Mesa (master): i965/blorp: Optimize manual_blend() for compressed multisampled surfaces.

[Paul Berry]

Module: Mesa
Branch: master
Commit: f91b4d92b97664e6354f66138705e93bec363ba0
URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=f91b4d92b97664e6354f66138705e93bec363ba0

Author: Paul Berry <stereotype441 at gmail.com>
Date:   Mon Jul  9 11:10:52 2012 -0700

i965/blorp: Optimize manual_blend() for compressed multisampled surfaces.

When downsampling a compressed multisampled surface, we can take a
shortcut to downsample any pixels that were completely covered by a
single primitive.  In this case, the first color value we fetch is the
correct final color for the downsampled pixel, so we can skip the rest
of the blending operation.

Reviewed-by: Anuj Phogat <anuj.phogat at gmail.com>

---

 src/mesa/drivers/dri/i965/brw_blorp_blit.cpp |   23 +++++++++++++++++++++++
 1 files changed, 23 insertions(+), 0 deletions(-)

diff --git a/src/mesa/drivers/dri/i965/brw_blorp_blit.cpp b/src/mesa/drivers/dri/i965/brw_blorp_blit.cpp
index 32fd48e..c8db662 100644
--- a/src/mesa/drivers/dri/i965/brw_blorp_blit.cpp
+++ b/src/mesa/drivers/dri/i965/brw_blorp_blit.cpp
@@ -1143,6 +1143,26 @@ brw_blorp_blit_program::manual_blend()
       }
       texel_fetch(texture_data[stack_depth++]);
 
+      if (i == 0 && key->tex_layout == INTEL_MSAA_LAYOUT_CMS) {
+         /* The Ivy Bridge PRM, Vol4 Part1 p27 (Multisample Control Surface)
+          * suggests an optimization:
+          *
+          *     "A simple optimization with probable large return in
+          *     performance is to compare the MCS value to zero (indicating
+          *     all samples are on sample slice 0), and sample only from
+          *     sample slice 0 using ld2dss if MCS is zero."
+          *
+          * Note that in the case where the MCS value is zero, sampling from
+          * sample slice 0 using ld2dss and sampling from sample 0 using
+          * ld2dms are equivalent (since all samples are on sample slice 0).
+          * Since we have already sampled from sample 0, all we need to do is
+          * skip the remaining fetches and averaging if MCS is zero.
+          */
+         brw_CMP(&func, vec16(brw_null_reg()), BRW_CONDITIONAL_NZ,
+                 mcs_data, brw_imm_ud(0));
+         brw_IF(&func, BRW_EXECUTE_16);
+      }
+
       /* Do count_trailing_one_bits(i) times */
       for (int j = count_trailing_one_bits(i); j-- > 0; ) {
          assert(stack_depth >= 2);
@@ -1169,6 +1189,9 @@ brw_blorp_blit_program::manual_blend()
                  brw_imm_f(1.0/num_samples));
       }
    }
+
+   if (key->tex_layout == INTEL_MSAA_LAYOUT_CMS)
+      brw_ENDIF(&func);
 }
 
 /**