1 files changed, 172 insertions, 40 deletions

diff --git a/drivers/gpu/drm/v3d/v3d_gem.c b/drivers/gpu/drm/v3d/v3d_gem.c
index 70c54774400b..05ca6319065e 100644
--- a/drivers/gpu/drm/v3d/v3d_gem.c
+++ b/drivers/gpu/drm/v3d/v3d_gem.c
@@ -207,32 +207,26 @@ v3d_flush_caches(struct v3d_dev *v3d)
 }
 
 static void
-v3d_attach_object_fences(struct v3d_exec_info *exec)
+v3d_attach_object_fences(struct v3d_bo **bos, int bo_count,
+			 struct dma_fence *fence)
 {
-	struct dma_fence *out_fence = &exec->render.base.s_fence->finished;
-	struct v3d_bo *bo;
 	int i;
 
-	for (i = 0; i < exec->bo_count; i++) {
-		bo = to_v3d_bo(&exec->bo[i]->base);
-
+	for (i = 0; i < bo_count; i++) {
 		/* XXX: Use shared fences for read-only objects. */
-		reservation_object_add_excl_fence(bo->resv, out_fence);
+		reservation_object_add_excl_fence(bos[i]->resv, fence);
 	}
 }
 
 static void
-v3d_unlock_bo_reservations(struct drm_device *dev,
-			   struct v3d_exec_info *exec,
+v3d_unlock_bo_reservations(struct v3d_bo **bos,
+			   int bo_count,
 			   struct ww_acquire_ctx *acquire_ctx)
 {
 	int i;
 
-	for (i = 0; i < exec->bo_count; i++) {
-		struct v3d_bo *bo = to_v3d_bo(&exec->bo[i]->base);
-
-		ww_mutex_unlock(&bo->resv->lock);
-	}
+	for (i = 0; i < bo_count; i++)
+		ww_mutex_unlock(&bos[i]->resv->lock);
 
 	ww_acquire_fini(acquire_ctx);
 }
@@ -245,19 +239,19 @@ v3d_unlock_bo_reservations(struct drm_device *dev,
  * to v3d, so we don't attach dma-buf fences to them.
  */
 static int
-v3d_lock_bo_reservations(struct drm_device *dev,
-			 struct v3d_exec_info *exec,
+v3d_lock_bo_reservations(struct v3d_bo **bos,
+			 int bo_count,
 			 struct ww_acquire_ctx *acquire_ctx)
 {
 	int contended_lock = -1;
 	int i, ret;
-	struct v3d_bo *bo;
 
 	ww_acquire_init(acquire_ctx, &reservation_ww_class);
 
 retry:
 	if (contended_lock != -1) {
-		bo = to_v3d_bo(&exec->bo[contended_lock]->base);
+		struct v3d_bo *bo = bos[contended_lock];
+
 		ret = ww_mutex_lock_slow_interruptible(&bo->resv->lock,
 						       acquire_ctx);
 		if (ret) {
@@ -266,23 +260,20 @@ retry:
 		}
 	}
 
-	for (i = 0; i < exec->bo_count; i++) {
+	for (i = 0; i < bo_count; i++) {
 		if (i == contended_lock)
 			continue;
 
-		bo = to_v3d_bo(&exec->bo[i]->base);
-
-		ret = ww_mutex_lock_interruptible(&bo->resv->lock, acquire_ctx);
+		ret = ww_mutex_lock_interruptible(&bos[i]->resv->lock,
+						  acquire_ctx);
 		if (ret) {
 			int j;
 
-			for (j = 0; j < i; j++) {
-				bo = to_v3d_bo(&exec->bo[j]->base);
-				ww_mutex_unlock(&bo->resv->lock);
-			}
+			for (j = 0; j < i; j++)
+				ww_mutex_unlock(&bos[j]->resv->lock);
 
 			if (contended_lock != -1 && contended_lock >= i) {
-				bo = to_v3d_bo(&exec->bo[contended_lock]->base);
+				struct v3d_bo *bo = bos[contended_lock];
 
 				ww_mutex_unlock(&bo->resv->lock);
 			}
@@ -302,12 +293,11 @@ retry:
 	/* Reserve space for our shared (read-only) fence references,
 	 * before we commit the CL to the hardware.
 	 */
-	for (i = 0; i < exec->bo_count; i++) {
-		bo = to_v3d_bo(&exec->bo[i]->base);
-
-		ret = reservation_object_reserve_shared(bo->resv);
+	for (i = 0; i < bo_count; i++) {
+		ret = reservation_object_reserve_shared(bos[i]->resv, 1);
 		if (ret) {
-			v3d_unlock_bo_reservations(dev, exec, acquire_ctx);
+			v3d_unlock_bo_reservations(bos, bo_count,
+						   acquire_ctx);
 			return ret;
 		}
 	}
@@ -409,6 +399,7 @@ v3d_exec_cleanup(struct kref *ref)
 	dma_fence_put(exec->render.done_fence);
 
