|
00:14
lizmat joined
01:22
FROGGS_ joined
01:57
klaas-janstol joined
03:10
_sri joined
04:33
woolfy joined
06:06
FROGGS[mobile] joined
06:38
Ven joined
08:18
zakharyas joined
09:11
Ven joined
09:31
Ven joined
09:46
Ven joined
10:01
lizmat joined
10:35
mj41 joined
10:44
Ven joined
10:54
lizmat_ joined
11:15
lizmat_ joined
11:31
lizmat_ joined
11:36
lizmat joined
11:45
tgt joined
11:47
mj41 joined
11:51
lizmat_ joined
11:54
lizmat__ joined
12:11
lizmat_ joined
12:12
lizmat__ joined
12:33
brrt joined
|
|||
| brrt | \o | 12:33 | |
| sleep & | 12:37 | ||
|
12:59
Ven joined
13:05
__sri joined
13:09
klaas-janstol joined
13:49
Ven joined
13:56
tgt joined
14:38
klaas-janstol joined
15:45
_sri joined
16:12
brrt joined
16:29
woolfy joined
16:34
lizmat joined
|
|||
| timotimo | oh hey brrt :) | 18:35 | |
| catching up on sleep after coming back from bulgaria? | |||
|
18:37
zakharyas joined
|
|||
| brrt | aye | 18:59 | |
| timotimo | ah you're back online :) | 19:00 | |
| brrt | yes :-) | ||
| uhm, you can haz segfault iirc? | |||
| timotimo | that's right | ||
| japhb | [Coke]: I know you said it days ago, but I still smile when I see your 'mentees: the frosh maker.' pun, having recently said goodbye to our summer mentee. | 19:01 | |
| timotimo | it happens about 11 minutes into running the spesh_diff script on the spesh log for the core setting :( | ||
| brrt | oooh... that.. blows | 19:02 | |
| timotimo | yup | ||
| but it was while i was trying to get a profile done | 19:03 | ||
| so it *could* have been due to that | |||
| brrt | hmm | ||
| timotimo | haven't tested it without the --profile, because i only ran the spesh_diff to profile it in the first place :) | ||
| did you get a chance to further your work on param_* stuff? | |||
| brrt | no, i had to leave home and hacking wednesday afternon aleady | 19:04 | |
| let me see if i can get that branch to crash earlier | 19:06 | ||
| timotimo | that's fine :) | ||
| brrt | it's possible we have a gc missassignment somewhere | 19:12 | |
| if the value you have is garbage | |||
| breaks quite a few spectests, too | 19:22 | ||
| timotimo | "that" branch refers to what exactly? | 19:24 | |
| brrt | the jit_getcodeobj branch | 19:25 | |
| timotimo | oh, that's mine isn't it? | 19:28 | |
| i couldn't get anyone else to reproduce failures with it :( | |||
| (except the failures that also happen without the branch) | 19:31 | ||
| (those also happen with the branch, of course) | |||
|
19:32
Ven joined
|
|||
| brrt | hmmm | 19:33 | |
| well, i can reproduce failures fortunately | |||
| :-) | |||
| albeit only in spectests | |||
| and if it compuer=dependent and it causes segfaults because of corrupted objects then i seriously suspect gc corruption | 19:34 | ||
| although that is still just suspicion :-) | |||
| timotimo | perlpunks.de/paste/show/53f4ba0e.5b59.14 ← so | ||
| brrt | i have more i think | 19:35 | |
| but i rebased it on top of master | |||
| timotimo | hm, ok | ||
| brrt | oh i do have a little bit of a nitpick, i'm going to fix that right now | 19:38 | |
| dalek | arVM: 74883b4 | (Bart Wiegmans)++ | src/jit/emit_x64.dasc: Use nz rather than ne when testing for NULL They are aliased to the same flag, so this is a nitpick; however nz implies that we want to test for non-zeroness and ne implies that we want to test for equality. |
19:43 | |
| brrt | tl;dr - use nz when testing for zero, ne for equality, it makes your intentions much clearer | 19:44 | |
| btw timotimo++ for fixing both sp_boolify_iter_arr and sp_boolify_iter_hash | |||
| brrt is afk | 19:46 | ||
|
19:46
brrt left
|
|||
| timotimo | i fixed them? | 19:52 | |
| didn't you fix them? | |||
| FROGGS | who cares when it works :o) | 19:55 | |
| timotimo | .o( ist doch egal, solang der sound gut ist! ) | 19:57 | |
| FROGGS | that covers it pretty much | 19:59 | |
|
20:00
klaas-janstol joined
|
|||
| timotimo | if you find sillyness acceptable, you can listen to the K.I.T.T. Cut, which is the source for that quote | 20:11 | |
| www.youtube.com/watch?v=CP_i5C4bByA | 20:12 | ||
| FROGGS | omg *g* | 20:20 | |
|
20:22
colomon joined
|
|||
| timotimo | battery is about to shut down | 20:41 | |
|
20:44
Ven joined
|
|||
| tgt | m: say do for 1..50 { default { $_ } } | 20:52 | |
| camelia | rakudo-moar c793d8: OUTPUT«1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17» | ||
| tgt | When using default it only prints 1 to 17 | 20:53 | |
| It works correctly with MVM_SPESH_INLINE_DISABLE=1 | |||
| when * works correctly | |||
| (Opened a bug: github.com/MoarVM/MoarVM/issues/127) | 21:03 | ||
| timotimo | i ... don't even know what default { ... } is for | 22:12 | |
| oh, that's for given/when, duh | |||
|
22:15
kjs_ joined
|
|||