K 10
svn:author
V 5
luigi
K 8
svn:date
V 27
2001-11-21T17:50:35.000000Z
K 7
svn:log
V 1306
Prevent panics when trying to kldunload a module which has been statically
compiled in and previously attempted to load (this is a very common
situation, because some programs e.g. ppp do try to load what they need).
This is done by having module_register() pass up any error it should find.

Not being familiar enough with the module handling code, i am not
sure if this is the best fix to the problem -- after spending a
bit of time on it I think there is some architectural problem in
that part of the kernel, which (as Mike Smith pointed out) still
confuses between "files" (loadable objects) and "modules" (subsystems
available in the kernel, which can reside in loadable object or be
statically compiled in).

On the other hand the problem is sufficiently serious to deserve
at least a short term fix, and this change is sufficiently small
(6 lines in a single file) to be easily backed out and replaced
with The Right Fix when/if this becomes available.

For the above reason, and because some postings i read seem to
suggest that the problem does not exist in current (though i couldn't
verify/find how it is fixed there).  I am only committing this to
RELENG_4 and not to HEAD.

For more details see PR kern/24392 and my message to -stable
with Message-ID <20011105111839.A75789@iguana.aciri.org>

END