The performance regression that Josef Bacik reported in the pathname
lookup (see commit 99d263d4c5 "vfs: fix bad hashing of dentries") made
me look at performance stability of the dcache code, just to verify that
the problem was actually fixed. That turned up a few other problems in
this area.
There are a few cases where we exit RCU lookup mode and go to the slow
serializing case when we shouldn't, Al has fixed those and they'll come
in with the next VFS pull.
But my performance verification also shows that link_path_walk() turns
out to have a very unfortunate 32-bit store of the length and hash of
the name we look up, followed by a 64-bit read of the combined hash_len
field. That screws up the processor store to load forwarding, causing
an unnecessary hickup in this critical routine.
It's caused by the ugly calling convention for the "hash_name()"
function, and easily fixed by just making hash_name() fill in the whole
'struct qstr' rather than passing it a pointer to just the hash value.
With that, the profile for this function looks much smoother.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Josef Bacik found a performance regression between 3.2 and 3.10 and
narrowed it down to commit bfcfaa77bd ("vfs: use 'unsigned long'
accesses for dcache name comparison and hashing"). He reports:
"The test case is essentially
for (i = 0; i < 1000000; i++)
mkdir("a$i");
On xfs on a fio card this goes at about 20k dir/sec with 3.2, and 12k
dir/sec with 3.10. This is because we spend waaaaay more time in
__d_lookup on 3.10 than in 3.2.
The new hashing function for strings is suboptimal for <
sizeof(unsigned long) string names (and hell even > sizeof(unsigned
long) string names that I've tested). I broke out the old hashing
function and the new one into a userspace helper to get real numbers
and this is what I'm getting:
Old hash table had 1000000 entries, 0 dupes, 0 max dupes
New hash table had 12628 entries, 987372 dupes, 900 max dupes
We had 11400 buckets with a p50 of 30 dupes, p90 of 240 dupes, p99 of 567 dupes for the new hash
My test does the hash, and then does the d_hash into a integer pointer
array the same size as the dentry hash table on my system, and then
just increments the value at the address we got to see how many
entries we overlap with.
As you can see the old hash function ended up with all 1 million
entries in their own bucket, whereas the new one they are only
distributed among ~12.5k buckets, which is why we're using so much
more CPU in __d_lookup".
The reason for this hash regression is two-fold:
- On 64-bit architectures the down-mixing of the original 64-bit
word-at-a-time hash into the final 32-bit hash value is very
simplistic and suboptimal, and just adds the two 32-bit parts
together.
In particular, because there is no bit shuffling and the mixing
boundary is also a byte boundary, similar character patterns in the
low and high word easily end up just canceling each other out.
- the old byte-at-a-time hash mixed each byte into the final hash as it
hashed the path component name, resulting in the low bits of the hash
generally being a good source of hash data. That is not true for the
word-at-a-time case, and the hash data is distributed among all the
bits.
The fix is the same in both cases: do a better job of mixing the bits up
and using as much of the hash data as possible. We already have the
"hash_32|64()" functions to do that.
Reported-by: Josef Bacik <jbacik@fb.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Chris Mason <clm@fb.com>
Cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull btrfs fixes from Chris Mason:
"Filipe is doing a careful pass through fsync problems, and these are
the fixes so far. I'll have one more for rc6 that we're still
testing.
My big commit is fixing up some inode hash races that Al Viro found
(thanks Al)"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs: use insert_inode_locked4 for inode creation
Btrfs: fix fsync data loss after a ranged fsync
Btrfs: kfree()ing ERR_PTRs
Btrfs: fix crash while doing a ranged fsync
Btrfs: fix corruption after write/fsync failure + fsync + log recovery
Btrfs: fix autodefrag with compression
Merge misc fixes from Andrew Morton:
"10 fixes"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
fs/notify: don't show f_handle if exportfs_encode_inode_fh failed
fsnotify/fdinfo: use named constants instead of hardcoded values
kcmp: fix standard comparison bug
mm/mmap.c: use pr_emerg when printing BUG related information
shm: add memfd.h to UAPI export list
checkpatch: allow commit descriptions on separate line from commit id
sh: get_user_pages_fast() must flush cache
eventpoll: fix uninitialized variable in epoll_ctl
kernel/printk/printk.c: fix faulty logic in the case of recursive printk
mem-hotplug: let memblock skip the hotpluggable memory regions in __next_mem_range()
Currently we handle only ENOSPC. In case of other errors the file_handle
variable isn't filled properly and we will show a part of stack.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Acked-by: Cyrill Gorcunov <gorcunov@openvz.org>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
MAX_HANDLE_SZ is equal to 128, but currently the size of pad is only 64
bytes, so exportfs_encode_inode_fh can return an error.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Acked-by: Cyrill Gorcunov <gorcunov@openvz.org>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When calling epoll_ctl with operation EPOLL_CTL_DEL, structure epds is
not initialized but ep_take_care_of_epollwakeup reads its event field.
