It is possible for corrupted filesystem images to produce very large
block offsets that may wrap when a length is added, and wrongly pass
the buffer size test.
Reported-by: Anatoly Trosinenko <anatoly.trosinenko@gmail.com>
Signed-off-by: Nicolas Pitre <nico@linaro.org>
Cc: stable@vger.kernel.org
fscache_set_key() can incur an out-of-bounds read, reported by KASAN:
BUG: KASAN: slab-out-of-bounds in fscache_alloc_cookie+0x5b3/0x680 [fscache]
Read of size 4 at addr ffff88084ff056d4 by task mount.nfs/32615
and also reported by syzbot at https://lkml.org/lkml/2018/7/8/236
BUG: KASAN: slab-out-of-bounds in fscache_set_key fs/fscache/cookie.c:120 [inline]
BUG: KASAN: slab-out-of-bounds in fscache_alloc_cookie+0x7a9/0x880 fs/fscache/cookie.c:171
Read of size 4 at addr ffff8801d3cc8bb4 by task syz-executor907/4466
This happens for any index_key_len which is not divisible by 4 and is
larger than the size of the inline key, because the code allocates exactly
index_key_len for the key buffer, but the hashing loop is stepping through
it 4 bytes (u32) at a time in the buf[] array.
Fix this by calculating how many u32 buffers we'll need by using
DIV_ROUND_UP, and then using kcalloc() to allocate a precleared allocation
buffer to hold the index_key, then using that same count as the hashing
index limit.
Fixes: ec0328e46d ("fscache: Maintain a catalogue of allocated cookies")
Reported-by: syzbot+a95b989b2dde8e806af8@syzkaller.appspotmail.com
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The inline key in struct rxrpc_cookie is insufficiently initialized,
zeroing only 3 of the 4 slots, therefore an index_key_len between 13 and 15
bytes will end up hashing uninitialized memory because the memcpy only
partially fills the last buf[] element.
Fix this by clearing fscache_cookie objects on allocation rather than using
the slab constructor to initialise them. We're going to pretty much fill
in the entire struct anyway, so bringing it into our dcache writably
shouldn't incur much overhead.
This removes the need to do clearance in fscache_set_key() (where we aren't
doing it correctly anyway).
Also, we don't need to set cookie->key_len in fscache_set_key() as we
already did it in the only caller, so remove that.
Fixes: ec0328e46d ("fscache: Maintain a catalogue of allocated cookies")
Reported-by: syzbot+a95b989b2dde8e806af8@syzkaller.appspotmail.com
Reported-by: Eric Sandeen <sandeen@redhat.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
the victim might've been rmdir'ed just before the lock_rename();
unlike the normal callers, we do not look the source up after the
parents are locked - we know it beforehand and just recheck that it's
still the child of what used to be its parent. Unfortunately,
the check is too weak - we don't spot a dead directory since its
->d_parent is unchanged, dentry is positive, etc. So we sail all
the way to ->rename(), with hosting filesystems _not_ expecting
to be asked renaming an rmdir'ed subdirectory.
The fix is easy, fortunately - the lock on parent is sufficient for
making IS_DEADDIR() on child safe.
Cc: stable@vger.kernel.org
Fixes: 9ae326a690 (CacheFiles: A cache that backs onto a mounted filesystem)
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The recent patch to fix the afs_server struct leak didn't actually fix the
bug, but rather fixed some of the symptoms. The problem is that an
asynchronous call that holds a resource pointed to by call->reply[0] will
find the pointer cleared in the call destructor, thereby preventing the
resource from being cleaned up.
In the case of the server record leak, the afs_fs_get_capabilities()
function in devel code sets up a call with reply[0] pointing at the server
record that should be altered when the result is obtained, but this was
being cleared before the destructor was called, so the put in the
destructor does nothing and the record is leaked.
Commit f014ffb025 removed the additional ref obtained by
afs_install_server(), but the removal of this ref is actually used by the
garbage collector to mark a server record as being defunct after the record
has expired through lack of use.
The offending clearance of call->reply[0] upon completion in
afs_process_async_call() has been there from the origin of the code, but
none of the asynchronous calls actually use that pointer currently, so it
should be safe to remove (note that synchronous calls don't involve this
function).
Fix this by the following means:
(1) Revert commit f014ffb025.
(2) Remove the clearance of reply[0] from afs_process_async_call().
Without this, afs_manage_servers() will suffer an assertion failure if it
sees a server record that didn't get used because the usage count is not 1.
Fixes: f014ffb025 ("afs: Fix afs_server struct leak")
Fixes: 08e0e7c82e ("[AF_RXRPC]: Make the in-kernel AFS filesystem use AF_RXRPC.")
