Make SelectionDAG::InferPtrAlignment use llvm::ComputeMaskedBits instead of duplicati...

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@145304 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Analysis/ValueTracking.cpp b/lib/Analysis/ValueTracking.cpp
index 22f1c144ea2680df17b281dc394d7a24658ba18d..8192c1471fc6e623096335a32a8546f06c8298f7 100644 (file)
--- a/lib/Analysis/ValueTracking.cpp
+++ b/lib/Analysis/ValueTracking.cpp
@@ -103,14 +103,16 @@ void llvm::ComputeMaskedBits(Value *V, const APInt &Mask,
   if (GlobalValue *GV = dyn_cast<GlobalValue>(V)) {
     unsigned Align = GV->getAlignment();
     if (Align == 0 && TD && GV->getType()->getElementType()->isSized()) {
   if (GlobalValue *GV = dyn_cast<GlobalValue>(V)) {
     unsigned Align = GV->getAlignment();
     if (Align == 0 && TD && GV->getType()->getElementType()->isSized()) {

-      Type *ObjectType = GV->getType()->getElementType();
-      // If the object is defined in the current Module, we'll be giving
-      // it the preferred alignment. Otherwise, we have to assume that it
-      // may only have the minimum ABI alignment.
-      if (!GV->isDeclaration() && !GV->mayBeOverridden())
-        Align = TD->getPrefTypeAlignment(ObjectType);
-      else
-        Align = TD->getABITypeAlignment(ObjectType);
+      if (GlobalVariable *GVar = dyn_cast<GlobalVariable>(GV)) {
+        Type *ObjectType = GVar->getType()->getElementType();
+        // If the object is defined in the current Module, we'll be giving
+        // it the preferred alignment. Otherwise, we have to assume that it
+        // may only have the minimum ABI alignment.
+        if (!GVar->isDeclaration() && !GVar->mayBeOverridden())
+          Align = TD->getPreferredAlignment(GVar);
+        else
+          Align = TD->getABITypeAlignment(ObjectType);
+      }

     }
     if (Align > 0)
       KnownZero = Mask & APInt::getLowBitsSet(BitWidth,
     }
     if (Align > 0)
       KnownZero = Mask & APInt::getLowBitsSet(BitWidth,

diff --git a/lib/CodeGen/SelectionDAG/SelectionDAG.cpp b/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
index 024a16373711f5b00bf0ed00e251cb4a2bc72be2..aa06955689ead145064ed91cda76155f6561e754 100644 (file)
--- a/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
+++ b/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
@@ -6553,24 +6553,15 @@ unsigned SelectionDAG::InferPtrAlignment(SDValue Ptr) const {
   const GlobalValue *GV;
   int64_t GVOffset = 0;
   if (TLI.isGAPlusOffset(Ptr.getNode(), GV, GVOffset)) {
   const GlobalValue *GV;
   int64_t GVOffset = 0;
   if (TLI.isGAPlusOffset(Ptr.getNode(), GV, GVOffset)) {

-    // If GV has specified alignment, then use it. Otherwise, use the preferred
-    // alignment.
-    unsigned Align = GV->getAlignment();
-    if (!Align) {
-      if (const GlobalVariable *GVar = dyn_cast<GlobalVariable>(GV)) {
-        if (GVar->hasInitializer()) {
-          const TargetData *TD = TLI.getTargetData();
-          Align = TD->getPreferredAlignment(GVar);
-        }
-      }
-      if (!Align)
-        // Conservatively returns zero here instead of using ABI alignment for
-        // type of the GV. If the type is a "packed" type, then the under-
-        // specified alignments is attached to the load / store instructions.
-        // In that case, the alignment of the type cannot be trusted.
-        return 0;
-    }
-    return MinAlign(Align, GVOffset);
+    unsigned PtrWidth = TLI.getPointerTy().getSizeInBits();
+    APInt AllOnes = APInt::getAllOnesValue(PtrWidth);
+    APInt KnownZero(PtrWidth, 0), KnownOne(PtrWidth, 0);
+    llvm::ComputeMaskedBits(const_cast<GlobalValue*>(GV), AllOnes,
+                            KnownZero, KnownOne, TLI.getTargetData());
+    unsigned AlignBits = KnownZero.countTrailingOnes();
+    unsigned Align = AlignBits ? 1 << std::min(31U, AlignBits) : 0;
+    if (Align)
+      return MinAlign(Align, GVOffset);

   }
 
   // If this is a direct reference to a stack slot, use information about the
   }
 
   // If this is a direct reference to a stack slot, use information about the

diff --git a/test/Transforms/InstCombine/constant-fold-gep.ll b/test/Transforms/InstCombine/constant-fold-gep.ll
index c679226d4a9d9d6884a22befa873851d7a699c67..e5b16ea0ffdc22a3ec262d98318cb8792da3aa83 100644 (file)
--- a/test/Transforms/InstCombine/constant-fold-gep.ll
+++ b/test/Transforms/InstCombine/constant-fold-gep.ll
@@ -9,7 +9,7 @@ target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 @Y = internal global [3 x %struct.X] zeroinitializer
 
 define void @frob() {
 @Y = internal global [3 x %struct.X] zeroinitializer
 
 define void @frob() {

-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 0), align 8
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 0), align 16

   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 0), align 4
 ; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 1), align 4
   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 1), align 4
   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 0), align 4
 ; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 1), align 4
   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 1), align 4


@@ -33,7 +33,7 @@ define void @frob() {
   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 10), align 4
 ; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 1, i32 1, i64 2), align 4
   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 11), align 4
   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 10), align 4
 ; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 1, i32 1, i64 2), align 4
   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 11), align 4

-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 2, i32 0, i64 0), align 8
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 2, i32 0, i64 0), align 16

   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 12), align 4
 ; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 2, i32 0, i64 1), align 4
   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 13), align 4
   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 12), align 4
 ; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 2, i32 0, i64 1), align 4
   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 13), align 4


@@ -47,7 +47,7 @@ define void @frob() {
   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 17), align 8
 ; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 1, i64 0, i32 0, i64 0), align 8
   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 18), align 8
   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 17), align 8
 ; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 1, i64 0, i32 0, i64 0), align 8
   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 18), align 8

-; CHECK: store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 2, i64 0, i32 0, i64 0), align 8
+; CHECK: store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 2, i64 0, i32 0, i64 0), align 16

   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 36), align 8
 ; CHECK: store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 1, i64 0, i32 0, i64 1), align 8
   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 19), align 8
   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 36), align 8
 ; CHECK: store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 1, i64 0, i32 0, i64 1), align 8
   store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 19), align 8