linux/fs
Josef Bacik 79d3d1d12e btrfs: don't allow large NOWAIT direct reads
Dylan and Jens reported a problem where they had an io_uring test that
was returning short reads, and bisected it to ee5b46a353 ("btrfs:
increase direct io read size limit to 256 sectors").

The root cause is their test was doing larger reads via io_uring with
NOWAIT and async.  This was triggering a page fault during the direct
read, however the first page was able to work just fine and thus we
submitted a 4k read for a larger iocb.

Btrfs allows for partial IO's in this case specifically because we don't
allow page faults, and thus we'll attempt to do any io that we can,
submit what we could, come back and fault in the rest of the range and
try to do the remaining IO.

However for !is_sync_kiocb() we'll call ->ki_complete() as soon as the
partial dio is done, which is incorrect.  In the sync case we can exit
the iomap code, submit more io's, and return with the amount of IO we
were able to complete successfully.

We were always doing short reads in this case, but for NOWAIT we were
getting saved by the fact that we were limiting direct reads to
sectorsize, and if we were larger than that we would return EAGAIN.

Fix the regression by simply returning EAGAIN in the NOWAIT case with
larger reads, that way io_uring can retry and get the larger IO and have
the fault logic handle everything properly.

This still leaves the AIO short read case, but that existed before this
change.  The way to properly fix this would be to handle partial iocb
completions, but that's a lot of work, for now deal with the regression
in the most straightforward way possible.