When this unintialized field has EPOLLWAKEUP bit set, a capability check
is done for CAP_BLOCK_SUSPEND in ep_take_care_of_epollwakeup. This
produces unexpected messages in the audit log, such as (on a system
running SELinux):
type=AVC msg=audit(1408212798.866:410): avc: denied
{ block_suspend } for pid=7754 comm="dbus-daemon" capability=36
scontext=unconfined_u:unconfined_r:unconfined_t
tcontext=unconfined_u:unconfined_r:unconfined_t
tclass=capability2 permissive=1
type=SYSCALL msg=audit(1408212798.866:410): arch=c000003e syscall=233
success=yes exit=0 a0=3 a1=2 a2=9 a3=7fffd4d66ec0 items=0 ppid=1
pid=7754 auid=1000 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0
fsgid=0 tty=(none) ses=3 comm="dbus-daemon"
exe="/usr/bin/dbus-daemon"
subj=unconfined_u:unconfined_r:unconfined_t key=(null)
("arch=c000003e syscall=233 a1=2" means "epoll_ctl(op=EPOLL_CTL_DEL)")
Remove use of epds in epoll_ctl when op == EPOLL_CTL_DEL.
Fixes: 4d7e30d989 ("epoll: Add a flag, EPOLLWAKEUP, to prevent suspend while epoll events are ready")
Signed-off-by: Nicolas Iooss <nicolas.iooss_linux@m4x.org>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Arve Hjønnevåg <arve@android.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull UDF fixes from Jan Kara:
"Fixes for UDF handling of NFS handles and one fix for proper handling
of corrupted media"
* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
udf: saner calling conventions for udf_new_inode()
udf: fix the udf_iget() vs. udf_new_inode() races
udf: merge the pieces inserting a new non-directory object into directory
udf: Set i_generation field
udf: Properly detect stale inodes
udf: Make udf_read_inode() and udf_iget() return error
udf: Avoid infinite loop when processing indirect ICBs
udf: Fold udf_fill_inode() into __udf_read_inode()
udf: Avoid dir link count to go negative
filelayout_retry_commit was recently split out from alloc_ds_commits,
but was done in such a way that the bucket pointer always starts at
index 0 no matter what the @idx argument is set to.
The intention of the @idx argument is to retry commits starting at
bucket @idx. This is called when alloc_ds_commits fails for a bucket.
Signed-off-by: Weston Andros Adamson <dros@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Pull cifs/smb3 fixes from Steve French:
"This includes various cifs and smb3 bug fixes including those for bugs
found with the recently updated xfstests.
Also I am working fixes for two additional cifs problems found by
xfstests which I plan to send later (when reviewed and run additional
tests)"
* 'for-next-3.17' of git://git.samba.org/sfrench/cifs-2.6:
Clarify Kconfig help text for CIFS and SMB2/SMB3
CIFS: Fix wrong filename length for SMB2
CIFS: Fix wrong restart readdir for SMB1
CIFS: Fix directory rename error
cifs: No need to send SIGKILL to demux_thread during umount
cifs: Allow directIO read/write during cache=strict
cifs: remove unneeded check of null checking in if condition
cifs: fix a possible use of uninit variable in SMB2_sess_setup
cifs: fix memory leak when password is supplied multiple times
cifs: fix a possible null pointer deref in decode_ascii_ssetup
Trivial whitespace fix
Pull ext4 bugfix from Ted Ts'o.
[ Hmm. It's possible we should make kfree() aware of error pointers,
and use IS_ERR_OR_NULL rather than a NULL check. But in the meantime
this is obviously the right fix. - Linus ]
* 'for_linus_urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
ext4: avoid trying to kfree an ERR_PTR pointer
Pull nfsd bugfixes from Bruce Fields:
"A couple minor nfsd bugfixes"
* 'for-3.17' of git://linux-nfs.org/~bfields/linux:
lockd: fix rpcbind crash on lockd startup failure
nfsd4: fix rd_dircount enforcement
Btrfs was inserting inodes into the hash table before we had fully
set the inode up on disk. This leaves us open to rare races that allow
two different inodes in memory for the same [root, inode] pair.
