Mesa (master): i965/nir: do int64 lowering before optimization

[Iago Toral Quiroga]

Module: Mesa
Branch: master
Commit: 1d20001d9711a7ea06f42292d3ed545d7ca0f50c
URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=1d20001d9711a7ea06f42292d3ed545d7ca0f50c

Author: Iago Toral Quiroga <itoral at igalia.com>
Date:   Fri Dec  1 13:46:23 2017 +0100

i965/nir: do int64 lowering before optimization

Otherwise loop unrolling will fail to see the actual cost of
the unrolling operations when the loop body contains 64-bit integer
instructions, and very specially when the divmod64 lowering applies,
since its lowering is quite expensive.

Without this change, some in-development CTS tests for int64
get stuck forever trying to register allocate a shader with
over 50K SSA values. The large number of SSA values is the result
of NIR first unrolling multiple seemingly simple loops that involve
int64 instructions, only to then lower these instructions to produce
a massive pile of code (due to the divmod64 lowering in the unrolled
instructions).

With this change, loop unrolling will see the loops with the int64
code already lowered and will realize that it is too expensive to
unroll.

v2: Run nir_algebraic first so we can hopefully get rid of some of
    the int64 instructions before we even attempt to lower them.

Reviewed-by: Matt Turner <mattst88 at gmail.com>

---

 src/intel/compiler/brw_nir.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/src/intel/compiler/brw_nir.c b/src/intel/compiler/brw_nir.c
index dbddef0d04..287bd4908d 100644
--- a/src/intel/compiler/brw_nir.c
+++ b/src/intel/compiler/brw_nir.c
@@ -634,6 +634,18 @@ brw_preprocess_nir(const struct brw_compiler *compiler, nir_shader *nir)
 
    OPT(nir_split_var_copies);
 
+   /* Run opt_algebraic before int64 lowering so we can hopefully get rid
+    * of some int64 instructions.
+    */
+   OPT(nir_opt_algebraic);
+
+   /* Lower int64 instructions before nir_optimize so that loop unrolling
+    * sees their actual cost.
+    */
+   nir_lower_int64(nir, nir_lower_imul64 |
+                        nir_lower_isign64 |
+                        nir_lower_divmod64);
+
    nir = brw_nir_optimize(nir, compiler, is_scalar);
 
    if (is_scalar) {
@@ -661,10 +673,6 @@ brw_preprocess_nir(const struct brw_compiler *compiler, nir_shader *nir)
       brw_nir_no_indirect_mask(compiler, nir->info.stage);
    nir_lower_indirect_derefs(nir, indirect_mask);
 
-   nir_lower_int64(nir, nir_lower_imul64 |
-                        nir_lower_isign64 |
-                        nir_lower_divmod64);
-
    /* Get rid of split copies */
    nir = brw_nir_optimize(nir, compiler, is_scalar);