I just downloaded the linux-2.5.40 tarball.

The kernel was built and tested on a box running Debian Unstable (refreshed today).
I had four (five if you include ALSA breaking make menuconfig) issues.

     - the kernel wouldn't link with XFS enabled due to some
       unreferenced symbols ("run_task_queue", etc).

     - configuring for the SCSI IOMega Parallel port drivers as modules,
       make modules_install fails as the 'imm.o' and 'ppa.o' files
       are missing. (i just 'touch'ed these files to get
       "make modules_install" to continue)

     - make modules_install runs depmod which fails with
depmod: cannot read ELF header from /lib/modules/2.5.40/kernel/drivers/scsi/imm.o
depmod: cannot read ELF header from /lib/modules/2.5.40/kernel/drivers/scsi/ppa.o
depmod: *** Unresolved symbols in
/lib/modules/2.5.40/kernel/drivers/usb/input/usbkbd.o
depmod: 	usb_kbd_free_buffers
depmod: *** Unresolved symbols in
/lib/modules/2.5.40/kernel/net/ipv4/netfilter/ipt_owner.o
depmod: 	next_thread
depmod: 	find_task_by_pid
depmod: *** Unresolved symbols in
/lib/modules/2.5.40/kernel/net/ipv6/netfilter/ip6t_owner.o
depmod: 	next_thread
depmod: 	find_task_by_pid

       The first two are naturally caused by me just touching up 0 byte
       files as mentioned above.

     - Booting up, i got several "Debug: sleeping function called from illegal
        context at slab.c:1374", and one nasty-looking BUG!:
       "kernel BUG at slab.c:1292!" which apparently killed off klogd.

The slab.c:1292 bug occurs during bootup, just after init goes
to runlevel 2 and starts syslogd and klogd, and is about to enable swap.

I tried booting twice, and got the same bug at the same time during bootup.
I'll attach "dmesg > dmesg.2.5.40.first" and "diff -u dmesg.2.5.40.first 
dmesg.2.5.40.second" to show the few differences between my two bootups...

As you can see the last dmesg line before the BUG! line mounts a 256mb swap-file 
residing on my ext3 "/" filesystem mounted with priority '-2'. I also have a 
160mb swap partition which is mounted with priority '-1'. I don't know if the 
bug is any way related to this, though. I'm also a bit unsure what to make of 
the stack trace, but it might be of interest that this machine is an NFS client 
to another linux NFS server (seeing "vfs" and "dgram_sendmsg" in there :) )


Apart from klogd dying, everything else seems fine (although I haven't tried 
anything fancy apart from running a few ssh sessions and playing some audio).
Restarting klogd via "/etc/init.d/klogd start" when the system is running
seems to bring up klogd just fine.


(I'm not subscribed to the list; I would appreciate being CC:ed replies.)