github.com/moarvm/moarvm | IRC logs at colabti.org/irclogger/irclogger_logs/moarvm
Set by AlexDaniel on 12 June 2018.
timotimo oh, i think the bit is logged inside the dispatcher itself, not the function that has invoked the dispatch 01:10
which is probably what jn meant by "the calculation exactly matches the way used for the [...] lookup" 01:18
Geth MoarVM/new-disp: 3613662b77 | (Timo Paulssen)++ | src/spesh/dump.c
output "cached" annotations properly in spesh log
 01:22
MoarVM/new-disp: 293d0246ed | (Timo Paulssen)++ | src/spesh/optimize.c
first draft for optimizing dispatch_*

a bit of code to lookup the right annotation, then go to the stats to find the "guard index" entry. however, the spesh log data currently lands in the dispatcher instead of the frame using the dispatcher.
MoarVM/new-disp: f78f20a69d | (Timo Paulssen)++ | src/disp/program.c
fix a typo in an error message
 01:23
timotimo so now i'm not sure, do i go via the sim stack when incorporating the log data, do i go via caller-> in interp.c to figure out the bytecode offset and spesh correlation number, or what? 01:27
those questions are jn-shaped :) :) 01:28
nwc10 good *, #moarvm 08:09
MasterDuke: s/ anymore// in your statement earlier 08:10
[and yes, that's Perl 5 :-)]
jnthn: last I tried, NQP happy, SEGV in rakudo. I should try again
actually, no, didn't run rakudo. did 8fb12697752c40e7f331f4e066c6ce40d2884cb9 / 23ba0e8d900e9075414e72e8046653b6464c13fd 08:11
MasterDuke nwc10: ah. i saw a bunch of moarvm commits from a nwc10 in the past and assumed you'd been in at some point 08:24
nwc10 jnthn: NQP still says All tests successful. 08:27
MasterDuke: I forget exactly how they arrived, but it wasn't via github
MasterDuke nwc10: heh, carrier pigeon? western union telegraph? anyway, does that PR look reasonable? 08:34
nwc10 "look reasonable" - yes 08:47
I only looked at it. Didn't test it 08:48
but thanks for following up on my comments faster than I managed to
Geth MoarVM: dogbert17++ created pull request #1302:
Update libuv to version 1.38.0		 14:25
timotimo "i haven't tried it, i only proved it correct" :D 14:48
[Coke] :)
jnthn Hmm, the azure builds all fail on github.com/MoarVM/MoarVM/pull/1302 with some git error... 23:00
I also don't understand why it does JVM builds, which aren't relevant for MoarVM at all
MasterDuke i mentioned the jmv thing to patrickb and left a .tell for him about the git error 23:01
yesterday i think it was, but haven't seen him around yet 23:02
Geth MoarVM: 57c8bb84e6 | (Dave Lewis)++ | src/6model/reprs/CStruct.c
Fix CStruct handling of inlined attributes
 23:03
MoarVM: 2aeadc12e2 | (Jonathan Worthington)++ (committed using GitHub Web editor) | src/6model/reprs/CStruct.c
Merge pull request #1295 from ddlws/cstructs

Fix CStruct handling of inlined attributes
MoarVM: 3f6e6582e6 | (Daniel Green)++ | 16 files
Convert MVM_free+NULL to MVM_free_null
 23:04
MoarVM: d00fb6fb21 | (Jonathan Worthington)++ (committed using GitHub Web editor) | 16 files
Merge pull request #1292 from MasterDuke17/convert_MVM_free_plus_null_to_MVM_free_null

Convert MVM_free+setting to NULL to MVM_free_null
jnthn MasterDuke: ah, thanks
.ask patrickb github.com/MoarVM/MoarVM/pull/1265 and its Rakudo counterpart seem reasonable to me, and I think I stuck an approving vote on the problem-solving solution long ago; any blockers you know keeping you from marking them not WIP so we can merge them? 23:06
tellable6 jnthn, I'll pass your message to patrickb

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!