i965: Enable nir_opt_idiv_const for 32 and 64-bit integers

The pass should work for all bit sizes but it's less clear that the
extra instructions are worth it on small integers.  Also, the hardware
doesn't do mul_high on anything other than 32-bit integers and, absent
any decent mechanism for testing the pass on 8 and 16-bit types, it's
probably best to just leave it disabled for now.

Shader-db results on Sky Lake:

    total instructions in shared programs: 15105795 -> 15111403 (0.04%)
    instructions in affected programs: 72774 -> 78382 (7.71%)
    helped: 0
    HURT: 265

Note that hurt here actually means helped because we're getting rid of
integer quotient operations (which are a send on some platforms!) and
replacing them with fairly cheap ALU ops.

Reviewed-by: Ian Romanick ian.d.romanick@intel.com

diff --git a/src/intel/compiler/brw_nir.c b/src/intel/compiler/brw_nir.c
index 2723ab6..44a720c 100644 (file)
--- a/src/intel/compiler/brw_nir.c
+++ b/src/intel/compiler/brw_nir.c
@@ -570,6 +570,7 @@ brw_nir_optimize(nir_shader *nir, const struct brw_compiler *compiler,
       OPT(nir_opt_cse);
       OPT(nir_opt_peephole_select, 0);
       OPT(nir_opt_intrinsics);
+      OPT(nir_opt_idiv_const, 32);
       OPT(nir_opt_algebraic);
       OPT(nir_opt_constant_folding);
       OPT(nir_opt_dead_cf);
@@ -679,7 +680,8 @@ brw_preprocess_nir(const struct brw_compiler *compiler, nir_shader *nir)
     */
    OPT(nir_lower_int64, nir_lower_imul64 |
                         nir_lower_isign64 |
-                        nir_lower_divmod64);
+                        nir_lower_divmod64 |
+                        nir_lower_imul_high64);
 
    nir = brw_nir_optimize(nir, compiler, is_scalar, true);