Re: 2.1.36 can't read System.map?
A Guy Called Tyketto (tyketto@zip.omnilinx.net)
Wed, 16 Jul 1997 00:05:10 -0600 (GMT-0600)
>
> Quoth Evan Jeffrey:
>
> >>Jul 15 16:29:13 maryland kernel: klogd 1.3-3, log source = /proc/kmsg started.
> >>Jul 15 16:29:14 maryland kernel: Error in symbol table input.
> >>Jul 15 16:29:14 maryland kernel: Cannot find map file.
> >>Jul 15 16:29:14 maryland kernel: Error seeking in /dev/kmem
> >>Jul 15 16:29:14 maryland kernel: Error adding kernel module table entry.
> >
> >It means that when you set up your new kernel, you didn't put the new
> >System.map file in /boot directory, probably, or couldn't find the right
> >file. klogd tries to find the correct file, but it isn't very intelligent.
>
> Errr, no. /boot/System.map is in the correct place, and it is the correct version and all that. And the problem isn't really related to klogd, who is just the messenger - looks more like some part of the kernel can't read the symbol table. AFAIK, klogd has no reason to mess with /dev/kmem.
>
It took me a while to figure out that problem when 2.1.36 was
initially released. The problem isn't you, nor the kernel, nor where
System.map needs to go. The problem is libc. Try upgrading your C library to
something higher than libc-5.4.23, then (optional) recompile. It should find
System.map no matter where you put it, or specify as the INSTALL_BOOT path in
the top level Kernel makefile.
Give that a shot, and see what you get.
BL.
--
Brad Littlejohn | Email: tyketto@omnilinx.net
Unix Systems Administrator, | tyketto@acm.unomaha.edu
WebMaster, NewsMaster.. Smeghead! :) | http://www.omnilinx.net/~tyketto
PGP: 1024/E9DF4D85 67 6B 33 D0 B9 95 F4 37 4B D1 CE BD 48 B0 06 93