linux-stable/fs
Qu Wenruo 0391c9085a btrfs: do not wait for short bulk allocation
commit 1db7959aac upstream.

[BUG]
There is a recent report that when memory pressure is high (including
cached pages), btrfs can spend most of its time on memory allocation in
btrfs_alloc_page_array() for compressed read/write.

[CAUSE]
For btrfs_alloc_page_array() we always go alloc_pages_bulk_array(), and
even if the bulk allocation failed (fell back to single page
allocation) we still retry but with extra memalloc_retry_wait().

If the bulk alloc only returned one page a time, we would spend a lot of
time on the retry wait.

The behavior was introduced in commit 395cb57e85 ("btrfs: wait between
incomplete batch memory allocations").

[FIX]
Although the commit mentioned that other filesystems do the wait, it's
not the case at least nowadays.

All the mainlined filesystems only call memalloc_retry_wait() if they
failed to allocate any page (not only for bulk allocation).
If there is any progress, they won't call memalloc_retry_wait() at all.

For example, xfs_buf_alloc_pages() would only call memalloc_retry_wait()
if there is no allocation progress at all, and the call is not for
metadata readahead.

So I don't believe we should call memalloc_retry_wait() unconditionally
for short allocation.

Call memalloc_retry_wait() if it fails to allocate any page for tree
block allocation (which goes with __GFP_NOFAIL and may not need the
special handling anyway), and reduce the latency for
btrfs_alloc_page_array().

