[01:50] lool, ping [01:55] horaceli: Now it's about 3AM for him, I don't think he's around [01:59] thanks, agoliveira, :-). === mjg59` is now known as mjg59 === asac_ is now known as asac === doko_ is now known as doko [07:30] horaceli: pong [07:31] hi, lool [07:31] good to see you [07:31] Same here; how are you doing? [07:31] I am good. thanks [07:31] Probably quite late at your place now [07:31] how are you? [07:31] Fine [07:31] not really, it is 3:13pm my time. [07:32] Oh ok [07:32] bfiller_ reported a bug on hildon-desktop-0.0.43 that home layout has some problem in 800x480 resolution [07:33] and I fixed in hildon-desktop, mainly for reading current screen resolution to determine the home layout size, it might need to be integrated to hildon-desktop [07:33] I have generated the patch, and will share it out soon. [07:33] Ok; bfiller_ has another problem which I couldn't reproduce until now [07:33] #2? [07:34] the 2nd description under the same bug? [07:35] I'm checking [07:35] It's in #172426 [07:35] Ah right, it's the same bug id you worked on [07:35] It's a bad idea to have multiple bugs described in the same report [07:36] horaceli: Anyway, how can I help you? [07:36] lool, here is the patch: http://moblin.org/repos/users/horace.li/hildon-desktop-fix-bug-172426.patch [07:36] and it is generated based on latest hildon-desktop-ubuntu [07:36] rev413 [07:36] I guess [07:37] Okay [07:37] thanks. :-) [07:37] any problem, just feel free to let me know [07:37] horaceli: Did you send it upstream? [07:38] not yet. [07:38] lool: I'm going to upload a new m-i-c to the PPA tonight, after I test things [07:38] should I? [07:38] horaceli: I like knowing the opinion of upstream when we merge things; they know the software well [07:38] that is a good point. [07:39] StevenK: Any particular issue I should be aware of? [07:39] lool: You added the PPA without adding -updates and -security. [07:39] StevenK: They didn't merge all changes at moblin, so it needs to be modified before upload if you take a new snapshot [07:40] I'm not going to take a new snapshot. [07:40] StevenK: Ok; I do know about this issue, didn't come to it yet [07:40] lool: Ah. Looks like I'll beat you to it. [07:41] * StevenK is testing image builds. [07:41] good morning [07:42] doko: is there a way to use the moblin lpia binary packages at the same time with using the i386? [07:43] doko: i cannot get the moblin-media package installed because it depends on moko which is only lpia [07:45] theseinfeld: well, you could do so with dpkg --force-architecture, but why is moko only built on lpia? [07:46] horaceli: Thanks for the fix BTW [07:47] doko: if I use the apt-get upgrade it sais: [07:47] doko: moblin-media kept back...blah blah moko which is a virtual package [07:49] doko: I even went further and did the apt.conf.d with APT::Architecture "lpia" :)) to see if I can get it [07:50] doko: but, is there a way that we can use both i386 and lpia moblin packages? or for every moblin.org lpia.deb we need to make a i386 package? [07:50] no [07:50] just build for i386 [07:51] doko, so you mean that everything that has lpia by moblin has to be redone :) great redundancy... === cprov is now known as cprov-out === cprov-out is now known as cprov === dholbach_ is now known as dholbach === cprov is now known as cprov-lunch === cprov-lunch is now known as cprov === robr__ is now known as robr === cpro1 is now known as cprov [16:57] HappyCamp_laptop: you got FIOS, sweet [16:58] Doesn't everone? ;) [16:58] I see comcast every once in a while [16:58] poor suckers [16:59] About to start the meeing [16:59] #startmeeting [16:59] Meeting started at 16:59. The chair is davidm. [16:59] Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE] [16:59] * lool waves [16:59] Good day everyone is everyone here? [17:00] Missing Don I see [17:01] here [17:01] Can anyone from Intel ping Don Johnson? [17:02] Anyway first topic: [17:02] [topic] lool to take a look at bug # 172426 and see if he can reproduce it. [17:02] New Topic: lool to take a look at bug # 172426 and see if he can reproduce it. [17:03] lool, status? [17:03] Didn't look [17:04] But need to merge patch by Horace which might fix it [17:04] So can you please carry on "lool to merge patch by Horace as an attempt to fix 172426" [17:04] Will do [17:04] good morning all [17:05] [action] carry over lool to take a look at bug # 172426 and see if he can reproduce it. until Dec 20th. [17:05] ACTION received: carry over lool to take a look at bug # 172426 and see if he can reproduce it. until Dec 20th. [17:05] I spoke to Don, he will not be joining today [17:05] next topic [17:05] [topic] bspencer & Don_Johnson to clarify the process and deliverables [17:05] New Topic: bspencer & Don_Johnson to clarify the process and deliverables [17:06] I know that Don has a process meeting set up for tomorrow so I think this is being done. [17:06] bspenser is not around so I don't know if he has anymore to say on the topic. [17:07] So I'm going to head to the next topic unless anyone has anything else on this? [17:07] [topic] bspencer & horaceli to verify the ubuntu branch works [17:07] New Topic: bspencer & horaceli to verify the ubuntu branch works [17:07] horaceli, any input on this? [17:08] looks like horaceli is not around either. [17:09] I'll carry it over again. [17:09] [action] # [17:09] continue to hold open: bspencer & horaceli to verify the ubuntu branch works [17:09] # [17:09] ACTION received: # [17:09] continue to hold open: bspencer & horaceli to verify the ubuntu branch works [17:09] [action] continue to hold open: bspencer & horaceli to verify the ubuntu branch works [17:09] ACTION received: continue to hold open: bspencer & horaceli to verify the ubuntu branch works [17:09] Horace is in PRC, he is most likely asleep [17:10] [topic] smagoun Since my group needs it, I'll take an action item to test overriding exa for a gutsy psb backport. In progress. [17:10] New Topic: smagoun Since my group needs it, I'll take an action item to test overriding exa for a gutsy psb backport. In progress. [17:10] We have a new drop of graphics drivers from Intel for our customer, I've moved on to working with those. I'll post things to the gutsy PPA if/when appropriate. We can close this issue. [17:10] OK, thanks smagoun it's closed then [17:10] mawhalen, thanks for the input on horaceli [17:11] [topic] amitk to update hardy ppa with drivers [17:11] New Topic: amitk to update hardy ppa with drivers [17:11] amitk, are you around? [17:12] I know he is working on that, I'll have to ping him off line, he is in India until the new year. [17:12] [action] davidm to ping amitk for status on # [17:12] update hardy ppa with drivers [17:12] # [17:12] ACTION received: davidm to ping amitk for status on # [17:13] [topic] davidm to query build team on best way to monitor the i386 builds and the LPIA builds to stay in sync [17:13] New Topic: davidm to query build team on best way to monitor the i386 builds and the LPIA builds to stay in sync [17:13] I'm still working this issue, Tollef and I will work this out. [17:13] I'm going to carry it over [17:14] [action] carry over, davidm to query build team on best way to monitor the i386 builds and the LPIA builds to stay in sync. [17:14] ACTION received: carry over, davidm to query build team on best way to monitor the i386 builds and the LPIA builds to stay in sync. [17:14] [topic] Michael Frey to post " Discuss reasons behind the decision to use acpid, custom moblin battery applet instead of using gnome-power-manager and its status bar applet" question to the list. [17:14] New Topic: Michael Frey to post " Discuss reasons behind the decision to use acpid, custom moblin battery applet instead of using gnome-power-manager and its status bar applet" question to the list. [17:14] you can close this -- I had a discussion with ToddBrandt [17:15] OK, good enough [17:15] [topic] agoliveira to report next week any findings on boot speed. [17:15] davidm: sorry, I got distracted [17:15] New Topic: agoliveira to report next week any findings on boot speed. [17:15] amitk, I'll reopen the topic next then [17:15] davidm: sorry - just lost network connection here. Did you discuss monitoring LPIA builds? [17:15] Yes, I sent an email about the first findings and I had some feedback. Nothing conclusive thought but 2 items: [17:16] smagoun_, Yes, I'm working that will Tollef today and tomorrow, it's on my list. [17:16] 1) The Q1 lost a good deal of time detecting and loading the modules: a brick kernel might help here but would require to create one for each device. [17:17] 2) Matt noticed that the booting process is very CPU bound. We could try to reduce this usiing an all ext3 booting scheme. [17:18] Apart from that I tried to manually simplify the booting scripts but and remove usplash didn't get more than 2 or 3 seconds. [17:18] agoliveira: do you have any concrete recommendations? [17:19] smagoun_: I don't see any real easy way to really reduce boot time. One thing would be do the all ext3 boot and see what happens. [17:19] all ext3 will conflict with the disk footprint requirements, but should be tested [17:19] patm: That's the problem. [17:20] Seems we need that tested now for two issues, boot speed and flash corruption [17:20] yes [17:20] smagoun_: didn't you find a problem with the console keyboard mapping taking many seconds [17:20] davidm: I don't see flash corruption as an issue. They are much more robust those days. [17:20] agoliveira: did you actually try a kernel with all modules compiled in, disabling udev, etc? [17:21] ChickenCutlass: I didn't see that. Did you noticed anything on the graph I sent? [17:21] smagoun_: No, didn't have time to try that. [17:21] ChickenCutlass: yes. There's a bug somewhere in the acton project in LP. Setting the console font is done twice, it takes 4sec each time [17:21] smagoun_: agoliveira that is 8 seconds -- because it is done twice on boot [17:21] agoliveira: davidm we still see disk corruption on occasion using unionfs [17:22] If the Q1 is the reference platform, we should focus on optimizing ext3 boots; optimizing squashfs boots would be a different job [17:22] Q1 is not the reference platform [17:22] ChickenCutlass: I'll have to check it out as I doin't remember that and I'm quite sure I would have noticed. [17:22] Ok :) [17:22] lool, menlow is the only target that counts [17:22] patm: Unfortunately that's what I have to use as a test base. [17:22] I understand [17:23] That's what IO have to report for now. [17:23] s/O// [17:23] patm, is there any chance your team will be testing the ext3 partition sometime soon? [17:23] patm: agoliveira many optimizations will benefit both squashfs and ext3-based systems [17:23] agoliveira: The bug is #141261 in LP [17:23] patm: It's not very interesting to look at Q1 boot speeds; I understood that menlow had an even slower CPU and had storage constraints; it's quite different :-/ [17:24] davidm: I tried it at one point. It booted, which was nice. Didn't have time beyond that, we're timesharing our CB among 6 developers [17:24] lool, it is different but I assume it shares some issues [17:24] Well, in this case, maybe someone with menlow should take care of that. [17:24] or send me one [17:25] what smagoun_ said earlier [17:25] smagoun_: Ouch [17:25] lool: I disagree. Q1 optimizations will help the Menlow systems too [17:25] smagoun_, is there any chance of leaving the ext3 build as the default for a while and see if the corruption goes away and we can get a quick look at boot speed too? [17:26] i'm with steve...the optimizations may be different, but the technique(s) used could be the same [17:26] smagoun_: Depends if we trade disk against CPU; the Q1 has more of both [17:26] We could well see improvements on the Q1 which are really killing the menlow or vice-versa [17:26] davidm: yes, that's reasonable. We have a tight schedule between now and the end of the year, but I'll see what I can do. [17:26] my original point is that we ultimately need to do timing on the menlow [17:27] +1 for patm [17:27] lool: actually, the Q1 has at least a lower clock speed than the menlow systems. [17:27] Mithrandir: I thought the CPU was slower on menlow [17:27] speed starts at 800 Mhz [17:27] which is the q1 clock [17:28] I didn't hear about clock speeds until now, but I recall I was told it was even slower due to a more energy efficient architecture [17:28] lool: The best way to make a system go faster is do less. In nearly every case that will benefit both platforms equally. [17:28] smagoun_: I'm not saying there are no changes that would benefit both; I'm saying we should bench the target and not infer optimizations from a non-target system [17:28] As the boot process is very cpu bound looks like the first shot would be an all ext3 boot [17:29] if you speed up ext3 how do you meet the storage footprint size? [17:30] robr2: One can't have it all :) [17:30] robr2, if ext3 makes a big difference then at least we know that squashfs is a part of the problem. [17:30] robr2: If we use ext3, we don't meet the disk footprint targets. [17:31] smagoun_: are you sure? [17:31] I think that we can squeeze that a bit by optimizing the installed packages. There must be room to manuver there. [17:31] the ask is for <500MB [17:31] at one point we did compare booting ext3 to squashfs/unionfs and there was a very small difference between the two -- but we also didn't optimize the boot for ext3 just swapped one for the other [17:32] smagoun_, right now we have two issues that need to be debugged, boot speed and flash corruption, ext3 testing will help with both. [17:32] Mithrandir: for our customer we're in danger of missing those targets even with squashfs right now :( [17:32] < 500MB not including OOo [17:32] you should say disk image corruption, nit just flash issue I think [17:32] patm, Ah, OK sorry [17:33] s/disk image/file system/, to nitpick [17:33] Still testing with ext3 will provide important info for both issues. [17:33] patm and I just talked off line, I think we have a way around some of his bottleneck. [17:34] it will give you data, not sure how that will help fix the corruption issue if there's a bug in either unionfs or squashfs and you need it meet the footprint requirement [17:34] davidm: are you going to clone Mithrandir ? [17:34] xmas present [17:34] (snowing HARD) [17:34] no snow here. :-( [17:34] robr2, it will at least tell us if the problem is in fact the unionfs or squashfs or not. [17:34] smagoun_, wish I could [17:35] *very* sunny outside :) [17:35] do you have a set of repeatable steps to reproduce the disk image corruption? [17:36] robr2 no if we did I'd have someone assigned to fix it. [17:36] robr2: I can get it to happen pretty much all the time by just hard shutting the system down [17:36] robr2: happens more often on the menlow system with SSD flash [17:36] robr2: no, I can't reproduce on command. It usually involves a hard power off though [17:37] So? [17:37] i'd like to have an engineer help with this issue, can we coordinate/share info? [17:38] robr2 yes, for sure. [17:38] [action] Lexington to test ext3 for boot speed and disk image corruption as soon as an extra CB shows up. [17:39] ACTION received: Lexington to test ext3 for boot speed and disk image corruption as soon as an extra CB shows up. [17:39] robr2: I've had some luck reproducing it by booting the system, then yanking the power cord midway through a couple of times. [17:39] [action] davidm to ship his cb to Lexington today overnight. [17:39] ACTION received: davidm to ship his cb to Lexington today overnight. [17:39] davidm: Thanks!! [17:39] hohoho [17:39] xmas [17:40] [action] Canonical to share all info with robr2 on disk image corruption as soon as available [17:40] ACTION received: Canonical to share all info with robr2 on disk image corruption as soon as available [17:40] is there a bug already in launchpad we could use to share info? [17:40] I believe there is, patm smagoun_ is it a public bug? [17:40] robr, not really, I can create one in moblin [17:40] looking now [17:40] If not we can open a public one [17:41] we could just enter what we find in the bug there and everyone can update/monitor it as needed [17:41] what project should it go against? [17:42] mobile I guess, until we can identify what is the issue [17:42] can't find a bug [17:43] smagoun_, kyleN don't get too excited, I'll need the unit back a some point after the new year but right now I think you can get more use from it solving this stuff. [17:43] Anyway so one more action to open a public bug? [17:43] Or is it done already :-) [17:44] who should open this bug? [17:45] smagoun_, can you since you care do it the most? [17:45] [action] davidm to follow up and make sure a bug is opened to share info with Intel [17:45] ACTION received: davidm to follow up and make sure a bug is opened to share info with Intel [17:46] its private bug#141249, but we should open a new one [17:46] patm, yes please [17:46] davidm: sure [17:46] https://bugs.edge.launchpad.net/moblin-other/+bug/176178 [17:46] Launchpad bug 176178 in moblin-other "Filesystem corruption on hard shutdown" [Undecided,New] [17:46] just filed it [17:46] Good enough, have we gone as far as we can today on