Teach SROA to cope with wrapper aggregates. These show up a lot in ABI

type coercion code, especially when targetting ARM. Things like [1
x i32] instead of i32 are very common there.

The goal of this logic is to ensure that when we are picking an alloca
type, we look through such wrapper aggregates and across any zero-length
aggregate elements to find the simplest type possible to form a type
partition.

This logic should (generally speaking) rarely fire. It only ends up
kicking in when an alloca is accessed using two different types (for
instance, i32 and float), and the underlying alloca type has wrapper
aggregates around it. I noticed a significant amount of this occurring
looking at stepanov_abstraction generated code for arm, and suspect it
happens elsewhere as well.

Note that this doesn't yet address truly heinous IR productions such as
PR14059 is concerning. Those result in mismatched *sizes* of types in
addition to mismatched access and alloca types.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@165870 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Transforms/Scalar/SROA.cpp b/lib/Transforms/Scalar/SROA.cpp
index cb6da3b95615415a169b9c122fd99295afe76c71..d6ddaaec573010c18986d45f3ccd9b3369930186 100644 (file)
--- a/lib/Transforms/Scalar/SROA.cpp
+++ b/lib/Transforms/Scalar/SROA.cpp
@@ -3049,6 +3049,36 @@ private:
 };
 }
 
+/// \brief Strip aggregate type wrapping.
+///
+/// This removes no-op aggregate types wrapping an underlying type. It will
+/// strip as many layers of types as it can without changing either the type
+/// size or the allocated size.
+static Type *stripAggregateTypeWrapping(const DataLayout &DL, Type *Ty) {
+  if (Ty->isSingleValueType())
+    return Ty;
+
+  uint64_t AllocSize = DL.getTypeAllocSize(Ty);
+  uint64_t TypeSize = DL.getTypeSizeInBits(Ty);
+
+  Type *InnerTy;
+  if (ArrayType *ArrTy = dyn_cast<ArrayType>(Ty)) {
+    InnerTy = ArrTy->getElementType();
+  } else if (StructType *STy = dyn_cast<StructType>(Ty)) {
+    const StructLayout *SL = DL.getStructLayout(STy);
+    unsigned Index = SL->getElementContainingOffset(0);
+    InnerTy = STy->getElementType(Index);
+  } else {
+    return Ty;
+  }
+
+  if (AllocSize > DL.getTypeAllocSize(InnerTy) ||
+      TypeSize > DL.getTypeSizeInBits(InnerTy))
+    return Ty;
+
+  return stripAggregateTypeWrapping(DL, InnerTy);
+}
+
 /// \brief Try to find a partition of the aggregate type passed in for a given
 /// offset and size.
 ///
@@ -3065,7 +3095,7 @@ private:
 static Type *getTypePartition(const DataLayout &TD, Type *Ty,
                               uint64_t Offset, uint64_t Size) {
   if (Offset == 0 && TD.getTypeAllocSize(Ty) == Size)
-    return Ty;
+    return stripAggregateTypeWrapping(TD, Ty);
 
   if (SequentialType *SeqTy = dyn_cast<SequentialType>(Ty)) {
     // We can't partition pointers...
@@ -3094,7 +3124,7 @@ static Type *getTypePartition(const DataLayout &TD, Type *Ty,
     assert(Offset == 0);
 
     if (Size == ElementSize)
-      return ElementTy;
+      return stripAggregateTypeWrapping(TD, ElementTy);
     assert(Size > ElementSize);
     uint64_t NumElements = Size / ElementSize;
     if (NumElements * ElementSize != Size)
@@ -3130,7 +3160,7 @@ static Type *getTypePartition(const DataLayout &TD, Type *Ty,
   assert(Offset == 0);
 
   if (Size == ElementSize)
-    return ElementTy;
+    return stripAggregateTypeWrapping(TD, ElementTy);
 
   StructType::element_iterator EI = STy->element_begin() + Index,
                                EE = STy->element_end();

diff --git a/test/Transforms/SROA/alignment.ll b/test/Transforms/SROA/alignment.ll
index 945ad9100210b43be34eaf95caad6c0e9c4b2a20..ad5fb6c4a5d8c1b4c1dd8f8c54760ba667da28fe 100644 (file)
--- a/test/Transforms/SROA/alignment.ll
+++ b/test/Transforms/SROA/alignment.ll
@@ -84,37 +84,6 @@ entry:
   ret void
 }
 
