mirror of
https://github.com/RPCS3/llvm.git
synced 2024-12-25 13:35:10 +00:00
Remove the type legality check from the SelectionDAGBuilder when it lowers @llvm.fmuladd to ISD::FMA nodes.
Performing this check unilaterally prevented us from generating FMAs when the incoming IR contained illegal vector types which would eventually be legalized to underlying types that *did* support FMA. For example, an @llvm.fmuladd on an OpenCL float16 should become a sequence of float4 FMAs, not float4 fmul+fadd's. NOTE: Because we still call the target-specific profitability hook, individual targets can reinstate the old behavior, if desired, by simply performing the legality check inside their callback hook. They can also perform more sophisticated legality checks, if, for example, some illegal vector types can be productively implemented as FMAs, but not others. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177820 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
parent
31881c235f
commit
7834c71433
@ -4914,7 +4914,6 @@ SelectionDAGBuilder::visitIntrinsicCall(const CallInst &I, unsigned Intrinsic) {
|
||||
case Intrinsic::fmuladd: {
|
||||
EVT VT = TLI.getValueType(I.getType());
|
||||
if (TM.Options.AllowFPOpFusion != FPOpFusion::Strict &&
|
||||
TLI.isOperationLegalOrCustom(ISD::FMA, VT) &&
|
||||
TLI.isFMAFasterThanMulAndAdd(VT)){
|
||||
setValue(&I, DAG.getNode(ISD::FMA, dl,
|
||||
getValue(I.getArgOperand(0)).getValueType(),
|
||||
|
20
test/CodeGen/X86/wide-fma-contraction.ll
Normal file
20
test/CodeGen/X86/wide-fma-contraction.ll
Normal file
@ -0,0 +1,20 @@
|
||||
; RUN: llc -march=x86 -mattr=+fma4 -mtriple=x86_64-apple-darwin < %s | FileCheck %s
|
||||
|
||||
; CHECK: fmafunc
|
||||
define <16 x float> @fmafunc(<16 x float> %a, <16 x float> %b, <16 x float> %c) {
|
||||
; CHECK-NOT: vmulps
|
||||
; CHECK-NOT: vaddps
|
||||
; CHECK: vfmaddps
|
||||
; CHECK-NOT: vmulps
|
||||
; CHECK-NOT: vaddps
|
||||
; CHECK: vfmaddps
|
||||
; CHECK-NOT: vmulps
|
||||
; CHECK-NOT: vaddps
|
||||
%ret = tail call <16 x float> @llvm.fmuladd.v16f32(<16 x float> %a, <16 x float> %b, <16 x float> %c)
|
||||
ret <16 x float> %ret
|
||||
}
|
||||
|
||||
declare <16 x float> @llvm.fmuladd.v16f32(<16 x float>, <16 x float>, <16 x float>) nounwind readnone
|
||||
|
||||
|
||||
|
Loading…
Reference in New Issue
Block a user