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 09:19
jnthnwrthngtn o/ 09:39
Nicholas \o 09:42
nine \o/ 09:53
[Coke] getting nativecall failures on windows build. 12:10
[Coke] as previously, fails with prove, passes if I run a singular test. 12:11
jnthnwrthngtn Haven't had chance to read it yet but users.cecs.anu.edu.au/~steveb/pubs...s-2022.pdf may be of interst here 12:50
MasterDuke some extra context here twitter.com/stevemblackburn/status...7621657600 15:53
japhb I have mixed feelings about that paper. On the one hand, the measurements they did do are careful, and I like that they have tables and graphs with confidence intervals. On the other hand, the reasoning in a couple places is handwavey (and wrong I think, in at least one case), and there's no small amount of self-hype embedded. 21:01
Gripping hand for me though is that the whole bit about measuring actual latency did a really poor job of talking about latency jitter. (I mean yes, you could kindof pull that out of the shape of figure 3, but otherwise it's not well covered.) 21:03