Friday, 2014-07-18

« Thursday, 2014-07-17 Index Saturday, 2014-07-19 »
sledgessitu: i9305 (and other sammys) don't have fastboot, they go through odin/heimdall01:02
sledgeshence that section01:02
*** ernesti has joined #sailfishos-porters06:16
*** furikku has joined #sailfishos-porters07:13
*** alin has quit IRC07:23
*** lbt_ has joined #sailfishos-porters08:07
*** lbt has quit IRC08:07
*** alin has joined #sailfishos-porters08:13
*** alin has quit IRC08:13
*** alin has joined #sailfishos-porters08:13
*** lbt_away has joined #sailfishos-porters08:59
*** lbt_away has quit IRC08:59
*** lbt_away has joined #sailfishos-porters08:59
*** lbt_ has quit IRC09:03
StskeepsHardware Adaptation Dev Kit 1.0 - http://releases.sailfishos.org/sfa-ea/2014-07-18_SailfishOSHardwareAdaptationDevelopmentKit.pdf09:04
*** lbt_away is now known as lbt09:05
lbtat last !! :D09:06
Stskeepsplease put it in #jollamobile and #sailfishos too, not there due to holiday reasons..09:06
lbtok09:06
lbtis this chan wide open now ?09:06
Stskeepsyes09:06
*** ChanServ sets mode: +o Stskeeps09:06
*** Stskeeps changes topic to "Official SailfishOS porters channel, see https://together.jolla.com/question/39552/what-is-the-participation-and-contribution-policy-for-jollas-open-source-contributors-in-open-source-projects/ - HADK 1.0: http://releases.sailfishos.org/sfa-ea/2014-07-18_SailfishOSHardwareAdaptationDevelopmentKit.pdf"09:07
*** ChanServ sets mode: -o Stskeeps09:07
*** Sailor2765 has joined #sailfishos-porters09:11
situAwesome !09:14
*** Anarky has joined #sailfishos-porters09:15
*** ryukafalz has joined #sailfishos-porters09:15
*** kimmoli has joined #sailfishos-porters09:16
Anarkylbt: page 16 the commands got eat09:20
lbtAnarky: ?09:20
sledgeseaten ;)09:21
sledgesis his intended comms about error :))09:21
lbtyeah - what do you think should be commented Anarky ?09:21
sledgesi found more, fixing now09:22
Anarkyyeah, sorry for my english, I mean the commands exceed the page width so latex cut it09:22
lbtnoooo09:22
lbtwell - you can cut'n'paste it :/09:23
Anarkynope, I tried a triple-click without success09:24
lbtclick/drag-down worked for me09:24
lbtI think we'll need to republish09:24
Anarkymaybe it's my pdf reader but the click/drag-down doesn't work for me09:26
*** lainwir3d has joined #sailfishos-porters09:31
ryukafalzThat... would be your PDF reader, yeah :P09:32
Anarkyfor the record, it's evince09:37
Anarky(same problem with epdfview)09:40
Anarkyok, with Firefox it works fine :)09:45
sledges^_^ glad you found a quick workaround09:45
sledgesfix is under way09:46
*** locusf has joined #sailfishos-porters09:50
* lbt uses okular and it wfm there10:02
*** Flyser has joined #sailfishos-porters10:02
*** zuh has joined #sailfishos-porters10:10
situsledges: Do you see RTC_SET_TIME being used anywhere in our build tree ?10:17
*** SK_work has joined #sailfishos-porters10:18
locusfis there supposed to be a contents page on HADK 1.0?10:23
locusfI can just see "Contents" on the page but nothing more10:23
sledgeslocusf: will be fixed in next ver10:23
sledgessitu: sec, need to boot up server10:24
locusfproblem in repo init in page 1110:29
locusfgpg: Signature made Thu May  1 20:34:18 2014 UTC using RSA key ID 692B382C10:29
locusfgpg: Can't check signature: public key not found10:29
locusferror: could not verify the tag 'v1.12.16'10:29
sledgeslocusf: checking10:32
locusfah, ancient repo command in my ~/bin10:33
sledges^_^10:33
locusfI played around with android in 2011 or something10:34
Flyseris it normal that I get "bash: no job control in this shell" when entering the mer sdk?10:57
*** lbt has quit IRC10:58
sledgesFlyser: probably some env variables10:58
sledgesas long as you enter sdk, all good10:58
*** lbt has joined #sailfishos-porters11:03
*** lbt has quit IRC11:03
*** lbt has joined #sailfishos-porters11:03
*** alin has quit IRC11:04
locusfsome pages have overflow on the sides11:11
locusfpage 19 for example11:11
locusfand 1611:11
sledgeslocusf: thanks, fixed in next releease11:13
sledgescan you cope paste?11:13
sledges(hehe interesting typo)11:13
locusfcannot copy-paste11:14
locusfis the guide done in latex?11:15
Anarkylocusf: I had the same problem, you can use Firefox's pdf.js to copy11:16
Anarkywith evince or epdfview it doesn't work11:16
ryukafalzWorks with okular11:18
ryukafalzah, already been mentioned11:18
locusfI use okular too11:19
* ryukafalz uses mostly KDE stuff so yeah11:19
Flyserthe variables on page 9 should probably be escaped, so that $MER_ROOT in $ANDROID_ROOT gets evaluated at runtime11:21
sledgesFlyser: ' instead of " ? i disagree, as you might want to change MER_ROOT on the fly (been there done that :)11:22
Flyseryeah, but ANDROID_ROOT is relative to MER_ROOT, so it should also change when you change MER_ROOT, right?11:23
sledgesnope11:24
sledgesandroid root can be (and should be) not within čer_root11:24
sledges*m11:24
Flyserhuh? page 9 says: ANDROID_ROOT="$MER_ROOT/android/droid"11:24
sledgesmer_root should be left homogeneous11:24
sledgestouche` ;D11:25
Flyser^^11:25
sledgesbut it's not like that on my setup :))11:25
sledgesso ok, it's fine to place it within mer_root, but escapes shouldn't be there, as mer_root can be  relocated (if you run out of space), together with its android_root if it's in11:26
sledgesso it evaluates on every startup11:26
sledgesand there are less places to change, if path changes11:26
Flyserthat's exactly my argument :P11:27
Flysercurrently it gets evaluated right within the cat << EOF part11:27
situsledges: Is this a good format for build instructions for Nexus 5 ? https://wiki.merproject.org/wiki/Building_Sailfish_OS_for_Nexus_511:29
situvgrade_: ^11:29
ryukafalzmerproject wiki cert expired D:11:29
sledgesFlyser: double quotes " should't get evaluated, as opposed to '11:30
ryukafalzwho manages that?11:30
sledgesryukafalz: cheers, known issue11:30
sledgeslbt11:30
Stskeepslbt: ^11:30
ryukafalzAhh, okay11:30
Flysersledges: uhm ... echo '$HOME' gives "$HOME"; echo "$HOME" gives "/home/flyser"11:32
Flyserthis is on a gentoo system running bash11:32
sledgesok, same here, it's been a long night ;)11:32
Flyser:)11:32
sledgesyes, you're right, i didn't remember what i did myself few weeks back: https://wiki.merproject.org/wiki/Platform_SDK#tl.3Bdr :))11:33
sledgeschanging to single quotes for next hadk rel, ta!11:33
lbtI renewed the wiki cert a while ago - it seems it needs to be 'aged' before I can deploy it11:39
ryukafalzO.o11:41
Flyserpage 11 could use some quotation: e.g. sudo mkdir -p "$ANDROID_ROOT"11:42
Flyserin case ANDROID_ROOT contains spaces11:42
FlyserUBUNTU_CHROOT on page 10 as well I guess11:43
sledgesooooh spaces, in linux?11:45
sledges;)11:46
sledgesi guess that's diggin your own hole11:46
ryukafalzugly ugly :P11:46
sledgesas android sdk would also go bananas11:46
FlyserI don't have spaces in the path, but others might11:46
*** nodevel has joined #sailfishos-porters11:51
* Flyser wishes for more information on "unsupported devices"12:10
sledgesthere are many more unofficial cm devices12:11
sledgeswhich have not 100% QA`d kernel/rootfs12:11
sledgesthe principle is the same, but expect dragons :)12:12
Flyserah ... I thought the italic phrase "List of Supported Devices" refers to the devices supported by sailfish (=the nexus devices) and not all cm-supported devices?12:13
*** vrutkovs has joined #sailfishos-porters12:14
sledgesFlyser: in that case, more information is in chapter 1412:15
FlyserI hope that is correct behaviour? https://gist.githubusercontent.com/flyser/42ad0872f961f4ce474f/raw/a9c6cf1753bacbab7107a44163f3b34e6bd42712/gistfile1.txt12:16
Flyserthis is page 12 ↑12:16
sledgesFlyser: if breakfast does not end with fatal error, you should continue12:17
Flysergreat :)12:17
Flyser"make -j4 hybris-hal" on page 12 fails, because gcc is not found only /usr/bin/gcc-4.7 is there12:20
FlyserI can fix this locally, but I thought I should mention it nevertheless12:21
sledgessudo apt-get install gcc ?12:21
sledgesor anything extra12:22
situsledges: Isn't he cross compiling and needs gcc for arm ?12:22
Flyseri just symlinked gcc-4.7 to gcc12:22
sledgeswhich gcc does it require?12:22
Flyseroh ... good point12:22
sledgescross compilers are within android sdk12:22
situgcc --version ?12:23
sledgesprobably some gcc for housekeeping12:23
sledgesi mean, which gcc does make -j4 hybris-hal require ?12:23
sledges( Flyser has/usr/bin/gcc-4.7 on board)12:23
sledgesstrange why there's no /usr/bin/gcc12:24
sledgesdid you try apt-get ?12:24
Flyserhttps://gist.github.com/flyser/a944245e9fadf0c6cd3712:24
Flyserthis is gcc-4.7 ↑12:24
* sledges nods12:24
sledgesexpected12:24
Flysershould I try apt-get?12:25
sledgesrm symlink12:25
sledgesand see if apt-get fixes12:25
sledgesooi12:25
Flyseryeah ... name resolution is broken :P12:26
sledgeso.O12:26
Flyseroh wait it isn't. it just fails to fetch gcc12:26
sledgesyeap, get some internet :)12:26
sledgesapt-get update ?12:26
sledges(i forgot ubuntu already :))12:26
Flyserhttps://gist.github.com/flyser/afd9f870f5a1572354e312:27
sledgesE: Unable to fetch some archives, maybe run apt-get update or try with --fix-missing?12:27
Flysereh .. yeah. I was just reading the --fix-missing part :P12:27
Flysersrsy12:27
Flysersry12:27
sledgesnp :)12:27
FlyserW: Failed to fetch http://archive.ubuntu.com/ubuntu/dists/quantal/universe/binary-amd64/Packages  404  Not Found12:28
sledgesping yahoo.com ;)12:28
Flyserworks. otherwise it would not be 40412:28
sledgessec12:29
sledgeswfm - DNS problems?12:29
sledgeshang on12:30
Flysercan you access http://archive.ubuntu.com/ubuntu/dists/quantal/universe/binary-amd64/Packages ?12:30
Flyserit does not work in my browser either12:30
sledgesit refreshed enough12:30
Stskeeps /etc/resolv.conf problem12:30
Stskeeps?12:30
sledgesnope, it updated most imporatant ones12:30
vgrade_situ: looks good to me12:30
sledgesthe other repos fail, but should be fine12:30
situvgrade_: ok, thanks.12:30
sledgessudo apt-get install gcc12:31
sledgesFlyser: try ^12:31
sledgesnow12:31
vgrade_Stskeeps: what's the status on distribution of images?12:31
sledges(i had 404 on last ~55% of multiverse etc repos)12:31
FlyserFailed to fetch http://archive.ubuntu.com/ubuntu/pool/main/g/gcc-defaults/gcc_4.7.2-1ubuntu2_amd64.deb  404  Not Found12:31
sledgesFlyser: which ubuntu chroot img did you dl ?12:33
