| *** mad_fitter <mad_fitter!~mad_fitte@rvr35113pubwifi023.nmu.edu> has quit IRC (Quit: irc2go) | 00:05 | |
| *** TheRealJohnGalt <TheRealJohnGalt!uid29986@gateway/web/irccloud.com/x-qnymczvxeljjzhwm> has quit IRC () | 01:32 | |
| *** TheRealJohnGalt <TheRealJohnGalt!uid29986@gateway/web/irccloud.com/x-evobimbbxlyqskdr> has joined #sailfishos-porters | 01:34 | |
| *** zhxt <zhxt!~zhxt@124.192.38.2> has quit IRC (Quit: Leaving) | 01:52 | |
| *** exadeci <exadeci!uid35778@gateway/web/irccloud.com/x-eantmjkynnfkbydx> has quit IRC () | 02:00 | |
| *** exadeci <exadeci!uid35778@gateway/web/irccloud.com/x-yvqildchcugjtmme> has joined #sailfishos-porters | 02:00 | |
| *** zhxt <zhxt!~zhxt@124.192.38.2> has joined #sailfishos-porters | 03:47 | |
| *** exadeci <exadeci!uid35778@gateway/web/irccloud.com/x-yvqildchcugjtmme> has quit IRC (Quit: Connection closed for inactivity) | 04:10 | |
| *** filippz <filippz!~filippz@93-143-195-115.adsl.net.t-com.hr> has joined #sailfishos-porters | 04:35 | |
| *** trt` <trt`!~trt@ip70-174-27-208.hr.hr.cox.net> has quit IRC (Read error: Connection reset by peer) | 04:37 | |
| *** filippz <filippz!~filippz@93-143-195-115.adsl.net.t-com.hr> has quit IRC (Ping timeout: 246 seconds) | 04:41 | |
| *** filippz <filippz!~filippz@93-143-195-115.adsl.net.t-com.hr> has joined #sailfishos-porters | 04:42 | |
| *** olafh_ <olafh_!~olafh@p4FF484CC.dip0.t-ipconnect.de> has joined #sailfishos-porters | 04:47 | |
| *** mariogrip <mariogrip!sid69323@gateway/web/irccloud.com/x-nrmdzvuiohmdawvn> has quit IRC () | 04:49 | |
| *** mariogrip <mariogrip!sid69323@gateway/web/irccloud.com/x-ofhwxhrhqrsanxrc> has joined #sailfishos-porters | 04:49 | |
| *** tbr <tbr!dm8tbr@cl-790.ham-01.de.sixxs.net> has quit IRC (Ping timeout: 246 seconds) | 05:01 | |
| *** vakkov <vakkov!~vakkov@vlan-94-190-247-69.comnet.bg> has joined #sailfishos-porters | 05:22 | |
| *** tbr <tbr!dm8tbr@cl-790.ham-01.de.sixxs.net> has joined #sailfishos-porters | 05:27 | |
| *** GodGinrai <GodGinrai!~godginrai@unaffiliated/godginrai> has quit IRC (Quit: g'night all) | 05:36 | |
| *** Makinit <Makinit!makinit@makinit.nl> has quit IRC (Ping timeout: 240 seconds) | 05:43 | |
| *** Makinit <Makinit!makinit@makinit.nl> has joined #sailfishos-porters | 05:45 | |
| locusf | morningles | 05:53 |
|---|---|---|
| muppis | Where you know when office building is too big? | 05:56 |
| muppis | When it rains on other side where on other side sun shines. | 05:56 |
| *** tanty_off is now known as tanty | 06:07 | |
| *** idej <idej!~idej@eduroam-89-175.zdv.uni-mainz.de> has joined #sailfishos-porters | 06:17 | |
| *** OrokuSaki <OrokuSaki!475bbaa0@gateway/web/freenode/ip.71.91.186.160> has joined #sailfishos-porters | 06:18 | |
| OrokuSaki | I have droid-sinc and fingerterm working in archlinux.. fingerterm is a lightdm option... I have a framebuffer console too... | 06:19 |
| OrokuSaki | and freedreno with x11 | 06:19 |
| OrokuSaki | dpi is off with fingerterm.. and I have been messing with "hawaii desktop" | 06:19 |
| OrokuSaki | and weston... which.. doesn't have working touchscreen but does have keyboard and mouse... that is all | 06:20 |
| OrokuSaki | =) | 06:20 |
| OrokuSaki | seeing if I can get gstreamer going.. it is broken.. just by itself | 06:20 |
| *** electrolux <electrolux!~Adium@192.194.199.34> has joined #sailfishos-porters | 06:28 | |
| *** tbr <tbr!dm8tbr@cl-790.ham-01.de.sixxs.net> has quit IRC (Ping timeout: 256 seconds) | 06:56 | |
| *** gabriel9|work <gabriel9|work!~bojan@9-133-241-92-ptp-stat.customer.blic.net> has joined #sailfishos-porters | 07:00 | |
| *** tbr <tbr!dm8tbr@cl-790.ham-01.de.sixxs.net> has joined #sailfishos-porters | 07:05 | |
| *** gabriel9 <gabriel9!~bojan@fasecuremail.com> has joined #sailfishos-porters | 07:05 | |
| *** gabriel9|work <gabriel9|work!~bojan@9-133-241-92-ptp-stat.customer.blic.net> has quit IRC (Ping timeout: 264 seconds) | 07:06 | |
| *** alin <alin!~alin@opensuse/member/ealin> has quit IRC (Quit: Konversation terminated!) | 07:06 | |
| * Nokius TGIF | 07:14 | |
| *** TheRealJohnGalt <TheRealJohnGalt!uid29986@gateway/web/irccloud.com/x-evobimbbxlyqskdr> has quit IRC (Quit: Connection closed for inactivity) | 07:42 | |
| *** alin <alin!~alin@opensuse/member/ealin> has joined #sailfishos-porters | 07:53 | |
| sledges | mroming! | 07:58 |
| sledges | muppis: lol! | 07:58 |
| sledges | and whilst you reach the other end, it stops raining?:)) | 07:59 |
| sledges | OrokuSaki: way to go foss man! | 07:59 |
| OrokuSaki | sledges: Thanks! my gst-droid got black listed.. trying again | 08:00 |
| sledges | PSA: i think jolla-camera should now work since 1.1.6, just need configuring: you need to adapt this file /etc/dconf/db/vendor.d/jolla-camera.hw.txt: https://pastee.org/9kmfm by looking at your cameraplus configs | 08:00 |
| sledges | have fun and report results pls! | 08:00 |
| sledges | maybe MSameer could confirm if i'm not talking jibberish ^ | 08:01 |
| muppis | sledges, yep. :D | 08:03 |
| sledges | s/jolla-camera.hw.txt/jolla-camera-hw.txt/ | 08:11 |
| *** alin <alin!~alin@opensuse/member/ealin> has quit IRC (Read error: Connection reset by peer) | 08:11 | |
| *** alin <alin!~alin@193.62.118.12> has joined #sailfishos-porters | 08:11 | |
| *** alin <alin!~alin@193.62.118.12> has quit IRC (Changing host) | 08:11 | |
| *** alin <alin!~alin@opensuse/member/ealin> has joined #sailfishos-porters | 08:11 | |
| *** gabriel9 <gabriel9!~bojan@fasecuremail.com> has quit IRC (Quit: Konversation terminated!) | 08:12 | |
| *** gabriel9|work <gabriel9|work!~bojan@9-133-241-92-ptp-stat.customer.blic.net> has joined #sailfishos-porters | 08:14 | |
| *** OrokuSaki <OrokuSaki!475bbaa0@gateway/web/freenode/ip.71.91.186.160> has quit IRC (Ping timeout: 246 seconds) | 08:15 | |
| *** tworaz <tworaz!~tworaz@212.180.237.71> has joined #sailfishos-porters | 08:15 | |
| *** gabriel9 <gabriel9!~bojan@fasecuremail.com> has joined #sailfishos-porters | 08:20 | |
| *** gabriel9|work <gabriel9|work!~bojan@9-133-241-92-ptp-stat.customer.blic.net> has quit IRC (Ping timeout: 255 seconds) | 08:21 | |
| *** TheRealJohnGalt <TheRealJohnGalt!uid29986@gateway/web/irccloud.com/x-ebedrmwqboivoeyx> has joined #sailfishos-porters | 08:25 | |
| carepack | ahoy | 08:36 |
| *** zajfy <zajfy!~zajfy@ip-144-188-230-46.dialup.ice.net> has joined #sailfishos-porters | 08:37 | |
| sledges | it's https://www.youtube.com/watch?v=kfVsfOSbJY0 ! \o/ | 08:38 |
| locusf | rr | 08:39 |
| locusf | shoulda known | 08:39 |
| locusf | :) | 08:39 |
| carepack | lol | 08:39 |
| carepack | hahahahha | 08:39 |
| *** alin <alin!~alin@opensuse/member/ealin> has quit IRC (Quit: Konversation terminated!) | 08:41 | |
| carepack | it's quiek, quiek, quiek, lllallla it's quiek quiek quiek | 08:42 |
| carepack | best comment on this: i never seen so much dislikes in my life | 08:44 |
| carepack | sledges: should I pr my changes and go for a look? | 08:46 |
| carepack | Before continue work on the next error.... | 08:46 |
| mal- | sledges: I'll try | 08:49 |
| mal- | the camera thing | 08:49 |
| mal- | sledges: I don't have jolla-camera-hw.txt only jolla-camera.txt, should I create that? | 08:53 |
| gabriel9 | is she porter? :D | 08:53 |
| dr_gogeta86 | sledges, i'm using legacy so i need the patch | 08:57 |
| sledges | carepack: sure! | 08:58 |
| sledges | mal-: jolla-camera.txt where? | 08:59 |
| dr_gogeta86 | sledges, i have some hint for get connected | 09:00 |
| dr_gogeta86 | use a proxy | 09:00 |
| dr_gogeta86 | like polipo or tinyproxy | 09:00 |
| dr_gogeta86 | is more platform indipendant | 09:00 |
| sledges | gabriel9: i hope she isn't, but carepack: i think the song has *so* many dislikes, just because it exploded into everyone just clicking "dislike" on a horrible song:)) confirmation bias | 09:00 |
| sledges | dr_gogeta86: i lost you | 09:00 |
| mal- | sledges: /etc/dconf/db/vendor.d/jolla-camera.txt | 09:01 |
| *** alin <alin!~alin@193.62.118.12> has joined #sailfishos-porters | 09:04 | |
| *** alin <alin!~alin@193.62.118.12> has quit IRC (Changing host) | 09:04 | |
