Hi On Mon, Oct 26, 2015 at 10:58 AM, Peter Maydell <peter.mayd...@linaro.org> wrote: > This will print a cryptic error message and then not fail the test, > which is not great. Maybe that's ok for the moment in the interests > of not keeping this huge patchset out of tree for too long[*], but > we should look at what glib's test framework provides in the way > of being able to report "skipped this test" outcomes.
g_test_skip() is since 2.38 (and can't be added in compat, because it uses internal variable etc) Furthermore, the shm error is a precondition for all the tests, it doesn't fit well with g_test_skip() which is inside the individual unit tests. > [*] Incidentally this whole saga demonstrates why my general > recommendation is to keep pull requests at much less than > 50 patches... > >> I rebased and updated the tag. > > If you mean by this "please retry the pull" you should send a fresh > coverletter email so my scripts will pick it up... ok -- Marc-André Lureau
