20100420

April 20th, 2010, 13:00 UTC in #ubuntu-meeting.

Agenda

Action Items from April 13th, 2010

  • dyfet to check update-manager sanity checks for ARM upgrades.
    • Carried Over.
  • GrueMaster to send a list of bugs that need attention to asac.

    • Carried Over.
  • JamieBennett to add StartupNofify support to webservice-office-zoho.

    • Done and uploaded. In the images now.
  • asac to work with roc on the firefox scrollbar issue.
    • Fix found, needs uploading.
  • NCommander to investigate gphoto FTBFS.
    • NCommander to produce fix.
  • asac to discuss the buildd situation with lamont.
    • Discussed, more buildd's now available.

Standing Items

Meeting Outcome

Action Items

  • dyfet to check update-manager sanity checks for ARM upgrades.
  • GrueMaster to determine what bugs that are good candidates to be fixed for lucid release

    • These are the most visible bugs on the live image currently
      • bug 561874 NULL Reference exception in F-Spot
      • bug 559065 ifconfig eth0 down will cause system hang after fec.c driver update
      • bug 558651 gvfs-afc-volume-monitor tries to launch two simultanious events when usb drive inserted
      • bug 558648 Netbook-launcher-efl occasionally displays cryptic error messages
      • bug 528887 maximus does not give default focus to newly started apps in combination with efl launcher
      • bug 539027 end_request: I/O error rebooting at end of install Lucid
      • bug 509006 [dove] hibernation failed to resume
  • NCommander, GrueMaster, plars, ogra to test firefox on new images

  • NCommander to give debdiff to {ogra|asac} for fixing libgphoto FTBFS
  • NCommander to work with plars, GrueMaster and ericm on dove suspend/resume

  • asac/ogra to fix -omap driver to check for /dev/fb0 too if /dev/fb isnt avail
  • persia and NCommander to remove evolution from ubuntu-netbook images
  • asac and JamieBennett to take email client category discussion offline and report back

Minutes

  • GrueMaster indicated that there were 121 outstanding bugs for ARMEL. He committed to producing a prioritized list of bugs so the team can concentrate on the most important ones.

  • The problem with scroll bars in firefox seems to be tool chain releated. asac reported that he had built firefox with -02 and saw the problem go away. This fix needs uploading but will first be tried out in asac's ppa. Several of the team announced that there were willing to test the fix on the various ARMEL images that are produced.
  • Talk moved onto the gphoto FTBFS. Initial thoughts were that it was hardware related with the buildd's in the data center causing problems. After some discussion ogra suggested that it was a packaging problem and that arch all was used to build the -doc package which caused issues on ARMEL. NCommander committed to producing a .debdiff to fix it.
  • asac was pleased to announce that the team now have 9 buildd's back in the data center. For a long time, actual good, physical hardware has been a bottleneck for the team but with the recent changes, the build queue is now empty.
  • The state of work items was next on the agenda. Although the team doesn't have many left, most items now outstanding are to be POSTPONED or re targeted for next cycle.
  • cooloney gave a status update on the iMX51 kernel. The regulator issue (bug 446140) was fixed in the latest kernel and GrueMaster was asked to test it.

  • The subject changed to the Marvell Dove kernel and the long running problem with suspend/resume returned. Although Marvell had reported that the issue was now fixed, the team were still seeing issues. Investigations continue.
  • plars reported that the TI server image was in good shape but the netbook image failed to install at the FlashKernel stage. ogra mentioned that he had produced a solution and that it should be available in the next image.

  • Continuing with the TI image, plars reported that xserver-xorg-video-*omap* didn't work. Again ogra was quick to offer a fix. Talk bounced around on the problem and asac and ogra committed to producing a proper fix.
  • JamieBennett voiced concern that Evolution was still in the images. persia offered to investigate.

  • The classification of the Evolution Email Client replacement, desktop-webmail, was briefly discussed. Being a replacement, JamieBennett thought that it should go in the same, familiar place, as Evolution was, in the Office menu. But, as it is a web based service, it could just as well be put under the Internet menu. asac and JamieBennett decided to talk about it offline. plars mentioned that it should go in favorites too which seems like a good idea.

  • Finally, the broken clock problem in util-linux (bug 563618) was discussed and persia committed to investigate.

Weekly Reports

Jamie Bennett (JamieBennett)

This Week

  • Final bit of webservice-office work. Added startup notification to ensure that on slower hardware, launching the app doesn't appear to do nothing.
  • More research for future projects.
  • UDS background prep.
  • Image testing.

Future

  • More of the same.

Paul Larson (plars)

This Week

  • found a fix for bug #561424 - console switching does not work with ti-omap, built a test kernel and tested it on omap
  • added beta2 testing options to pairwise iso tracker
  • Worked with Micah to get thunderbird segfault resolved
  • worked with desktop team to bring a11y crash to resolution
  • Performed all imx51 pairwise install tests and recorded results on the tracker

Future

  • RC testing should start next week
  • More work on TI image

Alexander Sack (asac)

Coordination/Planning:

  • UDS planning; reaching out for spec submissions in team, discussion of suggested specs for ubuntu on arm track
  • reach out for ofono upstream UDS participation
  • initial discussion of telephony stack options for ubuntu with dyfet
  • review final RC bug list; decide on actions and ensure things are on track
  • take a look at meego progress in git trees
  • cleanup team work items left for lucid
  • release team meeting and preparations
  • investigate new image building approaches for UDS: live-helper appears to be an interesting approach

ARM:

  • finish thumb2 review list for main; fix upx-ucl thumb2 bug, by using -marm
  • land patch in xserver-xorg to auto use omap_drv.so if the driver is installed and we are currently running on an omap platform.
  • finish desktop-webmail spec for lucid; get it uploaded, MIR'd and in the image
    • experimented a bit with maybe next-gen desktop-webmail approach to display webmail webview in a desktop-webmail UI rather than opening browser when clicking on mailto: links.
  • discuss options to not use gnome-panel for netbook-launcher-efl
  • started to investigate focus issue for netbook-launcher-efl; seems to be not a "simple" timestamp issue - lp:528887
  • investigate imx51 freescale enablement/optimization problems

Mozilla/Browser:

  • evaluate firefox rendering issue on armel: found that forcing -O2 in whole mozilla build fixes the scroll bar issue - lp:443147

  • fix most serious issues for ubufox before RC; bake 0.9rc1 release
  • discuss seemingless online/offline networking with community and chromium-browser upstream and presented ntrack as one option
  • help out updating/fixing localized search plugins for firefox
  • coordinate final mozilla items for lucid with mozillateam

