 0) + return newpage; + + put_page(newpage); + __free_page(newpage); +out: + /* + * Migration errors in case of get_user_pages() might not + * be fatal to CMA itself, so better don't fail here. + */ + return page; +} + +/** + * __get_user_pages() - pin user pages in memory + * @tsk: task_struct of target task + * @mm: mm_struct of target mm + * @start: starting user address + * @nr_pages: number of pages from start to pin + * @gup_flags: flags modifying pin behaviour + * @pages: array that receives pointers to the pages pinned. + * Should be at least nr_pages long. Or NULL, if caller + * only intends to ensure the pages are faulted in. + * @vmas: array of pointers to vmas corresponding to each page. + * Or NULL if the caller does not require them. + * @nonblocking: whether waiting for disk IO or mmap_sem contention + * + * Returns number of pages pinned. This may be fewer than the number + * requested. If nr_pages is 0 or negative, returns 0. If no pages + * were pinned, returns -errno. Each page returned must be released + * with a put_page() call when it is finished with. vmas will only + * remain valid while mmap_sem is held. + * + * Must be called with mmap_sem held for read or write. + * + * __get_user_pages walks a process's page tables and takes a reference to + * each struct page that each user address corresponds to at a given + * instant. That is, it takes the page that would be accessed if a user + * thread accesses the given user virtual address at that instant. + * + * This does not guarantee that the page exists in the user mappings when + * __get_user_pages returns, and there may even be a completely different + * page there in some cases (eg. if mmapped pagecache has been invalidated + * and subsequently re faulted). However it does guarantee that the page + * won't be freed completely. And mostly callers simply care that the page + * contains data that was valid *at some point in time*. Typically, an IO + * or similar operation cannot guarantee anything stronger anyway because + * locks can't be held over the syscall boundary. + * + * If @gup_flags & FOLL_WRITE == 0, the page must not be written to. If + * the page is written to, set_page_dirty (or set_page_dirty_lock, as + * appropriate) must be called after the page is finished with, and + * before put_page is called. + * + * If @nonblocking != NULL, __get_user_pages will not wait for disk IO + * or mmap_sem contention, and if waiting is needed to pin all pages, + * *@nonblocking will be set to 0. + * + * In most cases, get_user_pages or get_user_pages_fast should be used + * instead of __get_user_pages. __get_user_pages should be used only if + * you need some special @gup_flags. + */ +long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm, + unsigned long start, unsigned long nr_pages, + unsigned int gup_flags, struct page **pages, + struct vm_area_struct **vmas, int *nonblocking) +{ + long i; + unsigned long vm_flags; + unsigned int page_mask; + + if (!nr_pages) + return 0; + + VM_BUG_ON(!!pages != !!(gup_flags & FOLL_GET)); + + /* + * Require read or write permissions. + * If FOLL_FORCE is set, we only require the "MAY" flags. + */ + vm_flags = (gup_flags & FOLL_WRITE) ? + (VM_WRITE | VM_MAYWRITE) : (VM_READ | VM_MAYREAD); + vm_flags &= (gup_flags & FOLL_FORCE) ? + (VM_MAYREAD | VM_MAYWRITE) : (VM_READ | VM_WRITE); + + /* + * If FOLL_FORCE and FOLL_NUMA are both set, handle_mm_fault + * would be called on PROT_NONE ranges. We must never invoke + * handle_mm_fault on PROT_NONE ranges or the NUMA hinting + * page faults would unprotect the PROT_NONE ranges if + * _PAGE_NUMA and _PAGE_PROTNONE are sharing the same pte/pmd + * bitflag. So to avoid that, don't set FOLL_NUMA if + * FOLL_FORCE is set. + */ + if (!(gup_flags & FOLL_FORCE)) + gup_flags |= FOLL_NUMA; + + i = 0; + + do { + struct vm_area_struct *vma; + + vma = find_extend_vma(mm, start); + if (!vma && in_gate_area(mm, start)) { + unsigned long pg = start & PAGE_MASK; + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + + /* user gate pages are read-only */ + if (gup_flags & FOLL_WRITE) + return i ? : -EFAULT; + if (pg > TASK_SIZE) + pgd = pgd_offset_k(pg); + else + pgd = pgd_offset_gate(mm, pg); + BUG_ON(pgd_none(*pgd)); + pud = pud_offset(pgd, pg); + BUG_ON(pud_none(*pud)); + pmd = pmd_offset(pud, pg); + if (pmd_none(*pmd)) + return i ? : -EFAULT; + VM_BUG_ON(pmd_trans_huge(*pmd)); + pte = pte_offset_map(pmd, pg); + if (pte_none(*pte)) { + pte_unmap(pte); + return i ? : -EFAULT; + } + vma = get_gate_vma(mm); + if (pages) { + struct page *page; + + page = vm_normal_page(vma, start, *pte); + if (!page) { + if (!(gup_flags & FOLL_DUMP) && + is_zero_pfn(pte_pfn(*pte))) + page = pte_page(*pte); + else { + pte_unmap(pte); + return i ? : -EFAULT; + } + } + pages[i] = page; + get_page(page); + } + pte_unmap(pte); + page_mask = 0; + goto next_page; + } + + if (!vma || + (vma->vm_flags & (VM_IO | VM_PFNMAP)) || + !(vm_flags & vma->vm_flags)) + return i ? : -EFAULT; + + if (is_vm_hugetlb_page(vma)) { + i = follow_hugetlb_page(mm, vma, pages, vmas, + &start, &nr_pages, i, gup_flags); + continue; + } + + do { + struct page *page; + unsigned int foll_flags = gup_flags; + unsigned int page_increm; + + /* + * If we have a pending SIGKILL, don't keep faulting + * pages and potentially allocating memory. + */ + if (unlikely(fatal_signal_pending(current))) + return i ? i : -ERESTARTSYS; + + cond_resched(); + while (!(page = follow_page_mask(vma, start, + foll_flags, &page_mask))) { + int ret; + unsigned int fault_flags = 0; + + if (gup_flags & FOLL_DURABLE) + fault_flags = FAULT_FLAG_NO_CMA; + + /* For mlock, just skip the stack guard page. */ + if (foll_flags & FOLL_MLOCK) { + if (stack_guard_page(vma, start)) + goto next_page; + } + if (foll_flags & FOLL_WRITE) + fault_flags |= FAULT_FLAG_WRITE; + if (nonblocking) + fault_flags |= FAULT_FLAG_ALLOW_RETRY; + if (foll_flags & FOLL_NOWAIT) + fault_flags |= (FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_RETRY_NOWAIT); + + ret = handle_mm_fault(mm, vma, start, + fault_flags); + + if (ret & VM_FAULT_ERROR) { + if (ret & VM_FAULT_OOM) + return i ? i : -ENOMEM; + if (ret & (VM_FAULT_HWPOISON | + VM_FAULT_HWPOISON_LARGE)) { + if (i) + return i; + else if (gup_flags & FOLL_HWPOISON) + return -EHWPOISON; + else + return -EFAULT; + } + if (ret & VM_FAULT_SIGBUS) + return i ? i : -EFAULT; + BUG(); + } + + if (tsk) { + if (ret & VM_FAULT_MAJOR) + tsk->maj_flt++; + else + tsk->min_flt++; + } + + if (ret & VM_FAULT_RETRY) { + if (nonblocking) + *nonblocking = 0; + return i; + } + + /* + * The VM_FAULT_WRITE bit tells us that + * do_wp_page has broken COW when necessary, + * even if maybe_mkwrite decided not to set + * pte_write. We can thus safely do subsequent + * page lookups as if they were reads. But only + * do so when looping for pte_write is futile: + * in some cases userspace may also be wanting + * to write to the gotten user page, which a + * read fault here might prevent (a readonly + * page might get reCOWed by userspace write). + */ + if ((ret & VM_FAULT_WRITE) && + !(vma->vm_flags & VM_WRITE)) + foll_flags &= ~FOLL_WRITE; + + cond_resched(); + } + if (IS_ERR(page)) + return i ? i : PTR_ERR(page); + + if ((gup_flags & FOLL_DURABLE) && is_cma_page(page)) + page = migrate_replace_cma_page(page); + + if (pages) { + pages[i] = page; + + flush_anon_page(vma, page, start); + flush_dcache_page(page); + page_mask = 0; + } +next_page: + if (vmas) { + vmas[i] = vma; + page_mask = 0; + } + page_increm = 1 + (~(start >> PAGE_SHIFT) & page_mask); + if (page_increm > nr_pages) + page_increm = nr_pages; + i += page_increm; + start += page_increm * PAGE_SIZE; + nr_pages -= page_increm; + } while (nr_pages && start < vma->vm_end); + } while (nr_pages); + return i; +} +EXPORT_SYMBOL(__get_user_pages); + +/* + * fixup_user_fault() - manually resolve a user page fault + * @tsk: the task_struct to use for page fault accounting, or + * NULL if faults are not to be recorded. + * @mm: mm_struct of target mm + * @address: user address + * @fault_flags:flags to pass down to handle_mm_fault() + * + * This is meant to be called in the specific scenario where for locking reasons + * we try to access user memory in atomic context (within a pagefault_disable() + * section), this returns -EFAULT, and we want to resolve the user fault before + * trying again. + * + * Typically this is meant to be used by the futex code. + * + * The main difference with get_user_pages() is that this function will + * unconditionally call handle_mm_fault() which will in turn perform all the + * necessary SW fixup of the dirty and young bits in the PTE, while + * handle_mm_fault() only guarantees to update these in the struct page. + * + * This is important for some architectures where those bits also gate the + * access permission to the page because they are maintained in software. On + * such architectures, gup() will not be enough to make a subsequent access + * succeed. + * + * This should be called with the mm_sem held for read. + */ +int fixup_user_fault(struct task_struct *tsk, struct mm_struct *mm, + unsigned long address, unsigned int fault_flags) +{ + struct vm_area_struct *vma; + int ret; + + vma = find_extend_vma(mm, address); + if (!vma || address < vma->vm_start) + return -EFAULT; + + ret = handle_mm_fault(mm, vma, address, fault_flags); + if (ret & VM_FAULT_ERROR) { + if (ret & VM_FAULT_OOM) + return -ENOMEM; + if (ret & (VM_FAULT_HWPOISON | VM_FAULT_HWPOISON_LARGE)) + return -EHWPOISON; + if (ret & VM_FAULT_SIGBUS) + return -EFAULT; + BUG(); + } + if (tsk) { + if (ret & VM_FAULT_MAJOR) + tsk->maj_flt++; + else + tsk->min_flt++; + } + return 0; +} + +/** + * get_dump_page() - pin user page in memory while writing it to core dump + * @addr: user address + * + * Returns struct page pointer of user page pinned for dump, + * to be freed afterwards by page_cache_release() or put_page(). + * + * Returns NULL on any kind of failure - a hole must then be inserted into + * the corefile, to preserve alignment with its headers; and also returns + * NULL wherever the ZERO_PAGE, or an anonymous pte_none, has been found - + * allowing a hole to be left in the corefile to save diskspace. + * + * Called without mmap_sem, but after all other threads have been killed. + */ +#ifdef CONFIG_ELF_CORE +struct page *get_dump_page(unsigned long addr) +{ + struct vm_area_struct *vma; + struct page *page; + + if (__get_user_pages(current, current->mm, addr, 1, + FOLL_FORCE | FOLL_DUMP | FOLL_GET, &page, &vma, + NULL) < 1) + return NULL; + flush_cache_page(vma, addr, page_to_pfn(page)); + return page; +} +#endif /* CONFIG_ELF_CORE */ + +pte_t *__get_locked_pte(struct mm_struct *mm, unsigned long addr, + spinlock_t **ptl) +{ + pgd_t * pgd = pgd_offset(mm, addr); + pud_t * pud = pud_alloc(mm, pgd, addr); + if (pud) { + pmd_t * pmd = pmd_alloc(mm, pud, addr); + if (pmd) { + VM_BUG_ON(pmd_trans_huge(*pmd)); + return pte_alloc_map_lock(mm, pmd, addr, ptl); + } + } + return NULL; +} + +/* + * This is the old fallback for page remapping. + * + * For historical reasons, it only allows reserved pages. Only + * old drivers should use this, and they needed to mark their + * pages reserved for the old functions anyway. + */ +static int insert_page(struct vm_area_struct *vma, unsigned long addr, + struct page *page, pgprot_t prot) +{ + struct mm_struct *mm = vma->vm_mm; + int retval; + pte_t *pte; + spinlock_t *ptl; + + retval = -EINVAL; + if (PageAnon(page)) + goto out; + retval = -ENOMEM; + flush_dcache_page(page); + pte = get_locked_pte(mm, addr, &ptl); + if (!pte) + goto out; + retval = -EBUSY; + if (!pte_none(*pte)) + goto out_unlock; + + /* Ok, finally just insert the thing.. */ + get_page(page); + inc_mm_counter_fast(mm, MM_FILEPAGES); + page_add_file_rmap(page); + set_pte_at(mm, addr, pte, mk_pte(page, prot)); + + retval = 0; + pte_unmap_unlock(pte, ptl); + return retval; +out_unlock: + pte_unmap_unlock(pte, ptl); +out: + return retval; +} + +/** + * vm_insert_page - insert single page into user vma + * @vma: user vma to map to + * @addr: target user address of this page + * @page: source kernel page + * + * This allows drivers to insert individual pages they've allocated + * into a user vma. + * + * The page has to be a nice clean _individual_ kernel allocation. + * If you allocate a compound page, you need to have marked it as + * such (__GFP_COMP), or manually just split the page up yourself + * (see split_page()). + * + * NOTE! Traditionally this was done with "remap_pfn_range()" which + * took an arbitrary page protection parameter. This doesn't allow + * that. Your vma protection will have to be set up correctly, which + * means that if you want a shared writable mapping, you'd better + * ask for a shared writable mapping! + * + * The page does not need to be reserved. + * + * Usually this function is called from f_op->mmap() handler + * under mm->mmap_sem write-lock, so it can change vma->vm_flags. + * Caller must set VM_MIXEDMAP on vma if it wants to call this + * function from other places, for example from page-fault handler. + */ +int vm_insert_page(struct vm_area_struct *vma, unsigned long addr, + struct page *page) +{ + if (addr < vma->vm_start || addr >= vma->vm_end) + return -EFAULT; + if (!page_count(page)) + return -EINVAL; + if (!(vma->vm_flags & VM_MIXEDMAP)) { + BUG_ON(down_read_trylock(&vma->vm_mm->mmap_sem)); + BUG_ON(vma->vm_flags & VM_PFNMAP); + vma->vm_flags |= VM_MIXEDMAP; + } + return insert_page(vma, addr, page, vma->vm_page_prot); +} +EXPORT_SYMBOL(vm_insert_page); + +static int insert_pfn(struct vm_area_struct *vma, unsigned long addr, + unsigned long pfn, pgprot_t prot) +{ + struct mm_struct *mm = vma->vm_mm; + int retval; + pte_t *pte, entry; + spinlock_t *ptl; + + retval = -ENOMEM; + pte = get_locked_pte(mm, addr, &ptl); + if (!pte) + goto out; + retval = -EBUSY; + if (!pte_none(*pte)) + goto out_unlock; + + /* Ok, finally just insert the thing.. */ + entry = pte_mkspecial(pfn_pte(pfn, prot)); + set_pte_at(mm, addr, pte, entry); + update_mmu_cache(vma, addr, pte); /* XXX: why not for insert_page? */ + + retval = 0; +out_unlock: + pte_unmap_unlock(pte, ptl); +out: + return retval; +} + +/** + * vm_insert_pfn - insert single pfn into user vma + * @vma: user vma to map to + * @addr: target user address of this page + * @pfn: source kernel pfn + * + * Similar to vm_insert_page, this allows drivers to insert individual pages + * they've allocated into a user vma. Same comments apply. + * + * This function should only be called from a vm_ops->fault handler, and + * in that case the handler should return NULL. + * + * vma cannot be a COW mapping. + * + * As this is called only for pages that do not currently exist, we + * do not need to flush old virtual caches or the TLB. + */ +int vm_insert_pfn(struct vm_area_struct *vma, unsigned long addr, + unsigned long pfn) +{ + int ret; + pgprot_t pgprot = vma->vm_page_prot; + /* + * Technically, architectures with pte_special can avoid all these + * restrictions (same for remap_pfn_range). However we would like + * consistency in testing and feature parity among all, so we should + * try to keep these invariants in place for everybody. + */ + BUG_ON(!(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP))); + BUG_ON((vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP)) == + (VM_PFNMAP|VM_MIXEDMAP)); + BUG_ON((vma->vm_flags & VM_PFNMAP) && is_cow_mapping(vma->vm_flags)); + BUG_ON((vma->vm_flags & VM_MIXEDMAP) && pfn_valid(pfn)); + + if (addr < vma->vm_start || addr >= vma->vm_end) + return -EFAULT; + if (track_pfn_insert(vma, &pgprot, pfn)) + return -EINVAL; + + ret = insert_pfn(vma, addr, pfn, pgprot); + + return ret; +} +EXPORT_SYMBOL(vm_insert_pfn); + +int vm_insert_mixed(struct vm_area_struct *vma, unsigned long addr, + unsigned long pfn) +{ + BUG_ON(!(vma->vm_flags & VM_MIXEDMAP)); + + if (addr < vma->vm_start || addr >= vma->vm_end) + return -EFAULT; + + /* + * If we don't have pte special, then we have to use the pfn_valid() + * based VM_MIXEDMAP scheme (see vm_normal_page), and thus we *must* + * refcount the page if pfn_valid is true (hence insert_page rather + * than insert_pfn). If a zero_pfn were inserted into a VM_MIXEDMAP + * without pte special, it would there be refcounted as a normal page. + */ + if (!HAVE_PTE_SPECIAL && pfn_valid(pfn)) { + struct page *page; + + page = pfn_to_page(pfn); + return insert_page(vma, addr, page, vma->vm_page_prot); + } + return insert_pfn(vma, addr, pfn, vma->vm_page_prot); +} +EXPORT_SYMBOL(vm_insert_mixed); + +/* + * maps a range of physical memory into the requested pages. the old + * mappings are removed. any references to nonexistent pages results + * in null mappings (currently treated as "copy-on-access") + */ +static int remap_pte_range(struct mm_struct *mm, pmd_t *pmd, + unsigned long addr, unsigned long end, + unsigned long pfn, pgprot_t prot) +{ + pte_t *pte; + spinlock_t *ptl; + + pte = pte_alloc_map_lock(mm, pmd, addr, &ptl); + if (!pte) + return -ENOMEM; + arch_enter_lazy_mmu_mode(); + do { + BUG_ON(!pte_none(*pte)); + set_pte_at(mm, addr, pte, pte_mkspecial(pfn_pte(pfn, prot))); + pfn++; + } while (pte++, addr += PAGE_SIZE, addr != end); + arch_leave_lazy_mmu_mode(); + pte_unmap_unlock(pte - 1, ptl); + return 0; +} + +static inline int remap_pmd_range(struct mm_struct *mm, pud_t *pud, + unsigned long addr, unsigned long end, + unsigned long pfn, pgprot_t prot) +{ + pmd_t *pmd; + unsigned long next; + + pfn -= addr >> PAGE_SHIFT; + pmd = pmd_alloc(mm, pud, addr); + if (!pmd) + return -ENOMEM; + VM_BUG_ON(pmd_trans_huge(*pmd)); + do { + next = pmd_addr_end(addr, end); + if (remap_pte_range(mm, pmd, addr, next, + pfn + (addr >> PAGE_SHIFT), prot)) + return -ENOMEM; + } while (pmd++, addr = next, addr != end); + return 0; +} + +static inline int remap_pud_range(struct mm_struct *mm, pgd_t *pgd, + unsigned long addr, unsigned long end, + unsigned long pfn, pgprot_t prot) +{ + pud_t *pud; + unsigned long next; + + pfn -= addr >> PAGE_SHIFT; + pud = pud_alloc(mm, pgd, addr); + if (!pud) + return -ENOMEM; + do { + next = pud_addr_end(addr, end); + if (remap_pmd_range(mm, pud, addr, next, + pfn + (addr >> PAGE_SHIFT), prot)) + return -ENOMEM; + } while (pud++, addr = next, addr != end); + return 0; +} + +/** + * remap_pfn_range - remap kernel memory to userspace + * @vma: user vma to map to + * @addr: target user address to start at + * @pfn: physical address of kernel memory + * @size: size of map area + * @prot: page protection flags for this mapping + * + * Note: this is only safe if the mm semaphore is held when called. + */ +int remap_pfn_range(struct vm_area_struct *vma, unsigned long addr, + unsigned long pfn, unsigned long size, pgprot_t prot) +{ + pgd_t *pgd; + unsigned long next; + unsigned long end = addr + PAGE_ALIGN(size); + struct mm_struct *mm = vma->vm_mm; + int err; + + /* + * Physically remapped pages are special. Tell the + * rest of the world about it: + * VM_IO tells people not to look at these pages + * (accesses can have side effects). + * VM_PFNMAP tells the core MM that the base pages are just + * raw PFN mappings, and do not have a "struct page" associated + * with them. + * VM_DONTEXPAND + * Disable vma merging and expanding with mremap(). + * VM_DONTDUMP + * Omit vma from core dump, even when VM_IO turned off. + * + * There's a horrible special case to handle copy-on-write + * behaviour that some programs depend on. We mark the "original" + * un-COW'ed pages by matching them up with "vma->vm_pgoff". + * See vm_normal_page() for details. + */ + if (is_cow_mapping(vma->vm_flags)) { + if (addr != vma->vm_start || end != vma->vm_end) + return -EINVAL; + vma->vm_pgoff = pfn; + } + + err = track_pfn_remap(vma, &prot, pfn, addr, PAGE_ALIGN(size)); + if (err) + return -EINVAL; + + vma->vm_flags |= VM_IO | VM_PFNMAP | VM_DONTEXPAND | VM_DONTDUMP; + + BUG_ON(addr >= end); + pfn -= addr >> PAGE_SHIFT; + pgd = pgd_offset(mm, addr); + flush_cache_range(vma, addr, end); + do { + next = pgd_addr_end(addr, end); + err = remap_pud_range(mm, pgd, addr, next, + pfn + (addr >> PAGE_SHIFT), prot); + if (err) + break; + } while (pgd++, addr = next, addr != end); + + if (err) + untrack_pfn(vma, pfn, PAGE_ALIGN(size)); + + return err; +} +EXPORT_SYMBOL(remap_pfn_range); + +/** + * vm_iomap_memory - remap memory to userspace + * @vma: user vma to map to + * @start: start of area + * @len: size of area + * + * This is a simplified io_remap_pfn_range() for common driver use. The + * driver just needs to give us the physical memory range to be mapped, + * we'll figure out the rest from the vma information. + * + * NOTE! Some drivers might want to tweak vma->vm_page_prot first to get + * whatever write-combining details or similar. + */ +int vm_iomap_memory(struct vm_area_struct *vma, phys_addr_t start, unsigned long len) +{ + unsigned long vm_len, pfn, pages; + + /* Check that the physical memory area passed in looks valid */ + if (start + len < start) + return -EINVAL; + /* + * You *really* shouldn't map things that aren't page-aligned, + * but we've historically allowed it because IO memory might + * just have smaller alignment. + */ + len += start & ~PAGE_MASK; + pfn = start >> PAGE_SHIFT; + pages = (len + ~PAGE_MASK) >> PAGE_SHIFT; + if (pfn + pages < pfn) + return -EINVAL; + + /* We start the mapping 'vm_pgoff' pages into the area */ + if (vma->vm_pgoff > pages) + return -EINVAL; + pfn += vma->vm_pgoff; + pages -= vma->vm_pgoff; + + /* Can we fit all of the mapping? */ + vm_len = vma->vm_end - vma->vm_start; + if (vm_len >> PAGE_SHIFT > pages) + return -EINVAL; + + /* Ok, let it rip */ + return io_remap_pfn_range(vma, vma->vm_start, pfn, vm_len, vma->vm_page_prot); +} +EXPORT_SYMBOL(vm_iomap_memory); + +static int apply_to_pte_range(struct mm_struct *mm, pmd_t *pmd, + unsigned long addr, unsigned long end, + pte_fn_t fn, void *data) +{ + pte_t *pte; + int err; + pgtable_t token; + spinlock_t *uninitialized_var(ptl); + + pte = (mm == &init_mm) ? + pte_alloc_kernel(pmd, addr) : + pte_alloc_map_lock(mm, pmd, addr, &ptl); + if (!pte) + return -ENOMEM; + + BUG_ON(pmd_huge(*pmd)); + + arch_enter_lazy_mmu_mode(); + + token = pmd_pgtable(*pmd); + + do { + err = fn(pte++, token, addr, data); + if (err) + break; + } while (addr += PAGE_SIZE, addr != end); + + arch_leave_lazy_mmu_mode(); + + if (mm != &init_mm) + pte_unmap_unlock(pte-1, ptl); + return err; +} + +static int apply_to_pmd_range(struct mm_struct *mm, pud_t *pud, + unsigned long addr, unsigned long end, + pte_fn_t fn, void *data) +{ + pmd_t *pmd; + unsigned long next; + int err; + + BUG_ON(pud_huge(*pud)); + + pmd = pmd_alloc(mm, pud, addr); + if (!pmd) + return -ENOMEM; + do { + next = pmd_addr_end(addr, end); + err = apply_to_pte_range(mm, pmd, addr, next, fn, data); + if (err) + break; + } while (pmd++, addr = next, addr != end); + return err; +} + +static int apply_to_pud_range(struct mm_struct *mm, pgd_t *pgd, + unsigned long addr, unsigned long end, + pte_fn_t fn, void *data) +{ + pud_t *pud; + unsigned long next; + int err; + + pud = pud_alloc(mm, pgd, addr); + if (!pud) + return -ENOMEM; + do { + next = pud_addr_end(addr, end); + err = apply_to_pmd_range(mm, pud, addr, next, fn, data); + if (err) + break; + } while (pud++, addr = next, addr != end); + return err; +} + +/* + * Scan a region of virtual memory, filling in page tables as necessary + * and calling a provided function on each leaf page table. + */ +int apply_to_page_range(struct mm_struct *mm, unsigned long addr, + unsigned long size, pte_fn_t fn, void *data) +{ + pgd_t *pgd; + unsigned long next; + unsigned long end = addr + size; + int err; + + BUG_ON(addr >= end); + pgd = pgd_offset(mm, addr); + do { + next = pgd_addr_end(addr, end); + err = apply_to_pud_range(mm, pgd, addr, next, fn, data); + if (err) + break; + } while (pgd++, addr = next, addr != end); + + return err; +} +EXPORT_SYMBOL_GPL(apply_to_page_range); + +/* + * handle_pte_fault chooses page fault handler according to an entry + * which was read non-atomically. Before making any commitment, on + * those architectures or configurations (e.g. i386 with PAE) which + * might give a mix of unmatched parts, do_swap_page and do_nonlinear_fault + * must check under lock before unmapping the pte and proceeding + * (but do_wp_page is only called after already making such a check; + * and do_anonymous_page can safely check later on). + */ +static inline int pte_unmap_same(struct mm_struct *mm, pmd_t *pmd, + pte_t *page_table, pte_t orig_pte) +{ + int same = 1; +#if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT) + if (sizeof(pte_t) > sizeof(unsigned long)) { + spinlock_t *ptl = pte_lockptr(mm, pmd); + spin_lock(ptl); + same = pte_same(*page_table, orig_pte); + spin_unlock(ptl); + } +#endif + pte_unmap(page_table); + return same; +} + +static inline void cow_user_page(struct page *dst, struct page *src, unsigned long va, struct vm_area_struct *vma) +{ + /* + * If the source page was a PFN mapping, we don't have + * a "struct page" for it. We do a best-effort copy by + * just copying from the original user address. If that + * fails, we just zero-fill it. Live with it. + */ + if (unlikely(!src)) { + void *kaddr = kmap_atomic(dst); + void __user *uaddr = (void __user *)(va & PAGE_MASK); + + /* + * This really shouldn't fail, because the page is there + * in the page tables. But it might just be unreadable, + * in which case we just give up and fill the result with + * zeroes. + */ + if (__copy_from_user_inatomic(kaddr, uaddr, PAGE_SIZE)) + clear_page(kaddr); + kunmap_atomic(kaddr); + flush_dcache_page(dst); + } else + copy_user_highpage(dst, src, va, vma); +} + +/* + * This routine handles present pages, when users try to write + * to a shared page. It is done by copying the page to a new address + * and decrementing the shared-page counter for the old page. + * + * Note that this routine assumes that the protection checks have been + * done by the caller (the low-level page fault routine in most cases). + * Thus we can safely just mark it writable once we've done any necessary + * COW. + * + * We also mark the page dirty at this point even though the page will + * change only once the write actually happens. This avoids a few races, + * and potentially makes it more efficient. + * + * We enter with non-exclusive mmap_sem (to exclude vma changes, + * but allow concurrent faults), with pte both mapped and locked. + * We return with mmap_sem still held, but pte unmapped and unlocked. + */ +static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long address, pte_t *page_table, pmd_t *pmd, + spinlock_t *ptl, pte_t orig_pte, unsigned int flags) + __releases(ptl) +{ + struct page *old_page, *new_page = NULL; + pte_t entry; + int ret = 0; + int page_mkwrite = 0; + struct page *dirty_page = NULL; + unsigned long mmun_start = 0; /* For mmu_notifiers */ + unsigned long mmun_end = 0; /* For mmu_notifiers */ + gfp_t gfp = GFP_HIGHUSER_MOVABLE; + + if (IS_ENABLED(CONFIG_CMA) && (flags & FAULT_FLAG_NO_CMA)) + gfp &= ~__GFP_MOVABLE; + + old_page = vm_normal_page(vma, address, orig_pte); + if (!old_page) { + /* + * VM_MIXEDMAP !pfn_valid() case + * + * We should not cow pages in a shared writeable mapping. + * Just mark the pages writable as we can't do any dirty + * accounting on raw pfn maps. + */ + if ((vma->vm_flags & (VM_WRITE|VM_SHARED)) == + (VM_WRITE|VM_SHARED)) + goto reuse; + goto gotten; + } + + /* + * Take out anonymous pages first, anonymous shared vmas are + * not dirty accountable. + */ + if (PageAnon(old_page) && !PageKsm(old_page)) { + if (!trylock_page(old_page)) { + page_cache_get(old_page); + pte_unmap_unlock(page_table, ptl); + lock_page(old_page); + page_table = pte_offset_map_lock(mm, pmd, address, + &ptl); + if (!pte_same(*page_table, orig_pte)) { + unlock_page(old_page); + goto unlock; + } + page_cache_release(old_page); + } + if (reuse_swap_page(old_page)) { + /* + * The page is all ours. Move it to our anon_vma so + * the rmap code will not search our parent or siblings. + * Protected against the rmap code by the page lock. + */ + page_move_anon_rmap(old_page, vma, address); + unlock_page(old_page); + goto reuse; + } + unlock_page(old_page); + } else if (unlikely((vma->vm_flags & (VM_WRITE|VM_SHARED)) == + (VM_WRITE|VM_SHARED))) { + /* + * Only catch write-faults on shared writable pages, + * read-only shared pages can get COWed by + * get_user_pages(.write=1, .force=1). + */ + if (vma->vm_ops && vma->vm_ops->page_mkwrite) { + struct vm_fault vmf; + int tmp; + + vmf.virtual_address = (void __user *)(address & + PAGE_MASK); + vmf.pgoff = old_page->index; + vmf.flags = FAULT_FLAG_WRITE|FAULT_FLAG_MKWRITE; + vmf.page = old_page; + + /* + * Notify the address space that the page is about to + * become writable so that it can prohibit this or wait + * for the page to get into an appropriate state. + * + * We do this without the lock held, so that it can + * sleep if it needs to. + */ + page_cache_get(old_page); + pte_unmap_unlock(page_table, ptl); + + tmp = vma->vm_ops->page_mkwrite(vma, &vmf); + if (unlikely(tmp & + (VM_FAULT_ERROR | VM_FAULT_NOPAGE))) { + ret = tmp; + goto unwritable_page; + } + if (unlikely(!