Reported-by: Julian Taylor <julian.taylor@1und1.de>
Tested-by: Julian Taylor <julian.taylor@1und1.de>
Link: https://lore.kernel.org/all/8966c095-cbe7-4d22-9784-a647d1bf27c3@1und1.de/
Fixes: 395cb57e85 ("btrfs: wait between incomplete batch memory allocations")
CC: stable@vger.kernel.org # 6.1+
Reviewed-by: Sweet Tea Dorminy <sweettea-kernel@dorminy.me>
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2024-05-17 11:56:23 +02:00
..
9p fs/9p: drop inodes immediately on non-.L too 2024-05-17 11:56:09 +02:00
adfs
affs affs: initialize fsdata in affs_truncate() 2023-02-01 08:34:08 +01:00
afs afs: Revert "afs: Hide silly-rename files from userspace" 2024-03-26 18:20:57 -04:00
autofs autofs: fix memory leak of waitqueues in autofs_catatonic_mode 2023-09-23 11:10:59 +02:00
befs
bfs
btrfs btrfs: do not wait for short bulk allocation 2024-05-17 11:56:23 +02:00
cachefiles cachefiles: fix memory leak in cachefiles_add_cache() 2024-03-01 13:26:32 +01:00
ceph ceph: switch to corrected encoding of max_xattr_size in mdsmap 2024-03-15 10:48:13 -04:00
coda coda: Avoid partial allocation of sig_inputArgs 2023-03-10 09:33:52 +01:00
configfs configfs: fix possible memory leak in configfs_create_dir() 2022-12-31 13:32:22 +01:00
cramfs fs/cramfs/inode.c: initialize file_ra_state 2023-03-10 09:34:09 +01:00
crypto blk-crypto: add a blk_crypto_config_supported_natively helper 2023-05-11 23:03:00 +09:00
debugfs debugfs: fix automount d_fsdata usage 2024-01-20 11:50:04 +01:00
devpts
dlm fs: dlm: don't put dlm_local_addrs on heap 2024-02-16 19:06:29 +01:00
ecryptfs ecryptfs: Reject casefold directory inodes 2024-02-05 20:12:49 +00:00
efivarfs efivarfs: Request at most 512 bytes for variable names 2024-03-06 14:45:11 +00:00
efs
erofs erofs: apply proper VMA alignment for memory mapped files on THP 2024-03-15 10:48:16 -04:00
exfat exfat: support handle zero-size directory 2023-11-28 17:07:00 +00:00
exportfs exportfs: use pr_debug for unreachable debug statements 2024-03-06 14:45:15 +00:00
ext2 ext2: fix datatype of block number in ext2_xattr_set2() 2023-09-23 11:11:05 +02:00
ext4 ext4: forbid commit inconsistent quota data when errors=remount-ro 2024-04-13 13:05:09 +02:00
f2fs f2fs: truncate page cache before clearing flags when aborting atomic write 2024-04-03 15:19:37 +02:00
fat fat: fix uninitialized field in nostale filehandles 2024-04-03 15:19:27 +02:00
freevxfs
fscache netfs, fscache: Prevent Oops in fscache_put_cache() 2024-01-31 16:17:04 -08:00
fuse fuse: don't unhash root 2024-04-03 15:19:30 +02:00
gfs2 gfs2: Fix invalid metadata access in punch_hole 2024-05-17 11:56:04 +02:00
hfs hfs: fix missing hfs_bnode_get() in __hfs_bnode_create 2023-03-10 09:34:07 +01:00
hfsplus fs: hfsplus: remove WARN_ON() from hfsplus_cat_{read,write}_inode() 2023-05-24 17:32:34 +01:00
hostfs
hpfs
hugetlbfs mm/hugetlb: add hugetlb_folio_subpool() helpers 2024-05-17 11:55:51 +02:00
iomap iomap: update ki_pos a little later in iomap_dio_complete 2023-12-08 08:51:20 +01:00
isofs isofs: handle CDs with bad root inode but good Joliet root directory 2024-04-13 13:05:06 +02:00
jbd2 jbd2: fix soft lockup in journal_finish_inode_data_buffers() 2024-01-20 11:50:07 +01:00
jffs2 jffs2: reduce stack usage in jffs2_build_xattr_subsystem() 2023-07-19 16:22:11 +02:00
jfs jfs: fix array-index-out-of-bounds in diNewExt 2024-02-05 20:12:48 +00:00
kernfs fs/kernfs/dir: obey S_ISGID 2024-02-05 20:12:58 +00:00
lockd lockd: fix file selection in nlmsvc_cancel_blocked 2024-03-06 14:45:16 +00:00
minix
netfs netfs: Only call folio_start_fscache() one time for each folio 2023-10-06 14:56:32 +02:00
nfs nfs: Handle error of rpc_proc_register() in nfs_net_init(). 2024-05-17 11:55:54 +02:00
nfs_common
nfsd nfsd: hold a lighter-weight client reference over CB_RECALL_ANY 2024-04-10 16:28:32 +02:00
nilfs2 nilfs2: fix OOB in nilfs_set_de_type 2024-04-27 17:07:17 +02:00
nls fs/nls: make load_nls() take a const parameter 2023-09-13 09:42:22 +02:00
notify fanotify: disallow mount/sb marks on kernel internal pseudo fs 2023-07-19 16:22:05 +02:00
ntfs - hfs and hfsplus kmap API modernization from Fabio Francesco 2022-10-12 11:00:22 -07:00
ntfs3 fs/ntfs3: Fix NULL pointer dereference in 'ni_write_inode' 2024-03-06 14:45:06 +00:00
ocfs2 fs: ocfs2: namei: check return value of ocfs2_add_entry() 2023-09-13 09:42:33 +02:00
omfs
openpromfs
orangefs Julia Lawall reported this null pointer dereference, this should fix it. 2024-04-13 13:05:08 +02:00
overlayfs ima: detect changes to the backing overlay file 2023-11-28 17:07:12 +00:00
proc fs/proc: do_task_stat: use sig->stats_lock to gather the threads/children stats 2024-03-15 10:48:22 -04:00
pstore pstore/zone: Add a null pointer check to the psz_kmsg_read 2024-04-13 13:04:57 +02:00
qnx4
qnx6
quota quota: Fix rcu annotations of inode dquot pointers 2024-03-26 18:20:47 -04:00
ramfs shmem: use ramfs_kill_sb() for kill_sb method of ramfs-based tmpfs 2023-07-19 16:22:11 +02:00
reiserfs reiserfs: Check the return value from __getblk() 2023-09-13 09:42:27 +02:00
romfs
