This is an archive of the discontinued LLVM Phabricator instance.

[AArch64] Reduce vector insert/extract cost for Falkor.
ClosedPublic

Authored by mcrosier on Jan 6 2017, 10:08 AM.

Download Raw Diff

Details

Reviewers

Commits

rGe177185e796a: [AArch64] Reduce vector insert/extract cost for Falkor.
rL291254: [AArch64] Reduce vector insert/extract cost for Falkor.

Summary

PTAL,
Chad

Diff Detail

Event Timeline

mcrosier updated this revision to Diff 83380.Jan 6 2017, 10:08 AM

mcrosier retitled this revision from to [AArch64] Reduce vector insert/extract cost for Falkor..

mcrosier updated this object.

mcrosier added a reviewer: mssimpso.

mcrosier added a subscriber: llvm-commits.

Herald added subscribers: rengolin, aemerson. · View Herald TranscriptJan 6 2017, 10:08 AM

LGTM.

This revision is now accepted and ready to land.Jan 6 2017, 10:12 AM

Provide full context.

Closed by commit rL291254: [AArch64] Reduce vector insert/extract cost for Falkor. (authored by mcrosier). · Explain WhyJan 6 2017, 10:14 AM

This revision was automatically updated to reflect the committed changes.

Thanks, Matt.

Revision Contents

Path

Size

lib/

Target/

AArch64/

AArch64Subtarget.cpp

1 line

test/

Analysis/

CostModel/

AArch64/

falkor.ll

26 lines

Diff 83380

lib/Target/AArch64/AArch64Subtarget.cpp

Context not available.
	break;	break;
	case Falkor:	case Falkor:
	MaxInterleaveFactor = 4;	MaxInterleaveFactor = 4;
		VectorInsertExtractBaseCost = 2;
	break;	break;
	case Kryo:	case Kryo:
	MaxInterleaveFactor = 4;	MaxInterleaveFactor = 4;
Context not available.

test/Analysis/CostModel/AArch64/falkor.ll

This file was added.

				; RUN: opt < %s -cost-model -analyze -mcpu=falkor \| FileCheck %s

				target datalayout = "e-m:e-i64:64-i128:128-n32:64-S128"
				target triple = "aarch64--linux-gnu"

				; CHECK-LABEL: vectorInstrCost
				define void @vectorInstrCost() {

				; Vector extracts - extracting the first element should have a zero cost;
				; all other elements should have a cost of two.
				;
				; CHECK: cost of 0 {{.*}} extractelement <2 x i64> undef, i32 0
				; CHECK: cost of 2 {{.*}} extractelement <2 x i64> undef, i32 1
				%t1 = extractelement <2 x i64> undef, i32 0
				%t2 = extractelement <2 x i64> undef, i32 1

				; Vector inserts - inserting the first element should have a zero cost; all
				; other elements should have a cost of two.
				;
				; CHECK: cost of 0 {{.*}} insertelement <2 x i64> undef, i64 undef, i32 0
				; CHECK: cost of 2 {{.*}} insertelement <2 x i64> undef, i64 undef, i32 1
				%t3 = insertelement <2 x i64> undef, i64 undef, i32 0
				%t4 = insertelement <2 x i64> undef, i64 undef, i32 1

				ret void
				}