[ Upstream commit f2320f1dd6 ]
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: ac0b8b327a ("io_uring: fix use-after-free of sq->thread in __io_uring_show_fdinfo()")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 26ec15e4b0 upstream.
If peeking a bunch of buffers, normally io_ring_buffers_peek() will
truncate the end buffer. This isn't optimal as presumably more data will
be arriving later, and hence it's better to stop with the last full
buffer rather than truncate the end buffer.
Cc: stable@vger.kernel.org
Fixes: 35c8711c8f ("io_uring/kbuf: add helpers for getting/peeking multiple buffers")
Reported-by: Christian Mazakas <christian.mazakas@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit c538f400fa ]
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: ac0b8b327a ("io_uring: fix use-after-free of sq->thread in __io_uring_show_fdinfo()")
Signed-off-by: Keith Busch <kbusch@kernel.org>
Link: https://lore.kernel.org/r/20250611205343.1821117-1-kbusch@meta.com
[axboe: fold in fix for register.c]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit b53e523261 upstream.
There are a few spots where linked timeouts are armed, and not all of
them adhere to the pre-arm, attempt issue, post-arm pattern. This can
be problematic if the linked request returns that it will trigger a
callback later, and does so before the linked timeout is fully armed.
Consolidate all the linked timeout handling into __io_issue_sqe(),
rather than have it spread throughout the various issue entry points.
Cc: stable@vger.kernel.org
Link: https://github.com/axboe/liburing/issues/1390
Reported-by: Chase Hiltz <chase@path.net>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 92835cebab ]
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 af5d68f889 ("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: af5d68f889 ("io_uring/sqpoll: manage task_work privately")
Signed-off-by: Gabriel Krisman Bertazi <krisman@suse.de>
Link: https://lore.kernel.org/r/20250508181203.3785544-1-krisman@suse.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 687b2bae0e upstream.
Multishot normally uses io_req_post_cqe() to post completions, but when
stopping it, it may finish up with a deferred completion. This is fine,
except if another multishot event triggers before the deferred completions
get flushed. If this occurs, then CQEs may get reordered in the CQ ring,
as new multishot completions get posted before the deferred ones are
flushed. This can cause confusion on the application side, if strict
ordering is required for the use case.
When multishot posting via io_req_post_cqe(), flush any pending deferred
completions first, if any.
Cc: stable@vger.kernel.org # 6.1+
Reported-by: Norman Maurer <norman_maurer@apple.com>
Reported-by: Christian Mazakas <christian.mazakas@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 390513642e ]
io_uring always switches requests to atomic refcounting for iowq
execution before there is any parallilism by setting REQ_F_REFCOUNT,
and the flag is not cleared until the request completes. That should be
fine as long as the compiler doesn't make up a non existing value for
the flags, however KCSAN still complains when the request owner changes
oter flag bits:
BUG: KCSAN: data-race in io_req_task_cancel / io_wq_free_work
...
read to 0xffff888117207448 of 8 bytes by task 3871 on cpu 0:
req_ref_put_and_test io_uring/refs.h:22 [inline]
Skip REQ_F_REFCOUNT checks for iowq, we know it's set.
Reported-by: syzbot+903a2ad71fb3f1e47cf5@syzkaller.appspotmail.com
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/d880bc27fb8c3209b54641be4ff6ac02b0e5789a.1743679736.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit edd43f4d6f upstream.
A previous commit added a 'sync' parameter to io_fallback_tw(), which if
true, means the caller wants to wait on the fallback thread handling it.
But the logic is somewhat messed up, ensure that ctxs are swapped and
flushed appropriately.
Cc: stable@vger.kernel.org
Fixes: dfbe5561ae ("io_uring: flush offloaded and delayed task_work on exit")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 6889ae1b4d upstream.
