Wednesday, 2015-02-11

*** M4rtinK <M4rtinK!> has quit IRC (Ping timeout: 250 seconds)00:01
*** ruthenianboy <ruthenianboy!> has quit IRC (Quit: leaving)00:09
*** olafh <olafh!> has quit IRC (Ping timeout: 246 seconds)00:35
*** vakkov <vakkov!> has quit IRC (Ping timeout: 255 seconds)01:13
*** Tassadar <Tassadar!> has quit IRC (Ping timeout: 264 seconds)01:21
*** Sfiet_Konstantin <Sfiet_Konstantin!> has quit IRC (Quit: Konversation terminated!)01:27
*** johnnydavey <johnnydavey!~androirc@> has joined #sailfishos-porters01:35
johnnydaveygood am to everyone01:35
*** vakkov <vakkov!> has joined #sailfishos-porters01:36
johnnydaveyi would like to try sailfishos01:36
johnnydaveyhow can i port the os in myphone.01:37
johnnydaveymyphone is a rebranded cubot s20801:37
*** johnnydavey <johnnydavey!~androirc@> has quit IRC (Client Quit)01:39
*** iTune <iTune!> has joined #sailfishos-porters01:56
*** uhhimhere <uhhimhere!~uhhimhere@> has joined #sailfishos-porters02:42
*** RaYmAn <RaYmAn!> has quit IRC (Ping timeout: 245 seconds)04:41
*** faenil <faenil!> has quit IRC (Ping timeout: 245 seconds)04:41
*** faenil <faenil!> has joined #sailfishos-porters04:41
*** RaYmAn <RaYmAn!> has joined #sailfishos-porters04:43
*** RaYmAn is now known as Guest6550804:43
*** harha <harha!> has joined #sailfishos-porters05:04
uhhimhereso anybody know what dalvik turbo is and how you use it?05:12
*** trt` <trt`!> has joined #sailfishos-porters05:38
*** olafh <olafh!> has joined #sailfishos-porters05:47
*** Elephant454 <Elephant454!> has joined #sailfishos-porters05:49
Elephant454Can someone clear something up for me real quick? What is the difference between CyanogenMod 10.1 and 10.2?05:50
*** iTune <iTune!> has quit IRC (Ping timeout: 245 seconds)05:54
*** harha <harha!> has quit IRC (Quit: Leaving.)06:08
*** iTune <iTune!> has joined #sailfishos-porters06:09
Elephant454In the porting guide, it says that an Android device supported by CyanogenMod 10.1.x is required. What is the difference between 10.1 and 10.2? I am curious to see if the LG G2 could be ported.06:20
tbryou'd need to ask the CM people what the differences are.06:21
tbrthe thing is that the patches are written for 10.1, so they might not apply cleanly → more problems, more work06:21
* tbr has recently also seen 11.0 and L based stuff floating around here, not sure how usable06:22
Elephant454Ok... I see... That's what I was afraid of. I am facinated by this stuff, but I'm not in a position to be able to actively develop for my phone. I can't risk bricking it. Any idea how I could go about learning more and gaining experience without the threat of a bricked phone over my head?06:23
tbrdepends on the definition of "brick"06:24
Elephant454I'm sorry if it seems like a ridiculous question. I'm not some kid asking for party tricks. I have a passion for Linux and would love to be a part of the porting process...06:24
tbrif you need your phone to work at all times, then you can't develop on it anyway06:24
Elephant454Of course. I'm willing to go without it for a while. That's a given.06:25
tbrthere is no risk difference between 10.1 or 10.2, the difference is that it might be harder to make it work06:25
Elephant454I can't have no phone for the year and a half between now and when my contract expires. Not having it ever turn on after all possible recovery efforts is my fear. Doing something to the boot partition to render the phone useless...06:26
tbrby that you mean a "real" brick, not what today's kids claim a brick06:28
Elephant454The process as a whole isn't set in stone, though. Right? There are clear risks in using an OS that isn't a finished product. Sorry if these questions aren't very well suited for here. The kind of responses I get on Reddit and XDA are "Well as long as you follow the guides and do what your supposed to, nothing will go wrong."06:28
tbrthe classic meaning is: not recoverable from with any reasonable tools06:28
tbrtoday "phone doesn't boot because battery is empty - It's bricked, LOL!"06:29
Elephant454Drives me crazy...06:29
