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.
timo it's possible it didn't actually turn the thing on. there's a bunch of commented-out fprintf's, the one in moar.c should tell you if it's active or not 01:46
MasterDuke hm. is anyone else getting lots of fails in `make m-test`? 12:32
not the fails i'm talking about, but i also get these messages: 12:36
t/12-rakuast/xx-fixed-in-rakuast.rakutest                     (Wstat: 0 Tests: 104 Failed: 0)
  TODO passed:   8, 40, 104
lizmat yeah.. apparently some things got fixed in legacy as well 12:37
apart from that one, it's clean for me **after** a fresh recompile 12:40
having run with RAKUDO_RAKUAST=1 makes t/04-nativecall/24-cpp_param_letter.t fail for me
with a compilation error: Function 'NativeCall::Types::void' needs parens to avoid gobbling block (or perhaps it's a class that's not declared or available in this scope?) 12:41
MasterDuke timo: it looks like i'm pretty consistently getting a fail on your branch, without setting anything at all in the environment
t/06-telemetry/01-basic.t                                     (Wstat: 256 (exited 1) Tests: 42 Failed: 1)
  Failed test:  42
  Non-zero exit status: 1
but it passes with `MVM_NO_NURSERY_RANGE=1` 12:43
i also get intermittent: 12:50
t/02-rakudo/15-gh_1202.t                                      (Wstat: 256 (exited 1) Tests: 2 Failed: 1)
  Failed test:  1
  Non-zero exit status: 1
timo thanks i will look at it later today 13:25
if you're interested in seeing if the nursery range thing actually makes a difference on a "lower level" than "time spent", look at the output of "perf stat", maybe with -d or -dd for extra details, and look at the "page faults" number. also perf stat has -r5 which would run the command 5 times and give you average and std dev which is cool 13:26