Signed-off-by: David Howells <dhowells@redhat.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* Fix a livelock in dax_layout_busy_page() present since v4.18. The
lockup triggers when truncating an actively mapped huge page out of a
mapping pinned for direct-I/O.
* Fix mprotect() clobbers of _PAGE_DEVMAP. Broken since v4.5 mprotect()
clears this flag that is needed to communicate the liveness of device
pages to the get_user_pages() path.
-----BEGIN PGP SIGNATURE-----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=VXnU
-----END PGP SIGNATURE-----
Merge tag 'libnvdimm-fixes-4.19-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm
Dan writes:
"libnvdimm/dax 4.19-rc8
* Fix a livelock in dax_layout_busy_page() present since v4.18. The
lockup triggers when truncating an actively mapped huge page out of
a mapping pinned for direct-I/O.
* Fix mprotect() clobbers of _PAGE_DEVMAP. Broken since v4.5
mprotect() clears this flag that is needed to communicate the
liveness of device pages to the get_user_pages() path."
* tag 'libnvdimm-fixes-4.19-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
mm: Preserve _PAGE_DEVMAP across mprotect() calls
filesystem-dax: Fix dax_layout_busy_page() livelock
ubifs_assert() is not WARN_ON(), so we have to invert
the checks.
Randy faced this warning with UBIFS being a module, since
most users use UBIFS as builtin because UBIFS is the rootfs
nobody noticed so far. :-(
Including me.
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Fixes: 54169ddd38 ("ubifs: Turn two ubifs_assert() into a WARN_ON()")
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On non-preempt kernels this loop can take a long time (more than 50 ticks)
processing through entries.
Link: http://lkml.kernel.org/r/20181010172623.57033-1-khazhy@google.com
Signed-off-by: Khazhismel Kumykov <khazhy@google.com>
Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fix the following compile warning:
fs/ocfs2/dlmglue.c:99:30: warning: lockdep_keys defined but not used [-Wunused-variable]
static struct lock_class_key lockdep_keys[OCFS2_NUM_LOCK_TYPES];
Link: http://lkml.kernel.org/r/1536938148-32110-1-git-send-email-zhongjiang@huawei.com
Signed-off-by: zhong jiang <zhongjiang@huawei.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fix a leak of afs_server structs. The routine that installs them in the
various lookup lists and trees gets a ref on leaving the function, whether
it added the server or a server already exists. It shouldn't increment
the refcount if it added the server.
The effect of this that "rmmod kafs" will hang waiting for the leaked
server to become unused.
Fixes: d2ddc776a4 ("afs: Overhaul volume and server record caching and fileserver rotation")
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It turns out that the fix in commit 6636c3cc56 is bad; the assertion
that the iomap code no longer creates buffer heads is incorrect for
filesystems that set the IOMAP_F_BUFFER_HEAD flag.
Instead, what's happening is that gfs2_iomap_begin_write treats all
files that have the jdata flag set as journaled files, which is
incorrect as long as those files are inline ("stuffed"). We're handling
stuffed files directly via the page cache, which is why we ended up with
pages without buffer heads in gfs2_page_add_databufs.
Fix this by handling stuffed journaled files correctly in
gfs2_iomap_begin_write.
This reverts commit 6636c3cc5690c11631e6366cf9a28fb99c8b25bb.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Access to the list of cells by /proc/net/afs/cells has a couple of
problems:
(1) It should be checking against SEQ_START_TOKEN for the keying the
header line.
(2) It's only holding the RCU read lock, so it can't just walk over the
list without following the proper RCU methods.
Fix these by using an hlist instead of an ordinary list and using the
appropriate accessor functions to follow it with RCU.
Since the code that adds a cell to the list must also necessarily change,
sort the list on insertion whilst we're at it.
Fixes: 989782dcdc ("afs: Overhaul cell database management")
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Update for 4.19-rc7 to fix numerous file clone and deduplication issues.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=yj1P
-----END PGP SIGNATURE-----
Merge tag 'xfs-fixes-for-4.19-rc7' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux
Dave writes:
"xfs: fixes for 4.19-rc7
Update for 4.19-rc7 to fix numerous file clone and deduplication issues."
* tag 'xfs-fixes-for-4.19-rc7' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
xfs: fix data corruption w/ unaligned reflink ranges
xfs: fix data corruption w/ unaligned dedupe ranges
xfs: update ctime and remove suid before cloning files
xfs: zero posteof blocks when cloning above eof
xfs: refactor clonerange preparation into a separate helper
Commit 64bc06bb32 broke buffered writes to journaled files (chattr
+j): we'll try to journal the buffer heads of the page being written to
in gfs2_iomap_journaled_page_done. However, the iomap code no longer
creates buffer heads, so we'll BUG() in gfs2_page_add_databufs. Fix
that by creating buffer heads ourself when needed.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
In the presence of multi-order entries the typical
pagevec_lookup_entries() pattern may loop forever:
while (index < end && pagevec_lookup_entries(&pvec, mapping, index,
min(end - index, (pgoff_t)PAGEVEC_SIZE),
indices)) {
...
for (i = 0; i < pagevec_count(&pvec); i++) {
index = indices[i];
...
