Hello everyone,
Some of you might remember discussions earlier in 2017 (huh, long time ago) on centos-virt list about Linux kernel tty/ldisc related bug, which caused Xen dom0 kernel crashes for some people (and also more easy to trigger crashes on other arches, like on hp pa-risc), but the bug was difficult to reproduce for others/most people.
Fix for this issue was merged to centos virt-sig Linux 4.9.48-29 kernel earlier in 2017, but reverted/deleted in 4.9.75-30 due to problems with the patch. The patch in question was also merged to upstream Linux kernel, and later reverted there aswell.
But now finally the issue has been resolved in upstream Linux, and the patches have been already backported to Linux 4.9.152 longterm stable kernel. Changelog: https://cdn.kernel.org/pub/linux/kernel/v4.x/ChangeLog-4.9.152 You can find the tty/ldisc patches there.
So I propose we build the latest 4.9.155 kernel for CentOS virt-sig for centos6 and centos7.
Thanks,
-- Pasi
Hello,
I built 4.9.155 for both el6 and el7, you can test them here:
https://repo.aerisnetwork.com/archive/kernel-4.9.155-34.el6.x86_64.rpm https://repo.aerisnetwork.com/archive/kernel-4.9.155-34.el7.x86_64.rpm
It works fine for me so far on el6, I'll update my pull request later on GitHub with this version.
[root@node-tmp1 ~]# cat /proc/version Linux version 4.9.155-34.el6.x86_64 (mockbuild@build.aerisnetwork.net) (gcc version 7.3.1 20180303 (Red Hat 7.3.1-5) (GCC) ) #1 SMP Sun Feb 10 11:20:04 EST 2019
Karl
On Sat, Feb 9, 2019 at 10:32 AM Pasi Kärkkäinen pasik@iki.fi wrote:
Hello everyone,
Some of you might remember discussions earlier in 2017 (huh, long time ago) on centos-virt list about Linux kernel tty/ldisc related bug, which caused Xen dom0 kernel crashes for some people (and also more easy to trigger crashes on other arches, like on hp pa-risc), but the bug was difficult to reproduce for others/most people.
Fix for this issue was merged to centos virt-sig Linux 4.9.48-29 kernel earlier in 2017, but reverted/deleted in 4.9.75-30 due to problems with the patch. The patch in question was also merged to upstream Linux kernel, and later reverted there aswell.
But now finally the issue has been resolved in upstream Linux, and the patches have been already backported to Linux 4.9.152 longterm stable kernel. Changelog: https://cdn.kernel.org/pub/linux/kernel/v4.x/ChangeLog-4.9.152 You can find the tty/ldisc patches there.
So I propose we build the latest 4.9.155 kernel for CentOS virt-sig for centos6 and centos7.
Thanks,
-- Pasi
CentOS-virt mailing list CentOS-virt@centos.org https://lists.centos.org/mailman/listinfo/centos-virt
Hi,
On Sun, Feb 10, 2019 at 10:35:56PM -0500, Karl Johnson wrote:
Hello, I built 4.9.155 for both el6 and el7, you can test them here: [1]https://repo.aerisnetwork.com/archive/kernel-4.9.155-34.el6.x86_64.rpm [2]https://repo.aerisnetwork.com/archive/kernel-4.9.155-34.el7.x86_64.rpm It works fine for me so far on el6, I'll update my pull request later on GitHub with this version. [root@node-tmp1 ~]# cat /proc/version Linux version 4.9.155-34.el6.x86_64 ([3]mockbuild@build.aerisnetwork.net) (gcc version 7.3.1 20180303 (Red Hat 7.3.1-5) (GCC) ) #1 SMP Sun Feb 10 11:20:04 EST 2019
Great, thanks a lot! I'll give it a go.
