Don't remove side effecting instructions due to ConstantFoldInstruction

Just because we can constant fold the result of an instruction does not
imply that we can delete the instruction.  It may have side effects.

This fixes PR28655.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@276389 91177308-0d34-0410-b5e6-96231b3b80d8

lib/Transforms/IPO/GlobalOpt.cpp

@@ -44,6 +44,7 @@
 #include "llvm/Transforms/Utils/CtorUtils.h"
 #include "llvm/Transforms/Utils/Evaluator.h"
 #include "llvm/Transforms/Utils/GlobalStatus.h"
+#include "llvm/Transforms/Utils/Local.h"
 #include <algorithm>
 using namespace llvm;
 
@@ -779,7 +780,8 @@ static void ConstantPropUsersOf(Value *V, const DataLayout &DL,
         // Instructions could multiply use V.
         while (UI != E && *UI == I)
           ++UI;
-        I->eraseFromParent();
+        if (isInstructionTriviallyDead(I, TLI))
+          I->eraseFromParent();
       }
 }
 

lib/Transforms/InstCombine/InstructionCombining.cpp

@@ -2830,7 +2830,8 @@ bool InstCombiner::run() {
         // Add operands to the worklist.
         replaceInstUsesWith(*I, C);
         ++NumConstProp;
-        eraseInstFromFunction(*I);
+        if (isInstructionTriviallyDead(I, TLI))
+          eraseInstFromFunction(*I);
         MadeIRChange = true;
         continue;
       }
@@ -2851,7 +2852,8 @@ bool InstCombiner::run() {
         // Add operands to the worklist.
         replaceInstUsesWith(*I, C);
         ++NumConstProp;
-        eraseInstFromFunction(*I);
+        if (isInstructionTriviallyDead(I, TLI))
+          eraseInstFromFunction(*I);
         MadeIRChange = true;
         continue;
       }
@@ -3007,7 +3009,8 @@ static bool AddReachableCodeToWorklist(BasicBlock *BB, const DataLayout &DL,
                        << *Inst << '\n');
           Inst->replaceAllUsesWith(C);
           ++NumConstProp;
-          Inst->eraseFromParent();
+          if (isInstructionTriviallyDead(Inst, TLI))
+            Inst->eraseFromParent();
           continue;
         }
 

lib/Transforms/Scalar/ConstantProp.cpp

@@ -19,6 +19,7 @@
 //===----------------------------------------------------------------------===//
 
 #include "llvm/Transforms/Scalar.h"
+#include "llvm/Transforms/Utils/Local.h"
 #include "llvm/ADT/Statistic.h"
 #include "llvm/Analysis/ConstantFolding.h"
 #include "llvm/IR/Constant.h"
@@ -90,11 +91,13 @@ bool ConstantPropagation::runOnFunction(Function &F) {
 
         // Remove the dead instruction.
         WorkList.erase(I);
-        I->eraseFromParent();
+        if (isInstructionTriviallyDead(I, TLI)) {
+          I->eraseFromParent();
+          ++NumInstKilled;
+        }
 
         // We made a change to the function...
         Changed = true;
-        ++NumInstKilled;
       }
   }
   return Changed;

lib/Transforms/Scalar/JumpThreading.cpp

@@ -758,7 +758,8 @@ bool JumpThreadingPass::ProcessBlock(BasicBlock *BB) {
         ConstantFoldInstruction(I, BB->getModule()->getDataLayout(), TLI);
     if (SimpleVal) {
       I->replaceAllUsesWith(SimpleVal);
-      I->eraseFromParent();
+      if (isInstructionTriviallyDead(I, TLI))
+        I->eraseFromParent();
       Condition = SimpleVal;
     }
   }

lib/Transforms/Scalar/LICM.cpp

@@ -377,9 +377,11 @@ bool llvm::hoistRegion(DomTreeNode *N, AliasAnalysis *AA, LoopInfo *LI,
               &I, I.getModule()->getDataLayout(), TLI)) {
         DEBUG(dbgs() << "LICM folding inst: " << I << "  --> " << *C << '\n');
         CurAST->copyValue(&I, C);
-        CurAST->deleteValue(&I);
         I.replaceAllUsesWith(C);
-        I.eraseFromParent();
+        if (isInstructionTriviallyDead(&I, TLI)) {
+          CurAST->deleteValue(&I);
+          I.eraseFromParent();
+        }
         continue;
       }
 

test/Transforms/ConstProp/calls.ll

@@ -1,47 +1,47 @@
 ; RUN: opt < %s -constprop -S | FileCheck %s
 ; RUN: opt < %s -constprop -disable-simplify-libcalls -S | FileCheck %s --check-prefix=FNOBUILTIN
 