[ 114.987980][ T5313] WARNING: CPU: 6 PID: 5313 at io_uring/io_uring.c:872 io_req_post_cqe+0x12e/0x4f0
[ 114.991597][ T5313] RIP: 0010:io_req_post_cqe+0x12e/0x4f0
[ 115.001880][ T5313] Call Trace:
[ 115.002222][ T5313] <TASK>
[ 115.007813][ T5313] io_send+0x4fe/0x10f0
[ 115.009317][ T5313] io_issue_sqe+0x1a6/0x1740
[ 115.012094][ T5313] io_wq_submit_work+0x38b/0xed0
[ 115.013223][ T5313] io_worker_handle_work+0x62a/0x1600
[ 115.013876][ T5313] io_wq_worker+0x34f/0xdf0
As the comment states, io_req_post_cqe() should only be used by
multishot requests, i.e. REQ_F_APOLL_MULTISHOT, which bundled sends are
not. Add a flag signifying whether a request wants to post multiple
CQEs. Eventually REQ_F_APOLL_MULTISHOT should imply the new flag, but
that's left out for simplicity.
Cc: stable@vger.kernel.org
Fixes: a05d1f625c ("io_uring/net: support bundles for send")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/8b611dbb54d1cd47a88681f5d38c84d0c02bc563.1743067183.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 67c007d6c1 upstream.
refcount_t: underflow; use-after-free.
WARNING: CPU: 0 PID: 5823 at lib/refcount.c:28 refcount_warn_saturate+0x15a/0x1d0 lib/refcount.c:28
RIP: 0010:refcount_warn_saturate+0x15a/0x1d0 lib/refcount.c:28
Call Trace:
<TASK>
io_notif_flush io_uring/notif.h:40 [inline]
io_send_zc_cleanup+0x121/0x170 io_uring/net.c:1222
io_clean_op+0x58c/0x9a0 io_uring/io_uring.c:406
io_free_batch_list io_uring/io_uring.c:1429 [inline]
__io_submit_flush_completions+0xc16/0xd20 io_uring/io_uring.c:1470
io_submit_flush_completions io_uring/io_uring.h:159 [inline]
Before the blamed commit, sendzc relied on io_req_msg_cleanup() to clear
REQ_F_NEED_CLEANUP, so after the following snippet the request will
never hit the core io_uring cleanup path.
io_notif_flush();
io_req_msg_cleanup();
The easiest fix is to null the notification. io_send_zc_cleanup() can
still be called after, but it's tolerated.
Reported-by: syzbot+cf285a028ffba71b2ef5@syzkaller.appspotmail.com
Tested-by: syzbot+cf285a028ffba71b2ef5@syzkaller.appspotmail.com
Fixes: cc34d8330e ("io_uring/net: don't clear REQ_F_NEED_CLEANUP unconditionally")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/e1306007458b8891c88c4f20c966a17595f766b0.1742643795.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit cc34d8330e upstream.
io_req_msg_cleanup() relies on the fact that io_netmsg_recycle() will
always fully recycle, but that may not be the case if the msg cache
was already full. To ensure that normal cleanup always gets run,
let io_netmsg_recycle() deal with clearing the relevant cleanup flags,
as it knows exactly when that should be done.
Cc: stable@vger.kernel.org
Reported-by: David Wei <dw@davidwei.uk>
Fixes: 7519134178 ("io_uring/net: add iovec recycling")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 13918315c5 ]
When io_uring submission goes async for the first time on a given task,
we'll try to create a worker thread to handle the submission. Creating
this worker thread can fail due to various transient conditions, such as
an outstanding signal in the forking thread, so we have retry logic with
a limit of 3 retries. However, this retry logic appears to be too
aggressive/fast - we've observed a thread blowing through the retry
limit while having the same outstanding signal the whole time. Here's an
excerpt of some tracing that demonstrates the issue:
First, signal 26 is generated for the process. It ends up getting routed
to thread 92942.
0) cbd-92284 /* signal_generate: sig=26 errno=0 code=-2 comm=psblkdASD pid=92934 grp=1 res=0 */
This causes create_io_thread in the signalled thread to fail with
ERESTARTNOINTR, and thus a retry is queued.
