Sunday, 2014-09-21

*** marxistvegan has joined #sailfishos-porters00:02
*** Umeaboy has quit IRC00:03
sledgesalin: could you show a patch of what is broken so could be fixed?00:37
*** marxistvegan has quit IRC01:07
*** Sage__ has quit IRC01:43
*** Sage_ has joined #sailfishos-porters01:50
*** Sage_ has quit IRC01:50
*** Sage_ has joined #sailfishos-porters01:50
*** xwife has joined #sailfishos-porters01:53
*** zon is now known as zZz0n01:57
*** hurrian has quit IRC02:11
*** lbt has quit IRC02:51
xwifehi to all02:52
*** lbt has joined #sailfishos-porters02:52
xwifeis it possible to port sailfish in an mtk device?02:52
*** xwife has quit IRC04:14
*** hurrian has joined #sailfishos-porters04:21
*** hurrian has joined #sailfishos-porters04:21
*** dmt__ has quit IRC05:12
*** olafh has joined #sailfishos-porters05:16
*** Guhl99 has joined #sailfishos-porters07:28
*** Guhl99 is now known as guhl07:28
*** vrutkovs has joined #sailfishos-porters07:48
*** alin has quit IRC07:51
*** alin has joined #sailfishos-porters07:52
guhlmorning alin07:56
alinguhl: morning07:56
alinsledges: of course but a patch from what to what07:57
alinsledges: practically about what07:57
alinsledges: I could not find patterns07:57
guhlalin could you help me on how to install additional packages in the rom i created (or on the running system)07:59
alinsledges: what  I mean is against your version or the kosher one08:00
alinguhl: ok that is simple08:00
guhlpiggz asked if i can run qmlscene08:00
guhlhow would i install that08:00
alinguhl: I do not know what provides it08:02
guhlPACKAGES_TO_ADD is my friend?08:02
alinyes to put in that variable all the packages you want to add08:03
alinand $KSFL is your .ks file08:03
guhlyes i already figured out that the .ks file is where i add the repos and so but i must have missed the packages_to_add part08:04
guhlyour scripts are really great08:10
guhlespecially for learning08:10
piggzguhl: you can add the packages directly on the phone...saves building new images each time08:20
*** energycsdx has joined #sailfishos-porters08:20
guhlpiggz, morning08:22
alinpiggz: thought he did not boot...08:22
alinsledges: ok... problem solved...08:22
alinguhl: ssh into phone or telnet... then zypper in packagename08:22
piggzguhl: and if nwifi isnt working yet,
*** lbt has quit IRC08:23
guhlalin, i am really sorry for such a noob question but where does it get the package from08:23
*** lbt has joined #sailfishos-porters08:23
guhlah that answers this08:23
alinpiggz: guhl you may want usb-moded a newwer version... makes things work with tetheting08:24
guhlpiggz, which kernel did you actually use for you ace build?08:25
piggzguhl: 3.0.101 cm1108:25
guhlahh a fork from Mustaavalkosta who forked from KangBang / Flemmard08:27
*** Nokius has quit IRC08:28
guhlFlemmard is/was part of the andromadus team08:28
guhlso we share the basis as far as it comes to flemmards repo08:29
guhlpiggz which package would provide qmlscene?08:29
piggzguhl: and i used cm11 instead of cm10.1 becuase the wifi worked with no tweaking08:29
piggzguhl: qt5-qtdeclarative-qmlscene08:30
piggz(zypper se qmlscene)08:30
guhlzypper is not my friend yet :-)08:31
alinguhl: piggz so you do htc desire s too?08:33
alinvgrade:  problem solved08:34
alinvgrade: can you please have a go at the script now?08:34
alinas usual the problem was between the chair and screen08:34
piggzalin: i have desirehd, alin as the vision (with a keyboard, similar hw)08:36
guhlbut desire s would be another device08:36
guhlvision is desire z08:36
guhlsaga is desire s08:37
alinguhl: yes08:37
guhlbut if i get vision to work saga would be easy08:39
guhlat andromadus we did vision and saga08:39
guhlpiggz, i think i'll switch to the cm-11 kernel too as as you said wifi will work more easily08:41
guhlso you basically used the cm11 kernel but the hybris-10.1 mer stuff?08:41
piggzguhl: yes08:42
guhli assume that flemmard also changed to wifi to built in in the ace kernel or do you still have it as a module?08:42
*** spiiroin has quit IRC08:43
*** lbt has quit IRC08:43
piggzguhl: I dont have modules08:44
*** dmt__ has joined #sailfishos-porters08:46
guhlin my opinion we had a memory problem with the built in wifi approach on the vision because of the firmware loading08:46
guhlbut we'll see about that08:46
guhlif htc would have used the built in approach from the beginning you would not be installing sailfish os on your ace now :-)08:48
guhlwell loading the module give me a wlan0 that would probably work08:52