Stskeepsvgrade_: not really a legal person but i wouldn't terribly worry12:33
Stskeepsusual constraints apply, no 3rd party bits12:33
tbrvgrade_: if you need folders for images, let me know12:34
Flyserhttp://img.merproject.org/images/mer-hybris/ubu/ubuntu-quantal-android-rootfs.tar.bz2 as the guide suggested12:35
vgrade_Flyser: sledges host gcc was missing at one point12:36
vgrade_tbr: understood12:36
sledgesvgrade_: Flyser: ouch, i think they removed 4.7 recently, as i stll have it for when i did apt-get install: gcc (Ubuntu/Linaro 4.7.2-2ubuntu1) 4.7.212:37
Flysermaybe a LTS release of ubuntu would have been a better choice12:38
Flysershould I just symlink it again or do you want me to test some other workaround?12:39
sledgesRE LTS: absolutely, we'll still be able to switch to it12:39
sledgesFlyser: thanks, that's more than enough of info for now12:39
Flyserokay, lets see if symlinking works12:40
sledgesFlyser: so gcc-4.7 was already pre-installed, only symlink missing?12:46
Flyseryes12:47
Flyserlets see if gcc is enough. maybe it requires more binaries like ar or nm12:47
* sledges nods12:48
SazpaimonHey guys, thanks for releasing the HADK. I have a quick question. I want to do a Sailfish port to the Oneplus one, which comes with CyanogenMod 11. Should I use the hybris-11 branch?12:53
*** lbt_ has joined #sailfishos-porters12:53
*** lbt has quit IRC12:57
vrutkovsargh, I keep getting "sudo: effective uid is not 0, is sudo installed setuid root?" when I'm running and sudo command in SDK12:58
vrutkovsany ideas how to fix it?12:58
FlyserI suppose you didn't use sudo to extract the sdk12:59
vrutkovsFlyser: oops, my fault - messed up some ownership, thanks13:00
sledgesSazpaimon: yes13:01
vgrade_Sazpaimon: yes, I'm using repo init -u git://github.com/CyanogenMod/android.git -b refs/tags/cm-11.0-XNPH22R-bacon-03d77315ea13:01
vgrade_Sazpaimon: as a base13:02
Sazpaimonoh, you've already done some work on a oneplus port?13:02
vgrade_Sazpaimon: WIP13:02
Flyserbtw. which SoC does the Jolla use?13:02
Sazpaimongreat, I'll see what comes of that considering that I don't actually physically have a device yet13:03
Stskeeps8930aa13:03
Stskeepswell, msm8930aa13:03
Flysergreat. I am porting to a 8960, so sailfish should run fine13:04
Stskeepsyeah, n4 is a 8960 too13:04
FlyserI thought n4 is APQ8064?13:06
Stskeepshmm13:07
Stskeepsyeah, okay, but practically it's same thing :P apq just has modem seperate13:07
Stskeepsie the MDM attached over USB13:08
Flyserah. didn't know that13:08
Flyserhow long does the make hybris-hal take on a current average PC?13:13
sledgesmuch less than full android build13:13
Flysernever done a full android build :)13:13
Stskeepsbe glad.13:13
Stskeeps:P13:13
Flyser:D13:13
*** ernesti has quit IRC13:13
Flyserit fails after "Fixing mount-points for device xt897" ... no specific error message :-/13:16
sledgeslbt_: ^13:17
sledgesdebugy times ;)13:17
Flyserseems intended. "exit 1" in droid/hybris/hybris-boot/fixup-mountpoints13:19
lbt_ah13:19
*** lbt_ is now known as lbt13:20
sledgesFlyser: you need to provide your device there13:21
sledgesand it's mountpoint maps13:21
lbtFlyser: does it not print a big banner?13:21
Flyserno https://gist.github.com/flyser/8fff6d5088bc3dc7e55f13:22
lbtno13:22
lbthttps://github.com/lbt/hybris-boot/blob/hybris-10.1/fixup-mountpoints :/13:23
Flyserwouldn't it be possible to write a script that extracts the mount points from the CM source?13:24
lbtsledges: https://github.com/mer-hybris/hybris-boot/pull/22 *cough*13:25
locusf/android/droid/external/busybox/scripts/gcc-version.sh: line 11: cc: command not found13:26
locusfI get this error in make -j4 hybris-hal13:26
locusfhttps://gist.github.com/locusf/1283e902b83fe53c5bd313:27
locusffull output13:27
Flyserlocusf: yeah ... the same problem13:27
locusftrying to do samsung galaxy note build (n7000)13:27
Flyserlocusf: it fails in fixup-mountpoints13:28
Flyserand it *should* print "ERROR: $DEVICE does not have mountpoint fixup data - see SailfishOS HADK for details on how to fix this. Check: device/$VENDOR/$DEVICE/recovery.fstab"13:28
locusfFlyser: okay13:28
lbtthat cc problem ... I thought it was resolved - we had a tarball which had gcc missing iirc13:30
Flysermhhhh ... no recovery.fstab for my device13:31
Flyserweird13:31
lbtyeah - the fstab stuff is super annoyingly inconsistent13:31
locusfI do have it though13:32
sledgeslbt: *cough* *cough* my pardons :)13:32
lbtnp - it's been easy to overlook these things13:32
lbtalso I think we've missed something gcc-ish in the ubu tarball13:33
sledgesgcc case is odd , as gcc-4.7 binary -is- there13:35
sledgesso only symlink away?13:36
sledgesapt-get install gcc fails though as they removed 4.7 from repos13:36
lbtmultiarch ?13:36
lbtoh balls13:36
lbtanyhow .. gotta go to physio ... bbiab13:36
sledgeslbt: i want to merge that PR13:37
sledgesfeeling very confident since last comment13:37
sledgesif that kick Flyser ahead13:37
sledgeswith progress13:37
sledges*kicks13:37
sledgesFlyser: you could as well cherry-pick that ;)13:44
Flyserunless I understand the commit wrong, it does not fix the error, right?13:46
Flyserit just prints a better error message13:46
locusfhmm so my device is currently completely unsupported by fixup-mountpoints?13:50
*** phdeswer has quit IRC13:50
sledgeslocusf: Flyser: sub-section 14.3.213:51
sledgesFlyser: commit should do a better autodetection13:54
sledgeslots of stuff changed: https://github.com/mer-hybris/hybris-boot/pull/22/files13:54
locusfsledges: ok13:55
Flyserokay ... I cherry-picked it13:57
Flysernow the build process hangs due to perl13:58
locusfhow do I cherry-pick it?13:58
locusffrom lbt's repo?13:58
Flyserlocusf: apply https://github.com/mer-hybris/hybris-boot/pull/22.patch13:59
*** ParkerR has joined #sailfishos-porters13:59
ParkerRsledges, :)13:59
FlyserI think the awesome perl oneliner hangs if device/*/$(TARGET_DEVICE)/recovery.fstab does not exist13:59
locusfFlyser: cool14:01
locusf/bin/bash: fork: retry: Resource temporarily unavailable14:01
locusf/bin/bash: fork: retry: No child processes14:01
locusflol14:01
sledgesParkerR: o714:01
locusffail14:01
Stskeeps.. did you forkbomb your system?14:02
Flyser:D14:02
Flysersweet14:02
ParkerRsledges, How's it been?14:02
*** energycsdx has joined #sailfishos-porters14:02
locusfhmm I was as hosts root in android sources14:02
sledgesParkerR: busy ;)14:04
sledgesyou?14:04
ParkerRsledges, pretty good14:06
locusfI forgot that breakfast fails too https://gist.github.com/locusf/71bcd62f13f62e7e1f5c14:06
sledgesso, happy times have come ParkerR ;)14:06
locusfnew device is probably quite a messy thing to do with this HADK?14:07
sledgeslocusf: 5.3 ;)14:07
locusfI already did remove it !14:08
locusfI don't know how it got back14:08
locusfit gets recreated14:09
Flyseron page 13 ... where are the sb2-init and sb2 binaries supposed to come from?14:11
sledgeslocusf: yes14:11
locusfsledges: https://gist.github.com/locusf/71bcd62f13f62e7e1f5c14:11
sledgesFlyser: mer sdk14:11
sledgesexit android chroot14:12
sledgeslocusf: most likely needs investigating where that forbidden comes from14:12
sledgesFlyser: *ubuntu chroot14:13
Flyserthx14:14
energycsdxbtw any succes on running sailfish os on other devices?14:14
sledgesenergycsdx: all that had success before today: https://wiki.merproject.org/wiki/Adaptations/libhybris14:15
sledgesexpect the list to grow and greenify :)14:15
energycsdxsledges: certificate expired few days ago :(14:16
Flyser (ERROR)        zypper[8143/-147499264] connect: Mapped AF_UNIX address (/parentroot/home/flyser/playground/sailfish-hadk/mer/targets/motorola-xt897-armv7hl/run/dbus/system_bus_socket) is too long14:17
Flysersounds bad ... :(14:17
locusfsledges: API rate limit exceeded14:18
sledgesenergycsdx: working on that ta14:18
sledgesFlyser: max path or?14:18
Flyserguess so14:18
sledgesFlyser: which command14:18
sledgeslocusf: ?14:19
Flyserpage 13, sb2 -t $VENDOR-$DEVICE-armv7hl -m sdk-install -R zypper ref --force14:19
locusfsledges: github api rate limit14:19
locusfsledges: cyanogenmod's scripts use it14:20
sledgesgot over abused? :)14:20
locusfyeah14:20
locusfgotta wait ~hour14:20
energycsdx ok i have Sailfish OS partialy running on Sony Xperia L14:20
sledgesenergycsdx: sounds good14:20
locusfsledges: had to do curl -u "username" https://api.github.com before relaunching breakfast14:22
locusfroomservice.xml keeps regenerating14:24
sledgesyes, it does14:24
locusfwhen is the right time to remove it?14:24
sledgeswhen you get the error14:24
locusfI did and reran breakfast to come up with the same error14:24
sledgesFlyser: any other output on that command?14:25
sledgestry lunch14:25
energycsdxsledges: i started before HADK was published, i have running lipstick in chroot on android, and also phone boots from hybris boot but something wrong with lipstick, i guess it because i`m not installed droid-hal-$device.rpm, yesterday i managed to compile it, i`ll continue my fun with it on weekends14:25
* sledges nods energycsdx 14:25
energycsdxi`m using AOSP14:25
sledgesalso hwc will need to link against14:25
sledgesqpa14:25
sledgesenergycsdx: patches welcome ;)14:26
Flysersledges: https://gist.github.com/flyser/9fe03e83c38b7293a55214:26
energycsdxi know, lipstick works i chroot but still have problem when starting native, my guess it because missing udev rules and other configs14:27
sledgescould be yep14:28
sledgesaosp patches, i meant14:28
locusf             total       used       free     shared    buffers     cached14:28
locusfMem:         32133      31835        298        134       2137      1011514:28
locusflol I wondered why fork() doesn't work14:28
sledgeso.O14:28
locusfout of 32GB14:28
sledgeslul14:28
sledgeslooks pretty forked14:29
* sledges hides14:29
energycsdxsledges, i didn`t patch ASOP too match14:29
sledgesenergycsdx: we have lbt's effort14:30
sledgesyou could sync up14:30
sledgeswhich aosp btw?14:30
energycsdx4.4.414:31
sledgesenergycsdx: https://github.com/lbt/android/blob/hybris-aosp-4.4.2_r2/local_manifest.xml.template14:31
sledgesso it differs14:32
sledgesFlyser: any chance of moving MER_ROOT to $HOME (deja vu of our previous convo ;P )14:32
sledgesand retrying14:32
FlyserI am currently trying to ingnore the errors14:33
Flyserdo you think it might work?14:33
*** vrutkovs has quit IRC14:33
sledgesa14:34
sledgesthen carry on with another sb2 command14:34
