From: Mohammed Agabaria <mohammed.agabaria@intel.com>
Date: Wed, 25 Jan 2017 09:14:48 +0000 (+0000)
Subject: [X86] enable memory interleaving for X86\SLM arch.
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=e0bafdf0596fdc50eb272d49e345bea1425f8ab6;p=llvm

[X86] enable memory interleaving for X86\SLM arch.

Differential Revision: https://reviews.llvm.org/D28547


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@293040 91177308-0d34-0410-b5e6-96231b3b80d8
---

diff --git a/lib/Target/X86/X86TargetTransformInfo.cpp b/lib/Target/X86/X86TargetTransformInfo.cpp
index 073584870bb..bd7fa9ca617 100644
--- a/lib/Target/X86/X86TargetTransformInfo.cpp
+++ b/lib/Target/X86/X86TargetTransformInfo.cpp
@@ -2147,7 +2147,7 @@ bool X86TTIImpl::enableInterleavedAccessVectorization() {
   // TODO: We expect this to be beneficial regardless of arch,
   // but there are currently some unexplained performance artifacts on Atom.
   // As a temporary solution, disable on Atom.
-  return !(ST->isAtom() || ST->isSLM());
+  return !(ST->isAtom());
 }
 
 // Get estimation for interleaved load/store operations and strided load.
diff --git a/test/Transforms/LoopVectorize/X86/interleaving.ll b/test/Transforms/LoopVectorize/X86/interleaving.ll
index de5db532438..9294c92b575 100644
--- a/test/Transforms/LoopVectorize/X86/interleaving.ll
+++ b/test/Transforms/LoopVectorize/X86/interleaving.ll
@@ -1,4 +1,5 @@
 ; RUN: opt -S -mtriple=x86_64-pc_linux -loop-vectorize -instcombine < %s | FileCheck %s --check-prefix=NORMAL
+; RUN: opt -S -mtriple=x86_64-pc_linux -loop-vectorize -instcombine -mcpu=slm < %s | FileCheck %s --check-prefix=NORMAL
 ; RUN: opt -S -mtriple=x86_64-pc_linux -loop-vectorize -instcombine -mcpu=atom < %s | FileCheck %s --check-prefix=ATOM
 
 ; NORMAL-LABEL: foo