Hi,
currently ATrpms builds kmdls for the following RHEL5/CentOS5 kernels (actually for CentOS it also builds plus for plus kernels, but let's keep the discussion simple):
el5/2.6.18-92.1.6.el5 \ el5/2.6.18-92.1.1.el5 \ el5/2.6.18-92.el5 \ el5/2.6.18-53.1.21.el5 \ el5/2.6.18-53.1.19.el5 \ el5/2.6.18-53.1.14.el5 \ el5/2.6.18-53.1.13.el5 \ el5/2.6.18-53.1.6.el5 \ el5/2.6.18-53.1.4.el5 \ el5/2.6.18-53.el5 \ el5/2.6.18-8.1.15.el5 \ el5/2.6.18-8.1.14.el5 \ el5/2.6.18-8.1.10.el5 \ el5/2.6.18-8.1.8.el5 \ el5/2.6.18-8.1.6.el5 \ el5/2.6.18-8.1.4.el5 \ el5/2.6.18-8.1.3.el5 \ el5/2.6.18-8.1.1.el5 \ el5/2.6.18-8.el5 \
Similar for RHEL4/3. This slows down kmdl updates as say for a new nvidia driver one need to build kmdls for all these kernels in all flavours/archs etc.
I start to think whether these kernels are indeed being all used to the extend of justifying full kmdl support. Maybe it would make sense to keep the full last series (2.6.18-92* above) and the highest one from the series before (2.6.18-53.1.21.el5 and 2.6.18-8.1.15.el5).
Or is there any other idea? What are Red Hat's, CentOS' policies wrt support of kernels? ATrpms should probably just copy that policy and drop kernel support once the respective upstream support drops it.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Axel Thimm wrote: | I start to think whether these kernels are indeed being all used to | the extend of justifying full kmdl support. Maybe it would make sense | to keep the full last series (2.6.18-92* above) and the highest one | from the series before (2.6.18-53.1.21.el5 and 2.6.18-8.1.15.el5).
I would limit the kernels per distro: ~ - The one(s) shipped on the media ~ - The most recent kernel ~ - The one before the most recent kernel
BTW as far as I can tell there are no -92* kernel modules on ATrpms. And I have a number of issues with installing MythTV due to missing dependencies on Centos 5.2
Hugo.
- -- hvdkooij@vanderkooij.org http://hugo.vanderkooij.org/ PGP/GPG? Use: http://hugo.vanderkooij.org/0x58F19981.asc
A: Yes. >Q: Are you sure? >>A: Because it reverses the logical flow of conversation. >>>Q: Why is top posting frowned upon?
Bored? Click on http://spamornot.org/ and rate those images.
Hugo van der Kooij wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Axel Thimm wrote: | I start to think whether these kernels are indeed being all used to | the extend of justifying full kmdl support. Maybe it would make sense | to keep the full last series (2.6.18-92* above) and the highest one | from the series before (2.6.18-53.1.21.el5 and 2.6.18-8.1.15.el5).
I would limit the kernels per distro: ~ - The one(s) shipped on the media ~ - The most recent kernel ~ - The one before the most recent kernel
Yepp, sounds sane.
BTW as far as I can tell there are no -92* kernel modules on ATrpms.
IBTD :)
nvidia-graphics173.14.09-kmdl-2.6.18-92.1.1.el5-173.14.09-99.el5.x86_64
Ralph
On Thu, 3 Jul 2008, Axel Thimm wrote:
I start to think whether these kernels are indeed being all used to the extend of justifying full kmdl support. Maybe it would make sense to keep the full last series (2.6.18-92* above) and the highest one from the series before (2.6.18-53.1.21.el5 and 2.6.18-8.1.15.el5).
While Red Hat has promised initially a stable ABI within a version (including all updates), this was not always the case - or maybe it isn't clear to me what this promise means. However, the breakages are much more likely to occur when major updates (like RHEL 5.2) come while the minor kernel updates (like 2.6.18-53.1.19.el5) in between usually contain only security fixes (please note my usage of "usually" :-)). So I find it very likely that a module version compiled for one -8.1.x kernel will work on all other -8.1.x kernels and less likely to work on a 53.1.x one. Based on this assumption, keeping one kernel module for each major update and using weak-updates for the minor updates would probably work in a large percentage of cases.
For CentOS in particular, I can see that older kernels are actually not available anymore on mirrors, neither as separate packages nor as part of ISO images. The only way that I can think of to get those older kernels is to install from ISO images written some time ago on physical CD/DVDs - but these ISO images are only released for major updates, so only one kernel version per major update is actually available for those who choose this installation method and don't want to update to the latest and greatest. For this case, again, having one kernel module for each major update seems to be enough.
Disclaimer: I don't have any connection to Red Hat and I'm not a CentOS developer.
On Thu, 2008-07-03 at 18:07 +0200, Bogdan Costescu wrote:
On Thu, 3 Jul 2008, Axel Thimm wrote:
<snip>
For CentOS in particular, I can see that older kernels are actually not available anymore on mirrors, neither as separate packages nor as part of ISO images. The only way that I can think of to get those older kernels is to install from ISO images written some time ago on physical CD/DVDs - but these ISO images are only released for major updates, so only one kernel version per major update is actually available for those who choose this installation method and don't want to update to the latest and greatest. For this case, again, having one kernel module for each major update seems to be enough.
IIRC, older CentOS stuff can be found in archives at the CentOS site. If so, that's another alternative. But I don't know if that includes updates between major releases or not.
If not, the number of kernels that are even considered for support are reduced if one considers that availability of updates between major upgrades as unavailable and CDs that were created could not have these either (only major release snapshots should be on CDs).
snip>