timotimo a non-optimal design is fixable, too 00:42
are these objects actually supposed to be living this long? 00:52
because that's a metric fuckton of MVMCode objects
jnthn Well, I want to look into that too
timotimo maybe that'll turn out to be enough already?
jnthn I mean, the reason we get MVMCode objects is easy: you get one every time you take a closure.
That's how Moar represents a closure.
But the reason we have that many of them living for so long is worth looking into. 00:53
Thankfully, that's not overly hard to investigate: you can look inside the MVMCode and get the static frame, which gives you a name. And if the closure is anonymous you can look outwards until you find a name. :) 00:54
timotimo i kind of feel like i don't have the energy necessary to write the code to do that right now 01:12
i want a sleep soon
jnthn :)
Same :)
timotimo i have a weird mental thingie where not seeing perl6 be worked on much for longer than a day makes me a bit demotivated and slightly sad :\ 01:13
the last few weeks were not so terribly good in that regard :\ 01:17
jnthn Yeah, $dayjob threw me rather more teaching to do that I really wanted, and that combined with the travel was exhausting. 01:18
*than
I'd hoped to do some bits today but...really didn't have the concentration, unsurprisingly... 01:19
timotimo i'm not blaming you :) 01:20
since you do a whole alot per week most weeks
jnthn Yeah. In 2015, I'll be making very sure Perl 6 and MoarVM get a lot of me. :) 01:22
In no small part because it's what I'd really like to be doing. :) 01:23
timotimo i like the sound of that 01:28
dalek arVM: 7c6dd2e | (Jimmy Zhuo)++ | src/gc/worklist.h:
removed an unused macro
 02:22

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!