On Mon, Dec 21, 2015 at 04:45:20PM -0800, Joe Stringer wrote: > On 21 December 2015 at 16:11, Joe Stringer <j...@ovn.org> wrote: > > On 21 December 2015 at 16:01, Ben Pfaff <b...@ovn.org> wrote: > >> On Mon, Dec 21, 2015 at 03:56:40PM -0800, Joe Stringer wrote: > >>> The previous definitions of these variables using designated > >>> initializers caused a variety of issues when attempting to compile with > >>> MSVC, particularly if including these headers from C++ code. By defining > >>> them like this, we can appease MSVC and keep the definitions the same on > >>> all platforms. > >>> > >>> VMware-BZ: #1517163 > >>> Suggested-by: Yin Lin <li...@vmware.com> > >>> Signed-off-by: Joe Stringer <j...@ovn.org> > >> > >> Why not just use a new-enough MSVC? According to this discussion of the > >> topic, MSVC 2013 and newer support designated initializers in C++: > >> > >> http://stackoverflow.com/questions/18731707/why-does-c11-not-support-designated-initializer-list-as-c99 > > > > We have encountered this issue on MSVC2013u3 and MSVC2015u1. As far as > > I can tell, the compiler will accept them if it is compiling a .c > > file, but it fails if the header is included from a .cpp file. > > This page seems to confirm that designated initializers are only > supported in C mode: > http://blogs.msdn.com/b/vcblog/archive/2015/04/29/c-11-14-17-features-in-vs-2015-rc.aspx
It is a shame that Microsoft doesn't bother to actually document any of this, at least not accurately. (Their one page on "extensions" documents // comments as an extension, even though they've been part of C for the past 16 years!) Apparently we're supposed to refer to a blog for documentation? Ugh. You'd better add a comment so that no one changes this back to a sane definition. Acked-by: Ben Pfaff <b...@ovn.org> _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
