mirror of
https://github.com/torvalds/linux.git
synced 2026-05-02 05:22:49 -04:00
drm/amdkfd: allow users to target recommended SDMA engines
Certain GPUs have better copy performance over xGMI on specific SDMA engines depending on the source and destination GPU. Allow users to create SDMA queues on these recommended engines. Close to 2x overall performance has been observed with this optimization. Signed-off-by: Jonathan Kim <jonathan.kim@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This commit is contained in:
committed by
Alex Deucher
parent
60c30ba7ba
commit
e06b71b231
@@ -366,6 +366,7 @@ int pqm_create_queue(struct process_queue_manager *pqm,
|
||||
switch (type) {
|
||||
case KFD_QUEUE_TYPE_SDMA:
|
||||
case KFD_QUEUE_TYPE_SDMA_XGMI:
|
||||
case KFD_QUEUE_TYPE_SDMA_BY_ENG_ID:
|
||||
/* SDMA queues are always allocated statically no matter
|
||||
* which scheduler mode is used. We also do not need to
|
||||
* check whether a SDMA queue can be allocated here, because
|
||||
|
||||
Reference in New Issue
Block a user