 	dma_fence_put(exec->bin_done_fence);
+	dma_fence_put(exec->render_done_fence);
 
 	for (i = 0; i < exec->bo_count; i++)
 		drm_gem_object_put_unlocked(&exec->bo[i]->base);
@@ -429,6 +420,33 @@ void v3d_exec_put(struct v3d_exec_info *exec)
 	kref_put(&exec->refcount, v3d_exec_cleanup);
 }
 
+static void
+v3d_tfu_job_cleanup(struct kref *ref)
+{
+	struct v3d_tfu_job *job = container_of(ref, struct v3d_tfu_job,
+					       refcount);
+	struct v3d_dev *v3d = job->v3d;
+	unsigned int i;
+
+	dma_fence_put(job->in_fence);
+	dma_fence_put(job->done_fence);
+
+	for (i = 0; i < ARRAY_SIZE(job->bo); i++) {
+		if (job->bo[i])
+			drm_gem_object_put_unlocked(&job->bo[i]->base);
+	}
+
+	pm_runtime_mark_last_busy(v3d->dev);
+	pm_runtime_put_autosuspend(v3d->dev);
+
+	kfree(job);
+}
+
+void v3d_tfu_job_put(struct v3d_tfu_job *job)
+{
+	kref_put(&job->refcount, v3d_tfu_job_cleanup);
+}
+
 int
 v3d_wait_bo_ioctl(struct drm_device *dev, void *data,
 		  struct drm_file *file_priv)
@@ -503,6 +521,8 @@ v3d_submit_cl_ioctl(struct drm_device *dev, void *data,
 	struct drm_syncobj *sync_out;
 	int ret = 0;
 
+	trace_v3d_submit_cl_ioctl(&v3d->drm, args->rcl_start, args->rcl_end);
+
 	if (args->pad != 0) {
 		DRM_INFO("pad must be zero: %d\n", args->pad);
 		return -EINVAL;
@@ -521,12 +541,12 @@ v3d_submit_cl_ioctl(struct drm_device *dev, void *data,
 	kref_init(&exec->refcount);
 
 	ret = drm_syncobj_find_fence(file_priv, args->in_sync_bcl,
-				     0, &exec->bin.in_fence);
+				     0, 0, &exec->bin.in_fence);
 	if (ret == -EINVAL)
 		goto fail;
 
 	ret = drm_syncobj_find_fence(file_priv, args->in_sync_rcl,
-				     0, &exec->render.in_fence);
+				     0, 0, &exec->render.in_fence);
 	if (ret == -EINVAL)
 		goto fail;
 
@@ -546,7 +566,8 @@ v3d_submit_cl_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		goto fail;
 
-	ret = v3d_lock_bo_reservations(dev, exec, &acquire_ctx);
+	ret = v3d_lock_bo_reservations(exec->bo, exec->bo_count,
+				       &acquire_ctx);
 	if (ret)
 		goto fail;
 
@@ -572,20 +593,23 @@ v3d_submit_cl_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		goto fail_unreserve;
 
+	exec->render_done_fence =
+		dma_fence_get(&exec->render.base.s_fence->finished);
+
 	kref_get(&exec->refcount); /* put by scheduler job completion */
 	drm_sched_entity_push_job(&exec->render.base,
 				  &v3d_priv->sched_entity[V3D_RENDER]);
 	mutex_unlock(&v3d->sched_lock);
 
-	v3d_attach_object_fences(exec);
+	v3d_attach_object_fences(exec->bo, exec->bo_count,
+				 exec->render_done_fence);
 
-	v3d_unlock_bo_reservations(dev, exec, &acquire_ctx);
+	v3d_unlock_bo_reservations(exec->bo, exec->bo_count, &acquire_ctx);
 
 	/* Update the return sync object for the */
 	sync_out = drm_syncobj_find(file_priv, args->out_sync);
 	if (sync_out) {
-		drm_syncobj_replace_fence(sync_out, 0,
-					  &exec->render.base.s_fence->finished);
+		drm_syncobj_replace_fence(sync_out, exec->render_done_fence);
 		drm_syncobj_put(sync_out);
 	}
 
