As pointed out by Justing Forbes in the last FESCo Meeting [1], the kernel FTBFS in rawhide for x86 arch.
I fired a scratch build in koji to check the error [2]. Which throws
ERROR: "__udivdi3" [net/netfilter/xt_hashlimit.ko] undefined!
As Alexander Ploumistos pointed out in another thread, the issue is already known upstream.
A patch for the issue was already applied upstream. See [4] and [5].
I applied the patch to our kernel and fired a scratch koji build [6], which built successfully, and a COPR build [7], which was still running in the moment I wrote this message. The reason for the COPR build was that I needed to make some changes to the way I was applying the patch and I felt bad for keeping the build servers busy.
My question is, what is next? Do we just wait for some of the kernel packagers to pull the changes related to this issue, or should I submit this patch to the mailing list?
[1] http://meetbot-raw.fedoraproject.org/teams/fesco/fesco.2017-09-08-16.00.log.... [2] https://koji.fedoraproject.org/koji/taskinfo?taskID=21739202 [3] https://lists.fedoraproject.org/archives/list/x86@lists.fedoraproject.org/th... [4] https://patchwork.kernel.org/patch/9943187/ [5] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i... [6] https://koji.fedoraproject.org/koji/taskinfo?taskID=21765118 [7] https://copr.fedorainfracloud.org/coprs/athoscr/kernel/build/600404/
On Sun, Sep 10, 2017 at 5:36 PM, Athos Ribeiro athoscribeiro@gmail.com wrote:
My question is, what is next? Do we just wait for some of the kernel packagers to pull the changes related to this issue, or should I submit this patch to the mailing list?
I've had just one experience with kernel patches and it pretty much played out like this:
* I filed a bug for the problem I was seeing and I attached a patch I had ready for some kernel driver. * I was asked to submit the patch upstream, which I did. * At some point the patch went into something-next. * When it did, I posted it on the bug report and I think it was Josh Boyer who backported the patch into currently supported kernels.
As far as I can tell, we do not have a bug opened for this issue, or do we?
On Sun, Sep 10, 2017 at 05:57:04PM +0300, Alexander Ploumistos wrote:
As far as I can tell, we do not have a bug opened for this issue, or do we?
AFAIK, no, we don't.
On Sun, Sep 10, 2017 at 6:06 PM, Athos Ribeiro athoscribeiro@gmail.com wrote:
On Sun, Sep 10, 2017 at 05:57:04PM +0300, Alexander Ploumistos wrote:
As far as I can tell, we do not have a bug opened for this issue, or do we?
AFAIK, no, we don't.
I'll take care of it.
On Sun, Sep 10, 2017 at 11:29 AM, Alexander Ploumistos alex.ploumistos@gmail.com wrote:
Filed #1490123: https://bugzilla.redhat.com/show_bug.cgi?id=1490123 _______________________________________________ X86 mailing list -- x86@lists.fedoraproject.org To unsubscribe send an email to x86-leave@lists.fedoraproject.org
Nice work, folks!
Yes, I agree with the process. Definitely file a Fedora bug against the kernel. If it is FTBS, they can apply the patch when building the Fedora package until it gets merged in upstream.
jeff
On Sun, Sep 10, 2017 at 5:36 PM, Athos Ribeiro athoscribeiro@gmail.com wrote:
I applied the patch to our kernel and fired a scratch koji build [6], which built successfully, and a COPR build [7], which was still running in the moment I wrote this message. The reason for the COPR build was that I needed to make some changes to the way I was applying the patch and I felt bad for keeping the build servers busy.
And of course kudos for all the work!
-
Alexander Ploumistos -
Athos Ribeiro -
Jeff Backus