(tmp & VM_FAULT_LOCKED))) { + lock_page(old_page); + if (!old_page->mapping) { + ret = 0; /* retry the fault */ + unlock_page(old_page); + goto unwritable_page; + } + } else + VM_BUG_ON(!PageLocked(old_page)); + + /* + * Since we dropped the lock we need to revalidate + * the PTE as someone else may have changed it. If + * they did, we just return, as we can count on the + * MMU to tell us if they didn't also make it writable. + */ + page_table = pte_offset_map_lock(mm, pmd, address, + &ptl); + if (!pte_same(*page_table, orig_pte)) { + unlock_page(old_page); + goto unlock; + } + + page_mkwrite = 1; + } + dirty_page = old_page; + get_page(dirty_page); + +reuse: + flush_cache_page(vma, address, pte_pfn(orig_pte)); + entry = pte_mkyoung(orig_pte); + entry = maybe_mkwrite(pte_mkdirty(entry), vma); + if (ptep_set_access_flags(vma, address, page_table, entry,1)) + update_mmu_cache(vma, address, page_table); + pte_unmap_unlock(page_table, ptl); + ret |= VM_FAULT_WRITE; + + if (!dirty_page) + return ret; + + /* + * Yes, Virginia, this is actually required to prevent a race + * with clear_page_dirty_for_io() from clearing the page dirty + * bit after it clear all dirty ptes, but before a racing + * do_wp_page installs a dirty pte. + * + * __do_fault is protected similarly. + */ + if (!page_mkwrite) { + wait_on_page_locked(dirty_page); + set_page_dirty_balance(dirty_page, page_mkwrite); + /* file_update_time outside page_lock */ + if (vma->vm_file) + file_update_time(vma->vm_file); + } + put_page(dirty_page); + if (page_mkwrite) { + struct address_space *mapping = dirty_page->mapping; + + set_page_dirty(dirty_page); + unlock_page(dirty_page); + page_cache_release(dirty_page); + if (mapping) { + /* + * Some device drivers do not set page.mapping + * but still dirty their pages + */ + balance_dirty_pages_ratelimited(mapping); + } + } + + return ret; + } + + /* + * Ok, we need to copy. Oh, well.. + */ + page_cache_get(old_page); +gotten: + pte_unmap_unlock(page_table, ptl); + + if (unlikely(anon_vma_prepare(vma))) + goto oom; + + if (is_zero_pfn(pte_pfn(orig_pte))) { + new_page = alloc_zeroed_user_highpage(gfp, vma, address); + if (!new_page) + goto oom; + } else { + new_page = alloc_page_vma(gfp, vma, address); + if (!new_page) + goto oom; + cow_user_page(new_page, old_page, address, vma); + } + __SetPageUptodate(new_page); + + if (mem_cgroup_newpage_charge(new_page, mm, GFP_KERNEL)) + goto oom_free_new; + + mmun_start = address & PAGE_MASK; + mmun_end = mmun_start + PAGE_SIZE; + mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end); + + /* + * Re-check the pte - we dropped the lock + */ + page_table = pte_offset_map_lock(mm, pmd, address, &ptl); + if (likely(pte_same(*page_table, orig_pte))) { + if (old_page) { + if (!PageAnon(old_page)) { + dec_mm_counter_fast(mm, MM_FILEPAGES); + inc_mm_counter_fast(mm, MM_ANONPAGES); + } + } else + inc_mm_counter_fast(mm, MM_ANONPAGES); + flush_cache_page(vma, address, pte_pfn(orig_pte)); + entry = mk_pte(new_page, vma->vm_page_prot); + entry = maybe_mkwrite(pte_mkdirty(entry), vma); + /* + * Clear the pte entry and flush it first, before updating the + * pte with the new entry. This will avoid a race condition + * seen in the presence of one thread doing SMC and another + * thread doing COW. + */ + ptep_clear_flush(vma, address, page_table); + page_add_new_anon_rmap(new_page, vma, address); + /* + * We call the notify macro here because, when using secondary + * mmu page tables (such as kvm shadow page tables), we want the + * new page to be mapped directly into the secondary page table. + */ + set_pte_at_notify(mm, address, page_table, entry); + update_mmu_cache(vma, address, page_table); + if (old_page) { + /* + * Only after switching the pte to the new page may + * we remove the mapcount here. Otherwise another + * process may come and find the rmap count decremented + * before the pte is switched to the new page, and + * "reuse" the old page writing into it while our pte + * here still points into it and can be read by other + * threads. + * + * The critical issue is to order this + * page_remove_rmap with the ptp_clear_flush above. + * Those stores are ordered by (if nothing else,) + * the barrier present in the atomic_add_negative + * in page_remove_rmap. + * + * Then the TLB flush in ptep_clear_flush ensures that + * no process can access the old page before the + * decremented mapcount is visible. And the old page + * cannot be reused until after the decremented + * mapcount is visible. So transitively, TLBs to + * old page will be flushed before it can be reused. + */ + page_remove_rmap(old_page); + } + + /* Free the old page.. */ + new_page = old_page; + ret |= VM_FAULT_WRITE; + } else + mem_cgroup_uncharge_page(new_page); + + if (new_page) + page_cache_release(new_page); +unlock: + pte_unmap_unlock(page_table, ptl); + if (mmun_end > mmun_start) + mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end); + if (old_page) { + /* + * Don't let another task, with possibly unlocked vma, + * keep the mlocked page. + */ + if ((ret & VM_FAULT_WRITE) && (vma->vm_flags & VM_LOCKED)) { + lock_page(old_page); /* LRU manipulation */ + munlock_vma_page(old_page); + unlock_page(old_page); + } + page_cache_release(old_page); + } + return ret; +oom_free_new: + page_cache_release(new_page); +oom: + if (old_page) + page_cache_release(old_page); + return VM_FAULT_OOM; + +unwritable_page: + page_cache_release(old_page); + return ret; +} + +static void unmap_mapping_range_vma(struct vm_area_struct *vma, + unsigned long start_addr, unsigned long end_addr, + struct zap_details *details) +{ + zap_page_range_single(vma, start_addr, end_addr - start_addr, details); +} + +static inline void unmap_mapping_range_tree(struct rb_root *root, + struct zap_details *details) +{ + struct vm_area_struct *vma; + pgoff_t vba, vea, zba, zea; + + vma_interval_tree_foreach(vma, root, + details->first_index, details->last_index) { + + vba = vma->vm_pgoff; + vea = vba + ((vma->vm_end - vma->vm_start) >> PAGE_SHIFT) - 1; + /* Assume for now that PAGE_CACHE_SHIFT == PAGE_SHIFT */ + zba = details->first_index; + if (zba < vba) + zba = vba; + zea = details->last_index; + if (zea > vea) + zea = vea; + + unmap_mapping_range_vma(vma, + ((zba - vba) << PAGE_SHIFT) + vma->vm_start, + ((zea - vba + 1) << PAGE_SHIFT) + vma->vm_start, + details); + } +} + +static inline void unmap_mapping_range_list(struct list_head *head, + struct zap_details *details) +{ + struct vm_area_struct *vma; + + /* + * In nonlinear VMAs there is no correspondence between virtual address + * offset and file offset. So we must perform an exhaustive search + * across *all* the pages in each nonlinear VMA, not just the pages + * whose virtual address lies outside the file truncation point. + */ + list_for_each_entry(vma, head, shared.nonlinear) { + details->nonlinear_vma = vma; + unmap_mapping_range_vma(vma, vma->vm_start, vma->vm_end, details); + } +} + +/** + * unmap_mapping_range - unmap the portion of all mmaps in the specified address_space corresponding to the specified page range in the underlying file. + * @mapping: the address space containing mmaps to be unmapped. + * @holebegin: byte in first page to unmap, relative to the start of + * the underlying file. This will be rounded down to a PAGE_SIZE + * boundary. Note that this is different from truncate_pagecache(), which + * must keep the partial page. In contrast, we must get rid of + * partial pages. + * @holelen: size of prospective hole in bytes. This will be rounded + * up to a PAGE_SIZE boundary. A holelen of zero truncates to the + * end of the file. + * @even_cows: 1 when truncating a file, unmap even private COWed pages; + * but 0 when invalidating pagecache, don't throw away private data. + */ +void unmap_mapping_range(struct address_space *mapping, + loff_t const holebegin, loff_t const holelen, int even_cows) +{ + struct zap_details details; + pgoff_t hba = holebegin >> PAGE_SHIFT; + pgoff_t hlen = (holelen + PAGE_SIZE - 1) >> PAGE_SHIFT; + + /* Check for overflow. */ + if (sizeof(holelen) > sizeof(hlen)) { + long long holeend = + (holebegin + holelen + PAGE_SIZE - 1) >> PAGE_SHIFT; + if (holeend & ~(long long)ULONG_MAX) + hlen = ULONG_MAX - hba + 1; + } + + details.check_mapping = even_cows? NULL: mapping; + details.nonlinear_vma = NULL; + details.first_index = hba; + details.last_index = hba + hlen - 1; + if (details.last_index < details.first_index) + details.last_index = ULONG_MAX; + + + mutex_lock(&mapping->i_mmap_mutex); + if (unlikely(!RB_EMPTY_ROOT(&mapping->i_mmap))) + unmap_mapping_range_tree(&mapping->i_mmap, &details); + if (unlikely(!list_empty(&mapping->i_mmap_nonlinear))) + unmap_mapping_range_list(&mapping->i_mmap_nonlinear, &details); + mutex_unlock(&mapping->i_mmap_mutex); +} +EXPORT_SYMBOL(unmap_mapping_range); + +/* + * We enter with non-exclusive mmap_sem (to exclude vma changes, + * but allow concurrent faults), and pte mapped but not yet locked. + * We return with mmap_sem still held, but pte unmapped and unlocked. + */ +static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long address, pte_t *page_table, pmd_t *pmd, + unsigned int flags, pte_t orig_pte) +{ + spinlock_t *ptl; + struct page *page, *swapcache; + swp_entry_t entry; + pte_t pte; + int locked; + struct mem_cgroup *ptr; + int exclusive = 0; + int ret = 0; + + if (!pte_unmap_same(mm, pmd, page_table, orig_pte)) + goto out; + + entry = pte_to_swp_entry(orig_pte); + if (unlikely(non_swap_entry(entry))) { + if (is_migration_entry(entry)) { + migration_entry_wait(mm, pmd, address); + } else if (is_hwpoison_entry(entry)) { + ret = VM_FAULT_HWPOISON; + } else { + print_bad_pte(vma, address, orig_pte, NULL); + ret = VM_FAULT_SIGBUS; + } + goto out; + } + delayacct_set_flag(DELAYACCT_PF_SWAPIN); + page = lookup_swap_cache(entry); + if (!page) { + page = swapin_readahead(entry, + GFP_HIGHUSER_MOVABLE, vma, address); + if (!page) { + /* + * Back out if somebody else faulted in this pte + * while we released the pte lock. + */ + page_table = pte_offset_map_lock(mm, pmd, address, &ptl); + if (likely(pte_same(*page_table, orig_pte))) + ret = VM_FAULT_OOM; + delayacct_clear_flag(DELAYACCT_PF_SWAPIN); + goto unlock; + } + + /* Had to read the page from swap area: Major fault */ + ret = VM_FAULT_MAJOR; + count_vm_event(PGMAJFAULT); + mem_cgroup_count_vm_event(mm, PGMAJFAULT); + } else if (PageHWPoison(page)) { + /* + * hwpoisoned dirty swapcache pages are kept for killing + * owner processes (which may be unknown at hwpoison time) + */ + ret = VM_FAULT_HWPOISON; + delayacct_clear_flag(DELAYACCT_PF_SWAPIN); + swapcache = page; + goto out_release; + } + + swapcache = page; + locked = lock_page_or_retry(page, mm, flags); + + delayacct_clear_flag(DELAYACCT_PF_SWAPIN); + if (!locked) { + ret |= VM_FAULT_RETRY; + goto out_release; + } + + /* + * Make sure try_to_free_swap or reuse_swap_page or swapoff did not + * release the swapcache from under us. The page pin, and pte_same + * test below, are not enough to exclude that. Even if it is still + * swapcache, we need to check that the page's swap has not changed. + */ + if (unlikely(!PageSwapCache(page) || page_private(page) != entry.val)) + goto out_page; + + page = ksm_might_need_to_copy(page, vma, address); + if (unlikely(!page)) { + ret = VM_FAULT_OOM; + page = swapcache; + goto out_page; + } + + if (mem_cgroup_try_charge_swapin(mm, page, GFP_KERNEL, &ptr)) { + ret = VM_FAULT_OOM; + goto out_page; + } + + /* + * Back out if somebody else already faulted in this pte. + */ + page_table = pte_offset_map_lock(mm, pmd, address, &ptl); + if (unlikely(!pte_same(*page_table, orig_pte))) + goto out_nomap; + + if (unlikely(!PageUptodate(page))) { + ret = VM_FAULT_SIGBUS; + goto out_nomap; + } + + /* + * The page isn't present yet, go ahead with the fault. + * + * Be careful about the sequence of operations here. + * To get its accounting right, reuse_swap_page() must be called + * while the page is counted on swap but not yet in mapcount i.e. + * before page_add_anon_rmap() and swap_free(); try_to_free_swap() + * must be called after the swap_free(), or it will never succeed. + * Because delete_from_swap_page() may be called by reuse_swap_page(), + * mem_cgroup_commit_charge_swapin() may not be able to find swp_entry + * in page->private. In this case, a record in swap_cgroup is silently + * discarded at swap_free(). + */ + + inc_mm_counter_fast(mm, MM_ANONPAGES); + dec_mm_counter_fast(mm, MM_SWAPENTS); + pte = mk_pte(page, vma->vm_page_prot); + if ((flags & FAULT_FLAG_WRITE) && reuse_swap_page(page)) { + pte = maybe_mkwrite(pte_mkdirty(pte), vma); + flags &= ~FAULT_FLAG_WRITE; + ret |= VM_FAULT_WRITE; + exclusive = 1; + } + flush_icache_page(vma, page); + set_pte_at(mm, address, page_table, pte); + if (page == swapcache) + do_page_add_anon_rmap(page, vma, address, exclusive); + else /* ksm created a completely new copy */ + page_add_new_anon_rmap(page, vma, address); + /* It's better to call commit-charge after rmap is established */ + mem_cgroup_commit_charge_swapin(page, ptr); + + swap_free(entry); + if (vm_swap_full() || (vma->vm_flags & VM_LOCKED) || PageMlocked(page)) + try_to_free_swap(page); + unlock_page(page); + if (page != swapcache) { + /* + * Hold the lock to avoid the swap entry to be reused + * until we take the PT lock for the pte_same() check + * (to avoid false positives from pte_same). For + * further safety release the lock after the swap_free + * so that the swap count won't change under a + * parallel locked swapcache. + */ + unlock_page(swapcache); + page_cache_release(swapcache); + } + + if (flags & FAULT_FLAG_WRITE) { + ret |= do_wp_page(mm, vma, address, page_table, pmd, ptl, pte, flags); + if (ret & VM_FAULT_ERROR) + ret &= VM_FAULT_ERROR; + goto out; + } + + /* No need to invalidate - it was non-present before */ + update_mmu_cache(vma, address, page_table); +unlock: + pte_unmap_unlock(page_table, ptl); +out: + return ret; +out_nomap: + mem_cgroup_cancel_charge_swapin(ptr); + pte_unmap_unlock(page_table, ptl); +out_page: + unlock_page(page); +out_release: + page_cache_release(page); + if (page != swapcache) { + unlock_page(swapcache); + page_cache_release(swapcache); + } + return ret; +} + +/* + * This is like a special single-page "expand_{down|up}wards()", + * except we must first make sure that 'address{-|+}PAGE_SIZE' + * doesn't hit another vma. + */ +static inline int check_stack_guard_page(struct vm_area_struct *vma, unsigned long address) +{ + address &= PAGE_MASK; + if ((vma->vm_flags & VM_GROWSDOWN) && address == vma->vm_start) { + struct vm_area_struct *prev = vma->vm_prev; + + /* + * Is there a mapping abutting this one below? + * + * That's only ok if it's the same stack mapping + * that has gotten split.. + */ + if (prev && prev->vm_end == address) + return prev->vm_flags & VM_GROWSDOWN ? 0 : -ENOMEM; + + expand_downwards(vma, address - PAGE_SIZE); + } + if ((vma->vm_flags & VM_GROWSUP) && address + PAGE_SIZE == vma->vm_end) { + struct vm_area_struct *next = vma->vm_next; + + /* As VM_GROWSDOWN but s/below/above/ */ + if (next && next->vm_start == address + PAGE_SIZE) + return next->vm_flags & VM_GROWSUP ? 0 : -ENOMEM; + + expand_upwards(vma, address + PAGE_SIZE); + } + return 0; +} + +bool is_vma_temporary_stack(struct vm_area_struct *vma); +/* + * We enter with non-exclusive mmap_sem (to exclude vma changes, + * but allow concurrent faults), and pte mapped but not yet locked. + * We return with mmap_sem still held, but pte unmapped and unlocked. + */ +static int do_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long address, pte_t *page_table, pmd_t *pmd, + unsigned int flags) +{ + struct page *page; + spinlock_t *ptl; + pte_t entry; + + pte_unmap(page_table); + + /* Check if we need to add a guard page to the stack */ + if (check_stack_guard_page(vma, address) < 0) + return VM_FAULT_SIGBUS; + + /* Use the zero-page for reads */ + if (!(flags & FAULT_FLAG_WRITE)) { + entry = pte_mkspecial(pfn_pte(my_zero_pfn(address), + vma->vm_page_prot)); + page_table = pte_offset_map_lock(mm, pmd, address, &ptl); + if (!pte_none(*page_table)) + goto unlock; + goto setpte; + } + + /* Allocate our own private page. */ + if (unlikely(anon_vma_prepare(vma))) + goto oom; + if (vma->vm_flags & VM_LOCKED || flags & FAULT_FLAG_NO_CMA || + is_vma_temporary_stack(vma)) { + page = alloc_zeroed_user_highpage(GFP_HIGHUSER, vma, address); + } else { + page = alloc_zeroed_user_highpage_movable(vma, address); + } + if (!page) + goto oom; + /* + * The memory barrier inside __SetPageUptodate makes sure that + * preceeding stores to the page contents become visible before + * the set_pte_at() write. + */ + __SetPageUptodate(page); + + if (mem_cgroup_newpage_charge(page, mm, GFP_KERNEL)) + goto oom_free_page; + + entry = mk_pte(page, vma->vm_page_prot); + if (vma->vm_flags & VM_WRITE) + entry = pte_mkwrite(pte_mkdirty(entry)); + + page_table = pte_offset_map_lock(mm, pmd, address, &ptl); + if (!pte_none(*page_table)) + goto release; + + inc_mm_counter_fast(mm, MM_ANONPAGES); + page_add_new_anon_rmap(page, vma, address); +setpte: + set_pte_at(mm, address, page_table, entry); + + /* No need to invalidate - it was non-present before */ + update_mmu_cache(vma, address, page_table); +unlock: + pte_unmap_unlock(page_table, ptl); + return 0; +release: + mem_cgroup_uncharge_page(page); + page_cache_release(page); + goto unlock; +oom_free_page: + page_cache_release(page); +oom: + return VM_FAULT_OOM; +} + +/* + * __do_fault() tries to create a new page mapping. It aggressively + * tries to share with existing pages, but makes a separate copy if + * the FAULT_FLAG_WRITE is set in the flags parameter in order to avoid + * the next page fault. + * + * As this is called only for pages that do not currently exist, we + * do not need to flush old virtual caches or the TLB. + * + * We enter with non-exclusive mmap_sem (to exclude vma changes, + * but allow concurrent faults), and pte neither mapped nor locked. + * We return with mmap_sem still held, but pte unmapped and unlocked. + */ +static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long address, pmd_t *pmd, + pgoff_t pgoff, unsigned int flags, pte_t orig_pte) +{ + pte_t *page_table; + spinlock_t *ptl; + struct page *page; + struct page *cow_page; + pte_t entry; + int anon = 0; + struct page *dirty_page = NULL; + struct vm_fault vmf; + int ret; + int page_mkwrite = 0; + gfp_t gfp = GFP_HIGHUSER_MOVABLE; + + if (IS_ENABLED(CONFIG_CMA) && (flags & FAULT_FLAG_NO_CMA)) + gfp &= ~__GFP_MOVABLE; + + + /* + * If we do COW later, allocate page befor taking lock_page() + * on the file cache page. This will reduce lock holding time. + */ + if ((flags & FAULT_FLAG_WRITE) && !(vma->vm_flags & VM_SHARED)) { + + if (unlikely(anon_vma_prepare(vma))) + return VM_FAULT_OOM; + + cow_page = alloc_page_vma(gfp, vma, address); + if (!cow_page) + return VM_FAULT_OOM; + + if (mem_cgroup_newpage_charge(cow_page, mm, GFP_KERNEL)) { + page_cache_release(cow_page); + return VM_FAULT_OOM; + } + } else + cow_page = NULL; + + vmf.virtual_address = (void __user *)(address & PAGE_MASK); + vmf.pgoff = pgoff; + vmf.flags = flags; + vmf.page = NULL; + + ret = vma->vm_ops->fault(vma, &vmf); + if (unlikely(ret & (VM_FAULT_ERROR | VM_FAULT_NOPAGE | + VM_FAULT_RETRY))) + goto uncharge_out; + + if (unlikely(PageHWPoison(vmf.page))) { + if (ret & VM_FAULT_LOCKED) + unlock_page(vmf.page); + ret = VM_FAULT_HWPOISON; + goto uncharge_out; + } + + /* + * For consistency in subsequent calls, make the faulted page always + * locked. + */ + if (unlikely(!(ret & VM_FAULT_LOCKED))) + lock_page(vmf.page); + else + VM_BUG_ON(!PageLocked(vmf.page)); + + /* + * Should we do an early C-O-W break? + */ + page = vmf.page; + if (flags & FAULT_FLAG_WRITE) { + if (!(vma->vm_flags & VM_SHARED)) { + page = cow_page; + anon = 1; + copy_user_highpage(page, vmf.page, address, vma); + __SetPageUptodate(page); + } else { + /* + * If the page will be shareable, see if the backing + * address space wants to know that the page is about + * to become writable + */ + if (vma->vm_ops->page_mkwrite) { + int tmp; + + unlock_page(page); + vmf.flags = FAULT_FLAG_WRITE|FAULT_FLAG_MKWRITE; + tmp = vma->vm_ops->page_mkwrite(vma, &vmf); + if (unlikely(tmp & + (VM_FAULT_ERROR | VM_FAULT_NOPAGE))) { + ret = tmp; + goto unwritable_page; + } + if (unlikely(!(tmp & VM_FAULT_LOCKED))) { + lock_page(page); + if (!page->mapping) { + ret = 0; /* retry the fault */ + unlock_page(page); + goto unwritable_page; + } + } else + VM_BUG_ON(!PageLocked(page)); + page_mkwrite = 1; + } + } + + } + + page_table = pte_offset_map_lock(mm, pmd, address, &ptl); + + /* + * This silly early PAGE_DIRTY setting removes a race + * due to the bad i386 page protection. But it's valid + * for other architectures too. + * + * Note that if FAULT_FLAG_WRITE is set, we either now have + * an exclusive copy of the page, or this is a shared mapping, + * so we can make it writable and dirty to avoid having to + * handle that later. + */ + /* Only go through if we didn't race with anybody else... */ + if (likely(pte_same(*page_table, orig_pte))) { + flush_icache_page(vma, page); + entry = mk_pte(page, vma->vm_page_prot); + if (flags & FAULT_FLAG_WRITE) + entry = maybe_mkwrite(pte_mkdirty(entry), vma); + if (anon) { + inc_mm_counter_fast(mm, MM_ANONPAGES); + page_add_new_anon_rmap(page, vma, address); + } else { + inc_mm_counter_fast(mm, MM_FILEPAGES); + page_add_file_rmap(page); + if (flags & FAULT_FLAG_WRITE) { + dirty_page = page; + get_page(dirty_page); + } + } + set_pte_at(mm, address, page_table, entry); + + /* no need to invalidate: a not-present page won't be cached */ + update_mmu_cache(vma, address, page_table); + } else { + if (cow_page) + mem_cgroup_uncharge_page(cow_page); + if (anon) + page_cache_release(page); + else + anon = 1; /* no anon but release faulted_page */ + } + + pte_unmap_unlock(page_table, ptl); + + if (dirty_page) { + struct address_space *mapping = page->mapping; + int dirtied = 0; + + if (set_page_dirty(dirty_page)) + dirtied = 1; + unlock_page(dirty_page); + put_page(dirty_page); + if ((dirtied || page_mkwrite) && mapping) { + /* + * Some device drivers do not set page.mapping but still + * dirty their pages + */ + balance_dirty_pages_ratelimited(mapping); + } + + /* file_update_time outside page_lock */ + if (vma->vm_file && !page_mkwrite) + file_update_time(vma->vm_file); + } else { + unlock_page(vmf.page); + if (anon) + page_cache_release(vmf.page); + } + + return ret; + +unwritable_page: + page_cache_release(page); + return ret; +uncharge_out: + /* fs's fault handler get error */ + if (cow_page) { + mem_cgroup_uncharge_page(cow_page); + page_cache_release(cow_page); + } + return ret; +} + +static int do_linear_fault(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long address, pte_t *page_table, pmd_t *pmd, + unsigned int flags, pte_t orig_pte) +{ + pgoff_t pgoff = (((address & PAGE_MASK) + - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff; + + pte_unmap(page_table); + return __do_fault(mm, vma, address, pmd, pgoff, flags, orig_pte); +} + +/* + * Fault of a previously existing named mapping. Repopulate the pte + * from the encoded file_pte if possible. This enables swappable + * nonlinear vmas. + * + * We enter with non-exclusive mmap_sem (to exclude vma changes, + * but allow concurrent faults), and pte mapped but not yet locked. + * We return with mmap_sem still held, but pte unmapped and unlocked. + */ +static int do_nonlinear_fault(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long address, pte_t *page_table, pmd_t *pmd, + unsigned int flags, pte_t orig_pte) +{ + pgoff_t pgoff; + + flags |= FAULT_FLAG_NONLINEAR; + + if (!pte_unmap_same(mm, pmd, page_table, orig_pte)) + return 0; + + if (unlikely(!(vma->vm_flags & VM_NONLINEAR))) { + /* + * Page table corrupted: show pte and kill process. + */ + print_bad_pte(vma, address, orig_pte, NULL); + return VM_FAULT_SIGBUS; + } + + pgoff = pte_to_pgoff(orig_pte); + return __do_fault(mm, vma, address, pmd, pgoff, flags, orig_pte); +} + +int numa_migrate_prep(struct page *page, struct vm_area_struct *vma, + unsigned long addr, int page_nid) +{ + get_page(page); + + count_vm_numa_event(NUMA_HINT_FAULTS); + if (page_nid == numa_node_id()) + count_vm_numa_event(NUMA_HINT_FAULTS_LOCAL); + + return mpol_misplaced(page, vma, addr); +} + +int do_numa_page(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long addr, pte_t pte, pte_t *ptep, pmd_t *pmd) +{ + struct page *page = NULL; + spinlock_t *ptl; + int page_nid = -1; + int target_nid; + bool migrated = false; + + /* + * The "pte" at this point cannot be used safely without + * validation through pte_unmap_same(). It's of NUMA type but + * the pfn may be screwed if the read is non atomic. + * + * ptep_modify_prot_start is not called as this is clearing + * the _PAGE_NUMA bit and it is not really expected that there + * would be concurrent hardware modifications to the PTE. + */ + ptl = pte_lockptr(mm, pmd); + spin_lock(ptl); + if (unlikely(!pte_same(*ptep, pte))) { + pte_unmap_unlock(ptep, ptl); + goto out; + } + + pte = pte_mknonnuma(pte); + set_pte_at(mm, addr, ptep, pte); + update_mmu_cache(vma, addr, ptep); + + page = vm_normal_page(vma, addr, pte); + if (!page) { + pte_unmap_unlock(ptep, ptl); + return 0; + } + + page_nid = page_to_nid(page); + target_nid = numa_migrate_prep(page, vma, addr, page_nid); + pte_unmap_unlock(ptep, ptl); + if (target_nid == -1) { + put_page(page); + goto out; + } + + /* Migrate to the requested node */ + migrated = migrate_misplaced_page(page, target_nid); + if (migrated) + page_nid = target_nid; + +out: + if (page_nid != -1) + task_numa_fault(page_nid, 1, migrated); + return 0; +} + +/* NUMA hinting page fault entry point for regular pmds */ +#ifdef CONFIG_NUMA_BALANCING +static int do_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long addr, pmd_t *pmdp) +{ + pmd_t pmd; + pte_t *pte, *orig_pte; + unsigned long _addr = addr & PMD_MASK; + unsigned long offset; + spinlock_t *ptl; + bool numa = false; + + spin_lock(&mm->page_table_lock); + pmd = *pmdp; + if (pmd_numa(pmd)) { + set_pmd_at(mm, _addr, pmdp, pmd_mknonnuma(pmd)); + numa = true; + } + spin_unlock(&mm->page_table_lock); + + if (!numa) + return 0; + + /* we're in a page fault so some vma must be in the range */ + BUG_ON(!vma); + BUG_ON(vma->vm_start >= _addr + PMD_SIZE); + offset = max(_addr, vma->vm_start) & ~PMD_MASK; + VM_BUG_ON(offset >= PMD_SIZE); + orig_pte = pte = pte_offset_map_lock(mm, pmdp, _addr, &ptl); + pte += offset >> PAGE_SHIFT; + for (addr = _addr + offset; addr < _addr + PMD_SIZE; pte++, addr += PAGE_SIZE) { + pte_t pteval = *pte; + struct page *page; + int page_nid = -1; + int target_nid; + bool migrated = false; + + if (!pte_present(pteval)) + continue; + if (!pte_numa(pteval)) + continue; + if (addr >= vma->vm_end) { + vma = find_vma(mm, addr); + /* there's a pte present so there must be a vma */ + BUG_ON(!vma); + BUG_ON(addr < vma->vm_start); + } + if (pte_numa(pteval)) { + pteval = pte_mknonnuma(pteval); + set_pte_at(mm, addr, pte, pteval); + } + page = vm_normal_page(vma, addr, pteval); + if (unlikely(!page)) + continue; + /* only check non-shared pages */ + if (unlikely(page_mapcount(page) != 1)) + continue; + + page_nid = page_to_nid(page); + target_nid = numa_migrate_prep(page, vma, addr, page_nid); + pte_unmap_unlock(pte, ptl); + if (target_nid != -1) { + migrated = migrate_misplaced_page(page, target_nid); + if (migrated) + page_nid = target_nid; + } else { + put_page(page); + } + + if (page_nid != -1) + task_numa_fault(page_nid, 1, migrated); + + pte = pte_offset_map_lock(mm, pmdp, addr, &ptl); + } + pte_unmap_unlock(orig_pte, ptl); + + return 0; +} +#else +static int do_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long addr, pmd_t *pmdp) +{ + BUG(); + return 0; +} +#endif /* CONFIG_NUMA_BALANCING */ + +/* + * These routines also need to handle stuff like marking pages dirty + * and/or accessed for architectures that don't do it in hardware (most + * RISC architectures). The early dirtying is also good on the i386. + * + * There is also a hook called "update_mmu_cache()" that architectures + * with external mmu caches can use to update those (ie the Sparc or + * PowerPC hashed page tables that act as extended TLBs). + * + * We enter with non-exclusive mmap_sem (to exclude vma changes, + * but allow concurrent faults), and pte mapped but not yet locked. + * We return with mmap_sem still held, but pte unmapped and unlocked. + */ +int handle_pte_fault(struct mm_struct *mm, + struct vm_area_struct *vma, unsigned long address, + pte_t *pte, pmd_t *pmd, unsigned int flags) +{ + pte_t entry; + spinlock_t *ptl; + + entry = *pte; + if (!pte_present(entry)) { + if (pte_none(entry)) { + if (vma->vm_ops) { + if (likely(vma->vm_ops->fault)) + return do_linear_fault(mm, vma, address, + pte, pmd, flags, entry); + } + return do_anonymous_page(mm, vma, address, + pte, pmd, flags); + } + if (pte_file(entry)) + return do_nonlinear_fault(mm, vma, address, + pte, pmd, flags, entry); + return do_swap_page(mm, vma, address, + pte, pmd, flags, entry); + } + + if (pte_numa(entry)) + return do_numa_page(mm, vma, address, entry, pte, pmd); + + ptl = pte_lockptr(mm, pmd); + spin_lock(ptl); + if (unlikely(!pte_same(*pte, entry))) + goto unlock; + if (flags & FAULT_FLAG_WRITE) { + if (!pte_write(entry)) + return do_wp_page(mm, vma, address, + pte, pmd, ptl, entry, flags); + entry = pte_mkdirty(entry); + } + entry = pte_mkyoung(entry); + if (ptep_set_access_flags(vma, address, pte, entry, flags & FAULT_FLAG_WRITE)) { + update_mmu_cache(vma, address, pte); + } else { + /* + * This is needed only for protection faults but the arch code + * is not yet telling us if this is a protection fault or not. + * This still avoids useless tlb flushes for .text page faults + * with threads. + */ + if (flags & FAULT_FLAG_WRITE) + flush_tlb_fix_spurious_fault(vma, address); + } +unlock: + pte_unmap_unlock(pte, ptl); + return 0; +} + +/* + * By the time we get here, we already hold the mm semaphore + */ +static int __handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long address, unsigned int flags) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + + if (unlikely(is_vm_hugetlb_page(vma))) + return hugetlb_fault(mm, vma, address, flags); + +retry: + pgd = pgd_offset(mm, address); + pud = pud_alloc(mm, pgd, address); + if (!pud) + return VM_FAULT_OOM; + pmd = pmd_alloc(mm, pud, address); + if (!pmd) + return VM_FAULT_OOM; + if (pmd_none(*pmd) && transparent_hugepage_enabled(vma)) { + if (!vma->vm_ops) + return do_huge_pmd_anonymous_page(mm, vma, address, + pmd, flags); + } else { + pmd_t orig_pmd = *pmd; + int ret; + + barrier(); + if (pmd_trans_huge(orig_pmd)) { + unsigned int dirty = flags & FAULT_FLAG_WRITE; + + /* + * If the pmd is splitting, return and retry the + * the fault. Alternative: wait until the split + * is done, and goto retry. + */ + if (pmd_trans_splitting(orig_pmd)) + return 0; + + if (pmd_numa(orig_pmd)) + return do_huge_pmd_numa_page(mm, vma, address, + orig_pmd, pmd); + + if (dirty && !pmd_write(orig_pmd)) { + ret = do_huge_pmd_wp_page(mm, vma, address, pmd, + orig_pmd); + /* + * If COW results in an oom, the huge pmd will + * have been split, so retry the fault on the + * pte for a smaller charge. + */ + if (unlikely(ret & VM_FAULT_OOM)) + goto retry; + return ret; + } else { + huge_pmd_set_accessed(mm, vma, address, pmd, + orig_pmd, dirty); + } + + return 0; + } + } + + if (pmd_numa(*pmd)) + return do_pmd_numa_page(mm, vma, address, pmd); + + /* + * Use __pte_alloc instead of pte_alloc_map, because we can't + * run pte_offset_map on the pmd, if an huge pmd could + * materialize from under us from a different thread. + */ + if (unlikely(pmd_none(*pmd)) && + unlikely(__pte_alloc(mm, vma, pmd, address))) + return VM_FAULT_OOM; + /* if an huge pmd materialized from under us just retry later */ + if (unlikely(pmd_trans_huge(*pmd))) + return 0; + /* + * A regular pmd is established and it can't morph into a huge pmd + * from under us anymore at this point because we hold the mmap_sem + * read mode and khugepaged takes it in write mode. So now it's + * safe to run pte_offset_map(). + */ + pte = pte_offset_map(pmd, address); + + return handle_pte_fault(mm, vma, address, pte, pmd, flags); +} + +int handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long address, unsigned int flags) +{ + int ret; + + __set_current_state(TASK_RUNNING); + + count_vm_event(PGFAULT); + mem_cgroup_count_vm_event(mm, PGFAULT); + + /* do counter updates before entering really critical section. */ + check_sync_rss_stat(current); + + /* + * Enable the memcg OOM handling for faults triggered in user + * space. Kernel faults are handled more gracefully. + */ + if (flags & FAULT_FLAG_USER) + mem_cgroup_enable_oom(); + + ret = __handle_mm_fault(mm, vma, address, flags); + + if (flags & FAULT_FLAG_USER) + mem_cgroup_disable_oom(); + + if (WARN_ON(task_in_memcg_oom(current) && !