Post closed tickets in your report. | Note: This channel is for our weekly status meetings (Tuesdays at 19:30 UTC); you probably want #parrot instead. | irclog: irclog.perlgeek.de/
Set by moderator on 9 July 2014.
04:49 lizmat joined 06:36 basiliscos joined 08:08 basiliscos joined 08:32 wagle joined 13:29 rurban1 joined 13:31 bluescreen joined 14:04 lizmat joined 15:02 Chirag joined 17:29 lizmat joined 17:31 rurban1 joined 17:42 lizmat joined 18:23 lizmat joined 18:32 rurban1 joined
Util # Done 19:32
* Parrot release last week.
# Doing:
* smolder.parrot.org (still not fixed)
.end
Hello! Who all is in attendance?
Chirag me 19:33
19:33 rurban1 joined
Util Welcome, Chirag! 19:34
Ready to report?
Chirag thanks! in a minute, please 19:35
Util OK
Chirag Report- 19:37
#Done:
*change pmc_callcontext.h - inline the CallContext ATTR accessors to omit the obj check (Issue 1083 - part 3)
#Doing:
*other requirements of Issue 1083 (exporting direct calls)
gist.github.com/ZYROz/2b709dc74bd354143f81 -- has the benchmark data for my changes.. very weird data actually 19:40
Util Hmmm. very weird 19:42
Chirag It would be nice if rurban could run it on his machine to confirm .. so it has (the expanded add_invocant change + the changed pmc_callcontext.h) 19:43
Util Which branch are you working in? 19:46
Chirag ZYROz/pcc-gh1083 19:47
Util Thanks; I had just found it :) 19:48
Chirag: are your sure it is "ZYROz/pcc-gh1083", and not just "pcc-gh1083" ? 19:52
Chirag yes, sorry
github.com/ZYROz/parrot/tree/pcc-gh1083
Util And which two commit points need comparing? 19:54
Chirag gist.github.com/ZYROz/2b709dc74bd354143f81
I have only committed add_invocant expansion .. I am comparing this work+ .h file changes with the master branch 19:55
Util Chirag: Do you mean that the first commit point that needs benchmarking is '71d5afe', and the second is '71d5afe'+your_diff ? 19:56
Also, what platform are you benchmarking on? 19:57
Chirag I havent benchmarked 71d5afe alone.. just 71d5afe+my_new_.h_file against master 19:58
Ubuntu
Ubuntu 14* 20:00
Util and what are you using to benchmark (where is run-bench.sh from? I have not benchmarked Parrot before) 20:01
Chirag parrot-bench/do-bench.sh 20:02
Util I have EC2 access, and should be able to replicate.
Your fork of parrot-bench, or the original? 20:03
Chirag i was using my fork .. but no changes made
Util OK 20:04
Other than wanting confirmation of your weird data, do you have any blockers in the next week? 20:05
Are you comfortable with the next week's work?
rurban I was at the vet again, sorry 20:06
Chirag Yes, the only blocker was what I had asked previously regarding direct calls
So, comfortable 20:07
rurban I'l run Chirags's tryouts to see the results
Other than that I fixed a couple of minor stuff I detected
Not much else, still working on B::C (again) mostly 20:08
Util Thanks, rurban!
BTW, what Ubuntu package contains "perf"?
rurban I fear we'll only get big improvements on the method call front if we change back from args arrays to args in registers. 20:09
Util nm, I see it.
rurban linux-tools
Chirag linux-tools
Util thanks
rurban I hope I can benchmark in 2 days, right now my machines are too busy 20:10
Chirag no problem, I will finish the exporting of static methods by then 20:11
Util Anything else to discuss? From anyone? 20:15
Chirag nope, thanks! 20:19
Util OK. Thanks everyone! 20:20
Meeting adjourned.
20:35 rurban1 joined 21:13 basiliscos joined 21:29 lizmat joined 21:35 rurban1 joined 22:36 rurban1 joined 22:57 rurban1 joined 23:28 lizmat joined