This reverts commit 4db4b1a0

.

The IO threads allow and handle SIGSTOP now, so don't special case them
anymore in task_set_jobctl_pending().

Signed-off-by: Jens Axboe <axboe@kernel.dk>

This reverts commit 15b2219f

.

Before IO threads accepted signals, the freezer using take signals to wake
up an IO thread would cause them to loop without any way to clear the
pending signal. That is no longer the case, so stop special casing
PF_IO_WORKER in the freezer.

Signed-off-by: Jens Axboe <axboe@kernel.dk>

This reverts commit 6fb8f43c

.

The IO threads do allow signals now, including SIGSTOP, and we can allow
ptrace attach. Attaching won't reveal anything interesting for the IO
threads, but it will allow eg gdb to attach to a task with io_urings
and IO threads without complaining. And once attached, it will allow
the usual introspection into regular threads.

Signed-off-by: Jens Axboe <axboe@kernel.dk>

This reverts commit 5be28c8f

.

IO threads now take signals just fine, so there's no reason to limit them
specifically. Revert the change that prevented that from happening.

Signed-off-by: Jens Axboe <axboe@kernel.dk>

This is racy - move the blocking into when the task is created and
we're marking it as PF_IO_WORKER anyway. The IO threads are now
prepared to handle signals like SIGSTOP as well, so clear that from
the mask to allow proper stopping of IO threads.

Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Reported-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

We go through various hoops to disallow signals for the IO threads, but
there's really no reason why we cannot just allow them. The IO threads
never return to userspace like a normal thread, and hence don't go through
normal signal processing. Instead, just check for a pending signal as part
of the work loop, and call get_signal() to handle it for us if anything
is pending.

With that, we can support receiving signals, including special ones like
SIGSTOP.

Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Right now we're never calling get_signal() from PF_IO_WORKER threads, but
in preparation for doing so, don't handle a fatal signal for them. The
workers have state they need to cleanup when exiting, so just return
instead of calling do_exit() on their behalf. The threads themselves will
detect a fatal signal and do proper shutdown.

Signed-off-by: Jens Axboe <axboe@kernel.dk>

Arguably we want CQEs of linked requests be in a strict order of
submission as it always was. Now if init of a request fails its CQE may
be posted before all prior linked requests including the head of the
link. Fix it by failing it last.

Fixes: de59bc10

 ("io_uring: fail links more in io_submit_sqe()")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/b7a96b05832e7ab23ad55f84092a2548c4a888b0.1616699075.git.asml.silence@gmail.com


Signed-off-by: Jens Axboe <axboe@kernel.dk>

syzbot reports that it's triggering the warning condition on having
pending work on shutdown:

WARNING: CPU: 1 PID: 12346 at fs/io-wq.c:1061 io_wq_destroy fs/io-wq.c:1061 [inline]
WARNING: CPU: 1 PID: 12346 at fs/io-wq.c:1061 io_wq_put+0x153/0x260 fs/io-wq.c:1072
Modules linked in:
CPU: 1 PID: 12346 Comm: syz-executor.5 Not tainted 5.12.0-rc2-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:io_wq_destroy fs/io-wq.c:1061 [inline]
RIP: 0010:io_wq_put+0x153/0x260 fs/io-wq.c:1072
Code: 8d e8 71 90 ea 01 49 89 c4 41 83 fc 40 7d 4f e8 33 4d 97 ff 42 80 7c 2d 00 00 0f 85 77 ff ff ff e9 7a ff ff ff e8 1d 4d 97 ff <0f> 0b eb b9 8d 6b ff 89 ee 09 de bf ff ff ff ff e8 18 51 97 ff 09
RSP: 0018:ffffc90001ebfb08 EFLAGS: 00010293
RAX: ffffffff81e16083 RBX: ffff888019038040 RCX: ffff88801e86b780
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000040
RBP: 1ffff1100b2f8a80 R08: ffffffff81e15fce R09: ffffed100b2f8a82
R10: ffffed100b2f8a82 R11: 0000000000000000 R12: 0000000000000000
R13: dffffc0000000000 R14: ffff8880597c5400 R15: ffff888019038000
FS:  00007f8dcd89c700(0000) GS:ffff8880b9c00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 000055e9a054e160 CR3: 000000001dfb8000 CR4: 00000000001506f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 io_uring_clean_tctx+0x1b7/0x210 fs/io_uring.c:8802
 __io_uring_files_cancel+0x13c/0x170 fs/io_uring.c:8820
 io_uring_files_cancel include/linux/io_uring.h:47 [inline]
 do_exit+0x258/0x2340 kernel/exit.c:780
 do_group_exit+0x168/0x2d0 kernel/exit.c:922
 get_signal+0x1734/0x1ef0 kernel/signal.c:2773
 arch_do_signal_or_restart+0x3c/0x610 arch/x86/kernel/signal.c:811
 handle_signal_work kernel/entry/common.c:147 [inline]
 exit_to_user_mode_loop kernel/entry/common.c:171 [inline]
 exit_to_user_mode_prepare+0xac/0x1e0 kernel/entry/common.c:208
 __syscall_exit_to_user_mode_work kernel/entry/common.c:290 [inline]
 syscall_exit_to_user_mode+0x48/0x180 kernel/entry/common.c:301
 entry_SYSCALL_64_after_hwframe+0x44/0xae
RIP: 0033:0x465f69

