After 12 hours of testing it looks like this problem can be
solved by the following..

into groveronline.com:/root/bk/linux-acpi

Noticed by Martin Schwidefsky <schwidefsky@de.ibm.com>

dcache lock.

Found by Maneesh Soni <maneesh@in.ibm.com>

into home.transmeta.com:/home/torvalds/v2.5/linux

into home.transmeta.com:/home/torvalds/v2.5/linux

partly based on a patch from Tariq Shureih to the kernel janitors list.

into halogen.(none):/home/davej/src/kernel/2.5/watchdog

into home.transmeta.com:/home/torvalds/v2.5/linux

Patch from Nicolas Pitre

Patch from Eli Carter

Supercedes patch 1382
This patch copies the relevant documentation for the iop3xx support
from the 2.4.19-rmk4-ds2 kernel.

Patch from Dirk Behme

Cleanup arch/arm/kernel/head.S a little bit by removing the unused rambase from pgtbl and krnladr macros.

Patch from Jonas Larsson

The dc21285 mtd map driver supported RedBoot partitioning only. Kernel command line partitioning was not supported. This patch adds the missing functionality.

From Marc Zyngier <mzyngier@freesurf.fr>.

Hi.

This trivial patch switches the file to use C99 initializers.

Art Haas

(Bjorn Helgaas)

  - Add acpi_walk_resource support (Bjorn Helgaas)
  - Add S4BIOS support (Pavel Machek)
  - Rewrite GPE code to deal with future GPE block devices
  - Fix a bug that caused long delays when receiving SCIs
  - Fix a bug that causes oopses when battery strings are ""

into groveronline.com:/root/bk/linux-acpi

into home.transmeta.com:/home/torvalds/v2.5/linux

into halogen.(none):/home/davej/src/kernel/2.5/agpgart

Dammit! I'm not having a good morning. :(

Missing GFP_xxx argument to kmalloc.

This patch addresses two distinct races:

- Until now, the driver started the chip for Tx regardless of errors
  pending in the status register. Not good if an error occured while
  we were queueing packets -- the chip counter had not been reset,
  so Tx died. (We can't reliably get an interrupt for every error
  condition)

- The Rhine-II (when under load) frequently produces a Tx descriptor
  write-back race error. Failing to handle this means waiting for the
  netdev watchdog. Fixed.

  In addition, we must wait for the Tx engine to turn off on error
  conditions before we scavenge the descriptor entries. Failing to do
  so will typically lead to performance going down to about 10%: Burst,
  timeout, burst, timeout.. (again, with a Rhine-II under load).

Since Linus and Jeff raised the issue of PCI posted writes, I cleaned up
wait_for_reset() some more. Experiments show that with MMIO, a reset may
indeed take seemingly longer -- that is fixed by flushing that buffer.

Also, the driver now polls the appropriate register while waiting for the
reset to finish.