github.com/moarvm/moarvm | IRC logs at colabti.org/irclogger/irclogger_logs/moarvm
Set by AlexDaniel on 12 June 2018.
00:17 chansen_ left 00:19 chansen_ joined 00:35 Kaeipi left, Kaeipi joined 00:43 dogbert17 left 01:33 dogbert17 joined 05:04 Geth joined 06:31 squashable6 left 06:32 squashable6 joined 06:43 domidumont joined
nwc10 Beware - everything you read on the Internet today is unfunny and untrue. Including this statement. 06:47
06:59 patrickb joined 07:09 frost-lab joined 07:40 linkable6 left 07:41 linkable6 joined 07:56 Kaeipi left 07:58 Kaiepi joined 08:16 zakharyas joined 08:54 linkable6 left 08:55 linkable6 joined
lizmat I don't agree. It's not funny at all! And true! 09:10
MasterDuke jnthn: colabti.org/irclogger/irclogger_lo...-04-01#l44 has a question you might be able to answer 09:16
09:18 zakharyas left 09:35 linkable6 left 09:36 linkable6 joined
nwc10 paste.scsys.co.uk/595028 -- t/spec/S12-attributes/undeclared.t SEGVs spesh 09:47
attempted NULL pointer dereference. Nothing fishy reported by ASAN or valgrind before this 09:48
have not attempted to bisect why this now goes boom
or reduce it 09:49
(master/master/master as of right now) 09:50
AFK again 09:51
10:08 zakharyas joined 10:40 linkable6 left 10:41 linkable6 joined
Geth MoarVM: MasterDuke17++ created pull request #1454:
Free Uni property codes hashes with --full-cleanup
10:51
10:56 zakharyas left 11:01 frost-lab left, frost-lab joined 11:32 frost-lab left 12:17 sena_kun left, sena_kun joined 12:47 zakharyas joined
dogbert17 a SEGV, interesting 12:57
14:19 patrickb left
dogbert17 dogbert's replication attempts are a total failure :( 14:32
nwc10 I have the 3 environment variables set 14:33
plus
+#define MVM_ARRAY_CONC_DEBUG 1
+#define FSA_SIZE_DEBUG 1
+#define MVM_GC_DEBUG 1
+#define MVM_SPESH_CHECK_DU 1
dogbert17 you've it all it seems :)
nwc10 but it's probably some combination of MVM_SPESH_BLOCKING=1 and MVM_SPESH_NODELAY=1
and I've not had time to dig
dogbert17 will try for a while longer now that I have more info 14:35
tw, what does ROAST_TIMING_SCALE=8 do? I have never used that. 14:37
*btw
nine Prevents tests from timing out just because rakudo is slowed down by the debugging options 14:40
dogbert17 nine, thx
nine /win 10 16:06
dogbert17 ==655849==The signal is caused by a READ memory access. 16:19
==655849==Hint: address points to the zero page.
#0 0x7fc1ea34556b in MVM_spesh_arg_guard_run src/spesh/arg_guard.c:496
#1 0x7fc1ea0fb31a in MVM_frame_invoke src/core/frame.c:493
#2 0x7fc1ea221701 in invoke_handler src/6model/reprs/MVMCode.c:10
baam
oh, I can even get it to fail on every run 16:31
nine most important step to fix it :) 16:44
dogbert17 nine: I simply removed the first seven tests from the file so that there were only four left. That did the trick. 17:01
followed by 'MVM_SPESH_NODELAY=1 MVM_SPESH_BLOCKING=1 ./perl6-m -Ilib t/spec/S12-attributes/undeclared.t' 17:02
17:14 domidumont left 17:40 MasterDuke left 18:35 zakharyas left 20:33 zakharyas joined 20:45 MasterDuke joined, dogbert11 joined 20:49 dogbert17 left 22:03 zakharyas left