This patch fixes things by using insert_inode_locked4 to insert an I_NEW
inode and unlock_new_inode when we're ready for the rest of the kernel
to use the inode.
It also makes sure to init the operations pointers on the inode before
going into the error handling paths.
Signed-off-by: Chris Mason <clm@fb.com>
Reported-by: Al Viro <viro@zeniv.linux.org.uk>
While we're doing a full fsync (when the inode has the flag
BTRFS_INODE_NEEDS_FULL_SYNC set) that is ranged too (covers only a
portion of the file), we might have ordered operations that are started
before or while we're logging the inode and that fall outside the fsync
range.
Therefore when a full ranged fsync finishes don't remove every extent
map from the list of modified extent maps - as for some of them, that
fall outside our fsync range, their respective ordered operation hasn't
finished yet, meaning the corresponding file extent item wasn't inserted
into the fs/subvol tree yet and therefore we didn't log it, and we must
let the next fast fsync (one that checks only the modified list) see this
extent map and log a matching file extent item to the log btree and wait
for its ordered operation to finish (if it's still ongoing).
A test case for xfstests follows.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
The "inherit" in btrfs_ioctl_snap_create_v2() and "vol_args" in
btrfs_ioctl_rm_dev() are ERR_PTRs so we can't call kfree() on them.
These kind of bugs are "One Err Bugs" where there is just one error
label that does everything. I could set the "inherit = NULL" and keep
the single out label but it ends up being more complicated that way. It
makes the code simpler to re-order the unwind so it's in the mirror
order of the allocation and introduce some new error labels.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Chris Mason <clm@fb.com>
Commit 3b29970909 "nfsd4: enforce rd_dircount" totally misunderstood
rd_dircount; it refers to total non-attribute bytes returned, not number
of directory entries returned.
Bring the code into agreement with RFC 3530 section 14.2.24.
Cc: stable@vger.kernel.org
Fixes: 3b29970909 "nfsd4: enforce rd_dircount"
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Pull filesystem fixes from Al Viro:
"Several bugfixes (all of them -stable fodder).
Alexey's one deals with double mutex_lock() in UFS (apparently, nobody
has tried to test "ufs: sb mutex merge + mutex_destroy" on something
like file creation/removal on ufs). Mine deal with two kinds of
umount bugs, in umount propagation and in handling of automounted
submounts, both resulting in bogus transient EBUSY from umount"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
ufs: fix deadlocks introduced by sb mutex merge
fix EBUSY on umount() from MNT_SHRINKABLE
get rid of propagate_umount() mistakenly treating slaves as busy.
Commit 0244756edc ("ufs: sb mutex merge + mutex_destroy") introduces
deadlocks in ufs_new_inode() and ufs_free_inode().
Most callers of that functions acqure the mutex by themselves and
ufs_{new,free}_inode() do that via lock_ufs(),
i.e we have an unavoidable double lock.
The patch proposes to resolve the issue by making sure that
ufs_{new,free}_inode() are not called with the mutex held.
Found by Linux Driver Verification project (linuxtesting.org).
Cc: stable@vger.kernel.org # 3.16
Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Fix:
- a direct IO read/buffered read data corruption
- the associated fallout from the DIO data corruption fix
- collapse range bugs that are potential data corruption issues.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)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=tRQo
-----END PGP SIGNATURE-----
Merge tag 'xfs-for-linus-3.17-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/dgc/linux-xfs
Pull xfs fixes from Dave Chinner:
"The fixes all address recently discovered data corruption issues.
The original Direct IO issue was discovered by Chris Mason @ Facebook
on a production workload which mixed buffered reads with direct reads
and writes IO to the same file. The fix for that exposed other issues
with page invalidation (exposed by millions of fsx operations) failing
due to dirty buffers beyond EOF.
Finally, the collapse_range code could also cause problems due to
racing writeback changing the extent map while it was being shifted
around. The commits for that problem are simple mitigation fixes that
prevent the problem from occuring. A more robust fix for 3.18 that
addresses the underlying problem is currently being worked on by
Brian.
Summary of fixes:
- a direct IO read/buffered read data corruption
- the associated fallout from the DIO data corruption fix
- collapse range bugs that are potential data corruption issues"
* tag 'xfs-for-linus-3.17-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/dgc/linux-xfs:
xfs: trim eofblocks before collapse range
xfs: xfs_file_collapse_range is delalloc challenged
xfs: don't log inode unless extent shift makes extent modifications
xfs: use ranged writeback and invalidation for direct IO
xfs: don't zero partial page cache pages during O_DIRECT writes
xfs: don't zero partial page cache pages during O_DIRECT writes
xfs: don't dirty buffers beyond EOF
This patch changes sync_filesystem() to be EXPORT_SYMBOL().
