Re: Testers wanted: exorcised kmod.c: no more daemon!

Kirk Petersen (kirk@eidolon.speakeasy.org)
Tue, 14 Apr 1998 09:23:49 -0700


> hmmm didn't the old kerneld remove the modules that had a count=0 for a period
> of time, rather that just periodically remove the modules?
> let's say I stop using a device at 23:59:55 and start using it at 00:00:05,
> now
> only 10 seconds have elapsed, but my root cronjob might have "rmmod -a" at
> midnight and the module is unloaded.. hmm makes any sense? any easy way to
> fix this, while still having a userland unload?

But this is what basically happens. The first 'rmmod -a' doesn't
actually remove the modules, the second one does. I think it is because
the first time through, 'rmmod -a' figures out which modules have a 0
use count and markes thems as having been checked. The next time, if
the modules haven't been used in the mean time, 'rmmod -a' removes them.
So it will always be between N and 2N, where N is the interval
that you separate calls to delete_module(NULL), which is what
'rmmod -a' does.

-- 
Kirk Petersen
http://www.speakeasy.org/~kirk/

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu