github.com/moarvm/moarvm | IRC logs at colabti.org/irclogger/irclogger_logs/moarvm
Set by AlexDaniel on 12 June 2018.
Geth_ MoarVM/fix_spesh_plugin_guard_set_threading_issue: a3d18b2f2c | (Stefan Seifert)++ | 13 files
Fix access to freed memory in resolve_using_guards

Commit acb04a448bd7b067f1a0943b6fd521e21057c96e fixed outdated pointers in the guard set, but it was still possible that the guard set was freed while still in use by resolve_using_guards as the next GC safe point may occur in evaluate_guards. Fix this by turning SpeshPluginState into a 6model object, so we can let the GC figure out whether the state is still in use somewhere or not
07:07
nine The failures in t/05-messages/02-errors.t are just timing issues as well. This leaves t/05-messages/10-warnings.t, t/02-rakudo/repl.t and t/06-telemetry/01-basic.t to investigate 15:03
dogbert17 nine: are those errors specific to your branch? 15:12
nine dogbert17: that's what I want to find out. I'm very sure that they are not. I have never had a completely successful make test with extreme GC debugging measures active. My branch fixes some of the failures. The rest appears to be just timing and maybe additional, unknown issues. 15:14
I think repl.t has had interemittend failures for ages. Telemetry may have some specific issues just like we've found in the profiling code. 15:16
dogbert17 very interesting 15:17
lizmat looks forward to any findings wrt Telemetry :-) 15:27
MasterDuke nine: given you've been looking at profiling related stuff recently, do you have any thoughts/suggestions re the problem dogbert17++ found and then timotimo++ helped golf last night? 15:35
nine MasterDuke: no, nothing to add at this time 15:36
MasterDuke ah well 15:38
lizmat somehow, this sounds familiar: thenewstack.io/a-new-linux-memory-...ts-of-ram/ 20:00
MasterDuke lizmat: i remember seeing that a couple days ago. i still have the same question, does it only help if you're using multiple cgroups? 20:01
lizmat I have no idea... :-) 20:02
MasterDuke well, maybe i'll find out in a year or so if it gets merged 20:05
nine: completely changing topic, but after all your mast generation changes, is it as optimized as it could easily be? 22:32
Geth_ MoarVM: 01f85c8bf8 | (Daniel Green)++ | 19 files
Revert "Revert "Merge pull request #1139 from MasterDuke17/add_context_to_error_messages""

This reverts commit b2dc3f3ba2f0c4dc9c7bb35c0925a307b60b98d6.
This was reverted because a previous commit also had to be reverted and I wanted these changes to get their own NQP bump for easy bisecting.
22:35