| *** alin <alin!~alin@opensuse/member/ealin> has joined #sailfishos-porters | 09:04 | |
| sledges | mal-: name doesn't matter | 09:05 |
| sledges | jolla-camera.txt comes from jolla-camera package right? | 09:05 |
| sledges | so just add -hw next to it | 09:06 |
| mal- | ok | 09:06 |
| sledges | for hw specific bits | 09:06 |
| MSameer | sledges: 1.1.9 or greater | 09:07 |
| sledges | MSameer: dang, but commit about -hw.txt split is in since 1.1.4 | 09:08 |
| sledges | sorry guys :( | 09:08 |
| MSameer | sledges: but gstreamer 1.x is in 1.1.9 | 09:13 |
| *** zajfy <zajfy!~zajfy@ip-144-188-230-46.dialup.ice.net> has quit IRC (Ping timeout: 244 seconds) | 09:14 | |
| sledges | MSameer: in this community gstreamer1's been around since long tiem ;) | 09:15 |
| sledges | time | 09:15 |
| MSameer | sledges: gstreamer 1.x support in jolla camera | 09:15 |
| MSameer | gstreamer + qtmultimedia + jolla camera | 09:15 |
| sledges | understood | 09:16 |
| sledges | sorry for false alarm, but at least we'll know how to prepare for it;) | 09:17 |
| MSameer | sledges: the false alarm was not for me so no need to apologize :P | 09:19 |
| sledges | MSameer: i wasn't apologising to you :D | 09:19 |
| sledges | :P | 09:19 |
| *** zajfy <zajfy!~zajfy@ip-180-188-230-46.dialup.ice.net> has joined #sailfishos-porters | 09:28 | |
| *** phdeswer <phdeswer!~phdeswer@91-159-55-220.elisa-laajakaista.fi> has quit IRC (Ping timeout: 240 seconds) | 09:32 | |
| *** phdeswer <phdeswer!~phdeswer@91-159-55-220.elisa-laajakaista.fi> has joined #sailfishos-porters | 09:33 | |
| *** Nokius_ <Nokius_!~Nokius@p5DDB66CE.dip0.t-ipconnect.de> has joined #sailfishos-porters | 09:40 | |
| *** tbr <tbr!dm8tbr@cl-790.ham-01.de.sixxs.net> has quit IRC (Ping timeout: 240 seconds) | 09:40 | |
| *** Nokius <Nokius!~Nokius@p5DDB5135.dip0.t-ipconnect.de> has quit IRC (Ping timeout: 246 seconds) | 09:43 | |
| *** tbr <tbr!dm8tbr@cl-790.ham-01.de.sixxs.net> has joined #sailfishos-porters | 09:48 | |
| *** Tassadar <Tassadar!~tassadar@ip4-83-240-20-212.cust.nbox.cz> has joined #sailfishos-porters | 10:09 | |
| *** birdzhang <birdzhang!~birdzhang@183.61.112.4> has joined #sailfishos-porters | 10:20 | |
| *** idej <idej!~idej@eduroam-89-175.zdv.uni-mainz.de> has quit IRC (Ping timeout: 244 seconds) | 10:45 | |
| maikoool | is there a get.cm alternative for sailfish ports? | 10:46 |
| maikoool | with/without the jenkins infra behind it | 10:46 |
| sledges | nope | 10:47 |
| *** zhxt <zhxt!~zhxt@124.192.38.2> has quit IRC (Ping timeout: 264 seconds) | 10:47 | |
| maikoool | hm | 10:47 |
| maikoool | maybe it's time we got one? | 10:47 |
| sledges | http://images.devaamo.fi/sfe | 10:47 |
| sledges | is what we have only | 10:47 |
| sledges | and https://build.merproject.org/project/subprojects/nemo:devel:hw | 10:47 |
| maikoool | but obs doesn't build full zips, right? | 10:48 |
| sledges | but for complete jenkins (i.e. to use https://img.merproject.org/img/submit/ ) | 10:48 |
| sledges | we need to wait for lbt to re-add patterns support to obs | 10:49 |
| lbt | oh balls | 10:49 |
| sledges | o_O | 10:49 |
| lbt | just forgot about that :) | 10:49 |
| sledges | (/me didn't want to depict balls fwiw) | 10:49 |
| sledges | oh crumbs | 10:49 |
| sledges | just forgot to ping lbt daily about that :) | 10:49 |
| sledges | or thought mer merge needs to happen first ;p | 10:49 |
| *** zhxt <zhxt!~zhxt@124.192.38.2> has joined #sailfishos-porters | 10:51 | |
| maikoool | sledges: that jenkins server, can it handle the enormous amounts of diskspace required for continiously building images? | 10:55 |
| *** situ <situ!~situ@chordata.ucc.gu.uwa.edu.au> has quit IRC (Ping timeout: 255 seconds) | 10:57 | |
| birdzhang | When will Jolla release sailfish2.0 for nexus 5 :lovestruck: | 11:12 |
| Stskeeps | birdzhang: same with jolla1 rls i guess | 11:18 |
| birdzhang | months wait | 11:21 |
| carepack | do you think a rattling speaker on joll device is a case for warranty? It's hard to understand what's counterpart speaking during phone call.... | 11:28 |
| *** zajfy <zajfy!~zajfy@ip-180-188-230-46.dialup.ice.net> has quit IRC (Read error: Connection reset by peer) | 11:30 | |
| muppis | http://termite.apcdn.com/full/127001.jpg | 11:32 |
| *** gabriel9 <gabriel9!~bojan@fasecuremail.com> has quit IRC (Quit: Konversation terminated!) | 11:33 | |
| muppis | Hups, wrong channel. | 11:34 |
| *** zajfy <zajfy!~zajfy@ip-180-188-230-46.dialup.ice.net> has joined #sailfishos-porters | 11:35 | |
| *** birdzhang <birdzhang!~birdzhang@183.61.112.4> has quit IRC (Quit: Leaving.) | 11:36 | |
| *** gabriel9|work <gabriel9|work!~bojan@9-133-241-92-ptp-stat.customer.blic.net> has joined #sailfishos-porters | 11:38 | |
| *** gabriel9|work <gabriel9|work!~bojan@9-133-241-92-ptp-stat.customer.blic.net> has quit IRC (Ping timeout: 240 seconds) | 11:44 | |
| *** gabriel9 <gabriel9!~bojan@fasecuremail.com> has joined #sailfishos-porters | 11:44 | |
| Stskeeps | carepack: ask care | 11:57 |
| carepack | stskeeps: yep. Don't wanted to bug the care if somebody would definetely no warranty case. thx | 12:04 |
| mal- | I think I'll need to find a way for automatic keyboard backlight control, maybe to control it using the ambient light sensor | 12:06 |
| *** mad_fitter <mad_fitter!~mad_fitte@cpe890ripleyconstru1.nmu.edu> has joined #sailfishos-porters | 12:13 | |
| mal- | spiiroin: it seems that my keyboard backlight might need some modifications to mce, looks like it's quite device specific | 12:14 |
| maikoool | sledges: do you think there would be a use for an automated full build system, like get.cm with jenkins does, which would mainly provide zips and do builds if device repos are updated? | 12:15 |
| maikoool | cause my device porting isn't going well, i think i just don't have the required skills to do all this | 12:16 |
| maikoool | so i was thinks about building that buildsystem cause that's more up my street | 12:16 |
| mal- | spiiroin: my device has the backlight control in /sys/class/leds/keyboard-backlight/ instead of /sys/class/leds/keyboard/ | 12:18 |
| *** tbr <tbr!dm8tbr@cl-790.ham-01.de.sixxs.net> has quit IRC (Quit: maintenance) | 12:20 | |
| *** GodGinrai <GodGinrai!~godginrai@unaffiliated/godginrai> has joined #sailfishos-porters | 12:26 | |
| mad_fitter | Anyone know where all the hardware volume controls are in the system? I'd like to find a way to hack the system volumes, as all sounds (including message tone) are extremely quiet on my Nexus 5. | 12:33 |
| mal- | spiiroin: also there is a backlight for the back/home/settings buttons but I don't think that is very important | 12:55 |
| mal- | but anyway that is located at /sys/class/leds/button-backlight/ | 12:55 |
| *** situ <situ!~situ@chordata.ucc.gu.uwa.edu.au> has joined #sailfishos-porters | 13:05 | |
| *** vakkov <vakkov!~vakkov@vlan-94-190-247-69.comnet.bg> has quit IRC (Remote host closed the connection) | 13:07 | |
| *** vakkov <vakkov!~vakkov@vlan-94-190-247-69.comnet.bg> has joined #sailfishos-porters | 13:07 | |
| sledges | maikoool: sfa-mer already exists | 13:10 |
| sledges | could be maintained thiugh | 13:10 |
| sledges | maikoool: and you already done great with the port | 13:10 |
| maikoool | i was thinking more in terms of sfa-mer as a service | 13:10 |
| sledges | what problems? | 13:10 |
| maikoool | display wont work | 13:10 |
| maikoool | i just want to revisit device porting later | 13:10 |
| maikoool | i'm a devop so i like building websites and (build) infrastructure | 13:11 |
| maikoool | but if there's no use for a automated builder i wont spend time on it | 13:11 |
| maikoool | and when my device port is complete, i want to to autobuild anyway | 13:12 |
| maikoool | ;) | 13:12 |
| *** tbr <tbr!dm8tbr@cl-790.ham-01.de.sixxs.net> has joined #sailfishos-porters | 13:12 | |
| maikoool | so i may as well build that while waiting for new hadk-guide and 1.1.7 update | 13:12 |
| mal- | maikoool: when you added that QCOM_BSP to qt5-qpa-hwcomposer-plugin was it a clean source i.e. did you build it before, sometimes there might some compiled stuff left behind? | 13:13 |
