| IRC logs at
Set by AlexDaniel on 12 June 2018.
00:31 patrickb joined 00:35 patrickz left 01:39 patrickb left 01:51 Altai-man_ joined 01:54 sena_kun left 03:52 sena_kun joined 03:54 Altai-man_ left 05:51 Altai-man_ joined 05:54 sena_kun left 06:17 vesper11 left, vesper joined 07:52 sena_kun joined 07:54 Altai-man_ left 09:51 Altai-man_ joined 09:54 sena_kun left
Geth MoarVM: 8f2d0d7818 | (Vittore F. Scolari)++ | src/io/syncfile.c
Detect files that cannot be synched also on MacOS

MacOS reports ENOTSUP as error code when trying to fsync a pipe. After this commit, MoarVM detects this condition and ignores the error.
MoarVM: 03968485a1 | niner++ (committed using GitHub Web editor) | src/io/syncfile.c
Merge pull request #1226 from scovit/master

Detect files that cannot be synched also on MacOS
11:52 sena_kun joined
Geth MoarVM: 703f023d53 | (Stefan Seifert)++ | docs/ChangeLog
Update ChangeLog for 2020.01 release
11:54 Altai-man_ left
nine Can we please have a release? Over the past few weeks I've fixed all the issues that plagued us at work and when business opens again on Tuesday, I'd like us to be able to move past them. 11:55
sena_kun nine, is val-related blocker fixed? 11:58
or another option is to revert that
s/that/changes/ 11:59
nine I'm pretty sure that can be reverted in a release branch
sena_kun nine, ok. is there anyone in moarvm team who can do a release with github tags, signs and so on if I'll green light it?
if "yes", then I'd go for it 12:00
nine If there were, we'd probably have had a 2019.12 release :/
I know jnthn and samcv could do releases, but neither have been really here for some time 12:01
sena_kun ok, let me draft a plan of actions and starting preparations then... 12:02
nine, I am bumping moar/nqp 12:07
nine ok 12:15
sena_kun lizmat, ping? 12:19
lizmat pong 12:23
12:23 sivoais left
lizmat sena_kun: pong 12:23
sena_kun lizmat, o/ 12:24
I guess we are aiming for a release, will it be comfortable for you to revert val-related changes in a release branch and then continue to work on it after it'd be merged?
to clarify, the release branch is not created yet. :) 12:28
sena_kun run spec after bumping
lizmat sena_kun: if I revert the val work, then we need to revert the test changes as well 12:30
as otherwise we will see massive failures
both can be done
sena_kun lizmat, nice! I'll ping you when release branch will be available 12:31
lizmat grrr
sena_kun as for test changes - sure, roast revert is a sure option too
lizmat I seem to not be able to just revert it
it doesn't tell me why 12:32
ah, looks like I need to resolve some conflicts
sena_kun lizmat, you don't need to rush, we don't have a branch yet anyway 12:33
lizmat lemme see if I can fix the num stress test, I had an idea last night 12:34
sena_kun lizmat, as in, resolve the blocker without revert?
lizmat yeah 12:35
sena_kun lizmat, good luck!
fatal: unable to access '': The requested URL returned error: 403 12:36
can anyone please check if I have right to commit to nqp? 12:37
hmm, maybe that's my settings..
lizmat ok, that idea doesn't work :-( 12:44
nine sena_kun: sorry, don't seem to have admin privileges on any of our repos 12:45
sena_kun lizmat, sigh, that's sad. maybe you can help with checking rights on nqp repo though?
Geth MoarVM: 83393ef5a9 | (Stefan Seifert)++ | docs/ChangeLog
Update release name in ChangeLog
lizmat sena_kun: where should I look ?
sena_kun lizmat, nqp repo, settings are, hmmm...
lizmat and then?
I'm collaborators and teams 12:47
Seems I have "write" privileges on nqp
no maintain or admin
sena_kun :S 12:48
whatever, I'll do a PR, I guess... <- 12:50
I also can't accept PRs, so yes, it seems I don't have write access
lizmat merged the PR 12:51
nine lizmat: looks like the flapping is actually a JIT bug!
or not
sena_kun and rakudo is bumped too. \o/ 12:52
lizmat the flapping on the num stress test is my fault 12:55
nine Oh, the test is using random numbers
lizmat the flaps occur because it doesn't happen for all numbers
nine That's not good :* 12:56
Easiest way for our language implementation to pass that test is to ensure we create the right random numbers that pass... 12:57
lizmat :-) 13:00
nine For the failures it seems to either round in a different direction or add 2 more digits 13:01
lizmat the problem is an extra num --> Str cycle in the new val() logic 13:02
it appears the only way to create a right enough num from a string, is to actually use nqp::numify
rather than taking the parts and doing the calculation
samcv i can do a release if needed nine 13:03
nine samcv: please do :)
samcv okay. will do. give me a few hours 13:04
nine ++samcv
13:51 Altai-man_ joined 13:54 sena_kun left 15:52 sena_kun joined 15:54 Altai-man_ left 16:48 lucasb joined 17:16 domidumont joined 17:51 Altai-man_ joined 17:54 sena_kun left 18:23 domidumont left 19:52 sena_kun joined 19:54 Altai-man_ left
Geth MoarVM/refs/tags/2020.01: b49cb0fa76 | (Samantha McVey)++ | VERSION
Bump VERSION for release
samcv nine, released
lizmat tar file also pushed ? :-) 20:31
samcv lizmat, yep 20:40
21:51 Altai-man_ joined 21:54 sena_kun left
Geth MoarVM: b49cb0fa76 | (Samantha McVey)++ | VERSION
Bump VERSION for release
22:07 lucasb left
nine \o/ 22:09
samcv: many thanks!
23:52 sena_kun joined 23:54 Altai-man_ left