16:04:30 <bahamat> jclulow: I have a fix for 14697, but gerrit tells me I don't have create permission.
16:05:21 <bahamat> I'm not sure if my permissions got screwed up somehow, or if I got culled at some point.
16:08:36 <pmooney> bahamat: what was the error? (and what ref were you trying to push?)
16:09:00 <bahamat> pmooney: remote: You need 'Create' rights to create new references.
16:09:25 <bahamat> Pushing this branch: https://github.com/bahamat/illumos-joyent/commit/469640213ff339c80e379a8406971c6adb1b6938
16:09:45 <danmcd> bahamat: I can push to gerrit on your behalf if you wish.
16:09:53 <danmcd> (though getting perms is the better long-term solution.)
16:10:00 <bahamat> It used to work.
16:10:30 <danmcd> Is it because you're now @mnx.io instead of @joyent.com ?  That should've smoothly migrated...
16:10:36 <bahamat> If there was a policy change about who's allowed to push, I must have missed it.
16:10:51 <bahamat> gerrit has my mnx address in my profile.
16:10:56 <danmcd> Hmmph.
16:10:59 <andyf> Pushing HEAD:refs/for/master ?
16:11:11 <bahamat> remote: error: branch refs/heads/i_14967:
16:11:35 <andyf> For a new review, you need to use refs/for/master as far as I know
16:11:36 <pmooney> you're trying to push the branch name itself, instead of to the create-a-new-change-for-me ref
16:11:38 <danmcd> You did do "git push you⊙cio:/illumos-gate HEAD:refs/for/master" ?
16:11:39 <pmooney> yeah
16:11:51 <bahamat> Oh...ok.
16:11:59 <pmooney> HEAD assumes you have the branch in question checked out at the time
16:12:18 <pmooney> it'll also require a Change-id in the commit message for gerrit's tracking purposes
16:15:56 <bahamat> I always forget about that HEAD:refs/for/master
16:16:05 <bahamat> I think I ask about that every damn time :-/
16:32:09 <tsoome> alias gerrit='git push gerrit HEAD:refs/for/master' ;)
19:41:10 <gitomat> [illumos-gate] 14236 signed math leads getelfshdr astray -- Patrick Mooney <pmooney⊙pc>
19:41:39 <pmooney> *confetti*
19:42:43 <richlowe> one of the great advantages of using gerrit properly, will be not being allowed to push to master by mistake
19:42:49 <richlowe> which terrifies me
19:43:07 <pmooney> yeah, I'll appreciate that safety
19:43:22 <andyf> wrapper scripts with lots of checks ftw
19:43:24 <pmooney> (but am grumpy about how it means Change-id garbage in the commit messages)
19:43:54 <pmooney> andyf: doesn't prevent me from accidentlly doing `push cr master` instead of `push cr HEAD:refs/for/master`
19:44:48 <pmooney> I'm also not wild about scripts that do automatic attribution, as some folks have gerrit-specific emails configured
19:46:02 <andyf> I was not really suggesting scripts are better than automation there..
19:46:29 <andyf> but I have hidden the refs/for/master thing in a git alias, so just `git pg` which means I never even remember it without looking
19:47:39 <pmooney> I'm a git weirdo who does a lot of stuff by hand
19:47:51 <pmooney> (like always fetch/merge instead of pull)
19:47:52 <richlowe> I'm a git weirdo who doesn't trust magic near illumos repos
19:48:05 <richlowe> and even then I've fucked it up at least twice
19:48:09 <andyf> and then I borrowed jlevon's "push-to-gate" script which does the check things
20:24:34 <nahamu> Is there a plan to put gerrit in charge of landing things?
20:24:41 <richlowe> what do you mean by "plan"?
20:24:52 <richlowe> there's like, 8 people going "Gosh, I wish I had time for that"
20:24:56 <richlowe> is that a plan?
20:24:56 <nahamu> Haha
20:25:12 <nahamu> I was looking for a mailing list thread or IPD or something.
20:26:21 <jclulow> There is not a concrete plan
20:26:35 <jclulow> At some point we could just rip the band-aid off tbh
20:27:37 <nahamu> As someone who rarely integrates anything, I was more curious and surprised rather than feeling like I have an "Important Opinion" (TM)
20:28:17 <nahamu> I think I interpreted pmooney's "I'll appreciate that safety" as implying that there was a concrete plan.
20:28:30 <nahamu> I'll go back to being quiet now. :)
20:28:41 <richlowe> the safety is that when you do it, none of us can push to master anymore
20:28:43 <richlowe> except jclulow
20:28:46 <richlowe> and unfortunately probably me :(
20:29:08 <jclulow> richlowe: I have a pattern for a sudo mode
20:29:15 <jclulow> so you can have a switch to turn on/off for doing it
20:54:03 <richlowe> does anyone know off the top of my head if I mark something __unused then use it, can I get gcc to tell me?
20:58:12 <tsoome> I do not think so...  
21:04:22 <richlowe> doh
21:08:15 <yuripv> heh, https://gcc.gnu.org/bugzilla/show_bug.cgi?id=51579, looks like no one was interested
23:14:24 <gitomat> [illumos-gate] 14867 panic in smb_request_alloc / smb_oplock_ind_break during shutdown -- Gordon Ross <gordon.ross⊙tc>