|
00:45
Ven joined
|
|||
| timotimo | oh, the "attempted to unlock a mutex by a thread not holding it" or whatever was from when i added a new op and forgot ot recompile rakudo ... i think? | 00:49 | |
| or when i did the reverse; threw away the changes and forgot to recompile rakudo | |||
| Please report this as a bug (mail to [email@hidden.address] | 00:50 | ||
| and re-run with the --ll-exception command line option | |||
| to get more information about your error> | |||
| where is this coming from? o_O | |||
| Backtrace.pm, eh? huh. | |||
| Geth | MoarVM: MasterDuke17++ created pull request #578: Fix mp_get_int64 and mp_get_uint64 for realz |
01:02 | |
|
01:48
ilbot3 joined
02:00
leego joined
02:19
samcv joined
02:35
geekosaur joined
03:12
brrt joined
05:37
domidumont joined
05:44
domidumont joined
06:32
geekosaur joined
07:13
brrt joined
|
|||
| nine | blog.ffwll.ch/2017/04/review-howto.html | 07:33 | |
| yoleaux | 25 Mar 2017 07:50Z <samcv> nine: getting those failing tests again. maybe because i'm using libffi under moarvm and not dyncall? | ||
| 30 Mar 2017 03:59Z <AlexDaniel> nine: any ideas? irclog.perlgeek.de/perl6/2017-03-30#i_14348415 | |||
| samcv | that was a while ago. hm | 07:34 | |
| nine | Oops... Shouldn't I have gotten these on #perl6-dev, too? | ||
| samcv | probably | 07:35 | |
| anyway i have not 100% confirmed the abovee, but that's what i suspect the issue may be | 07:37 | ||
|
07:38
domidumont joined
|
|||
| nine | Well dyncall is certainly the better tested one. | 07:58 | |
|
08:00
spebern joined
08:12
spebern joined
|
|||
| brrt | good *, #moarvm | 08:24 | |
| ohai nine | |||
| lol 'giving less focus on top committer stats' | |||
| well, if you're open source community is so screwed that people regard it as a points-gathering exercise... | 08:25 | ||
| *your | |||
| samcv | hi brrt | 08:27 | |
| brrt | hi samcv | 08:28 | |
| samcv | that's a good article | 08:33 | |
| i need to invent one of those watches that counts your steps or something except it tracks your points | 08:34 | ||
| and it will make exciting noisese when you gain points | |||
| also soon we will have coverage reports with every commit i expect | 08:37 | ||
| since i've now learned how to get travis ci to make commits to repositories | |||
| and we will have a github pages page which will have the coverage infos | |||
| brrt | it's a decent article, but I can't help but grumble like an old fart about the… assumingness of it all | 08:38 | |
| i'm not sure how to say it correctly | |||
| samcv | i think I get what you mean | ||
| brrt | well, that's better than I could expect :-) | 08:39 | |
| that said | 08:40 | ||
| how is your experience as a new contributor for moar / rakudo? | |||
| i do wonder | |||
| samcv | good | 08:42 | |
| it makes me very happy | |||
| to be improving a language I love for all the people that use it | 08:53 | ||
| is very satisfying | |||
|
09:42
zakharyas joined
09:47
brrt joined
|
|||
| brrt | well, that's good to hear :-) | 09:48 | |
| nine | brrt: the article does mention that it is written with a kernel focus | ||
| brrt | oh, i didn't notice tha | 09:52 | |
| nine | brrt: s/better of re-engineering/better off re-engineering/ in your latest blog post | 09:54 | |
| And very interesting bit about setjmp/longjmp :) | 09:55 | ||
| brrt | i'll fix it, thanks :-) | 09:58 | |
| nine | brrt: hahaha, that's cool. Just reading your Register Allocator Update blog post and when I came to the paragraph about squeezing in new tiles into the tile array, I though "couldn't you just leave some space between the tiles?". And that seems to be exactly what you did :) | 10:05 | |
|
10:21
stmuk joined
|
|||
| brrt | yes, it is :-) | 10:26 | |
| samcv | night moarvm :) | 10:47 | |
|
11:20
stmuk_ joined
|
|||
| Geth | MoarVM: 6346f44c70 | (Daniel Green)++ | src/6model/reprs/P6bigint.c Fix mp_get_int64 and mp_get_uint64 for realz mp_get_long_long does in fact overflow/wrap around when the mp_int is too large. So bring back checking the number of bits in the mp_int before calling it. Also, remove the redundant check for whether the mp_int is 0, since the exact same check is done in mp_get_long_long. |
11:23 | |
| MoarVM: 4d8e140ccb | (Daniel Green)++ | src/6model/reprs/P6bigint.c Make bits const and explain signed max bits better |
|||
| MoarVM: 66dd8c932d | niner++ | src/6model/reprs/P6bigint.c Merge pull request #578 from MasterDuke17/fix_mp_get_int64_and_mp_get_uint64_for_realz Fix mp_get_int64 and mp_get_uint64 for realz |
|||
|
11:30
stmuk joined
13:34
brrt joined
14:00
brrt joined
14:04
spebern joined
14:26
stmuk_ joined
14:30
stmuk joined
14:40
stmuk_ joined
14:43
stmuk joined
14:50
Geth joined
14:51
Geth joined
14:52
Geth joined
14:53
stmuk_ joined
14:59
stmuk joined
15:02
stmuk_ joined
15:08
brrt joined
|
|||
| Geth | MoarVM: c9ab59c6ba | (Jonathan Worthington)++ | src/io/io.c Remove arbitrary and small length range check. This meant we could not slurp a file over 100MB in binary mode without getting an error. Removing this, such a file is very easily slurped. |
15:21 | |
|
15:22
brrt joined
16:35
ilmari joined
17:03
Geth joined
17:04
Geth joined
17:06
stmuk joined
17:18
domidumont joined
17:57
AlexDaniel joined
18:03
AlexDaniel joined
18:41
geekosaur joined
19:23
AlexDaniel joined
20:30
AlexDaniel joined
21:37
AlexDaniel joined
22:45
Geth joined
22:46
Geth joined
|
|||