kernel_optimize_test/fs
Jens Axboe dd3db2a34c io_uring: drop file set ref put/get on switch
Dan reports that he triggered a warning on ring exit doing some testing:

percpu ref (io_file_data_ref_zero) <= 0 (0) after switching to atomic
WARNING: CPU: 3 PID: 0 at lib/percpu-refcount.c:160 percpu_ref_switch_to_atomic_rcu+0xe8/0xf0
Modules linked in:
CPU: 3 PID: 0 Comm: swapper/3 Not tainted 5.6.0-rc3+ #5648
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1ubuntu1 04/01/2014
RIP: 0010:percpu_ref_switch_to_atomic_rcu+0xe8/0xf0
Code: e7 ff 55 e8 eb d2 80 3d bd 02 d2 00 00 75 8b 48 8b 55 d8 48 c7 c7 e8 70 e6 81 c6 05 a9 02 d2 00 01 48 8b 75 e8 e8 3a d0 c5 ff <0f> 0b e9 69 ff ff ff 90 55 48 89 fd 53 48 89 f3 48 83 ec 28 48 83
RSP: 0018:ffffc90000110ef8 EFLAGS: 00010292
RAX: 0000000000000045 RBX: 7fffffffffffffff RCX: 0000000000000000
RDX: 0000000000000045 RSI: ffffffff825be7a5 RDI: ffffffff825bc32c
RBP: ffff8881b75eac38 R08: 000000042364b941 R09: 0000000000000045
R10: ffffffff825beb40 R11: ffffffff825be78a R12: 0000607e46005aa0
R13: ffff888107dcdd00 R14: 0000000000000000 R15: 0000000000000009
FS:  0000000000000000(0000) GS:ffff8881b9d80000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f49e6a5ea20 CR3: 00000001b747c004 CR4: 00000000001606e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <IRQ>
 rcu_core+0x1e4/0x4d0
 __do_softirq+0xdb/0x2f1
 irq_exit+0xa0/0xb0
 smp_apic_timer_interrupt+0x60/0x140
 apic_timer_interrupt+0xf/0x20
 </IRQ>
RIP: 0010:default_idle+0x23/0x170
Code: ff eb ab cc cc cc cc 0f 1f 44 00 00 41 54 55 53 65 8b 2d 10 96 92 7e 0f 1f 44 00 00 e9 07 00 00 00 0f 00 2d 21 d0 51 00 fb f4 <65> 8b 2d f6 95 92 7e 0f 1f 44 00 00 5b 5d 41 5c c3 65 8b 05 e5 95

Turns out that this is due to percpu_ref_switch_to_atomic() only
grabbing a reference to the percpu refcount if it's not already in
atomic mode. io_uring drops a ref and re-gets it when switching back to
percpu mode. We attempt to protect against this with the FFD_F_ATOMIC
bit, but that isn't reliable.

We don't actually need to juggle these refcounts between atomic and
percpu switch, we can just do them when we've switched to atomic mode.
This removes the need for FFD_F_ATOMIC, which wasn't reliable.

Fixes: 05f3fb3c53 ("io_uring: avoid ring quiesce for fixed file set unregister and update")
Reported-by: Dan Melnic <dmm@fb.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
2020-02-26 10:53:33 -07:00
..
9p
adfs fs/adfs: bigdir: Fix an error code in adfs_fplus_read() 2020-01-25 11:31:59 -05:00
affs affs: fix a memory leak in affs_remount 2019-11-18 14:26:43 +01:00
afs Merge branch 'merge.nfs-fs_parse.1' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2020-02-08 13:26:41 -08:00
autofs Merge branch 'next.autofs' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2019-12-05 17:11:48 -08:00
befs
bfs
btrfs for-5.6-rc2-tag 2020-02-23 09:43:50 -08:00
cachefiles cachefiles: drop direct usage of ->bmap method. 2020-02-03 08:05:56 -05:00
ceph ceph: noacl mount option is effectively ignored 2020-02-11 17:04:40 +01:00
cifs cifs: make sure we do not overflow the max EA buffer size 2020-02-14 11:10:24 -06:00
coda
configfs utimes: Clamp the timestamps in notify_change() 2019-12-08 19:10:50 -05:00
cramfs cramfs: switch to use of errofc() et.al. 2020-02-07 14:48:41 -05:00
crypto fscrypt: improve format of no-key names 2020-01-22 14:50:03 -08:00
debugfs Merge branch 'work.recursive_removal' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2020-02-05 05:09:46 +00:00
devpts
dlm dlm: use SO_SNDTIMEO_NEW instead of SO_SNDTIMEO_OLD 2019-12-18 18:07:31 +01:00
ecryptfs eCryptfs fixes for 5.6-rc3 2020-02-17 21:08:37 -08:00
efivarfs
efs
erofs erofs: clean up z_erofs_submit_queue() 2020-01-21 16:46:23 +08:00
exportfs race in exportfs_decode_fh() 2019-11-11 09:21:59 -05:00
ext2 dax fixes 5.6-rc1 2020-02-11 16:52:08 -08:00
ext4 ext4: fix mount failure with quota configured as module 2020-02-21 19:32:07 -05:00