| maikoool | i created a new vm two days ago | 13:14 |
| mal- | ok | 13:14 |
| maikoool | so everything has been done clean at least once | 13:14 |
| maikoool | i'll recollect all the logs and post them here in a bit to see if maybe something comes up | 13:14 |
| *** mgrover <mgrover!d4fafbb5@gateway/web/freenode/ip.212.250.251.181> has joined #sailfishos-porters | 13:15 | |
| mal- | maikoool: do you still have the problem with droid-hal-init? | 13:16 |
| maikoool | yes | 13:16 |
| maikoool | segfaults | 13:16 |
| mgrover | gotta love the segfaults | 13:16 |
| mal- | you should find a fix to that first because you never know what it might cause | 13:16 |
| maikoool | lot of time when i remove or comment a line which is failing it fails a line before that | 13:17 |
| maikoool | which is strange | 13:17 |
| maikoool | i'm reflashing my device to ensure latest firmware is installed to rule out that (i wasnt sure which firmware version it had) | 13:18 |
| maikoool | so i'll go on debugging in a few minutes | 13:18 |
| *** beidl_ <beidl_!~quassel@194-118-188-88.adsl.highway.telekom.at> has joined #sailfishos-porters | 13:23 | |
| *** beidl <beidl!~quassel@188-22-92-40.adsl.highway.telekom.at> has quit IRC (Ping timeout: 240 seconds) | 13:25 | |
| *** mad_fitter <mad_fitter!~mad_fitte@cpe890ripleyconstru1.nmu.edu> has quit IRC (Quit: irc2go) | 13:27 | |
| *** mgrover <mgrover!d4fafbb5@gateway/web/freenode/ip.212.250.251.181> has quit IRC (Quit: Page closed) | 13:28 | |
| sledges | maikoool: automated local build already exists: https://github.com/dmt4/sfa-mer | 13:44 |
| sledges | just needs going modular properly, and then adding devices to it | 13:44 |
| maikoool | yeah, i'd like to provide that as a managed and hosted service, that also publicly makes the resulting zip available | 13:46 |
| maikoool | in a nice list per device with build logs and history | 13:46 |
| maikoool | i got myself a way too big dedi some time ago, still need to put that too god use | 13:47 |
| sledges | dedi? | 13:50 |
| sledges | when patterns are in obs, we'll just use img | 13:50 |
| sledges | i guess you'll be able to POST to that one | 13:50 |
| maikoool | dedicated server | 13:51 |
| maikoool | but img makes zipfile from rpm packages built on obs? | 13:54 |
| maikoool | and make hybris-hal is done on obs? | 13:54 |
| alin | sledges: the modular shall be in my branch | 13:56 |
| alin | maikoool: no at the moment the hybris is done locally and then we upload them to obs | 13:57 |
| sledges | alin: yes, needs a push, but first its behaviour needs to reach consent | 13:59 |
| sledges | we should do this build unsupervised, at least after initial questions (go modular, select pixel ratio) | 13:59 |
| sledges | for all subsequent builds | 13:59 |
| *** electrolux <electrolux!~Adium@192.194.199.34> has quit IRC (Quit: Leaving.) | 14:01 | |
| *** mad_fitter <mad_fitter!~mad_fitte@rvr35114pubwifi177.nmu.edu> has joined #sailfishos-porters | 14:04 | |
| sledges | mad_fitter: with volume up to the max and ringtone volume up to the max it sounds pretty loud | 14:06 |
| mad_fitter | sledges: Everything else seems fine except my notifications tone. Weird, but not a big deal. | 14:07 |
| sledges | mad_fitter: i just tested on settings->sounds->jolla-ringtone->play button | 14:07 |
| sledges | (=click on "current ringtone") | 14:07 |
| mad_fitter | Odd. Played there its louet.d as hell, but when I actually get a text, its super qui | 14:08 |
| maikoool | building hybris is probably also the most heavy as in buildsystem load and diskspace | 14:10 |
| sledges | mad_fitter: you tried same text ringtone? | 14:10 |
| sledges | and volume rocker is to the max? | 14:10 |
| maikoool | assuming that you dont need the android tree after making that package | 14:10 |
| mad_fitter | sledges: Indeed. | 14:10 |
| sledges | maikoool: i just managed to reduce fresh aosp5 checkout to 14G, with only the needed repos | 14:12 |
| sledges | same can be done for 11.0 etc | 14:12 |
| maikoool | how much disk in use after building? | 14:12 |
| mad_fitter | sledges: Figured it out. When it's set from the system, it works fine...but when you apply an ambience, it somehow limits the volume. | 14:12 |
| mal- | mad_fitter: ambience has it's own volume setting also | 14:14 |
| sledges | maikoool: 14G sources, 2.2G kernel+build products, 353M out/ (without kernel, as in aosp kernel is built out-of-tree) = 19G | 14:14 |
| mad_fitter | sledges: Really? Didn't know that, lol. Good to know ;) | 14:14 |
| maikoool | hm | 14:15 |
| maikoool | mine's 32G | 14:15 |
| maikoool | includes kernel | 14:15 |
| maikoool | but that's cm11 | 14:15 |
| mal- | cm11 is not reduced very well | 14:15 |
| sledges | yep | 14:15 |
| sledges | the technique of reducing is being polished | 14:16 |
| sledges | let me push it out a bit | 14:16 |
| mal- | sledges: just curious, have you tried to compile hwcomposer with those reduced versions? | 14:17 |
| sledges | mal-: reduced tree? yes | 14:18 |
| mal- | ok | 14:18 |
| *** idej <idej!~idej@eduroam-89-175.zdv.uni-mainz.de> has joined #sailfishos-porters | 14:38 | |
| sledges | PSA: WIP reduce manifest HOWTO: http://piratepad.net/port-reduce-manifest | 14:42 |
| maikoool | k so here output from droid-hal-init strace | 14:48 |
| maikoool | https://pastee.org/tnesm | 14:48 |
| maikoool | i'm assuming the error is in default.prop | 14:48 |
| *** mad_fitter <mad_fitter!~mad_fitte@rvr35114pubwifi177.nmu.edu> has quit IRC (Quit: irc2go) | 14:48 | |
| alin | sledges: but even that 14GiB checkout is still a lot | 14:50 |
| alin | i wonder is is not a smarter move to clone all with depth=0 or 1 | 14:51 |
| alin | that hall reduce the size of the initial checkout | 14:51 |
| alin | the problem is if one needs to change branches and other points back into history | 14:52 |
| mal- | maikoool: just to be sure, is /system mounted properly before droid-hal-init? | 14:53 |
| maikoool | i'm running it manually | 14:54 |
| maikoool | /system is mounted : /dev/mmcblk0p13 on /system type ext4 (ro,relatime,discard,data=ordered) | 14:54 |
| maikoool | it is ro though | 14:54 |
| mal- | it should be ro | 14:55 |
| mal- | maikoool: and /system seems to contain correct files? | 14:55 |
| mal- | maikoool: when you manually start droid-hal-init, how do you do it, from command line? are all of these done before https://github.com/mer-hybris/droid-system-packager/blob/master/bin/droid-hal-startup.sh | 14:57 |
| maikoool | k so i downloaded that script | 14:58 |
| maikoool | ran it | 14:58 |
| maikoool | same error | 14:58 |
| maikoool | /system looks fine to me | 14:59 |
| maikoool | /dev/mmcblk0p13 1.2G 449M 763M 38% /system | 14:59 |
| sledges | alin: i'm happy with reduction from 57G full aosp5 | 14:59 |
| mal- | sledges: where is the source of droid-hal-init? | 15:00 |
| vakkov | mal-: https://github.com/mer-hybris/droid-hal-device/pull/38 | 15:01 |
| vakkov | agh, shit it was not that :D | 15:02 |
| alin | sledges: the repo is dead | 15:03 |
| vakkov | mal-: hybris-boot maybe | 15:03 |
| sledges | alin: no | 15:04 |
| sledges | mal- droid-hal-init is renamed android's init | 15:05 |
| sledges | simples:) | 15:05 |
| sledges | built in-tree | 15:05 |
| sledges | in HABUILD_SDK | 15:05 |
| maikoool | https://github.com/mer-hybris/android_system_core ? | 15:06 |
| mal- | sledges: ok | 15:06 |
| alin | sledges: ? | 15:06 |
| alin | I tried to clone and says does not exist | 15:06 |
| *** jocarog <jocarog!4937b789@gateway/web/freenode/ip.73.55.183.137> has joined #sailfishos-porters | 15:07 | |
| sledges | alin: because syntax is wrong | 15:08 |
| sledges | alin: fixed | 15:08 |
| mal- | maikoool: are there any other services that fail? pastebin whole kernel and journal logs | 15:08 |
| alin | sledges: and in this one... can you put it to master rather than modular https://github.com/mer-hybris/dhd2modular | 15:09 |
| alin | is tested now | 15:09 |
