[00:25] Project daily-hammerhead build #340: SUCCESS in 11 min: http://ci.ubports.com/job/daily-hammerhead/340/ [00:27] Project ota-push build #1111: SUCCESS in 1 min 52 sec: http://ci.ubports.com/job/ota-push/1111/ [00:34] Xuyznaet was added by: Xuyznaet [01:05] @Xuyznaet, welcome [01:07] hi ! Реально скачать прошивку для bacon [01:07] okay so the kernel boots but I had to disable an if check that is there for optimzation to get it to compile with gcc 6, and I haven't tested to make sure everything works yet [01:08] @Xuyznaet, this supergroup is English only, would like to point out [01:08] built out-of-tree using gcc 6 [01:08] you can get exclusive support from me на русском, but for that you will need to PM me (@vanyasem) [01:09] Now to add in the freqeuncy and voltage scaling modules [01:10] still don't know if it's stable though, just know it boots and seems to run no worse than the version i downloaded from official source [01:11] "It's not obviously worse" is good enough [01:11] :P [01:12] lol, not by my standards [01:13] but if bugs are there they will show up eventually [01:14] moving from gcc 4 to 6 might cause some issues, as 4 has more undefined behaviour that could be used in ways that cause errors in 6 [01:15] but moving to 6, means we'll be ready to more to 7 when it gets mature in 6-8 months without issues, and 7.2 currently outperforms all previous versions(for arm archs especailly) in terms of efficient output [01:15] right now 7.2 works pretty well but they need to add more unit tests for better code coverage [01:16] I fixed all the unit test fails, but they need more unit tests [01:17] Like i can't fix errors I don't know are there [01:17] and I don't know gcc internals enough to write unit tests for all the stuff I think needs it [01:17] Oh, you're talking about the tests for 7.2 [01:17] yeah [01:17] Not the tests for the kernel under 7.2 [01:17] correct [01:17] I'm talking about gcc itself [01:18] As a coding noob, I respect and fear what you're talking about [01:18] lol, give it another 20 years, you'll get there [01:19] Getting to where i'm at, takes a long time, but it's not as hard as it seems, just a lot of stuff to learn and practice [01:20] I just learned how to downcast in Java. [01:20] I got a long ways to go [01:20] I've been studying asm/os dev for 14 years, and compiler theory for 4-6 years [01:20] coding for 25 years total [01:20] so it took me 11 years before I even thought about learning this stuff [01:21] I always dreamed when I was first learning to code of making my own OS [01:23] so eventually when I felt ready... i devoted a LOT of time to learning it [01:25] there's many areas of computer science I am not good at though, I just have niche skills that are rare because they are rarely needed [01:25] Like, not many asm coders with compiler theory and os dev experience are needed, compared to normal software devs [01:26] You only need a few hundreds in the world really... and then there's the 200-300 of us who have our own os projects we do as a hobby [01:27] No one wants to hire you for embedded devices? [01:27] You'd think that'd be pretty killer for this new IoT thing. [01:27] they might if I had credentials [01:27] but I don't have any [01:28] Damn. Pieces of paper are killers. [01:29] yeah... I also never wanted a job in the field until recently so I never kept anything I made for a portfolio [01:29] so I have no piece of paper and very little work to show my skills to employers [01:29] oh... [01:30] MS interviewed me over skype.. made it to the second interview, then they asked me about my education... they lost interest when i told them I left school in the middle of grade 5 so I have a grade 4 education [01:30] Sad, seeing as not being in school just meant that I spend all day every day coding and learning more about code [01:31] but whatever, they won't be getting any more vuln research from me [01:31] I reported 8 0 days in 2 months and they didn't give me anything [01:32] Okay, depending on severity that's ridiculous [01:32] I should have sold them as exploits on the black market [01:32] But that's unethical and gets you disappeared by the FBI. :/ [01:32] Oh, I guess we're off topic. [01:32] lol I wouldn't actually do that, but I wish i was the kind of person that would [01:33] yeah true [01:33] Normally I'm the one who tells people to move to the off topic group. :P [01:34] haha, i'm always getting off topic because I'm using to being on a small network where off-topic is allowed unless topic is actively being discussed [01:34] old habits die hard [01:34] if you wanna chat more pm me, otherwise I'll go back to coding [01:36] On Telegram we have our offtopic group, @ubports_ot. It's a neat place. [01:37] @UniversalSuperBox Keeping that one Telegram only? [01:38] I guess if there was enough demand