Skip to content

Commit

Permalink
[X86][AVX] Fix handling of out-of-bounds shift amounts in AVX2 vector…
Browse files Browse the repository at this point in the history 
… logical shift nodes llvm#83840
  • Loading branch information
@SahilPatidar
SahilPatidar committed Apr 19, 2024
1 parent affcaf6 commit c11493d
Show file tree
Hide file tree
Showing 2 changed files with 145 additions and 0 deletions.
29 changes: 29 additions & 0 deletions llvm/lib/Target/X86/X86ISelLowering.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -45655,6 +45655,19 @@ static SDValue combineSelect(SDNode *N, SelectionDAG &DAG,
}
}

if (N->getOpcode() == ISD::VSELECT && LHS.getOpcode() == ISD::SRL &&
supportedVectorVarShift(VT, Subtarget, ISD::SRL)) {
APInt SV;
if (Cond.getOpcode() == ISD::SETCC && Cond.getOperand(0) == LHS.getOperand(1) &&
cast<CondCodeSDNode>(Cond.getOperand(2))->get() == ISD::SETULT &&
ISD::isConstantSplatVector(Cond.getOperand(1).getNode(), SV) &&
ISD::isConstantSplatVectorAllZeros(RHS.getNode()) &&
SV == VT.getScalarSizeInBits()) {
SDLoc DL(LHS);
return DAG.getNode(X86ISD::VSRLV, DL, LHS->getVTList(), LHS.getOperand(0), LHS.getOperand(1));
}
}

// Early exit check
if (!TLI.isTypeLegal(VT) || isSoftF16(VT, Subtarget))
return SDValue();
Expand Down Expand Up @@ -47536,6 +47549,22 @@ static SDValue combineShiftRightLogical(SDNode *N, SelectionDAG &DAG,
if (SDValue V = combineShiftToPMULH(N, DAG, Subtarget))
return V;

if (N0.getOpcode() == ISD::VSELECT &&
supportedVectorVarShift(VT, Subtarget, ISD::SRL)) {
SDValue Cond = N0.getOperand(0);
SDValue N00 = N0.getOperand(1);
SDValue N01 = N0.getOperand(2);
APInt SV;
if (Cond.getOpcode() == ISD::SETCC && Cond.getOperand(0) == N1 &&
cast<CondCodeSDNode>(Cond.getOperand(2))->get() == ISD::SETULT &&
ISD::isConstantSplatVector(Cond.getOperand(1).getNode(), SV) &&
ISD::isConstantSplatVectorAllZeros(N01.getNode()) &&
SV == VT.getScalarSizeInBits()) {
SDLoc DL(N);
return DAG.getNode(X86ISD::VSRLV, DL, N->getVTList(), N00, N1);
}
}

// Only do this on the last DAG combine as it can interfere with other
// combines.
if (!DCI.isAfterLegalizeDAG())
Expand Down
116 changes: 116 additions & 0 deletions llvm/test/CodeGen/X86/combine-srl.ll
View file
Original file line number Diff line number Diff line change
Expand Up @@ -606,3 +606,119 @@ define <4 x i32> @combine_vec_lshr_trunc_and(<4 x i32> %x, <4 x i64> %y) {
%3 = lshr <4 x i32> %x, %2
ret <4 x i32> %3
}

