On Fri, Jul 31, 2015 at 01:19:11PM +0100, Peter Maydell wrote: > On 31 July 2015 at 13:06, Daniel P. Berrange <berra...@redhat.com> wrote: > > In fixing the mingw64 problem wrt to localtime_r availability, I relied > > on the fact that qemu-common.h is supposed to be included everywhere, > > to guarantee that we always have unistd.h included before time.h: > > It's not really supposed to be included everywhere. It's just a > convenient way for a .c file to get a lot of stuff all at once. > It's a bit of a mess. > > > So I'm wondering if there is appetite for cleaning this and and > > introducing standard practice for inclusion of qemu-common.h ? > > You might want to have a look at my series which tries to > make osdep.h the "anybody can include this to get the really > key 'breaks if this isn't here' stuff". > > http://lists.nongnu.org/archive/html/qemu-devel/2015-07/msg04593.html
Ah, interesting, I'll check out that series - it sounds like it is similar to what i was thinking Regards, Daniel -- |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