sledgessee if it succeeds at all14:34
sledges(involving zypper/ssu)14:34
sledgesFlyser: struct sockaddr_un { unsigned short sun_family;  /* AF_UNIX */ char sun_path[108];14:35
sledgesand your path is 111 chars14:36
sledgescrazy :)14:36
*** vrutkovs has joined #sailfishos-porters14:36
energycsdxsledges: i checked, i patched only frameworks/base/Android.mk and cherypicked bionic hybris patch, i think maybe possible to not patch Android.mk14:37
* sledges nods14:37
sledgesit would just be nice to finally have it mer-hybris upstream wing14:37
sledgesbut more tests i guess ;)14:38
sledgeswhen we get your xperia up14:38
sledgesyou can start properly now with hadk14:38
Flysersledges: the second zypper ref fails as well, the other commands work14:38
sledgesFlyser: ^14:38
Flysercreating and running the hello world works as well14:38
sledgesthat doesn't require repos14:38
sledgeswait till you bump the wall14:38
sledgesthough i wish you wouldn't ;)14:38
sledgesmaybe in sb2 those dbus socks aren't critical14:39
sledgesso yeap, go ahead14:39
Flyser*bump* :D14:39
Flyserbuilding the spec file does not work14:39
Flyserhttps://gist.github.com/flyser/5bef6e3e3fbea09dc06314:40
FlyserI presume I need kernel 3.4? :(14:40
locusfsledges: put somewhere that one might need to create ~/.netrc in order to avoid the rate-limit things in github api14:41
sledgeslocusf: and what's inside that netrc?14:50
locusfhttps://gist.github.com/locusf/9f91a614398c59f5188b <- now su is needed14:50
sledgesyour account?14:50
locusfsledges: yes14:50
locusffor api.github.com14:50
locusfsledges: https://confluence.atlassian.com/display/STASH/Permanently+authenticating+with+Git+repositories#PermanentlyauthenticatingwithGitrepositories-Usingthe.netrcfile14:50
sledgeslocusf: check if we didn't remove su as redundant by chance first14:53
locusfsledges: how do I check your mistakes :) ?14:53
sledgeslol, i'd rather call it intentional cleanup of manifest ;)14:54
sledgessee mako as most worked-on reference14:54
sledgesmer-hybris/android hybris-10.1 branch14:54
sledgesdefault.xml14:54
sledgescompare with originals14:54
sledgeslook for "su"14:54
Flysersledges: what is the minimum kernel version requirement of sailfish os?14:57
locusfwhere is mer-hybris/android in the tree?14:57
sledgesFlyser: 2.6.32 ?14:57
Flyserokay, good14:57
sledgesbut with later sailfish os it might have gone up14:58
sledges:/14:58
energycsdxFlyser: usually android kernel have some config turned off, this configs is critical for Sailfish OS (systemd), you should recompile kernel for you device but with proper configs14:58
sledgesbecause of sustemd14:58
sledgeslocusf: it's on github14:58
Flyserenergycsdx: yeah, I already figured that out :) I just asked, so I don't waste my time14:58
Flyserwhich kernel version is the jolla running on?14:58
locusfsledges: there's no su in that default.xml?14:59
energycsdxi guess jolla run version that is in BSP, it is fork from CAF and there is version 3.415:00
Flyserok. lets hope my 3.0 is current enough15:00
locusfsledges: https://github.com/mer-hybris/android/blob/hybris-10.1/cm-10.1.3.xml <- it is here15:00
sledgeslocusf: means can be gone15:00
locusfsledges: CM_VERSION=10.1-20140718-UNOFFICIAL-n700015:02
locusfthats the version I'm trying to create, do I need to have a corresponding .xml?15:02
sledgeslocusf: you are at testing phase of your CM15:02
sledgeswrt hadk15:02
sledgesfor sfos you won't need su15:03
sledgesit will tell you how to create manifests (i hope :D )15:03
sledgesbut if you want to be sure to be sure15:03
sledgesput that kosch's su in and compile cm and deploy it15:03
sledgesso you have confidence you haven't missed anything15:03
sledgesor go happy-go-lucky ;)15:04
locusfin the manifest file?15:04
sledgesyes15:04
locusfok15:04
locusf/bin/sh: 1: gcc: not found15:05
Flyser:)15:05
locusfI prolly need that symlink now Flyser ?15:05
Flyserlocusf: yes15:05
Flyserln -s /usr/bin/gcc-4.7 /usr/bin/gcc15:06
sledgeslocusf: hang on15:06
sledgestry15:06
sledgessudo apt-get install build-essentials15:07
locusfsledges: https://gist.github.com/locusf/070f0857364c7cd921b115:08
sledgeslocusf: symlink15:09
sledgesthanks for that btw15:09
locusfsledges: yeah symlinked but the build only lasts about 5 seconds after the symlink15:10
locusfsledges: https://gist.github.com/locusf/b001e0d29c68765e5eef15:10
sledgeslocusf: pull in su15:11
sledgesrepo sync15:11
locusfsledges: ok15:12
Flyserwhere is the kernel config stored and is it possible to edit it with make menuconfig? if so how? (would be cool to have this in the HADK)15:12
FlyserHADK guide*15:13
locusfsu needs again sqlite15:13
locusfadding as I go15:14
sledgeslocusf: story of my life :)15:15
sledgesFlyser: check commits to mako (lge_mako_kernel repo)15:16
locusfok another fixup error15:16
sledgescmdline mer kernel checker is all you need15:16
sledgesdon't try to touch other parts that cyanogenmod/aosp has decided for you before15:16
Flyserfound it thx15:16
Flyserkernel checker is all I need? but if it fails, I do have to modify the config, no?15:17
sledgesyes, by using VIM15:18
sledges;)15:18
Flyserokay :)15:18
sledgesit would take ages to go through all those oldconfigs/gens etc15:18
Flyserhow to regenerate droid/out/target/product/$DEVICE/obj/KERNEL_OBJ/.config15:21
sledgesdarn i did that once; does it not pickup?15:23
Flyserno. deleting doesn't work either15:24
sledgesFlyser: make -j16 hybris-boot15:24
sledgesgets the kernel going15:24
Flyserthx15:25
sledgeshelped?15:25
Flysersecond ...15:25
sledges^^15:25
Flyser-j4 for me. -j8 box is on its way :P15:26
sledges;)15:26
sledgeslucky to have a rack over here :)15:26
*** lbt has quit IRC15:28
Flyserperl hangs again15:29
*** GranPC_ has joined #sailfishos-porters15:29
*** alterego_ has joined #sailfishos-porters15:33
locusfsigh, llvm compiling now :/15:34
sledgeso.O15:35
sledgesis normal or you still get shivers? :)15:36
locusfI have no idea why its compiling15:36
*** lbt has joined #sailfishos-porters15:38
*** lbt has quit IRC15:38
*** lbt has joined #sailfishos-porters15:38
sledgeslocusf: full cm build?15:39
sledgesi get llvm compiled too for hadk-hal15:39
locusfsledges: no, just for hybris-hal15:40
sledgesyes that ;p15:40
sledgesi mean, it's not mesa, so don't fret at it :))15:41
locusfokay15:42
*** lbt has quit IRC15:43
*** lbt has joined #sailfishos-porters15:44
*** tbr has quit IRC15:44
locusfI see that make -j4 misses out on stuff so gonna build with -j1 instead15:44
*** alterego_ has quit IRC15:46
*** Sailor2765 has quit IRC15:48
*** Sailor2765 has joined #sailfishos-porters15:48
locusfFlyser: did you do the fixup-mountpoints manually?15:51
locusfsledges: https://gist.github.com/locusf/096dd78befe6cc8f1d2615:52
Flyserlocusf: yes15:52
locusfFlyser: how?15:53
locusfjust modified that script with your fstab?15:53
Flyserexactly15:53
*** Sailor2765 has quit IRC15:56
locusfFlyser: did you need to have anything replaced in the sed?15:57
Flyseryes http://bpaste.net/show/475397/15:58
FlyserI doubt that all lines are neccessary, but it wont hurt :)15:59
locusfhow did you figure them out?15:59
Flyserls -l /dev/block/platform/*/by-name16:00
Flyseron the device16:00
Flysersomehow my kernel config changes do not take effect ...16:02
FlyserI wonder why16:02
sledgesFlyser: you do modify *_defconfig equivalent right?16:03
Flyseryes16:03
Flysermsm8960_defconfig16:03
sledgesare you sure it's the one cm picks up?16:03
sledges(o think there's BOARD_*CONFIG value somewhere in .mk)16:04
locusfso the fstab is going to be wrong if I don't add the substitutions to the fixup-mountpoints?16:06
sledges.zip installation will certainly fail16:08
sledges(good measure to see where else fixup-mountpoints is referenced)16:08
*** tbr has joined #sailfishos-porters16:13
FlyserI don't know if its the right one16:14
Flysercant seem to find out ...16:14
locusfreran with new fstab16:14
locusfis it bad that there is only dw_mmc in cyanogenmod console?16:14
locusfthe other block device didn't list anything by-name16:15
Flysermaybe your device is like the i9305?16:15
Flyserand you only need to 's /block/ / ' \16:16
locusfthat just replaces every /block/ with nothing, right ?16:17
Flyseryes16:17
locusfwhere should I see the result of that fixup?16:17
Flyserdont know sorry16:18
locusfwell fingers crossed then :D16:18
Flysersame here :P16:18
locusfI have both kernel and recovery images now <316:19
Flyserme too *highfive* :D16:20
locusfJolla boys don't even have to bother having other toolchains, eg. for other armv8 arches16:20
locusfwe just use the one for Jolla and n9 etc16:20
sledgesthey're all armv7hl thankfully16:22
sledgesfor now ;P16:22
locusfyeah and note is too16:22
locusfFlyser is your cpu armv7hl too?16:22
sledgeslocusf: you need to see if an mmc partition is under same path in cwm and in fully booted droid16:22
Flyserlocusf: yes. I guess almost all android devices run on armv716:23
locusfsledges: ok16:23
locusfsledges: which one of them btw?16:24
*** tbr has quit IRC16:24
locusfsionce you said an, I guess that means that there is more than one16:24
Flysersledges: do you have any idea how I can figure out which defconfig is selected?16:24
FlyserI can't find a config file or anything ...16:25
sledgeslocusf: userdata and boot16:26
sledgesany will do really, you just need to see path naming consistency16:26
sledgesFlyser: find -maxdepth 6 -name "*.mk" -exec grep -H defconfig {} \;16:26
locusfMerSDK locusf@locusf-desktop:~$ cd $ANDROID_ROOT16:27
locusfbash: cd: /android/droid: No such file or directory16:27
locusfwhoops16:28
sledgeslocusf: check $MER_ROOT in ~/.*.env16:28
sledgesalso running hadk [DEVICE] helps16:28
locusfexport MER_ROOT="/srv/mer"16:29
sledgesand ANDROID_ROOT ?16:29
locusfexport ANDROID_ROOT="/android/droid"16:29
sledges;P16:29
sledgesfirst, use ' quotes16:29
Flyserthanks its msm8960_mmi_defconfig16:30
sledgesFlyser: you can also see it being picked up if you analyse cm kernel build log16:30
locusfaah16:30
locusfI need $MER_ROOT/android/droid16:31
locusfand $MER_ROOT needs to be something else16:31
Flyserfor the record: it was defined in $MER_SDK/android/droid/device/$VENDOR/*-common/BoardConfigCommon.mk16:31
sledgesFlyser: that :")16:31
FlyserI was searching through the kernel directory ^^16:32
sledges:)16:32
sledgesif Torvalds finds an .mk there.... x)16:32
locusfsledges: so single quotes?16:32
Flysersledges: :D16:33
