[Mesa-dev] [PATCH] Revert "i965/fs: Allow scalar source regions on SNB math instructions."

[Francisco Jerez]

This reverts commit c1107cec44ab030c7fcc97c67baa12df1cc9d7b5.
Apparently the hardware spec text I quoted in the commit message was
outright lying about scalar source math being supported on SNB, the
hardware seems to load 32 contiguous bits of data for each channel
regardless of the regioning mode.  Fixes regressions in the following
CTS tests (which we didn't catch early due to CTS being temporarily
disabled in our CI system):

   es2-cts.gtf.gl.atan.atan_vec3_frag_xvary
   es2-cts.gtf.gl.cos.cos_vec2_frag_xvary
   es2-cts.gtf.gl.atan.atan_vec2_frag_xvary
   es2-cts.gtf.gl.pow.pow_vec2_frag_xvary_yconsthalf
   es2-cts.gtf.gl.cos.cos_float_frag_xvary
   es2-cts.gtf.gl.pow.pow_float_frag_xvary_yconsthalf
   es2-cts.gtf.gl.atan.atan_vec3_frag_xvaryyvary
   es2-cts.gtf.gl.pow.pow_vec3_frag_xvary_yconsthalf
   es2-cts.gtf.gl.cos.cos_vec3_frag_xvary
   es2-cts.gtf.gl.atan.atan_vec2_frag_xvaryyvary

Cc: mesa-stable at lists.freedesktop.org
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96346
Reported-by: Mark Janes <mark.a.janes at intel.com>
---
 src/mesa/drivers/dri/i965/brw_eu_emit.c               |  6 ++----
 src/mesa/drivers/dri/i965/brw_fs_builder.h            | 10 ++++++++--
 src/mesa/drivers/dri/i965/brw_fs_copy_propagation.cpp |  9 +++++++--
 3 files changed, 17 insertions(+), 8 deletions(-)

diff --git a/src/mesa/drivers/dri/i965/brw_eu_emit.c b/src/mesa/drivers/dri/i965/brw_eu_emit.c
index 8218f9c..2a8e661 100644
--- a/src/mesa/drivers/dri/i965/brw_eu_emit.c
+++ b/src/mesa/drivers/dri/i965/brw_eu_emit.c
@@ -2000,10 +2000,8 @@ void gen6_math(struct brw_codegen *p,
 
    assert(dest.hstride == BRW_HORIZONTAL_STRIDE_1);
    if (devinfo->gen == 6) {
-      assert(has_scalar_region(src0) ||
-             src0.hstride == BRW_HORIZONTAL_STRIDE_1);
-      assert(has_scalar_region(src1) ||
-             src1.hstride == BRW_HORIZONTAL_STRIDE_1);
+      assert(src0.hstride == BRW_HORIZONTAL_STRIDE_1);
+      assert(src1.hstride == BRW_HORIZONTAL_STRIDE_1);
    }
 
    if (function == BRW_MATH_FUNCTION_INT_DIV_QUOTIENT ||
diff --git a/src/mesa/drivers/dri/i965/brw_fs_builder.h b/src/mesa/drivers/dri/i965/brw_fs_builder.h
index aef35f3..f22903e 100644
--- a/src/mesa/drivers/dri/i965/brw_fs_builder.h
+++ b/src/mesa/drivers/dri/i965/brw_fs_builder.h
@@ -621,14 +621,20 @@ namespace brw {
       src_reg
       fix_math_operand(const src_reg &src) const
       {
-         /* Gen6 hardware ignores source modifiers (negate and abs) on math
+         /* Can't do hstride == 0 args on gen6 math, so expand it out. We
+          * might be able to do better by doing execsize = 1 math and then
+          * expanding that result out, but we would need to be careful with
+          * masking.
+          *
+          * Gen6 hardware ignores source modifiers (negate and abs) on math
           * instructions, so we also move to a temp to set those up.
           *
           * Gen7 relaxes most of the above restrictions, but still can't use IMM
           * operands to math
           */
          if ((shader->devinfo->gen == 6 &&
-              (src.file == IMM || src.abs || src.negate)) ||
+              (src.file == IMM || src.file == UNIFORM ||
+               src.abs || src.negate)) ||
              (shader->devinfo->gen == 7 && src.file == IMM)) {
             const dst_reg tmp = vgrf(src.type);
             MOV(tmp, src);
diff --git a/src/mesa/drivers/dri/i965/brw_fs_copy_propagation.cpp b/src/mesa/drivers/dri/i965/brw_fs_copy_propagation.cpp
index 2a83eb9..d88d62b 100644
--- a/src/mesa/drivers/dri/i965/brw_fs_copy_propagation.cpp
+++ b/src/mesa/drivers/dri/i965/brw_fs_copy_propagation.cpp
@@ -578,9 +578,14 @@ fs_visitor::try_constant_propagate(fs_inst *inst, acp_entry *entry)
             break;
          /* fallthrough */
       case SHADER_OPCODE_POW:
-         /* Allow constant propagation into src1, and let constant combining
-          * promote the constant on Gen < 8.
+         /* Allow constant propagation into src1 (except on Gen 6), and let
+          * constant combining promote the constant on Gen < 8.
+          *
+          * While Gen 6 MATH can take a scalar source, its source and
+          * destination offsets must be equal and we cannot ensure that.
           */
+         if (devinfo->gen == 6)
+            break;
          /* fallthrough */
       case BRW_OPCODE_BFI1:
       case BRW_OPCODE_ASR:
-- 
2.7.3