The reason this is needed is that starting with 3.15 kernel, due to
Theodore Ts'o's commit 02b9984d64 ("fs: push sync_filesystem() down to
the file system's remount_fs()"), all file systems that have dirty data
to be written out need to call sync_filesystem() from their
->remount_fs() method when remounting read-only.
As this is now a generically required function rather than an internal
only function it should be EXPORT_SYMBOL() so that all file systems can
call it.
Signed-off-by: Anton Altaparmakov <aia21@cantab.net>
Acked-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull aio bugfixes from Ben LaHaise:
"Two small fixes"
* git://git.kvack.org/~bcrl/aio-fixes:
aio: block exit_aio() until all context requests are completed
aio: add missing smp_rmb() in read_events_ring
It seems that exit_aio() also needs to wait for all iocbs to complete (like
io_destroy), but we missed the wait step in current implemention, so fix
it in the same way as we did in io_destroy.
Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
Signed-off-by: Benjamin LaHaise <bcrl@kvack.org>
Cc: stable@vger.kernel.org
Currently udf_iget() (triggered by NFS) can race with udf_new_inode()
leading to two inode structures with the same inode number:
nfsd: iget_locked() creates inode
nfsd: try to read from disk, block on that.
udf_new_inode(): allocate inode with that inumber
udf_new_inode(): insert it into icache, set it up and dirty
udf_write_inode(): write inode into buffer cache
nfsd: get CPU again, look into buffer cache, see nice and sane on-disk
inode, set the in-core inode from it
Fix the problem by putting inode into icache in locked state (I_NEW set)
and unlocking it only after it's fully set up.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jan Kara <jack@suse.cz>
boilerplate code in udf_{create,mknod,symlink} taken to new helper
symlink case converted to unique id calculated by udf_new_inode() - no
point finding a new one.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jan Kara <jack@suse.cz>
Currently UDF doesn't initialize i_generation in any way and thus NFS
can easily get reallocated inodes from stale file handles. Luckily UDF
already has a unique object identifier associated with each inode -
i_unique. Use that for initialization of i_generation.
Signed-off-by: Jan Kara <jack@suse.cz>
NFS can easily ask for inodes that are already deleted. Currently UDF
happily returns such inodes which is a bug. Return -ESTALE if
udf_read_inode() is asked to read deleted inode.
Signed-off-by: Jan Kara <jack@suse.cz>
Currently __udf_read_inode() wasn't returning anything and we found out
whether we succeeded reading inode by checking whether inode is bad or
not. udf_iget() returned NULL on failure and inode pointer otherwise.
Make these two functions properly propagate errors up the call stack and
use the return value in callers.
Signed-off-by: Jan Kara <jack@suse.cz>
We did not implement any bound on number of indirect ICBs we follow when
loading inode. Thus corrupted medium could cause kernel to go into an
infinite loop, possibly causing a stack overflow.
Fix the possible stack overflow by removing recursion from
__udf_read_inode() and limit number of indirect ICBs we follow to avoid
infinite loops.
Signed-off-by: Jan Kara <jack@suse.cz>
There's no good reason to separate these since udf_fill_inode() is
called only from __udf_read_inode() and both do part of the same thing.
Signed-off-by: Jan Kara <jack@suse.cz>
If we are writing back inode of unlinked directory, its link count ends
up being (u16)-1. Although the inode is deleted, udf_iget() can load the
inode when NFS uses stale file handle and get confused.
Signed-off-by: Jan Kara <jack@suse.cz>
Pull f2fs bug fixes from Jaegeuk Kim:
"This series includes patches to:
- fix recovery routines
- fix bugs related to inline_data/xattr
- fix when casting the dentry names
- handle EIO or ENOMEM correctly
- fix memory leak
- fix lock coverage"
* tag 'for-f2fs-3.17-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (28 commits)
f2fs: reposition unlock_new_inode to prevent accessing invalid inode
f2fs: fix wrong casting for dentry name
f2fs: simplify by using a literal
f2fs: truncate stale block for inline_data
f2fs: use macro for code readability
f2fs: introduce need_do_checkpoint for readability
f2fs: fix incorrect calculation with total/free inode num
f2fs: remove rename and use rename2
f2fs: skip if inline_data was converted already
f2fs: remove rewrite_node_page
f2fs: avoid double lock in truncate_blocks
f2fs: prevent checkpoint during roll-forward
f2fs: add WARN_ON in f2fs_bug_on
f2fs: handle EIO not to break fs consistency
f2fs: check s_dirty under cp_mutex
f2fs: unlock_page when node page is redirtied out
f2fs: introduce f2fs_cp_error for readability
f2fs: give a chance to mount again when encountering errors
f2fs: trigger release_dirty_inode in f2fs_put_super
f2fs: don't skip checkpoint if there is no dirty node pages
...
