This is an archive of the discontinued LLVM Phabricator instance.

Differential D111064

[X86][Costmodel] Load/store i32/f32 Stride=4 VF=16 interleaving costs
ClosedPublic

Authored by lebedev.ri on Oct 4 2021, 8:27 AM.

Details

Reviewers
RKSimon
Commits
rG7d91037fd2f7: [X86][Costmodel] Load/store i32/f32 Stride=4 VF=16 interleaving costs
Summary

This one required quite a bit of assembly surgery, but the trend continues, so i think this is right.

The only sched models that for cpu's that support avx2
but not avx512 are: haswell, broadwell, skylake, zen1-3

For load we have:
https://godbolt.org/z/EKWdj8cKT - for intels Block RThroughput: <=32.0; for ryzens, Block RThroughput: <=24.0
So could pick cost of 32.

For store we have:
https://godbolt.org/z/zj4bb9P75 - for intels Block RThroughput: =32.0; for ryzens, Block RThroughput: <=16.0
So we could pick cost of 32.

I'm directly using the shuffling asm the llc produced,
without any manual fixups that may be needed
to ensure sequential execution.

Diff Detail

Event Timeline

lebedev.ri created this revision.Oct 4 2021, 8:27 AM
Herald added subscribers: pengfei, hiraditya. · View Herald TranscriptOct 4 2021, 8:27 AM
lebedev.ri requested review of this revision.Oct 4 2021, 8:27 AM
lebedev.ri added a parent revision: D111063: [X86][Costmodel] Load/store i32/f32 Stride=4 VF=8 interleaving costs.
lebedev.ri updated this revision to Diff 376907.Oct 4 2021, 8:30 AM
lebedev.ri edited the summary of this revision. (Show Details)

Rebased, NFC.

RKSimon accepted this revision.Oct 4 2021, 8:57 AM

LGTM

This revision is now accepted and ready to land.Oct 4 2021, 8:57 AM
lebedev.ri added a comment.Oct 4 2021, 9:29 AM
In D111064#3040063, @RKSimon wrote:

LGTM

Thank you for the reviews!

Closed by commit rG7d91037fd2f7: [X86][Costmodel] Load/store i32/f32 Stride=4 VF=16 interleaving costs (authored by lebedev.ri). · Explain WhyOct 5 2021, 6:59 AM
This revision was automatically updated to reflect the committed changes.
lebedev.ri added a commit: rG7d91037fd2f7: [X86][Costmodel] Load/store i32/f32 Stride=4 VF=16 interleaving costs.

