maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

R600/SI: use patterns for clamp, fabs, fneg 

Instead of using custom inserters, it's simpler and
should make DAG folding easier.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 175755
This commit is contained in:
Christian Konig 2013-02-21 15:17:27 +00:00
parent bf114b42a8
commit 8dbe6f617c
2 changed files with 22 additions and 40 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

36
llvm/lib/Target/R600/SIISelLowering.cpp
View File

@ -62,7 +62,6 @@ SITargetLowering::SITargetLowering(TargetMachine &TM) :
MachineBasicBlock * SITargetLowering::EmitInstrWithCustomInserter( MachineBasicBlock * SITargetLowering::EmitInstrWithCustomInserter(
MachineInstr * MI, MachineBasicBlock * BB) const { MachineInstr * MI, MachineBasicBlock * BB) const {
const TargetInstrInfo * TII = getTargetMachine().getInstrInfo();
MachineRegisterInfo & MRI = BB->getParent()->getRegInfo(); MachineRegisterInfo & MRI = BB->getParent()->getRegInfo();
MachineBasicBlock::iterator I = MI; MachineBasicBlock::iterator I = MI;
@ -70,41 +69,6 @@ MachineBasicBlock * SITargetLowering::EmitInstrWithCustomInserter(
default: default:
return AMDGPUTargetLowering::EmitInstrWithCustomInserter(MI, BB); return AMDGPUTargetLowering::EmitInstrWithCustomInserter(MI, BB);
case AMDGPU::BRANCH: return BB; case AMDGPU::BRANCH: return BB;
case AMDGPU::CLAMP_SI:
BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::V_ADD_F32_e64))
.addOperand(MI->getOperand(0))
.addOperand(MI->getOperand(1))
.addImm(0x80) // SRC1
.addImm(0) // ABS
.addImm(1) // CLAMP
.addImm(0) // OMOD
.addImm(0); // NEG
MI->eraseFromParent();
break;
case AMDGPU::FABS_SI:
BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::V_ADD_F32_e64))
.addOperand(MI->getOperand(0))
.addOperand(MI->getOperand(1))
.addImm(0x80) // SRC1
.addImm(1) // ABS
.addImm(0) // CLAMP
.addImm(0) // OMOD
.addImm(0); // NEG
MI->eraseFromParent();
break;
case AMDGPU::FNEG_SI:
BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::V_ADD_F32_e64))
.addOperand(MI->getOperand(0))
.addOperand(MI->getOperand(1))
.addImm(0x80) // SRC1
.addImm(0) // ABS
.addImm(0) // CLAMP
.addImm(0) // OMOD
.addImm(1); // NEG
MI->eraseFromParent();
break;
case AMDGPU::SHADER_TYPE: case AMDGPU::SHADER_TYPE:
BB->getParent()->getInfo<SIMachineFunctionInfo>()->ShaderType = BB->getParent()->getInfo<SIMachineFunctionInfo>()->ShaderType =
MI->getOperand(0).getImm(); MI->getOperand(0).getImm();

26
llvm/lib/Target/R600/SIInstructions.td
View File

@ -1187,10 +1187,6 @@ defm : SamplePatterns<VReg_128, v4i32>;
defm : SamplePatterns<VReg_256, v8i32>; defm : SamplePatterns<VReg_256, v8i32>;
defm : SamplePatterns<VReg_512, v16i32>; defm : SamplePatterns<VReg_512, v16i32>;
def CLAMP_SI : CLAMP<VReg_32>;
def FABS_SI : FABS<VReg_32>;
def FNEG_SI : FNEG<VReg_32>;
def : Extract_Element <f32, v4f32, VReg_128, 0, sub0>; def : Extract_Element <f32, v4f32, VReg_128, 0, sub0>;
def : Extract_Element <f32, v4f32, VReg_128, 1, sub1>; def : Extract_Element <f32, v4f32, VReg_128, 1, sub1>;
def : Extract_Element <f32, v4f32, VReg_128, 2, sub2>; def : Extract_Element <f32, v4f32, VReg_128, 2, sub2>;
@ -1214,6 +1210,28 @@ def : BitConvert <i32, f32, VReg_32>;
def : BitConvert <f32, i32, SReg_32>; def : BitConvert <f32, i32, SReg_32>;
def : BitConvert <f32, i32, VReg_32>; def : BitConvert <f32, i32, VReg_32>;
/********** =================== **********/
/********** Src & Dst modifiers **********/
/********** =================== **********/
def : Pat <
(int_AMDIL_clamp VReg_32:$src, (f32 FP_ZERO), (f32 FP_ONE)),
(V_ADD_F32_e64 VReg_32:$src, (i32 0x80 /* SRC1 */),
0 /* ABS */, 1 /* CLAMP */, 0 /* OMOD */, 0 /* NEG */)
>;
def : Pat <
(fabs VReg_32:$src),
(V_ADD_F32_e64 VReg_32:$src, (i32 0x80 /* SRC1 */),
1 /* ABS */, 0 /* CLAMP */, 0 /* OMOD */, 0 /* NEG */)
>;
def : Pat <
(fneg VReg_32:$src),
(V_ADD_F32_e64 VReg_32:$src, (i32 0x80 /* SRC1 */),
0 /* ABS */, 0 /* CLAMP */, 0 /* OMOD */, 1 /* NEG */)
>;
/********** ================== **********/ /********** ================== **********/
/********** Immediate Patterns **********/ /********** Immediate Patterns **********/
/********** ================== **********/ /********** ================== **********/