Home
release-team@conference.openafs.org
Thursday, August 11, 2022< ^ >
kaduk@jabber.openafs.org/barnowl has set the subject to: openafs release team
Room Configuration
Room Occupants

GMT+0
[15:05:10] wiesand joins the room
[15:56:24] meffie joins the room
[15:56:46] <meffie> happy thursday
[15:57:05] <kaduk@jabber.openafs.org/barnowl> greetings
[16:00:10] Cheyenne joins the room
[16:01:35] <Cheyenne> Hello all
[16:03:41] <wiesand> Hello
[16:04:55] <wiesand> So I "merged away" Cheyenne's stack. Huge thanks for preparing this!
[16:05:17] <kaduk@jabber.openafs.org/barnowl> huge thanks to Cheyenne for preparing it; huge thanks to wiesand for
making it happen :)
[16:05:29] <meffie> yay!
[16:08:09] <wiesand> I'm behind on pulling up a few more changes .
[16:09:12] <Cheyenne> The 5.19 changes should fit fairly easily if you want to include those.
[16:11:13] <meffie> shall we push 1.8.x versions of those to the openafs-stable-1_8_x branch on gerrit?
[16:11:23] <wiesand> 15406 (trivial), 15068 (not that trivial). And the Linux 5.20 ones.
[16:13:19] <wiesand> Yes, at this point we should probably include the 5.19 changes. Please push those.
[16:13:32] <Cheyenne> Okay.  I'll take care of those
[16:13:36] <meffie> thanks Cheyenne!
[16:13:37] <kaduk@jabber.openafs.org/barnowl> Maybe I haven't had enough tea yet this morning, but ... which Linux
5.20 changes?
[16:14:35] <Cheyenne> there will be some 5.20 changes.. I'm just waiting for at least the first RC tag .  The change seems pretty simple
[16:15:07] <kaduk@jabber.openafs.org/barnowl> Ah, okay.
[16:15:42] <wiesand> I was probably talking about 5.19, with 5.20 in my mind. Sorry for that.
[16:15:51] <Cheyenne> Ah.. okay :)
[16:16:20] <kaduk@jabber.openafs.org/barnowl> Fair enough :)
[16:18:10] <Cheyenne> so far for 5.20 it simply looks like a change in the name of one of the linux api's that is called from 2 spots
[16:18:34] <kaduk@jabber.openafs.org/barnowl> With a side helping of future GPLONLY worries?
[16:18:55] <Cheyenne> correct
[16:19:11] <Cheyenne> possible GPLONLY worries I guess.  
[16:19:16] <wiesand> Tiring…
[16:20:58] <Cheyenne> Oh... the gentoo folks came up with a nice twist -- they tried to turn on LTO (link-time-optimization) -- which causes some errors due to some type mismatches found during linking
[16:21:33] <meffie> what does that mean?
[16:22:21] <Cheyenne> the one instance that I found (without much digging) was one extern was a char *x and another was a struct foo *x
[16:22:54] <meffie> ok
[16:23:25] <wiesand> Sounds like a can of worms though…
[16:23:32] <Cheyenne> yes
[16:24:05] <kaduk@jabber.openafs.org/barnowl> it's legal to access a struct foo *x via a char * pointer ... but
maybe you have to give it a different name for that access.
[16:24:29] <meffie> agreed
[16:24:49] <Cheyenne> prdir is the offending external (see ptprocs.c and ptubik.c)
[16:25:37] <meffie> thanks
[16:26:33] <Cheyenne> I didn't dig any deeper however.. I was just curious on what type of errors cropped up and that was the first I stumbled on.
[16:26:58] <kaduk@jabber.openafs.org/barnowl> Understood.  Still, an interesting topic to think about.
[16:27:35] <Cheyenne> yeah.. I was thinking of adding that to a static analysis type of scan
[16:27:44] <wiesand> good job, anyway
[16:27:53] <meffie> looking , it might be externed for testing or somesuch.