Revision Contents

Diff 377210

llvm/lib/Target/X86/X86TargetTransformInfo.cpp

Show First 20 Lines • Show All 5,125 Lines • ▼ Show 20 Lines static const CostTblEntry AVX2InterleavedLoadTbl[] = {
{4, MVT::v4i16, 17}, // (load 16i16 and) deinterleave into 4 x 4i16 {4, MVT::v4i16, 17}, // (load 16i16 and) deinterleave into 4 x 4i16
{4, MVT::v8i16, 33}, // (load 32i16 and) deinterleave into 4 x 8i16 {4, MVT::v8i16, 33}, // (load 32i16 and) deinterleave into 4 x 8i16
{4, MVT::v16i16, 75}, // (load 64i16 and) deinterleave into 4 x 16i16 {4, MVT::v16i16, 75}, // (load 64i16 and) deinterleave into 4 x 16i16
{4, MVT::v32i16, 150}, // (load 128i16 and) deinterleave into 4 x 32i16 {4, MVT::v32i16, 150}, // (load 128i16 and) deinterleave into 4 x 32i16
{4, MVT::v2i32, 4}, // (load 8i32 and) deinterleave into 4 x 2i32 {4, MVT::v2i32, 4}, // (load 8i32 and) deinterleave into 4 x 2i32
{4, MVT::v4i32, 8}, // (load 16i32 and) deinterleave into 4 x 4i32 {4, MVT::v4i32, 8}, // (load 16i32 and) deinterleave into 4 x 4i32
{4, MVT::v8i32, 16}, // (load 32i32 and) deinterleave into 4 x 8i32 {4, MVT::v8i32, 16}, // (load 32i32 and) deinterleave into 4 x 8i32
{4, MVT::v16i32, 32}, // (load 64i32 and) deinterleave into 4 x 16i32
{6, MVT::v2i8, 6}, // (load 12i8 and) deinterleave into 6 x 2i8 {6, MVT::v2i8, 6}, // (load 12i8 and) deinterleave into 6 x 2i8
{6, MVT::v4i8, 14}, // (load 24i8 and) deinterleave into 6 x 4i8 {6, MVT::v4i8, 14}, // (load 24i8 and) deinterleave into 6 x 4i8
{6, MVT::v8i8, 18}, // (load 48i8 and) deinterleave into 6 x 8i8 {6, MVT::v8i8, 18}, // (load 48i8 and) deinterleave into 6 x 8i8
{6, MVT::v16i8, 43}, // (load 96i8 and) deinterleave into 6 x 16i8 {6, MVT::v16i8, 43}, // (load 96i8 and) deinterleave into 6 x 16i8
{6, MVT::v32i8, 82}, // (load 192i8 and) deinterleave into 6 x 32i8 {6, MVT::v32i8, 82}, // (load 192i8 and) deinterleave into 6 x 32i8
{6, MVT::v2i16, 13}, // (load 12i16 and) deinterleave into 6 x 2i16 {6, MVT::v2i16, 13}, // (load 12i16 and) deinterleave into 6 x 2i16
▲ Show 20 Lines • Show All 60 Lines • ▼ Show 20 Lines static const CostTblEntry AVX2InterleavedStoreTbl[] = {
{4, MVT::v4i16, 6}, // interleave 4 x 4i16 into 16i16 (and store) {4, MVT::v4i16, 6}, // interleave 4 x 4i16 into 16i16 (and store)
{4, MVT::v8i16, 10}, // interleave 4 x 8i16 into 32i16 (and store) {4, MVT::v8i16, 10}, // interleave 4 x 8i16 into 32i16 (and store)
{4, MVT::v16i16, 32}, // interleave 4 x 16i16 into 64i16 (and store) {4, MVT::v16i16, 32}, // interleave 4 x 16i16 into 64i16 (and store)
{4, MVT::v32i16, 64}, // interleave 4 x 32i16 into 128i16 (and store) {4, MVT::v32i16, 64}, // interleave 4 x 32i16 into 128i16 (and store)
{4, MVT::v2i32, 5}, // interleave 4 x 2i32 into 8i32 (and store) {4, MVT::v2i32, 5}, // interleave 4 x 2i32 into 8i32 (and store)
{4, MVT::v4i32, 6}, // interleave 4 x 4i32 into 16i32 (and store) {4, MVT::v4i32, 6}, // interleave 4 x 4i32 into 16i32 (and store)
{4, MVT::v8i32, 16}, // interleave 4 x 8i32 into 32i32 (and store) {4, MVT::v8i32, 16}, // interleave 4 x 8i32 into 32i32 (and store)
{4, MVT::v16i32, 32}, // interleave 4 x 16i32 into 64i32 (and store)
{6, MVT::v2i8, 7}, // interleave 6 x 2i8 into 12i8 (and store) {6, MVT::v2i8, 7}, // interleave 6 x 2i8 into 12i8 (and store)
{6, MVT::v4i8, 9}, // interleave 6 x 4i8 into 24i8 (and store) {6, MVT::v4i8, 9}, // interleave 6 x 4i8 into 24i8 (and store)
{6, MVT::v8i8, 16}, // interleave 6 x 8i8 into 48i8 (and store) {6, MVT::v8i8, 16}, // interleave 6 x 8i8 into 48i8 (and store)
{6, MVT::v16i8, 27}, // interleave 6 x 16i8 into 96i8 (and store) {6, MVT::v16i8, 27}, // interleave 6 x 16i8 into 96i8 (and store)
{6, MVT::v32i8, 90}, // interleave 6 x 32i8 into 192i8 (and store) {6, MVT::v32i8, 90}, // interleave 6 x 32i8 into 192i8 (and store)
{6, MVT::v2i16, 10}, // interleave 6 x 2i16 into 12i16 (and store) {6, MVT::v2i16, 10}, // interleave 6 x 2i16 into 12i16 (and store)
▲ Show 20 Lines • Show All 175 Lines • Show Last 20 Lines

