mirror of
https://github.com/RPCSX/llvm.git
synced 2024-12-13 14:46:53 +00:00
e874345be4
If pseudoToMCOpcode failed, we would return the original opcode, so operands would be swapped, but the instruction would remain the same. It resulted in LSHLREV a, b ---> LSHLREV b, a. This fixes Glamor text rendering and piglit/arb_sample_shading-builtin-gl-sample-mask on VI. This is a candidate for stable branches. v2: the test was simplified by Tom Stellard git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@240824 91177308-0d34-0410-b5e6-96231b3b80d8
34 lines
1.3 KiB
LLVM
34 lines
1.3 KiB
LLVM
; RUN: llc -march=amdgcn -mcpu=verde -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=SI %s
|
|
; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=VI %s
|
|
|
|
; GCN-LABEL: {{^}}main:
|
|
; SI: v_lshl_b32_e32 v{{[0-9]+}}, 1, v{{[0-9]+}}
|
|
; VI: v_lshlrev_b32_e64 v{{[0-9]+}}, v{{[0-9]+}}, 1
|
|
|
|
define void @main() #0 {
|
|
main_body:
|
|
%0 = fptosi float undef to i32
|
|
%1 = call <4 x i32> @llvm.SI.imageload.v4i32(<4 x i32> undef, <32 x i8> undef, i32 2)
|
|
%2 = extractelement <4 x i32> %1, i32 0
|
|
%3 = and i32 %0, 7
|
|
%4 = shl i32 1, %3
|
|
%5 = and i32 %2, %4
|
|
%6 = icmp eq i32 %5, 0
|
|
%.10 = select i1 %6, float 0.000000e+00, float undef
|
|
%7 = call i32 @llvm.SI.packf16(float undef, float %.10)
|
|
%8 = bitcast i32 %7 to float
|
|
call void @llvm.SI.export(i32 15, i32 1, i32 1, i32 0, i32 1, float undef, float %8, float undef, float %8)
|
|
ret void
|
|
}
|
|
|
|
; Function Attrs: nounwind readnone
|
|
declare <4 x i32> @llvm.SI.imageload.v4i32(<4 x i32>, <32 x i8>, i32) #1
|
|
|
|
; Function Attrs: nounwind readnone
|
|
declare i32 @llvm.SI.packf16(float, float) #1
|
|
|
|
declare void @llvm.SI.export(i32, i32, i32, i32, i32, float, float, float, float)
|
|
|
|
attributes #0 = { "ShaderType"="0" "enable-no-nans-fp-math"="true" }
|
|
attributes #1 = { nounwind readnone }
|