WHen the ring exits, cleanup is done and the final cancelation and
waiting on completions is done by io_ring_exit_work. That function is
invoked by kworker, which doesn't take any signals. Because of that, it
doesn't really matter if we wait for completions in TASK_INTERRUPTIBLE
or TASK_UNINTERRUPTIBLE state. However, it does matter to the hung task
detection checker!
Normally we expect cancelations and completions to happen rather
quickly. Some test cases, however, will exit the ring and park the
owning task stopped (eg via SIGSTOP). If the owning task needs to run
task_work to complete requests, then io_ring_exit_work won't make any
progress until the task is runnable again. Hence io_ring_exit_work can
trigger the hung task detection, which is particularly problematic if
panic-on-hung-task is enabled.
As the ring exit doesn't take signals to begin with, have it wait
interruptibly rather than uninterruptibly. io_uring has a separate
stuck-exit warning that triggers independently anyway, so we're not
really missing anything by making this switch.
Cc: stable@vger.kernel.org # 5.10+
Link: https://lore.kernel.org/r/b0e4aaef-7088-56ce-244c-976edeac0e66@kernel.dk
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Everybody is passing in the request, so get rid of the io_ring_ctx and
explicit user_data pass-in. Both the ctx and user_data can be deduced
from the request at hand.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We use task_work for a variety of reasons, but doing completions or
triggering rety after poll are by far the hottest two. Use the indirect
funtion call wrappers to avoid the indirect function call if
CONFIG_RETPOLINE is set.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We use lockless lists for the local and deferred task_work, which means
that when we queue up events for processing, we ultimately process them
in reverse order to how they were received. This usually doesn't matter,
but for some cases, it does seem to make a big difference. Do the right
thing and reverse the list before processing it, so that we know it's
processed in the same order in which it was received.
This makes a rather big difference for some medium load network tests,
where consistency of performance was a bit all over the place. Here's
a case that has 4 connections each doing two sends and receives:
io_uring port=10002: rps:161.13k Bps: 1.45M idle=256ms
io_uring port=10002: rps:107.27k Bps: 0.97M idle=413ms
io_uring port=10002: rps:136.98k Bps: 1.23M idle=321ms
io_uring port=10002: rps:155.58k Bps: 1.40M idle=268ms
and after the change:
io_uring port=10002: rps:205.48k Bps: 1.85M idle=140ms user=40ms
io_uring port=10002: rps:203.57k Bps: 1.83M idle=139ms user=20ms
io_uring port=10002: rps:218.79k Bps: 1.97M idle=106ms user=30ms
io_uring port=10002: rps:217.88k Bps: 1.96M idle=110ms user=20ms
io_uring port=10002: rps:222.31k Bps: 2.00M idle=101ms user=0ms
io_uring port=10002: rps:218.74k Bps: 1.97M idle=102ms user=20ms
io_uring port=10002: rps:208.43k Bps: 1.88M idle=125ms user=40ms
using more of the time to actually process work rather than sitting
idle.
No effects have been observed at the peak end of the spectrum, where
performance is still the same even with deep batch depths (and hence
more items to sort).
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If we're doing multishot receives, then we always end up doing two trips
through sock_recvmsg(). For protocols that sanely set msghdr->msg_inq,
then we don't need to waste time picking a new buffer and attempting a
new receive if there's nothing there.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Rather than have this logic in both io_recv() and io_recvmsg_multishot(),
push it into the handler they both call when finishing a receive
operation.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We can't currently tell if ->msg_inq was set when we ask for msg_get_inq,
initialize it to -1U so we can tell apart if it was set and there's
no data left, or if it just wasn't set at all by the protocol.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
With IORING_REGISTER_USE_REGISTERED_RING, an application can register
the ring fd and use it via registered index rather than installed fd.
This allows using a registered ring for everything *except* the initial
mmap.
With IORING_SETUP_NO_MMAP, io_uring_setup uses buffers allocated by the
user, rather than requiring a subsequent mmap.
The combination of the two allows a user to operate *entirely* via a
registered ring fd, making it unnecessary to ever install the fd in the
first place. So, add a flag IORING_SETUP_REGISTERED_FD_ONLY to make
io_uring_setup register the fd and return a registered index, without
installing the fd.
This allows an application to avoid touching the fd table at all, and
allows a library to never even momentarily install a file descriptor.
This splits out an io_ring_add_registered_file helper from
io_ring_add_registered_fd, for use by io_uring_setup.
