i965: Disable auxiliary buffers when there are self-dependencies.

Jason and I investigated several OpenGL CTS failures where the tests
bind the same texture for rendering and texturing, at the same time.
This has defined results as long as the reads happen before writes,
or the regions are non-overlapping.  Normally, this just works out.

However, CCS can cause problems.  If the shader is reading one set of
pixels, and writing to different pixels that are adjacent, they may end
up being covered by the same CCS block.  So rendering may be writing a
CCS block, while the sampler is trying to read it.  Corruption ensues.

Disabling CCS is unfortunate, but safe.

Fixes several KHR-GL45.texture_barrier.* subtests.

Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>

diff --git a/src/mesa/drivers/dri/i965/brw_draw.c b/src/mesa/drivers/dri/i965/brw_draw.c
index 0b1c4d998e390ddcdab3522488696d14eeaae500..d6aa95b6f60b904f3d2a27c629ae334f7be08dbe 100644 (file)
--- a/src/mesa/drivers/dri/i965/brw_draw.c
+++ b/src/mesa/drivers/dri/i965/brw_draw.c
@@ -336,21 +336,38 @@ brw_merge_inputs(struct brw_context *brw,
    }
 }
 
+/* Disable auxiliary buffers if a renderbuffer is also bound as a texture
+ * or shader image.  This causes a self-dependency, where both rendering
+ * and sampling may concurrently read or write the CCS buffer, causing
+ * incorrect pixels.
+ */
 static bool
-intel_disable_rb_aux_buffer(struct brw_context *brw, const struct brw_bo *bo)
+intel_disable_rb_aux_buffer(struct brw_context *brw,
+                            struct intel_mipmap_tree *tex_mt,
+                            const char *usage)
 {
    const struct gl_framebuffer *fb = brw->ctx.DrawBuffer;
    bool found = false;
 
+   /* We only need to worry about color compression and fast clears. */
+   if (tex_mt->aux_usage != ISL_AUX_USAGE_CCS_D &&
+       tex_mt->aux_usage != ISL_AUX_USAGE_CCS_E)
+      return false;
+
    for (unsigned i = 0; i < fb->_NumColorDrawBuffers; i++) {
       const struct intel_renderbuffer *irb =
          intel_renderbuffer(fb->_ColorDrawBuffers[i]);
 
-      if (irb && irb->mt->bo == bo) {
+      if (irb && irb->mt->bo == tex_mt->bo) {
          found = brw->draw_aux_buffer_disabled[i] = true;
       }
    }
 
+   if (found) {
+      perf_debug("Disabling CCS because a renderbuffer is also bound %s.\n",
+                 usage);
+   }
+
    return found;
 }
 
@@ -363,7 +380,6 @@ intel_disable_rb_aux_buffer(struct brw_context *brw, const struct brw_bo *bo)
 void
 brw_predraw_resolve_inputs(struct brw_context *brw)
 {
-   const struct gen_device_info *devinfo = &brw->screen->devinfo;
    struct gl_context *ctx = &brw->ctx;
    struct intel_texture_object *tex_obj;
 
@@ -383,15 +399,11 @@ brw_predraw_resolve_inputs(struct brw_context *brw)
       enum isl_format view_format =
          translate_tex_format(brw, tex_obj->_Format, sampler->sRGBDecode);
 
-      bool aux_supported;
-      intel_miptree_prepare_texture(brw, tex_obj->mt, view_format,
-                                    &aux_supported);
+      const bool disable_aux =
+         intel_disable_rb_aux_buffer(brw, tex_obj->mt, "for sampling");
 
-      if (!aux_supported && devinfo->gen >= 9 &&
-          intel_disable_rb_aux_buffer(brw, tex_obj->mt->bo)) {
-         perf_debug("Sampling renderbuffer with non-compressible format - "
-                    "turning off compression\n");
-      }
+      intel_miptree_prepare_texture(brw, tex_obj->mt, view_format,
+                                    disable_aux);
 
       brw_render_cache_set_check_flush(brw, tex_obj->mt->bo);
 
@@ -412,13 +424,10 @@ brw_predraw_resolve_inputs(struct brw_context *brw)
             tex_obj = intel_texture_object(u->TexObj);
 
             if (tex_obj && tex_obj->mt) {
-               intel_miptree_prepare_image(brw, tex_obj->mt);
+               intel_disable_rb_aux_buffer(brw, tex_obj->mt,
+                                           "as a shader image");
 
