On Tue, Nov 8, 2016 at 12:22 PM, Hans de Goede <hdegoede@xxxxxxxxxx> wrote: > Hi all, > > I'm getting a bunch of broken dependencies > mails every day due to a few games I maintain > depending on pygame (during runtime) and > that is not available on ppc64 / ppc64le. A fix should be landing shortly. Peter > Checking pygame I noticed that Björn Esser > has tried to build pygame for ppc but that > failed to build. > > Björn, thank you for working on this. > > Looking at the build failure, it seems that > there is may be an endian issue, which makes sense > given that the build failure is on ppc64le: > > FAIL: make sure the color key is not changed when saving. > ---------------------------------------------------------------------- > Traceback (most recent call last): > File "test/image_test.py", line 284, in test_save_colorkey > self.assertEqual(p1, s2.get_at((0,0))) > AssertionError: (23, 23, 23, 255) != (23, 23, 255, 23) > ---------------------------------------------------------------------- > > Although only the last 2 bytes of the 32 bit are swapped, which > is weird. To make sure this really is an endian-ness issue > I've also started a scratchbuild for just ppc64 and that > did complete successfully. > > Looking at pygame itself, there seems to be no ppc specific > code in there, all endian specific bits are done based on > SDL_BYTEORDER which does get set correctly on ppc64le. > > Looking at SDL there also is not a whole lot of ppc specific > code (I've grepped for both ppc and powerpc) and certainly > nothing which would explain this. > > Any help with debugging this would be appreciated. In the mean > time I've filed a bug for tracking this: > https://bugzilla.redhat.com/show_bug.cgi?id=1392887 > > I've also made the bug block the ppc tracker (1071880). > > And I'm going to kick of a build with the image test test-results > ignored on ppc64le, to fix the broken deps issues. > > Regards, > > Hans > _______________________________________________ > devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx > To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx _______________________________________________ devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
