llvm.org GIT mirror llvm / efa4fe4
Merging r232386: ------------------------------------------------------------------------ r232386 | thomas.stellard | 2015-03-16 11:53:55 -0400 (Mon, 16 Mar 2015) | 8 lines R600/SI: don't try min3/max3/med3 with f64 There are no opcodes for this. This also adds a test case. v2: make test more robust Patch by: Grigori Goronzy ------------------------------------------------------------------------ git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_36@236040 91177308-0d34-0410-b5e6-96231b3b80d8 Tom Stellard 5 years ago
2 changed file(s) with 25 addition(s) and 0 deletion(s). Raw diff Collapse all Expand all
15831583 case AMDGPUISD::UMAX:
15841584 case AMDGPUISD::UMIN: {
15851585 if (DCI.getDAGCombineLevel() >= AfterLegalizeDAG &&
1586 N->getValueType(0) != MVT::f64 &&
15861587 getTargetMachine().getOptLevel() > CodeGenOpt::None)
15871588 return performMin3Max3Combine(N, DCI);
15881589 break;
0 ; RUN: llc -march=amdgcn -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
1 ; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
2
3 declare double @llvm.maxnum.f64(double, double) nounwind readnone
4
5 ; SI-LABEL: {{^}}test_fmax3_f64:
6 ; SI-DAG: buffer_load_dwordx2 [[REGA:v\[[0-9]+:[0-9]+\]]], s[{{[0-9]+:[0-9]+}}], 0{{$}}
7 ; SI-DAG: buffer_load_dwordx2 [[REGB:v\[[0-9]+:[0-9]+\]]], s[{{[0-9]+:[0-9]+}}], 0 offset:8
8 ; SI-DAG: buffer_load_dwordx2 [[REGC:v\[[0-9]+:[0-9]+\]]], s[{{[0-9]+:[0-9]+}}], 0 offset:16
9 ; SI: v_max_f64 [[REGA]], [[REGA]], [[REGB]]
10 ; SI: v_max_f64 [[RESULT:v\[[0-9]+:[0-9]+\]]], [[REGA]], [[REGC]]
11 ; SI: buffer_store_dwordx2 [[RESULT]],
12 ; SI: s_endpgm
13 define void @test_fmax3_f64(double addrspace(1)* %out, double addrspace(1)* %aptr) nounwind {
14 %bptr = getelementptr double addrspace(1)* %aptr, i32 1
15 %cptr = getelementptr double addrspace(1)* %aptr, i32 2
16 %a = load double addrspace(1)* %aptr, align 8
17 %b = load double addrspace(1)* %bptr, align 8
18 %c = load double addrspace(1)* %cptr, align 8
19 %f0 = call double @llvm.maxnum.f64(double %a, double %b) nounwind readnone
20 %f1 = call double @llvm.maxnum.f64(double %f0, double %c) nounwind readnone
21 store double %f1, double addrspace(1)* %out, align 8
22 ret void
23 }