}
index++; /* BUG */
}
The loop updates 'index' for each index found and then increments to the
next possible page to continue the lookup. However, if the last entry in
the pagevec is multi-order then the next possible page index is more
than 1 page away. Fix this locally for the filesystem-dax case by
checking for dax-multi-order entries. Going forward new users of
multi-order entries need to be similarly careful, or we need a generic
way to report the page increment in the radix iterator.
Fixes: 5fac7408d8 ("mm, fs, dax: handle layout changes to pinned dax...")
Cc: <stable@vger.kernel.org>
Cc: Ross Zwisler <zwisler@kernel.org>
Cc: Matthew Wilcox <willy@infradead.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
When reflinking sub-file ranges, a data corruption can occur when
the source file range includes a partial EOF block. This shares the
unknown data beyond EOF into the second file at a position inside
EOF, exposing stale data in the second file.
XFS only supports whole block sharing, but we still need to
support whole file reflink correctly. Hence if the reflink
request includes the last block of the souce file, only proceed with
the reflink operation if it lands at or past the destination file's
current EOF. If it lands within the destination file EOF, reject the
entire request with -EINVAL and make the caller go the hard way.
This avoids the data corruption vector, but also avoids disruption
of returning EINVAL to userspace for the common case of whole file
cloning.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
A deduplication data corruption is Exposed by fstests generic/505 on
XFS. It is caused by extending the block match range to include the
partial EOF block, but then allowing unknown data beyond EOF to be
considered a "match" to data in the destination file because the
comparison is only made to the end of the source file. This corrupts
the destination file when the source extent is shared with it.
XFS only supports whole block dedupe, but we still need to appear to
support whole file dedupe correctly. Hence if the dedupe request
includes the last block of the souce file, don't include it in the
actual XFS dedupe operation. If the rest of the range dedupes
successfully, then report the partial last block as deduped, too, so
that userspace sees it as a successful dedupe rather than return
EINVAL because we can't dedupe unaligned blocks.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
In dlm_init_lockres() we access and modify res->tracking and
dlm->tracking_list without holding dlm->track_lock. This can cause list
corruptions and can end up in kernel panic.
Fix this by locking res->tracking and dlm->tracking_list with
dlm->track_lock instead of dlm->spinlock.
Link: http://lkml.kernel.org/r/1529951192-4686-1-git-send-email-ashish.samant@oracle.com
Signed-off-by: Ashish Samant <ashish.samant@oracle.com>
Reviewed-by: Changwei Ge <ge.changwei@h3c.com>
Acked-by: Joseph Qi <jiangqi903@gmail.com>
Acked-by: Jun Piao <piaojun@huawei.com>
Cc: Mark Fasheh <mark@fasheh.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Cc: Changwei Ge <ge.changwei@h3c.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently, you can use /proc/self/task/*/stack to cause a stack walk on
a task you control while it is running on another CPU. That means that
the stack can change under the stack walker. The stack walker does
have guards against going completely off the rails and into random
kernel memory, but it can interpret random data from your kernel stack
as instruction pointers and stack pointers. This can cause exposure of
kernel stack contents to userspace.
Restrict the ability to inspect kernel stacks of arbitrary tasks to root
in order to prevent a local attacker from exploiting racy stack unwinding
to leak kernel task stack contents. See the added comment for a longer
rationale.
There don't seem to be any users of this userspace API that can't
gracefully bail out if reading from the file fails. Therefore, I believe
that this change is unlikely to break things. In the case that this patch
does end up needing a revert, the next-best solution might be to fake a
single-entry stack based on wchan.
Link: http://lkml.kernel.org/r/20180927153316.200286-1-jannh@google.com
Fixes: 2ec220e27f ("proc: add /proc/*/stack")
Signed-off-by: Jann Horn <jannh@google.com>
Acked-by: Kees Cook <keescook@chromium.org>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Ken Chen <kenchen@google.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Laura Abbott <labbott@redhat.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H . Peter Anvin" <hpa@zytor.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
ocfs2_duplicate_clusters_by_page() may crash if one of the extent's pages
is dirty. When a page has not been written back, it is still in dirty
state. If ocfs2_duplicate_clusters_by_page() is called against the dirty
page, the crash happens.
To fix this bug, we can just unlock the page and wait until the page until
its not dirty.
The following is the backtrace:
kernel BUG at /root/code/ocfs2/refcounttree.c:2961!