-%struct.S = type { i8, { i64 } }
-
-define void @test4() {
-; This test case triggered very strange alignment behavior with memcpy due to
-; strange splitting. Reported by Duncan.
-; CHECK: @test4
-
-entry:
-  %D.2113 = alloca %struct.S
-  %Op = alloca %struct.S
-  %D.2114 = alloca %struct.S
-  %gep1 = getelementptr inbounds %struct.S* %Op, i32 0, i32 0
-  store i8 0, i8* %gep1, align 8
-  %gep2 = getelementptr inbounds %struct.S* %Op, i32 0, i32 1, i32 0
-  %cast = bitcast i64* %gep2 to double*
-  store double 0.000000e+00, double* %cast, align 8
-  store i64 0, i64* %gep2, align 8
-  %dst1 = bitcast %struct.S* %D.2114 to i8*
-  %src1 = bitcast %struct.S* %Op to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst1, i8* %src1, i32 16, i32 8, i1 false)
-  %dst2 = bitcast %struct.S* %D.2113 to i8*
-  %src2 = bitcast %struct.S* %D.2114 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst2, i8* %src2, i32 16, i32 8, i1 false)
-; We get 3 memcpy calls with various reasons to shrink their alignment to 1.
-; CHECK: @llvm.memcpy.p0i8.p0i8.i32(i8* %{{.*}}, i8* %{{.*}}, i32 3, i32 1, i1 false)
-; CHECK: @llvm.memcpy.p0i8.p0i8.i32(i8* %{{.*}}, i8* %{{.*}}, i32 8, i32 1, i1 false)
-; CHECK: @llvm.memcpy.p0i8.p0i8.i32(i8* %{{.*}}, i8* %{{.*}}, i32 11, i32 1, i1 false)
-
-  ret void
-}
-
 define void @test5() {
 ; Test that we preserve underaligned loads and stores when splitting.
 ; CHECK: @test5

diff --git a/test/Transforms/SROA/basictest.ll b/test/Transforms/SROA/basictest.ll
index e7767ef5e967dd5665d23d00e0b0f642714f9dec..7d8e5cdf86f2ac71d2c6b86e79aee8a58fbcde08 100644 (file)
--- a/test/Transforms/SROA/basictest.ll
+++ b/test/Transforms/SROA/basictest.ll
@@ -968,3 +968,47 @@ entry:
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %cast0, i8* %cast1, i32 12, i32 0, i1 false)
   ret void
 }
+
+define i32 @test22(i32 %x) {
+; Test that SROA and promotion is not confused by a grab bax mixture of pointer
+; types involving wrapper aggregates and zero-length aggregate members.
+; CHECK: @test22
+
+entry:
+  %a1 = alloca { { [1 x { i32 }] } }
+  %a2 = alloca { {}, { float }, [0 x i8] }
+  %a3 = alloca { [0 x i8], { [0 x double], [1 x [1 x <4 x i8>]], {} }, { { {} } } }
+; CHECK-NOT: alloca
+
+  %wrap1 = insertvalue [1 x { i32 }] undef, i32 %x, 0, 0
+  %gep1 = getelementptr { { [1 x { i32 }] } }* %a1, i32 0, i32 0, i32 0
+  store [1 x { i32 }] %wrap1, [1 x { i32 }]* %gep1
+
+  %gep2 = getelementptr { { [1 x { i32 }] } }* %a1, i32 0, i32 0
+  %ptrcast1 = bitcast { [1 x { i32 }] }* %gep2 to { [1 x { float }] }*
+  %load1 = load { [1 x { float }] }* %ptrcast1
+  %unwrap1 = extractvalue { [1 x { float }] } %load1, 0, 0
+
+  %wrap2 = insertvalue { {}, { float }, [0 x i8] } undef, { float } %unwrap1, 1
+  store { {}, { float }, [0 x i8] } %wrap2, { {}, { float }, [0 x i8] }* %a2
+
+  %gep3 = getelementptr { {}, { float }, [0 x i8] }* %a2, i32 0, i32 1, i32 0
+  %ptrcast2 = bitcast float* %gep3 to <4 x i8>*
+  %load3 = load <4 x i8>* %ptrcast2
+  %valcast1 = bitcast <4 x i8> %load3 to i32
+
+  %wrap3 = insertvalue [1 x [1 x i32]] undef, i32 %valcast1, 0, 0
+  %wrap4 = insertvalue { [1 x [1 x i32]], {} } undef, [1 x [1 x i32]] %wrap3, 0
+  %gep4 = getelementptr { [0 x i8], { [0 x double], [1 x [1 x <4 x i8>]], {} }, { { {} } } }* %a3, i32 0, i32 1
+  %ptrcast3 = bitcast { [0 x double], [1 x [1 x <4 x i8>]], {} }* %gep4 to { [1 x [1 x i32]], {} }*
+  store { [1 x [1 x i32]], {} } %wrap4, { [1 x [1 x i32]], {} }* %ptrcast3
+
+  %gep5 = getelementptr { [0 x i8], { [0 x double], [1 x [1 x <4 x i8>]], {} }, { { {} } } }* %a3, i32 0, i32 1, i32 1, i32 0
+  %ptrcast4 = bitcast [1 x <4 x i8>]* %gep5 to { {}, float, {} }*
+  %load4 = load { {}, float, {} }* %ptrcast4
+  %unwrap2 = extractvalue { {}, float, {} } %load4, 1
+  %valcast2 = bitcast float %unwrap2 to i32
+
+  ret i32 %valcast2
+; CHECK: ret i32
+}