I defined a separate ovs_mutex_init_recursive(). I sent out a v2 in
case you want to look:
http://openvswitch.org/pipermail/dev/2013-August/030895.html
On Sat, Aug 17, 2013 at 05:21:41PM +0800, Ethan Jackson wrote:
> Oops yes you're right. It might be worth defining a separate function for
> that, but its your call. The patch is good as is.
>
> Ethan (iPhone)
>
> On Aug 17, 2013, at 13:08, Ben Pfaff <b...@nicira.com> wrote:
>
> >
> > On Aug 16, 2013 9:47 PM, "Ethan Jackson" <et...@nicira.com> wrote:
> > > Personally I'd prefer we change ovs_mutex_init() to not take a mutex
> > > type as an argument and simply use the error checking mutex always. I
> > > think it's going to be a long time before we actually need to
> > > configure this on a per mutex basis, and I don't think we know what
> > > abstraction we'll need at that point today.
> >
> > Don't we need the ability to initialize a recursive mutex? It didn't occur
> > to me to drop the parameter but that's the only current reason to keep it.
> >
> > Thanks,
> >
> > Ben.
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