-- Pasi
Karl
On Sat, Feb 9, 2019 at 10:32 AM Pasi KÀrkkÀinen <[4]pasik@iki.fi> wrote:
Hello everyone, Some of you might remember discussions earlier in 2017 (huh, long time ago) on centos-virt list about Linux kernel tty/ldisc related bug, which caused Xen dom0 kernel crashes for some people (and also more easy to trigger crashes on other arches, like on hp pa-risc), but the bug was difficult to reproduce for others/most people. Fix for this issue was merged to centos virt-sig Linux 4.9.48-29 kernel earlier in 2017, but reverted/deleted in 4.9.75-30 due to problems with the patch. The patch in question was also merged to upstream Linux kernel, and later reverted there aswell. But now finally the issue has been resolved in upstream Linux, and the patches have been already backported to Linux 4.9.152 longterm stable kernel. Changelog: [5]https://cdn.kernel.org/pub/linux/kernel/v4.x/ChangeLog-4.9.152 You can find the tty/ldisc patches there. So I propose we build the latest 4.9.155 kernel for CentOS virt-sig for centos6 and centos7. Thanks, -- Pasi _______________________________________________ CentOS-virt mailing list [6]CentOS-virt@centos.org [7]https://lists.centos.org/mailman/listinfo/centos-virtReferences
Visible links
- https://repo.aerisnetwork.com/archive/kernel-4.9.155-34.el6.x86_64.rpm
- https://repo.aerisnetwork.com/archive/kernel-4.9.155-34.el7.x86_64.rpm
- mailto:mockbuild@build.aerisnetwork.net
- mailto:pasik@iki.fi
- https://cdn.kernel.org/pub/linux/kernel/v4.x/ChangeLog-4.9.152
- mailto:CentOS-virt@centos.org
- https://lists.centos.org/mailman/listinfo/centos-virt
CentOS-virt mailing list CentOS-virt@centos.org https://lists.centos.org/mailman/listinfo/centos-virt
Hi,
On Mon, Feb 11, 2019 at 12:49:30PM +0200, Pasi Kärkkäinen wrote:
Hi,
On Sun, Feb 10, 2019 at 10:35:56PM -0500, Karl Johnson wrote:
Hello, I built 4.9.155 for both el6 and el7, you can test them here: [1]https://repo.aerisnetwork.com/archive/kernel-4.9.155-34.el6.x86_64.rpm [2]https://repo.aerisnetwork.com/archive/kernel-4.9.155-34.el7.x86_64.rpm It works fine for me so far on el6, I'll update my pull request later on GitHub with this version. [root@node-tmp1 ~]# cat /proc/version Linux version 4.9.155-34.el6.x86_64 ([3]mockbuild@build.aerisnetwork.net) (gcc version 7.3.1 20180303 (Red Hat 7.3.1-5) (GCC) ) #1 SMP Sun Feb 10 11:20:04 EST 2019
Great, thanks a lot! I'll give it a go.
With initial testing seems to work OK for me! Meanwhile upstream kernel is up to v4.9.161, so let's update the sig-virt kernel to that version..
Karl: Can you prepare a PR?
Thanks,
-- Pasi
-- Pasi
Karl
On Sat, Feb 9, 2019 at 10:32 AM Pasi KÀrkkÀinen <[4]pasik@iki.fi> wrote:
Hello everyone, Some of you might remember discussions earlier in 2017 (huh, long time ago) on centos-virt list about Linux kernel tty/ldisc related bug, which caused Xen dom0 kernel crashes for some people (and also more easy to trigger crashes on other arches, like on hp pa-risc), but the bug was difficult to reproduce for others/most people. Fix for this issue was merged to centos virt-sig Linux 4.9.48-29 kernel earlier in 2017, but reverted/deleted in 4.9.75-30 due to problems with the patch. The patch in question was also merged to upstream Linux kernel, and later reverted there aswell. But now finally the issue has been resolved in upstream Linux, and the patches have been already backported to Linux 4.9.152 longterm stable kernel. Changelog: [5]https://cdn.kernel.org/pub/linux/kernel/v4.x/ChangeLog-4.9.152 You can find the tty/ldisc patches there. So I propose we build the latest 4.9.155 kernel for CentOS virt-sig for centos6 and centos7. Thanks, -- Pasi
On Wed, Feb 27, 2019 at 4:44 AM Pasi Kärkkäinen pasik@iki.fi wrote:
Hi,
On Mon, Feb 11, 2019 at 12:49:30PM +0200, Pasi Kärkkäinen wrote:
Hi,
On Sun, Feb 10, 2019 at 10:35:56PM -0500, Karl Johnson wrote:
Hello, I built 4.9.155 for both el6 and el7, you can test them here: [1]
https://repo.aerisnetwork.com/archive/kernel-4.9.155-34.el6.x86_64.rpm
[2]
https://repo.aerisnetwork.com/archive/kernel-4.9.155-34.el7.x86_64.rpm
It works fine for me so far on el6, I'll update my pull request
later on
GitHub with this version. [root@node-tmp1 ~]# cat /proc/version Linux version 4.9.155-34.el6.x86_64 ([3]
mockbuild@build.aerisnetwork.net)
(gcc version 7.3.1 20180303 (Red Hat 7.3.1-5) (GCC) ) #1 SMP Sun
Feb 10
11:20:04 EST 2019
Great, thanks a lot! I'll give it a go.