| sledges | mal-: so we're going with keeping these for aosp5 (possibly cm12). feel free to remove the lines of repos that aren't needed from your findings when you reduced cm11 back in a day | 15:10 |
| sledges | https://github.com/sledges/android/blob/hybris-aosp-5.1.0_r5/keep.pathlist.d/base.pathlist | 15:10 |
| *** gabriel9 <gabriel9!~bojan@fasecuremail.com> has quit IRC (Quit: Konversation terminated!) | 15:12 | |
| maikoool | mal-: dmesg output: http://pastebin.com/ZDtiakNU, journalctl output: http://pastebin.com/zzmabRMg, systemctl output: http://pastebin.com/CNq51Pj0 | 15:13 |
| jocarog | quit | 15:14 |
| *** jocarog <jocarog!4937b789@gateway/web/freenode/ip.73.55.183.137> has left #sailfishos-porters | 15:14 | |
| sledges | alin: done on dhd2modular | 15:15 |
| alin | sledges: thanks | 15:15 |
| alin | I think I shall add my new functions and helpers there | 15:15 |
| alin | on a blue moon | 15:15 |
| mal- | maikoool: those logs don't show any errors with droid-hal-init | 15:16 |
| maikoool | i could swear it said failing two boots ago | 15:19 |
| maikoool | and the stracktrace at line 127 is the same as when i run droid-hal-init manually | 15:21 |
| sledges | maikoool: just to be on same page: what did you change that from a working port you went to a crashing one? | 15:22 |
| maikoool | it never worked | 15:22 |
| maikoool | lipstick still wont boot the first two times | 15:23 |
| maikoool | and the apps keep crashing if it does boot | 15:23 |
| mal- | sledges: if I have understood correctly lipstick causes oom-killer to activate for some reason | 15:23 |
| *** nh1402 <nh1402!~nh1402@95.148.89.71> has joined #sailfishos-porters | 15:23 | |
| maikoool | i had a port that lipstick worked on but that is not this device | 15:24 |
| maikoool | working port device battery died or something, it was already wonky but now it's really dead | 15:24 |
| *** exadeci <exadeci!uid35778@gateway/web/irccloud.com/x-buncxvdmilkhgmnb> has joined #sailfishos-porters | 15:24 | |
| sledges | maikoool: which devices is this then? | 15:25 |
| maikoool | sony xperia sp | 15:25 |
| sledges | and the one before? | 15:25 |
| maikoool | codenamed huashan | 15:25 |
| maikoool | htc sensation (pyramid) | 15:25 |
| sledges | ah ok | 15:25 |
| sledges | thanks for refreshing my mem:) | 15:25 |
| maikoool | and to complete the story | 15:25 |
| maikoool | i also have a xperia z3 compact and for that i need aosp5 | 15:26 |
| sledges | maikoool: you said lipstick/apps crashing, at random. how does d-h-i come into this picture then? | 15:26 |
| maikoool | so busy with xperia sp | 15:26 |
| maikoool | well, i started looking for other failed services | 15:26 |
| sledges | sure, we'll get there | 15:26 |
| sledges | maikoool: does d-h-i fail on every boot? | 15:26 |
| mal- | sledges: minimer works fine on maikoool's device, only lipstick causes problems | 15:27 |
| sledges | mal-: maikoool: have you tried qml-composer ? | 15:27 |
| maikoool | ? | 15:28 |
| sledges | as lipstick is doing compositing | 15:28 |
| sledges | minimer doesn't, it's just a test for hwc | 15:28 |
| sledges | qml-composer does compositing | 15:28 |
| sledges | and source is simplified to isolate the case | 15:28 |
| mal- | sledges: no, didn't know about that | 15:28 |
| maikoool | neither did i | 15:29 |
| maikoool | it's not in the hadk-faq piratepad, right? | 15:29 |
| sledges | it's not | 15:30 |
| sledges | because usually lipstick problems can be tackled | 15:30 |
| sledges | by inspecting fence leaks | 15:30 |
| sledges | so let's get fence leaks out of the way first | 15:30 |
| sledges | maikoool: did you run the ... | wc -l i pasted earler? | 15:31 |
| mal- | sledges: do fence leaks prevent lipstick from starting? | 15:31 |
| sledges | fence leaks start when lipstick starts | 15:31 |
| sledges | and when it runs out of them, it gets killed | 15:31 |
| sledges | first couple of apps get | 15:31 |
| *** tworaz <tworaz!~tworaz@212.180.237.71> has quit IRC (Ping timeout: 260 seconds) | 15:31 | |
| sledges | ^ this is mako on cm11.0 | 15:32 |
| sledges | mako on cm10.1 has no problems | 15:32 |
| *** tanty is now known as tanty_off | 15:32 | |
| maikoool | gets up to 96 | 15:32 |
| sledges | but let's first ensure those are the fences that leak | 15:32 |
| maikoool | then crashes | 15:32 |
| sledges | hmmmm | 15:32 |
| sledges | what are the contents? | 15:32 |
| sledges | do they mention fences? | 15:32 |
| sledges | im sure it will get to different numbers every time | 15:33 |
| mal- | 96 doesn't sound much, my device shows 123 | 15:34 |
| maikoool | so i'm running ls /proc/$(pidof lipstick)/fd/ | wc -l | 15:35 |
| maikoool | can i cat those files | 15:35 |
| maikoool | in that folder? | 15:35 |
| mal- | same command here | 15:35 |
| sledges | maikoool: i mean, what names of those files | 15:37 |
| maikoool | theyre all numbers | 15:37 |
| sledges | hmmm | 15:37 |
| maikoool | ah | 15:37 |
| sledges | then you have no fence leaks | 15:37 |
| maikoool | ls -la | 15:37 |
| maikoool | 4 occurences | 15:37 |
| maikoool | of sync_fence | 15:38 |
| sledges | no leaks then | 15:38 |
| maikoool | no, since i've now ran lipstick two times | 15:38 |
| maikoool | the next time it runs | 15:38 |
| sledges | maikoool: mal-: then: http://www.merproject.org/logs/%23sailfishos-porters/%23sailfishos-porters.2015-07-22.log.html#t2015-07-22T20:19:04 | 15:38 |
| sledges | (worth putting this nicely into faq) | 15:39 |
| maikoool | folder doesnt exist | 15:40 |
| maikoool | /usr/lib/qt5/examples does | 15:40 |
| maikoool | no contents in there | 15:40 |
| mal- | maikoool: zypper in qt5-qtwayland-wayland_egl-examples | 15:42 |
| *** Antartica <Antartica!~Antartica@90.69.50.20> has joined #sailfishos-porters | 15:42 | |
| Antartica | mal-: I have a .kmap for the XperiaPro (converted from my layout for Android) | 15:43 |
| mal- | Antartica: I have also made that | 15:43 |
| mal- | partial at least | 15:44 |
| Antartica | mal-: pm8058.kmap at http://pastebin.com/YFMjY0vj | 15:44 |
| Antartica | Hope mine helps to complet yours :) | 15:45 |
| Antartica | ...complete... | 15:45 |
| Antartica | I've done a lua script to convert a .kl/.kcm pair for android to a linux .kmap | 15:46 |
| Antartica | mal-: The script is at http://mangaroboto.com/notmangaroboto/klkcm2kmap-v0.tgz | 15:47 |
| mal- | Antartica: http://pastebin.com/L2NsVhiV this is my version | 15:48 |
| maikoool | sledges: when i run ./qml-compositor -platform hwcomposer it says Fatal: Failed to open server socket | 15:49 |
| *** idej <idej!~idej@eduroam-89-175.zdv.uni-mainz.de> has quit IRC (Quit: Leaving) | 15:51 | |
| Antartica | mal-: perfect. Now checking if I've lost any keycode in the conversion... | 15:52 |
| maikoool | ah nvm | 15:52 |
| maikoool | was not running as user nemo | 15:52 |
| mal- | Antartica: I made that manually, some of the ones at the end are lines from the version I used as a base | 15:52 |
| mal- | and some of the extra keys are in a separate file | 15:53 |
| Antartica | mal-: ok. Just spotted a bug in my converter... (lines start with key instead of keycode). Arg! | 15:55 |
| maikoool | k so i ran that, it shows stuff on screen | 15:58 |
| maikoool | ls /fd | wc -l on that process shows 53 | 15:58 |
| maikoool | after a few moments though, its killed | 15:58 |
| sledges | Antartica: hi, are you also running sfos? | 16:00 |
| maikoool | however, same behaviour as from lipstick applies | 16:01 |
| Antartica | mal: fixing the key/keycode mismatch, both cover the same keycodes (yours have Insert/PageUp/Remove/Ens/PageDown that I suppose are from the template you used) | 16:01 |
| maikoool | the third+ time i start it, it keeps running | 16:01 |
| sledges | maikoool: cool, though i presume you have had BT of lipstick already? | 16:01 |
| Antartica | sledges: no. I want to run sfos in my Xperia Pro. Patiently waiting for mal- to release a image :) | 16:02 |
| maikoool | i did a strace, gdb bt does not show much | 16:02 |
| sledges | mal-: you see, pressure's mounting:) | 16:04 |
| sledges | maikoool: you could bt on qml-compositor now | 16:04 |
