drm/amdkfd: Improve concurrency of event handling

Browse source

Use rcu_read_lock to read p->event_idr concurrently with other readers
and writers. Use p->event_mutex only for creating and destroying events
and in kfd_wait_on_events.

Protect the contents of the kfd_event structure with a per-event
spinlock that can be taken inside the rcu_read_lock critical section.

This eliminates contention of p->event_mutex in set_event, which tends
to be on the critical path for dispatch latency even when busy waiting
is used. It also eliminates lock contention in event interrupt handlers.
Since the p->event_mutex is now used much less, the impact of requiring
it in kfd_wait_on_events should also be much smaller.

This should improve event handling latency for processes using multiple
GPUs concurrently.

v2: Reschedule the worker periodically to avoid soft lockup warnings

Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Sean Keely <Sean.Keely@amd.com> # v1
Tested-by: Sanjay Tripathi <sanjay.tripathi@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

...

This commit is contained in:

Felix Kuehling 2022-03-01 20:40:45 -05:00 • committed by Alex Deucher

parent 8d2aad983d

commit 5273e82c5f

3 changed files with 88 additions and 43 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download patch file Download diff file Expand all files Collapse all files

119

drivers/gpu/drm/amd/amdkfd/kfd_events.c

View file

1

drivers/gpu/drm/amd/amdkfd/kfd_events.h

View file

11

drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c

View file