Signed-off-by: Josh Triplett <josh@joshtriplett.org>
Link: https://lore.kernel.org/r/bc8f431bada371c183b95a83399628b605e978a3.1682699803.git.josh@joshtriplett.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently io_uring applications must call mmap(2) twice to map the rings
themselves, and the sqes array. This works fine, but it does not support
using huge pages to back the rings/sqes.
Provide a way for the application to pass in pre-allocated memory for
the rings/sqes, which can then suitably be allocated from shmfs or
via mmap to get huge page support.
Particularly for larger rings, this reduces the TLBs needed.
If an application wishes to take advantage of that, it must pre-allocate
the memory needed for the sq/cq ring, and the sqes. The former must
be passed in via the io_uring_params->cq_off.user_data field, while the
latter is passed in via the io_uring_params->sq_off.user_data field. Then
it must set IORING_SETUP_NO_MMAP in the io_uring_params->flags field,
and io_uring will then map the existing memory into the kernel for shared
use. The application must not call mmap(2) to map rings as it otherwise
would have, that will now fail with -EINVAL if this setup flag was used.
The pages used for the rings and sqes must be contigious. The intent here
is clearly that huge pages should be used, otherwise the normal setup
procedure works fine as-is. The application may use one huge page for
both the rings and sqes.
Outside of those initialization changes, everything works like it did
before.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We do rings and sqes separately, move them into a helper that does both
the freeing and clearing of the memory.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In preparation for having more than one time of ring allocator, make the
existing one return valid/error-pointer rather than just NULL.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We only have two reserved members we're not clearing, do so manually
instead. This is in preparation for using one of these members for
a new feature.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Now that we have both sockets and block devices setting FMODE_NOWAIT
appropriately, we can get rid of all the odd special casing in
__io_file_supports_nowait() and rely soley on FMODE_NOWAIT and
O_NONBLOCK rather than special case sockets and (in particular) bdevs.
Link: https://lore.kernel.org/r/20230509151910.183637-4-axboe@kernel.dk
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Pull compute express link fixes from Dan Williams:
- Fix a compilation issue with DEFINE_STATIC_SRCU() in the unit tests
- Fix leaking kernel memory to a root-only sysfs attribute
* tag 'cxl-fixes-6.4-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/cxl/cxl:
cxl: Add missing return to cdat read error path
tools/testing/cxl: Use DEFINE_STATIC_SRCU()
Pull parisc architecture fixes from Helge Deller:
- Fix encoding of swp_entry due to added SWP_EXCLUSIVE flag
- Include reboot.h to avoid gcc-12 compiler warning
* tag 'parisc-for-6.4-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
parisc: Fix encoding of swp_entry due to added SWP_EXCLUSIVE flag
parisc: kexec: include reboot.h
Pull ARM fixes from Russell King:
- fix unwinder for uleb128 case
- fix kernel-doc warnings for HP Jornada 7xx
- fix unbalanced stack on vfp success path
* tag 'for-linus' of git://git.armlinux.org.uk/~rmk/linux-arm:
ARM: 9297/1: vfp: avoid unbalanced stack on 'success' return path
ARM: 9296/1: HP Jornada 7XX: fix kernel-doc warnings
ARM: 9295/1: unwind:fix unwind abort for uleb128 case
Pull locking fix from Borislav Petkov:
- Make sure __down_read_common() is always inlined so that the callers'
names land in traceevents output and thus the blocked function can be
identified
* tag 'locking_urgent_for_v6.4_rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
locking/rwsem: Add __always_inline annotation to __down_read_common() and inlined callers
Pull perf fixes from Borislav Petkov:
- Make sure the PEBS buffer is flushed before reprogramming the
hardware so that the correct record sizes are used
- Update the sample size for AMD BRS events
- Fix a confusion with using the same on-stack struct with different
events in the event processing path
* tag 'perf_urgent_for_v6.4_rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/x86/intel/ds: Flush PEBS DS when changing PEBS_DATA_CFG
perf/x86: Fix missing sample size update on AMD BRS
perf/core: Fix perf_sample_data not properly initialized for different swevents in perf_tp_event()
Pull scheduler fix from Borislav Petkov:
- Fix a couple of kernel-doc warnings
