**** BEGIN LOGGING AT Sat Apr 30 02:59:58 2016 Apr 30 05:33:21 Garfonso: You're busy at funny hours :P Apr 30 05:33:32 Is Settings working now with your PR? Apr 30 06:18:24 morphis: ping Apr 30 07:32:30 nizovn: Your PR sorts the issue I mentioned yesterday? Apr 30 07:32:40 Herrie: yes, it should Apr 30 07:32:54 nizovn: OK thnx :) Apr 30 07:33:39 nizovn: Needed some distraction from studying LOL ? Apr 30 07:36:46 yes, maybe. but i can easily lose all motivation to study... :) Apr 30 08:29:25 Tofe: ping Apr 30 08:29:28 morphis: ping Apr 30 08:55:46 Morning Apr 30 08:55:54 Herrie: my voicecall MR is now merged Apr 30 08:56:56 Tofe: Nice :) Apr 30 08:57:19 Tofe: You familiar with the porting process a bit? I got 2 suspicious messages and not sure I can ignore those... Apr 30 08:58:50 Herrie: settings is working on recent mako Apr 30 09:00:04 ah.. I see... download manager needs sqlite3, too ;) Apr 30 09:02:17 Garfonso: Yeah we're trying to sort sqlite3 issues Apr 30 09:03:24 yes, I read some chatting about this... but I didn't have time to look into it and probably don't have a clue anyway. ;) Apr 30 09:10:00 Herrie: it's been a very long time since I last did a port :p Apr 30 09:10:21 Herrie: but try me Apr 30 09:12:32 Tofe: https://bpaste.net/show/2e1d4c715ca2 Apr 30 09:12:38 Not sure I can ignore this? Apr 30 09:13:22 If it's only about recovery I'd say go on Apr 30 09:13:52 Also https://bpaste.net/show/3233dbbb726c, seems morphis explicity removed this dumpkey.jar with https://github.com/webOS-ports/android_system_core/commit/6b732c308a47a8821742fd83de5117455205f397 Apr 30 09:17:38 So I gues we don't need it, just I'm not familiar with the process :P Apr 30 09:17:51 We don't need it, but the error seems fatal Apr 30 09:18:06 maybe the dependency is now in several places Apr 30 11:11:41 Herrie: it looks like the fdinfo directory in /proc//fdinfo has a quite buggy behavior on mako Apr 30 11:12:01 Herrie: could you try something on maguro on adb? Apr 30 11:12:36 do a "ls -l /proc/self/fdinfo", and see if you get plenty of errors about symlinks Apr 30 11:14:12 looks fine on TP Apr 30 11:17:41 So either we find the commit in the kernel that would fix it, or we disable the call in systemd Apr 30 11:19:25 There's plenty of commits to choose from CM13: https://github.com/CyanogenMod/lge-kernel-mako/commits/cm-13.0/fs/proc so I'll just try disabling it in systemd :) Apr 30 11:22:25 The latter works Apr 30 11:23:55 Herrie: the only thing to do is to replace this line https://github.com/CyanogenMod/lge-kernel-mako/commits/cm-13.0/fs/proc with "r = -ENOENT;" Apr 30 11:24:12 sorry, this line: https://github.com/systemd/systemd/blob/v229/src/basic/mount-util.c#L57 Apr 30 11:44:59 Tofe: Charging my maguro ;) Apr 30 11:45:01 Will give it a go Apr 30 11:47:59 Herrie: I really don't see what is the root issue with the mako kernel, and I'm not that fond of debugging it... Let's do a patch for mako for systemd 229, shall we? Apr 30 11:48:49 Tofe: Yeah but it's good to know if it occurs on Maguro as well Apr 30 11:48:56 Since it may be a deeper issue Apr 30 11:49:18 As systemd 229 boots on maguro, I think you won't have the issue Apr 30 11:50:03 Tofe: Maybe not obviously but curious to see Apr 30 11:50:13 I'm going to paste output from Maguro anyway Apr 30 11:50:19 ok Apr 30 12:43:02 Tofe: You feel like giving updated webengine/chromium a go or should I build locally? Saw some interesting fixes, might solve the html5test.com crash :) Apr 30 12:43:11 I already prepared the branches Apr 30 12:51:07 https://github.com/webOS-ports/qtwebengine/commit/8704965f88a5ce44b182fe77dee27245415dccd9 and https://github.com/webOS-ports/qtwebengine-chromium/commit/6172a587e3bca2041953736e215b6bda06e11037 Apr 30 12:54:58 Tofe: This is on Maguro: https://bpaste.net/show/811c28348837 Apr 30 12:56:42 Herrie|2: maguro is fine Apr 30 12:57:07 Well, let's try the updated qtwebengine Apr 30 12:57:15 it's already merged anyway ;) Apr 30 12:58:17 Tofe: It's a separate branch for now to test Apr 30 12:58:26 Was able to clean merge Apr 30 12:58:33 Might also solve DougReaders file issue Apr 30 12:58:40 Based on some of the commits Apr 30 12:59:51 ok Apr 30 13:01:28 And maybe even pulseaudio not working, saw some changes to .pri as well if I remember correctly Apr 30 13:12:46 seems there is another bug in qtwebengine - some sites are rendered with wrong scale, e.g. reddit.com, also this can be seen in pivotce app. Apr 30 13:17:22 nizovn: That might be addressed too Apr 30 13:18:05 ok Apr 30 13:18:43 See https://github.com/webOS-ports/qtwebengine/commit/8704965f88a5ce44b182fe77dee27245415dccd9 and search for zoom. They're changing a few bits Apr 30 13:19:10 Setting a m_defaultZoomFactor of 1.0 for example Apr 30 14:13:56 hey, look. it's Herrie|2 Apr 30 14:36:51 He has quite a couple of doppelgangers Apr 30 14:45:53 Tofe: the build instructions totally fall apart once you get to the Kernel and OpenEmbedded part Apr 30 14:47:39 but i did manage to build the hammerhead portion Apr 30 14:48:37 https://gofile.me/2y9J3/YErQA9Ihm Apr 30 14:49:05 so, better instructions for the rest of the bits would be super crazy helpful Apr 30 14:50:48 HaDAk: what falls apart? what did you need to adapt? if you have access to the wiki, maybe you can note what needs attention/rewrite? Apr 30 14:51:55 i mean to say Apr 30 14:51:58 i don't know what to do next. Apr 30 14:52:07 the instructions are ambiguous Apr 30 14:52:31 i'll try to fix the defconfig part. that was awful to figure out. Apr 30 14:52:42 but now i'm on to creating the openembedded machine layer Apr 30 14:52:48 and it's just like "haha, good luck. read some docs." Apr 30 14:53:02 no, man. no. i don't want to "just go read some docs." i want some help in getting this shit going. Apr 30 14:53:45 i managed to "find" the mako.conf file. i had to dig through some repos that weren't really linked well (just a tiiiiny footnote). i copied that out, and started poking at it. Apr 30 14:53:51 took me a while to find the /boot partition Apr 30 14:53:57 which, by the way,... Apr 30 14:53:58 ls -al /dev/block/platform/msm_sdcc.1/by-name/|grep ' boot ' Apr 30 14:54:19 that'll give you the partition labeled /boot. unlikely to work on ALL devices, but worked for me Apr 30 14:54:32 now the part i'm stuck on is trying to figure out the rest of the values in this file Apr 30 14:55:12 i'm stuck trying to determine the partition size and flash block size Apr 30 14:55:20 and i don't know how to figure out what the serial console should be. Apr 30 14:55:37 otherwise, i *think* the rest of the information is correct...but again, no real reference to make sure i'm pulling the right into Apr 30 14:55:39 info* Apr 30 14:56:48 ok. friend is here. we're meeting for breakfast. bbl. Apr 30 16:11:52 * DougReeder checks in Apr 30 17:28:16 Tofe: Any luck with the webengine or should I kick off a build locally? Apr 30 17:38:56 morphis: ping Apr 30 18:48:00 Herrie: oh, I didn't understand I was to do something :p Apr 30 18:59:10 Tofe: I can put a build to run shortly so it'll be done tomorrow ;) Apr 30 19:00:54 Tofe: Seems they'll drop dependency on com.nokia.meego 2.0 from FingerTerm. Looks OK on qemux86, running IPK for Mako now. Can you test it once i have it? Apr 30 19:09:25 not sure if I'll be able to test tonight (some kind of weird party they do in Finland last day of April), but tomorrow yes Apr 30 19:10:58 Tofe: I can test it on Maguro, if it behaves there it should be fine :) Apr 30 19:11:12 yup Apr 30 19:11:25 Seems Yocto released Krogoth yesterday. I guess we can move shortly since you sorted systemd & PA8 issues :) Apr 30 19:11:38 I'll run a copy of unstable + latest QT 5.6 updates now Apr 30 19:11:42 And test tomorrow May 01 00:22:38 DougReeder: get your life in order, man. in/out/in/out/in/out...my window is filled with join/part messages **** ENDING LOGGING AT Sun May 01 02:59:58 2016