Thanks to Dan Carpenter for extending smatch to find bugs like this.
(This was found using a development version of smatch.)
Fixes: 36de928641
Reported-by: Dan Carpenter <dan.carpenter@oracle.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@vger.kernel.org
While doing a ranged fsync, that is, one whose range doesn't cover the
whole possible file range (0 to LLONG_MAX), we can crash under certain
circumstances with a trace like the following:
[41074.641913] invalid opcode: 0000 [#1] SMP DEBUG_PAGEALLOC
(...)
[41074.642692] CPU: 0 PID: 24580 Comm: fsx Not tainted 3.16.0-fdm-btrfs-next-45+ #1
(...)
[41074.643886] RIP: 0010:[<ffffffffa01ecc99>] [<ffffffffa01ecc99>] btrfs_ordered_update_i_size+0x279/0x2b0 [btrfs]
(...)
[41074.644919] Stack:
(...)
[41074.644919] Call Trace:
[41074.644919] [<ffffffffa01db531>] btrfs_truncate_inode_items+0x3f1/0xa10 [btrfs]
[41074.644919] [<ffffffffa01eb54f>] ? btrfs_get_logged_extents+0x4f/0x80 [btrfs]
[41074.644919] [<ffffffffa02137a9>] btrfs_log_inode+0x2f9/0x970 [btrfs]
[41074.644919] [<ffffffff81090875>] ? sched_clock_local+0x25/0xa0
[41074.644919] [<ffffffff8164a55e>] ? mutex_unlock+0xe/0x10
[41074.644919] [<ffffffff810af51d>] ? trace_hardirqs_on+0xd/0x10
[41074.644919] [<ffffffffa0214b4f>] btrfs_log_inode_parent+0x1ef/0x560 [btrfs]
[41074.644919] [<ffffffff811d0c55>] ? dget_parent+0x5/0x180
[41074.644919] [<ffffffffa0215d11>] btrfs_log_dentry_safe+0x51/0x80 [btrfs]
[41074.644919] [<ffffffffa01e2d1a>] btrfs_sync_file+0x1ba/0x3e0 [btrfs]
[41074.644919] [<ffffffff811eda6b>] vfs_fsync_range+0x1b/0x30
(...)
The necessary conditions that lead to such crash are:
* an incremental fsync (when the inode doesn't have the
BTRFS_INODE_NEEDS_FULL_SYNC flag set) happened for our file and it logged
a file extent item ending at offset X;
* the file got the flag BTRFS_INODE_NEEDS_FULL_SYNC set in its inode, due
to a file truncate operation that reduces the file to a size smaller
than X;
* a ranged fsync call happens (via an msync for example), with a range that
doesn't cover the whole file and the end of this range, lets call it Y, is
smaller than X;
* btrfs_log_inode, sees the flag BTRFS_INODE_NEEDS_FULL_SYNC set and
calls btrfs_truncate_inode_items() to remove all items from the log
tree that are associated with our file;
* btrfs_truncate_inode_items() removes all of the inode's items, and the lowest
file extent item it removed is the one ending at offset X, where X > 0 and
X > Y - before returning, it calls btrfs_ordered_update_i_size() with an offset
parameter set to X;
* btrfs_ordered_update_i_size() sees that X is greater then the current ordered
size (btrfs_inode's disk_i_size) and then it assumes there can't be any ongoing
ordered operation with a range covering the offset X, calling a BUG_ON() if
such ordered operation exists. This assumption is made because the disk_i_size
is only increased after the corresponding file extent item is added to the
btree (btrfs_finish_ordered_io);
* But because our fsync covers only a limited range, such an ordered extent might
exist, and our fsync callback (btrfs_sync_file) doesn't wait for such ordered
extent to finish when calling btrfs_wait_ordered_range();
And then by the time btrfs_ordered_update_i_size() is called, via:
btrfs_sync_file() ->
btrfs_log_dentry_safe() ->
btrfs_log_inode_parent() ->
btrfs_log_inode() ->
btrfs_truncate_inode_items() ->
btrfs_ordered_update_i_size()
We hit the BUG_ON(), which could never happen if the fsync range covered the whole
possible file range (0 to LLONG_MAX), as we would wait for all ordered extents to
finish before calling btrfs_truncate_inode_items().
