diff options
| author |   Nirmoy Das <nirmoy.aiemd@gmail.com> | 2020-06-25 14:07:23 +0200 |
|---|---|---|
| committer |   Christian König <christian.koenig@amd.com> | 2020-06-26 14:16:29 +0200 |
| commit | d41a39dda1407ff50c7c4bfbd307c188f1ae364b (patch) | |
| tree | f668464bee4abac9f7f2f5ff445b80ad499acda0 /include/drm/gpu_scheduler.h | |
| parent | drm/nouveau: don't use ttm bo->offset v3 (diff) | |
| download | linux-dev-d41a39dda1407ff50c7c4bfbd307c188f1ae364b.tar.xz linux-dev-d41a39dda1407ff50c7c4bfbd307c188f1ae364b.zip | |
drm/scheduler: improve job distribution with multiple queues
This patch uses score to select a new drm scheduler for better
loadbalance between multiple drm schedulers instead of num_jobs.
Below are test results after running amdgpu_test for ~10 times.
Before this patch:
sched_name num of many times it got schedule
========= ==================================
sdma0 1463
sdma1 198
comp_1.0.1 280
After this patch:
sched_name num of many times it got schedule
========= ==================================
sdma0 925
sdma1 928
comp_1.0.1 177
comp_1.1.1 44
comp_1.2.1 43
comp_1.3.1 44
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/373000/
Signed-off-by: Christian König <christian.koenig@amd.com>
Diffstat (limited to 'include/drm/gpu_scheduler.h')
| -rw-r--r-- | include/drm/gpu_scheduler.h | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/include/drm/gpu_scheduler.h b/include/drm/gpu_scheduler.h index a21b3b92135a..b9780ae9dd26 100644 --- a/include/drm/gpu_scheduler.h +++ b/include/drm/gpu_scheduler.h @@ -263,7 +263,7 @@ struct drm_sched_backend_ops { * @job_list_lock: lock to protect the ring_mirror_list. * @hang_limit: once the hangs by a job crosses this limit then it is marked * guilty and it will be considered for scheduling further. - * @num_jobs: the number of jobs in queue in the scheduler + * @score: score to help loadbalancer pick a idle sched * @ready: marks if the underlying HW is ready to work * @free_guilty: A hit to time out handler to free the guilty job. * @@ -284,8 +284,8 @@ struct drm_gpu_scheduler { struct list_head ring_mirror_list; spinlock_t job_list_lock; int hang_limit; - atomic_t num_jobs; - bool ready; + atomic_t score; + bool ready; bool free_guilty; }; |