[llvm-bugs] [Bug 34810] New: Mangled name for std::is_same within function identifier changes when including <complex>

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=34810

            Bug ID: 34810
           Summary: Mangled name for std::is_same within function
                    identifier changes when including <complex>
           Product: libc++
           Version: 5.0
          Hardware: PC
                OS: All
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: All Bugs
          Assignee: unassignedclangbugs at nondot.org
          Reporter: yaghmour.shafik at gmail.com
                CC: llvm-bugs at lists.llvm.org, mclow.lists at gmail.com

Given the following code:

#include <complex>
#include <type_traits>

template <typename T>
std::enable_if_t<std::is_same<T, float>::value, float> f(const void *g) 
{ return {}; }

template float f<float>(const void *g);

int main(){ return 0; }



and using the following command line options to clang++:

-stdlib=libc++ -std=c++14

The mangled name for the f(const void *g) drops std:: from std::is_same, we
get:

__Z1fIfENSt3__19enable_ifIXsr7is_sameIT_fEE5valueEfE4typeEPKv
                            ^^^

See gotbolt example: https://godbolt.org/g/RehUpq



while dropping dropping the include of <complex> we get: 

__Z1fIfENSt9enable_ifIXsr3std7is_sameIT_fEE5valueEfE4typeEPKv
                          ^^^^

See godbolt example: https://godbolt.org/g/CkEwJw



In our larger application this is causing a linker failure. 

I don't believe this is expected behavior, we should expect the same type for
std::is_same<float,float> whether we include <complex> or not

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20171002/d5bc30bc/attachment.html>