On Thu, May 04, 2023 at 09:24:27AM -0700, Keith Busch wrote:
> From: Keith Busch <kbusch@xxxxxxxxxx>
>
> io_uring tries to optimize allocating tags by hinting to the plug how
> many it expects to need for a batch instead of allocating each tag
> individually. But io_uring submission queueus may have a mix of many
> devices for io, so the number of io's counted may be overestimated. This
> can lead to allocating too many tags, which adds overhead to finding
> that many contiguous tags, freeing up the ones we didn't use, and may
> starve out other users that can actually use them.
When running batched IO to multiple nvme drives, like with t/io_uring,
this shows a tiny improvement in CPU utilization from avoiding the
unlikely clean up condition in __blk_flush_plug() shown below:
if (unlikely(!rq_list_empty(plug->cached_rq)))
blk_mq_free_plug_rqs(plug);