Meeting Log

[14:01] <NCommander> #startmeeting
[14:01] <MootBot> Meeting started at 08:01. The chair is NCommander.
[14:01] <MootBot> Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE]
[14:01] <NCommander> If your not here, say I
[14:02] <cooloney> I
[14:02] <cooloney> heh
[14:03] <asac> hey cooloney
[14:03] <NCommander> [link] https://wiki.ubuntu.com/MobileTeam/Meeting/2010/20100413
[14:03] <MootBot> LINK received:  https://wiki.ubuntu.com/MobileTeam/Meeting/2010/20100413
[14:03] <cooloney> asac: hey, man
[14:03] <JamieBennett> NCommander: wrong link
[14:03] <NCommander> james_w: I just noticed you bolded the again on OOo being broken on the last meeting notes. That made me chuckle :-)
[14:04] <NCommander> oh, whoops
[14:04] <NCommander> [link] https://wiki.ubuntu.com/MobileTeam/Meeting/2010/20100420
[14:04] <MootBot> LINK received:  https://wiki.ubuntu.com/MobileTeam/Meeting/2010/20100420
[14:04] <asac> NCommander: i think the actions from last week are wrong
[14:04] <asac> oh ;)
[14:04] <asac> the url was wrong
[14:04] <JamieBennett> :)
[14:04]  * davidm now has coffee and is here
[14:05] <NCommander> davidm: coffee is a good thing. Its also a dependency of the mobile-team package
[14:05] <NCommander> :-)
[14:05] <asac> heh
[14:06]  * JamieBennett kicks the meeting chair to get going ;)
[14:06] <NCommander> so are we all here?
[14:06] <asac> NCommander: how about action items from last meeting?
[14:06] <NCommander> I was making sure we were all here
[14:06] <NCommander> [topic] AI Review
[14:06] <MootBot> New Topic:  AI Review
[14:06] <NCommander> [topic] GrueMaster and plars to continue to keep working with crimsun on ARM PulseAudio
[14:06] <MootBot> New Topic:  GrueMaster and plars to continue to keep working with crimsun on ARM PulseAudio
[14:06] <NCommander> er
[14:06] <NCommander> sorry
[14:06]  * NCommander fixs the page he's looking on
[14:06] <ogra> morning NCommander
[14:06] <ogra> :P
[14:06] <NCommander> [topic] dyfet to check update-manager sanity checks for ARM upgrades.
[14:06] <MootBot> New Topic:  dyfet to check update-manager sanity checks for ARM upgrades.
[14:06] <dyfet> Okay
[14:07] <JamieBennett> the checks were fine?
[14:07] <JamieBennett> no string changed needed?
[14:07] <ogra> surely not
[14:07] <asac> dyfet: ?
[14:07] <dyfet> I have to check :)
[14:07] <asac> can you do that today?
[14:07] <dyfet> I did not realize I had this task until now :).  Yes I can
[14:08] <NCommander> c/o I guess
[14:08] <asac> yes
[14:08] <NCommander> [action] dyfet to check update-manager sanity checks for ARM upgrades.
[14:08] <MootBot> ACTION received:  dyfet to check update-manager sanity checks for ARM upgrades.
[14:08] <plars> on a side note, I did attempt update from karmic
[14:08] <NCommander> [action] GrueMaster to send a list of bugs that need attention to asac.
[14:08] <MootBot> ACTION received:  GrueMaster to send a list of bugs that need attention to asac.
[14:08] <plars> and it worked on imx51 where I tested it at least
[14:08] <plars> but this would not have needed the sanity check
[14:08]  * mvo mumbles something about string freeze
[14:08] <NCommander> plars: it should also work on dove if you shoehorn an X0 kernel onto a karmic installation
[14:08] <ogra> mvo, well, even if we only change the check thats better than nothing
[14:08] <GrueMaster> Shortly after last meeting, plars beat me to the punch with http://people.canonical.com/~plars/dashboard.html.
[14:09] <ogra> mvo, keeipng the old error
[14:09] <mvo> ogra: yeah
[14:09] <asac> GrueMaster: hmm. ok
[14:09] <plars> GrueMaster: not sure that's a replacement.. it's not done, and not on a cron job to update yet
[14:09] <asac> GrueMaster: but thats not a short list
[14:09] <asac> f candidates that might need more attention etc.
[14:09] <asac> anyway. ok
[14:09] <plars> GrueMaster: that's also for a somewhat different purpose
[14:09] <asac> right
[14:10] <GrueMaster> May be for a different purpose, but it generated the same list.
[14:10] <asac> which list is that?
[14:10] <asac> the "Bugs tagged for armel"?
[14:10] <GrueMaster> Bugs that need attention.
[14:10] <GrueMaster> For armel.
[14:10] <NCommander> anything else on this?
[14:11] <asac> GrueMaster: where is that on that dashboard?
[14:11] <GrueMaster> Only to note that none of the bugs directly affecting live images have been looked at.
[14:11] <GrueMaster> Bugs tagged armel 121Â 
[14:11] <asac> htats the full list
[14:11] <asac> we wanted a short list for prioritization ;)
[14:12] <GrueMaster> Ok, I'll whip something together today.
[14:12] <asac> anyway. lets move on. if you see anything that isnt on RC bug list and that is a good candidate for a fix let me know
[14:12] <asac> thanks
[14:13] <asac> NCommander: maybe action on that and move on ;)
[14:13] <NCommander> [action] GrueMaster to priorize the 121 bugs and send them to asac
[14:13] <MootBot> ACTION received:  GrueMaster to priorize the 121 bugs and send them to asac
[14:13] <asac> JamieBennett to add StartupNofify  support to webservice-office-zoho.  -> that happened