Elephant454"My phone won't turn on, gaiz, halp"06:29
Elephant454"Oh, it just wasn't charged. Sorry. Thx."06:29
tbranywoo, if your phone can boot CM and if there is a sensible recovery boot option for it, then you should be fine and risk should be very low06:30
tbrand by sensible I mean something along the lines of that clockworkmod stuff06:31
Elephant454I should be good to go through the rest of the guide without fear?06:31
Elephant454Already have a TWRP backup. Willing to do fastboot if necessary.06:31
Elephant454Or wait... can that even be done on the G2?06:32
* tbr isn't familiar with that phone06:32
Elephant454LG G2: d80006:33
locusfhmm if its different from the g3 which has locked bootloader, then it might be doable06:33
locusfremember that following hadk and building the HA layer is just 25% of the work :)06:34
Elephant454It has a locked bootloader. Recovery and flashing is only accessable through a LOKI exploit.06:34
Elephant454Will this present problems of it's own?06:35
locusfnot sure06:36
locusfcan you flash custom kernel on it?06:36
Elephant454Trying to remember if I ever have in the past...06:36
locusfcm zip usually does that06:37
Elephant454I have flashed CM 11 through recovery and run that on there before.06:37
locusfah and its at&t g206:38
tbrwell, if you have a working recovery, then things should be possible06:39
Elephant454This seemed far-fetched at first, but now I feel very excited by the posibilities...06:39
tbrbtw: if there is cm11, then it might be easier to go with the cm11 version of the hybris patches?06:39
locusfnot just easier but required06:40
locusfeg. sailfishos compatibility layer which gives you linux glibc on android libraries06:40
Elephant454So I would be using the pieces of CM 11 that I would need to port Sailfish as opposed to the CM 10.1 pieces? I haven't read over the rest of the porting guide yet. I wanted to make sure that the CM version wouldn't present a roadblock, first.06:44
tbrif there is no cm10.1 for your phone, then using CM11 as the base is the sensible choice06:45
Elephant454Ok. Thank you guys so much for your help. I should hit the hay so I can tackle this with a clear head later. I'll let you know how it goes.06:46
*** Elephant454 <Elephant454!> has quit IRC (Quit: leaving)06:46
uhhimherewhat is dalvik turbo ?07:00
locusfno idea07:01
*** tanty_off is now known as tanty07:02
uhhimhereThis article's factual accuracy may be compromised due to out-of-date information.07:05
uhhimherei guess its a dead thing07:05
*** ruthenianboy <ruthenianboy!> has joined #sailfishos-porters07:10
ruthenianboyI successfuly build hybris boot and recovery images. THey are both the same in size. When I flashed them to phone, after reboot phone booted into fastboot mode (download mode).07:11
*** ruthenianboy <ruthenianboy!> has quit IRC (Quit: leaving)07:19
*** harha <harha!> has joined #sailfishos-porters07:26
*** ruthenianboy <ruthenianboy!> has joined #sailfishos-porters07:31
*** uhhimhere <uhhimhere!~uhhimhere@> has quit IRC (Quit: Leaving)07:56
*** cxl000 <cxl000!> has joined #sailfishos-porters07:58
*** plfiorini <plfiorini!> has quit IRC (Quit: Sto andando via)07:59
*** eleroux <eleroux!~Adium@> has joined #sailfishos-porters08:04
sledgesruthenianboy: fastboot boot hybris-recovery.img08:10
*** carepack <carepack!> has joined #sailfishos-porters08:13
carepackgood morning08:13
ruthenianboysledges: i flashed it via fastboot flash recovery hybris-recovery.img. Why to boot image via fastboot. Isnt't it supposed to be pernamently flashed in phone?08:14
tbrruthenianboy: for testing it makes sense to NOT make it permanent...08:16
*** alin_ <alin_!> has quit IRC (Quit: Konversation terminated!)08:20
ruthenianboyWill try and see. Maybe it is really not working and phone is stucked in fastboot instead of booting08:21
*** BluesLee <BluesLee!> has joined #sailfishos-porters08:23
BluesLeesledges: can you point me to the jolla camera app sources?08:24
sledgesBluesLee: what are you up to?;)08:26
BluesLeesledges: or point me to the changes which have been made to get the cam running on the nexus4?08:27