guhlwithout module loading support in the kernel we would never have been able to bring down the power of the emmc to remove the write protection08:53
guhland the htc phone would still be locked down08:54
piggzguhl: you clearly know more about this than me!!08:58
guhlwell thats my home ground08:59
*** alin has quit IRC09:05
*** alin has joined #sailfishos-porters09:05
alinsledges: testing them image with your branch09:05
alinsledges: ok it boots09:09
guhlback to building android, ...09:10
*** lbt has joined #sailfishos-porters09:11
guhlpiggz, the funny things about our devices is that they are obviously still in use09:21
guhlin September i still had 600 downloads of gfree from my site alone09:22
guhland gfree is for sure mirrored to a lot of other sites09:22
*** spiiroin has joined #sailfishos-porters09:25
guhlpiggz, what was the reason for the pmem hack in your kernel?09:27
piggzguhl: apps wont load09:27
piggzguhl: pmem gives a mmap no memory fault09:28
guhlok so i will for sure have to add that too09:28
piggzguhl: afaict, pmem reserves some mem on startup ... the first mmap succeeds but maps the entire reservation, so all others fail09:29
*** piggz has quit IRC09:35
*** appiah has quit IRC09:37
*** piggz has joined #sailfishos-porters09:45
*** guhl has quit IRC09:52
*** carepack has joined #sailfishos-porters09:58
*** Nokius has joined #sailfishos-porters10:16
*** filippz has joined #sailfishos-porters10:22
*** Nokius has quit IRC10:30
*** Guhl has joined #sailfishos-porters10:33
*** lbt_ has joined #sailfishos-porters10:39
*** lbt_away has joined #sailfishos-porters10:40
*** lbt has quit IRC10:43
*** lbt_ has quit IRC10:43
*** zetaz has joined #sailfishos-porters10:44
*** arcean has joined #sailfishos-porters10:53
piggzGuhl: it will be good that we can split the work load in getting the hardware working10:55
piggzit would be nice to fix my pmem hack properly ... i think its responsible for the graphics corruption i see with the vkb10:55
piggznext i need to looad at audio and camera10:55
alinpiggz: camera is a long story speak with MSameer about11:08
*** piggz has quit IRC11:13
*** Nokius has joined #sailfishos-porters11:21
*** Nokius has quit IRC11:25
*** filippz has quit IRC11:26
*** piggz has joined #sailfishos-porters11:26
piggzalin: how do i go about debugging the audio?11:30
*** alin_ has joined #sailfishos-porters11:33
alin_piggz: pulseaudio general guide?11:34
*** alin has quit IRC11:37
sledgesalin_: thanks. what's the status now of images/branches/patterns?11:42
alin_sledges: works11:42
*** lbt_away is now known as lbt11:42
alin_sledges: tested with the dhd from your branch and the image builds and boots11:43
alin_sledges: the one you passed me last week11:43
sledgesalin_: and upstream?11:43
alin_sledges: and with upstream too11:43
alin_sledges: the mistake was mine... remember i swapped the Jolla configuration with the one suggested in page 2411:44
alin_that is a bad idea... does nto pull all the packages..11:44
sledgesah ok11:44
alin_sledges: the only thing I had to to by hand  was to install in the mer sdk-utils -.6511:44
sledgeshappened in the past:)11:44
sledgesisn't that in :relase already?11:44
alin_sledges: I do not know I did not check11:45
alin_sledges: let me see11:45
piggzalin_: well, yes, and linking that with an android kernel/hybris11:45
sledgescool thanks for your test, i'll continue work on my branch11:45
alin_sledges: is not yet... in release... is only in testing11:46
sledgesi guess lbt could promote that thanks to you11:47
sledgesso we unbreak the porting11:48
* lbt needs some more feedback on mb2 in non-hadk before going to stable11:48
lbtbit busy today though11:48
alin_sledges: other thing when building the mw11:50
alin_  sb2 -t $VENDOR-$DEVICE-armv7hl -R -m sdk-install zypper in qt5-qttools-kmap2qmap repomd-pattern-builder11:50
alin_you see the two packages needed11:50
alin_and of course rfkill...11:50
alin_sledges: now i do not know the status of the usb-moded for the september update... but probably will be in base the correct one11:51
alin_will a september update happen?11:52
alin_sledges: by the way... now the store works out of the box11:52
alin_piggz: but the debugging methods will be more or less the same11:53
*** alin_ has quit IRC11:58
*** zetaz has left #sailfishos-porters12:02
*** lbt has quit IRC12:02
*** lbt has joined #sailfishos-porters12:03
*** alin has joined #sailfishos-porters12:06
*** Nokius has joined #sailfishos-porters12:06
*** dmt__ has quit IRC12:18
Guhlsorry but does somebody know this?12:21