[14:13] <JamieBennett> done and in the images :)
[14:13] <NCommander> [topic] JamieBennett to add StartupNofify support to webservice-office-zoho
[14:13] <MootBot> New Topic:  JamieBennett to add StartupNofify support to webservice-office-zoho
[14:13] <NCommander> oh good
[14:13] <NCommander> [topic] asac to work with roc on the firefox scrollbar issue
[14:13] <MootBot> New Topic:  asac to work with roc on the firefox scrollbar issue
[14:13] <asac> i dont want 121 prioritized ;) ... just good candidates for a fix
[14:14] <asac> yes... this turned out to be toolchain
[14:14] <asac> a certain mix of optimization flags seem to trigger it
[14:14] <asac> i dont see it with all -O2 which we will go for
[14:14] <JamieBennett> fixed now?
[14:14] <NCommander> [action] Correction: GrueMaster to determine what bugs that are good candidates to be fixed for lucid release
[14:14] <MootBot> ACTION received:  Correction: GrueMaster to determine what bugs that are good candidates to be fixed for lucid release
[14:14] <ogra> fix in sight ?
[14:14] <asac> JamieBennett: not in archive.
[14:14] <asac> ogra: yes. -O2 ;)
[14:14] <ogra> cool
[14:14] <asac> i will push that to some ppa so we can verify without busting RC images
[14:15] <NCommander> [action] NCommander, GrueMaster, plars to test firefox images
[14:15] <MootBot> ACTION received:  NCommander, GrueMaster, plars to test firefox images
[14:15] <ogra> add me too :)
[14:15] <ogra> for omap ;)
[14:15] <asac> hehe
[14:15] <asac> ogra: does ffox start on beagle?
[14:15] <ogra> sure
[14:15] <asac> good
[14:15] <NCommander> ogra: how many days does it take?
[14:15] <ogra> did even when we still had a live image
[14:15] <NCommander> [action] NCommander, GrueMaster, plars, ogra to test firefox images
[14:15] <dmart> asac: can you let me know the details in the firefox issue later?  Out tools guys may be interested...
[14:15] <MootBot> ACTION received:  NCommander, GrueMaster, plars, ogra to test firefox images
[14:16] <ogra> NCommander, faster than imx51 :P
[14:16] <ogra> (in live image at least)
[14:16] <asac> dmart: lets talk about this after meeting. yes.
[14:16] <NCommander> [topic] NCommander to investigate gphoto FTBFS
[14:16] <MootBot> New Topic:  NCommander to investigate gphoto FTBFS
[14:16] <asac> ogra: wow
[14:16] <NCommander> Short story: our buildds suck
[14:16] <asac> NCommander: so i had this here too and its hanging in dot
[14:16] <ogra> but we dont have live images anymore so i cant show off with that :)
[14:16] <NCommander> Slightly longer story: doxygen hangs on imx51 :-/
[14:16] <asac> from graphviz package
[14:16] <asac> is looping
[14:16] <ogra> NCommander, the package sucks
[14:16] <NCommander> asac: yeah, it doesn't do it on dove though.
[14:17] <ogra> has nothing to do with our buildds
[14:17] <NCommander> ogra: builds fine on dove
[14:17] <persia> The buildds also aren't great.  I can build hdf5 locally, but it fails on buildds.
[14:17] <ogra> its simply building the arch all portion under armel
[14:17] <ogra> which it shouldnt
[14:17] <asac> NCommander: maybe its in libimagemagick
[14:17] <asac> does that use something special?
[14:17] <asac> ;)
[14:17] <ogra> just fix gphoto
[14:18] <asac> ogra: ?
[14:18] <ogra> the packaging is clearly broken
[14:18] <NCommander> ogra: I'm going to work around the issue, to change the build behavior will be an invasive patch. I discussed this with slangasek so I just need to kick a debdiff
[14:18] <ogra> asac, it builds arch all stuff under armel
[14:18] <asac> ogra: generting a graph on the builder hangs
[14:18] <NCommander> ogra: its not the packaging, its the package. configure builds docs if doxygen is detected
[14:18] <ogra> asac, for -doc packages that are arch all
[14:18] <ogra> NCommander, right ...
[14:18] <ogra> for lucid -> quick fix
[14:19] <ogra> for the meerkat fix it *right*
[14:19] <NCommander> ogra: yes, that was my plan
[14:19] <ogra> already uploaded ?
[14:19] <NCommander> ogra: no, haven't fired a test build into a devirtualized PPA yet. Will do today
[14:19] <ogra> (we know that since last wed.)
[14:19] <NCommander> ogra: know what?
[14:20] <ogra> the packaging issue with all vs any
[14:20] <asac> NCommander: i see it hanging here on imx51 on a plain install
[14:20] <asac> why would a devirtualized ppa better?
[14:20] <NCommander> asac: it didn't hang here, but I think I let it build for 20 hours without checking on it :-)
[14:20] <persia> asac: Tests the actual buildd environment, rather than something else.
[14:20] <asac> NCommander: i left it running over a night hanging in the "dot"
[14:21] <asac> persia: well. it hangs here at the same place where it hangs on the builders ;)
[14:21] <NCommander> so I'll finish this today
[14:21] <ogra> thanks :)
[14:21] <asac> ok .ext
[14:21] <asac> next
[14:21] <NCommander> [action] NCommander to give debdiff to {ogra|asac} for fixing libgphoto FTBFS
[14:21] <MootBot> ACTION received:  NCommander to give debdiff to {ogra|asac} for fixing libgphoto FTBFS
