[02:09] hi === duflu_ is now known as duflu === chihchun_afk is now known as chihchun [05:08] hi [05:13] lol === chihchun is now known as chihchun_afk [05:32] hi [05:33] ubuntu phone works in motorola razr d3? [05:33] sorry my english [05:48] good morning :) === jppiiroinen_ is now known as jppiiroinen [06:47] good morning [07:05] good morning! [07:23] Does anybody know how I can reach vicamo? [07:25] SturmFlut, https://launchpad.net/~vicamo === chriadam is now known as chriadam|away [08:24] zubozrout: hi - got your ping [08:25] hi mcphail [08:26] zubozrout: I think is is fair to say SDL is a "work in progress" rather than at "just works" status [08:26] zubozrout: do you want to discuss this in #ubuntu-app-devel? [08:27] mcphail: ah, ok. But Neverball and Neverputt works - how was that done? [08:27] mcphail: Do you think that it is a better place? [08:28] zubozrout: yes - see you there [08:28] * seb128 isn't a fan of the new icons :-/ [08:29] seb128, same here ... [08:29] and summing up the reactions i see on IRC nobody really is [08:29] I wonder if that's worth opening a bug about [08:29] * mcphail wishes the browser icon would change [08:30] or telling tiheum at least [08:30] i dont really mind the app icons though ... but the panel ones [08:31] I'm okay with the new app icons, but the "Notifications" panel icon gets green when there are notifications, otherwise it's black. By that logic, shouldn't the WiFi icon also get green when it's connected? [08:32] the new camera app icon.. ewwww [08:33] ogra_: I think they look exactly the same as the android 5.0 icons [08:33] yeah [08:33] hmm, i have never seen them live [08:33] i also don't like android m icons === dbarth_ is now known as dbarth [08:33] ogra_: wifey has the nexus 5 they are identical [08:34] bah, thats even worse [08:34] http://blogs-images.forbes.com/gordonkelly/files/2015/05/1.png [08:34] ^^ [08:35] well, BT and battery are definitely different ... [08:35] but yeah, wifi is the same [08:35] still [08:36] I sort of like the new app icons but I don't like the browser icon. [08:36] Also did the panel icons change? Maybe I need to update my tablet and look again. [08:36] yes, massively [08:37] Oh, I remember why I didn't update my tablet. I thought I might want to demonstrate windowed and staged mode with my Bluetooth mouse tomorrow. [08:38] uuu nice [08:38] ogra_: and cellular battery is only rotated 90 ° so the only real change is bluetooth. [08:38] is there any way now to get a screenshot via hardware buttons? [08:38] volume up/down? [08:38] Isotop7: There has been for a long time, press both volume keys at the same time [08:38] Isotop7, since ages, press both volume buttons [08:39] ogra_: the issue is there is only so much you can do with a monochrome icon that 16x16 or whatever size it is [08:39] ah thank you...shouldve followed the last time...sorry :) [08:39] what i actually find worse than the icons is that the date is gone from the today scope [08:39] that makes it a lot less usable [08:41] saw an update land on how they calculate grid units (wily) [08:42] and then someone posted this image http://i.imgur.com/OUWhgUV.jpg [08:42] 5 icons / row on opo [08:42] yeah, looks like they turned it to high [08:43] I feel like that camera icon is staring into my soul. [08:43] :)) yeah [08:43] i really really don't like the camera icon === conyoo is now known as conyoo|AW [08:45] The only one I really dislike is the new Dekko icon. [08:46] nhaines: I like that one [08:46] davmor2: aside from the aspect ratio of the envelope feeling really wrong, the stripes on the side seem too thin. [08:47] (As in, don't extend far enough away from the edge of the icon.) [08:47] ogra_: the Today scope still shows the date here. [08:48] weird [08:48] ogra_: I have no date either [08:48] well, are we supposed to have one or not ? [08:49] date shown here, but I'm not on a fancy channel [08:50] mcphail, we are not on fancy channels either :P [08:50] Huh, Contacts tries to sidestage on the N7, and it's just sort of floating weirdly. [08:50] ogra_: you're not on the secret magical Canonical-only channel? :) [08:51] lol, no, there is no such thing [08:52] No, no, of course not. Say no more. *wink* [08:52] Rule #1 of the Canonical Channel - don't talk about the Canonical Channel [09:02] Good morning all; happy Sugar Cookie Day! 