forked from OSchip/llvm-project
[ARM] Combines for VMOVN
This adds two combines for VMOVN, one to fold VMOVN[tb](c, VQMOVNb(a, b)) => VQMOVN[tb](c, b) The other to perform demand bits analysis on the lanes of a VMOVN. We know that only the bottom lanes of the second operand and the top or bottom lanes of the Qd operand are needed in the result, depending on if the VMOVN is bottom or top. Differential Revision: https://reviews.llvm.org/D77718
This commit is contained in:
David Green
parent
2e1fbf85b6
commit
72f1fb2edf
|
|
@ -14421,6 +14421,41 @@ static SDValue PerformVECREDUCE_ADDCombine(SDNode *N, SelectionDAG &DAG,
|
|||
return SDValue();
|
||||
}
|
||||
|
||||
static SDValue PerformVMOVNCombine(SDNode *N,
|
||||
TargetLowering::DAGCombinerInfo &DCI) {
|
||||
SDValue Op0 = N->getOperand(0);
|
||||
SDValue Op1 = N->getOperand(1);
|
||||
unsigned IsTop = N->getConstantOperandVal(2);
|
||||
|
||||
// VMOVNt(c, VQMOVNb(a, b)) => VQMOVNt(c, b)
|
||||
// VMOVNb(c, VQMOVNb(a, b)) => VQMOVNb(c, b)
|
||||
if ((Op1->getOpcode() == ARMISD::VQMOVNs ||
|
||||
Op1->getOpcode() == ARMISD::VQMOVNu) &&
|
||||
Op1->getConstantOperandVal(2) == 0)
|
||||
return DCI.DAG.getNode(Op1->getOpcode(), SDLoc(Op1), N->getValueType(0),
|
||||
Op0, Op1->getOperand(1), N->getOperand(2));
|
||||
|
||||
// Only the bottom lanes from Qm (Op1) and either the top or bottom lanes from
|
||||
// Qd (Op0) are demanded from a VMOVN, depending on whether we are inserting
|
||||
// into the top or bottom lanes.
|
||||
unsigned NumElts = N->getValueType(0).getVectorNumElements();
|
||||
APInt Op1DemandedElts = APInt::getSplat(NumElts, APInt::getLowBitsSet(2, 1));
|
||||
APInt Op0DemandedElts =
|
||||
IsTop ? Op1DemandedElts
|
||||
: APInt::getSplat(NumElts, APInt::getHighBitsSet(2, 1));
|
||||
|
||||
APInt KnownUndef, KnownZero;
|
||||
const TargetLowering &TLI = DCI.DAG.getTargetLoweringInfo();
|
||||
if (TLI.SimplifyDemandedVectorElts(Op0, Op0DemandedElts, KnownUndef,
|
||||
KnownZero, DCI))
|
||||
return SDValue(N, 0);
|
||||
if (TLI.SimplifyDemandedVectorElts(Op1, Op1DemandedElts, KnownUndef,
|
||||
KnownZero, DCI))
|
||||
return SDValue(N, 0);
|
||||
|
||||
return SDValue();
|
||||
}
|
||||
|
||||
static SDValue PerformLongShiftCombine(SDNode *N, SelectionDAG &DAG) {
|
||||
SDLoc DL(N);
|
||||
SDValue Op0 = N->getOperand(0);
|
||||
|
|
@ -15555,6 +15590,8 @@ SDValue ARMTargetLowering::PerformDAGCombine(SDNode *N,
|
|||
return PerformVCMPCombine(N, DCI, Subtarget);
|
||||
case ISD::VECREDUCE_ADD:
|
||||
return PerformVECREDUCE_ADDCombine(N, DCI.DAG, Subtarget);
|
||||
case ARMISD::VMOVN:
|
||||
return PerformVMOVNCombine(N, DCI);
|
||||
case ARMISD::ASRL:
|
||||
case ARMISD::LSRL:
|
||||
case ARMISD::LSLL:
|
||||
|
|
|
|||
|
|
@ -1418,13 +1418,11 @@ define arm_aapcs_vfpcc void @ssatmul_8i_q15(i16* nocapture readonly %pSrcA, i16*
|
|||
; CHECK-NEXT: vldrh.u16 q1, [r1], #16
|
||||
; CHECK-NEXT: vmullt.s16 q2, q1, q0
|
||||
; CHECK-NEXT: vmullb.s16 q0, q1, q0
|
||||
; CHECK-NEXT: vshr.s32 q2, q2, #15
|
||||
; CHECK-NEXT: vshr.s32 q0, q0, #15
|
||||
; CHECK-NEXT: vqmovnb.s32 q2, q2
|
||||
; CHECK-NEXT: vshr.s32 q2, q2, #15
|
||||
; CHECK-NEXT: vqmovnb.s32 q0, q0
|
||||
; CHECK-NEXT: vmovlb.s16 q2, q2
|
||||
; CHECK-NEXT: vmovlb.s16 q0, q0
|
||||
; CHECK-NEXT: vmovnt.i32 q0, q2
|
||||
; CHECK-NEXT: vqmovnt.s32 q0, q2
|
||||
; CHECK-NEXT: vstrb.8 q0, [r2], #16
|
||||
; CHECK-NEXT: le lr, .LBB7_4
|
||||
; CHECK-NEXT: @ %bb.5: @ %middle.block
|
||||
|
|
@ -1867,11 +1865,9 @@ define arm_aapcs_vfpcc void @ssatmul_8ti_q15(i16* nocapture readonly %pSrcA, i16
|
|||
; CHECK-NEXT: vmullb.s16 q5, q6, q5
|
||||
; CHECK-NEXT: vshr.s32 q7, q7, #15
|
||||
; CHECK-NEXT: vshr.s32 q5, q5, #15
|
||||
; CHECK-NEXT: vqmovnb.s32 q7, q7
|
||||
; CHECK-NEXT: vqmovnb.s32 q5, q5
|
||||
; CHECK-NEXT: vmovlb.s16 q7, q7
|
||||
; CHECK-NEXT: vmovlb.s16 q5, q5
|
||||
; CHECK-NEXT: vmovnt.i32 q5, q7
|
||||
; CHECK-NEXT: vqmovnt.s32 q5, q7
|
||||
; CHECK-NEXT: vpst
|
||||
; CHECK-NEXT: vstrht.16 q5, [r2], #16
|
||||
; CHECK-NEXT: le lr, .LBB10_2
|
||||
|
|
@ -2664,13 +2660,11 @@ define arm_aapcs_vfpcc void @ssatmul_16i_q7(i8* nocapture readonly %pSrcA, i8* n
|
|||
; CHECK-NEXT: vldrb.u8 q1, [r1], #16
|
||||
; CHECK-NEXT: vmullt.s8 q2, q1, q0
|
||||
; CHECK-NEXT: vmullb.s8 q0, q1, q0
|
||||
; CHECK-NEXT: vshr.s16 q2, q2, #7
|
||||
; CHECK-NEXT: vshr.s16 q0, q0, #7
|
||||
; CHECK-NEXT: vqmovnb.s16 q2, q2
|
||||
; CHECK-NEXT: vshr.s16 q2, q2, #7
|
||||
; CHECK-NEXT: vqmovnb.s16 q0, q0
|
||||
; CHECK-NEXT: vmovlb.s8 q2, q2
|
||||
; CHECK-NEXT: vmovlb.s8 q0, q0
|
||||
; CHECK-NEXT: vmovnt.i16 q0, q2
|
||||
; CHECK-NEXT: vqmovnt.s16 q0, q2
|
||||
; CHECK-NEXT: vstrb.8 q0, [r2], #16
|
||||
; CHECK-NEXT: le lr, .LBB16_4
|
||||
; CHECK-NEXT: @ %bb.5: @ %middle.block
|
||||
|
|
@ -3336,13 +3330,11 @@ define arm_aapcs_vfpcc void @ssatmul_16ti_q7(i8* nocapture readonly %pSrcA, i8*
|
|||
; CHECK-NEXT: vldrbt.u8 q4, [r1], #16
|
||||
; CHECK-NEXT: vmullt.s8 q5, q4, q0
|
||||
; CHECK-NEXT: vmullb.s8 q0, q4, q0
|
||||
; CHECK-NEXT: vshr.s16 q5, q5, #7
|
||||
; CHECK-NEXT: vshr.s16 q0, q0, #7
|
||||
; CHECK-NEXT: vqmovnb.s16 q5, q5
|
||||
; CHECK-NEXT: vshr.s16 q5, q5, #7
|
||||
; CHECK-NEXT: vqmovnb.s16 q0, q0
|
||||
; CHECK-NEXT: vmovlb.s8 q5, q5
|
||||
; CHECK-NEXT: vmovlb.s8 q0, q0
|
||||
; CHECK-NEXT: vmovnt.i16 q0, q5