[14:21] <asac> buildd situation was discussed
[14:21] <NCommander> [topic] asac to discuss the buildd situation with lamont
[14:21] <MootBot> New Topic:  asac to discuss the buildd situation with lamont
[14:21] <NCommander> asac: results?
[14:21] <ogra> we have them
[14:22] <asac> NCommander: we have 9 builders
[14:22] <asac> err 8 now it seems
[14:22] <asac> anyway ... all bbg3 is my last update on that
[14:22] <ogra> polus 1 livefs builder
[14:22] <ogra> *plus
[14:22] <asac> and the build queue is empty
[14:22] <NCommander> cool
[14:22] <ogra> yeah
[14:22] <NCommander> [topic] Current Items
[14:22] <MootBot> New Topic:  Current Items
[14:23] <NCommander> [topic] http://people.canonical.com/~pitti/workitems/canonical-mobile.html
[14:23] <MootBot> New Topic:  http://people.canonical.com/~pitti/workitems/canonical-mobile.html
[14:23] <ogra> we'Re touching the trendline again ?
[14:23] <asac> so there are a bunch of work items
[14:23] <asac> we should go through and clean them up
[14:23] <asac> we want to be at ZERO
[14:23] <ogra> ++
[14:24] <asac> i am not sure what to do with my firefox tasks
[14:24] <asac> maybe i will throw them over the fence to chris coulson ;)
[14:24] <asac> but i think they might even be done by now
[14:24] <asac> i have to check
[14:24]  * ogra wonders why uboot imx still shows up 
[14:25] <ogra> do i need to set POSTPONED to DROPPED or something ?
[14:25] <asac> ogra: yeah. let me check
[14:25] <asac> ogra: i can take that off the lucid radar completely
[14:25] <asac> but that would make all DONE items disappear
[14:25] <ogra> yeah, do that
[14:25] <asac> let me see
[14:26] <ogra> it would fix the red stuff in the graph
[14:26] <ogra> devicetree seems similar
[14:26] <asac> ogra: i made that now deferred
[14:26] <asac> not sure if that changes a thing
[14:26] <ogra> we'll see in 1h :)
[14:26] <asac> ogra: if it doesnt disappear set the reset to POSTPONE
[14:26] <ogra> it *is* all postponed
[14:27] <asac> persia: there are a couple of itms on that list for you
[14:27] <asac> about desktop-lucid-xorg-enable-century-usb-screen
[14:27] <asac> have never heard of that spec before ;)
[14:27] <ogra> yay for last minute specs :)
[14:27] <GrueMaster> I thought that was hinging on bryce doing something.
[14:28] <persia> These are brand-new for me for now.  They will not be done for lucid.
[14:28] <ogra> yeah, bryce commented in the spec
[14:28] <persia> I don't even have hardware to test that driver, and won't until bryce gives me back my device at UDS.
[14:28] <ogra> all postpone
[14:29] <persia> postpone?  Is that how we cancel stuff entirely?
[14:29] <ogra> "During the sprint I received a Century USB monitor device thingee from Persia which needs to have X support added for it."
[14:29] <ogra> thats the spec description
[14:29] <ogra> so you dont have it anymore ?
[14:30] <persia> No.  Bryce has my hardware.
[14:30] <davidm> I heard from bryce, it would take a lot of work (coding) to make these devices work correctly so he is not doing it.
[14:30] <ogra> ah
[14:30] <ogra> davidm, mine works but needs an xorg.conf
[14:30] <davidm> Perhaps we can chase upstream or do it ourselves at some point.
[14:30] <persia> Yeah, I got it to that stage by packaging upstream.
[14:30] <davidm> Yea, the autodetection is hard
[14:30]  * ogra has a displaylink adapter but no time to do anything with it atm
=== gnomefreak76 is now known as gnomefreak
[14:31] <persia> upstream is mostly inactive, but the xrandr stuff is potentially possible to add.
[14:31] <persia> autodetection is something that nobody should expect from me :)
[14:31] <ogra> salcker !
[14:31] <ogra> :)
[14:31]  * NCommander hides from X
[14:32] <asac> ok. for the workitem parrt of this discussion: ... persia can you confirm with bryce that nothing can be done for release and then postpone and drop a comment?
[14:32] <ogra> it has three TODO items that need to be changed
[14:33] <persia> I've confirmed already.  Is "POSTPONED" correct?  Is there no "CANCELLED" or "DROPPED"?
[14:33]  * ogra isnt sure
[14:33] <NCommander> persia: I think if you just remove the item, it popped off the list
[14:33] <asac> persia: i think DROPPPED was discussed, but wasnt implemented yet
[14:33] <asac> so just POSTPONE
[14:33] <asac> or POSTPONED ... i think both are ok
[14:34] <persia> OK.  POSTPONE it is, but if someone wants it done, they should do it :)
[14:34] <asac> persia: so to summarize: all the items for you on that spec are postpone?
[14:34] <persia> Yes.
[14:34] <asac> ok that keeps us below the trend again ;)
[14:35] <NCommander> \o/
[14:35] <asac> ok ...  i think thats it for work items
[14:35] <NCommander> can I mov eon?
[14:35] <asac> lets move on
[14:35] <asac> yep
[14:36] <persia> Folks should tell me if they want me to do stuff, rather than just updating random whiteboards :)
[14:36] <NCommander> [topic] Kernel Status (cooloney, ericm)
[14:36] <MootBot> New Topic:  Kernel Status (cooloney, ericm)
[14:36] <asac> persia: so the problem was that you didnt know that those were your items ;)?
[14:37] <cooloney> for fsl-imx51 kernel, 2 fec patches were applied. they fixed the fec performance drop issue
[14:37] <ogra> \o/
[14:37] <cooloney> i posted a patch fixing ifdown system hang issue. it was reviewed, need some updates and testing again.
[14:37] <cooloney> i filed a bug #567157 about regulator issue.
[14:37] <ubottu> Launchpad bug 567157 in linux-fsl-imx51 "regulators enabled at boot and also print error messages at boot." [High,In progress] https://launchpad.net/bugs/567157
[14:37] <cooloney> need some test, heh
[14:37]  * ogra thought we had a bug for that already
[14:38] <cooloney> and too bad, my bb2.5 strikes now
[14:38] <cooloney> can not boot
[14:38] <cooloney> ogra: really?
[14:38] <ogra> Bug 446140
[14:38] <ubottu> Launchpad bug 446140 in linux-fsl-imx51 "regulator issues with 2.6.31 vs 2.6.28 with the same patchset" [High,Fix released] https://launchpad.net/bugs/446140
[14:38]  * ogra wonders why thats fix released though
[14:39] <cooloney> yeah, that's why i did not find it
[14:39] <cooloney> it was closed as fix released
[14:40] <asac> ok ... GrueMaster can you test the kernel from cooloney on bbg to verify the regulartor issue?
[14:40] <asac> cooloney: thats what is needed, right?
[14:40] <ogra> cooloney, right, but same issue
[14:40] <GrueMaster> planned on it after the meeting.
[14:41] <asac> ok thanks. i dont think we need an action, do we?
[14:41] <cooloney> asac: right.
[14:41] <asac> ok anything else on imx?
[14:41] <cooloney> asac: no, i am done
[14:41] <asac> ericm: ... any update on dove kernel?
[14:41] <ericm> not much update for mvl-dove, just confirmed with Marvell that suspend/resume works with our recent kernel on their side, but I need to confirm with GrueMaster and plars as well
[14:41] <asac> thanks cooloney
[14:42] <asac> ericm: how does marvell confirm that if you ask them? do they use our image etc.?
[14:42] <plars> ericm: istr one of the kernels I tested for you recently had it working again, I'll confirm with the latest image today
[14:42] <ericm> hibernation/resume is reported to work with explict resume=/dev/sda2 on kernel command line
[14:42] <ogra> erm
[14:42] <ericm> plars, thanks
[14:42] <ogra> we set that in initramfs
[14:42] <ericm> asac, they are supposed to use lucid-beta
[14:42] <ogra> but based on UUID i think
[14:42] <GrueMaster> testing it now on my dove with RC build.
[14:43] <asac> ericm: right. just wonder what that means ;) ... do they use their own kernel etc.?
[14:43] <ericm> ogra, I know - that's why I'm debugging
[14:43] <ericm> asac, nope - they are using our kernel
[14:43] <ogra> ericm, /usr/share/initramfs-tools/scripts/local-premount/resume
[14:43] <asac> very good. so chances are hhigh that this is really fixed, which probably makes a great dove release ;)
[14:43] <asac> thanks ericm
[14:44] <asac> (and NCommander of course )
[14:44] <ericm> ogra, from the log - it seems that restoring from swap is actually OK in both cases
[14:44] <GrueMaster> Unfortunately, it is not working on my dove.
[14:44] <ericm> but weird that with initramfs it failed to resume at the last step
[14:44] <asac> oh ... we want to fix resume= in inintramfs ogra ?
[14:45] <ogra> asac, iirc the installer writes the UUID somewhere and the initramfs script reads that
[14:45] <ericm> GrueMaster, I doubt it's a HW issue as I'm also unable to resume from suspend with my original dove board til I replaced one with Marvell shanghai
[14:45] <ogra> so it might be an issue with dove subarch missing something wrt installation
[14:45] <ericm> ogra, that's /etc/initramfs/conf.d/resume
[14:45] <ericm> /etc/initramfs-tools/conf.d/resume, sorry
[14:45] <ogra>  /etc/initramfs-tools/conf.d/resume
[14:45] <ogra> yeah
[14:45] <ogra> do you have the swap UUID in there in your dove install ?
[14:46] <ericm> ogra, the resuming from swap is actually OK in both cases
[14:46] <ericm> ogra, yes
[14:46] <ogra> ah, k
[14:46]  * ogra doesnt understand the issue then
[14:46] <ericm> ogra, yeah it's weird
[14:46] <ogra> isnt resuming from swpa all we want ?
[14:46] <ogra> *swap
[14:46] <ogra> if that works, whats the issue ?
[14:47] <ericm> ogra, apart from that, the system will still undergo a full resume process and a SoC specific resume routine to restore the PC to what it was before
[14:47] <ogra> so it resumes twice ?
[14:47]  * ogra doesnt understand
[14:48] <ericm> ogra, not necessarily, the first is reboot + restoring ram from swap, the 2nd part is a resume process + restoring of PC
[14:48] <ericm> ogra, we can put it offline maybe
[14:48] <ogra> yeah
[14:48] <ogra> move on
[14:48] <ericm> nothing more, I think we can move on
[14:48] <plars> of the two, I would thinkg suspend/resume is much more important
[14:49] <asac> ok action: ericm and ogra and NCommander to sort out resume on dove ;)
[14:49] <asac> NCommander: ^^
[14:49] <ogra> asac,  i dont have the HW
[14:49] <asac> ack ... suspend/resume is important
