david@prism.demon.co.uk (David Metcalfe) writes:

quoted text
>I am getting null pointer traps in 0.98pl2.  I have the kernel
>compiled with the tcp/ip code and I have a scsi disk.  I first
>noticed the problem while rc.net was running.  Judicious use of
>commenting out suggested that the problem was occurring when config
>was being run, so I decided to try recompiling config.

The problem is in the tcp kernel code, you'll just have to do a
chmod -x /usr/etc/inet/rc.net
for now.

What version of login are you using? and where is the source  for it?

quoted text
>gcc -O3 -s -N -o config config.c produced the following:

quoted text
>Unable to handle kernel paging request at address c0000000
>Oops: 0000
>EIP: 0008:002CD5E
>EFLAGS: 00010202
>fs: 0017
>base: 00000000, limit: c0000000
>Pid: 24, process nr: 7
>83 3e 00 75 08 66 8b 76 04 66
>Segmentation fault

quoted text
>How are you supposed to tell in which part of the kernel the problem
>lies.  Linus' own testing would tend to indicate the problem lies in
>the scsi code.

Are you using a scsi disk?

quoted text
>David

quoted text
>-----------------------------------------------------
>David Metcalfe          david@prism.demon.co.uk     
>London, England         dmetcalf@cix.compulink.co.uk
>     
>                       
>