openSUSE Kernel July 2009

kernel@lists.opensuse.org

14 participants
12 discussions

[opensuse-kernel] KOTD release numbering

by Michal Marek

Hi, (sorry for the long mail, jump to "== Examples ==" below for the important part) I'm thinking about changing the versioning of KOTD packages (http://ftp.suse.com/pub/projects/kernel/kotd/) as the current scheme is a bit of mess. I'd like to 1) make the release string shorter, 2) make the release string sort with official distro kernels nicely, so that with some luck, an older kotd will be replaced by a newer update kernel and vice versa. I'm thinking about the following: drop the branch name and timestamp from the release string (same information is in rpm -qi output, no need to duplicate it here). Instead of the timestamp, use the number of commits (see below) so that the release string increases. To get proper ordering with official kernels, prepend the release string of the last official kernel. More precisely: last update kernel: $version-$release current kotd: $version2-??? if $version == $version2 $release2 = $release.$ncommits.$hash else $release2 = 0.0.0.$hash remember "$version2-0.0" as a fake update kernel and count from there next time endif $ncommits would be the number of commits to the kernel repository since the last update, $hash would be the commit id (same as today). == Examples == The current SLE11_BRANCH kotd would be named kernel-default-2.6.27.25-0.0.22.8dd1bfd.x86_64.rpm because the last released update was 2.6.27.23 and Greg updated to 2.6.27.25 22 commits ago. The next update (if the version stays at .25) will be 2.6.27.25-0.1, i.e. correctly newer than the kotd. The SL110_BRANCH kotd would be named kernel-default-2.6.25.20-0.4.23.4d35a02.x86_64.rpm because the last released update was 2.6.25.20-0.4, 23 commits ago. The next update will be 2.6.25.20-0.5 at least, again newer than the kotd. Maybe the commit id should be prefixed with a "g", just like git-describe does, to make it clear what it is. Does this make sense to everyone? Does anyone depend on the current numbering scheme? Michal -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-kernel+help(a)opensuse.org

11 years, 10 months

[opensuse-kernel] RPM Lints on Kernel Module

by aledr

Good day! I got these lints while packaging a kernel module: module-without-depmod-postin module-without-depmod-postun suse-policy-kmp-missing-enhances suse-policy-kmp-missing-requires Can someone advice? Thanks. -- [ ]'s Aledr - Alexandre "OpenSource Solutions for SmallBusiness Problems" -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-kernel+help(a)opensuse.org

11 years, 11 months

Re: [opensuse-kernel] Backporting 2.6.30?

by Francis Giannaros

2009/7/27 Jeff Mahoney <jeffm(a)suse.de>: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Francis Giannaros wrote: >> 2009/7/27 Greg KH <gregkh(a)suse.de>: >>> On Sat, Jul 25, 2009 at 10:14:00PM +0100, Francis Giannaros wrote: >>>>>> Just try it. I'am running 3.6.30 kernels on 11.1 for some time now without >>>>>> problems. Most error messages are caused by a configuration change. The >>>>>> driver is no longer build as module but compiled into the kernel. >>>>> Yes, that is why those warnings show up now. For some reason I thought >>>>> we had fixed the printing of those messages, maybe that hasn't made it >>>>> out into FACTORY just yet... >>>>> >>>>> thanks, >>>> Seems to be working very well so far...I guess then it would be great >>>> if someone triggered the builds for openSUSE 11.1 as well. >>> Why do you need something different, the above should work fine. We >>> aren't going to add it to the "normal" 11.1 update server. >> >> Of course not, it would just be better if it was in an openSUSE 11.1 >> repository under the kernel:head project (rather than under factory), >> because at least then the packages are being built against 11.1 >> (though I admit it's working fine for me now). >> >> Looks like someone already did it since my original mail, so no worries. > > Yeah, I did it but wasn't going to announce it until it was populated. After updating to the .31rc4 my X doesn't start (but it works in failsafe); if I run sax2 -r it freezes (the screen, I can still input stuff even though it's not shown), but here's the error log: http://francis.giannaros.org/temp/SaX.log -- any ideas? Regards, -- Francis Giannaros http://francis.giannaros.org -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-kernel+help(a)opensuse.org

11 years, 11 months

[opensuse-kernel] KOTD 2.6.31-rc4-master_20090729140126_ddfe58ae x86:64 and virtualbox-3.0.2

by Markus Koßmann

If I try to load the vboxdrv kernel module conpiled for KOTD 2.6.31-rc4- master_20090729140126_ddfe58ae-default x86_64 it fails with: Jul 31 19:11:23 PC1 kernel: [ 414.624889] vboxdrv: Trying to deactivate the NMI watchdog permanently... Jul 31 19:11:23 PC1 kernel: [ 414.624893] vboxdrv: NMI watchdog either active or at least initialized. Please disable the NMI Jul 31 19:11:23 PC1 kernel: [ 414.624894] vboxdrv: watchdog by specifying 'nmi_watchdog=0' at kernel command line. With KOTD 2.6.30-master_20090718013908_97e05475 I didn't saw that problem and there seems to be no kernel configuration change that obviously triggers that problem. Any ideas what's broken ? Should I file a bug against virtualbox or is this a kernel problem ? -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-kernel+help(a)opensuse.org

11 years, 11 months

[opensuse-kernel] openSUSE 11.0 Joystick re-calibration bug in kernel

by Michael Kostka

Hello, I think I've found a bug in the current kernel (2.6.25.20-0.4-default) in openSuSE 11.0. The bug has been since openSuSE 11.0 has released. It seems like the problem is already known and fixed in a newer kernel: http://lkml.indiana.edu/hypermail/linux/kernel/0705.2/3006.html I've already tried to make an entry in the bugzilla but I am not able to access it. It would be very nice if someone could look after this problem or could make an entry in the bugzilla. Yours sincerely, Michael Kostka -- Entek Systems E-Mail: info(a)enteksystems.de Homepage: www.enteksystems.de -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-kernel+help(a)opensuse.org

11 years, 11 months

[opensuse-kernel] Backporting 2.6.30?

by Francis Giannaros

Hi guys, Is there any chance that the .30 kernel will be backported to openSUSE 11.1, perhaps in the OBS? In general building the factory kernel for the current stable distribution could allow a lot of people to give it extensive testing, so it might really help. No idea if it would require too much work Regards, -- Francis Giannaros http://francis.giannaros.org -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-kernel+help(a)opensuse.org

11 years, 12 months

Re: [opensuse-kernel] Backporting 2.6.30?

by Francis Giannaros

2009/7/27 Jeff Mahoney <jeffm(a)suse.de>: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Francis Giannaros wrote: >> 2009/7/27 Greg KH <gregkh(a)suse.de>: >>> On Sat, Jul 25, 2009 at 10:14:00PM +0100, Francis Giannaros wrote: >>>>>> Just try it. I'am running 3.6.30 kernels on 11.1 for some time now without >>>>>> problems. Most error messages are caused by a configuration change. The >>>>>> driver is no longer build as module but compiled into the kernel. >>>>> Yes, that is why those warnings show up now. For some reason I thought >>>>> we had fixed the printing of those messages, maybe that hasn't made it >>>>> out into FACTORY just yet... >>>>> >>>>> thanks, >>>> Seems to be working very well so far...I guess then it would be great >>>> if someone triggered the builds for openSUSE 11.1 as well. >>> Why do you need something different, the above should work fine. We >>> aren't going to add it to the "normal" 11.1 update server. >> >> Of course not, it would just be better if it was in an openSUSE 11.1 >> repository under the kernel:head project (rather than under factory), >> because at least then the packages are being built against 11.1 >> (though I admit it's working fine for me now). >> >> Looks like someone already did it since my original mail, so no worries. > > Yeah, I did it but wasn't going to announce it until it was populated. Thanks Jeff, I'll test them regularly. Regards, -- Francis Giannaros http://francis.giannaros.org -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-kernel+help(a)opensuse.org

11 years, 12 months

[opensuse-kernel] btrfs in our 2.6.30 kernel

by Andreas Jaeger

Just read on lwn: "Be aware that there was a recent flag day in the btrfs on-disk format: A commit shortly after the 2.6.30 release changed the on disk format in a way that isn't compatible with older kernels. If you create your btrfs file system using the old, 2.6.30 or earlier kernel and tools, and boot into a newer kernel with the new format, you won't be able to use your file system with a 2.6.30 or older kernel any longer. Linus Torvalds found this out the hard way. But if this does happen to you, don't panic - you can find rescue images and other helpful information on the the btrfs wiki. " Could we backport this change to our 2.6.30 code for openSUSE 11.2, please? Btw. Arvin just implemented feature 306587 - minimal support for creation of a btrfs filesystem in the YaST partitioner, Thanks, Andreas -- Andreas Jaeger, aj(a){novell.com,opensuse.org} SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg) Maxfeldstr. 5, 90409 Nürnberg, Germany GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126

12 years

[opensuse-kernel] experiencing corruption in ext4 partitions

by Dirk Müller

Hi, for a couple of months now my /abuild partition is running as ext4 filesystem (previously with extents enabled). Starting with the 2.6.30-10 update, I have regular filesystem corruption within hours (and the associated machine crash soon afterwards) on this partition. downgrading to 2.6.30-8 fixes this issue for me. opensuse Factory, x86_64 Did we change something ? Greetings, Dirk -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-kernel+help(a)opensuse.org

12 years

[opensuse-kernel] Workaround if your UMTS card fails to work

by Gerald Pfeifer

If your UMTS card does not work with current FACTORY, especially if the card is Nozomi based, you may want to give an older kernel a try. 2.6.30 apparently introduced a regression here, and 2.6.27 runs just fine so far on my FACTORY system. This took me a bit of poking and playing around, which is why I am posting it here in the hope it helps someone. ( https://bugzilla.novell.com/show_bug.cgi?id=507321 ) Gerald -- Dr. Gerald Pfeifer E gp(a)novell.com SUSE Linux Products GmbH Director Product Management F +49(911)74053-483 HRB 16746 (AG Nuremberg) SUSE Linux Enterprise, openSUSE, Appliances GF Markus Rex -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe(a)opensuse.org For additional commands, e-mail: opensuse-kernel+help(a)opensuse.org

12 years

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

openSUSE Kernel July 2009