😃 [09:02] Now there's a holiday I could get behind. [09:10] I'm really looking forward to see which day JamesTait will choose on July 14 [09:11] SturmFlut, do you have a preference? [09:12] is it possible to just download files without connecting.them to apps? [09:13] JamesTait: I can't decide between chaos/madness on one side and the artiodactyl on the other [09:13] Isotop7: you mean through web browser app? [09:13] i mean via webbrowser app? [09:14] SturmFlut, not a fan of sharks? [09:15] Isotop7: Not as far as I know. It would be great though - Content hub should allow that - well, maybe it is a filemanager problem as it should associate with every file extension. [09:15] JamesTait: I'm a scuba diver, it's Shark Awareness day every day for us [09:16] okay...i agree with that...downloading is pretty useless atm except your downloading media content... [09:16] Well played, SturmFlut. ☺ [09:16] Isotop7: If I need to download something special, I am forced to use wget, but that is not a solution :) [09:17] especially if you have to browse the site first to get the link... [09:18] Isotop7: Yes, but even that wont help for sites like Google drive :), you can't copy a link there. [09:18] Isotop7: Well, installing filemanager that would be able to handle every extension would help - maybe we should fill a bug there if it isn't already. [09:20] JamesTait: Hm, but have you ever looked a cow in the eye? These creatures are on the verge to chaos and madness all the time. "Pandemonium" and "Cow" is basically the same thing. Let's call it "Laser-Shark Cow Pandemonium Day" then. [09:20] should i file it? [09:20] Another problem solved using questionable logic! [09:21] Isotop7: There is a bug report I think, https://bugs.launchpad.net/ubuntu/+source/webbrowser-app/+bug/1354391 [09:21] Launchpad bug 1354391 in webbrowser-app (Ubuntu) "[browser] Can't download random files" [High,Triaged] [09:21] Isotop7: Planned for OTA-6 [09:22] nice...since otas for arale isnt compatible: at what ota can i expect it in arale? [09:23] Isotop7: What do you mean by "not compatible"? OTA updates are shipped for all devices [09:23] SturmFlut, I like to go geocaching, and once found myself in a field surrounded by cows, with my two boys. That was pretty scary. [09:23] Isotop7: No, I just did it :) https://bugs.launchpad.net/ubuntu-filemanager-app/+bug/1472978 [09:23] Launchpad bug 1472978 in Ubuntu File Manager App "Associate with every extension to allow downloading every file using Web Browser" [Undecided,New] [09:24] It turned out that wasn't the way to get to the geocache. [09:24] how can one disable app's life cycle? :/ i want to be able to run all of them in foreground in windowed mode on a desktop [09:25] zubozrout: There was already a bug report, I just posted it, https://bugs.launchpad.net/ubuntu/+source/webbrowser-app/+bug/1354391 [09:25] Launchpad bug 1354391 in webbrowser-app (Ubuntu) "[browser] Can't download random files" [High,Triaged] [09:25] SturmFlut: Well, thanks for the info, I didn't know that. Then there is a philosophical debate where is the best place to solve the bug. [09:25] conyoo|AW: Doesn't the Life Cycle only exist on the phone, and not on the desktop [09:26] i always read ota 4 is available but i only got ota 2 on normal.channel... [09:26] SturmFlut, nope, it's the same on desktop [09:26] wily desktop next [09:29] SturmFlut: only when your in the water, surely you don't go looking for sharks in your house ;) [09:30] seb128: do you know about the lang packs? I'm trying to understand why something that is translated in es.po in messaging-app repo is not in the messaging-app.mo that comes with language-pack-touch-es === conyoo|AW is now known as conyoo [09:32] tsdgeos, read backlog from #ubuntu-devel from the past half an hour [09:33] okidoki [09:33] tsdgeos, or just move to something else, it's being handled [09:33] okidoki [09:33] :-) [09:34] can someone enlighten me regarding the otas