Some of our developers have been running into this problem and I'm wondering
how others deal with it.
It seems if we have a recipe, (A) that depends upon header files staged by
another recipe (B) then:
1) if the source for recipe (B) is modified and recompiled with:
bitbake -f -c compile B
bitbake B,
I would expect that when recipe A is run:
bitbake A
It too would be recompiled, however that does not seem to be the case. In
fact even if recipe B is cleaned up:
bitbake -c clean B
and then we run recipe A
bitbake A
only recipe B is run and recipe A is NOT rerun! the dependency is satisfied
by rerunning recipe B and then bitbake stops..
is this expected behaviour? Is there away to force rerunning of recipes that
would be "out of date" due to one of its dependent recipes being return and
re-staging (potentially) new headers and libraries?
Any help of suggestions would be greatly appreciated.
Mike
_______________________________________________
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