which shouldn't happen, but seems to be possible due to a race on whether
or not the io-wq manager sees a fatal signal first, or whether the io-wq
workers do. If we race with queueing work and then send a fatal signal to
the owning task, and the io-wq worker sees that before the manager sets
IO_WQ_BIT_EXIT, then it's possible to have the worker exit and leave work
behind.

Just turn the WARN_ON_ONCE() into a cancelation condition instead.

Reported-by:  <syzbot+77a738a6bc947bf639ca@syzkaller.appspotmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

WARNING: CPU: 1 PID: 27907 at fs/io_uring.c:7147 io_sq_thread_park+0xb5/0xd0 fs/io_uring.c:7147
CPU: 1 PID: 27907 Comm: iou-sqp-27905 Not tainted 5.12.0-rc4-syzkaller #0
RIP: 0010:io_sq_thread_park+0xb5/0xd0 fs/io_uring.c:7147
Call Trace:
 io_ring_ctx_wait_and_kill+0x214/0x700 fs/io_uring.c:8619
 io_uring_release+0x3e/0x50 fs/io_uring.c:8646
 __fput+0x288/0x920 fs/file_table.c:280
 task_work_run+0xdd/0x1a0 kernel/task_work.c:140
 io_run_task_work fs/io_uring.c:2238 [inline]
 io_run_task_work fs/io_uring.c:2228 [inline]
 io_uring_try_cancel_requests+0x8ec/0xc60 fs/io_uring.c:8770
 io_uring_cancel_sqpoll+0x1cf/0x290 fs/io_uring.c:8974
 io_sqpoll_cancel_cb+0x87/0xb0 fs/io_uring.c:8907
 io_run_task_work_head+0x58/0xb0 fs/io_uring.c:1961
 io_sq_thread+0x3e2/0x18d0 fs/io_uring.c:6763
 ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:294

May happen that last ctx ref is killed in io_uring_cancel_sqpoll(), so
fput callback (i.e. io_uring_release()) is enqueued through task_work,
and run by same cancellation. As it's deeply nested we can't do parking
or taking sqd->lock there, because its state is unclear. So avoid
ctx ejection from sqd list from io_ring_ctx_wait_and_kill() and do it
in a clear context in io_ring_exit_work().

Fixes: f6d54255

 ("io_uring: halt SQO submission on ctx exit")
Reported-by:  <syzbot+e3a3f84f5cecf61f0583@syzkaller.appspotmail.com>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/e90df88b8ff2cabb14a7534601d35d62ab4cb8c7.1616496707.git.asml.silence@gmail.com


Signed-off-by: Jens Axboe <axboe@kernel.dk>

io_provide_buffers_prep()'s "p->len * p->nbufs" to sign extension
problems. Not a huge problem as it's only used for access_ok() and
increases the checked length, but better to keep typing right.