(ret & VM_FAULT_OOM))) + mem_cgroup_oom_synchronize(); + + return ret; +} + +#ifndef __PAGETABLE_PUD_FOLDED +/* + * Allocate page upper directory. + * We've already handled the fast-path in-line. + */ +int __pud_alloc(struct mm_struct *mm, pgd_t *pgd, unsigned long address) +{ + pud_t *new = pud_alloc_one(mm, address); + if (!new) + return -ENOMEM; + + smp_wmb(); /* See comment in __pte_alloc */ + + spin_lock(&mm->page_table_lock); + if (pgd_present(*pgd)) /* Another has populated it */ + pud_free(mm, new); + else + pgd_populate(mm, pgd, new); + spin_unlock(&mm->page_table_lock); + return 0; +} +#endif /* __PAGETABLE_PUD_FOLDED */ + +#ifndef __PAGETABLE_PMD_FOLDED +/* + * Allocate page middle directory. + * We've already handled the fast-path in-line. + */ +int __pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address) +{ + pmd_t *new = pmd_alloc_one(mm, address); + if (!new) + return -ENOMEM; + + smp_wmb(); /* See comment in __pte_alloc */ + + spin_lock(&mm->page_table_lock); +#ifndef __ARCH_HAS_4LEVEL_HACK + if (pud_present(*pud)) /* Another has populated it */ + pmd_free(mm, new); + else + pud_populate(mm, pud, new); +#else + if (pgd_present(*pud)) /* Another has populated it */ + pmd_free(mm, new); + else + pgd_populate(mm, pud, new); +#endif /* __ARCH_HAS_4LEVEL_HACK */ + spin_unlock(&mm->page_table_lock); + return 0; +} +#endif /* __PAGETABLE_PMD_FOLDED */ + +#if !defined(__HAVE_ARCH_GATE_AREA) + +#if defined(AT_SYSINFO_EHDR) +static struct vm_area_struct gate_vma; + +static int __init gate_vma_init(void) +{ + gate_vma.vm_mm = NULL; + gate_vma.vm_start = FIXADDR_USER_START; + gate_vma.vm_end = FIXADDR_USER_END; + gate_vma.vm_flags = VM_READ | VM_MAYREAD | VM_EXEC | VM_MAYEXEC; + gate_vma.vm_page_prot = __P101; + + return 0; +} +__initcall(gate_vma_init); +#endif + +struct vm_area_struct *get_gate_vma(struct mm_struct *mm) +{ +#ifdef AT_SYSINFO_EHDR + return &gate_vma; +#else + return NULL; +#endif +} + +int in_gate_area_no_mm(unsigned long addr) +{ +#ifdef AT_SYSINFO_EHDR + if ((addr >= FIXADDR_USER_START) && (addr < FIXADDR_USER_END)) + return 1; +#endif + return 0; +} + +#endif /* __HAVE_ARCH_GATE_AREA */ + +static int __follow_pte(struct mm_struct *mm, unsigned long address, + pte_t **ptepp, spinlock_t **ptlp) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *ptep; + + pgd = pgd_offset(mm, address); + if (pgd_none(*pgd) || unlikely(pgd_bad(*pgd))) + goto out; + + pud = pud_offset(pgd, address); + if (pud_none(*pud) || unlikely(pud_bad(*pud))) + goto out; + + pmd = pmd_offset(pud, address); + VM_BUG_ON(pmd_trans_huge(*pmd)); + if (pmd_none(*pmd) || unlikely(pmd_bad(*pmd))) + goto out; + + /* We cannot handle huge page PFN maps. Luckily they don't exist. */ + if (pmd_huge(*pmd)) + goto out; + + ptep = pte_offset_map_lock(mm, pmd, address, ptlp); + if (!ptep) + goto out; + if (!pte_present(*ptep)) + goto unlock; + *ptepp = ptep; + return 0; +unlock: + pte_unmap_unlock(ptep, *ptlp); +out: + return -EINVAL; +} + +static inline int follow_pte(struct mm_struct *mm, unsigned long address, + pte_t **ptepp, spinlock_t **ptlp) +{ + int res; + + /* (void) is needed to make gcc happy */ + (void) __cond_lock(*ptlp, + !(res = __follow_pte(mm, address, ptepp, ptlp))); + return res; +} + +/** + * follow_pfn - look up PFN at a user virtual address + * @vma: memory mapping + * @address: user virtual address + * @pfn: location to store found PFN + * + * Only IO mappings and raw PFN mappings are allowed. + * + * Returns zero and the pfn at @pfn on success, -ve otherwise. + */ +int follow_pfn(struct vm_area_struct *vma, unsigned long address, + unsigned long *pfn) +{ + int ret = -EINVAL; + spinlock_t *ptl; + pte_t *ptep; + + if (!(vma->vm_flags & (VM_IO | VM_PFNMAP))) + return ret; + + ret = follow_pte(vma->vm_mm, address, &ptep, &ptl); + if (ret) + return ret; + *pfn = pte_pfn(*ptep); + pte_unmap_unlock(ptep, ptl); + return 0; +} +EXPORT_SYMBOL(follow_pfn); + +#ifdef CONFIG_HAVE_IOREMAP_PROT +int follow_phys(struct vm_area_struct *vma, + unsigned long address, unsigned int flags, + unsigned long *prot, resource_size_t *phys) +{ + int ret = -EINVAL; + pte_t *ptep, pte; + spinlock_t *ptl; + + if (!(vma->vm_flags & (VM_IO | VM_PFNMAP))) + goto out; + + if (follow_pte(vma->vm_mm, address, &ptep, &ptl)) + goto out; + pte = *ptep; + + if ((flags & FOLL_WRITE) && !pte_write(pte)) + goto unlock; + + *prot = pgprot_val(pte_pgprot(pte)); + *phys = (resource_size_t)pte_pfn(pte) << PAGE_SHIFT; + + ret = 0; +unlock: + pte_unmap_unlock(ptep, ptl); +out: + return ret; +} + +int generic_access_phys(struct vm_area_struct *vma, unsigned long addr, + void *buf, int len, int write) +{ + resource_size_t phys_addr; + unsigned long prot = 0; + void __iomem *maddr; + int offset = addr & (PAGE_SIZE-1); + + if (follow_phys(vma, addr, write, &prot, &phys_addr)) + return -EINVAL; + + maddr = ioremap_prot(phys_addr, PAGE_SIZE, prot); + if (write) + memcpy_toio(maddr + offset, buf, len); + else + memcpy_fromio(buf, maddr + offset, len); + iounmap(maddr); + + return len; +} +EXPORT_SYMBOL_GPL(generic_access_phys); +#endif + +/* + * Access another process' address space as given in mm. If non-NULL, use the + * given task for page fault accounting. + */ +static int __access_remote_vm(struct task_struct *tsk, struct mm_struct *mm, + unsigned long addr, void *buf, int len, int write) +{ + struct vm_area_struct *vma; + void *old_buf = buf; + + down_read(&mm->mmap_sem); + /* ignore errors, just check how much was successfully transferred */ + while (len) { + int bytes, ret, offset; + void *maddr; + struct page *page = NULL; + + ret = get_user_pages(tsk, mm, addr, 1, + write, 1, &page, &vma); + if (ret <= 0) { + /* + * Check if this is a VM_IO | VM_PFNMAP VMA, which + * we can access using slightly different code. + */ +#ifdef CONFIG_HAVE_IOREMAP_PROT + vma = find_vma(mm, addr); + if (!vma || vma->vm_start > addr) + break; + if (vma->vm_ops && vma->vm_ops->access) + ret = vma->vm_ops->access(vma, addr, buf, + len, write); + if (ret <= 0) +#endif + break; + bytes = ret; + } else { + bytes = len; + offset = addr & (PAGE_SIZE-1); + if (bytes > PAGE_SIZE-offset) + bytes = PAGE_SIZE-offset; + + maddr = kmap(page); + if (write) { + copy_to_user_page(vma, page, addr, + maddr + offset, buf, bytes); + set_page_dirty_lock(page); + } else { + copy_from_user_page(vma, page, addr, + buf, maddr + offset, bytes); + } + kunmap(page); + page_cache_release(page); + } + len -= bytes; + buf += bytes; + addr += bytes; + } + up_read(&mm->mmap_sem); + + return buf - old_buf; +} + +/** + * access_remote_vm - access another process' address space + * @mm: the mm_struct of the target address space + * @addr: start address to access + * @buf: source or destination buffer + * @len: number of bytes to transfer + * @write: whether the access is a write + * + * The caller must hold a reference on @mm. + */ +int access_remote_vm(struct mm_struct *mm, unsigned long addr, + void *buf, int len, int write) +{ + return __access_remote_vm(NULL, mm, addr, buf, len, write); +} + +/* + * Access another process' address space. + * Source/target buffer must be kernel space, + * Do not walk the page table directly, use get_user_pages + */ +int access_process_vm(struct task_struct *tsk, unsigned long addr, + void *buf, int len, int write) +{ + struct mm_struct *mm; + int ret; + + mm = get_task_mm(tsk); + if (!mm) + return 0; + + ret = __access_remote_vm(tsk, mm, addr, buf, len, write); + mmput(mm); + + return ret; +} + +/* + * Print the name of a VMA. + */ +void print_vma_addr(char *prefix, unsigned long ip) +{ + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma; + + /* + * Do not print if we are in atomic + * contexts (in exception stacks, etc.): + */ + if (preempt_count()) + return; + + down_read(&mm->mmap_sem); + vma = find_vma(mm, ip); + if (vma && vma->vm_file) { + struct file *f = vma->vm_file; + char *buf = (char *)__get_free_page(GFP_KERNEL); + if (buf) { + char *p; + + p = d_path(&f->f_path, buf, PAGE_SIZE); + if (IS_ERR(p)) + p = "?"; + printk("%s%s[%lx+%lx]", prefix, kbasename(p), + vma->vm_start, + vma->vm_end - vma->vm_start); + free_page((unsigned long)buf); + } + } + up_read(&mm->mmap_sem); +} + +#ifdef CONFIG_PROVE_LOCKING +void might_fault(void) +{ + /* + * Some code (nfs/sunrpc) uses socket ops on kernel memory while + * holding the mmap_sem, this is safe because kernel memory doesn't + * get paged out, therefore we'll never actually fault, and the + * below annotations will generate false positives. + */ + if (segment_eq(get_fs(), KERNEL_DS)) + return; + + might_sleep(); + /* + * it would be nicer only to annotate paths which are not under + * pagefault_disable, however that requires a larger audit and + * providing helpers like get_user_atomic. + */ + if (!in_atomic() && current->mm) + might_lock_read(¤t->mm->mmap_sem); +} +EXPORT_SYMBOL(might_fault); +#endif + +#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLBFS) +static void clear_gigantic_page(struct page *page, + unsigned long addr, + unsigned int pages_per_huge_page) +{ + int i; + struct page *p = page; + + might_sleep(); + for (i = 0; i < pages_per_huge_page; + i++, p = mem_map_next(p, page, i)) { + cond_resched(); + clear_user_highpage(p, addr + i * PAGE_SIZE); + } +} +void clear_huge_page(struct page *page, + unsigned long addr, unsigned int pages_per_huge_page) +{ + int i; + + if (unlikely(pages_per_huge_page > MAX_ORDER_NR_PAGES)) { + clear_gigantic_page(page, addr, pages_per_huge_page); + return; + } + + might_sleep(); + for (i = 0; i < pages_per_huge_page; i++) { + cond_resched(); + clear_user_highpage(page + i, addr + i * PAGE_SIZE); + } +} + +static void copy_user_gigantic_page(struct page *dst, struct page *src, + unsigned long addr, + struct vm_area_struct *vma, + unsigned int pages_per_huge_page) +{ + int i; + struct page *dst_base = dst; + struct page *src_base = src; + + for (i = 0; i < pages_per_huge_page; ) { + cond_resched(); + copy_user_highpage(dst, src, addr + i*PAGE_SIZE, vma); + + i++; + dst = mem_map_next(dst, dst_base, i); + src = mem_map_next(src, src_base, i); + } +} + +void copy_user_huge_page(struct page *dst, struct page *src, + unsigned long addr, struct vm_area_struct *vma, + unsigned int pages_per_huge_page) +{ + int i; + + if (unlikely(pages_per_huge_page > MAX_ORDER_NR_PAGES)) { + copy_user_gigantic_page(dst, src, addr, vma, + pages_per_huge_page); + return; + } + + might_sleep(); + for (i = 0; i < pages_per_huge_page; i++) { + cond_resched(); + copy_user_highpage(dst + i, src + i, addr + i*PAGE_SIZE, vma); + } +} +#endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLBFS */ diff -Naur kernel.21.3.orig/mm/mmap.c kernel.21.3.new/mm/mmap.c --- kernel.21.3.orig/mm/mmap.c 2015-02-03 01:44:33.000000000 +0000 +++ kernel.21.3.new/mm/mmap.c 2015-04-25 23:28:45.398480745 +0000 @@ -251,7 +251,7 @@ if (vma->vm_ops && vma->vm_ops->close) vma->vm_ops->close(vma); if (vma->vm_file) - fput(vma->vm_file); + vma_fput(vma); mpol_put(vma_policy(vma)); kmem_cache_free(vm_area_cachep, vma); return next; @@ -861,7 +861,7 @@ if (remove_next) { if (file) { uprobe_munmap(next, next->vm_start, next->vm_end); - fput(file); + vma_fput(vma); } if (next->anon_vma) anon_vma_merge(vma, next); @@ -1628,8 +1628,8 @@ unmap_and_free_vma: if (correct_wcount) atomic_inc(&inode->i_writecount); + vma_fput(vma); vma->vm_file = NULL; - fput(file); /* Undo any partial mapping done by a device driver. */ unmap_region(mm, vma, prev, vma->vm_start, vma->vm_end); @@ -2452,7 +2452,7 @@ goto out_free_mpol; if (new->vm_file) - get_file(new->vm_file); + vma_get_file(new); if (new->vm_ops && new->vm_ops->open) new->vm_ops->open(new); @@ -2471,7 +2471,7 @@ if (new->vm_ops && new->vm_ops->close) new->vm_ops->close(new); if (new->vm_file) - fput(new->vm_file); + vma_fput(new); unlink_anon_vmas(new); out_free_mpol: mpol_put(pol); @@ -2873,7 +2873,7 @@ if (anon_vma_clone(new_vma, vma)) goto out_free_mempol; if (new_vma->vm_file) - get_file(new_vma->vm_file); + vma_get_file(new_vma); if (new_vma->vm_ops && new_vma->vm_ops->open) new_vma->vm_ops->open(new_vma); vma_link(mm, new_vma, prev, rb_link, rb_parent); diff -Naur kernel.21.3.orig/mm/mmap.c.orig kernel.21.3.new/mm/mmap.c.orig --- kernel.21.3.orig/mm/mmap.c.orig 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/mm/mmap.c.orig 2015-02-03 01:44:33.000000000 +0000 @@ -0,0 +1,3281 @@ +/* + * mm/mmap.c + * + * Written by obz. + * + * Address space accounting code + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +#include "internal.h" + +#ifndef arch_mmap_check +#define arch_mmap_check(addr, len, flags) (0) +#endif + +#ifndef arch_rebalance_pgtables +#define arch_rebalance_pgtables(addr, len) (addr) +#endif + +static void unmap_region(struct mm_struct *mm, + struct vm_area_struct *vma, struct vm_area_struct *prev, + unsigned long start, unsigned long end); + +/* description of effects of mapping type and prot in current implementation. + * this is due to the limited x86 page protection hardware. The expected + * behavior is in parens: + * + * map_type prot + * PROT_NONE PROT_READ PROT_WRITE PROT_EXEC + * MAP_SHARED r: (no) no r: (yes) yes r: (no) yes r: (no) yes + * w: (no) no w: (no) no w: (yes) yes w: (no) no + * x: (no) no x: (no) yes x: (no) yes x: (yes) yes + * + * MAP_PRIVATE r: (no) no r: (yes) yes r: (no) yes r: (no) yes + * w: (no) no w: (no) no w: (copy) copy w: (no) no + * x: (no) no x: (no) yes x: (no) yes x: (yes) yes + * + */ +pgprot_t protection_map[16] = { + __P000, __P001, __P010, __P011, __P100, __P101, __P110, __P111, + __S000, __S001, __S010, __S011, __S100, __S101, __S110, __S111 +}; + +pgprot_t vm_get_page_prot(unsigned long vm_flags) +{ + return __pgprot(pgprot_val(protection_map[vm_flags & + (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]) | + pgprot_val(arch_vm_get_page_prot(vm_flags))); +} +EXPORT_SYMBOL(vm_get_page_prot); + +int sysctl_overcommit_memory __read_mostly = OVERCOMMIT_GUESS; /* heuristic overcommit */ +int sysctl_overcommit_ratio __read_mostly = 50; /* default is 50% */ +int sysctl_max_map_count __read_mostly = DEFAULT_MAX_MAP_COUNT; +unsigned long sysctl_user_reserve_kbytes __read_mostly = 1UL << 17; /* 128MB */ +unsigned long sysctl_admin_reserve_kbytes __read_mostly = 1UL << 13; /* 8MB */ +/* + * Make sure vm_committed_as in one cacheline and not cacheline shared with + * other variables. It can be updated by several CPUs frequently. + */ +struct percpu_counter vm_committed_as ____cacheline_aligned_in_smp; + +/* + * The global memory commitment made in the system can be a metric + * that can be used to drive ballooning decisions when Linux is hosted + * as a guest. On Hyper-V, the host implements a policy engine for dynamically + * balancing memory across competing virtual machines that are hosted. + * Several metrics drive this policy engine including the guest reported + * memory commitment. + */ +unsigned long vm_memory_committed(void) +{ + return percpu_counter_read_positive(&vm_committed_as); +} +EXPORT_SYMBOL_GPL(vm_memory_committed); + +/* + * Check that a process has enough memory to allocate a new virtual + * mapping. 0 means there is enough memory for the allocation to + * succeed and -ENOMEM implies there is not. + * + * We currently support three overcommit policies, which are set via the + * vm.overcommit_memory sysctl. See Documentation/vm/overcommit-accounting + * + * Strict overcommit modes added 2002 Feb 26 by Alan Cox. + * Additional code 2002 Jul 20 by Robert Love. + * + * cap_sys_admin is 1 if the process has admin privileges, 0 otherwise. + * + * Note this is a helper function intended to be used by LSMs which + * wish to use this logic. + */ +int __vm_enough_memory(struct mm_struct *mm, long pages, int cap_sys_admin) +{ + unsigned long free, allowed, reserve; + + vm_acct_memory(pages); + + /* + * Sometimes we want to use more memory than we have + */ + if (sysctl_overcommit_memory == OVERCOMMIT_ALWAYS) + return 0; + + if (sysctl_overcommit_memory == OVERCOMMIT_GUESS) { + free = global_page_state(NR_FREE_PAGES); + free += global_page_state(NR_FILE_PAGES); + + /* + * shmem pages shouldn't be counted as free in this + * case, they can't be purged, only swapped out, and + * that won't affect the overall amount of available + * memory in the system. + */ + free -= global_page_state(NR_SHMEM); + + free += get_nr_swap_pages(); + + /* + * Any slabs which are created with the + * SLAB_RECLAIM_ACCOUNT flag claim to have contents + * which are reclaimable, under pressure. The dentry + * cache and most inode caches should fall into this + */ + free += global_page_state(NR_SLAB_RECLAIMABLE); + + /* + * Leave reserved pages. The pages are not for anonymous pages. + */ + if (free <= totalreserve_pages) + goto error; + else + free -= totalreserve_pages; + + /* + * Reserve some for root + */ + if (!cap_sys_admin) + free -= sysctl_admin_reserve_kbytes >> (PAGE_SHIFT - 10); + + if (free > pages) + return 0; + + goto error; + } + + allowed = (totalram_pages - hugetlb_total_pages()) + * sysctl_overcommit_ratio / 100; + /* + * Reserve some for root + */ + if (!cap_sys_admin) + allowed -= sysctl_admin_reserve_kbytes >> (PAGE_SHIFT - 10); + allowed += total_swap_pages; + + /* + * Don't let a single process grow so big a user can't recover + */ + if (mm) { + reserve = sysctl_user_reserve_kbytes >> (PAGE_SHIFT - 10); + allowed -= min(mm->total_vm / 32, reserve); + } + + if (percpu_counter_read_positive(&vm_committed_as) < allowed) + return 0; +error: + vm_unacct_memory(pages); + + return -ENOMEM; +} + +/* + * Requires inode->i_mapping->i_mmap_mutex + */ +static void __remove_shared_vm_struct(struct vm_area_struct *vma, + struct file *file, struct address_space *mapping) +{ + if (vma->vm_flags & VM_DENYWRITE) + atomic_inc(&file_inode(file)->i_writecount); + if (vma->vm_flags & VM_SHARED) + mapping->i_mmap_writable--; + + flush_dcache_mmap_lock(mapping); + if (unlikely(vma->vm_flags & VM_NONLINEAR)) + list_del_init(&vma->shared.nonlinear); + else + vma_interval_tree_remove(vma, &mapping->i_mmap); + flush_dcache_mmap_unlock(mapping); +} + +/* + * Unlink a file-based vm structure from its interval tree, to hide + * vma from rmap and vmtruncate before freeing its page tables. + */ +void unlink_file_vma(struct vm_area_struct *vma) +{ + struct file *file = vma->vm_file; + + if (file) { + struct address_space *mapping = file->f_mapping; + mutex_lock(&mapping->i_mmap_mutex); + __remove_shared_vm_struct(vma, file, mapping); + mutex_unlock(&mapping->i_mmap_mutex); + } +} + +/* + * Close a vm structure and free it, returning the next. + */ +static struct vm_area_struct *remove_vma(struct vm_area_struct *vma) +{ + struct vm_area_struct *next = vma->vm_next; + + might_sleep(); + if (vma->vm_ops && vma->vm_ops->close) + vma->vm_ops->close(vma); + if (vma->vm_file) + fput(vma->vm_file); + mpol_put(vma_policy(vma)); + kmem_cache_free(vm_area_cachep, vma); + return next; +} + +static unsigned long do_brk(unsigned long addr, unsigned long len); + +SYSCALL_DEFINE1(brk, unsigned long, brk) +{ + unsigned long rlim, retval; + unsigned long newbrk, oldbrk; + struct mm_struct *mm = current->mm; + unsigned long min_brk; + bool populate; + + down_write(&mm->mmap_sem); + +#ifdef CONFIG_COMPAT_BRK + /* + * CONFIG_COMPAT_BRK can still be overridden by setting + * randomize_va_space to 2, which will still cause mm->start_brk + * to be arbitrarily shifted + */ + if (current->brk_randomized) + min_brk = mm->start_brk; + else + min_brk = mm->end_data; +#else + min_brk = mm->start_brk; +#endif + if (brk < min_brk) + goto out; + + /* + * Check against rlimit here. If this check is done later after the test + * of oldbrk with newbrk then it can escape the test and let the data + * segment grow beyond its set limit the in case where the limit is + * not page aligned -Ram Gupta + */ + rlim = rlimit(RLIMIT_DATA); + if (rlim < RLIM_INFINITY && (brk - mm->start_brk) + + (mm->end_data - mm->start_data) > rlim) + goto out; + + newbrk = PAGE_ALIGN(brk); + oldbrk = PAGE_ALIGN(mm->brk); + if (oldbrk == newbrk) + goto set_brk; + + /* Always allow shrinking brk. */ + if (brk <= mm->brk) { + if (!do_munmap(mm, newbrk, oldbrk-newbrk)) + goto set_brk; + goto out; + } + + /* Check against existing mmap mappings. */ + if (find_vma_intersection(mm, oldbrk, newbrk+PAGE_SIZE)) + goto out; + + /* Ok, looks good - let it rip. */ + if (do_brk(oldbrk, newbrk-oldbrk) != oldbrk) + goto out; + +set_brk: + mm->brk = brk; + populate = newbrk > oldbrk && (mm->def_flags & VM_LOCKED) != 0; + up_write(&mm->mmap_sem); + if (populate) + mm_populate(oldbrk, newbrk - oldbrk); + return brk; + +out: + retval = mm->brk; + up_write(&mm->mmap_sem); + return retval; +} + +static long vma_compute_subtree_gap(struct vm_area_struct *vma) +{ + unsigned long max, subtree_gap; + max = vma->vm_start; + if (vma->vm_prev) + max -= vma->vm_prev->vm_end; + if (vma->vm_rb.rb_left) { + subtree_gap = rb_entry(vma->vm_rb.rb_left, + struct vm_area_struct, vm_rb)->rb_subtree_gap; + if (subtree_gap > max) + max = subtree_gap; + } + if (vma->vm_rb.rb_right) { + subtree_gap = rb_entry(vma->vm_rb.rb_right, + struct vm_area_struct, vm_rb)->rb_subtree_gap; + if (subtree_gap > max) + max = subtree_gap; + } + return max; +} + +#ifdef CONFIG_DEBUG_VM_RB +static int browse_rb(struct rb_root *root) +{ + int i = 0, j, bug = 0; + struct rb_node *nd, *pn = NULL; + unsigned long prev = 0, pend = 0; + + for (nd = rb_first(root); nd; nd = rb_next(nd)) { + struct vm_area_struct *vma; + vma = rb_entry(nd, struct vm_area_struct, vm_rb); + if (vma->vm_start < prev) { + printk("vm_start %lx prev %lx\n", vma->vm_start, prev); + bug = 1; + } + if (vma->vm_start < pend) { + printk("vm_start %lx pend %lx\n", vma->vm_start, pend); + bug = 1; + } + if (vma->vm_start > vma->vm_end) { + printk("vm_end %lx < vm_start %lx\n", + vma->vm_end, vma->vm_start); + bug = 1; + } + if (vma->rb_subtree_gap != vma_compute_subtree_gap(vma)) { + printk("free gap %lx, correct %lx\n", + vma->rb_subtree_gap, + vma_compute_subtree_gap(vma)); + bug = 1; + } + i++; + pn = nd; + prev = vma->vm_start; + pend = vma->vm_end; + } + j = 0; + for (nd = pn; nd; nd = rb_prev(nd)) + j++; + if (i != j) { + printk("backwards %d, forwards %d\n", j, i); + bug = 1; + } + return bug ? -1 : i; +} + +static void validate_mm_rb(struct rb_root *root, struct vm_area_struct *ignore) +{ + struct rb_node *nd; + + for (nd = rb_first(root); nd; nd = rb_next(nd)) { + struct vm_area_struct *vma; + vma = rb_entry(nd, struct vm_area_struct, vm_rb); + BUG_ON(vma != ignore && + vma->rb_subtree_gap != vma_compute_subtree_gap(vma)); + } +} + +void validate_mm(struct mm_struct *mm) +{ + int bug = 0; + int i = 0; + unsigned long highest_address = 0; + struct vm_area_struct *vma = mm->mmap; + while (vma) { + struct anon_vma_chain *avc; + vma_lock_anon_vma(vma); + list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) + anon_vma_interval_tree_verify(avc); + vma_unlock_anon_vma(vma); + highest_address = vma->vm_end; + vma = vma->vm_next; + i++; + } + if (i != mm->map_count) { + printk("map_count %d vm_next %d\n", mm->map_count, i); + bug = 1; + } + if (highest_address != mm->highest_vm_end) { + printk("mm->highest_vm_end %lx, found %lx\n", + mm->highest_vm_end, highest_address); + bug = 1; + } + i = browse_rb(&mm->mm_rb); + if (i != mm->map_count) { + printk("map_count %d rb %d\n", mm->map_count, i); + bug = 1; + } + BUG_ON(bug); +} +#else +#define validate_mm_rb(root, ignore) do { } while (0) +#define validate_mm(mm) do { } while (0) +#endif + +RB_DECLARE_CALLBACKS(static, vma_gap_callbacks, struct vm_area_struct, vm_rb, + unsigned long, rb_subtree_gap, vma_compute_subtree_gap) + +/* + * Update augmented rbtree rb_subtree_gap values after vma->vm_start or + * vma->vm_prev->vm_end values changed, without modifying the vma's position + * in the rbtree. + */ +static void vma_gap_update(struct vm_area_struct *vma) +{ + /* + * As it turns out, RB_DECLARE_CALLBACKS() already created a callback + * function that does exacltly what we want. + */ + vma_gap_callbacks_propagate(&vma->vm_rb, NULL); +} + +static inline void vma_rb_insert(struct vm_area_struct *vma, + struct rb_root *root) +{ + /* All rb_subtree_gap values must be consistent prior to insertion */ + validate_mm_rb(root, NULL); + + rb_insert_augmented(&vma->vm_rb, root, &vma_gap_callbacks); +} + +static void vma_rb_erase(struct vm_area_struct *vma, struct rb_root *root) +{ + /* + * All rb_subtree_gap values must be consistent prior to erase, + * with the possible exception of the vma being erased. + */ + validate_mm_rb(root, vma); + + /* + * Note rb_erase_augmented is a fairly large inline function, + * so make sure we instantiate it only once with our desired + * augmented rbtree callbacks. + */ + rb_erase_augmented(&vma->vm_rb, root, &vma_gap_callbacks); +} + +/* + * vma has some anon_vma assigned, and is already inserted on that + * anon_vma's interval trees. + * + * Before updating the vma's vm_start / vm_end / vm_pgoff fields, the + * vma must be removed from the anon_vma's interval trees using + * anon_vma_interval_tree_pre_update_vma(). + * + * After the update, the vma will be reinserted using + * anon_vma_interval_tree_post_update_vma(). + * + * The entire update must be protected by exclusive mmap_sem and by + * the root anon_vma's mutex. + */ +static inline void +anon_vma_interval_tree_pre_update_vma(struct vm_area_struct *vma) +{ + struct anon_vma_chain *avc; + + list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) + anon_vma_interval_tree_remove(avc, &avc->anon_vma->rb_root); +} + +static inline void +anon_vma_interval_tree_post_update_vma(struct vm_area_struct *vma) +{ + struct anon_vma_chain *avc; + + list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) + anon_vma_interval_tree_insert(avc, &avc->anon_vma->rb_root); +} + +static int find_vma_links(struct mm_struct *mm, unsigned long addr, + unsigned long end, struct vm_area_struct **pprev, + struct rb_node ***rb_link, struct rb_node **rb_parent) +{ + struct rb_node **__rb_link, *__rb_parent, *rb_prev; + + __rb_link = &mm->mm_rb.rb_node; + rb_prev = __rb_parent = NULL; + + while (*__rb_link) { + struct vm_area_struct *vma_tmp; + + __rb_parent = *__rb_link; + vma_tmp = rb_entry(__rb_parent, struct vm_area_struct, vm_rb); + + if (vma_tmp->vm_end > addr) { + /* Fail if an existing vma overlaps the area */ + if (vma_tmp->vm_start < end) + return -ENOMEM; + __rb_link = &__rb_parent->rb_left; + } else { + rb_prev = __rb_parent; + __rb_link = &__rb_parent->rb_right; + } + } + + *pprev = NULL; + if (rb_prev) + *pprev = rb_entry(rb_prev, struct vm_area_struct, vm_rb); + *rb_link = __rb_link; + *rb_parent = __rb_parent; + return 0; +} + +static unsigned long count_vma_pages_range(struct mm_struct *mm, + unsigned long addr, unsigned long end) +{ + unsigned long nr_pages = 0; + struct vm_area_struct *vma; + + /* Find first overlaping mapping */ + vma = find_vma_intersection(mm, addr, end); + if (!vma) + return 0; + + nr_pages = (min(end, vma->vm_end) - + max(addr, vma->vm_start)) >> PAGE_SHIFT; + + /* Iterate over the rest of the overlaps */ + for (vma = vma->vm_next; vma; vma = vma->vm_next) { + unsigned long overlap_len; + + if (vma->vm_start > end) + break; + + overlap_len = min(end, vma->vm_end) - vma->vm_start; + nr_pages += overlap_len >> PAGE_SHIFT; + } + + return nr_pages; +} + +void __vma_link_rb(struct mm_struct *mm, struct vm_area_struct *vma, + struct rb_node **rb_link, struct rb_node *rb_parent) +{ + /* Update tracking information for the gap following the new vma. */ + if (vma->vm_next) + vma_gap_update(vma->vm_next); + else + mm->highest_vm_end = vma->vm_end; + + /* + * vma->vm_prev wasn't known when we followed the rbtree to find the + * correct insertion point for that vma. As a result, we could not + * update the vma vm_rb parents rb_subtree_gap values on the way down. + * So, we first insert the vma with a zero rb_subtree_gap value + * (to be consistent with what we did on the way down), and then + * immediately update the gap to the correct value. Finally we + * rebalance the rbtree after all augmented values have been set. + */ + rb_link_node(&vma->vm_rb, rb_parent, rb_link); + vma->rb_subtree_gap = 0; + vma_gap_update(vma); + vma_rb_insert(vma, &mm->mm_rb); +} + +static void __vma_link_file(struct vm_area_struct *vma) +{ + struct file *file; + + file = vma->vm_file; + if (file) { + struct address_space *mapping = file->f_mapping; + + if (vma->vm_flags & VM_DENYWRITE) + atomic_dec(&file_inode(file)->i_writecount); + if (vma->vm_flags & VM_SHARED) + mapping->i_mmap_writable++; + + flush_dcache_mmap_lock(mapping); + if (unlikely(vma->vm_flags & VM_NONLINEAR)) + vma_nonlinear_insert(vma, &mapping->i_mmap_nonlinear); + else + vma_interval_tree_insert(vma, &mapping->i_mmap); + flush_dcache_mmap_unlock(mapping); + } +} + +static void +__vma_link(struct mm_struct *mm, struct vm_area_struct *vma, + struct vm_area_struct *prev, struct rb_node **rb_link, + struct rb_node *rb_parent) +{ + __vma_link_list(mm, vma, prev, rb_parent); + __vma_link_rb(mm, vma, rb_link, rb_parent); +} + +static void vma_link(struct mm_struct *mm, struct vm_area_struct *vma, + struct vm_area_struct *prev, struct rb_node **rb_link, + struct rb_node *rb_parent) +{ + struct address_space *mapping = NULL; + + if (vma->vm_file) + mapping = vma->vm_file->f_mapping; + + if (mapping) + mutex_lock(&mapping->i_mmap_mutex); + + __vma_link(mm, vma, prev, rb_link, rb_parent); + __vma_link_file(vma); + + if (mapping) + mutex_unlock(&mapping->i_mmap_mutex); + + mm->map_count++; + validate_mm(mm); +} + +/* + * Helper for vma_adjust() in the split_vma insert case: insert a vma into the + * mm's list and rbtree. It has already been inserted into the interval tree. + */ +static void __insert_vm_struct(struct mm_struct *mm, struct vm_area_struct *vma) +{ + struct vm_area_struct *prev; + struct rb_node **rb_link, *rb_parent; + + if (find_vma_links(mm, vma->vm_start, vma->vm_end, + &prev, &rb_link, &rb_parent)) + BUG(); + __vma_link(mm, vma, prev, rb_link, rb_parent); + mm->map_count++; +} + +static inline void +__vma_unlink(struct mm_struct *mm, struct vm_area_struct *vma, + struct vm_area_struct *prev) +{ + struct vm_area_struct *next; + + vma_rb_erase(vma, &mm->mm_rb); + prev->vm_next = next = vma->vm_next; + if (next) + next->vm_prev = prev; + if (mm->mmap_cache == vma) + mm->mmap_cache = prev; +} + +/* + * We cannot adjust vm_start, vm_end, vm_pgoff fields of a vma that + * is already present in an i_mmap tree without adjusting the tree. + * The following helper function should be used when such adjustments + * are necessary. The "insert" vma (if any) is to be inserted + * before we drop the necessary locks. + */ +int vma_adjust(struct vm_area_struct *vma, unsigned long start, + unsigned long end, pgoff_t pgoff, struct vm_area_struct *insert) +{ + struct mm_struct *mm = vma->vm_mm; + struct vm_area_struct *next = vma->vm_next; + struct vm_area_struct *importer = NULL; + struct address_space *mapping = NULL; + struct rb_root *root = NULL; + struct anon_vma *anon_vma = NULL; + struct file *file = vma->vm_file; + bool start_changed = false, end_changed = false; + long adjust_next = 0; + int remove_next = 0; + + if (next && !insert) { + struct vm_area_struct *exporter = NULL; + + if (end >= next->vm_end) { + /* + * vma expands, overlapping all the next, and + * perhaps the one after too (mprotect case 6). + */ +again: remove_next = 1 + (end > next->vm_end); + end = next->vm_end; + exporter = next; + importer = vma; + } else if (end > next->vm_start) { + /* + * vma expands, overlapping part of the next: + * mprotect case 5 shifting the boundary up. + */ + adjust_next = (end - next->vm_start) >> PAGE_SHIFT; + exporter = next; + importer = vma; + } else if (end < vma->vm_end) { + /* + * vma shrinks, and !insert tells it's not + * split_vma inserting another: so it must be + * mprotect case 4 shifting the boundary down. + */ + adjust_next = - ((vma->vm_end - end) >> PAGE_SHIFT); + exporter = vma; + importer = next; + } + + /* + * Easily overlooked: when mprotect shifts the boundary, + * make sure the expanding vma has anon_vma set if the + * shrinking vma had, to cover any anon pages imported. + */ + if (exporter && exporter->anon_vma && !importer->anon_vma) { + if (anon_vma_clone(importer, exporter)) + return -ENOMEM; + importer->anon_vma = exporter->anon_vma; + } + } + + if (file) { + mapping = file->f_mapping; + if (!(vma->vm_flags & VM_NONLINEAR)) { + root = &mapping->i_mmap; + uprobe_munmap(vma, vma->vm_start, vma->vm_end); + + if (adjust_next) + uprobe_munmap(next, next->vm_start, + next->vm_end); + } + + mutex_lock(&mapping->i_mmap_mutex); + if (insert) { + /* + * Put into interval tree now, so instantiated pages + * are visible to arm/parisc __flush_dcache_page + * throughout; but we cannot insert into address + * space until vma start or end is updated. + */ + __vma_link_file(insert); + } + } + + vma_adjust_trans_huge(vma, start, end, adjust_next); + + anon_vma = vma->anon_vma; + if (!anon_vma && adjust_next) + anon_vma = next->anon_vma; + if (anon_vma) { + VM_BUG_ON(adjust_next && next->anon_vma && + anon_vma != next->anon_vma); + anon_vma_lock_write(anon_vma); + anon_vma_interval_tree_pre_update_vma(vma); + if (adjust_next) + anon_vma_interval_tree_pre_update_vma(next); + } + + if (root) { + flush_dcache_mmap_lock(mapping); + vma_interval_tree_remove(vma, root); + if (adjust_next) + vma_interval_tree_remove(next, root); + } + + if (start != vma->vm_start) { + vma->vm_start = start; + start_changed = true; + } + if (end != vma->vm_end) { + vma->vm_end = end; + end_changed = true; + } + vma->vm_pgoff = pgoff; + if (adjust_next) { + next->vm_start += adjust_next << PAGE_SHIFT; + next->vm_pgoff += adjust_next; + } + + if (root) { + if (adjust_next) + vma_interval_tree_insert(next, root); + vma_interval_tree_insert(vma, root); + flush_dcache_mmap_unlock(mapping); + } + + if (remove_next) { + /* + * vma_merge has merged next into vma, and needs + * us to remove next before dropping the locks. + */ + __vma_unlink(mm, next, vma); + if (file) + __remove_shared_vm_struct(next, file, mapping); + } else if (insert) { + /* + * split_vma has split insert from vma, and needs + * us to insert it before dropping the locks + * (it may either follow vma or precede it). + */ + __insert_vm_struct(mm, insert); + } else { + if (start_changed) + vma_gap_update(vma); + if (end_changed) { + if (!next) + mm->highest_vm_end = end; + else if (!adjust_next) + vma_gap_update(next); + } + } + + if (anon_vma) { + anon_vma_interval_tree_post_update_vma(vma); + if (adjust_next) + anon_vma_interval_tree_post_update_vma(next); + anon_vma_unlock_write(anon_vma); + } + if (mapping) + mutex_unlock(&mapping->i_mmap_mutex); + + if (root) { + uprobe_mmap(vma); + + if (adjust_next) + uprobe_mmap(next); + } + + if (remove_next) { + if (file) { + uprobe_munmap(next, next->vm_start, next->vm_end); + fput(file); + } + if (next->anon_vma) + anon_vma_merge(vma, next); + mm->map_count--; + mpol_put(vma_policy(next)); + kmem_cache_free(vm_area_cachep, next); + /* + * In mprotect's case 6 (see comments on vma_merge), + * we must remove another next too. It would clutter + * up the code too much to do both in one go. + */ + next = vma->vm_next; + if (remove_next == 2) + goto again; + else if (next) + vma_gap_update(next); + else + mm->highest_vm_end = end; + } + if (insert && file) + uprobe_mmap(insert); + + validate_mm(mm); + + return 0; +} + +/* + * If the vma has a ->close operation then the driver probably needs to release + * per-vma resources, so we don't attempt to merge those. + */ +static inline int is_mergeable_vma(struct vm_area_struct *vma, + struct file *file, unsigned long vm_flags, + const char __user *anon_name) +{ + if (vma->vm_flags ^ vm_flags) + return 0; + if (vma->vm_file != file) + return 0; + if (vma->vm_ops && vma->vm_ops->close) + return 0; + if (vma_get_anon_name(vma) != anon_name) + return 0; + return 1; +} + +static inline int is_mergeable_anon_vma(struct anon_vma *anon_vma1, + struct anon_vma *anon_vma2, + struct vm_area_struct *vma) +{ + /* + * The list_is_singular() test is to avoid merging VMA cloned from + * parents. This can improve scalability caused by anon_vma lock. + */ + if ((!anon_vma1 || !anon_vma2) && (!vma || + list_is_singular(&vma->anon_vma_chain))) + return 1; + return anon_vma1 == anon_vma2; +} + +/* + * Return true if we can merge this (vm_flags,anon_vma,file,vm_pgoff) + * in front of (at a lower virtual address and file offset than) the vma. + * + * We cannot merge two vmas if they have differently assigned (non-NULL) + * anon_vmas, nor if same anon_vma is assigned but offsets incompatible. + * + * We don't check here for the merged mmap wrapping around the end of pagecache + * indices (16TB on ia32) because do_mmap_pgoff() does not permit mmap's which + * wrap, nor mmaps which cover the final page at index -1UL. + */ +static int +can_vma_merge_before(struct vm_area_struct *vma, unsigned long vm_flags, + struct anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff, + const char __user *anon_name) +{ + if (is_mergeable_vma(vma, file, vm_flags, anon_name) && + is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) { + if (vma->vm_pgoff == vm_pgoff) + return 1; + } + return 0; +} + +/* + * Return true if we can merge this (vm_flags,anon_vma,file,vm_pgoff) + * beyond (at a higher virtual address and file offset than) the vma. + * + * We cannot merge two vmas if they have differently assigned (non-NULL) + * anon_vmas, nor if same anon_vma is assigned but offsets incompatible. + */ +static int +can_vma_merge_after(struct vm_area_struct *vma, unsigned long vm_flags, + struct anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff, + const char __user *anon_name) +{ + if (is_mergeable_vma(vma, file, vm_flags, anon_name) && + is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) { + pgoff_t vm_pglen; + vm_pglen = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT; + if (vma->vm_pgoff + vm_pglen == vm_pgoff) + return 1; + } + return 0; +} + +/* + * Given a mapping request (addr,end,vm_flags,file,pgoff,anon_name), + * figure out whether that can be merged with its predecessor or its + * successor. Or both (it neatly fills a hole). + * + * In most cases - when called for mmap, brk or mremap - [addr,end) is + * certain not to be mapped by the time vma_merge is called; but when + * called for mprotect, it is certain to be already mapped (either at + * an offset within prev, or at the start of next), and the flags of + * this area are about to be changed to vm_flags - and the no-change + * case has already been eliminated. + * + * The following mprotect cases have to be considered, where AAAA is + * the area passed down from mprotect_fixup, never extending beyond one + * vma, PPPPPP is the prev vma specified, and NNNNNN the next vma after: + * + * AAAA AAAA AAAA AAAA + * PPPPPPNNNNNN PPPPPPNNNNNN PPPPPPNNNNNN PPPPNNNNXXXX + * cannot merge might become might become might become + * PPNNNNNNNNNN PPPPPPPPPPNN PPPPPPPPPPPP 6 or + * mmap, brk or case 4 below case 5 below PPPPPPPPXXXX 7 or + * mremap move: PPPPNNNNNNNN 8 + * AAAA + * PPPP NNNN PPPPPPPPPPPP PPPPPPPPNNNN PPPPNNNNNNNN + * might become case 1 below case 2 below case 3 below + * + * Odd one out? Case 8, because it extends NNNN but needs flags of XXXX: + * mprotect_fixup updates vm_flags & vm_page_prot on successful return. + */ +struct vm_area_struct *vma_merge(struct mm_struct *mm, + struct vm_area_struct *prev, unsigned long addr, + unsigned long end, unsigned long vm_flags, + struct anon_vma *anon_vma, struct file *file, + pgoff_t pgoff, struct mempolicy *policy, + const char __user *anon_name) +{ + pgoff_t pglen = (end - addr) >> PAGE_SHIFT; + struct vm_area_struct *area, *next; + int err; + + /* + * We later require that vma->vm_flags == vm_flags, + * so this tests vma->vm_flags & VM_SPECIAL, too. + */ + if (vm_flags & VM_SPECIAL) + return NULL; + + if (prev) + next = prev->vm_next; + else + next = mm->mmap; + area = next; + if (next && next->vm_end == end) /* cases 6, 7, 8 */ + next = next->vm_next; + + /* + * Can it merge with the predecessor? + */ + if (prev && prev->vm_end == addr && + mpol_equal(vma_policy(prev), policy) && + can_vma_merge_after(prev, vm_flags, anon_vma, + file, pgoff, anon_name)) { + /* + * OK, it can. Can we now merge in the successor as well? + */ + if (next && end == next->vm_start && + mpol_equal(policy, vma_policy(next)) && + can_vma_merge_before(next, vm_flags, anon_vma, + file, pgoff+pglen, anon_name) && + is_mergeable_anon_vma(prev->anon_vma, + next->anon_vma, NULL)) { + /* cases 1, 6 */ + err = vma_adjust(prev, prev->vm_start, + next->vm_end, prev->vm_pgoff, NULL); + } else /* cases 2, 5, 7 */ + err = vma_adjust(prev, prev->vm_start, + end, prev->vm_pgoff, NULL); + if (err) + return NULL; + khugepaged_enter_vma_merge(prev); + return prev; + } + + /* + * Can this new request be merged in front of next? + */ + if (next && end == next->vm_start && + mpol_equal(policy, vma_policy(next)) && + can_vma_merge_before(next, vm_flags, anon_vma, + file, pgoff+pglen, anon_name)) { + if (prev && addr < prev->vm_end) /* case 4 */ + err = vma_adjust(prev, prev->vm_start, + addr, prev->vm_pgoff, NULL); + else /* cases 3, 8 */ + err = vma_adjust(area, addr, next->vm_end, + next->vm_pgoff - pglen, NULL); + if (err) + return NULL; + khugepaged_enter_vma_merge(area); + return area; + } + + return NULL; +} + +/* + * Rough compatbility check to quickly see if it's even worth looking + * at sharing an anon_vma. + * + * They need to have the same vm_file, and the flags can only differ + * in things that mprotect may change. + * + * NOTE! The fact that we share an anon_vma doesn't _have_ to mean that + * we can merge the two vma's. For example, we refuse to merge a vma if + * there is a vm_ops->close() function, because that indicates that the + * driver is doing some kind of reference counting. But that doesn't + * really matter for the anon_vma sharing case. + */ +static int anon_vma_compatible(struct vm_area_struct *a, struct vm_area_struct *b) +{ + return a->vm_end == b->vm_start && + mpol_equal(vma_policy(a), vma_policy(b)) && + a->vm_file == b->vm_file && + !((a->vm_flags ^ b->vm_flags) & ~(VM_READ|VM_WRITE|VM_EXEC)) && + b->vm_pgoff == a->vm_pgoff + ((b->vm_start - a->vm_start) >> PAGE_SHIFT); +} + +/* + * Do some basic sanity checking to see if we can re-use the anon_vma + * from 'old'. The 'a'/'b' vma's are in VM order - one of them will be + * the same as 'old', the other will be the new one that is trying + * to share the anon_vma. + * + * NOTE! This runs with mm_sem held for reading, so it is possible that + * the anon_vma of 'old' is concurrently in the process of being set up + * by another page fault trying to merge _that_. But that's ok: if it + * is being set up, that automatically means that it will be a singleton + * acceptable for merging, so we can do all of this optimistically. But + * we do that ACCESS_ONCE() to make sure that we never re-load the pointer. + * + * IOW: that the "list_is_singular()" test on the anon_vma_chain only + * matters for the 'stable anon_vma' case (ie the thing we want to avoid + * is to return an anon_vma that is "complex" due to having gone through + * a fork). + * + * We also make sure that the two vma's are compatible (adjacent, + * and with the same memory policies). That's all stable, even with just + * a read lock on the mm_sem. + */ +static struct anon_vma *reusable_anon_vma(struct vm_area_struct *old, struct vm_area_struct *a, struct vm_area_struct *b) +{ + if (anon_vma_compatible(a, b)) { + struct anon_vma *anon_vma = ACCESS_ONCE(old->anon_vma); + + if (anon_vma && list_is_singular(&old->anon_vma_chain)) + return anon_vma; + } + return NULL; +} + +/* + * find_mergeable_anon_vma is used by anon_vma_prepare, to check + * neighbouring vmas for a suitable anon_vma, before it goes off + * to allocate a new anon_vma. It checks because a repetitive + * sequence of mprotects and faults may otherwise lead to distinct + * anon_vmas being allocated, preventing vma merge in subsequent + * mprotect. + */ +struct anon_vma *find_mergeable_anon_vma(struct vm_area_struct *vma) +{ + struct anon_vma *anon_vma; + struct vm_area_struct *near; + + near = vma->vm_next; + if (!near) + goto try_prev; + + anon_vma = reusable_anon_vma(near, vma, near); + if (anon_vma) + return anon_vma; +try_prev: + near = vma->vm_prev; + if (!near) + goto none; + + anon_vma = reusable_anon_vma(near, near, vma); + if (anon_vma) + return anon_vma; +none: + /* + * There's no absolute need to look only at touching neighbours: + * we could search further afield for "compatible" anon_vmas. + * But it would probably just be a waste of time searching, + * or lead to too many vmas hanging off the same anon_vma. + * We're trying to allow mprotect remerging later on, + * not trying to minimize memory used for anon_vmas. + */ + return NULL; +} + +#ifdef CONFIG_PROC_FS +void vm_stat_account(struct mm_struct *mm, unsigned long flags, + struct file *file, long pages) +{ + const unsigned long stack_flags + = VM_STACK_FLAGS & (VM_GROWSUP|VM_GROWSDOWN); + + mm->total_vm += pages; + + if (file) { + mm->shared_vm += pages; + if ((flags & (VM_EXEC|VM_WRITE)) == VM_EXEC) + mm->exec_vm += pages; + } else if (flags & stack_flags) + mm->stack_vm += pages; +} +#endif /* CONFIG_PROC_FS */ + +/* + * If a hint addr is less than mmap_min_addr change hint to be as + * low as possible but still greater than mmap_min_addr + */ +static inline unsigned long round_hint_to_min(unsigned long hint) +{ + hint &= PAGE_MASK; + if (((void *)hint != NULL) && + (hint < mmap_min_addr)) + return PAGE_ALIGN(mmap_min_addr); + return hint; +} + +/* + * The caller must hold down_write(¤t->mm->mmap_sem). + */ + +unsigned long do_mmap_pgoff(struct file *file, unsigned long addr, + unsigned long len, unsigned long prot, + unsigned long flags, unsigned long pgoff, + unsigned long *populate) +{ + struct mm_struct * mm = current->mm; + struct inode *inode; + vm_flags_t vm_flags; + + *populate = 0; + + /* + * Does the application expect PROT_READ to imply PROT_EXEC? + * + * (the exception is when the underlying filesystem is noexec + * mounted, in which case we dont add PROT_EXEC.) + */ + if ((prot & PROT_READ) && (current->personality & READ_IMPLIES_EXEC)) + if (!(file && (file->f_path.mnt->mnt_flags & MNT_NOEXEC))) + prot |= PROT_EXEC; + + if (!len) + return -EINVAL; + + if (!(flags & MAP_FIXED)) + addr = round_hint_to_min(addr); + + /* Careful about overflows.. */ + len = PAGE_ALIGN(len); + if (!len) + return -ENOMEM; + + /* offset overflow? */ + if ((pgoff + (len >> PAGE_SHIFT)) < pgoff) + return -EOVERFLOW; + + /* Too many mappings? */ + if (mm->map_count > sysctl_max_map_count) + return -ENOMEM; + + /* Obtain the address to map to. we verify (or select) it and ensure + * that it represents a valid section of the address space. + */ + addr = get_unmapped_area(file, addr, len, pgoff, flags); + if (addr & ~PAGE_MASK) + return addr; + + /* Do simple checking here so the lower-level routines won't have + * to. we assume access permissions have been handled by the open + * of the memory object, so we don't do any here. + */ + vm_flags = calc_vm_prot_bits(prot) | calc_vm_flag_bits(flags) | + mm->def_flags | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC; + + if (flags & MAP_LOCKED) + if (!can_do_mlock()) + return -EPERM; + + /* mlock MCL_FUTURE? */ + if (vm_flags & VM_LOCKED) { + unsigned long locked, lock_limit; + locked = len >> PAGE_SHIFT; + locked += mm->locked_vm; + lock_limit = rlimit(RLIMIT_MEMLOCK); + lock_limit >>= PAGE_SHIFT; + if (locked > lock_limit && !capable(CAP_IPC_LOCK)) + return -EAGAIN; + } + + inode = file ? file_inode(file) : NULL; + + if (file) { + switch (flags & MAP_TYPE) { + case MAP_SHARED: + if ((prot&PROT_WRITE) && !(file->f_mode&FMODE_WRITE)) + return -EACCES; + + /* + * Make sure we don't allow writing to an append-only + * file.. + */ + if (IS_APPEND(inode) && (file->f_mode & FMODE_WRITE)) + return -EACCES; + + /* + * Make sure there are no mandatory locks on the file. + */ + if (locks_verify_locked(inode)) + return -EAGAIN; + + vm_flags |= VM_SHARED | VM_MAYSHARE; + if (!(file->f_mode & FMODE_WRITE)) + vm_flags &= ~(VM_MAYWRITE | VM_SHARED); + + /* fall through */ + case MAP_PRIVATE: + if (!(file->f_mode & FMODE_READ)) + return -EACCES; + if (file->f_path.mnt->mnt_flags & MNT_NOEXEC) { + if (vm_flags & VM_EXEC) + return -EPERM; + vm_flags &= ~VM_MAYEXEC; + } + + if (!file->f_op || !file->f_op->mmap) + return -ENODEV; + break; + + default: + return -EINVAL; + } + } else { + switch (flags & MAP_TYPE) { + case MAP_SHARED: + /* + * Ignore pgoff. + */ + pgoff = 0; + vm_flags |= VM_SHARED | VM_MAYSHARE; + break; + case MAP_PRIVATE: + /* + * Set pgoff according to addr for anon_vma. + */ + pgoff = addr >> PAGE_SHIFT; + break; + default: + return -EINVAL; + } + } + + /* + * Set 'VM_NORESERVE' if we should not account for the + * memory use of this mapping. + */ + if (flags & MAP_NORESERVE) { + /* We honor MAP_NORESERVE if allowed to overcommit */ + if (sysctl_overcommit_memory != OVERCOMMIT_NEVER) + vm_flags |= VM_NORESERVE; + + /* hugetlb applies strict overcommit unless MAP_NORESERVE */ + if (file && is_file_hugepages(file)) + vm_flags |= VM_NORESERVE; + } + + addr = mmap_region(file, addr, len, vm_flags, pgoff); + if (!IS_ERR_VALUE(addr) && + ((vm_flags & VM_LOCKED) || + (flags & (MAP_POPULATE | MAP_NONBLOCK)) == MAP_POPULATE)) + *populate = len; + return addr; +} + +SYSCALL_DEFINE6(mmap_pgoff, unsigned long, addr, unsigned long, len, + unsigned long, prot, unsigned long, flags, + unsigned long, fd, unsigned long, pgoff) +{ + struct file *file = NULL; + unsigned long retval = -EBADF; + + if (!(flags & MAP_ANONYMOUS)) { + audit_mmap_fd(fd, flags); + if (unlikely(flags & MAP_HUGETLB)) + return -EINVAL; + file = fget(fd); + if (!file) + goto out; + if (is_file_hugepages(file)) + len = ALIGN(len, huge_page_size(hstate_file(file))); + } else if (flags & MAP_HUGETLB) { + struct user_struct *user = NULL; + struct hstate *hs = hstate_sizelog((flags >> MAP_HUGE_SHIFT) & + SHM_HUGE_MASK); + + if (!hs) + return -EINVAL; + + len = ALIGN(len, huge_page_size(hs)); + /* + * VM_NORESERVE is used because the reservations will be + * taken when vm_ops->mmap() is called + * A dummy user value is used because we are not locking + * memory so no accounting is necessary + */ + file = hugetlb_file_setup(HUGETLB_ANON_FILE, len, + VM_NORESERVE, + &user, HUGETLB_ANONHUGE_INODE, + (flags >> MAP_HUGE_SHIFT) & MAP_HUGE_MASK); + if (IS_ERR(file)) + return PTR_ERR(file); + } + + flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE); + + retval = vm_mmap_pgoff(file, addr, len, prot, flags, pgoff); + if (file) + fput(file); +out: + return retval; +} + +#ifdef __ARCH_WANT_SYS_OLD_MMAP +struct mmap_arg_struct { + unsigned long addr; + unsigned long len; + unsigned long prot; + unsigned long flags; + unsigned long fd; + unsigned long offset; +}; + +SYSCALL_DEFINE1(old_mmap, struct mmap_arg_struct __user *, arg) +{ + struct mmap_arg_struct a; + + if (copy_from_user(&a, arg, sizeof(a))) + return -EFAULT; + if (a.offset & ~PAGE_MASK) + return -EINVAL; + + return sys_mmap_pgoff(a.addr, a.len, a.prot, a.flags, a.fd, + a.offset >> PAGE_SHIFT); +} +#endif /* __ARCH_WANT_SYS_OLD_MMAP */ + +/* + * Some shared mappigns will want the pages marked read-only + * to track write events. If so, we'll downgrade vm_page_prot + * to the private version (using protection_map[] without the + * VM_SHARED bit). + */ +int vma_wants_writenotify(struct vm_area_struct *vma) +{ + vm_flags_t vm_flags = vma->vm_flags; + + /* If it was private or non-writable, the write bit is already clear */ + if ((vm_flags & (VM_WRITE|VM_SHARED)) != ((VM_WRITE|VM_SHARED))) + return 0; + + /* The backer wishes to know when pages are first written to? */ + if (vma->vm_ops && vma->vm_ops->page_mkwrite) + return 1; + + /* The open routine did something to the protections already? */ + if (pgprot_val(vma->vm_page_prot) != + pgprot_val(vm_get_page_prot(vm_flags))) + return 0; + + /* Specialty mapping? */ + if (vm_flags & VM_PFNMAP) + return 0; + + /* Can the mapping track the dirty pages? */ + return vma->vm_file && vma->vm_file->f_mapping && + mapping_cap_account_dirty(vma->vm_file->f_mapping); +} + +/* + * We account for memory if it's a private writeable mapping, + * not hugepages and VM_NORESERVE wasn't set. + */ +static inline int accountable_mapping(struct file *file, vm_flags_t vm_flags) +{ + /* + * hugetlb has its own accounting separate from the core VM + * VM_HUGETLB may not be set yet so we cannot check for that flag. + */ + if (file && is_file_hugepages(file)) + return 0; + + return (vm_flags & (VM_NORESERVE | VM_SHARED | VM_WRITE)) == VM_WRITE; +} + +unsigned long mmap_region(struct file *file, unsigned long addr, + unsigned long len, vm_flags_t vm_flags, unsigned long pgoff) +{ + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma, *prev; + int correct_wcount = 0; + int error; + struct rb_node **rb_link, *rb_parent; + unsigned long charged = 0; + struct inode *inode = file ? file_inode(file) : NULL; + + /* Check against address space limit. */ + if (!may_expand_vm(mm, len >> PAGE_SHIFT)) { + unsigned long nr_pages; + + /* + * MAP_FIXED may remove pages of mappings that intersects with + * requested mapping. Account for the pages it would unmap. + */ + if (!(vm_flags & MAP_FIXED)) + return -ENOMEM; + + nr_pages = count_vma_pages_range(mm, addr, addr + len); + + if (!may_expand_vm(mm, (len >> PAGE_SHIFT) - nr_pages)) + return -ENOMEM; + } + + /* Clear old maps */ + error = -ENOMEM; +munmap_back: + if (find_vma_links(mm, addr, addr + len, &prev, &rb_link, &rb_parent)) { + if (do_munmap(mm, addr, len)) + return -ENOMEM; + goto munmap_back; + } + + /* + * Private writable mapping: check memory availability + */ + if (accountable_mapping(file, vm_flags)) { + charged = len >> PAGE_SHIFT; + if (security_vm_enough_memory_mm(mm, charged)) + return -ENOMEM; + vm_flags |= VM_ACCOUNT; + } + + /* + * Can we just expand an old mapping? + */ + vma = vma_merge(mm, prev, addr, addr + len, vm_flags, NULL, file, pgoff, + NULL, NULL); + if (vma) + goto out; + + /* + * Determine the object being mapped and call the appropriate + * specific mapper. the address has already been validated, but + * not unmapped, but the maps are removed from the list. + */ + vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); + if (!vma) { + error = -ENOMEM; + goto unacct_error; + } + + vma->vm_mm = mm; + vma->vm_start = addr; + vma->vm_end = addr + len; + vma->vm_flags = vm_flags; + vma->vm_page_prot = vm_get_page_prot(vm_flags); + vma->vm_pgoff = pgoff; + INIT_LIST_HEAD(&vma->anon_vma_chain); + + error = -EINVAL; /* when rejecting VM_GROWSDOWN|VM_GROWSUP */ + + if (file) { + if (vm_flags & (VM_GROWSDOWN|VM_GROWSUP)) + goto free_vma; + if (vm_flags & VM_DENYWRITE) { + error = deny_write_access(file); + if (error) + goto free_vma; + correct_wcount = 1; + } + vma->vm_file = get_file(file); + error = file->f_op->mmap(file, vma); + if (error) + goto unmap_and_free_vma; + + /* Can addr have changed?? + * + * Answer: Yes, several device drivers can do it in their + * f_op->mmap method. -DaveM + * Bug: If addr is changed, prev, rb_link, rb_parent should + * be updated for vma_link() + */ + WARN_ON_ONCE(addr != vma->vm_start); + + addr = vma->vm_start; + pgoff = vma->vm_pgoff; + vm_flags = vma->vm_flags; + } else if (vm_flags & VM_SHARED) { + if (unlikely(vm_flags & (VM_GROWSDOWN|VM_GROWSUP))) + goto free_vma; + error = shmem_zero_setup(vma); + if (error) + goto free_vma; + } + + if (vma_wants_writenotify(vma)) { + pgprot_t pprot = vma->vm_page_prot; + + /* Can vma->vm_page_prot have changed?? + * + * Answer: Yes, drivers may have changed it in their + * f_op->mmap method. + * + * Ensures that vmas marked as uncached stay that way. + */ + vma->vm_page_prot = vm_get_page_prot(vm_flags & ~VM_SHARED); + if (pgprot_val(pprot) == pgprot_val(pgprot_noncached(pprot))) + vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); + } + + vma_link(mm, vma, prev, rb_link, rb_parent); + file = vma->vm_file; + + /* Once vma denies write, undo our temporary denial count */ + if (correct_wcount) + atomic_inc(&inode->i_writecount); +out: + perf_event_mmap(vma); + quadd_event_mmap(vma); + + vm_stat_account(mm, vm_flags, file, len >> PAGE_SHIFT); + if (vm_flags & VM_LOCKED) { + if (!