This is an archive of the discontinued LLVM Phabricator instance.

Differential D132922

[RISCV][NFC] Add cost model for llvm.fmuladd
ClosedPublic

Authored by Miss_Grape on Aug 30 2022, 2:17 AM.

Details

Reviewers
jacquesguan
benshi001
Commits
rG62a238a1f839: [RISCV][NFC] Add cost model tests of llvm.fmuladd

Diff Detail

Event Timeline

Miss_Grape created this revision.Aug 30 2022, 2:17 AM
Herald added a project: Restricted Project. · View Herald TranscriptAug 30 2022, 2:17 AM
Herald added subscribers: sunshaoce, VincentWu, luke957 and 28 others. · View Herald Transcript
Miss_Grape requested review of this revision.Aug 30 2022, 2:17 AM
Herald added a project: Restricted Project. · View Herald TranscriptAug 30 2022, 2:17 AM
Herald added subscribers: llvm-commits, • pcwang-thead, eopXD and 2 others. · View Herald Transcript
Miss_Grape updated this revision to Diff 456581.Aug 30 2022, 2:36 AM
Miss_Grape added a reviewer: jacquesguan.
Harbormaster completed remote builds in B184101: Diff 456581.Aug 30 2022, 3:19 AM
reames requested changes to this revision.Aug 30 2022, 7:32 AM

Diff contains only tests which differs from review description.

This revision now requires changes to proceed.Aug 30 2022, 7:32 AM
Miss_Grape updated this revision to Diff 456832.Aug 30 2022, 6:23 PM
In D132922#3758432, @reames wrote:

Diff contains only tests which differs from review description.

Done

Harbormaster completed remote builds in B184268: Diff 456832.Aug 30 2022, 7:27 PM
reames resigned from this revision.Aug 31 2022, 2:26 PM
Miss_Grape edited reviewers, added: benshi001; removed: reames.Aug 31 2022, 5:54 PM
benshi001 added a comment.Aug 31 2022, 8:38 PM

Change the title to [RISCV][NFC] Add more cost model tests of llvm.fmuladd

benshi001 accepted this revision.Aug 31 2022, 8:39 PM
This revision is now accepted and ready to land.Aug 31 2022, 8:39 PM
Closed by commit rG62a238a1f839: [RISCV][NFC] Add cost model tests of llvm.fmuladd (authored by Miss_Grape). · Explain WhySep 1 2022, 12:33 AM
This revision was automatically updated to reflect the committed changes.
Miss_Grape added a commit: rG62a238a1f839: [RISCV][NFC] Add cost model tests of llvm.fmuladd.

