On 2014-09-30 11:36:55, John Johansen wrote: > On 09/30/2014 10:28 AM, Steve Beattie wrote: > > On Wed, Sep 24, 2014 at 04:00:12PM -0700, Seth Arnold wrote: > >> On Wed, Sep 24, 2014 at 03:45:46PM -0700, Steve Beattie wrote: > >>> Not at all. Honestly, my preferences are: > >>> > >>> 1) __unused — simple, clear, straightforward annotation. Alas, > >>> conflicts with a symbol in older glibcs and with convention of __ > >>> prefixed symbols being system synbols. I wish gcc just went ahead > >>> and defined this. > >>> > >>> 2) __aa_unused — nearly as simple as __unused; the underscores > >>> indicate specialness, the __aa_ prefix makes it less likely to > >>> conflict with any actual system symbol name, despite intruding on > >>> the system symbol convention. NACK'ed by Christian, so would need > >>> at least one more ACK to be acceptable. (_aa_unused has the same > >>> problem, standards-wise,) > >> > >> Sorry Christian. :) > >> > >> __aa_unused please. > >> > >> Acked-by: Seth Arnold <[email protected]> > > > > Right, I have your ack, which counter balances Christian's > > nack. But I'd need one more to commit (I have mine and yours, if > > I'm understanding http://wiki.apparmor.net/index.php/CommitPolicy > > correctly). > > > > What do people think about using just the bare word? > unused > > Since the macro is not being exposed by a public header file the > potential set of namespace collisions is limited to within apparmor > it self, and we can deal with that.
I'm ok with using the bare word. I agree that we should not expose such a macro publicly. Tyler
signature.asc
Description: Digital signature
-- AppArmor mailing list [email protected] Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/apparmor
