[Mesa-dev] [PATCH 08/25] SQUASH: i965/fs: Add basic dataflow check to opt_sampler_eot().

[Francisco Jerez]

The sampler EOT optimization pass naively assumes that the texturing
instruction provides all the data used by the FB write just because
they're standing next to each other.  The least we should be checking
is whether the source and destination regions of the FB write and
texturing instructions match.  Without this the previous seemingly
harmless commit would have caused opt_sampler_eot() to misoptimize a
shader from dota-2 causing DCE to eliminate all of its 78 instructions
except for the final sampler EOT message (!).
---
 src/mesa/drivers/dri/i965/brw_defines.h |  1 +
 src/mesa/drivers/dri/i965/brw_fs.cpp    | 14 ++++++++++++++
 2 files changed, 15 insertions(+)

diff --git a/src/mesa/drivers/dri/i965/brw_defines.h b/src/mesa/drivers/dri/i965/brw_defines.h
index 8794d44..60b215d 100644
--- a/src/mesa/drivers/dri/i965/brw_defines.h
+++ b/src/mesa/drivers/dri/i965/brw_defines.h
@@ -1395,6 +1395,7 @@ enum fb_write_logical_srcs {
    FB_WRITE_LOGICAL_SRC_SRC_STENCIL, /* gl_FragStencilRefARB */
    FB_WRITE_LOGICAL_SRC_OMASK,       /* Sample Mask (gl_SampleMask) */
    FB_WRITE_LOGICAL_SRC_COMPONENTS,  /* REQUIRED */
+   FB_WRITE_LOGICAL_NUM_SRCS
 };
 
 enum tex_logical_srcs {
diff --git a/src/mesa/drivers/dri/i965/brw_fs.cpp b/src/mesa/drivers/dri/i965/brw_fs.cpp
index 6abf776..1574cd9 100644
--- a/src/mesa/drivers/dri/i965/brw_fs.cpp
+++ b/src/mesa/drivers/dri/i965/brw_fs.cpp
@@ -2647,6 +2647,20 @@ fs_visitor::opt_sampler_eot()
    if (tex_inst->prev->is_head_sentinel())
       return false;
 
+   /* Check that the FB write sources are fully initialized by the single
+    * texturing instruction.
+    */
+   for (unsigned i = 0; i < FB_WRITE_LOGICAL_NUM_SRCS; i++) {
+      if (i == FB_WRITE_LOGICAL_SRC_COLOR0) {
+         if (!fb_write->src[i].equals(tex_inst->dst) ||
+             fb_write->regs_read(i) != tex_inst->regs_written)
+         return false;
+      } else if (i != FB_WRITE_LOGICAL_SRC_COMPONENTS) {
+         if (fb_write->src[i].file != BAD_FILE)
+            return false;
+      }
+   }
+
    assert(!tex_inst->eot); /* We can't get here twice */
    assert((tex_inst->offset & (0xff << 24)) == 0);
 
-- 
2.7.3