13) task_th-92942 /* io_uring_queue_async_work: ring 000000007325c9ae, request 0000000080c96d8e, user_data 0x0, opcode URING_CMD, flags 0x8240001, normal queue, work 000000006e96dd3f */
13) task_th-92942 io_wq_enqueue() {
13) task_th-92942 _raw_spin_lock();
13) task_th-92942 io_wq_activate_free_worker();
13) task_th-92942 _raw_spin_lock();
13) task_th-92942 create_io_worker() {
13) task_th-92942 __kmalloc_cache_noprof();
13) task_th-92942 __init_swait_queue_head();
13) task_th-92942 kprobe_ftrace_handler() {
13) task_th-92942 get_kprobe();
13) task_th-92942 aggr_pre_handler() {
13) task_th-92942 pre_handler_kretprobe();
13) task_th-92942 /* create_enter: (create_io_thread+0x0/0x50) fn=0xffffffff8172c0e0 arg=0xffff888996bb69c0 node=-1 */
13) task_th-92942 } /* aggr_pre_handler */
...
13) task_th-92942 } /* copy_process */
13) task_th-92942 } /* create_io_thread */
13) task_th-92942 kretprobe_rethook_handler() {
13) task_th-92942 /* create_exit: (create_io_worker+0x8a/0x1a0 <- create_io_thread) arg1=0xfffffffffffffdff */
13) task_th-92942 } /* kretprobe_rethook_handler */
13) task_th-92942 queue_work_on() {
...
The CPU is then handed to a kworker to process the queued retry:
------------------------------------------
13) task_th-92942 => kworker-54154
------------------------------------------
13) kworker-54154 io_workqueue_create() {
13) kworker-54154 io_queue_worker_create() {
13) kworker-54154 task_work_add() {
13) kworker-54154 wake_up_state() {
13) kworker-54154 try_to_wake_up() {
13) kworker-54154 _raw_spin_lock_irqsave();
13) kworker-54154 _raw_spin_unlock_irqrestore();
13) kworker-54154 } /* try_to_wake_up */
13) kworker-54154 } /* wake_up_state */
13) kworker-54154 kick_process();
13) kworker-54154 } /* task_work_add */
13) kworker-54154 } /* io_queue_worker_create */
13) kworker-54154 } /* io_workqueue_create */
And then we immediately switch back to the original task to try creating
a worker again. This fails, because the original task still hasn't
handled its signal.
-----------------------------------------
13) kworker-54154 => task_th-92942
------------------------------------------
13) task_th-92942 create_worker_cont() {
13) task_th-92942 kprobe_ftrace_handler() {
13) task_th-92942 get_kprobe();
13) task_th-92942 aggr_pre_handler() {
13) task_th-92942 pre_handler_kretprobe();
13) task_th-92942 /* create_enter: (create_io_thread+0x0/0x50) fn=0xffffffff8172c0e0 arg=0xffff888996bb69c0 node=-1 */
13) task_th-92942 } /* aggr_pre_handler */
13) task_th-92942 } /* kprobe_ftrace_handler */
13) task_th-92942 create_io_thread() {
13) task_th-92942 copy_process() {
13) task_th-92942 task_active_pid_ns();
13) task_th-92942 _raw_spin_lock_irq();
13) task_th-92942 recalc_sigpending();
13) task_th-92942 _raw_spin_lock_irq();
13) task_th-92942 } /* copy_process */
13) task_th-92942 } /* create_io_thread */
13) task_th-92942 kretprobe_rethook_handler() {
13) task_th-92942 /* create_exit: (create_worker_cont+0x35/0x1b0 <- create_io_thread) arg1=0xfffffffffffffdff */
13) task_th-92942 } /* kretprobe_rethook_handler */
13) task_th-92942 io_worker_release();
13) task_th-92942 queue_work_on() {
13) task_th-92942 clear_pending_if_disabled();
13) task_th-92942 __queue_work() {
13) task_th-92942 } /* __queue_work */
13) task_th-92942 } /* queue_work_on */
13) task_th-92942 } /* create_worker_cont */
The pattern repeats another couple times until we blow through the retry
counter, at which point we give up. All outstanding work is canceled,
and the io_uring command which triggered all this is failed with
ECANCELED:
13) task_th-92942 io_acct_cancel_pending_work() {
...
13) task_th-92942 /* io_uring_complete: ring 000000007325c9ae, req 0000000080c96d8e, user_data 0x0, result -125, cflags 0x0 extra1 0 extra2 0 */
Finally, the task gets around to processing its outstanding signal 26,
but it's too late.