Reported-by: Colin Ian King <colin.king@canonical.com>
Fixes: efe68c1c

 ("io_uring: validate the full range of provided buffers for access")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Colin Ian King <colin.king@canonical.com>
Link: https://lore.kernel.org/r/562376a39509e260d8532186a06226e56eb1f594.1616149233.git.asml.silence@gmail.com


Signed-off-by: Jens Axboe <axboe@kernel.dk>

Don't miss to call kiocb_end_write() from __io_complete_rw() on reissue.
Shouldn't be much of a problem as the function actually does some work
only for ISREG, and NONBLOCK won't be reissued.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/32af9b77c5b874e1bee1a3c46396094bd969e577.1616366969.git.asml.silence@gmail.com


Signed-off-by: Jens Axboe <axboe@kernel.dk>

Request's io-wq work is hashed in io_prep_async_link(), so
as trace_io_uring_queue_async_work() looks at it should follow after
prep has been done.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/709c9f872f4d2e198c7aed9c49019ca7095dd24d.1616366969.git.asml.silence@gmail.com


Signed-off-by: Jens Axboe <axboe@kernel.dk>

Linus correctly points out that this is both unnecessary and generates
much worse code on some archs as going from current to thread_info is
actually backwards - and obviously just wasteful, since the thread_info
is what we care about.

Since io_uring only operates on current for these operations, just use
test_thread_flag() instead. For io-wq, we can further simplify and use
tracehook_notify_signal() to handle the TIF_NOTIFY_SIGNAL work and clear
the flag. The latter isn't an actual bug right now, but it may very well
be in the future if we place other work items under TIF_NOTIFY_SIGNAL.

Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/io-uring/CAHk-=wgYhNck33YHKZ14mFB5MzTTk8gqXHcfj=RWTAXKwgQJgg@mail.gmail.com/


Signed-off-by: Jens Axboe <axboe@kernel.dk>

Without that it's not safe to use them in a linked combination with
others.

Now combinations like IORING_OP_SENDMSG followed by IORING_OP_SPLICE
should be possible.

We already handle short reads and writes for the following opcodes:

- IORING_OP_READV
- IORING_OP_READ_FIXED
- IORING_OP_READ
- IORING_OP_WRITEV
- IORING_OP_WRITE_FIXED
- IORING_OP_WRITE
- IORING_OP_SPLICE
- IORING_OP_TEE

Now we have it for these as well:

- IORING_OP_SENDMSG
- IORING_OP_SEND
- IORING_OP_RECVMSG
- IORING_OP_RECV

For IORING_OP_RECVMSG we also check for the MSG_TRUNC and MSG_CTRUNC
flags in order to call req_set_fail_links().

There might be applications arround depending on the behavior
that even short send[msg]()/recv[msg]() retuns continue an
IOSQE_IO_LINK chain.

It's very unlikely that such applications pass in MSG_WAITALL,
which is only defined in 'man 2 recvmsg', but not in 'man 2 sendmsg'.

It's expected that the low level sock_sendmsg() call just ignores
MSG_WAITALL, as MSG_ZEROCOPY is also ignored without explicitly set
SO_ZEROCOPY.

We also expect the caller to know about the implicit truncation to
MAX_RW_COUNT, which we don't detect.

cc: netdev@vger.kernel.org
Link: https://lore.kernel.org/r/c4e1a4cc0d905314f4d5dc567e65a7b09621aab3.1615908477.git.metze@samba.org


Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Mark the current task as running if we need to run task_work from the
io-wq threads as part of work handling. If that is the case, then return
as such so that the caller can appropriately loop back and reset if it
was part of a going-to-sleep flush.

Fixes: 3bfe6106

 ("io-wq: fork worker threads from original task")
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Just like we don't allow normal signals to IO threads, don't deliver a
STOP to a task that has PF_IO_WORKER set. The IO threads don't take
signals in general, and have no means of flushing out a stop either.

Longer term, we may want to look into allowing stop of these threads,
as it relates to eg process freezing. For now, this prevents a spin
issue if a SIGSTOP is delivered to the parent task.

