user/sven/linux.git/fs/io_uring.c, branch v5.18.3

io_uring: only wake when the correct events are set

2022-06-09T08:30:14Z

[ Upstream commit 1b1d7b4bf1d9948c8dba5ee550459ce7c65ac019 ] The check for waking up a request compares the poll_t bits, however this will always contain some common flags so this always wakes up. For files with single wait queues such as sockets this can cause the request to be sent to the async worker unnecesarily. Further if it is non-blocking will complete the request with EAGAIN which is not desired. Here exclude these common events, making sure to not exclude POLLERR which might be important. Fixes: d7718a9d25a6 ("io_uring: use poll driven retry for files that support it") Signed-off-by: Dylan Yudaken Link: https://lore.kernel.org/r/20220512091834.728610-3-dylany@fb.com Signed-off-by: Jens Axboe Signed-off-by: Sasha Levin

io_uring: avoid io-wq -EAGAIN looping for !IOPOLL

2022-06-09T08:30:13Z

[ Upstream commit e0deb6a025ae8c850dc8685be39fb27b06c88736 ] If an opcode handler semi-reliably returns -EAGAIN, io_wq_submit_work() might continue busily hammer the same handler over and over again, which is not ideal. The -EAGAIN handling in question was put there only for IOPOLL, so restrict it to IOPOLL mode only where there is no other recourse than to retry as we cannot wait. Fixes: def596e9557c9 ("io_uring: support for IO polling") Signed-off-by: Pavel Begunkov Link: https://lore.kernel.org/r/f168b4f24181942f3614dd8ff648221736f572e6.1652433740.git.asml.silence@gmail.com Signed-off-by: Jens Axboe Signed-off-by: Sasha Levin

io_uring: don't attempt to IOPOLL for MSG_RING requests

2022-05-17T18:46:04Z

We gate whether to IOPOLL for a request on whether the opcode is allowed on a ring setup for IOPOLL and if it's got a file assigned. MSG_RING is the only one that allows a file yet isn't pollable, it's merely supported to allow communication on an IOPOLL ring, not because we can poll for completion of it. Put the assigned file early and clear it, so we don't attempt to poll for it. Reported-by: syzbot+1a0a53300ce782f8b3ad@syzkaller.appspotmail.com Fixes: 3f1d52abf098 ("io_uring: defer msg-ring file validity check until command issue") Signed-off-by: Jens Axboe

io_uring: assign non-fixed early for async work

2022-05-02T14:09:39Z

We defer file assignment to ensure that fixed files work with links between a direct accept/open and the links that follow it. But this has the side effect that normal file assignment is then not complete by the time that request submission has been done. For deferred execution, if the file is a regular file, assign it when we do the async prep anyway. Signed-off-by: Jens Axboe

io_uring: check that data field is 0 in ringfd unregister

2022-04-29T14:39:43Z

Only allow data field to be 0 in struct io_uring_rsrc_update user arguments to allow for future possible usage. Fixes: e7a6c00dc77a ("io_uring: add support for registering ring file descriptors") Signed-off-by: Eugene Syromiatnikov Link: https://lore.kernel.org/r/20220429142218.GA28696@asgard.redhat.com Signed-off-by: Jens Axboe

io_uring: fix uninitialized field in rw io_kiocb

2022-04-28T19:13:43Z

io_rw_init_file does not initialize kiocb->private, so when iocb_bio_iopoll reads kiocb->private it can contain uninitialized data. Fixes: 3e08773c3841 ("block: switch polling to be bio based") Signed-off-by: Joseph Ravichandran Signed-off-by: Jens Axboe

io_uring: check reserved fields for recv/recvmsg

2022-04-27T02:48:37Z

We should check unused fields for non-zero and -EINVAL if they are set, making it consistent with other opcodes. Fixes: aa1fa28fc73e ("io_uring: add support for recvmsg()") Signed-off-by: Jens Axboe

io_uring: check reserved fields for send/sendmsg

2022-04-27T02:48:31Z

We should check unused fields for non-zero and -EINVAL if they are set, making it consistent with other opcodes. Fixes: 0fa03c624d8f ("io_uring: add support for sendmsg()") Signed-off-by: Jens Axboe

io_uring: fix leaks on IOPOLL and CQE_SKIP

2022-04-17T12:54:11Z

If all completed requests in io_do_iopoll() were marked with REQ_F_CQE_SKIP, we'll not only skip CQE posting but also io_free_batch_list() leaking memory and resources. Move @nr_events increment before REQ_F_CQE_SKIP check. We'll potentially return the value greater than the real one, but iopolling will deal with it and the userspace will re-iopoll if needed. In anyway, I don't think there are many use cases for REQ_F_CQE_SKIP + IOPOLL. Fixes: 83a13a4181b0e ("io_uring: tweak iopoll CQE_SKIP event counting") Signed-off-by: Pavel Begunkov Link: https://lore.kernel.org/r/5072fc8693fbfd595f89e5d4305bfcfd5d2f0a64.1650186611.git.asml.silence@gmail.com Signed-off-by: Jens Axboe

io_uring: free iovec if file assignment fails

2022-04-17T03:14:00Z

We just return failure in this case, but we need to release the iovec first. If we're doing IO with more than FAST_IOV segments, then the iovec is allocated and must be freed. Reported-by: syzbot+96b43810dfe9c3bb95ed@syzkaller.appspotmail.com Fixes: 584b0180f0f4 ("io_uring: move read/write file prep state into actual opcode handler") Signed-off-by: Jens Axboe