13) task_th-92942 /* signal_deliver: sig=26 errno=0 code=-2 sa_handler=59566a0 sa_flags=14000000 */
Try to address this issue by adding a small scaling delay when retrying
worker creation. This should give the forking thread time to handle its
signal in the above case. This isn't a particularly satisfying solution,
as sufficiently paradoxical scheduling would still have us hitting the
same issue, and I'm open to suggestions for something better. But this
is likely to prevent this (already rare) issue from hitting in practice.
Signed-off-by: Uday Shankar <ushankar@purestorage.com>
Link: https://lore.kernel.org/r/20250208-wq_retry-v2-1-4f6f5041d303@purestorage.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 5e0e02f0d7 ]
futex_queue() -> __futex_queue() uses 'current' as the task to store in
the struct futex_q->task field. This is fine for synchronous usage of
the futex infrastructure, but it's not always correct when used by
io_uring where the task doing the initial futex_queue() might not be
available later on. This doesn't lead to any issues currently, as the
io_uring side doesn't support PI futexes, but it does leave a
potentially dangling pointer which is never a good idea.
Have futex_queue() take a task_struct argument, and have the regular
callers pass in 'current' for that. Meanwhile io_uring can just pass in
NULL, as the task should never be used off that path. In theory
req->tctx->task could be used here, but there's no point populating it
with a task field that will never be used anyway.
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/22484a23-542c-4003-b721-400688a0d055@kernel.dk
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 8802766324 upstream.
IORING_REGISTER_PBUF_RING can reuse an old struct io_buffer_list if it
was created for legacy selected buffer and has been emptied. It violates
the requirement that most of the field should stay stable after publish.
Always reallocate it instead.
Cc: stable@vger.kernel.org
Reported-by: Pumpkin Chang <pumpkin@devco.re>
Fixes: 2fcabce2d7 ("io_uring: disallow mixed provided buffer group registrations")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 0edf1283a9 ]
Any uring_cmd always has async data allocated now, there's no reason to
check and clear a cached copy of the SQE.
Fixes: d10f19dff5 ("io_uring/uring_cmd: switch to always allocating async data")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 06521ac048 ]
struct io_tw_state is managed by core io_uring, and opcode handling code
must never try to cheat and create their own instances, it's plain
incorrect.
io_waitid_complete() attempts exactly that outside of the task work
context, and even though the ring is locked, there would be no one to
reap the requests from the defer completion list. It only works now
because luckily it's called before io_uring_try_cancel_uring_cmd(),
which flushes completions.
Fixes: f31ecf671d ("io_uring: add IORING_OP_WAITID support")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 8c8492ca64 upstream.
If a socket is shutdown before the connection completes, POLLERR is set
in the poll mask. However, connect ignores this as it doesn't know, and
attempts the connection again. This may lead to a bogus -ETIMEDOUT
result, where it should have noticed the POLLERR and just returned
-ECONNRESET instead.
Have the poll logic check for whether or not POLLERR is set in the mask,
and if so, mark the request as failed. Then connect can appropriately
fail the request rather than retry it.
Reported-by: Sergey Galas <ssgalas@cloud.ru>
Cc: stable@vger.kernel.org
Link: https://github.com/axboe/liburing/discussions/1335
Fixes: 3fb1bd6881 ("io_uring/net: handle -EINPROGRESS correct for IORING_OP_CONNECT")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit d58d82bd0e ]
io_uring_cmd_sock() does a normal read of cmd->sqe->cmd_op, where it
really should be using a READ_ONCE() as ->sqe may still be pointing to
the original SQE. Since the prep side already does this READ_ONCE() and
stores it locally, use that value rather than re-read it.
Fixes: 8e9fad0e70 ("io_uring: Add io_uring command support for sockets")
Link: https://lore.kernel.org/r/20250121-uring-sockcmd-fix-v1-1-add742802a29@google.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit bd2703b42d upstream.