sledgesBluesLee: they are in the .ks08:27
BluesLeesledges: i followed the gst0.1 -> gst1.0 discussions on jolla together, it makes no sense to wait08:29
sledgesBluesLee: and will work only with android camera HAL v108:29
sledgesBluesLee: linky?08:29
BluesLeesledges: and what is the current hal version?08:29
sledgesBluesLee: depends on device08:30
BluesLeesledges: nexus508:30
sledges v208:30
BluesLeesledges: any chances to pull sone gst1.0 libraries from ubuntu touch and write a small qt/qml app to take rudimentary cam shots?08:31
BluesLeesledges: is there an overview what is open-source and what not in sfos?08:33
sledgesyou need to collab with MSameer who's rolling out gst1.0 (they should already be there to compile and natively integrate side-by-side), and then to write the HAL adaptor for gst1.0, hope he can alrady parallelise the effort08:33
BluesLeesledges: he is already delivering gst1.0 with his own camera app, the work has to be done on the hal adaptor i guess08:35
sledgesBluesLee: it's not that simple as it sounds08:35
sledgesto bundle gst in a separate app, as opposed system-wide, but like i said, you can have gst1.0 libs already on sfos08:36
BluesLeesledges: the introduction of gst1.0 has a big impact on the whole system, thats the reason why it takes so.long08:36
sledgesit can stay side-by-side with no side-effects (sic):)08:36
BluesLeeas on some desktop distros08:36
sledgesas per gst design itself, it's possible08:37
BluesLeesledges: i read an ubuntu touch introduction for devs, parallels to sfos are interesting08:38
BluesLeesledges: is sfos cooperating with the utouch guys to build synergies?08:38
sledgesubuntu had camera's working via compay layer, via which we never took on08:39
*** jerpelea <jerpelea!d5730a62@gateway/web/freenode/ip.> has joined #sailfishos-porters08:45
*** harha <harha!> has quit IRC (Quit: Leaving.)08:56
*** Guest65508 is now known as RaYmAn09:00
*** plfiorini <plfiorini!> has joined #sailfishos-porters09:04
*** BluesLee_ <BluesLee_!> has joined #sailfishos-porters09:04
*** BluesLee_ <BluesLee_!> has quit IRC (Client Quit)09:04
*** BluesLee <BluesLee!> has quit IRC (Ping timeout: 250 seconds)09:05
*** PhompAng <PhompAng!> has joined #sailfishos-porters09:11
*** iTune <iTune!> has quit IRC (Ping timeout: 252 seconds)09:13
*** BluesLee_ <BluesLee_!> has joined #sailfishos-porters09:16
*** BluesLee_ <BluesLee_!> has quit IRC (Client Quit)09:17
*** ruthenianboy <ruthenianboy!> has quit IRC (Ping timeout: 245 seconds)09:39
*** harha <harha!> has joined #sailfishos-porters09:42
*** gogeta <gogeta!~gogeta@> has joined #sailfishos-porters10:11
*** beidl_ <beidl_!> has quit IRC (Read error: Connection reset by peer)10:14
*** piggz <piggz!> has joined #sailfishos-porters10:14
*** alboe <alboe!> has joined #sailfishos-porters10:15
*** gogeta <gogeta!~gogeta@> has quit IRC (Ping timeout: 252 seconds)10:28
*** Lipevakala <Lipevakala!> has joined #sailfishos-porters10:31
*** Lipevakala <Lipevakala!> has quit IRC (Ping timeout: 256 seconds)10:38
*** Lipevakala <Lipevakala!> has joined #sailfishos-porters10:42
*** Lipevakala <Lipevakala!> has quit IRC (Client Quit)10:44
*** Lipevakala <Lipevakala!> has joined #sailfishos-porters10:45
*** krnlyng <krnlyng!~liar@> has quit IRC (Ping timeout: 240 seconds)10:49
mal-sledges: I was looking at the source code to see if that modification would work or not, I don't see why it would help in my case since in test_hwcomposer.cpp all function calls have HWC_NUM_DISPLAY_TYPES which is also used in hwc.cpp reset-function which was to be patched10:54
*** uhhimhere <uhhimhere!~uhhimhere@> has joined #sailfishos-porters10:54
sledgesmal-: but has other values being passed (so a test for qmlscene would then also be in order)10:55
sledgesi.e. qpa-hwcomposer-plugin is sending different values depending on hwc hw version10:55
sledgesand those may be also not the only codepaths taken10:56
sledgesit is a long shot, but have you got a better plan?;)10:56
mal-no :)10:56
sledgesand a shot in the dark too, i'd say ;)10:56
