[arch-dev-public] Minimal kernel version
Hi guys, As far as I understand the minimal kernel version we support is 2.6.27. This is at least due to glibc and udev. I thought it would be worth noting that as of its next release udev will require kernel version 2.6.32: <http://git.kernel.org/?p=linux/hotplug/udev.git;a=commit;h=67a77c8bf299f6264f001677becd056316ebce2f>. Maybe this could be an opportunity to raise the requirement across the board, and also to make a news item about this? Cheers, Tom
On Thu, May 5, 2011 at 12:58 PM, Tom Gundersen <teg@jklm.no> wrote:
Hi guys,
As far as I understand the minimal kernel version we support is 2.6.27. This is at least due to glibc and udev.
I thought it would be worth noting that as of its next release udev will require kernel version 2.6.32: <http://git.kernel.org/?p=linux/hotplug/udev.git;a=commit;h=67a77c8bf299f6264f001677becd056316ebce2f>.
Maybe this could be an opportunity to raise the requirement across the board, and also to make a news item about this?
Not to be the thorn in the side of progress, but we can't keep changing this every six months on a whim unless we truly don't care about some semblance of stability and uptime. We last changed this less than 5 months ago[1], and before that, 21 months ago[2]. I think there needs to be an extremely valid reason someone can't have a machine with uptime approaching a year in duration. I'm sure there are people running pre-2.6.32 kernels out there even though we don't currently package one- it was only released in December 2009, so 18 months ago. So obviously if you update udev on your system, you should be expected to run a kernel that satisfies the requirement, but I don't think we should force people to jump from the 2.6.27 level we established only 5 months ago just yet if at all possible. -Dan [1] http://www.archlinux.org/news/minimum-required-kernel-version-increased-1/ [2] http://www.archlinux.org/news/udev-minimum-kernel-version/
Le 5 mai 2011 14:20:31, Dan McGee a écrit :
On Thu, May 5, 2011 at 12:58 PM, Tom Gundersen <teg@jklm.no> wrote:
Hi guys,
As far as I understand the minimal kernel version we support is 2.6.27. This is at least due to glibc and udev.
I thought it would be worth noting that as of its next release udev will require kernel version 2.6.32: <http://git.kernel.org/?p=linux/hotplug/udev.git;a=commit;h=67a77c8bf299f 6264f001677becd056316ebce2f>.
Maybe this could be an opportunity to raise the requirement across the board, and also to make a news item about this?
Not to be the thorn in the side of progress, but we can't keep changing this every six months on a whim unless we truly don't care about some semblance of stability and uptime. We last changed this less than 5 months ago[1], and before that, 21 months ago[2]. I think there needs to be an extremely valid reason someone can't have a machine with uptime approaching a year in duration. I'm sure there are people running pre-2.6.32 kernels out there even though we don't currently package one- it was only released in December 2009, so 18 months ago.
So obviously if you update udev on your system, you should be expected to run a kernel that satisfies the requirement, but I don't think we should force people to jump from the 2.6.27 level we established only 5 months ago just yet if at all possible.
-Dan
[1] http://www.archlinux.org/news/minimum-required-kernel-version-increased-1/ [2] http://www.archlinux.org/news/udev-minimum-kernel-version/
I might be wrong on this because I do not have all the information, but I think that one of the consequence of our choice to set minimum version to 2.6.27 is that we need to patch the glibc to avoid a threading issue on x86_64[1]. This problem do not seems to happen if glibc is builded for 2.6.32. Again, I do not have all information on this. Allan might have a more informed opinion on this issue. Stéphane [1] http://sourceware.org/bugzilla/show_bug.cgi?id=12403
On 06/05/11 05:06, Stéphane Gaudreault wrote:
Le 5 mai 2011 14:20:31, Dan McGee a écrit :
On Thu, May 5, 2011 at 12:58 PM, Tom Gundersen<teg@jklm.no> wrote:
Hi guys,
As far as I understand the minimal kernel version we support is 2.6.27. This is at least due to glibc and udev.
I thought it would be worth noting that as of its next release udev will require kernel version 2.6.32: <http://git.kernel.org/?p=linux/hotplug/udev.git;a=commit;h=67a77c8bf299f 6264f001677becd056316ebce2f>.
Maybe this could be an opportunity to raise the requirement across the board, and also to make a news item about this?
Not to be the thorn in the side of progress, but we can't keep changing this every six months on a whim unless we truly don't care about some semblance of stability and uptime. We last changed this less than 5 months ago[1], and before that, 21 months ago[2]. I think there needs to be an extremely valid reason someone can't have a machine with uptime approaching a year in duration. I'm sure there are people running pre-2.6.32 kernels out there even though we don't currently package one- it was only released in December 2009, so 18 months ago.
So obviously if you update udev on your system, you should be expected to run a kernel that satisfies the requirement, but I don't think we should force people to jump from the 2.6.27 level we established only 5 months ago just yet if at all possible.
-Dan
[1] http://www.archlinux.org/news/minimum-required-kernel-version-increased-1/ [2] http://www.archlinux.org/news/udev-minimum-kernel-version/
I might be wrong on this because I do not have all the information, but I think that one of the consequence of our choice to set minimum version to 2.6.27 is that we need to patch the glibc to avoid a threading issue on x86_64[1]. This problem do not seems to happen if glibc is builded for 2.6.32.
Again, I do not have all information on this. Allan might have a more informed opinion on this issue.
Stéphane
That is a small issue that is fixed and no longer a concern, so it not a reason to increase minimum required kernel version. When I increased the required version to 2.6.27, I selected that version because that was the latest supported upstream (i.e. by the kernel developers). That situation still has not changed. I will note that Fedora (and RHEL6?) uses 2.6.32 as their minimum. Allan
On Thu, May 5, 2011 at 8:20 PM, Dan McGee <dpmcgee@gmail.com> wrote:
Not to be the thorn in the side of progress, but we can't keep changing this every six months on a whim unless we truly don't care about some semblance of stability and uptime.