define <4 x i32> @combine_vec_lshr_clamped1(<4 x i32> %sh, <4 x i32> %amt) {
; SSE2-LABEL: combine_vec_lshr_clamped1:
; SSE2: # %bb.0:
; SSE2-NEXT: movdqa {{.*#+}} xmm2 = [2147483648,2147483648,2147483648,2147483648]
; SSE2-NEXT: pxor %xmm1, %xmm2
; SSE2-NEXT: pcmpgtd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm2
; SSE2-NEXT: pshuflw {{.*#+}} xmm3 = xmm1[2,3,3,3,4,5,6,7]
; SSE2-NEXT: movdqa %xmm0, %xmm4
; SSE2-NEXT: psrld %xmm3, %xmm4
; SSE2-NEXT: pshuflw {{.*#+}} xmm3 = xmm1[0,1,1,1,4,5,6,7]
; SSE2-NEXT: movdqa %xmm0, %xmm5
; SSE2-NEXT: psrld %xmm3, %xmm5
; SSE2-NEXT: punpcklqdq {{.*#+}} xmm5 = xmm5[0],xmm4[0]
; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm1[2,3,2,3]
; SSE2-NEXT: pshuflw {{.*#+}} xmm3 = xmm1[2,3,3,3,4,5,6,7]
; SSE2-NEXT: movdqa %xmm0, %xmm4
; SSE2-NEXT: psrld %xmm3, %xmm4
; SSE2-NEXT: pshuflw {{.*#+}} xmm1 = xmm1[0,1,1,1,4,5,6,7]
; SSE2-NEXT: psrld %xmm1, %xmm0
; SSE2-NEXT: punpckhqdq {{.*#+}} xmm0 = xmm0[1],xmm4[1]
; SSE2-NEXT: shufps {{.*#+}} xmm5 = xmm5[0,3],xmm0[0,3]
; SSE2-NEXT: pandn %xmm5, %xmm2
; SSE2-NEXT: movdqa %xmm2, %xmm0
; SSE2-NEXT: retq
;
; SSE41-LABEL: combine_vec_lshr_clamped1:
; SSE41: # %bb.0:
; SSE41-NEXT: pmovsxbd {{.*#+}} xmm2 = [31,31,31,31]
; SSE41-NEXT: pminud %xmm1, %xmm2
; SSE41-NEXT: pcmpeqd %xmm1, %xmm2
; SSE41-NEXT: pshuflw {{.*#+}} xmm3 = xmm1[2,3,3,3,4,5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm4
; SSE41-NEXT: psrld %xmm3, %xmm4
; SSE41-NEXT: pshufd {{.*#+}} xmm3 = xmm1[2,3,2,3]
; SSE41-NEXT: pshuflw {{.*#+}} xmm5 = xmm3[2,3,3,3,4,5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm6
; SSE41-NEXT: psrld %xmm5, %xmm6
; SSE41-NEXT: pblendw {{.*#+}} xmm6 = xmm4[0,1,2,3],xmm6[4,5,6,7]
; SSE41-NEXT: pshuflw {{.*#+}} xmm1 = xmm1[0,1,1,1,4,5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm4
; SSE41-NEXT: psrld %xmm1, %xmm4
; SSE41-NEXT: pshuflw {{.*#+}} xmm1 = xmm3[0,1,1,1,4,5,6,7]
; SSE41-NEXT: psrld %xmm1, %xmm0
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm4[0,1,2,3],xmm0[4,5,6,7]
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm0[0,1],xmm6[2,3],xmm0[4,5],xmm6[6,7]
; SSE41-NEXT: pand %xmm2, %xmm0
; SSE41-NEXT: retq
;
; AVX-LABEL: combine_vec_lshr_clamped1:
; AVX: # %bb.0:
; AVX-NEXT: vpsrlvd %xmm1, %xmm0, %xmm0
; AVX-NEXT: retq
%cmp.i = icmp ult <4 x i32> %amt, <i32 32, i32 32, i32 32, i32 32>
%shr = lshr <4 x i32> %sh, %amt
%1 = select <4 x i1> %cmp.i, <4 x i32> %shr, <4 x i32> zeroinitializer
ret <4 x i32> %1
}