Revision Contents

PathSize
llvm/
test/
Analysis/
CostModel/
RISCV/
89 lines

Diff 456577

llvm/test/Analysis/CostModel/RISCV/arith-fp.ll

Show First 20 Lines • Show All 637 Lines • ▼ Show 20 Lines;
%NXV1F64 = call <vscale x 1 x double> @llvm.fma.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x double> undef, <vscale x 1 x double> undef) %NXV1F64 = call <vscale x 1 x double> @llvm.fma.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x double> undef, <vscale x 1 x double> undef)
%NXV2F64 = call <vscale x 2 x double> @llvm.fma.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x double> undef, <vscale x 2 x double> undef) %NXV2F64 = call <vscale x 2 x double> @llvm.fma.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x double> undef, <vscale x 2 x double> undef)
%NXV4F64 = call <vscale x 4 x double> @llvm.fma.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x double> undef, <vscale x 4 x double> undef) %NXV4F64 = call <vscale x 4 x double> @llvm.fma.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x double> undef, <vscale x 4 x double> undef)
%NXV8F64 = call <vscale x 8 x double> @llvm.fma.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x double> undef, <vscale x 8 x double> undef) %NXV8F64 = call <vscale x 8 x double> @llvm.fma.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x double> undef, <vscale x 8 x double> undef)
ret i32 undef ret i32 undef
} }
define void @fmuladd() {
; CHECK-LABEL: 'fmuladd'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %1 = call half @llvm.fmuladd.f16(half undef, half undef, half undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call float @llvm.fmuladd.f32(float undef, float undef, float undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call double @llvm.fmuladd.f64(double undef, double undef, double undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %4 = call <2 x half> @llvm.fmuladd.v2f16(<2 x half> undef, <2 x half> undef, <2 x half> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %5 = call <4 x half> @llvm.fmuladd.v4f16(<4 x half> undef, <4 x half> undef, <4 x half> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %6 = call <8 x half> @llvm.fmuladd.v8f16(<8 x half> undef, <8 x half> undef, <8 x half> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %7 = call <16 x half> @llvm.fmuladd.v16f16(<16 x half> undef, <16 x half> undef, <16 x half> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %8 = call <2 x float> @llvm.fmuladd.v2f32(<2 x float> undef, <2 x float> undef, <2 x float> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %9 = call <4 x float> @llvm.fmuladd.v4f32(<4 x float> undef, <4 x float> undef, <4 x float> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %10 = call <8 x float> @llvm.fmuladd.v8f32(<8 x float> undef, <8 x float> undef, <8 x float> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %11 = call <16 x float> @llvm.fmuladd.v16f32(<16 x float> undef, <16 x float> undef, <16 x float> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %12 = call <2 x double> @llvm.fmuladd.v2f64(<2 x double> undef, <2 x double> undef, <2 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %13 = call <4 x double> @llvm.fmuladd.v4f64(<4 x double> undef, <4 x double> undef, <4 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %14 = call <8 x double> @llvm.fmuladd.v8f64(<8 x double> undef, <8 x double> undef, <8 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %15 = call <16 x double> @llvm.fmuladd.v16f64(<16 x double> undef, <16 x double> undef, <16 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %16 = call <vscale x 2 x half> @llvm.fmuladd.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x half> undef, <vscale x 2 x half> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %17 = call <vscale x 4 x half> @llvm.fmuladd.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x half> undef, <vscale x 4 x half> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %18 = call <vscale x 8 x half> @llvm.fmuladd.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x half> undef, <vscale x 8 x half> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %19 = call <vscale x 16 x half> @llvm.fmuladd.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x half> undef, <vscale x 16 x half> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %20 = call <vscale x 2 x float> @llvm.fmuladd.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x float> undef, <vscale x 2 x float> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %21 = call <vscale x 4 x float> @llvm.fmuladd.