Mesa (master): aco: use nir_lower_idiv_precise

[GitLab Mirror]

Module: Mesa
Branch: master
Commit: 132ae89b1908dc1ba8483e72ded9380764d8435c
URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=132ae89b1908dc1ba8483e72ded9380764d8435c

Author: Rhys Perry <pendingchaos02 at gmail.com>
Date:   Wed Sep 18 20:45:05 2019 +0100

aco: use nir_lower_idiv_precise

v7: rename _nv50/_llvm to _fast/_precise

Signed-off-by: Rhys Perry <pendingchaos02 at gmail.com>
Reviewed-by: Daniel Schürmann <daniel at schuermann.dev>

---

 src/amd/compiler/aco_instruction_selection_setup.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/amd/compiler/aco_instruction_selection_setup.cpp b/src/amd/compiler/aco_instruction_selection_setup.cpp
index fa457e2e246..9727020a8a9 100644
--- a/src/amd/compiler/aco_instruction_selection_setup.cpp
+++ b/src/amd/compiler/aco_instruction_selection_setup.cpp
@@ -1328,7 +1328,7 @@ setup_isel_context(Program* program,
                                                       nir_lower_iabs64));
 
       nir_opt_idiv_const(nir, 32);
-      nir_lower_idiv(nir, nir_lower_idiv_fast); // TODO: use the LLVM path once !1239 is merged
+      nir_lower_idiv(nir, nir_lower_idiv_precise);
 
       /* optimize the lowered ALU operations */
       nir_copy_prop(nir);