locusfI'm at page 15 btwe16:33
Flyseractually you need double quotes there16:33
sledgeslocusf: where is your android root physically? point android_root there (it doesn't have to be under mer_root, it was just an instance)16:33
sledgeswha? :DD16:34
Flysersingle quotes are required on page 916:34
locusf/parentroot/srv/mer/sdks/ubuntu/android/droid16:34
Flyserbut if you edit the file manually you need double quotes16:34
sledgeslocusf: ah it's the case of /parentroot16:35
sledges"/parentroot/$MER_ROOT/android/droid"16:35
sledgeswill do16:35
sledges"/parentroot/$MER_ROOT/skds/ubuntu/android/droid"16:36
sledges:P16:36
sledgesdon't know why you put it under ubuntu, but as i say, it can be virtually anywhere16:36
sledges*sdks not skds (sounds like skidmarks :)))16:36
locusfsledges: https://gist.github.com/locusf/62c7304a51a1050cbcbd16:36
sledgeslocusf: adjust kernel flags16:37
locusfCan't open ./out/target/product/%{device}/obj/KERNEL_OBJ/.config: No such file or directory at /usr/bin/mer_verify_kernel_config line 58.16:37
sledgessec 14.3.516:37
sledgescommand used?16:38
locusfsledges: in the rpm postinst?16:38
locusfas can be seen in the gist16:38
sledgesrun manually in section 14.3.5 for now16:38
sledgesthanks for that16:38
locusfok16:38
* Flyser cries while compiling the kernel a fifth time16:39
*** vakkov has joined #sailfishos-porters16:41
sledgesonly fifth? ;P16:41
Flyser:P16:41
sledgesvakkov: o/16:41
FlyserI should have done this on my quadcore desktop pc and not on my dualcore notebook ..16:42
vakkovhi there!  :)16:42
*** pat_o has joined #sailfishos-porters16:43
sledgeslocusf: sec 7.1.1: # THE COMMAND BELOW WILL FAIL. It's normal, carry on with the next one.16:44
sledges;)16:44
sledgesexplanation right below :)16:44
locusfsledges: yeah but then I don't have the droid-hal-$DEVICE.spec16:45
sledgesvakkov: how you've been?16:45
Flyserlocusf: copy the mako one16:45
Flyserand adjust16:45
sledgeslocusf: 14.4.116:45
locusfFlyser: okay16:47
sledgessimply, you follow chapter 14, and then where apropriate, it swings you back to needed route (like in 14.4.3 )16:47
sledgesas you're adding a new device16:48
Flyserits a bit weird to have this split between multiple chapters tbo16:48
locusfsledges: where is the kernel config to modify?16:48
sledgesFlyser: proposals welcome :)16:49
sledgeslocusf: 19:31 < Flyser> for the record: it was defined in $MER_SDK/android/droid/device/$VENDOR/*-common/BoardConfigCommon.mk16:49
locusfsledges: ok16:49
Flysersledges: proposal: create a reference in 7.1.1 saying "If you are adding a device also see 14.4.3" or something16:50
sledgeshadk is existing-device-centric, as we aim to improve current HAs16:50
sledgesotherwise ch 7 would be called "add a new device" :)16:51
locusfsigh, text overflow on page 4416:51
*** lainwir3d has quit IRC16:51
locusfsledges: it wasn't there16:52
locusfsledges: the kernel config16:52
locusfthat was simply a makefile16:52
FlyserI understand that, but it's difficult to know if I should look for a solution in chapter 14 or if there is some other problem16:52
Flyserlocusf: the makefile tells you the filename of the config16:53
Flysersomething_defconfig16:53
sledgeslocusf: aware of overflow, don't know what to do about that, just use a smarter pdf reader for now.. soz16:53
sledgesFlyser: probably table of contents could help ;)16:53
Flysersledges: :P16:54
sledgestotally my fault16:54
sledges(goes to buy everyone beer)16:54
*** SK_work has quit IRC16:54
Flyserits over 30° here, so a cold beer would be appreciated :)16:55
vrutkovshalp! http://pastebin.com/USXZrjL416:55
sledgesit's 24 here and that's at least 4 deg above summer average :))16:55
locusfFlyser: no defconfig found :/16:57
sledgesvrutkovs: on hybris/hybris-boot cherry-pick  .... (wait a min)16:57
locusfor any *defconfig*16:57
sledgesvrutkovs: https://github.com/mer-hybris/hybris-boot/pull/22.patch (thanks to Flyser )16:57
vrutkovssledges: ah, thanks16:57
sledgesif that gets rid of your error (pls tell), i'll merge PR16:57
locusfhmm everyone is trying to create new device and it seems that it is quite difficult :p16:58
Flyserlocusf: maybe n7000_defconfig?16:58
sledgeslocusf: and you thought creating new device will be easy?16:58
* sledges thinks about months we brought mako & co to where they are16:58
Flysersledges: but as I said this pullrequest contains an endless loop in the perl magic16:59
sledgesFlyser: oh?16:59
locusfsledges: nope :p16:59
sledgesdid you fix it locally?16:59
Flyserit happens at least when the recovery.fstab is non-existant16:59
Flysermaybe other cases too16:59
sledgesohk16:59
FlyserI don't know much perl, so I am not of much help here16:59
sledgeslbt: ^16:59
sledgesFlyser: so how do you solve it? Ctrl+C? ;)17:00
Flyseryes and then creating the recovery.fstab17:00
sledgesblank?17:00
Flyserno, I picked one from github17:00
Flyserfor my device17:00
sledgesfrom where?17:00
sledgeswhy is it not shipped together?17:00
Flyserhttps://raw.githubusercontent.com/SkrilaxCZ/android_device_moto_asanti_c/jellybean/recovery.fstab17:01
sledges(locusf: welcome to the world of android devices reading ^ ;))17:01
FlyserI don't know17:01
sledgeshmmmmm17:01
Flyserits not here https://github.com/cyanogenmod/android_device_motorola_xt897/tree/cm-10.117:01
sledgesand does it contain any crucial info?17:01
sledgesor just a placeholder17:01
locusfsledges: lol17:01
locusfsledges: I'm already like "I have no idea what I'm doing but I'm doing it anyways"17:02
sledges;) you'll get around once past initial androidy hurdles (if you have .img files, the worst of device bringup phase is in the past)17:02
sledgesFlyser: and where is fstab in https://github.com/cyanogenmod/android_device_motorola_xt897/tree/cm-10.1 ?17:03
sledgesor there's nothing at all? o.O17:03
Flyserthat's what I am wondering as well17:03
locusfbloody hell, now my HABUILD_SDK no longer points to my device17:04
sledgeslol, no wonder script halts17:04
sledgeslocusf: hadk DEVICE17:04
sledges?17:04
vrutkovssledges: yep, that one worked (had to also add my device u8815 along with "encore")17:04
locusfsledges: I just run hadk, it gives it correctly17:04
Flysermaybe the fstab gets pulled from the device? like the binary blobs?17:05
sledgeslocusf: HABUILD_SDK is a prompt, not an env var17:05
locusfsledges: I know17:05
locusfits the ubuntu-chroot one?17:05
sledgesvrutkovs: yep, fixup-mountpoints has dedicated paragraph in hadk ;)17:05
sledgeslocusf: yes, i do ubu-chroot -r /parentroot/$MER_ROOT/sdks/ubuntu17:06
locusfI'm just going to say screw the kernel config and head to target installation for my device17:06
sledges on my build env17:06
locusfyes17:06
Flysersledges: When I say "Error: Pattern dir './rpm/patterns/xt897' doesn't exist.", will you respond with a chapter number? :P17:06
sledgeslocusf: analyse output of cm kernel compile17:06
sledgeslook out for defconfig17:07
sledges(first lines before actual compilation)17:07
sledgesFlyser: yes17:07
sledges:D17:07
sledgesbut, tell me which one are you coming from? ;)17:07
Flyser7.1.1: mb2 -t $VENDOR-$DEVICE-armv7hl -s rpm/droid-hal-$DEVICE.spec build17:08
*** tbr has joined #sailfishos-porters17:08
sledgesaha! thanks, i overlooked that ;)17:09
sledgesa proper bug in doc you found :P17:09
Flyseryay17:09
sledgesFlyser: for now, refer to bottom of 8.317:09
sledgesi'll put up a fix, cheers again17:09
sledgesi see Flyser is winning the porting race atm ;)17:11
Flyser:P17:11
vakkovexport MER_ROOT  should be the path where my sdk is installed, right?17:11
locusfI'm too at 7.1.117:11
sledgesvakkov: yes, top level like /srv/mer (see https://wiki.merproject.org/wiki/Platform_SDK#tl.3Bdr )17:12
sledgeslocusf: are you building RPMs already ;)17:12
locusfsledges: https://gist.github.com/locusf/8dc4cfe2b33a3f88832817:12
locusfthat probably fails because of my git config has pager as most and its not available in the mer repos17:12
sledgesah17:13
sledgesyap17:13
Flyserokay, it created some patterns/xt897/*.yaml what do I need to modify there?17:13
sledgesnothing17:14
sledges:D17:14
Flyserbest answer!17:14
Flysersledges: I think there is a typo in 14.4.217:17
FlyserCOMPOSITOR_CONFIGS_DIR == COMPOSITOR_CONFIGS, right=17:17
Flyser?17:17
locusferror: File not found: /parentroot/srv/mer/sdks/ubuntu/android/droid/installroot/var/lib/environment/compositor/droid-hal-device.conf17:18
sledgeslocusf: copy from mako17:18
sledgesunder rpm/device-YOURVENDOR-YOURDEVICE-configs17:19
Flysercopy from mako? 14.4.2 says something different?17:19
sledgesFlyser: is the same file17:20
sledgesand yes, thanks for finding another bug ;)17:20
sledgesyou get another start17:20
sledgesand first porter gets a badge(r)17:20
sledges(i just made it all up)17:20
locusfhttps://gist.github.com/locusf/bc9bbf35a88adc2a76e417:20
Flysersledges: "gets a Jolla" would be much more encouraging :P17:21
sledgesbbbbut why porting then? :DD17:21
locusfproblems with processing patterns-rpm17:21
vrutkovsreference device, to compare!17:22
Flysersledges: for fun :)17:22
vrutkovs(10 Jolla's, to be sure)17:22
sledgesto enable world to experience sailfish17:22
sledgesis the answer ;)17:22
Flyser:D17:22
sledgesif you like fish meat that is17:22
sledges:D17:22
sledgeslocusf: 20:09 < sledges> Flyser: for now, refer to bottom of 8.317:23
locusfsledges: thx17:24
sledgeswe :)17:24
locusfwhee rpms17:25
Flyseractually I get a similar error even after reading chapter 8.3: https://gist.github.com/flyser/c9a449294855ac1b3d0217:26
Flyserlocusf: really? what did you do?17:26
locusfsledges: http://pastie.org/940307417:26
locusfFlyser:  rpm/helpers/add_new_device.sh.17:27
Flyserdid that as well17:27
locusfstill no rpms?17:27
Flysernope17:28
Flysersee the gist link17:28
locusfhmm17:29
sledgeslocusf: do 7.1.3 first, then come back to 712 and see if it fixed17:35
sledgesorder might be wrong17:35
Flysersledges: any idea regarding my error?17:36
sledgesFlyser: yes17:37
sledgesyou have duplicates like ./device-$VENDOR-$DEVICE-configs/lib/udev/platform-device and ./device-configs-all/lib/udev/platform-device17:37
sledgesremove the device-specific ones17:37
locusfsledges: is page 16 rest supposed to be "build" ?17:39
sledgesvgrade_: situ: why is this failing? they copied from hammerhead17:39
Flysersledges: I have one ./device-lge-hammerhead-configs/lib/udev/platform-device and ./device-configs-all/lib/udev/platform-device17:39
