kernel 2.6.32-rc5.dmz.1-liquorix-686

[mdmarmer]

In addition to the strange behavior I saw in an iso built with remastersys, this kernel does not shut down all the way in my installed system. It will reboot ok, but on shutdown it hangs after shutting down alsa.

Mike

[damentz]

Hmm, that's strange, next RC I'll test with CFQ as default and see what difference it makes. I still haven't included the IO-Delay port change from 2.6.31 for this version.

[reiner]

Hello again,
I am currently experimenting with 2.6.32-rc5.dmz.1. on my Dual-Core AMD64 machine.
It looks like there is no nfs-support compiled in yet (nfs-kernel-server complains heavily about that).
Is that supposed to be so ?

regards

Reiner

[damentz]

You're right, a warning is given of no kernel support. However, the config does have NFS enabled - this must be a kernel bug that will be fixed before the final version of 2.6.32 comes out.

I'll keep an eye on this and ask others.

[reiner]

Hmm,
found another one on the 66 version. Madwifi (ath_pci) seems not to be included yet. Will it be available in the final version ?

regards

Reiner

[damentz]

The zen guys may have taken this out on purpose, that I'll need to check too. Ath5k is supposed to deprecated madwifi in the long term, maybe 2.6.32's source of ath5k is considered "good enough" or better?

If you check gitweb for their sources, git.zen-kernel.org, they don't have a branch to merge for madwifi at the moment in zen.git

[damentz]

Ok, try 2.6.32-rc5.dmz.3 when it hits the repos, there may have been a problem with AUFS that's corrected.

[reiner]

@damentz:
Thanks for the info I will try that.

regards

Reiner

[reiner]

@damentz:

the ath5k driver in 2.6.32-5rc really has improved seriously, but not reached completely the performance of madwifi yet. The annoying error messages that came up with my cards have completely disapeared now.
I will compile and add madwifi myself until ath5k finally has grown up (Maybe with 2.6.32 final this is the case).

regards

Reiner

[damentz]

It appears I made a mistake while uploading new packages to my repository.

Some of you may have downloaded the 2.6.32 kernel through the main repository which was supposed to reside in the future repo.

You can remove it if you wish, the main repo should only contain stable kernels.