Hi all, > attach always 'succeeds' in the sense that after attach has been called, > detach will always be called. The detach routine should tear down > everything that needs tearing down and not do things that will fail. > Perhaps the init could simply be done before the attach routine gets the > chance to fail?
Some drivers mark the progress through attach so that detach will work correctly in all cases. Can we do the same here? For example, see sc->sc_att_stage in: https://nxr.netbsd.org/xref/src/sys/dev/ic/gem.c and also how gem_partial_detach() is called when the attach fails part way through. Regards, Julian