[exception RIP: ocfs2_duplicate_clusters_by_page+822]
__ocfs2_move_extent+0x80/0x450 [ocfs2]
? __ocfs2_claim_clusters+0x130/0x250 [ocfs2]
ocfs2_defrag_extent+0x5b8/0x5e0 [ocfs2]
__ocfs2_move_extents_range+0x2a4/0x470 [ocfs2]
ocfs2_move_extents+0x180/0x3b0 [ocfs2]
? ocfs2_wait_for_recovery+0x13/0x70 [ocfs2]
ocfs2_ioctl_move_extents+0x133/0x2d0 [ocfs2]
ocfs2_ioctl+0x253/0x640 [ocfs2]
do_vfs_ioctl+0x90/0x5f0
SyS_ioctl+0x74/0x80
do_syscall_64+0x74/0x140
entry_SYSCALL_64_after_hwframe+0x3d/0xa2
Once we find the page is dirty, we do not wait until it's clean, rather we
use write_one_page() to write it back
Link: http://lkml.kernel.org/r/20180829074740.9438-1-lchen@suse.com
[lchen@suse.com: update comments]
Link: http://lkml.kernel.org/r/20180830075041.14879-1-lchen@suse.com
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Larry Chen <lchen@suse.com>
Acked-by: Changwei Ge <ge.changwei@h3c.com>
Cc: Mark Fasheh <mark@fasheh.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Cc: Joseph Qi <jiangqi903@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-----BEGIN PGP SIGNATURE-----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=4PHD
-----END PGP SIGNATURE-----
Merge tag '4.19-rc6-smb3-fixes' of git://git.samba.org/sfrench/cifs-2.6
Steve writes:
"SMB3 fixes
four small SMB3 fixes: one for stable, the others to address a more
recent regression"
* tag '4.19-rc6-smb3-fixes' of git://git.samba.org/sfrench/cifs-2.6:
smb3: fix lease break problem introduced by compounding
cifs: only wake the thread for the very last PDU in a compound
cifs: add a warning if we try to to dequeue a deleted mid
smb2: fix missing files in root share directory listing
Before cloning into a file, update the ctime and remove sensitive
attributes like suid, just like we'd do for a regular file write.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
When we're reflinking between two files and the destination file range
is well beyond the destination file's EOF marker, zero any posteof
speculative preallocations in the destination file so that we don't
expose stale disk contents. The previous strategy of trying to clear
the preallocations does not work if the destination file has the
PREALLOC flag set.
Uncovered by shared/010.
Reported-by: Zorro Lang <zlang@redhat.com>
Bugzilla-id: https://bugzilla.kernel.org/show_bug.cgi?id=201259
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Refactor all the reflink preparation steps into a separate helper
that we'll use to land all the upcoming fixes for insufficient input
checks.
This rework also moves the invalidation of the destination range to
the prep function so that it is done before the range is remapped.
This ensures that nobody can access the data in range being remapped
until the remap is complete.
[dgc: fix xfs_reflink_remap_prep() return value and caller check to
handle vfs_clone_file_prep_inodes() returning 0 to mean "nothing to
do". ]
[dgc: make sure length changed by vfs_clone_file_prep_inodes() gets
propagated back to XFS code that does the remapping. ]
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQSQHSd0lITzzeNWNm3h3BK/laaZPAUCW7YOCQAKCRDh3BK/laaZ
PAmGAQC1/3+mkfHXqyDx+zv4f7A348ZULW26EWFWwMwlEljWowD+PlRFBgUu5v5c
194mCApZU7hl6o0u07aijjz6m81e6QE=
=dM89
-----END PGP SIGNATURE-----
Merge tag 'ovl-fixes-4.19-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs
Miklos writes:
"overlayfs fixes for 4.19-rc7
This update fixes a couple of regressions in the stacked file update
added in this cycle, as well as some older bugs uncovered by
syzkaller.
There's also one trivial naming change that touches other parts of
the fs subsystem."