f2fs Merge branch 'work.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2020-02-08 13:04:49 -08:00
fat Merge branch 'imm.timestamp' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2020-02-05 05:02:42 +00:00
freevxfs
fscache proc: convert everything to "struct proc_ops" 2020-02-04 03:05:26 +00:00
fuse Merge branch 'merge.nfs-fs_parse.1' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2020-02-08 13:26:41 -08:00
gfs2 Merge branch 'merge.nfs-fs_parse.1' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2020-02-08 13:26:41 -08:00
hfs hfs/hfsplus: use 64-bit inode timestamps 2019-12-18 18:07:32 +01:00
hfsplus hfs/hfsplus: use 64-bit inode timestamps 2019-12-18 18:07:32 +01:00
hostfs hostfs: pass 64-bit timestamps to/from user space 2019-12-18 18:07:32 +01:00
hpfs fs: compat_ioctl: move FITRIM emulation into file systems 2019-10-23 17:23:46 +02:00
hugetlbfs hugetlbfs: switch to use of invalfc() 2020-02-07 14:48:42 -05:00
iomap fs: Fix page_mkwrite off-by-one errors 2020-01-06 08:58:23 -08:00
isofs
jbd2 jbd2: fix ocfs2 corrupt when clearing block group bits 2020-02-21 19:32:07 -05:00
jffs2 fs_parse: fold fs_parameter_desc/fs_parameter_spec 2020-02-07 14:48:37 -05:00
jfs Trivial cleanup for jfs 2020-02-05 05:28:20 +00:00
kernfs Merge branch 'imm.timestamp' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2020-02-05 05:02:42 +00:00
lockd proc: convert everything to "struct proc_ops" 2020-02-04 03:05:26 +00:00
minix
nfs NFSv4: Ensure the delegation cred is pinned when we call delegreturn 2020-02-13 16:23:02 -05:00
nfs_common
nfsd Highlights: 2020-02-07 17:50:21 -08:00
nilfs2 fs: compat_ioctl: move FITRIM emulation into file systems 2019-10-23 17:23:46 +02:00
nls
notify fs: call fsnotify_sb_delete after evict_inodes 2019-12-18 00:03:01 -05:00
ntfs utimes: Clamp the timestamps in notify_change() 2019-12-08 19:10:50 -05:00
ocfs2 treewide: remove redundant IS_ERR() before error code check 2020-02-04 03:05:27 +00:00
omfs
openpromfs
orangefs help_next should increase position index 2020-02-04 15:22:04 -05:00
overlayfs ovl: fix lseek overflow on 32bit 2020-02-03 11:41:53 +01:00
proc Merge branch 'merge.nfs-fs_parse.1' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2020-02-08 13:26:41 -08:00
pstore pstore/ram: Regularize prz label allocation lifetime 2020-01-08 17:05:45 -08:00
qnx4
qnx6
quota \n 2020-01-30 15:37:41 -08:00
ramfs fs_parse: fold fs_parameter_desc/fs_parameter_spec 2020-02-07 14:48:37 -05:00
reiserfs Merge branch 'akpm' (patches from Andrew) 2020-01-31 12:16:36 -08:00
romfs
squashfs
sysfs treewide: remove redundant IS_ERR() before error code check 2020-02-04 03:05:27 +00:00
sysv
tracefs simple_recursive_removal(): kernel-side rm -rf for ramfs-style filesystems 2019-12-10 22:29:58 -05:00
ubifs Merge branch 'imm.timestamp' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2020-02-05 05:02:42 +00:00
udf udf: Clarify meaning of f_files in udf_statfs 2020-01-20 13:59:41 +01:00
ufs
unicode kbuild: rename hostprogs-y/always to hostprogs/always-y 2020-02-04 01:53:07 +09:00
vboxsf fs: Add VirtualBox guest shared folder (vboxsf) support 2020-02-08 17:34:58 -05:00
verity fs-verity: use u64_to_user_ptr() 2020-01-14 13:28:28 -08:00
xfs dax fixes 5.6-rc1 2020-02-11 16:52:08 -08:00
zonefs fs: New zonefs file system 2020-02-07 14:39:38 +09:00
aio.c aio: prevent potential eventfd recursion on poll 2020-02-03 17:27:47 -07:00
anon_inodes.c
attr.c utimes: Clamp the timestamps in notify_change() 2019-12-08 19:10:50 -05:00
bad_inode.c
binfmt_aout.c
binfmt_elf_fdpic.c y2038: elfcore: Use __kernel_old_timeval for process times 2019-11-15 14:38:29 +01:00
binfmt_elf.c fs/binfmt_elf.c: coredump: allow process with empty address space to coredump 2020-01-31 10:30:41 -08:00
binfmt_em86.c
binfmt_flat.c
binfmt_misc.c
binfmt_script.c
block_dev.c block: don't send uevent for empty disk when not invalidating 2019-12-02 18:49:30 -07:00
buffer.c smp: Remove allocation mask from on_each_cpu_cond.*() 2020-01-24 20:40:09 +01:00