Reported-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>

They don't take signals individually, and even if they share signals with
the parent task, don't allow them to be delivered through the worker
thread. Linux does allow this kind of behavior for regular threads, but
it's really a compatability thing that we need not care about for the IO
threads.

Reported-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Attaching to already dead/dying SQPOLL task is disallowed in
io_sq_offload_create(), but cleanup is hand coded by calling
io_put_sq_data()/etc., that miss to put ctx->sq_creds.

Defer everything to error-path io_sq_thread_finish(), adding
ctx->sqd_list in the error case as well as finish will handle it.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Link: https://lore.kernel.org/r/ce2a598e66e48347bb04afbaf2acc67c0cc7971a.1615809009.git.metze@samba.org


Signed-off-by: Jens Axboe <axboe@kernel.dk>

Link: https://lore.kernel.org/r/8c1d14f3748105f4caeda01716d47af2fa41d11c.1615809009.git.metze@samba.org


Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

We never want to generate any SIGPIPE, -EPIPE only is much better.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Link: https://lore.kernel.org/r/38961085c3ec49fd21550c7788f214d1ff02d2d4.1615908477.git.metze@samba.org


Signed-off-by: Jens Axboe <axboe@kernel.dk>

Running sqpoll cancellations via task_work_run() is a bad idea because
it depends on other task works to be run, but those may be locked in
currently running task_work_run() because of how it's (splicing the list
in batches).

Enqueue and run them through a separate callback head, namely
struct io_sq_data::park_task_work. As a nice bonus we now precisely
control where it's run, that's much safer than guessing where it can
happen as it was before.

Reported-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

We already have helpers to run/add callback_head but taking ctx and
working with ctx->exit_task_work. Extract generic versions of them
implemented in terms of struct callback_head, it will be used later.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

If io_sq_thread_park() of one task got rescheduled right after
set_bit(), before it gets back to mutex_lock() there can happen
park()/unpark() by another task with SQPOLL locking again and
continuing running never seeing that first set_bit(SHOULD_PARK),
so won't even try to put the mutex down for parking.

It will get parked eventually when SQPOLL drops the lock for reschedule,
but may be problematic and will get in the way of further fixes.

Account number of tasks waiting for parking with a new atomic variable
park_pending and adjust SHOULD_PARK accordingly. It doesn't entirely
replaces SHOULD_PARK bit with this atomic var because it's convenient
to have it as a bit in the state and will help to do optimisations
later.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

io_sq_thread_finish() is called in io_ring_ctx_free(), so SQPOLL task is
potentially running submitting new requests. It's not a disaster because
of using a "try" variant of percpu_ref_get, but is far from nice.

Remove ctx from the sqd ctx list earlier, before cancellation loop, so
SQPOLL can't find it and so won't submit new requests.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

The only user of read-locking of sqd->rw_lock is sq_thread itself, which
is by definition alone, so we don't really need rw_semaphore, but mutex
will do. Replace it with a mutex, and kill read-to-write upgrading and
extra task_work handling in io_sq_thread().

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

If io_req_complete_post() put not a final ref, we can't rely on the
request's ctx ref, and so ctx may potentially be freed while
complete_post() is in io_cqring_ev_posted()/etc.

In that case get an additional ctx reference, and put it in the end, so
protecting following io_cqring_ev_posted(). And also prolong ctx
lifetime until spin_unlock happens, as we do with mutexes, so added
percpu_ref_get() doesn't race with ctx free.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

It's racy to modify req->flags from a not owning context, e.g. linked
timeout calling req_set_fail_links() for the master request might race
with that request setting/clearing flags while being executed
concurrently. Just remove req_set_fail_links(prev) from
io_link_timeout_fn(), io_async_find_and_cancel() and functions down the
line take care of setting the fail bit.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Like we did for the personality idr, convert the IO buffer idr to use
XArray. This avoids a use-after-free on removal of entries, since idr
doesn't like doing so from inside an iterator, and it nicely reduces
the amount of code we need to support this feature.