this topic? [17:47] This is the kind of situation that should never actually happen, but... [17:47] davidm: I guess so. [17:47] OK then next topic [17:47] [topic] robr to arrange a meeting around USB and VFAT issues. In this case all parties to this meeting are scattered around the world so someone will be inconvenienced. [17:47] New Topic: robr to arrange a meeting around USB and VFAT issues. In this case all parties to this meeting are scattered around the world so someone will be inconvenienced. [17:48] This happened but has been continued, robr2 is it confirmed for tomorrow or not? [17:48] do you know? [17:48] no, i don't know [17:49] i can send an invite out again to the team, but i need to make sure rustyl and alek are available from intel's side [17:50] OK, thanks, I have not heard anything more since the meeting. [17:50] robr2: I think you can simply mail ubuntu-mobile@ when yuo have a date [17:50] It needs not be private mail [17:51] [action] robr2 to check if meeting will happen tomorrow. [17:51] ACTION received: robr2 to check if meeting will happen tomorrow. [17:51] (since the meeting is public anyway) [17:51] amitk, you here? [17:51] yup [17:51] ok then next topic i [17:51] [topic] amitk to update hardy ppa with drivers [17:51] New Topic: amitk to update hardy ppa with drivers [17:51] I setup the continued USB meeting for 10am PDT [17:52] Friday [17:52] davidm: hardy ppa? [17:52] OK, thanks bspencer [17:52] are we agreed that it will work for those here? [17:52] cause I don't want to roll out of bed at that early hour for nothing ;) [17:52] I have to reschedule a meeting with Don Johnson on process but === sodarock is now known as HappyCamp_laptop [17:53] silence is affirmation [17:53] bspencer: And that time would be what UTC (I'm lazy to google for it) [17:53] He had taken that time slot but I'll contact him after this [17:53] agoliveira, +9hrs I think [17:53] or close to that...maybe +8 or +10 after daylight savings [17:53] I think it's 10 [17:53] I think 1800 [17:54] fine. see you then [17:54] PDT is either -7 or -8 [17:54] amitk it was a carry over from the meeting before last week. [17:54] agoliveira: it's 4pm your time [17:54] US/Pacific is -08:00 [17:54] smagoun_: Cool. I was adding PDT to my world clock :) [17:54] smagoun_: Thanks. [17:55] davidm: you mean gutsy ppa though, right? [17:55] We have a hardy ppa to be symmetrical with gutsy ppa [17:55] (agoliveira you might want to double-check, just to be sure!) [17:55] Hmm http://www.world-time-zones.org/zones/pacific-daylight-time.htm says -7 [17:55] I'm lost [17:55] So now I'm confused, amitk I think we need to go back to the old minutes and track this down. [17:55] * amitk learns something new. The vacation was successful :) [17:55] lool, we are on standard time right now, not daylight [17:56] since we are almost out of time I'll leave the carry over in place and we can see what the item was. [17:56] PST [17:56] davidm: sure [17:56] Ok next topic [17:56] http://www.timeanddate.com/worldclock is helpful for this sort of thing [17:56] LINK received: http://www.timeanddate.com/worldclock is helpful for this sort of thing [17:56] thanks mootbot [17:56] Ok so -8 [17:56] [topic] smagoun - discuss tagging moblin projects in git [17:56] New Topic: smagoun - discuss tagging moblin projects in git [17:56] I'd like the intel folks to tag their git trees when doing a release. It makes finding the source that corresponds to a given release much easier. The alternative is wading through commit diffs trying to find the SHA-1 commit ID that corresponds to a release. [17:57] smagoun_, I hear your request. Not the first time [17:57] thanks for your patience :-\ [17:57] Some trees are already tagged, but some - like moblin-media - aren't tagged, and the commit messages don't always announce a new release [17:57] I talked to the media guys about starting this practice when I was visiting in PRC [17:57] sounds like a good idea, we have also discussed having our guys create tarballs at release time [17:57] I'll encourage them a little stronger [17:57] bspencer: thanks, I appreciate it. [17:57] let me clarify my understanding [17:57] after git push, you get the tag [17:58] then you git tag version (or something) [17:58] then you can pull from that tag. [17:58] bspencer: you have to push the