I could convince the bot's keeper to put it in another group [01:38] No worries, if I'm that curious I could just fire up the Telegram client. [01:45] I have telegram but I prefer IRC... old habits i guess [01:45] I use telegram when I'm on the phone away from computer [01:45] otherwise I'd rather all my chats be in the same window [01:46] I'm one of those matrix weirdos, which also means I'm on IRC too. Oh wait I'm offtopic still. :p [01:46] matrix weirdos? [01:46] https://matrix.org [01:47] ahh, I'm on i2p myself [01:53] #ubports:matrix.org === Piece_Maker is now known as Acou_Bass [04:31] @wayneoutthere, +1! Exactly my point even if I am not in a Christ church. [04:38] @xorpad, Wow! 5 days battery life would be really great. Hopefully those hammerhead improvements would also be able to use on other devices like my fp 2 😁 [04:39] bastos777 every chip is different in terms of how you code it for non-ARM-standard stuff such as controlling volatage and frequency, but it should certainly be a feature in all kernels [04:41] like, ARM is just a standard made by ARM Holdings, the chips are not related. unless an architecture/driver programming documentation is available we can only do things android already does by reverse engineering what they have [04:41] I can reverse engineer stuff but it's very time consuming [04:42] and I don't think voltage scaling is normal in android kernels so we'd have to either fuzz around for it, likely bricking a lot of devices, or give sexual favors to someone at the manufacturers organization in exchange for their docs [04:43] the docs do exist, we may be able to get them under an NDA if we ask nicely and make ourselves look really official and professional, but no guarentees [04:43] Like, if the chip maker publishes the programming/architecture docs, it will be easy [04:44] if they don't... it will be next to impossible without bricking a lot of devices trying to figure out what opcodes change the voltage [04:45] I've fuzzed a driver before on a netbook gpu, killed my own netbook, then killed 20 other peoples netbooks, then got it narrowed down enough I stopped bricking them and just had to find the right options from the ones I had in the pool [04:47] We can steal code from other custom kernels if they have the features though [04:47] but if they have the features, the docs are probably public and we should make our own code [04:47] if you want a job done right you gotta do it yourself [04:52] @xorpad, This is quite a good rule. Would vote for practicing it like this. Anyway we have the @ubports_of groups. [04:54] what si ubports_of ? [04:55] @xorpad, 😂 [04:57] @xorpad, Sorry. @ubports_ot . autocorrect bug. [04:58] ot is what/ [04:58] This is the off topics group from our community in telegram. [04:58] Try it out... [04:58] oh, well I'm talking on topic, about kernel dev [04:58] and whats required to add that to all kernels [04:58] I know. [04:58] but I will join it... [04:59] gimmie a min to get onto telegram [04:59] I replied to your previous post [05:00] I am far away from being a coder, but I like your comments a lot @Xorpad ! [05:01] Thank you for contributing. [05:01] bastos777: https://zerorax.imgur.com/all/ <--- I'm the second one [05:01] oops https://imgur.com/E17yO5H [05:01] lulz [05:07] That one? ' I code and practice martial arts and play guitar and that's pretty much my whole life.' [05:08] @UniversalSuperBox, There's this amazing UI kit that you can open up in inkscape https://github.com/halfsail/Ubuntu-UI-Toolkit [05:10] bastos777 yeah, that's a very brief desc of my life [05:10] but I meant the image.. the one with the coder memes [05:44] J was added by: J [06:15] hmmm... it seems when you remove tasks via swiping from the active task slide-out menu, the memory they are using doesn't get freed [06:22] and, the latest devel release update soft-bricked my phone [06:23] so there's that [07:01] Project click-import build #14: SUCCESS in 6.6 sec: http://ci.ubports.com/job/click-import/14/ [07:01] Project vivid-rootfs-armhf build #149: SUCCESS in 1.3 sec: http://ci.ubports.com/job/vivid-rootfs-armhf/149/ [07:01] Project ota-push build #1112: SUCCESS in 2.5 sec: http://ci.ubports.com/job/ota-push/1112/ [07:39] We need to work on multirom so people don't have to give up android to use ubuntu touch [07:40] it makes it easy to install ubuntu touch from inside a rooted phone to a place in /data/media that can be mounted instead of the normal partitions and booted as if they were flashed to the device normally [07:41] Download MultiROM app - > install ubuntu touch -> have dual-boot android + ubport -> ??? -> profit!!! [07:44] If we work on multirom, we can increase user base significantly and make the level of skill/comfort required to install it much less [07:45] @xorpad, you completely skip the part where UT