[llvm-commits] [llvm] r44353 - Re: [llvm] r44353 /llvm/trunk/test/CFrontend/2005-02-20-AggregateSAVEEXPR.c
zhousheng00 at gmail.com
Sun Dec 2 18:13:41 PST 2007
> Message: 3
> Date: Thu, 29 Nov 2007 13:02:39 -0800
> From: Eric Christopher <echristo at apple.com>
> Subject: Re: [llvm-commits] [llvm] r44353 -
> To: CVS Commit Messages for LLVM repository <llvm-commits at cs.uiuc.edu>
> Message-ID: <D9AB5C0B-5BAC-4EEA-A9C1-A590283A06AA at apple.com>
> Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
> On Nov 27, 2007, at 1:22 AM, Eric Christopher wrote:
> >> This pretty much looks like cygwin-related stuff. Actually it needs
> >> to
> >> store two sets of headers - "native" and mingw, thus such directory.
> >> My mingw32 installation has complex.h, not mingw/complex.h.
> >> However, I
> >> don't understand, why cygwin doesn't have it's complex.h and test
> >> should
> >> use mingw-provided one.
> > Agreed, I thought it did. It has been a while since I've had a cygwin
> > box though. I can ask those guys tomorrow though.
> Anyhow, I talked to some of the cygwin guys and there's no complex
> really at all so the test should probably be disabled for cygwin right
Although cygwin has no its own complex support, it does integrate the
mingw's complex into itself.
Why not just use it?
(or xfailed so that it starts xpassing when/if someone turns on complex
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the llvm-commits