Re: [PATCH 5/6] io_uring/futex: use generic io_cancel_remove() helper

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 2/6/25 5:56 AM, lizetao wrote:
> Hi,
> 
>> -----Original Message-----
>> From: Jens Axboe <axboe@xxxxxxxxx>
>> Sent: Thursday, February 6, 2025 4:26 AM
>> To: io-uring@xxxxxxxxxxxxxxx
>> Cc: Jens Axboe <axboe@xxxxxxxxx>
>> Subject: [PATCH 5/6] io_uring/futex: use generic io_cancel_remove() helper
>>
>> Don't implement our own loop rolling and checking, just use the generic helper to
>> find and cancel requests.
>>
>> Signed-off-by: Jens Axboe <axboe@xxxxxxxxx>
>> ---
>>  io_uring/futex.c | 24 +-----------------------
>>  1 file changed, 1 insertion(+), 23 deletions(-)
>>
>> diff --git a/io_uring/futex.c b/io_uring/futex.c index
>> 808eb57f1210..54b9760f2aa6 100644
>> --- a/io_uring/futex.c
>> +++ b/io_uring/futex.c
>> @@ -116,29 +116,7 @@ static bool __io_futex_cancel(struct io_kiocb *req)  int
>> io_futex_cancel(struct io_ring_ctx *ctx, struct io_cancel_data *cd,
>>  		    unsigned int issue_flags)
>>  {
>> -	struct hlist_node *tmp;
>> -	struct io_kiocb *req;
>> -	int nr = 0;
>> -
>> -	if (cd->flags &
>> (IORING_ASYNC_CANCEL_FD|IORING_ASYNC_CANCEL_FD_FIXED))
>> -		return -ENOENT;
> 
> Why remove this check?

It isn't really necessary. Yes we could loop pointlessly if they are set
and not find anything, but it's really just a bad use case from the
application. End result should be the same, that -ENOENT is returned on
trying to lookup a futex operation based on the fd.

-- 
Jens Axboe




[Index of Archives]     [Linux Samsung SoC]     [Linux Rockchip SoC]     [Linux Actions SoC]     [Linux for Synopsys ARC Processors]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]


  Powered by Linux