|
00:33
nativecallable6 joined,
coverable6 joined,
notable6 joined,
greppable6 joined,
evalable6 joined,
reportable6 joined,
benchable6 joined,
releasable6 joined,
bloatable6 joined,
squashable6 joined
00:34
statisfiable6 joined,
unicodable6 joined,
committable6 joined,
shareable6 joined,
bisectable6 joined,
quotable6 joined
|
|||
| samcv | brrt++ | 01:56 | |
| MasterDuke | samcv: do we use memchr for finding a single char in a string? | 01:59 | |
| samcv | MasterDuke: for 8bit strings, yes | 02:00 | |
| MasterDuke: err actually technically no. but we do use it through libc or glibc | |||
| let me check | |||
| MasterDuke | cool. i just saw something on HN about the new release of rust using it for some dramatic speedup | 02:01 | |
| samcv | MasterDuke: freebsd memmem uses memchr as needed | 02:02 | |
| and i'm 95% sure glibc does as well | |||
| MasterDuke | so we get it "for free"? nice | 02:03 | |
| samcv | yeah it uses the distro memmem. so windows will use window's memmem, even though freebsd's memmem | 02:04 | |
| no clue about difference in speed between platforms of memmem | |||
| *memchr | 02:07 | ||
|
02:55
ilbot3 joined
03:45
greppable6 joined
04:27
evalable6 joined
04:53
AlexDaniel joined
05:33
harrow joined
07:24
brrt joined
|
|||
| brrt | . | 07:24 | |
| yoleaux | 15 Feb 2018 22:37Z <MasterDuke> brrt: does github.com/MoarVM/MoarVM/commit/c7774e37d2 fix the getlexstatic_o template also? | ||
| brrt | .tell MasterDuke to be honest I'd expect not | 07:25 | |
| yoleaux | brrt: I'll pass your message to MasterDuke. | ||
| brrt | maybe the box_i template | ||
| symptoms of the getlexstatic_o call were quite different | |||
|
07:48
Ven`` joined
08:00
bisectable6 joined
|
|||
| nine | brrt++ # benchmarking is fun again | 08:13 | |
| brrt | :-) | 08:21 | |
| nine: do you recall what broke from the box_i template? | 08:31 | ||
| nine | sorry, don't remember anything specific | 08:32 | |
| brrt | hmm | 08:33 | |
| i think maybe nwc10 was complaining about that at some point | 08:34 | ||
| nwc10 | I didn't mean to be "complaining" - "reporting that it's broken, and that the cause seems to be X" | 08:41 | |
| and, IIRC, yes, something broke, forgot what | |||
| is there an easy way to re-enable it and hence test? | |||
| "be X, and I don't have any good idea how to fix it" | |||
| brrt | hehe sorry, i used the wrong phrase | 08:43 | |
| was *reporting* very usefully, that stuff was broken | |||
| unfortunately I couldn't figure out what was | |||
| yeah, sure, i can re-enable that template | |||
|
08:44
Ven`` joined
|
|||
| nwc10 | brrt: was more - "Is there a commit I can revert locally and then test?" | 08:47 | |
| brrt | oh yes | ||
| dd72e1f1ecc16ce75e78db3c813bb6cf43b574ba | 08:48 | ||
| this one | |||
|
08:49
Ven`` joined
|
|||
| nwc10 | OK, started, but I have other things to do so might not be checking back for a while | 08:50 | |
| brrt | ok, thanks | 08:52 | |
|
09:21
zakharyas joined
|
|||
| brrt | samcv: i was looking at the unicode.c file | 09:28 | |
| i'm wondering if we can replace the concatenation with a #include | |||
|
10:40
AlexDaniel joined
10:42
Ven`` joined
|
|||
| dogbert2 | brrt++ # bug fix | 11:04 | |
|
11:06
evalable6 joined
11:29
brrt joined
11:39
zakharyas joined
11:57
zakharyas joined
12:24
Ven`` joined
12:26
tbrowder joined
13:35
zakharyas joined
14:11
zakharyas joined
14:24
Ven`` joined
14:40
Ven`` joined
15:21
Ven`` joined
15:27
zakharyas joined
15:46
AlexDaniel joined
15:52
bart__ joined
16:33
nwc10 joined
16:41
Ven`` joined
16:43
zakharyas joined
17:27
Ven`` joined
18:09
Ven`` joined
18:32
Ven`` joined
|
|||
| samcv | .tell brrt that sounds like a good idea | 19:20 | |
| yoleaux | samcv: I'll pass your message to brrt. | ||
|
19:21
releasable6 joined
20:01
Ven`` joined
20:57
MasterDuke joined
21:14
bisectable6 joined
21:26
Kaiepi joined
21:56
dogbert17 joined
22:08
greppable6 joined
22:51
japhb joined
23:16
japhb joined
|
|||