On Wed, Sep 25, 2019 at 7:37 PM Chris Johns <chr...@rtems.org> wrote: > > On 26/9/19 3:25 am, Gedare Bloom wrote: > > I just meant that the view of the running tests is not useful for > > comparison, > > Yes. It lets you see if something major is wrong with a test run. The very > original implementation listed the failing tests by name but this added little > value and you could not track the state of the run so it was removed. > > > it is exactly this summary (the end result) that helps. > > Great, this is the important bit. > > > If we had regular testing, > > parsing the results and producing a status matrix could help for > > understanding > > the tiers. I'm not saying I know how this would be accomplished, and it > > seems it > > would require coordination among community members who test on different > > bsps. > > A score board? This would be really nice to have. I think something that hooks > into procmail and monitors the emails posted to the bu...@rtems.org list would > work. I would to encourage anyone and everyone to post results for BSPs they > have. A score board can then be used to maintain the tiers. > > Also there is a ticket to have the tester take the console output and generate > the results. If this also posted the results email it would help make the test > results more widely available. > > The tester work so far has been unfunded or GSoC projects and I cannot see > this > changing soon however it is vital to our users, the community and a wider > audience that we have quality current published results. > Speaking of unfunded work, this idea might translate reasonably well into several GCI coding tasks.
Just a thought, in case we participate. Gedare _______________________________________________ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel
