15:00:03 #startmeeting 15:00:03 Meeting started Thu Sep 29 15:00:03 2011 UTC. The chair is NCommander. Information about MeetBot at http://wiki.ubuntu.com/AlanBell/mootbot. 15:00:03 15:00:03 Available commands: #accept #accepted #action #agree #agreed #chair #commands #endmeeting #endvote #halp #help #idea #info #link #lurk #meetingname #meetingtopic #nick #progress #rejected #replay #restrictlogs #save #startmeeting #subtopic #topic #unchair #undo #unlurk #vote #voters #votesrequired 15:00:52 * janimo waves 15:00:53 [link] https://wiki.ubuntu.com/ARM/Meeting/2011/20110929 15:01:09 there are no outstanding action items 15:01:17 [link] http://people.canonical.com/~platform/workitems/oneiric/ubuntu-arm.html 15:01:31 [link] http://people.canonical.com/~platform/workitems/oneiric/ubuntu-arm-ubuntu-11.10-beta-2.html 15:01:43 that doersnt make any sense :) 15:01:46 B2 is gone 15:01:54 argh 15:02:10 ENEEDCAFFIENE 15:02:21 :) 15:02:50 the burndown chart looks good, we completed pretty much all major work items this cycle 15:03:06 [action] Everyone to postpone any remaining work items 15:03:06 * meetingology Everyone to postpone any remaining work items 15:03:18 yeah, doko trashes our statisticts 15:03:28 i think the postponing happened already 15:03:40 i only see three Wi left 15:03:51 two from doko, one assigned to the team ? 15:04:14 why the heck is that assigned to the team 15:04:17 quite a few of my WIs are testing/research which are not affected by a particular release date 15:04:25 I need to postpone the lava work. Seems the system was reengineered and will require me to rethink how I deploy it (again). 15:04:33 janimo, didnt you add the 4M boundary stuff already ? 15:04:52 I just postedponed my one oustanding work item 15:05:02 ogra_, add it where? I think that was done first in Oneiric 15:05:16 janimo, yes, me too, thats why i ask you :) 15:05:25 there is an open WI for it 15:05:41 align rootfs to 4M boundary on OMAP preinstalled images (debian-cd) 15:05:43 GrueMaster, yes, I had two lava WIs . One took too long for little benefit, the other I just postponed. No more yak-shaving for me 15:05:48 Pretty sure I merged that in debian-cd for janimo months ago. 15:05:52 i think i merged code from you that had this 15:06:00 or reviwed 15:06:02 heh 15:06:07 *reviewed 15:06:08 ogra_, so it is done, not sure where it is to be added. You mean it does not show up in the tracker? 15:06:22 janimo, it does, but not assigned to you 15:06:24 that may be because silly syntax in the whiteboard then 15:06:32 ah. weird. Anyway it is done :) 15:06:37 * ogra_ will fix that post meeting, i just wanted to be sure it happened 15:06:51 anything else here or can I move? 15:06:56 move 15:07:04 it happened. OTOH I saw no perceptible speed gain in booting panda with that partition alignment 15:07:07 [topic] ARM Server Status (NCommander, Daviey) 15:07:32 ogra_: I'll mark the bug as fix released. 15:07:51 yeah, thx 15:08:10 with the exception of guided partitioning (will cover in image status), I'm very happy with the state of our server images and feelthey are ready for release 15:09:10 anyone got anything else? 15:09:22 Daviey, ?? 15:09:46 doesnt look like 15:09:56 [topic]Kernel Status (cooloney, ppisati) 15:10:19 audio !!! 15:10:20 a new omap4 kernel has been pushed yesterday 15:10:39 with 4460 support and some other goodies 15:10:46 today the audio fix should go in 15:10:51 \o/ 15:11:00 ppisati, also support for pvr is in? 15:11:13 or rather did not regress 15:11:16 janimo, i think rsalveti cared for that 15:11:20 ok 15:11:21 in the tree we merged 15:11:24 janimo: the new gfx driver you mean? 15:11:27 * rsalveti waves 15:11:30 not that I have an omap4 to test on but ... 15:11:32 ppisati, yes 15:11:58 should be a separate pkg, but the kernel we pulled had a couple of reverts to accomodate it 15:12:01 janimo: in theory it should work with the driver already available at the ppa 15:12:04 just didn't test it yet 15:12:07 ok 15:12:22 would be good to know before release if it works ;) 15:12:39 but thats still weeks :) 15:13:11 janimo, what about ac100 ? 15:13:21 it works 15:13:22 do you plan another upload with the quietening patch ? 15:13:31 or do we keep that for post release 15:13:40 if there's any more fixes needed they should be batched in an upload 15:13:51 ogra_, what does it quiet exactly? nvec? 15:13:53 k, i agree 15:14:03 janimo, the horrtible niosy boot 15:14:21 ogra_, I'd rather have it before release. Also enable MD/DM as those were dropped vs your package 15:14:36 janimo, k, then go ahead :) 15:14:44 ogra_, hmm, nosiy boot, isn't that taken care of by quiet on the cmdline? 15:14:50 ok, we'll discuss it on ac100 15:14:51 yes 15:14:58 but it wasnt acxtually quiet 15:15:08 marvins tree has the fix already 15:15:10 * janimo needs to check. 15:15:22 * ogra_ is done 15:15:32 ok, but he also has crazy zcache stuff, so I may need to cherry pick . We'll see 15:15:43 ppisati: and remember that to have proper 4460 support we still need to fix bug 851974 15:15:45 Launchpad bug 851974 in u-boot-linaro (Ubuntu) "FFe: u-boot-linaro 11.09, misc fixes for new silicon and boards" [Undecided,New] https://launchpad.net/bugs/851974 15:15:48 zcache needs to be explicitly enabled on kernel cmdline 15:15:53 dont bother 15:16:15 rsalveti: is uboot going to be updated? 15:16:25 i mean, the pkg in the archive 15:16:26 ppisati: it needs to be updated :-) 15:16:27 rsalveti, jcrigby is already working on the upload (i acked the bug) 15:16:28 ok 15:16:59 ogra_: great, thanks 15:17:09 does he need a sponsor ? 15:17:14 * ogra_ forgot to ask 15:17:16 nops 15:17:19 k 15:17:30 will include the release team then 15:18:56 NCommander, ? move ? 15:19:06 [topic] ARM Porting/FTBFS status (NCommander, janimo) 15:19:18 nothing to report 15:19:25 currently looking at ido ftbfs 15:20:04 So the guided partition on omap isa bit of a problem. Specifically, you can't set a device in a partman reciepe. The documentation is misleading since the device{} tag only works for LVM 15:20:23 topic ? 15:20:40 This doesn't fit into ARM porting? 15:20:57 i thought you wanted to save it for image :) 15:20:59 anyway 15:21:07 * ogra_ throws bug 857299 in the room 15:21:08 Launchpad bug 857299 in banshee (Ubuntu) "banshee window remain white on startup on pandaboard" [Undecided,New] https://launchpad.net/bugs/857299 15:21:27 i fear that needs serious caretaking 15:21:37 and it happens on ac100 too 15:21:40 ah, banshee again 15:22:13 That being said, partman-uboot is done, and works extremely well, so it adds sanity checking. I'm unsure though if I should upload it this late (although it will properly add warnings to an unbootable configuration) 15:22:47 NCommander, whats its status now (i.e. the package we have in the archive) ? 15:22:57 i assume its worse ? 15:23:58 we have nothing in the archive, I was holding off uploading. Uploading partman-uboot will get the netboot installer to do error checking, but I don't know how to force the installer straight to the manual partitioning screen 15:24:15 as such there is no error or sanity checking during partitioning on OMAP 15:24:31 partman-uboot isnt there from the dove times ? 15:24:40 oh, no its there 15:24:48 Theres just no OMAP code in it as of writing 15:24:50 and its unusable atm 15:25:00 right and we dont support dove 15:25:05 upload away ! 15:25:07 right the current partman-uboot does nothing on OMAP. The new version fixes it 15:25:11 k 15:25:19 (actually, I left dove support in incase it ever comes back :-)) 15:25:21 uploading now 15:25:29 awesome 15:25:30 [topic] ARM Image Status (ogra, NCommander) 15:25:48 well, archive is skewed 15:25:54 * NCommander headthunks 15:25:57 up to yesterday they built fine 15:26:01 mx53 boot and installer is very slow 15:26:07 hey, x86 skew this time 15:26:09 :) 15:26:12 slower than I remember the omap3 to be 15:27:05 how is the system after install ? 15:27:18 i think we can live with a slow install if the system itself is ok 15:28:08 (and with a slow boot fwiw) 15:28:46 ogra_, it worked apart from trying to start ubiquity again. Need to see again with oem-config out 15:28:48 * GrueMaster proposes going back to live/installer images for P. 15:28:52 this was yesterday's image 15:28:58 GrueMaster, +1 15:29:09 GrueMaster, make the install faster than 2h and we can talk 15:29:14 if that measn installers like x86 ubuntu uses 15:29:29 it also means extra hours for testing 15:29:32 ogra_, something worth at least investigating. spec? 15:29:39 and we will likely have more arches next round again 15:29:47 janimo, definitely 15:30:01 ogra_, only if we keep preinstalled. Which is a misnomer as it still asks questions and takes time 15:30:18 * ogra_ is all for not diverging (though preinstalled fulfills a need as well) 15:30:25 The point is we need some way to get off of SD if the system has a better medium available (USB, SATA, etc). 15:30:35 We just need to sort out how to add in-place/resize as an option to d-i. 15:30:37 agreed 15:30:38 unlike linaros which is really a dd-able preinstall, albeit without the luxury of picking name for user 1000 15:30:40 I think I may have a spec for that. 15:30:44 We'll spec it for UDS 15:30:52 yeah, there is a d-i spec for that 15:31:23 infinity, did you solve the slideshow issue ? 