needs to be ported to each and every phone first. even if UT supports Multirom, it doesn't mean it would automatically run on everything [07:46] YougoChats: I'm aware... lets go one step at a time here [07:46] In very past I had windows together with Ubuntu, and it was some how stressing, so I have made the choice: ubuntu! and since that times I did not switched back [07:47] I had the courage, they can find it also [07:47] YougoChats I can port multirom to some devices, but only ones i have or can get cheaply for testing on [07:47] and UBPorts is swamped as it is. adding demand for multirom, and by extension all devices that run multirom doesn't help [07:48] well, I'm not saying someone else should do it [07:48] UT is still at a stage where it needs to prove it can run well on a limited set of devices [07:48] i'm saying I should do it [07:48] and yeah, just a limited set of devices, [07:48] the ones officially supported for now, we can add to that list later [07:48] s/we/I [07:50] even if we just get it working on one popular device, I have a nexus 5 and plan to get all the devices supported by ubports when I can afford to [07:50] so for now I can start with the nexus 5 [07:50] even if it only works for nexus 5, it's good to have it working and maintained [07:50] ofc when I get those other devices i can port it to them too [07:51] and I would like to make some more kernels for more devices too [07:51] right now i'm working on adding voltage scaling to the nexus 5 kernel [07:53] @xorpad, Of course I do not want you to stop it is your decision. Any help and effort is welcome. [07:53] Well you can expect a lot of effort [07:53] :D [07:55] well, if you can make Multirom sit in between without UT noticing, or submit patches to UT that don't otherwise affect UT's functioning, i'd say go for it. [07:55] I am pretty sure I can... and if I can't, I'll research how and then I'll be able to [07:56] I mean... MultiROM working should be easy [07:58] porting ubports to new devices will be hard, but i'm up for it in the semi-near future... For now I'm just gonna work on the nexus 5 kernel though and multirom [07:58] the hard stuff is the fun stuff:D [07:59] thinking of making a replacement for the android bootloader though... and having the bootloader as something we load and branch over to if we need it, and then have a proper linux bootloader, and we can repartition the device with the bootloader so that we don't have to deal with androids partition layouts [08:00] I dunno, I got a lot of plans, we'll see what you guys accept out of what I make i guess [08:01] we like pioneers and dreamers [08:01] yay! [08:02] And whatever you don't accept, I'll put into a fork and you can merge it back if you ever want to [08:03] Mostly I'm interested in working on the kernels really, but I also want to have a light-weight alternative to unity 8 [08:04] Also, porting to other devices, and making the system not dependant on anything from android other than the android linux kernel and drivers [08:05] Basically I'll do whatever I feel is a good idea to do, from app dev to kernel modding [08:06] I'd like to make a new keyboard... the camera needs some algorithms to increase the contrast and sharpen the image because it doesn't look as nice as when I had android on the same phone... [08:06] Bellow the list of core https://ubports.com/page/website.aboutus [08:07] cool [08:07] maybe I'll earn a spot on that developers list eventually:D [08:08] but that's not really important, i just like OS development [08:08] I don't care if I'm an official dev or not tbh, I just want to write code [08:09] They are also here, but not now [08:09] 👍 [08:09] I know:D I hope they are interested in all my ideas:D [08:10] does ubports have a repository for apt? [08:10] whenever i try to use apt it errors on me [08:10] saying nothing found [08:11] I could maybe host a repository if there isn't one [08:12] all I know is I want this project to be awesome and I want to help make it awesome [08:14] as UBorts plans to move UT onto Halium, perhaps it is wise to get Halium running on Multirom? [08:15] YougoChats... Probably a good idea... but that's trivial [08:15] you'd upen up a whole range of mobile distros to Multirom [08:16] Multirom is just a kernel used as a bootloader which mounts the partitions from the images in the filesystem and calls kexec to boot the boot the kernel and begin the rom booting [08:16] so multirom already basically supports halium [08:16] Halium might need some mods, but multirom will work as is [08:17] well, then you're almost there 😊 [08:17] Yep... this is easy peasy for me, just time consuming [08:17] but I enjoy it [08:18] it's taken me 14 years to learn os dev and compiler dev and how to write secure code and how to reverse engineer stuff so I can use it for my own projects... but