2 files changed, 22 insertions, 3 deletions
diff --git a/clang/lib/CodeGen/TargetInfo.cpp b/clang/lib/CodeGen/TargetInfo.cpp
index 7c850afbf8f..6d5e38daa89 100644
--- a/clang/lib/CodeGen/TargetInfo.cpp
+++ b/clang/lib/CodeGen/TargetInfo.cpp
@@ -1198,6 +1198,13 @@ void X86_64ABIInfo::classify(QualType Ty, uint64_t OffsetBase,
     Current = NoClass;
     uint64_t EltSize = getContext().getTypeSize(AT->getElementType());
     uint64_t ArraySize = AT->getSize().getZExtValue();
+
+    // The only case a 256-bit wide vector could be used is when the array
+    // contains a single 256-bit element. Since Lo and Hi logic isn't extended
+    // to work for sizes wider than 128, early check and fallback to memory.
+    if (Size > 128 && EltSize != 256)
+      return;
+
     for (uint64_t i=0, Offset=OffsetBase; i<ArraySize; ++i, Offset += EltSize) {
       Class FieldLo, FieldHi;
       classify(AT->getElementType(), Offset, FieldLo, FieldHi);
diff --git a/clang/test/CodeGen/x86_64-arguments.c b/clang/test/CodeGen/x86_64-arguments.c
index 3d4016db829..b0218351657 100644
--- a/clang/test/CodeGen/x86_64-arguments.c
+++ b/clang/test/CodeGen/x86_64-arguments.c
@@ -277,9 +277,9 @@ void f38(s256 x);
 void f37(__m256 x);
 void f39() { f38(x38); f37(x37); }
 
-// Make sure that the struct below is passed in the same way
-// regardless of avx being used
-//
+// The two next tests make sure that the struct below is passed
+// in the same way regardless of avx being used
+
 // CHECK: declare void @func40(%struct.t128* byval align 16)
 typedef float __m128 __attribute__ ((__vector_size__ (16)));
 typedef struct t128 {
@@ -292,3 +292,15 @@ void func41(two128 s) {
   func40(s);
 }
 
+// CHECK: declare void @func42(%struct.t128_2* byval align 16)
+typedef struct xxx {
+  __m128 array[2];
+} Atwo128;
+typedef struct t128_2 {
+  Atwo128 x;
+} SA;
+
+extern void func42(SA s);
+void func43(SA s) {
+  func42(s);
+}