Guhlwhen doing the android build12:21
Guhlit starts a process that is:12:21
Guhl/bin/bash -c /usr/bin/perl -w -e '$fs=shift; while (<>) { next unless /^$fs\s|\s$fs\s/;for (split) {next unless m(^/dev); print "$_\n"; }}' /boot  | sort -u12:22
Guhlwhich first starts:12:22
Guhl/usr/bin/perl -w -e $fs=shift; while (<>) { next unless /^$fs\s|\s$fs\s/;for (split) {next unless m(^/dev); print "$_\n"; }} /boot12:22
Guhland this hangs forever12:23
sledgesGuhl: known issue when recovery.fstab is nowhere to be found12:23
sledges(or another accompanying *fstab* file can't remember the name)12:23
Guhlit hangs for /boot and if i kill that prozess then for /data12:25
Guhland if i kill that i get the message12:28
Guhlhybris/hybris-boot/ ********************* /boot appears to live on12:28
Guhlhybris/hybris-boot/ ********************* /data appears to live on12:28
Guhlhybris/hybris-boot/ *** There should be a one and only one device entry for HYBRIS_BOOT_PART and HYBRIS_DATA_PART.  Stop.12:28
Guhlon the console12:28
*** alin has quit IRC12:32
Guhli think i get the point12:33
sledgesGuhl: that's what i mean12:33
sledgesfstab files are poorly12:33
sledgeshow were you able to install image then?:))12:33
*** alin has joined #sailfishos-porters12:33
*** alin has quit IRC12:33
*** alin has joined #sailfishos-porters12:33
Guhlsledges, yes i was able to install and start it12:33
Guhlbut no screen at all12:34
Guhlbut before i continue to debug that i want to switch to our cm-11 kernel12:34
Guhlthat is much cleaner12:34
sledgesGuhl: your /boot and /data should be valid, otherwise .zip wouldn't install itself..12:36
*** javispedro has joined #sailfishos-porters12:38
Guhlyes, now that i see from the file that it's sed'ing the updater-script i also understand the point of it12:38
Guhljust still don't know why it hangs12:38
sledgesdue to *fstab* in our device/* files being the format they want, and not what is most common (as we expect)12:38
Guhli just wonder why i did not have that error before12:39
sledgess/our/your/ ;)12:39
Guhli somehow thougt that this is what fixup-mountpoints is for12:40
Guhlaha seems that i do not have a file that is called *fstab* in device/*/vision anymore12:48
Guhlyeah i switched also the device repos from our cm-10.1 to the cm-11.0 branch in my local_manifest12:50
Guhlthe fstab files are really gone in the cm-11.0 device directory12:55
*** arcean has quit IRC12:56
sledgesso you are switching to 11.0?12:56
sledgesbrave move sir, as you can tell now;)12:56
Guhlhmm hard decisions13:02
sledgeswell, i utterly advise to stay on 10.1.13:03
sledgesand get graphics going first13:03
Guhlyes this seems reasonable13:05
*** dmt__ has joined #sailfishos-porters13:05
Guhlbut i will switch to the 4.4. kernel13:05
alinsledges: what are you breaking in your branck?13:15
sledgessplit version13:17
alinsledges: yes... I mean what is the final goal for that?13:17
alinsledges: except world domination13:17
*** dmt__ has quit IRC13:17
sledgesnothing except that13:17
sledgesdhd needs various parts split off, because when everything is on obs, a smallest change would trigger the world to rebuild13:19
sledgesand we don't want that13:19
sledgesalso having separate changelogs per-device is a goal13:19
alinsledges: ok I see13:46
alinsledges: one I have seen... dhd shall conflict the packages that provides configs for13:47
alinone exmaple is this13:47
sledgeshw-release ? ;)13:47
sledgesthat has been solved...ish13:48
alinthat dile is provided by some other package...13:48
alinsledges: I do not know I did not check if was solved or not...13:48
sledgesneeds to be put to Prefer:13:48
sledgeson OBS13:48
sledgesbut what to do with local builds?13:48
alinsledges: there is a spec file.... for the settings one...13:49
alinsledges: just stick a conflict in it13:49
sledgescould you propose a proper fix pls? im afk abit13:51
alinsledges: the prj configs are fine but I think nicer is to have conflicts inside the .spec for the package13:52
*** beidl_ has joined #sailfishos-porters14:04
*** beidl has quit IRC14:05
*** energycsdx has quit IRC14:18
*** beidl_ is now known as beidl14:22
beidlhello friends14:22
sledgeshoy beidl, how're goggelen!?14:22
beidlsledges: gogg.. what?! :D14:23
beidlsledges: eggs? :D14:23
beidlsledges: it was hard to read, written it's a little different from it's pronounciation14:24
beidlsledges: but yes, they are scrambled as HELL14:25
sledgeshow do you spell eggs in Tirolean then?;D14:25