tag too, which I think requires a flag to git push [17:58] bspencer: On your side: git commit, git tag , git push, git push --tags [17:59] cool [17:59] * HappyCamp_laptop wonders if he can work it into the build system some how [17:59] * bspencer likes HappyCamp's wonderings [17:59] * HappyCamp_laptop but probably won't try for now [17:59] bspencer, are you willing to take an action on this [17:59] bspencer: at UDS we briefly talked about having a -commits mailing list to track activity on moblin [18:00] We do have a commit mailing list [18:00] HappyCamp_laptop: you do? [18:00] we've had from day 1 ;-) [18:00] HappyCamp_laptop: git-bp can tag the configured branch with the configured formatted tag and message [18:00] Not well publicized :( [18:00] https://www.moblin.org/mailman/listinfo/commits [18:00] amitk: i'm on it... [18:00] [link] https://www.moblin.org/mailman/listinfo/commits [18:00] LINK received: https://www.moblin.org/mailman/listinfo/commits [18:01] davidm, my action is a little soft because we will discuss it and start the practice, but it has no end date. I'm happy to have the action though [18:01] Thanks how about just report status in next meeting? [18:02] np [18:02] * agoliveira wants some freaking lasers... [18:02] oops [18:02] [action] bspencer to talk with deveopers to push get tags when making a release, and report back on status of this [18:02] ACTION received: bspencer to talk with deveopers to push get tags when making a release, and report back on status of this [18:02] amitk: I have also emailed our web content person to add the commit mailing list to the website info. [18:02] OK we are out of time and there are no new items on the agenda [18:03] HappyCamp_laptop: thanks [18:03] lool: I will see if I have git-bp on our systems [18:03] davidm, can I ask about PPM ? [18:03] Sure go ahead [18:03] HappyCamp_laptop: That's git-buildpackage BTW, I'm just lazy typing it without tab :) [18:03] basic question is what is the plan and current status for integrating PPM [18:03] k [18:04] PPM = power policy manager [18:04] Ah, thanks [18:06] patm: who was supposed to answer this question? [18:06] Even better: which question? :) [18:06] the right person to answer that question is rustyl [18:07] [topic] what is the plan and current status for integrating PPM [18:07] New Topic: what is the plan and current status for integrating PPM [18:07] * agoliveira is being picky about the question mark, that's all :) [18:08] * HappyCamp_laptop thinks agoliveira wants question marks in front of and at the end of all questions. Wierd! [18:08] i know PPM is committed on moblin and in images, but i'm not sure the latest status [18:08] amitk, do you have any input? [18:08] HappyCamp_laptop: No, we don't speak spanish around here, as you know :) [18:09] speak only geek [18:09] mohammed is the engineer on our side doing the work with rusty coordinating/directing the effort [18:09] davidm: not really. My plate has been full [18:09] amitk, OK, thanks [18:10] patm, can we send the question to rusty in email? [18:10] agoliveira: predicted high temperatures this week in Portland: 6, 7, 6, 7, 7 in Celsius, and it is wet :( [18:10] [action] davidm to query rustyl on status of PPM [18:10] ACTION received: davidm to query rustyl on status of PPM [18:11] Oregon forecast: http://www.wrh.noaa.gov/forecast/MapClick.php?site=pqr&smap=1&textField1=45.53861&textField2=-122.9596°rees=C [18:11] OK unless there is anything else I'm going to close the meeting? [18:11] +1 [18:12] going once........... [18:12] going twice....... [18:12] gone [18:12] #endmeeting [18:12] Meeting finished at 18:12. [18:13] HappyCamp_laptop: I can just hope for a better weather in january. In any case I already bought a heavy and water-proof jacket. [18:14] davidm: are we going to discuss the driver updates now? [18:14] agoliveira: january is usually colder than December :( [18:15] sodarock_home: I hope it's at least drier or you will see a very ill-tempered version of me there. [18:17] sodarock_home: it's 6-7°C there now? That's nice and warm.. [18:17] LOL, I don't think agoliveira thinks so :) [18:17] it's -9°C here now. [18:18] Brrr [18:18] Okay, I got to go drive into work now === Jayc__ is now known as Jayc_ === Jayc__ is now known as Jayc_ === robr__ is now known as robr [21:29] praj