[Intel-xe] [PATCH 4/5] drm/xe: Allow num_batch_buffer == 0 in exec IOCTL

[Matthew Brost]

The idea being out-syncs can signal indicating all previous operations
on the exec queue are complete. An example use case of this would be
support for implementing vkQueueWaitForIdle easily.

Signed-off-by: Matthew Brost <matthew.brost at intel.com>
---
 drivers/gpu/drm/xe/xe_engine.c       |  5 ++++-
 drivers/gpu/drm/xe/xe_engine_types.h |  2 +-
 drivers/gpu/drm/xe/xe_exec.c         | 21 ++++++++++++++++++---
 drivers/gpu/drm/xe/xe_sync.c         |  5 ++++-
 drivers/gpu/drm/xe/xe_sync.h         |  2 +-
 drivers/gpu/drm/xe/xe_vm.c           |  2 +-
 6 files changed, 29 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_engine.c b/drivers/gpu/drm/xe/xe_engine.c
index bef82b40a097..26cc058ca734 100644
--- a/drivers/gpu/drm/xe/xe_engine.c
+++ b/drivers/gpu/drm/xe/xe_engine.c
@@ -839,7 +839,10 @@ int xe_engine_set_property_ioctl(struct drm_device *dev, void *data,
 static void xe_engine_last_fence_lockdep_assert(struct xe_engine *e,
 						struct xe_vm *vm)
 {
-	lockdep_assert_held_write(&vm->lock);
+	if (e->flags & ENGINE_FLAG_VM)
+		lockdep_assert_held_write(&vm->lock);
+	else
+		xe_vm_assert_held(vm);
 }
 
 /**
diff --git a/drivers/gpu/drm/xe/xe_engine_types.h b/drivers/gpu/drm/xe/xe_engine_types.h
index 9ca19942d72b..8e324be80792 100644
--- a/drivers/gpu/drm/xe/xe_engine_types.h
+++ b/drivers/gpu/drm/xe/xe_engine_types.h
@@ -54,7 +54,7 @@ struct xe_engine {
 
 	/**
 	 * @last_fence: last fence on engine, protected by vm->lock in write
-	 * mode if bind engine
+	 * mode if bind engine, protected by dma resv lock if exec engine
 	 */
 	struct dma_fence *last_fence;
 
diff --git a/drivers/gpu/drm/xe/xe_exec.c b/drivers/gpu/drm/xe/xe_exec.c
index 58af1e0ef4d1..fae99c2e81e3 100644
--- a/drivers/gpu/drm/xe/xe_exec.c
+++ b/drivers/gpu/drm/xe/xe_exec.c
@@ -196,7 +196,8 @@ int xe_exec_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	if (XE_IOCTL_DBG(xe, engine->flags & ENGINE_FLAG_VM))
 		return -EINVAL;
 
-	if (XE_IOCTL_DBG(xe, engine->width != args->num_batch_buffer))
+	if (XE_IOCTL_DBG(xe, args->num_batch_buffer &&
+			 engine->width != args->num_batch_buffer))
 		return -EINVAL;
 
 	if (XE_IOCTL_DBG(xe, engine->flags & ENGINE_FLAG_BANNED)) {
@@ -217,12 +218,13 @@ int xe_exec_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	for (i = 0; i < args->num_syncs; i++) {
 		err = xe_sync_entry_parse(xe, xef, &syncs[num_syncs++],
 					  &syncs_user[i], true,
-					  xe_vm_no_dma_fences(vm));
+					  xe_vm_no_dma_fences(vm),
+					  !args->num_batch_buffer);
 		if (err)
 			goto err_syncs;
 	}
 
-	if (xe_engine_is_parallel(engine)) {
+	if (args->num_batch_buffer && xe_engine_is_parallel(engine)) {
 		err = __copy_from_user(addresses, addresses_user, sizeof(u64) *
 				       engine->width);
 		if (err) {
@@ -261,6 +263,17 @@ int xe_exec_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 		goto err_engine_end;
 	}
 
+	if (!args->num_batch_buffer) {
+		for (i = 0; i < num_syncs; i++) {
+			struct dma_fence *fence =
+				xe_engine_last_fence_get(engine, engine->vm);
+
+			xe_sync_entry_signal(&syncs[i], NULL, fence);
+		}
+
+		goto err_engine_end;
+	}
+
 	if (xe_engine_is_lr(engine) && xe_engine_ring_full(engine)) {
 		err = -EWOULDBLOCK;
 		goto err_engine_end;
@@ -354,6 +367,8 @@ int xe_exec_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 
 	if (xe_engine_is_lr(engine))
 		engine->ring_ops->emit_job(job);
+	xe_engine_last_fence_set(engine, engine->vm,
+				 &job->drm.s_fence->finished);
 	xe_sched_job_push(job);
 	xe_vm_reactivate_rebind(vm);
 
diff --git a/drivers/gpu/drm/xe/xe_sync.c b/drivers/gpu/drm/xe/xe_sync.c
index 73ef259aa387..2461e7d4814c 100644
--- a/drivers/gpu/drm/xe/xe_sync.c
+++ b/drivers/gpu/drm/xe/xe_sync.c
@@ -100,7 +100,7 @@ static void user_fence_cb(struct dma_fence *fence, struct dma_fence_cb *cb)
 int xe_sync_entry_parse(struct xe_device *xe, struct xe_file *xef,
 			struct xe_sync_entry *sync,
 			struct drm_xe_sync __user *sync_user,
-			bool exec, bool no_dma_fences)
+			bool exec, bool no_dma_fences, bool exec_nop)
 {
 	struct drm_xe_sync sync_in;
 	int err;
@@ -171,6 +171,9 @@ int xe_sync_entry_parse(struct xe_device *xe, struct xe_file *xef,
 		break;
 
 	case DRM_XE_SYNC_USER_FENCE:
+		if (XE_IOCTL_DBG(xe, exec_nop))
+			return -EOPNOTSUPP;
+
 		if (XE_IOCTL_DBG(xe, !signal))
 			return -EOPNOTSUPP;
 
diff --git a/drivers/gpu/drm/xe/xe_sync.h b/drivers/gpu/drm/xe/xe_sync.h
index 30958ddc4cdc..98f02bb34637 100644
--- a/drivers/gpu/drm/xe/xe_sync.h
+++ b/drivers/gpu/drm/xe/xe_sync.h
@@ -15,7 +15,7 @@ struct xe_sched_job;
 int xe_sync_entry_parse(struct xe_device *xe, struct xe_file *xef,
 			struct xe_sync_entry *sync,
 			struct drm_xe_sync __user *sync_user,
-			bool exec, bool compute_mode);
+			bool exec, bool compute_mode, bool exec_nop);
 int xe_sync_entry_wait(struct xe_sync_entry *sync);
 int xe_sync_entry_add_deps(struct xe_sync_entry *sync,
 			   struct xe_sched_job *job);
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 874ef9afbecf..879378a5ca8b 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -2994,7 +2994,7 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	for (num_syncs = 0; num_syncs < args->num_syncs; num_syncs++) {
 		err = xe_sync_entry_parse(xe, xef, &syncs[num_syncs],
 					  &syncs_user[num_syncs], false,
-					  xe_vm_no_dma_fences(vm));
+					  xe_vm_no_dma_fences(vm), false);
 		if (err)
 			goto free_syncs;
 	}
-- 
2.34.1