github.com/moarvm/moarvm | IRC logs at colabti.org/irclogger/irclogger_logs/moarvm
Set by AlexDaniel on 12 June 2018.
nwc10 good *, #moarvm 06:59
lizmat nwc10 o/ 07:04
nwc10 finally we have the rain 07:06
after 3 days of weather forecasts saying "it will rain this afternoon" or similar values of "soon"
dogbert17 is still dabbling with the eval bug 13:56
seems as if it is this variable attribute which is somehow getting busted: github.com/Raku/nqp/blob/master/sr...W.nqp#L742 14:00
the code often fails with the msg 'Iteration past end of iterator' and --ll-exception points to github.com/Raku/nqp/blob/master/sr...W.nqp#L750 14:04
at the MoarVM level we're at this line github.com/MoarVM/MoarVM/blob/mast...Iter.c#L56 14:05
when the program fails more violently, i.e. SEGV or ASAN barfage it happens in array related code, i.e. VMArray.c 14:07
==219346==ERROR: AddressSanitizer: heap-use-after-free on address 0x61d000770410 at pc 0x7f4b86173fc2 bp 0x7f4b7dbbcc70 sp 0x7f4b7dbbcc60 READ of size 8 at 0x61d000770410 thread T4 14:08
#0 0x7f4b86173fc1 in at_pos src/6model/reprs/VMArray.c:161
#1 0x7f4b861be309 in shift src/6model/reprs/MVMIter.c:57
ok, enough babbling :)
lizmat nwc10++ # fixing the JS build 16:08
nwc10 well, yes-and-no 16:31
half fixing it
lizmat well, noticing it and working on it then :-) 16:34
nwc10 unfortunately, I don't think that I can fix the second problem that I found -- github.com/Raku/nqp/issues/656 16:36
so not working further on that.