char_dev.c chardev: Avoid potential use-after-free in 'chrdev_open()' 2020-01-06 20:10:26 +01:00
compat_binfmt_elf.c y2038: elfcore: Use __kernel_old_timeval for process times 2019-11-15 14:38:29 +01:00
compat.c
coredump.c pipe: use exclusive waits when reading or writing 2020-02-08 11:39:19 -08:00
d_path.c
dax.c dax: pass NOWAIT flag to iomap_apply 2020-02-05 20:34:32 -08:00
dcache.c Merge branch 'work.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2019-12-08 11:08:28 -08:00
dcookies.c
direct-io.c fs/direct-io.c: include fs/internal.h for missing prototype 2020-01-04 13:55:09 -08:00
drop_caches.c fs: avoid softlockups in s_inodes iterators 2019-12-18 00:03:01 -05:00
eventfd.c eventfd: track eventfd_signal() recursion depth 2020-02-03 17:27:38 -07:00
eventpoll.c eventpoll: support non-blocking do_epoll_ctl() calls 2020-01-29 15:45:47 -07:00
exec.c Merge branch 'akpm' (patches from Andrew) 2020-01-31 12:16:36 -08:00
fcntl.c Merge branch 'work.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2019-12-08 11:08:28 -08:00
fhandle.c
file_table.c
file.c threads-v5.6 2020-01-29 19:38:34 -08:00
filesystems.c fs_parser: remove fs_parameter_description name field 2020-02-07 14:48:36 -05:00
fs_context.c add prefix to fs_context->log 2020-02-07 14:48:35 -05:00
fs_parser.c turn fs_param_is_... into functions 2020-02-07 14:48:38 -05:00
fs_pin.c
fs_struct.c
fs_types.c
fs-writeback.c memcg: fix a crash in wb_workfn when a device disappears 2020-01-31 10:30:36 -08:00
fsopen.c add prefix to fs_context->log 2020-02-07 14:48:35 -05:00
inode.c Merge branch 'work.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2020-02-08 13:04:49 -08:00
internal.h for-5.6/io_uring-vfs-2020-01-29 2020-01-29 18:53:37 -08:00
io_uring.c io_uring: drop file set ref put/get on switch 2020-02-26 10:53:33 -07:00
io-wq.c io-wq: remove spin-for-work optimization 2020-02-25 08:57:37 -07:00
io-wq.h io-wq: ensure work->task_pid is cleared on init 2020-02-25 13:23:48 -07:00
ioctl.c compat-ioctl fix for v5.6 2020-02-08 13:44:41 -08:00
Kconfig fs: New zonefs file system 2020-02-09 15:51:46 -08:00
Kconfig.binfmt
libfs.c simple_recursive_removal(): kernel-side rm -rf for ramfs-style filesystems 2019-12-10 22:29:58 -05:00
locks.c locks: print unsigned ino in /proc/locks 2019-12-29 09:00:58 -05:00
Makefile fs: New zonefs file system 2020-02-09 15:51:46 -08:00
mbcache.c
mount.h
mpage.c fs: move guard_bio_eod() after bio_set_op_attrs 2020-01-09 08:16:12 -07:00
namei.c vfs: fix do_last() regression 2020-02-01 10:36:49 -08:00
namespace.c saner copy_mount_options() 2020-02-03 21:23:33 -05:00
no-block.c
nsfs.c Merge branch 'work.openat2' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2020-01-29 11:20:24 -08:00
open.c fs: make build_open_flags() available internally 2020-01-20 17:01:53 -07:00
pipe.c pipe: make sure to wake up everybody when the last reader/writer closes 2020-02-18 14:34:36 -08:00
pnode.c
pnode.h
posix_acl.c fs/posix_acl.c: fix kernel-doc warnings 2020-01-04 13:55:09 -08:00
proc_namespace.c
read_write.c overlayfs update for 5.6 2020-02-04 11:45:21 +00:00
readdir.c readdir: make user_access_begin() use the real access range 2020-01-23 10:15:28 -08:00
select.c y2038: syscalls: change remaining timeval to __kernel_old_timeval 2019-11-15 14:38:29 +01:00
seq_file.c
signalfd.c
splice.c pipe: use exclusive waits when reading or writing 2020-02-08 11:39:19 -08:00
stack.c sched/rt, fs: Use CONFIG_PREEMPTION 2019-12-08 14:37:36 +01:00
stat.c fs: make two stat prep helpers available 2020-01-20 17:03:54 -07:00
statfs.c
super.c fs: call fsnotify_sb_delete after evict_inodes 2019-12-18 00:03:01 -05:00
sync.c
timerfd.c timerfd: Make timerfd_settime() time namespace aware 2020-01-14 12:20:53 +01:00
userfaultfd.c Merge branch 'akpm' (patches from Andrew) 2019-12-01 20:36:41 -08:00
utimes.c utimes: Clamp the timestamps in notify_change() 2019-12-08 19:10:50 -05:00
xattr.c