/srv/irclogs.ubuntu.com/2011/03/02/#ubuntu-mozillateam.txt

=== asac_ is now known as asac
julius__hello guys, the nightly builds for firefox have failed again ...07:38
julius__1 out of 4 hunks FAILED -- rejects in file uriloader/exthandler/unix/nsMIMEInfoUnix.cpp07:39
julius__Patch mozilla-kde.patch does not apply (enforce with -f)07:40
micahgjulius__: feel free to propose a merge fixing it if you like07:40
julius__micahg: I would gladly do so, but my system is not prepared for builds and trying do set it up will probably take longer than having someone else looking over the issue07:42
micahgjulius__: we'll try to get to it soon, I won't be able to look at it until tomorrow night07:46
julius__micahg: OK, no problem, just wanted to inform you; thank you!07:47
micahgjulius__: we could use someone to help keep the dailies going if you'd be interested in doing that07:48
julius__micahg: I could give it a try, for the next 10 days I have more time available07:49
micahgjulius__: sure, it's purely volunteer, so you can spend as much time as *you* want07:50
julius__micahg: OK then07:50
* micahg actually started with that almost 2 years ago :)07:50
julius__I see ... :)07:51
julius__micahg: so, how do I start?07:51
micahgthe branch is lp:firefox07:51
micahgyou can grab the .orig.tar.gz from the daily PPA07:52
julius__OK07:53
julius__micahg: I found some differences but I don't know what to do with them08:57
julius__micahg: where and how I can test the new patch?09:06
micahgjulius__: sorry, it's 3AM and I have to finish something before bed, can we continue later?09:15
julius__micahg: sure, no problem; sleep well!09:15
micahgjulius__: there should be a README.source in the debian dir in lp:firefox if you're feeling adventerous09:16
julius__micahg: OK, I'll look there; have a good night sleep!09:17
micahgjulius__: thanks09:19
fta2micahg, http://launchpadlibrarian.net/65426925/buildlog_ubuntu-natty-armel.chromium-browser_9.0.597.107~r75357-0ubuntu1_FAILEDTOBUILD.txt.gz  :(09:20
fta2i should drop arm, i'm just poking in the dark without any way to build or test :(09:21
julius__fta2: do you know how to integrate a patch into the firefox nightly build?09:27
fta2julius__, sure, but i'm no longer a mozilla maintainer, ask chrisccoulson or micahg09:28
julius__fta2: OK, thanks09:29
micahgfta2: well, there's a task assigned to linaro to help with that09:39
julius__chrisccoulson: hi, I saw that the firefox nightly did not build and I think I found the issue; how can I test the patch?10:45
=== m_conley_away is now known as m_conley
fta2asac, help!14:08
asacfta: i am here ;)14:13
* asac hopes fta didnt drown already14:13
chrisccoulsonhi fta214:23
chrisccoulsoni might have the thunderbird.head branch ready to start 3.3 dailies later14:24
=== m_conley is now known as m_conley_away
ftachrisccoulson, cool, i need the new branch names and locations15:57
ftaasac, still there?15:57
ftaasac, http://paste.ubuntu.com/574525/  i need to know where i should enable/disable armv7, arm_neon, arm_thumb for lucid, maverick, natty15:58
ftaasac, i'm sick of poking in the dark, i can't build, i can't test15:59
ftaasac, my last update failed, i found out why by digging into my bzr branches, but it's not a good solution, i always find the problems afterwards16:00
ftaasac, my only other option is to simply stop caring about arm :(16:00
=== m_conley_away is now known as m_conley
=== m_conley is now known as m_conley_away
jdstrandfta: fyi- https://launchpad.net/~ubuntu-security-proposed/+archive/ppa/+buildjob/229442219:28
jdstrandfta: maverick/armel ftbfs19:28
ftajdstrand, i know19:28
jdstrandah19:28
jdstrandok19:28
ftasame on natty19:28
jdstrandfta: I'm somewhat inclined to push out i386 and amd64 and have armel get fixed in the next update. what is your opinion?19:29
jdstrandmicahg: ^19:29
ftajdstrand, http://irclogs.ubuntu.com/2011/03/02/%23ubuntu-mozillateam.txt  9:20 and 15:5819:30
ftajdstrand, yep, please do19:30
jdstrandfta: ok19:31
ftajdstrand, i really don't know how i can really support arm. we don't have porter boxes like debian, i don't have that h/w here, i'm not allowed to have a native ppa (which won't solve the testing part anyway), and no-one from the arm team seem to care enough to help.19:33
