|
github.com/moarvm/moarvm | IRC logs at colabti.org/irclogger/irclogger_logs/moarvm Set by AlexDaniel on 12 June 2018. |
|||
|
00:33
lucasb left
01:27
lizmat_ joined
01:30
lizmat left
02:19
lizmat joined
02:22
lizmat_ left,
Kaiepi left
03:27
AlexDani` joined
03:29
AlexDaniel left
04:20
Kaiepi joined
04:26
lizmat left
05:23
MasterDuke left
06:11
Kaiepi left
06:12
Kaiepi joined
06:42
AlexDani` is now known as AlexDaniel
06:53
hoelzro_ joined
06:57
hoelzro left,
synopsebot left
07:57
zakharyas joined
08:41
lizmat joined
08:53
lizmat left
09:15
zakharyas left
12:05
lucasb joined
12:32
lucasb left,
lucasb joined
12:44
AlexDaniel left
13:33
AlexDaniel joined
13:38
andrzejku joined
|
|||
| andrzejku | hello | 13:38 | |
| I back with the same question 'load_setting' on a null object | |||
| how can I inspect that bug? | |||
| github.com/perl6/nqp/blob/50bb7de4...Loader.nqp | 13:42 | ||
| timotimo | what did you do to provoke the bug? :) | 13:54 | |
| that's the implementation of load_setting, but it isn't called, because the ModuleLoader object (or whatever else) is undefined at the place it's trying to call that module | 13:55 | ||
| have you tried --ll-exception yet to get a fuller backtrace? | |||
| andrzejku | timotimo, wait I will try | 13:58 | |
| I build my own package and after installing I get that error | 14:00 | ||
| Unhandled exception: Cannot call method 'load_setting' on a null object | |||
| at <unknown>:1 (/usr/lib64/perl6/runtime/ModuleLoader.moarvm:<dependencies+deserialize>) | |||
| from <unknown>:1 (/usr/lib64/perl6/runtime/perl6.moarvm:<dependencies+deserialize>) | |||
| got the newest version 2019.03 | 14:02 | ||
| timotimo | have you tried nuking the install directory and reinstalling? | 14:06 | |
| andrzejku | nuking? | 14:10 | |
| I have one idea | 14:11 | ||
| wait :D | |||
| as I understand those nqp files they are generated to kind of binary moarvm file which is interpreted by vm | 14:13 | ||
| and they are loaded in sequence somewhere | |||
| that's the base of language | 14:14 | ||
| timotimo, xD | 14:15 | ||
| timotimo, > say "hello, world" | 14:16 | ||
| hello, world | |||
| it is working | |||
|
16:33
domidumont joined
16:48
andrzejku left
16:51
MasterDuke joined,
MasterDuke left,
MasterDuke joined
17:05
andrzejku joined
17:42
andrzejku left
17:46
andrzejku joined
18:31
AlexDaniel left
18:37
andrzejku left
18:41
lizmat joined
19:53
domidumont left
20:03
lucasb left
20:14
AlexDaniel joined
20:19
AlexDaniel left
21:09
AlexDaniel joined
21:22
AlexDaniel left
|
|||
| timotimo | MasterDuke: would you like to look at the heap snapshot format that i came up with by looking at one of the files and trying to figure it out? %) | 21:35 | |
| actually it's currently not working right i think? | 21:37 | ||
| MasterDuke | is there some sort of specific feedback you're looking for? | ||
| i haven't used the heap profiler recently, you think it's broken right now? | 21:38 | ||
| timotimo | it is not | 21:39 | |
| the files are just way too big | |||
| and the format reading code is a bit too complex for my tastes | 21:40 | ||
| i was silently dropping some exceptions, that caused the output to not be quite correct | 22:05 | ||
| MasterDuke: hack.p6c.org/~timo/bloop.mvmheap | 22:10 | ||
| check it out | |||
| the main concern so far has been the framing data format | 22:13 | ||
| total compressed size 7123.40332 KiB | 22:16 | ||
| total uncompressed size 123252.400391 KiB | |||
| that seems rather worth a lot | |||
| i think it's still wrong somewhere | 22:23 | ||
| only the new format, of course | |||
| i'll tweak the offsets a bit ... | 22:36 | ||
| MasterDuke | current moarperf can read it? | 22:38 | |
| timotimo | not at all | 22:39 | |
| MasterDuke | ah. current heap analyzer? | ||
| timotimo | nope | 22:40 | |
| it's not even written by moar itself :D | |||
|
22:41
Kaiepi left
|
|||
| MasterDuke | ah, we have some artisinal, bespoke heap snapshots then... | 22:41 | |
| timotimo | at the moment moarvm is hacked to spit out all the values to gzipped files in /tmp | 22:42 | |
| and my tool takes them and spits out the new format | |||
| the next step will be to implement the reading of the format and writing the "one gzipped file per kind of data" to /tmp in perl6 so i can compare if it's correct | 22:43 | ||
| and then the writing part gets ported to C | |||
| did you have a look at it in a hex editor? | 22:44 | ||
| MasterDuke | it's opened in one, but my daughter wants to watch videos of dogs... | 22:45 | |
| timotimo | that is very important | 22:46 | |
| i've re-uploaded the file, it's a bit more sensible now | 22:55 | ||
|
23:26
Kaiepi joined
|
|||