Re: [PATCH] io_uring: no read-retry on -EAGAIN error and O_NONBLOCK marked file

[Jens Axboe <axboe@kernel.dk>]

On 9/2/20 10:09 AM, Pavel Begunkov wrote:
> On 02/09/2020 19:00, Jens Axboe wrote:
>>     Actually two things that need fixing up here:
>>     
>>     - The io_rw_reissue() -EAGAIN retry is explicit to block devices and
>>       regular files, so don't ever attempt to do that on other types of
>>       files.
>>     
>>     - If we hit -EAGAIN on a nonblock marked file, don't arm poll handler for
>>       it. It should just complete with -EAGAIN.
>>     
>>     Cc: stable@vger.kernel.org
>>     Reported-by: Norman Maurer <norman.maurer@googlemail.com>
>>     Signed-off-by: Jens Axboe <axboe@kernel.dk>
>>
>> ---
>>
>> diff --git a/fs/io_uring.c b/fs/io_uring.c
>> index b1ccd7072d93..65656102bbeb 100644
>> --- a/fs/io_uring.c
>> +++ b/fs/io_uring.c
>> @@ -2300,8 +2300,11 @@ static bool io_resubmit_prep(struct io_kiocb *req, int error)
>>  static bool io_rw_reissue(struct io_kiocb *req, long res)
>>  {
>>  #ifdef CONFIG_BLOCK
>> +	umode_t mode = file_inode(req->file)->i_mode;
>>  	int ret;
>>  
>> +	if (!S_ISBLK(mode) && !S_ISREG(mode))
>> +		return false;
>>  	if ((res != -EAGAIN && res != -EOPNOTSUPP) || io_wq_current_is_worker())
>>  		return false;
>>  
>> @@ -3146,6 +3149,9 @@ static int io_read(struct io_kiocb *req, bool force_nonblock,
>>  		/* IOPOLL retry should happen for io-wq threads */
>>  		if (!force_nonblock && !(req->ctx->flags & IORING_SETUP_IOPOLL))
>>  			goto done;
>> +		/* no retry on NONBLOCK marked file */
>> +		if (req->file->f_flags & O_NONBLOCK)
>> +			goto done;
> 
> Looks like it works with open(O_NONBLOCK) but not with IOCB_NOWAIT in the
> request's flags. Is that so?

The IOCB_NOWAIT case should already be fine, it's just the O_NONBLOCK that
isn't covered. IOCB_NOWAIT on a socket unhelpfully fails anyway, since
FMODE_NOWAIT isn't available. Arguably something that should be fixed
separately.

-- 
Jens Axboe