On 04/05/2012 07:05 AM, Daniel P. Berrange wrote:
> On Thu, Apr 05, 2012 at 06:07:20AM -0600, Eric Blake wrote:
>> On 04/05/2012 04:10 AM, Bruno Haible wrote:
>> You forgot to check one (64-bit mingw64):
>>
>> $ x86_64-w64-mingw32-gcc -E foo.c | grep size_t
>> __extension__ typedef unsigned long long size_t;
>> __extension__ typedef long long ssize_t;
>> size_t a;
>> ssize_t b;
>>
>> but apparently that is also working.  Then how do I explain the compiler
>> warning that Daniel was seeing under mingw64?  Daniel, were you testing
>> with x86_64-w64-mingw32-gcc or i686-w64-mingw32-gcc?  Can you come up
>> with a minimal test case that shows the issue you had?
> 
> I was testing mingw32, not mingw64, since the former is what is in
> Fedora 16

That explains it, then.  As Bruno pointed out, the bug is in mingw32's
<sys/types.h>, and is NOT present in either the 32-bit or 64-bit mingw64
project.  mingw32 is starting to lose ground, now that Fedora 17 ships
mingw64.

-- 
Eric Blake   ebl...@redhat.com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org

Attachment: signature.asc
Description: OpenPGP digital signature

Reply via email to