Reported-by: Dylan Yudaken <dylany@fb.com>
Fixes: ee5b46a353 ("btrfs: increase direct io read size limit to 256 sectors")
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-08-22 18:08:07 +02:00
..
9p 9p: fix EBADF errors in cached mode 2022-06-17 06:03:30 +09:00
adfs fs: Convert block_read_full_page() to block_read_full_folio() 2022-05-09 16:21:44 -04:00
affs affs: Convert affs to read_folio 2022-05-09 16:21:44 -04:00
afs netfs: do not unlock and put the folio twice 2022-07-14 10:10:12 +02:00
autofs
befs befs: Convert befs to read_folio 2022-05-09 16:21:45 -04:00
bfs fs: Convert block_read_full_page() to block_read_full_folio() 2022-05-09 16:21:44 -04:00
btrfs btrfs: don't allow large NOWAIT direct reads 2022-08-22 18:08:07 +02:00
cachefiles cachefiles: narrow the scope of flushed requests when releasing fd 2022-07-05 16:12:21 +01:00
ceph netfs: do not unlock and put the folio twice 2022-07-14 10:10:12 +02:00
cifs smb3: workaround negprot bug in some Samba servers 2022-07-13 19:59:47 -05:00
coda coda: Convert coda to read_folio 2022-05-09 16:21:45 -04:00
configfs configfs: fix a race in configfs_{,un}register_subsystem() 2022-02-22 18:30:28 +01:00
cramfs cramfs: Convert cramfs to read_folio 2022-05-09 16:21:45 -04:00
crypto fscrypt: add new helper functions for test_dummy_encryption 2022-05-09 16:18:54 -07:00
debugfs debugfs: Document that debugfs_create functions need not be error checked 2022-02-25 11:56:13 +01:00
devpts fsnotify: fix fsnotify hooks in pseudo filesystems 2022-01-24 14:17:02 +01:00
dlm dlm: use kref_put_lock in __put_lkb 2022-05-02 11:23:49 -05:00
ecryptfs ecryptfs: Convert ecryptfs to read_folio 2022-05-09 16:21:45 -04:00
efivarfs
efs efs: Convert efs symlinks to read_folio 2022-05-09 16:21:45 -04:00
erofs Changes since last update: 2022-06-01 11:54:29 -07:00
exfat exfat: use updated exfat_chain directly during renaming 2022-06-09 21:26:32 +09:00
exportfs exportfs: support idmapped mounts 2022-04-28 16:31:10 +02:00
ext2 ext2: fix fs corruption when trying to remove a non-empty directory with IO error 2022-06-16 10:55:45 +02:00
ext4 ext4: fix a doubled word "need" in a comment 2022-06-18 19:36:20 -04:00
f2fs f2fs: do not count ENOENT for error case 2022-06-21 08:29:56 -07:00
fat Not a lot of material this cycle. Many singleton patches against various 2022-05-27 11:22:03 -07:00
freevxfs SPDX changes for 5.19-rc1 2022-06-03 10:34:34 -07:00
fscache fscache: Fix invalidation/lookup race 2022-07-05 16:12:55 +01:00
fuse libnvdimm for 5.19 2022-05-27 15:49:30 -07:00
gfs2 Page cache changes for 5.19 2022-05-24 19:55:07 -07:00
hfs fs: Change try_to_free_buffers() to take a folio 2022-05-09 23:12:34 -04:00
hfsplus fs: Change try_to_free_buffers() to take a folio 2022-05-09 23:12:34 -04:00
hostfs hostfs: Convert hostfs to read_folio 2022-05-09 16:21:45 -04:00
hpfs hpfs: Convert symlinks to read_folio 2022-05-09 16:21:45 -04:00
hugetlbfs hugetlbfs: zero partial pages during fallocate hole punch 2022-06-16 19:11:32 -07:00
iomap Page cache changes for 5.19 2022-05-24 19:55:07 -07:00
isofs isofs: Convert symlinks and zisofs to read_folio 2022-05-09 16:21:45 -04:00
jbd2 fs: fix jbd2_journal_try_to_free_buffers() kernel-doc comment 2022-06-16 10:36:09 -04:00
jffs2 This pull request contains fixes for JFFS2, UBI and UBIFS 2022-06-03 14:42:24 -07:00
jfs JFS: One bug fix and some code cleanup 2022-05-27 15:59:21 -07:00
kernfs kernfs: Separate kernfs_pr_cont_buf and rename_lock. 2022-05-19 19:37:06 +02:00
ksmbd vfs: fix copy_file_range() regression in cross-fs copies 2022-06-30 15:16:38 -07:00
lockd lockd: fix nlm_close_files 2022-07-11 15:49:56 -04:00
minix fs: Convert block_read_full_page() to block_read_full_folio() 2022-05-09 16:21:44 -04:00
netfs netfs: do not unlock and put the folio twice 2022-07-14 10:10:12 +02:00