Flyserthe hammerhead file should not be relevant, right?17:39
sledgesFlyser: hammerhead should be ignored17:39
locusfso close ... fork: Resource temporarily unavailable17:39
sledgeslocusf: still no decent pdfreader ? ;)17:40
situMay I know the exact error ?17:40
locusfsledges: which one do you recommend?17:40
Flyserpdf.js works fine17:40
sledgessitu: https://gist.github.com/flyser/c9a449294855ac1b3d0217:40
sledgeslocusf: how did you get rid of "File listed twice"?17:41
locusfsledges: yes17:41
sledgessitu: apparently they didn't copy it from hammerhaed :o17:41
locusfsledges: http://pastie.org/private/7gevoo5soyoggo8ohwb5qq17:41
locusfwhy do I get these forking errors ...17:41
sledgeslocusf: okular, Firefox PDF reader17:42
Flyserhttps://gist.github.com/flyser/86acea511d84126b22f6 is correct, right?17:42
situsledges: We should check from where these files are getting picked up and are they really required17:42
sledgesFlyser: sure there's no 3rd platform-device under your device -configs/ dir?17:42
locusfsledges: I am using okular ...17:42
sledgeslocusf: free ?17:43
sledgeslocusf: then you can tripleclick, or click-drag-down to select-> the invisible17:43
Flyserhttps://gist.github.com/flyser/d727579f9b2fc39b0e8817:43
sledgeshummmmmm17:43
sledgeslocusf: how did you get rid of "File listed twice"?17:43
locusfsledges: I didn't even have it17:43
sledgesyes you did: https://gist.github.com/locusf/bc9bbf35a88adc2a76e417:43
sledges:)17:43
locusfhmm I don't know actually17:44
sledges(argument clinic? :)) http://www.youtube.com/watch?v=kQFKtI6gn9Y )17:44
sledgesFlyser: try mv`ing ./device-configs-all/lib/udev/platform-device away for a laugh first17:47
FlyserFile not found: /home/flyser/playground/sailfish-hadk/mer/android/droid/installroot/lib/udev/platform-device17:48
locusfrebootig17:48
energycsdxwhat should be at /system ?17:49
sledgesFlyser: and putting it under your device -configs/ now instead?17:49
sledgesenergycsdx: CM's system from its .zup17:49
sledgesuntouched17:49
sledgesit gets mounted by systemd lib/systemd/system/system.mount17:50
energycsdxand under /usr/libexec/droid-hybris/system? only patched bionic?17:51
sledgesyes, things locally built like logcat17:52
sledgesby hybris-hal17:52
sledgesthey are collected by dhd main rpm17:52
sledgesand get installed there through it17:52
Flysersledges: same error17:52
energycsdxand blobs like libEGL.so are under /system?17:53
sledgesenergycsdx: ./usr/libexec/droid-hybris/system/lib/libEGL.so17:54
energycsdxsorry libEGL.so is not vendor blob17:56
sledgesenergycsdx: best if you follow hadk..17:56
locusfsledges: Error <creator>: URLGrabber error: https://releases.jollamobile.com/releases/1.0.8.19/jolla-hw/adaptation-samsung-n7000/armv7hl/repodata/repomd.xml17:56
locusfoh I can already see the error lo17:57
energycsdxsledges: i reviewed it17:57
sledgeslocusf: ok ;)17:58
energycsdxit is not clear what is difference between /system and /usr/libexec/droid-hybris17:58
locusfthe sed fails17:58
vgrade_files listed twice is benign I think17:58
sledgeslocusf: you shouldn't have that repo in17:58
sledgesvgrade_: tru!17:59
sledgesFlyser: vgrade_: then we need to get rid of these: /init.qcom.class_core.sh17:59
sledgesi guess some perls pulled them in ;)17:59
sledgeslocusf: yes, sed fail, but why?17:59
vgrade_sledges: or package them17:59
locusfsledges: newline in copypaste18:01
sledgesa18:01
energycsdxsledges: forexample this http://code.metager.de/source/xref/android/4.4/frameworks/native/opengl/libs/EGL/Loader.cpp#264 ignores HYBRIS_LD_LIBRARY_PATH and require to have /system/lib/egl/* things but actual libs will be loaded from another location18:01
sledgesi wonder when those *.sh are executed, Flyser ?18:01
sledgesFlyser: patch droid-hal-device.inc , add after # Droid config droppings18:01
sledges /init*.sh18:02
locusfsledges: http://pastie.org/940314318:02
vgrade_energycsdx: /system is original CM libs, /usr/libexec/droid-hybris contains hybris stub libs to allow calls to bionic drivers from linux side18:02
sledgeslocusf: your .ks is not sedded enough18:02
locusfsledges: which sed to apply?18:04
energycsdxso /usr/libexec/droid-hybris may not contain vendor specific libs and can be same for same android version?18:04
locusfI haven't run ssuks18:04
sledgeslocusf: 82 first one18:05
sledgeswith proper copypaste ;)18:05
sledgeslocusf: i wanted to ask about outcome of order of sections mixed earlier18:05
locusfsledges: could you please paste it somewhere?18:05
locusfthat first sed18:06
vgrade_energycsdx: the libs in libexec need to be built with the same android tree the /system libs were18:06
locusfsledges: http://pastie.org/private/f2bkakptkysuovplubdwiq <- this is after the first sed unmodified18:08
locusfsledges: second sed I modified: sed -e "s|^$HA_REPO.*$|$HA_REPO --baseurl=file://$ANDROID_ROOT/droid-local-repo/$DEVICE|" $ANDROID_ROOT/installroot/usr/share/kickstarts/Jolla-@RELEASE@-$DEVICE-@ARCH@.ks > tmp/Jolla-@RELEASE@-$DEVICE-@ARCH@.ks18:08
sledgeslocusf: 2nd set should not be run18:08
sledgesas there's no obs repo for your device18:08
Flysergot to go. be back in an hour or so18:09
sledgesFlyser: o/18:09
locusfI mean second sed == my modified sed18:09
sledgeslocusf: not having hadk here18:09
sledgeslocusf: first sed is needed only18:09
locusfyes18:09
sledgesand no mods18:09
locusfas I have done18:09
sledges;p18:09
locusfno second sed as I can read :p18:09
sledgesso your repo jollamobile.com should be gon18:09
locusfat least hopefully18:09
locusfI did forget to run process_patterns18:11
locusfas mentioned in 8.318:12
locusfWarning: repo problem: nothing provides jolla-rnd-device needed by pattern:jolla-hw-adaptation-n7000-0.0.6-201407181724.noarch,18:14
sledgeslocusf: progress!18:15
sledgesremove that from patterns18:15
sledgesand follow all steps to regenerate them18:15
locusfI commented out the repos in .ks18:15
sledgesrepos?18:15
sledgesno just that pkg18:15
sledgesyou could try -jolla-rnd-device18:15
sledgesin %packages18:15
locusfhmm nope18:17
sledges21:15 < sledges> remove that from patterns18:17
sledges21:15 < sledges> and follow all steps to regenerate them18:17
locusfI edited patterns and then ran process_patterns18:18
sledgesone of last 8.x sections say the sequence you have to go through18:18
sledgesjust regen patts is not enough iirc18:18
locusfwhich one?18:20
sledgeswith loads of symlinks18:20
locusfwhich section?18:20
locusfin HADK manual18:20
sledges8x18:21
sledgesdont' have at hand18:21
sledgesdineer now ;P18:22
*** Nokius has joined #sailfishos-porters18:25
sledgeswelcome new porters :)18:26
locusfit was the mb2 build18:27
NokiusThanks for the Adaptation Development Kit Jolla18:30
sledgeswe :)18:30
locusfsledges: in 13.8 now18:34
sledgeswhat will you hack Nokius ? :)18:36
sledgeslocusf: the race is on ;)18:36
NokiusWill try the find5 (hope i have the skills ;-/ and the time )18:37
sledgesNokius: looks like there's stable 10.1.3 cm build for it, so will ease your efforts; we're here to help18:39
Nokiussledges: yeah its a HADK device18:40
sledgeswell, we have cm-11 work ongoing too (as well as aosp), not far from well tested18:41
locusfsledges: http://pastie.org/private/6re8uzdmakzjxapf2e8wg18:41
sledgesbroaden the horizons, namely nexus518:41
locusfa bunch of those rpmdb things too, had to cut to the few remaining ones18:42
sledgeslocusf: sb2 .... --sdk-install .... zypper in libtool18:43
sledgesto isolate the case18:43
locusflots of these error: rpmdbNextIterator: skipping h#      47 Header V3 DSA/SHA1 Signature, key ID f2633ee0: BAD18:44
locusfand then fails in (with --nodeps --force) Error: Subprocess failed. Error: RPM failed: error: /var/cache/zypp/packages/jolla/core/armv7hl/libtool-2.4.2-1.1.1.armv7hl.rpm: Header V3 DSA/SHA1 Signature, key ID f2633ee0: BAD18:44
locusferror: /var/cache/zypp/packages/jolla/core/armv7hl/libtool-2.4.2-1.1.1.armv7hl.rpm cannot be installed18:44
locusfprobably something screwed in the rpmdb18:44
sledgestry wayland-devel18:45
locusf(with --nodeps --force) Error: Subprocess failed. Error: RPM failed: error: /var/cache/zypp/packages/jolla/core/armv7hl/wayland-devel-1.1.0+git2-1.2.1.armv7hl.rpm: Header V3 DSA/SHA1 Signature, key ID f2633ee0: BAD18:45
locusferror: /var/cache/zypp/packages/jolla/core/armv7hl/wayland-devel-1.1.0+git2-1.2.1.armv7hl.rpm cannot be installed18:45
locusfhad to do sb2 -t $VENDOR-$DEVICE-armv7hl -m sdk-install -R rpm --rebuilddb18:46
locusfbut still failing18:47
locusfhttp://pastie.org/940328318:48
sledgeshmmmm18:48
*** vrutkovs has quit IRC18:49
sledgessuggest recreating target18:50
locusfmissing zypper ref fixed it18:51
locusfnope18:51
sledgeslocusf:18:52
sledgesrm -f /var/lib/rpm/__db*18:52
sledgesthen18:52
sledgesrpm --rebuilddb18:52
sledgeson target ofc18:52
Nokiussledges: is anyone else working on a find5 build?18:52
locusfsledges: nope18:52
locusfsledges: how to uninstall target?18:53
sledgeslocusf: examing what sb* commands you have18:56
Flysersledges: adding the /init*.sh worked \o/18:56
sledges\o/18:56
sledgesNokius: nope, which chipset is it?18:56
locusfsledges: which should I have?18:56
sledgesNokius: also strengthen your crew by creating xda find5 sailfishos thread18:57
locusfsledges: http://pastie.org/940330218:57
sledgeslocusf: many18:57
sledgesx)18:57
Nokiussledges: Snapdragon S4 APQ806418:59
Nokiussledges: soource wikipedia18:59
*** furikku has quit IRC19:00
sledgesNokius: same as n4 - best supported hadk device to date ;)19:04
situNokius: Isn't it Qualcomm APQ8064 Quad Core 1.5GHz ?19:04
situNokius: http://global.oppo.com/products/find5/specs/19:04
situare they both same ?19:05
sledgeslocusf: mv`ing existing target and recreating->sb2-init`ing new one should suffice19:05
Nokiussledges: ture the same cpu as the n419:07
*** alin has joined #sailfishos-porters19:09
sledgesalin: hi, the "soon" day has come ;)19:10
alinsledges: yes19:10
sledgesalways keeping promises :)19:10
alinsledges: I got it... just been pretty busy... now I was reading... and I already have a question19:10
sledgesfire away ;)19:10
alinsledges: the guide says you like cyanogem 10.x only... but if I go to mer-hybris I see repos for hammerhead19:11
alinsledges: there is some kind of contradiction...19:11
sledges11.0 is not ready as it has been undertested19:11
sledgesbut last days changed a lot19:11
sledgesand we already have nexus5 instructions following 11 branch:19:11
