On 10/13/09, Alan Jenkins <sourcejedi.lkml@xxxxxxxxxxxxxx> wrote: > On 10/13/09, Andreas Robinson <andr345@xxxxxxxxx> wrote: >> I've come as far as I can go at the moment, I think. >> >> Since the last update, tests and docs have been added and a few bugs >> have been fixed. Replacing the install commands with their softdep >> equivalents in /etc/modprobe.d/alsa-base.conf have been tested as >> well. >> >> These are the relevant lines from alsa-base.conf, for my system. I've >> omitted the ones that aren't used on my system, for brevity: >> >> softdep snd-pcm post: snd-pcm-oss >> softdep snd-mixer post: snd-mixer-oss >> softdep snd-seq post: snd-seq-midi snd-seq-oss >> softdep snd-rawmidi post: snd-seq-midi >> >> Running lsmod before and after the change (and a reboot) yields the >> same result so it must have worked. :) > > Great :). I haven't looked at the code yet, but here's my reply to > this message - > >> Finally, someone suggested in a private mail that we implement >> softdeps in .modinfo, much the same way as aliases are. If you don't >> have any objections, I'll go ahead and start working on that. > > It needs a good reason. IMO it needs a very good reason... > > Is this about the uhci v.s. ehci mess? It would definitely be nice to > resolve that. > > I think we should hear the opinion of the maintainer of whichever > kernel module(s) are supposed to benefit. > >> commit 5fd23bcc409045894152bc2e58cd978dcba5140a >> Author: Andreas Robinson <andr345@xxxxxxxxx> >> Date: Sat Oct 10 20:00:22 2009 +0200 >> >> modprobe: softdep now handles mock modules >> >> do_softdep skips the main module if it does not exist, following >> the behaviour of the install and remove commands. >> >> Signed-off-by: Andreas Robinson <andr345@xxxxxxxxx> > > Hum. Why? > > Install / remove commands are stupidly powerful. Also - this affects _all_ softdep rules, right? If you write an install command for a real module e.g. eeepc-laptop, which includes "modprobe --ignore-install eeepc-laptop", you still get a "FATAL" message if the module doesn't exist. The exit code will even be correct if you write the rule correctly :-). But this commit means that if you add softdeps for a module, it won't complain if the module doesn't exist when you try to load it. That seems like a step backwards Regards Alan -- To unsubscribe from this list: send the line "unsubscribe linux-modules" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
