On Thu, Jun 12, 2008 at 02:13:06AM -0400, Jeff Mahoney wrote:

Hi all -

Just a quick FYI that after the 11.0 branch announcement, the HEAD kernel has been rebased to 2.6.26-rc5-git5. Updates will be ongoing.

Thanks a lot for doing this.

Once 11.0 is officially released, this will be updated as the FACTORY kernel. Until then, packages will be available via the KOTD site at http://ftp.suse.com/pub/projects/kernel/kotd/HEAD/

I'm considering what would be involved in updating the version that is in the "normal" 11.0 update repos to 2.6.26, when it is out and we seem to have tested it pretty well in FACTORY. It would be nice to offer this for the increased number of bug fixes, and new hardware support. But it would be something new from what we have done before, and I don't know how people would react to it. Any opinions? thanks, greg k-h -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-kernel+help@opensuse.org

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Andreas Jaeger wrote:

Jeff Mahoney writes:

...

Hi all -

Just a quick FYI that after the 11.0 branch announcement, the HEAD kernel has been rebased to 2.6.26-rc5-git5. Updates will be ongoing.

I tried booting it on my x86-64 machine and it seemed to boot successfully - but then I saw "bootsplash turned off" (not exact message) and the system was frozen completely.

Has anybody had better success with the kernel? I can attach on monday a serial console if needed,

I have 2.6.26-rc6-git1 booted on an x86_64 node, but it's not a desktop. - -Jeff - -- Jeff Mahoney SUSE Labs -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org iEYEARECAAYFAkhT5XkACgkQLPWxlyuTD7JFJACeNgpwa+rCThtitUzZiZj/naTg Q7EAn0Ln2sNFUdItL4xzE16fwr/flnZ+ =XpJT -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-kernel+help@opensuse.org

On Sat, Jun 14, 2008 at 07:22:28PM +0200, Andreas Jaeger wrote:

* dmesg tells me: Driver 'sd' needs updating - please use bus_type methods

That should have been saying this for the 11.0 kernel as well, it's an old message :) thanks, greg k-h -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-kernel+help@opensuse.org

At Mon, 16 Jun 2008 10:32:30 +0200, Jean Delvare wrote:

Hi Andreas,

Le samedi 14 juin 2008, Andreas Jaeger a écrit :

...

* hda intel does not work (phonon tells me it's broken and uses pcspeaker instead), dmesg shows: hda_intel: probe_mask set to 0x1 for device 17aa:20ac PCI: Setting latency timer of device 0000:00:1b.0 to 64 input: PC Speaker as /devices/platform/pcspkr/input/input8

Taking a blind shot... Could it be that the PC speaker driver registers as a regular ALSA device now and comes up before hda_intel, becoming the default audio device? Check if "aplay -l" returns more than one card. Maybe you have to tell ALSA which "card" is #0 and which one is #1. I think Yast has an interface to set this.

It's not about YaST but rather auto-loading mechanism. As a workaround, just add snd-pcsp to /etc/modprobe.d/blacklist. Takashi -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-kernel+help@opensuse.org

At Mon, 16 Jun 2008 13:13:20 +0200, Jean Delvare wrote:

Hi Takashi,

Le lundi 16 juin 2008, Takashi Iwai a écrit :

...

At Mon, 16 Jun 2008 10:32:30 +0200, Jean Delvare wrote:

...

Hi Andreas,

Le samedi 14 juin 2008, Andreas Jaeger a écrit :

...

* hda intel does not work (phonon tells me it's broken and uses pcspeaker instead), dmesg shows: hda_intel: probe_mask set to 0x1 for device 17aa:20ac PCI: Setting latency timer of device 0000:00:1b.0 to 64 input: PC Speaker as /devices/platform/pcspkr/input/input8

Taking a blind shot... Could it be that the PC speaker driver registers as a regular ALSA device now and comes up before hda_intel, becoming the default audio device? Check if "aplay -l" returns more than one card. Maybe you have to tell ALSA which "card" is #0 and which one is #1. I think Yast has an interface to set this.

It's not about YaST but rather auto-loading mechanism. As a workaround, just add snd-pcsp to /etc/modprobe.d/blacklist.

I mentioned Yast just as an easy way to change the audio device index values, so that snd-hda-intel remains the main audio device. I guess that editing /etc/modprobe.d/sound manually the following way would work as well:

options snd-hda-intel enable=1 index=0 options snd-pcsp enable=1 index=2

(or even enable=0 for snd-pcsp)

On 11.0, it works better with snd slots option. Now YaST sets options snd slots=snd-hda-intel which will reserve the first slot for snd-hda-intel driver, so this problem won't occur. Maybe the module config of Andreas is from the older version.

I am curious what are the integration plans for the snd-pcsp driver. We certainly want to include this driver in our products, for the users who want it. But OTOH we have to make sure that it will not be the default sound device. Does this mean that we blacklist the snd-pcsp driver if a real sound card is found during installation? Do we blacklist it by default and let it up to the few interested users to remove the blacklist statement manually? Or do we just give the PC speaker a greater index value as I suggested above, so that other devices take precedence?

I mean to blacklist it forever. It's almost a fun project and not for serious usage. So, either disabling in the kernel config or addint it to blacklist would make both users and us happy. Takashi -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-kernel+help@opensuse.org