github.com/moarvm/moarvm | IRC logs at colabti.org/irclogger/irclogger_logs/moarvm
Set by AlexDaniel on 12 June 2018.
timotimo fixes heap snapshots 14:01
spesh graphs weren't given a "describe" function when spesh became interruptible
brrt \o 14:03
timotimo wow, the heapanalyzer somehow became very unhappy in the mean time 14:05
free(): invalid pointer 14:07
brrt hmmm 14:09
damn those developers, always breaking things
timotimo also, perhaps it's not understanding the heap snapshot format correctly any more? 14:10
brrt I know, as usual, nothing of the matter 14:13
timotimo :) 14:14
Geth MoarVM: 4666ff437e | (Timo Paulssen)++ | 3 files
make heap snapshot profiler functional again
14:15
timotimo looking forward to having automated sanity tests for the profilers at some point %)
brrt I think we're nearing the stage that we need to have tests for MoarVM.... 14:16
This morning, I had the mad plan of having inline expression templates 14:17
does ` do anything in C?
brrt should get himself K&R at some point
timotimo ` didn't exist back then :P 14:38
that commit id is beautiful 14:39
dogbert2_ didn't it use to be possible to run a single spectest file using make, e.g. 'make t/spec/S32-list/seq.t' ? 14:42
timotimo yeah i think so 14:51
man, i need better programs to run the profiler on 14:56
Geth MoarVM: ugexe++ created pull request #985:
Use larger buffer for gethostname when available
timotimo i want something that has multiple threads, runs a bunch of GC, and actually also has major collections 15:01
Zoffix .ask brrt does R#2395 (Proc.in.close hang/long time) sound like anything you may have worked on or do you have any ideas about that ticket? I saw some commits about event loops in MoarVM. Dunno if that's what's involved in handling closing of STDIN for a proc (it uses Proc::Async under the hood) 16:10
yoleaux Zoffix: I'll pass your message to brrt.
synopsebot R#2395 [open]: github.com/rakudo/rakudo/issues/2395 [⚠ blocker ⚠] hangs in t/spec/S32-io/io-cathandle.t
Zoffix ins->operands[2].lit_i16 = repr_data->attribute_offsets[repr_data->unbox_num_slot]; 18:36
Does that look right? "lit_i16" being assigned a "unbox_num_slot"? Like int vs num innit? 18:37
looks like that's used everywhere… never mfind.
Geth MoarVM/fix-blocker: ff8c03790b | (Zoffix Znet)++ | src/6model/reprs/P6opaque.c
Comment out problematic case in speshing unboxing of _n

Fixes R#2395 github.com/rakudo/rakudo/issues/2395 but I've no idea why.
18:57
synopsebot R#2395 [open]: github.com/rakudo/rakudo/issues/2395 [⚠ blocker ⚠] hangs in t/spec/S32-io/io-cathandle.t
Geth MoarVM: zoffixznet++ created pull request #986:
Comment out problematic case in speshing unboxing of _n
18:58