nfs NFSv4: Add an fattr allocation to _nfs4_discover_trunking() 2022-06-30 16:13:00 -04:00
nfs_common
nfsd Notable regression fixes: 2022-07-14 12:29:43 -07:00
nilfs2 nilfs2: fix incorrect masking of permission flags for symlinks 2022-07-03 15:42:33 -07:00
nls
notify fanotify: refine the validation checks on non-dir inode mask 2022-06-28 11:18:13 +02:00
ntfs Not a lot of material this cycle. Many singleton patches against various 2022-05-27 11:22:03 -07:00
ntfs3 Ntfs3 for 5.19 2022-06-03 16:57:16 -07:00
ocfs2 Not a lot of material this cycle. Many singleton patches against various 2022-05-27 11:22:03 -07:00
omfs fs: Convert block_read_full_page() to block_read_full_folio() 2022-05-09 16:21:44 -04:00
openpromfs fs: allocate inode by using alloc_inode_sb() 2022-03-22 15:57:03 -07:00
orangefs orangefs: Convert to free_folio 2022-05-09 23:12:53 -04:00
overlayfs ovl: turn of SB_POSIXACL with idmapped layers temporarily 2022-07-08 15:48:31 +02:00
proc Not a lot of material this cycle. Many singleton patches against various 2022-05-27 11:22:03 -07:00
pstore pstore: Don't use semaphores in always-atomic-context code 2022-03-15 11:08:23 -07:00
qnx4 fs: Convert block_read_full_page() to block_read_full_folio() 2022-05-09 16:21:44 -04:00
qnx6 fs: Convert mpage_readpage to mpage_read_folio 2022-05-09 16:21:44 -04:00
quota quota: Prevent memory allocation recursion while holding dq_lock 2022-06-06 10:08:10 +02:00
ramfs Merge branch 'akpm' (patches from Andrew) 2021-11-09 10:11:53 -08:00
reiserfs fs: Change try_to_free_buffers() to take a folio 2022-05-09 23:12:34 -04:00
romfs romfs: Convert romfs to read_folio 2022-05-09 16:21:46 -04:00
smbfs_common Add various fsctl structs 2022-05-23 20:24:12 -05:00
squashfs Page cache changes for 5.19 2022-05-24 19:55:07 -07:00
sysfs kobject: kobj_type: remove default_attrs 2022-04-05 15:39:19 +02:00
sysv Not a lot of material this cycle. Many singleton patches against various 2022-05-27 11:22:03 -07:00
tracefs tracefs: Fix syntax errors in comments 2022-06-17 19:01:28 -04:00
ubifs This pull request contains fixes for JFFS2, UBI and UBIFS 2022-06-03 14:42:24 -07:00
udf Page cache changes for 5.19 2022-05-24 19:55:07 -07:00
ufs fs: Convert block_read_full_page() to block_read_full_folio() 2022-05-09 16:21:44 -04:00
unicode kbuild: unify cmd_copy and cmd_shipped 2022-02-14 10:37:32 +09:00
vboxsf vboxsf: Convert vboxsf to read_folio 2022-05-09 16:21:46 -04:00
verity Page cache changes for 5.19 2022-05-24 19:55:07 -07:00
xfs xfs: prevent a UAF when log IO errors race with unmount 2022-07-01 09:09:52 -07:00
zonefs zonefs: fix zonefs_iomap_begin() for reads 2022-06-08 19:13:55 +09:00
aio.c Merge branch 'work.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2022-04-01 19:57:03 -07:00
anon_inodes.c
attr.c fs: account for group membership 2022-06-14 12:18:47 +02:00
bad_inode.c
binfmt_aout.c
binfmt_elf_fdpic.c coredump: Snapshot the vmas in do_coredump 2022-03-08 12:55:29 -06:00
binfmt_elf_test.c binfmt_elf: Introduce KUnit test 2022-03-03 20:38:56 -08:00
binfmt_elf.c revert "fs/binfmt_elf: use PT_LOAD p_align values for static PIE" 2022-04-15 14:49:56 -07:00
binfmt_flat.c binfmt_flat: Remove shared library support 2022-04-22 10:57:18 -07:00
binfmt_misc.c Fix regression due to "fs: move binfmt_misc sysctl to its own file" 2022-02-09 09:50:02 -08:00
binfmt_script.c
buffer.c fs: Convert drop_buffers() to use a folio 2022-05-09 23:12:34 -04:00
char_dev.c
compat_binfmt_elf.c binfmt_elf: Introduce KUnit test 2022-03-03 20:38:56 -08:00
coredump.c ptrace: Cleanups for v5.18 2022-03-28 17:29:53 -07:00
d_path.c d_path: fix Kernel doc validator complaining 2021-11-06 13:30:32 -07:00
dax.c libnvdimm for 5.19 2022-05-27 15:49:30 -07:00
dcache.c mm: dcache: use kmem_cache_alloc_lru() to allocate dentry 2022-03-22 15:57:03 -07:00
