kernel/linux.git/io_uring/sqpoll.c, branch v6.12.80

io_uring/sqpoll: be smarter on when to update the stime usage

2025-10-29T13:08:59+00:00

Commit a94e0657269c5b8e1a90b17aa2c048b3d276e16d upstream. The current approach is a bit naive, and hence calls the time querying way too often. Only start the "doing work" timer when there's actual work to do, and then use that information to terminate (and account) the work time once done. This greatly reduces the frequency of these calls, when they cannot have changed anyway. Running a basic random reader that is setup to use SQPOLL, a profile before this change shows these as the top cycle consumers: + 32.60% iou-sqp-1074 [kernel.kallsyms] [k] thread_group_cputime_adjusted + 19.97% iou-sqp-1074 [kernel.kallsyms] [k] thread_group_cputime + 12.20% io_uring io_uring [.] submitter_uring_fn + 4.13% iou-sqp-1074 [kernel.kallsyms] [k] getrusage + 2.45% iou-sqp-1074 [kernel.kallsyms] [k] io_submit_sqes + 2.18% iou-sqp-1074 [kernel.kallsyms] [k] __pi_memset_generic + 2.09% iou-sqp-1074 [kernel.kallsyms] [k] cputime_adjust and after this change, top of profile looks as follows: + 36.23% io_uring io_uring [.] submitter_uring_fn + 23.26% iou-sqp-819 [kernel.kallsyms] [k] io_sq_thread + 10.14% iou-sqp-819 [kernel.kallsyms] [k] io_sq_tw + 6.52% iou-sqp-819 [kernel.kallsyms] [k] tctx_task_work_run + 4.82% iou-sqp-819 [kernel.kallsyms] [k] nvme_submit_cmds.part.0 + 2.91% iou-sqp-819 [kernel.kallsyms] [k] io_submit_sqes [...] 0.02% iou-sqp-819 [kernel.kallsyms] [k] cputime_adjust where it's spending the cycles on things that actually matter. Reported-by: Fengnan Chang Cc: stable@vger.kernel.org Fixes: 3fcb9d17206e ("io_uring/sqpoll: statistics of the true utilization of sq threads") Signed-off-by: Jens Axboe Signed-off-by: Greg Kroah-Hartman

io_uring/sqpoll: switch away from getrusage() for CPU accounting

2025-10-29T13:08:59+00:00

Commit 8ac9b0d33e5c0a995338ee5f25fe1b6ff7d97f65 upstream. getrusage() does a lot more than what the SQPOLL accounting needs, the latter only cares about (and uses) the stime. Rather than do a full RUSAGE_SELF summation, just query the used stime instead. Cc: stable@vger.kernel.org Fixes: 3fcb9d17206e ("io_uring/sqpoll: statistics of the true utilization of sq threads") Reviewed-by: Gabriel Krisman Bertazi Signed-off-by: Jens Axboe Signed-off-by: Greg Kroah-Hartman

io_uring/sqpoll: don't put task_struct on tctx setup failure

2025-06-27T10:11:42+00:00

[ Upstream commit f2320f1dd6f6f82cb2c7aff23a12bab537bdea89 ] A recent commit moved the error handling of sqpoll thread and tctx failures into the thread itself, as part of fixing an issue. However, it missed that tctx allocation may also fail, and that io_sq_offload_create() does its own error handling for the task_struct in that case. Remove the manual task putting in io_sq_offload_create(), as io_sq_thread() will notice that the tctx did not get setup and hence it should put itself and exit. Reported-by: syzbot+763e12bbf004fb1062e4@syzkaller.appspotmail.com Fixes: ac0b8b327a56 ("io_uring: fix use-after-free of sq->thread in __io_uring_show_fdinfo()") Signed-off-by: Jens Axboe Signed-off-by: Sasha Levin