Side-note: Changing the minimum regularly does not mean that people need to upgrade regularly. Hopefully the people who were forced to upgrade before, did not upgrade to the new minimum, but to something relatively recent. In other words, the people who upgraded five months ago, should be good to go for at least another year (fingers crossed).
So obviously if you update udev on your system, you should be expected to run a kernel that satisfies the requirement
This was exactly my concern. If people hold back udev to avoid updating the kernel, they would also need to hold back other packages that might depend on changes in udev, and so on. Which packages to hold back will probably not be at all clear (as backwards compatibility might be unintentionally broken). What do we recommend people to do in these situations? Cheers, Tom
Am 05.05.2011 20:20, schrieb Dan McGee:
Maybe this could be an opportunity to raise the requirement across the board, and also to make a news item about this?
Not to be the thorn in the side of progress, but we can't keep changing this every six months on a whim unless we truly don't care about some semblance of stability and uptime.
Either we are bleeding edge or we aren't. udev doesn't work completely with kernels older than 2.6.32, and we want the latest version. Not updating udev is not an option for me.
Am Fri, 06 May 2011 09:57:12 +0200 schrieb Thomas Bächler <thomas@archlinux.org>:
Am 05.05.2011 20:20, schrieb Dan McGee:
Maybe this could be an opportunity to raise the requirement across the board, and also to make a news item about this?
Not to be the thorn in the side of progress, but we can't keep changing this every six months on a whim unless we truly don't care about some semblance of stability and uptime.
Either we are bleeding edge or we aren't. udev doesn't work completely with kernels older than 2.6.32, and we want the latest version. Not updating udev is not an option for me.
Is there no udev-compat anymore to work with older kernels? -Andy
Am 06.05.2011 18:08, schrieb Andreas Radke:
Either we are bleeding edge or we aren't. udev doesn't work completely with kernels older than 2.6.32, and we want the latest version. Not updating udev is not an option for me.
Is there no udev-compat anymore to work with older kernels?
-Andy
The commit references in Tom's mail does not include any changes to the compat rules.
On Fri, May 6, 2011 at 6:08 PM, Andreas Radke <a.radke@arcor.de> wrote:
Am Fri, 06 May 2011 09:57:12 +0200 schrieb Thomas Bächler <thomas@archlinux.org>:
Am 05.05.2011 20:20, schrieb Dan McGee:
Maybe this could be an opportunity to raise the requirement across the board, and also to make a news item about this?
Not to be the thorn in the side of progress, but we can't keep changing this every six months on a whim unless we truly don't care about some semblance of stability and uptime.
Either we are bleeding edge or we aren't. udev doesn't work completely with kernels older than 2.6.32, and we want the latest version. Not updating udev is not an option for me.
Is there no udev-compat anymore to work with older kernels?
There is, but it no longer contains rules to support kernels pre 2.6.32 (in fact it only contains a few rules to support 2.6.32, 2.6.33 and newer does not need it). Cheers, Tom
On Fri, May 6, 2011 at 6:08 PM, Andreas Radke <a.radke@arcor.de> wrote:
Am Fri, 06 May 2011 09:57:12 +0200
schrieb Thomas Bächler <thomas@archlinux.org>:
Am 05.05.2011 20:20, schrieb Dan McGee:
Maybe this could be an opportunity to raise the requirement across the board, and also to make a news item about this?
Not to be the thorn in the side of progress, but we can't keep changing this every six months on a whim unless we truly don't care about some semblance of stability and uptime.
Either we are bleeding edge or we aren't. udev doesn't work completely with kernels older than 2.6.32, and we want the latest version. Not updating udev is not an option for me.
Is there no udev-compat anymore to work with older kernels?
There is, but it no longer contains rules to support kernels pre 2.6.32 (in fact it only contains a few rules to support 2.6.32, 2.6.33 and newer does not need it).
Cheers,
Tom If someone has the need to run older kernels, those should be smooth enough to keep an older udev version. I agree with Thomas, we are bleeding egde and if users need stuff that is older
Am Freitag 06 Mai 2011 schrieb Tom Gundersen: than half a yea, you should be able to manage this yourself. greetings tpowa -- Tobias Powalowski Archlinux Developer & Package Maintainer (tpowa) http://www.archlinux.org tpowa@archlinux.org
On Sat, 2011-05-07 at 07:53 +0200, Tobias Powalowski wrote:
If someone has the need to run older kernels, those should be smooth enough to keep an older udev version. I agree with Thomas, we are bleeding egde and if users need stuff that is older than half a yea, you should be able to manage this yourself.
The problem with raising the kernel version is that glibc won't operate on older kernels at all. In some situations you don't even need udev, or you can get away with not updating udev. Not updating glibc is not an option for anyone. At this moment I have a chroot setup which runs on a debian system with 2.6.26 still (yes, I should update the kernel of that box to 2.6.32). For that, I already need the "openvz" repository with glibc packages created by Ionut. Note that as soon as you update glibc on an outdated system that does not run 2.6.32 yet, all dynamic linked binaries on your system will fail to run, meaning that all post_* scripts will fail and a clean reboot of the system is not even possible anymore. We should be aware of that before just raising the kernel version in glibc to just another version.
participants (8)
-
Allan McRae
-
Andreas Radke
-
Dan McGee
-
Jan de Groot
-
Stéphane Gaudreault
-
Thomas Bächler
-
Tobias Powalowski
-
Tom Gundersen