[llvm] r241897 - Disable loop re-rotation for -Oz (patch by Andrey Turetsky)

Alexey Bataev a.bataev at hotmail.com
Fri Jul 10 03:37:10 PDT 2015


Author: abataev
Date: Fri Jul 10 05:37:09 2015
New Revision: 241897

URL: http://llvm.org/viewvc/llvm-project?rev=241897&view=rev
Log:
Disable loop re-rotation for -Oz (patch by Andrey Turetsky)

After changes in rL231820 loop re-rotation is performed even in -Oz mode. Since loop rotation is disabled for -Oz, it seems loop re-rotation should be disabled too.
Differential Revision: http://reviews.llvm.org/D10961

Added:
    llvm/trunk/test/Transforms/LoopRotate/oz-disable.ll
Modified:
    llvm/trunk/lib/Transforms/IPO/PassManagerBuilder.cpp

Modified: llvm/trunk/lib/Transforms/IPO/PassManagerBuilder.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Transforms/IPO/PassManagerBuilder.cpp?rev=241897&r1=241896&r2=241897&view=diff
==============================================================================
--- llvm/trunk/lib/Transforms/IPO/PassManagerBuilder.cpp (original)
+++ llvm/trunk/lib/Transforms/IPO/PassManagerBuilder.cpp Fri Jul 10 05:37:09 2015
@@ -320,8 +320,8 @@ void PassManagerBuilder::populateModuleP
 
   // Re-rotate loops in all our loop nests. These may have fallout out of
   // rotated form due to GVN or other transformations, and the vectorizer relies
-  // on the rotated form.
-  MPM.add(createLoopRotatePass());
+  // on the rotated form. Disable header duplication at -Oz.
+  MPM.add(createLoopRotatePass(SizeLevel == 2 ? 0 : -1));
 
   // Distribute loops to allow partial vectorization.  I.e. isolate dependences
   // into separate loop that would otherwise inhibit vectorization.

Added: llvm/trunk/test/Transforms/LoopRotate/oz-disable.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Transforms/LoopRotate/oz-disable.ll?rev=241897&view=auto
==============================================================================
--- llvm/trunk/test/Transforms/LoopRotate/oz-disable.ll (added)
+++ llvm/trunk/test/Transforms/LoopRotate/oz-disable.ll Fri Jul 10 05:37:09 2015
@@ -0,0 +1,30 @@
+; REQUIRES: asserts
+; RUN: opt < %s -S -Os -debug -debug-only=loop-rotate 2>&1 | FileCheck %s -check-prefix=OS
+; RUN: opt < %s -S -Oz -debug -debug-only=loop-rotate 2>&1 | FileCheck %s -check-prefix=OZ
+
+; Loop should be rotated for -Os but not for -Oz.
+; OS: rotating Loop at depth 1
+; OZ-NOT: rotating Loop at depth 1
+
+ at e = global i32 10
+
+declare void @use(i32)
+
+define void @test() {
+entry:
+  %end = load i32, i32* @e
+  br label %loop
+
+loop:
+  %n.phi = phi i32 [ %n, %loop.fin ], [ 0, %entry ]
+  %cond = icmp eq i32 %n.phi, %end
+  br i1 %cond, label %exit, label %loop.fin
+
+loop.fin:
+  %n = add i32 %n.phi, 1
+  call void @use(i32 %n)
+  br label %loop
+
+exit:
+  ret void
+}





More information about the llvm-commits mailing list