((vm_flags & VM_SPECIAL) || is_vm_hugetlb_page(vma) || + vma == get_gate_vma(current->mm))) + mm->locked_vm += (len >> PAGE_SHIFT); + else + vma->vm_flags &= ~VM_LOCKED; + } + + if (file) + uprobe_mmap(vma); + + return addr; + +unmap_and_free_vma: + if (correct_wcount) + atomic_inc(&inode->i_writecount); + vma->vm_file = NULL; + fput(file); + + /* Undo any partial mapping done by a device driver. */ + unmap_region(mm, vma, prev, vma->vm_start, vma->vm_end); + charged = 0; +free_vma: + kmem_cache_free(vm_area_cachep, vma); +unacct_error: + if (charged) + vm_unacct_memory(charged); + return error; +} + +unsigned long unmapped_area(struct vm_unmapped_area_info *info) +{ + /* + * We implement the search by looking for an rbtree node that + * immediately follows a suitable gap. That is, + * - gap_start = vma->vm_prev->vm_end <= info->high_limit - length; + * - gap_end = vma->vm_start >= info->low_limit + length; + * - gap_end - gap_start >= length + */ + + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma; + unsigned long length, low_limit, high_limit, gap_start, gap_end; + + /* Adjust search length to account for worst case alignment overhead */ + length = info->length + info->align_mask; + if (length < info->length) + return -ENOMEM; + + /* Adjust search limits by the desired length */ + if (info->high_limit < length) + return -ENOMEM; + high_limit = info->high_limit - length; + + if (info->low_limit > high_limit) + return -ENOMEM; + low_limit = info->low_limit + length; + + /* Check if rbtree root looks promising */ + if (RB_EMPTY_ROOT(&mm->mm_rb)) + goto check_highest; + vma = rb_entry(mm->mm_rb.rb_node, struct vm_area_struct, vm_rb); + if (vma->rb_subtree_gap < length) + goto check_highest; + + while (true) { + /* Visit left subtree if it looks promising */ + gap_end = vma->vm_start; + if (gap_end >= low_limit && vma->vm_rb.rb_left) { + struct vm_area_struct *left = + rb_entry(vma->vm_rb.rb_left, + struct vm_area_struct, vm_rb); + if (left->rb_subtree_gap >= length) { + vma = left; + continue; + } + } + + gap_start = vma->vm_prev ? vma->vm_prev->vm_end : 0; +check_current: + /* Check if current node has a suitable gap */ + if (gap_start > high_limit) + return -ENOMEM; + if (gap_end >= low_limit && gap_end - gap_start >= length) + goto found; + + /* Visit right subtree if it looks promising */ + if (vma->vm_rb.rb_right) { + struct vm_area_struct *right = + rb_entry(vma->vm_rb.rb_right, + struct vm_area_struct, vm_rb); + if (right->rb_subtree_gap >= length) { + vma = right; + continue; + } + } + + /* Go back up the rbtree to find next candidate node */ + while (true) { + struct rb_node *prev = &vma->vm_rb; + if (!rb_parent(prev)) + goto check_highest; + vma = rb_entry(rb_parent(prev), + struct vm_area_struct, vm_rb); + if (prev == vma->vm_rb.rb_left) { + gap_start = vma->vm_prev->vm_end; + gap_end = vma->vm_start; + goto check_current; + } + } + } + +check_highest: + /* Check highest gap, which does not precede any rbtree node */ + gap_start = mm->highest_vm_end; + gap_end = ULONG_MAX; /* Only for VM_BUG_ON below */ + if (gap_start > high_limit) + return -ENOMEM; + +found: + /* We found a suitable gap. Clip it with the original low_limit. */ + if (gap_start < info->low_limit) + gap_start = info->low_limit; + + /* Adjust gap address to the desired alignment */ + gap_start += (info->align_offset - gap_start) & info->align_mask; + + VM_BUG_ON(gap_start + info->length > info->high_limit); + VM_BUG_ON(gap_start + info->length > gap_end); + return gap_start; +} + +unsigned long unmapped_area_topdown(struct vm_unmapped_area_info *info) +{ + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma; + unsigned long length, low_limit, high_limit, gap_start, gap_end; + + /* Adjust search length to account for worst case alignment overhead */ + length = info->length + info->align_mask; + if (length < info->length) + return -ENOMEM; + + /* + * Adjust search limits by the desired length. + * See implementation comment at top of unmapped_area(). + */ + gap_end = info->high_limit; + if (gap_end < length) + return -ENOMEM; + high_limit = gap_end - length; + + if (info->low_limit > high_limit) + return -ENOMEM; + low_limit = info->low_limit + length; + + /* Check highest gap, which does not precede any rbtree node */ + gap_start = mm->highest_vm_end; + if (gap_start <= high_limit) + goto found_highest; + + /* Check if rbtree root looks promising */ + if (RB_EMPTY_ROOT(&mm->mm_rb)) + return -ENOMEM; + vma = rb_entry(mm->mm_rb.rb_node, struct vm_area_struct, vm_rb); + if (vma->rb_subtree_gap < length) + return -ENOMEM; + + while (true) { + /* Visit right subtree if it looks promising */ + gap_start = vma->vm_prev ? vma->vm_prev->vm_end : 0; + if (gap_start <= high_limit && vma->vm_rb.rb_right) { + struct vm_area_struct *right = + rb_entry(vma->vm_rb.rb_right, + struct vm_area_struct, vm_rb); + if (right->rb_subtree_gap >= length) { + vma = right; + continue; + } + } + +check_current: + /* Check if current node has a suitable gap */ + gap_end = vma->vm_start; + if (gap_end < low_limit) + return -ENOMEM; + if (gap_start <= high_limit && gap_end - gap_start >= length) + goto found; + + /* Visit left subtree if it looks promising */ + if (vma->vm_rb.rb_left) { + struct vm_area_struct *left = + rb_entry(vma->vm_rb.rb_left, + struct vm_area_struct, vm_rb); + if (left->rb_subtree_gap >= length) { + vma = left; + continue; + } + } + + /* Go back up the rbtree to find next candidate node */ + while (true) { + struct rb_node *prev = &vma->vm_rb; + if (!rb_parent(prev)) + return -ENOMEM; + vma = rb_entry(rb_parent(prev), + struct vm_area_struct, vm_rb); + if (prev == vma->vm_rb.rb_right) { + gap_start = vma->vm_prev ? + vma->vm_prev->vm_end : 0; + goto check_current; + } + } + } + +found: + /* We found a suitable gap. Clip it with the original high_limit. */ + if (gap_end > info->high_limit) + gap_end = info->high_limit; + +found_highest: + /* Compute highest gap address at the desired alignment */ + gap_end -= info->length; + gap_end -= (gap_end - info->align_offset) & info->align_mask; + + VM_BUG_ON(gap_end < info->low_limit); + VM_BUG_ON(gap_end < gap_start); + return gap_end; +} + +/* Get an address range which is currently unmapped. + * For shmat() with addr=0. + * + * Ugly calling convention alert: + * Return value with the low bits set means error value, + * ie + * if (ret & ~PAGE_MASK) + * error = ret; + * + * This function "knows" that -ENOMEM has the bits set. + */ +#ifndef HAVE_ARCH_UNMAPPED_AREA +unsigned long +arch_get_unmapped_area(struct file *filp, unsigned long addr, + unsigned long len, unsigned long pgoff, unsigned long flags) +{ + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma; + struct vm_unmapped_area_info info; + + if (len > TASK_SIZE - mmap_min_addr) + return -ENOMEM; + + if (flags & MAP_FIXED) + return addr; + + if (addr) { + addr = PAGE_ALIGN(addr); + vma = find_vma(mm, addr); + if (TASK_SIZE - len >= addr && addr >= mmap_min_addr && + (!vma || addr + len <= vma->vm_start)) + return addr; + } + + info.flags = 0; + info.length = len; + info.low_limit = mm->mmap_base; + info.high_limit = TASK_SIZE; + info.align_mask = 0; + return vm_unmapped_area(&info); +} +#endif + +void arch_unmap_area(struct mm_struct *mm, unsigned long addr) +{ + /* + * Is this a new hole at the lowest possible address? + */ + if (addr >= TASK_UNMAPPED_BASE && addr < mm->free_area_cache) + mm->free_area_cache = addr; +} + +/* + * This mmap-allocator allocates new areas top-down from below the + * stack's low limit (the base): + */ +#ifndef HAVE_ARCH_UNMAPPED_AREA_TOPDOWN +unsigned long +arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, + const unsigned long len, const unsigned long pgoff, + const unsigned long flags) +{ + struct vm_area_struct *vma; + struct mm_struct *mm = current->mm; + unsigned long addr = addr0; + struct vm_unmapped_area_info info; + + /* requested length too big for entire address space */ + if (len > TASK_SIZE - mmap_min_addr) + return -ENOMEM; + + if (flags & MAP_FIXED) + return addr; + + /* requesting a specific address */ + if (addr) { + addr = PAGE_ALIGN(addr); + vma = find_vma(mm, addr); + if (TASK_SIZE - len >= addr && addr >= mmap_min_addr && + (!vma || addr + len <= vma->vm_start)) + return addr; + } + + info.flags = VM_UNMAPPED_AREA_TOPDOWN; + info.length = len; + info.low_limit = max(PAGE_SIZE, mmap_min_addr); + info.high_limit = mm->mmap_base; + info.align_mask = 0; + addr = vm_unmapped_area(&info); + + /* + * A failed mmap() very likely causes application failure, + * so fall back to the bottom-up function here. This scenario + * can happen with large stack limits and large mmap() + * allocations. + */ + if (addr & ~PAGE_MASK) { + VM_BUG_ON(addr != -ENOMEM); + info.flags = 0; + info.low_limit = TASK_UNMAPPED_BASE; + info.high_limit = TASK_SIZE; + addr = vm_unmapped_area(&info); + } + + return addr; +} +#endif + +void arch_unmap_area_topdown(struct mm_struct *mm, unsigned long addr) +{ + /* + * Is this a new hole at the highest possible address? + */ + if (addr > mm->free_area_cache) + mm->free_area_cache = addr; + + /* dont allow allocations above current base */ + if (mm->free_area_cache > mm->mmap_base) + mm->free_area_cache = mm->mmap_base; +} + +unsigned long +get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, + unsigned long pgoff, unsigned long flags) +{ + unsigned long (*get_area)(struct file *, unsigned long, + unsigned long, unsigned long, unsigned long); + + unsigned long error = arch_mmap_check(addr, len, flags); + if (error) + return error; + + /* Careful about overflows.. */ + if (len > TASK_SIZE) + return -ENOMEM; + + get_area = current->mm->get_unmapped_area; + if (file && file->f_op && file->f_op->get_unmapped_area) + get_area = file->f_op->get_unmapped_area; + addr = get_area(file, addr, len, pgoff, flags); + if (IS_ERR_VALUE(addr)) + return addr; + + if (addr > TASK_SIZE - len) + return -ENOMEM; + if (addr & ~PAGE_MASK) + return -EINVAL; + + addr = arch_rebalance_pgtables(addr, len); + error = security_mmap_addr(addr); + return error ? error : addr; +} + +EXPORT_SYMBOL(get_unmapped_area); + +/* Look up the first VMA which satisfies addr < vm_end, NULL if none. */ +struct vm_area_struct *find_vma(struct mm_struct *mm, unsigned long addr) +{ + struct vm_area_struct *vma = NULL; + + /* Check the cache first. */ + /* (Cache hit rate is typically around 35%.) */ + vma = ACCESS_ONCE(mm->mmap_cache); + if (!(vma && vma->vm_end > addr && vma->vm_start <= addr)) { + struct rb_node *rb_node; + + rb_node = mm->mm_rb.rb_node; + vma = NULL; + + while (rb_node) { + struct vm_area_struct *vma_tmp; + + vma_tmp = rb_entry(rb_node, + struct vm_area_struct, vm_rb); + + if (vma_tmp->vm_end > addr) { + vma = vma_tmp; + if (vma_tmp->vm_start <= addr) + break; + rb_node = rb_node->rb_left; + } else + rb_node = rb_node->rb_right; + } + if (vma) + mm->mmap_cache = vma; + } + return vma; +} + +EXPORT_SYMBOL(find_vma); + +/* + * Same as find_vma, but also return a pointer to the previous VMA in *pprev. + */ +struct vm_area_struct * +find_vma_prev(struct mm_struct *mm, unsigned long addr, + struct vm_area_struct **pprev) +{ + struct vm_area_struct *vma; + + vma = find_vma(mm, addr); + if (vma) { + *pprev = vma->vm_prev; + } else { + struct rb_node *rb_node = mm->mm_rb.rb_node; + *pprev = NULL; + while (rb_node) { + *pprev = rb_entry(rb_node, struct vm_area_struct, vm_rb); + rb_node = rb_node->rb_right; + } + } + return vma; +} + +/* + * Verify that the stack growth is acceptable and + * update accounting. This is shared with both the + * grow-up and grow-down cases. + */ +static int acct_stack_growth(struct vm_area_struct *vma, unsigned long size, unsigned long grow) +{ + struct mm_struct *mm = vma->vm_mm; + struct rlimit *rlim = current->signal->rlim; + unsigned long new_start; + + /* address space limit tests */ + if (!may_expand_vm(mm, grow)) + return -ENOMEM; + + /* Stack limit test */ + if (size > ACCESS_ONCE(rlim[RLIMIT_STACK].rlim_cur)) + return -ENOMEM; + + /* mlock limit tests */ + if (vma->vm_flags & VM_LOCKED) { + unsigned long locked; + unsigned long limit; + locked = mm->locked_vm + grow; + limit = ACCESS_ONCE(rlim[RLIMIT_MEMLOCK].rlim_cur); + limit >>= PAGE_SHIFT; + if (locked > limit && !capable(CAP_IPC_LOCK)) + return -ENOMEM; + } + + /* Check to ensure the stack will not grow into a hugetlb-only region */ + new_start = (vma->vm_flags & VM_GROWSUP) ? vma->vm_start : + vma->vm_end - size; + if (is_hugepage_only_range(vma->vm_mm, new_start, size)) + return -EFAULT; + + /* + * Overcommit.. This must be the final test, as it will + * update security statistics. + */ + if (security_vm_enough_memory_mm(mm, grow)) + return -ENOMEM; + + /* Ok, everything looks good - let it rip */ + if (vma->vm_flags & VM_LOCKED) + mm->locked_vm += grow; + vm_stat_account(mm, vma->vm_flags, vma->vm_file, grow); + return 0; +} + +#if defined(CONFIG_STACK_GROWSUP) || defined(CONFIG_IA64) +/* + * PA-RISC uses this for its stack; IA64 for its Register Backing Store. + * vma is the last one with address > vma->vm_end. Have to extend vma. + */ +int expand_upwards(struct vm_area_struct *vma, unsigned long address) +{ + int error; + + if (!(vma->vm_flags & VM_GROWSUP)) + return -EFAULT; + + /* + * We must make sure the anon_vma is allocated + * so that the anon_vma locking is not a noop. + */ + if (unlikely(anon_vma_prepare(vma))) + return -ENOMEM; + vma_lock_anon_vma(vma); + + /* + * vma->vm_start/vm_end cannot change under us because the caller + * is required to hold the mmap_sem in read mode. We need the + * anon_vma lock to serialize against concurrent expand_stacks. + * Also guard against wrapping around to address 0. + */ + if (address < PAGE_ALIGN(address+4)) + address = PAGE_ALIGN(address+4); + else { + vma_unlock_anon_vma(vma); + return -ENOMEM; + } + error = 0; + + /* Somebody else might have raced and expanded it already */ + if (address > vma->vm_end) { + unsigned long size, grow; + + size = address - vma->vm_start; + grow = (address - vma->vm_end) >> PAGE_SHIFT; + + error = -ENOMEM; + if (vma->vm_pgoff + (size >> PAGE_SHIFT) >= vma->vm_pgoff) { + error = acct_stack_growth(vma, size, grow); + if (!error) { + /* + * vma_gap_update() doesn't support concurrent + * updates, but we only hold a shared mmap_sem + * lock here, so we need to protect against + * concurrent vma expansions. + * vma_lock_anon_vma() doesn't help here, as + * we don't guarantee that all growable vmas + * in a mm share the same root anon vma. + * So, we reuse mm->page_table_lock to guard + * against concurrent vma expansions. + */ + spin_lock(&vma->vm_mm->page_table_lock); + anon_vma_interval_tree_pre_update_vma(vma); + vma->vm_end = address; + anon_vma_interval_tree_post_update_vma(vma); + if (vma->vm_next) + vma_gap_update(vma->vm_next); + else + vma->vm_mm->highest_vm_end = address; + spin_unlock(&vma->vm_mm->page_table_lock); + + perf_event_mmap(vma); + quadd_event_mmap(vma); + } + } + } + vma_unlock_anon_vma(vma); + khugepaged_enter_vma_merge(vma); + validate_mm(vma->vm_mm); + return error; +} +#endif /* CONFIG_STACK_GROWSUP || CONFIG_IA64 */ + +/* + * vma is the first one with address < vma->vm_start. Have to extend vma. + */ +int expand_downwards(struct vm_area_struct *vma, + unsigned long address) +{ + int error; + + /* + * We must make sure the anon_vma is allocated + * so that the anon_vma locking is not a noop. + */ + if (unlikely(anon_vma_prepare(vma))) + return -ENOMEM; + + address &= PAGE_MASK; + error = security_mmap_addr(address); + if (error) + return error; + + vma_lock_anon_vma(vma); + + /* + * vma->vm_start/vm_end cannot change under us because the caller + * is required to hold the mmap_sem in read mode. We need the + * anon_vma lock to serialize against concurrent expand_stacks. + */ + + /* Somebody else might have raced and expanded it already */ + if (address < vma->vm_start) { + unsigned long size, grow; + + size = vma->vm_end - address; + grow = (vma->vm_start - address) >> PAGE_SHIFT; + + error = -ENOMEM; + if (grow <= vma->vm_pgoff) { + error = acct_stack_growth(vma, size, grow); + if (!error) { + /* + * vma_gap_update() doesn't support concurrent + * updates, but we only hold a shared mmap_sem + * lock here, so we need to protect against + * concurrent vma expansions. + * vma_lock_anon_vma() doesn't help here, as + * we don't guarantee that all growable vmas + * in a mm share the same root anon vma. + * So, we reuse mm->page_table_lock to guard + * against concurrent vma expansions. + */ + spin_lock(&vma->vm_mm->page_table_lock); + anon_vma_interval_tree_pre_update_vma(vma); + vma->vm_start = address; + vma->vm_pgoff -= grow; + anon_vma_interval_tree_post_update_vma(vma); + vma_gap_update(vma); + spin_unlock(&vma->vm_mm->page_table_lock); + + perf_event_mmap(vma); + quadd_event_mmap(vma); + } + } + } + vma_unlock_anon_vma(vma); + khugepaged_enter_vma_merge(vma); + validate_mm(vma->vm_mm); + return error; +} + +/* + * Note how expand_stack() refuses to expand the stack all the way to + * abut the next virtual mapping, *unless* that mapping itself is also + * a stack mapping. We want to leave room for a guard page, after all + * (the guard page itself is not added here, that is done by the + * actual page faulting logic) + * + * This matches the behavior of the guard page logic (see mm/memory.c: + * check_stack_guard_page()), which only allows the guard page to be + * removed under these circumstances. + */ +#ifdef CONFIG_STACK_GROWSUP +int expand_stack(struct vm_area_struct *vma, unsigned long address) +{ + struct vm_area_struct *next; + + address &= PAGE_MASK; + next = vma->vm_next; + if (next && next->vm_start == address + PAGE_SIZE) { + if (!(next->vm_flags & VM_GROWSUP)) + return -ENOMEM; + } + return expand_upwards(vma, address); +} + +struct vm_area_struct * +find_extend_vma(struct mm_struct *mm, unsigned long addr) +{ + struct vm_area_struct *vma, *prev; + + addr &= PAGE_MASK; + vma = find_vma_prev(mm, addr, &prev); + if (vma && (vma->vm_start <= addr)) + return vma; + if (!prev || expand_stack(prev, addr)) + return NULL; + if (prev->vm_flags & VM_LOCKED) + __mlock_vma_pages_range(prev, addr, prev->vm_end, NULL); + return prev; +} +#else +int expand_stack(struct vm_area_struct *vma, unsigned long address) +{ + struct vm_area_struct *prev; + + address &= PAGE_MASK; + prev = vma->vm_prev; + if (prev && prev->vm_end == address) { + if (!(prev->vm_flags & VM_GROWSDOWN)) + return -ENOMEM; + } + return expand_downwards(vma, address); +} + +struct vm_area_struct * +find_extend_vma(struct mm_struct * mm, unsigned long addr) +{ + struct vm_area_struct * vma; + unsigned long start; + + addr &= PAGE_MASK; + vma = find_vma(mm,addr); + if (!vma) + return NULL; + if (vma->vm_start <= addr) + return vma; + if (!(vma->vm_flags & VM_GROWSDOWN)) + return NULL; + start = vma->vm_start; + if (expand_stack(vma, addr)) + return NULL; + if (vma->vm_flags & VM_LOCKED) + __mlock_vma_pages_range(vma, addr, start, NULL); + return vma; +} +#endif + +/* + * Ok - we have the memory areas we should free on the vma list, + * so release them, and do the vma updates. + * + * Called with the mm semaphore held. + */ +static void remove_vma_list(struct mm_struct *mm, struct vm_area_struct *vma) +{ + unsigned long nr_accounted = 0; + + /* Update high watermark before we lower total_vm */ + update_hiwater_vm(mm); + do { + long nrpages = vma_pages(vma); + + if (vma->vm_flags & VM_ACCOUNT) + nr_accounted += nrpages; + vm_stat_account(mm, vma->vm_flags, vma->vm_file, -nrpages); + vma = remove_vma(vma); + } while (vma); + vm_unacct_memory(nr_accounted); + validate_mm(mm); +} + +/* + * Get rid of page table information in the indicated region. + * + * Called with the mm semaphore held. + */ +static void unmap_region(struct mm_struct *mm, + struct vm_area_struct *vma, struct vm_area_struct *prev, + unsigned long start, unsigned long end) +{ + struct vm_area_struct *next = prev? prev->vm_next: mm->mmap; + struct mmu_gather tlb; + + lru_add_drain(); + tlb_gather_mmu(&tlb, mm, start, end); + update_hiwater_rss(mm); + unmap_vmas(&tlb, vma, start, end); + free_pgtables(&tlb, vma, prev ? prev->vm_end : FIRST_USER_ADDRESS, + next ? next->vm_start : USER_PGTABLES_CEILING); + tlb_finish_mmu(&tlb, start, end); +} + +/* + * Create a list of vma's touched by the unmap, removing them from the mm's + * vma list as we go.. + */ +static void +detach_vmas_to_be_unmapped(struct mm_struct *mm, struct vm_area_struct *vma, + struct vm_area_struct *prev, unsigned long end) +{ + struct vm_area_struct **insertion_point; + struct vm_area_struct *tail_vma = NULL; + unsigned long addr; + + insertion_point = (prev ? &prev->vm_next : &mm->mmap); + vma->vm_prev = NULL; + do { + vma_rb_erase(vma, &mm->mm_rb); + mm->map_count--; + tail_vma = vma; + vma = vma->vm_next; + } while (vma && vma->vm_start < end); + *insertion_point = vma; + if (vma) { + vma->vm_prev = prev; + vma_gap_update(vma); + } else + mm->highest_vm_end = prev ? prev->vm_end : 0; + tail_vma->vm_next = NULL; + if (mm->unmap_area == arch_unmap_area) + addr = prev ? prev->vm_end : mm->mmap_base; + else + addr = vma ? vma->vm_start : mm->mmap_base; + mm->unmap_area(mm, addr); + mm->mmap_cache = NULL; /* Kill the cache. */ +} + +/* + * __split_vma() bypasses sysctl_max_map_count checking. We use this on the + * munmap path where it doesn't make sense to fail. + */ +static int __split_vma(struct mm_struct * mm, struct vm_area_struct * vma, + unsigned long addr, int new_below) +{ + struct mempolicy *pol; + struct vm_area_struct *new; + int err = -ENOMEM; + + if (is_vm_hugetlb_page(vma) && (addr & + ~(huge_page_mask(hstate_vma(vma))))) + return -EINVAL; + + new = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); + if (!new) + goto out_err; + + /* most fields are the same, copy all, and then fixup */ + *new = *vma; + + INIT_LIST_HEAD(&new->anon_vma_chain); + + if (new_below) + new->vm_end = addr; + else { + new->vm_start = addr; + new->vm_pgoff += ((addr - vma->vm_start) >> PAGE_SHIFT); + } + + pol = mpol_dup(vma_policy(vma)); + if (IS_ERR(pol)) { + err = PTR_ERR(pol); + goto out_free_vma; + } + vma_set_policy(new, pol); + + if (anon_vma_clone(new, vma)) + goto out_free_mpol; + + if (new->vm_file) + get_file(new->vm_file); + + if (new->vm_ops && new->vm_ops->open) + new->vm_ops->open(new); + + if (new_below) + err = vma_adjust(vma, addr, vma->vm_end, vma->vm_pgoff + + ((addr - new->vm_start) >> PAGE_SHIFT), new); + else + err = vma_adjust(vma, vma->vm_start, addr, vma->vm_pgoff, new); + + /* Success. */ + if (!err) + return 0; + + /* Clean everything up if vma_adjust failed. */ + if (new->vm_ops && new->vm_ops->close) + new->vm_ops->close(new); + if (new->vm_file) + fput(new->vm_file); + unlink_anon_vmas(new); + out_free_mpol: + mpol_put(pol); + out_free_vma: + kmem_cache_free(vm_area_cachep, new); + out_err: + return err; +} + +/* + * Split a vma into two pieces at address 'addr', a new vma is allocated + * either for the first part or the tail. + */ +int split_vma(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long addr, int new_below) +{ + if (mm->map_count >= sysctl_max_map_count) + return -ENOMEM; + + return __split_vma(mm, vma, addr, new_below); +} + +/* Munmap is split into 2 main parts -- this part which finds + * what needs doing, and the areas themselves, which do the + * work. This now handles partial unmappings. + * Jeremy Fitzhardinge + */ +int do_munmap(struct mm_struct *mm, unsigned long start, size_t len) +{ + unsigned long end; + struct vm_area_struct *vma, *prev, *last; + + if ((start & ~PAGE_MASK) || start > TASK_SIZE || len > TASK_SIZE-start) + return -EINVAL; + + if ((len = PAGE_ALIGN(len)) == 0) + return -EINVAL; + + /* Find the first overlapping VMA */ + vma = find_vma(mm, start); + if (!vma) + return 0; + prev = vma->vm_prev; + /* we have start < vma->vm_end */ + + /* if it doesn't overlap, we have nothing.. */ + end = start + len; + if (vma->vm_start >= end) + return 0; + + /* + * If we need to split any vma, do it now to save pain later. + * + * Note: mremap's move_vma VM_ACCOUNT handling assumes a partially + * unmapped vm_area_struct will remain in use: so lower split_vma + * places tmp vma above, and higher split_vma places tmp vma below. + */ + if (start > vma->vm_start) { + int error; + + /* + * Make sure that map_count on return from munmap() will + * not exceed its limit; but let map_count go just above + * its limit temporarily, to help free resources as expected. + */ + if (end < vma->vm_end && mm->map_count >= sysctl_max_map_count) + return -ENOMEM; + + error = __split_vma(mm, vma, start, 0); + if (error) + return error; + prev = vma; + } + + /* Does it split the last one? */ + last = find_vma(mm, end); + if (last && end > last->vm_start) { + int error = __split_vma(mm, last, end, 1); + if (error) + return error; + } + vma = prev? prev->vm_next: mm->mmap; + + /* + * unlock any mlock()ed ranges before detaching vmas + */ + if (mm->locked_vm) { + struct vm_area_struct *tmp = vma; + while (tmp && tmp->vm_start < end) { + if (tmp->vm_flags & VM_LOCKED) { + mm->locked_vm -= vma_pages(tmp); + munlock_vma_pages_all(tmp); + } + tmp = tmp->vm_next; + } + } + + /* + * Remove the vma's, and unmap the actual pages + */ + detach_vmas_to_be_unmapped(mm, vma, prev, end); + unmap_region(mm, vma, prev, start, end); + + /* Fix up all other VM information */ + remove_vma_list(mm, vma); + + return 0; +} + +int vm_munmap(unsigned long start, size_t len) +{ + int ret; + struct mm_struct *mm = current->mm; + + down_write(&mm->mmap_sem); + ret = do_munmap(mm, start, len); + up_write(&mm->mmap_sem); + return ret; +} +EXPORT_SYMBOL(vm_munmap); + +SYSCALL_DEFINE2(munmap, unsigned long, addr, size_t, len) +{ + profile_munmap(addr); + return vm_munmap(addr, len); +} + +static inline void verify_mm_writelocked(struct mm_struct *mm) +{ +#ifdef CONFIG_DEBUG_VM + if (unlikely(down_read_trylock(&mm->mmap_sem))) { + WARN_ON(1); + up_read(&mm->mmap_sem); + } +#endif +} + +/* + * this is really a simplified "do_mmap". it only handles + * anonymous maps. eventually we may be able to do some + * brk-specific accounting here. + */ +static unsigned long do_brk(unsigned long addr, unsigned long len) +{ + struct mm_struct * mm = current->mm; + struct vm_area_struct * vma, * prev; + unsigned long flags; + struct rb_node ** rb_link, * rb_parent; + pgoff_t pgoff = addr >> PAGE_SHIFT; + int error; + + len = PAGE_ALIGN(len); + if (!len) + return addr; + + flags = VM_DATA_DEFAULT_FLAGS | VM_ACCOUNT | mm->def_flags; + + error = get_unmapped_area(NULL, addr, len, 0, MAP_FIXED); + if (error & ~PAGE_MASK) + return error; + + /* + * mlock MCL_FUTURE? + */ + if (mm->def_flags & VM_LOCKED) { + unsigned long locked, lock_limit; + locked = len >> PAGE_SHIFT; + locked += mm->locked_vm; + lock_limit = rlimit(RLIMIT_MEMLOCK); + lock_limit >>= PAGE_SHIFT; + if (locked > lock_limit && !capable(CAP_IPC_LOCK)) + return -EAGAIN; + } + + /* + * mm->mmap_sem is required to protect against another thread + * changing the mappings in case we sleep. + */ + verify_mm_writelocked(mm); + + /* + * Clear old maps. this also does some error checking for us + */ + munmap_back: + if (find_vma_links(mm, addr, addr + len, &prev, &rb_link, &rb_parent)) { + if (do_munmap(mm, addr, len)) + return -ENOMEM; + goto munmap_back; + } + + /* Check against address space limits *after* clearing old maps... */ + if (!may_expand_vm(mm, len >> PAGE_SHIFT)) + return -ENOMEM; + + if (mm->map_count > sysctl_max_map_count) + return -ENOMEM; + + if (security_vm_enough_memory_mm(mm, len >> PAGE_SHIFT)) + return -ENOMEM; + + /* Can we just expand an old private anonymous mapping? */ + vma = vma_merge(mm, prev, addr, addr + len, flags, + NULL, NULL, pgoff, NULL, NULL); + if (vma) + goto out; + + /* + * create a vma struct for an anonymous mapping + */ + vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); + if (!vma) { + vm_unacct_memory(len >> PAGE_SHIFT); + return -ENOMEM; + } + + INIT_LIST_HEAD(&vma->anon_vma_chain); + vma->vm_mm = mm; + vma->vm_start = addr; + vma->vm_end = addr + len; + vma->vm_pgoff = pgoff; + vma->vm_flags = flags; + vma->vm_page_prot = vm_get_page_prot(flags); + vma_link(mm, vma, prev, rb_link, rb_parent); +out: + perf_event_mmap(vma); + quadd_event_mmap(vma); + + mm->total_vm += len >> PAGE_SHIFT; + if (flags & VM_LOCKED) + mm->locked_vm += (len >> PAGE_SHIFT); + return addr; +} + +unsigned long vm_brk(unsigned long addr, unsigned long len) +{ + struct mm_struct *mm = current->mm; + unsigned long ret; + bool populate; + + down_write(&mm->mmap_sem); + ret = do_brk(addr, len); + populate = ((mm->def_flags & VM_LOCKED) != 0); + up_write(&mm->mmap_sem); + if (populate) + mm_populate(addr, len); + return ret; +} +EXPORT_SYMBOL(vm_brk); + +/* Release all mmaps. */ +void exit_mmap(struct mm_struct *mm) +{ + struct mmu_gather tlb; + struct vm_area_struct *vma; + unsigned long nr_accounted = 0; + + /* mm's last user has gone, and its about to be pulled down */ + mmu_notifier_release(mm); + + if (mm->locked_vm) { + vma = mm->mmap; + while (vma) { + if (vma->vm_flags & VM_LOCKED) + munlock_vma_pages_all(vma); + vma = vma->vm_next; + } + } + + arch_exit_mmap(mm); + + vma = mm->mmap; + if (!vma) /* Can happen if dup_mmap() received an OOM */ + return; + + lru_add_drain(); + flush_cache_mm(mm); + tlb_gather_mmu(&tlb, mm, 0, -1); + /* update_hiwater_rss(mm) here? but nobody should be looking */ + /* Use -1 here to ensure all VMAs in the mm are unmapped */ + unmap_vmas(&tlb, vma, 0, -1); + + free_pgtables(&tlb, vma, FIRST_USER_ADDRESS, USER_PGTABLES_CEILING); + tlb_finish_mmu(&tlb, 0, -1); + + /* + * Walk the list again, actually closing and freeing it, + * with preemption enabled, without holding any MM locks. + */ + while (vma) { + if (vma->vm_flags & VM_ACCOUNT) + nr_accounted += vma_pages(vma); + vma = remove_vma(vma); + } + vm_unacct_memory(nr_accounted); + + WARN_ON(mm->nr_ptes > (FIRST_USER_ADDRESS+PMD_SIZE-1)>>PMD_SHIFT); +} + +/* Insert vm structure into process list sorted by address + * and into the inode's i_mmap tree. If vm_file is non-NULL + * then i_mmap_mutex is taken here. + */ +int insert_vm_struct(struct mm_struct *mm, struct vm_area_struct *vma) +{ + struct vm_area_struct *prev; + struct rb_node **rb_link, *rb_parent; + + /* + * The vm_pgoff of a purely anonymous vma should be irrelevant + * until its first write fault, when page's anon_vma and index + * are set. But now set the vm_pgoff it will almost certainly + * end up with (unless mremap moves it elsewhere before that + * first wfault), so /proc/pid/maps tells a consistent story. + * + * By setting it to reflect the virtual start address of the + * vma, merges and splits can happen in a seamless way, just + * using the existing file pgoff checks and manipulations. + * Similarly in do_mmap_pgoff and in do_brk. + */ + if (!vma->vm_file) { + BUG_ON(vma->anon_vma); + vma->vm_pgoff = vma->vm_start >> PAGE_SHIFT; + } + if (find_vma_links(mm, vma->vm_start, vma->vm_end, + &prev, &rb_link, &rb_parent)) + return -ENOMEM; + if ((vma->vm_flags & VM_ACCOUNT) && + security_vm_enough_memory_mm(mm, vma_pages(vma))) + return -ENOMEM; + + vma_link(mm, vma, prev, rb_link, rb_parent); + return 0; +} + +/* + * Copy the vma structure to a new location in the same mm, + * prior to moving page table entries, to effect an mremap move. + */ +struct vm_area_struct *copy_vma(struct vm_area_struct **vmap, + unsigned long addr, unsigned long len, pgoff_t pgoff, + bool *need_rmap_locks) +{ + struct vm_area_struct *vma = *vmap; + unsigned long vma_start = vma->vm_start; + struct mm_struct *mm = vma->vm_mm; + struct vm_area_struct *new_vma, *prev; + struct rb_node **rb_link, *rb_parent; + struct mempolicy *pol; + bool faulted_in_anon_vma = true; + + /* + * If anonymous vma has not yet been faulted, update new pgoff + * to match new location, to increase its chance of merging. + */ + if (unlikely(!vma->vm_file && !vma->anon_vma)) { + pgoff = addr >> PAGE_SHIFT; + faulted_in_anon_vma = false; + } + + if (find_vma_links(mm, addr, addr + len, &prev, &rb_link, &rb_parent)) + return NULL; /* should never get here */ + new_vma = vma_merge(mm, prev, addr, addr + len, vma->vm_flags, + vma->anon_vma, vma->vm_file, pgoff, vma_policy(vma), + vma_get_anon_name(vma)); + if (new_vma) { + /* + * Source vma may have been merged into new_vma + */ + if (unlikely(vma_start >= new_vma->vm_start && + vma_start < new_vma->vm_end)) { + /* + * The only way we can get a vma_merge with + * self during an mremap is if the vma hasn't + * been faulted in yet and we were allowed to + * reset the dst vma->vm_pgoff to the + * destination address of the mremap to allow + * the merge to happen. mremap must change the + * vm_pgoff linearity between src and dst vmas + * (in turn preventing a vma_merge) to be + * safe. It is only safe to keep the vm_pgoff + * linear if there are no pages mapped yet. + */ + VM_BUG_ON(faulted_in_anon_vma); + *vmap = vma = new_vma; + } + *need_rmap_locks = (new_vma->vm_pgoff <= vma->vm_pgoff); + } else { + new_vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); + if (new_vma) { + *new_vma = *vma; + new_vma->vm_start = addr; + new_vma->vm_end = addr + len; + new_vma->vm_pgoff = pgoff; + pol = mpol_dup(vma_policy(vma)); + if (IS_ERR(pol)) + goto out_free_vma; + vma_set_policy(new_vma, pol); + INIT_LIST_HEAD(&new_vma->anon_vma_chain); + if (anon_vma_clone(new_vma, vma)) + goto out_free_mempol; + if (new_vma->vm_file) + get_file(new_vma->vm_file); + if (new_vma->vm_ops && new_vma->vm_ops->open) + new_vma->vm_ops->open(new_vma); + vma_link(mm, new_vma, prev, rb_link, rb_parent); + *need_rmap_locks = false; + } + } + return new_vma; + + out_free_mempol: + mpol_put(pol); + out_free_vma: + kmem_cache_free(vm_area_cachep, new_vma); + return NULL; +} + +/* + * Return true if the calling process may expand its vm space by the passed + * number of pages + */ +int may_expand_vm(struct mm_struct *mm, unsigned long npages) +{ + unsigned long cur = mm->total_vm; /* pages */ + unsigned long lim; + + lim = rlimit(RLIMIT_AS) >> PAGE_SHIFT; + + if (cur + npages > lim) + return 0; + return 1; +} + + +static int special_mapping_fault(struct vm_area_struct *vma, + struct vm_fault *vmf) +{ + pgoff_t pgoff; + struct page **pages; + + /* + * special mappings have no vm_file, and in that case, the mm + * uses vm_pgoff internally. So we have to subtract it from here. + * We are allowed to do this because we are the mm; do not copy + * this code into drivers! + */ + pgoff = vmf->pgoff - vma->vm_pgoff; + + for (pages = vma->vm_private_data; pgoff && *pages; ++pages) + pgoff--; + + if (*pages) { + struct page *page = *pages; + get_page(page); + vmf->page = page; + return 0; + } + + return VM_FAULT_SIGBUS; +} + +/* + * Having a close hook prevents vma merging regardless of flags. + */ +static void special_mapping_close(struct vm_area_struct *vma) +{ +} + +static const struct vm_operations_struct special_mapping_vmops = { + .close = special_mapping_close, + .fault = special_mapping_fault, +}; + +/* + * Called with mm->mmap_sem held for writing. + * Insert a new vma covering the given region, with the given flags. + * Its pages are supplied by the given array of struct page *. + * The array can be shorter than len >> PAGE_SHIFT if it's null-terminated. + * The region past the last page supplied will always produce SIGBUS. + * The array pointer and the pages it points to are assumed to stay alive + * for as long as this mapping might exist. + */ +int install_special_mapping(struct mm_struct *mm, + unsigned long addr, unsigned long len, + unsigned long vm_flags, struct page **pages) +{ + int ret; + struct vm_area_struct *vma; + + vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); + if (unlikely(vma == NULL)) + return -ENOMEM; + + INIT_LIST_HEAD(&vma->anon_vma_chain); + vma->vm_mm = mm; + vma->vm_start = addr; + vma->vm_end = addr + len; + + vma->vm_flags = vm_flags | mm->def_flags | VM_DONTEXPAND; + vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); + + vma->vm_ops = &special_mapping_vmops; + vma->vm_private_data = pages; + + ret = insert_vm_struct(mm, vma); + if (ret) + goto out; + + mm->total_vm += len >> PAGE_SHIFT; + + perf_event_mmap(vma); + quadd_event_mmap(vma); + + return 0; + +out: + kmem_cache_free(vm_area_cachep, vma); + return ret; +} + +static DEFINE_MUTEX(mm_all_locks_mutex); + +static void vm_lock_anon_vma(struct mm_struct *mm, struct anon_vma *anon_vma) +{ + if (!test_bit(0, (unsigned long *) &anon_vma->root->rb_root.rb_node)) { + /* + * The LSB of head.next can't change from under us + * because we hold the mm_all_locks_mutex. + */ + down_write_nest_lock(&anon_vma->root->rwsem, &mm->mmap_sem); + /* + * We can safely modify head.next after taking the + * anon_vma->root->rwsem. If some other vma in this mm shares + * the same anon_vma we won't take it again. + * + * No need of atomic instructions here, head.next + * can't change from under us thanks to the + * anon_vma->root->rwsem. + */ + if (__test_and_set_bit(0, (unsigned long *) + &anon_vma->root->rb_root.rb_node)) + BUG(); + } +} + +static void vm_lock_mapping(struct mm_struct *mm, struct address_space *mapping) +{ + if (!test_bit(AS_MM_ALL_LOCKS, &mapping->flags)) { + /* + * AS_MM_ALL_LOCKS can't change from under us because + * we hold the mm_all_locks_mutex. + * + * Operations on ->flags have to be atomic because + * even if AS_MM_ALL_LOCKS is stable thanks to the + * mm_all_locks_mutex, there may be other cpus + * changing other bitflags in parallel to us. + */ + if (test_and_set_bit(AS_MM_ALL_LOCKS, &mapping->flags)) + BUG(); + mutex_lock_nest_lock(&mapping->i_mmap_mutex, &mm->mmap_sem); + } +} + +/* + * This operation locks against the VM for all pte/vma/mm related + * operations that could ever happen on a certain mm. This includes + * vmtruncate, try_to_unmap, and all page faults. + * + * The caller must take the mmap_sem in write mode before calling + * mm_take_all_locks(). The caller isn't allowed to release the + * mmap_sem until mm_drop_all_locks() returns. + * + * mmap_sem in write mode is required in order to block all operations + * that could modify pagetables and free pages without need of + * altering the vma layout (for example populate_range() with + * nonlinear vmas). It's also needed in write mode to avoid new + * anon_vmas to be associated with existing vmas. + * + * A single task can't take more than one mm_take_all_locks() in a row + * or it would deadlock. + * + * The LSB in anon_vma->rb_root.rb_node and the AS_MM_ALL_LOCKS bitflag in + * mapping->flags avoid to take the same lock twice, if more than one + * vma in this mm is backed by the same anon_vma or address_space. + * + * We can take all the locks in random order because the VM code + * taking i_mmap_mutex or anon_vma->rwsem outside the mmap_sem never + * takes more than one of them in a row. Secondly we're protected + * against a concurrent mm_take_all_locks() by the mm_all_locks_mutex. + * + * mm_take_all_locks() and mm_drop_all_locks are expensive operations + * that may have to take thousand of locks. + * + * mm_take_all_locks() can fail if it's interrupted by signals. + */ +int mm_take_all_locks(struct mm_struct *mm) +{ + struct vm_area_struct *vma; + struct anon_vma_chain *avc; + + BUG_ON(down_read_trylock(&mm->mmap_sem)); + + mutex_lock(&mm_all_locks_mutex); + + for (vma = mm->mmap; vma; vma = vma->vm_next) { + if (signal_pending(current)) + goto out_unlock; + if (vma->vm_file && vma->vm_file->f_mapping) + vm_lock_mapping(mm, vma->vm_file->f_mapping); + } + + for (vma = mm->mmap; vma; vma = vma->vm_next) { + if (signal_pending(current)) + goto out_unlock; + if (vma->anon_vma) + list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) + vm_lock_anon_vma(mm, avc->anon_vma); + } + + return 0; + +out_unlock: + mm_drop_all_locks(mm); + return -EINTR; +} + +static void vm_unlock_anon_vma(struct anon_vma *anon_vma) +{ + if (test_bit(0, (unsigned long *) &anon_vma->root->rb_root.rb_node)) { + /* + * The LSB of head.next can't change to 0 from under + * us because we hold the mm_all_locks_mutex. + * + * We must however clear the bitflag before unlocking + * the vma so the users using the anon_vma->rb_root will + * never see our bitflag. + * + * No need of atomic instructions here, head.next + * can't change from under us until we release the + * anon_vma->root->rwsem. + */ + if (!__test_and_clear_bit(0, (unsigned long *) + &anon_vma->root->rb_root.rb_node)) + BUG(); + anon_vma_unlock_write(anon_vma); + } +} + +static void vm_unlock_mapping(struct address_space *mapping) +{ + if (test_bit(AS_MM_ALL_LOCKS, &mapping->flags)) { + /* + * AS_MM_ALL_LOCKS can't change to 0 from under us + * because we hold the mm_all_locks_mutex. + */ + mutex_unlock(&mapping->i_mmap_mutex); + if (!test_and_clear_bit(AS_MM_ALL_LOCKS, + &mapping->flags)) + BUG(); + } +} + +/* + * The mmap_sem cannot be released by the caller until + * mm_drop_all_locks() returns. + */ +void mm_drop_all_locks(struct mm_struct *mm) +{ + struct vm_area_struct *vma; + struct anon_vma_chain *avc; + + BUG_ON(down_read_trylock(&mm->mmap_sem)); + BUG_ON(!mutex_is_locked(&mm_all_locks_mutex)); + + for (vma = mm->mmap; vma; vma = vma->vm_next) { + if (vma->anon_vma) + list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) + vm_unlock_anon_vma(avc->anon_vma); + if (vma->vm_file && vma->vm_file->f_mapping) + vm_unlock_mapping(vma->vm_file->f_mapping); + } + + mutex_unlock(&mm_all_locks_mutex); +} + +/* + * initialise the VMA slab + */ +void __init mmap_init(void) +{ + int ret; + + ret = percpu_counter_init(&vm_committed_as, 0); + VM_BUG_ON(ret); +} + +/* + * Initialise sysctl_user_reserve_kbytes. + * + * This is intended to prevent a user from starting a single memory hogging + * process, such that they cannot recover (kill the hog) in OVERCOMMIT_NEVER + * mode. + * + * The default value is min(3% of free memory, 128MB) + * 128MB is enough to recover with sshd/login, bash, and top/kill. + */ +static int init_user_reserve(void) +{ + unsigned long free_kbytes; + + free_kbytes = global_page_state(NR_FREE_PAGES) << (PAGE_SHIFT - 10); + + sysctl_user_reserve_kbytes = min(free_kbytes / 32, 1UL << 17); + return 0; +} +module_init(init_user_reserve) + +/* + * Initialise sysctl_admin_reserve_kbytes. + * + * The purpose of sysctl_admin_reserve_kbytes is to allow the sys admin + * to log in and kill a memory hogging process. + * + * Systems with more than 256MB will reserve 8MB, enough to recover + * with sshd, bash, and top in OVERCOMMIT_GUESS. Smaller systems will + * only reserve 3% of free pages by default. + */ +static int init_admin_reserve(void) +{ + unsigned long free_kbytes; + + free_kbytes = global_page_state(NR_FREE_PAGES) << (PAGE_SHIFT - 10); + + sysctl_admin_reserve_kbytes = min(free_kbytes / 32, 1UL << 13); + return 0; +} +module_init(init_admin_reserve) + +/* + * Reinititalise user and admin reserves if memory is added or removed. + * + * The default user reserve max is 128MB, and the default max for the + * admin reserve is 8MB. These are usually, but not always, enough to + * enable recovery from a memory hogging process using login/sshd, a shell, + * and tools like top. It may make sense to increase or even disable the + * reserve depending on the existence of swap or variations in the recovery + * tools. So, the admin may have changed them. + * + * If memory is added and the reserves have been eliminated or increased above + * the default max, then we'll trust the admin. + * + * If memory is removed and there isn't enough free memory, then we + * need to reset the reserves. + * + * Otherwise keep the reserve set by the admin. + */ +static int reserve_mem_notifier(struct notifier_block *nb, + unsigned long action, void *data) +{ + unsigned long tmp, free_kbytes; + + switch (action) { + case MEM_ONLINE: + /* Default max is 128MB. Leave alone if modified by operator. */ + tmp = sysctl_user_reserve_kbytes; + if (0 < tmp && tmp < (1UL << 17)) + init_user_reserve(); + + /* Default max is 8MB. Leave alone if modified by operator. */ + tmp = sysctl_admin_reserve_kbytes; + if (0 < tmp && tmp < (1UL << 13)) + init_admin_reserve(); + + break; + case MEM_OFFLINE: + free_kbytes = global_page_state(NR_FREE_PAGES) << (PAGE_SHIFT - 10); + + if (sysctl_user_reserve_kbytes > free_kbytes) { + init_user_reserve(); + pr_info("vm.user_reserve_kbytes reset to %lu\n", + sysctl_user_reserve_kbytes); + } + + if (sysctl_admin_reserve_kbytes > free_kbytes) { + init_admin_reserve(); + pr_info("vm.admin_reserve_kbytes reset to %lu\n", + sysctl_admin_reserve_kbytes); + } + break; + default: + break; + } + return NOTIFY_OK; +} + +static struct notifier_block reserve_mem_nb = { + .notifier_call = reserve_mem_notifier, +}; + +static int __meminit init_reserve_notifier(void) +{ + if (register_hotmemory_notifier(&reserve_mem_nb)) + printk("Failed registering memory add/remove notifier for admin reserve"); + + return 0; +} +module_init(init_reserve_notifier) diff -Naur kernel.21.3.orig/mm/msync.c kernel.21.3.new/mm/msync.c --- kernel.21.3.orig/mm/msync.c 2015-02-03 01:44:33.000000000 +0000 +++ kernel.21.3.new/mm/msync.c 2015-04-25 23:28:45.400480725 +0000 @@ -80,10 +80,10 @@ start = vma->vm_end; if ((flags & MS_SYNC) && file && (vma->vm_flags & VM_SHARED)) { - get_file(file); + vma_get_file(vma); up_read(&mm->mmap_sem); error = vfs_fsync(file, 0); - fput(file); + vma_fput(vma); if (error || start >= end) goto out; down_read(&mm->mmap_sem); diff -Naur kernel.21.3.orig/mm/nommu.c kernel.21.3.new/mm/nommu.c --- kernel.21.3.orig/mm/nommu.c 2015-02-03 01:44:33.000000000 +0000 +++ kernel.21.3.new/mm/nommu.c 2015-04-25 23:28:45.401480715 +0000 @@ -650,7 +650,7 @@ up_write(&nommu_region_sem); if (region->vm_file) - fput(region->vm_file); + vmr_fput(region); /* IO memory and memory shared directly out of the pagecache * from ramfs/tmpfs mustn't be released here */ @@ -808,7 +808,7 @@ if (vma->vm_ops && vma->vm_ops->close) vma->vm_ops->close(vma); if (vma->vm_file) - fput(vma->vm_file); + vma_fput(vma); put_nommu_region(vma->vm_region); kmem_cache_free(vm_area_cachep, vma); } @@ -1374,7 +1374,7 @@ goto error_just_free; } } - fput(region->vm_file); + vmr_fput(region); kmem_cache_free(vm_region_jar, region); region = pregion; result = start; @@ -1450,10 +1450,10 @@ up_write(&nommu_region_sem); error: if (region->vm_file) - fput(region->vm_file); + vmr_fput(region); kmem_cache_free(vm_region_jar, region); if (vma->vm_file) - fput(vma->vm_file); + vma_fput(vma); kmem_cache_free(vm_area_cachep, vma); kleave(" = %d", ret); return ret; diff -Naur kernel.21.3.orig/mm/prfile.c kernel.21.3.new/mm/prfile.c --- kernel.21.3.orig/mm/prfile.c 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/mm/prfile.c 2015-04-25 23:28:45.405480675 +0000 @@ -0,0 +1,86 @@ +/* + * Mainly for aufs which mmap(2) diffrent file and wants to print different path + * in /proc/PID/maps. + * Call these functions via macros defined in linux/mm.h. + * + * See Documentation/filesystems/aufs/design/06mmap.txt + * + * Copyright (c) 2014 Junjro R. Okajima + * Copyright (c) 2014 Ian Campbell + */ + +#include +#include +#include + +/* #define PRFILE_TRACE */ +static inline void prfile_trace(struct file *f, struct file *pr, + const char func[], int line, const char func2[]) +{ +#ifdef PRFILE_TRACE + if (pr) + pr_info("%s:%d: %s, %p\n", func, line, func2, + f ? (char *)f->f_dentry->d_name.name : "(null)"); +#endif +} + +#ifdef CONFIG_MMU +void vma_do_file_update_time(struct vm_area_struct *vma, const char func[], + int line) +{ + struct file *f = vma->vm_file, *pr = vma->vm_prfile; + + prfile_trace(f, pr, func, line, __func__); + file_update_time(f); + if (f && pr) + file_update_time(pr); +} + +struct file *vma_do_pr_or_file(struct vm_area_struct *vma, const char func[], + int line) +{ + struct file *f = vma->vm_file, *pr = vma->vm_prfile; + + prfile_trace(f, pr, func, line, __func__); + return (f && pr) ? pr : f; +} + +void vma_do_get_file(struct vm_area_struct *vma, const char func[], int line) +{ + struct file *f = vma->vm_file, *pr = vma->vm_prfile; + + prfile_trace(f, pr, func, line, __func__); + get_file(f); + if (f && pr) + get_file(pr); +} + +void vma_do_fput(struct vm_area_struct *vma, const char func[], int line) +{ + struct file *f = vma->vm_file, *pr = vma->vm_prfile; + + prfile_trace(f, pr, func, line, __func__); + fput(f); + if (f && pr) + fput(pr); +} +#else +struct file *vmr_do_pr_or_file(struct vm_region *region, const char func[], + int line) +{ + struct file *f = region->vm_file, *pr = region->vm_prfile; + + prfile_trace(f, pr, func, line, __func__); + return (f && pr) ? pr : f; +} + +void vmr_do_fput(struct vm_region *region, const char func[], int line) +{ + struct file *f = region->vm_file, *pr = region->vm_prfile; + + prfile_trace(f, pr, func, line, __func__); + fput(f); + if (f && pr) + fput(pr); +} +#endif /* CONFIG_MMU */ diff -Naur kernel.21.3.orig/security/commoncap.c kernel.21.3.new/security/commoncap.c --- kernel.21.3.orig/security/commoncap.c 2015-02-03 01:44:35.000000000 +0000 +++ kernel.21.3.new/security/commoncap.c 2015-04-25 23:28:54.091393815 +0000 @@ -999,9 +999,11 @@ } return ret; } +EXPORT_SYMBOL(cap_mmap_addr); int cap_mmap_file(struct file *file, unsigned long reqprot, unsigned long prot, unsigned long flags) { return 0; } +EXPORT_SYMBOL(cap_mmap_file); diff -Naur kernel.21.3.orig/security/commoncap.c.orig kernel.21.3.new/security/commoncap.c.orig --- kernel.21.3.orig/security/commoncap.c.orig 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/security/commoncap.c.orig 2015-02-03 01:44:35.000000000 +0000 @@ -0,0 +1,1007 @@ +/* Common capabilities, needed by capability.o. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#ifdef CONFIG_ANDROID_PARANOID_NETWORK +#include +#endif + +/* + * If a non-root user executes a setuid-root binary in + * !secure(SECURE_NOROOT) mode, then we raise capabilities. + * However if fE is also set, then the intent is for only + * the file capabilities to be applied, and the setuid-root + * bit is left on either to change the uid (plausible) or + * to get full privilege on a kernel without file capabilities + * support. So in that case we do not raise capabilities. + * + * Warn if that happens, once per boot. + */ +static void warn_setuid_and_fcaps_mixed(const char *fname) +{ + static int warned; + if (!warned) { + printk(KERN_INFO "warning: `%s' has both setuid-root and" + " effective capabilities. Therefore not raising all" + " capabilities.\n", fname); + warned = 1; + } +} + +int cap_netlink_send(struct sock *sk, struct sk_buff *skb) +{ + return 0; +} + +/** + * cap_capable - Determine whether a task has a particular effective capability + * @cred: The credentials to use + * @ns: The user namespace in which we need the capability + * @cap: The capability to check for + * @audit: Whether to write an audit message or not + * + * Determine whether the nominated task has the specified capability amongst + * its effective set, returning 0 if it does, -ve if it does not. + * + * NOTE WELL: cap_has_capability() cannot be used like the kernel's capable() + * and has_capability() functions. That is, it has the reverse semantics: + * cap_has_capability() returns 0 when a task has a capability, but the + * kernel's capable() and has_capability() returns 1 for this case. + */ +int cap_capable(const struct cred *cred, struct user_namespace *targ_ns, + int cap, int audit) +{ + struct user_namespace *ns = targ_ns; + +#ifdef CONFIG_ANDROID_PARANOID_NETWORK + if (cap == CAP_NET_RAW && in_egroup_p(AID_NET_RAW)) + return 0; + if (cap == CAP_NET_ADMIN && in_egroup_p(AID_NET_ADMIN)) + return 0; +#endif + + /* See if cred has the capability in the target user namespace + * by examining the target user namespace and all of the target + * user namespace's parents. + */ + for (;;) { + /* Do we have the necessary capabilities? */ + if (ns == cred->user_ns) + return cap_raised(cred->cap_effective, cap) ? 0 : -EPERM; + + /* Have we tried all of the parent namespaces? */ + if (ns == &init_user_ns) + return -EPERM; + + /* + * The owner of the user namespace in the parent of the + * user namespace has all caps. + */ + if ((ns->parent == cred->user_ns) && uid_eq(ns->owner, cred->euid)) + return 0; + + /* + * If you have a capability in a parent user ns, then you have + * it over all children user namespaces as well. + */ + ns = ns->parent; + } + + /* We never get here */ +} + +/** + * cap_settime - Determine whether the current process may set the system clock + * @ts: The time to set + * @tz: The timezone to set + * + * Determine whether the current process may set the system clock and timezone + * information, returning 0 if permission granted, -ve if denied. + */ +int cap_settime(const struct timespec *ts, const struct timezone *tz) +{ + if (!capable(CAP_SYS_TIME)) + return -EPERM; + return 0; +} + +/** + * cap_ptrace_access_check - Determine whether the current process may access + * another + * @child: The process to be accessed + * @mode: The mode of attachment. + * + * If we are in the same or an ancestor user_ns and have all the target + * task's capabilities, then ptrace access is allowed. + * If we have the ptrace capability to the target user_ns, then ptrace + * access is allowed. + * Else denied. + * + * Determine whether a process may access another, returning 0 if permission + * granted, -ve if denied. + */ +int cap_ptrace_access_check(struct task_struct *child, unsigned int mode) +{ + int ret = 0; + const struct cred *cred, *child_cred; + + rcu_read_lock(); + cred = current_cred(); + child_cred = __task_cred(child); + if (cred->user_ns == child_cred->user_ns && + cap_issubset(child_cred->cap_permitted, cred->cap_permitted)) + goto out; + if (ns_capable(child_cred->user_ns, CAP_SYS_PTRACE)) + goto out; + ret = -EPERM; +out: + rcu_read_unlock(); + return ret; +} + +/** + * cap_ptrace_traceme - Determine whether another process may trace the current + * @parent: The task proposed to be the tracer + * + * If parent is in the same or an ancestor user_ns and has all current's + * capabilities, then ptrace access is allowed. + * If parent has the ptrace capability to current's user_ns, then ptrace + * access is allowed. + * Else denied. + * + * Determine whether the nominated task is permitted to trace the current + * process, returning 0 if permission is granted, -ve if denied. + */ +int cap_ptrace_traceme(struct task_struct *parent) +{ + int ret = 0; + const struct cred *cred, *child_cred; + + rcu_read_lock(); + cred = __task_cred(parent); + child_cred = current_cred(); + if (cred->user_ns == child_cred->user_ns && + cap_issubset(child_cred->cap_permitted, cred->cap_permitted)) + goto out; + if (has_ns_capability(parent, child_cred->user_ns, CAP_SYS_PTRACE)) + goto out; + ret = -EPERM; +out: + rcu_read_unlock(); + return ret; +} + +/** + * cap_capget - Retrieve a task's capability sets + * @target: The task from which to retrieve the capability sets + * @effective: The place to record the effective set + * @inheritable: The place to record the inheritable set + * @permitted: The place to record the permitted set + * + * This function retrieves the capabilities of the nominated task and returns + * them to the caller. + */ +int cap_capget(struct task_struct *target, kernel_cap_t *effective, + kernel_cap_t *inheritable, kernel_cap_t *permitted) +{ + const struct cred *cred; + + /* Derived from kernel/capability.c:sys_capget. */ + rcu_read_lock(); + cred = __task_cred(target); + *effective = cred->cap_effective; + *inheritable = cred->cap_inheritable; + *permitted = cred->cap_permitted; + rcu_read_unlock(); + return 0; +} + +/* + * Determine whether the inheritable capabilities are limited to the old + * permitted set. Returns 1 if they are limited, 0 if they are not. + */ +static inline int cap_inh_is_capped(void) +{ + + /* they are so limited unless the current task has the CAP_SETPCAP + * capability + */ + if (cap_capable(current_cred(), current_cred()->user_ns, + CAP_SETPCAP, SECURITY_CAP_AUDIT) == 0) + return 0; + return 1; +} + +/** + * cap_capset - Validate and apply proposed changes to current's capabilities + * @new: The proposed new credentials; alterations should be made here + * @old: The current task's current credentials + * @effective: A pointer to the proposed new effective capabilities set + * @inheritable: A pointer to the proposed new inheritable capabilities set + * @permitted: A pointer to the proposed new permitted capabilities set + * + * This function validates and applies a proposed mass change to the current + * process's capability sets. The changes are made to the proposed new + * credentials, and assuming no error, will be committed by the caller of LSM. + */ +int cap_capset(struct cred *new, + const struct cred *old, + const kernel_cap_t *effective, + const kernel_cap_t *inheritable, + const kernel_cap_t *permitted) +{ + if (cap_inh_is_capped() && + !cap_issubset(*inheritable, + cap_combine(old->cap_inheritable, + old->cap_permitted))) + /* incapable of using this inheritable set */ + return -EPERM; + + if (!cap_issubset(*inheritable, + cap_combine(old->cap_inheritable, + old->cap_bset))) + /* no new pI capabilities outside bounding set */ + return -EPERM; + + /* verify restrictions on target's new Permitted set */ + if (!cap_issubset(*permitted, old->cap_permitted)) + return -EPERM; + + /* verify the _new_Effective_ is a subset of the _new_Permitted_ */ + if (!cap_issubset(*effective, *permitted)) + return -EPERM; + + new->cap_effective = *effective; + new->cap_inheritable = *inheritable; + new->cap_permitted = *permitted; + return 0; +} + +/* + * Clear proposed capability sets for execve(). + */ +static inline void bprm_clear_caps(struct linux_binprm *bprm) +{ + cap_clear(bprm->cred->cap_permitted); + bprm->cap_effective = false; +} + +/** + * cap_inode_need_killpriv - Determine if inode change affects privileges + * @dentry: The inode/dentry in being changed with change marked ATTR_KILL_PRIV + * + * Determine if an inode having a change applied that's marked ATTR_KILL_PRIV + * affects the security markings on that inode, and if it is, should + * inode_killpriv() be invoked or the change rejected? + * + * Returns 0 if granted; +ve if granted, but inode_killpriv() is required; and + * -ve to deny the change. + */ +int cap_inode_need_killpriv(struct dentry *dentry) +{ + struct inode *inode = dentry->d_inode; + int error; + + if (!inode->i_op->getxattr) + return 0; + + error = inode->i_op->getxattr(dentry, XATTR_NAME_CAPS, NULL, 0); + if (error <= 0) + return 0; + return 1; +} + +/** + * cap_inode_killpriv - Erase the security markings on an inode + * @dentry: The inode/dentry to alter + * + * Erase the privilege-enhancing security markings