O_TMPFILE is actually __O_TMPFILE|O_DIRECTORY. This means that the old
fast-path check for RESOLVE_CACHED would reject all users passing
O_DIRECTORY with -EAGAIN, when in fact the intended test was to check
for __O_TMPFILE.

Cc: stable@vger.kernel.org # v5.12+
Fixes: 99668f618062 ("fs: expose LOOKUP_CACHED through openat2() RESOLVE_CACHED")
Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>
Message-Id: <20230806-resolve_cached-o_tmpfile-v1-1-7ba16308465e@cyphar.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>

Pull fsnotify updates from Jan Kara:

 - Support for fanotify events returning file handles for filesystems
   not exportable via NFS

 - Improved error handling exportfs functions

 - Add missing FS_OPEN events when unusual open helpers are used

* tag 'fsnotify_for_v6.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
  fsnotify: move fsnotify_open() hook into do_dentry_open()
  exportfs: check for error return value from exportfs_encode_*()
  fanotify: support reporting non-decodeable file handles
  exportfs: allow exporting non-decodeable file handles to userspace
  exportfs: add explicit flag to request non-decodeable file handles
  exportfs: change connectable argument to bit flags

Pull vfs file handling updates from Christian Brauner:
 "This contains Amir's work to fix a long-standing problem where an
  unprivileged overlayfs mount can be used to avoid fanotify permission
  events that were requested for an inode or superblock on the
  underlying filesystem.

  Some background about files opened in overlayfs. If a file is opened
  in overlayfs @file->f_path will refer to a "fake" path. What this
  means is that while @file->f_inode will refer to inode of the
  underlying layer, @file->f_path refers to an overlayfs
  {dentry,vfsmount} pair. The reasons for doing this are out of scope
  here but it is the reason why the vfs has been providing the
  open_with_fake_path() helper for overlayfs for very long time now. So
  nothing new here.

  This is for sure not very elegant and everyone including the overlayfs
  maintainers agree. Improving this significantly would involve more
  fragile and potentially rather invasive changes.

  In various codepaths access to the path of the underlying filesystem
  is needed for such hybrid file. The best example is fsnotify where
  this becomes security relevant. Passing the overlayfs
  @file->f_path->dentry will cause fsnotify to skip generating fsnotify
  events registered on the underlying inode or superblock.

  To fix this we extend the vfs provided open_with_fake_path() concept
  for overlayfs to create a backing file container that holds the real
  path and to expose a helper that can be used by relevant callers to
  get access to the path of the underlying filesystem through the new
  file_real_path() helper. This pattern is similar to what we do in
  d_real() and d_real_inode().

  The first beneficiary is fsnotify and fixes the security sensitive
  problem mentioned above.

  There's a couple of nice cleanups included as well.

  Over time, the old open_with_fake_path() helper added specifically for
  overlayfs a long time ago started to get used in other places such as
  cachefiles. Even though cachefiles have nothing to do with hybrid
  files.

  The only reason cachefiles used that concept was that files opened
  with open_with_fake_path() aren't charged against the caller's open
  file limit by raising FMODE_NOACCOUNT. It's just mere coincidence that
  both overlayfs and cachefiles need to ensure to not overcharge the
  caller for their internal open calls.

  So this work disentangles FMODE_NOACCOUNT use cases and backing file
  use-cases by adding the FMODE_BACKING flag which indicates that the
  file can be used to retrieve the backing file of another filesystem.
  (Fyi, Jens will be sending you a really nice cleanup from Christoph
  that gets rid of 3 FMODE_* flags otherwise this would be the last
  fmode_t bit we'd be using.)

  So now overlayfs becomes the sole user of the renamed
  open_with_fake_path() helper which is now named backing_file_open().
  For internal kernel users such as cachefiles that are only interested
  in FMODE_NOACCOUNT but not in FMODE_BACKING we add a new
  kernel_file_open() helper which opens a file without being charged
  against the caller's open file limit. All new helpers are properly
  documented and clearly annotated to mention their special uses.

  We also rename vfs_tmpfile_open() to kernel_tmpfile_open() to clearly
  distinguish it from vfs_tmpfile() and align it the other kernel_*()
  internal helpers"

* tag 'v6.5/vfs.file' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
  ovl: enable fsnotify events on underlying real files
  fs: use backing_file container for internal files with "fake" f_path
  fs: move kmem_cache_zalloc() into alloc_empty_file*() helpers
  fs: use a helper for opening kernel internal files
  fs: rename {vfs,kernel}_tmpfile_open()

Overlayfs uses open_with_fake_path() to allocate internal kernel files,
with a "fake" path - whose f_path is not on the same fs as f_inode.

Allocate a container struct backing_file for those internal files, that
is used to hold the "fake" ovl path along with the real path.

backing_file_real_path() can be used to access the stored real path.

Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Message-Id: <20230615112229.2143178-5-amir73il@gmail.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>

cachefiles uses kernel_open_tmpfile() to open kernel internal tmpfile
without accounting for nr_files.

cachefiles uses open_with_fake_path() for the same reason without the
need for a fake path.

Fork open_with_fake_path() to kernel_file_open() which only does the
noaccount part and use it in cachefiles.

Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Message-Id: <20230615112229.2143178-3-amir73il@gmail.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>

fsnotify_open() hook is called only from high level system calls
context and not called for the very many helpers to open files.

This may makes sense for many of the special file open cases, but it is
inconsistent with fsnotify_close() hook that is called for every last
fput() of on a file object with FMODE_OPENED.

As a result, it is possible to observe ACCESS, MODIFY and CLOSE events
without ever observing an OPEN event.

Fix this inconsistency by replacing all the fsnotify_open() hooks with
a single hook inside do_dentry_open().

If there are special cases that would like to opt-out of the possible
overhead of fsnotify() call in fsnotify_open(), they would probably also
want to avoid the overhead of fsnotify() call in the rest of the fsnotify
hooks, so they should be opening that file with the __FMODE_NONOTIFY flag.

However, in the majority of those cases, the s_fsnotify_connectors
optimization in fsnotify_parent() would be sufficient to avoid the
overhead of fsnotify() call anyway.

Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Message-Id: <20230611122429.1499617-1-amir73il@gmail.com>

Use __FMODE_NONOTIFY instead of FMODE_NONOTIFY to fixes
the following sparce warnings:
fs/overlayfs/file.c:48:37: sparse: warning: restricted fmode_t degrades to integer
fs/overlayfs/file.c:128:13: sparse: warning: restricted fmode_t degrades to integer
fs/open.c:1159:21: sparse: warning: restricted fmode_t degrades to integer

Signed-off-by: Min-Hua Chen <minhuadotchen@gmail.com>
Message-Id: <20230502232210.119063-1-minhuadotchen@gmail.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>

fs/open.c: In functions 'setattr_vfsuid' and 'setattr_vfsgid':
 warning: Function parameter or member 'attr' not described
 - Fix warning by removing kernel-doc for these as they are static
   inline functions and not required to be exposed via kernel-doc.

fs/open.c:
 warning: Excess function parameter 'opened' description in 'finish_open'
 warning: Excess function parameter 'cred' description in 'vfs_open'
 - Fix by removing the parameters from the kernel-doc as they are no
   longer required by the function.

Signed-off-by: Anuradha Weeraman <anuradha@debian.org>
Message-Id: <20230506182928.384105-1-anuradha@debian.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>

After a couple of years and multiple LTS releases we received a report
that the behavior of O_DIRECTORY | O_CREAT changed starting with v5.7.

On kernels prior to v5.7 combinations of O_DIRECTORY, O_CREAT, O_EXCL
had the following semantics:

(1) open("/tmp/d", O_DIRECTORY | O_CREAT)
    * d doesn't exist:                create regular file
    * d exists and is a regular file: ENOTDIR
    * d exists and is a directory:    EISDIR

(2) open("/tmp/d", O_DIRECTORY | O_CREAT | O_EXCL)
    * d doesn't exist:                create regular file
    * d exists and is a regular file: EEXIST
    * d exists and is a directory:    EEXIST

(3) open("/tmp/d", O_DIRECTORY | O_EXCL)
    * d doesn't exist:                ENOENT
    * d exists and is a regular file: ENOTDIR
    * d exists and is a directory:    open directory

On kernels since to v5.7 combinations of O_DIRECTORY, O_CREAT, O_EXCL
have the following semantics:

(1) open("/tmp/d", O_DIRECTORY | O_CREAT)
    * d doesn't exist:                ENOTDIR (create regular file)
    * d exists and is a regular file: ENOTDIR
    * d exists and is a directory:    EISDIR

(2) open("/tmp/d", O_DIRECTORY | O_CREAT | O_EXCL)
    * d doesn't exist:                ENOTDIR (create regular file)
    * d exists and is a regular file: EEXIST
    * d exists and is a directory:    EEXIST

(3) open("/tmp/d", O_DIRECTORY | O_EXCL)
    * d doesn't exist:                ENOENT
    * d exists and is a regular file: ENOTDIR
    * d exists and is a directory:    open directory

This is a fairly substantial semantic change that userspace didn't
notice until Pedro took the time to deliberately figure out corner
cases. Since no one noticed this breakage we can somewhat safely assume
that O_DIRECTORY | O_CREAT combinations are likely unused.

The v5.7 breakage is especially weird because while ENOTDIR is returned
indicating failure a regular file is actually created. This doesn't make
a lot of sense.

Time was spent finding potential users of this combination. Searching on
codesearch.debian.net showed that codebases often express semantical
expectations about O_DIRECTORY | O_CREAT which are completely contrary
to what our code has done and currently does.

