- Thursday, September 9, 2021

release-team@conference.openafs.org

Thursday, September 9, 2021< ^ >

kaduk@jabber.openafs.org/barnowl has set the subject to: openafs release team

Room Configuration

Room Occupants

GMT+0
[01:10:57] Cheyenne leaves the room
[11:54:57] mbarbosa joins the room
[13:18:55] Cheyenne joins the room
[13:53:50] meffie joins the room
[14:57:17] meffie leaves the room
[15:39:40] meffie joins the room
[16:00:46] <kaduk@jabber.openafs.org/barnowl> greetings
[16:00:55] <Cheyenne> Hello all
[16:02:02] <meffie> hello
[16:02:43] <kaduk@jabber.openafs.org/barnowl> [my previous meeting went to the end, so I will be in and out,
refreshing beverage/etc.]
[16:05:45] wiesand joins the room
[16:06:47] <wiesand> Hello. Sorry, I had troubles connecting. Started 7m ago…
[16:07:51] <wiesand> Another apology: I made false promises last week. The week was crazy again…
[16:10:47] <wiesand> But I am watching the progress regarding 1.8.x and FBSD - wow, there was a lot more missing than I thought. If we're serious about FBSD support, https://gerrit.openafs.org/#/q/status:open+project:openafs+branch:master+topic:fbsd12-updates clearly should go in.
[16:11:33] <meffie> yes, maybe we should have a 1.8.8.1 point release just for the freebsd stuff to get it merged?
[16:12:39] <wiesand> That (FBSD) is #2 on my priority list - #1 is Linux 5.15
[16:13:41] <meffie> i think all the freebsd12 patches needed to build and run freebsd12 have been submitted by andrew , but we should verify.
[16:13:44] <kaduk@jabber.openafs.org/barnowl> I regret to report that I did not look at the linux 5.15 changes yet
[16:14:04] <kaduk@jabber.openafs.org/barnowl> He commented on the "tip" commit that he had verified that all the
changes needed for freebsd12 were there
[16:14:36] <wiesand> Mike: maybe, but I think the FBSD stuff is mostly straightforward and I'm not sure it warrants a point release. In any case it will take review review review - and that looks a bit meager right now.
[16:15:05] <Cheyenne> the 5.15 problem itself was simple to identify (and nothing major).  but there is a backward compiler compatibility question.
[16:15:08] <meffie> ok, i'm not really the person to review freebsd changes tho.
[16:15:35] <kaduk@jabber.openafs.org/barnowl> right, with variadic macros
[16:16:13] <kaduk@jabber.openafs.org/barnowl> I do remember it being an issue in the past, but not the
details...which means I might be thinking of, e.g., krb5 rather than
openafs
[16:16:18] <Cheyenne> correct.  One alternative is to change osi_Msg to a static inline (though it requires an extra include in the "right" spot)
[16:16:42] <Cheyenne> .. need stdarg.h somewhere..
[16:17:25] <Cheyenne> -or- to have a couple of osi_Msg macros each with a specific number of args
[16:17:51] <kaduk@jabber.openafs.org/barnowl> Ugh, I hope it doesn't come to that.
[16:18:06] <wiesand> Thanks Cheyenne. My stance on backward compiler compatibility is: let's break it. If it does turn out to be a real problem, we (well, you ;-) can fix it later. Especially since even Andrew isn't sure what the change might break.
[16:18:19] <meffie> we are not using __LINE__ nor __FUNCTION__, why is it a macro??
[16:18:31] <kaduk@jabber.openafs.org/barnowl> I am reminded of
src/afs/afs_segments.c:    /* would like a Trace5, but it doesn't
exist... */
[16:19:35] <Cheyenne> meffie: I can't answer that one.. to me it's one of the mysteries of the old times.
[16:19:48] <meffie> why did you keep it a define?
[16:20:10] <meffie> osi_Msg() should be a function
[16:20:36] <meffie> (sorry for the rant, this old code drives me crazy)
[16:20:45] <Cheyenne> least change.. but I have no problem making it a function (which to me actually makes more sense)
[16:21:27] <kaduk@jabber.openafs.org/barnowl> Please go ahead and prepare a revision to do that
[16:21:36] <wiesand> +1
[16:21:37] <kaduk@jabber.openafs.org/barnowl> we seem to have good support here for that approach
[16:21:38] <meffie> thanks ben!
[16:23:08] <Cheyenne> will do (it's mostly done actually .. just trying to figure out the "correct" spot to insert the prereq #include for stdarg (or I could make osi_Msg an exported function instead of an static inline)
[16:23:19] <meffie> i'm glad this change is limited to just a few files in rx. i was afraid it was going to be a lot more invasive
[16:23:53] <kaduk@jabber.openafs.org/barnowl> wouldn't it be afssysincludes.h ?
[16:23:56] <Cheyenne> every new release drop that first build is "interesting"
[16:24:12] <meffie> i not sure it needs to be inline. we shouldnt be hammering printk :)
[16:25:08] <wiesand> I'm very embarrassed to say that I don't have anything else regarding 1.8.x for today.
[16:25:28] <Cheyenne> Okay.. I'll find a good "home" for it and make it a function.
[16:25:41] <wiesand> Anyone? Or on to master/1.9.x?
[16:25:55] <meffie> i dont have anything else for 1.8.x
[16:26:34] <kaduk@jabber.openafs.org/barnowl> I merged a few of the static-analysis changes and a couple of FBSD
changes, but nothing terribly exciting.
[16:26:56] <kaduk@jabber.openafs.org/barnowl> Not much of a structure to what I did, though, just grabbing at what I
happened to see first.
[16:27:09] <kaduk@jabber.openafs.org/barnowl> So linux 5.15 should be prioritized for this coming week
[16:27:42] <wiesand> That would be good.
[16:27:43] <meffie> thanks Ben.
[16:27:45] <kaduk@jabber.openafs.org/barnowl> not too much else to report from my side, too, I'm afraid
[16:29:54] <wiesand> So maybe it will be a brief one today. Please review the FBSD changes - getting those merged away will help making  further pullups less complicated.
[16:30:32] <meffie> i'll ping ted c and mans to review those.
[16:30:54] <wiesand> Thanks Mike! Anything else to discuss today?
[16:30:56] <kaduk@jabber.openafs.org/barnowl> I will also try to look again at the fbsd changes with the 1.8 view
[16:30:59] <meffie> and do my best as well
[16:31:45] <meffie> ted said he is running with a superset of those patches, so he should be able to give good feedback
[16:32:28] <wiesand> NB the å on a Mac keyboard is Alt-a ;-)
[16:33:19] <meffie> ha! thank you. (this is a ubuntu vm, i'll have to sort out how to spell it properly here)
[16:34:03] <kaduk@jabber.openafs.org/barnowl> I have nothing else for today
[16:34:30] <wiesand> And is pronounced a bit like the "O" in "door". But it depends a bit on which scandinavian country you're in.
[16:34:31] <meffie> i have nothing else
[16:34:42] <wiesand> So, adjourn?
[16:35:16] <Cheyenne> sounds good.
[16:36:09] <meffie> thanks all!
[16:36:18] <Cheyenne> thanks
[16:36:24] <wiesand> Let's adjourn then. Thanks a lot everybody!
[16:36:34] <Cheyenne> have a good week everyone
[16:36:41] <kaduk@jabber.openafs.org/barnowl> thanks everyone; have a good week!
[17:33:29] meffie leaves the room
[19:41:54] mvita2 joins the room
[19:42:43] mvita2 leaves the room
[19:55:38] wiesand leaves the room
[21:53:32] mbarbosa leaves the room