github.com/moarvm/moarvm | IRC logs at colabti.org/irclogger/irclogger_logs/moarvm
Set by AlexDaniel on 12 June 2018.
lizmat and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2018/10/22/...g-contest/ 09:27
moritz lizmat++ 09:35
brrt \o 18:28
timotimo o/ 18:47
brrt ohai timotiom 19:04
timotimo
timotimo omit timo
brrt lol
I'm kind of out of ideas to hack on ... I'm a bit stuck with the nativeinvoke expr jit 19:05
timotimo i have an idea that's perhaps too "out of reach" for you to try?
the idea would be to have a "if we know the STable, we can generate better jit trees for a given op" 19:06
timotimo it's not my idea, nor is it new 19:07
but it'd be cool to have it some time soon-ish
timotimo having that would allow us to get even bettar than the lego jit currently, especially for the nativeref ops that still have a switch-case for operand size or something after the devirtualization 19:07
Zoffix brrt: do you have the skills to fix R#2400 ? That could be an idea :) 19:16
synopsebot R#2400 [open]: github.com/rakudo/rakudo/issues/2400 [fudged tests committed][regression][⚠ blocker ⚠] Module Crane is failing tests on HEAD
Zoffix brrt: basically making failing test pass: github.com/perl6/roast/commit/6c01...78d9b2b292
c: HEAD use lib </home/bisectable/git/whateverable/sandbox>; use Foo; dd foo
committable6 Zoffix, ¦HEAD(5c457d9): «Type check failed for return value; expected Hash[Any:D,List:D] but got $(my Any:D %{List:D})␤ in sub foo at /home/bisectable/git/whateverable/sandbox/Foo.pm6 (Foo) line 2␤ in block <unit> at /tmp/og9pgo3xy2 line 1␤␤ «exit code = 1»»
Zoffix And a very similar bug was fixed recently. Maybe the fix for that gives a hint: github.com/rakudo/rakudo/issues/23...-431033154 19:17
brrt timotimo: yeah, that is a good idea. and no, it's not new :-)
and it's on the list, but it's a biggish project
well, doesn't have to be a giant project
timotimo yeah, i'm also afraid it's big 19:18
brrt Zoffix: I'll have a look
<brag> I'm decently good in breaking big things into small ones </brag>
But I'm not quite sure how I'd tackle it 19:19
Zoffix: I'm sorry, but I don't even have an idea where to start 19:25
AlexDaniel brrt: I believe in you ♥ :) 19:26
brrt hehe
with some hope I'll be able to make more time for MoarVM soon :-) 19:46
Zoffix Thanks for trying :) 19:56
It's kinda scary tho that jnthn being away for a few days kinda leaves us stranded. 19:57
Zoffix makes a mental note to git gud at MoarVM in 2019 19:58
brrt this does kind of seem a rakudo issue to me 20:01
but, that's just me
and I know fairly little about the rakudo compiler 20:02
I mean, for sure, give us segfaults, we can handle those :-D
AlexDaniel samcv: There are still two blockers, but sometimes blockers disappear rather quickly. Please start working on the changelog, if you haven't already. 👍 20:33
samcv ok :) 20:53