[LLVMbugs] [Bug 18322] New: test/Headers/cxx11.cpp is failing on OS X 10.9 Mavericks

bugzilla-daemon at llvm.org bugzilla-daemon at llvm.org
Tue Dec 24 12:40:47 PST 2013


            Bug ID: 18322
           Summary: test/Headers/cxx11.cpp is failing on OS X 10.9
           Product: clang
           Version: trunk
          Hardware: PC
                OS: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++11
          Assignee: unassignedclangbugs at nondot.org
          Reporter: nicolasweber at gmx.de
                CC: dgregor at apple.com, llvmbugs at cs.uiuc.edu
    Classification: Unclassified


On Mon, Dec 2, 2013 at 5:58 PM, Rafael EspĂ­ndola <rafael.espindola at gmail.com>
> looks like this found a real issue with OS X 10.9's headers.

I agree that this looks like a bug in the headers; it looks like
sys/_types/_wchar_t.h should probably not be specified as a separate submodule,
since including it in an 'extern "C" {' suggests that it's just a fragment of
text rather than an actual submodule.

> Should we just xfail the test?

That fixes the test, but does nothing for people who are using Clang SVN on Mac
OS with older (incompatible) modules. We need to figure out how to handle that
case. Maybe we should include a version number in module.map files, and ignore
module.maps that don't specify the right version?

> Including a system header (<stdint.h>) in a
> clang test seems like a bad idea, no?

Well... the point of this test is to ensure that we do the right thing when our
own <stdint.h> #include_next's the system one. We should probably change the
test to get that effect a different way (for instance, by providing a fake
sysroot with a fake <stdint.h>).

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/20131224/ae1f9c92/attachment.html>

More information about the llvm-bugs mailing list