llvm/test/Analysis/CostModel/X86/interleaved-load-f32-stride-4.ll

Show All 21 Lines
; AVX1: LV: Found an estimated cost of 34 for VF 4 For instruction: %v0 = load float, float* %in0, align 4 ; AVX1: LV: Found an estimated cost of 34 for VF 4 For instruction: %v0 = load float, float* %in0, align 4
; AVX1: LV: Found an estimated cost of 76 for VF 8 For instruction: %v0 = load float, float* %in0, align 4 ; AVX1: LV: Found an estimated cost of 76 for VF 8 For instruction: %v0 = load float, float* %in0, align 4
; AVX1: LV: Found an estimated cost of 152 for VF 16 For instruction: %v0 = load float, float* %in0, align 4 ; AVX1: LV: Found an estimated cost of 152 for VF 16 For instruction: %v0 = load float, float* %in0, align 4
; ;
; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load float, float* %in0, align 4 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load float, float* %in0, align 4
; AVX2: LV: Found an estimated cost of 5 for VF 2 For instruction: %v0 = load float, float* %in0, align 4 ; AVX2: LV: Found an estimated cost of 5 for VF 2 For instruction: %v0 = load float, float* %in0, align 4
; AVX2: LV: Found an estimated cost of 10 for VF 4 For instruction: %v0 = load float, float* %in0, align 4 ; AVX2: LV: Found an estimated cost of 10 for VF 4 For instruction: %v0 = load float, float* %in0, align 4
; AVX2: LV: Found an estimated cost of 20 for VF 8 For instruction: %v0 = load float, float* %in0, align 4 ; AVX2: LV: Found an estimated cost of 20 for VF 8 For instruction: %v0 = load float, float* %in0, align 4
; AVX2: LV: Found an estimated cost of 152 for VF 16 For instruction: %v0 = load float, float* %in0, align 4 ; AVX2: LV: Found an estimated cost of 40 for VF 16 For instruction: %v0 = load float, float* %in0, align 4
; ;
; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load float, float* %in0, align 4 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load float, float* %in0, align 4
; AVX512: LV: Found an estimated cost of 5 for VF 2 For instruction: %v0 = load float, float* %in0, align 4 ; AVX512: LV: Found an estimated cost of 5 for VF 2 For instruction: %v0 = load float, float* %in0, align 4
; AVX512: LV: Found an estimated cost of 5 for VF 4 For instruction: %v0 = load float, float* %in0, align 4 ; AVX512: LV: Found an estimated cost of 5 for VF 4 For instruction: %v0 = load float, float* %in0, align 4
; AVX512: LV: Found an estimated cost of 8 for VF 8 For instruction: %v0 = load float, float* %in0, align 4 ; AVX512: LV: Found an estimated cost of 8 for VF 8 For instruction: %v0 = load float, float* %in0, align 4
; AVX512: LV: Found an estimated cost of 22 for VF 16 For instruction: %v0 = load float, float* %in0, align 4 ; AVX512: LV: Found an estimated cost of 22 for VF 16 For instruction: %v0 = load float, float* %in0, align 4
; AVX512: LV: Found an estimated cost of 92 for VF 32 For instruction: %v0 = load float, float* %in0, align 4 ; AVX512: LV: Found an estimated cost of 92 for VF 32 For instruction: %v0 = load float, float* %in0, align 4
; ;
Show All 40 Lines

