[cfe-commits] [PATCH] Support gcc 4.4 C++0x builtins

[Howard Hinnant]

On Feb 12, 2011, at 2:03 AM, Jeffrey Yasskin wrote:

> Thanks! I used this to build some more google code and ran into a
> couple more missing builtins from <complex>. Here's a patch that adds
> those too, which I've also attached to your bug.
> 
> On Wed, Feb 9, 2011 at 1:53 AM, John Bytheway <jbytheway at gmail.com> wrote:
>> The attached patch adds support for the builtins defined in gcc 4.4's
>> <cmath> header in C++0x mode.  There is ample opportunity for error in
>> this patch, but it's been checked by both me and Marius Wachtler.  See
>> 
>> http://llvm.org/bugs/show_bug.cgi?id=9138
>> 
>> (I was asked to send the patch here after attaching it to the bug; this
>> one corrects the error Marius noticed in the one attached to the bug.)
>> 
>> Note that I've added builtins purely on the basis of error messages
>> (either because their lack was indicated by an error, or because they
>> were the "<foo>" version where the "<foo>f" and "<foo>l" versions
>> appeared in errors).  I haven't looked at the standards to see what
>> functions really should exist.
>> 
>> This work was inspired by
>> <http://lists.cs.uiuc.edu/pipermail/cfe-dev/2011-February/013168.html>
>> and related messages.
>> 
>> John Bytheway
>> 
>> _______________________________________________
>> cfe-commits mailing list
>> cfe-commits at cs.uiuc.edu
>> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
>> 
>> 
> <cxx0x_builtins.patch>_______________________________________________
> cfe-commits mailing list
> cfe-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits

Here's a boat-load of C++ complex transcendental tests that could be adapted to C99 complex:

http://llvm.org/svn/llvm-project/libcxx/trunk/test/numerics/complex.number/complex.transcendentals/

-Howard