RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-25 04:02:41 +01:00
Code Issues Projects Releases Wiki Activity

R600: Expand vector flog2

Browse Source 
llvm-svn: 211376
This commit is contained in:
Tom Stellard 2014-06-20 17:06:07 +00:00
parent f2ad4b6bb8
commit 0c64d4a322
2 changed files with 80 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
lib/Target/R600/AMDGPUISelLowering.cpp
View File

@ -321,6 +321,7 @@ AMDGPUTargetLowering::AMDGPUTargetLowering(TargetMachine &TM) :
setOperationAction(ISD::FCOS, VT, Expand); setOperationAction(ISD::FCOS, VT, Expand);
setOperationAction(ISD::FDIV, VT, Expand); setOperationAction(ISD::FDIV, VT, Expand);
setOperationAction(ISD::FEXP2, VT, Expand); setOperationAction(ISD::FEXP2, VT, Expand);
setOperationAction(ISD::FLOG2, VT, Expand);
setOperationAction(ISD::FPOW, VT, Expand); setOperationAction(ISD::FPOW, VT, Expand);
setOperationAction(ISD::FFLOOR, VT, Expand); setOperationAction(ISD::FFLOOR, VT, Expand);
setOperationAction(ISD::FTRUNC, VT, Expand); setOperationAction(ISD::FTRUNC, VT, Expand);

79
test/CodeGen/R600/llvm.log2.ll Normal file
View File

@ -0,0 +1,79 @@
;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s --check-prefix=EG-CHECK --check-prefix=FUNC
;RUN: llc < %s -march=r600 -mcpu=cayman | FileCheck %s --check-prefix=CM-CHECK --check-prefix=FUNC
;RUN: llc < %s -march=r600 -mcpu=SI | FileCheck %s --check-prefix=SI-CHECK --check-prefix=FUNC
;FUNC-LABEL: @test
;EG-CHECK: LOG_IEEE
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}}
;SI-CHECK: V_LOG_F32
define void @test(float addrspace(1)* %out, float %in) {
entry:
%0 = call float @llvm.log2.f32(float %in)
store float %0, float addrspace(1)* %out
ret void
}
;FUNC-LABEL: @testv2
;EG-CHECK: LOG_IEEE
;EG-CHECK: LOG_IEEE
; FIXME: We should be able to merge these packets together on Cayman so we
; have a maximum of 4 instructions.
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}}
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}}
;SI-CHECK: V_LOG_F32
;SI-CHECK: V_LOG_F32
define void @testv2(<2 x float> addrspace(1)* %out, <2 x float> %in) {
entry:
%0 = call <2 x float> @llvm.log2.v2f32(<2 x float> %in)
store <2 x float> %0, <2 x float> addrspace(1)* %out
ret void
}
;FUNC-LABEL: @testv4
;EG-CHECK: LOG_IEEE
;EG-CHECK: LOG_IEEE
;EG-CHECK: LOG_IEEE
;EG-CHECK: LOG_IEEE
; FIXME: We should be able to merge these packets together on Cayman so we
; have a maximum of 4 instructions.
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}}
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}}
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}}
;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}}
;SI-CHECK: V_LOG_F32
;SI-CHECK: V_LOG_F32
;SI-CHECK: V_LOG_F32
;SI-CHECK: V_LOG_F32
define void @testv4(<4 x float> addrspace(1)* %out, <4 x float> %in) {
entry:
%0 = call <4 x float> @llvm.log2.v4f32(<4 x float> %in)
store <4 x float> %0, <4 x float> addrspace(1)* %out
ret void
}
declare float @llvm.log2.f32(float) readnone
declare <2 x float> @llvm.log2.v2f32(<2 x float>) readnone
declare <4 x float> @llvm.log2.v4f32(<4 x float>) readnone