[CodeGenPrepare] Don't create dead instructions in addrmode sinking

When we fail to sink an instruction, we must make sure not to modify
the function; otherwise, we end up in an infinite loop because
CodeGenPrepare iterates until it doesn't make any changes.

Fixes https://bugs.llvm.org/show_bug.cgi?id=33608 .

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@307866 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/CodeGen/CodeGenPrepare.cpp b/lib/CodeGen/CodeGenPrepare.cpp
index 9351f2d..b7155ac 100644 (file)
--- a/lib/CodeGen/CodeGenPrepare.cpp
+++ b/lib/CodeGen/CodeGenPrepare.cpp
@@ -4393,6 +4393,20 @@ bool CodeGenPrepare::optimizeMemoryInst(Instruction *MemoryInst, Value *Addr,
       AddrMode.Scale = 0;
     }
 
+    // It is only safe to sign extend the BaseReg if we know that the math
+    // required to create it did not overflow before we extend it. Since
+    // the original IR value was tossed in favor of a constant back when
+    // the AddrMode was created we need to bail out gracefully if widths
+    // do not match instead of extending it.
+    //
+    // (See below for code to add the scale.)
+    if (AddrMode.Scale) {
+      Type *ScaledRegTy = AddrMode.ScaledReg->getType();
+      if (cast<IntegerType>(IntPtrTy)->getBitWidth() >
+          cast<IntegerType>(ScaledRegTy)->getBitWidth())
+        return false;
+    }
+
     if (AddrMode.BaseGV) {
       if (ResultPtr)
         return false;
@@ -4443,19 +4457,11 @@ bool CodeGenPrepare::optimizeMemoryInst(Instruction *MemoryInst, Value *Addr,
         Value *V = AddrMode.ScaledReg;
         if (V->getType() == IntPtrTy) {
           // done.
-        } else if (cast<IntegerType>(IntPtrTy)->getBitWidth() <
-                   cast<IntegerType>(V->getType())->getBitWidth()) {
-          V = Builder.CreateTrunc(V, IntPtrTy, "sunkaddr");
         } else {
-          // It is only safe to sign extend the BaseReg if we know that the math
-          // required to create it did not overflow before we extend it. Since
-          // the original IR value was tossed in favor of a constant back when
-          // the AddrMode was created we need to bail out gracefully if widths
-          // do not match instead of extending it.
-          Instruction *I = dyn_cast_or_null<Instruction>(ResultIndex);
-          if (I && (ResultIndex != AddrMode.BaseReg))
-            I->eraseFromParent();
-          return false;
+          assert(cast<IntegerType>(IntPtrTy)->getBitWidth() <
+                 cast<IntegerType>(V->getType())->getBitWidth() &&
+                 "We can't transform if ScaledReg is too narrow");
+          V = Builder.CreateTrunc(V, IntPtrTy, "sunkaddr");
         }
 
         if (AddrMode.Scale != 1)

diff --git a/test/Transforms/CodeGenPrepare/X86/sink-addrmode.ll b/test/Transforms/CodeGenPrepare/X86/sink-addrmode.ll
index 9d6e668..b6b7757 100644 (file)
--- a/test/Transforms/CodeGenPrepare/X86/sink-addrmode.ll
+++ b/test/Transforms/CodeGenPrepare/X86/sink-addrmode.ll
@@ -4,6 +4,8 @@ target datalayout =
 "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128"
 target triple = "x86_64-unknown-linux-gnu"
 
+@x = external global [1 x [2 x <4 x float>]]
+
 ; Can we sink single addressing mode computation to use?
 define void @test1(i1 %cond, i64* %base) {
 ; CHECK-LABEL: @test1
@@ -194,3 +196,25 @@ rare.2:
 
 
 declare void @slowpath(i32, i32*)
+
+; Make sure we don't end up in an infinite loop after we fail to sink.
+; CHECK-LABEL: define void @test8
+; CHECK: %ptr = getelementptr i8, i8* %aFOO_load_ptr2int_2void, i32 undef
+define void @test8() {
+allocas:
+  %aFOO_load = load float*, float** undef
+  %aFOO_load_ptr2int = ptrtoint float* %aFOO_load to i64
+  %aFOO_load_ptr2int_broadcast_init = insertelement <4 x i64> undef, i64 %aFOO_load_ptr2int, i32 0
+  %aFOO_load_ptr2int_2void = inttoptr i64 %aFOO_load_ptr2int to i8*
+  %ptr = getelementptr i8, i8* %aFOO_load_ptr2int_2void, i32 undef
+  br label %load.i145
+
+load.i145:
+  %ptr.i143 = bitcast i8* %ptr to <4 x float>*
+  %valall.i144 = load <4 x float>, <4 x float>* %ptr.i143, align 4
+  %x_offset = getelementptr [1 x [2 x <4 x float>]], [1 x [2 x <4 x float>]]* @x, i32 0, i64 0
+  br label %pl_loop.i.i122
+
+pl_loop.i.i122:
+  br label %pl_loop.i.i122
+}