With IORING_SETUP_SQPOLL all requests are created by the SQPOLL task,
which means that req->task should always match sqd->thread. Since
accesses to sqd->thread should be separately protected, use req->task
in io_req_normal_work_add() instead.
Note, in the eyes of io_req_normal_work_add(), the SQPOLL task struct
is always pinned and alive, and sqd->thread can either be the task or
NULL. It's only problematic if the compiler decides to reload the value
after the null check, which is not so likely.
Cc: stable@vger.kernel.org
Cc: Bui Quang Minh <minhquangbui99@gmail.com>
Reported-by: lizetao <lizetao1@huawei.com>
Fixes: 78f9b61bd8 ("io_uring: wake SQPOLL task when task_work is added to an empty queue")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/1cbbe72cf32c45a8fee96026463024cd8564a7d7.1736541357.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit c9a40292a4 upstream.
io_eventfd_do_signal() is invoked from an RCU callback, but when
dropping the reference to the io_ev_fd, it calls io_eventfd_free()
directly if the refcount drops to zero. This isn't correct, as any
potential freeing of the io_ev_fd should be deferred another RCU grace
period.
Just call io_eventfd_put() rather than open-code the dec-and-test and
free, which will correctly defer it another RCU grace period.
Fixes: 21a091b970 ("io_uring: signal registered eventfd to process deferred task work")
Reported-by: Jann Horn <jannh@google.com>
Cc: stable@vger.kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit ed123c948d upstream.
For non-pollable files, buffer ring consumption will commit upfront.
This is fine, but io_ring_buffer_select() will return the address of the
buffer after having committed it. For incrementally consumed buffers,
this is incorrect as it will modify the buffer address.
Store the pre-committed value and return that. If that isn't done, then
the initial part of the buffer is not used and the application will
correctly assume the content arrived at the start of the userspace
buffer, but the kernel will have put it later in the buffer. Or it can
cause a spurious -EFAULT returned in the CQE, depending on the buffer
size. As bounds are suitably checked for doing the actual IO, no adverse
side effects are possible - it's just a data misplacement within the
existing buffer.
Reported-by: Gwendal Fernet <gwendalfernet@gmail.com>
Cc: stable@vger.kernel.org
Fixes: ae98dbf43d ("io_uring/kbuf: add support for incremental buffer consumption")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit c6e60a0a68 ]
syzbot reports that ->msg_inq may get used uinitialized from the
following path:
BUG: KMSAN: uninit-value in io_recv_buf_select io_uring/net.c:1094 [inline]
BUG: KMSAN: uninit-value in io_recv+0x930/0x1f90 io_uring/net.c:1158
io_recv_buf_select io_uring/net.c:1094 [inline]
io_recv+0x930/0x1f90 io_uring/net.c:1158
io_issue_sqe+0x420/0x2130 io_uring/io_uring.c:1740
io_queue_sqe io_uring/io_uring.c:1950 [inline]
io_req_task_submit+0xfa/0x1d0 io_uring/io_uring.c:1374
io_handle_tw_list+0x55f/0x5c0 io_uring/io_uring.c:1057
tctx_task_work_run+0x109/0x3e0 io_uring/io_uring.c:1121
tctx_task_work+0x6d/0xc0 io_uring/io_uring.c:1139
task_work_run+0x268/0x310 kernel/task_work.c:239
io_run_task_work+0x43a/0x4a0 io_uring/io_uring.h:343
io_cqring_wait io_uring/io_uring.c:2527 [inline]
__do_sys_io_uring_enter io_uring/io_uring.c:3439 [inline]
__se_sys_io_uring_enter+0x204f/0x4ce0 io_uring/io_uring.c:3330
__x64_sys_io_uring_enter+0x11f/0x1a0 io_uring/io_uring.c:3330
x64_sys_call+0xce5/0x3c30 arch/x86/include/generated/asm/syscalls_64.h:427
do_syscall_x64 arch/x86/entry/common.c:52 [inline]
do_syscall_64+0xcd/0x1e0 arch/x86/entry/common.c:83
entry_SYSCALL_64_after_hwframe+0x77/0x7f
and it is correct, as it's never initialized upfront. Hence the first
submission can end up using it uninitialized, if the recv wasn't
successful and the networking stack didn't honor ->msg_get_inq being set
and filling in the output value of ->msg_inq as requested.
