shylie/llvm-project
1
0
Fork 0
Code Issues Pull Requests Actions 6 Packages Projects Releases Wiki Activity

[lld][MachO] Enable LoopVectorization and SLPVectorization for ThinLTO (#182748)

Browse Source 
Commit 21a4710c67a97838dd75cf60ed24da11280800f8 previously enabled
LoopVectorization and SLPVectorization CodeGen options for the ELF and
COFF LTO backends. Since the Mach-O LTO port did not exist at the time,
it missed this configuration.

This patch adds these options to the Mach-O LTO setup for consistency
with the other backends. Without this, SLP and loop vectorization passes
are silently skipped during Mach-O LTO for O2 and O3 builds.
This commit is contained in:
Tal Keren 2026-02-23 23:30:41 +02:00 committed by GitHub
parent 9829d082af
commit cb0b13d9d7
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 52 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
lld/MachO/LTO.cpp
View File

@ -61,6 +61,10 @@ static lto::Config createConfig() {
c.DisableVerify = config->disableVerify;
c.OptLevel = config->ltoo;
c.CGOptLevel = config->ltoCgo;
c.PTO.LoopVectorization = c.OptLevel > 1;
c.PTO.SLPVectorization = c.OptLevel > 1;
if (config->saveTemps)
checkError(c.addSaveTemps(config->outputFile.str() + ".",
/*UseInputModulePath=*/true));

48
lld/test/MachO/lto-slp-vectorize-pm.ll Normal file
View File

@ -0,0 +1,48 @@
; REQUIRES: x86
; RUN: opt -module-summary %s -o %t.o
; Test SLP and Loop Vectorization are enabled by default at O2 and O3
; RUN: %lld --lto-debug-pass-manager --lto-O0 -save-temps -dylib -o %t1.o %t.o 2>&1 | FileCheck %s --check-prefix=CHECK-O0-SLP
; RUN: llvm-dis %t.o.4.opt.bc -o - | FileCheck %s --check-prefix=CHECK-O0-LPV
; RUN: %lld --lto-debug-pass-manager --lto-O1 -save-temps -dylib -o %t2.o %t.o 2>&1 | FileCheck %s --check-prefix=CHECK-O1-SLP
; RUN: llvm-dis %t.o.4.opt.bc -o - | FileCheck %s --check-prefix=CHECK-O1-LPV
; RUN: %lld --lto-debug-pass-manager --lto-O2 -save-temps -dylib -o %t3.o %t.o 2>&1 | FileCheck %s --check-prefix=CHECK-O2-SLP
; RUN: llvm-dis %t.o.4.opt.bc -o - | FileCheck %s --check-prefix=CHECK-O2-LPV
; RUN: %lld --lto-debug-pass-manager --lto-O3 -save-temps -dylib -o %t4.o %t.o 2>&1 | FileCheck %s --check-prefix=CHECK-O3-SLP
; RUN: llvm-dis %t.o.4.opt.bc -o - | FileCheck %s --check-prefix=CHECK-O3-LPV
; CHECK-O0-SLP-NOT: Running pass: SLPVectorizerPass
; CHECK-O1-SLP-NOT: Running pass: SLPVectorizerPass
; CHECK-O2-SLP: Running pass: SLPVectorizerPass
; CHECK-O3-SLP: Running pass: SLPVectorizerPass
; CHECK-O0-LPV-NOT: = !{!"llvm.loop.isvectorized", i32 1}
; CHECK-O1-LPV-NOT: = !{!"llvm.loop.isvectorized", i32 1}
; CHECK-O2-LPV: = !{!"llvm.loop.isvectorized", i32 1}
; CHECK-O3-LPV: = !{!"llvm.loop.isvectorized", i32 1}
target datalayout = "e-m:o-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-apple-macosx10.15.0"
define i32 @foo(ptr %a) {
entry:
br label %for.body
for.body:
%indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
%red.05 = phi i32 [ 0, %entry ], [ %add, %for.body ]
%arrayidx = getelementptr inbounds i32, ptr %a, i64 %indvars.iv
%0 = load i32, ptr %arrayidx, align 4
%add = add nsw i32 %0, %red.05
%indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
%exitcond = icmp eq i64 %indvars.iv.next, 255
br i1 %exitcond, label %for.end, label %for.body, !llvm.loop !0
for.end:
ret i32 %add
}
!0 = distinct !{!0, !1}
!1 = !{!"llvm.loop.unroll.disable", i1 true}