on an inode. + * + * Returns 0 if successful, -ve on error. + */ +int cap_inode_killpriv(struct dentry *dentry) +{ + struct inode *inode = dentry->d_inode; + + if (!inode->i_op->removexattr) + return 0; + + return inode->i_op->removexattr(dentry, XATTR_NAME_CAPS); +} + +/* + * Calculate the new process capability sets from the capability sets attached + * to a file. + */ +static inline int bprm_caps_from_vfs_caps(struct cpu_vfs_cap_data *caps, + struct linux_binprm *bprm, + bool *effective, + bool *has_cap) +{ + struct cred *new = bprm->cred; + unsigned i; + int ret = 0; + + if (caps->magic_etc & VFS_CAP_FLAGS_EFFECTIVE) + *effective = true; + + if (caps->magic_etc & VFS_CAP_REVISION_MASK) + *has_cap = true; + + CAP_FOR_EACH_U32(i) { + __u32 permitted = caps->permitted.cap[i]; + __u32 inheritable = caps->inheritable.cap[i]; + + /* + * pP' = (X & fP) | (pI & fI) + */ + new->cap_permitted.cap[i] = + (new->cap_bset.cap[i] & permitted) | + (new->cap_inheritable.cap[i] & inheritable); + + if (permitted & ~new->cap_permitted.cap[i]) + /* insufficient to execute correctly */ + ret = -EPERM; + } + + /* + * For legacy apps, with no internal support for recognizing they + * do not have enough capabilities, we return an error if they are + * missing some "forced" (aka file-permitted) capabilities. + */ + return *effective ? ret : 0; +} + +/* + * Extract the on-exec-apply capability sets for an executable file. + */ +int get_vfs_caps_from_disk(const struct dentry *dentry, struct cpu_vfs_cap_data *cpu_caps) +{ + struct inode *inode = dentry->d_inode; + __u32 magic_etc; + unsigned tocopy, i; + int size; + struct vfs_cap_data caps; + + memset(cpu_caps, 0, sizeof(struct cpu_vfs_cap_data)); + + if (!inode || !inode->i_op->getxattr) + return -ENODATA; + + size = inode->i_op->getxattr((struct dentry *)dentry, XATTR_NAME_CAPS, &caps, + XATTR_CAPS_SZ); + if (size == -ENODATA || size == -EOPNOTSUPP) + /* no data, that's ok */ + return -ENODATA; + if (size < 0) + return size; + + if (size < sizeof(magic_etc)) + return -EINVAL; + + cpu_caps->magic_etc = magic_etc = le32_to_cpu(caps.magic_etc); + + switch (magic_etc & VFS_CAP_REVISION_MASK) { + case VFS_CAP_REVISION_1: + if (size != XATTR_CAPS_SZ_1) + return -EINVAL; + tocopy = VFS_CAP_U32_1; + break; + case VFS_CAP_REVISION_2: + if (size != XATTR_CAPS_SZ_2) + return -EINVAL; + tocopy = VFS_CAP_U32_2; + break; + default: + return -EINVAL; + } + + CAP_FOR_EACH_U32(i) { + if (i >= tocopy) + break; + cpu_caps->permitted.cap[i] = le32_to_cpu(caps.data[i].permitted); + cpu_caps->inheritable.cap[i] = le32_to_cpu(caps.data[i].inheritable); + } + + return 0; +} + +/* + * Attempt to get the on-exec apply capability sets for an executable file from + * its xattrs and, if present, apply them to the proposed credentials being + * constructed by execve(). + */ +static int get_file_caps(struct linux_binprm *bprm, bool *effective, bool *has_cap) +{ + struct dentry *dentry; + int rc = 0; + struct cpu_vfs_cap_data vcaps; + + bprm_clear_caps(bprm); + + if (!file_caps_enabled) + return 0; + + if (bprm->file->f_path.mnt->mnt_flags & MNT_NOSUID) + return 0; + + dentry = dget(bprm->file->f_dentry); + + rc = get_vfs_caps_from_disk(dentry, &vcaps); + if (rc < 0) { + if (rc == -EINVAL) + printk(KERN_NOTICE "%s: get_vfs_caps_from_disk returned %d for %s\n", + __func__, rc, bprm->filename); + else if (rc == -ENODATA) + rc = 0; + goto out; + } + + rc = bprm_caps_from_vfs_caps(&vcaps, bprm, effective, has_cap); + if (rc == -EINVAL) + printk(KERN_NOTICE "%s: cap_from_disk returned %d for %s\n", + __func__, rc, bprm->filename); + +out: + dput(dentry); + if (rc) + bprm_clear_caps(bprm); + + return rc; +} + +/** + * cap_bprm_set_creds - Set up the proposed credentials for execve(). + * @bprm: The execution parameters, including the proposed creds + * + * Set up the proposed credentials for a new execution context being + * constructed by execve(). The proposed creds in @bprm->cred is altered, + * which won't take effect immediately. Returns 0 if successful, -ve on error. + */ +int cap_bprm_set_creds(struct linux_binprm *bprm) +{ + const struct cred *old = current_cred(); + struct cred *new = bprm->cred; + bool effective, has_cap = false; + int ret; + kuid_t root_uid; + + effective = false; + ret = get_file_caps(bprm, &effective, &has_cap); + if (ret < 0) + return ret; + + root_uid = make_kuid(new->user_ns, 0); + + if (!issecure(SECURE_NOROOT)) { + /* + * If the legacy file capability is set, then don't set privs + * for a setuid root binary run by a non-root user. Do set it + * for a root user just to cause least surprise to an admin. + */ + if (has_cap && !uid_eq(new->uid, root_uid) && uid_eq(new->euid, root_uid)) { + warn_setuid_and_fcaps_mixed(bprm->filename); + goto skip; + } + /* + * To support inheritance of root-permissions and suid-root + * executables under compatibility mode, we override the + * capability sets for the file. + * + * If only the real uid is 0, we do not set the effective bit. + */ + if (uid_eq(new->euid, root_uid) || uid_eq(new->uid, root_uid)) { + /* pP' = (cap_bset & ~0) | (pI & ~0) */ + new->cap_permitted = cap_combine(old->cap_bset, + old->cap_inheritable); + } + if (uid_eq(new->euid, root_uid)) + effective = true; + } +skip: + + /* if we have fs caps, clear dangerous personality flags */ + if (!cap_issubset(new->cap_permitted, old->cap_permitted)) + bprm->per_clear |= PER_CLEAR_ON_SETID; + + + /* Don't let someone trace a set[ug]id/setpcap binary with the revised + * credentials unless they have the appropriate permit. + * + * In addition, if NO_NEW_PRIVS, then ensure we get no new privs. + */ + if ((!uid_eq(new->euid, old->uid) || + !gid_eq(new->egid, old->gid) || + !cap_issubset(new->cap_permitted, old->cap_permitted)) && + bprm->unsafe & ~LSM_UNSAFE_PTRACE_CAP) { + /* downgrade; they get no more than they had, and maybe less */ + if (!capable(CAP_SETUID) || + (bprm->unsafe & LSM_UNSAFE_NO_NEW_PRIVS)) { + new->euid = new->uid; + new->egid = new->gid; + } + new->cap_permitted = cap_intersect(new->cap_permitted, + old->cap_permitted); + } + + new->suid = new->fsuid = new->euid; + new->sgid = new->fsgid = new->egid; + + if (effective) + new->cap_effective = new->cap_permitted; + else + cap_clear(new->cap_effective); + bprm->cap_effective = effective; + + /* + * Audit candidate if current->cap_effective is set + * + * We do not bother to audit if 3 things are true: + * 1) cap_effective has all caps + * 2) we are root + * 3) root is supposed to have all caps (SECURE_NOROOT) + * Since this is just a normal root execing a process. + * + * Number 1 above might fail if you don't have a full bset, but I think + * that is interesting information to audit. + */ + if (!cap_isclear(new->cap_effective)) { + if (!cap_issubset(CAP_FULL_SET, new->cap_effective) || + !uid_eq(new->euid, root_uid) || !uid_eq(new->uid, root_uid) || + issecure(SECURE_NOROOT)) { + ret = audit_log_bprm_fcaps(bprm, new, old); + if (ret < 0) + return ret; + } + } + + new->securebits &= ~issecure_mask(SECURE_KEEP_CAPS); + return 0; +} + +/** + * cap_bprm_secureexec - Determine whether a secure execution is required + * @bprm: The execution parameters + * + * Determine whether a secure execution is required, return 1 if it is, and 0 + * if it is not. + * + * The credentials have been committed by this point, and so are no longer + * available through @bprm->cred. + */ +int cap_bprm_secureexec(struct linux_binprm *bprm) +{ + const struct cred *cred = current_cred(); + kuid_t root_uid = make_kuid(cred->user_ns, 0); + + if (!uid_eq(cred->uid, root_uid)) { + if (bprm->cap_effective) + return 1; + if (!cap_isclear(cred->cap_permitted)) + return 1; + } + + return (!uid_eq(cred->euid, cred->uid) || + !gid_eq(cred->egid, cred->gid)); +} + +/** + * cap_inode_setxattr - Determine whether an xattr may be altered + * @dentry: The inode/dentry being altered + * @name: The name of the xattr to be changed + * @value: The value that the xattr will be changed to + * @size: The size of value + * @flags: The replacement flag + * + * Determine whether an xattr may be altered or set on an inode, returning 0 if + * permission is granted, -ve if denied. + * + * This is used to make sure security xattrs don't get updated or set by those + * who aren't privileged to do so. + */ +int cap_inode_setxattr(struct dentry *dentry, const char *name, + const void *value, size_t size, int flags) +{ + if (!strcmp(name, XATTR_NAME_CAPS)) { + if (!capable(CAP_SETFCAP)) + return -EPERM; + return 0; + } + + if (!strncmp(name, XATTR_SECURITY_PREFIX, + sizeof(XATTR_SECURITY_PREFIX) - 1) && + !capable(CAP_SYS_ADMIN)) + return -EPERM; + return 0; +} + +/** + * cap_inode_removexattr - Determine whether an xattr may be removed + * @dentry: The inode/dentry being altered + * @name: The name of the xattr to be changed + * + * Determine whether an xattr may be removed from an inode, returning 0 if + * permission is granted, -ve if denied. + * + * This is used to make sure security xattrs don't get removed by those who + * aren't privileged to remove them. + */ +int cap_inode_removexattr(struct dentry *dentry, const char *name) +{ + if (!strcmp(name, XATTR_NAME_CAPS)) { + if (!capable(CAP_SETFCAP)) + return -EPERM; + return 0; + } + + if (!strncmp(name, XATTR_SECURITY_PREFIX, + sizeof(XATTR_SECURITY_PREFIX) - 1) && + !capable(CAP_SYS_ADMIN)) + return -EPERM; + return 0; +} + +/* + * cap_emulate_setxuid() fixes the effective / permitted capabilities of + * a process after a call to setuid, setreuid, or setresuid. + * + * 1) When set*uiding _from_ one of {r,e,s}uid == 0 _to_ all of + * {r,e,s}uid != 0, the permitted and effective capabilities are + * cleared. + * + * 2) When set*uiding _from_ euid == 0 _to_ euid != 0, the effective + * capabilities of the process are cleared. + * + * 3) When set*uiding _from_ euid != 0 _to_ euid == 0, the effective + * capabilities are set to the permitted capabilities. + * + * fsuid is handled elsewhere. fsuid == 0 and {r,e,s}uid!= 0 should + * never happen. + * + * -astor + * + * cevans - New behaviour, Oct '99 + * A process may, via prctl(), elect to keep its capabilities when it + * calls setuid() and switches away from uid==0. Both permitted and + * effective sets will be retained. + * Without this change, it was impossible for a daemon to drop only some + * of its privilege. The call to setuid(!=0) would drop all privileges! + * Keeping uid 0 is not an option because uid 0 owns too many vital + * files.. + * Thanks to Olaf Kirch and Peter Benie for spotting this. + */ +static inline void cap_emulate_setxuid(struct cred *new, const struct cred *old) +{ + kuid_t root_uid = make_kuid(old->user_ns, 0); + + if ((uid_eq(old->uid, root_uid) || + uid_eq(old->euid, root_uid) || + uid_eq(old->suid, root_uid)) && + (!uid_eq(new->uid, root_uid) && + !uid_eq(new->euid, root_uid) && + !uid_eq(new->suid, root_uid)) && + !issecure(SECURE_KEEP_CAPS)) { + cap_clear(new->cap_permitted); + cap_clear(new->cap_effective); + } + if (uid_eq(old->euid, root_uid) && !uid_eq(new->euid, root_uid)) + cap_clear(new->cap_effective); + if (!uid_eq(old->euid, root_uid) && uid_eq(new->euid, root_uid)) + new->cap_effective = new->cap_permitted; +} + +/** + * cap_task_fix_setuid - Fix up the results of setuid() call + * @new: The proposed credentials + * @old: The current task's current credentials + * @flags: Indications of what has changed + * + * Fix up the results of setuid() call before the credential changes are + * actually applied, returning 0 to grant the changes, -ve to deny them. + */ +int cap_task_fix_setuid(struct cred *new, const struct cred *old, int flags) +{ + switch (flags) { + case LSM_SETID_RE: + case LSM_SETID_ID: + case LSM_SETID_RES: + /* juggle the capabilities to follow [RES]UID changes unless + * otherwise suppressed */ + if (!issecure(SECURE_NO_SETUID_FIXUP)) + cap_emulate_setxuid(new, old); + break; + + case LSM_SETID_FS: + /* juggle the capabilties to follow FSUID changes, unless + * otherwise suppressed + * + * FIXME - is fsuser used for all CAP_FS_MASK capabilities? + * if not, we might be a bit too harsh here. + */ + if (!issecure(SECURE_NO_SETUID_FIXUP)) { + kuid_t root_uid = make_kuid(old->user_ns, 0); + if (uid_eq(old->fsuid, root_uid) && !uid_eq(new->fsuid, root_uid)) + new->cap_effective = + cap_drop_fs_set(new->cap_effective); + + if (!uid_eq(old->fsuid, root_uid) && uid_eq(new->fsuid, root_uid)) + new->cap_effective = + cap_raise_fs_set(new->cap_effective, + new->cap_permitted); + } + break; + + default: + return -EINVAL; + } + + return 0; +} + +/* + * Rationale: code calling task_setscheduler, task_setioprio, and + * task_setnice, assumes that + * . if capable(cap_sys_nice), then those actions should be allowed + * . if not capable(cap_sys_nice), but acting on your own processes, + * then those actions should be allowed + * This is insufficient now since you can call code without suid, but + * yet with increased caps. + * So we check for increased caps on the target process. + */ +static int cap_safe_nice(struct task_struct *p) +{ + int is_subset; + + rcu_read_lock(); + is_subset = cap_issubset(__task_cred(p)->cap_permitted, + current_cred()->cap_permitted); + rcu_read_unlock(); + + if (!is_subset && !capable(CAP_SYS_NICE)) + return -EPERM; + return 0; +} + +/** + * cap_task_setscheduler - Detemine if scheduler policy change is permitted + * @p: The task to affect + * + * Detemine if the requested scheduler policy change is permitted for the + * specified task, returning 0 if permission is granted, -ve if denied. + */ +int cap_task_setscheduler(struct task_struct *p) +{ + return cap_safe_nice(p); +} + +/** + * cap_task_ioprio - Detemine if I/O priority change is permitted + * @p: The task to affect + * @ioprio: The I/O priority to set + * + * Detemine if the requested I/O priority change is permitted for the specified + * task, returning 0 if permission is granted, -ve if denied. + */ +int cap_task_setioprio(struct task_struct *p, int ioprio) +{ + return cap_safe_nice(p); +} + +/** + * cap_task_ioprio - Detemine if task priority change is permitted + * @p: The task to affect + * @nice: The nice value to set + * + * Detemine if the requested task priority change is permitted for the + * specified task, returning 0 if permission is granted, -ve if denied. + */ +int cap_task_setnice(struct task_struct *p, int nice) +{ + return cap_safe_nice(p); +} + +/* + * Implement PR_CAPBSET_DROP. Attempt to remove the specified capability from + * the current task's bounding set. Returns 0 on success, -ve on error. + */ +static long cap_prctl_drop(struct cred *new, unsigned long cap) +{ + if (!capable(CAP_SETPCAP)) + return -EPERM; + if (!cap_valid(cap)) + return -EINVAL; + + cap_lower(new->cap_bset, cap); + return 0; +} + +/** + * cap_task_prctl - Implement process control functions for this security module + * @option: The process control function requested + * @arg2, @arg3, @arg4, @arg5: The argument data for this function + * + * Allow process control functions (sys_prctl()) to alter capabilities; may + * also deny access to other functions not otherwise implemented here. + * + * Returns 0 or +ve on success, -ENOSYS if this function is not implemented + * here, other -ve on error. If -ENOSYS is returned, sys_prctl() and other LSM + * modules will consider performing the function. + */ +int cap_task_prctl(int option, unsigned long arg2, unsigned long arg3, + unsigned long arg4, unsigned long arg5) +{ + struct cred *new; + long error = 0; + + new = prepare_creds(); + if (!new) + return -ENOMEM; + + switch (option) { + case PR_CAPBSET_READ: + error = -EINVAL; + if (!cap_valid(arg2)) + goto error; + error = !!cap_raised(new->cap_bset, arg2); + goto no_change; + + case PR_CAPBSET_DROP: + error = cap_prctl_drop(new, arg2); + if (error < 0) + goto error; + goto changed; + + /* + * The next four prctl's remain to assist with transitioning a + * system from legacy UID=0 based privilege (when filesystem + * capabilities are not in use) to a system using filesystem + * capabilities only - as the POSIX.1e draft intended. + * + * Note: + * + * PR_SET_SECUREBITS = + * issecure_mask(SECURE_KEEP_CAPS_LOCKED) + * | issecure_mask(SECURE_NOROOT) + * | issecure_mask(SECURE_NOROOT_LOCKED) + * | issecure_mask(SECURE_NO_SETUID_FIXUP) + * | issecure_mask(SECURE_NO_SETUID_FIXUP_LOCKED) + * + * will ensure that the current process and all of its + * children will be locked into a pure + * capability-based-privilege environment. + */ + case PR_SET_SECUREBITS: + error = -EPERM; + if ((((new->securebits & SECURE_ALL_LOCKS) >> 1) + & (new->securebits ^ arg2)) /*[1]*/ + || ((new->securebits & SECURE_ALL_LOCKS & ~arg2)) /*[2]*/ + || (arg2 & ~(SECURE_ALL_LOCKS | SECURE_ALL_BITS)) /*[3]*/ + || (cap_capable(current_cred(), + current_cred()->user_ns, CAP_SETPCAP, + SECURITY_CAP_AUDIT) != 0) /*[4]*/ + /* + * [1] no changing of bits that are locked + * [2] no unlocking of locks + * [3] no setting of unsupported bits + * [4] doing anything requires privilege (go read about + * the "sendmail capabilities bug") + */ + ) + /* cannot change a locked bit */ + goto error; + new->securebits = arg2; + goto changed; + + case PR_GET_SECUREBITS: + error = new->securebits; + goto no_change; + + case PR_GET_KEEPCAPS: + if (issecure(SECURE_KEEP_CAPS)) + error = 1; + goto no_change; + + case PR_SET_KEEPCAPS: + error = -EINVAL; + if (arg2 > 1) /* Note, we rely on arg2 being unsigned here */ + goto error; + error = -EPERM; + if (issecure(SECURE_KEEP_CAPS_LOCKED)) + goto error; + if (arg2) + new->securebits |= issecure_mask(SECURE_KEEP_CAPS); + else + new->securebits &= ~issecure_mask(SECURE_KEEP_CAPS); + goto changed; + + default: + /* No functionality available - continue with default */ + error = -ENOSYS; + goto error; + } + + /* Functionality provided */ +changed: + return commit_creds(new); + +no_change: +error: + abort_creds(new); + return error; +} + +/** + * cap_vm_enough_memory - Determine whether a new virtual mapping is permitted + * @mm: The VM space in which the new mapping is to be made + * @pages: The size of the mapping + * + * Determine whether the allocation of a new virtual mapping by the current + * task is permitted, returning 0 if permission is granted, -ve if not. + */ +int cap_vm_enough_memory(struct mm_struct *mm, long pages) +{ + int cap_sys_admin = 0; + + if (cap_capable(current_cred(), &init_user_ns, CAP_SYS_ADMIN, + SECURITY_CAP_NOAUDIT) == 0) + cap_sys_admin = 1; + return __vm_enough_memory(mm, pages, cap_sys_admin); +} + +/* + * cap_mmap_addr - check if able to map given addr + * @addr: address attempting to be mapped + * + * If the process is attempting to map memory below dac_mmap_min_addr they need + * CAP_SYS_RAWIO. The other parameters to this function are unused by the + * capability security module. Returns 0 if this mapping should be allowed + * -EPERM if not. + */ +int cap_mmap_addr(unsigned long addr) +{ + int ret = 0; + + if (addr < dac_mmap_min_addr) { + ret = cap_capable(current_cred(), &init_user_ns, CAP_SYS_RAWIO, + SECURITY_CAP_AUDIT); + /* set PF_SUPERPRIV if it turns out we allow the low mmap */ + if (ret == 0) + current->flags |= PF_SUPERPRIV; + } + return ret; +} + +int cap_mmap_file(struct file *file, unsigned long reqprot, + unsigned long prot, unsigned long flags) +{ + return 0; +} diff -Naur kernel.21.3.orig/security/device_cgroup.c kernel.21.3.new/security/device_cgroup.c --- kernel.21.3.orig/security/device_cgroup.c 2015-02-03 01:44:35.000000000 +0000 +++ kernel.21.3.new/security/device_cgroup.c 2015-04-25 23:28:54.097393755 +0000 @@ -7,6 +7,7 @@ #include #include #include +#include #include #include #include @@ -789,6 +790,7 @@ return __devcgroup_check_permission(type, imajor(inode), iminor(inode), access); } +EXPORT_SYMBOL(__devcgroup_inode_permission); int devcgroup_inode_mknod(int mode, dev_t dev) { diff -Naur kernel.21.3.orig/security/security.c kernel.21.3.new/security/security.c --- kernel.21.3.orig/security/security.c 2015-02-03 01:44:35.000000000 +0000 +++ kernel.21.3.new/security/security.c 2015-04-25 23:28:54.121393514 +0000 @@ -416,6 +416,7 @@ return 0; return security_ops->path_rmdir(dir, dentry); } +EXPORT_SYMBOL(security_path_rmdir); int security_path_unlink(struct path *dir, struct dentry *dentry) { @@ -432,6 +433,7 @@ return 0; return security_ops->path_symlink(dir, dentry, old_name); } +EXPORT_SYMBOL(security_path_symlink); int security_path_link(struct dentry *old_dentry, struct path *new_dir, struct dentry *new_dentry) @@ -440,6 +442,7 @@ return 0; return security_ops->path_link(old_dentry, new_dir, new_dentry); } +EXPORT_SYMBOL(security_path_link); int security_path_rename(struct path *old_dir, struct dentry *old_dentry, struct path *new_dir, struct dentry *new_dentry) @@ -458,6 +461,7 @@ return 0; return security_ops->path_truncate(path); } +EXPORT_SYMBOL(security_path_truncate); int security_path_chmod(struct path *path, umode_t mode) { @@ -465,6 +469,7 @@ return 0; return security_ops->path_chmod(path, mode); } +EXPORT_SYMBOL(security_path_chmod); int security_path_chown(struct path *path, kuid_t uid, kgid_t gid) { @@ -472,6 +477,7 @@ return 0; return security_ops->path_chown(path, uid, gid); } +EXPORT_SYMBOL(security_path_chown); int security_path_chroot(struct path *path) { @@ -548,6 +554,7 @@ return 0; return security_ops->inode_readlink(dentry); } +EXPORT_SYMBOL(security_inode_readlink); int security_inode_follow_link(struct dentry *dentry, struct nameidata *nd) { @@ -562,6 +569,7 @@ return 0; return security_ops->inode_permission(inode, mask); } +EXPORT_SYMBOL(security_inode_permission); int security_inode_setattr(struct dentry *dentry, struct iattr *attr) { @@ -683,6 +691,7 @@ return fsnotify_perm(file, mask); } +EXPORT_SYMBOL(security_file_permission); int security_file_alloc(struct file *file) { @@ -743,6 +752,7 @@ return ret; return ima_file_mmap(file, prot); } +EXPORT_SYMBOL(security_mmap_file); int security_mmap_addr(unsigned long addr) { diff -Naur kernel.21.3.orig/security/security.c.orig kernel.21.3.new/security/security.c.orig --- kernel.21.3.orig/security/security.c.orig 1970-01-01 00:00:00.000000000 +0000 +++ kernel.21.3.new/security/security.c.orig 2015-02-03 01:44:35.000000000 +0000 @@ -0,0 +1,1450 @@ +/* + * Security plug functions + * + * Copyright (C) 2001 WireX Communications, Inc + * Copyright (C) 2001-2002 Greg Kroah-Hartman + * Copyright (C) 2001 Networks Associates Technology, Inc + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define MAX_LSM_EVM_XATTR 2 + +/* Boot-time LSM user choice */ +static __initdata char chosen_lsm[SECURITY_NAME_MAX + 1] = + CONFIG_DEFAULT_SECURITY; + +static struct security_operations *security_ops; +static struct security_operations default_security_ops = { + .name = "default", +}; + +static inline int __init verify(struct security_operations *ops) +{ + /* verify the security_operations structure exists */ + if (!ops) + return -EINVAL; + security_fixup_ops(ops); + return 0; +} + +static void __init do_security_initcalls(void) +{ + initcall_t *call; + call = __security_initcall_start; + while (call < __security_initcall_end) { + (*call) (); + call++; + } +} + +/** + * security_init - initializes the security framework + * + * This should be called early in the kernel initialization sequence. + */ +int __init security_init(void) +{ + printk(KERN_INFO "Security Framework initialized\n"); + + security_fixup_ops(&default_security_ops); + security_ops = &default_security_ops; + do_security_initcalls(); + + return 0; +} + +void reset_security_ops(void) +{ + security_ops = &default_security_ops; +} + +/* Save user chosen LSM */ +static int __init choose_lsm(char *str) +{ + strncpy(chosen_lsm, str, SECURITY_NAME_MAX); + return 1; +} +__setup("security=", choose_lsm); + +/** + * security_module_enable - Load given security module on boot ? + * @ops: a pointer to the struct security_operations that is to be checked. + * + * Each LSM must pass this method before registering its own operations + * to avoid security registration races. This method may also be used + * to check if your LSM is currently loaded during kernel initialization. + * + * Return true if: + * -The passed LSM is the one chosen by user at boot time, + * -or the passed LSM is configured as the default and the user did not + * choose an alternate LSM at boot time. + * Otherwise, return false. + */ +int __init security_module_enable(struct security_operations *ops) +{ + return !strcmp(ops->name, chosen_lsm); +} + +/** + * register_security - registers a security framework with the kernel + * @ops: a pointer to the struct security_options that is to be registered + * + * This function allows a security module to register itself with the + * kernel security subsystem. Some rudimentary checking is done on the @ops + * value passed to this function. You'll need to check first if your LSM + * is allowed to register its @ops by calling security_module_enable(@ops). + * + * If there is already a security module registered with the kernel, + * an error will be returned. Otherwise %0 is returned on success. + */ +int __init register_security(struct security_operations *ops) +{ + if (verify(ops)) { + printk(KERN_DEBUG "%s could not verify " + "security_operations structure.\n", __func__); + return -EINVAL; + } + + if (security_ops != &default_security_ops) + return -EAGAIN; + + security_ops = ops; + + return 0; +} + +/* Security operations */ + +int security_binder_set_context_mgr(struct task_struct *mgr) +{ + return security_ops->binder_set_context_mgr(mgr); +} + +int security_binder_transaction(struct task_struct *from, struct task_struct *to) +{ + return security_ops->binder_transaction(from, to); +} + +int security_binder_transfer_binder(struct task_struct *from, struct task_struct *to) +{ + return security_ops->binder_transfer_binder(from, to); +} + +int security_binder_transfer_file(struct task_struct *from, struct task_struct *to, struct file *file) +{ + return security_ops->binder_transfer_file(from, to, file); +} + +int security_ptrace_access_check(struct task_struct *child, unsigned int mode) +{ +#ifdef CONFIG_SECURITY_YAMA_STACKED + int rc; + rc = yama_ptrace_access_check(child, mode); + if (rc) + return rc; +#endif + return security_ops->ptrace_access_check(child, mode); +} + +int security_ptrace_traceme(struct task_struct *parent) +{ +#ifdef CONFIG_SECURITY_YAMA_STACKED + int rc; + rc = yama_ptrace_traceme(parent); + if (rc) + return rc; +#endif + return security_ops->ptrace_traceme(parent); +} + +int security_capget(struct task_struct *target, + kernel_cap_t *effective, + kernel_cap_t *inheritable, + kernel_cap_t *permitted) +{ + return security_ops->capget(target, effective, inheritable, permitted); +} + +int security_capset(struct cred *new, const struct cred *old, + const kernel_cap_t *effective, + const kernel_cap_t *inheritable, + const kernel_cap_t *permitted) +{ + return security_ops->capset(new, old, + effective, inheritable, permitted); +} + +int security_capable(const struct cred *cred, struct user_namespace *ns, + int cap) +{ + return security_ops->capable(cred, ns, cap, SECURITY_CAP_AUDIT); +} + +int security_capable_noaudit(const struct cred *cred, struct user_namespace *ns, + int cap) +{ + return security_ops->capable(cred, ns, cap, SECURITY_CAP_NOAUDIT); +} + +int security_quotactl(int cmds, int type, int id, struct super_block *sb) +{ + return security_ops->quotactl(cmds, type, id, sb); +} + +int security_quota_on(struct dentry *dentry) +{ + return security_ops->quota_on(dentry); +} + +int security_syslog(int type) +{ + return security_ops->syslog(type); +} + +int security_settime(const struct timespec *ts, const struct timezone *tz) +{ + return security_ops->settime(ts, tz); +} + +int security_vm_enough_memory_mm(struct mm_struct *mm, long pages) +{ + return security_ops->vm_enough_memory(mm, pages); +} + +int security_bprm_set_creds(struct linux_binprm *bprm) +{ + return security_ops->bprm_set_creds(bprm); +} + +int security_bprm_check(struct linux_binprm *bprm) +{ + int ret; + + ret = security_ops->bprm_check_security(bprm); + if (ret) + return ret; + return ima_bprm_check(bprm); +} + +void security_bprm_committing_creds(struct linux_binprm *bprm) +{ + security_ops->bprm_committing_creds(bprm); +} + +void security_bprm_committed_creds(struct linux_binprm *bprm) +{ + security_ops->bprm_committed_creds(bprm); +} + +int security_bprm_secureexec(struct linux_binprm *bprm) +{ + return security_ops->bprm_secureexec(bprm); +} + +int security_sb_alloc(struct super_block *sb) +{ + return security_ops->sb_alloc_security(sb); +} + +void security_sb_free(struct super_block *sb) +{ + security_ops->sb_free_security(sb); +} + +int security_sb_copy_data(char *orig, char *copy) +{ + return security_ops->sb_copy_data(orig, copy); +} +EXPORT_SYMBOL(security_sb_copy_data); + +int security_sb_remount(struct super_block *sb, void *data) +{ + return security_ops->sb_remount(sb, data); +} + +int security_sb_kern_mount(struct super_block *sb, int flags, void *data) +{ + return security_ops->sb_kern_mount(sb, flags, data); +} + +int security_sb_show_options(struct seq_file *m, struct super_block *sb) +{ + return security_ops->sb_show_options(m, sb); +} + +int security_sb_statfs(struct dentry *dentry) +{ + return security_ops->sb_statfs(dentry); +} + +int security_sb_mount(const char *dev_name, struct path *path, + const char *type, unsigned long flags, void *data) +{ + return security_ops->sb_mount(dev_name, path, type, flags, data); +} + +int security_sb_umount(struct vfsmount *mnt, int flags) +{ + return security_ops->sb_umount(mnt, flags); +} + +int security_sb_pivotroot(struct path *old_path, struct path *new_path) +{ + return security_ops->sb_pivotroot(old_path, new_path); +} + +int security_sb_set_mnt_opts(struct super_block *sb, + struct security_mnt_opts *opts) +{ + return security_ops->sb_set_mnt_opts(sb, opts); +} +EXPORT_SYMBOL(security_sb_set_mnt_opts); + +int security_sb_clone_mnt_opts(const struct super_block *oldsb, + struct super_block *newsb) +{ + return security_ops->sb_clone_mnt_opts(oldsb, newsb); +} +EXPORT_SYMBOL(security_sb_clone_mnt_opts); + +int security_sb_parse_opts_str(char *options, struct security_mnt_opts *opts) +{ + return security_ops->sb_parse_opts_str(options, opts); +} +EXPORT_SYMBOL(security_sb_parse_opts_str); + +int security_inode_alloc(struct inode *inode) +{ + inode->i_security = NULL; + return security_ops->inode_alloc_security(inode); +} + +void security_inode_free(struct inode *inode) +{ + integrity_inode_free(inode); + security_ops->inode_free_security(inode); +} + +int security_inode_init_security(struct inode *inode, struct inode *dir, + const struct qstr *qstr, + const initxattrs initxattrs, void *fs_data) +{ + struct xattr new_xattrs[MAX_LSM_EVM_XATTR + 1]; + struct xattr *lsm_xattr, *evm_xattr, *xattr; + int ret; + + if (unlikely(IS_PRIVATE(inode))) + return 0; + + memset(new_xattrs, 0, sizeof new_xattrs); + if (!initxattrs) + return security_ops->inode_init_security(inode, dir, qstr, + NULL, NULL, NULL); + lsm_xattr = new_xattrs; + ret = security_ops->inode_init_security(inode, dir, qstr, + &lsm_xattr->name, + &lsm_xattr->value, + &lsm_xattr->value_len); + if (ret) + goto out; + + evm_xattr = lsm_xattr + 1; + ret = evm_inode_init_security(inode, lsm_xattr, evm_xattr); + if (ret) + goto out; + ret = initxattrs(inode, new_xattrs, fs_data); +out: + for (xattr = new_xattrs; xattr->name != NULL; xattr++) { + kfree(xattr->name); + kfree(xattr->value); + } + return (ret == -EOPNOTSUPP) ? 