Revision 78076bb64aa8ba5b7207c38b2660a9e10ffa8cc7 authored by Jens Axboe on 05 December 2019, 02:56:40 UTC, committed by Jens Axboe on 05 December 2019, 03:12:58 UTC
We recently changed this from a single list to an rbtree, but for some real life workloads, the rbtree slows down the submission/insertion case enough so that it's the top cycle consumer on the io_uring side. In testing, using a hash table is a more well rounded compromise. It is fast for insertion, and as long as it's sized appropriately, it works well for the cancellation case as well. Running TAO with a lot of network sockets, this removes io_poll_req_insert() from spending 2% of the CPU cycles. Reported-by: Dan Melnic <dmm@fb.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
1 parent 08bdcc3