[04/10] drm/i915: Choose partial chunksize based on tile row size

Commit Message

Chris Wilson Aug. 12, 2016, 10:28 a.m. UTC

In order to support setting up fences for partial mappings of an object,
we have to align those mappings with the fence. The minimum chunksize we
choose is at least the size of a single tile row.

Signed-off-by: Chris Wilson <[e-mail]>
---
 drivers/gpu/drm/i915/i915_gem.c | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)

Comments

Joonas Lahtinen Aug. 12, 2016, 10:38 a.m. UTC | #1

On pe, 2016-08-12 at 11:28 +0100, Chris Wilson wrote:
> In order to support setting up fences for partial mappings of an object,
> we have to align those mappings with the fence. The minimum chunksize we
> choose is at least the size of a single tile row.
> 
> Signed-off-by: Chris Wilson <[e-mail]>

Reviewed-by: Joonas Lahtinen <[e-mail]>

Regards, Joonas

Patch

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index b386efaa5aa2..b0e73c5a29c6 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1650,6 +1650,16 @@  i915_gem_mmap_ioctl(struct drm_device *dev, void *data,
 	return 0;
 }
 
+static unsigned int tile_row_pages(struct drm_i915_gem_object *obj)
+{
+	u64 size;
+
+	size = i915_gem_object_get_stride(obj);
+	size *= i915_gem_object_get_tiling(obj) == I915_TILING_Y ? 32 : 8;
+
+	return size >> PAGE_SHIFT;
+}
+
 /**
  * i915_gem_fault - fault a page into the GTT
  * @area: CPU VMA in question
@@ -1709,7 +1719,11 @@  int i915_gem_fault(struct vm_area_struct *area, struct vm_fault *vmf)
 	/* Use a partial view if the object is bigger than the aperture. */
 	if (obj->base.size >= ggtt->mappable_end &&
 	    !i915_gem_object_is_tiled(obj)) {
-		static const unsigned int chunk_size = 256; // 1 MiB
+		unsigned int chunk_size;
+
+		chunk_size = 256; /* 1 MiB */
+		if (i915_gem_object_is_tiled(obj))
+			chunk_size = max(chunk_size, tile_row_pages(obj));
 
 		memset(&view, 0, sizeof(view));
 		view.type = I915_GGTT_VIEW_PARTIAL;