-               if (tex_obj->mt->aux_usage == ISL_AUX_USAGE_CCS_E &&
-                   intel_disable_rb_aux_buffer(brw, tex_obj->mt->bo)) {
-                  perf_debug("Using renderbuffer as shader image - turning "
-                             "off lossless compression\n");
-               }
+               intel_miptree_prepare_image(brw, tex_obj->mt);
 
                brw_render_cache_set_check_flush(brw, tex_obj->mt->bo);
             }

diff --git a/src/mesa/drivers/dri/i965/intel_mipmap_tree.c b/src/mesa/drivers/dri/i965/intel_mipmap_tree.c
index 670a92c1168a5affc9e337d1df7c06d53f7366ee..95bc48e6529417c97e3d8bd46a004a27a83fd627 100644 (file)
--- a/src/mesa/drivers/dri/i965/intel_mipmap_tree.c
+++ b/src/mesa/drivers/dri/i965/intel_mipmap_tree.c
@@ -2630,9 +2630,9 @@ intel_miptree_prepare_texture_slices(struct brw_context *brw,
                                      enum isl_format view_format,
                                      uint32_t start_level, uint32_t num_levels,
                                      uint32_t start_layer, uint32_t num_layers,
-                                     bool *aux_supported_out)
+                                     bool disable_aux)
 {
-   enum isl_aux_usage aux_usage =
+   enum isl_aux_usage aux_usage = disable_aux ? ISL_AUX_USAGE_NONE :
       intel_miptree_texture_aux_usage(brw, mt, view_format);
    bool clear_supported = aux_usage != ISL_AUX_USAGE_NONE;
 
@@ -2646,20 +2646,18 @@ intel_miptree_prepare_texture_slices(struct brw_context *brw,
    intel_miptree_prepare_access(brw, mt, start_level, num_levels,
                                 start_layer, num_layers,
                                 aux_usage, clear_supported);
-   if (aux_supported_out)
-      *aux_supported_out = aux_usage != ISL_AUX_USAGE_NONE;
 }
 
 void
 intel_miptree_prepare_texture(struct brw_context *brw,
                               struct intel_mipmap_tree *mt,
                               enum isl_format view_format,
-                              bool *aux_supported_out)
+                              bool disable_aux)
 {
    intel_miptree_prepare_texture_slices(brw, mt, view_format,
                                         0, INTEL_REMAINING_LEVELS,
                                         0, INTEL_REMAINING_LAYERS,
-                                        aux_supported_out);
+                                        disable_aux);
 }
 
 void
@@ -2677,8 +2675,13 @@ intel_miptree_prepare_fb_fetch(struct brw_context *brw,
                                struct intel_mipmap_tree *mt, uint32_t level,
                                uint32_t start_layer, uint32_t num_layers)
 {
+   /* This is only used for non-coherent framebuffer fetch, so we don't
+    * need to worry about CCS_E and can simply pass 'false' below.
+    */
+   assert(brw->screen->devinfo.gen < 9);
+
    intel_miptree_prepare_texture_slices(brw, mt, mt->surf.format, level, 1,
-                                        start_layer, num_layers, NULL);
+                                        start_layer, num_layers, false);
 }
 
 enum isl_aux_usage

diff --git a/src/mesa/drivers/dri/i965/intel_mipmap_tree.h b/src/mesa/drivers/dri/i965/intel_mipmap_tree.h
index 439b0f66aeb0d29e20a435c6647d4655bd5ee297..5ab929bb40b1ed5a35b185a17ceba13fb9474f0d 100644 (file)
--- a/src/mesa/drivers/dri/i965/intel_mipmap_tree.h
+++ b/src/mesa/drivers/dri/i965/intel_mipmap_tree.h
@@ -640,7 +640,7 @@ void
 intel_miptree_prepare_texture(struct brw_context *brw,
                               struct intel_mipmap_tree *mt,
                               enum isl_format view_format,
-                              bool *aux_supported_out);
+                              bool disable_aux);
 void
 intel_miptree_prepare_image(struct brw_context *brw,
                             struct intel_mipmap_tree *mt);