Skip to content

K
Kernel

openeuler / Kernel
1 年多 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 20 8月, 2020 4 次提交
    • P
      io_uring: kill extra iovec=NULL in import_iovec() · 867a23ea
      由 Pavel Begunkov 提交于 
      If io_import_iovec() returns an error, return iovec is undefined and
must not be used, so don't set it to NULL when failing.
Signed-off-by: NPavel Begunkov <asml.silence@gmail.com>
Signed-off-by: NJens Axboe <axboe@kernel.dk>
      867a23ea
    • P
      io_uring: comment on kfree(iovec) checks · f261c168
      由 Pavel Begunkov 提交于 
      kfree() handles NULL pointers well, but io_{read,write}() checks it
because of performance reasons. Leave a comment there for those who are
tempted to patch it.
Signed-off-by: NPavel Begunkov <asml.silence@gmail.com>
Signed-off-by: NJens Axboe <axboe@kernel.dk>
      f261c168
    • P
      io_uring: fix racy req->flags modification · bb175342
      由 Pavel Begunkov 提交于 
      Setting and clearing REQ_F_OVERFLOW in io_uring_cancel_files() and
io_cqring_overflow_flush() are racy, because they might be called
asynchronously.

REQ_F_OVERFLOW flag in only needed for files cancellation, so if it can
be guaranteed that requests _currently_ marked inflight can't be
overflown, the problem will be solved with removing the flag
altogether.

That's how the patch works, it removes inflight status of a request
in io_cqring_fill_event() whenever it should be thrown into CQ-overflow
list. That's Ok to do, because no opcode specific handling can be done
after io_cqring_fill_event(), the same assumption as with "struct
io_completion" patches.
And it already have a good place for such cleanups, which is
io_clean_op(). A nice side effect of this is removing this inflight
check from the hot path.

note on synchronisation: now __io_cqring_fill_event() may be taking two
spinlocks simultaneously, completion_lock and inflight_lock. It's fine,
because we never do that in reverse order, and CQ-overflow of inflight
requests shouldn't happen often.
Signed-off-by: NPavel Begunkov <asml.silence@gmail.com>
Signed-off-by: NJens Axboe <axboe@kernel.dk>
      bb175342
    • J
      io_uring: use system_unbound_wq for ring exit work · fc666777
      由 Jens Axboe 提交于 
      We currently use system_wq, which is unbounded in terms of number of
workers. This means that if we're exiting tons of rings at the same
time, then we'll briefly spawn tons of event kworkers just for a very
short blocking time as the rings exit.

Use system_unbound_wq instead, which has a sane cap on the concurrency
level.
Signed-off-by: NJens Axboe <axboe@kernel.dk>
      fc666777
  3. 19 8月, 2020 1 次提交
    • J
      io_uring: cleanup io_import_iovec() of pre-mapped request · 8452fd0c
      由 Jens Axboe 提交于 
      io_rw_prep_async() goes through a dance of clearing req->io, calling
the iovec import, then re-setting req->io. Provide an internal helper
that does the right thing without needing state tweaked to get there.

This enables further cleanups in io_read, io_write, and
io_resubmit_prep(), but that's left for another time.
Signed-off-by: NJens Axboe <axboe@kernel.dk>
      8452fd0c
  5. 17 8月, 2020 2 次提交
    • J
      io_uring: get rid of kiocb_wait_page_queue_init() · 3b2a4439
      由 Jens Axboe 提交于 
      The 5.9 merge moved this function io_uring, which means that we don't
need to retain the generic nature of it. Clean up this part by removing
redundant checks, and just inlining the small remainder in
io_rw_should_retry().

No functional changes in this patch.
Signed-off-by: NJens Axboe <axboe@kernel.dk>
      3b2a4439
    • J
      io_uring: find and cancel head link async work on files exit · b711d4ea
      由 Jens Axboe 提交于 
      Commit f254ac04 ("io_uring: enable lookup of links holding inflight files")
only handled 2 out of the three head link cases we have, we also need to
lookup and cancel work that is blocked in io-wq if that work has a link
that's holding a reference to the files structure.

Put the "cancel head links that hold this request pending" logic into
io_attempt_cancel(), which will to through the motions of finding and
canceling head links that hold the current inflight files stable request
pending.