[16:30:33] <wiesand> I don't have more on 1.8.x for today. The to-do-list is clear. Alas, the drought is keeping to occupy my time :-((
[16:31:01] <kaduk@jabber.openafs.org/barnowl> I have little on master/1.9; I've been looking at a couple things,
just slowly.
[16:31:04] <Cheyenne> I'll get the 5.19 stuff pushed up
[16:31:20] <wiesand> Thanks!
[16:32:11] <kaduk@jabber.openafs.org/barnowl> Thanks!
[16:32:17] <kaduk@jabber.openafs.org/barnowl> Oh!
[16:32:27] <kaduk@jabber.openafs.org/barnowl> Were we going to talk about 1.8.8.2 vs 1.8.9 today?
[16:33:49] <meffie> i guess i dont have a strong preference for the version number, as long as we get it out soon :)
[16:33:51] <Cheyenne> the only items I can think for master are .. 14993 (minor cleanup), 14903 (minor fix), and 14889 (build for fix for opensuse)
[16:33:59] <wiesand> I'm behind on the release notes too. But my feeling is that we should call it 1.8.9 .
[16:34:10] <kaduk@jabber.openafs.org/barnowl> (Looking over the `git log --oneline` output, I would be inclined
towards 1.8.9 as well)
[16:34:12] <Cheyenne> yeah either is fine.
[16:34:34] <meffie> 1.8.9 sounds reasonable
[16:35:06] <wiesand> Fine, 1.8.9 it shall be.
[16:35:47] <meffie> thanks wiesand
[16:36:35] <kaduk@jabber.openafs.org/barnowl> Cheyenne: thanks for the pointers re master changes ... sad to see I
looked at two of them already and then failed to follow up on the
replies/updates :(
[16:36:41] <kaduk@jabber.openafs.org/barnowl> thanks weisand
[16:37:14] <meffie> there is also the change for the bug pointed out by IBM at the workshop...
[16:37:30] <meffie> (looking for the gerrit number)
[16:38:12] <kaduk@jabber.openafs.org/barnowl> https://gerrit.openafs.org/#/c/15038/ I think?
[16:38:20] <meffie> 15038 Avoid more out of bounds indexing when checking volume names
[16:38:25] <meffie> yes
[16:38:28] <meffie> yes
[16:38:33] <kaduk@jabber.openafs.org/barnowl> :high-five:
[16:38:38] <meffie> lol
[16:39:02] <wiesand> er… ?
[16:39:42] <kaduk@jabber.openafs.org/barnowl> ok.
Wiesand has a plan for 1.8.9.
I have a plan for master changes to look at.
Wiesand is confused about what just happened.  Let's fix that, then
move to adjourn...
[16:40:09] <meffie> heh, i have nothing more. thanks.
[16:40:16] <kaduk@jabber.openafs.org/barnowl> There was a workshop talk "memory error detector" that pointed out
some real segfaults that can occur, and Mike put together 15038 with a
fix.
[16:41:02] <kaduk@jabber.openafs.org/barnowl> When talking about changes for master, we remembered that there was
"that fix triggered by the workshop talk", and both Mike and I went to
try to find it at the same time, and we found it at the same time, and
both posted the link/info about it.
[16:41:23] <meffie> yes, ok have a good day all, need to logout.
[16:41:38] meffie leaves the room
[16:41:40] <Cheyenne> Okay.. 5.19 for 1.8.x should be pushed soon
[16:42:00] <kaduk@jabber.openafs.org/barnowl> So I think the "lol"-worthy/amusing bit was just that we found it at
the same time.  Or maybe that the "yes" double-sent.  But not anything
of great import, certainly.
[16:42:51] <kaduk@jabber.openafs.org/barnowl> Is wiesand less confused now?
[16:42:59] <wiesand> Thanks. CU.
[16:43:04] wiesand leaves the room
[16:43:08] <kaduk@jabber.openafs.org/barnowl> Great.
Have a good week everyone!
[22:35:45] Cheyenne leaves the room