With initial testing seems to work OK for me! Meanwhile upstream kernel is up to v4.9.161, so let's update the sig-virt kernel to that version..
Karl: Can you prepare a PR?
Thanks,
-- Pasi
I built 4.9.161, works for me on el6:
[root@node-tmp1 ~]# cat /proc/version Linux version 4.9.161-34.el6.x86_64 (mockbuild@build.aerisnetwork.net) (gcc version 7.3.1 20180303 (Red Hat 7.3.1-5) (GCC) ) #1 SMP Wed Feb 27 09:48:40 EST 2019
Only one strange thing, I had to run grub-install because Grub couldn't find root device after reboot, not sure this has something to do with the kernel upgrade. Kernel 4.9.161 doesn't have the patch for CVE-2019-8912 yet so we might wait until it get released.
I updated my PR: https://github.com/CentOS-virt7/xen-kernel/pull/18
https://repo.aerisnetwork.com/archive/kernel-4.9.161-34.el6.x86_64.rpm https://repo.aerisnetwork.com/archive/kernel-4.9.161-34.el7.x86_64.rpm
Anthony Perard already tried to build my PR on the CBS but he was not able to enable the SCL repo ( http://mirror.centos.org/centos/$releasever/sclo/$basearch/rh/) to get devtoolset-7. We might need to get help from centos-devel.
Karl
-- Pasi
Karl
On Sat, Feb 9, 2019 at 10:32 AM Pasi KÀrkkÀinen <[4]pasik@iki.fi>
wrote:
Hello everyone, Some of you might remember discussions earlier in 2017 (huh, longtime
ago) on centos-virt list about Linux kernel tty/ldisc relatedbug, which
caused Xen dom0 kernel crashes for some people (and also moreeasy to
trigger crashes on other arches, like on hp pa-risc), but the bugwas
difficult to reproduce for others/most people. Fix for this issue was merged to centos virt-sig Linux 4.9.48-29kernel
earlier in 2017, but reverted/deleted in 4.9.75-30 due toproblems with
the patch. The patch in question was also merged to upstream Linux kernel, and later reverted there aswell. But now finally the issue has been resolved in upstream Linux,and the
patches have been already backported to Linux 4.9.152 longtermstable
kernel. Changelog: [5]https://cdn.kernel.org/pub/linux/kernel/v4.x/ChangeLog-4.9.152 You can find the tty/ldisc patches there. So I propose we build the latest 4.9.155 kernel for CentOSvirt-sig for
centos6 and centos7. Thanks, -- Pasi
Hi,
On Wed, Feb 27, 2019 at 03:12:40PM -0500, Karl Johnson wrote:
On Wed, Feb 27, 2019 at 4:44 AM Pasi KÀrkkÀinen <[1]pasik@iki.fi> wrote:
Hi, On Mon, Feb 11, 2019 at 12:49:30PM +0200, Pasi KÀrkkÀinen wrote: > Hi, > > On Sun, Feb 10, 2019 at 10:35:56PM -0500, Karl Johnson wrote: > >  Hello, > >  I built 4.9.155 for both el6 and el7, you can test them here: > >  [1][2]https://repo.aerisnetwork.com/archive/kernel-4.9.155-34.el6.x86_64.rpm > >  [2][3]https://repo.aerisnetwork.com/archive/kernel-4.9.155-34.el7.x86_64.rpm > >  It works fine for me so far on el6, I'll update my pull request later on > >  GitHub with this version. > >  [root@node-tmp1 ~]# cat /proc/version > >  Linux version 4.9.155-34.el6.x86_64 ([3][4]mockbuild@build.aerisnetwork.net) > >  (gcc version 7.3.1 20180303 (Red Hat 7.3.1-5) (GCC) ) #1 SMP Sun Feb 10 > >  11:20:04 EST 2019 > > Great, thanks a lot! I'll give it a go. > With initial testing seems to work OK for me! Meanwhile upstream kernel is up to v4.9.161, so let's update the sig-virt kernel to that version.. Karl: Can you prepare a PR? Thanks, -- PasiI built 4.9.161, works for me on el6:
Nice!
