github.com/moarvm/moarvm | IRC logs at irclog.perlgeek.de/moarvm/today
Set by moderator on 2 May 2018.
00:16 quotable6 joined, nativecallable6 joined, coverable6 joined, evalable6 joined, bloatable6 joined 01:03 Kaypie joined 01:58 ilbot3 joined
moderator github.com/moarvm/moarvm | IRC logs at irclog.perlgeek.de/moarvm/today
04:23 unicodable6 joined 05:54 domidumont joined 05:57 domidumont joined 06:03 AlexDaniel joined 06:05 domidumont joined 07:10 robertle joined 07:16 ZofBot joined, huggable joined, buggable joined 07:25 AlexDaniel joined 07:56 zakharyas joined
Geth MoarVM: df91ba76d0 | (Bart Wiegmans)++ | src/spesh/dump.c
[Spesh] Don't leek CUUID/name

ASAN would complain bitterly about this
08:05
08:17 brrt joined
brrt good * 08:17
i have a new problem for y'all
a mysterious JIT fail that only fails when MVM_SPESH_BLOCKING is false 08:18
AlexDaniel squashable6: next 08:25
squashable6 AlexDaniel, ⚠🍕 Next SQUASHathon in 1 day and ≈1 hour (2018-05-05 UTC-12⌁UTC+14). See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
09:08 Ven`` joined 09:09 Ven`` joined
brrt hmm, it maybe, hopefully, entirely my fault 10:01
10:52 brrt joined 12:38 brrt joined 12:47 undersightable6 joined
AlexDaniel samcv: is updating en.wikipedia.org/wiki/MoarVM part of the release process? 12:51
if not, maybe it should be
“Stable release 2017.02” 12:52
that's not even 2018
jnthn Is this scripted as part of the Rakudo release process? If so, can we just nab the code?
Also, "stable release" doesn't exist. We just have (monthly) releases. 12:53
AlexDaniel there's no code, I did attempt to automate it but never finished it
NeuralAnomaly also doesn't have it I think 12:54
brrt wow, that page is out of date
nwc10 xkcd.com/386/ 12:55
jnthn :)
I wonder if we can be considered not notable enough, and thus problem solved? :P 12:56
AlexDaniel I'm surprised that it's not deleted yet… 12:58
nwc10 but then how do you know if a project is "special biologist word for stable" or alive, if there is no Wikipedia page observing it? 12:59
AlexDaniel Categories: Stub-Class Computing articles | Low-importance Computing articles
:)
nwc10 where "alive" is "under active development" but I coudln't see how to allude to dead/alive with both terms substituted
13:00 zakharyas joined
nwc10 en.wikipedia.org/wiki/MoarVM 13:01
13:01 bisectable6 joined
nwc10 gets a " may not meet Wikipedia's notability guidelines for products and services" whist en.wikipedia.org/wiki/Parrot_virtual_machine does not 13:01
lizmat has at least updated the release info 13:09
13:10 bisectable6 joined 13:11 zakharyas joined
nine "However, Rakudo on the Java Virtual Machine is still faster for longer running programs." [Citation needed!] 13:11
13:13 committable6 joined
lizmat hehe once it runs, I guess that's probably still true for some value of"longer" :-) 13:14
13:15 committable6 joined 13:22 committable6 joined 13:26 undersightable6 joined 13:31 undersightable6 joined 14:11 zakharyas joined 14:17 zakharyas1 joined
brrt implmeenting getlexcaller in the jit blows up :-( 14:59
jnthn :( 15:03
brrt yeah, and its not quite obvious why 15:04
jnthn hopes his SEGV in the spesh plugin mechanism will turn out to be more obvious 15:08
brrt oh, it is more obvious than i thought :-) 15:12
i should've expected a NULL and returned a VMNull if so 15:13
hang on, res->o is a deref, innit? 15:15
jnthn yes
brrt good :-)
jnthn Well yeah, mine was silly. Used unsigned ints where shoulda used signed :P
brrt at what point, i sometimes wonder, do we actually start blaming the programming language 15:16
Geth MoarVM/pluggable-spesh: 6cb9624a5e | (Jonathan Worthington)++ | src/spesh/plugin.c
Use unsigned ints in binary search
15:19
brrt jnthn: docs.microsoft.com/en-us/cpp/c-run...ce/bsearch 15:22
or 'man bsearch'
Geth MoarVM: 1ef4aaf736 | (Bart Wiegmans)++ | tools/jit-bisect.pl
[jit-bisect.pl] spesh bisect is 1-based

We'd request a log for a frame just one after the one that was supposedly broken.
15:32
MoarVM/pluggable-spesh: 0f52872be2 | (Jonathan Worthington)++ | 6 files
Aggregate spesh plugin resolution statistics
15:53
jnthn Yay, now - at last - nearly at the point where the magic happens. :) 15:54
15:59 robertle joined
samcv AlexDaniel`: ah that sounds like a good idea 16:16
so we need some pages to cite so moarvm becomes notable 16:17
also However, Rakudo on the Java Virtual Machine is still faster for longer running programs. << i want this line removed
16:21 domidumont joined
lizmat samcv: done 16:23
samcv also since we are developers. we could always write a blog post. and then can site it and it's a notable citation because a developer of the project wrote it :) 16:25
though that won't do for proving that moarvm itself is notable. just that what we say can be true about the project
timotimo i don't think that's how that works 16:27
samcv really? 16:28
timotimo i think notability requires "other people" to write about you?
samcv hmm
yeah that's what i mean
timotimo well, core devs aren't "other people" :)
samcv i mean developers writing something could be ok for citing some facts about moarvm
but we'd need some outside article to show that moarvm is notable
timotimo what we need is someone to interview one of us :P
unless that is me, in which case there should be two people being interviewed 16:29
because i tend to get stuff wrong :P
(like how wikipedia works)
lizmat news.ycombinator.com/item?id=7806723 16:30
something that comes up for me in a google search 16:31
Geth MoarVM/pluggable-spesh: 4138b806c4 | (Jonathan Worthington)++ | src/spesh/facts.c
Analyze spesh plugin guard indices for stability

Currently we don't do anything with that determination, but this is what will allow us to carry on and rewrite the instruction and add guards.
16:37
lizmat www.josetteorama.com/all-about-perl...rt-1-of-3/ 16:38
jnthn dinner & 16:39
16:41 Ven`` joined
samcv lizmat: i added a citation to that article now 17:18
18:27 AlexDaniel joined
Geth MoarVM/make_builds_reproducible_again: 6899b0a172 | (Stefan Seifert)++ | 2 files
Serialize hashes sorted for more reproducible builds
20:05
20:08 Kaiepi joined 20:12 Kaiepi joined 20:49 brrt joined
brrt good * 20:55
Geth MoarVM/jit-moar-ops: 6 commits pushed by (Bart Wiegmans)++ 21:26
lizmat whee!
brrt i can run make test only after running make realclean on rakudo. 21:29
hmm
lizmat calls it a night 21:31
21:47 MasterDuke joined
timotimo yay, MVM_JIT_SPESH_SLOT_VALUE! 22:11
22:26 Kaiepi joined 22:55 Kaiepi joined