hardenedbsd/HardenedBSD
1
0
Fork 0
mirror of https://git.hardenedbsd.org/hardenedbsd/HardenedBSD.git synced 2024-11-22 03:04:34 +01:00
Code Issues Actions Packages Projects Releases Wiki Activity

Schedule fast taskqueue callouts on right CPU.

Browse Source 
With fast taskqueues using direct callouts we can reduce number of
CPU wakeups by scheduling callout on current CPU if taskqueue calls
taskqueue_enqueue_timeout() on itself.  The trick won't work for
regular taskqueues, since the callout thread will occupy the CPU.
It also may not work in case of multiple threads since we do not
know which thread will pick the task, and we do not want excessive
callout migrations.  So we optimize only the other cases we can.

In practice this allows iichid(4) taskqueue to stay on CPU where
underlying ig4(4) interrupts are routed and to not kick CPU 0 with
timer interrupts on each sampling period (every 2nd/3rd sleep).

MFC after:	1 month
This commit is contained in:
Alexander Motin 2023-12-26 22:30:56 -05:00
parent a8b70cf260
commit 7bbac6419d
1 changed files with 8 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
sys/kern/subr_taskqueue.c
View File

@ -363,8 +363,14 @@ taskqueue_enqueue_timeout_sbt(struct taskqueue *queue,
if (sbt > 0) { if (sbt > 0) {
if (queue->tq_spin) if (queue->tq_spin)
flags |= C_DIRECT_EXEC; flags |= C_DIRECT_EXEC;
callout_reset_sbt(&timeout_task->c, sbt, pr, if (queue->tq_spin && queue->tq_tcount == 1 &&
taskqueue_timeout_func, timeout_task, flags); queue->tq_threads[0] == curthread) {
callout_reset_sbt_curcpu(&timeout_task->c, sbt, pr,
taskqueue_timeout_func, timeout_task, flags);
} else {
callout_reset_sbt(&timeout_task->c, sbt, pr,
taskqueue_timeout_func, timeout_task, flags);
}
} }
TQ_UNLOCK(queue); TQ_UNLOCK(queue);
} }