sledgesif it fails (or if we want to have test_hwcomposer working), we'll still be looking at that assert10:57
sledgesbut before that, I'll need to check how test_hwcomposer and test_eglfs(or _fbdev? can't remember now) work on a working UI device, and what env variables they take10:57
* sledges afk or a bit10:58
*** carepack <carepack!> has quit IRC (Quit: Lost terminal)11:05
*** dr_gogeta86 <dr_gogeta86!~gogeta@> has quit IRC (Ping timeout: 256 seconds)11:12
*** beidl <beidl!> has joined #sailfishos-porters11:17
*** iGhost <iGhost!> has joined #sailfishos-porters11:21
*** krnlyng <krnlyng!~liar@> has joined #sailfishos-porters11:22
*** dr_gogeta86 <dr_gogeta86!~gogeta@> has joined #sailfishos-porters11:24
*** PhompAng <PhompAng!> has quit IRC (Ping timeout: 252 seconds)11:24
*** beidl <beidl!> has quit IRC (Ping timeout: 245 seconds)11:26
mal-sledges: just wondering what is the actual error code in the failure, I only see that there is some error but which one11:47
sledgesmal-: strace?11:48
*** carepack <carepack!> has joined #sailfishos-porters11:48
sledgesor gdb ;)11:48
mal-at least I cannot figure it out from strace11:49
mal-looking at the code, in most failures there should be something in android logs since there are calls to ALOGE11:51
mal-I'll need to look into that this evening11:52
*** phdeswer <phdeswer!> has quit IRC (Ping timeout: 252 seconds)11:59
*** birdzhang <birdzhang!~kvirc@> has joined #sailfishos-porters12:00
*** beidl <beidl!> has joined #sailfishos-porters12:00
*** birdzhang <birdzhang!~kvirc@> has left #sailfishos-porters ("Once you know what it is you want to be true, instinct is a very useful device for enabling you to know that it is")12:11
*** zhxt <zhxt!~zhxt@> has joined #sailfishos-porters12:14
*** zhxt <zhxt!~zhxt@> has quit IRC (Read error: Connection reset by peer)12:17
*** zhxt <zhxt!~zhxt@> has joined #sailfishos-porters12:17
*** phdeswer <phdeswer!~phdeswer@2001:998:2a:dead:6267:20ff:fe47:e16c> has joined #sailfishos-porters12:38
vakkovDeveloper Taixzo needs a Jolla phone to port his Google now-like client Saera t Sailfish.
*** cmazieri <cmazieri!> has joined #sailfishos-porters12:45
*** harha1 <harha1!~harha@> has joined #sailfishos-porters13:03
*** phdeswer <phdeswer!~phdeswer@2001:998:2a:dead:6267:20ff:fe47:e16c> has quit IRC (Ping timeout: 250 seconds)13:04
*** harha <harha!> has quit IRC (Ping timeout: 250 seconds)13:06
*** Tassadar <Tassadar!> has joined #sailfishos-porters13:50
*** zhxt <zhxt!~zhxt@> has quit IRC (Quit: Konversation terminated!)13:50
*** phdeswer <phdeswer!> has joined #sailfishos-porters13:52
*** zhxt <zhxt!~zhxt@> has joined #sailfishos-porters13:54
*** trt` <trt`!> has quit IRC (Quit: AndroIRC - Android IRC Client ( ))13:55
*** trt` <trt`!> has joined #sailfishos-porters13:58
*** beidl_ <beidl_!> has joined #sailfishos-porters14:06
*** beidl <beidl!> has quit IRC (Ping timeout: 246 seconds)14:09
*** harha1 <harha1!~harha@> has quit IRC (Quit: Leaving.)14:10
*** eleroux <eleroux!~Adium@> has quit IRC (Quit: Leaving.)14:11
*** zhxt <zhxt!~zhxt@> has quit IRC (Quit: Konversation terminated!)14:21
*** zanac <zanac!> has joined #sailfishos-porters14:45
*** harha <harha!> has joined #sailfishos-porters15:12
*** harha <harha!> has quit IRC (Ping timeout: 250 seconds)15:23
*** harha <harha!> has joined #sailfishos-porters15:26
*** harha <harha!> has quit IRC (Ping timeout: 245 seconds)15:33
*** harha <harha!> has joined #sailfishos-porters15:36
*** zama <zama!~zama@unaffiliated/stryx/x-3871776> has quit IRC (Ping timeout: 245 seconds)15:39
*** zama <zama!~zama@unaffiliated/stryx/x-3871776> has joined #sailfishos-porters15:41
*** Lipevakala <Lipevakala!> has quit IRC (Ping timeout: 250 seconds)15:55
*** Lipevakala_ <Lipevakala_!> has joined #sailfishos-porters15:55
*** harha <harha!> has quit IRC (Quit: Leaving.)15:59
*** Lipevakala <Lipevakala!> has joined #sailfishos-porters16:00
*** Lipevakala_ <Lipevakala_!> has quit IRC (Ping timeout: 256 seconds)16:02
*** jerpelea <jerpelea!d5730a62@gateway/web/freenode/ip.