io_poll_double_wake() is called for both request types - both pure poll
requests, and internal polls. This means that we should be using the
right handler based on the request type. Use the one that the original
caller already assigned for the waitqueue handling, that will always
match the correct type.

Cc: stable@vger.kernel.org # v5.8+
Reported-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Restructure __io_queue_sqe() so it follows simple if/else if/else
control flow. It's more readable and removes extra goto/labels.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Don't overuse goto's, complex control flow doesn't make compilers happy
and makes code harder to read.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Set IO_WQ_WORK_CONCURRENT for all REQ_F_FORCE_ASYNC requests, do that in
that is also looks better.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Inline io_link_cancel_timeout() and __io_kill_linked_timeout() into
io_kill_linked_timeout(). That allows to easily move a put of a cancelled
linked timeout out of completion_lock and to not deferring it. It is also
much more readable when not scattered across three different functions.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Move REQ_F_LINK_TIMEOUT clearing out of __io_kill_linked_timeout()
because it might return early and leave the flag set. It's not a
problem, but may be confusing.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

An armed linked timeout can never be a head of a link, so we don't need
to clear REQ_F_LINK_HEAD for it.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

__io_kill_linked_timeout() already checks for REQ_F_LTIMEOUT_ACTIVE and
it's set only for linked timeouts. No need to verify next request's
opcode.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

With the set_fs change, we can no longer rely on copy_{to,from}_user()
accepting a kernel pointer, and it was bad form to do so anyway. Clean
this up and change the internal helper that io_uring uses to deal with
kernel pointers instead. This puts the offset copy in/out in __do_splice()
instead, which just calls the same helper.
Signed-off-by: Jens Axboe <axboe@kernel.dk>

We jump through a hoop for fixed buffers, where we first map these to
a bvec(), then kmap() the bvec to obtain the pointer we copy to/from.
This was always a bit ugly, and with the set_fs changes, it ends up
being practically problematic as well.

There's no need to jump through these hoops, just use the original user
pointers and length for the non iter based read/write.
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Every close(io_uring) causes cancellation of all inflight requests
carrying ->files. That's not nice but was neccessary up until recently.
Now task->files removal is handled in the core code, so that part of
flush can be removed.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

We correctly set io-wq NUMA node affinities when the io-wq context is
setup, but if an entire node CPU set is offlined and then brought back
online, the per node affinities are broken. Ensure that we set them
again whenever a CPU comes online. This ensures that we always track
the right node affinity. The usual cpuhp notifiers are used to drive it.
Reported-by: Zhang Qiang <qiang.zhang@windriver.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Clear linked_timeout for next requests in __io_queue_sqe() so we won't
queue it up unnecessary when it's going to be punted.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Cc: stable@vger.kernel.org # v5.9
Signed-off-by: Jens Axboe <axboe@kernel.dk>

This one was missed in the earlier conversion, should be included like
any of the other IO identity flags. Make sure we restore to RLIM_INIFITY
when dropping the personality again.

Fixes: 98447d65 ("io_uring: move io identity items into separate struct")
Signed-off-by: Jens Axboe <axboe@kernel.dk>

io_link_timeout_fn() removes REQ_F_LINK_TIMEOUT from the link head's
flags, it's not atomic and may race with what the head is doing.

If io_link_timeout_fn() doesn't clear the flag, as forced by this patch,
then it may happen that for "req -> link_timeout1 -> link_timeout2",
__io_kill_linked_timeout() would find link_timeout2 and try to cancel
it, so miscounting references. Teach it to ignore such double timeouts
by marking the active one with a new flag in io_prep_linked_timeout().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Move INIT_HLIST_NODE(&req->hash_node) into __io_arm_poll_handler(), so
that it doesn't duplicated and common poll code would be responsible for
it.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

io_poll_task_handler() doesn't add clarity, inline it in its only user.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

io_poll_add_prep() doesn't need to verify ->file because it's already
done in io_init_req().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

ctx->cached_cq_overflow is changed only under completion_lock. Convert
it from atomic_t to just int, and mark all places when it's read without
lock with READ_ONCE, which guarantees atomicity (relaxed ordering).
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Inline io_fail_links() and kill extra io_cqring_ev_posted().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Don't take an identity on personality/creds init only to drop it a few
lines after. Extract a function which prepares req->work but leaves it
without identity.

Note: it's safe to not check REQ_F_WORK_INITIALIZED there because it's
nobody had a chance to init it before io_init_req().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Use IO_WQ_WORK_CREDS to figure out if req has creds to be used.
Since recently it should rely only on flags, but not value of
work.creds.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

commit 021a2446 ("block: add QUEUE_FLAG_NOWAIT") adds a new helper
function blk_queue_nowait() to check if the bdev supports handling of
REQ_NOWAIT or not. Since then bio-based dm device can also support
REQ_NOWAIT, and currently only dm-linear supports that since
commit 6abc4946 ("dm: add support for REQ_NOWAIT and enable it for
linear target").
Signed-off-by: Jeffle Xu <jefflexu@linux.alibaba.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

For the case where read-ahead is disabled on the file, or if the cgroup
is congested, ensure that we can at least do 1 page of read-ahead to
make progress on the read in an async fashion. This could potentially be
larger, but it's not needed in terms of functionality, so let's error on
the side of caution as larger counts of pages may run into reclaim
issues (particularly if we're congested).

This makes sure we're not hitting the potentially sync ->readpage() path
for IO that is marked IOCB_WAITQ, which could cause us to block. It also
means we'll use the same path for IO, regardless of whether or not
read-ahead happens to be disabled on the lower level device.
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Reported-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reported-by: Hao_Xu <haoxu@linux.alibaba.com>
[axboe: updated for new ractl API]
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Once we've copied some data for an iocb that is marked with IOCB_WAITQ,
we should no longer attempt to async lock a new page. Instead make sure
we return the copied amount, and let the caller retry, instead of
returning -EIOCBQUEUED for a new page.

This should only be possible with read-ahead disabled on the below
device, and multiple threads racing on the same file. Haven't been able
to reproduce on anything else.

Cc: stable@vger.kernel.org # v5.9
Fixes: 1a0a7853 ("mm: support async buffered reads in generic_file_buffered_read()")
Reported-by: Kent Overstreet <kent.overstreet@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

__io_queue_proc() is used by both, poll reqs and apoll. Don't use
req->poll.events to copy poll mask because for apoll it aliases with
private data of the request.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Make sure the async io-wq workers inherit the loginuid and sessionid from
the original task, and restore them to unset once we're done with the
async work item.

While at it, disable the ability for kernel threads to write to their own
loginuid.
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Even though we place the req_issued and req_complete in separate
cachelines, there's considerable overhead in doing the atomics
particularly on the completion side.

Get rid of having the two counters, and just use a percpu_counter for
this. That's what it was made for, after all. This considerably
reduces the overhead in __io_free_req().
Signed-off-by: Jens Axboe <axboe@kernel.dk>

This avoids doing a copy for each new async IO, if some parts of the
io_identity has changed. We avoid reference counting for the normal
fast path of nothing ever changing.
Signed-off-by: Jens Axboe <axboe@kernel.dk>

This is, by definition, a per-task structure. So store it in the
task context, instead of doing carrying it in each io_kiocb. We're being
a bit inefficient if members have changed, as that requires an alloc and
copy of a new io_identity struct. The next patch will fix that up.
Signed-off-by: Jens Axboe <axboe@kernel.dk>

If the io_identity doesn't completely match the task, then create a
copy of it and use that. The existing copy remains valid until the last
user of it has gone away.

This also changes the personality lookup to be indexed by io_identity,
instead of creds directly.
Signed-off-by: Jens Axboe <axboe@kernel.dk>

io-wq contains a pointer to the identity, which we just hold in io_kiocb
for now. This is in preparation for putting this outside io_kiocb. The
only exception is struct files_struct, which we'll need different rules
for to avoid a circular dependency.

No functional changes in this patch.
Signed-off-by: Jens Axboe <axboe@kernel.dk>

We solely rely on work->work_flags now, so use that for proper checking
and clearing/dropping of various identity items.
Signed-off-by: Jens Axboe <axboe@kernel.dk>

We have a number of bits that decide what context to inherit. Set up
io-wq flags for these instead. This is in preparation for always having
the various members set, but not always needing them for all requests.

No intended functional changes in this patch.
Signed-off-by: Jens Axboe <axboe@kernel.dk>

There was an assumption that kthread_create_on_node() would properly set
NUMA affinities in terms of CPUs allowed, but it doesn't. Make sure we
do this when creating an io-wq context on NUMA.

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

syzbot reports the following crash:

general protection fault, probably for non-canonical address 0xdffffc0000000000: 0000 [#1] PREEMPT SMP KASAN
KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
CPU: 1 PID: 8927 Comm: syz-executor.3 Not tainted 5.9.0-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:io_file_from_index fs/io_uring.c:5963 [inline]
RIP: 0010:io_sqe_files_register fs/io_uring.c:7369 [inline]
RIP: 0010:__io_uring_register fs/io_uring.c:9463 [inline]
RIP: 0010:__do_sys_io_uring_register+0x2fd2/0x3ee0 fs/io_uring.c:9553
Code: ec 03 49 c1 ee 03 49 01 ec 49 01 ee e8 57 61 9c ff 41 80 3c 24 00 0f 85 9b 09 00 00 4d 8b af b8 01 00 00 4c 89 e8 48 c1 e8 03 <80> 3c 28 00 0f 85 76 09 00 00 49 8b 55 00 89 d8 c1 f8 09 48 98 4c
RSP: 0018:ffffc90009137d68 EFLAGS: 00010246
RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffffc9000ef2a000
RDX: 0000000000040000 RSI: ffffffff81d81dd9 RDI: 0000000000000005
RBP: dffffc0000000000 R08: 0000000000000001 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000000 R12: ffffed1012882a37
R13: 0000000000000000 R14: ffffed1012882a38 R15: ffff888094415000
FS:  00007f4266f3c700(0000) GS:ffff8880ae500000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 000000000118c000 CR3: 000000008e57d000 CR4: 00000000001506e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46
 entry_SYSCALL_64_after_hwframe+0x44/0xa9
RIP: 0033:0x45de59
Code: 0d b4 fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 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 0f 83 db b3 fb ff c3 66 2e 0f 1f 84 00 00 00 00
RSP: 002b:00007f4266f3bc78 EFLAGS: 00000246 ORIG_RAX: 00000000000001ab
RAX: ffffffffffffffda RBX: 00000000000083c0 RCX: 000000000045de59
RDX: 0000000020000280 RSI: 0000000000000002 RDI: 0000000000000005
RBP: 000000000118bf68 R08: 0000000000000000 R09: 0000000000000000
R10: 40000000000000a1 R11: 0000000000000246 R12: 000000000118bf2c
R13: 00007fff2fa4f12f R14: 00007f4266f3c9c0 R15: 000000000118bf2c
Modules linked in:
---[ end trace 2a40a195e2d5e6e6 ]---
RIP: 0010:io_file_from_index fs/io_uring.c:5963 [inline]
RIP: 0010:io_sqe_files_register fs/io_uring.c:7369 [inline]
RIP: 0010:__io_uring_register fs/io_uring.c:9463 [inline]
RIP: 0010:__do_sys_io_uring_register+0x2fd2/0x3ee0 fs/io_uring.c:9553
Code: ec 03 49 c1 ee 03 49 01 ec 49 01 ee e8 57 61 9c ff 41 80 3c 24 00 0f 85 9b 09 00 00 4d 8b af b8 01 00 00 4c 89 e8 48 c1 e8 03 <80> 3c 28 00 0f 85 76 09 00 00 49 8b 55 00 89 d8 c1 f8 09 48 98 4c
RSP: 0018:ffffc90009137d68 EFLAGS: 00010246
RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffffc9000ef2a000
RDX: 0000000000040000 RSI: ffffffff81d81dd9 RDI: 0000000000000005
RBP: dffffc0000000000 R08: 0000000000000001 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000000 R12: ffffed1012882a37
R13: 0000000000000000 R14: ffffed1012882a38 R15: ffff888094415000
FS:  00007f4266f3c700(0000) GS:ffff8880ae400000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 000000000074a918 CR3: 000000008e57d000 CR4: 00000000001506f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400

which is a copy of fget failure condition jumping to cleanup, but the
cleanup requires ctx->file_data to be assigned. Assign it when setup,
and ensure that we clear it again for the error path exit.

Fixes: 5398ae69 ("io_uring: clean file_data access in files_register")
Reported-by: syzbot+f4ebcc98223dafd8991e@syzkaller.appspotmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

This reverts commit 738277ad.

This change didn't make a lot of sense, and as Linus reports, it actually
fails on clang:

   /tmp/io_uring-dd40c4.s:26476: Warning: ignoring changed section
   attributes for .data..read_mostly

The arrays are already marked const so, by definition, they are not
just read-mostly, they are read-only.
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

REQ_F_COMP_LOCKED is used and implemented in a buggy way. The problem is
that the flag is set before io_put_req() but not cleared after, and if
that wasn't the final reference, the request will be freed with the flag
set from some other context, which may not hold a spinlock. That means
possible races with removing linked timeouts and unsynchronised
completion (e.g. access to CQ).

Instead of fixing REQ_F_COMP_LOCKED, kill the flag and use
task_work_add() to move such requests to a fresh context to free from
it, as was done with __io_free_req_finish().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

io_req_clean_work() checks REQ_F_COMP_LOCK to pass this two layers up.
Move the check up into __io_free_req(), so at least it doesn't looks so
ugly and would facilitate further changes.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Move io_put_req() in io_poll_task_handler() from under spinlock. This
eliminates the need to use REQ_F_COMP_LOCKED, at the expense of
potentially having to grab the lock again. That's still a better trade
off than relying on the locked flag.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>