for arale and the other phones? [09:34] SturmFlut: So what is the resolution of that web browser unable to download anything bug report? As I understand it is long time reported. Is this a solution in works? https://bugs.launchpad.net/ubuntu/+source/webbrowser-app/+bug/1354391/comments/4 [09:34] Launchpad bug 1354391 in webbrowser-app (Ubuntu) "[browser] Can't download random files" [High,Triaged] [09:35] zubozrout: The last status update says "Milestone ww34-2015", that means that it is planned to be solved with the next OTA update in six weeks. [09:35] SturmFlut: and sorry for reporting the other bug, apparently I was too quick, although I still think File Manager should manage everything, not only some selected file extensions. [09:36] SturmFlut: ok, thanks [09:38] okay...but what will be the version number for this ota for arale? also 6? atm im stuck at ota2 on main channel... [09:41] Isotop7: The OTA is just a virtual number that maps to a specific image release number for every device. For example OTA-4 was r23 on krillin, but a different image number on other devices. [09:44] yeah i thought that...so how is ota 6 comparable to arale (which actuak status is ota 2) and mako (ota 4)? [09:45] Isotop7: you'll note that the state says "r2" (release 2), and not OTA. [09:45] arale shipped after several other OTA releases. That's why it's only had one update--it shipped with improvements already. [09:58] An Ubuntu phone will launch in India! [09:58] \o/ [09:59] great ... [09:59] Still just waiting on a California launch. :) [09:59] [citation needed] [10:00] * ogra_ waits for the "how do i dualboot with tizen, how do i replace ubuntu with tizen" questions [10:00] :P [10:00] "Dear lazyweb, I wanted an Ubuntu phone so I bought a phone I knew didn't come with it. halp!" [10:01] heh, yeah [10:01] while that is the topic on the german ubuntu channel, again. any news on whatsapp client for ubuntu? :/ [10:02] k1l_: all that info will come from whatsapp. [10:04] ok. so its about getting to the breaking point of getting a market share or getting enough users complaining to whatsapp guys of not making a client [10:04] Yup! [10:04] And the second one doesn't necessarily have to wait for the first one. ;) [10:05] was there some communication between ubuntu/canonical and whatsapp so far? [10:06] No one will ever know, because that would be confidential. [10:06] Probably. :) [10:06] Stskeeps: Can't give you a citation :/ You'll have to believe us. [10:06] http://www.reactiongifs.us/wp-content/uploads/2014/06/its_classified_barack_obama.gif [10:06] ;p === dholbach_ is now known as dholbach [10:08] k1l_: yes [10:09] k1l_: we have a team who have those kinds of conversations all the time === greyback__ is now known as greyback === chihchun_afk is now known as chihchun === MacSlow is now known as MacSlow|lunch [11:22] hum... I have an offset when it comes to appointments in the timedate indicator - did anyone else notice something like this? [11:23] that's with rc-proposed [11:27] dholbach: What kind of offset? I haven't noticed it yet - on devel-proposed. I have 3 events shown there and all seem to match their proper dates. [11:28] zubozrout, they are off by two hours [11:28] I'm in UTC+2, maybe that's why(?) [11:29] dholbach: same here on devel-proposed, off by one hour (just created a testing event) [11:30] popey, ^ have you seen this happening? [11:30] err, sorry [11:30] I'm on rc-proposed [11:30] not devel-proposed [11:31] dholbach: Hmm, no, it is showing me the time right ... I was looking at a wrong place, sorry [11:31] dholbach: Yep, I know you are on rc-proposed, ... it is just that I am on devel-proposed and was trying to reproduce your bug, but I failed. [11:32] dholbach: Also UTC+2 here, so maybe it is already fixed [11:33] dholbach: Do you have the events saved locally or is it shared with Google calendar? [11:35] dholbach: In my case both events (one local, one in Google calendar) I created using Calendar app shows right. [11:36] abeato, hi, quick question. When list-modems echos no ServiceNumbers and the user expects service numbers, is that ofono's fault? [11:37] dholbach: Reboot