03:26 apogee_ntv left 03:27 apogee_ntv joined
patrickb timo: In the current debugger impl, I'm essentially processing incoming events as a single stream of events. I rely on events not jumping the queue. This is made difficult by MVM-Remote splitting up incoming events into multiple streams. What's your opinion on adding a all-events supply that could be used alternatively? 05:46
06:39 kjp left 06:41 kjp joined 06:42 kjp left 06:43 kjp joined 12:04 [Coke]_ joined 12:07 [Coke] left 12:43 [Coke]_ is now known as [Coke][, [Coke][ is now known as [Coke]
timo feel free to put something like that in if it would help you 15:06
19:33 lizmat left 19:34 lizmat joined 20:44 lizmat left 20:46 lizmat joined
timo 1st time I'm seeing spesh complain that the reverse postorder calculation failed :) 21:46
I'm now fuzzing JSON::Fast and JSON::Tiny, which as you may notice are both written in raku, not nqp :) 22:59
with defered forkserver and persistent mode, the stability is garbage but the speed goes up and down between 1k and 3k execs per second 23:00
that's decent enough to actually fuzz something!
[Coke] are you fuzzing the raku modules to catch stuff in moar? 23:01
timo right now i'm really just developing the ability of fuzzing code in general 23:26
i think i saw two lyrids!
I currently don't have a commandline to build a nqp-m or rakudo-m that has ASan turned on, so it's not going to be very good at finding real hard crashes at all 23:29
do you have any idea in particular for what I should point the fuzzer at to maybe find something interesting in the fuzzing implementation itself, or the fuzzed thing? 23:33