* tag 'ovl-fixes-4.19-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs:
ovl: fix format of setxattr debug
ovl: fix access beyond unterminated strings
ovl: make symbol 'ovl_aops' static
vfs: swap names of {do,vfs}_clone_file_range()
ovl: fix freeze protection bypass in ovl_clone_file_range()
ovl: fix freeze protection bypass in ovl_write_iter()
ovl: fix memory leak on unlink of indexed file
Accumlated regression and bug fixes for 4.19-rc6, including:
o make iomap correctly mark dirty pages for sub-page block sizes
o fix regression in handling extent-to-btree format conversion errors
o fix torn log wrap detection for new logs
o various corrupt inode detection fixes
o various delalloc state fixes
o cleanup all the missed transaction cancel cases missed from changes merged
in 4.19-rc1
o fix lockdep false positive on transaction allocation
o fix locking and reference counting on buffer log items
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=5ylA
-----END PGP SIGNATURE-----
Merge tag 'xfs-fixes-for-4.19-rc6' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux
Dave writes:
"XFS fixes for 4.19-rc6
Accumlated regression and bug fixes for 4.19-rc6, including:
o make iomap correctly mark dirty pages for sub-page block sizes
o fix regression in handling extent-to-btree format conversion errors
o fix torn log wrap detection for new logs
o various corrupt inode detection fixes
o various delalloc state fixes
o cleanup all the missed transaction cancel cases missed from changes merged
in 4.19-rc1
o fix lockdep false positive on transaction allocation
o fix locking and reference counting on buffer log items"
* tag 'xfs-fixes-for-4.19-rc6' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
xfs: fix error handling in xfs_bmap_extents_to_btree
iomap: set page dirty after partial delalloc on mkwrite
xfs: remove invalid log recovery first/last cycle check
xfs: validate inode di_forkoff
xfs: skip delalloc COW blocks in xfs_reflink_end_cow
xfs: don't treat unknown di_flags2 as corruption in scrub
xfs: remove duplicated include from alloc.c
xfs: don't bring in extents in xfs_bmap_punch_delalloc_range
xfs: fix transaction leak in xfs_reflink_allocate_cow()
xfs: avoid lockdep false positives in xfs_trans_alloc
xfs: refactor xfs_buf_log_item reference count handling
xfs: clean up xfs_trans_brelse()
xfs: don't unlock invalidated buf on aborted tx commit
xfs: remove last of unnecessary xfs_defer_cancel() callers
xfs: don't crash the vfs on a garbage inline symlink
Format has a typo: it was meant to be "%.*s", not "%*s". But at some point
callers grew nonprintable values as well, so use "%*pE" instead with a
maximized length.
Reported-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Fixes: 3a1e819b4e ("ovl: store file handle of lower inode on copy up")
Cc: <stable@vger.kernel.org> # v4.12
Al writes:
"xattrs regression fix from Andreas; sat in -next for quite a while."
* 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
sysfs: Do not return POSIX ACL xattrs via listxattr
Fixes problem (discovered by Aurelien) introduced by recent commit:
commit b24df3e30c
("cifs: update receive_encrypted_standard to handle compounded responses")
which broke the ability to respond to some lease breaks
(lease breaks being ignored is a problem since can block
server response for duration of the lease break timeout).
Signed-off-by: Steve French <stfrench@microsoft.com>
Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
Reviewed-by: Pavel Shilovsky <pshilov@microsoft.com>
For compounded PDUs we whould only wake the waiting thread for the
very last PDU of the compound.
We do this so that we are guaranteed that the demultiplex_thread will
not process or access any of those MIDs any more once the send/recv
thread starts processing.
Else there is a race where at the end of the send/recv processing we
will try to delete all the mids of the compound. If the multiplex
thread still has other mids to process at this point for this compound
this can lead to an oops.
Needed to fix recent commit:
commit 730928c8f4
("cifs: update smb2_queryfs() to use compounding")
Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Reviewed-by: Pavel Shilovsky <pshilov@microsoft.com>
cifs_delete_mid() is called once we are finished handling a mid and we
expect no more work done on this mid.
Needed to fix recent commit:
commit 730928c8f4
("cifs: update smb2_queryfs() to use compounding")
Add a warning if someone tries to dequeue a mid that has already been
flagged to be deleted.
Also change list_del() to list_del_init() so that if we have similar bugs
resurface in the future we will not oops.
Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Reviewed-by: Pavel Shilovsky <pshilov@microsoft.com>
When mounting a Windows share that is the root of a drive (eg. C$)
the server does not return . and .. directory entries. This results in
the smb2 code path erroneously skipping the 2 first entries.
Pseudo-code of the readdir() code path:
cifs_readdir(struct file, struct dir_context)
initiate_cifs_search <-- if no reponse cached yet
server->ops->query_dir_first
dir_emit_dots
dir_emit <-- adds "." and ".." if we're at pos=0
find_cifs_entry
initiate_cifs_search <-- if pos < start of current response
(restart search)
server->ops->query_dir_next <-- if pos > end of current response
(fetch next search res)
for(...) <-- loops over cur response entries
starting at pos
cifs_filldir <-- skip . and .., emit entry
cifs_fill_dirent
dir_emit
pos++
A) dir_emit_dots() always adds . & ..
and sets the current dir pos to 2 (0 and 1 are done).
Therefore we always want the index_to_find to be 2 regardless of if
the response has . and ..