Cc: stable@vger.kernel.org
Reported-by: NPavel Begunkov <asml.silence@gmail.com>
Signed-off-by: NJens Axboe <axboe@kernel.dk>
      b711d4ea
  7. 16 8月, 2020 2 次提交
    • J
      io_uring: short circuit -EAGAIN for blocking read attempt · f91daf56
      由 Jens Axboe 提交于 
      One case was missed in the short IO retry handling, and that's hitting
-EAGAIN on a blocking attempt read (eg from io-wq context). This is a
problem on sockets that are marked as non-blocking when created, they
don't carry any REQ_F_NOWAIT information to help us terminate them
instead of perpetually retrying.

Fixes: 227c0c96 ("io_uring: internally retry short reads")
Signed-off-by: NJens Axboe <axboe@kernel.dk>
      f91daf56
    • J
      io_uring: sanitize double poll handling · d4e7cd36
      由 Jens Axboe 提交于 
      There's a bit of confusion on the matching pairs of poll vs double poll,
depending on if the request is a pure poll (IORING_OP_POLL_ADD) or
poll driven retry.

Add io_poll_get_double() that returns the double poll waitqueue, if any,
and io_poll_get_single() that returns the original poll waitqueue. With
that, remove the argument to io_poll_remove_double().

Finally ensure that wait->private is cleared once the double poll handler
has run, so that remove knows it's already been seen.

Cc: stable@vger.kernel.org # v5.8
Reported-by: syzbot+7f617d4a9369028b8a2c@syzkaller.appspotmail.com
Fixes: 18bceab1 ("io_uring: allow POLL_ADD with double poll_wait() users")
Signed-off-by: NJens Axboe <axboe@kernel.dk>
      d4e7cd36
  9. 15 8月, 2020 2 次提交
  11. 14 8月, 2020 3 次提交
    • S
      SMB3: Fix mkdir when idsfromsid configured on mount · c8c412f9
      由 Steve French 提交于 
      mkdir uses a compounded create operation which was not setting
the security descriptor on create of a directory. Fix so
mkdir now sets the mode and owner info properly when idsfromsid
and modefromsid are configured on the mount.
Signed-off-by: NSteve French <stfrench@microsoft.com>
CC: Stable <stable@vger.kernel.org> # v5.8
Reviewed-by: NPaulo Alcantara (SUSE) <pc@cjr.nz>
Reviewed-by: NPavel Shilovsky <pshilov@microsoft.com>
      c8c412f9
    • J
      io_uring: internally retry short reads · 227c0c96
      由 Jens Axboe 提交于 
      We've had a few application cases of not handling short reads properly,
and it is understandable as short reads aren't really expected if the
application isn't doing non-blocking IO.

Now that we retain the iov_iter over retries, we can implement internal
retry pretty trivially. This ensures that we don't return a short read,
even for buffered reads on page cache conflicts.

Cleanup the deep nesting and hard to read nature of io_read() as well,
it's much more straight forward now to read and understand. Added a
few comments explaining the logic as well.
Signed-off-by: NJens Axboe <axboe@kernel.dk>
      227c0c96
    • J
      io_uring: retain iov_iter state over io_read/io_write calls · ff6165b2
      由 Jens Axboe 提交于 
      Instead of maintaining (and setting/remembering) iov_iter size and
segment counts, just put the iov_iter in the async part of the IO
structure.

This is mostly a preparation patch for doing appropriate internal retries
for short reads, but it also cleans up the state handling nicely and
simplifies it quite a bit.
Signed-off-by: NJens Axboe <axboe@kernel.dk>
      ff6165b2
  13. 13 8月, 2020 25 次提交
  15. 12 8月, 2020 1 次提交
    • T
      NFS: Fix flexfiles read failover · 563c53e7
      由 Trond Myklebust 提交于 
      The current mirrored read failover code is correctly resetting the mirror
index between failed reads, however it is not able to actually flip the
RPC call over to the next RPC client.
The end result is that we keep resending the RPC call to the same client
over and over.

The fix is to use the pnfs_read_resend_pnfs() mechanism to schedule a
new RPC call, but we need to add the ability to pass in a mirror
index so that we always retry the next mirror in the list.

Fixes: 166bd5b8 ("pNFS/flexfiles: Fix layoutstats handling during read failovers")
Signed-off-by: NTrond Myklebust <trond.myklebust@hammerspace.com>
      563c53e7