> has quit IRC (Ping timeout: 246 seconds)16:21
*** Tassadar <Tassadar!> has quit IRC (Ping timeout: 240 seconds)16:24
*** arcean <arcean!> has joined #sailfishos-porters16:37
*** vakkov <vakkov!> has quit IRC (Ping timeout: 245 seconds)16:48
*** plfiorini <plfiorini!> has quit IRC (Quit: Sto andando via)17:01
*** vakkov <vakkov!~vakkov@> has joined #sailfishos-porters17:01
*** tanty is now known as tanty_off17:02
*** ruthenianboy <ruthenianboy!> has joined #sailfishos-porters17:07
ruthenianboyDo someone know what's the problem when I am trying to boot via "fastboot boot hybris-recovery.img" and I am getting following error: "FAILED (remote: dtb not found)" ? I think that image was not packed correctly17:09
ruthenianboyPorting journey will be very complicated for me as I see...17:11
sledgesyour boot.img should be made via these commands:
sledgesor it can be as simple as
sledgesruthenianboy: see how hybris-*.img are made:
sledgesand adjust  accordingly. to be sure, doublecheck with how the boot.img from looks like via split_bootimg command in mersdk17:14
*** carepack <carepack!> has quit IRC (Quit: Lost terminal)17:15
locusfyeah the dtb is apparently vital for some devices17:25
*** ruthenianboy <ruthenianboy!> has quit IRC (Ping timeout: 246 seconds)17:31
*** ruthenianboy <ruthenianboy!> has joined #sailfishos-porters17:34
*** uhhimhere <uhhimhere!~uhhimhere@> has quit IRC (Ping timeout: 255 seconds)17:34
ruthenianboyunfortunatelly this is to much complicated for me. I do not know where to place that files17:35
*** alin <alin!~alin@opensuse/member/ealin> has joined #sailfishos-porters17:38
*** gogeta <gogeta!> has joined #sailfishos-porters17:39
*** uhhimhere <uhhimhere!~uhhimhere@> has joined #sailfishos-porters17:40
ruthenianboycan I make image manually from files in out/target/product/w7? There is kernel, initrd, etc.17:41
*** uhhimhere <uhhimhere!~uhhimhere@> has quit IRC (Max SendQ exceeded)17:43
*** iGhost <iGhost!> has quit IRC (Ping timeout: 244 seconds)17:43
sledgesruthenianboy: just split_bootimg is your friend17:44
sledgesuse it with boot.img from your, and it will give you cmdline how to re-create it (substituting kernel, initrd and dt)17:45
*** uhhimhere <uhhimhere!~uhhimhere@> has joined #sailfishos-porters17:46
sledgesruthenianboy: where it says "second", that means device tree image17:46
*** gogeta <gogeta!> has quit IRC (Ping timeout: 264 seconds)17:48
ruthenianboygenerated files are not usable. They do not have device tree (I tried to boot kernel and initrd directly). I must recompile everything from begining. How can split_bootimg help me?17:51
sledges19:45 < sledges> use it with boot.img from your cm.zip17:52
*** uhhimhere <uhhimhere!~uhhimhere@> has quit IRC (Ping timeout: 252 seconds)17:52
sledgesyour device tree file should reside after kernel build:
sledgesis the missing piece17:53
*** piggz <piggz!> has quit IRC (Ping timeout: 264 seconds)17:53
*** uhhimhere <uhhimhere!~uhhimhere@> has joined #sailfishos-porters17:54
*** ruthenianboy <ruthenianboy!> has quit IRC (Ping timeout: 246 seconds)18:00
*** ruthenianboy <ruthenianboy!> has joined #sailfishos-porters18:16
ruthenianboyI used split_bootimg to see how to build image. Then I used mkbootimg with parameters from split_bootimg to build boot.img but file again do not contain dt.18:17
locusfruthenianboy: whats your device?18:18
locusfmkqcdbootimage usually takes in the dtbs18:18
*** trt` <trt`!> has quit IRC (Ping timeout: 252 seconds)18:18
ruthenianboylge w7 - LG L90 D405n18:19
*** plfiorini <plfiorini!> has joined #sailfishos-porters18:24
*** trt` <trt`!~trt@> has joined #sailfishos-porters18:24
*** plfiorini <plfiorini!> has quit IRC (Remote host closed the connection)18:24
*** alin <alin!~alin@opensuse/member/ealin> has quit IRC (Quit: Konversation terminated!)18:26
*** plfiorini <plfiorini!> has joined #sailfishos-porters18:30
*** alboe <alboe!> has quit IRC (Quit: :tuiQ)18:35
*** trt__ <trt__!~trt@> has joined #sailfishos-porters18:36
*** trt` <trt`!~trt@> has quit IRC (Ping timeout: 252 seconds)18:38
locusfruthenianboy: <- try this18:38
