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.
nine I'm pretty sure now that the issue is indeed, that we take one deopting op (dispatch_o) and replace it with multiple guards. Only the first such guard re-uses the deopt annotation (thus retaining the deopt idx). The others will get clones of the annotation with new deopt_idxs. 08:06
nine We won't find deopt usages for these new deopt_idxs. 08:06
Aren't MVM_SPESH_ANN_DEOPT_SYNTH annotations the mechanism intended for precicely such situations? 08:07
timo i mean you would assume, right? 15:57
nine Kinda yes :) 19:03
MasterDuke why do they get new deopt_idxs? 19:05
lizmat and yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2021/08/09/2021-...tcon-done/ 19:22
timo already using Comma already, lizmat? :) 19:48
lizmat no, as it still chokes on handling the core setting afaik
but am now convinced I should use it while developing my modules
but deeply ingrained muscle memory is hard to change :-) 19:49
MasterDuke yeah, i really would love for it to handle the core setting. i don't develop modules much, and vim is still what i instinctively reach for first for small scripts 19:52
iirc it complains about something in the t/ directory. i tried excluding that directory in a *.iml file, but i couldn't get it working 19:53
timo liz, i was pointing out a typo in your post :D 19:54
its all right that you dont use comma for core rakudo wor 19:55
lizmat timo: ah, that's more a grammo :-) 19:56
fixed, timo++
timo the typo system is not strong enough to detect grammar mistakes
lizmat afk for a bit& 19:57

Powered by Cro and the Raku® Programming Language.

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

Thank you!