On 25/11/2015 23:33, Burton, Ross wrote: > > On 23 November 2015 at 23:41, Tudor Florea <tudor.flo...@enea.com > <mailto:tudor.flo...@enea.com>> wrote: > > +--- a/tests/run.sh 2015-11-18 15:09:05.517018566 +0100 ++++ > b/tests/run.sh 2015-11-18 20:02:15.492749213 +0100 > > > Would it be possible to keep run.sh as it is but pass --parallel 1 > from run-ptest so it doesn't run the jobs in parallel? Short answer is no. In the latest version the package run.sh has been adapted to work with parallel test driver (only). Ptest however works with serial test driver only at this moment.
Partly off-topic: This automake patch [1] would we a good step toward having Ptest working in parallel. However the patch was not yet applied in poky and upstream still haven't answered the query [2] > Or, do we care if it runs the tests in parallel? Running tests in parallel have the benefit of spending less time with testing. Otherwise, the results should be the same. > I'm looking at that patch and thinking it's going to be a nightmare > to maintain when upstream modifies their script. I agree is not the best patch. I simply revert to the file used in version 2.24.2 ( plus a minor grep option addition) to make it work. We will get rid of this patch when we eventually make Ptest run with parallel test driver. Another option is to ignore run.sh wrapper and add the logic "run all testcases" into run-ptest. > > Ross [1] http://lists.openembedded.org/pipermail/openembedded-core/2014-November/098932.html [2] https://lists.gnu.org/archive/html/bug-automake/2014-11/msg00005.html Regards, Tudor. -- _______________________________________________ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