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x float> undef, <vscale x 4 x float> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %22 = call <vscale x 8 x float> @llvm.fmuladd.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x float> undef, <vscale x 8 x float> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %23 = call <vscale x 16 x float> @llvm.fmuladd.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x float> undef, <vscale x 16 x float> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %24 = call <vscale x 2 x double> @llvm.fmuladd.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x double> undef, <vscale x 2 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %25 = call <vscale x 4 x double> @llvm.fmuladd.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x double> undef, <vscale x 4 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %26 = call <vscale x 8 x double> @llvm.fmuladd.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x double> undef, <vscale x 8 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %27 = call <vscale x 16 x double> @llvm.fmuladd.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x double> undef, <vscale x 16 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret void
;
call half @llvm.fmuladd.f16(half undef, half undef, half undef)
call float @llvm.fmuladd.f32(float undef, float undef, float undef)
call double @llvm.fmuladd.f64(double undef, double undef, double undef)
call <2 x half> @llvm.fmuladd.v2f16(<2 x half> undef, <2 x half> undef, <2 x half> undef)
call <4 x half> @llvm.fmuladd.v4f16(<4 x half> undef, <4 x half> undef, <4 x half> undef)
call <8 x half> @llvm.fmuladd.v8f16(<8 x half> undef, <8 x half> undef, <8 x half> undef)
call <16 x half> @llvm.fmuladd.v16f16(<16 x half> undef, <16 x half> undef, <16 x half> undef)
call <2 x float> @llvm.fmuladd.v2f32(<2 x float> undef, <2 x float> undef, <2 x float> undef)
call <4 x float> @llvm.fmuladd.v4f32(<4 x float> undef, <4 x float> undef, <4 x float> undef)
call <8 x float> @llvm.fmuladd.v8f32(<8 x float> undef, <8 x float> undef, <8 x float> undef)
call <16 x float> @llvm.fmuladd.v16f32(<16 x float> undef, <16 x float> undef, <16 x float> undef)
call <2 x double> @llvm.fmuladd.v2f64(<2 x double> undef, <2 x double> undef, <2 x double> undef)
call <4 x double> @llvm.fmuladd.v4f64(<4 x double> undef, <4 x double> undef, <4 x double> undef)
call <8 x double> @llvm.fmuladd.v8f64(<8 x double> undef, <8 x double> undef, <8 x double> undef)
call <16 x double> @llvm.fmuladd.v16f64(<16 x double> undef, <16 x double> undef, <16 x double> undef)
call <vscale x 2 x half> @llvm.fmuladd.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x half> undef, <vscale x 2 x half> undef)
call <vscale x 4 x half> @llvm.fmuladd.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x half> undef, <vscale x 4 x half> undef)
call <vscale x 8 x half> @llvm.fmuladd.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x half> undef, <vscale x 8 x half> undef)
call <vscale x 16 x half> @llvm.fmuladd.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x half> undef, <vscale x 16 x half> undef)
call <vscale x 2 x float> @llvm.fmuladd.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x float> undef, <vscale x 2 x float> undef)
call <vscale x 4 x float> @llvm.fmuladd.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x float> undef, <vscale x 4 x float> undef)
call <vscale x 8 x float> @llvm.fmuladd.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x float> undef, <vscale x 8 x float> undef)
call <vscale x 16 x float> @llvm.fmuladd.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x float> undef, <vscale x 16 x float> undef)
call <vscale x 2 x double> @llvm.fmuladd.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x double> undef, <vscale x 2 x double> undef)
call <vscale x 4 x double> @llvm.fmuladd.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x double> undef, <vscale x 4 x double> undef)
call <vscale x 8 x double> @llvm.fmuladd.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x double> undef, <vscale x 8 x double> undef)
call <vscale x 16 x double> @llvm.fmuladd.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x double> undef, <vscale x 16 x double> undef)
ret void
}
declare half @llvm.copysign.f16(half, half) declare half @llvm.copysign.f16(half, half)
declare float @llvm.copysign.f32(float, float) declare float @llvm.copysign.f32(float, float)