*** trt__ is now known as trt`18:38
ruthenianboylocusf: that tools are working, bootable img was created18:55
*** Tassadar <Tassadar!> has joined #sailfishos-porters18:55
*** trt` <trt`!~trt@> has quit IRC (Ping timeout: 250 seconds)18:57
*** trt` <trt`!> has joined #sailfishos-porters18:58
ruthenianboyi flashed newly created boot.img to boot. Now when I power on the phone, LG logo appear and aftere some time it will boot to recovery menu. It is correct behavior?19:05
*** ruthenianboy <ruthenianboy!> has quit IRC (Quit: leaving)19:06
*** CardenalJimenez <CardenalJimenez!~CardenalJ@2a01:e35:2422:60d0:b054:75a6:4b11:8364> has joined #sailfishos-porters19:08
*** piggz <piggz!> has joined #sailfishos-porters19:19
mal-sledges: I might have found the actual error in hwc.cpp, the condition is otherwise ok, but it has <= instead of <19:23
mal-in other places, such as the one that was patched before, it is the correct <19:24
sledgesmal-: oh cool, in the for(;;) loop?19:25
mal-since I could not see any errors in log there is no other way for it to fail19:25
mal-now I just need to compile a new version of the library19:28
*** zanac <zanac!> has quit IRC (Remote host closed the connection)19:28
mal-that is not the only place where there is similar mistake, but in the other place it does not cause any problems, it hwc_set the count goes 0->3 but in the other place it goes 3->0 and therefore the final handled situation is actually successful19:31
sledgesbut if 3 also get's in the loop (when it should be max=3-1=2)19:32
sledgeson the first reversing iteration19:32
mal-yes, for other devices it is like that19:33
mal-for msm8960 and msm8974 at least19:33
mal-those have for (int32_t i = numDisplays - 1; i >= 0; i--)19:34
mal-I have for (int32_t i = numDisplays; i >= 0; i--)19:35
sledgesso it's 2->0, all good;)19:35
sledgesyep, you are missing - 119:35
mal-but 3->0 also works in that case, even though it's incorrect19:35
sledgesdoes it call set() inside the loop?19:35
mal-no, set is called separately19:36
mal-that other error is in hwc_prepare19:36
sledgesprobably non critical on MAX19:36
sledges(when it should be MAX-1)19:37
mal-there it is19:37
mal-the error is critical only when going 0->319:38
sledgeson the same structs?19:39
sledgesi mean maybe displays is allocated [MAX>numDisplays] ?19:39
*** Nekron_ <Nekron_!> has joined #sailfishos-porters19:39
mal-what do you mean?19:40
sledgeshwc_display_contents_1_t** displays is allocated somewhere as displays[NUM_DISPLAYS]19:40
*** Nokius_ <Nokius_!> has joined #sailfishos-porters19:40
sledgesand NUM_DISPLAYS is 419:40
sledgesand numDisplays is passed as 319:40
sledgesthat's why things don't blow up19:40
sledgeswith displays structure19:40
sledgesand with _prepare19:40
sledgesbut blowsup elsewhere19:41
sledges^ a speculation19:41
mal-look at the switch after that place19:41
mal-it only has 3 options19:41
sledgesand it's strict 1:1 mapping?19:41
sledgesand it never returns -EINVAL?19:42
sledgesso maybe the alloc is filled with zeroes, and it goes 0,2,1,019:42
mal-it does in hwc_set, that is the condition I get19:42
mal-that causes the assert to fail19:43
sledgesyes, we're talking about the loop which doesn't blow up19:43
*** Nokius <Nokius!> has quit IRC (Ping timeout: 240 seconds)19:43
sledgesbut there's nothing better to do now than to debug;)19:44
mal-in the hwc_prepare, the loop just continues even after failure and since the next loops are ok, it eventually return success19:44
sledgesmal-: well spotted, ofc!:)19:45
locusfI'm getting this error while compiling libhybris for lollipop: nativewindowbase.cpp:10:23: fatal error: sync/sync.h: No such file or directory19:47
mal-we'll see what problems I run into after I fix this19:47
sledgesStskeeps: ^^19:47
locusfsledges: hes off to bed, just asked him19:47
*** piggz <piggz!> has quit IRC (Quit: Konversation terminated!)19:48
*** piggz <piggz!> has joined #sailfishos-porters19:48
locusfI've got system/core/libsync/sync/sync.h but its not picked up to droid-headers19:49