beidlsledges: I'm from Burgenland, here we'd probably write "gogal" :P14:25
* beidl tries an image build with OBS repo + sfa-mer scripts14:26
beidllets see where it goes. I'm too lazy to rebuild everything manually for the target14:27
alinbeidl: update the script14:29
alinbeidl: shall work14:29
beidlalin: I pulled updates at 3am this morning14:30
alinbeidl: probably I did some changes int he morning14:30
alinbeidl: pull again14:30
beidlalin: damn quick, I must admit14:31
alinbeidl: what is quicl?14:31
beidlalin: updates to the script :)14:31
alinbeidl: he he...14:32
alinbeidl: ./ -mer-root $HOME/lavello/NEXUS5/mer -android-root $HOME/lavello/NEXUS5/droid  -device hammerhead -vendor lge -extraname ame -sfrelease -jobs 4 -branch hybris-11.0 -dest $HOME/lavello/nexus5 -mwrepo "x" -extrarepo ""14:32
alinthis is what I use for n514:32
beidlalin: does the script replace adaptation0 with EXTRA_REPO?14:32
alinthat builds the middleware... and dhd...14:32
alinbeidl: no... mw-repo is the adaptation... extra is extra14:33
beidlalin: in the target or the actual image?14:33
alinin the image you still have the old ones14:34
alinalso you may need to install sdk-utils 0.65 in mer14:34
beidlalin: ok :) will play around with the script and maybe add a patch to do it14:34
beidlalin: is it in mer-rolling already?14:35
alinbeidl: is not...14:35
alinbeidl: if would be there would be great14:35
alinrpm -U
alinwhat do you want to patch?14:35
beidlalin: thanks14:35
beidlalin: adding a way to add adaptationXY to the image ssu repos14:36
alinbeidl: godo one... I was looking for that...14:37
beidlalin: until now I've added a sed command to the kickstart file to replace the existing adaptation repo with a working OBS one14:37
alinbeidl: I suspect the simplest is a ssu ar14:37
alinbeidl: have a look in the the same I do14:37
alinbut that does not keep the repos in image14:37
beidlalin: I will, thanks14:37
alini think they are used just for build14:38
alinbeidl: this is what is in image - adaptation0 ...
beidlalin: I'd like to push updates to running installations via the repo14:39
beidlthats why I'd need it14:39
alinbeidl: yes... i do that too but i add it after by hand14:39
alinssu ar...14:39
alinso  if you tell me how one can run a command in the image... I will add the things in no time14:40
alinsledges: can you add something in ks to run in the image?14:41
beidlalin: after "export SSU_RELEASE_TYPE=release"14:41
beidlalin: or rather, before ### begin 01_arch-hack14:41
beidlalin: thats where I added the sed command (thanks to sledges for that one)14:42
alinbeidl: ok... so post is safe14:42
beidlalin: yup14:42
alinbeidl: perfect I will add it now14:42
beidlalin: nice, thank you :D14:43
alinbeidl: there is a ssu section I will try to add the magic there14:43
alinbeidl: now if I call it adaptation0 may be written by ssu ur14:44
beidlalin: it wouldn't14:45
beidlalin: it would default to the sed'ed repo14:46
beidlalin: sry, I read "overwritten"14:46
beidlalin: that might work.14:46
alinbeidl: yes ssu ur is bad14:47
beidlalin: sed -i -e 's|^adaptation=.*$|adaptation=|' /usr/share/ssu/repos.ini14:47
beidlalin: that should work for middleware packages. additional dhd repo could probaby be added with ssu ar14:48
beidlas adaptation1 or something14:48
alinbeidl: ssu ar adds in it14:50
alinso I just add that command14:50
alinin the post14:50
beidlalin: yes14:50
alinbeidl: also added the extra repo if any14:50
alinthe same with adaptation14:50
alincalled him adaptation1 not to conflict with the adaptation014:51
beidlalin: perfect. needed as the mw packages and dhd packages are in separate repos14:51
alinbeidl: yap14:51
beidlalin: as long as one of them overrides adaptation0 its fine :)14:51
alinnone will overwrite it14:51
alinssu ur14:52
alinwill restore to whatever was initially... at least this is my understandign now14:52
alinbuilding the image14:53
alinif goes without error I will commit14:53
alinthese names are taken store-auth-repos=aliendalvik,thirdparty,adaptation,adaptation0,adaptation1,eas,xt914:54
alinwe will need to find a better one14:55
beidlalin: the sed script I sent makes sure it won't get overriden14:55
beidlas it is already in repos.ini14:55
alinyap... destroys the ssu configuration... may be difficult to test the officials ones14:57
alinbetter we just add a new one and give more flexibility14:57
beidlalin: what do you mean "destroy"? I'm pretty sure there wont be an official adaptation repo for maguro as I'm not employed by Jolla14:58
alinbeidl:  - adaptation0 ...
