On Tue, 30 Aug 2022 05:50:06 -0700, Dylan Yudaken wrote:
> We have seen workloads which suffer due to the way task work is currently
> scheduled. This scheduling can cause non-trivial tasks to run interrupting
> useful work on the workload. For example in network servers, a large async
> recv may run, calling memcpy on a large packet, interrupting a send. Which
> would add latency.
>
> This series adds an option to defer async work until user space calls
> io_uring_enter with the GETEVENTS flag. This allows the workload to choose
> when to schedule async work and have finer control (at the expense of
> complexity of managing this) of scheduling.
>
> [...]
Applied, thanks!
[1/7] io_uring: remove unnecessary variable
commit: e2f73811084cb6f4821cbe6e53bb0d9407af52f5
[2/7] io_uring: introduce io_has_work
commit: cb0101159452eb63d0facf806e0a868963a11035
[3/7] io_uring: do not run task work at the start of io_uring_enter
commit: 415f872f5f3aedb62bf857c6e315a88e26c08aaa
[4/7] io_uring: add IORING_SETUP_DEFER_TASKRUN
commit: 2a23c7a52405c17183f36ef4aefb1fcf44425c2d
[5/7] io_uring: move io_eventfd_put
commit: 488a41100f4b9d7e10f05edead4e4d6d1c0c85a9
[6/7] io_uring: signal registered eventfd to process deferred task work
commit: 3403692af0c6d5ae826f536a2c08f4e7c0b8b163
[7/7] io_uring: trace local task work run
commit: 73100da58e5b1f6d8ba85e134463fb2aaca59978
Best regards,
--
Jens Axboe
