Perl 6 - the future is here, just unevenly distributed

IRC log for #moarvm, 2014-09-22

| Channels | #moarvm index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:30 colomon joined #moarvm
01:07 FROGGS__ joined #moarvm
01:18 colomon joined #moarvm
02:20 diakopter left #moarvm
03:09 colomon joined #moarvm
06:24 sergot o/
07:08 zakharyas joined #moarvm
07:12 brrt joined #moarvm
07:21 brrt timotimo: becauase mvmarray checks negative indices, array bounds, and a start index
07:26 brrt the only thing i'd think of is proving bounds on the index
07:31 FROGGS joined #moarvm
07:31 timotimo brrt: when we have an iterator, we know the indices cannot be negative
07:32 timotimo as they start at 0 and only ever get incremented
07:32 timotimo how's that? :)
07:33 brrt that would be decent i think :-)
07:49 brrt joined #moarvm
07:51 brrt but the special case  nature of it suggests special casing on iterators. hence, shift
07:52 timotimo er ... yeah, that'? what i was suggesting to do
07:52 timotimo well, i suggested to do it in spesh
08:09 brrt joined #moarvm
08:37 kjs_ joined #moarvm
09:21 kjs_ joined #moarvm
09:50 kjs_ joined #moarvm
10:50 kjs_ joined #moarvm
11:03 leont joined #moarvm
11:33 camelia joined #moarvm
11:33 oetiker joined #moarvm
11:34 LLamaRider joined #moarvm
12:35 kjs_ joined #moarvm
12:35 ggoebel2 joined #moarvm
12:46 kjs_ joined #moarvm
13:46 JimmyZ joined #moarvm
13:53 kjs_ joined #moarvm
14:36 cognome joined #moarvm
14:55 brrt joined #moarvm
15:06 cognome joined #moarvm
15:40 kjs_ joined #moarvm
15:56 leont joined #moarvm
16:15 FROGGS joined #moarvm
16:15 leont According to Reini "parrot was destroyed by a different set of poisonous people. Now without them progress can continue, even of most of devs choose to switch to the new vm moar"…
16:18 FROGGS hmmm, I only agree with parts of this sentence
16:19 FROGGS might be just because I'm a newbie still
17:15 FROGGS[mobile] joined #moarvm
17:31 nwc10 apparently, I've retired. This is news to me. I just thought that I was very busy with real life (TM)
17:32 nwc10 it's amazing what you learn from the Internet
17:47 [Coke] . ask diakopter - think I figured out the problem - can you up the swap space on host08 to match host07? (I can probably stumble my way through this if needed)
17:47 [Coke] (the problem == things exploding when we have two big java things going on at the same time)
17:48 [Coke] .ask diakopter - think I figured out the problem - can you up the swap space on host08 to match host07? (I can probably stumble my way through this if needed)
17:48 [Coke] bah, wrong window.
18:45 Ven joined #moarvm
18:54 kjs_ joined #moarvm
18:55 zakharyas joined #moarvm
19:14 brrt joined #moarvm
19:14 camelia joined #moarvm
19:17 kjs_ joined #moarvm
19:18 brrt leont: i don't think parrot was destroyed by poison though
19:33 japhb Reference link?
19:34 nwc10 thread here, I think http://www.nntp.perl.org/group/perl.perl5.porters/2014/09/msg220297.html
19:34 nwc10 (not that message
19:34 nwc10 please note that all the archives I can find online seem to be a bit behind
19:40 japhb Hrmm, yeah, I don't see the above quotes in that thread on www.nntp.perl.org, so I'll check back another time
19:41 nwc10 this is the only current archive I can find: https://groups.google.com/d/msg/perl.perl5.porters/-ASCEIPVgGU/Q8PS6pM2AakJ
19:47 japhb Oh ouch, banned?
19:49 lizmat yup
19:49 lizmat for 1 month
19:50 * japhb sighs at the whole oft-repeated cycle (not him in particular, I dunno, just ... among projects I have tracked at one time or another)
20:01 nine Great way to "solve" disagreements. On both sides.
20:29 timotimo i'm not sure i understand all of what brrt and jnthn discussed in #perl6 recently ...
20:38 Ven joined #moarvm
20:43 beppu joined #moarvm
21:01 brrt joined #moarvm
21:10 japhb timotimo: What have they been discussing?
21:17 timotimo optimization opportunities, changes to make MVMArray less dynamic/changable
21:21 brrt timotimo: meaning? :-) i'm not sure i understand all either, but i'd like to hear
21:21 brrt and discuss
21:23 brrt basically, the way i get it, both jnthn and i would like much cheaper arrays
21:23 brrt one reason for that would be that our string type could be implemented higher up
21:23 brrt i may be hugely misrepresenting stuff right now and if so i apologise :-)
21:30 timotimo ah
21:30 timotimo well, the QPMC stuff, which is "we have a starting index which we'll shuffle around back to 0 only when we need more space", was a really gigantic improvement to lots of use cases
21:32 kjs_ joined #moarvm
21:34 brrt qpmc?
21:35 timotimo er
21:36 timotimo QRPA
21:36 timotimo sorry
21:36 timotimo quantùm resizable power array
21:45 brrt uhm
21:45 brrt google doesn't know that acronym clearly
21:46 brrt but i guess that's what we have now?
21:47 [Coke] qast resizable pmc array.
21:47 timotimo well, not "qast" i think
21:47 [Coke] where pmc is "parrot magic cookie" or "native object in parrot"
21:47 [Coke] but rpa stuck in some cases.
21:48 [Coke] ah, whoops.
21:49 [Coke] ETOOMANYQs.
21:49 timotimo :)
21:49 [Coke] 25c0b53 in nqp says "quicker"
21:51 brrt oh i see
21:52 timotimo Q is also "comes after P" often, like in QAST as the successor to PAST
21:52 brrt but my suspicion was that resizable pmc array was sparse by default?
21:53 timotimo hm, i'm not sure when having null slots in there happened
21:54 [Coke] I don't remember parrot internals well enough to answer that.
21:55 japhb [Coke]: I'd consider that a good thing, at this point.  (And frankly I'm in the same situation.)
21:58 kjs_ joined #moarvm
22:02 FROGGS[mobile]2 joined #moarvm
22:05 * brrt afk
22:05 brrt left #moarvm
22:15 leont joined #moarvm
22:31 leont joined #moarvm

| Channels | #moarvm index | Today | | Search | Google Search | Plain-Text | summary