On Thu, 2010-04-08 at 16:33 -0500, Mike Westerhof wrote: > Antonio Ospite wrote: > > Having these around can cause some conflict in recipes depending on > > recent glib-2.0-native, see [1]. > > > > What can happen is that, when no preferred versions are set (like in > > DISTRO=minimal), bitbake picks up an older -native recipe even if a more > > recent one using the BBCLASSEXTEND = "native" mechanism exists. > > > > [1] http://tinderbox.openembedded.org/packages/540390/ > > > > Signed-off-by: Antonio Ospite <[email protected]> > > --- > > Please check if the commit message is clear enough, I think I've figured > > out what's happening behind the scenes, but I have not that deep > > knowledge of bitbake. > > > > Thanks, > > Antonio Ospite > > http://ao2.it > > > > > > .../glib-gettextize-dir.patch | 11 -- > > recipes/glib-2.0/glib-2.0-native_2.12.4.bb | 49 ---------- > > recipes/glib-2.0/glib-2.0-native_2.16.1.bb | 40 -------- > > recipes/glib-2.0/glib-2.0-native_2.18.0.bb | 40 -------- > > recipes/glib-2.0/glib-2.0-native_2.2.3.bb | 100 > > -------------------- > > recipes/glib-2.0/glib-2.0-native_2.22.1.bb | 39 -------- > > recipes/glib-2.0/glib-2.0-native_2.4.6.bb | 49 ---------- > > recipes/glib-2.0/glib-2.0-native_2.6.5.bb | 51 ---------- > > recipes/glib-2.0/glib-2.0-native_2.6.6.bb | 51 ---------- > > 9 files changed, 0 insertions(+), 430 deletions(-) > > delete mode 100644 > > recipes/glib-2.0/glib-2.0-native-2.12.4/glib-gettextize-dir.patch > > delete mode 100644 recipes/glib-2.0/glib-2.0-native_2.12.4.bb > > delete mode 100644 recipes/glib-2.0/glib-2.0-native_2.16.1.bb > > delete mode 100644 recipes/glib-2.0/glib-2.0-native_2.18.0.bb > > delete mode 100644 recipes/glib-2.0/glib-2.0-native_2.2.3.bb > > delete mode 100644 recipes/glib-2.0/glib-2.0-native_2.22.1.bb > > NACK. This glib version is still in use; don't delete it: > /u/slug/slugos/tmp/work/x86_64-linux/glib-2.0-native-2.22.1-r6.2 > > (Surely there must a better way to fix this than removal of active recipes?)
If 2.22.1 was converted to BBCLASSEXTEND, the rest of the mechanisms would work right again. -- Tom Rini <[email protected]> Mentor Graphics Corporation _______________________________________________ Openembedded-devel mailing list [email protected] http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