sledgeshttps://wiki.merproject.org/wiki/Building_Sailfish_OS_for_Nexus_519:12
sledgesby situ ^19:12
alinsledges: good... next week I am on holiday...19:13
alinsledges: so I can tinker19:13
sledgesbril, also less-headachy next hadk version with fixes will be out by then19:15
Flyserany idea? https://gist.github.com/flyser/4abb50c2c02daa943c6919:16
alinsledges: in the https://wiki.merproject.org/wiki/Adaptations/libhybris seems camera is listed as not working...19:16
FlyserI already did 7.1.3 before19:16
Flyserand used zypper ar -G instead of zypper ar19:16
alinsledges: I have abused my phone and but ubuntu on it... and there the camera was workign19:16
alincan't we borrow their fix?19:16
locusfsledges: back in business19:17
sledgesalin: but hadk is ;) so now enabling camera and other tinkering is in the hands of the whole community19:17
sledgeslocusf: me too, dinner was good :)19:17
sledgeslocusf: you can help Flyser with his last paste ;P as you've been there today19:18
locusfsledges: ok19:21
sledgesFlyser: b2 -t $VENDOR-$DEVICE-armv7hl -R -m sdk-install zypper lr -u19:21
sledges*sb219:21
sledgeslooks like something not done till the end..19:21
locusfsledges: the pdf produces newlines when there is a * and . character19:21
sledgeshuh?19:21
locusfat least for me it is19:23
Flyserhttps://gist.github.com/flyser/95187ef5ffa15492b0ff19:23
Flyserlocusf: same here19:23
sledgesFlyser: sb2 -t $VENDOR-$DEVICE-armv7hl -R -m sdk-install ssu lr19:23
sledgesguys what do you mean . * newlines?19:24
sledgespage?19:24
sledgeshlp me reproduce :)19:24
Flysertry to copy the rm -f $ANDROID_ROOT/droid-local-repo/$DEVICE/droid-hal-*rpm on page 1619:24
Flyseror the line below that one19:25
sledgesrm -f $ANDROID_ROOT/droid-local-repo/$DEVICE/droid-hal-*rpm19:25
sledges(using evince :")19:25
Flysersledges: seems like I am bumping the said wall19:25
sledgesFlyser: ssu lr pls ;)19:25
Flyserconnect: Mapped AF_UNIX address (/parentroot/home/flyser/playground/sailfish-hadk/mer/targets/motorola-xt897-armv7hl/run/dbus/system_bus_socket) is too long19:25
sledgesFlyser: hopefully those are benign19:26
sledgeswe'll see19:26
locusfsledges: http://pastie.org/private/h6rdbwee5pplnbr1fcvxg19:27
Flyserhttps://gist.github.com/flyser/89b2a3952cfac41873ec19:27
sledgesga!19:27
locusfyeah no local repo enabled19:27
sledgesFlyser: local repo should be there (and yes, enabled)19:28
sledgeswhy did you use -G ?19:28
Flyserotherwise it would spit out an error saying that something wasn't signed19:28
locusfsb2 -t $VENDOR-$DEVICE-armv7hl -R -m sdk-install zypper ar $ANDROID_ROOT/droid-local-repo/$DEVICE local-$DEVICE-hal19:28
sledgessignage is also not critical19:28
sledgesif it can't find it19:29
sledgessign into target19:29
sledgesand look for it first19:29
sledgesmounts / parentroot might be bonkers19:29
locusfalso zypper ref afterwards19:29
sledges(target shell: `sb2 -t $VENDOR-$DEVICE-armv7hl -R -m sdk-install`)19:29
sledgessee if $ANDROID_ROOT is reachable from within target19:30
locusfsledges: any comments on the pastie or should we help Flyser first :) ?19:32
sledgescoming ;)19:32
locusfah19:32
FlyserANDROID_ROOT is reachable19:32
sledgeslocusf: you should have this in your .ks: repo --name=jolla-@RELEASE@ --baseurl=https://releases.jolla.com/releases/@RELEASE@/jolla/@ARCH@/19:32
sledgesalso this is not good: Unknown command: -jolla-rnd-device19:32
sledgeslocusf: ^ (it should be in %packages )19:32
sledgesFlyser: then just do within shell: zypper ar $ANDROID_ROOT/droid-local-repo/$DEVICE local-$DEVICE-ha219:33
sledges*hal19:33
sledgesand subst DEVICE ofc19:33
locusfError <creator>: URLGrabber error: https://releases.jollamobile.com/releases/1.0.8.19/jolla/armv7hl/repodata/repomd.xml19:33
Flyserwhy is zypper showing it and ssu not? https://gist.github.com/flyser/f0f2af3d0fccf6cefe4c19:34
sledgeslocusf: only jolla.com domains should be in your ks19:34
*** vrutkovs has joined #sailfishos-porters19:34
locusfsledges: http://pastie.org/940337919:34
locusfthats my .ks19:35
sledgesFlyser: i don't like that dir19:35
sledgeszypper dr local-xt897-hal19:35
sledgesand re-add withotu -G19:35
vgrade_locusf: try  https://releases.jolla.com/releases/1.0.8.19/jolla/armv7hl/repodata/repomd.xml19:35
sledgesit should say 3 | local     | local     | Yes     | No      | plugin:ssu?repo=local19:35
sledgeslocusf: you're on a wrong domain :/19:36
locusfsledges: yes19:36
sledgesvgrade_: he needs all repos renamed19:36
sledgesi.e not good atm19:36
sledgessec19:36
Flysersledges: I already readded without the -G19:37
Flyserzypper dr does not exist. do you mean rr?19:37
sledgesyes19:37
Flyseryeah ... did that19:37
sledgeslocusf: did you generate .ks via hybris/droid-hal-configs/rpm/droid-hal-configs.spec ?19:38
locusfsledges: nope, with 8.219:38
sledgeslocusf: well $ANDROID_ROOT/installroot/usr/share/kickstarts/Jolla-@RELEASE@-$DEVICE-@ARCH@.ks  had to already be there19:39
sledgesfrom prev steps19:39
sledgessoz guys, all text in hadk italia should be hyperlink19:39
sledgesnew version will be a sip of fresh air ;)19:40
locusfnice19:40
FlyserI think I should fix the dbus error first19:41
sledgeslocusf: ks should come from 71219:41
sledgesFlyser: hold on, i see some naming changes that i might have not retesteed19:41
sledgeslet me git log/blame first ;P :))19:42
locusfsledges: I have run it19:42
sledgeslocusf: and resulting ks has jollamobile all over the place?19:42
sledgesin that case, you'll need to test me a hack ^_^19:42
locusfsledges: yes19:43
sledgesopen rpm/droid-hal-configs.spec19:43
sledgescomment out #  %{!?ssu_override: %define ssu_override domain=sales}19:43
sledgesand under it add19:43
sledges%define ssu_override domain=sales19:43
sledgesthen re-run19:43
sledgesand check guts of .ks again19:44
locusfno such file rpm/droid-hal-configs.spec19:45
energycsdxdo i need /usr/libexec/droid if i can fully compile AOSP and place to system?19:45
energycsdxto /system19:45
sledgeslocusf: hybris/droid-hal-configs/rpm/droid-hal-configs.spec19:46
sledgesenergycsdx: yes 'cause they're stubs (thanks vgrade_ )19:46
sledgeslibhybris will look under libexec/droid19:46
locusfsledges: worked19:47
sledgesnot good19:47
sledges:D19:47
sledgesssu_override should not be already defined in a local env19:48
locusfhttp://pastie.org/940340119:48
locusfnow that error19:48
sledgeslocusf: 13.8.3 ;)19:49
locusfye19:49
sledges(from 14.4.4 don't say we didn't tell ;P)19:49
sledgeslocusf: the end is nigh nw ;)19:49
sledgesFlyser: you're genious, -G -is- needed :D19:50
sledgesthe patch got reverted and someone forgot -G ;)19:50
sledgesoups19:50
sledgeshmmm19:51
sledgeswait that's for mer-tools-testing19:51
sledgesinvestigation still on ;P19:51
Flyser^^19:51
sledgesah ok19:51
sledgesssu ar is needed19:51
sledgesnot zypper19:51
* sledges takes out a whip ;)19:51
sledgesbut we won't mention names :))19:51
energycsdxby doing that sailfish os is not depend on android version flashed on phone, because now i do it with AOSP and there is unmodified stock on phone19:52
sledgesvgrade_: ^ ?19:52
locusfsledges: nice19:53
vgrade_energycsdx: do you have full source for all libs in /system19:53
vgrade_stock system19:53
locusfWarning: repo problem: nothing provides qt5-feedback-haptics-droid-vibrator needed by pattern:jolla-hw-adaptation-hybris-droid-vibrator-0.0.6-201407181827.noarch,19:54
locusfngfd-plugin-droid-vibrator is installed into repos19:54
*** lbt has quit IRC19:55
*** ParkerR has quit IRC19:55
vrutkovsfolks, got moar problems: http://pastebin.com/6gx6ZuNw19:55
energycsdxvgrade_: no, and i know that it is different from AOSP, but i can build AOSP, it is officialy available19:55
*** ParkerR has joined #sailfishos-porters19:55
*** lbt has joined #sailfishos-porters19:55
*** lbt has quit IRC19:55
*** lbt has joined #sailfishos-porters19:55
energycsdxvrutkovs: had simular problem, patched out this19:56
sledgesFlyser: rr that repo and add it like it says in 13.8.119:57
sledgesi will sort out the zypper ar mess19:57
vrutkovsenergycsdx: removing those dvb entries?19:58
vgrade_energycsdx: so you plan to build an ASOP versiosn of /system libs and install.  Then use the same ASOP version to build the hybris libs?19:58
locusfbuilding image19:59
sledgeslocusf: qt5-feedback-haptics-droid-vibrator is also needed19:59
sledgesto be built19:59
locusfsledges: did so20:00
locusfbut its not documented20:00
locusfafaik20:00
sledgeslocusf: yes, good job; added to todo20:00
energycsdxvgrade: i already build ASOP version of /system and usualy just copy it to jolla root /system, but now i started to pack it to dhd.rpm and only part of this system goes to /usr/libexec/droid/ and that puzled me20:01
Nokiussledges: sorry for bothering but on page 9 (HADK) does mean Host the Bulid engine or the chroot20:03
sledgesNokius: your build host20:04
sledgesphysical20:04
sledgesLinux ;)20:04
locusfeww I'm sweaty and sticky -> shower20:04
Nokiussledges: okay so i was right :)20:04
sledgeslocusf: is this what hadk+weather do to you ;)20:05
sledgesmic's busy dl'ing rpms? ;)20:05
vgrade_energycsdx: the libs and bins in libexec are the ones used by the linux part of sailfish (ie built) against libc and they call into the ones in /system which are built against bionic using libhybris.  So as long as the version of CM / ASOP you use to build libexec libs is the same as the ones in /system (even those you don't have source code for) as the interfaces are matched20:08
vgrade_energycsdx: maybe its just that dhd only packs the libs/bins it needs to provide the functionality for sailfishos20:12
Flysersledges: works :)20:12
sledgesFlyser: ;)20:13
Flyserwait ... it doesn't20:16
alinsledges: few comments on guide... there are some sentences like install repo command... or as seen in mer wiki... can you add a footnote with the links20:16
alinsledges: some stupid pdf viewers do not follow link20:16
alinalso maybe in the hadk commands for mersk and ubuntu chroot add ~/bin in path20:17
sledgesalin: which page?20:17
alinas may be needed if one follows without thinking the repo install guide20:17
Flysersledges: mb2 -t $VENDOR-$DEVICE-armv7hl -s hybris/droid-hal-configs/rpm/droid-hal-configs.spec build is still needed, right? (7.1.2)20:17
alinsledges: page 11 for repo20:17
alinsledges: pake 9 for mer wiki20:18
alinsledges: the PATH on page 920:18