[root@node-tmp1 ~]# cat /proc/version Linux version 4.9.161-34.el6.x86_64 ([5]mockbuild@build.aerisnetwork.net) (gcc version 7.3.1 20180303 (Red Hat 7.3.1-5) (GCC) ) #1 SMP Wed Feb 27 09:48:40 EST 2019 Only one strange thing, I had to run grub-install because Grub couldn't find root device after reboot, not sure this has something to do with the kernel upgrade. Kernel 4.9.161 doesn't have the patch for CVE-2019-8912 yet so we might wait until it get released. I updated my PR: [6]https://github.com/CentOS-virt7/xen-kernel/pull/18 [7]https://repo.aerisnetwork.com/archive/kernel-4.9.161-34.el6.x86_64.rpm [8]https://repo.aerisnetwork.com/archive/kernel-4.9.161-34.el7.x86_64.rpm
I'll give it a go soon.
Anthony Perard already tried to build my PR on the CBS but he was not able to enable the SCL repo ([9]http://mirror.centos.org/centos/$releasever/sclo/$basearch/rh/) to get devtoolset-7. We might need to get help from centos-devel.
Karl
OK. Something to discuss on #centos-devel perhaps?
Thanks a lot!
-- Pasi
On Wed, Feb 27, 2019 at 03:12:40PM -0500, Karl Johnson wrote:
I built 4.9.161, works for me on el6:
[root@node-tmp1 ~]# cat /proc/version Linux version 4.9.161-34.el6.x86_64 (mockbuild@build.aerisnetwork.net) (gcc version 7.3.1 20180303 (Red Hat 7.3.1-5) (GCC) ) #1 SMP Wed Feb 27 09:48:40 EST 2019
Only one strange thing, I had to run grub-install because Grub couldn't find root device after reboot, not sure this has something to do with the kernel upgrade. Kernel 4.9.161 doesn't have the patch for CVE-2019-8912 yet so we might wait until it get released.
I updated my PR: https://github.com/CentOS-virt7/xen-kernel/pull/18
Hi Karl, I've build and tested your PR, but without the GCC7 patch, and when I've tested it on el6, none of the guests had network access. I had to revert the bridge changes to have them working (CONFIG_BRIDGE_NETFILTER=m and CONFIG_BRIDGE=m).
Why did you want the BRIDGE built-in instead of a module?
Maybe you have something to setup the bridge (probably filter) properly, or maybe something is different in the xen package between el6 and el7. Any idea?
I still have to investigate the Xen pkg, and have a closer look at those tests I'm running.
Thanks,
On Mon, Mar 4, 2019 at 6:13 AM Anthony PERARD anthony.perard@citrix.com wrote:
On Wed, Feb 27, 2019 at 03:12:40PM -0500, Karl Johnson wrote:
I built 4.9.161, works for me on el6:
[root@node-tmp1 ~]# cat /proc/version Linux version 4.9.161-34.el6.x86_64 (mockbuild@build.aerisnetwork.net)
(gcc
version 7.3.1 20180303 (Red Hat 7.3.1-5) (GCC) ) #1 SMP Wed Feb 27
09:48:40
EST 2019
Only one strange thing, I had to run grub-install because Grub couldn't find root device after reboot, not sure this has something to do with the kernel upgrade. Kernel 4.9.161 doesn't have the patch for CVE-2019-8912 yet so we might wait until it get released.
I updated my PR: https://github.com/CentOS-virt7/xen-kernel/pull/18
Hi Karl, I've build and tested your PR, but without the GCC7 patch, and when I've tested it on el6, none of the guests had network access. I had to revert the bridge changes to have them working (CONFIG_BRIDGE_NETFILTER=m and CONFIG_BRIDGE=m).
Why did you want the BRIDGE built-in instead of a module?
Maybe you have something to setup the bridge (probably filter) properly, or maybe something is different in the xen package between el6 and el7. Any idea?
I still have to investigate the Xen pkg, and have a closer look at those tests I'm running.
Hum strange. I switched from module to built-in because I always had a deprecated warning about bridge when booting CentOS 6 (and I guess everyone needs bridge?). I don't remember what was the exact warning because it's been a long time since I switched it. I don't mind to put it back as a module, both works for me but the warning is gone when built-in.
