|
github.com/moarvm/moarvm | IRC logs at colabti.org/irclogger/irclogger_logs/moarvm Set by AlexDaniel on 12 June 2018. |
|||
|
00:59
MasterDuke joined,
p6bannerbot sets mode: +v MasterDuke
01:00
MasterDuke left,
MasterDuke joined,
herbert.freenode.net sets mode: +v MasterDuke,
p6bannerbot sets mode: +v MasterDuke
01:32
nwc10 left
03:00
leont left
04:35
MasterDuke left
06:58
nwc10 joined
06:59
p6bannerbot sets mode: +v nwc10
07:11
domidumont joined
07:12
p6bannerbot sets mode: +v domidumont
08:07
leont joined
08:08
p6bannerbot sets mode: +v leont
08:23
leont left
08:24
zakharyas joined
08:25
p6bannerbot sets mode: +v zakharyas
10:00
statisfiable6 left
10:10
lizmat left
10:28
lizmat joined,
p6bannerbot sets mode: +v lizmat
10:52
zakharyas left
11:03
statisfiable6 joined
11:04
p6bannerbot sets mode: +v statisfiable6
11:06
leont joined
11:07
p6bannerbot sets mode: +v leont
11:31
domidumont left
11:39
statisfiable6 left
11:54
brrt joined,
p6bannerbot sets mode: +v brrt
|
|||
| brrt | \o | 11:55 | |
| nwc10 | o/ | 11:58 | |
|
12:14
brrt left
12:37
statisfiable6 joined,
p6bannerbot sets mode: +v statisfiable6
12:54
domidumont joined
12:55
p6bannerbot sets mode: +v domidumont
14:39
reportable6 left,
reportable6 joined
14:40
p6bannerbot sets mode: +v reportable6
15:00
dogbert2_ joined,
p6bannerbot sets mode: +v dogbert2_
15:25
zakharyas joined,
p6bannerbot sets mode: +v zakharyas
15:52
brrt joined
15:53
p6bannerbot sets mode: +v brrt
16:18
avar left,
avar joined,
avar left,
avar joined,
p6bannerbot sets mode: +v avar
16:19
p6bannerbot sets mode: +v avar
16:21
zakharyas left
16:29
robertle joined,
p6bannerbot sets mode: +v robertle
16:41
brrt left
|
|||
| Geth | MoarVM: 7ba02259bc | (Jonathan Worthington)++ | src/6model/reprs/MVMMultiCache.c Correct named argument type tuple index calc Thus fixing an out-of-bounds memory access. Resolves #996. |
16:50 | |
|
16:52
domidumont left
17:03
robertle left
17:06
domidumont joined,
p6bannerbot sets mode: +v domidumont
17:21
domidumont left
17:55
brrt joined
17:56
p6bannerbot sets mode: +v brrt
17:57
domidumont joined
|
|||
| brrt | \o | 17:57 | |
| so. I think I've figured out a fix for one of dogberts' issues | |||
| specifically the one that was related to trunc | |||
|
17:58
p6bannerbot sets mode: +v domidumont
18:04
brrt left
18:06
brrt joined
18:07
p6bannerbot sets mode: +v brrt
18:44
brrt left
|
|||
| dogbert11 | brrt, sounds great. Is it a difficult fix? | 18:49 | |
|
19:03
brrt joined
19:04
p6bannerbot sets mode: +v brrt
|
|||
| brrt | no, but it is a typical case of 'what is actually the desired behaviour here' | 19:06 | |
| if I have a 64 bit integer, and I truncate it to 16 bits, and then I reuse it as a 64 bit integer, do I want that to be a): signed b): unsigned | |||
| does trunc mean that the upper bits are 0, or sign-extended? | 19:07 | ||
| officially this shouldn't matter. but in practice it clearly does | 19:08 | ||
|
19:09
brrt left
19:10
brrt joined
19:11
p6bannerbot sets mode: +v brrt
19:26
squashable6 left
19:28
squashable6 joined
19:29
p6bannerbot sets mode: +v squashable6
19:31
emerson left
19:50
domidumont left
20:06
brrt left
20:07
brrt joined
20:08
p6bannerbot sets mode: +v brrt
|
|||
| Geth | MoarVM: 9468b7f750 | (Bart Wiegmans)++ | src/jit/core_templates.expr [JIT] Fix trunc to cast upwards This is a bit confusing altogether. My original theory of trunc was that we'd do a downcast, clearing the upper bits, storing only the lower bits, but treating the number as being smaller. But all sorts of downstream code didn't really agree. So now we do the opposite - zero-extend a smaller number into a bigger one (and store that instead). And that seems to fix a few bugs. |
20:09 | |
| dogbert11 | brrt++ | 20:33 | |
| lizmat | And yet another weekly hits the Net: p6weekly.wordpress.com/2018/11/19/...en-perl-6/ | ||
| brrt | lizmat++ | 20:52 | |
|
20:53
brrt left
21:17
Kaiepi left
21:18
Kaiepi joined
21:19
p6bannerbot sets mode: +v Kaiepi
22:38
TimToady_ joined
22:39
p6bannerbot sets mode: +v TimToady_,
TimToady left
|
|||