sledgesalin: sorry you need to source a better pdf reader. we also have linewrap problems on page 16 and 19 and until next version of hadk is out with them fixed, we ask people to use pdf readers where still can copy-paste oob20:19
sledgesbut would like to keep the document clean wrt to hyperlinks20:19
sledgesand set some minimal pdf reader requirements :)20:19
alinsledges: I use okular20:20
sledgesand okular does not follow hyperlink, not even with ctrl? nor shows tooltip when hovered?20:20
sledgeshmmm20:20
sledgesFlyser: yes20:20
alinsledges: only if I am in browse mode20:20
Flysersledges: it still fails :(20:21
sledgesFlyser: and i guess it works only after 713 (or in your case 13.8.1; i'll change order, thanks)20:21
Nokiussledges: just saw that is written before |-)20:21
sledgesFlyser: same?20:21
sledgesalin: you see sometimes (not always) it's about when there is will there is way ;)20:21
sledgesNokius: o.O20:22
alinsledges: is just annoting to change modes... that is all...20:22
Nokiussledges: that is to do on the physical engine20:22
Flysersledges: https://gist.github.com/flyser/64259263918617c8039620:23
sledgesalin: there will be many hyperlinks in the document (also soon i will add internal ref hyperlinks to jump to sections), i can't footnote each one and simply reduce the cleanliness of the doc20:23
FlyserI did all of 13.8.120:23
sledgesFlyser: did you do createrepo?20:24
locusfpushed to device and now installing :)20:24
sledgesand did you do zupper ref?20:24
sledgeslocusf: wow!20:24
locusfwithout the stock image20:24
locusfbtw20:24
sledgeserm?20:24
sledges;P20:24
sledgesbut wit hunderlying cm?20:24
locusfthe stock is 4.1.220:24
locusfyes20:24
sledgesso, what was before cm?20:24
sledgesdid you try if cm works)?20:24
sledgeswhich cm? ;P20:24
locusfthat 4.1.220:24
locusfyes I tried that it worked20:24
Flysersledges: yes20:24
locusf10.1.320:24
locusfCM20:24
sledgeslocusf: ok20:25
sledgesFlyser: zypper ref inside target again?20:25
alinalso a way to checkout only some nice repos rather than all will be useful20:25
locusfinstallation script apparently takes a while20:25
sledgesalso, check `zypper se droid-hal` in target20:25
sledgeslocusf: yes, big compression rate (bz2)20:25
sledgesbut congrats you passed first phase on .zip finding boot and data partitions ;)20:25
Flyserhttps://gist.github.com/flyser/fa2a0b0c07f0453e2eb320:26
sledgesyou can see their values being written now now in recovery20:26
sledgesFlyser: and se ?20:26
sledgesalin: which device?20:26
Flyserhttps://gist.github.com/flyser/7ea2db75a77e028d38c820:26
alinsledges: nexus520:27
alinsledges: also a -depth 0 may help to reduce the size20:27
alinin the git clone20:27
sledgesalin: they already been cleaned up (quite) a bit https://github.com/mer-hybris/android/commit/54060f5faa8b9e0414a80ede476b1cee19dbe4ec20:27
sledgesalin: more plans to clean-up are in situ and lbt heads ;)20:28
sledgesmako has been trimmed the most20:28
sledgesto the bone hehe20:28
sledges(nexus4)20:29
alinsledges: yap I know make20:29
alinsledges: I missed it by a day20:29
sledgesjust for those who read us ;)20:29
alinsledges: i was trying to make my mind... and by the time i decided bloody google released nexus 520:29
sledgesFlyser: inside target, try `zypper in droid-hal`20:29
sledges;)20:30
Flyserhttps://gist.github.com/flyser/d08efc1c6b4323e7e39d20:30
locusfgalaxy note stuck on device screen20:31
sledgeslocusf: usb cable time20:31
alinsledges: 7GiB cloned... let us hope I will finish by time I have have to go20:31
sledgesand `dmesg -w` on host20:31
sledgesalin: a porter's life ;)20:32
sledgesFlyser: hang on20:32
alinsledges: exactly a real porter... I need to go to airport... and collect my gf20:32
locusfsledges: ?20:32
sledgesFlyser: sb2 -t $VENDOR-$DEVICE-armv7hl -R -msdk-install \20:33
sledgeszypper install droid-hal-$DEVICE-devel20:33
sledgeslocusf: plug usb cable to your host and note20:33
sledgesand watch dmesg on host as you do20:33
sledgeslocusf: you're lucky it doesn't bootloop ;)20:33
Flyser'droid-hal-xt897-devel' is already installed. O_o20:34
sledgesFlyser: moment20:34
sledgesFlyser: is all because of those .sh files20:35
sledgeslol20:35
locusfsledges: dmesg is silent, nothing happens on the screen either20:35
sledgeslocusf: fastboot boot hybris-recovery.img20:36
vgrade_locusf: what deveice are you on?20:36
locusfvgrade_: Samsung Galaxy Note20:36
locusfsledges: is there such a command in Mer SDK?20:37
sledgeslocusf: yes20:37
*** teotwaki has joined #sailfishos-porters20:37
locusfsledges: where is hybris-recovery?20:37
sledgeslocusf: $ANDROID_ROOT/out/.....20:37
situsledges: We have removed all the apps from repo, still it's taking so long to clone.20:38
sledgesFlyser: you need to gut those .sh file out for a moment (or do you have a clue why they start asking for /system/sh vgrade_ ? )20:38
sledgessitu: you mean packages? well there's more20:38
sledgessee lbt's filtered branch20:38
locusfI don't think this is fastboot capable20:38
sledgeslocusf: sudo fastboot devices20:38
sledgeslocusf: how did you install cwm?20:39
sledgessammy yes, it might be odin/heimdall thing20:39
sledgeslocusf:20:39
sledgescan you access rootfs in cwm?20:39
locusfsledges: http://wiki.cyanogenmod.org/w/Install_CM_for_n700020:39
sledgesdo `touch /data/.stowaways/sailfishos/init_enter_debug`20:40
locusfsledges: how can I try?20:40
sledgesboot cwm20:40
sledgeschoose install zip from sd card20:40
sledges(it will then mount)20:40
sledgesand then do adb shell20:40
sledgesinto it20:40
*** jahof has joined #sailfishos-porters20:42
locusfcwm doesn't work anymore :/20:42
sledgesFlyser: or do in target `mkdir -p /system/bin/; cp /bin/sh /system/bin/` :DD20:42
*** nodevel has quit IRC20:42
sledgesit must have gotten overwritten by hybris-boot if you got the boot/recovery partition numbers wrong20:43
sledgesjust reinstate cwm then20:43
locusfo20:43
locusfok20:43
sledgesjahof welcome :)20:43
sledgesteotwaki: too ^ :)20:44
jahofhello :)20:44
Flysersledges: and then?20:44
sledgesgot some crazy devices ready fellows? :)20:44
sledgesFlyser: zypper in droid-hal20:44
locusfback on cwm20:45
Flysernothing provides /system/bin/sh needed by droid-hal-xt897-0.0.6-201407181911.armv7hl20:45
sledgesbut it's there now? /system/bin/sh20:45
Flyseryes20:46
sledgesmeh20:46
vgrade_but zypper / rpm database does not have a record20:46
sledgesyeahp... got you vgrade_20:46
sledgeswas a cheeky attempt anyway :)20:46
* vgrade_ looks at backlog for those .sh files20:47
teotwakisledges: ta20:47
locusf~ $ touch /data/.stowaways/sailfishos/init_enter_debug20:47
locusftouch: /data/.stowaways/sailfishos/init_enter_debug: Permission denied20:47
teotwakilooking through the pdf20:47
teotwakisounds like my phone isn't officially supported by CM just yet (Moto G 4G)20:48
sledgeslocusf: su20:49
vgrade_Flyser: what device you on?20:49
Flysermotorola photon q (xt897)20:49
sledgesvgrade_: https://gist.github.com/flyser/c9a449294855ac1b3d0220:49
locusf~ $ su20:49
locusf/system/bin/sh: su: not found20:49
vgrade_sledges: thanks20:49
teotwakigood to see a mobile-foss channel so active again, haven't seen that in a few years since the good days of maemo. I'll just idle in here for a while, take care.20:50
sledgeslocusf: you'll need to master odin/heimdall20:50
sledgesteotwaki: cheers, feel free to lurk about :)20:51
locusfheimdall only here :/20:51
vgrade_teotwaki: I thought 4g moto was supported20:51
sledgeslocusf: way to go Linux ;)20:51
vgrade_was looking at it before I cam about my one+1's20:51
vgrade_came20:51
locusfsledges: so access /data via heimdall?20:52
teotwakivgrade_: only the regular moto g, 4g version (XT1039) not just yet (only been out for a couple of weeks, so to be expected really)20:52
vgrade_Flyser: do you have a link to your android device tree github repo20:52
sledgeslocusf: that or try to boot an .img (flash hybris-recovery.img instead of hybris-boot.img to boot partition, by means you can find with heimdall)20:53
Flyserhttps://github.com/cyanogenmod/android_device_motorola_xt897/tree/cm-10.120:53
locusfsledges: booted now20:53
locusfblack screen20:53
vgrade_locusf: \o20:54
sledgeslocusf: dmesg?20:54
locusfvgrade_: \o?20:54
sledges( vgrade_ is half \o/ :))20:54
locusfhttp://pastie.org/940350520:54
vgrade_locusf: means something is running to clear screen from recovery20:54
locusfvgrade_: ok20:55
sledgeslsusb -v | grep iSerial20:56
sledgesiSerial20:56
sledgeslocusf: ^20:56
locusf  iSerial                 020:56
locusf  iSerial                 4 0019da475a6c3f20:56
locusf  iSerial                 020:56
locusf  iSerial                 1 0000:00:1d.020:56
locusf  iSerial                 1 0000:0b:00.020:56
locusf  iSerial                 1 0000:0b:00.020:56
locusf  iSerial                 1 0000:0a:00.020:56
Nokiussledges: are you in charge of the HADK because the wiki set an entry in the .mersdk.profile that user can see in CLI that a MER SDK pointer but in the KIT removes it / does not help the user to have an easier overview where he is :-/20:57
sledgesNokius: thanks will add to todo20:57
sledgesNokius: the whole team is in charge of modifying hadk20:58
sledgeslocusf: i expect iSerial20:58
sledges3 Mer Debug telnet on port 23 on rnd ...20:58
sledgesmeans hybris*.img not booted20:58
sledgesjust like with sony xperia z, one needs to flash for booting to happen20:59
sledgesvolatile doesn't work20:59
Nokiussledges: not sure if its me but after editing the mersdk.profile as shown in hadk is gone ….20:59
Nokiusrealy love the hadk Thanks20:59
sledgesNokius: just add PS1 setting to the bottom /;)20:59
sledgesalso make sure you >> into profile21:00
Nokiussledges: was also my guess ^^ repo sync is running21:01
sledgescool21:01
vrutkovsargh, got stuck at 7.1.2: http://pastebin.com/6MwQgjn2 - no droid-hal-config.spec there21:01
vrutkovssledges: ^ ?21:01
sledgesvrutkovs: `cd hybris/droid-hal-configs; git status`21:03
vgrade_Flyser: sledges those scripts come from https://github.com/CyanogenMod/android_device_motorola_msm8960-common/tree/cm-10.2/rootdir/etc21:03
Flyservgrade_: are you still having a look?21:03
vgrade_https://github.com/CyanogenMod/android_device_motorola_msm8960-common/blob/cm-10.2/rootdir/etc/init.qcom.sh#L121:03
Flyseroops21:03
sledges^_^21:03
vrutkovssledges: weird, it says its not a git repo...21:03
sledgesvrutkovs: repo sync went wrong21:04
sledgesis it in manifest ?21:04
