On Mon, Jun 19, 2000, Frank van Maarseveen <F.vanMaarseveen@inter.NL.net> wrote:
> On Fri, Jun 16, 2000 at 10:29:03AM +0100, Tim Waugh wrote:
> > On Fri, Jun 16, 2000 at 08:58:47AM +1000, Keith Owens wrote:
> >
> > > man modules.conf. Look at the "before" and "after" options, they are
> > > designed for this sort of inter-module dependency when symbols alone
> > > will not do the job.
> >
> > Does that work when modules are autounloaded one by one?
> Not tested, but I expect problems.
>
> I have given it some thought but the fact that all drivers must
> be loaded before uhci is not good. When uhci cannot find a driver
> for something on the bus it should first try to load the driver
> instead of dumping all the USB info to syslog. Now it seems that I'm
> forced to load the complete USB driver suite in advance. Who knows
> what kind of device I'm going to attach...
>
> (actually one can think of two policies: one where plugging in the device
> causes the modprobe and one where device inode use causes the modprobe. I
> guess that the first is the coolest and also compatible with PCMCIA and
> family, but correct me when I'm wrong)
This has been a topic of discussion for months now. There have been a
couple of threads about this specifically. You can find them in the
linux-usb archives.
The short of it is, it's unfortunately not that easy.
Look for something from me soon expanding on it a great deal more.
> Also, I think it is best for usbdevfs to make it part of /proc.
> The name is a bit misleading because there are no device nodes
> such as with devfs and mounting it on /proc/bus/usb is awful.
This is incorrect. Just because the files don't have a major/minor
doesn't mean they don't have the same functionality. Having a virtual FS
means we don't need major/minor numbers.
Although I do agree having it on /proc is not the best place, but this
is another long thread of discussion. I don't have a strong enough
opinion to get involved :)
Part of the reason for making it a different VFS is because of some
locking issues according to Thomas Sailer (the original author of the
patch). I can't expand any further than that.
> Oh, and I still see:
> Jun 19 22:55:21 iapetus kernel: hub.c: giving up on killing khubd
> Jun 19 22:55:21 iapetus kernel: usb.c: deregistering driver hub
>
> Easy to reproduce:
> Alias usbdevfs to usbcore
> mount usbdevfs somewhere
> umount it
> modprobe -r usbcore
>
> (no USB devices attached)
I'll have to test this out. I don't have this problem with 2.3 or 2.4
kernels.
JE
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Fri Jun 23 2000 - 21:00:18 EST