-declare double @acos(double)
-declare double @asin(double)
-declare double @atan(double)
-declare double @atan2(double, double)
-declare double @ceil(double)
-declare double @cos(double)
-declare double @cosh(double)
-declare double @exp(double)
-declare double @exp2(double)
-declare double @fabs(double)
-declare double @floor(double)
-declare double @fmod(double, double)
-declare double @log(double)
-declare double @log10(double)
-declare double @pow(double, double)
-declare double @sin(double)
-declare double @sinh(double)
-declare double @sqrt(double)
-declare double @tan(double)
-declare double @tanh(double)
+declare double @acos(double) readnone nounwind
+declare double @asin(double) readnone nounwind
+declare double @atan(double) readnone nounwind
+declare double @atan2(double, double) readnone nounwind
+declare double @ceil(double) readnone nounwind
+declare double @cos(double) readnone nounwind
+declare double @cosh(double) readnone nounwind
+declare double @exp(double) readnone nounwind
+declare double @exp2(double) readnone nounwind
+declare double @fabs(double) readnone nounwind
+declare double @floor(double) readnone nounwind
+declare double @fmod(double, double) readnone nounwind
+declare double @log(double) readnone nounwind
+declare double @log10(double) readnone nounwind
+declare double @pow(double, double) readnone nounwind
+declare double @sin(double) readnone nounwind
+declare double @sinh(double) readnone nounwind
+declare double @sqrt(double) readnone nounwind
+declare double @tan(double) readnone nounwind
+declare double @tanh(double) readnone nounwind
 
-declare float @acosf(float)
-declare float @asinf(float)
-declare float @atanf(float)
-declare float @atan2f(float, float)
-declare float @ceilf(float)
-declare float @cosf(float)
-declare float @coshf(float)
-declare float @expf(float)
-declare float @exp2f(float)
-declare float @fabsf(float)
-declare float @floorf(float)
-declare float @fmodf(float, float)
-declare float @logf(float)
-declare float @log10f(float)
-declare float @powf(float, float)
-declare float @sinf(float)
-declare float @sinhf(float)
-declare float @sqrtf(float)
-declare float @tanf(float)
-declare float @tanhf(float)
+declare float @acosf(float) readnone nounwind
+declare float @asinf(float) readnone nounwind
+declare float @atanf(float) readnone nounwind
+declare float @atan2f(float, float) readnone nounwind
+declare float @ceilf(float) readnone nounwind
+declare float @cosf(float) readnone nounwind
+declare float @coshf(float) readnone nounwind
+declare float @expf(float) readnone nounwind
+declare float @exp2f(float) readnone nounwind
+declare float @fabsf(float) readnone nounwind
+declare float @floorf(float) readnone nounwind
+declare float @fmodf(float, float) readnone nounwind
+declare float @logf(float) readnone nounwind
+declare float @log10f(float) readnone nounwind
+declare float @powf(float, float) readnone nounwind
+declare float @sinf(float) readnone nounwind
+declare float @sinhf(float) readnone nounwind
+declare float @sqrtf(float) readnone nounwind
+declare float @tanf(float) readnone nounwind
+declare float @tanhf(float) readnone nounwind
 
 define double @T() {
 ; CHECK-LABEL: @T(

test/Transforms/InstCombine/call.ll

@@ -276,3 +276,14 @@ define <2 x i16> @test16() {
   %X = call <2 x i16> bitcast (i32 ()* @test16a to <2 x i16> ()*)( )
   ret <2 x i16> %X
 }
+
+declare i32 @pr28655(i32 returned %V)
+
+define i32 @test17() {
+entry:
+  %C = call i32 @pr28655(i32 0)
+  ret i32 %C
+}
+; CHECK-LABEL: @test17(
+; CHECK: call i32 @pr28655(i32 0)
+; CHECK: ret i32 0

test/Transforms/InstCombine/log-pow.ll

@@ -55,7 +55,8 @@ define double @log_exp2_not_fast(double %x) {
 ; CHECK-NEXT:  %call3 = call fast double @log(double %call2)
 ; CHECK-NEXT:  ret double %call3
 
-declare double @log(double)
+declare double @log(double) #0
 declare double @exp2(double)
 declare double @llvm.pow.f64(double, double)
 
+attributes #0 = { nounwind readnone }