0 : ret; +} +EXPORT_SYMBOL(security_inode_init_security); + +int security_old_inode_init_security(struct inode *inode, struct inode *dir, + const struct qstr *qstr, char **name, + void **value, size_t *len) +{ + if (unlikely(IS_PRIVATE(inode))) + return -EOPNOTSUPP; + return security_ops->inode_init_security(inode, dir, qstr, name, value, + len); +} +EXPORT_SYMBOL(security_old_inode_init_security); + +#ifdef CONFIG_SECURITY_PATH +int security_path_mknod(struct path *dir, struct dentry *dentry, umode_t mode, + unsigned int dev) +{ + if (unlikely(IS_PRIVATE(dir->dentry->d_inode))) + return 0; + return security_ops->path_mknod(dir, dentry, mode, dev); +} +EXPORT_SYMBOL(security_path_mknod); + +int security_path_mkdir(struct path *dir, struct dentry *dentry, umode_t mode) +{ + if (unlikely(IS_PRIVATE(dir->dentry->d_inode))) + return 0; + return security_ops->path_mkdir(dir, dentry, mode); +} +EXPORT_SYMBOL(security_path_mkdir); + +int security_path_rmdir(struct path *dir, struct dentry *dentry) +{ + if (unlikely(IS_PRIVATE(dir->dentry->d_inode))) + return 0; + return security_ops->path_rmdir(dir, dentry); +} + +int security_path_unlink(struct path *dir, struct dentry *dentry) +{ + if (unlikely(IS_PRIVATE(dir->dentry->d_inode))) + return 0; + return security_ops->path_unlink(dir, dentry); +} +EXPORT_SYMBOL(security_path_unlink); + +int security_path_symlink(struct path *dir, struct dentry *dentry, + const char *old_name) +{ + if (unlikely(IS_PRIVATE(dir->dentry->d_inode))) + return 0; + return security_ops->path_symlink(dir, dentry, old_name); +} + +int security_path_link(struct dentry *old_dentry, struct path *new_dir, + struct dentry *new_dentry) +{ + if (unlikely(IS_PRIVATE(old_dentry->d_inode))) + return 0; + return security_ops->path_link(old_dentry, new_dir, new_dentry); +} + +int security_path_rename(struct path *old_dir, struct dentry *old_dentry, + struct path *new_dir, struct dentry *new_dentry) +{ + if (unlikely(IS_PRIVATE(old_dentry->d_inode) || + (new_dentry->d_inode && IS_PRIVATE(new_dentry->d_inode)))) + return 0; + return security_ops->path_rename(old_dir, old_dentry, new_dir, + new_dentry); +} +EXPORT_SYMBOL(security_path_rename); + +int security_path_truncate(struct path *path) +{ + if (unlikely(IS_PRIVATE(path->dentry->d_inode))) + return 0; + return security_ops->path_truncate(path); +} + +int security_path_chmod(struct path *path, umode_t mode) +{ + if (unlikely(IS_PRIVATE(path->dentry->d_inode))) + return 0; + return security_ops->path_chmod(path, mode); +} + +int security_path_chown(struct path *path, kuid_t uid, kgid_t gid) +{ + if (unlikely(IS_PRIVATE(path->dentry->d_inode))) + return 0; + return security_ops->path_chown(path, uid, gid); +} + +int security_path_chroot(struct path *path) +{ + return security_ops->path_chroot(path); +} +#endif + +int security_inode_create(struct inode *dir, struct dentry *dentry, umode_t mode) +{ + if (unlikely(IS_PRIVATE(dir))) + return 0; + return security_ops->inode_create(dir, dentry, mode); +} +EXPORT_SYMBOL_GPL(security_inode_create); + +int security_inode_link(struct dentry *old_dentry, struct inode *dir, + struct dentry *new_dentry) +{ + if (unlikely(IS_PRIVATE(old_dentry->d_inode))) + return 0; + return security_ops->inode_link(old_dentry, dir, new_dentry); +} + +int security_inode_unlink(struct inode *dir, struct dentry *dentry) +{ + if (unlikely(IS_PRIVATE(dentry->d_inode))) + return 0; + return security_ops->inode_unlink(dir, dentry); +} + +int security_inode_symlink(struct inode *dir, struct dentry *dentry, + const char *old_name) +{ + if (unlikely(IS_PRIVATE(dir))) + return 0; + return security_ops->inode_symlink(dir, dentry, old_name); +} + +int security_inode_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +{ + if (unlikely(IS_PRIVATE(dir))) + return 0; + return security_ops->inode_mkdir(dir, dentry, mode); +} +EXPORT_SYMBOL_GPL(security_inode_mkdir); + +int security_inode_rmdir(struct inode *dir, struct dentry *dentry) +{ + if (unlikely(IS_PRIVATE(dentry->d_inode))) + return 0; + return security_ops->inode_rmdir(dir, dentry); +} + +int security_inode_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev) +{ + if (unlikely(IS_PRIVATE(dir))) + return 0; + return security_ops->inode_mknod(dir, dentry, mode, dev); +} + +int security_inode_rename(struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry) +{ + if (unlikely(IS_PRIVATE(old_dentry->d_inode) || + (new_dentry->d_inode && IS_PRIVATE(new_dentry->d_inode)))) + return 0; + return security_ops->inode_rename(old_dir, old_dentry, + new_dir, new_dentry); +} + +int security_inode_readlink(struct dentry *dentry) +{ + if (unlikely(IS_PRIVATE(dentry->d_inode))) + return 0; + return security_ops->inode_readlink(dentry); +} + +int security_inode_follow_link(struct dentry *dentry, struct nameidata *nd) +{ + if (unlikely(IS_PRIVATE(dentry->d_inode))) + return 0; + return security_ops->inode_follow_link(dentry, nd); +} + +int security_inode_permission(struct inode *inode, int mask) +{ + if (unlikely(IS_PRIVATE(inode))) + return 0; + return security_ops->inode_permission(inode, mask); +} + +int security_inode_setattr(struct dentry *dentry, struct iattr *attr) +{ + int ret; + + if (unlikely(IS_PRIVATE(dentry->d_inode))) + return 0; + ret = security_ops->inode_setattr(dentry, attr); + if (ret) + return ret; + return evm_inode_setattr(dentry, attr); +} +EXPORT_SYMBOL_GPL(security_inode_setattr); + +int security_inode_getattr(struct vfsmount *mnt, struct dentry *dentry) +{ + if (unlikely(IS_PRIVATE(dentry->d_inode))) + return 0; + return security_ops->inode_getattr(mnt, dentry); +} + +int security_inode_setxattr(struct dentry *dentry, const char *name, + const void *value, size_t size, int flags) +{ + int ret; + + if (unlikely(IS_PRIVATE(dentry->d_inode))) + return 0; + ret = security_ops->inode_setxattr(dentry, name, value, size, flags); + if (ret) + return ret; + ret = ima_inode_setxattr(dentry, name, value, size); + if (ret) + return ret; + return evm_inode_setxattr(dentry, name, value, size); +} + +void security_inode_post_setxattr(struct dentry *dentry, const char *name, + const void *value, size_t size, int flags) +{ + if (unlikely(IS_PRIVATE(dentry->d_inode))) + return; + security_ops->inode_post_setxattr(dentry, name, value, size, flags); + evm_inode_post_setxattr(dentry, name, value, size); +} + +int security_inode_getxattr(struct dentry *dentry, const char *name) +{ + if (unlikely(IS_PRIVATE(dentry->d_inode))) + return 0; + return security_ops->inode_getxattr(dentry, name); +} + +int security_inode_listxattr(struct dentry *dentry) +{ + if (unlikely(IS_PRIVATE(dentry->d_inode))) + return 0; + return security_ops->inode_listxattr(dentry); +} + +int security_inode_removexattr(struct dentry *dentry, const char *name) +{ + int ret; + + if (unlikely(IS_PRIVATE(dentry->d_inode))) + return 0; + ret = security_ops->inode_removexattr(dentry, name); + if (ret) + return ret; + ret = ima_inode_removexattr(dentry, name); + if (ret) + return ret; + return evm_inode_removexattr(dentry, name); +} + +int security_inode_need_killpriv(struct dentry *dentry) +{ + return security_ops->inode_need_killpriv(dentry); +} + +int security_inode_killpriv(struct dentry *dentry) +{ + return security_ops->inode_killpriv(dentry); +} + +int security_inode_getsecurity(const struct inode *inode, const char *name, void **buffer, bool alloc) +{ + if (unlikely(IS_PRIVATE(inode))) + return -EOPNOTSUPP; + return security_ops->inode_getsecurity(inode, name, buffer, alloc); +} + +int security_inode_setsecurity(struct inode *inode, const char *name, const void *value, size_t size, int flags) +{ + if (unlikely(IS_PRIVATE(inode))) + return -EOPNOTSUPP; + return security_ops->inode_setsecurity(inode, name, value, size, flags); +} + +int security_inode_listsecurity(struct inode *inode, char *buffer, size_t buffer_size) +{ + if (unlikely(IS_PRIVATE(inode))) + return 0; + return security_ops->inode_listsecurity(inode, buffer, buffer_size); +} + +void security_inode_getsecid(const struct inode *inode, u32 *secid) +{ + security_ops->inode_getsecid(inode, secid); +} + +int security_file_permission(struct file *file, int mask) +{ + int ret; + + ret = security_ops->file_permission(file, mask); + if (ret) + return ret; + + return fsnotify_perm(file, mask); +} + +int security_file_alloc(struct file *file) +{ + return security_ops->file_alloc_security(file); +} + +void security_file_free(struct file *file) +{ + security_ops->file_free_security(file); +} + +int security_file_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + return security_ops->file_ioctl(file, cmd, arg); +} + +static inline unsigned long mmap_prot(struct file *file, unsigned long prot) +{ + /* + * Does we have PROT_READ and does the application expect + * it to imply PROT_EXEC? If not, nothing to talk about... + */ + if ((prot & (PROT_READ | PROT_EXEC)) != PROT_READ) + return prot; + if (!(current->personality & READ_IMPLIES_EXEC)) + return prot; + /* + * if that's an anonymous mapping, let it. + */ + if (!file) + return prot | PROT_EXEC; + /* + * ditto if it's not on noexec mount, except that on !MMU we need + * BDI_CAP_EXEC_MMAP (== VM_MAYEXEC) in this case + */ + if (!(file->f_path.mnt->mnt_flags & MNT_NOEXEC)) { +#ifndef CONFIG_MMU + unsigned long caps = 0; + struct address_space *mapping = file->f_mapping; + if (mapping && mapping->backing_dev_info) + caps = mapping->backing_dev_info->capabilities; + if (!(caps & BDI_CAP_EXEC_MAP)) + return prot; +#endif + return prot | PROT_EXEC; + } + /* anything on noexec mount won't get PROT_EXEC */ + return prot; +} + +int security_mmap_file(struct file *file, unsigned long prot, + unsigned long flags) +{ + int ret; + ret = security_ops->mmap_file(file, prot, + mmap_prot(file, prot), flags); + if (ret) + return ret; + return ima_file_mmap(file, prot); +} + +int security_mmap_addr(unsigned long addr) +{ + return security_ops->mmap_addr(addr); +} + +int security_file_mprotect(struct vm_area_struct *vma, unsigned long reqprot, + unsigned long prot) +{ + return security_ops->file_mprotect(vma, reqprot, prot); +} + +int security_file_lock(struct file *file, unsigned int cmd) +{ + return security_ops->file_lock(file, cmd); +} + +int security_file_fcntl(struct file *file, unsigned int cmd, unsigned long arg) +{ + return security_ops->file_fcntl(file, cmd, arg); +} + +int security_file_set_fowner(struct file *file) +{ + return security_ops->file_set_fowner(file); +} + +int security_file_send_sigiotask(struct task_struct *tsk, + struct fown_struct *fown, int sig) +{ + return security_ops->file_send_sigiotask(tsk, fown, sig); +} + +int security_file_receive(struct file *file) +{ + return security_ops->file_receive(file); +} + +int security_file_open(struct file *file, const struct cred *cred) +{ + int ret; + + ret = security_ops->file_open(file, cred); + if (ret) + return ret; + + return fsnotify_perm(file, MAY_OPEN); +} + +int security_task_create(unsigned long clone_flags) +{ + return security_ops->task_create(clone_flags); +} + +void security_task_free(struct task_struct *task) +{ +#ifdef CONFIG_SECURITY_YAMA_STACKED + yama_task_free(task); +#endif + security_ops->task_free(task); +} + +int security_cred_alloc_blank(struct cred *cred, gfp_t gfp) +{ + return security_ops->cred_alloc_blank(cred, gfp); +} + +void security_cred_free(struct cred *cred) +{ + security_ops->cred_free(cred); +} + +int security_prepare_creds(struct cred *new, const struct cred *old, gfp_t gfp) +{ + return security_ops->cred_prepare(new, old, gfp); +} + +void security_transfer_creds(struct cred *new, const struct cred *old) +{ + security_ops->cred_transfer(new, old); +} + +int security_kernel_act_as(struct cred *new, u32 secid) +{ + return security_ops->kernel_act_as(new, secid); +} + +int security_kernel_create_files_as(struct cred *new, struct inode *inode) +{ + return security_ops->kernel_create_files_as(new, inode); +} + +int security_kernel_module_request(char *kmod_name) +{ + return security_ops->kernel_module_request(kmod_name); +} + +int security_kernel_module_from_file(struct file *file) +{ + int ret; + + ret = security_ops->kernel_module_from_file(file); + if (ret) + return ret; + return ima_module_check(file); +} + +int security_task_fix_setuid(struct cred *new, const struct cred *old, + int flags) +{ + return security_ops->task_fix_setuid(new, old, flags); +} + +int security_task_setpgid(struct task_struct *p, pid_t pgid) +{ + return security_ops->task_setpgid(p, pgid); +} + +int security_task_getpgid(struct task_struct *p) +{ + return security_ops->task_getpgid(p); +} + +int security_task_getsid(struct task_struct *p) +{ + return security_ops->task_getsid(p); +} + +void security_task_getsecid(struct task_struct *p, u32 *secid) +{ + security_ops->task_getsecid(p, secid); +} +EXPORT_SYMBOL(security_task_getsecid); + +int security_task_setnice(struct task_struct *p, int nice) +{ + return security_ops->task_setnice(p, nice); +} + +int security_task_setioprio(struct task_struct *p, int ioprio) +{ + return security_ops->task_setioprio(p, ioprio); +} + +int security_task_getioprio(struct task_struct *p) +{ + return security_ops->task_getioprio(p); +} + +int security_task_setrlimit(struct task_struct *p, unsigned int resource, + struct rlimit *new_rlim) +{ + return security_ops->task_setrlimit(p, resource, new_rlim); +} + +int security_task_setscheduler(struct task_struct *p) +{ + return security_ops->task_setscheduler(p); +} + +int security_task_getscheduler(struct task_struct *p) +{ + return security_ops->task_getscheduler(p); +} + +int security_task_movememory(struct task_struct *p) +{ + return security_ops->task_movememory(p); +} + +int security_task_kill(struct task_struct *p, struct siginfo *info, + int sig, u32 secid) +{ + return security_ops->task_kill(p, info, sig, secid); +} + +int security_task_wait(struct task_struct *p) +{ + return security_ops->task_wait(p); +} + +int security_task_prctl(int option, unsigned long arg2, unsigned long arg3, + unsigned long arg4, unsigned long arg5) +{ +#ifdef CONFIG_SECURITY_YAMA_STACKED + int rc; + rc = yama_task_prctl(option, arg2, arg3, arg4, arg5); + if (rc != -ENOSYS) + return rc; +#endif + return security_ops->task_prctl(option, arg2, arg3, arg4, arg5); +} + +void security_task_to_inode(struct task_struct *p, struct inode *inode) +{ + security_ops->task_to_inode(p, inode); +} + +int security_ipc_permission(struct kern_ipc_perm *ipcp, short flag) +{ + return security_ops->ipc_permission(ipcp, flag); +} + +void security_ipc_getsecid(struct kern_ipc_perm *ipcp, u32 *secid) +{ + security_ops->ipc_getsecid(ipcp, secid); +} + +int security_msg_msg_alloc(struct msg_msg *msg) +{ + return security_ops->msg_msg_alloc_security(msg); +} + +void security_msg_msg_free(struct msg_msg *msg) +{ + security_ops->msg_msg_free_security(msg); +} + +int security_msg_queue_alloc(struct msg_queue *msq) +{ + return security_ops->msg_queue_alloc_security(msq); +} + +void security_msg_queue_free(struct msg_queue *msq) +{ + security_ops->msg_queue_free_security(msq); +} + +int security_msg_queue_associate(struct msg_queue *msq, int msqflg) +{ + return security_ops->msg_queue_associate(msq, msqflg); +} + +int security_msg_queue_msgctl(struct msg_queue *msq, int cmd) +{ + return security_ops->msg_queue_msgctl(msq, cmd); +} + +int security_msg_queue_msgsnd(struct msg_queue *msq, + struct msg_msg *msg, int msqflg) +{ + return security_ops->msg_queue_msgsnd(msq, msg, msqflg); +} + +int security_msg_queue_msgrcv(struct msg_queue *msq, struct msg_msg *msg, + struct task_struct *target, long type, int mode) +{ + return security_ops->msg_queue_msgrcv(msq, msg, target, type, mode); +} + +int security_shm_alloc(struct shmid_kernel *shp) +{ + return security_ops->shm_alloc_security(shp); +} + +void security_shm_free(struct shmid_kernel *shp) +{ + security_ops->shm_free_security(shp); +} + +int security_shm_associate(struct shmid_kernel *shp, int shmflg) +{ + return security_ops->shm_associate(shp, shmflg); +} + +int security_shm_shmctl(struct shmid_kernel *shp, int cmd) +{ + return security_ops->shm_shmctl(shp, cmd); +} + +int security_shm_shmat(struct shmid_kernel *shp, char __user *shmaddr, int shmflg) +{ + return security_ops->shm_shmat(shp, shmaddr, shmflg); +} + +int security_sem_alloc(struct sem_array *sma) +{ + return security_ops->sem_alloc_security(sma); +} + +void security_sem_free(struct sem_array *sma) +{ + security_ops->sem_free_security(sma); +} + +int security_sem_associate(struct sem_array *sma, int semflg) +{ + return security_ops->sem_associate(sma, semflg); +} + +int security_sem_semctl(struct sem_array *sma, int cmd) +{ + return security_ops->sem_semctl(sma, cmd); +} + +int security_sem_semop(struct sem_array *sma, struct sembuf *sops, + unsigned nsops, int alter) +{ + return security_ops->sem_semop(sma, sops, nsops, alter); +} + +void security_d_instantiate(struct dentry *dentry, struct inode *inode) +{ + if (unlikely(inode && IS_PRIVATE(inode))) + return; + security_ops->d_instantiate(dentry, inode); +} +EXPORT_SYMBOL(security_d_instantiate); + +int security_getprocattr(struct task_struct *p, char *name, char **value) +{ + return security_ops->getprocattr(p, name, value); +} + +int security_setprocattr(struct task_struct *p, char *name, void *value, size_t size) +{ + return security_ops->setprocattr(p, name, value, size); +} + +int security_netlink_send(struct sock *sk, struct sk_buff *skb) +{ + return security_ops->netlink_send(sk, skb); +} + +int security_secid_to_secctx(u32 secid, char **secdata, u32 *seclen) +{ + return security_ops->secid_to_secctx(secid, secdata, seclen); +} +EXPORT_SYMBOL(security_secid_to_secctx); + +int security_secctx_to_secid(const char *secdata, u32 seclen, u32 *secid) +{ + return security_ops->secctx_to_secid(secdata, seclen, secid); +} +EXPORT_SYMBOL(security_secctx_to_secid); + +void security_release_secctx(char *secdata, u32 seclen) +{ + security_ops->release_secctx(secdata, seclen); +} +EXPORT_SYMBOL(security_release_secctx); + +int security_inode_notifysecctx(struct inode *inode, void *ctx, u32 ctxlen) +{ + return security_ops->inode_notifysecctx(inode, ctx, ctxlen); +} +EXPORT_SYMBOL(security_inode_notifysecctx); + +int security_inode_setsecctx(struct dentry *dentry, void *ctx, u32 ctxlen) +{ + return security_ops->inode_setsecctx(dentry, ctx, ctxlen); +} +EXPORT_SYMBOL(security_inode_setsecctx); + +int security_inode_getsecctx(struct inode *inode, void **ctx, u32 *ctxlen) +{ + return security_ops->inode_getsecctx(inode, ctx, ctxlen); +} +EXPORT_SYMBOL(security_inode_getsecctx); + +#ifdef CONFIG_SECURITY_NETWORK + +int security_unix_stream_connect(struct sock *sock, struct sock *other, struct sock *newsk) +{ + return security_ops->unix_stream_connect(sock, other, newsk); +} +EXPORT_SYMBOL(security_unix_stream_connect); + +int security_unix_may_send(struct socket *sock, struct socket *other) +{ + return security_ops->unix_may_send(sock, other); +} +EXPORT_SYMBOL(security_unix_may_send); + +int security_socket_create(int family, int type, int protocol, int kern) +{ + return security_ops->socket_create(family, type, protocol, kern); +} + +int security_socket_post_create(struct socket *sock, int family, + int type, int protocol, int kern) +{ + return security_ops->socket_post_create(sock, family, type, + protocol, kern); +} + +int security_socket_bind(struct socket *sock, struct sockaddr *address, int addrlen) +{ + return security_ops->socket_bind(sock, address, addrlen); +} + +int security_socket_connect(struct socket *sock, struct sockaddr *address, int addrlen) +{ + return security_ops->socket_connect(sock, address, addrlen); +} + +int security_socket_listen(struct socket *sock, int backlog) +{ + return security_ops->socket_listen(sock, backlog); +} + +int security_socket_accept(struct socket *sock, struct socket *newsock) +{ + return security_ops->socket_accept(sock, newsock); +} + +int security_socket_sendmsg(struct socket *sock, struct msghdr *msg, int size) +{ + return security_ops->socket_sendmsg(sock, msg, size); +} + +int security_socket_recvmsg(struct socket *sock, struct msghdr *msg, + int size, int flags) +{ + return security_ops->socket_recvmsg(sock, msg, size, flags); +} + +int security_socket_getsockname(struct socket *sock) +{ + return security_ops->socket_getsockname(sock); +} + +int security_socket_getpeername(struct socket *sock) +{ + return security_ops->socket_getpeername(sock); +} + +int security_socket_getsockopt(struct socket *sock, int level, int optname) +{ + return security_ops->socket_getsockopt(sock, level, optname); +} + +int security_socket_setsockopt(struct socket *sock, int level, int optname) +{ + return security_ops->socket_setsockopt(sock, level, optname); +} + +int security_socket_shutdown(struct socket *sock, int how) +{ + return security_ops->socket_shutdown(sock, how); +} + +int security_sock_rcv_skb(struct sock *sk, struct sk_buff *skb) +{ + return security_ops->socket_sock_rcv_skb(sk, skb); +} +EXPORT_SYMBOL(security_sock_rcv_skb); + +int security_socket_getpeersec_stream(struct socket *sock, char __user *optval, + int __user *optlen, unsigned len) +{ + return security_ops->socket_getpeersec_stream(sock, optval, optlen, len); +} + +int security_socket_getpeersec_dgram(struct socket *sock, struct sk_buff *skb, u32 *secid) +{ + return security_ops->socket_getpeersec_dgram(sock, skb, secid); +} +EXPORT_SYMBOL(security_socket_getpeersec_dgram); + +int security_sk_alloc(struct sock *sk, int family, gfp_t priority) +{ + return security_ops->sk_alloc_security(sk, family, priority); +} + +void security_sk_free(struct sock *sk) +{ + security_ops->sk_free_security(sk); +} + +void security_sk_clone(const struct sock *sk, struct sock *newsk) +{ + security_ops->sk_clone_security(sk, newsk); +} +EXPORT_SYMBOL(security_sk_clone); + +void security_sk_classify_flow(struct sock *sk, struct flowi *fl) +{ + security_ops->sk_getsecid(sk, &fl->flowi_secid); +} +EXPORT_SYMBOL(security_sk_classify_flow); + +void security_req_classify_flow(const struct request_sock *req, struct flowi *fl) +{ + security_ops->req_classify_flow(req, fl); +} +EXPORT_SYMBOL(security_req_classify_flow); + +void security_sock_graft(struct sock *sk, struct socket *parent) +{ + security_ops->sock_graft(sk, parent); +} +EXPORT_SYMBOL(security_sock_graft); + +int security_inet_conn_request(struct sock *sk, + struct sk_buff *skb, struct request_sock *req) +{ + return security_ops->inet_conn_request(sk, skb, req); +} +EXPORT_SYMBOL(security_inet_conn_request); + +void security_inet_csk_clone(struct sock *newsk, + const struct request_sock *req) +{ + security_ops->inet_csk_clone(newsk, req); +} + +void security_inet_conn_established(struct sock *sk, + struct sk_buff *skb) +{ + security_ops->inet_conn_established(sk, skb); +} + +int security_secmark_relabel_packet(u32 secid) +{ + return security_ops->secmark_relabel_packet(secid); +} +EXPORT_SYMBOL(security_secmark_relabel_packet); + +void security_secmark_refcount_inc(void) +{ + security_ops->secmark_refcount_inc(); +} +EXPORT_SYMBOL(security_secmark_refcount_inc); + +void security_secmark_refcount_dec(void) +{ + security_ops->secmark_refcount_dec(); +} +EXPORT_SYMBOL(security_secmark_refcount_dec); + +int security_tun_dev_alloc_security(void **security) +{ + return security_ops->tun_dev_alloc_security(security); +} +EXPORT_SYMBOL(security_tun_dev_alloc_security); + +void security_tun_dev_free_security(void *security) +{ + security_ops->tun_dev_free_security(security); +} +EXPORT_SYMBOL(security_tun_dev_free_security); + +int security_tun_dev_create(void) +{ + return security_ops->tun_dev_create(); +} +EXPORT_SYMBOL(security_tun_dev_create); + +int security_tun_dev_attach_queue(void *security) +{ + return security_ops->tun_dev_attach_queue(security); +} +EXPORT_SYMBOL(security_tun_dev_attach_queue); + +int security_tun_dev_attach(struct sock *sk, void *security) +{ + return security_ops->tun_dev_attach(sk, security); +} +EXPORT_SYMBOL(security_tun_dev_attach); + +int security_tun_dev_open(void *security) +{ + return security_ops->tun_dev_open(security); +} +EXPORT_SYMBOL(security_tun_dev_open); + +void security_skb_owned_by(struct sk_buff *skb, struct sock *sk) +{ + security_ops->skb_owned_by(skb, sk); +} + +#endif /* CONFIG_SECURITY_NETWORK */ + +#ifdef CONFIG_SECURITY_NETWORK_XFRM + +int security_xfrm_policy_alloc(struct xfrm_sec_ctx **ctxp, struct xfrm_user_sec_ctx *sec_ctx) +{ + return security_ops->xfrm_policy_alloc_security(ctxp, sec_ctx); +} +EXPORT_SYMBOL(security_xfrm_policy_alloc); + +int security_xfrm_policy_clone(struct xfrm_sec_ctx *old_ctx, + struct xfrm_sec_ctx **new_ctxp) +{ + return security_ops->xfrm_policy_clone_security(old_ctx, new_ctxp); +} + +void security_xfrm_policy_free(struct xfrm_sec_ctx *ctx) +{ + security_ops->xfrm_policy_free_security(ctx); +} +EXPORT_SYMBOL(security_xfrm_policy_free); + +int security_xfrm_policy_delete(struct xfrm_sec_ctx *ctx) +{ + return security_ops->xfrm_policy_delete_security(ctx); +} + +int security_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *sec_ctx) +{ + return security_ops->xfrm_state_alloc_security(x, sec_ctx, 0); +} +EXPORT_SYMBOL(security_xfrm_state_alloc); + +int security_xfrm_state_alloc_acquire(struct xfrm_state *x, + struct xfrm_sec_ctx *polsec, u32 secid) +{ + if (!polsec) + return 0; + /* + * We want the context to be taken from secid which is usually + * from the sock. + */ + return security_ops->xfrm_state_alloc_security(x, NULL, secid); +} + +int security_xfrm_state_delete(struct xfrm_state *x) +{ + return security_ops->xfrm_state_delete_security(x); +} +EXPORT_SYMBOL(security_xfrm_state_delete); + +void security_xfrm_state_free(struct xfrm_state *x) +{ + security_ops->xfrm_state_free_security(x); +} + +int security_xfrm_policy_lookup(struct xfrm_sec_ctx *ctx, u32 fl_secid, u8 dir) +{ + return security_ops->xfrm_policy_lookup(ctx, fl_secid, dir); +} + +int security_xfrm_state_pol_flow_match(struct xfrm_state *x, + struct xfrm_policy *xp, + const struct flowi *fl) +{ + return security_ops->xfrm_state_pol_flow_match(x, xp, fl); +} + +int security_xfrm_decode_session(struct sk_buff *skb, u32 *secid) +{ + return security_ops->xfrm_decode_session(skb, secid, 1); +} + +void security_skb_classify_flow(struct sk_buff *skb, struct flowi *fl) +{ + int rc = security_ops->xfrm_decode_session(skb, &fl->flowi_secid, 0); + + BUG_ON(rc); +} +EXPORT_SYMBOL(security_skb_classify_flow); + +#endif /* CONFIG_SECURITY_NETWORK_XFRM */ + +#ifdef CONFIG_KEYS + +int security_key_alloc(struct key *key, const struct cred *cred, + unsigned long flags) +{ + return security_ops->key_alloc(key, cred, flags); +} + +void security_key_free(struct key *key) +{ + security_ops->key_free(key); +} + +int security_key_permission(key_ref_t key_ref, + const struct cred *cred, key_perm_t perm) +{ + return security_ops->key_permission(key_ref, cred, perm); +} + +int security_key_getsecurity(struct key *key, char **_buffer) +{ + return security_ops->key_getsecurity(key, _buffer); +} + +#endif /* CONFIG_KEYS */ + +#ifdef CONFIG_AUDIT + +int security_audit_rule_init(u32 field, u32 op, char *rulestr, void **lsmrule) +{ + return security_ops->audit_rule_init(field, op, rulestr, lsmrule); +} + +int security_audit_rule_known(struct audit_krule *krule) +{ + return security_ops->audit_rule_known(krule); +} + +void security_audit_rule_free(void *lsmrule) +{ + security_ops->audit_rule_free(lsmrule); +} + +int security_audit_rule_match(u32 secid, u32 field, u32 op, void *lsmrule, + struct audit_context *actx) +{ + return security_ops->audit_rule_match(secid, field, op, lsmrule, actx); +} + +#endif /* CONFIG_AUDIT */