@@ -595,13 +619,121 @@ v3d_submit_cl_ioctl(struct drm_device *dev, void *data,
 
 fail_unreserve:
 	mutex_unlock(&v3d->sched_lock);
-	v3d_unlock_bo_reservations(dev, exec, &acquire_ctx);
+	v3d_unlock_bo_reservations(exec->bo, exec->bo_count, &acquire_ctx);
 fail:
 	v3d_exec_put(exec);
 
 	return ret;
 }
 
+/**
+ * v3d_submit_tfu_ioctl() - Submits a TFU (texture formatting) job to the V3D.
+ * @dev: DRM device
+ * @data: ioctl argument
+ * @file_priv: DRM file for this fd
+ *
+ * Userspace provides the register setup for the TFU, which we don't
+ * need to validate since the TFU is behind the MMU.
+ */
+int
+v3d_submit_tfu_ioctl(struct drm_device *dev, void *data,
+		     struct drm_file *file_priv)
+{
+	struct v3d_dev *v3d = to_v3d_dev(dev);
+	struct v3d_file_priv *v3d_priv = file_priv->driver_priv;
+	struct drm_v3d_submit_tfu *args = data;
+	struct v3d_tfu_job *job;
+	struct ww_acquire_ctx acquire_ctx;
+	struct drm_syncobj *sync_out;
+	struct dma_fence *sched_done_fence;
+	int ret = 0;
+	int bo_count;
+
+	trace_v3d_submit_tfu_ioctl(&v3d->drm, args->iia);
+
+	job = kcalloc(1, sizeof(*job), GFP_KERNEL);
+	if (!job)
+		return -ENOMEM;
+
+	ret = pm_runtime_get_sync(v3d->dev);
+	if (ret < 0) {
+		kfree(job);
+		return ret;
+	}
+
+	kref_init(&job->refcount);
+
+	ret = drm_syncobj_find_fence(file_priv, args->in_sync,
+				     0, 0, &job->in_fence);
+	if (ret == -EINVAL)
+		goto fail;
+
+	job->args = *args;
+	job->v3d = v3d;
+
+	spin_lock(&file_priv->table_lock);
+	for (bo_count = 0; bo_count < ARRAY_SIZE(job->bo); bo_count++) {
+		struct drm_gem_object *bo;
+
+		if (!args->bo_handles[bo_count])
+			break;
+
+		bo = idr_find(&file_priv->object_idr,
+			      args->bo_handles[bo_count]);
+		if (!bo) {
+			DRM_DEBUG("Failed to look up GEM BO %d: %d\n",
+				  bo_count, args->bo_handles[bo_count]);
+			ret = -ENOENT;
+			spin_unlock(&file_priv->table_lock);
+			goto fail;
+		}
+		drm_gem_object_get(bo);
+		job->bo[bo_count] = to_v3d_bo(bo);
+	}
+	spin_unlock(&file_priv->table_lock);
+
+	ret = v3d_lock_bo_reservations(job->bo, bo_count, &acquire_ctx);
+	if (ret)
+		goto fail;
+
+	mutex_lock(&v3d->sched_lock);
+	ret = drm_sched_job_init(&job->base,
+				 &v3d_priv->sched_entity[V3D_TFU],
+				 v3d_priv);
+	if (ret)
+		goto fail_unreserve;
+
+	sched_done_fence = dma_fence_get(&job->base.s_fence->finished);
+
+	kref_get(&job->refcount); /* put by scheduler job completion */
+	drm_sched_entity_push_job(&job->base, &v3d_priv->sched_entity[V3D_TFU]);
+	mutex_unlock(&v3d->sched_lock);
+
+	v3d_attach_object_fences(job->bo, bo_count, sched_done_fence);
+
+	v3d_unlock_bo_reservations(job->bo, bo_count, &acquire_ctx);
+
+	/* Update the return sync object */
+	sync_out = drm_syncobj_find(file_priv, args->out_sync);
+	if (sync_out) {
+		drm_syncobj_replace_fence(sync_out, sched_done_fence);
+		drm_syncobj_put(sync_out);
+	}
+	dma_fence_put(sched_done_fence);
+
+	v3d_tfu_job_put(job);
+
+	return 0;
+
+fail_unreserve:
+	mutex_unlock(&v3d->sched_lock);
+	v3d_unlock_bo_reservations(job->bo, bo_count, &acquire_ctx);
+fail:
+	v3d_tfu_job_put(job);
+
+	return ret;
+}
+
 int
 v3d_gem_init(struct drm_device *dev)
 {