sledges(should be, if not, means you checked out before today ;))21:04
Flyservgrade_: so what should I do? just delete them?21:04
sledges(repo synced)21:04
vgrade_Flyser: sec, thinking21:04
vgrade_looking where they are run from21:07
locusfsledges: I'll continue tomorrow, now some sleep :)21:07
vgrade_locusf: nn21:07
vgrade_Flyser: so they are run from https://github.com/CyanogenMod/android_device_motorola_msm8960-common/blob/cm-10.2/rootdir/etc/init.qcom.rc21:08
sledgeslocusf: o/21:08
sledgesvgrade_: sed /system/bin -> /bin ?21:08
sledgesin the place which scavenges them (droid-hal-device.inc )21:09
vgrade_could do, just wondering if there are any instructions in HADK about adapting your .rc files21:10
sledgeshmm, for now they have just been pulled in no questions asked21:10
vgrade_yes21:11
sledgeslbt's script crawls .rc files and comments out # mount lines by externalising those into /systemd/21:13
vgrade_sledges: ok good21:13
vgrade_if we look at where the scripts are called https://github.com/CyanogenMod/android_device_motorola_msm8960-common/blob/cm-10.2/rootdir/etc/init.qcom.rc21:13
vgrade_then driod-init specifically calls /system/bin/sh <script>21:14
alinsledges: ok.. went in trouble but I need to go too21:14
alinsledges: see you later21:14
sledgesalin: laters!ciao21:14
vgrade_alin: nn21:14
*** alin has quit IRC21:14
vgrade_sledges: so we may just be able to remove the #!/system/bin/sh altogether21:15
vgrade_so rpmbuild won't catch the dependenxy21:16
vrutkovssledges: weird, had to check it out manually - can't find it in any manifest21:17
* vgrade_ thinks that every device he's looked at so far is differrnt21:17
sledgesvrutkovs: did you repo sync recently?21:17
sledgesah, it's in manifest, but probably other branch21:17
sledgesvgrade_: so what would we need to replace with what?21:18
sledgesvgrade_: you think rpmdb catches it from `exec /system/bin/sh ...` line ?21:20
sledgesi think it's L1 (#!/system/bin/sh)21:20
sledgesof the script21:20
vgrade_sledges: yea catches the dep from #!21:20
sledgesyep21:20
sledgesas it was fine as long as we didn't have .sh files in21:21
sledgesah21:22
sledges00:15 < vgrade_> sledges: so we may just be able to remove the #!/system/bin/sh altogether21:22
sledgesmissed that21:22
sledges:D21:22
vgrade_busy channel tonight,21:22
sledgeswell, i was /lastlogging quite a bit ^_^21:22
sledgeswhy not system/bin->bin ?21:23
sledgesat any rate21:23
sledgesFlyser is our guinea pig ^_^21:23
Flyser:-)21:23
sledgesFlyser: pick one method, modify those files and re-run mb2 of dhd21:23
FlyserL1 is not evaluated at all. so it doesn't matter if it's removed or #!/bin/sh21:24
vgrade_that would work for installing in target, would it work when called from droid-init21:24
sledgesFlyser: rpmbuild evaluates it....21:24
sledgesis where can of worms starts21:24
sledgesvgrade_: droid-init would exec /system/bin/sh regardless21:24
sledgesand there is one21:24
FlyserI know, I meant its not evaluated when the scripts are executed21:24
sledgesFlyser: yes21:25
vgrade_ok happy for /bin.sh21:25
vgrade_ok happy for /bin/sh21:25
sledgesFlyser: the stage is yours :))21:26
FlyserI am on it21:26
Flyserits way too warm here and its getting late, give me a minute ;)21:27
sledgeswell i'll need to make a thermal shutdown soon21:29
sledgesas russians say, morning is wiser than the evening;)21:29
Flyser^^21:30
Flyserso which steps do I need to repeat?21:31
sledgesFlyser: mb2 ... rpm/droid-hal-$DEVICE.spec21:32
Flyser*sigh* Problem: nothing provides /system/bin/sh needed by droid-hal-xt897-0.0.6-201407182132.armv7hl21:35
sledgesthis is old timestamp droid-hal-xt897-0.0.6-201407182132.armv7hl21:36
Flyseroh ... good catch21:36
sledgesrm *rpms21:36
sledgesthem mv RPMS/* droid-local....21:36
sledgesthen zypper ref21:36
Flyserwait ... it is the current timestamp21:38
Flyserthats 21:32 UTC21:38
sledgesoh21:38
sledges:P21:38
sledgesare you from east coast? ;)21:38
Flysergermany21:39
sledges201407182132 ?21:39
sledgesnoway ;P21:39
Flyserthats UTC21:39
sledgesit's 23:39 in germany now21:39
sledgesand 21:39 utc21:39
Flyserthe chroot has different timezone settings21:39
sledgesoh lol21:40
sledgesx))))21:40
Flyser^^21:40
sledgeswhat does date say?21:40
vgrade_197021:40
Flyser21:39:4621:40
sledgesso21:41
sledges-20140718213221:41
sledgeshmph21:41
sledgesat any rate, i believe you had rm`d old rpms ;)21:41
sledges(date on target ;)21:41
sledgesFlyser: yeap, i mean full date on target, with TZ21:41
Flyserdo I need to rerun make hybris-hal?21:42
sledgesFlyser: did you remove/rename #!/system/bin in all those .sh files?21:42
sledgesah21:42
sledgesyes21:42
sledgesand no21:42
sledgesyea but no but yea but no :D21:42
sledges(is some local thing here, little britain i think)21:42
vgrade_:)21:42
FlyserFri Jul 18 21:42:26 UTC 201421:42
sledgeso.O21:42
sledgesanyway21:42
sledgesfind those file under out/21:42
sledges*files21:42
sledgesand re-run mb221:43
Flyserbtw ... have look at L1 of droid/device/lge/mako/init.mako.bt.sh:21:43
Nokiussledges: :-/  "make -j4 hybris-hal" returns errors21:44
Nokiushttp://goo.gl/CePXAs21:44
sledgesNokius: symlink /usr/bin/gcc-4.7 to /usr/bin/gcc21:46
vgrade_Flyser: so that .sh can't be being picked21:46
sledgesfix is on its way21:46
* vgrade_ looks for line in https://github.com/mer-hybris/droid-hal-device/blob/hybris-10.1/droid-hal-device.inc which copies those sh files21:49
sledgesvgrade_: https://github.com/mer-hybris/droid-hal-device/blob/hybris-10.1/droid-hal-device.inc#L52721:50
sledgesdroppings ^_^21:50
sledgesthey appear there after lbt's mount smart script does its job iirc21:51
Nokiussledges:  out/ looks still empty21:51
Nokiushttp://goo.gl/8dMLTx21:51
vgrade_but in mako case the sh script does not get copied21:52
sledgesNokius: what's output of make?21:53
sledgesvgrade_: is there a .sh script? o.O21:53
vgrade_Flyser: I assume you created a device-motorola-xt897-configs dir and copied the scripts from etc into it21:53
sledgesi put money on, he didn't21:53
sledges:D21:53
vgrade_00:43 < Flyser> btw ... have look at L1 of droid/device/lge/mako/init.mako.bt.sh:21:53
Nokiussledges:   http://goo.gl/ODkZBq21:54
sledgesah21:54
sledgesNokius: so gcc fix didn't work?21:54
sledgesNokius: you need to add your find5 mountpoints to fixup-mountpoints ile21:55
sledgesNokius: see 14.3.221:55
Flyservgrade_: no21:56
sledges(ka-ching!)21:56
sledgesok guys, thermal shutown imminent, see you in the morning light!21:57
Flyser^^21:57
Flysersledges: good night21:57
sledgesnn21:57
vgrade_nn sledges21:57
situGood night21:58
* vgrade_ wonders how the qcom scripts get into the droid hal rpm21:58
Nokiussledges:  thanks gn21:59
situvgrade_: same here21:59
vgrade_Flyser: can you pastie your rpm build log21:59
vgrade_situ: seen announcements about jolla for IND today22:00
situvgrade_: Yep, I hope they do well here.22:00
Flyservgrade_: this? https://gist.github.com/flyser/1b43bf2fae7710a4377c22:00
vgrade_Flyser: yup , tanks22:01
vgrade_situ: I'm sure you will be marketing22:01
situvgrade_: heh.. I am bad at it but will try ;)22:01
vgrade_https://gist.github.com/flyser/1b43bf2fae7710a4377c#file-gistfile1-txt-L263 still has /system/bin/sh22:12
vgrade_as required22:12
Flyseryes I am currently grepping to find out why22:13
vgrade_grep in out/target/product...22:14
vgrade_Flyser: did you make hybros-hal in the end?22:15
vgrade_before reruning rpm build22:16
Flysernot since I replaced /system/bin/sh22:16
Nokiusone question about 14.3.2 Configure mountpoint information in HADK22:17
Nokiusmake -j4 hybris-hal returns22:17
Nokiushybris/hybris-boot/Android.mk:48: ********************* /boot should live on /dev/block/platform/msm_sdcc.1/by-name/boot22:17
Nokiushybris/hybris-boot/Android.mk:49: ********************* /data should live on /dev/block/platform/msm_sdcc.1/by-name/userdata22:17
Nokiusnow is the way to go to patch the hybris/hybris-boot/fixup-mountpoints22:17
Nokiusthe patch has to look like this right?22:17
Nokius    "find5")22:17
Nokius        sed -i \22:17
Nokius            -e 's block/platform/msm_sdcc.1/by-name/boot mmcblkXXXX ' \22:17
Nokius            -e 's block/platform/msm_sdcc.1/by-name/userdata mmcblkXXX ' \22:17
Nokius            "$@"22:17
Nokius        ;;22:17
Nokiusbut how i can find the exect location?22:17
FlyserI used ls -l /dev/block/platform/*/by-name/ on the device22:17
FlyserNokius: ↑22:17
vgrade_Flyser: that might be the issue, as those scripts might get copied to another location and then picked up from there by repm build22:18
NokiusFlyser: okay thanks22:18
Nokiusis some one taking track about the patch for hybris/hybris-boot/fixup-mountpoints22:24
vgrade_Nokius: how do you mean "taking track"22:27
Nokiusis some one interested in the patch like takeing it into the hybris master ?22:29
vgrade_Nokius: oh yes, sure, pull requestes very welcom22:31
vgrade_situ and I have been working on N5 any now have most of that upstream now22:32
*** jahof has quit IRC22:33
Nokiuscat find hybris/hybris-boot/fixup-mountpoints here  https://github.com/libhybris/libhybris/tree/master/hybris22:34
Nokiusbut adding it here https://github.com/libhybris/libhybris/pulls is correct?22:34
Nokiussorry not familiar with git :-(22:34
vgrade_Nokius: sec22:35
vgrade_from https://github.com/mer-hybris/hybris-boot click fork to get your own copy22:36
Nokiusvgrade_:  okay will try it … newby22:36
vgrade_then you can work on your version then once you've tested etc you can do a pull request22:37
vgrade_Nokius: no probs, I'm not the best on git, situ helps me out lots22:37
Nokius^^thanks22:37
vgrade_just shout if you get stuck22:38
vgrade_night guys22:40
Flyser*finally*22:42
Flyserbuilding hybris/droid-hal-configs/rpm/droid-hal-configs.spec worked22:42
Flysergn822:42
vgrade_Flyser: nn22:42
vgrade_Flyser: just needed a rebuild of droid-hal ?22:43
vgrade_hybris-hal22:43
Nokiusvgrade_: gn822:44
Flyservgrade_: grepped through $ANDROID_ROOT and replaced the path in all copies of the shell scripts22:54
Nokiussee us tomorrow gn8 Thanks22:57
*** teotwaki has quit IRC22:58
*** vrutkovs has quit IRC23:17
*** Bulder has joined #sailfishos-porters23:21
*** Bulder has quit IRC23:22
*** Flyser has quit IRC23:29
*** Flyser has joined #sailfishos-porters23:35
*** Flyser has joined #sailfishos-porters23:35
« Thursday, 2014-07-17 Index Saturday, 2014-07-19 »

Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!