Stskeepsnever say never..14:59
alinbeidl: just in case jolla wants to use that repo for whatever reason for a device14:59
Stskeepswe're not meego, we do think there could be a path for 'official' images14:59
beidlalin: true. the fish-names for nexus devices fit in the jolla/sailfish scheme :)15:00
beidlStskeeps: thats nice indeed15:00
alinbeidl: so my idea is not to clash with their names15:00
beidlalin: alright. I could still modify the ks manually for the time being15:00
alinbeidl: I see that ssu ur does not activate repos I deactivated via ssu dr15:01
alinbeidl: so maybe I shall add one more to disable adaptation015:01
alinbeidl: in the image15:01
beidlalin: sounds reasonable15:02
alinnow i need to test the changes without installing the image15:03
alinso now I did like this... added what you pass in -mwrepo as dhd repo, and in extra as extra repo15:03
beidlalin: open the bz2 archive and check if repos.ini points to the respective repos15:04
beidlalin: ok, sounds great15:04
beidlI'll test it shortly on my machine :)15:04
alinStskeeps: do you know in which file one keeps what is added via ssu ar?15:05
alinsledges: ^15:05
alinbeidl: pushed15:10
beidlalin: \o/ thanks15:10
alinbeidl: i think it worked I see the extra repo in the image....15:13
alinbeidl: so please let me know15:13
beidlalin: perfect, trying now15:14
alinbeidl: Stskeeps we need the backup to work...15:14
alinthen I can flash as many times i like15:14
alinbut recreating all the bloody accounts puts me off15:15
*** marxistvegan has joined #sailfishos-porters15:23
beidlalin: definitely. but tar'ing /home and preserving permissions might do for now15:25
beidllet's see how long such a rebuild takes on the new SSD15:25
piggzwhy cant I select 'allow installation from untrusted sources' in settings?  I click accept and it stays unchecked15:26
piggzused rpm to manually install the harbour warehouse15:26
sledgespiggz: zypper in devicelock-plugin-encsfa15:33
sledgesalin: /etc/ssu ?15:34
*** phdeswer has joined #sailfishos-porters15:38
alinsledges: yap dounf them15:46
alinbeidl: not really15:46
*** arcean has joined #sailfishos-porters15:47
alinbeidl: so any success?15:47
beidlalin: had to fix issues with the manifest. it's at building clang now15:47
alinbeidl: I see... manifest has nothing to with the script I hope15:48
beidlalin: no worries :)15:48
alinbeidl: that is huge and I do not get it do we really need it?15:48
alinbeidl: the clang15:48
beidlalin: I'm starting to think that we absolutely don't need it. the build uses gcc, and that is prebuilt anyways15:49
alinsledges: Stskeeps ^ do we really need to build clang?15:49
* sledges shrugs15:53
sledgesjust chuck it out and see;)15:53
*** zZz0n is now known as zon15:56
beidlsledges: I'm in the middle of building, I'll try it later :)15:56
beidlif it works I'll push changes to the manifest (additionally with maguro repos)15:57
alinbeidl: cool15:57
alinso I will not duplicate your efforts15:58
beidlalin: I can't waste other peoples time ;)15:58
beidl"I'm used to wasting time" #thingssoldierssay15:59
*** dmt has quit IRC16:04
*** dmt has joined #sailfishos-porters16:07
*** dmt has quit IRC16:07
*** marxistvegan has quit IRC16:17
*** zetaz has joined #sailfishos-porters16:30
alinbeidl: finshed?16:33
beidlalin: mv: cannot stat `nemo:devel:hw:samsung:maguro/droid-hal-maguro/*.rpm': No such file or directory16:35
beidlalin: which REPO is it looking in? MW_ or EXTRA_ ?16:35
alinbeidl: MW16:35
beidlalin: I figured that, switched those two, rebuilding16:35
alinbeidl: ok shall be fast now16:36
beidlrepo metadata refreshing takes some time though16:36
alinbeidl: yes16:36
alinbut compared with 30min full16:36
alinnow shall be no more thna 516:36
*** arcean has quit IRC16:37
* beidl gets himself another cup of coffee16:37
beidlI might play around with the hwcomposer qpa again after testing16:42
beidlStskeeps: is it possible that rendering on OMAP4/hwcomposer 1.0 is slow with an application window being focused because the app and the shell are rendering into the same HW layer?16:43
beidlalin: seems like the osc command doesn't do what its supposed to do with the dhd subproject16:49
beidlalin: I had to add the dhd subproject to the osc command in, still the rpm packages don't get pulled, instead the _service file is downloaded16:50
*** phdeswer has quit IRC16:50
beidlalin: also, why pull rpms via osc if the repository is already in the target, it would just pick up the newer versions of packages16:54
alinbeidl: build the patterns16:59
alinbeidl: woeking around a bug16:59
alinbeidl: getbinaries17:00
alinbeidl: this was because of our specific structure17:01