| sledges | debuginfo you can download too for that | 16:05 |
| mal- | Antartica: oh, then I might release one soon, I didn't remember that anyone was waiting :) | 16:05 |
| *** electrolux <electrolux!~Adium@d0rktbvlyrqx80tfj9k-3.rev.dnainternet.fi> has joined #sailfishos-porters | 16:05 | |
| sledges | maikoool: also as mal- said it's oom kicking in, so worth running valgrind on qml-compositor (it just sounds scary;) | 16:05 |
| Antartica | mal-: that would be greatly appreciated :) | 16:06 |
| Antartica | mal-: my layout, fixed: http://pastebin.com/Zgkm0ZyV | 16:06 |
| mal- | I have been doing final tinkering of basically insignificant problems | 16:06 |
| Antartica | mal-: would you mind testing it? | 16:06 |
| maikoool | sledges: Program terminated with signal SIGKILL, Killed., bt: No stack | 16:06 |
| mal- | Antartica: keycode 102 = Delete looks odd | 16:07 |
| Antartica | mal-: my bad. 102 is HOME | 16:08 |
| sledges | mal-: you'd be surprised ;) the number of times people come by and we have to say "sorry the port has had an image done, but porter left and took all porting source with him, so you can't even reproduce", even though i keep poking everyone to upstream asap ;) | 16:08 |
| sledges | submit patches | 16:08 |
| sledges | maikoool: valgrind time | 16:08 |
| maikoool | it's installing | 16:09 |
| maikoool | assuming zypper in valgrind is enough | 16:09 |
| sledges | yes | 16:09 |
| Antartica | mal-: what would be the correct mapping for MENU/HOME/BACK? I have it as an odd SuperL/Delete/Escape, but don't know what are the best keys of sfos... | 16:09 |
| Antartica | ...for sfos... | 16:10 |
| mal- | sledges: btw, is there some way to get my image to devaamo? not sure who manages that | 16:11 |
| sledges | tbr: ^ pls | 16:11 |
| mal- | Antartica: I have pageup, home, pagedown | 16:12 |
| sledges | oleid was an exception to the porters-gone trend: https://wiki.merproject.org/wiki/Adaptations/libhybris/Install_SailfishOS_for_yuga ;) | 16:12 |
| tbr | sledges: r | 16:12 |
| tbr | mal-: please message me your preferred user name and the directory name that I should create | 16:13 |
| mal- | tbr: mal, and I think folder should be iyokan, although I have plans for images other devices of the 2011 sonys also | 16:15 |
| tbr | well we'll see once you have those other images. | 16:17 |
| Antartica | mal-: layout fixed with that; http://pastebin.com/tpxx4naD ; sources at http://mangaroboto.com/notmangaroboto/klkcm2kmap-v2.tgz | 16:17 |
| vakkov | tbr: I'd like to ask you for a log in/account or whatever it is to devaamo so that i can upload a maguro image | 16:18 |
| mal- | Antartica: I was also thinking what would be good usage for those buttons | 16:18 |
| *** r0kk3rz <r0kk3rz!~chatzilla@host-92-30-149-11.as13285.net> has joined #sailfishos-porters | 16:20 | |
| maikoool | sledges: i'm not sure how to interpret the output of valgrind, i ran EGL_PLATFORM=hwcomposer QT_QPA_PLATFORM=hwcomposer valgrind --leak-check=yes ./qml-compositor -platform hwcomposer -plugin evdevtouch:/dev/input/event7 -plugin evdevkeyboard:keymap=/usr/share/qt5/keymaps/droid.qmap 2>&1 | tee ~/valgrind.log and the output is here: http://pastebin.com/aTXXWdCg | 16:21 |
| tbr | one time passwords messaged | 16:23 |
| tbr | please change upon logging in via ssh | 16:23 |
| tbr | you can copy over images with scp, sftp, rsync to /var/www/devaamo_images/sfe/$yourdirectory | 16:23 |
| tbr | I'll leave it to sledges to explain to you both about the requirements to such images: basically, no proprietary binaries, nothing that's not redistributable | 16:24 |
| tbr | bbl | 16:25 |
| sledges | thanks tbr r :) | 16:25 |
| sledges | yep, no sw mp3 codecs, nor non-free text prediction, neither non-free exchange support | 16:27 |
| Antartica | mal-: The keys for next page and previous page in fbreader would be good candidates for those keys O:) | 16:27 |
| sledges | oh and no bundles from /vendor/ | 16:27 |
| mal- | Antartica: does that work with page up/down? | 16:28 |
| mal- | Antartica: I'll see if I manage to build this weekend an image that work well enough, or maybe I should just lower my standards | 16:30 |
| sledges | maikoool: while i run similar test on nexus5, could you just run "free -m" continuously during qml-compositor launch, to see if memory is eaten so oom killer is justified | 16:32 |
| Antartica | mal-: pageUp/PageDown seems as the safest bet. | 16:32 |
| sledges | maikoool: also: - To disable it for a particular process, set value of the VM parameter "/proc/{pid}/oom_adj" to "-17". This will prevent this process from getting selected by OOM killer under low memory conditions. | 16:33 |
| maikoool | oom killer is not justified | 16:34 |
| maikoool | 250mb ram in use of 840 | 16:34 |
| maikoool | after lipstick crash 182 in use of 840 | 16:34 |
| Antartica | mal-: Don't lower your standards too much. If you manage to have an image by the end of August I'll be very happy. | 16:35 |
| maikoool | sledges: when oom killer is disabled, device reboots when starting lipstick instead of failing | 16:36 |
| mal- | Antartica: last image I built was almost ready (except for keyboard layout stuff) but missed some preinstalled packages | 16:36 |
| maikoool | well not reboot | 16:36 |
| maikoool | just hangs | 16:36 |
| *** ced117_ <ced117_!~ced117@AStrasbourg-552-1-10-109.w92-130.abo.wanadoo.fr> has joined #sailfishos-porters | 16:36 | |
| maikoool | telnet is closed, nothing responds, nothing on screen | 16:36 |
| mal- | maikoool: that sounds still like oom-killer | 16:37 |
| Antartica | mal-: of course if you put an image this weekend I'll try it ASAP :) | 16:37 |
| maikoool | i did run echo -17 > /proc/$(pidof lipstick)/oom_adj | 16:37 |
| maikoool | as suggested | 16:37 |
| mal- | Antartica: which language layout do you have? | 16:38 |
| *** ced117_ is now known as ced117 | 16:38 | |
| *** ced117 <ced117!~ced117@AStrasbourg-552-1-10-109.w92-130.abo.wanadoo.fr> has quit IRC (Changing host) | 16:38 | |
| *** ced117 <ced117!~ced117@opensuse/member/ced117> has joined #sailfishos-porters | 16:38 | |
| Antartica | mal-: That layout I've posted is mostly a US layout with some spanish specific symbols added (Ñ¡ dead_acute). | 16:39 |
| mal- | ok, maybe I'll leave the default layout for now and let users replace it with what they want until proper support for changing layout is available | 16:40 |
| Antartica | mal-: But the layout you are using is very basebones, it lacks a lot of useful symbols for using a shell | 16:41 |
| sledges | maikoool: could you run while 1; do sleep 0.1; free -m; done | 16:43 |
| sledges | in a separate window? | 16:43 |
| sledges | with oom disabled | 16:43 |
| Antartica | mal-: Perhaps mine is a little overkill, but at least "|<>$" are needed (IMHO) | 16:44 |
| mal- | Antartica: my layout only includes the keys seen on my keyboard | 16:44 |
| maikoool | ah yep | 16:44 |
| Antartica | s/basebones/barebones/g | 16:44 |
| maikoool | it shoots up like crazy | 16:44 |
| maikoool | from 300 to 400 to 750 in 3 ticks | 16:45 |
| maikoool | then device closes telnet | 16:45 |
| mal- | something is leaking badly | 16:45 |
| maikoool | i was running watch free -m | 16:45 |
| maikoool | 1 second is not enough to notice the ram use going up | 16:46 |
| maikoool | damn | 16:46 |
| mal- | maikoool: if you run minimer does it seem to be leaking? | 16:46 |
| maikoool | lemme try | 16:46 |
| mal- | to narrow down the reason | 16:46 |
| Antartica | mal-: the xperia pro I have has a german keyboard so it lacks adiaeresis aring odiaeresis | 16:47 |
| maikoool | mal-: minimer ram use is stable | 16:48 |
| maikoool | uses less than 30 mb ram | 16:48 |
| sledges | crumbs | 16:49 |
| mal- | interesting, since german language has ä and ö, maybe it has the key for adding the dots | 16:49 |
| maikoool | qml-compositor does the same | 16:49 |
| maikoool | ram shoots up then oom-killer comes | 16:49 |
| sledges | maikoool: have you tried fbdev instead of hwcomposer ? | 16:50 |
| maikoool | doesnt work | 16:50 |
