Claes Bäckström wrote:

As of today I recomend against updating the kernel, I have done this and run in to some problems. Yesterday it just swithed to runlevel 0 every time I tried to start it. Claiming that the temperature in my laptop was over 5000 C and after some reboots all worked again (And my laptop isn't a pile of melted plastic). Not sure what happened. And the next problem I'm facing is that I can't seem to compile any drivers for it. And this have worked before like a charm. Not sure but perhaps this is 'coz kernel-headers, kernel-source and kernel-syms is different versions. But these kind of troubles is why I love using none-released pre versions of Linux. Ahhh my weekly doses of breakage. :-) Bring it on!

Warm Regards, Claes Backstrom

PS. The drivers I can't compile is simply that I try to rebuild the madwifi source rpm from factory. DS.

On 1/15/07, Dominique Leuenberger <Dominique.Leuenberger@tmf-group.com> wrote:

...

Hi everybody,

As I have the ftp factory tree in my installation sources, of course My OS is constantly trying to update around 1000 packages. So far no problem at att :-) (except my low bandwidth.. but that's another story).

Just one question: Does somebody know when the kernel upgrade to 2.6.20 might be possible? So far, the kernel (in rc4) is there, but an update is not possible es for example the wlan-kmp is missing (and yes, I'm on a wireless connection). But still, I would like to test on the new kernel, as some features of my notebook (Flash Media) should finally be supported :-)

Or is there something I can contribute in this case?

Dominique

I did give a heads up on this list some time a week or 2 ago - vanilla kernels. I did have a couple of problems. On 2.6.19-git and later, the Gigabit ethernet (forecedeth) on a 64x2 box (Foxconn mobo nVidia Corporation MCP51 Ethernet Controller (rev a1)) gave strange problems, another box could ping it, but ping from the box saw nothing, I thought it was a bad receiver on the on-board ethernet, swapped in a eepro100, reconfigured and it's still up and running. Some days later, amazingly I had exactly the same happen on the x86_64 laptop (BCM5788 with tg3 driver). Whilst working the problem with LKML, one suggestion amongst others was to turn off SuSEfirewall2, that fixed the problem. Because of the workqueue changes in >=2.6.19-git, acx111 and rt2500 modules wouldn't build. acx-20070101.tar.bz2 downloaded yesterday is working with 2.6.20-rc4-git6 on x86 box, rt2x00-cvs-daily.tar.gz downloaded earlier built fine against 2.6.20-rc3 on the x86_64 laptop and on 2.6.20-rc4-git6 it built, but modprobe of rt2500pci gives unknown symbols rt2x00debug_register and rt2x00debug_deregister. The other problem I had, yet to be checked out with SuSEfirewall2 disabled, booting with SuSEfirewall2 enabled and acpi=off stopped the ethernet being configured. Regards Sid. -- Sid Boyce ... Hamradio License G3VBV, Licensed Private Pilot Emeritus IBM/Amdahl Mainframes and Sun/Fujitsu Servers Tech Support Specialist, Cricket Coach Microsoft Windows Free Zone - Linux used for all Computing Tasks --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-factory+help@opensuse.org