alinbeidl: one shall use getbinarries and grap the rpms17:01
beidlalin: I'll let it build the stuff locally first then, as long as it's in the local repo17:01
alinbeidl: ok17:03
alinbeidl: the idea was that we get the rpm from that repo17:06
alinbeidl: one can use them17:06
*** phdeswer has joined #sailfishos-porters17:06
alinbeidl: you do not have rpms loaded for maguro?17:06
beidlalin: things have changed with OBS-built dhd packages17:07
beidlalin: resetup after switching from an HDD to an SSD.17:07
beidlthe backup tool didn't back up root-owned files, so I decided to start from scratch17:07
alinbeidl: strange they did nit change for hammerhhead17:07
alinshow me what repo you use17:07
alinbeidl: i see you build the rpms... we do not17:10
alinbeidl: thought was not possible to build them on obs17:10
alinbeidl: then I suspect you may no need to copy them at all17:11
*** arcean has joined #sailfishos-porters17:11
alinbeidl: you shall have the repo added as remote17:11
beidlalin: all thanks to lbt17:11
alinbeidl: comment the else branch at 44 or so17:12
alinI am curious17:12
alinbeidl: so we need sledges to do the same for hammerhead17:14
*** mispp has quit IRC17:37
*** Nokius has quit IRC17:50
*** Nokius has joined #sailfishos-porters17:53
*** alin_ has joined #sailfishos-porters17:54
*** alin has quit IRC17:54
*** Nokius has quit IRC17:56
*** Nokius has joined #sailfishos-porters18:02
*** Umeaboy has joined #sailfishos-porters18:12
Umeaboysledges or lbt: When I sudo rpm -U..............rpm I don't need to run the sb2 command with inc build, right?18:14
UmeaboyJust the second one.18:14
UmeaboyOr have I got it wrong?18:14
*** lbt has quit IRC18:39
*** lbt has joined #sailfishos-porters18:40
*** lbt_ has joined #sailfishos-porters18:43
*** lbt has quit IRC18:46
*** piggz has quit IRC18:53
*** javispedro has quit IRC18:53
*** piggz has joined #sailfishos-porters18:53
*** piggz has quit IRC19:03
*** piggz has joined #sailfishos-porters19:06
*** zon is now known as zZz0n19:06
*** Umeaboy has quit IRC19:16
*** javispedro has joined #sailfishos-porters19:35
*** alin__ has joined #sailfishos-porters19:38
*** alin_ has quit IRC19:39
*** marxistvegan has joined #sailfishos-porters19:50
*** javispedro has quit IRC19:52
*** Umeaboy has joined #sailfishos-porters20:02
UmeaboyCan the awk-issue be fixed soon?20:02
UmeaboyOr is that something I have to export?20:02
Umeaboylbt_: ^^20:03
*** carepack has quit IRC20:22
Guhlpiggz, time for noob questions?20:27
piggzgo on20:27
Guhlhow is the libhybris hwcomposer architecture actually working?20:28
Guhldoes it need the proprietary android libraries?20:28
piggzyes i think so, libhybris is an android-bionic -> 'regular linux'bridge20:31
piggzso it converts all android servcices into regular ones20:31
piggzbut im no export20:31
Guhlok let's assume they are needed - where are they supposed to be20:32
Guhlthey are going to be on the system partition20:32
Guhlbut as far as i see it system (in our case /dev/mmcblk0p25) is not mounted20:33
Guhlthe only thing that i have mounted is20:35
Guhl/dev/mmcblk0p26 on / type ext4 (rw,noatime,user_xattr,barrier=1,data=ordered)20:35
Guhl/dev/mmcblk0p26 on /data type ext4 (rw,relatime,user_xattr,barrier=1,data=ordered)20:35
Guhlso basically data once as / and once as /data20:35
piggzlet me check mine..20:37
Guhli would not ask that stupid if i would have found any kind of usefull documentation on that on the web20:38
Guhlpiggz, btw20:39
Guhl/usr/lib/qt5/bin/qmlscene -platform hwcomposer20:39
Guhlalso just gives me20:39
GuhlEGLFS: Screen Info20:40
Guhl - Physical size: QSizeF(4.29497e+09, 4.29497e+09)20:40
Guhl - Screen size: QSize(480, 800)20:40
Guhl - Screen depth: 3220:40
GuhlSegmentation fault20:40
Guhlwell it did the first 3 time, now running it again the phone rebooted20:40
piggzGuhl: i have p25 mountd on /system20:52
Guhlwell that makes sense20:53
Guhli do not even have a mountpoint for /system20:54
piggzGuhl: try EGL_PLATFORM=fbdev /usr/lib/qt5/bin/qmlscene -platform hwcomposer20:54
Guhlyes thats what i did20:54
*** javispedro has joined #sailfishos-porters20:55
*** Sage_ has quit IRC20:58
*** Sage_ has joined #sailfishos-porters20:58
piggzGuhl: i would think you need /system/lib/libEGL.so20:59
Guhli think so too but the question is where and what am i missing in the init that system is not mouted21:00
Guhlcan you show me your init script from ramdisk somewhere21:01
Guhllike init.ace.rc ...21:01
Guhlor are they not used anyway21:03
Guhlpiggz, who is owner of /system21:09
Guhlfor now i just created /system and chowned it system:system21:09
Guhland mounted /dev/mmcblk0p25 on it21:10
piggzGuhl: root:root21:10
