[Intel-xe] [PATCH v2 3/4] drm/xe: Rename exec_queue_kill_compute to xe_vm_remove_compute_exec_queue

[Niranjana Vishwanathapura]

On Fri, Sep 15, 2023 at 04:34:44PM -0700, Matthew Brost wrote:
>Much better name and aligns with xe_vm_add_compute_exec_queue. As part
>of the rename, move the implementation from xe_exec_queue.c to xe_vm.c.
>
>Suggested-by: Niranjana Vishwanathapura <niranjana.vishwanathapura at intel.com>
>Signed-off-by: Matthew Brost <matthew.brost at intel.com>
>---
> drivers/gpu/drm/xe/xe_exec_queue.c | 20 ++------------------
> drivers/gpu/drm/xe/xe_vm.c         | 21 +++++++++++++++++++++
> drivers/gpu/drm/xe/xe_vm.h         |  1 +
> 3 files changed, 24 insertions(+), 18 deletions(-)
>
>diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
>index 93d2f0591c34..b95b62a2538f 100644
>--- a/drivers/gpu/drm/xe/xe_exec_queue.c
>+++ b/drivers/gpu/drm/xe/xe_exec_queue.c
>@@ -775,22 +775,6 @@ int xe_exec_queue_get_property_ioctl(struct drm_device *dev, void *data,
> 	return ret;
> }
>
>-static void exec_queue_kill_compute(struct xe_exec_queue *q)
>-{
>-	if (!xe_vm_in_compute_mode(q->vm))
>-		return;
>-
>-	down_write(&q->vm->lock);
>-	list_del(&q->compute.link);
>-	--q->vm->preempt.num_exec_queues;
>-	if (q->compute.pfence) {
>-		dma_fence_enable_sw_signaling(q->compute.pfence);
>-		dma_fence_put(q->compute.pfence);
>-		q->compute.pfence = NULL;
>-	}
>-	up_write(&q->vm->lock);
>-}
>-
> /**
>  * xe_exec_queue_is_lr() - Whether an exec_queue is long-running
>  * @q: The exec_queue
>@@ -861,11 +845,11 @@ void xe_exec_queue_kill(struct xe_exec_queue *q)
> 	list_for_each_entry_safe(eq, next, &eq->multi_gt_list,
> 				 multi_gt_link) {
> 		q->ops->kill(eq);
>-		exec_queue_kill_compute(eq);
>+		xe_vm_remove_compute_exec_queue(q->vm, eq);
> 	}
>
> 	q->ops->kill(q);
>-	exec_queue_kill_compute(q);
>+	xe_vm_remove_compute_exec_queue(q->vm, q);
> }
>
> int xe_exec_queue_destroy_ioctl(struct drm_device *dev, void *data,
>diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
>index 03ee6993e7c7..433bdfae6759 100644
>--- a/drivers/gpu/drm/xe/xe_vm.c
>+++ b/drivers/gpu/drm/xe/xe_vm.c
>@@ -380,6 +380,27 @@ int xe_vm_add_compute_exec_queue(struct xe_vm *vm, struct xe_exec_queue *q)
> 	return err;
> }
>
>+/**
>+ * xe_vm_remove_compute_exec_queue() - Remove compute exec queue from VM
>+ * @vm: The VM.
>+ * @q: The exec_queue
>+ */
>+void xe_vm_remove_compute_exec_queue(struct xe_vm *vm, struct xe_exec_queue *q)

Actually wonder if we need 'vm' as input param here as we can always use q->vm.
If we do keep it, it will force use to assert 'vm == q->vm' here, so probably
better remove it.
In either case,
Reviewed-by: Niranjana Vishwanathapura <niranjana.vishwanathapura at intel.com>

Niranjana

>+{
>+	if (!xe_vm_in_compute_mode(vm))
>+		return;
>+
>+	down_write(&vm->lock);
>+	list_del(&q->compute.link);
>+	--vm->preempt.num_exec_queues;
>+	if (q->compute.pfence) {
>+		dma_fence_enable_sw_signaling(q->compute.pfence);
>+		dma_fence_put(q->compute.pfence);
>+		q->compute.pfence = NULL;
>+	}
>+	up_write(&vm->lock);
>+}
>+
> /**
>  * __xe_vm_userptr_needs_repin() - Check whether the VM does have userptrs
>  * that need repinning.
>diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h
>index 5608e4e33169..813328ad3f7f 100644
>--- a/drivers/gpu/drm/xe/xe_vm.h
>+++ b/drivers/gpu/drm/xe/xe_vm.h
>@@ -168,6 +168,7 @@ static inline bool xe_vm_no_dma_fences(struct xe_vm *vm)
> }
>
> int xe_vm_add_compute_exec_queue(struct xe_vm *vm, struct xe_exec_queue *q);
>+void xe_vm_remove_compute_exec_queue(struct xe_vm *vm, struct xe_exec_queue *q);
>
> int xe_vm_userptr_pin(struct xe_vm *vm);
>
>-- 
>2.34.1
>