[02:51:58] --- Jeffrey Altman has left: Replaced by new connection [02:52:00] --- Jeffrey Altman has become available [03:10:03] --- Jeffrey Altman has left: Replaced by new connection [03:10:10] --- Jeffrey Altman has become available [05:14:49] --- meffie has become available [06:06:29] --- squinney has become available [06:54:46] --- Simon Wilkinson has become available [06:55:32] --- Stephan Wiesand has become available [07:01:06] Good morning/afternoon. Thanks for being here. [07:01:14] hi [07:01:40] Who else actually is? [07:01:54] hi [07:02:37] I will be here and not. [07:02:52] hi [07:03:26] Let's start with 1.6.2.1? [07:04:30] Is that not done yet? :) [07:05:03] An OS X problem in 1.6.2 was found and addressed in 9747. Does anyone think we should branch and include that fix in 1.6.2.1? [07:05:16] No [07:05:49] I thought 1.6.2.1 had been released [07:06:24] I don't think an official announcment has been sent, but I thought it was tagged and everything? [07:06:27] The web changes haven't been merged, and the announcement is sitting in the moderator's custody. [07:06:51] Other than that: yes, it's been "released". [07:07:22] --- deason has become available [07:07:45] I think if there is going to be an update for osx it should be independent of 1.6.2.1. 1.6.2.2 perhaps but 1.6.2.1 should fly [07:07:50] So, even if we wanted to slip something in, we couldn't... [07:08:04] If noone thinks anything else is wrong with 1.6.2.1, no need to discuss it here. [07:08:11] And I think we're close enough to 1.6.3 (and 1.6.2.1 took long enough), that I don't think there's any value to a 1.6.2.2 [07:08:22] +1 [07:08:33] That's what I think. Just wanted to ask... [07:08:57] Ok, on to 1.6.3. [07:09:25] The hot topic is certainly 9695. [07:10:23] --- Jeffrey Altman has left: Disconnected [07:10:59] There doesn't seem to be anything specifically negative about the code. [07:11:15] That's something ;-) [07:11:35] vos certainly has no need to use vlib.a [07:12:10] I have to admit I didn't quite get that one. [07:12:21] do we really need to be worrying about that [07:12:58] It's just about an example in the commit message? [07:13:22] --- Jeffrey Altman has become available [07:14:04] What will it take to get 9694 merged on master? [07:14:10] And how long? [07:15:26] Or is it debatable whether it should go in at all? [07:16:32] The change seems completely fine. [07:16:43] i though there was a consensus it was to go in, as a way forward. [07:16:50] I haven't heard anyone objecting to it going in... chaskiel hasn't said a whole lot, but you can only wait so long [07:16:55] I don't actually think that the function needs to be inline, anyway, regardless of linker madness. [07:17:06] From what I remember it seemed fine. (I was bad and haven't done my code review yet.) [07:17:52] Derrick would need to comment on whether it's suitable for inclusion or not, but I really can't see there being a problem. It looks like a perfect middle ground. [07:18:35] Thanks Simon. You were the one I expected to obect, maybe. [07:19:11] So it just needs to be done. [07:19:46] I built 1.6 + 9695 today. It's not running an a server yet, though. [07:20:37] note that 9695 hasn't been updated yet to reflect the changes in the 9694 submission [07:20:57] which is basically just changing the option names... but it's just something that "needs to be done" once 9694 is in [07:21:34] Ok, thanks for the heads up. [07:21:59] Other changes I'm inclined to merge before pre1: [07:22:19] 8946/7 [07:22:27] 8818 [07:22:34] 9661 [07:23:08] 9597 + 9770 (need more review) [07:23:33] Thoughts on those? [07:23:51] --- Jeffrey Altman has left: Replaced by new connection [07:24:19] --- Jeffrey Altman has become available [07:24:59] Mike, I'm not sure, but 9065 could be required to make some of the other xstat changes work? [07:25:07] WRT 8818 - can Ben give it to whoever he knows that's trying to build on Irix and check that its okay? [07:26:38] we can also build it on the irix builder [07:26:45] actually, I should do that now [07:27:02] Stephan: nothing from me on those; those are fine choices by me [07:27:12] Andrew: Good. But hadn't that been done already? [07:27:12] or "... to me", whatever [07:27:26] 9065 was to to fix the length checks and so we can safely add more cm call stats (as fs flushall does) [07:28:16] Stephan Wiesand: that's why I said I submitted that originally, but I didn't actually say anywhere if it built; I'm not sure if I forgot to do it or just forgot to say what happened [07:28:51] also, I'm not sure if there are more desired openbsd changes from antoine? if anything further comes in, I say they can just go in if they are obsd-only, though [07:28:55] I think that in addition to 8818 we needed another patch, but my guy got frustrated and built 1.4.x instead. Let me look. [07:29:31] EGOTO2 