Karl
Hi,
On Mon, Mar 04, 2019 at 04:22:19PM -0500, Karl Johnson wrote:
On Mon, Mar 4, 2019 at 6:13 AM Anthony PERARD <[1]anthony.perard@citrix.com> wrote:
On Wed, Feb 27, 2019 at 03:12:40PM -0500, Karl Johnson wrote: > I built 4.9.161, works for me on el6: > > [root@node-tmp1 ~]# cat /proc/version > Linux version 4.9.161-34.el6.x86_64 ([2]mockbuild@build.aerisnetwork.net) (gcc > version 7.3.1 20180303 (Red Hat 7.3.1-5) (GCC) ) #1 SMP Wed Feb 27 09:48:40 > EST 2019 > > Only one strange thing, I had to run grub-install because Grub couldn't > find root device after reboot, not sure this has something to do with the > kernel upgrade. > Kernel 4.9.161 doesn't have the patch for CVE-2019-8912 yet so we might > wait until it get released. > > I updated my PR: [3]https://github.com/CentOS-virt7/xen-kernel/pull/18 Hi Karl, I've build and tested your PR, but without the GCC7 patch, and when I've tested it on el6, none of the guests had network access. I had to revert the bridge changes to have them working (CONFIG_BRIDGE_NETFILTER=m and CONFIG_BRIDGE=m). Why did you want the BRIDGE built-in instead of a module? Maybe you have something to setup the bridge (probably filter) properly, or maybe something is different in the xen package between el6 and el7. Any idea? I still have to investigate the Xen pkg, and have a closer look at those tests I'm running.Hum strange. I switched from module to built-in because I always had a deprecated warning about bridge when booting CentOS 6 (and I guess everyone needs bridge?). I don't remember what was the exact warning because it's been a long time since I switched it. I don't mind to put it back as a module, both works for me but the warning is gone when built-in.
Yeah better to keep things as modules, like they have been.
btw upstream Linux 4.9 LTS kernel is now at 4.9.162 :)
Karl
-- Pasi
Hi,
On Tue, Mar 05, 2019 at 11:16:20PM +0200, Pasi Kärkkäinen wrote:
Hi,
On Mon, Mar 04, 2019 at 04:22:19PM -0500, Karl Johnson wrote:
> > I updated my PR: [3]https://github.com/CentOS-virt7/xen-kernel/pull/18 Hi Karl, I've build and tested your PR, but without the GCC7 patch, and when I've tested it on el6, none of the guests had network access. I had to revert the bridge changes to have them working (CONFIG_BRIDGE_NETFILTER=m and CONFIG_BRIDGE=m). Why did you want the BRIDGE built-in instead of a module? Maybe you have something to setup the bridge (probably filter) properly, or maybe something is different in the xen package between el6 and el7. Any idea? I still have to investigate the Xen pkg, and have a closer look at those tests I'm running.Hum strange. I switched from module to built-in because I always had a deprecated warning about bridge when booting CentOS 6 (and I guess everyone needs bridge?). I don't remember what was the exact warning because it's been a long time since I switched it. I don't mind to put it back as a module, both works for me but the warning is gone when built-in.
Yeah better to keep things as modules, like they have been.
btw upstream Linux 4.9 LTS kernel is now at 4.9.162 :)
I suggest we:
1) Rebase to linux 4.9.162. 2) Drop the "set bridge to built-in instead of module" changes. Let's keep those settings like they were earlier. 3) Drop the use of gcc7 for now.
How does that sound?
Thanks,
-- Pasi
On Tue, Mar 12, 2019 at 01:21:01PM +0200, Pasi Kärkkäinen wrote:
On Mon, Mar 04, 2019 at 04:22:19PM -0500, Karl Johnson wrote:
> > I updated my PR: [3]https://github.com/CentOS-virt7/xen-kernel/pull/18 Hi Karl, I've build and tested your PR, but without the GCC7 patch, and when I've tested it on el6, none of the guests had network access. I had to revert the bridge changes to have them working (CONFIG_BRIDGE_NETFILTER=m and CONFIG_BRIDGE=m). Why did you want the BRIDGE built-in instead of a module? Maybe you have something to setup the bridge (probably filter) properly, or maybe something is different in the xen package between el6 and el7. Any idea? I still have to investigate the Xen pkg, and have a closer look at those tests I'm running.Hum strange. I switched from module to built-in because I always had a deprecated warning about bridge when booting CentOS 6 (and I guess everyone needs bridge?). I don't remember what was the exact warning because it's been a long time since I switched it. I don't mind to put it back as a module, both works for me but the warning is gone when built-in.