sledgeslocusf: how does a 10.1/11.0 tree looks like wrt?19:49
locusfI don't have one right now19:49
*** M4rtinK <M4rtinK!> has joined #sailfishos-porters19:51
*** SamZaNemesis <SamZaNemesis!~Sam@> has joined #sailfishos-porters19:53
locusfah it has changed location19:54 expects it to be in system/core/include/sync19:54
mal-sledges: the easy way for me to see what happens after fixing that problme would be to just remove the assert from test_hwcomposer, since that error is really not fatal19:54
sledgesmal-: go ahead19:55
mal-I'll of course fix the real cause later but it might take a while to compile it19:56
*** ruthenianboy <ruthenianboy!> has joined #sailfishos-porters19:56
locusfaand we don'tt have bionic headers either19:57
locusfand I really don't know if your bionic is patched19:57
locusfsurely some work tomorrow for me then19:57
locusflaters, nigh19:57
*** saidinesh5 <saidinesh5!> has quit IRC (Quit: - Chat comfortably. Anywhere.)19:57
*** saidinesh5 <saidinesh5!> has joined #sailfishos-porters19:57
sledgesbionic patches is part of hadk, not aosp:)19:58
*** vakkov <vakkov!~vakkov@> has quit IRC (Ping timeout: 265 seconds)19:59
sledgesruthenianboy: the correct behaviour is when you see same events that happen via fastboot boot boot.img and your re-created boot.img duplicate19:59
ruthenianboyLooks like my attempt to build hybris boot was not successful. I think it something to do with local manifest. I do not know why it did not created proper image (without dtb).19:59
sledgesruthenianboy: the time-consuming method would be to build cm on your own20:00
sledgesin a separate tree20:00
sledgesyour device is not officially supported on cm wiki20:01
locusfsledges: well won't we still use aosp bionic?20:01
locusfit is compiled like the rest of the system20:01
sledgeslocusf: yes, but if you checked out aosp, surely it wouldn't have bionic patches out of the box20:01
sledgesfor mer20:01
locusfyes sure20:02
ruthenianboyBut it is included in cyanogenmod git as android_device_lge_w7. I will try to repack original boot.img from phone with kernel generated in step 5.2 to see if kernel is functional. Or I will try to use kernel and device source provided from LG on their site. I am in impasse now as my knowledge is limited. Any other ideas where I can start looking? Thanks20:05
sledgesruthenianboy: that's a good try - swapping in and out, given the situation that it's not officially supported cm device (presence of git repos doesn't mean you have instructions to reproduce the build sanely at hand)?20:06
sledges(no question mark:)20:06
piggzis there anything like planet-sailfish ... or just use planet-maemo?20:07
piggzthe planet-maemo rss feed is rejected by owncloud news :/20:07
mal-sledges: it works!!!20:08
sledgesmal-: \o/20:08
mal-there is some nice animation on screen when I run test_hwcomposer20:09
sledgesyou get picture!? 8D20:09
sledges\\o// rock'n'roll!! \m/20:09
sledgesbloody caf :DDD20:09
mal-indeed :D20:09
sledgesEGL_PLATFORM=hwcomposer ?20:10
sledgesso, next victim - qmlscene :))20:11
mal-yes :)20:11
sledgeswhich now might need intervention into hwcomposer.msm*.so...20:12
ruthenianboysledges: I will try to build CM on my own from lg sources or from existing gits. I still do not understand difference between CM build and step 5.2 HADK. What is the purpose of hybris-boot and hybris-recovery? Hybris-boot contain same kernel but different initrd which provide debug via telnet and modified init script which boot SFOS?20:12
sledgesruthenianboy: hybris-boot contains modified kernel config too, which needs to pass mer-kernel-check20:12
mal-sledges: looking at the qmlscene with HYBRIS_LOGGING_LEVEL=debug, it could be failing in the same place20:18
mal-but it just crashes for some reason20:19
sledgesit asks for more20:19
mal-in both cases last debug output is EGL hwcomposer_window.cpp:220 (virtual int HWComposerNativeWindow::queueBuffer(BaseNativeWindowBuffer*, int))20:19
mal-onw I need to look for the source of qmlscene20:20
mal-or whatever is actually implementing HWComposerNativeWindow when you run qmlscene20:23