The expectation often is that this particular combination would create
and open a directory. This suggests users who tried to use that
combination would stumble upon the counterintuitive behavior no matter
if pre-v5.7 or post v5.7 and quickly realize neither semantics give them
what they want. For some examples see the code examples in [1] to [3]
and the discussion in [4].

There are various ways to address this issue. The lazy/simple option
would be to restore the pre-v5.7 behavior and to just live with that bug
forever. But since there's a real chance that the O_DIRECTORY | O_CREAT
quirk isn't relied upon we should try to get away with murder(ing bad
semantics) first. If we need to Frankenstein pre-v5.7 behavior later so
be it.

So let's simply return EINVAL categorically for O_DIRECTORY | O_CREAT
combinations. In addition to cleaning up the old bug this also opens up
the possiblity to make that flag combination do something more intuitive
in the future.

Starting with this commit the following semantics apply:

(1) open("/tmp/d", O_DIRECTORY | O_CREAT)
    * d doesn't exist:                EINVAL
    * d exists and is a regular file: EINVAL
    * d exists and is a directory:    EINVAL

(2) open("/tmp/d", O_DIRECTORY | O_CREAT | O_EXCL)
    * d doesn't exist:                EINVAL
    * d exists and is a regular file: EINVAL
    * d exists and is a directory:    EINVAL

(3) open("/tmp/d", O_DIRECTORY | O_EXCL)
    * d doesn't exist:                ENOENT
    * d exists and is a regular file: ENOTDIR
    * d exists and is a directory:    open directory

One additional note, O_TMPFILE is implemented as:

    #define __O_TMPFILE    020000000
    #define O_TMPFILE      (__O_TMPFILE | O_DIRECTORY)
    #define O_TMPFILE_MASK (__O_TMPFILE | O_DIRECTORY | O_CREAT)

For older kernels it was important to return an explicit error when
O_TMPFILE wasn't supported. So O_TMPFILE requires that O_DIRECTORY is
raised alongside __O_TMPFILE. It also enforced that O_CREAT wasn't
specified. Since O_DIRECTORY | O_CREAT could be used to create a regular
allowing that combination together with __O_TMPFILE would've meant that
false positives were possible, i.e., that a regular file was created
instead of a O_TMPFILE. This could've been used to trick userspace into
thinking it operated on a O_TMPFILE when it wasn't.

Now that we block O_DIRECTORY | O_CREAT completely the check for O_CREAT
in the __O_TMPFILE branch via if ((flags & O_TMPFILE_MASK) != O_TMPFILE)
can be dropped. Instead we can simply check verify that O_DIRECTORY is
raised via if (!(flags & O_DIRECTORY)) and explain this in two comments.

As Aleksa pointed out O_PATH is unaffected by this change since it
always returned EINVAL if O_CREAT was specified - with or without
O_DIRECTORY.

Link: https://lore.kernel.org/lkml/20230320071442.172228-1-pedro.falcato@gmail.com
Link: https://sources.debian.org/src/flatpak/1.14.4-1/subprojects/libglnx/glnx-dirfd.c/?hl=324#L324 [1]
Link: https://sources.debian.org/src/flatpak-builder/1.2.3-1/subprojects/libglnx/glnx-shutil.c/?hl=251#L251 [2]
Link: https://sources.debian.org/src/ostree/2022.7-2/libglnx/glnx-dirfd.c/?hl=324#L324 [3]
Link: https://www.openwall.com/lists/oss-security/2014/11/26/14 [4]
Reported-by: Pedro Falcato <pedro.falcato@gmail.com>
Cc: Aleksa Sarai <cyphar@cyphar.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>

access(2) remains commonly used, for example on exec:
access("/etc/ld.so.preload", R_OK)

or when running gcc: strace -c gcc empty.c

  % time     seconds  usecs/call     calls    errors syscall
  ------ ----------- ----------- --------- --------- ----------------
    0.00    0.000000           0        42        26 access

It falls down to do_faccessat without the AT_EACCESS flag, which in turn
results in allocation of new creds in order to modify fsuid/fsgid and
caps.  This is a very expensive process single-threaded and most notably
multi-threaded, with numerous structures getting refed and unrefed on
imminent new cred destruction.

Turns out for typical consumers the resulting creds would be identical
and this can be checked upfront, avoiding the hard work.

An access benchmark plugged into will-it-scale running on Cascade Lake
shows:

    test     proc     before       after
    access1     1    1310582     2908735    (+121%) # distinct files
    access1    24    4716491    63822173   (+1353%) # distinct files
    access2    24    2378041     5370335    (+125%) # same file

The above benchmarks are not integrated into will-it-scale, but can be
found in a pull request:

  https://github.com/antonblanchard/will-it-scale/pull/36/files

Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>