2.6.8-rc1 -ck snapshot

[Con Kolivas <kernel@kolivas.org>]

[-- Attachment #1: Type: text/plain, Size: 1698 bytes --]

I've posted a snapshot of the current -ck development against 2.6.8-rc1

website: http://kernel.kolivas.org
rc1 patches: http://ck.kolivas.org/patches/2.6/2.6.7/2.6.8-rc1/

The staircase scheduler has been basically unchanged for a couple of 
weeks now with no bugs showing up (woohoo). I've resynced with the 
current versions of the vm hacks, added Ingo's voluntary preempt patch, 
added Zwane's low latency irq locks, the suse writeback latency patch, 
badram, lufs and the fixed version of William Lee Irwin III's preempt 
test. Bootsplash is currently on the backburner.

To preempt test is a debugging tool for latency and not intended for 
regular usage. Enable it in the config and set the bootparam
preempt_thresh=1
for checking for when there are parts of kernel code that are not 
preemptible for >1ms (set it to any desired number or dont set it to 
disable it).

It outputs to the syslog something like this:
4ms non-preemptible critical section violated 1 ms preempt threshold 
starting at exit_mmap+0x1c/0x188 and ending at exit_mmap+0x118/0x188
  [<c011d769>] dec_preempt_count+0x14f/0x151
  [<c014d0d4>] exit_mmap+0x118/0x188
etc.

I'm running this kernel currently without problems. The badram and lufs 
features are compile tested only.

Full patches:
from_2.6.7_to_staircase7.A
schedrange.diff
schedbatch2.3.diff
schediso2.3.diff
autotune_swappiness.diff
autotune_inactivation.diff
supermount-ng204.diff
defaultcfq.diff
config_hz.diff
voluntary-preempt-2.6.7-bk20-H3
patch-i386-irq_enable_spinlocks2
patch-ool-spinlocks
9000_SuSE-117-writeback-lat.patch
wli_preempttest2
wli_preempttest2.1
BadRAM-2.6.5.2.patch.bz2
lufs-0.9.7-2.6.0-test9.patch.bz2
ck5version.diff


Cheers,
Con

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 256 bytes --]