| maikoool | permission denied | 16:51 |
| maikoool | ERROR: failed to open framebuffer: (Operation not permitted) | 16:51 |
| sledges | eglfs? | 16:51 |
| *** electrolux <electrolux!~Adium@d0rktbvlyrqx80tfj9k-3.rev.dnainternet.fi> has quit IRC (Quit: Leaving.) | 16:51 | |
| sledges | or something | 16:51 |
| sledges | need to grep recent logs | 16:51 |
| mal- | Antartica: anywayl, a single layout is not enough so I'll just add instructions how to set the correct one manually for now | 16:51 |
| Antartica | mal-: yes, it is strange. perhaps they went for a "clean" look. Anyway, in android the keyboard has some "bonus symbols" in fn+key :) | 16:52 |
| Antartica | mal-: perfect. | 16:52 |
| mal- | Antartica: I'll need to google how all of the layouts look and then define those manually | 16:53 |
| mal- | Antartica: btw there was an idea that the symbol key would bring up a symbol keyboard to the display like it does on android (is I remember correctly) | 16:53 |
| mal- | but this was just an idea, nothing implemented yet | 16:54 |
| Antartica | mal-: I always though that "that symbol key" was very cumbersome. I mapped mine to control (so I could type Control+C). Even my wife (has another Xperia Pro) agreed. | 16:55 |
| maikoool | [F] init_platform:917 - This application failed to start because it could not find or load the Qt platform plugin "eglfs". | 16:55 |
| maikoool | ah nvm | 16:56 |
| maikoool | theres a package for that | 16:56 |
| mal- | Antartica: could be, but still there is the problem of keyboard not visually showing what extra symbol for example you have defined to the kmap | 16:56 |
| Antartica | It is easier to do a key combo than using SYM+select the symbol on screen. | 16:56 |
| mal- | might be difficult to remember where everything is | 16:56 |
| sledges | yes | 16:57 |
| mal- | I agree that the selection screen on android was not very practical always | 16:57 |
| mal- | the real question is why the keys don't have a third usage like normal keyboard often have | 16:59 |
| *** zajfy <zajfy!~zajfy@ip-180-188-230-46.dialup.ice.net> has quit IRC (Read error: Connection reset by peer) | 16:59 | |
| Antartica | mal-: in my layout I've put things where the label reminds of the extra symbol ('' in 'fn+n', '' in 'FN+?', etc) but I know that is not possible (for example, < and > is on FN+a and FN+z...) | 16:59 |
| Antartica | ..not possible in a lot of cases | 17:00 |
| giucam | maikoool: can you try the qwindow-compositor example? | 17:00 |
| mal- | Antartica: I would have put < and > to ( and ), although [ and ] are also useful for that place | 17:01 |
| sledges | giucam: wow that's new, thanks:) | 17:03 |
| giucam | not really :P | 17:03 |
| sledges | to this channel;) | 17:03 |
| *** zajfy <zajfy!~zajfy@ip-180-188-230-46.dialup.ice.net> has joined #sailfishos-porters | 17:05 | |
| maikoool | giucam: does the same | 17:05 |
| Antartica | mal-: There was a thread in XDA with key layouts: http://forum.xda-developers.com/showthread.php?t=1690889 | 17:05 |
| maikoool | runs a few seconds, then memory goes out of control and process gets killed | 17:05 |
| maikoool | trying with QT_QPA_PLATFORM=eglfs results in a segfault around send(30, "\2\0\0\0debug.egl.force_msaa\0\0\0\0\0\0\0\0"..., 128, 0) = 128 | 17:06 |
| giucam | maikoool: can you start it in gdb and break after a second? | 17:07 |
| Antartica | mal-: I think my script should work on those (the layouts posted there seem to be in .kl/.kcm format) | 17:07 |
| maikoool | euhm | 17:08 |
| maikoool | how do i do this? | 17:08 |
| sledges | Ctrl+C, and go back to =hwcomposer | 17:09 |
| maikoool | k did that | 17:09 |
| maikoool | i'm now in gdb interactive prompt | 17:09 |
| giucam | bt | 17:10 |
| maikoool | https://pastee.org/ptqey | 17:11 |
| maikoool | hm screen won't clear | 17:12 |
| giucam | ok, try t a a bt | 17:12 |
| maikoool | isn't 1 second a bit random? | 17:12 |
| maikoool | would it be better to wait before something appears on screen? | 17:12 |
| giucam | if it usually does, probably yes | 17:13 |
| giucam | yeah, with 1 second i mean before it goes out of control | 17:13 |
| maikoool | check | 17:13 |
| giucam | but the latest the better | 17:13 |
| maikoool | that's good | 17:14 |
| maikoool | wont make much sense to break before it shows the hwcomposer information, right | 17:14 |
| giucam | yeah | 17:14 |
| maikoool | https://pastee.org/f8ug4 | 17:17 |
| maikoool | if i run it for longer then it's too late to press control c before oom kicks in | 17:17 |
| maikoool | oh wait | 17:17 |
| maikoool | that was my third attempt | 17:17 |
| maikoool | need to reboot first | 17:18 |
| maikoool | -__- | 17:18 |
| maikoool | after two tries leak does not appear so quickly | 17:18 |
| giucam | weird stuff | 17:18 |
| maikoool | yep | 17:18 |
| maikoool | lipstick behaves the same | 17:18 |
| maikoool | first two starts -> oom killed | 17:18 |
| maikoool | third and up -> homescreen shows but if app is started -> oom kill | 17:19 |
| maikoool | https://pastee.org/6pssv | 17:20 |
| maikoool | for comparison: this is the output shown when i don't break it after a few secs: https://pastee.org/qg7wj | 17:20 |
| mal- | Antartica: ok, I will try using those | 17:20 |
| giucam | i was hoping for something better :/ | 17:21 |
| mal- | Antartica: your script makes my life easier, thanks | 17:21 |
| alin | vgrade: ping | 17:23 |
| mal- | Antartica: there is a little problem with installing sailfish image, it requires replacing hwcomposer with a patched one in /system, I'll try a way to remove that neeed | 17:25 |
| Antartica | mal-: Happy to be of help. | 17:28 |
| Antartica | mal-: If the instructions are clear, I don't mind having to fix things. | 17:29 |
| mal- | Antartica: I know a way that should work so I can install hwcomposer to another place but it needs testing | 17:31 |
| giucam | maikoool: i don't really know... i would grab qwindow-compositor and comment out stuff until it stops going crazy | 17:31 |
| Antartica | mal-: ok | 17:33 |
| maikoool | i found a mozilla bugreport about memory leaking, they also use hwcomposer | 17:33 |
| maikoool | it suggest checking cat /sys/class/kgsl/kgsl/page_alloc | 17:33 |
| maikoool | and that indeed also increases a lot just before crash | 17:33 |
| mal- | maikoool: was there a fix? | 17:34 |
| maikoool | https://bugzilla.mozilla.org/show_bug.cgi?id=1029856 | 17:34 |
| merbot | Mozilla bug 1029856 in Graphics: Layers "Potential memory leak in b2g process." [Major,Resolved: fixed] | 17:34 |
| maikoool | not sure if related enough though | 17:34 |
| Antartica | mal-: I'm leaving. I will monitor the logs of the channel; if you have any problem with my scripts or anything (or publish the image!), just say so... | 17:42 |
| Antartica | Thanks to all the porters :-b | 17:42 |
| alin | Nokius_: ping | 17:42 |
| mal- | Antartica: ok | 17:42 |
| Antartica | Bye! | 17:43 |
| *** Antartica <Antartica!~Antartica@90.69.50.20> has left #sailfishos-porters ("Saliendo") | 17:43 | |
| maikoool | #ifdef USE_COPYBIT_COMPOSITION_FALLBACK // Use Copybit, when MDP comp fails (only for 8960 which has dedicated 2D core) | 17:44 |
| maikoool | i wonder where that variable is initialized | 17:44 |
| maikoool | my device is based on the 8960 | 17:44 |
| maikoool | but officially it's a 8960t | 17:44 |
| maikoool | maybe there's a subtle difference | 17:44 |
| maikoool | or that variables is not filled in properly | 17:44 |
| *** alin <alin!~alin@opensuse/member/ealin> has quit IRC (Quit: Konversation terminated!) | 17:49 | |
| *** piggz <piggz!~piggz@195.147.231.206> has quit IRC (Ping timeout: 240 seconds) | 18:09 | |
| *** nh1402_ <nh1402_!~nh1402@95.148.89.74> has joined #sailfishos-porters | 18:09 | |
| sledges | maikoool: interesting find indeed | 18:09 |
| *** exadeci <exadeci!uid35778@gateway/web/irccloud.com/x-buncxvdmilkhgmnb> has quit IRC (Quit: Connection closed for inactivity) | 18:10 | |
| *** nh1402 <nh1402!~nh1402@95.148.89.71> has quit IRC (Ping timeout: 244 seconds) | 18:12 | |