B) smb1 code initializes index_of_last_entry with a +2 offset
in cifssmb.c CIFSFindFirst():
psrch_inf->index_of_last_entry = 2 /* skip . and .. */ +
psrch_inf->entries_in_buffer;
Later in find_cifs_entry() we want to find the next dir entry at pos=2
as a result of (A)
first_entry_in_buffer = cfile->srch_inf.index_of_last_entry -
cfile->srch_inf.entries_in_buffer;
This var is the dir pos that the first entry in the buffer will
have therefore it must be 2 in the first call.
If we don't offset index_of_last_entry by 2 (like in (B)),
first_entry_in_buffer=0 but we were instructed to get pos=2 so this
code in find_cifs_entry() skips the 2 first which is ok for non-root
shares, as it skips . and .. from the response but is not ok for root
shares where the 2 first are actual files
pos_in_buf = index_to_find - first_entry_in_buffer;
// pos_in_buf=2
// we skip 2 first response entries :(
for (i = 0; (i < (pos_in_buf)) && (cur_ent != NULL); i++) {
/* go entry by entry figuring out which is first */
cur_ent = nxt_dir_entry(cur_ent, end_of_smb,
cfile->srch_inf.info_level);
}
C) cifs_filldir() skips . and .. so we can safely ignore them for now.
Sample program:
int main(int argc, char **argv)
{
const char *path = argc >= 2 ? argv[1] : ".";
DIR *dh;
struct dirent *de;
printf("listing path <%s>\n", path);
dh = opendir(path);
if (!dh) {
printf("opendir error %d\n", errno);
return 1;
}
while (1) {
de = readdir(dh);
if (!de) {
if (errno) {
printf("readdir error %d\n", errno);
return 1;
}
printf("end of listing\n");
break;
}
printf("off=%lu <%s>\n", de->d_off, de->d_name);
}
return 0;
}
Before the fix with SMB1 on root shares:
<.> off=1
<..> off=2
<$Recycle.Bin> off=3
<bootmgr> off=4
and on non-root shares:
<.> off=1
<..> off=4 <-- after adding .., the offsets jumps to +2 because
<2536> off=5 we skipped . and .. from response buffer (C)
<411> off=6 but still incremented pos
<file> off=7
<fsx> off=8
Therefore the fix for smb2 is to mimic smb1 behaviour and offset the
index_of_last_entry by 2.
Test results comparing smb1 and smb2 before/after the fix on root
share, non-root shares and on large directories (ie. multi-response
dir listing):
PRE FIX
=======
pre-1-root VS pre-2-root:
ERR pre-2-root is missing [bootmgr, $Recycle.Bin]
pre-1-nonroot VS pre-2-nonroot:
OK~ same files, same order, different offsets
pre-1-nonroot-large VS pre-2-nonroot-large:
OK~ same files, same order, different offsets
POST FIX
========
post-1-root VS post-2-root:
OK same files, same order, same offsets
post-1-nonroot VS post-2-nonroot:
OK same files, same order, same offsets
post-1-nonroot-large VS post-2-nonroot-large:
OK same files, same order, same offsets
REGRESSION?
===========
pre-1-root VS post-1-root:
OK same files, same order, same offsets
pre-1-nonroot VS post-1-nonroot:
OK same files, same order, same offsets
BugLink: https://bugzilla.samba.org/show_bug.cgi?id=13107
Signed-off-by: Aurelien Aptel <aaptel@suse.com>
Signed-off-by: Paulo Alcantara <palcantara@suse.deR>
Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
CC: Stable <stable@vger.kernel.org>
Commit 01239d77b9 ("xfs: fix a null pointer dereference in
xfs_bmap_extents_to_btree") attempted to fix a null pointer
dreference when a fuzzing corruption of some kind was found.
This fix was flawed, resulting in assert failures like:
XFS: Assertion failed: ifp->if_broot == NULL, file: fs/xfs/libxfs/xfs_bmap.c, line: 715
.....
Call Trace:
xfs_bmap_extents_to_btree+0x6b9/0x7b0
__xfs_bunmapi+0xae7/0xf00
? xfs_log_reserve+0x1c8/0x290
xfs_reflink_remap_extent+0x20b/0x620
xfs_reflink_remap_blocks+0x7e/0x290
xfs_reflink_remap_range+0x311/0x530
vfs_dedupe_file_range_one+0xd7/0xe0
vfs_dedupe_file_range+0x15b/0x1a0
do_vfs_ioctl+0x267/0x6c0
The problem is that the error handling code now asserts that the
inode fork is not in btree format before the error handling code
undoes the modifications that put the fork back in extent format.
Fix this by moving the assert back to after the xfs_iroot_realloc()
call that returns the fork to extent format, and clean up the jump
labels to be meaningful.
Also, returning ENOSPC when xfs_btree_get_bufl() fails to
instantiate the buffer that was allocated (the actual fix in the
commit mentioned above) is incorrect. This is a fatal error - only
an invalid block address or a filesystem shutdown can result in
failing to get a buffer here.