llvm/test/Analysis/CostModel/X86/interleaved-load-i32-stride-4.ll

Show All 21 Lines
; AVX1: LV: Found an estimated cost of 42 for VF 4 For instruction: %v0 = load i32, i32* %in0, align 4 ; AVX1: LV: Found an estimated cost of 42 for VF 4 For instruction: %v0 = load i32, i32* %in0, align 4
; AVX1: LV: Found an estimated cost of 92 for VF 8 For instruction: %v0 = load i32, i32* %in0, align 4 ; AVX1: LV: Found an estimated cost of 92 for VF 8 For instruction: %v0 = load i32, i32* %in0, align 4
; AVX1: LV: Found an estimated cost of 184 for VF 16 For instruction: %v0 = load i32, i32* %in0, align 4 ; AVX1: LV: Found an estimated cost of 184 for VF 16 For instruction: %v0 = load i32, i32* %in0, align 4
; ;
; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i32, i32* %in0, align 4 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i32, i32* %in0, align 4
; AVX2: LV: Found an estimated cost of 5 for VF 2 For instruction: %v0 = load i32, i32* %in0, align 4 ; AVX2: LV: Found an estimated cost of 5 for VF 2 For instruction: %v0 = load i32, i32* %in0, align 4
; AVX2: LV: Found an estimated cost of 10 for VF 4 For instruction: %v0 = load i32, i32* %in0, align 4 ; AVX2: LV: Found an estimated cost of 10 for VF 4 For instruction: %v0 = load i32, i32* %in0, align 4
; AVX2: LV: Found an estimated cost of 20 for VF 8 For instruction: %v0 = load i32, i32* %in0, align 4 ; AVX2: LV: Found an estimated cost of 20 for VF 8 For instruction: %v0 = load i32, i32* %in0, align 4
; AVX2: LV: Found an estimated cost of 184 for VF 16 For instruction: %v0 = load i32, i32* %in0, align 4 ; AVX2: LV: Found an estimated cost of 40 for VF 16 For instruction: %v0 = load i32, i32* %in0, align 4
; ;
; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i32, i32* %in0, align 4 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i32, i32* %in0, align 4
; AVX512: LV: Found an estimated cost of 5 for VF 2 For instruction: %v0 = load i32, i32* %in0, align 4 ; AVX512: LV: Found an estimated cost of 5 for VF 2 For instruction: %v0 = load i32, i32* %in0, align 4
; AVX512: LV: Found an estimated cost of 5 for VF 4 For instruction: %v0 = load i32, i32* %in0, align 4 ; AVX512: LV: Found an estimated cost of 5 for VF 4 For instruction: %v0 = load i32, i32* %in0, align 4
; AVX512: LV: Found an estimated cost of 8 for VF 8 For instruction: %v0 = load i32, i32* %in0, align 4 ; AVX512: LV: Found an estimated cost of 8 for VF 8 For instruction: %v0 = load i32, i32* %in0, align 4
; AVX512: LV: Found an estimated cost of 22 for VF 16 For instruction: %v0 = load i32, i32* %in0, align 4 ; AVX512: LV: Found an estimated cost of 22 for VF 16 For instruction: %v0 = load i32, i32* %in0, align 4
; AVX512: LV: Found an estimated cost of 92 for VF 32 For instruction: %v0 = load i32, i32* %in0, align 4 ; AVX512: LV: Found an estimated cost of 92 for VF 32 For instruction: %v0 = load i32, i32* %in0, align 4
; ;
Show All 40 Lines

llvm/test/Analysis/CostModel/X86/interleaved-store-f32-stride-4.ll

Show All 21 Lines
; AVX1: LV: Found an estimated cost of 30 for VF 4 For instruction: store float %v3, float* %out3, align 4 ; AVX1: LV: Found an estimated cost of 30 for VF 4 For instruction: store float %v3, float* %out3, align 4
; AVX1: LV: Found an estimated cost of 76 for VF 8 For instruction: store float %v3, float* %out3, align 4 ; AVX1: LV: Found an estimated cost of 76 for VF 8 For instruction: store float %v3, float* %out3, align 4
; AVX1: LV: Found an estimated cost of 152 for VF 16 For instruction: store float %v3, float* %out3, align 4 ; AVX1: LV: Found an estimated cost of 152 for VF 16 For instruction: store float %v3, float* %out3, align 4
; ;
; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: store float %v3, float* %out3, align 4 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: store float %v3, float* %out3, align 4
; AVX2: LV: Found an estimated cost of 6 for VF 2 For instruction: store float %v3, float* %out3, align 4 ; AVX2: LV: Found an estimated cost of 6 for VF 2 For instruction: store float %v3, float* %out3, align 4
; AVX2: LV: Found an estimated cost of 8 for VF 4 For instruction: store float %v3, float* %out3, align 4 ; AVX2: LV: Found an estimated cost of 8 for VF 4 For instruction: store float %v3, float* %out3, align 4
; AVX2: LV: Found an estimated cost of 20 for VF 8 For instruction: store float %v3, float* %out3, align 4 ; AVX2: LV: Found an estimated cost of 20 for VF 8 For instruction: store float %v3, float* %out3, align 4
; AVX2: LV: Found an estimated cost of 152 for VF 16 For instruction: store float %v3, float* %out3, align 4 ; AVX2: LV: Found an estimated cost of 40 for VF 16 For instruction: store float %v3, float* %out3, align 4
; ;
; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: store float %v3, float* %out3, align 4 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: store float %v3, float* %out3, align 4
; AVX512: LV: Found an estimated cost of 5 for VF 2 For instruction: store float %v3, float* %out3, align 4 ; AVX512: LV: Found an estimated cost of 5 for VF 2 For instruction: store float %v3, float* %out3, align 4
; AVX512: LV: Found an estimated cost of 5 for VF 4 For instruction: store float %v3, float* %out3, align 4 ; AVX512: LV: Found an estimated cost of 5 for VF 4 For instruction: store float %v3, float* %out3, align 4
; AVX512: LV: Found an estimated cost of 11 for VF 8 For instruction: store float %v3, float* %out3, align 4 ; AVX512: LV: Found an estimated cost of 11 for VF 8 For instruction: store float %v3, float* %out3, align 4
; AVX512: LV: Found an estimated cost of 22 for VF 16 For instruction: store float %v3, float* %out3, align 4 ; AVX512: LV: Found an estimated cost of 22 for VF 16 For instruction: store float %v3, float* %out3, align 4
; AVX512: LV: Found an estimated cost of 44 for VF 32 For instruction: store float %v3, float* %out3, align 4 ; AVX512: LV: Found an estimated cost of 44 for VF 32 For instruction: store float %v3, float* %out3, align 4
; AVX512: LV: Found an estimated cost of 88 for VF 64 For instruction: store float %v3, float* %out3, align 4 ; AVX512: LV: Found an estimated cost of 88 for VF 64 For instruction: store float %v3, float* %out3, align 4
▲ Show 20 Lines • Show All 42 Lines • Show Last 20 Lines