| maikoool | i'm now trying this first: https://android-review.googlesource.com/#/c/111273/1/msm8974/libhwcomposer/hwc_mdpcomp.cpp | 18:19 |
| maikoool | we'll see if that fixes anything | 18:19 |
| maikoool | are there hybris branches for aosp < 5 btw? | 18:26 |
| sledges | maikoool: only cm - 11.0 and 10.1 | 18:26 |
| maikoool | maybe using sony official 4.3 sources will fix this | 18:26 |
| maikoool | firefox os for xperia sp is built with 4.3 and recaf aosp 4.4 | 18:26 |
| maikoool | that works fine, i flashed it a time ago | 18:27 |
| sledges | are you using legacyxperia? | 18:27 |
| maikoool | no | 18:27 |
| sledges | recaf? | 18:27 |
| sledges | but cm works | 18:27 |
| sledges | hmm... | 18:27 |
| maikoool | there's this special aosp 4.4 recaf branch by adfad666 | 18:27 |
| maikoool | not sure what that's all about | 18:27 |
| maikoool | the repo has dissapeared from his personal github account but there's a fork under the incendiummod project (ffos for huashan) | 18:28 |
| *** alin <alin!~alin@vpn15.dl.ac.uk> has joined #sailfishos-porters | 18:28 | |
| *** alin <alin!~alin@vpn15.dl.ac.uk> has quit IRC (Changing host) | 18:28 | |
| *** alin <alin!~alin@opensuse/member/ealin> has joined #sailfishos-porters | 18:28 | |
| maikoool | legacyxperia is for 2011 devices it says | 18:28 |
| maikoool | mine's 2013 | 18:28 |
| sledges | fair | 18:34 |
| sledges | but cm boots and works | 18:34 |
| sledges | why you think it'd differ? | 18:34 |
| maikoool | yep | 18:34 |
| maikoool | cm10.1, 11.0 and 12.1 all three work | 18:34 |
| maikoool | because firefox os built from aosp | 18:35 |
| maikoool | why wouldn't they just build from cm? | 18:35 |
| *** CarlosMazieri <CarlosMazieri!~b35201@gate-tx3.freescale.com> has quit IRC (Quit: using sirc version 2.211+KSIRC/1.3.12) | 18:44 | |
| sledges | maikoool: aosp is closer to the core, less mods, but questionable hw support, because cm patch some bits to working state, in case ODM doesn't unleash all the sources (known for many non-nexus ODMs, except sony) | 18:56 |
| sledges | probably that's why ff went with aosp | 18:57 |
| sledges | yet hybris wants to support widest range, so went the cm route initially. we're going aosp5 as you see though | 18:57 |
| sledges | for many reasons | 18:57 |
| alin | sledges: MSameer ffmpegthumbailer | 18:57 |
| sledges | one of them - cm looking weird | 18:57 |
| alin | did you migrate it in the end? | 18:57 |
| sledges | alin: didn't | 18:58 |
| alin | ok that is not good | 18:58 |
| sledges | alin: it has no original sources anymore, so just sits there like a zombie | 19:01 |
| sledges | we need to see how free is MSameer to help us with thumbnailers | 19:01 |
| alin | sledges: but is needed | 19:01 |
| sledges | yes | 19:01 |
| sledges | the only copy of source - snapshot on obs | 19:01 |
| alin | sledges: ok till then lbt is our only change | 19:01 |
| sledges | waaaait | 19:02 |
| sledges | :D | 19:02 |
| sledges | https://github.com/cameraplus/ffmpegthumbnailer | 19:02 |
| sledges | done :D | 19:03 |
| * sledges points warheads to correct places | 19:03 | |
| sledges | done | 19:04 |
| alin | sledges: libav too I think | 19:04 |
| alin | maybe MSameerknows | 19:04 |
| sledges | yes | 19:05 |
| sledges | maybe | 19:05 |
| sledges | ;) | 19:05 |
| sledges | mal-: did you know about this: Possibility to choose different keyboard layouts for Bluetooth keyboards if paired with one for different languages via Settings > System > Text input > Active layout (from https://together.jolla.com/question/98953/release-notes-117-bjorntrasket/?sort=votes&page=1 ) | 19:05 |
| mal- | sledges: yes, I've been waiting for the release :) | 19:06 |
| mal- | sledges: is that only for bt keyboards? | 19:06 |
| sledges | tohkbd is treated as bt too | 19:08 |
| mal- | ok | 19:08 |
| alin | sledges: the spec for libav is focked | 19:08 |
| mal- | sledges: what was it about libav and not distritutable, in faq it says about gstreamer1.0-libav, is it only that or also libav? | 19:08 |
| mal- | just thinking about that ffmpegthumbnailer libav dependency | 19:09 |
| sledges | is what we need MSameer to confirm | 19:11 |
| mal- | I was thinking about including cameraplus to the image | 19:12 |
| *** alin <alin!~alin@opensuse/member/ealin> has quit IRC (Remote host closed the connection) | 19:14 | |
| MSameer | sledges: might be possible to build against platform libav | 19:15 |
| sledges | MSameer: could you elaborate a bit? | 19:28 |
| sledges | MSameer: we've already been building libav for cam+ since ages: https://build.merproject.org/project/monitor/nemo:devel:hw:common | 19:28 |
| sledges | alin: libav .spec is exactly the same that we're building, nothing is fscked | 19:30 |
| *** tworaz <tworaz!~tworaz@212.180.237.71> has joined #sailfishos-porters | 19:36 | |
| MSameer | sledges: what is the problem then? | 19:36 |
| sledges | can we redist it? | 19:36 |
| *** filippz <filippz!~filippz@93-143-195-115.adsl.net.t-com.hr> has quit IRC (Remote host closed the connection) | 19:37 | |
| *** alin <alin!~alin@opensuse/member/ealin> has joined #sailfishos-porters | 19:46 | |
| MSameer | sledges: what do you mean? | 19:48 |
| MSameer | sledges: ofc no | 19:49 |
| sledges | is https://github.com/cameraplus/libav redistributable? | 19:49 |
| sledges | because gst-libav is 100% not redist | 19:49 |
| sledges | what are the differences and why there are two? | 19:50 |
| mal- | sledges: the libhardware search path hack works, still need find out why libhardware is not looked from /usr/libexec/droid-hybris/system/lib | 19:53 |
| MSameer | sledges: libav is libav, gst-libav is gstreamer plugin for libav | 19:58 |
| alin | MSameer: yes but does not build | 20:04 |
| dr_gogeta86 | good evening | 20:08 |
| MSameer | alin: please use a full sentence that describes the issue | 20:09 |
| MSameer | alin: I don't understand anything from what you have just said | 20:09 |
| alin | MSameer: libav I have tried to build it using the spec file but failt to build | 20:10 |
| alin | fails | 20:10 |
| MSameer | alin: libav from which repos? which spec file? | 20:10 |
| alin | MSameer: the one sledges pointed to you | 20:10 |
| MSameer | alin: I am not going to continue pulling info from you | 20:11 |
| MSameer | alin: --verbose | 20:11 |
| dr_gogeta86 | who did display-legacy patch ? | 20:11 |
| mal- | dr_gogeta86: if you mean line 31 on faq then me | 20:13 |
| dr_gogeta86 | yep | 20:13 |
| mal- | seems that it has expired | 20:13 |
| dr_gogeta86 | i need it for display-legacy | 20:14 |
| *** r0kk3rz <r0kk3rz!~chatzilla@host-92-30-149-11.as13285.net> has quit IRC (Ping timeout: 244 seconds) | 20:14 | |
| dr_gogeta86 | is completelly different | 20:15 |
| sledges | alin: libav builds fine on obs | 20:15 |
| dr_gogeta86 | some peeps did it with caf on my device | 20:15 |
| sledges | if you have local build issues, paste logs, we can't mindread yet:) | 20:16 |
| mal- | dr_gogeta86: where is the source for display-legacy? | 20:17 |
| *** phdeswer <phdeswer!~phdeswer@91-159-55-220.elisa-laajakaista.fi> has quit IRC (Ping timeout: 264 seconds) | 20:18 | |
| alin | sledges: MSameer mae | 20:20 |
| alin | make | 20:20 |
| alin | make: *** No targets specified and no makefile found. Stop. | 20:20 |
| dr_gogeta86 | github c8813q-dev/android_hardware_qcom_media-legacy | 20:20 |
| dr_gogeta86 | https://github.com/c8813q-dev/android_hardware_qcom_media-legacy | 20:20 |
| alin | trying to build rpm/libav.spec | 20:20 |
| dr_gogeta86 | sorry | 20:20 |
| dr_gogeta86 | c8813q-dev/android_hardware_qcom_display-legacy | 20:20 |
| mal- | dr_gogeta86: which device did you have? | 20:21 |
| dr_gogeta86 | huawei g525/c8813q | 20:21 |
| mal- | I'll check that source | 20:22 |
| dr_gogeta86 | https://github.com/c8813q-dev/android_hardware_qcom_display-legacy/tree/cm-11.0/libhwcomposer | 20:22 |
| alin | sledges: MSameer seems ot be no source | 20:23 |
| *** piggz <piggz!~piggz@195.147.220.200> has joined #sailfishos-porters | 20:24 | |
| mal- | dr_gogeta86: that source doesn't seem to be needing patching | 20:24 |