define <4 x i32> @combine_vec_lshr_clamped2(<4 x i32> %sh, <4 x i32> %amt) {
; SSE2-LABEL: combine_vec_lshr_clamped2:
; SSE2: # %bb.0:
; SSE2-NEXT: movdqa {{.*#+}} xmm2 = [2147483648,2147483648,2147483648,2147483648]
; SSE2-NEXT: pxor %xmm1, %xmm2
; SSE2-NEXT: pcmpgtd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm2
; SSE2-NEXT: pandn %xmm0, %xmm2
; SSE2-NEXT: pshufd {{.*#+}} xmm0 = xmm1[2,3,2,3]
; SSE2-NEXT: pshuflw {{.*#+}} xmm3 = xmm0[2,3,3,3,4,5,6,7]
; SSE2-NEXT: movdqa %xmm2, %xmm4
; SSE2-NEXT: psrld %xmm3, %xmm4
; SSE2-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[0,1,1,1,4,5,6,7]
; SSE2-NEXT: movdqa %xmm2, %xmm3
; SSE2-NEXT: psrld %xmm0, %xmm3
; SSE2-NEXT: punpckhqdq {{.*#+}} xmm3 = xmm3[1],xmm4[1]
; SSE2-NEXT: pshuflw {{.*#+}} xmm0 = xmm1[2,3,3,3,4,5,6,7]
; SSE2-NEXT: movdqa %xmm2, %xmm4
; SSE2-NEXT: psrld %xmm0, %xmm4
; SSE2-NEXT: pshuflw {{.*#+}} xmm0 = xmm1[0,1,1,1,4,5,6,7]
; SSE2-NEXT: psrld %xmm0, %xmm2
; SSE2-NEXT: punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm4[0]
; SSE2-NEXT: shufps {{.*#+}} xmm2 = xmm2[0,3],xmm3[0,3]
; SSE2-NEXT: movaps %xmm2, %xmm0
; SSE2-NEXT: retq
;
; SSE41-LABEL: combine_vec_lshr_clamped2:
; SSE41: # %bb.0:
; SSE41-NEXT: pmovsxbd {{.*#+}} xmm2 = [31,31,31,31]
; SSE41-NEXT: pminud %xmm1, %xmm2
; SSE41-NEXT: pcmpeqd %xmm1, %xmm2
; SSE41-NEXT: pand %xmm2, %xmm0
; SSE41-NEXT: pshufd {{.*#+}} xmm2 = xmm1[2,3,2,3]
; SSE41-NEXT: pshuflw {{.*#+}} xmm3 = xmm2[2,3,3,3,4,5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm4
; SSE41-NEXT: psrld %xmm3, %xmm4
; SSE41-NEXT: pshuflw {{.*#+}} xmm3 = xmm1[2,3,3,3,4,5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm5
; SSE41-NEXT: psrld %xmm3, %xmm5
; SSE41-NEXT: pblendw {{.*#+}} xmm5 = xmm5[0,1,2,3],xmm4[4,5,6,7]
; SSE41-NEXT: pshuflw {{.*#+}} xmm2 = xmm2[0,1,1,1,4,5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm3
; SSE41-NEXT: psrld %xmm2, %xmm3
; SSE41-NEXT: pshuflw {{.*#+}} xmm1 = xmm1[0,1,1,1,4,5,6,7]
; SSE41-NEXT: psrld %xmm1, %xmm0
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm3[4,5,6,7]
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm0[0,1],xmm5[2,3],xmm0[4,5],xmm5[6,7]
; SSE41-NEXT: retq
;
; AVX-LABEL: combine_vec_lshr_clamped2:
; AVX: # %bb.0:
; AVX-NEXT: vpsrlvd %xmm1, %xmm0, %xmm0
; AVX-NEXT: retq
%cmp.i = icmp ult <4 x i32> %amt, <i32 32, i32 32, i32 32, i32 32>
%1 = select <4 x i1> %cmp.i, <4 x i32> %sh, <4 x i32> zeroinitializer
%shr = lshr <4 x i32> %1, %amt
ret <4 x i32> %shr
}

0 comments on commit c11493d

Please sign in to comment.