| gnomefreak | they have to change code anyway since rc1 should say that in help and on top panel | 00:00 |
|---|---|---|
| gnomefreak | so in theory the code is gonna change anyway | 00:00 |
| gnomefreak | weather they call it final or rc2 noone knows until they do it | 00:01 |
| gnomefreak | but i agree 100% that bug is a blocker | 00:01 |
| gnomefreak | ok rc1 is gonna be blessed final as in it wont be branded as rc | 00:02 |
| gnomefreak | i think that is where the original post came from | 00:03 |
| gnomefreak | so they were just talking within the build being branded as final :) | 00:03 |
| gnomefreak | ok back to work around here so i can get to sleep tonight | 00:05 |
| danage | hello! you will hate this question, but is there an ETA to fix flash? | 01:43 |
| === asac_ is now known as asac | ||
| futtta | morning all | 07:26 |
| futtta | quick question: i noticed ff3b5 in hardy is faster then ff3rc1 from mozilla in the sunspider tests. is this because the hardy ff3b5 has been optimized? | 07:31 |
| futtta | no-one awaky? :) | 07:34 |
| futtta | awake that is | 07:35 |
| asac | futtta: it is? | 08:05 |
| asac | interesting | 08:05 |
| asac | maybe its because b5 still used jemalloc | 08:05 |
| mconnor | we turned off jemalloc on Linux? | 08:13 |
| futtta | back | 08:13 |
| futtta | asac; yep, according to a quick-and-dirty test i did anyhow | 08:14 |
| futtta | cfr. http://blog.futtta.be/2008/05/19/firefox-3rc1-shines-in-javascript-benchmark/ | 08:14 |
| mconnor | hmm | 08:14 |
| === fta2 is now known as fta | ||
| mconnor | oh | 08:14 |
| mconnor | asac: what' gcc are you guys shipping? | 08:14 |
| mconnor | 4.3 is supposed to be about 10% faster than the 4.1 we're using | 08:15 |
| mconnor | futtta: about:buildconfig in b5? | 08:16 |
| asac | mconnor: not sure if you turned jemalloc off ... at least its not linked statically into libxul anymore | 08:17 |
| mconnor | oh, didn't we fix that for you guys? | 08:17 |
| asac | right. but in b5 we still kept it enabled | 08:17 |
| asac | well .. at that point it was still linked in | 08:17 |
| asac | in hardy we have gcc (GCC) 4.2.3 (Ubuntu 4.2.3-2ubuntu7) | 08:18 |
| asac | in intrepid (current development) its 4.3 | 08:18 |
| mconnor | hmm | 08:18 |
| mconnor | that's still going to be faster | 08:18 |
| mconnor | 4.1 kinda sucked | 08:18 |
| asac | yes. afaik ew somewhat skipped 4.1 | 08:19 |
| asac | futtta: could you try the preview of RC1 we have in https://edge.launchpad.net/~fta/+archive to see if the performance gap is gone? | 08:21 |
| futtta | mconnor: i'm at work now, but i'll reboot into ubuntu later today to check about:buildconfig | 08:21 |
| futtta | asac: same, i'll check that later today | 08:21 |
| asac | thanks | 08:21 |
| asac | let us know | 08:22 |
| fta | we have issues with mozilla tools built with gcc 4.3 in intrepid. See the bottom of https://wiki.ubuntu.com/CompilerFlags | 08:22 |
| asac | yeah .... but those compiler flags were added by ubuntu to improve security mostly | 08:23 |
| mconnor | futtta: its ok, 4.2.3 would explain it | 08:24 |
| fta | asac, yet it shows issues either in xul or in libc | 08:24 |
| asac | right | 08:24 |
| asac | i guess we should get a backtrace with debug symbols from it | 08:25 |
| mconnor | if there's an upstream bug, bsmedberg is pretty decent about fixing stuff like that | 08:25 |
| fta | those are aborts, not crashes | 08:25 |
| asac | right | 08:26 |
| asac | buffer overflow ;) | 08:26 |
| mconnor | and, really, one wonders if we should use those flags ourselves | 08:26 |
| fta | the 1st abort has a decent trace, the other aborts had corrupted stacks | 08:26 |
| mconnor | but I don't know how well supported they are | 08:26 |
| asac | mconnor: i can try to get our security guy in contact with you ... or with whoever is best suiteable to discuss this | 08:27 |
| asac | s/try// | 08:27 |
| futtta | some people seem to be arriving on my blog when searching for 'ubuntu firefox 3rc1' in google, what are the plans on pushing rc1 to hardy-users? | 08:44 |
| asac | futtta: working on it .... takes quite a bit of QA to do this transition obviously | 08:44 |
| futtta | that makes sense :) | 08:46 |
| asac | futtta: basically the ffox and xul bits are ready, but we still need to transition some rdepends as it breaks ABI and translations | 08:47 |
| wgrant | asac: Are fta's the most officially blessed packages? I might as well upgrade and try to break things. | 08:47 |
| wgrant | asac: Oh dear, please don't tell me xulrunner broke ABI.... | 08:47 |
| asac | wgrant: we dont really encourage to use them if you are not a developer | 08:47 |
| asac | but otherwise yes. those are more or less the most officially blessed packages we have from the mozillateam that track the trunk | 08:48 |
| wgrant | I was in ~motu last time I checked. Or do you mean Mozilla devs? | 08:48 |
| asac | wgrant: no thats fine. just remember to check here if you see wierd things | 08:51 |
| asac | ok moving to another room | 08:51 |
| wgrant | asac: Of course. | 08:52 |
| [reed] | https://bugzilla.mozilla.org/show_bug.cgi?id=434592 | 08:59 |
| ubottu | Mozilla bug 434592 in Phishing Protection "Safe browsing doesn't work on libxul-sdk builds" [Major,New] | 08:59 |
| wgrant | The AwesomeBar seems faster. | 09:03 |
| mconnor | wgrant: it probably is | 09:05 |
| mconnor | also, breaking ABI between beta and final, well... | 09:06 |
| e-gandalf | hi | 09:24 |
| e-gandalf | [reed]: ping | 09:30 |
| [reed] | e-gandalf: pong | 09:31 |
| e-gandalf | will you help me write a short summary of the SSO idea for weave team? | 09:31 |
| [reed] | for .moz or ubuntu.net stuff? | 09:32 |
| e-gandalf | yea | 09:33 |
| e-gandalf | and if you're attending the mozilla upstream talk? | 09:36 |
| e-gandalf | I could you some help here too | 09:36 |
| [reed] | e-gandalf: yeah, I'm in the Mozilla upstream talk | 09:43 |
| [reed] | if you have questions for your current talk, you can ask here, and I can try to answer | 09:43 |
| e-gandalf | well, I'm just planning to write up a short summary of each talk for our team | 09:44 |
| e-gandalf | what was there important for Mozilla, what have we learnt | 09:44 |
| e-gandalf | etc | 09:44 |
| e-gandalf | I'm in the SSO talk right now | 09:44 |
| e-gandalf | but it's not too active I must say | 09:44 |
| e-gandalf | they probably discussed it yesterday | 09:44 |
| [reed] | yeah | 09:45 |
| [reed] | two hour talk | 09:46 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!