15:31:37 * ogra_ saw all the other fixes but cant remember that one 15:31:41 ogra_: Fix committed to ubiquity. Was an upload done while I was asleep? 15:31:50 i dont think so 15:31:54 thanks ! 15:33:22 NCommander, move ? 15:33:32 [topic] QA Status (GrueMaster, mahmoh) 15:34:20 With the current state of the daily images, I really shouldn't comment. 15:34:35 So much is broken (at least for me). 15:34:51 Server images are fine. Desktop.... 15:35:46 desktop up to date from a week ago works relatively fine here on the ac100 15:36:30 Where to start. yesterday's daily omap4. Oem-config recycles instead of finishing install. One indicator was flash-kernel not syncing befor umount. 15:36:42 Changed flash-kernel, but it doesn't appear to be the only issue. 15:37:05 I can't reproduce that looping business. :/ 15:37:06 well, there are a bunch of fixes in the archive already but we didnt get images yet 15:37:13 (I could reproduce the f-k thing, and it's fixed) 15:37:29 Enabled debugging in oem-config only increases the useless messaging by 10-20x. Still no failure indicator, except now it crashes. 15:37:35 the looping wont happen if oem-config is uninstalled at the right timwe 15:37:40 so thats a moot bug 15:37:56 So why isn't it uninstalling? 15:37:57 since the deinstallation shoudl be fixed now 15:38:05 preseed issue 15:38:21 But there was no indication. 15:38:30 Nothing in recent logs. 15:38:32 inifnity fixed it 15:38:39 in live-build i think 15:39:15 If that was the oem-config-gtk, it is in yesterday's image. 15:39:21 Yeah. 15:39:33 But you won't get to oem-config-remove if oem-config is crashing for you. 15:39:36 was that supposed to be fixed in yesterdays ? 15:39:39 So, that might still need investigation. 15:39:43 oh 15:39:45 indeed 15:40:05 And I still don't know why oem-config is crashing. 15:40:20 I have looked through every log. 15:41:38 anything else? 15:41:47 not from me 15:42:00 At any rate, other issues I have been seeing are sound (analog audio device renamed from SDP4430 to Panda in the kernel without alsaucm conf files being updated). And I haven't been able to get much further in daily testing. 15:42:24 Nothing else here. 15:42:40 well, and dont expect images for the next two days given the upload rate 15:43:01 I don't expect images until Tuesday, earliest. 15:43:06 heh 15:43:23 i guess sat or sun images should work again 15:43:39 * GrueMaster won't hold his breath. 15:44:03 indeed and i didnt mean to ask you to work on sat or sun :) 15:44:41 If they are there, I may give them a cursory glance over morning coffee. 15:45:06 GrueMaster: The audio device name was reverted. 15:45:20 s/was/is just being/ 15:45:40 infinity: I know. I am just reporting it for the meeting minutes. 15:45:43 I'll stick with was, I accpted that last night! 15:45:53 I am actively in that discussion even now during the meeting. 15:46:01 Heh. 15:46:21 infinity, well, reading the #ubuntu-arm backlog it is just being changed 15:46:25 And until it is in an image, the issue remains open. 15:46:53 anyway 15:46:56 NCommander, move :) 15:47:08 [topic] Linaro Updates (rsalveti) 15:47:34 Oh, nevermind, the revert last night was DRM names. I guess we're changing drivers all over the place. :P 15:47:34 * rsalveti waves 15:48:09 guess the main things we have to report are kind at ubuntu already :-) 15:48:19 the work on the tilt tree, to make it work with 4460 15:48:32 new u-boot-linaro release, fixing 4430 ES2.0 and bringing support for 4460 15:48:42 libjpeg-turbo is now at the archive 15:49:15 other than that we have a release today 15:49:27 and plan to start moving to oneiric next week already 15:49:51 that's most from my side 15:50:32 rsalveti: What changes to u-boot are there on the beagle side? Is it just to sync the source tree? 15:50:56 GrueMaster: boot timeout is now 2 secs 15:51:07 and generally just code sync with upstream 15:51:09 Ah, good. 15:51:27 10 seconds was a bit excessive. 15:51:43 anything else, we're almost out of time 15:52:36 [topic] AOB 15:52:55 * ogra_ has something 15:53:18 since we have our own ARM track at UDS again now ... and will have to give a summaryx at the end again ... 15:53:57 which always costed us half the day to collect on friday ... davidm thought it would be good if we had a closing meeting every day after the last session 15:53:59 I'll be working at putting together a draft of all server blueprints based off the PBLueprintIdeas Page 15:54:21 so everyone will have to write down stuff about the sessions he was in at end of the day 15:54:35 tahts all ... just wanted to announce that 15:54:40 ok 15:54:55 closing the meeting out in 3 15:54:57 2 15:54:59 1 15:55:01 #endmeeting