So just don't call btrfs_ordered_update_i_size() if we're removing the inode's items
from a log tree, which isn't supposed to change the in memory inode's disk_i_size.
Issue found while running xfstests/generic/127 (happens very rarely for me), more
specifically via the fsx calls that use memory mapped IO (and issue msync calls).
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
While writing to a file, in inode.c:cow_file_range() (and same applies to
submit_compressed_extents()), after reserving an extent for the file data,
we create a new extent map for the written range and insert it into the
extent map cache. After that, we create an ordered operation, but if it
fails (due to a transient/temporary-ENOMEM), we return without dropping
that extent map, which points to a reserved extent that is freed when we
return. A subsequent incremental fsync (when the btrfs inode doesn't have
the flag BTRFS_INODE_NEEDS_FULL_SYNC) considers this extent map valid and
logs a file extent item based on that extent map, which points to a disk
extent that doesn't contain valid data - it was freed by us earlier, at this
point it might contain any random/garbage data.
Therefore, if we reach an error condition when cowing a file range after
we added the new extent map to the cache, drop it from the cache before
returning.
Some sequence of steps that lead to this:
$ mkfs.btrfs -f /dev/sdd
$ mount -o commit=9999 /dev/sdd /mnt
$ cd /mnt
$ xfs_io -f -c "pwrite -S 0x01 -b 4096 0 4096" -c "fsync" foo
$ xfs_io -c "pwrite -S 0x02 -b 4096 4096 4096"
$ sync
$ od -t x1 foo
0000000 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01
*
0010000 02 02 02 02 02 02 02 02 02 02 02 02 02 02 02 02
*
0020000
$ xfs_io -c "pwrite -S 0xa1 -b 4096 0 4096" foo
# Now this write + fsync fail with -ENOMEM, which was returned by
# btrfs_add_ordered_extent() in inode.c:cow_file_range().
$ xfs_io -c "pwrite -S 0xff -b 4096 4096 4096" foo
$ xfs_io -c "fsync" foo
fsync: Cannot allocate memory
# Now do a new write + fsync, which will succeed. Our previous
# -ENOMEM was a transient/temporary error.
$ xfs_io -c "pwrite -S 0xee -b 4096 16384 4096" foo
$ xfs_io -c "fsync" foo
# Our file content (in page cache) is now:
$ od -t x1 foo
0000000 a1 a1 a1 a1 a1 a1 a1 a1 a1 a1 a1 a1 a1 a1 a1 a1
*
0010000 ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
*
0020000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
*
0040000 ee ee ee ee ee ee ee ee ee ee ee ee ee ee ee ee
*
0050000
# Now reboot the machine, and mount the fs, so that fsync log replay
# takes place.
# The file content is now weird, in particular the first 8Kb, which
# do not match our data before nor after the sync command above.
$ od -t x1 foo
0000000 ee ee ee ee ee ee ee ee ee ee ee ee ee ee ee ee
*
0010000 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01
*
0020000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
*
0040000 ee ee ee ee ee ee ee ee ee ee ee ee ee ee ee ee
*
0050000
# In fact these first 4Kb are a duplicate of the last 4kb block.
# The last write got an extent map/file extent item that points to
# the same disk extent that we got in the write+fsync that failed
# with the -ENOMEM error. btrfs-debug-tree and btrfsck allow us to
# verify that:
$ btrfs-debug-tree /dev/sdd
(...)
