|
Post closed tickets in your report. | Note: This channel is for our weekly status meetings (Fridays at 13:30 UTC); you probably want #parrot instead. | irclog: irclog.perlgeek.de/ Set by moderator on 12 November 2014. |
|||
|
07:52
Reini joined
12:23
ilbot2 joined
|
|||
| moderator | Post closed tickets in your report. | Note: This channel is for our weekly status meetings (Fridays at 13:30 UTC); you probably want #parrot instead. | irclog: irclog.perlgeek.de/ | ||
| Reini | Who is in attendance today? | 13:32 | |
| Util | I am, but late. | 13:33 | |
| Reini | I only did one branch, but this is a big one | ||
| DONE: | 13:34 | ||
| * smoke-me/rpa-offset-gh1152 OPT | |||
| + Rewrote ResizablePMCArray with moving offset, esp on shift. | |||
| Even better than nqp/qrpa. 20% faster overall. #1152 | |||
| + Change ResizablePMCArray.delete to throw out of bounds errors, | |||
| but just warn on deprecated delete[0] with size 0 for backcompat. #1153 | |||
| + Add check for illegal delete of capture blocks in PGE.match #1154 | |||
| + Add new traceflag -t20 for ResizablePMCArray states. | |||
| + Rewrote docs/pmc/array.pod | |||
| + Added tests to t/pmc/resizablepmcarray.t to cover all offset cases. #1152 | |||
| WORK | |||
| * smoke-me/embed-gh829 FIX | |||
| * analyze performance loss with --gc gms #1141 (20%), see parrot-bench OPT | |||
| BLOCKED | |||
| * smoke-me/signal-tests + smoke-me/test-capture-gh892 need windows smokes CAGE | |||
| * smoke-me/vms-gh866 needs a vms smoke FEAT | |||
| Too much wait. I'll just merge them today. Look good to me. | |||
| So basically parrot is now the fastest in all the releases covered in parrot-bench | |||
| earlier release do not build for me on 64bit | |||
| nor -m32 | 13:35 | ||
| Util | Done: Attended Atlanta.pm.org until very late last night. Briefly discussed the 2015 Perl 6 release. Mostly had discussions on archives from the Perl 5 Advent calendars. | 13:37 | |
| .end | |||
| Reini | We are also organizing the next big German Perl Workshop in Dresden next year. Thinking what talks to do, or not | 13:38 | |
| Had a long meeting with them last night until 1am :) | |||
| Util | Reini: After merge, please leave smoke-me/signal-tests + smoke-me/test-capture-gh892 in place (instead of deleting them) for a while, in case a Win32 problem turns up. | 13:39 | |
| German Workshop: around when next year? | |||
| Reini | Middle of May | ||
| yes, will do. travis is acting a bit weird lately, They updated to Ubuntu precise, hence I removed some wrong deps | 13:40 | ||
| Util | Thx | ||
| Reini: Congratulations, and Reini++ , for performance gains! | 13:41 | ||
| lizmat_ | Reini++ :-) | 13:42 | |
| lizmat goes back to lurking | 13:43 | ||
| Reini | And we still have 20% for the GC, ~15% for the pcc and maybe 80-100% for the jit lateron | ||
| moar is still better and faster, but they don't support lock-less threading and pass less tests. | |||
| so we have some time left on our clock | 13:44 | ||
| esp. their gcc and calling convention is okay | |||
| GC, not gcc | |||
| after this release I will do less, switch to p2 a bit, then perl5 updates for 5.22 | 13:45 | ||
| Util | I understand doing less. Your sprint has been very impressive. | 13:47 | |
| Reini | I had not much else to do. the perl5 compiler is finished, and I had to wait for the next projects | ||
| the GC slowdown would be important also. its either the generations, but I doubt that, or the stringpool | 13:48 | ||
| I wonder how many new failures perl6 will get when delete[invalid index] is now called. I only saw 2 fails, but I'm not if they are regressions | 13:52 | ||
| not sure if | 13:53 | ||
| Util | I wonder also. I will kep an eye out for it. | 13:56 | |
| Anything else to discuss? From anyone? | 13:57 | ||
| Reini | nope | ||
| Util | Happy Holidays, everyone! | 13:58 | |
| Meeting adjourned. | |||
| Reini | oops, t/01-sanity/08-var-array.t test 11 looks like one | ||
| I'll check | |||
|
15:36
Reini joined
|
|||