[PATCH] D40948: Switch Clang's default C++ language target to C++14

[Tim Northover via Phabricator via cfe-commits]

t.p.northover added inline comments.


================
Comment at: clang/test/CodeGenCXX/new-overflow.cpp:88
   // CHECK:      [[N:%.*]] = sext i16 {{%.*}} to i32
-  // CHECK-NEXT: [[T0:%.*]] = icmp slt i32 [[N]], 0
-  // CHECK-NEXT: [[T1:%.*]] = select i1 [[T0]], i32 -1, i32 [[N]]
-  // CHECK-NEXT: call i8* @_Znaj(i32 [[T1]])
+  // CHECK-NEXT: call i8* @_Znaj(i32 [[N]])
   // CHECK:      getelementptr inbounds {{.*}}, i32 [[N]]
----------------
rsmith wrote:
> The changes in this file are a regression; C++14 requires us to check whether the array bound prior to promotion is negative. Can you file a bug on that?
I've filed https://llvm.org/PR35573. Not quite sure what to do about this test until it's fixed though. Add a second RUN line to check both variants and then XFAIL it?


================
Comment at: clang/test/CodeGenCXX/vtable-available-externally.cpp:275
 struct C {
   virtual D& operator=(const D&);
 };
----------------
rsmith wrote:
> To make this test work in C++11 onwards, you need to add a virtual move assignment operator here:
> 
> ```
> virtual D& operator=(D&&);
> ```
That didn't quite work. The issue appears to be that D has both of those implicitly defined in C++14 mode, but only the move assignment operator is used below. Speculative VTable emission requires all of them to be used.

So adding a "d = d;" line to the second g function fixes the test. Does that sound sane to you, or am I missing the point?


================
Comment at: clang/test/SemaCXX/unknown-type-name.cpp:98
 
-template<typename T> int junk1(T::junk); // expected-warning{{variable templates are a C++14 extension}}
+template<typename T> int junk1(T::junk);
+#if __cplusplus <= 201103L
----------------
rsmith wrote:
> Huh, we should probably have a `-Wvexing-parse` warning for this. Can you file a bug?
I've filed https://llvm.org/PR35576. You may want to sanity check it though, I was pretty light on the detail because I wasn't sure of the exact diagnostic being proposed.


Repository:
  rC Clang

https://reviews.llvm.org/D40948