|
||||
; CHECK-NEXT: vqmovnt.s16 q0, q5
|
||||
; CHECK-NEXT: vpst
|
||||
; CHECK-NEXT: vstrbt.8 q0, [r2], #16
|
||||
; CHECK-NEXT: le lr, .LBB19_2
|
||||
|
|
|
|||
|
|
@ -4,9 +4,7 @@
|
|||
define arm_aapcs_vfpcc <8 x i16> @vqmovni32_sminmax_t1(<4 x i32> %s0, <8 x i16> %src1) {
|
||||
; CHECK-LABEL: vqmovni32_sminmax_t1:
|
||||
; CHECK: @ %bb.0: @ %entry
|
||||
; CHECK-NEXT: vqmovnb.s32 q0, q0
|
||||
; CHECK-NEXT: vmovlb.s16 q0, q0
|
||||
; CHECK-NEXT: vmovnt.i32 q1, q0
|
||||
; CHECK-NEXT: vqmovnt.s32 q1, q0
|
||||
; CHECK-NEXT: vmov q0, q1
|
||||
; CHECK-NEXT: bx lr
|
||||
entry:
|
||||
|
|
@ -23,7 +21,6 @@ define arm_aapcs_vfpcc <8 x i16> @vqmovni32_sminmax_t2(<4 x i32> %s0, <8 x i16>
|
|||
; CHECK-LABEL: vqmovni32_sminmax_t2:
|
||||
; CHECK: @ %bb.0: @ %entry
|
||||
; CHECK-NEXT: vqmovnb.s32 q0, q0
|
||||
; CHECK-NEXT: vmovlb.s16 q0, q0
|
||||
; CHECK-NEXT: vmovnt.i32 q0, q1
|
||||
; CHECK-NEXT: bx lr
|
||||
entry:
|
||||
|
|
@ -39,9 +36,7 @@ entry:
|
|||
define arm_aapcs_vfpcc <8 x i16> @vqmovni32_sminmax_b1(<4 x i32> %s0, <8 x i16> %src1) {
|
||||
; CHECK-LABEL: vqmovni32_sminmax_b1:
|
||||
; CHECK: @ %bb.0: @ %entry
|
||||
; CHECK-NEXT: vqmovnb.s32 q0, q0
|
||||
; CHECK-NEXT: vmovlb.s16 q0, q0
|
||||
; CHECK-NEXT: vmovnb.i32 q1, q0
|
||||
; CHECK-NEXT: vqmovnb.s32 q1, q0
|
||||
; CHECK-NEXT: vmov q0, q1
|
||||
; CHECK-NEXT: bx lr
|
||||
entry:
|
||||
|
|
@ -137,9 +132,7 @@ entry:
|
|||
define arm_aapcs_vfpcc <16 x i8> @vqmovni16_sminmax_t1(<8 x i16> %s0, <16 x i8> %src1) {
|
||||
; CHECK-LABEL: vqmovni16_sminmax_t1:
|
||||
; CHECK: @ %bb.0: @ %entry
|
||||
; CHECK-NEXT: vqmovnb.s16 q0, q0
|
||||
; CHECK-NEXT: vmovlb.s8 q0, q0
|
||||
; CHECK-NEXT: vmovnt.i16 q1, q0
|
||||
; CHECK-NEXT: vqmovnt.s16 q1, q0
|
||||
; CHECK-NEXT: vmov q0, q1
|
||||
; CHECK-NEXT: bx lr
|
||||
entry:
|
||||
|
|
@ -156,7 +149,6 @@ define arm_aapcs_vfpcc <16 x i8> @vqmovni16_sminmax_t2(<8 x i16> %s0, <16 x i8>
|
|||
; CHECK-LABEL: vqmovni16_sminmax_t2:
|
||||
; CHECK: @ %bb.0: @ %entry
|
||||
; CHECK-NEXT: vqmovnb.s16 q0, q0
|
||||
; CHECK-NEXT: vmovlb.s8 q0, q0
|
||||
; CHECK-NEXT: vmovnt.i16 q0, q1
|
||||
; CHECK-NEXT: bx lr
|
||||
entry:
|
||||
|
|
@ -172,9 +164,7 @@ entry:
|
|||
define arm_aapcs_vfpcc <16 x i8> @vqmovni16_sminmax_b1(<8 x i16> %s0, <16 x i8> %src1) {
|
||||
; CHECK-LABEL: vqmovni16_sminmax_b1:
|
||||
; CHECK: @ %bb.0: @ %entry
|
||||
; CHECK-NEXT: vqmovnb.s16 q0, q0
|
||||
; CHECK-NEXT: vmovlb.s8 q0, q0
|
||||
; CHECK-NEXT: vmovnb.i16 q1, q0
|
||||
; CHECK-NEXT: vqmovnb.s16 q1, q0
|
||||
; CHECK-NEXT: vmov q0, q1
|
||||
; CHECK-NEXT: bx lr
|
||||
entry:
|
||||
|
|
|
|||
Loading…
Reference in New Issue