[14:49] <ogra> happy to give tips though
[14:49] <plars> and it sounds like suspend/resume may be still broken as well
[14:49] <ogra> where i can
[14:49] <asac> ogra: well, but you have valuable input even without that it seems
[14:49] <NCommander> asac: probably not going to happen for lucid release, might be able to SRU it
[14:49]  * plars will test also today
[14:49] <ericm> plars, thanks
[14:49] <asac> ok NCommander give action and move on ;)
[14:49] <ogra> ++
[14:50] <asac> (10 minutes left)
[14:50] <NCommander> [action] NCommander to work with plars, GrueMaster and ericm on dove suspend/resume
[14:50] <MootBot> ACTION received:  NCommander to work with plars, GrueMaster and ericm on dove suspend/resume
[14:50] <NCommander> [topic] QA Status (GrueMaster, plars)
[14:50] <MootBot> New Topic:  QA Status (GrueMaster, plars)
[14:50]  * ogra listens 
[14:50]  * asac hands mic to GrueMaster 
[14:50] <plars> Looks like GrueMaster cranked through all of the iso images yesterday
[14:50] <GrueMaster> Still a lot of bugs on live/installed image.  Will detail list and email it out today.
[14:51] <asac> GrueMaster: thanks. maybe add that to the meeting wiki too
[14:51] <asac> that gets send to everyone
[14:51] <GrueMaster> Will do.
[14:51] <asac> GrueMaster: at best under the standing item (e.g. not in activity)
[14:51] <plars> I was spending some time with the TI image yesterday, server install seems to work pretty well for me, but had some problems with the netbook image at the end with FlashKernel
[14:51] <plars> so install did not complete
[14:51] <ogra> plars, solved today
[14:52] <ogra> but needs an update of livecd-rootfs on the livefs builder
[14:52] <ogra> not sure when the cronjob runs
[14:52] <plars> also, the xserver-xorg-video-*omap* doesn't seem to work, but I think I saw an email from ogra that may indicate the reason
[14:52] <ogra> so tomorrows or tomorrow+1 image will work
[14:52] <GrueMaster> ogra: I think the cron jobs are shut off during release week.
[14:52] <ogra> plars, yeah, i looked at that today
[14:53] <ogra> plars, sudo ln -s /dev/fb0 /dev/fb should fix it :)
[14:53] <plars> ogra: is it the same problem with both of them?
[14:53] <asac> plars: do you only have /dev/fb0?
[14:53] <ogra> GrueMaster, not the one on the livefs builder that keeps the packages up to date :)
[14:53] <asac> or also fb1?
[14:53] <ogra> asac, both
[14:53] <asac> i want to fix the -omap package to check for those too
[14:53] <ogra> fb1 is used for xv
[14:53] <plars> asac: yes
[14:53] <asac> ogra: do you know what omap board had /dev/fb?
[14:53] <plars> asac: linking to fb does fix it
[14:54] <asac> because thats clearly hard coded as the default in the driver ... while it shouldprobably also ceck for fb0 etc.
[14:54] <ogra> asac, no idea, beagle has fb0 and 1
[14:54] <ogra> asac, i can give you a debdiff with fix for lucid ... just didnt have time to test yet
[14:54] <plars> ogra: actually I even have fb2
[14:54] <NCommander> ogra: dove has fb0, 1, and sometimes 2
[14:54] <asac> ACTION: asac/ogra to fix -omap driver to check for /dev/fb0 too if /dev/fb isnt avail
[14:54] <ogra> right, the driver uses fb0 and 1
[14:54] <asac> can we have a bug?
[14:54] <ogra> but has /dev/fb hardcoded
[14:55] <plars> asac: was just working on that before the meeting started
[14:55] <ogra> i'll file one after meeting
[14:55] <ogra> unless plars beats me to it :)
[14:55] <asac> ok ... let me know ... and attach patch; if its as i wanted it that means its probably right ;)
[14:55] <asac> ogra: so you alreadcy have the patch?
[14:55] <asac> otherwise i can also do it.
[14:55] <ogra> #define DEFAULT_DEVICE "/dev/fb"
[14:55] <asac> well
[14:56] <asac> i would prefer to check fb0 on top
[14:56] <asac> ;)
[14:56] <asac> rather than just changeing DEFAULT_DEVICE
[14:56] <ogra> well, lool had an idea to change the code completely
[14:56] <ogra> but thats rather intrusive
[14:56] <asac> I would suspect it was added there for some reason ;) ... and we could avoid breaking that use case
[14:56] <asac> i will look then
[14:56] <asac> i already looked. didnt feel that intrusive
[14:57] <ogra> using fbdev_open() and fbdevHWInit()
[14:57] <ogra> see xorg-server/hw/xfree86/fbdevhw/fbdevhw.c
[14:57] <asac> yes. i will see if that makes sense
[14:57] <asac> otherwise we can fall back to plain hacky approach ;)
[14:57] <ogra> yeah
[14:57] <lool> ogra: "change the code completely" indeed sounds intrusive...
[14:57] <asac> ok moving on ...
[14:58] <asac> "ARM Application status (JamieBennett, dyfet) "
[14:58] <plars> we are about out of time
[14:58] <ogra> lool, adding code to use these functions vs a one char change :)
[14:58] <asac> yes.
[14:58] <JamieBennett> Evolution still in the image :(
[14:58] <asac> just quick:
[14:58] <asac> persia: can you please take the action to get evolution of that image?
[14:58] <ogra> NCommander, proposed to check the deps
[14:58] <asac> you basically volunteered to do that ;)
