v3d: Drop in a bunch of notes about performance improvement opportunities.

[mesa.git] / src / gallium / drivers / v3d / v3dx_rcl.c

diff --git a/src/gallium/drivers/v3d/v3dx_rcl.c b/src/gallium/drivers/v3d/v3dx_rcl.c
index 3a76b0f3b24ff2b972859009f57eae01132939d3..17b30465c9d7cbe2ce1950196b746ec7ebdd07dc 100644 (file)
--- a/src/gallium/drivers/v3d/v3dx_rcl.c
+++ b/src/gallium/drivers/v3d/v3dx_rcl.c
@@ -74,6 +74,7 @@ load_general(struct v3d_cl *cl, struct pipe_surface *psurf, int buffer,
                         load.input_image_format = V3D_OUTPUT_IMAGE_FORMAT_S8;
                 else
                         load.input_image_format = surf->format;
+                load.r_b_swap = surf->swap_rb;
 
                 if (surf->tiling == VC5_TILING_UIF_NO_XOR ||
                     surf->tiling == VC5_TILING_UIF_XOR) {
@@ -137,6 +138,7 @@ store_general(struct v3d_job *job,
                 else
                         store.output_image_format = surf->format;
 
+                store.r_b_swap = surf->swap_rb;
                 store.memory_format = surf->tiling;
 
                 if (surf->tiling == VC5_TILING_UIF_NO_XOR ||
@@ -759,7 +761,10 @@ v3dX(emit_rcl)(struct v3d_job *job)
 
         v3d_rcl_emit_generic_per_tile_list(job, nr_cbufs - 1);
 
-        /* XXX: Use Morton order */
+        /* XXX perf: We should expose GL_MESA_tile_raster_order to improve X11
+         * performance, but we should use Morton order otherwise to improve
+         * cache locality.
+         */
         uint32_t supertile_w_in_pixels = job->tile_width * supertile_w;
         uint32_t supertile_h_in_pixels = job->tile_height * supertile_h;
         uint32_t min_x_supertile = job->draw_min_x / supertile_w_in_pixels;