RSB build issue with rtems-libbsd

[Vijay Kumar Banerjee]

On Wed, Mar 10, 2021 at 11:01 AM Chris Johns <chrisj at rtems.org> wrote:
>
> On 9/3/21 3:53 pm, Vijay Kumar Banerjee wrote:
> > On Mon, Mar 8, 2021 at 7:57 PM Chris Johns <chrisj at rtems.org> wrote:
> >>
> >> On 9/3/21 1:44 pm, Vijay Kumar Banerjee wrote:
> >>> On Mon, Mar 8, 2021 at 3:14 PM Joel Sherrill <joel at rtems.org> wrote:
> >>>> On Mon, Mar 8, 2021, 4:02 PM Chris Johns <chrisj at rtems.org> wrote:
> >>>>>
> >>>>> On 6/3/21 2:24 pm, Vijay Kumar Banerjee wrote:
> >>>>>> Hello,
> >>>>>>
> >>>>>> I'm getting the following error when trying to build rtems-libbsd from
> >>>>>> RSB. Is there anything that I'm missing?
> >>>>>>
> >>>>>> ```
> >>>>>> ../source-builder/sb-set-builder --prefix=$RTEMS6_PREFIX
> >>>>>> 6/rtems-libbsd --host=powerpc-rtems6 --with-rtems-bsp=beatnik
> >>>>>> RTEMS Source Builder - Set Builder, 6 (102cb1e6450f)
> >>>>>> warning: exe: absolute exe found in path: (__chown) /usr/sbin/chown
> >>>>>> Build Set: 6/rtems-libbsd
> >>>>>> config: tools/rtems-libbsd-6.cfg
> >>>>>> error: rtems-bsp.cfg:155: invalid %if operator:  -mcpu=7400
> >>>>>> -I/home/vijay/development/rtems/6/powerpc-rtems6/beatnik/lib/include
> >>>>>
> >>>>> This looks like a bug. Please raise a ticket.
> >>>>>
> >>>>>> ==
> >>>>>> Build FAILED
> >>>>>> Build Set: Time 0:00:00.046141
> >>>>>> Build FAILED
> >>>>>>
> >>>>>> ```
> >>>>>>
> >>>>>> Commenting these checks from rtems-bsp.cfg gets it building.
> >>>>>
> >>>>> Unfortunately the checks are needed and this is not the answer.
> >>>>>
> >>>>> I wonder how Jennifer has been building libbsd. She is working on this BSP and
> >>>>> libbsd.
> >>>>
> >>>
> >>> This issue is happening with other bsps as well.
> >>
> >> Great and thanks for checking. I asked this question in the ticket. :)
> >>
> >>> This error is also being a blocker to building rtems-net-legacy with
> >>> the new patch. I can have a look at it, do you have any hints on what
> >>> might be going wrong?
> >>
> >> Yes in the ticket :) (in short for the thread run with --trace and compare with
> >> libbsd).
> >>
> > Hi,
> >
> > I think I found it! :
>
> I saw in the patch. A great find. It has been a long time since we have had a
> logic problem in the config parser. This is excellent debugging work.
>
Thanks! I always wanted to be able to find a single character bug with
a big impact. :D

Best regards,
Vijay
> Chris