* tag 'sched_urgent_for_v6.4_rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched: fix cid_lock kernel-doc warnings
Pull x86 fix from Borislav Petkov:
- Add the required PCI IDs so that the generic SMN accesses provided by
amd_nb.c work for drivers which switch to them. Add a PCI device ID
to k10temp's table so that latter is loaded on such systems too
* tag 'x86_urgent_for_v6.4_rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
hwmon: (k10temp) Add PCI ID for family 19, model 78h
x86/amd_nb: Add PCI ID for family 19h model 78h
Pull timer fix from Borislav Petkov:
- Prevent CPU state corruption when an active clockevent broadcast
device is replaced while the system is already in oneshot mode
* tag 'timers_urgent_for_v6.4_rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
tick/broadcast: Make broadcast device replacement work correctly
Pull ext4 fixes from Ted Ts'o:
"Some ext4 bug fixes (mostly to address Syzbot reports)"
* tag 'ext4_for_linus_stable' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
ext4: bail out of ext4_xattr_ibody_get() fails for any reason
ext4: add bounds checking in get_max_inline_xattr_value_size()
ext4: add indication of ro vs r/w mounts in the mount message
ext4: fix deadlock when converting an inline directory in nojournal mode
ext4: improve error recovery code paths in __ext4_remount()
ext4: improve error handling from ext4_dirhash()
ext4: don't clear SB_RDONLY when remounting r/w until quota is re-enabled
ext4: check iomap type only if ext4_iomap_begin() does not fail
ext4: avoid a potential slab-out-of-bounds in ext4_group_desc_csum
ext4: fix data races when using cached status extents
ext4: avoid deadlock in fs reclaim with page writeback
ext4: fix invalid free tracking in ext4_xattr_move_to_block()
ext4: remove a BUG_ON in ext4_mb_release_group_pa()
ext4: allow ext4_get_group_info() to fail
ext4: fix lockdep warning when enabling MMP
ext4: fix WARNING in mb_find_extent
Pull SCSI fix from James Bottomley:
"A single small fix for the UFS driver to fix a power management
failure"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: ufs: core: Fix I/O hang that occurs when BKOPS fails in W-LUN suspend
In ext4_update_inline_data(), if ext4_xattr_ibody_get() fails for any
reason, it's best if we just fail as opposed to stumbling on,
especially if the failure is EFSCORRUPTED.
Cc: stable@kernel.org
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Whether the file system is mounted read-only or read/write is more
important than the quota mode, which we are already printing. Add the
ro vs r/w indication since this can be helpful in debugging problems
from the console log.
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
If there are failures while changing the mount options in
__ext4_remount(), we need to restore the old mount options.
This commit fixes two problem. The first is there is a chance that we
will free the old quota file names before a potential failure leading
to a use-after-free. The second problem addressed in this commit is
if there is a failed read/write to read-only transition, if the quota
has already been suspended, we need to renable quota handling.
Cc: stable@kernel.org
Link: https://lore.kernel.org/r/20230506142419.984260-2-tytso@mit.edu
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
When a file system currently mounted read/only is remounted
read/write, if we clear the SB_RDONLY flag too early, before the quota
is initialized, and there is another process/thread constantly
attempting to create a directory, it's possible to trigger the
WARN_ON_ONCE(dquot_initialize_needed(inode));
in ext4_xattr_block_set(), with the following stack trace:
WARNING: CPU: 0 PID: 5338 at fs/ext4/xattr.c:2141 ext4_xattr_block_set+0x2ef2/0x3680
RIP: 0010:ext4_xattr_block_set+0x2ef2/0x3680 fs/ext4/xattr.c:2141
Call Trace:
ext4_xattr_set_handle+0xcd4/0x15c0 fs/ext4/xattr.c:2458
ext4_initxattrs+0xa3/0x110 fs/ext4/xattr_security.c:44
security_inode_init_security+0x2df/0x3f0 security/security.c:1147
__ext4_new_inode+0x347e/0x43d0 fs/ext4/ialloc.c:1324
ext4_mkdir+0x425/0xce0 fs/ext4/namei.c:2992
vfs_mkdir+0x29d/0x450 fs/namei.c:4038
do_mkdirat+0x264/0x520 fs/namei.c:4061
__do_sys_mkdirat fs/namei.c:4076 [inline]
__se_sys_mkdirat fs/namei.c:4074 [inline]
__x64_sys_mkdirat+0x89/0xa0 fs/namei.c:4074
Cc: stable@kernel.org
Link: https://lore.kernel.org/r/20230506142419.984260-1-tytso@mit.edu
Reported-by: syzbot+6385d7d3065524c5ca6d@syzkaller.appspotmail.com
Link: https://syzkaller.appspot.com/bug?id=6513f6cb5cd6b5fc9f37e3bb70d273b94be9c34c
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
