Welcome to the main channel on the development of MoarVM, a virtual machine for NQP and Rakudo (moarvm.org). This channel is being logged for historical purposes.
Set by lizmat on 24 May 2021.
Nicholas good *, #moarvm 06:28
japhb Good *, Nicholas 06:31
lizmat probably of interest here: github.com/rakudo/rakudo/issues/4741 09:23
and good *, #moarvm
Nicholas (gdb) p param 09:30
$1 = (MVMObject *) 0x1
This looks suspiciously similar to github.com/MoarVM/MoarVM/issues/1656 09:31
but I can't recreate it (yet) on a build on a machine where rr works
done so. 10:09
issues updated.
I don't think that I can take this further.
Nicholas what's the best way to "hit" MVM_coerce_s_i with an NQP testcase? 16:31
m: say 6 * "7" 16:35
camelia 42
Nicholas aha, that seems good enough. (Yes, that's not NQP...)
Nicholas m: use nqp; nqp::say(nqp::coerce_si("-")); 18:32
camelia Iteration past end of grapheme iterator
in block <unit> at <tmp> line 1
Nicholas that's not what we want, is it?
(I don't know how to show the same code running on the most recent release) 18:33
if it wasn't obvious - I've figured this out from code inspection. So I have an idea how to fix it. 18:34
MasterDuke committable6: 2021.12 use nqp; nqp::say(nqp::coerce_si("-")); 22:58
committable6 MasterDuke, ¦2021.12: «0␤»
MasterDuke whoops. didn't intend that 22:59
i was actually wondering how much of the strtoll functionality is specced, and if there was any further simplifications that could be done. guess that's one thing that isn't specced 23:01