[04/39] drm/i915: Prelude to splitting i915_gem_do_execbuffer in two

Commit Message

John Harrison Nov. 23, 2015, 11:38 a.m. UTC

From: John Harrison <[e-mail]>

The scheduler decouples the submission of batch buffers to the driver with their
submission to the hardware. This basically means splitting the execbuffer()
function in half. This change rearranges some code ready for the split to occur.

Change-Id: Icc9c8afaac18821f3eb8a151a49f918f90c068a3
For: VIZ-1587
Signed-off-by: John Harrison <[e-mail]>
---
 drivers/gpu/drm/i915/i915_gem_execbuffer.c | 64 ++++++++++++++++++------------
 drivers/gpu/drm/i915/intel_lrc.c           | 18 ++++++---
 2 files changed, 51 insertions(+), 31 deletions(-)

Patch

diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 5f629f8..f7f1057 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -935,10 +935,7 @@  i915_gem_execbuffer_move_to_gpu(struct drm_i915_gem_request *req,
 	if (flush_domains & I915_GEM_DOMAIN_GTT)
 		wmb();
 
-	/* Unconditionally invalidate gpu caches and ensure that we do flush
-	 * any residual writes from the previous batch.
-	 */
-	return intel_ring_invalidate_all_caches(req);
+	return 0;
 }
 
 static bool
@@ -1191,17 +1188,6 @@  i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 	u32 instp_mask;
 	int ret;
 
-	ret = i915_gem_execbuffer_move_to_gpu(params->request, vmas);
-	if (ret)
-		return ret;
-
-	ret = i915_switch_context(params->request);
-	if (ret)
-		return ret;
-
-	WARN(params->ctx->ppgtt && params->ctx->ppgtt->pd_dirty_rings & (1<<ring->id),
-	     "%s didn't clear reload\n", ring->name);
-
 	instp_mode = args->flags & I915_EXEC_CONSTANTS_MASK;
 	instp_mask = I915_EXEC_CONSTANTS_MASK;
 	switch (instp_mode) {
@@ -1235,11 +1221,37 @@  i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 		return -EINVAL;
 	}
 
+	ret = i915_gem_execbuffer_move_to_gpu(params->request, vmas);
+	if (ret)
+		return ret;
+
+	i915_gem_execbuffer_move_to_active(vmas, params->request);
+
+	/* To be split into two functions here... */
+
+	intel_runtime_pm_get(dev_priv);
+
+	/*
+	 * Unconditionally invalidate gpu caches and ensure that we do flush
+	 * any residual writes from the previous batch.
+	 */
+	ret = intel_ring_invalidate_all_caches(params->request);
+	if (ret)
+		goto error;
+
+	/* Switch to the correct context for the batch */
+	ret = i915_switch_context(params->request);
+	if (ret)
+		goto error;
+
+	WARN(params->ctx->ppgtt && params->ctx->ppgtt->pd_dirty_rings & (1<<ring->id),
+	     "%s didn't clear reload\n", ring->name);
+
 	if (ring == &dev_priv->ring[RCS] &&
 	    instp_mode != dev_priv->relative_constants_mode) {
 		ret = intel_ring_begin(params->request, 4);
 		if (ret)
-			return ret;
+			goto error;
 
 		intel_ring_emit(ring, MI_NOOP);
 		intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
@@ -1253,7 +1265,7 @@  i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 	if (args->flags & I915_EXEC_GEN7_SOL_RESET) {
 		ret = i915_reset_gen7_sol_offsets(dev, params->request);
 		if (ret)
-			return ret;
+			goto error;
 	}
 
 	exec_len   = args->batch_len;
@@ -1264,14 +1276,20 @@  i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 					exec_start, exec_len,
 					params->dispatch_flags);
 	if (ret)
-		return ret;
+		goto error;
 
 	trace_i915_gem_ring_dispatch(params->request, params->dispatch_flags);
 
-	i915_gem_execbuffer_move_to_active(vmas, params->request);
 	i915_gem_execbuffer_retire_commands(params);
 
-	return 0;
+error:
+	/*
+	 * intel_gpu_busy should also get a ref, so it will free when the device
+	 * is really idle.
+	 */
+	intel_runtime_pm_put(dev_priv);
+
+	return ret;
 }
 
 /**
@@ -1480,8 +1498,6 @@  i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 		args->flags &= ~I915_EXEC_WAIT_FENCE;
 	}
 
-	intel_runtime_pm_get(dev_priv);
-
 	ret = i915_mutex_lock_interruptible(dev);
 	if (ret)
 		goto pre_mutex_err;
@@ -1688,10 +1704,6 @@  err:
 	mutex_unlock(&dev->struct_mutex);
 
 pre_mutex_err:
-	/* intel_gpu_busy should also get a ref, so it will free when the device
-	 * is really idle. */
-	intel_runtime_pm_put(dev_priv);
-
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index e510730..4bf0ee6 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -647,10 +647,7 @@  static int execlists_move_to_gpu(struct drm_i915_gem_request *req,
 	if (flush_domains & I915_GEM_DOMAIN_GTT)
 		wmb();
 
-	/* Unconditionally invalidate gpu caches and ensure that we do flush
-	 * any residual writes from the previous batch.
-	 */
-	return logical_ring_invalidate_all_caches(req);
+	return 0;
 }
 
 int intel_logical_ring_alloc_request_extras(struct drm_i915_gem_request *request)
@@ -913,6 +910,18 @@  int intel_execlists_submission(struct i915_execbuffer_params *params,
 	if (ret)
 		return ret;
 
+	i915_gem_execbuffer_move_to_active(vmas, params->request);
+
+	/* To be split into two functions here... */
+
+	/*
+	 * Unconditionally invalidate gpu caches and ensure that we do flush
+	 * any residual writes from the previous batch.
+	 */
+	ret = logical_ring_invalidate_all_caches(params->request);
+	if (ret)
+		return ret;
+
 	if (ring == &dev_priv->ring[RCS] &&
 	    instp_mode != dev_priv->relative_constants_mode) {
 		ret = intel_logical_ring_begin(params->request, 4);
@@ -937,7 +946,6 @@  int intel_execlists_submission(struct i915_execbuffer_params *params,
 
 	trace_i915_gem_ring_dispatch(params->request, params->dispatch_flags);
 
-	i915_gem_execbuffer_move_to_active(vmas, params->request);
 	i915_gem_execbuffer_retire_commands(params);
 
 	return 0;