Hence change this to EFSCORRUPTED so that the higher layer knows
this was a corruption related failure and should not treat it as an
ENOSPC error. This should result in a shutdown (via cancelling a
dirty transaction) which is necessary as we do not attempt to clean
up the (invalid) block that we have already allocated.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
As reported by nixiaoming, with some minor clarifications:
1) memory leak in ramoops_register_dummy():
dummy_data = kzalloc(sizeof(*dummy_data), GFP_KERNEL);
but no kfree() if platform_device_register_data() fails.
2) memory leak in ramoops_init():
Missing platform_device_unregister(dummy) and kfree(dummy_data)
if platform_driver_register(&ramoops_driver) fails.
I've clarified the purpose of ramoops_register_dummy(), and added a
common cleanup routine for all three failure paths to call.
Reported-by: nixiaoming <nixiaoming@huawei.com>
Cc: stable@vger.kernel.org
Cc: Anton Vorontsov <anton@enomsg.org>
Cc: Colin Cross <ccross@android.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Geliang Tang <geliangtang@gmail.com>
Signed-off-by: Kees Cook <keescook@chromium.org>
Fix a deadlock in the new for 4.19 dax_lock_mapping_entry() routine.
-----BEGIN PGP SIGNATURE-----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=3l8a
-----END PGP SIGNATURE-----
Merge tag 'libnvdimm-fixes2-4.19-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm
Dan writes:
"filesystem-dax for 4.19-rc6
Fix a deadlock in the new for 4.19 dax_lock_mapping_entry() routine."
* tag 'libnvdimm-fixes2-4.19-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
dax: Fix deadlock in dax_lock_mapping_entry()
The iomap page fault mechanism currently dirties the associated page
after the full block range of the page has been allocated. This
leaves the page susceptible to delayed allocations without ever
being set dirty on sub-page block sized filesystems.
For example, consider a page fault on a page with one preexisting
real (non-delalloc) block allocated in the middle of the page. The
first iomap_apply() iteration performs delayed allocation on the
range up to the preexisting block, the next iteration finds the
preexisting block, and the last iteration attempts to perform
delayed allocation on the range after the prexisting block to the
end of the page. If the first allocation succeeds and the final
allocation fails with -ENOSPC, iomap_apply() returns the error and
iomap_page_mkwrite() fails to dirty the page having already
performed partial delayed allocation. This eventually results in the
page being invalidated without ever converting the delayed
allocation to real blocks.
This problem is reliably reproduced by generic/083 on XFS on ppc64
systems (64k page size, 4k block size). It results in leaked
delalloc blocks on inode reclaim, which triggers an assert failure
in xfs_fs_destroy_inode() and filesystem accounting inconsistency.
Move the set_page_dirty() call from iomap_page_mkwrite() to the
actor callback, similar to how the buffer head implementation works.
The actor callback is called iff ->iomap_begin() returns success, so
ensures the page is dirtied as soon as possible after an allocation.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
One of the first steps of log recovery is to check for the special
case of a zeroed log. If the first cycle in the log is zero or the
tail portion of the log is zeroed, the head is set to the first
instance of cycle 0. xlog_find_zeroed() includes a sanity check that
enforces that the first cycle in the log must be 1 if the last cycle
is 0. While this is true in most cases, the check is not totally
valid because it doesn't consider the case where the filesystem
crashed after a partial/out of order log buffer completion that
wraps around the end of the physical log.
For example, consider a filesystem that has completed most of the
first cycle of the log, reaches the end of the physical log and
splits the next single log buffer write into two in order to wrap
around the end of the log. If these I/Os are reordered, the second
(wrapped) I/O completes and the first happens to fail, the log is
left in a state where the last cycle of the log is 0 and the first
cycle is 2. This causes the xlog_find_zeroed() sanity check to fail
and prevents the filesystem from mounting. This situation has been
reproduced on particular systems via repeated runs of generic/475.
This is an expected state that log recovery already knows how to
deal with, however. Since the log is still partially zeroed, the
head is detected correctly and points to a valid tail. The
subsequent stale block detection clears blocks beyond the head up to
the tail (within a maximum range), with the express purpose of
clearing such out of order writes. As expected, this removes the out
of order cycle 2 blocks at the physical start of the log.
In other words, the only thing that prevents a clean mount and
recovery of the filesystem in this scenario is the specific (last ==
0 && first != 1) sanity check in xlog_find_zeroed(). Since the log
head/tail are now independently validated via cycle, log record and
CRC checks, this highly specific first cycle check is of dubious
value. Remove it and rely on the higher level validation to
determine whether log content is sane and recoverable.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Verify the inode di_forkoff, lifted from xfs_repair's
process_check_inode_forkoff().