Fixes: 5a2e745d

 ("io_uring: buffer registration infrastructure")
Cc: stable@vger.kernel.org
Cc: Matthew Wilcox <willy@infradead.org>
Cc: yangerkun <yangerkun@huawei.com>
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

With the freezer using the proper signaling to notify us of when it's
time to freeze a thread, we can re-enable normal freezer usage for the
IO threads. Ensure that SQPOLL, io-wq, and the io-wq manager call
try_to_freeze() appropriately, and remove the default setting of
PF_NOFREEZE from create_io_thread().

Signed-off-by: Jens Axboe <axboe@kernel.dk>

Don't send fake signals to PF_IO_WORKER threads, they don't accept
signals. Just treat them like kthreads in this regard, all they need
is a wakeup as no forced kernel/user transition is needed.

Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

IORING_OP_ASYNC_CANCEL tries io-wq cancellation only for current task.
If it fails go over tctx_list and try it out for every single tctx.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

1) The first problem is io_uring_cancel_sqpoll() ->
io_uring_cancel_task_requests() basically doing park(); park(); and so
hanging.

2) Another one is more subtle, when the master task is doing cancellations,
but SQPOLL task submits in-between the end of the cancellation but
before finish() requests taking a ref to the ctx, and so eternally
locking it up.

3) Yet another is a dying SQPOLL task doing io_uring_cancel_sqpoll() and
same io_uring_cancel_sqpoll() from the owner task, they race for
tctx->wait events. And there probably more of them.

Instead do SQPOLL cancellations from within SQPOLL task context via
task_work, see io_sqpoll_cancel_sync(). With that we don't need temporal
park()/unpark() during cancellation, which is ugly, subtle and anyway
doesn't allow to do io_run_task_work() properly.

io_uring_cancel_sqpoll() is called only from SQPOLL task context and
under sqd locking, so all parking is removed from there. And so,
io_sq_thread_[un]park() and io_sq_thread_stop() are not used now by
SQPOLL task, and that spare us from some headache.

Also remove ctx->sqd_list early to avoid 2). And kill tctx->sqpoll,
which is not used anymore.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

SQPOLL thread to which we're trying to attach may be going away, it's
not nice but a more serious problem is if io_sq_offload_create() sees
sqd->thread==NULL, and tries to init it with a new thread. There are
tons of ways it can be exploited or fail.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

We always do complete(&sqd->startup) almost right after sqd->thread
creation, either in the success path or in io_sq_thread_finish(). It's
specifically created not started for us to be able to set some stuff
like sqd->thread and io_uring_alloc_task_context() before following
right after wake_up_new_task().

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

As io_uring_cancel_files() and others let SQO to run between
io_uring_try_cancel_requests(), SQO may generate new deferred requests,
so it's safer to try to cancel them in it.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

We bypass IOPOLL completion polling (and reaping) for the SQPOLL thread,
but if it's the thread itself invoking cancelations, then we still need
to perform it or no one will.

Fixes: 9936c7c2

 ("io_uring: deduplicate core cancellations sequence")
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Earlier kernels had SQPOLL threads that could share across anything, as
we grabbed the context we needed on a per-ring basis. This is no longer
the case, so only allow attaching directly if we're in the same thread
group. That is the common use case. For non-group tasks, just setup a
new context and thread as we would've done if sharing wasn't set. This
isn't 100% ideal in terms of CPU utilization for the forked and share
case, but hopefully that isn't much of a concern. If it is, there are
plans in motion for how to improve that. Most importantly, we want to
avoid app side regressions where sharing worked before and now doesn't.
With this patch, functionality is equivalent to previous kernels that
supported IORING_SETUP_ATTACH_WQ with SQPOLL.

Reported-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

We use ->ctx_new_list to notify sqo about new ctx pending, then sqo
should stop and splice it to its sqd->ctx_list, paired with
->sq_thread_comp.

The last one is broken because nobody reinitialises it, and trying to
fix it would only add more complexity and bugs. And the first isn't
really needed as is done under park(), that protects from races well.
Add ctx into sqd->ctx_list directly (under park()), it's much simpler
and allows to kill both, ctx_new_list and sq_thread_comp.

note: apparently there is no real problem at the moment, because
sq_thread_comp is used only by io_sq_thread_finish() followed by
parking, where list_del(&ctx->sqd_list) removes it well regardless
whether it's in the new or the active list.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>