[PATCH 16/16] manual-waits

Chris Wilson chris at chris-wilson.co.uk
Fri Jan 1 10:52:43 UTC 2021


---
 drivers/gpu/drm/i915/gem/i915_gem_domain.c    | 77 +++++++------------
 .../i915/gem/selftests/i915_gem_coherency.c   | 26 +++++++
 2 files changed, 53 insertions(+), 50 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_domain.c b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
index 666e4c60d610..61031897da28 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_domain.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
@@ -49,7 +49,7 @@ flush_write_domain(struct drm_i915_gem_object *obj, unsigned int flush_domains)
 		break;
 
 	case I915_GEM_DOMAIN_CPU:
-		i915_gem_clflush_object(obj, I915_CLFLUSH_SYNC);
+		i915_gem_clflush_object(obj, 0);
 		break;
 
 	case I915_GEM_DOMAIN_RENDER:
@@ -104,13 +104,6 @@ i915_gem_object_set_to_wc_domain(struct drm_i915_gem_object *obj, bool write)
 
 	assert_object_held(obj);
 
-	ret = i915_gem_object_wait(obj,
-				   I915_WAIT_INTERRUPTIBLE |
-				   (write ? I915_WAIT_ALL : 0),
-				   MAX_SCHEDULE_TIMEOUT);
-	if (ret)
-		return ret;
-
 	if (obj->write_domain == I915_GEM_DOMAIN_WC)
 		return 0;
 
@@ -165,13 +158,6 @@ i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)
 
 	assert_object_held(obj);
 
-	ret = i915_gem_object_wait(obj,
-				   I915_WAIT_INTERRUPTIBLE |
-				   (write ? I915_WAIT_ALL : 0),
-				   MAX_SCHEDULE_TIMEOUT);
-	if (ret)
-		return ret;
-
 	if (obj->write_domain == I915_GEM_DOMAIN_GTT)
 		return 0;
 
@@ -444,17 +430,8 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
 int
 i915_gem_object_set_to_cpu_domain(struct drm_i915_gem_object *obj, bool write)
 {
-	int ret;
-
 	assert_object_held(obj);
 
-	ret = i915_gem_object_wait(obj,
-				   I915_WAIT_INTERRUPTIBLE |
-				   (write ? I915_WAIT_ALL : 0),
-				   MAX_SCHEDULE_TIMEOUT);
-	if (ret)
-		return ret;
-
 	flush_write_domain(obj, ~I915_GEM_DOMAIN_CPU);
 
 	/* Flush the CPU cache if it's still invalid. */
@@ -512,19 +489,6 @@ i915_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 	if (!obj)
 		return -ENOENT;
 
-	/*
-	 * Try to flush the object off the GPU without holding the lock.
-	 * We will repeat the flush holding the lock in the normal manner
-	 * to catch cases where we are gazumped.
-	 */
-	err = i915_gem_object_wait(obj,
-				   I915_WAIT_INTERRUPTIBLE |
-				   I915_WAIT_PRIORITY |
-				   (write_domain ? I915_WAIT_ALL : 0),
-				   MAX_SCHEDULE_TIMEOUT);
-	if (err)
-		goto out;
-
 	/*
 	 * Proxy objects do not control access to the backing storage, ergo
 	 * they cannot be used as a means to manipulate the cache domain
@@ -575,6 +539,19 @@ i915_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 
 	i915_gem_object_unlock(obj);
 
+	/*
+	 * Try to flush the object off the GPU without holding the lock.
+	 * We will repeat the flush holding the lock in the normal manner
+	 * to catch cases where we are gazumped.
+	 */
+	err = i915_gem_object_wait(obj,
+				   I915_WAIT_INTERRUPTIBLE |
+				   I915_WAIT_PRIORITY |
+				   (write_domain ? I915_WAIT_ALL : 0),
+				   MAX_SCHEDULE_TIMEOUT);
+	if (err)
+		goto out;
+
 	if (write_domain)
 		i915_gem_object_invalidate_frontbuffer(obj, ORIGIN_CPU);
 
@@ -601,12 +578,6 @@ int i915_gem_object_prepare_read(struct drm_i915_gem_object *obj,
 
 	assert_object_held(obj);
 
-	ret = i915_gem_object_wait(obj,
-				   I915_WAIT_INTERRUPTIBLE,
-				   MAX_SCHEDULE_TIMEOUT);
-	if (ret)
-		return ret;
-
 	ret = i915_gem_object_pin_pages(obj);
 	if (ret)
 		return ret;
@@ -622,6 +593,12 @@ int i915_gem_object_prepare_read(struct drm_i915_gem_object *obj,
 
 	flush_write_domain(obj, ~I915_GEM_DOMAIN_CPU);
 
+	ret = i915_gem_object_wait(obj,
+				   I915_WAIT_INTERRUPTIBLE,
+				   MAX_SCHEDULE_TIMEOUT);
+	if (ret)
+		goto err_unpin;
+
 	/* If we're not in the cpu read domain, set ourself into the gtt
 	 * read domain and manually flush cachelines (if required). This
 	 * optimizes for the case when the gpu will dirty the data
@@ -651,13 +628,6 @@ int i915_gem_object_prepare_write(struct drm_i915_gem_object *obj,
 
 	assert_object_held(obj);
 
-	ret = i915_gem_object_wait(obj,
-				   I915_WAIT_INTERRUPTIBLE |
-				   I915_WAIT_ALL,
-				   MAX_SCHEDULE_TIMEOUT);
-	if (ret)
-		return ret;
-
 	ret = i915_gem_object_pin_pages(obj);
 	if (ret)
 		return ret;
@@ -673,6 +643,13 @@ int i915_gem_object_prepare_write(struct drm_i915_gem_object *obj,
 
 	flush_write_domain(obj, ~I915_GEM_DOMAIN_CPU);
 
+	ret = i915_gem_object_wait(obj,
+				   I915_WAIT_INTERRUPTIBLE |
+				   I915_WAIT_ALL,
+				   MAX_SCHEDULE_TIMEOUT);
+	if (ret)
+		goto err_unpin;
+
 	/* If we're not in the cpu write domain, set ourself into the
 	 * gtt write domain and manually flush cachelines (as required).
 	 * This optimizes for the case when the gpu will use the data
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c
index 1117d2a44518..050062191f51 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c
@@ -95,6 +95,13 @@ static int gtt_set(struct context *ctx, unsigned long offset, u32 v)
 	if (err)
 		return err;
 
+	err = i915_gem_object_wait(ctx->obj,
+				   I915_WAIT_ALL |
+				   I915_WAIT_INTERRUPTIBLE,
+				   HZ / 2);
+	if (err)
+		return err;
+
 	vma = i915_gem_object_ggtt_pin(ctx->obj, NULL, 0, 0, PIN_MAPPABLE);
 	if (IS_ERR(vma))
 		return PTR_ERR(vma);
@@ -128,6 +135,12 @@ static int gtt_get(struct context *ctx, unsigned long offset, u32 *v)
 	if (err)
 		return err;
 
+	err = i915_gem_object_wait(ctx->obj,
+				   I915_WAIT_INTERRUPTIBLE,
+				   HZ / 2);
+	if (err)
+		return err;
+
 	vma = i915_gem_object_ggtt_pin(ctx->obj, NULL, 0, 0, PIN_MAPPABLE);
 	if (IS_ERR(vma))
 		return PTR_ERR(vma);
@@ -160,6 +173,13 @@ static int wc_set(struct context *ctx, unsigned long offset, u32 v)
 	if (err)
 		return err;
 
+	err = i915_gem_object_wait(ctx->obj,
+				   I915_WAIT_ALL |
+				   I915_WAIT_INTERRUPTIBLE,
+				   HZ / 2);
+	if (err)
+		return err;
+
 	map = i915_gem_object_pin_map(ctx->obj, I915_MAP_WC);
 	if (IS_ERR(map))
 		return PTR_ERR(map);
@@ -183,6 +203,12 @@ static int wc_get(struct context *ctx, unsigned long offset, u32 *v)
 	if (err)
 		return err;
 
+	err = i915_gem_object_wait(ctx->obj,
+				   I915_WAIT_INTERRUPTIBLE,
+				   HZ / 2);
+	if (err)
+		return err;
+
 	map = i915_gem_object_pin_map(ctx->obj, I915_MAP_WC);
 	if (IS_ERR(map))
 		return PTR_ERR(map);
-- 
2.20.1



More information about the Intel-gfx-trybot mailing list