Hi,(CC me)
As a newbie i'm kinda expecting flames here... but, I just had to share
my thoughts =/
Couldn't the kernel module subsystem work something like this:
When a module is loaded it gets it's own mutex (for shared locks).
Each program/module/kernel part that uses it obtains a lock.
If we load a new module that replaces a older module we could have lock
migrations to the new module.
This would make sure that all use of the module has ceased before it's
removed.
This would also require a "usage count" per module/kernel/program
though. But that could be included in the module subsystem imho or am i
wrong?
If a module is replaced by a newer/other module how do we manage the
transgression to the new module? Well depending on how the sub system
works it could be done automagically i think...
One problem with this is that it'll use more memory...
And i bet that there are better less complicated ways of doing this but
i really think that the modules should be removable, it might take a
hour or so (worst case senario) but it should be removed eventually...
//Ian Kumlien
PS. This is more or less a brainfart... so there might be huge holes in
the idea etc. And don't forget to CC me since i'm not on the list.
DS.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Mon Jul 15 2002 - 22:00:13 EST