Jens Axboe 65521ecb86 io_uring/cancel: de-unionize file and user_data in struct io_cancel_data ...

[ Upstream commit 22dbb0987b ]

By having them share the same space in struct io_cancel_data, it ends up
disallowing IORING_ASYNC_CANCEL_FD|IORING_ASYNC_CANCEL_USERDATA from
working. Eg you cannot match on both a file and user_data for
cancelation purposes. This obviously isn't a common use case as nobody
has reported this, but it does result in -ENOENT potentially being
returned when trying to match on both, rather than actually doing what
the API says it would.

Fixes: 4bf94615b8 ("io_uring: allow IORING_OP_ASYNC_CANCEL with 'fd' key")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>

2026-03-04 07:20:34 -05:00

..

advise.c

…

advise.h

…

alloc_cache.h

…

cancel.c

…

cancel.h

io_uring/cancel: de-unionize file and user_data in struct io_cancel_data

2026-03-04 07:20:34 -05:00

epoll.c

…

epoll.h

…

eventfd.c

io_uring/eventfd: ensure io_eventfd_signal() defers another RCU period

2025-01-17 13:40:58 +01:00

eventfd.h

…

fdinfo.c

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

2025-10-29 14:08:59 +01:00

fdinfo.h

…

filetable.c

io_uring: correct __must_hold annotation in io_install_fixed_file

2025-10-29 14:08:59 +01:00

filetable.h

…

fs.c

…

fs.h

…

futex.c

io_uring/futex: ensure io_futex_wait() cleans up properly on failure

2025-08-28 16:31:05 +02:00

futex.h

…

io_uring.c

io_uring: use release-acquire ordering for IORING_SETUP_R_DISABLED

2026-03-04 07:19:27 -05:00

io_uring.h

io_uring: include dying ring in task_work "should cancel" state

2025-09-25 11:13:47 +02:00

io-wq.c

io_uring/io-wq: check IO_WQ_BIT_EXIT inside work run loop

2026-01-30 10:28:41 +01:00

io-wq.h

…

kbuf.c

io_uring/kbuf: flag partial buffer mappings

2025-07-06 11:01:48 +02:00

kbuf.h

io_uring/kbuf: drop WARN_ON_ONCE() from incremental length check

2025-09-25 11:13:48 +02:00

Makefile

…

memmap.c

…

memmap.h

…

msg_ring.c

io_uring: use release-acquire ordering for IORING_SETUP_R_DISABLED

2026-03-04 07:19:27 -05:00

msg_ring.h

…

napi.c

io_uring/napi: fix io_napi_entry RCU accesses

2025-11-24 10:36:06 +01:00

napi.h

…

net.c

io_uring/net: commit partial buffers on retry

2025-08-28 16:31:04 +02:00

net.h

…

nop.c

…

nop.h

…

notif.c

io_uring/zctx: check chained notif contexts

2025-11-13 15:34:03 -05:00

notif.h

…

opdef.c

io_uring: make fallocate be hashed work

2025-07-17 18:37:21 +02:00

opdef.h

…

openclose.c

io_uring: fix filename leak in __io_openat_prep()

2026-01-08 10:14:13 +01:00

openclose.h

…

poll.c

io_uring/poll: correctly handle io_poll_add() return value on update

2026-01-08 10:14:25 +01:00

poll.h

…

refs.h

io_uring: always do atomic put from iowq

2025-05-02 07:59:21 +02:00

register.c

io_uring: use release-acquire ordering for IORING_SETUP_R_DISABLED

2026-03-04 07:19:27 -05:00

register.h

…

rsrc.c

io_uring/rsrc: don't rely on user vaddr alignment

2025-07-06 11:01:47 +02:00

rsrc.h

io_uring/rsrc: don't rely on user vaddr alignment

2025-07-06 11:01:47 +02:00

rw.c

io_uring/rw: recycle buffers manually for non-mshot reads

2026-02-12 13:14:56 +01:00

rw.h

…

slist.h

…

splice.c

…

splice.h

…

sqpoll.c

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

2025-10-29 14:08:59 +01:00

sqpoll.h

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

2025-10-29 14:08:59 +01:00

statx.c

…

statx.h

…

sync.c

io_uring/sync: validate passed in offset

2026-03-04 07:19:28 -05:00

sync.h

…

tctx.c

…

tctx.h

…

timeout.c

io_uring: include dying ring in task_work "should cancel" state

2025-09-25 11:13:47 +02:00

timeout.h

…

truncate.c

…

truncate.h

…

uring_cmd.c

io_uring: include dying ring in task_work "should cancel" state

2025-09-25 11:13:47 +02:00

uring_cmd.h

…

waitid.c

io_uring/waitid: always prune wait queue entry in io_waitid_wait()

2025-10-15 12:00:20 +02:00

waitid.h

…

xattr.c

…

xattr.h

…