16:00:09 <pjgeorg> #startmeeting Kmods SIG
16:00:09 <centbot> Meeting started Mon Mar  7 16:00:09 2022 UTC.  The chair is pjgeorg. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:09 <centbot> Useful Commands: #action #agreed #help #info #idea #link #topic.
16:00:15 <jcpunk> hello
16:00:30 <pjgeorg> Hi
16:01:54 <pjgeorg> Some report about what happened till last meeting.
16:02:04 <pjgeorg> Concerning issue #19
16:02:11 <pjgeorg> #link https://pagure.io/centos-sig-kmods/sig/issue/19
16:02:30 <pjgeorg> It seems Red Hat by now considers this a bug
16:02:36 <jcpunk> excellent!
16:02:38 <pjgeorg> #link https://bugzilla.redhat.com/show_bug.cgi?id=2054595
16:03:21 <pjgeorg> Honestly any other discussion would have been really bad for us and anybody else working on out-of-tree drivers for EL.
16:03:36 <pjgeorg> s/discussion/decision/
16:04:06 <pjgeorg> But nothing official, all comments are private. Like with most kernel bug reports :(
16:04:24 <jcpunk> I was a bit shocked that it ended up localized to __pci_register_driver  as I thought that was a mostly stable symbol upstream
16:05:23 <pjgeorg> Same for me.
16:06:24 <pjgeorg> Once I found out it is due to __pci_register_driver I had to check the stablelist twice to believe it is also not listed there.
16:07:25 <jcpunk> I'm not even sure what sort of new features/behavior it would be getting
16:08:52 <pjgeorg> I can't say. I haven't checked why this new member **dev_groups has been added, i.e., what it is being used for.
16:11:55 <pjgeorg> Probably a good example of an advantage possible thanks to CentOS Stream: With this change first landing in CentOS Stream we have been able to detect and report the bug and Red Hat kernel developers are now able to fix it before the next EL minor release. I.e. the bug never made it into an EL released kernel.
16:12:07 <jcpunk> :)
16:12:41 <pjgeorg> Concerning issue #20
16:12:45 <pjgeorg> #link https://pagure.io/centos-sig-kmods/sig/issue/20
16:12:55 <pjgeorg> I haven't had time to look into it yet.
16:13:30 <pjgeorg> Same for #22
16:13:34 <pjgeorg> #link https://pagure.io/centos-sig-kmods/sig/issue/22
16:14:00 <pjgeorg> However, according to the description, at least this one is not time critical.
16:16:29 <pjgeorg> I also still need to update our centos-release-kmods package including update SIG key.
16:16:33 <pjgeorg> #link https://lists.centos.org/pipermail/centos-devel/2022-March/120262.html
16:17:35 <pjgeorg> #action pjgeorg Update centos-release-kmods with re-signed SIG key
16:20:28 <pjgeorg> I'd also like to change the schedule of our meeting from Monday every other week to first Monday of every month.
16:20:36 <jcpunk> works for me
16:22:25 <pjgeorg> #agreed Regular meeting on the first Monday (16:00 UTC) of each month
16:23:51 <pjgeorg> Nothing more to report from my side, except that it seems that we are now past RHLE 9.0 freeze.
16:24:17 <jcpunk> I've been getting that feeling based on some of the bugzillas I'm watching
16:24:40 <pjgeorg> That + EPEL has frozen the buildroot
16:25:05 <pjgeorg> Any probably the most obvious one:
16:25:08 <pjgeorg> #link https://gitlab.com/redhat/centos-stream/src/kernel/centos-stream-9/-/merge_requests/566
16:25:28 <jcpunk> ohh, didn't see that one
16:26:52 <pjgeorg> For us it means that the kernel development will now probably slow down and it is not that easy anymore to get changes upstream.
16:27:05 <pjgeorg> At least not for 9.0 as these do afaik now require an exception.
16:47:20 <pjgeorg> #link https://github.com/CentOS/Calendar/pull/54
16:54:52 <pjgeorg> Time is almost up.
17:00:01 <pjgeorg> #endmeeting