Skip to content

Commit db37270

Browse files
LiqinWengLiqinWeng
committed
[RISCV][TTI] Add vp.fneg intrinsic cost with functionalOP
1 parent d1b311d commit db37270

File tree

2 files changed

+65
-1
lines changed

2 files changed

+65
-1
lines changed

‎llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1123,7 +1123,8 @@ RISCVTTIImpl::getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
11231123
case Intrinsic::vp_fsub:
11241124
case Intrinsic::vp_fmul:
11251125
case Intrinsic::vp_fdiv:
1126-
case Intrinsic::vp_frem: {
1126+
case Intrinsic::vp_frem:
1127+
case Intrinsic::vp_fneg: {
11271128
std::optional<unsigned> FOp =
11281129
VPIntrinsic::getFunctionalOpcodeForVP(ICA.getID());
11291130
assert(FOp.has_value());

‎llvm/test/Analysis/CostModel/RISCV/rvv-intrinsics.ll

Lines changed: 63 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1501,6 +1501,68 @@ define void @vp_fadd(){
15011501
}
15021502

15031503

1504+
define void @vp_fneg() {
1505+
; CHECK-LABEL: 'vp_fneg'
1506+
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t0 = call <2 x float> @llvm.vp.fneg.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
1507+
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t2 = call <4 x float> @llvm.vp.fneg.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
1508+
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t4 = call <8 x float> @llvm.vp.fneg.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
1509+
; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t6 = call <16 x float> @llvm.vp.fneg.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
1510+
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t8 = call <2 x double> @llvm.vp.fneg.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
1511+
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t10 = call <4 x double> @llvm.vp.fneg.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
1512+
; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t12 = call <8 x double> @llvm.vp.fneg.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
1513+
; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t14 = call <16 x double> @llvm.vp.fneg.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
1514+
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t17 = call <vscale x 2 x float> @llvm.vp.fneg.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
1515+
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t19 = call <vscale x 4 x float> @llvm.vp.fneg.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
1516+
; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t21 = call <vscale x 8 x float> @llvm.vp.fneg.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
1517+
; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t23 = call <vscale x 16 x float> @llvm.vp.fneg.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
1518+
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t25 = call <vscale x 2 x double> @llvm.vp.fneg.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
1519+
; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t27 = call <vscale x 4 x double> @llvm.vp.fneg.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
1520+
; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t29 = call <vscale x 8 x double> @llvm.vp.fneg.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
1521+
; CHECK-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %t31 = call <vscale x 16 x double> @llvm.vp.fneg.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
1522+
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
1523+
;
1524+
; TYPEBASED-LABEL: 'vp_fneg'
1525+
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t0 = call <2 x float> @llvm.vp.fneg.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
1526+
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t2 = call <4 x float> @llvm.vp.fneg.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
1527+
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t4 = call <8 x float> @llvm.vp.fneg.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
1528+
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t6 = call <16 x float> @llvm.vp.fneg.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
1529+
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t8 = call <2 x double> @llvm.vp.fneg.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
1530+
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t10 = call <4 x double> @llvm.vp.fneg.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
1531+
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t12 = call <8 x double> @llvm.vp.fneg.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
1532+
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t14 = call <16 x double> @llvm.vp.fneg.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
1533+
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t17 = call <vscale x 2 x float> @llvm.vp.fneg.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
1534+
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t19 = call <vscale x 4 x float> @llvm.vp.fneg.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
1535+
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t21 = call <vscale x 8 x float> @llvm.vp.fneg.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
1536+
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t23 = call <vscale x 16 x float> @llvm.vp.fneg.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
1537+
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t25 = call <vscale x 2 x double> @llvm.vp.fneg.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
1538+
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t27 = call <vscale x 4 x double> @llvm.vp.fneg.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
1539+
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t29 = call <vscale x 8 x double> @llvm.vp.fneg.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
1540+
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %t31 = call <vscale x 16 x double> @llvm.vp.fneg.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
1541+
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
1542+
;
1543+
%t0 = call <2 x float> @llvm.vp.fneg.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
1544+
%t2 = call <4 x float> @llvm.vp.fneg.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
1545+
%t4 = call <8 x float> @llvm.vp.fneg.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
1546+
%t6 = call <16 x float> @llvm.vp.fneg.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
1547+
1548+
%t8 = call <2 x double> @llvm.vp.fneg.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
1549+
%t10 = call <4 x double> @llvm.vp.fneg.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
1550+
%t12 = call <8 x double> @llvm.vp.fneg.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
1551+
%t14 = call <16 x double> @llvm.vp.fneg.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
1552+
1553+
%t17 = call <vscale x 2 x float> @llvm.vp.fneg.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
1554+
%t19 = call <vscale x 4 x float> @llvm.vp.fneg.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
1555+
%t21 = call <vscale x 8 x float> @llvm.vp.fneg.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
1556+
%t23 = call <vscale x 16 x float> @llvm.vp.fneg.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
1557+
1558+
%t25 = call <vscale x 2 x double> @llvm.vp.fneg.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
1559+
%t27 = call <vscale x 4 x double> @llvm.vp.fneg.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
1560+
%t29 = call <vscale x 8 x double> @llvm.vp.fneg.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
1561+
%t31 = call <vscale x 16 x double> @llvm.vp.fneg.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
1562+
1563+
ret void
1564+
}
1565+
15041566
declare <2 x i8> @llvm.vp.add.v2i8(<2 x i8>, <2 x i8>, <2 x i1>, i32)
15051567
declare <4 x i8> @llvm.vp.add.v4i8(<4 x i8>, <4 x i8>, <4 x i1>, i32)
15061568
declare <8 x i8> @llvm.vp.add.v8i8(<8 x i8>, <8 x i8>, <8 x i1>, i32)
@@ -1772,3 +1834,4 @@ declare <vscale x 1 x i64> @llvm.vp.fshl.nxv1i64(<vscale x 1 x i64>, <vscale x 1
17721834
declare <vscale x 2 x i64> @llvm.vp.fshl.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i1>, i32)
17731835
declare <vscale x 4 x i64> @llvm.vp.fshl.nxv4i64(<vscale x 4 x i64>, <vscale x 4 x i64>, <vscale x 4 x i64>, <vscale x 4 x i1>, i32)
17741836
declare <vscale x 8 x i64> @llvm.vp.fshl.nxv8i64(<vscale x 8 x i64>, <vscale x 8 x i64>, <vscale x 8 x i64>, <vscale x 8 x i1>, i32)
1837+

0 commit comments

Comments
 (0)