My questions are:
Is it normal to have those XFAIL instances in the test?
Should it have been a perfect check? In such case: How do I proceed in order fix the XFAILs?
What about the panic: Segmentation fault?
Am 24.01.2017 um 14:33 schrieb dariodematties:
> To tell you the truth, I don't know how to interpret that information.
> Yet, it seems there is no NVIDIA tag there.
> What is the best way to proceed?
I also use the i915 driver without problems. Perhaps you've hit this bug?
Everything seems to be ok so far.
Yet, when I run make check, I keep getting:
libinterp/dldfcn/__osmesa_print__.cc-tst ....................panic: Segmentation fault -- stopping myself...
octave exited with signal 11
Should I try with make distclean?
In fact, I used make distclean and ./configure --without-osmesa, make and then make check again. I got the same error.
I did what you suggested.
I am in ~/Downloads.
Inside of ~/Downloads is octave-4.2.0 (sources).
Then mkdir octave-build.
then cd octave-build and then
../octave-4.2.0/configure --without-osmesa I've just realized about the following message from ../octave-4.2.0/configure --without-osmesa:
configure: WARNING: unrecognized options: --without-osmesa How to proceed now?
Am 25.01.2017 um 20:04 schrieb dariodematties:
> then *cd octave-build* and then
> *../octave-4.2.0/configure --without-osmesa*
> I've just realized about the following message from
> *../octave-4.2.0/configure --without-osmesa*:
> *configure: WARNING: unrecognized options: --without-osmesa*
> How to proceed now?
I'm a little bit puzzled because the many * in your output.
Of course the call to configure has to be:
without * at the end
what does "../octave-4.2.0/configure --help | grep osmesa"
without quotes return when executed?