| japhb | jnthn: I had an excuse for the crickets. I was teaching a promising young programmer a short history of interpreters and JITs for his future study. | 00:02 | |
| Also: \o/ for the merge. :-) | |||
| jnthn | japhb: Cool :) | 00:07 | |
|
00:45
rurban joined
00:55
colomon joined
01:15
colomon joined
|
|||
| timotimo | yay merging moar-conc | 01:20 | |
| jnthn: is now-ish a good point in time to consider implementing a bit of the async i/o stuff we already have on the jvm on moar? | 01:22 | ||
| (that is: if i can figure out how to make it work at all :P) | 01:25 | ||
|
01:26
rurban joined
|
|||
| timotimo | it's especially clever to rebuild nqp and *then* git pull the important changes :P | 01:35 | |
|
02:18
colomon joined
06:42
FROGGS joined
07:03
ggoebel11118 joined
08:06
woolfy joined
08:19
woolfy left
|
|||
| dalek | arVM: 78f9ab0 | (Tobias Leich)++ | src/6model/reprs/P6opaque. (2 files): make helper public that gives us a replaced MVMP6opaqueBody We need this helper in rakudo too, for example when we need the replaced body of a list after something get mixed into it, which results in a growth (and therefor a replacement) of the body. |
09:11 | |
|
09:37
colomon joined
10:02
JimmyZ joined
|
|||
| dalek | arVM: ba1c829 | jimmy++ | src/6model/reprs/P6opaque.c: Minor optimization This helps improve startup time and reduces CORE.setting about 9KB |
12:23 | |
| JimmyZ | This patch needs to rebuild nqp/radkuo ... | 12:25 | |
| *rakudo | |||
| dalek | arVM: eb2ba07 | jimmy++ | src/6model/reprs/P6opaque. (2 files): s/MVMP6opaque_real_data/MVM_p6opaque_real_data/g |
12:33 | |
| FROGGS | JimmyZ: why do you rename it to MVM_p6opaque* when everything starts with MVMP6opaque? | 12:39 | |
| dalek | arVM: 0b08db7 | jimmy++ | src/6model/serialization.c: Small cleanup to serialization [read|write]_int[16|32]_func will be removed in the near future once nqp-m gets a new bootstrap |
12:40 | |
| JimmyZ | FROGGS: Global function always starts with MVM_* | 12:42 | |
| FROGGS | hmmm, okay | ||
| JimmyZ | FROGGS: I think it needs s/MVMP6opaque_initialize/MVMP_6opaque_initialize/ too | 12:43 | |
| jnthn: ^^ how do you think it? | |||
| jnthn | JimmyZ: Well, apart from the REPR initialize ones, which don't. | ||
| But yeah, the rest do | |||
| JimmyZ | OK :) | 12:44 | |
| jnthn | MVM_iter_blah, for example | ||
| I'd leave the initialize ones, tbh. | |||
| JimmyZ | fine to me | ||
| jnthn | But yeah, MVM_p6opaque_real_data fits better. | ||
|
14:37
FROGGS[mobile] joined
15:42
zakharyas joined
16:18
zakharyas joined
|
|||
| timotimo | JimmyZ: did you measure startup time speedup for that optimization patch? | 18:15 | |
|
18:55
lizmat joined
19:04
zakharyas joined
19:34
woolfy joined
19:36
colomon joined
21:24
masak joined
22:14
sergot joined
|
|||
| jnthn has been making progress on his bytecode specialization preparations for Moar. | 23:26 | ||
| Expect I'll have something to push/show within the next week, perhaps within a couple of days. | |||
| FROGGS | I guess not only brrt is eager to see that :o) | 23:27 | |
|
23:41
colomon joined
|
|||
| timotimo | ooooooh | 23:44 | |