item 6 key (257 EXTENT_DATA 0) itemoff 15819 itemsize 53
extent data disk byte 12582912 nr 8192
extent data offset 0 nr 8192 ram 8192
item 7 key (257 EXTENT_DATA 8192) itemoff 15766 itemsize 53
extent data disk byte 0 nr 0
extent data offset 0 nr 8192 ram 8192
item 8 key (257 EXTENT_DATA 16384) itemoff 15713 itemsize 53
extent data disk byte 12582912 nr 4096
extent data offset 0 nr 4096 ram 4096
$ umount /dev/sdd
$ btrfsck /dev/sdd
Checking filesystem on /dev/sdd
UUID: db5e60e1-050d-41e6-8c7f-3d742dea5d8f
checking extents
extent item 12582912 has multiple extent items
ref mismatch on [12582912 4096] extent item 1, found 2
Backref bytes do not match extent backref, bytenr=12582912, ref bytes=4096, backref bytes=8192
backpointer mismatch on [12582912 4096]
Errors found in extent allocation tree or chunk allocation
checking free space cache
checking fs roots
root 5 inode 257 errors 1000, some csum missing
found 131074 bytes used err is 1
total csum bytes: 4
total tree bytes: 131072
total fs tree bytes: 32768
total extent tree bytes: 16384
btree space waste bytes: 123404
file data blocks allocated: 274432
referenced 274432
Btrfs v3.14.1-96-gcc7fd5a-dirty
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
We ran into a case on ppc64 running mariadb where io_getevents would
return zeroed out I/O events. After adding instrumentation, it became
clear that there was some missing synchronization between reading the
tail pointer and the events themselves. This small patch fixes the
problem in testing.
Thanks to Zach for helping to look into this, and suggesting the fix.
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Benjamin LaHaise <bcrl@kvack.org>
Cc: stable@vger.kernel.org
As the race condition on the inode cache, following scenario can appear:
[Thread a] [Thread b]
->f2fs_mkdir
->f2fs_add_link
->__f2fs_add_link
->init_inode_metadata failed here
->gc_thread_func
->f2fs_gc
->do_garbage_collect
->gc_data_segment
->f2fs_iget
->iget_locked
->wait_on_inode
->unlock_new_inode
->move_data_page
->make_bad_inode
->iput
When we fail in create/symlink/mkdir/mknod/tmpfile, the new allocated inode
should be set as bad to avoid being accessed by other thread. But in above
scenario, it allows f2fs to access the invalid inode before this inode was set
as bad.
This patch fix the potential problem, and this issue was found by code review.
change log from v1:
o Add condition judgment in gc_data_segment() suggested by Changman Lee.
o use iget_failed to simplify code.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
xfs_collapse_file_space() currently writes back the entire file
undergoing collapse range to settle things down for the extent shift
algorithm. While this prevents changes to the extent list during the
collapse operation, the writeback itself is not enough to prevent
unnecessary collapse failures.
The current shift algorithm uses the extent index to iterate the in-core
extent list. If a post-eof delalloc extent persists after the writeback
(e.g., a prior zero range op where the end of the range aligns with eof
can separate the post-eof blocks such that they are not written back and
converted), xfs_bmap_shift_extents() becomes confused over the encoded
br_startblock value and fails the collapse.
As with the full writeback, this is a temporary fix until the algorithm
is improved to cope with a volatile extent list and avoid attempts to
shift post-eof extents.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dave Chinner <david@fromorbit.com>
If we have delalloc extents on a file before we run a collapse range
opertaion, we sync the range that we are going to collapse to
convert delalloc extents in that region to real extents to simplify
the shift operation.
However, the shift operation then assumes that the extent list is
not going to change as it iterates over the extent list moving
things about. Unfortunately, this isn't true because we can't hold
the ILOCK over all the operations. We can prevent new IO from
modifying the extent list by holding the IOLOCK, but that doesn't
prevent writeback from running....
And when writeback runs, it can convert delalloc extents is the
range of the file prior to the region being collapsed, and this
changes the indexes of all the extents in the file. That causes the
collapse range operation to Go Bad.
The right fix is to rewrite the extent shift operation not to be
dependent on the extent list not changing across the entire
operation, but this is a fairly significant piece of work to do.
Hence, as a short-term workaround for the problem, sync the entire
file before starting a collapse operation to remove all delalloc
ranges from the file and so avoid the problem of concurrent
writeback changing the extent list.
Diagnosed-and-Reported-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dave Chinner <david@fromorbit.com>
The file collapse mechanism uses xfs_bmap_shift_extents() to collapse
all subsequent extents down into the specified, previously punched out,
region. This function performs some validation, such as whether a
sufficient hole exists in the target region of the collapse, then shifts
the remaining exents downward.
The exit path of the function currently logs the inode unconditionally.
While we must log the inode (and abort) if an error occurs and the
transaction is dirty, the initial validation paths can generate errors
before the transaction has been dirtied. This creates an unnecessary
filesystem shutdown scenario, as the caller will cancel a transaction
that has been marked dirty.
Modify xfs_bmap_shift_extents() to OR the logflags bits as modifications
are made to the inode bmap. Only log the inode in the exit path if
logflags has been set. This ensures we only have to cancel a dirty
transaction if modifications have been made and prevents an unnecessary
filesystem shutdown otherwise.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Now we are not doing silly things with dirtying buffers beyond EOF
and using invalidation correctly, we can finally reduce the ranges of
writeback and invalidation used by direct IO to match that of the IO
being issued.