Yeah better to keep things as modules, like they have been.
btw upstream Linux 4.9 LTS kernel is now at 4.9.162 :)
I suggest we:
- Rebase to linux 4.9.162.
- Drop the "set bridge to built-in instead of module" changes. Let's keep those settings like they were earlier.
- Drop the use of gcc7 for now.
How does that sound?
Karl: Would you like to update the github PR ? Or should someone else pick up the kernel update work?
Thanks,
-- Pasi
Thanks,
-- Pasi
On Fri, Mar 22, 2019 at 4:43 AM Pasi Kärkkäinen pasik@iki.fi wrote:
On Tue, Mar 12, 2019 at 01:21:01PM +0200, Pasi Kärkkäinen wrote:
On Mon, Mar 04, 2019 at 04:22:19PM -0500, Karl Johnson wrote:
> > I updated my PR: [3]https://github.com/CentOS-virt7/xen-kernel/pull/18
Hi Karl, I've build and tested your PR, but without the GCC7patch, and
when I've tested it on el6, none of the guests had networkaccess.
I had to revert the bridge changes to have them working (CONFIG_BRIDGE_NETFILTER=m and CONFIG_BRIDGE=m). Why did you want the BRIDGE built-in instead of a module? Maybe you have something to setup the bridge (probably filter)properly,
or maybe something is different in the xen package between el6and el7.
Any idea? I still have to investigate the Xen pkg, and have a closer lookat those
tests I'm running.Hum strange. I switched from module to built-in because I always
had a
deprecated warning about bridge when booting CentOS 6 (and I guess everyone needs bridge?). I don't remember what was the exact
warning
because it's been a long time since I switched it. I don't mind
to put it
back as a module, both works for me but the warning is gone when
built-in.
Yeah better to keep things as modules, like they have been.
btw upstream Linux 4.9 LTS kernel is now at 4.9.162 :)
I suggest we:
- Rebase to linux 4.9.162.
- Drop the "set bridge to built-in instead of module" changes. Let's
keep those settings like they were earlier.
- Drop the use of gcc7 for now.
How does that sound?
Karl: Would you like to update the github PR ? Or should someone else pick up the kernel update work?
Thanks,
-- Pasi
Sure, I will update the PR in the next few days.
Karl
On Sat, Mar 23, 2019 at 2:07 PM Karl Johnson karljohnson.it@gmail.com wrote:
On Fri, Mar 22, 2019 at 4:43 AM Pasi Kärkkäinen pasik@iki.fi wrote:
On Tue, Mar 12, 2019 at 01:21:01PM +0200, Pasi Kärkkäinen wrote:
On Mon, Mar 04, 2019 at 04:22:19PM -0500, Karl Johnson wrote:
> > I updated my PR: [3]https://github.com/CentOS-virt7/xen-kernel/pull/18
Hi Karl, I've build and tested your PR, but without the GCC7patch, and
when I've tested it on el6, none of the guests had networkaccess.
I had to revert the bridge changes to have them working (CONFIG_BRIDGE_NETFILTER=m and CONFIG_BRIDGE=m). Why did you want the BRIDGE built-in instead of a module? Maybe you have something to setup the bridge (probably filter)properly,
or maybe something is different in the xen package between el6and el7.
Any idea? I still have to investigate the Xen pkg, and have a closerlook at those
tests I'm running.Hum strange. I switched from module to built-in because I always
had a
deprecated warning about bridge when booting CentOS 6 (and I
guess
everyone needs bridge?). I don't remember what was the exact
warning
because it's been a long time since I switched it. I don't mind
to put it
back as a module, both works for me but the warning is gone when
built-in.
Yeah better to keep things as modules, like they have been.
btw upstream Linux 4.9 LTS kernel is now at 4.9.162 :)
I suggest we:
- Rebase to linux 4.9.162.
- Drop the "set bridge to built-in instead of module" changes. Let's
keep those settings like they were earlier.
- Drop the use of gcc7 for now.
How does that sound?
Karl: Would you like to update the github PR ? Or should someone else pick up the kernel update work?
Thanks,
-- Pasi
Sure, I will update the PR in the next few days.