Set it to 0 upfront when it's allocated, just to silence this KMSAN
warning. There's no side effect of using it uninitialized, it'll just
potentially cause the next receive to use a recv value hint that's not
accurate.
Fixes: c6f32c7d9e ("io_uring/net: get rid of ->prep_async() for receive side")
Reported-by: syzbot+068ff190354d2f74892f@syzkaller.appspotmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit e33ac68e5e upstream.
BUG: KASAN: slab-use-after-free in __lock_acquire+0x370b/0x4a10 kernel/locking/lockdep.c:5089
Call Trace:
<TASK>
...
_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 <huk23@m.fudan.edu.cn>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/0f2f1aa5729332612bd01fe0f2f385fd1f06ce7c.1735231717.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit dbd2ca9367 upstream.
task work can be executed after the task has gone through io_uring
termination, whether it's the final task_work run or the fallback path.
In this case, task work will find ->io_wq being already killed and
null'ed, which is a problem if it then tries to forward the request to
io_queue_iowq(). Make io_queue_iowq() fail requests in this case.
Note that it also checks PF_KTHREAD, because the user can first close
a DEFER_TASKRUN ring and shortly after kill the task, in which case
->iowq check would race.
Cc: stable@vger.kernel.org
Fixes: 50c52250e2 ("block: implement async io_uring discard cmd")
Fixes: 773af69121 ("io_uring: always reissue from task_work context")
Reported-by: Will <willsroot@protonmail.com>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/63312b4a2c2bb67ad67b857d17a300e1d3b078e8.1734637909.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 12d908116f upstream.
Currently, io_uring_unreg_ringfd() (which cleans up registered rings) is
only called on exit, but __io_uring_free (which frees the tctx in which the
registered ring pointers are stored) is also called on execve (via
begin_new_exec -> io_uring_task_cancel -> __io_uring_cancel ->
io_uring_cancel_generic -> __io_uring_free).
This means: A process going through execve while having registered rings
will leak references to the rings' `struct file`.
Fix it by zapping registered rings on execve(). This is implemented by
moving the io_uring_unreg_ringfd() from io_uring_files_cancel() into its
callee __io_uring_cancel(), which is called from io_uring_task_cancel() on
execve.
This could probably be exploited *on 32-bit kernels* by leaking 2^32
references to the same ring, because the file refcount is stored in a
pointer-sized field and get_file() doesn't have protection against
refcount overflow, just a WARN_ONCE(); but on 64-bit it should have no
impact beyond a memory leak.
Cc: stable@vger.kernel.org
Fixes: e7a6c00dc7 ("io_uring: add support for registering ring file descriptors")
Signed-off-by: Jann Horn <jannh@google.com>
Link: https://lore.kernel.org/r/20241218-uring-reg-ring-cleanup-v1-1-8f63e999045b@google.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 7eb75ce752 ]
syzbot triggered the following WARN_ON:
WARNING: CPU: 0 PID: 16 at io_uring/tctx.c:51 __io_uring_free+0xfa/0x140 io_uring/tctx.c:51
which is the
WARN_ON_ONCE(!xa_empty(&tctx->xa));
sanity check in __io_uring_free() when a io_uring_task is going through
its final put. The syzbot test case includes injecting memory allocation
failures, and it very much looks like xa_store() can fail one of its
memory allocations and end up with ->head being non-NULL even though no
entries exist in the xarray.
Until this issue gets sorted out, work around it by attempting to
iterate entries in our xarray, and WARN_ON_ONCE() if one is found.
Reported-by: syzbot+cc36d44ec9f368e443d3@syzkaller.appspotmail.com
Link: https://lore.kernel.org/io-uring/673c1643.050a0220.87769.0066.GAE@google.com/
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 43eef70e7e upstream.
io_pages_unmap() is a bit tricky in trying to figure whether the pages
were previously vmap'ed or not. In particular If there is juts one page
it belives there is no need to vunmap. Paired io_pages_map(), however,
could've failed io_mem_alloc_compound() and attempted to
io_mem_alloc_single(), which does vmap, and that leads to unpaired vmap.
