RPCSX/llvm
1
0
Fork 0
mirror of https://github.com/RPCSX/llvm.git synced 2025-02-26 16:05:54 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

[X86][SSE] Ensure BLENDPD/BLENDPS/PBLEND inputs are both of the correct input type

Browse Source 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@256782 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Simon Pilgrim 2016-01-04 21:41:11 +00:00
parent 8abd63fcbb
commit 472704020c
2 changed files with 32 additions and 55 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
lib/Target/X86/X86ISelLowering.cpp
View File

@ -4557,6 +4557,7 @@ static SDValue Insert128BitVector(SDValue Result, SDValue Vec, unsigned IdxVal,
MVT CastVT = Subtarget.hasAVX2() ? MVT::v8i32 : MVT::v8f32;
SDValue Mask = DAG.getConstant(0x0f, dl, MVT::i8);
Result = DAG.getBitcast(CastVT, Result);
Vec256 = DAG.getBitcast(CastVT, Vec256);
Vec256 = DAG.getNode(X86ISD::BLENDI, dl, CastVT, Result, Vec256, Mask);
return DAG.getBitcast(ResultVT, Vec256);
@ -27467,6 +27468,8 @@ static SDValue PerformBLENDICombine(SDNode *N, SelectionDAG &DAG) {
SDValue V1 = N->getOperand(1);
SDLoc DL(N);
EVT VT = N->getValueType(0);
assert(VT == V0.getValueType() && VT == V1.getValueType() &&
"Unexpected input vector types");
// Canonicalize a v2f64 blend with a mask of 2 by swapping the vector
// operands and changing the mask to 1. This saves us a bunch of

84
test/CodeGen/X86/avx-cast.ll
View File

@ -1,38 +1,27 @@
; RUN: llc < %s -mtriple=x86_64-apple-darwin -mattr=+avx | FileCheck %s --check-prefix=AVX1
; RUN: llc < %s -mtriple=x86_64-apple-darwin -mattr=+avx2 | FileCheck %s --check-prefix=AVX2
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=x86_64-apple-darwin -mattr=+avx | FileCheck %s --check-prefix=AVX --check-prefix=AVX1
; RUN: llc < %s -mtriple=x86_64-apple-darwin -mattr=+avx2 | FileCheck %s --check-prefix=AVX --check-prefix=AVX2
; Prefer a blend instruction to a vinsert128 instruction because blends
; are simpler (no lane changes) and therefore will have equal or better
; performance.
define <8 x float> @castA(<4 x float> %m) nounwind uwtable readnone ssp {
; AVX1-LABEL: castA:
; AVX1: vxorps %ymm1, %ymm1, %ymm1
; AVX1-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
; AVX1-NEXT: retq
;
; AVX2-LABEL: castA:
; AVX2: vxorps %ymm1, %ymm1, %ymm1
; AVX2-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
; AVX2-NEXT: retq
entry:
; AVX-LABEL: castA:
; AVX: ## BB#0:
; AVX-NEXT: vxorps %ymm1, %ymm1, %ymm1
; AVX-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
; AVX-NEXT: retq
%shuffle.i = shufflevector <4 x float> %m, <4 x float> zeroinitializer, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 4, i32 4, i32 4>
ret <8 x float> %shuffle.i
}
define <4 x double> @castB(<2 x double> %m) nounwind uwtable readnone ssp {
; AVX1-LABEL: castB:
; AVX1: vxorpd %ymm1, %ymm1, %ymm1
; AVX1-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
; AVX1-NEXT: retq
;
; AVX2-LABEL: castB:
; AVX2: vxorpd %ymm1, %ymm1, %ymm1
; AVX2-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
; AVX2-NEXT: retq
entry:
; AVX-LABEL: castB:
; AVX: ## BB#0:
; AVX-NEXT: vxorpd %ymm1, %ymm1, %ymm1
; AVX-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
; AVX-NEXT: retq
%shuffle.i = shufflevector <2 x double> %m, <2 x double> zeroinitializer, <4 x i32> <i32 0, i32 1, i32 2, i32 2>
ret <4 x double> %shuffle.i
}
@ -41,16 +30,16 @@ entry:
define <4 x i64> @castC(<2 x i64> %m) nounwind uwtable readnone ssp {
; AVX1-LABEL: castC:
; AVX1: vxorps %xmm1, %xmm1, %xmm1
; AVX1: ## BB#0:
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
; AVX1-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
; AVX1-NEXT: retq
;
; AVX2-LABEL: castC:
; AVX2: vpxor %ymm1, %ymm1, %ymm1
; AVX2: ## BB#0:
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
; AVX2-NEXT: retq
entry:
%shuffle.i = shufflevector <2 x i64> %m, <2 x i64> zeroinitializer, <4 x i32> <i32 0, i32 1, i32 2, i32 2>
ret <4 x i64> %shuffle.i
}
@ -59,43 +48,28 @@ entry:
; vzeroupper before the return, so just check for the absence of shuffles.
define <4 x float> @castD(<8 x float> %m) nounwind uwtable readnone ssp {
; AVX1-LABEL: castD:
; AVX1-NOT: extract
; AVX1-NOT: blend
;
; AVX2-LABEL: castD:
; AVX2-NOT: extract
; AVX2-NOT: blend
entry:
; AVX-LABEL: castD:
; AVX: ## BB#0:
; AVX-NEXT: vzeroupper
; AVX-NEXT: retq
%shuffle.i = shufflevector <8 x float> %m, <8 x float> %m, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
ret <4 x float> %shuffle.i
}
define <2 x i64> @castE(<4 x i64> %m) nounwind uwtable readnone ssp {
; AVX1-LABEL: castE:
; AVX1-NOT: extract
; AVX1-NOT: blend
;
; AVX2-LABEL: castE:
; AVX2-NOT: extract
; AVX2-NOT: blend
entry:
; AVX-LABEL: castE:
; AVX: ## BB#0:
; AVX-NEXT: vzeroupper
; AVX-NEXT: retq
%shuffle.i = shufflevector <4 x i64> %m, <4 x i64> %m, <2 x i32> <i32 0, i32 1>
ret <2 x i64> %shuffle.i
}
define <2 x double> @castF(<4 x double> %m) nounwind uwtable readnone ssp {
; AVX1-LABEL: castF:
; AVX1-NOT: extract
; AVX1-NOT: blend
;
; AVX2-LABEL: castF:
; AVX2-NOT: extract
; AVX2-NOT: blend
entry:
; AVX-LABEL: castF:
; AVX: ## BB#0:
; AVX-NEXT: vzeroupper
; AVX-NEXT: retq
%shuffle.i = shufflevector <4 x double> %m, <4 x double> %m, <2 x i32> <i32 0, i32 1>
ret <2 x double> %shuffle.i
}