i feel like I'm finally ready to stop with the hobby projects and work on a serious one [08:18] I am driving [08:18] I am sitting on the couch [08:18] :D [08:19] i don't drive... Don't have one of those car things [08:19] my family bought me one once, but I totaled it the first day I had it drifting around a parking lot [08:19] so they said never again [08:19] Morning [08:19] Hi Flohack! [08:19] xorpad are you never asleep 😆 [08:20] Just catching up with the night [08:20] I skipped now 300+ messages or so [08:20] Flohack, not usually... I got brain damage in a muay thai fight that makes me never get tired [08:20] the part of my brain that should trigger sleep has 0 activity on FMRI [08:21] so I have a cocktail of meds for the occasional time I decide to sleep, but mostly i'm just coding all day and night [08:21] @Flohack thanks for coming [08:21] Usually my days last 60-80 hours, followed by 10 hours of sleep [08:21] it makes me really productive [08:22] I got the nexus 5 kernel building with gcc 7.2 without the android build tree [08:22] still need to fully test everything [08:23] Project daily-fp2 build #358: SUCCESS in 13 min: http://ci.ubports.com/job/daily-fp2/358/ [08:23] that should alone give like 20% or more performance boost, because 7.2 has so much optimzing compared to 4/5/6 [08:24] Project ota-push build #1113: SUCCESS in 1 min 12 sec: http://ci.ubports.com/job/ota-push/1113/ [08:24] 20% is conservative, I've seen 40% boosts on certain perfomrance aspects when I build AOSP with 7.2 [08:25] it's insane how much 7.2 optimizes compared to anything before it... well 7 and up do it, but 7.2 is the first 7.x that's reliable enough to use [08:25] @xorpad, Hmm on my N5 I have a really bad experience with the battery. I dont expect that 900% will work with Ubuntu Touch but if you can double or triple it, would be fine ;) [08:26] Flohack, have you tried franco kernel on your nexus 5? [08:27] it literally gives you 9x the standby time, although screen time can't be fixed with anything in the kernel and will still only be 4-5 hours [08:28] but franco kernel just isn't very effective without the control app that costs $5... We will have the settings menu as our control app... if you accept my code that is [08:29] @xorpad, No I did not use any Android on this phone [08:29] I will wait patiently for your version then ;) [08:29] Flohack cool:D [08:29] Expect it in a couple weeks i'm guessing [08:29] or at least a small preview of whats to come [08:30] I do hope my work meets your standards... I'm even commenting everything, and i never write comments [08:30] but i figure if I'm working on a project others also work on, they will need comments [08:30] so I'm making an effort to comment everything so it's more likely to get accepted [08:31] my last linux kernel contribution got rejected because it was pure assembly with no comments and they said they couldn't understand it [08:31] lol [08:31] Well for commenting, if a line of modification is clear from the code it does not need a comment. Just give a line of intention or reasoning, its easier for merging the pull request, to understand whats going on. But dont overcomment it ;) [08:31] yes this what you said 😆 [08:33] anyways I gotta reboot back into my other linux install to get back to work on the kernel... I had to boot into this one to reflash ubuntu touch because the update soft-bricked my phone(developemnt branch) [08:33] brb [08:35] Im in my office now so dont expect too much following the next hours ;) [08:35] Flohack, cool [08:52] To use apt you need libertine or to put your filesystem in r/w mode..you will lose your changes at every upgrade.. [08:53] I see... [09:36] Davy was added by: Davy [09:46] @Davy, welcome [09:49] Hi there, thank you! It took a while, but now even I know, that Cannonical doesn't support Touch anymore. That's why I am here [09:52] Cannonical aint got nothing on ubports [09:54] anyone who would abandon a new phone OS with ELF binaries instead of highly abstracted runtimes and vms with optimzing routines to make them seem less bloated, is insane [09:56] this OS is awesome [09:56] I do count on that ;) [09:57] man i made a big mess of the kernel lol [10:01] Davy, you a user or a coder? [10:01] Just a user. No help for the kernel from me ;) [10:02] haha, I don't expect help, I just wondering [10:02] have you got it working yet? [10:03] It is rebooting just yet. Gotta tell you in a minute [10:03] Ahh:D [10:03] which device you got? [10:03] Aquaris m10 [10:05] Oh, cool [10:05] I've got a Nexus 5 with Ubports on it [10:06] I love it [10:07] it's my new favorite toy [10:08] You know, the last few months I was wondering, why there weren't any updates. Didn't had the time to get informed. Today I read: no support anymore. Still like it so i am glad, that there is a comunityproject [10:08] Davy, you can choose between stable, RC and devel after initial install [10:10] In system settings. It is hidden down the bottom. Not obvious [10:10] okay I fixed the mess [10:11] now I have resulting mess related to the new code [10:11] That equals progress [10:11] Thanks for the hin, lionelb. Do I get that right? The ubuntustore is still the old unsupported one, right? so I should get a hang of the open store? [10:12] One mess at a time [10:13] Ubuntu Store is closing and not maintained. Check out Dekko2 and uMatriks on OpenStore [10:13] Alright, thanks! So far it looks pretty good! [10:14] Note pull down flashlight [10:15] Not that you have a flashlight on M10 haha [10:18] Well no use for that ;) [10:21] There are memory leak issues on M10 so may be performance deterioration over time. Dialler will go in OTA 3 I think [10:23] A alright, good to know. So far it still looking great! So much better than before. [10:25] It's awesome to the max yo [10:26] yay, another bug squashed... i wonder how many more pop up before i end this hunt [10:26] fixed, on, got 7 more once that one worked [10:26] lol [10:29] Good luck to you! gotta go do some working. Bye! [10:30] later:D [10:39] is there a generic carddav? Like calendars, or only by owncloud. [10:53] and it will be great if can sync contacts and task by nextcloud. [11:13] @Davy, Yes, it's great. Like you I am just a user. Run Touch on my M10 and my Nexus 5. Couldn't be happier. Can't imagine going back to Android now. It doesn't suit everybody but is ideal for me. Knowing that people who are cleverer than me are improving it all the time is very encouraging. [12:08] Is anyone also having issues conn3cting to the forum? [12:10] @demokrit atomos, Looks OK from here [12:10] Hm ok i'll check again [12:11] Nice, works again, thx for checking Dalton! [12:38] It seems that every other update (devel) to my OPO crashes or disables the indicator-rotation-lock. Does anybody know the terminal command to restart it? [12:42] (Photo, 1080x1920) https://irc.ubports.com/GVRHiZgx/file_960.jpg [12:44] Doesn't look right and the screen doesn't rotate. [12:48] devel version is the in development not even fully tested release [12:49] if you want stable don't go devel [12:49] I had devel and the last update soft-bricked my phone [12:49] I had to re-flash it [12:49] you gotta expect that from non-stable releases with RC updates [12:50] sometimes (often but not every time) my install is wonky after an update. no rotation, wifi is funny, no screen dimming etc. usually a reboot will fix it. life on th eedge is fun [12:51] I am aware of that and is taking the calculated risk. I switched to rc a few days ago, but it also had the problem with rotation, etc. Agreed, living on the edge is fun. Especially when eagerly awaiting fixes... [12:52] I haven't tried stable yet because so far (fingers crossed) had no soft brick with devel. Maybe odds are not in my favor... [12:53] I have been thinking rc sometimes, but I want to see that weekly updates are coming first. [12:55] Project daily-mako build #133: SUCCESS in 12 min: http://ci.ubports.com/job/daily-mako/133/ [12:55] Project ota-push build #1114: SUCCESS in 0.54 sec: http://ci.ubports.com/job/ota-push/1114/ [12:58] Does Ubuntu Touch comes with terminal. (Jk) 🙈🙈🙈 [12:59] Yes. The terminal is installed by default [13:03] @TomasOqvist, It rotates now after a restart. This seems somehow connected to the issue that the screen doesn't turn on with the power button. In those cases. power on followed by touching the screen turns it on. [13:17] @Jaume81, Only through syncevolution in cli so far, there is a script link somewhere in the wiki or forum [13:34] since OTA2 it can sync a generic caldav, it could be for contacts and task, too. In future. By nextcloud, I tried by owncloud and it don't sync but I will try again. [14:30] For Nextcloud/owncloud there is a bug on the login, it's has been fixed and will be released. [14:41] @xorpad, +1 Thank you for this attitude ! [14:41] bastos777, I'm just doing what I love to do, but glad you appreciate it [14:42] I've been merging in code from latest stable mainline linux kernel all night getting it to compile with gcc 6.4.0, which will make way for 7.x when it gets a little more mature and reliable [14:43] 7.x has way more performance than 4.8 which this code is meant for and 5/6 which this code doesn't work under because of stricter enforcment of behaviour and restrictions on undefined compiler behaviour [14:43] So when 7.3 hits, and we're ready to use it, we can get that 25-30% performance boost easily [14:44] 7.1 was hella buggy, 7.2 fixed 99% of the problems, and the next release, as x.3.0 always is, will be a stable reliable version we can use without worrying about certain things not working under certain circumstances [14:45]