diff --git a/lib/CodeGen/TargetInfo.cpp b/lib/CodeGen/TargetInfo.cpp
index 604e6cb..09aec44 100644
--- a/lib/CodeGen/TargetInfo.cpp
+++ b/lib/CodeGen/TargetInfo.cpp
@@ -2484,7 +2484,8 @@
     // FIXME: Cleanup.
     assert(AI.isDirect() && "Unexpected ABI info for mixed regs");
     llvm::StructType *ST = cast<llvm::StructType>(AI.getCoerceToType());
-    llvm::Value *Tmp = CGF.CreateTempAlloca(ST);
+    llvm::Value *Tmp = CGF.CreateMemTemp(Ty);
+    Tmp = CGF.Builder.CreateBitCast(Tmp, ST->getPointerTo());
     assert(ST->getNumElements() == 2 && "Unexpected ABI info for mixed regs");
     llvm::Type *TyLo = ST->getElementType(0);
     llvm::Type *TyHi = ST->getElementType(1);
@@ -2508,6 +2509,18 @@
     RegAddr = CGF.Builder.CreateGEP(RegAddr, gp_offset);
     RegAddr = CGF.Builder.CreateBitCast(RegAddr,
                                         llvm::PointerType::getUnqual(LTy));
+
+    // Copy to a temporary if necessary to ensure the appropriate alignment.
+    std::pair<CharUnits, CharUnits> SizeAlign =
+        CGF.getContext().getTypeInfoInChars(Ty);
+    uint64_t TySize = SizeAlign.first.getQuantity();
+    unsigned TyAlign = SizeAlign.second.getQuantity();
+    if (TyAlign > 8) {
+      RegAddr = CGF.Builder.CreateGEP(RegAddr, gp_offset);
+      llvm::Value *Tmp = CGF.CreateMemTemp(Ty);
+      CGF.Builder.CreateMemCpy(Tmp, RegAddr, TySize, 8, false);
+      RegAddr = Tmp;
+    }
   } else if (neededSSE == 1) {
     RegAddr = CGF.Builder.CreateGEP(RegAddr, fp_offset);
     RegAddr = CGF.Builder.CreateBitCast(RegAddr,
@@ -2521,9 +2534,9 @@
     llvm::Type *DoubleTy = CGF.DoubleTy;
     llvm::Type *DblPtrTy =
       llvm::PointerType::getUnqual(DoubleTy);
-    llvm::StructType *ST = llvm::StructType::get(DoubleTy,
-                                                       DoubleTy, NULL);
-    llvm::Value *V, *Tmp = CGF.CreateTempAlloca(ST);
+    llvm::StructType *ST = llvm::StructType::get(DoubleTy, DoubleTy, NULL);
+    llvm::Value *V, *Tmp = CGF.CreateMemTemp(Ty);
+    Tmp = CGF.Builder.CreateBitCast(Tmp, ST->getPointerTo());
     V = CGF.Builder.CreateLoad(CGF.Builder.CreateBitCast(RegAddrLo,
                                                          DblPtrTy));
     CGF.Builder.CreateStore(V, CGF.Builder.CreateStructGEP(Tmp, 0));
diff --git a/test/CodeGen/x86_64-arguments.c b/test/CodeGen/x86_64-arguments.c
index 518ee84..e94ddb9 100644
--- a/test/CodeGen/x86_64-arguments.c
+++ b/test/CodeGen/x86_64-arguments.c
@@ -392,3 +392,13 @@
 // CHECK:      [[T0:%.*]] = load double*
 // CHECK-NEXT: [[T1:%.*]] = load double*
 // CHECK-NEXT: call void (double, double, ...)* bitcast (void (...)* @test50_helper to void (double, double, ...)*)(double [[T0]], double [[T1]])
+
+struct test51_s { __uint128_t intval; };
+void test51(struct test51_s *s, __builtin_va_list argList) {
+    *s = __builtin_va_arg(argList, struct test51_s);
+}
+
+// CHECK: define void @test51
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* {{.*}}, i8* {{.*}}, i64 16, i32 8, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* {{.*}}, i8* {{.*}}, i64 16, i32 16, i1 false)
+// CHECK-NEXT: ret void