On Wed, Oct 07, 2009 at 03:08:24PM +0100, Daniel P. Berrange wrote: > On Wed, Oct 07, 2009 at 03:56:12PM +0200, Daniel Veillard wrote: > > On Wed, Oct 07, 2009 at 11:42:19AM +0100, Daniel P. Berrange wrote: > > > * daemon/Makefile.am: Fix missing sasl rule > > > * src/datatypes.c: Add unistd.h to avoid gnulib bug > > > * src/util/cgroup.c: Disable mntent if not available > > > --- > > > > ACK, also unification of type we use long long in libvirt to denotate > > 64 bits ints, it's actually the largest change in that patch. > > Opps, I should have mentioned why I did that too... > > The PRIi64 constants is intended to expand to the OS specific printf > format required for a 64 bit integer type. All Win32 runtimes have > broken 64-bit integer printf support - their definition of PRIi64 > in fact wants a type of 'long' which is only 32bits. libvirt though > uses GNULIB which provides a replacement printf impl which works > properly, but it does not replace the PRIi64 constant which still > refers to the Microsoft impl's requirements. Thus we switch to just > using the fixed %llu which we know GNULIB guarentees to support, so > does not need "portability" constants. Of course using int64_t > with %llu then generates a warning, unless you cast to unsigned long > long. So in the end the simple solution is just to say > > %llu and unsigned long long > > everywhere, and ignore all this portability stuff which is not in fact > portable or reliable Okay, I basically guessed so except I didn;t know the details of the PRIx64 breakages, thanks ! Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel@xxxxxxxxxxxx | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/ -- Libvir-list mailing list Libvir-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libvir-list