github.com/moarvm/moarvm | IRC logs at colabti.org/irclogger/irclogger_logs/moarvm
Set by AlexDaniel on 12 June 2018.
nwc10 good *, #moarvm 09:42
All tests successful. 10:20
Files=145, Tests=13359, 63104 wallclock secs (17.15 usr 3.74 sys + 178776.99 cusr 506.76 csys = 179304.64 CPU)
that's NQP, with GC_DEBUG=3
so, most of a day
for the tests, and more than a day for the build.
what's with 10:21
+++ Compiling NQPP5QRegex.moarvm
something must be wrong in the dependencies, as that seems to get recompiled more often than it shoudl
(some hours later, my install might complete)
brrt \o 13:59
nwc10 o/ 14:00
timotimo oo 19:20
MasterDuke pp 19:21
timotimo qp 19:22
MasterDuke qr? 19:28
timotimo dwop
qwop
MasterDuke great game 19:29
bartolin I've just stumbled above this segfault: 20:42
m: use nqp; my %h; say nqp::existskey(%h, "Foo")
camelia (signal SEGV)
bartolin it seems to happen on line 1478 in src/6model/reprs/P6opaque.c (del is NULL) 20:43
there is similar code for other functions (at_key, etc.) 20:44
should I open an issue in MoarVM for this? 20:45
timotimo missing concreteness check, eh? 21:26
in the future i think we'll be using the dispatch mechanism so that moar 1) does the checking without us putting it everywhere and 2) can toss out the check if spesh can prove it's not needed 21:27
^- use the dispatcher mechanism for repr ops like {at,exists,bind}{key,pos} and friends 21:28
bartolin that sounds promising. so, it's not worth to look at this now? 21:32
timotimo possibly 21:33

Powered by Cro and the Raku® Programming Language.

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

Thank you!