[Intel-xe] [PATCH] drm/xe: Fix exec queue usage for unbinds

Niranjana Vishwanathapura niranjana.vishwanathapura at intel.com
Fri Sep 29 22:05:06 UTC 2023


On Fri, Sep 29, 2023 at 01:06:18PM -0700, Matthew Brost wrote:
>Passing in a NULL exec queue to __xe_pt_unbind_vma results in the
>migrate exec queue being used. This is not the intent from the VM bind
>IOCTL, rather a NULL exec queue should use default VM exec queue.
>
>Signed-off-by: Matthew Brost <matthew.brost at intel.com>
>---
> drivers/gpu/drm/xe/xe_vm.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
>diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
>index 57ffac324564..26b9f2310c5e 100644
>--- a/drivers/gpu/drm/xe/xe_vm.c
>+++ b/drivers/gpu/drm/xe/xe_vm.c
>@@ -1732,7 +1732,8 @@ xe_vm_unbind_vma(struct xe_vma *vma, struct xe_exec_queue *q,
> 		if (!(vma->tile_present & BIT(id)))
> 			goto next;
>
>-		fence = __xe_pt_unbind_vma(tile, vma, q, first_op ? syncs : NULL,
>+		fence = __xe_pt_unbind_vma(tile, vma, q ? q : vm->q[id],
>+					   first_op ? syncs : NULL,
> 					   first_op ? num_syncs : 0);

Reviewed-by: Niranjana Vishwanathapura <niranjana.vishwanathapura at intel.com>

> 		if (IS_ERR(fence)) {
> 			err = PTR_ERR(fence);
>-- 
>2.34.1
>


More information about the Intel-xe mailing list