On 3/8/21 6:20 AM, Pavel Begunkov wrote: > There are enough of problems with IORING_SETUP_R_DISABLED, including the > burden of checking and kicking off the SQO task all over the codebase -- > for exit/cancel/etc. > > Rework it, always start the thread but don't do submit unless the flag > is gone, that's much easier. This is a good simplification, much better than having the weird state that is not started. -- Jens Axboe