Karl
I've updated my PR and tested a build, it boots well:
[root@node-tmp1 ~]# cat /proc/version Linux version 4.9.165-35.el6.x86_64 (mockbuild@build.aerisnetwork.net) (gcc version 4.4.7 20120313 (Red Hat 4.4.7-23) (GCC) ) #1 SMP Sat Mar 23 14:44:16 EDT 2019
Test rpms: https://repo.aerisnetwork.com/archive/
Anthony will be able to build it on the CBS as devtoolset has been removed from the spec.
Karl
On Sat, Mar 23, 2019 at 05:00:00PM -0400, Karl Johnson wrote:
On Sat, Mar 23, 2019 at 2:07 PM Karl Johnson <[1]karljohnson.it@gmail.com> wrote:
Sure, I will update the PR in the next few days. KarlI've updated my PR and tested a build, it boots well: [root@node-tmp1 ~]# cat /proc/version Linux version 4.9.165-35.el6.x86_64 ([4]mockbuild@build.aerisnetwork.net) (gcc version 4.4.7 20120313 (Red Hat 4.4.7-23) (GCC) ) #1 SMP Sat Mar 23 14:44:16 EDT 2019 Test rpms: [5]https://repo.aerisnetwork.com/archive/ Anthony will be able to build it on the CBS as devtoolset has been removed from the spec.
With quick testing the 4.9.165-35.el6 kernel seems to work OK for me!
Anthony: Can you please review the updated PR on github, merge it and kick of the builds for el6 and el7 ?
Karl
Thanks,
-- Pasi
On Sun, Mar 24, 2019 at 01:12:53PM +0200, Pasi Kärkkäinen wrote:
On Sat, Mar 23, 2019 at 05:00:00PM -0400, Karl Johnson wrote:
On Sat, Mar 23, 2019 at 2:07 PM Karl Johnson <[1]karljohnson.it@gmail.com> wrote:
Sure, I will update the PR in the next few days. KarlI've updated my PR and tested a build, it boots well: [root@node-tmp1 ~]# cat /proc/version Linux version 4.9.165-35.el6.x86_64 ([4]mockbuild@build.aerisnetwork.net) (gcc version 4.4.7 20120313 (Red Hat 4.4.7-23) (GCC) ) #1 SMP Sat Mar 23 14:44:16 EDT 2019 Test rpms: [5]https://repo.aerisnetwork.com/archive/ Anthony will be able to build it on the CBS as devtoolset has been removed from the spec.
With quick testing the 4.9.165-35.el6 kernel seems to work OK for me!
Anthony: Can you please review the updated PR on github, merge it and kick of the builds for el6 and el7 ?
Done, I've pushed kernel-4.9.165-35.el7 and kernel-4.9.165-35.el6 to testing.
Hi,
On Tue, Mar 26, 2019 at 05:36:42PM +0000, Anthony PERARD wrote:
On Sun, Mar 24, 2019 at 01:12:53PM +0200, Pasi Kärkkäinen wrote:
On Sat, Mar 23, 2019 at 05:00:00PM -0400, Karl Johnson wrote:
On Sat, Mar 23, 2019 at 2:07 PM Karl Johnson <[1]karljohnson.it@gmail.com> wrote:
Sure, I will update the PR in the next few days. KarlI've updated my PR and tested a build, it boots well: [root@node-tmp1 ~]# cat /proc/version Linux version 4.9.165-35.el6.x86_64 ([4]mockbuild@build.aerisnetwork.net) (gcc version 4.4.7 20120313 (Red Hat 4.4.7-23) (GCC) ) #1 SMP Sat Mar 23 14:44:16 EDT 2019 Test rpms: [5]https://repo.aerisnetwork.com/archive/ Anthony will be able to build it on the CBS as devtoolset has been removed from the spec.
With quick testing the 4.9.165-35.el6 kernel seems to work OK for me!
Anthony: Can you please review the updated PR on github, merge it and kick of the builds for el6 and el7 ?
Done, I've pushed kernel-4.9.165-35.el7 and kernel-4.9.165-35.el6 to testing.
Great, thanks a lot! I grabbed kernel-4.9.165-35.el6 from testing, and it seems to boot and work OK for me with quick testing.
-- Anthony PERARD
Thanks,
-- Pasi
-
Anthony PERARD -
Karl Johnson -
Pasi Kärkkäinen