Parrot 6.4.0 "Double-eyed Fig Parrot" | parrot.org/ | Log: irclog.perlgeek.de/parrot | #parrotsketch meeting Tuesday 19:30 UTC
Set by moderator on 25 May 2014.
Coke saw a commit to master from ZYROz; intentional? 00:08
rurban that was the merge into master 00:20
Util Coke: Looks like rurban merged the pmc2c_orig2-gh1069 branch into master.
rurban It's stable now
We want to release it next week, and rakudo gets a 4% speed boost 00:21
parrot -O2 was not faster with rakudo when I tested it. only parrot got faster 00:22
01:11 rurban joined 01:17 rurban joined 01:20 FROGGS_ joined 04:14 nnunley joined 05:58 awwaiid joined 06:05 awwaiid joined 06:17 rurban joined 06:19 rurban1 joined 06:27 rurban joined 06:42 awwaiid joined 07:02 FROGGS_ joined 07:28 rurban joined 08:29 rurban joined 08:31 rurban1 joined 08:40 basiliscos joined 09:33 rurban joined 09:47 Psyche^ joined 10:33 rurban joined 11:09 agre joined 11:34 rurban joined 12:35 FROGGS joined 12:36 rurban joined 12:56 rurban joined 13:00 Chirag joined 13:37 bighugedog joined 14:03 rurban joined 18:00 Chirag joined
Chirag rurban: Hey! 18:08
rurban_ Hey. did you see? 18:17
And I think we can delete our branches now. Next steps 18:19
Chirag yes I saw 18:22
but I am finding many more
I am not making any changes just noting them down 18:23
Coke opbots, trust Chirag 18:27
slavorg Ok
rurban_ put them in a new branch please. I'll test them. 18:33
Chirag alright..
rurban_ git branch -d <branch>; git push origin :<branch>
to delete it locally and remote 18:34
There's still time to put more optimizations in until Saturday
Chirag yeah.. i am trying to check all again 18:35
I deleted from my repo(origin) but I dont have permissions for parrot's repo(upstream) 18:44
rurban_ just push it to yours 18:49
Chirag i did that
but it still shows in parrot/parrot
rurban_ done 18:50
you can add a 3 for the new WB optims
Chirag i can name it anything? 18:52
rurban_ well, ZYROz/pmc2_orig3-gh1069 would be nice. Something with optim_wb? 18:54
We'll delete it anyway soon
Chirag i ll keep that.. 18:55
rurban_ Only if we don't delete it (stalled old work) they name should reflect somehow the context. So it's good to have the gh#issue in it
Chirag ok.. 18:56
there are some sure shot optims but i am not sure about some METHODS
rurban_ Just look at the generated C then
Basically they are all :write, but you can mark them as not (:manual|no_wb) 18:57
And they have all RETURN, no return. Which is good 18:58
Chirag but all RETURNS would be doing a manual_wb
rurban_ no. if you annotate it with :manual_wb no WB will be added there 19:02
I think so. need to check
technically :no_wb and :manual_wb are the same, WBs are not inserted automatically 19:03
It's just better for the eye. Does not need vs have it already
Chirag yes
rurban_ And there's a third option. Just add the WB ine or in a comment manually, and pmc2c will detect it and mark as :manual_wb 19:04
Needs to be added to the docs...
19:04 bighugedog joined
Chirag hmm 19:05
I will make two commits... One with all the ones I am sure of and the other for all the METHODS... then u can check and comment
actually the fact that all METHODS are :write helps.. 19:06
rurban_ sure 19:11
Chirag i was wondering why sort() and reverse() need a WB 19:12
rurban_ METHOD's are really special cases, like MULTI also.
You need to improve them in the nest task. signature handling, in pmc2c
Should be even easier than the WBs 19:13
Chirag When do I start with that?
rurban_ When you got time :)
Let's look over the WB's first 19:14
Chirag I have no other priorities..
:)
ya I am looking at them now
rurban_ It will not be ready for 6.5.0 for sure. But maybe 6.6.0
Chirag Will you be creating another issue# for this task? 19:16
rurban_ Yes, just too busy with other stuff now 19:17
Chirag sure.. np.. will continue with the WBs 19:18