iv/linux
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

io_uring: don't punt files update to io-wq unconditionally

Browse Source 
There's no reason to punt it unconditionally, we just need to ensure that
the submit lock grabbing is conditional.

Fixes: 05f3fb3c53 ("io_uring: avoid ring quiesce for fixed file set unregister and update")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This commit is contained in:
Jens Axboe 2021-09-24 08:43:54 -06:00
parent 9990da93d2
commit cdb31c29d3
1 changed files with 2 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
fs/io_uring.c
View File

@ -6340,19 +6340,16 @@ static int io_files_update(struct io_kiocb *req, unsigned int issue_flags)
struct io_uring_rsrc_update2 up;
int ret;
if (issue_flags & IO_URING_F_NONBLOCK)
return -EAGAIN;
up.offset = req->rsrc_update.offset;
up.data = req->rsrc_update.arg;
up.nr = 0;
up.tags = 0;
up.resv = 0;
mutex_lock(&ctx->uring_lock);
io_ring_submit_lock(ctx, !(issue_flags & IO_URING_F_NONBLOCK));
ret = __io_register_rsrc_update(ctx, IORING_RSRC_FILE,
&up, req->rsrc_update.nr_args);
mutex_unlock(&ctx->uring_lock);
io_ring_submit_unlock(ctx, !(issue_flags & IO_URING_F_NONBLOCK));
if (ret < 0)
req_set_fail(req);