direct-io.c direct-io: remove random prefetches 2022-04-17 19:50:02 -06:00
drop_caches.c
eventfd.c
eventpoll.c eventpoll: simplify sysctl declaration with register_sysctl() 2022-01-22 08:33:35 +02:00
exec.c fix race between exit_itimers() and /proc/pid/timers 2022-07-11 09:52:59 -07:00
fcntl.c VFS: add FMODE_CAN_ODIRECT file flag 2022-05-09 18:20:49 -07:00
fhandle.c
file_table.c Descriptor handling cleanups 2022-06-04 18:52:00 -07:00
file.c fix the breakage in close_fd_get_file() calling conventions change 2022-06-05 15:03:03 -04:00
filesystems.c
fs_context.c vfs: fs_context: fix up param length parsing in legacy_parse_param 2022-01-18 09:23:19 +02:00
fs_parser.c fs_parse: allow parameter value to be empty 2021-12-09 14:09:36 -05:00
fs_pin.c
fs_struct.c
fs_types.c
fs-writeback.c writeback: Fix inode->i_io_list not be protected by inode->i_lock error 2022-06-06 09:54:30 +02:00
fsopen.c uninline may_mount() and don't opencode it in fspick(2)/fsopen(2) 2022-05-19 23:25:10 -04:00
init.c
inode.c writeback: Fix inode->i_io_list not be protected by inode->i_lock error 2022-06-06 09:54:30 +02:00
internal.h Cleanups (and one fix) around struct mount handling. 2022-06-04 19:00:05 -07:00
io_uring.c io_uring: do not recycle buffer in READV 2022-07-21 08:31:31 -06:00
io-wq.c io-wq: use __set_notify_signal() to wake workers 2022-04-30 08:39:54 -06:00
io-wq.h io_uring: add support for IORING_ASYNC_CANCEL_ALL 2022-04-24 18:18:18 -06:00
ioctl.c Fixes for 5.18-rc1: 2022-04-01 19:35:56 -07:00
Kconfig mm: hugetlb_vmemmap: cleanup CONFIG_HUGETLB_PAGE_FREE_VMEMMAP* 2022-04-28 23:16:15 -07:00
Kconfig.binfmt m68knommu: changes for linux 5.19 2022-05-30 10:56:18 -07:00
kernel_read_file.c
libfs.c fs: Convert simple_readpage to simple_read_folio 2022-05-09 16:21:44 -04:00
locks.c fs/lock: add 2 callbacks to lock_manager_operations to resolve conflict 2022-05-19 12:25:39 -04:00
Makefile Fix from Christoph Hellwig merging the CONFIG_UNICODE_UTF8_DATA into the 2022-02-01 11:13:24 -08:00
mbcache.c
mount.h
mpage.c fs: Change try_to_free_buffers() to take a folio 2022-05-09 23:12:34 -04:00
namei.c Several cleanups in fs/namei.c. 2022-06-04 19:07:15 -07:00
namespace.c Cleanups (and one fix) around struct mount handling. 2022-06-04 19:00:05 -07:00
no-block.c
nsfs.c
open.c RISC-V Patches for the 5.19 Merge Window, Part 1 2022-05-31 14:10:54 -07:00
pipe.c Not a lot of material this cycle. Many singleton patches against various 2022-05-27 11:22:03 -07:00
pnode.c
pnode.h
posix_acl.c fs: fix acl translation 2022-04-19 10:19:02 -07:00
proc_namespace.c fs: add is_idmapped_mnt() helper 2021-12-03 18:44:06 +01:00
read_write.c vfs: fix copy_file_range() regression in cross-fs copies 2022-06-30 15:16:38 -07:00
readdir.c
remap_range.c Revert "vf/remap: return the amount of bytes actually deduplicated" 2022-07-14 15:35:24 -07:00
select.c select: Fix indefinitely sleeping task in poll_schedule_timeout() 2022-01-11 09:03:05 -08:00
seq_file.c rxrpc: Fix locking issue 2022-05-22 21:03:01 +01:00
signalfd.c Merge branch 'signal-for-v5.17' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace 2022-01-17 05:49:30 +02:00
splice.c mm: Convert remove_mapping() to take a folio 2022-03-21 12:59:01 -04:00
stack.c
stat.c RISC-V Patches for the 5.19 Merge Window, Part 1 2022-05-31 14:10:54 -07:00
statfs.c
super.c block: add a bdev_stable_writes helper 2022-04-17 19:49:59 -06:00
sync.c riscv: compat: syscall: Add compat_sys_call_table implementation 2022-04-26 13:36:25 -07:00
sysctls.c fs: move namespace sysctls and declare fs base directory 2022-01-22 08:33:36 +02:00
timerfd.c
userfaultfd.c mm/uffd: enable write protection for shmem & hugetlbfs 2022-05-13 07:20:11 -07:00
utimes.c
xattr.c fs: split off do_getxattr from getxattr 2022-04-24 18:18:37 -06:00