Merge tag 'drm-msm-next-2017-11-01' of git://people.freedesktop.org/~robclark/linux into drm-next

 + preemption support for a5xx[1][2]

 + display fixes for 8x96 (snapdragon 820) including fixes for 4k scanout
   (hwpipe assignment re-work to handle multiple hwpipe assigned to plane
   for wide scanout)

 + async cursor plane updates and fixes

 + refactor adreno_bind/hwinit.. still defer fw loading until device open,
   but move clk/irq/etc to probe/bind time to fix issues when fw isn't
   present in filesys

 + clk/dt bindings cleanups w/ backward compat via msm_clk_get() (dt docs
   part ack'ed by Rob Herring)

 + fw loading re-work with helper to handle either /lib/firmware/qcom/$fw
   or /lib/firmware/$fw.. background, we've started landing fw for some of
   generations in linux-firmware, but there is a preference to put fw files
   under 'qcom' subdirectory, which is not what was done on android or for
   people who copied fw from android.  So now we first look in qcom subdir
   and then fallback to the original location.

 + bunch of GPU debugging enhancements, to dump full cmdline of processes
   that trigger faults, and to add a new debugfs to capture cmdstream of
   just submits that triggered faults.. both quite useful for piglit ;-)

* tag 'drm-msm-next-2017-11-01' of git://people.freedesktop.org/~robclark/linux: (38 commits)
  drm/msm: use %z format modifier for printing size_t
  drm/msm/mdp5: Don't use async plane update path if plane visibility changes
  drm/msm/mdp5: mdp5_crtc: Restore cursor state only if LM cursors are enabled
  drm/msm/mdp5: Update mdp5_pipe_assign to spit out both planes
  drm/msm/mdp5: Prepare mdp5_pipe_assign for some rework
  drm/msm: remove mdp5_cursor_plane_funcs
  drm/msm: update cursors asynchronously through atomic
  drm/msm/atomic: switch to drm_atomic_helper_check
  drm/msm/mdp5: restore cursor state when enabling crtc
  drm/msm/mdp5: don't use autosuspend
  drm/msm/mdp5: ignore planes that are not visible
  drm/msm: dump submits which triggered gpu hang
  drm/msm: preserve IOVAs in submit's bo table
  drm/msm/rd: allow adding addition msg to top of dump
  drm/msm: split rd debugfs file
  drm/msm: add special _get_vaddr_active() for cmdstream dumps
  drm/msm: show task cmdline in gpu recovery messages
  drm/msm: dump a rd GPUADDR header for all buffers in the command
  drm/msm: Removed unused struct_mutex_task
  drm/msm: Implement preemption for A5XX targets
  ...

1 files changed, 43 insertions, 18 deletions

diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c
index 5d0a75d4b249..b8dc8f96caf2 100644
--- a/drivers/gpu/drm/msm/msm_gem_submit.c
+++ b/drivers/gpu/drm/msm/msm_gem_submit.c
@@ -31,7 +31,8 @@
 #define BO_PINNED   0x2000
 
 static struct msm_gem_submit *submit_create(struct drm_device *dev,
-		struct msm_gpu *gpu, uint32_t nr_bos, uint32_t nr_cmds)
+		struct msm_gpu *gpu, struct msm_gpu_submitqueue *queue,
+		uint32_t nr_bos, uint32_t nr_cmds)
 {
 	struct msm_gem_submit *submit;
 	uint64_t sz = sizeof(*submit) + ((u64)nr_bos * sizeof(submit->bos[0])) +
@@ -49,6 +50,8 @@ static struct msm_gem_submit *submit_create(struct drm_device *dev,
 	submit->fence = NULL;
 	submit->pid = get_pid(task_pid(current));
 	submit->cmd = (void *)&submit->bos[nr_bos];
+	submit->queue = queue;
+	submit->ring = gpu->rb[queue->prio];
 
 	/* initially, until copy_from_user() and bo lookup succeeds: */
 	submit->nr_bos = 0;
@@ -66,6 +69,8 @@ void msm_gem_submit_free(struct msm_gem_submit *submit)
 	dma_fence_put(submit->fence);
 	list_del(&submit->node);
 	put_pid(submit->pid);
+	msm_submitqueue_put(submit->queue);
+
 	kfree(submit);
 }
 
@@ -156,7 +161,8 @@ out:
 	return ret;
 }
 
-static void submit_unlock_unpin_bo(struct msm_gem_submit *submit, int i)
+static void submit_unlock_unpin_bo(struct msm_gem_submit *submit,
+		int i, bool backoff)
 {
 	struct msm_gem_object *msm_obj = submit->bos[i].obj;
 
@@ -166,7 +172,7 @@ static void submit_unlock_unpin_bo(struct msm_gem_submit *submit, int i)
 	if (submit->bos[i].flags & BO_LOCKED)
 		ww_mutex_unlock(&msm_obj->resv->lock);
 