Guhlbtw mounting system changed the output to21:11
GuhlEGLFS: Screen Info21:11
Guhl - Physical size: QSizeF(4.29497e+09, 4.29497e+09)21:11
Guhl - Screen size: QSize(480, 800)21:11
Guhl - Screen depth: 3221:11
GuhlQPA-HWC: hw_get_module(HWC_HARDWARE_MODULE_ID, (const hw_module_t **)(&hwc_module)) in create returned -221:11
piggzGuhl: do you have /vendor for firmwares21:11
Guhlyes as this is a link to /system/vendor21:12
Guhllrwxrwxrwx   1 root   root        14 Sep 21 22:01 vendor -> /system/vendor21:13
UmeaboyHmmmmmmmmm. The kernel verify script seems to have found a value that doesn't exist in the .config21:18
UmeaboyHow's that possible?21:18
UmeaboyCONFIG_CGROUP_PERF does not exist.21:18
sledgesGuhl: /system needs to be mounted on startup, check /lib/systemd/system/system.moutn21:19
sledgesUmeaboy: _PERF has been introduced two weeks ago21:19
Umeaboysledges: And?21:22
Guhlsledges, yes that says21:23
Guhlbut the  mountpoint /system did not exist ?21:23
Umeaboysledges: I'm changing what the script told me to change.21:27
sledgesGuhl: then create it:)21:29
sledgesand reboot21:29
zetazI am struggling with the boot-recovery kernel21:29
zetazethernet over USB works ok, I get udhcp too, but no way to find where the telnet is21:30
*** VDVsx has quit IRC21:31
*** phdeswer has quit IRC21:31
*** hurrian has quit IRC21:31
zetazbut lsusb on my PC tells me  the device has this serial "Mer Debug telnet on port 23 on rndis0 - also running udhcpd"21:31
zetazlooking with wireshark, the device answers to ARP requests, but that's all. nmap can't find a single port opened.21:31
zetazdoes that problem rings a bell to someone ?21:32
*** VDVsx has joined #sailfishos-porters21:32
sledgeszetaz: so your host is configured for e.g., try pinging and then (yes, there's a catch, possible;)21:33
*** dmt__ has joined #sailfishos-porters21:35
zetazsledges: udhcp give the address to my computer. Trying with .121:35
zetaz"telnet: connect to address No route to host"21:36
piggzGuhl: good luck, off to zzzZZZ21:36
*** piggz has quit IRC21:37
*** phdeswer has joined #sailfishos-porters21:37
zetazsledges : so this is not only an IP address problem...21:38
zetazlaunching an nmap search for all the devices in the range 192.168.2.* only returns two hosts up : my computer (.20) and the device (.15)....21:41
zetazI also tried previously manual configuration with IP .10 (as found somewhere in the logs of this channel), but results are the same as with DHCP21:42
Guhlsledges, systemctl start system.mount21:43
Guhlgives me21:43
GuhlA dependency job for system.mount failed. See 'journalctl -xn' for details.21:43
Guhlbut there is nothing meaningful in journal21:44
alin__zetaz: 232321:46
alin__sledges: now that dhd can be build on obs... can we do it for hammerhead too21:48
zetazalin__: same results (I ran it again to be sure). And at the step I am, it should be 23 (hybris-boot). 2323 should only be used after the boot is done21:48
sledgeszetaz: what alin said21:49
sledgesyet it's recovery, should be 2321:49
zetazhow can dhcp works, but not telnet... I am puzzled21:49
sledgesGuhl: it should work on startup. is it symlinked?21:49
sledgeszetaz: i'm puzzled that there is dhcp. should be static, and your host should statically set to 2.2 (or what you like)21:50
sledgesalin__: not up my street, bus pls try;)21:50
sledgesGuhl: find /lib | grep system.mount21:51
sledgesGuhl: find /lib/systemd | grep system.mount21:51
sledgeszetaz: nmap ?21:51
zetazsledges : "Host is up (0.0011s latency). All 1000 scanned ports on are filtered"21:52
zetazsledges: which should dhcp be there ? It is configured and launched from here :
zetaz(by DHCP I mean a udhcpd server running on the device)21:54
sledgeszetaz: does it ping back?21:55
zetazsledges: no ping back. forgot to say that21:55
zetazwith wireshark I can see some ARP requests, where the device answers with its MAC address when a request is made with its IP. but all ICMP requests sent for the ping are not answered...21:57
Umeaboysledges: Am I supposed to change the values in the .config-file or somewhere else?21:58
UmeaboyNow changing the content of /home/kristoffer/mer/android/droid/kernel/samsung/smdk4412/arch/arm/configs/cyanogenmod_i9305_defconfig21:59
zetazI am booting my laptop to check it is not an issue only with my computer. I can safely stop the firewall on the laptop too.21:59
Guhlbut i'll give it up for tonight22:01
UmeaboySeems like CONFIG_NETFILTER_XT_MATCH_NFACCT isn't in that either.22:04
sledgeszetaz: ok, if other laptop won't help, we'll need to swap usb0 and rndis0 inside init-debug22:06
sledgesand re-deploy22:06
