On Fri, Nov 29, 2019 at 12:43 PM Neil Horman <nhor...@tuxdriver.com> wrote:
>
> On Wed, Nov 27, 2019 at 09:45:46PM +0100, David Marchand wrote:
> > On Tue, Nov 26, 2019 at 3:22 PM Neil Horman <nhor...@tuxdriver.com> wrote:
> > > On Mon, Nov 25, 2019 at 05:13:14PM +0100, David Marchand wrote:
> > > > So far, we did not pay attention to direct access to variables but they
> > > > are part of the API/ABI too and should be clearly identified.
> > > >
> > > > Introduce a __rte_experimental_var tag and mark existing variables.
> > > >
> > > > Fixes: a4bcd61de82d ("buildtools: add script to check experimental API
> > > > exports")
> > > > Cc: sta...@dpdk.org
> > > >
> > > > Signed-off-by: David Marchand <david.march...@redhat.com>
> > > > ---
> > > > Quick patch to try to catch experimental variables.
> > > > Not sure if we could use a single section, so please advise if there is
> > > > better to do about this.
> > > >
> > > I don't see any definition of __rte_experimental_var here, won't the
> > > preprocessor choke on this when you try to compile without that?
> >
> > Sorry, not getting your point.
> > If there is an issue, then it is the same as __rte_experimental.
> >
> No, there is no issue, I'm just blind. For some reason cscope wasn't finding
> the definition of __rte_experimental_var when I applied your patch for me.
> Its
> clear you have it below.
Ok, thanks for confirming.
> Acked-by: Neil Horman <nhor...@tuxdriver.com>
I will do an extra pass and submit a non RFC patch with your ack if I
have no change.
Thanks Neil.
--
David Marchand
