- Wednesday, July 16, 2014

release-team@conference.openafs.org

Wednesday, July 16, 2014< ^ >

Room Configuration

Room Occupants

GMT+0
[06:33:25] shadow@gmail.com/barnowl853202E0 leaves the room
[06:33:37] shadow@gmail.com/barnowl853202E0 joins the room
[06:48:18] shadow@gmail.com/barnowl853202E0 leaves the room
[06:48:30] shadow@gmail.com/barnowl853202E0 joins the room
[11:18:05] wiesand joins the room
[12:39:58] wiesand joins the room
[13:36:31] meffie joins the room
[13:48:21] deason joins the room
[13:49:06] kaduk joins the room
[14:01:03] <wiesand> Hello
[14:01:34] <meffie> greetings
[14:01:38] <kaduk> g'day
[14:03:45] <kaduk> I started looking at the stack-reduction pullups.
[14:03:54] <wiesand> Thanks.
[14:04:08] <kaduk> gerrit does not always show enough context to verify the correctness of the error handling, so I end up pulling up the file in a terminal alongside.
[14:04:22] <wiesand> (and generally thanks for the review that happened in the last days)
[14:04:55] <wiesand> I know reviewing changes is serious work.
[14:05:32] <wiesand> Indispensable though. Think heartbleed.
[14:06:57] <meffie> thank you kaduk
[14:07:38] <kaduk> I am slightly tempted to hack up nl(1) to accept an environment variable making it default to -b a, because it feels silly to always have to pass those arguments.
[14:09:08] <kaduk> I was poking at the jabber logs, and it looked like Simon poked his head into the openafs room a few days ago.
[14:09:31] <deason> alias nl='nl -b a' ?
[14:10:25] <meffie> yeah, i was going to say, sounds like alias would do.
[14:10:38] <kaduk> Probably.
[14:10:54] <kaduk> At least the nl(1) on the machine in front of me DTRT with repeated -b arguments.
[14:12:46] <meffie> (and yes, -b a should be the default :)
[14:15:06] <kaduk> Now you're going to tell me that I can have an alias for 'git show origin/openafs-stable-1_6_x:src/afs/afs_osi_pag.c', too ;)
[14:17:43] <meffie> git config aliases :)
[14:20:27] <wiesand> I'm looking at some older changes... what should we do with 11075? It seems a replacement isn't coming up.
[14:28:04] <kaduk> The affected population seems small enough that I don't see a need to wait for a "complete" fix.
[14:31:06] <wiesand> Yes, we should probably include it unless we have anything better when we get closer to a prerelease.
[14:31:58] <wiesand> Regarding your -1 on 10946, I'm not sure it's a bad idea to have the information that it wasn't a clean pullup in the commit message.
[14:32:45] <kaduk> It's probably not a bad idea; I would put it in a different format, though.
[14:33:49] <meffie> oh, yeah, newer versions of git does that, if here was a merge confict in the cherry pick, and then you do --continue.
[14:34:15] <wiesand> It's what git gives you automatically, so it seems kind of standard to me.
[14:34:16] <meffie> i think they would be nice to keep.
[14:35:05] <kaduk> Well, if people want to keep them like that, I can go remove the -1...
[14:36:08] <wiesand> I'm wondering what others think. And please remove your -1 if and only if you really don't object.
[14:36:10] <deason> I thought git always did that; I always removed it because they're never there and it seems like noise
[14:36:43] <meffie> it's one less thing to hand edit, and it shows which files needed to be resolved.
[14:38:06] <meffie> i've only seen it where there are conflicts, then you run cherry-pick --continue
[14:38:58] <shadow@gmail.com/barnowl853202E0> we've never left the conflicts in any commit messages
[14:39:10] <kaduk> I'm pretty sure I've seen it other places, but I don't remember where.
[14:39:10] <meffie> (i've been removing them because i dont want any -1's from ben)
[14:39:17] <kaduk> :)
[14:41:30] <wiesand> Looks like a gatekeeper ruled that they should be removed. Fine. I'll repush the change.
[14:42:32] <shadow@gmail.com/barnowl853202E0> "ruled"
[14:43:08] <wiesand> Isn't that what gatekeepers do?
[14:43:28] <shadow@gmail.com/barnowl853202E0> rule. drool. whatever.
[14:43:56] <meffie> well, i dont feel strongly about it. we need to edit the commit message anyway.
[14:44:32] <meffie> what's the harm in just one more straw ;)
[14:52:43] <deason> it seems like something that could be noted in gerrit, but staying around with the commit forever seems pointless
[14:53:14] <deason> but even noting in gerrit I would be concerned about over-emphasizing review on just files with marked conflicts
[14:55:27] <wiesand> On the other hand, "cherry picked from commit ..." simply isn't the (whole) truth.
[14:56:19] <meffie> exactly
[14:56:57] <kaduk> It's the truth ... from a certain point of view. "Just like Darth Vader killed Anakin Skywalker."
[14:59:32] <meffie> heh
[15:01:22] meffie leaves the room
[15:03:17] <deason> cherry-picked from a commit doesn't mean it's exactly identical to the original commit, even if there are no conflicts
[15:05:40] <wiesand> Patch set 2 is up there for review now ;-)
[15:11:43] <wiesand> I should be able to merge quite a few changes now.
[15:12:02] <wiesand> Stack reduction passed only partially though.
[15:21:43] meffie joins the room
[15:56:14] meffie leaves the room
[15:56:14] meffie joins the room
[16:12:26] meffie leaves the room
[16:23:18] wiesand leaves the room
[16:32:49] <kaduk> Do we know when we expect Simon to return?
[16:32:49] wiesand leaves the room
[17:44:28] Jeffrey Altman leaves the room
[17:44:29] Jeffrey Altman joins the room
[17:49:08] Jeffrey Altman leaves the room
[17:49:08] Jeffrey Altman joins the room
[18:36:32] kaduk leaves the room
[22:55:12] deason leaves the room