Re: [PATCH 08/10] io_uring/cancel: drop sqd->lock while SQPOLL waits

[Jens Axboe <axboe@kernel.dk>]

On 10/30/26 1:11 PM, Karl Mehltretter wrote:
> On Fri, Oct 30, 2026 at 07:34:06AM +0100, Jens Axboe wrote:
>> But no reason not to make the check see it. I've queued up the below,
>> what do you think? It sets IO_SQ_THREAD_SHOULD_STOP in io_sq_thread()
>> before it cancels, and tests that bit in io_sq_offload_create()
>> alongside sqd->thread. Then no new state needed.
> 
> That looks good, except for the WARN_ON_ONCE() in io_sq_thread_stop()
> below.
> 
>> diff --git a/io_uring/sqpoll.c b/io_uring/sqpoll.c
>> index 49c52284ecdd..c833e776320d 100644
>> --- a/io_uring/sqpoll.c
>> +++ b/io_uring/sqpoll.c
>> @@ -433,6 +433,12 @@ static int io_sq_thread(void *data)
>>  	if (io_sq_tw_pending())
>>  		io_sq_tw(UINT_MAX);
>>  
>> +	/*
>> +	 * Cancelation drops sqd->lock while it waits, and a ring could attach
>> +	 * to a thread that is on its way out. A fatal signal doesn't set the
>> +	 * bit, so set it here for io_sq_offload_create() to see.
>> +	 */
>> +	set_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state);
>>  	io_uring_cancel_generic(true, sqd);
> 
> Setting IO_SQ_THREAD_SHOULD_STOP here on a fatal signal means
> io_sq_thread_stop() will see it already set when the last ring drops
> sqd->refs and trigger:
>   
>     WARN_ON_ONCE(test_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state));

Want to send a patch against current master?

-- 
Jens Axboe