| dr_gogeta86 | but doesn't work | 20:25 |
| dr_gogeta86 | i always got many windows error | 20:25 |
| dr_gogeta86 | see backlog | 20:25 |
| dr_gogeta86 | QPA-HWC: There can only be one window, someone tried to create more | 20:26 |
| mal- | could be that the legacy version used on your device is too different | 20:28 |
| mal- | the bugs in hwcomposer should not cause that kind of problem | 20:28 |
| mal- | the known bugs | 20:28 |
| sledges | alin: you are not pasting exact errors, i can't decrypt otherwise what you are saying | 20:30 |
| alin | sledges: I told you that is the exact error | 20:30 |
| alin | of one clones that repo there is nothing but the spec file | 20:30 |
| sledges | git clone -r | 20:30 |
| sledges | or | 20:30 |
| sledges | git submodule init | 20:30 |
| sledges | git submodule update | 20:30 |
| mal- | sledges: it works! had to patch libhybris also to find /usr/libexec/droid-hybris/system/lib/libhardware.so | 20:33 |
| sledges | yaay! | 20:34 |
| mal- | now we can sideload patched android libraries without changing /system | 20:34 |
| sledges | oh man! | 20:34 |
| sledges | (some of those we couldn't;) | 20:34 |
| sledges | sorry | 20:35 |
| sledges | some of those we already could | 20:35 |
| dr_gogeta86 | :\ | 20:35 |
| mal- | sledges: which ones? since anything used through libhardware we couldn't | 20:36 |
| sledges | well | 20:36 |
| sledges | all those that already are under /usr/libexec/droid-hybris/system/lib/ | 20:36 |
| mal- | anyway the reason was faulty libhybris, it had android_dlopen("/system/lib/libhardware.so", RTLD_LAZY); instead of "libhardware.so" | 20:38 |
| sledges | libhybris was LAZY | 20:40 |
| sledges | :D | 20:40 |
| sledges | now i wonder what regressions we'll encounter after your patch ;) | 20:40 |
| mal- | what does LAZY do? | 20:41 |
| sledges | noidea | 20:41 |
| maikoool | mal-: does your patch mean i wont have to copy over hwcomposer.msm8960.so manually? | 20:43 |
| maikoool | if so | 20:43 |
| maikoool | can i try it? | 20:43 |
| mal- | maikoool: yes, it consists of two patches, one for libhybris and one for android libhardware | 20:45 |
| maikoool | they're on github? | 20:45 |
| mal- | no | 20:46 |
| mal- | sledges: I don't really see what regressions we could encounter, since it always reverts back to /vendor or /system | 20:47 |
| mal- | maikoool: I'll fork the needed repos, just a moment | 20:49 |
| alin | MSameer: foudn the issue... in the spec.. configure needs to be in the %build section not in %prep | 20:49 |
| maikoool | np, no rush | 20:49 |
| alin | MSameer: rpm/libav.spec | 20:49 |
| sledges | mal-: in case if there is a .so under libexec, which has never been picked before due to bug you fixes, and would result in differing behaviour | 20:51 |
| sledges | but since it's under libexec, 99.999 times out of 100 it should've been there on there on the first place, so all should be ok | 20:51 |
| sledges | and even if behaviour would change, it should change for beter (just like GLESv1) | 20:52 |
| sledges | so PR away :) | 20:52 |
| alin | nh1402_: ping | 21:00 |
| nh1402_ | alin: pong | 21:01 |
| mal- | maikoool: https://github.com/mlehtima/android_hardware_libhardware/tree/hybris-11.0-44S and https://github.com/mlehtima/libhybris | 21:02 |
| maikoool | cool | 21:03 |
| mal- | sledges: ^ aren't those quite nice patches | 21:05 |
| sledges | mal-: careful, i might develop a fetish for them :D | 21:07 |
| sledges | mal-: libhybris PR upstream pls | 21:08 |
| mal- | sledges: PR sent | 21:14 |
| sledges | mal-: i mean | 21:14 |
| sledges | libhybris upstream | 21:14 |
| sledges | not downstream | 21:14 |
| mal- | sledges: oh | 21:15 |
| mal- | sledges: PR sent upstream | 21:20 |
| sledges | Stskeeps: https://github.com/libhybris/libhybris/pull/283 ^ | 21:21 |
| mal- | sorry, I forgot again to paste the link | 21:21 |
| sledges | np:) | 21:21 |
| dr_gogeta86 | sledges, lipstick got some kind | 21:24 |
| dr_gogeta86 | start without dbus ? | 21:24 |
| mal- | https://github.com/mer-hybris/android_hardware_libhardware/pull/3 | 21:24 |
| mal- | sledges: ^ | 21:24 |
| sledges | mal-: reference libhybris PR? | 21:32 |
| alin | MSameer: sledges https://github.com/cameraplus/libav/pull/1 | 21:38 |
| *** nh1402 <nh1402!~nh1402@95.148.89.74> has joined #sailfishos-porters | 21:38 | |
| MSameer | alin: merged | 21:39 |
| alin | MSameer: thanks | 21:40 |
| MSameer | alin: thanks too | 21:40 |
| *** nh1402_ <nh1402_!~nh1402@95.148.89.74> has quit IRC (Ping timeout: 255 seconds) | 21:41 | |
| mal- | sledges: better now? | 21:42 |
| sledges | mal-: yes | 21:48 |
| mal- | tomorrow's task will be to build a new image | 21:49 |
| sledges | mal-: much thanks, let's wait till libhybris is accepted | 21:52 |
| sledges | or we can merge libhardware already now | 21:52 |
| mal- | no hurry, I can always build my image from local versions | 21:53 |
| *** Nokius_ is now known as Nokius | 21:56 | |
| Nokius | evening | 21:56 |
| *** M4rtinK <M4rtinK!~M4rtinK@ip-89-177-124-246.net.upcbroadband.cz> has joined #sailfishos-porters | 21:59 | |
| Nokius | alin: pong | 22:00 |
| alin | Nokius: ping | 22:01 |
| alin | sorted out | 22:01 |
| Nokius | alin: evening | 22:02 |
| Nokius | alin: ah ok | 22:03 |
| Nokius | alin: was at work | 22:03 |
| sledges | o/ | 22:03 |
| Nokius | sledges: hi | 22:04 |
| Nokius | sledges: I will try the patch tomorrow what I have to build after patching? | 22:09 |
| Nokius | alin: u can start tranlating :) | 22:18 |
| Nokius | alin: https://www.transifex.com/projects/p/sailorgram/ | 22:19 |
| * Nokius gn8 | 22:25 | |
| alin | Nokius: night | 22:27 |
| sledges | Nokius: make hybris-hal | 22:30 |
| *** Tassadar <Tassadar!~tassadar@ip4-83-240-20-212.cust.nbox.cz> has quit IRC (Ping timeout: 260 seconds) | 22:49 | |
| *** OrokuSaki <OrokuSaki!475bbaa0@gateway/web/freenode/ip.71.91.186.160> has joined #sailfishos-porters | 22:55 | |
| OrokuSaki | @MSameer So I compiled gst-droid on archlinux... I get the same error as before that patch you made... Not sure why... Might update sailfish to 2.0 on my device and try again... just odd.... figured it might work.. haven't tried it in weston.. I read firefox uses gstreamer and etc... had to rip out droidcamsrc though... qt5.4 didn't quite match up..... | 22:57 |
| MSameer | what patch? | 22:58 |
| MSameer | what error? | 22:58 |
| OrokuSaki | we last did this on: http://www.merproject.org/logs/%23sailfishos-porters/%23sailfishos-porters.2015-04-13.log.html 4-13 | 22:59 |
| OrokuSaki | this is the last commit: | 23:00 |
| OrokuSaki | https://github.com/foolab/gst-droid/commit/bcf07558190bdc6dbd7319402571eaf2bd3fee63 | 23:00 |
| OrokuSaki | have the error shortly | 23:00 |
| MSameer | I don't understand | 23:00 |
| OrokuSaki | I took my gst-droid from sailfish and thought I would try it on arch... using the same stuff I have from sailfish that does work | 23:01 |
| OrokuSaki | recompiled and etc | 23:01 |
| OrokuSaki | when I run logcat I get this: ( 1141): DroidMediaCodec: Error 0x3d reading from codec | 23:02 |
| OrokuSaki | I tried different gst buffers... 1-8 =) | 23:02 |
| OrokuSaki | curious if gst-droid works in X11 =) | 23:06 |
| MSameer | arch linux? | 23:15 |
| *** nh1402 <nh1402!~nh1402@95.148.89.74> has quit IRC (Ping timeout: 265 seconds) | 23:25 | |
| *** M4rtinK <M4rtinK!~M4rtinK@ip-89-177-124-246.net.upcbroadband.cz> has quit IRC (Ping timeout: 240 seconds) | 23:33 | |
| *** olafh_ <olafh_!~olafh@p4FF484CC.dip0.t-ipconnect.de> has quit IRC (Ping timeout: 250 seconds) | 23:35 | |
| *** tworaz <tworaz!~tworaz@212.180.237.71> has quit IRC (Ping timeout: 272 seconds) | 23:36 | |
| *** zajfy <zajfy!~zajfy@ip-180-188-230-46.dialup.ice.net> has quit IRC (Read error: Connection reset by peer) | 23:43 | |
| OrokuSaki | nod arch linux... I got android and hybris stuff working... I have pulse and droidsink working so I thought I would try this... | 23:48 |
| *** tbr <tbr!dm8tbr@cl-790.ham-01.de.sixxs.net> has quit IRC (Ping timeout: 240 seconds) | 23:54 | |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