io_uring: consistently use rcu semantics with sqpoll thread

2025-06-19T13:32:33+00:00

[ Upstream commit c538f400fae22725580842deb2bef546701b64bd ] The sqpoll thread is dereferenced with rcu read protection in one place, so it needs to be annotated as an __rcu type, and should consistently use rcu helpers for access and assignment to make sparse happy. Since most of the accesses occur under the sqd->lock, we can use rcu_dereference_protected() without declaring an rcu read section. Provide a simple helper to get the thread from a locked context. Fixes: ac0b8b327a5677d ("io_uring: fix use-after-free of sq->thread in __io_uring_show_fdinfo()") Signed-off-by: Keith Busch Link: https://lore.kernel.org/r/20250611205343.1821117-1-kbusch@meta.com [axboe: fold in fix for register.c] Signed-off-by: Jens Axboe Signed-off-by: Sasha Levin

io_uring: fix use-after-free of sq->thread in __io_uring_show_fdinfo()

2025-06-19T13:32:33+00:00

[ Upstream commit ac0b8b327a5677dc6fecdf353d808161525b1ff0 ] syzbot reports: BUG: KASAN: slab-use-after-free in getrusage+0x1109/0x1a60 Read of size 8 at addr ffff88810de2d2c8 by task a.out/304 CPU: 0 UID: 0 PID: 304 Comm: a.out Not tainted 6.16.0-rc1 #1 PREEMPT(voluntary) Hardware name: QEMU Ubuntu 24.04 PC (i440FX + PIIX, 1996), BIOS 1.16.3-debian-1.16.3-2 04/01/2014 Call Trace: dump_stack_lvl+0x53/0x70 print_report+0xd0/0x670 ? __pfx__raw_spin_lock_irqsave+0x10/0x10 ? getrusage+0x1109/0x1a60 kasan_report+0xce/0x100 ? getrusage+0x1109/0x1a60 getrusage+0x1109/0x1a60 ? __pfx_getrusage+0x10/0x10 __io_uring_show_fdinfo+0x9fe/0x1790 ? ksys_read+0xf7/0x1c0 ? do_syscall_64+0xa4/0x260 ? vsnprintf+0x591/0x1100 ? __pfx___io_uring_show_fdinfo+0x10/0x10 ? __pfx_vsnprintf+0x10/0x10 ? mutex_trylock+0xcf/0x130 ? __pfx_mutex_trylock+0x10/0x10 ? __pfx_show_fd_locks+0x10/0x10 ? io_uring_show_fdinfo+0x57/0x80 io_uring_show_fdinfo+0x57/0x80 seq_show+0x38c/0x690 seq_read_iter+0x3f7/0x1180 ? inode_set_ctime_current+0x160/0x4b0 seq_read+0x271/0x3e0 ? __pfx_seq_read+0x10/0x10 ? __pfx__raw_spin_lock+0x10/0x10 ? __mark_inode_dirty+0x402/0x810 ? selinux_file_permission+0x368/0x500 ? file_update_time+0x10f/0x160 vfs_read+0x177/0xa40 ? __pfx___handle_mm_fault+0x10/0x10 ? __pfx_vfs_read+0x10/0x10 ? mutex_lock+0x81/0xe0 ? __pfx_mutex_lock+0x10/0x10 ? fdget_pos+0x24d/0x4b0 ksys_read+0xf7/0x1c0 ? __pfx_ksys_read+0x10/0x10 ? do_user_addr_fault+0x43b/0x9c0 do_syscall_64+0xa4/0x260 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f0f74170fc9 Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 8 RSP: 002b:00007fffece049e8 