On Mon, 24 Jan 2000, Ville Herva wrote:
> On Mon, Jan 24, 2000 at 08:52:39AM +1100, you [Keith Owens] claimed:
> > On Sun, 23 Jan 2000 11:44:16 +0200,
> > Ville Herva <vherva@niksula.hut.fi> wrote:
> > >Warning (compare_ksyms_lsmod): module es1370 is in lsmod but not in ksyms,
> > >probably no symbols exported
> > >
> > >Doesn't that mean that lsmod lists those (as it certainly does), but
> > >there are no symbols present for them in /proc/ksyms? Browsing the
> > >ksymoops docs, I gather that modules that do not export any symbols are
> > >not listed at all by ksymoops. In /proc/ksyms there where no lines for the
> > >above modules, as there were for others.
> >
> > The current modutils adds some generated symbols for *all* modules,
> > specifically to get around this problem. You msut have loaded the
> > modules using an old modutils.
>
> Okay, that is the case. After reload they appear just fine.
>
> But the addresses in the ksymoops output are still nowhere near C5D28000
> (the PC of the unkillable process). I'm making this assumtion that they
> are not even _supposed_ to be, since I can find a single match or nearly
> match comparing the PC's of other processes and ksymoops output. So can I
> get any sense out of that PC address, or should I just apply the kernel
> debugger patch and wait a couple of weeks?
On my system, modules always load in the C5xxxxxx range. Obviously not
the case with your setup. Unless someone can suggest a better way, I
think kdb is the easiest way to find out what's going on. (A diag-patch
ala echo pid > /proc/oops would be handy for [really running?] problems
like this one.. but doesn't exist afaik)
-Mike
-
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 : Mon Jan 31 2000 - 21:00:11 EST