On Wed, Dec 11, 2013 at 09:16:44AM +0200, Baruch Siach wrote:
> On Wed, Dec 11, 2013 at 07:28:35AM +0100, Thomas De Schampheleire wrote:
> > Baruch Siach wrote:
> > >On Tue, Dec 10, 2013 at 08:30:07AM +0100, Thomas Petazzoni wrote:
> > >> xtensa | m4-1.4.17 | NOK | http://autobuild.buildroot.net/results/0b9b3bb1791bbc9f902a23360c4540d90a0e05ca/
> > >> xtensa | m4-1.4.17 | NOK | http://autobuild.buildroot.net/results/ffda6f3f64c612da999229226f699c01ccebe4c2/
> > >
> > >I'm struggling with this one. The configure script test for
> > >fails because it lacks '-lrt'. This causes m4 to use an internal
> > >implementation that fails at build time for some other reason (haven't
> > >the reason for that yet).
> > >
> > >How can I add -lrt to the build test command line?
> > Isn't this by adding/setting the LIBS variable before
> > calling the macro that executes the test?
> Thanks, I'll try that. This macro seems to be called from m4/spawn_h.m4.
> failures seems to be xtensa specific. I wonder how other archs handle
Well, this check seems to come from gnulib. How can I fix it then?
BTW, coreutils shows the same problem
I still don't understand why the autobuilder doesn't fail on other archs. I
can reproduce this failure on ARM with the following defconfig:
Thomas P, is there a way to see the config.log file from a successful build
m4 on the autobuilder?
These successes are quite hard to find, it seems. Of the latest 50
the m4 package is selected in only 3 configs with the following results:
Same failure on arc
Failure for other reason on x86_64
Success on sh4a, apparently using glibc newer than 2.16 that moved
into the main libc.so
~. .~ Tk Open Systems
- [email protected] - tel: +972.2.679.5364,