EFLAGS: 00000206 ORIG_RAX: 0000000000000000 RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f0f74170fc9 RDX: 0000000000001000 RSI: 00007fffece049f0 RDI: 0000000000000004 RBP: 00007fffece05ad0 R08: 0000000000000000 R09: 00007fffece04d90 R10: 0000000000000000 R11: 0000000000000206 R12: 00005651720a1100 R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000 Allocated by task 298: kasan_save_stack+0x33/0x60 kasan_save_track+0x14/0x30 __kasan_slab_alloc+0x6e/0x70 kmem_cache_alloc_node_noprof+0xe8/0x330 copy_process+0x376/0x5e00 create_io_thread+0xab/0xf0 io_sq_offload_create+0x9ed/0xf20 io_uring_setup+0x12b0/0x1cc0 do_syscall_64+0xa4/0x260 entry_SYSCALL_64_after_hwframe+0x77/0x7f Freed by task 22: kasan_save_stack+0x33/0x60 kasan_save_track+0x14/0x30 kasan_save_free_info+0x3b/0x60 __kasan_slab_free+0x37/0x50 kmem_cache_free+0xc4/0x360 rcu_core+0x5ff/0x19f0 handle_softirqs+0x18c/0x530 run_ksoftirqd+0x20/0x30 smpboot_thread_fn+0x287/0x6c0 kthread+0x30d/0x630 ret_from_fork+0xef/0x1a0 ret_from_fork_asm+0x1a/0x30 Last potentially related work creation: kasan_save_stack+0x33/0x60 kasan_record_aux_stack+0x8c/0xa0 __call_rcu_common.constprop.0+0x68/0x940 __schedule+0xff2/0x2930 __cond_resched+0x4c/0x80 mutex_lock+0x5c/0xe0 io_uring_del_tctx_node+0xe1/0x2b0 io_uring_clean_tctx+0xb7/0x160 io_uring_cancel_generic+0x34e/0x760 do_exit+0x240/0x2350 do_group_exit+0xab/0x220 __x64_sys_exit_group+0x39/0x40 x64_sys_call+0x1243/0x1840 do_syscall_64+0xa4/0x260 entry_SYSCALL_64_after_hwframe+0x77/0x7f The buggy address belongs to the object at ffff88810de2cb00 which belongs to the cache task_struct of size 3712 The buggy address is located 1992 bytes inside of freed 3712-byte region [ffff88810de2cb00, ffff88810de2d980) which is caused by the task_struct pointed to by sq->thread being released while it is being used in the function __io_uring_show_fdinfo(). Holding ctx->uring_lock does not prevent ehre relase or exit of sq->thread. Fix this by assigning and looking up ->thread under RCU, and grabbing a reference to the task_struct. This ensures that it cannot get released while fdinfo is using it. Reported-by: syzbot+531502bbbe51d2f769f4@syzkaller.appspotmail.com Closes: https://lore.kernel.org/all/682b06a5.a70a0220.3849cf.00b3.GAE@google.com Fixes: 3fcb9d17206e ("io_uring/sqpoll: statistics of the true utilization of sq threads") Signed-off-by: Penglei Jiang Link: https://lore.kernel.org/r/20250610171801.70960-1-superman.xpt@gmail.com [axboe: massage commit message] Signed-off-by: Jens Axboe Signed-off-by: Sasha Levin

