linux/fs
Brian Foster c3155097ad xfs: sync eofblocks scans under iolock are livelock prone
The xfs_eofblocks.eof_scan_owner field is an internal field to
facilitate invoking eofb scans from the kernel while under the iolock.
This is necessary because the eofb scan acquires the iolock of each
inode. Synchronous scans are invoked on certain buffered write failures
while under iolock. In such cases, the scan owner indicates that the
context for the scan already owns the particular iolock and prevents a
double lock deadlock.

eofblocks scans while under iolock are still livelock prone in the event
of multiple parallel scans, however. If multiple buffered writes to
different inodes fail and invoke eofblocks scans at the same time, each
scan avoids a deadlock with its own inode by virtue of the
eof_scan_owner field, but will never be able to acquire the iolock of
the inode from the parallel scan. Because the low free space scans are
invoked with SYNC_WAIT, the scan will not return until it has processed
every tagged inode and thus both scans will spin indefinitely on the
iolock being held across the opposite scan. This problem can be
reproduced reliably by generic/224 on systems with higher cpu counts
(x16).

To avoid this problem, simplify the semantics of eofblocks scans to
never invoke a scan while under iolock. This means that the buffered
write context must drop the iolock before the scan. It must reacquire
the lock before the write retry and also repeat the initial write
checks, as the original state might no longer be valid once the iolock
was dropped.

Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2017-01-30 16:32:25 -08:00
..
9p Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
adfs Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2016-10-10 20:16:43 -07:00
affs vfs: remove ".readlink = generic_readlink" assignments 2016-12-09 16:45:04 +01:00
afs Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
autofs4 Merge uncontroversial parts of branch 'readlink' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs 2016-12-17 19:16:12 -08:00
befs befs: add NFS export support 2016-12-22 11:25:24 +00:00
bfs Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
btrfs Merge branch 'for-linus-4.10' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs 2017-01-27 12:41:46 -08:00
cachefiles Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2016-10-10 20:16:43 -07:00
ceph ceph: fix bad endianness handling in parse_reply_info_extra 2017-01-18 17:58:45 +01:00
cifs Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
coda vfs: remove ".readlink = generic_readlink" assignments 2016-12-09 16:45:04 +01:00
configfs Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
cramfs
crypto fscrypt: fix renaming and linking special files 2016-12-31 00:47:05 -05:00
debugfs Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2016-10-10 20:16:43 -07:00
devpts Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2016-10-10 20:16:43 -07:00
dlm ktime: Get rid of ktime_equal() 2016-12-25 17:21:23 +01:00
ecryptfs vfs: remove ".readlink = generic_readlink" assignments 2016-12-09 16:45:04 +01:00
efivarfs Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2016-10-10 20:16:43 -07:00
efs Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
exofs exofs: don't mess with simple_write_{begin,end} 2016-12-10 14:25:19 -05:00
exportfs exportfs: be careful to only return expected errors. 2016-10-06 09:07:44 -04:00
ext2 iomap: constify struct iomap_ops 2017-01-30 16:32:25 -08:00
ext4 iomap: constify struct iomap_ops 2017-01-30 16:32:25 -08:00
f2fs block: Rename blk_queue_zone_size and bdev_zone_size 2017-01-12 07:58:32 -07:00
fat Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2016-10-10 20:16:43 -07:00
freevxfs
fscache
fuse fuse: fix time_to_jiffies nsec sanity check 2017-01-13 17:20:47 +01:00
gfs2 ktime: Cleanup ktime_set() usage 2016-12-25 17:21:22 +01:00
hfs Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
hfsplus Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
hostfs vfs: remove ".readlink = generic_readlink" assignments 2016-12-09 16:45:04 +01:00
hpfs Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2016-10-10 20:16:43 -07:00
hugetlbfs Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
isofs Merge branch 'for-4.10/block' of git://git.kernel.dk/linux-block 2016-12-13 10:19:16 -08:00
jbd2 Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
jffs2 vfs: remove ".readlink = generic_readlink" assignments 2016-12-09 16:45:04 +01:00
jfs Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
kernfs Merge uncontroversial parts of branch 'readlink' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs 2016-12-17 19:16:12 -08:00
lockd netns: make struct pernet_operations::id unsigned int 2016-11-18 10:59:15 -05:00
minix vfs: remove ".readlink = generic_readlink" assignments 2016-12-09 16:45:04 +01:00
ncpfs Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
nfs pNFS: Fix a reference leak in _pnfs_return_layout 2017-01-26 15:50:41 -05:00
nfs_common netns: make struct pernet_operations::id unsigned int 2016-11-18 10:59:15 -05:00
nfsd nfsd: fix supported attributes for acl & labels 2017-01-12 15:55:51 -05:00
nilfs2 Merge uncontroversial parts of branch 'readlink' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs 2016-12-17 19:16:12 -08:00
nls
notify Merge branch 'stable-4.10' of git://git.infradead.org/users/pcmoore/audit 2017-01-05 23:06:06 -08:00
ntfs Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
ocfs2 ocfs2: fix crash caused by stale lvb with fsdlm plugin 2017-01-10 18:31:54 -08:00
omfs Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2016-10-10 20:16:43 -07:00
openpromfs Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
orangefs Merge uncontroversial parts of branch 'readlink' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs 2016-12-17 19:16:12 -08:00
overlayfs ovl: fix possible use after free on redirect dir lookup 2017-01-18 15:19:54 +01:00
proc proc: add a schedule point in proc_pid_readdir() 2017-01-24 16:26:14 -08:00
pstore Improvements and fixes to pstore subsystem: 2016-12-13 09:16:11 -08:00
qnx4
qnx6
quota Merge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs 2016-12-19 08:23:53 -08:00
ramfs Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
reiserfs Merge uncontroversial parts of branch 'readlink' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs 2016-12-17 19:16:12 -08:00
romfs romfs: use different way to generate fsid for BLOCK or MTD 2017-01-24 16:26:14 -08:00
squashfs Merge uncontroversial parts of branch 'readlink' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs 2016-12-17 19:16:12 -08:00
sysfs Merge branch 'for-4.9' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup 2016-10-14 12:18:50 -07:00
sysv vfs: remove ".readlink = generic_readlink" assignments 2016-12-09 16:45:04 +01:00
tracefs fs: Replace CURRENT_TIME with current_time() for inode timestamps 2016-09-27 21:06:21 -04:00
ubifs ubifs: Fix journal replay wrt. xattr nodes 2017-01-17 14:35:58 +01:00
udf block,fs: untangle fs.h and blk_types.h 2016-11-01 09:43:26 -06:00
ufs Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
xfs xfs: sync eofblocks scans under iolock are livelock prone 2017-01-30 16:32:25 -08:00
aio.c aio: fix lock dep warning 2017-01-14 19:31:40 -05:00
anon_inodes.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
attr.c Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2016-10-10 20:16:43 -07:00
bad_inode.c bad_inode: add missing i_op initializers 2016-12-09 11:57:43 +01:00
binfmt_aout.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
binfmt_elf.c coredump: Ensure proper size of sparse core files 2017-01-14 19:32:40 -05:00
binfmt_elf_fdpic.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
binfmt_em86.c fs/binfmt_em86.c: fix incompatible pointer type 2016-08-02 19:35:15 -04:00
binfmt_flat.c
binfmt_misc.c fs: Replace current_fs_time() with current_time() 2016-09-27 21:06:22 -04:00
binfmt_script.c
block_dev.c block: fix use after free in __blkdev_direct_IO 2017-01-24 07:55:53 -07:00
buffer.c clean_bdev_aliases: Prevent cleaning blocks that are not in block range 2017-01-02 09:35:14 -07:00
char_dev.c dax: define a unified inode/address_space for device-dax mappings 2016-08-23 22:58:51 -07:00
compat.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
compat_binfmt_elf.c
compat_ioctl.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
coredump.c coredump: Ensure proper size of sparse core files 2017-01-14 19:32:40 -05:00
dax.c iomap: constify struct iomap_ops 2017-01-30 16:32:25 -08:00
dcache.c mnt: Protect the mountpoint hashtable with mount_lock 2017-01-10 13:34:43 +13:00
dcookies.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
direct-io.c do_direct_IO: Use inode->i_blkbits to compute block count to be cleaned 2017-01-10 13:29:54 -07:00
drop_caches.c
eventfd.c
eventpoll.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
exec.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
fcntl.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
fhandle.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
file.c fs/file: more unsigned file descriptors 2016-09-27 18:47:38 -04:00
file_table.c constify alloc_file() 2016-12-05 19:01:16 -05:00
filesystems.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
fs-writeback.c fs/fs-writeback.c: remove redundant if check 2016-12-12 18:55:08 -08:00
fs_pin.c
fs_struct.c
inode.c Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2016-10-10 20:16:43 -07:00
internal.h iomap: constify struct iomap_ops 2017-01-30 16:32:25 -08:00
ioctl.c vfs: call vfs_clone_file_range() under freeze protection 2016-12-16 11:02:54 +01:00
iomap.c iomap: constify struct iomap_ops 2017-01-30 16:32:25 -08:00
Kconfig dax: fix build warnings with FS_DAX and !FS_IOMAP 2017-01-24 16:26:14 -08:00
Kconfig.binfmt docs: fix locations of several documents that got moved 2016-10-24 08:12:35 -02:00
libfs.c libfs: Modify mount_pseudo_xattr to be clear it is not a userspace mount 2017-01-10 13:34:55 +13:00
locks.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
Makefile logfs: remove from tree 2016-12-14 23:48:11 -05:00
mbcache.c mbcache: document that "find" functions only return reusable entries 2016-12-03 15:55:01 -05:00
mount.h vfs: add path_is_mountpoint() helper 2016-12-03 20:51:35 -05:00
mpage.c fs: Add helper to clean bdev aliases under a bh and use it 2016-11-04 14:34:47 -06:00
namei.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
namespace.c mnt: Protect the mountpoint hashtable with mount_lock 2017-01-10 13:34:43 +13:00
no-block.c
nsfs.c net: add an ioctl to get a socket network namespace 2016-10-31 10:56:36 -04:00
open.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
pipe.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
pnode.c reorganize do_make_slave() 2016-12-16 16:30:49 -05:00
pnode.h mnt: Add a per mount namespace limit on the number of mounts 2016-09-30 12:46:48 -05:00
posix_acl.c tmpfs: clear S_ISGID when setting posix ACLs 2017-01-10 01:29:48 -05:00
proc_namespace.c
read_write.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
readdir.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
select.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
seq_file.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
signalfd.c
splice.c splice: reinstate SIGPIPE/EPIPE handling 2016-12-21 10:59:34 -08:00
stack.c
stat.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
statfs.c vfs: misc struct path constification 2016-12-05 19:03:49 -05:00
super.c quota: Remove dqonoff_mutex 2016-11-30 08:38:07 +01:00
sync.c
timerfd.c ktime: Cleanup ktime_set() usage 2016-12-25 17:21:22 +01:00
userfaultfd.c userfaultfd: fix SIGBUS resulting from false rwsem wakeups 2017-01-24 16:26:14 -08:00
utimes.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
xattr.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00