smb ksmbd: do not grant v2 lease if parent lease key and epoch are not set 2024-05-17 11:56:22 +02:00
squashfs revert "squashfs: harden sanity check in squashfs_read_xattr_id_table" 2023-02-22 12:59:50 +01:00
sysfs fs: sysfs: Fix reference leak in sysfs_break_active_protection() 2024-04-27 17:07:16 +02:00
sysv sysv: don't call sb_bread() with pointers_lock held 2024-04-13 13:05:05 +02:00
tracefs tracefs: Add missing lockdown check to tracefs_create_dir() 2023-09-23 11:11:12 +02:00
ubifs ubifs: Set page uptodate in the correct place 2024-04-03 15:19:27 +02:00
udf udf: initialize newblock to 0 2023-09-13 09:43:05 +02:00
ufs
unicode
vboxsf vboxsf: Avoid an spurious warning if load_nls_xxx() fails 2024-04-10 16:28:25 +02:00
verity fsverity: skip PKCS#7 parser when keyring is empty 2023-09-13 09:43:03 +02:00
xfs xfs: read only mounts with fsopen mount API are busted 2024-01-31 16:17:08 -08:00
zonefs zonefs: Improve error handling 2024-02-23 09:12:45 +01:00
Kconfig smb: move client and server files to common directory fs/smb 2023-06-28 11:12:40 +02:00
Kconfig.binfmt Xtensa updates for v6.1 2022-10-10 14:21:11 -07:00
Makefile smb: move client and server files to common directory fs/smb 2023-06-28 11:12:40 +02:00
aio.c fs/aio: Check IOCB_AIO_RW before the struct aio_kiocb conversion 2024-04-03 15:19:42 +02:00
anon_inodes.c
attr.c attr: block mode changes of symlinks 2023-09-23 11:11:10 +02:00
bad_inode.c
binfmt_elf.c mm: always expand the stack with the mmap write lock held 2023-07-01 13:16:25 +02:00
binfmt_elf_fdpic.c fs: binfmt_elf_efpic: fix personality for ELF-FDPIC 2023-10-06 14:57:06 +02:00
binfmt_elf_test.c
binfmt_flat.c
binfmt_misc.c binfmt_misc: fix shift-out-of-bounds in check_special_flags 2022-12-31 13:32:57 +01:00
binfmt_script.c
buffer.c - hfs and hfsplus kmap API modernization from Fabio Francesco 2022-10-12 11:00:22 -07:00
char_dev.c chardev: fix error handling in cdev_device_add() 2022-12-31 13:32:41 +01:00
compat_binfmt_elf.c
coredump.c coredump: Move dump_emit_page() to kill unused warning 2023-02-22 12:59:50 +01:00
d_path.c
dax.c
dcache.c fast_dput(): handle underflows gracefully 2024-02-05 20:12:54 +00:00
direct-io.c
drop_caches.c
eventfd.c eventfd: prevent underflow for eventfd semaphores 2023-09-13 09:42:27 +02:00
eventpoll.c epoll: ep_autoremove_wake_function should use list_del_init_careful 2023-06-21 16:00:54 +02:00
exec.c exec: Fix NOMMU linux_binprm::exec in transfer_args_to_stack() 2024-04-03 15:19:48 +02:00
fcntl.c fs: Fix rw_hint validation 2024-03-26 18:20:28 -04:00
fhandle.c do_sys_name_to_handle(): use kzalloc() to fix kernel-infoleak 2024-03-26 18:20:27 -04:00
file.c file: reinstate f_pos locking optimization for regular files 2023-08-11 12:08:23 +02:00
file_table.c
filesystems.c
fs-writeback.c writeback, cgroup: switch inodes with dirty timestamps to release dying cgwbs 2023-11-20 11:51:50 +01:00
fs_context.c vfs, security: Fix automount superblock LSM init problem, preventing NFS sb sharing 2023-09-13 09:42:28 +02:00
fs_parser.c ext4: journal_path mount options should follow links 2023-01-07 11:11:59 +01:00
fs_pin.c
fs_struct.c
fs_types.c
fsopen.c
init.c
inode.c filemap: add a per-mapping stable writes flag 2024-01-10 17:10:32 +01:00
internal.h nfs: use vfs setgid helper 2023-08-30 16:11:10 +02:00
ioctl.c lsm: new security_file_ioctl_compat() hook 2024-01-31 16:17:00 -08:00
kernel_read_file.c
libfs.c libfs: add DEFINE_SIMPLE_ATTRIBUTE_SIGNED for signed value 2022-12-31 13:31:58 +01:00
locks.c filelock: add a new locks_inode_context accessor function 2024-03-06 14:45:16 +00:00
mbcache.c ext4: fix deadlock due to mbcache entry corruption 2023-01-07 11:12:02 +01:00
mount.h
mpage.c
namei.c rename(): fix the locking of subdirectories 2024-01-31 16:17:02 -08:00
namespace.c fs: relax mount_setattr() permission checks 2024-02-23 09:12:46 +01:00
no-block.c
nsfs.c
open.c open: make RESOLVE_CACHED correctly test for O_TMPFILE 2023-08-11 12:08:22 +02:00
pipe.c fs/pipe: Fix lockdep false-positive in watchqueue pipe_write() 2024-04-10 16:28:30 +02:00
pnode.c pnode: terminate at peers of source 2023-01-04 11:29:01 +01:00
pnode.h
posix_acl.c - Yu Zhao's Multi-Gen LRU patches are here. They've been under test in 2022-10-10 17:53:04 -07:00
proc_namespace.c
read_write.c use less confusing names for iov_iter direction initializers 2023-02-09 11:28:04 +01:00
readdir.c
remap_range.c
select.c fs/select: rework stack allocation hack for clang 2024-03-26 18:20:28 -04:00
seq_file.c use less confusing names for iov_iter direction initializers 2023-02-09 11:28:04 +01:00
signalfd.c
splice.c mm: merge folio_has_private()/filemap_release_folio() call pairs 2024-01-10 17:10:31 +01:00
stack.c
stat.c
statfs.c statfs: enforce statfs[64] structure initialization 2023-05-24 17:32:51 +01:00
super.c fs: Protect reconfiguration of sb read-write from racing writes 2023-08-11 12:08:24 +02:00
sync.c
sysctls.c
timerfd.c
userfaultfd.c Revert "userfaultfd: don't fail on unrecognized features" 2023-04-26 14:28:37 +02:00
utimes.c
xattr.c fs: don't audit the capability check in simple_xattr_list() 2022-12-31 13:31:55 +01:00