Bring the writeback and invalidation ranges back to match the
generic direct IO code - this will greatly reduce the perturbation
of cached data when direct IO and buffered IO are mixed, but still
provide the same buffered vs direct IO coherency behaviour we
currently have.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Similar to direct IO reads, direct IO writes are using
truncate_pagecache_range to invalidate the page cache. This is
incorrect due to the sub-block zeroing in the page cache that
truncate_pagecache_range() triggers.
This patch fixes things by using invalidate_inode_pages2_range
instead. It preserves the page cache invalidation, but won't zero
any pages.
cc: stable@vger.kernel.org
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dave Chinner <david@fromorbit.com>
xfs is using truncate_pagecache_range to invalidate the page cache
during DIO reads. This is different from the other filesystems who
only invalidate pages during DIO writes.
truncate_pagecache_range is meant to be used when we are freeing the
underlying data structs from disk, so it will zero any partial
ranges in the page. This means a DIO read can zero out part of the
page cache page, and it is possible the page will stay in cache.
buffered reads will find an up to date page with zeros instead of
the data actually on disk.
This patch fixes things by using invalidate_inode_pages2_range
instead. It preserves the page cache invalidation, but won't zero
any pages.
[dchinner: catch error and warn if it fails. Comment.]
cc: stable@vger.kernel.org
Signed-off-by: Chris Mason <clm@fb.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dave Chinner <david@fromorbit.com>
generic/263 is failing fsx at this point with a page spanning
EOF that cannot be invalidated. The operations are:
1190 mapwrite 0x52c00 thru 0x5e569 (0xb96a bytes)
1191 mapread 0x5c000 thru 0x5d636 (0x1637 bytes)
1192 write 0x5b600 thru 0x771ff (0x1bc00 bytes)
where 1190 extents EOF from 0x54000 to 0x5e569. When the direct IO
write attempts to invalidate the cached page over this range, it
fails with -EBUSY and so any attempt to do page invalidation fails.
The real question is this: Why can't that page be invalidated after
it has been written to disk and cleaned?
Well, there's data on the first two buffers in the page (1k block
size, 4k page), but the third buffer on the page (i.e. beyond EOF)
is failing drop_buffers because it's bh->b_state == 0x3, which is
BH_Uptodate | BH_Dirty. IOWs, there's dirty buffers beyond EOF. Say
what?
OK, set_buffer_dirty() is called on all buffers from
__set_page_buffers_dirty(), regardless of whether the buffer is
beyond EOF or not, which means that when we get to ->writepage,
we have buffers marked dirty beyond EOF that we need to clean.
So, we need to implement our own .set_page_dirty method that
doesn't dirty buffers beyond EOF.
This is messy because the buffer code is not meant to be shared
and it has interesting locking issues on the buffer dirty bits.
So just copy and paste it and then modify it to suit what we need.
Note: the solutions the other filesystems and generic block code use
of marking the buffers clean in ->writepage does not work for XFS.
It still leaves dirty buffers beyond EOF and invalidations still
fail. Hence rather than play whack-a-mole, this patch simply
prevents those buffers from being dirtied in the first place.
cc: <stable@kernel.org>
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=Wb1t
-----END PGP SIGNATURE-----
Merge tag 'locks-v3.17-3' of git://git.samba.org/jlayton/linux
Pull file locking bugfx from Jeff Layton:
"Just a bugfix for a bug that crept in to v3.15. It's in a rather rare
error path, and I'm not aware of anyone having hit it, but it's worth
fixing for v3.17"
* tag 'locks-v3.17-3' of git://git.samba.org/jlayton/linux:
locks: pass correct "before" pointer to locks_unlink_lock in generic_add_lease
We need the parents of victims alive until namespace_unlock() gets to
dput() of the (ex-)mountpoints. However, that screws up the "is it
busy" checks in case when we have shrinkable mounts that need to be
killed. Solution: go ahead and decrement refcounts of parents right
in umount_tree(), increment them again just before dropping rwsem in
namespace_unlock() (and let the loop in the end of namespace_unlock()
finally drop those references for good, as we do now). Parents can't
get freed until we drop rwsem - at least one reference is kept until
then, both in case when parent is among the victims and when it is
not. So they'll still be around when we get to namespace_unlock().
Cc: stable@vger.kernel.org # 3.12+
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