The solution is to fail if io_mem_alloc_compound() can't allocate a
single page. That's the easiest way to deal with it, and those two
functions are getting removed soon, so no need to overcomplicate it.
Cc: stable@vger.kernel.org
Fixes: 3ab1db3c60 ("io_uring: get rid of remap_pfn_range() for mapping rings/sqes")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/477e75a3907a2fe83249e49c0a92cd480b2c60e0.1732569842.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When io_uring starts a write, it'll call kiocb_start_write() to bump the
super block rwsem, preventing any freezes from happening while that
write is in-flight. The freeze side will grab that rwsem for writing,
excluding any new writers from happening and waiting for existing writes
to finish. But io_uring unconditionally uses kiocb_start_write(), which
will block if someone is currently attempting to freeze the mount point.
This causes a deadlock where freeze is waiting for previous writes to
complete, but the previous writes cannot complete, as the task that is
supposed to complete them is blocked waiting on starting a new write.
This results in the following stuck trace showing that dependency with
the write blocked starting a new write:
task:fio state:D stack:0 pid:886 tgid:886 ppid:876
Call trace:
__switch_to+0x1d8/0x348
__schedule+0x8e8/0x2248
schedule+0x110/0x3f0
percpu_rwsem_wait+0x1e8/0x3f8
__percpu_down_read+0xe8/0x500
io_write+0xbb8/0xff8
io_issue_sqe+0x10c/0x1020
io_submit_sqes+0x614/0x2110
__arm64_sys_io_uring_enter+0x524/0x1038
invoke_syscall+0x74/0x268
el0_svc_common.constprop.0+0x160/0x238
do_el0_svc+0x44/0x60
el0_svc+0x44/0xb0
el0t_64_sync_handler+0x118/0x128
el0t_64_sync+0x168/0x170
INFO: task fsfreeze:7364 blocked for more than 15 seconds.
Not tainted 6.12.0-rc5-00063-g76aaf945701c #7963
with the attempting freezer stuck trying to grab the rwsem:
task:fsfreeze state:D stack:0 pid:7364 tgid:7364 ppid:995
Call trace:
__switch_to+0x1d8/0x348
__schedule+0x8e8/0x2248
schedule+0x110/0x3f0
percpu_down_write+0x2b0/0x680
freeze_super+0x248/0x8a8
do_vfs_ioctl+0x149c/0x1b18
__arm64_sys_ioctl+0xd0/0x1a0
invoke_syscall+0x74/0x268
el0_svc_common.constprop.0+0x160/0x238
do_el0_svc+0x44/0x60
el0_svc+0x44/0xb0
el0t_64_sync_handler+0x118/0x128
el0t_64_sync+0x168/0x170
Fix this by having the io_uring side honor IOCB_NOWAIT, and only attempt a
blocking grab of the super block rwsem if it isn't set. For normal issue
where IOCB_NOWAIT would always be set, this returns -EAGAIN which will
have io_uring core issue a blocking attempt of the write. That will in
turn also get completions run, ensuring forward progress.
Since freezing requires CAP_SYS_ADMIN in the first place, this isn't
something that can be triggered by a regular user.
Cc: stable@vger.kernel.org # 5.10+
Reported-by: Peter Mann <peter.mann@sh.cz>
Link: https://lore.kernel.org/io-uring/38c94aec-81c9-4f62-b44e-1d87f5597644@sh.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
A previous commit improved how !FMODE_NOWAIT is dealt with, but
inadvertently negated a check whilst doing so. This caused -EAGAIN to be
returned from reading files with O_NONBLOCK set. Fix up the check for
REQ_F_SUPPORT_NOWAIT.
Reported-by: Julian Orth <ju.orth@gmail.com>
Link: https://github.com/axboe/liburing/issues/1270
Fixes: f7c9134385 ("io_uring/rw: allow pollable non-blocking attempts for !FMODE_NOWAIT")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe
Penglei Jiang
Pavel Begunkov
Keith Busch
Gabriel Krisman Bertazi
Christoph Hellwig
Uday Shankar
Jann Horn
Bernd Schubert