[ltp] custom kernel for SuSE 9.1

Eben King linux-thinkpad@linux-thinkpad.org
Tue, 3 Aug 2004 17:36:10 -0400 (EDT)


I'm trying to make a custom kernel for my SuSE 9.1 system, which is an IBM
Thinkpad T40 (1.4 GHz P4-M, 768 MB RAM, 25 (30?) GB HD).  The HD has 4
primary partitions (no extended partitions).  The reason I'm doing this is
to eliminate the drivers and options it uses for which I don't have the
hardware support (i.e., almost all of them).  For instance, I don't have 
RAID (nor is it feasible on this machine).

I've done this many times in the past, and it's (usually) no big deal.  
This time, however, it's being a bear.

I'm starting with the SuSE stock .config (gzip -d < /proc/config.gz >
/usr/src/linux-2.6.7/.config), and (after I get that to work) modifying it
to fit my needs.  But I get many errors as soon as the root fs (reiserfs)  
is mounted during the boot sequence.  The errors are similar to (similar
because I'm copying by hand):

hda: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error}
hda: read_intr: error=0x10 { SectorIdNotFound }, LBASect=58542742, sector=58542742

That pair repeated, with occasional

ide0: reset: success

or

end_request: I/O error, dev hda, sector 58542742

thrown in.  LBASect and the two sectors are the same number.  (Is 58542742
decimal or hex?  How big are the sectors?)  I didn't see them changing.  
but they went by rather fast (even with a 60-line screen).  I don't
suppose there's any way I could log that and retrieve it...

This drive works fine under the stock SuSE kernel and under XP, so it's
not a hardware issue.

I've tried a few changes I found (on the LKML archives) as suggested
solutions for people with similar problems -- don't share IDE interrupts,
don't use the old MFM/RLL driver (I wasn't), and maybe a few others I've
forgotten.  None helped.  I did have to make reiserfs built in, not 
modularized; SuSE uses an initrd but I didn't want to (or know how).  
Before I did that, it couldn't mount the root fs at all.  I'm not sure 
this is an improvement.

Does anybody have any clues or pointers?  Thanks.

-- 
-eben    ebQenW1@EtaRmpTabYayU.rIr.OcoPm    home.tampabay.rr.com/hactar

A: Because it looks dumb and is hard to read.
Q: Why is top-posting wrong?  -- from lots42@xxx.com