-	if (!(submit->bos[i].flags & BO_VALID))
+	if (backoff && !(submit->bos[i].flags & BO_VALID))
 		submit->bos[i].iova = 0;
 
 	submit->bos[i].flags &= ~(BO_LOCKED | BO_PINNED);
@@ -201,10 +207,10 @@ retry:
 
 fail:
 	for (; i >= 0; i--)
-		submit_unlock_unpin_bo(submit, i);
+		submit_unlock_unpin_bo(submit, i, true);
 
 	if (slow_locked > 0)
-		submit_unlock_unpin_bo(submit, slow_locked);
+		submit_unlock_unpin_bo(submit, slow_locked, true);
 
 	if (ret == -EDEADLK) {
 		struct msm_gem_object *msm_obj = submit->bos[contended].obj;
@@ -221,7 +227,7 @@ fail:
 	return ret;
 }
 
-static int submit_fence_sync(struct msm_gem_submit *submit)
+static int submit_fence_sync(struct msm_gem_submit *submit, bool no_implicit)
 {
 	int i, ret = 0;
 
@@ -229,7 +235,22 @@ static int submit_fence_sync(struct msm_gem_submit *submit)
 		struct msm_gem_object *msm_obj = submit->bos[i].obj;
 		bool write = submit->bos[i].flags & MSM_SUBMIT_BO_WRITE;
 
-		ret = msm_gem_sync_object(&msm_obj->base, submit->gpu->fctx, write);
+		if (!write) {
+			/* NOTE: _reserve_shared() must happen before
+			 * _add_shared_fence(), which makes this a slightly
+			 * strange place to call it.  OTOH this is a
+			 * convenient can-fail point to hook it in.
+			 */
+			ret = reservation_object_reserve_shared(msm_obj->resv);
+			if (ret)
+				return ret;
+		}
+
+		if (no_implicit)
+			continue;
+
+		ret = msm_gem_sync_object(&msm_obj->base, submit->ring->fctx,
+			write);
 		if (ret)
 			break;
 	}
@@ -373,7 +394,7 @@ static void submit_cleanup(struct msm_gem_submit *submit)
 
 	for (i = 0; i < submit->nr_bos; i++) {
 		struct msm_gem_object *msm_obj = submit->bos[i].obj;
-		submit_unlock_unpin_bo(submit, i);
+		submit_unlock_unpin_bo(submit, i, false);
 		list_del_init(&msm_obj->submit_entry);
 		drm_gem_object_unreference(&msm_obj->base);
 	}
@@ -391,6 +412,8 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	struct msm_gpu *gpu = priv->gpu;
 	struct dma_fence *in_fence = NULL;
 	struct sync_file *sync_file = NULL;
+	struct msm_gpu_submitqueue *queue;
+	struct msm_ringbuffer *ring;
 	int out_fence_fd = -1;
 	unsigned i;
 	int ret;
@@ -407,6 +430,12 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	if (MSM_PIPE_FLAGS(args->flags) & ~MSM_SUBMIT_FLAGS)
 		return -EINVAL;
 
+	queue = msm_submitqueue_get(ctx, args->queueid);
+	if (!queue)
+		return -ENOENT;
+
+	ring = gpu->rb[queue->prio];
+
 	if (args->flags & MSM_SUBMIT_FENCE_FD_IN) {
 		in_fence = sync_file_get_fence(args->fence_fd);
 
@@ -417,7 +446,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 		 * Wait if the fence is from a foreign context, or if the fence
 		 * array contains any fence from a foreign context.
 		 */
-		if (!dma_fence_match_context(in_fence, gpu->fctx->context)) {
+		if (!dma_fence_match_context(in_fence, ring->fctx->context)) {
 			ret = dma_fence_wait(in_fence, true);
 			if (ret)
 				return ret;
@@ -435,9 +464,8 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 			goto out_unlock;
 		}
 	}
-	priv->struct_mutex_task = current;
 
-	submit = submit_create(dev, gpu, args->nr_bos, args->nr_cmds);
+	submit = submit_create(dev, gpu, queue, args->nr_bos, args->nr_cmds);
 	if (!submit) {
 		ret = -ENOMEM;
 		goto out_unlock;
@@ -451,11 +479,9 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	if (ret)
 		goto out;
 
-	if (!(args->flags & MSM_SUBMIT_NO_IMPLICIT)) {
-		ret = submit_fence_sync(submit);
-		if (ret)
-			goto out;
-	}
+	ret = submit_fence_sync(submit, !!(args->flags & MSM_SUBMIT_NO_IMPLICIT));
+	if (ret)
+		goto out;
 
 	ret = submit_pin_objects(submit);
 	if (ret)
@@ -522,7 +548,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 
 	submit->nr_cmds = i;
 
-	submit->fence = msm_fence_alloc(gpu->fctx);
+	submit->fence = msm_fence_alloc(ring->fctx);
 	if (IS_ERR(submit->fence)) {
 		ret = PTR_ERR(submit->fence);
 		submit->fence = NULL;
@@ -555,7 +581,6 @@ out:
 out_unlock:
 	if (ret && (out_fence_fd >= 0))
 		put_unused_fd(out_fence_fd);
-	priv->struct_mutex_task = NULL;
 	mutex_unlock(&dev->struct_mutex);
 	return ret;
 }