[14:58] <NCommander> [topic] sudo landscape-config --url https://staging.landscape.canonical.com/message-system --ping-url http://staging.landscape.canonical.com/ping
[14:58] <MootBot> New Topic:  sudo landscape-config --url https://staging.landscape.canonical.com/message-system --ping-url http://staging.landscape.canonical.com/ping
[14:58] <NCommander> er
[14:58]  * NCommander coughs
[14:58] <ogra> huh ?
[14:58] <NCommander> [topic] ARM Application status (JamieBennett, dyfet)
[14:58] <MootBot> New Topic:  ARM Application status (JamieBennett, dyfet)
[14:58] <NCommander> C*P fail
[14:59]  * ogra wonders if there is a meeting after us
[14:59] <asac> NCommander: right. see action for persia above
[14:59] <JamieBennett> not much else apart from desktop-webmail needs moving to Office category
[14:59] <asac> JamieBennett: i dont think it has to
[14:59] <plars> JamieBennett: it should probably go in favorites as well
[14:59] <asac> other mailers are in internet too
[14:59] <NCommander> [action] persia and NCommander to remove evolution from ubuntu-netbook images
[14:59] <MootBot> ACTION received:  persia and NCommander to remove evolution from ubuntu-netbook images
[14:59] <plars> JamieBennett: especially if it's going to be replacing evo on that image
[14:59] <asac> thanks
[14:59] <JamieBennett> asac: its where people will go looking for email since they are used to evolution being there
[15:00] <asac> well. debatable ;) ... all tbird users (which are 50%) go for internet
[15:00] <asac> lets take this offline
[15:00] <asac> image status?
[15:00] <JamieBennett> asac: OK, not a big deal
[15:00] <asac> NCommander: ^
[15:00] <NCommander> [action] asac and JamieBennett to take email client category discussion offline and report back
[15:00] <MootBot> ACTION received:  asac and JamieBennett to take email client category discussion offline and report back
[15:00] <ogra> images seem to work ok though bug 563618 seems to hit JamieBennett on imx51 now too, omap misses a working netbook image (the others work, netbook should be ready tomorrow as mentioned above)
[15:00] <ubottu> Launchpad bug 563618 in util-linux "Ignoring a broken clock results in infinite reboots; not ignoring results in fsck failure; no solution to this problem" [High,Triaged] https://launchpad.net/bugs/563618
[15:01]  * ogra points to https://wiki.ubuntu.com/ARM/BeagleNetInstall
[15:01] <asac> sigh
[15:01] <ogra> i'll create the same for server ... and netbook
[15:01] <asac> JamieBennett: i think you can hit "M" key when its looping
[15:01] <asac> to go to maintenance shell
[15:01] <JamieBennett> ogra: after setting the clock correctly it still hangs for me
[15:01] <ogra> slangasek told me he would take a look at the approach but i think he is to busy atm
[15:01] <JamieBennett> (on next boot)
[15:02] <JamieBennett> (with power still applied)
[15:02] <persia> Sorry.  Laptop hung.  I'm happy with my action.
[15:02] <ogra> JamieBennett, did you edit the two fstab files ?
[15:02] <asac> persia: thx
[15:02] <persia> omap fb may be complicated: there's a known workaround in case it is.
[15:02] <JamieBennett> ogra: not yet, will after the meeting
[15:02] <asac> ok lets take that offline
[15:02] <asac> i dont think we will find a solution during meeing for this bad bug
[15:02] <ogra> NCommander, AOB ... move !
[15:02] <NCommander> [action] persia and asac to take omap fb issue offline and report back
[15:02] <MootBot> ACTION received:  persia and asac to take omap fb issue offline and report back
[15:02] <NCommander> [topic] AOB
[15:02] <MootBot> New Topic:  AOB
[15:03] <asac> NCommander: omap fb?
[15:03] <asac> that was about  bug 563618
[15:03] <ogra> heh
[15:03] <ubottu> Launchpad bug 563618 in util-linux "Ignoring a broken clock results in infinite reboots; not ignoring results in fsck failure; no solution to this problem" [High,Triaged] https://launchpad.net/bugs/563618
[15:03] <asac> and didnt involve me ;)
[15:03] <asac> but no action neeeded imo
[15:03] <NCommander> [action] persia to take omap fb issue offline and report back
[15:03] <asac> so AOB?
[15:03] <ogra> no
[15:03] <MootBot> ACTION received:  persia to take omap fb issue offline and report back
[15:03] <ogra> no
[15:03] <persia> No
[15:03] <ogra> lol
[15:03] <asac> omg
[15:03] <asac> ;)
[15:03] <NCommander> [topic] ARM Image Status (ogra, persia)
[15:03] <MootBot> New Topic:  ARM Image Status (ogra, persia)
[15:03] <plars> omap fb bug is bug #567260
[15:03] <ubottu> Launchpad bug 567260 in xf86-video-omapfb "xserver-xorg-video-omap* fail due to no /dev/fb" [Undecided,New] https://launchpad.net/bugs/567260
[15:03] <ogra> NCommander, we had that
[15:03] <NCommander> right
[15:03] <NCommander> soryr
[15:04] <NCommander> [topic] AOB
[15:04] <MootBot> New Topic:  AOB
[15:04] <asac> ok lets wrap up ;)
[15:04] <ogra> we had AOB too already ... you just got confused
[15:04] <ogra> #endmeeting :)
[15:04] <asac> anything else?
[15:04] <asac> ok thanks!!
[15:04] <NCommander> #endmeeting

ARM/Meeting/2010/20100420 (last edited 2011-07-28 17:59:01 by davidm)