won't change it - Time indicator still shows the events right. [11:38] dholbach: not myself, no [11:39] dholbach: are these events from your google calendar? [11:43] dholbach: do they show at the right time in the calendar app? [11:52] jgdx, depends on the user :p === chihchun is now known as chihchun_afk [11:52] jgdx, ServiceNumbers are simply a list of numbers that come in a SIM file [11:53] jgdx, so they are either there or not. Each operator has different criteria for inserting numbers there === _salem is now known as salem_ [11:56] abeato, okay, thanks! [11:56] jgdx, np [12:02] popey, yep, at the right time in the app [12:02] hmm, odd. [12:02] popey, and yes, from the google cal [12:02] maybe a bug in indicator-datetime? [12:02] poke charles [12:07] hmm [12:08] dholbach, popey, that sounds like https://bugs.launchpad.net/ubuntu/+source/indicator-datetime/+bug/1456281 which is fix released, let's confirm that landed in rc-proposed [12:08] Launchpad bug 1456281 in indicator-datetime (Ubuntu) "Alarms not going off in ubuntu-touch/rc-proposed/bq-aquaris.en" [Critical,Fix released] [12:11] charles, 13.10.0+15.04.20150618-0ubuntu1 is on arale rc-proposed [12:11] dholbach, I don't have an arale, I can test on ubuntu-touch/rc-proposed/bq-aquaris.en [12:12] let's take a look at 13.10.0+15.04.20150618-0ubuntu1... [12:16] o/ [12:19] dholbach, do the events show the right time in the indicator pulldown? [12:19] charles, no, that's where they're off [12:20] in the calendar app and in the google cal they're right === MacSlow|lunch is now known as MacSlow [13:27] hey seb128, thanks for noticing that the bug I reported is a duplicate of bug #1420500 [13:27] bug 1420500 in ubuntu-system-settings (Ubuntu) "Screen lock delay values are too high" [Wishlist,Triaged] https://launchpad.net/bugs/1420500 [13:27] dgadomski, yw :-) [13:27] seb128: from the bug history I understand that the design for that bug is ready [13:27] do you? [13:28] it's assigned to a designer (mpt) but there is no design recommendation [13:28] I was thinking about the "no longer affects: ubuntu-ux" comment [13:28] it's because mpt doesn't like that project [13:28] he prefers to keep the component bug and assign it to the designer [13:28] oh, so it's clear know :) [13:28] thanks [13:28] but https://wiki.ubuntu.com/SecurityAndPrivacySettings#Phone still has the "n minutes" list [13:28] https://wiki.ubuntu.com/SecurityAndPrivacySettings?action=AttachFile&do=get&target=phone-security-privacy-idle.png [13:29] mpt is on holidays for the next week or so now [13:29] going to need to wait for him to be back I guess [13:29] cool, let's see what he decides [13:37] kenvandine, you said upstream libqofono added modemtechs somewhere, but I can't find it in your 0.79 branch? [13:37] jgdx, let me look again [13:38] popey, zubozrout, followup to the indicator-datetime question dholbach asked above, looks like this is a fixed bug #1456281, we just need to get the fix in ubuntu-touch/rc-proposed/ubuntu [13:38] bug 1456281 in indicator-datetime (Ubuntu) "Alarms not going off in ubuntu-touch/rc-proposed/bq-aquaris.en" [Critical,Fix released] https://launchpad.net/bugs/1456281 [13:38] go go go! :) [13:39] ship it === howefield is now known as howefield_afk [13:52] kenvandine, upstream ofono has added the availabletechnologies property, but I can't see that libqofono has added support for it. [13:52] abeato, hi, is ModemTechnologies === AvailableTechnologies (from upstream)? [13:53] abeato, if so, why the conflict? :) [13:53] jgdx, right, I submitted upstream and got the patches accepted, but they preferred to change the name [13:53] abeato, cool, but also uncool. How long will we have to carry that patch? [13:54] jgdx, no worries, I'll keep you updated when we move to 1.16 [13:54] okay :) [13:54] kenvandine, so we'll need to unpop that patch for now :) [13:55] jgdx, sorry, i got distracted :) [13:56] jgdx, i guess that's what i saw [13:56] so what we need to do is update settings to match [13:56] kenvandine, … eventually yes [13:59] but we're not using upstream ofono, so we'll have to wait until abeato breaks api in 1.16 :) [13:59] oh right [13:59] jgdx, ok i'll port that old patch [13:59] I can propose a patch for libqofono in the mean time [13:59] jgdx, cool [14:00] i'll port the old patch [14:00] i think I have the old one for modemTechnologies somewhere [14:00] i have it [14:00] it didn't apply cleanly [14:00] jgdx, kenvandine must be a matter of just changing the name [14:00] abeato, yeah but our ofono doesn't have it === vrruiz_ is now known as rvr [14:31] kenvandine, done testing. Can't find any regressions, except the connection type one. [14:32] Hello === dandrader is now known as dandrader|afk === charles_ is now known as charles === dandrader|afk is now known as dandrader [15:55] beuno: When will Panda Love hit the front page of the click store scope? ;) [15:56] SturmFlut, I don't make those decisions! you have to bribe someone else [15:57] beuno: The panda has enough love to bribe everybody. Even Canonical management loves the panda. === dandrader is now known as dandrader|lunch === pstolowski is now known as pstolowski|afk [16:59] hi all.. :) [16:59] someone who knows if touch is ok with samsung s5? [17:01] !devices | bayoumi [17:01] bayoumi: You can find the full list of devices, official images, community images, and works in progress at https://wiki.ubuntu.com/Touch/Devices [17:03] popey, thx.. Samsung Galaxy S5 (SKTelecom) | galaxyz | WIP | WIP | N/A | Sam W. [17:03] Does it mean that its only work on SKTelecom? [17:03] i dont think it's done [17:04] no, cause its still in work in progress... [17:44] Hello. [17:44] I have just installed Ubuntu Touch on my Nexus 4, watched the convergence video [17:45] How I can use the "convergence" on it? [17:48] do we have any info on whatsapp on ubuntu touch? i hate that this is a showstopper for the most people... (for info: http://news.softpedia.com/news/Canonical-Working-to-Add-Whatsapp-and-Dropbox-as-Services-and-Not-Apps-483172.shtml) [17:52] bump? [17:53] Akritor: i dont know if the convergence is out in the wild already [17:54] Isotop7: i asked the same question today and that was the conclusion: ok. so its about getting to the breaking point of getting a market share or getting enough users complaining to whatsapp guys of not making a client [17:54] Isotop7: but its the whatsapp guys that need to make the client [17:55] oh okay :D [17:55] maybe we can do some kind of petition? :D [17:55] Isotop7: you need to ask whatsapp about that [17:55] you know what? I'll do that right away :) [17:57] Isotop7: yeah, its not something we can do [18:00] k...thanks anyway :) [18:09] hi all! I have purchased Cut the Rope, but something is weird with is. Is it normal If I play with it and exit when I start to play again I need to complete all level in the boxes? So it resets my box progress everytime. [18:11] it should retain your progress [18:11] is there already a bug for the pixelated photo preview on arale? Dont want to file it twice... [18:11] let me test that here [18:12] Thanks, and sorry for my English :) [18:14] robin-hero: yeah, it's not saving progress here either. [18:14] robin-hero: I'll report it [18:14] although dunno where :) [18:14] popey: :) [18:18] guess ill just report my bug :) [18:18] what bug? [18:19] robin-hero: https://bugs.launchpad.net/canonical-devices-system-image/+bug/1471569 [18:19] Launchpad bug 1471569 in Canonical System Image "cut the rope does not save after 2. box, 4. level" [Undecided,New] [18:22] the camera preview on my arale is pixelated and looks like the camera is more like 2mpx...the picture itself after it is taken is very good... [18:22] popey: Thanks [18:22] oh, odd. [18:22] not seen that [18:26] Hello Everybody one stupid question: How is it possibe to modify a file with nano and ADB ? I cannot save... [18:27] pillepop: the core system is read only [18:28] pillepop: if you really need to do it then you can make the system writable but it will break updates if you do that [18:28] I need to modify the configfile from syncevolution. Any Chance? [18:28] pillepop: theres an issue with nano over adb shell, either use phablet-shell then nano, or adb shell then vi [18:28] or just pull, edit, push... [18:28] Ok, I will