io_uring/sqpoll: Increase task_work submission batch size

2025-05-18T06:24:57+00:00

[ Upstream commit 92835cebab120f8a5f023a26a792a2ac3f816c4f ] Our QA team reported a 10%-23%, throughput reduction on an io_uring sqpoll testcase doing IO to a null_blk, that I traced back to a reduction of the device submission queue depth utilization. It turns out that, after commit af5d68f8892f ("io_uring/sqpoll: manage task_work privately"), we capped the number of task_work entries that can be completed from a single spin of sqpoll to only 8 entries, before the sqpoll goes around to (potentially) sleep. While this cap doesn't drive the submission side directly, it impacts the completion behavior, which affects the number of IO queued by fio per sqpoll cycle on the submission side, and io_uring ends up seeing less ios per sqpoll cycle. As a result, block layer plugging is less effective, and we see more time spent inside the block layer in profilings charts, and increased submission latency measured by fio. There are other places that have increased overhead once sqpoll sleeps more often, such as the sqpoll utilization calculation. But, in this microbenchmark, those were not representative enough in perf charts, and their removal didn't yield measurable changes in throughput. The major overhead comes from the fact we plug less, and less often, when submitting to the block layer. My benchmark is: fio --ioengine=io_uring --direct=1 --iodepth=128 --runtime=300 --bs=4k \ --invalidate=1 --time_based --ramp_time=10 --group_reporting=1 \ --filename=/dev/nullb0 --name=RandomReads-direct-nullb-sqpoll-4k-1 \ --rw=randread --numjobs=1 --sqthread_poll In one machine, tested on top of Linux 6.15-rc1, we have the following baseline: READ: bw=4994MiB/s (5236MB/s), 4994MiB/s-4994MiB/s (5236MB/s-5236MB/s), io=439GiB (471GB), run=90001-90001msec With this patch: READ: bw=5762MiB/s (6042MB/s), 5762MiB/s-5762MiB/s (6042MB/s-6042MB/s), io=506GiB (544GB), run=90001-90001msec which is a 15% improvement in measured bandwidth. The average submission latency is noticeably lowered too. As measured by fio: Baseline: lat (usec): min=20, max=241, avg=99.81, stdev=3.38 Patched: lat (usec): min=26, max=226, avg=86.48, stdev=4.82 If we look at blktrace, we can also see the plugging behavior is improved. In the baseline, we end up limited to plugging 8 requests in the block layer regardless of the device queue depth size, while after patching we can drive more io, and we manage to utilize the full device queue. In the baseline, after a stabilization phase, an ordinary submission looks like: 254,0 1 49942 0.016028795 5977 U N [iou-sqp-5976] 7 After patching, I see consistently more requests per unplug. 254,0 1 4996 0.001432872 3145 U N [iou-sqp-3144] 32 Ideally, the cap size would at least be the deep enough to fill the device queue, but we can't predict that behavior, or assume all IO goes to a single device, and thus can't guess the ideal batch size. We also don't want to let the tw run unbounded, though I'm not sure it would really be a problem. Instead, let's just give it a more sensible value that will allow for more efficient batching. I've tested with different cap values, and initially proposed to increase the cap to 1024. Jens argued it is too big of a bump and I observed that, with 32, I'm no longer able to observe this bottleneck in any of my machines. Fixes: af5d68f8892f ("io_uring/sqpoll: manage task_work privately") Signed-off-by: Gabriel Krisman Bertazi Link: https://lore.kernel.org/r/20250508181203.3785544-1-krisman@suse.de Signed-off-by: Jens Axboe Signed-off-by: Sasha Levin

io_uring/sqpoll: zero sqd->thread on tctx errors

2025-01-17T12:40:52+00:00

commit 4b7cfa8b6c28a9fa22b86894166a1a34f6d630ba upstream. Syzkeller reports: BUG: KASAN: slab-use-after-free in thread_group_cputime+0x409/0x700 kernel/sched/cputime.c:341 Read of size 8 at addr ffff88803578c510 by task syz.2.3223/27552 Call Trace: ... kasan_report+0x143/0x180 mm/kasan/report.c:602 thread_group_cputime+0x409/0x700 kernel/sched/cputime.c:341 thread_group_cputime_adjusted+0xa6/0x340 kernel/sched/cputime.c:639 getrusage+0x1000/0x1340 kernel/sys.c:1863 io_uring_show_fdinfo+0xdfe/0x1770 io_uring/fdinfo.c:197 seq_show+0x608/0x770 fs/proc/fd.c:68 ... That's due to sqd->task not being cleared properly in cases where SQPOLL task tctx setup fails, which can essentially only happen with fault injection to insert allocation errors. Cc: stable@vger.kernel.org Fixes: 1251d2025c3e1 ("io_uring/sqpoll: early exit thread if task_context wasn't allocated") Reported-by: syzbot+3d92cfcfa84070b0a470@syzkaller.appspotmail.com Signed-off-by: Pavel Begunkov Link: https://lore.kernel.org/r/efc7ec7010784463b2e7466d7b5c02c2cb381635.1736519461.git.asml.silence@gmail.com Signed-off-by: Jens Axboe Signed-off-by: Greg Kroah-Hartman

io_uring/sqpoll: fix sqpoll error handling races

2025-01-02T09:34:22+00:00

commit e33ac68e5e21ec1292490dfe061e75c0dbdd3bd4 upstream. BUG: KASAN: slab-use-after-free in __lock_acquire+0x370b/0x4a10 kernel/locking/lockdep.c:5089 Call Trace: ... _raw_spin_lock_irqsave+0x3d/0x60 kernel/locking/spinlock.c:162 class_raw_spinlock_irqsave_constructor include/linux/spinlock.h:551 [inline] try_to_wake_up+0xb5/0x23c0 kernel/sched/core.c:4205 io_sq_thread_park+0xac/0xe0 io_uring/sqpoll.c:55 io_sq_thread_finish+0x6b/0x310 io_uring/sqpoll.c:96 io_sq_offload_create+0x162/0x11d0 io_uring/sqpoll.c:497 io_uring_create io_uring/io_uring.c:3724 [inline] io_uring_setup+0x1728/0x3230 io_uring/io_uring.c:3806 ... Kun Hu reports that the SQPOLL creating error path has UAF, which happens if io_uring_alloc_task_context() fails and then io_sq_thread() manages to run and complete before the rest of error handling code, which means io_sq_thread_finish() is looking at already killed task. Note that this is mostly theoretical, requiring fault injection on the allocation side to trigger in practice. Cc: stable@vger.kernel.org Reported-by: Kun Hu Signed-off-by: Pavel Begunkov Link: https://lore.kernel.org/r/0f2f1aa5729332612bd01fe0f2f385fd1f06ce7c.1735231717.git.asml.silence@gmail.com Signed-off-by: Jens Axboe Signed-off-by: Greg Kroah-Hartman

Merge tag 'for-6.12/io_uring-20240922' of git://git.kernel.dk/linux

2024-09-24T18:11:38+00:00

Pull more io_uring updates from Jens Axboe: "Mostly just a set of fixes in here, or little changes that didn't get included in the initial pull request. This contains: - Move the SQPOLL napi polling outside the submission lock (Olivier) - Rename of the "copy buffers" API that got added in the 6.12 merge window. There's really no copying going on, it's just referencing the buffers. After a bit of consideration, decided that it was better to simply rename this to avoid potential confusion (me) - Shrink struct io_mapped_ubuf from 48 to 32 bytes, by changing it to start + len tracking rather than having start / end in there, and by removing the caching of folio_mask when we can just calculate it from folio_shift when we need it (me) - Fixes for the SQPOLL affinity checking (me, Felix) - Fix for how cqring waiting checks for the presence of task_work. Just check it directly rather than check for a specific notification mechanism (me) - Tweak to how request linking is represented in tracing (me) - Fix a syzbot report that deliberately sets up a huge list of overflow entries, and then hits rcu stalls when flushing this list. Just check for the need to preempt, and drop/reacquire locks in the loop. There's no state maintained over the loop itself, and each entry is yanked from head-of-list (me)" * tag 'for-6.12/io_uring-20240922' of git://git.kernel.dk/linux: io_uring: check if we need to reschedule during overflow flush io_uring: improve request linking trace io_uring: check for presence of task_work rather than TIF_NOTIFY_SIGNAL io_uring/sqpoll: do the napi busy poll outside the submission block io_uring: clean up a type in io_uring_register_get_file() io_uring/sqpoll: do not put cpumask on stack io_uring/sqpoll: retain test for whether the CPU is valid io_uring/rsrc: change ubuf->ubuf_end to length tracking io_uring/rsrc: get rid of io_mapped_ubuf->folio_mask io_uring: rename "copy buffers" to "clone buffers"

Merge tag 'pull-stable-struct_fd' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs

2024-09-23T16:35:36+00:00

Pull 'struct fd' updates from Al Viro: "Just the 'struct fd' layout change, with conversion to accessor helpers" * tag 'pull-stable-struct_fd' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: add struct fd constructors, get rid of __to_fd() struct fd: representation change introduce fd_file(), convert all accessors to it.