[llvm] r212629 - Fix for PR20059 (instcombine reorders shufflevector after instruction that may trap)

[Sanjay Patel]

Author: spatel
Date: Wed Jul  9 11:34:54 2014
New Revision: 212629

URL: http://llvm.org/viewvc/llvm-project?rev=212629&view=rev
Log:
Fix for PR20059 (instcombine reorders shufflevector after instruction that may trap)

In PR20059 ( http://llvm.org/pr20059 ), instcombine eliminates shuffles that are necessary before performing an operation that can trap (srem).

This patch calls isSafeToSpeculativelyExecute() and bails out of the optimization in SimplifyVectorOp() if needed.

Differential Revision: http://reviews.llvm.org/D4424


Added:
    llvm/trunk/test/Transforms/InstCombine/pr20059.ll
Modified:
    llvm/trunk/lib/Transforms/InstCombine/InstructionCombining.cpp

Modified: llvm/trunk/lib/Transforms/InstCombine/InstructionCombining.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Transforms/InstCombine/InstructionCombining.cpp?rev=212629&r1=212628&r2=212629&view=diff
==============================================================================
--- llvm/trunk/lib/Transforms/InstCombine/InstructionCombining.cpp (original)
+++ llvm/trunk/lib/Transforms/InstCombine/InstructionCombining.cpp Wed Jul  9 11:34:54 2014
@@ -42,6 +42,7 @@
 #include "llvm/Analysis/ConstantFolding.h"
 #include "llvm/Analysis/InstructionSimplify.h"
 #include "llvm/Analysis/MemoryBuiltins.h"
+#include "llvm/Analysis/ValueTracking.h"
 #include "llvm/IR/CFG.h"
 #include "llvm/IR/DataLayout.h"
 #include "llvm/IR/GetElementPtrTypeIterator.h"
@@ -1195,6 +1196,11 @@ static Value *CreateBinOpAsGiven(BinaryO
 Value *InstCombiner::SimplifyVectorOp(BinaryOperator &Inst) {
   if (!Inst.getType()->isVectorTy()) return nullptr;
 
+  // It may not be safe to reorder shuffles and things like div, urem, etc.
+  // because we may trap when executing those ops on unknown vector elements.
+  // See PR20059.
+  if (!isSafeToSpeculativelyExecute(&Inst)) return nullptr;
+
   unsigned VWidth = cast<VectorType>(Inst.getType())->getNumElements();
   Value *LHS = Inst.getOperand(0), *RHS = Inst.getOperand(1);
   assert(cast<VectorType>(LHS->getType())->getNumElements() == VWidth);

Added: llvm/trunk/test/Transforms/InstCombine/pr20059.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Transforms/InstCombine/pr20059.ll?rev=212629&view=auto
==============================================================================
--- llvm/trunk/test/Transforms/InstCombine/pr20059.ll (added)
+++ llvm/trunk/test/Transforms/InstCombine/pr20059.ll Wed Jul  9 11:34:54 2014
@@ -0,0 +1,32 @@
+; RUN: opt -S -instcombine < %s | FileCheck %s
+
+; In PR20059 ( http://llvm.org/pr20059 ), shufflevector operations are reordered/removed
+; for an srem operation. This is not a valid optimization because it may cause a trap
+; on div-by-zero.
+
+; CHECK-LABEL: @do_not_reorder
+; CHECK: %splat1 = shufflevector <4 x i32> %p1, <4 x i32> undef, <4 x i32> zeroinitializer
+; CHECK-NEXT: %splat2 = shufflevector <4 x i32> %p2, <4 x i32> undef, <4 x i32> zeroinitializer
+; CHECK-NEXT: %retval = srem <4 x i32> %splat1, %splat2
+define <4 x i32> @do_not_reorder(<4 x i32> %p1, <4 x i32> %p2) {
+  %splat1 = shufflevector <4 x i32> %p1, <4 x i32> undef, <4 x i32> zeroinitializer
+  %splat2 = shufflevector <4 x i32> %p2, <4 x i32> undef, <4 x i32> zeroinitializer
+  %retval = srem <4 x i32> %splat1, %splat2
+  ret <4 x i32> %retval
+}
+; RUN: opt -S -instcombine < %s | FileCheck %s
+
+; In PR20059 ( http://llvm.org/pr20059 ), shufflevector operations are reordered/removed
+; for an srem operation. This is not a valid optimization because it may cause a trap
+; on div-by-zero.
+
+; CHECK-LABEL: @do_not_reorder
+; CHECK: %splat1 = shufflevector <4 x i32> %p1, <4 x i32> undef, <4 x i32> zeroinitializer
+; CHECK-NEXT: %splat2 = shufflevector <4 x i32> %p2, <4 x i32> undef, <4 x i32> zeroinitializer
+; CHECK-NEXT: %retval = srem <4 x i32> %splat1, %splat2
+define <4 x i32> @do_not_reorder(<4 x i32> %p1, <4 x i32> %p2) {
+  %splat1 = shufflevector <4 x i32> %p1, <4 x i32> undef, <4 x i32> zeroinitializer
+  %splat2 = shufflevector <4 x i32> %p2, <4 x i32> undef, <4 x i32> zeroinitializer
+  %retval = srem <4 x i32> %splat1, %splat2
+  ret <4 x i32> %retval
+}