have a try thank you. [18:29] or ssh [18:29] or that [18:29] out of interest, why are you fiddling with syncevolution files? [18:30] pillepop: ^ popey question I think is for you [18:30] I hav a problem with my selb singned Certificate and would like to swich of temporarly the Safety ... [18:33] ahhh [18:44] how about customizable led notifications on arale? is this a thing were going to see? [18:45] Thank you for your help, I've got it with vi. Now its time for Bed. Greatings from germany and byby [18:45] see you pillepop ! [18:45] Isotop7: the arale LED is white [18:45] hard to customise [18:45] unless I'm mistaken [18:48] Isotop7: It sure is all you need is a felt tip pen ;) No seriously popey is right it's a white led [18:56] popey: i rather meant it regarding pulsating and periods...for example emails are a long pulse with long pauses and telegram can be 3 fast pulses with a short pause...also correct me if i'm wrong but last night i thought the led had a green shade when i charged it while being off... [19:19] Isotop7: i see a green pulse when there's a system update available [19:20] dobey: so you can confirm there is more than a white led? [19:21] yes, it used to be blue for system updates [19:21] but it's green again now that the notifications envelope icon was changed back to green [19:22] not for arale [19:22] it's a white led, that's the only color it does [19:24] oh [19:24] is the hardware only white? or is it just a "driver" issue? [19:25] hw i think [19:27] kay...then it must be pcb or something shining through... [19:32] okay...thats the solution to my question... (http://meizufans.eu:4567/forum/topic/2140/home-led-colour-module/2) sorry! [20:37] Can anyone help me resolve the Ubuntu phone MX4 mount issue? It mounted once but never again since on any computer or port. It just shows blank [20:41] Fragy, hi what do you mean it shows blank? [20:41] Hello [20:41] How can I delete my apps from https://myapps.developer.ubuntu.com/dev ? [20:42] pmcgowan: well, nothing's there. [20:42] pmcgowan: if I do 'ls' [20:43] Fragy, what image is installed [20:43] I am seeing an issue here too [20:43] pmcgowan: how do I check? [20:43] oh wait need to unlcok it [20:44] yes, I've tried it with unlocked, it's the same, even enabled the developer's mode [20:44] Fragy, if you have a password or pincode you need to enter that on the phone and unlock - until then it shows blank === BOHverkill_ is now known as BOHverkill [20:44] hmm [20:46] Fragy, that usb is sensitive, sometimes only works with the cable that came with it [20:46] and on usb 2.0 port [20:47] Fragy, are you plugging into ubuntu desktop [20:47] I have the original cable and using usb 2.0 [20:47] well, linux mint. I've restarted the phone now, it doesn't mount anymore. [20:47] I'll check on a windows machine. brb [20:48] fwiw this is what I see in syslog when I plug it in and it works http://pastebin.ubuntu.com/11851358/ [20:50] same in windows, it doesn't mount [20:51] pmcgowan: where can I check the log? [20:51] I did tail -f /var/log/syslog then plug it in [20:51] alright, checking... [20:55] ncme: click on the app, and then on the left click the "Unpublish application" button [20:56] pmcgowan: I noticed that mounting arale via MTP usually worked with r2, at least wit my device, but since switching to rc-proposed r52 it seems to have gotten worse. Can't get it to cooperate most of the time now. Same USB cable, same USB port on the notebook, no USB hub. [20:57] pmcgowan: I see the same error in the logs as you do, http://paste.ubuntu.com/11851382/ [20:57] "PTP: reading event an error 0x02ff occurred" [20:57] SturmFlut, but mine is moutning fine here [20:58] pmcgowan: I don't get any of that. I get: musb-hdrc: BUS RESET as b_peripheral [U3D] [U2 RESET_INTR] [U3D][USB Speed = High speed [21:00] Fragy, there are some tips here but I don't think you need the adb file http://askubuntu.com/questions/632444/meizu-mx4-ubuntu-edition-not-mounting-as-usb-drive-and-not-available-to-adb [21:00] pmcgowan: It is followed by "ignoring libusb_claim_interface() = -6PTP_ERROR_IO" most of the time, which results in an USB reset, and after