*** _Serial has joined #sailfishos-porters22:06
sledgesalso check for *stderr* file on physical target sf rootfs22:07
beidllbt_: hey maestro, any progress on the GLIBC symbol version issue when building an image using OBS repos?22:08
beidlWarning: [droid-hal-maguro-0.0.6-10.4.1.jolla] Requires [], which is not provided22:08
beidlWarning: [droid-hal-maguro-0.0.6-10.4.1.jolla] Requires [], which is not provided22:08
beidlWarning: [droid-hal-maguro-0.0.6-10.4.1.jolla] Requires [], which is not provided22:08
Umeaboysledges: The kernel verify script must be showing me the wrong stuff to change as that doesn't exist in the defconfig.22:08
alin__beidl: so you made some progress...22:09
sledgesUmeaboy: defconfig does not show all values, just changed ones, all others are default22:10
alin__beidl: did the script work when building everything?22:10
beidlalin__: I gave up on using the script, sorry. it doesn't work the way maguro repos are set up22:10
alin__beidl: but -mwrepo "x" you need no repo22:11
Umeaboysledges: So, I move on to the next one then. I hope that this will be more accurate in the future.22:11
beidlalin__: for some reason it didn't even build the local rpms, even though mwrepo was 'x'22:11
alin__beidl: did you installt sdk-utils?22:11
beidlalin__: yup22:11
alin__what was the error?22:12
sledgesUmeaboy: new rules apply, will make it to next hadk rev:
sledges(read commit desc)22:12
beidlalin__: there was no specific error output iirc. it did mka the droid-side source code, but it wasn't packaged up22:13
Umeaboysledges: What about the defconfig for i9305 then? Not the same rule for that?22:14
alin__beidl: strange...22:15
sledgesUmeaboy: exactly the same rule, needs regenerating22:18
*** lbt_ has quit IRC22:19
*** lbt_ has joined #sailfishos-porters22:20
*** lbt_ has joined #sailfishos-porters22:20
sledgesUmeaboy: and so then ready for your first ever PR;)22:22
zetazsledges: same on the other computer (it runs Fedora too..., but without firewall for this test). Did not found a *stderr* with something interesting in it. there is an old ".stowaways/sailfishos/init-debug-stderrout" which contains only "+ exec".22:22
zetazso I will try  swaping usb0 and rndis0 as you said22:23
sledgeszetaz: ok22:23
zetazsledges: you mean swapping the order in which they are called, or swapping the USB_IFACE names ?22:24
Guhlnight guys - thanks again22:25
sledgeszetaz: order22:27
sledges01:06 < sledges> nn22:28
zetazsledges: ok. Thanks22:28
beidlsledges: nice saying :D22:28
UmeaboyAfter I edited the defconfig.22:28
sledgesUmeaboy: ubu-chroot22:28
sledgesbeidl :)22:28
*** lbt_ is now known as lbt22:29
UmeaboyRunning now...... ;)22:29
UmeaboyAre you having any plan on fixing this?22:30
Umeaboyawk: cannot open sdk/files/ (No such file or directory)22:30
Umeaboyfind: `device/*/generic': No such file or directory22:30
*** Guhl has quit IRC22:30
*** vrutkovs has quit IRC22:32
beidlsledges: Gott schuf die Tugend, der Mensch den Anschein davon. :P22:32
UmeaboyWould I have to wait long for the make process to finish?22:37
sledgesbeidl: translation?;)22:37
*** _Serial has quit IRC22:38
UmeaboyNothing seems to be happening, but I may be wrong.22:38
beidlsledges: god created virtue, mankind the appearance of it :)22:38
Umeaboybeidl: And I created a VIRTUal Linux-machine. ;)22:40
Umeaboysledges: It seems that nothing is happening, but I can leave it until mid day.22:45
sledgesUmeaboy: paste22:50
sledgesdid you eat lunch/breakfast etc?22:50
Umeaboysledges: Nope.22:51
sledgespsl do22:52
zetazsledges: Yeah ! ;)22:58
zetazinverting usb0 and rndis0 worked for telnet. I'm in !22:58
zetazbut I lost the udhcpd server, and have to set the IP by hand now (not a problem, but strange)22:58
UmeaboyI think the   <project path="device/samsung/smdk4412-qcom-common" name="CyanogenMod/android_device_samsung_smdk4412-qcom-common" /> should be either uncommented or removed from the .repo/manifest.xml23:00
UmeaboyCan you fix that, sledges?23:01
UmeaboyIt gives me this error when I wrote breakfast i9305: fatal: duplicate path device/samsung/smdk4412-common in /home/kristoffer/mer/android/droid/.repo/manifest.xml23:02
zetazsledges : thanks again for you time. I'll document this issue.23:06
zetazgood night.23:06
*** zetaz has left #sailfishos-porters23:07
UmeaboyAny way. I have to go.23:10
*** Umeaboy has quit IRC23:10
*** javispedro has quit IRC23:19
*** _Serial has joined #sailfishos-porters23:20
beidlwill have to go now as well. bye everyone.23:20
*** cxl000 has quit IRC23:26
*** olafh has quit IRC23:35
*** RealJohnGalt has quit IRC23:39
*** RealJohnGalt has joined #sailfishos-porters23:44

Generated by 2.17.1 by Marius Gedminas - find it at!