declare double @llvm.copysign.f64(double, double) declare double @llvm.copysign.f64(double, double)
declare <1 x half> @llvm.copysign.v1f16(<1 x half>, <1 x half>) declare <1 x half> @llvm.copysign.v1f16(<1 x half>, <1 x half>)
declare <2 x half> @llvm.copysign.v2f16(<2 x half>, <2 x half>) declare <2 x half> @llvm.copysign.v2f16(<2 x half>, <2 x half>)
declare <4 x half> @llvm.copysign.v4f16(<4 x half>, <4 x half>) declare <4 x half> @llvm.copysign.v4f16(<4 x half>, <4 x half>)
declare <8 x half> @llvm.copysign.v8f16(<8 x half>, <8 x half>) declare <8 x half> @llvm.copysign.v8f16(<8 x half>, <8 x half>)
▲ Show 20 Lines • Show All 63 Lines • ▼ Show 20 Lines
declare <2 x double> @llvm.fma.v2f64(<2 x double>, <2 x double>, <2 x double>) declare <2 x double> @llvm.fma.v2f64(<2 x double>, <2 x double>, <2 x double>)
declare <4 x double> @llvm.fma.v4f64(<4 x double>, <4 x double>, <4 x double>) declare <4 x double> @llvm.fma.v4f64(<4 x double>, <4 x double>, <4 x double>)
declare <8 x double> @llvm.fma.v8f64(<8 x double>, <8 x double>, <8 x double>) declare <8 x double> @llvm.fma.v8f64(<8 x double>, <8 x double>, <8 x double>)
declare <vscale x 1 x double> @llvm.fma.nxv1f64(<vscale x 1 x double>, <vscale x 1 x double>, <vscale x 1 x double>) declare <vscale x 1 x double> @llvm.fma.nxv1f64(<vscale x 1 x double>, <vscale x 1 x double>, <vscale x 1 x double>)
declare <vscale x 2 x double> @llvm.fma.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>) declare <vscale x 2 x double> @llvm.fma.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>)
declare <vscale x 4 x double> @llvm.fma.nxv4f64(<vscale x 4 x double>, <vscale x 4 x double>, <vscale x 4 x double>) declare <vscale x 4 x double> @llvm.fma.nxv4f64(<vscale x 4 x double>, <vscale x 4 x double>, <vscale x 4 x double>)
declare <vscale x 8 x double> @llvm.fma.nxv8f64(<vscale x 8 x double>, <vscale x 8 x double>, <vscale x 8 x double>) declare <vscale x 8 x double> @llvm.fma.nxv8f64(<vscale x 8 x double>, <vscale x 8 x double>, <vscale x 8 x double>)
declare half @llvm.fmuladd.f16(half, half, half)
declare float @llvm.fmuladd.f32(float, float, float)
declare double @llvm.fmuladd.f64(double, double, double)
declare <2 x half> @llvm.fmuladd.v2f16(<2 x half>, <2 x half>, <2 x half>)
declare <4 x half> @llvm.fmuladd.v4f16(<4 x half>, <4 x half>, <4 x half>)
declare <8 x half> @llvm.fmuladd.v8f16(<8 x half>, <8 x half>, <8 x half>)
declare <16 x half> @llvm.fmuladd.v16f16(<16 x half>, <16 x half>, <16 x half>)
declare <2 x float> @llvm.fmuladd.v2f32(<2 x float>, <2 x float>, <2 x float>)
declare <4 x float> @llvm.fmuladd.v4f32(<4 x float>, <4 x float>, <4 x float>)
declare <8 x float> @llvm.fmuladd.v8f32(<8 x float>, <8 x float>, <8 x float>)
declare <16 x float> @llvm.fmuladd.v16f32(<16 x float>, <16 x float>, <16 x float>)
declare <2 x double> @llvm.fmuladd.v2f64(<2 x double>, <2 x double>, <2 x double>)
declare <4 x double> @llvm.fmuladd.v4f64(<4 x double>, <4 x double>, <4 x double>)
declare <8 x double> @llvm.fmuladd.v8f64(<8 x double>, <8 x double>, <8 x double>)
declare <16 x double> @llvm.fmuladd.v16f64(<16 x double>, <16 x double>, <16 x double>)
declare <vscale x 2 x half> @llvm.fmuladd.nxv2f16(<vscale x 2 x half>, <vscale x 2 x half>, <vscale x 2 x half>)
declare <vscale x 4 x half> @llvm.fmuladd.nxv4f16(<vscale x 4 x half>, <vscale x 4 x half>, <vscale x 4 x half>)
declare <vscale x 8 x half> @llvm.fmuladd.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>)
declare <vscale x 16 x half> @llvm.fmuladd.nxv16f16(<vscale x 16 x half>, <vscale x 16 x half>, <vscale x 16 x half>)
declare <vscale x 2 x float> @llvm.fmuladd.nxv2f32(<vscale x 2 x float>, <vscale x 2 x float>, <vscale x 2 x float>)
declare <vscale x 4 x float> @llvm.fmuladd.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>)
declare <vscale x 8 x float> @llvm.fmuladd.nxv8f32(<vscale x 8 x float>, <vscale x 8 x float>, <vscale x 8 x float>)
declare <vscale x 16 x float> @llvm.fmuladd.nxv16f32(<vscale x 16 x float>, <vscale x 16 x float>, <vscale x 16 x float>)
declare <vscale x 2 x double> @llvm.fmuladd.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>)
declare <vscale x 4 x double> @llvm.fmuladd.nxv4f64(<vscale x 4 x double>, <vscale x 4 x double>, <vscale x 4 x double>)
declare <vscale x 8 x double> @llvm.fmuladd.nxv8f64(<vscale x 8 x double>, <vscale x 8 x double>, <vscale x 8 x double>)
declare <vscale x 16 x double> @llvm.fmuladd.nxv16f64(<vscale x 16 x double>, <vscale x 16 x double>, <vscale x 16 x double>)