makes my eyes hurt... [07:30:34] I think most of the OpenBSD-only changes are in already. [07:31:00] We would also need 9608 pulled up to build IRIX, I believe. [07:32:07] 9608 looks trivial enough that including it shouldn't be a problem [07:33:09] It's one of those not obviously trivial to me. [07:34:01] I'm learning a lot about what's going on in the code these days. Just give me another decade... [07:34:36] 9608 just removes a duplicate definition of the variable. [07:34:46] no, I don't think that change is actually correct, unless b64_string_t and afs_ino_str_t are the same or similar enough for that not to be a problem [07:35:11] That's what I was wondering about. [07:35:17] the change it references, 38cf3146, added an 'stmp', but there was an existing 'stmp' in that function that was for something else [07:36:01] Who'll pull it up? [07:37:13] I wouldn't want to delay a pre1 for IRIX. [07:37:18] ...but it doesn't technically matter, since afs_ino_str_t is larger than b64_string_t, so that should not be a problem [07:37:27] looks like derrick just did [07:37:31] afs_ino_str_t is a char [AFS_INO_STR_LENGTH], a b64_string_t is a char[8]. AFS_INO_STR_LENGTH is 32 [07:38:19] Yes, he did. 9771. If the experts +1 it, I have no problem merging it. [07:40:07] Simon, Derrick, thanks a lot for the reviews. [07:40:35] sorry i was late. i was actually in front of the computer but ignoring this window. [07:40:49] WRT 9065, I don't particularly like it, but I should have raised my objections before it was merged on master. [07:40:49] 9771 would be more useful if it depended on 8818 [07:41:48] --- Jeffrey Altman has left: Disconnected [07:42:06] Well, they're independent. [07:42:25] he presumably means for verification [07:42:30] Yeah, but Andrew wants to be able to ask the Irix builder to build with both of them. [07:42:50] there. rebased both. [07:43:04] --- jaltman/t61p has become available [07:43:38] Ah, ok. [07:43:54] Buildbot is still a bit of a mystery to me. [07:44:26] When I push a series of patches, it seems to build the last one first? [07:45:27] Yeah. Our builbot is running with a load of glue scripts that Jason put together. [07:46:28] in this case, though, we're not talking about the stuff that build things automatically, but we can submit things to manually [07:47:46] What's new in version 5 of 9694? [07:48:25] documentation mistakes I -1'd in the previous patchset [07:49:33] Ah. [07:53:41] It seems everything is on its way. A couple of changes got sufficient review during the meeting. 9694/5 will set the pace. [07:53:56] Time to start working on release notes. [07:54:44] I asked last week whether it really makes sense to work on them on master and not 1.6.x, but got no response. Anyone? [07:55:55] The question is what we should ship 1.8 (or whatever) with - do they have the 1.6 series release notes or not? [07:57:55] when we stopped tracking version numbers on master first I stopped tracking the release notes on master. for 1.7 release notes only go on the 1.7 branch. if we ever start to issue builds from master on a regular basis those release notes will need to be different [07:59:12] If they're relevant to the next release, they can still be added when the new branch is created? [07:59:29] And if they aren't, well... [08:02:07] I think I'll push the draft to 1_6_x. Feel free to -2 it then :-) [08:02:19] Sounds fine to me [08:02:35] i'm sure it will be fine [08:02:53] Thanks. Any other business today? [08:04:23] (latency is considerable today - let's wait a minute) [08:05:13] what will happen in a minute? ;) [08:05:23] no further business from me [08:05:32] hearing none [08:05:50] Derrick: I'll stop waiting for anyone bringing up anything else, and call it a day. [08:06:04] go on home. enjoy your evening. [08:06:17] But it seems everyone is happy. [08:06:24] Thanks a lot everyone! [08:06:29] thanks [08:06:55] I'll send minutes later. Bye. [08:07:03] --- Stephan Wiesand has left [08:24:11] --- meffie has left [08:25:43] --- Jeffrey Altman has become available [08:26:09] --- Jeffrey Altman has left: Disconnected [08:33:30] --- squinney has left [09:11:49] --- Jeffrey Altman has become available [09:12:16] --- Jeffrey Altman has left: Disconnected [09:19:26] --- Jeffrey Altman has become available [09:20:26] --- Jeffrey Altman has left: Disconnected [10:01:50] --- Jeffrey Altman has become available [10:01:55] --- Jeffrey Altman has left: Disconnected [10:15:29] --- Jeffrey Altman has become available [10:15:33] --- Jeffrey Altman has left: Disconnected [10:26:45] --- Jeffrey Altman has become available [10:26:57] --- jaltman/t61p has left: Disconnected [11:17:36] --- Simon Wilkinson has left [11:36:23] --- Simon Wilkinson has become available [15:14:19] --- deason has left [15:15:08] --- deason has become available [15:50:19] --- deason has left