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
The iomap direct I/O code issues a single ->end_io call for the whole
I/O request, and if some of the extents cowered needed a COW operation
it will call xfs_reflink_end_cow over the whole range.
When we do AIO writes we drop the iolock after doing the initial setup,
but before the I/O completion. Between dropping the lock and completing
the I/O we can have a racing buffered write create new delalloc COW fork
extents in the region covered by the outstanding direct I/O write, and
thus see delalloc COW fork extents in xfs_reflink_end_cow. As
concurrent writes are fundamentally racy and no guarantees are given we
can simply skip those.
This can be easily reproduced with xfstests generic/208 in always_cow
mode.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
xchk_inode_flags2() currently treats any di_flags2 values that the
running kernel doesn't recognize as corruption, and calls
xchk_ino_set_corrupt() if they are set. However, it's entirely possible
that these flags were set in some newer kernel and are quite valid,
but ignored in this kernel.
(Validators don't care one bit about unknown di_flags2.)
Call xchk_ino_set_warning instead, because this may or may not actually
indicate a problem.
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Remove duplicated include xfs_alloc.h
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
This function is only used to punch out delayed allocations on I/O
failure, which means we need to have read the extents earlier.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
When xfs_reflink_allocate_cow() allocates a transaction, it drops
the ILOCK to perform the operation. This Introduces a race condition
where another thread modifying the file can perform the COW
allocation operation underneath us. This result in the retry loop
finding an allocated block and jumping straight to the conversion
code. It does not, however, cancel the transaction it holds and so
this gets leaked. This results in a lockdep warning:
================================================
WARNING: lock held when returning to user space!
4.18.5 #1 Not tainted
------------------------------------------------
worker/6123 is leaving the kernel with locks still held!
1 lock held by worker/6123:
#0: 000000009eab4f1b (sb_internal#2){.+.+}, at: xfs_trans_alloc+0x17c/0x220
And eventually the filesystem deadlocks because it runs out of log
space that is reserved by the leaked transaction and never gets
released.
The logic flow in xfs_reflink_allocate_cow() is a convoluted mess of
gotos - it's no surprise that it has bug where the flow through
several goto jumps then fails to clean up context from a non-obvious
logic path. CLean up the logic flow and make sure every path does
the right thing.
Reported-by: Alexander Y. Fomichev <git.user@gmail.com>
Tested-by: Alexander Y. Fomichev <git.user@gmail.com>
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=200981
Signed-off-by: Dave Chinner <dchinner@redhat.com>
[hch: slight refactor]
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
We've had a few reports of lockdep tripping over memory reclaim
context vs filesystem freeze "deadlocks". They all have looked
to be false positives on analysis, but it seems that they are
being tripped because we take freeze references before we run
a GFP_KERNEL allocation for the struct xfs_trans.
We can avoid this false positive vector just by re-ordering the
operations in xfs_trans_alloc(). That is. we need allocate the
structure before we take the freeze reference and enter the GFP_NOFS
allocation context that follows the xfs_trans around. This prevents
lockdep from seeing the GFP_KERNEL allocation inside the transaction
context, and that prevents it from triggering the freeze level vs
alloc context vs reclaim warnings.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
The xfs_buf_log_item structure has a reference counter with slightly
tricky semantics. In the common case, a buffer is logged and
committed in a transaction, committed to the on-disk log (added to
the AIL) and then finally written back and removed from the AIL. The
bli refcount covers two potentially overlapping timeframes:
1. the bli is held in an active transaction
2. the bli is pinned by the log
The caveat to this approach is that the reference counter does not
purely dictate the lifetime of the bli. IOW, when a dirty buffer is
physically logged and unpinned, the bli refcount may go to zero as
the log item is inserted into the AIL. Only once the buffer is
written back can the bli finally be freed.
The above semantics means that it is not enough for the various
refcount decrementing contexts to release the bli on decrement to
zero. xfs_trans_brelse(), transaction commit (->iop_unlock()) and
unpin (->iop_unpin()) must all drop the associated reference and
make additional checks to determine if the current context is
responsible for freeing the item.
For example, if a transaction holds but does not dirty a particular
bli, the commit may drop the refcount to zero. If the bli itself is
clean, it is also not AIL resident and must be freed at this time.
The same is true for xfs_trans_brelse(). If the transaction dirties
a bli and then aborts or an unpin results in an abort due to a log
I/O error, the last reference count holder is expected to explicitly
remove the item from the AIL and release it (since an abort means
filesystem shutdown and metadata writeback will never occur).
This leads to fairly complex checks being replicated in a few
different places. Since ->iop_unlock() and xfs_trans_brelse() are
nearly identical, refactor the logic into a common helper that
implements and documents the semantics in one place. This patch does
not change behavior.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