llvm/test/Analysis/CostModel/X86/interleaved-store-i32-stride-4.ll

Show All 21 Lines
; AVX1: LV: Found an estimated cost of 38 for VF 4 For instruction: store i32 %v3, i32* %out3, align 4 ; AVX1: LV: Found an estimated cost of 38 for VF 4 For instruction: store i32 %v3, i32* %out3, align 4
; AVX1: LV: Found an estimated cost of 92 for VF 8 For instruction: store i32 %v3, i32* %out3, align 4 ; AVX1: LV: Found an estimated cost of 92 for VF 8 For instruction: store i32 %v3, i32* %out3, align 4
; AVX1: LV: Found an estimated cost of 184 for VF 16 For instruction: store i32 %v3, i32* %out3, align 4 ; AVX1: LV: Found an estimated cost of 184 for VF 16 For instruction: store i32 %v3, i32* %out3, align 4
; ;
; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: store i32 %v3, i32* %out3, align 4 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: store i32 %v3, i32* %out3, align 4
; AVX2: LV: Found an estimated cost of 6 for VF 2 For instruction: store i32 %v3, i32* %out3, align 4 ; AVX2: LV: Found an estimated cost of 6 for VF 2 For instruction: store i32 %v3, i32* %out3, align 4
; AVX2: LV: Found an estimated cost of 8 for VF 4 For instruction: store i32 %v3, i32* %out3, align 4 ; AVX2: LV: Found an estimated cost of 8 for VF 4 For instruction: store i32 %v3, i32* %out3, align 4
; AVX2: LV: Found an estimated cost of 20 for VF 8 For instruction: store i32 %v3, i32* %out3, align 4 ; AVX2: LV: Found an estimated cost of 20 for VF 8 For instruction: store i32 %v3, i32* %out3, align 4
; AVX2: LV: Found an estimated cost of 184 for VF 16 For instruction: store i32 %v3, i32* %out3, align 4 ; AVX2: LV: Found an estimated cost of 40 for VF 16 For instruction: store i32 %v3, i32* %out3, align 4
; ;
; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: store i32 %v3, i32* %out3, align 4 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: store i32 %v3, i32* %out3, align 4
; AVX512: LV: Found an estimated cost of 5 for VF 2 For instruction: store i32 %v3, i32* %out3, align 4 ; AVX512: LV: Found an estimated cost of 5 for VF 2 For instruction: store i32 %v3, i32* %out3, align 4
; AVX512: LV: Found an estimated cost of 5 for VF 4 For instruction: store i32 %v3, i32* %out3, align 4 ; AVX512: LV: Found an estimated cost of 5 for VF 4 For instruction: store i32 %v3, i32* %out3, align 4
; AVX512: LV: Found an estimated cost of 11 for VF 8 For instruction: store i32 %v3, i32* %out3, align 4 ; AVX512: LV: Found an estimated cost of 11 for VF 8 For instruction: store i32 %v3, i32* %out3, align 4
; AVX512: LV: Found an estimated cost of 22 for VF 16 For instruction: store i32 %v3, i32* %out3, align 4 ; AVX512: LV: Found an estimated cost of 22 for VF 16 For instruction: store i32 %v3, i32* %out3, align 4
; AVX512: LV: Found an estimated cost of 44 for VF 32 For instruction: store i32 %v3, i32* %out3, align 4 ; AVX512: LV: Found an estimated cost of 44 for VF 32 For instruction: store i32 %v3, i32* %out3, align 4
; AVX512: LV: Found an estimated cost of 88 for VF 64 For instruction: store i32 %v3, i32* %out3, align 4 ; AVX512: LV: Found an estimated cost of 88 for VF 64 For instruction: store i32 %v3, i32* %out3, align 4
▲ Show 20 Lines • Show All 42 Lines • Show Last 20 Lines