ftaso i can only flip switches in the dark with a long stick, and hope for the best.19:34
ftadepressing19:34
jdstrandfta: I hear you. micahg may be able to help going forward. one option could be to have him funnel your packages to the ubuntu-mozilla-security ppa19:34
jdstrandfta: but that is only part of the problem, I realize19:35
micahgthe problem looks like it just needs an ifdef, I'm just not sure what the ifdef should be19:36
micahgI might be able to get janimo to take a look19:36
ftanonix4, it's just that i forgot a arm_thumb=0 during the codecs merge, but i found another while trying to fix that19:36
ftanonix4, n-m19:37
ftai meant "no, it's..."19:37
micahgah, ok19:37
* micahg should've checked that as well19:37
ftabut i had armv7 set in one package and unset in the other, for months, no-one noticed19:38
ftai'm wondering if anyone is really testing those19:39
chrisccoulsonfta - for arm - i feel exactly the same as you ;)19:39
ftahence my request to asac19:39
chrisccoulson(although i can build it, i have no way of testing anything on arm)19:39
chrisccoulsonand the porter boxes are super slow ;)19:39
ftai'm accepting donations ;)19:39
chrisccoulsonheh :)19:40
ftai should blog about this19:40
ftai think i will19:41
=== m_conley_away is now known as m_conley
asacfta: dont bother about lucid ... maverick and natty are important. both are armv7=1 and neon=0 and arm_thumb=1 ... but for performance intense hot spots of code you might want to keep arm_thumb=0 if you have an option to do that for parts19:57
asacwe are adding "neon if available" for skia so then you can use neon=1 ... but you would be notified i guess19:58
asaclucid is the same btw19:58
asacfta: do you want hardware?19:59
asaci can get you a board19:59
ftathat would solve the problem19:59
asacdefinitly a beagle for now ... a panda probably soon19:59
asacawesome i will work on that19:59
asaclet me see if i can get you access to a porter box as well sooner20:02
ftaexcellent20:05
=== debfx_ is now known as debfx
m_conleychrisccoulson: ping20:37
m_conleychrisccoulson: ping20:58
chrisccoulsonhi m_conley, how are you?20:58
m_conleychrisccoulson: things are good!  Have things started to calm down for you a bit yet?20:58
chrisccoulsonheh :)20:58
chrisccoulsoni wish ;)20:58
m_conleychrisccoulson: well, I was wondering if you wanted to give my add-on a spin:  https://addons.mozilla.org/en-US/thunderbird/addon/messaging-menu-integration/20:59
chrisccoulsonyeah, i will do. i'm just trying to build thunderbird 3.3 atm20:59
m_conleynice21:00
m_conleyit's a good version.  :)21:00
m_conley(the add-on is currently only supporting 3.1.* though)21:00
chrisccoulsonah, ok21:00
chrisccoulsonthat should be easy for me to test then :)21:00
ftachrisccoulson, still in the a3 crashes?21:40
chrisccoulsonfta - not now \o/21:56
chrisccoulsonjust getting tbird 3.3 ready :)21:56
ftachrisccoulson, good, maybe you'll have time for breakpad soon then ;)21:57
chrisccoulsonfta - yes :)21:57
ftachrisccoulson, http://ftagada.wordpress.com/2011/03/02/blindly-poking-at-arm-bugs-with-a-stick/21:57
chrisccoulsoni keep thinking about buying some arm hardware21:59
chrisccoulsonif i can find any ;)21:59
=== m_conley is now known as m_conley_away
chrisccoulsonit is a bit of a pain. and firefox takes nearly 10 hours to build on arm now!22:00
chrisccoulsonactually22:00
chrisccoulsonit's 12 hours :(22:00
chrisccoulsonthat sucks22:01
fta21 for ch, before the codecs merge22:01
chrisccoulsoneek, that's pretty long22:01
chrisccoulsonare you running a test suite too?22:01
ftano22:01
ftanot anymore22:01
chrisccoulsonthat *is* quite long ;)22:01
chrisccoulsonthe 12 hours for firefox includes the full set of tests, which take up most of the time22:01
ftaa couple of hours just for linking the binary22:02
chrisccoulsonor, at least i *think* it's most of the time. i guess i should instrument the build really22:02
ftaPERF: 'chrome chrome_sandbox ' built in 1184 sec  (amd64)22:06
ftaPERF: 'chrome chrome_sandbox ' built in 69396 sec (armel, without the codecs)22:06
ftajust the build, not the deb, strip, etc22:07
ftaand that's v9, v11 is biggest22:07
ftabigger22:07
fta1h20 spent outside of the real build22:10

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!