[Mesa-dev] [PATCH 4/9] radeon/llvm: improve ABS_i32 lowering

[Vadim Girlin]

We can save one instruction by lowering it to:
  SUB_INT tmp, 0, src
  MAX_INT dst, src, tmp

Signed-off-by: Vadim Girlin <vadimgirlin at gmail.com>
---
 src/gallium/drivers/radeon/R600LowerInstructions.cpp |   18 +++++-------------
 1 file changed, 5 insertions(+), 13 deletions(-)

diff --git a/src/gallium/drivers/radeon/R600LowerInstructions.cpp b/src/gallium/drivers/radeon/R600LowerInstructions.cpp
index 41bf365..42c9766 100644
--- a/src/gallium/drivers/radeon/R600LowerInstructions.cpp
+++ b/src/gallium/drivers/radeon/R600LowerInstructions.cpp
@@ -90,24 +90,16 @@ bool R600LowerInstructionsPass::runOnMachineFunction(MachineFunction &MF)
 
       case AMDIL::ABS_i32:
         {
-          unsigned setgt = MRI->createVirtualRegister(
+          unsigned neg = MRI->createVirtualRegister(
                            &AMDIL::R600_TReg32RegClass);
-          BuildMI(MBB, I, MBB.findDebugLoc(I), TII->get(AMDIL::SETGE_INT),
-                  setgt)
+          BuildMI(MBB, I, MBB.findDebugLoc(I), TII->get(AMDIL::SUB_INT),neg)
                   .addReg(AMDIL::ZERO)
                   .addOperand(MI.getOperand(1));
 
-          unsigned add_int = MRI->createVirtualRegister(
-                             &AMDIL::R600_TReg32RegClass);
-          BuildMI(MBB, I, MBB.findDebugLoc(I), TII->get(AMDIL::ADD_INT),
-                  add_int)
-                  .addReg(setgt)
-                  .addOperand(MI.getOperand(1));
-
-          BuildMI(MBB, I, MBB.findDebugLoc(I), TII->get(AMDIL::XOR_INT))
+          BuildMI(MBB, I, MBB.findDebugLoc(I), TII->get(AMDIL::MAX_INT))
                   .addOperand(MI.getOperand(0))
-                  .addReg(setgt)
-                  .addReg(add_int);
+                  .addOperand(MI.getOperand(1))
+                  .addReg(neg);
 
           break;
         }
-- 
1.7.10.1