Perl 6 language and compiler development | Logs at colabti.org/irclogger/irclogger_log/perl6-dev | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
Set by Zoffix on 27 July 2018.
MasterDuke i'm getting a lot of one-off failures in spectests. i can never get the tests to fail again when run in a loop. i suspect some sort of moarvm problem that's randomly manifesting itself 03:20
anyone else seeing anything similar?
fwiw, i've seen failures in t/spec/S06-traits/misc.t, t/spec/S29-context/die.t, and t/spec/S05-modifier/ignorecase.t, 03:23
also t/spec/S17-supply/batch.t 03:44
lizmat Files=1262, Tests=107941, 409 wallclock secs (30.14 usr 7.23 sys + 2909.12 cusr 248.78 csys = 3195.27 CPU) 08:29
[Tux] Rakudo version 2019.03.1-317-gb2e72d6d7 - MoarVM version 2019.05-4-g807b92f92
csv-ip5xs0.763 - 0.783
csv-ip5xs-206.004 - 6.521
csv-parser23.888 - 25.443
csv-test-xs-200.427 - 0.440
test7.530 - 7.935
test-t1.711 - 1.731
test-t --race0.822 - 0.872
test-t-2030.341 - 30.925
test-t-20 --race9.208 - 10.792
 09:07
gfldex should an LTA for "STable conflict detected during deserialization." go to NQP or Rakudo? 12:34
timotimo i think the error occurs inside of moarvm; if any more info should be made available, it'd have to come from C code and be "improved" upon in nqp or rakudo 12:44
gfldex It would be nice to list the modules that have been loaded so far when the error occurs. 12:47
timotimo ah, that could be done in nqp i'd think 12:48
dogbert17 m: say (for ^4 { .say }).WHAT # is this the correct behavior or should the parenthesis imply lazyness? 14:32
camelia 0
1
2
3
(List)
jnthn dogbert17: Correct, I think; you'd have to stick lazy in for that 14:59
m: say (lazy for ^4 { .say }).WHAT
camelia (Seq)
dogbert17 jnthn: many thanks, I'm looking into some old RT's and your comment will enable me to close one if I can find a test 15:01
Geth nqp/jvm_beyond_java_8: c88c64add0 | usev6++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java
[JVM] Don't warn about illegal reflective access

  ... on Java 9 and newer.		 16:31
bartolin if noone objects, I'll merge github.com/perl6/nqp/pull/442 soon-ish (which requires at least Java 9). Afterwards, I'll add the above workaround. I'm still waiting for a running spectest (with Java 11) to complete, but it looks good so far. 16:38
I wonder if we should just require Java 11. (I doubt someone will use Rakudo or NQP with Java 9 or 10, since they have already reached end of support. I will only test with Java 11.) 16:41
ugexe bartolin++ 16:46
MasterDuke bartolin: +1 on bumping to java 11 17:25

Powered by Cro and the Raku® Programming Language.

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

Thank you!