that reset arale should probably reconnect to the bus, which doesn't seem to happen [21:01] pmcgowan: It mounted the first time I plugged it in just fine. I was able to browse the phone. [21:01] but that was the only time.. [21:01] hmm [21:01] Fragy, what kind of USB is that ? the MX4 has probs with 3.0 ... the slower the better ... (i use an 1.1 hub here) [21:02] ogra_: 2.0, original cable [21:03] it "could" be hardware we have seen some of the older ones lose connection [21:04] pmcgowan: I have noticed a couple of things regarding USB/ADB/MTP on my device [21:04] 1. phablet-shell seems to work 99% of the time on my device, even when MTP just failed, and the phablet-shell session hardly ever randomly disconnects. [21:05] aren't MTP and PTP different things? Why would you be getting a PTP error? [21:05] 2. If I want to transfer large files via ADB, e.g. when using u-d-f, the ADB connection just silently stops working. [21:05] we do know that at some point meizu replaced the usb part in production, but that was some time ago [21:05] yeah, PTP should not be enabled at all [21:06] 3. If I boot the ADB-enabled recovery image from tangxi, ADB always seems to work, even with large transfers [21:06] yep [21:06] SturmFlut, yes to all of those [21:06] in recovery it draws less power [21:06] but mtp always works with a good cable for me [21:08] I have no ideas other than to restart everything [21:08] So could it maybe be the case that our production kernels have problems with large USB frames? Because phablet-shell uses an ADB TCP port forward and very small packets, but ADB push/pull uses full-sizes ones. [21:08] it could be ... [21:09] Let me check that [21:10] it could also be a discrepancy how the little kernel and the actual kernel initialize the HW [21:11] Recovery runs a full kernel, Little Kernel just talks Fastboot [21:11] yes and initializes the USB port for that ... [21:12] and then hands over to uboot which starts the main kernel who then initializes USB again [21:12] there are three drivers in that chain ... [21:16] There is no uboot, and Little Kernel initialises the USB regardless of which Linux kernel comes next, Normal or Recovery [21:17] oh ? i thought there is a uboot between the two kernels [21:18] No, uboot wouldn't add any value. The Boot ROM loads the Preloader, the Preloader loads Little Kernel, Little Kernel does Fastboot or directly loads the final Linux kernel [21:19] http://sturmflut.github.io/ubuntu/touch/2015/05/07/hacking-ubuntu-touch-index/ At the bottom, spent a whole weekend reverse engineering the whole MediaTek SoC boot process [21:20] * SturmFlut always has a blog article for everything :P [21:22] heh [21:22] Okay, captured a full USB trace during a failure, let's see [21:24] Hmmmmmm [21:29] anyone know about the ASUS Transformer Infinty? [21:30] If I interpret this correctly, there is not a word of MTP spoken on the wire, this fails at a lower level [21:33] SturmFlut, well, you should see HW disconnects on your desktop when that happens [21:33] (at least i did in the past) [21:36] I see something I don't like [21:37] sturmflut@fire:~$ lsmod | grep cdc_acm [21:37] cdc_acm 36864 46 [21:37] 46? [21:39] Well, because it apparently reacts to the USB ID of the MediaTek preloader [21:39] and then fails to communicate and the module is not unloaded [21:43] I found a security issue [21:43] ahoneybun: Which one? [21:44] so when the phone is locked right [21:44] You can pull down the indicator bar [21:44] but if I build and deploy my app from the SDK it bypasses the lock [21:44] then the system is free to use [21:44] Yeah, but that requires Developer Mode enabled [21:45] or doesn't it [21:45] true [21:45] it should definitely remove the lockfile that enables this when you disconnect [21:45] (it should only work during one session) [21:46] SturmFlut, even with developer mode enabled the screen unlock is mandatory [21:46] disconnect the usb? [21:46] well, whatever you can do in the sdk to disable/disconnect it [21:47] there is a file you can put in place to override the lockscreen checking ... that file needs to be removed when the sdk disconnects [21:47]