Nokius_sledges: did you find a way to get the problem solved? may its find5 time again :)20:24
*** Nekron_ <Nekron_!> has quit IRC (Remote host closed the connection)20:24
sledgesNokius_: patience my friend:))20:25
Nokius_sledges: all day long :P20:29
Nokius_looks like you had similar effect last night20:30
*** Sfiet_Konstantin <Sfiet_Konstantin!> has joined #sailfishos-porters20:32
mal-sledges: I cannot understart why that error could cause segfault, maybe there is something else wrong also20:32
Nokius_so thats why ask ;) I'm not interested to rush annoy anyone20:32
sledgesNokius_: no worries:)20:32
mal-but still it would quite a coincidence that the crash just happens to happen at the same place20:32
*** ruthenianboy <ruthenianboy!> has quit IRC (Ping timeout: 252 seconds)20:34
mal-hmm, there could be a reason why might segfault, it does access an out-of-bounds access to an array20:36
*** vakkov <vakkov!> has joined #sailfishos-porters20:45
sledgesmal-: found all sources you needed?20:52
mal-not yet20:53
mal-well, I haven't looked much yet20:54
mal-needed some food20:54
mal-could this "hwc_display_contents_1_t *list = displays[i];" cause a segfault if "i" is too large20:55
sledgestrace where displays is alloced20:55
mal-sledges: I knew about that source, but that calls this->present, which was implemented in test_hwcomposer20:57
mal-queueBuffer calls that, and that was the location of the problem20:58
mal-virtual void present(HWComposerNativeWindowBuffer *buffer)20:59
mal-so that has to be implemented somewhere else also21:00
sledgeszypper in qt5-qtdeclarative-qmlscene21:00
mal-it seems qt5-qpa-hwcomposer-plugin21:00
sledgesqtdeclarative is to look for21:00
sledgesor that21:01
mal-there is HWC_PLUGIN_EXPECT_ZERO(err);21:01
mal-so a similar check is done, now I need check what HWC_PLUGIN_EXPECT_ZERO does21:01
*** piggz <piggz!> has quit IRC (Ping timeout: 256 seconds)21:02
mal-that should only print a warning, so it never reaches that far, so my guess for the reason of segfault look likely21:03
mal-actually I'm not sure where it outputs the warning since it calls qWarning21:04
mal-I'll try to compile the fixed hwcomposer tomorrow to see if that fixes things21:10
sledgesok, good progress today21:10
sledgesit might boil down to modifying caf bits, but you'll confirm that from a distant approach;)21:11
mal-I'm very pleased with today's progress, thanks a lot21:12
mal-I was starting to lose hope already with the lack of graphics21:13
sledgeswe finally dug the tunnel into the rabbit hole of android gpu stack :D21:14
sledgesthis should've been done 8 months ago, but oh well;p21:14
mal-it was Stskeeps that made the first discovery which enabled all this progress21:15
sledgeshe gave a push, but i still wonder how test_hwcomposer would look like without QCOM_BSP bits;)21:16
mal-what does that do actually?21:17
sledgesion mem alloc, might be related, might be not (yet) affecting21:17
* sledges afk dinner21:17
sledgesas numdisplays has been issue in the past, with flagships like nexus521:20
mal-it would be interesting to look at how android uses hwcomposer, since it does not fail with those bugs in hwc.cpp21:21
*** Tassadar <Tassadar!> has quit IRC (Ping timeout: 245 seconds)21:24
*** alin <alin!> has joined #sailfishos-porters21:35
*** alin <alin!> has quit IRC (Changing host)21:35
*** alin <alin!~alin@opensuse/member/ealin> has joined #sailfishos-porters21:35
*** Sfiet_Konstantin <Sfiet_Konstantin!> has quit IRC (Quit: Konversation terminated!)21:43
*** piggz <piggz!> has joined #sailfishos-porters21:44
alinsledges: Stskeeps lbt this is funny...
*** vakkov <vakkov!> has quit IRC (Ping timeout: 256 seconds)22:01
*** cxl000 <cxl000!> has quit IRC (Quit: Leaving)22:04
M4rtinKarmor support :)22:11
alinM4rtinK: yap... intercontinental sailfish...22:19
M4rtinKNemo Mobile attack submarine! :)22:20
alinM4rtinK: with nuclear battery22:25
M4rtinK"Sailfish Class" would make a nice ship or submarine class :)22:26
klopsi-u3hey cool toy22:46
alinsledges: got the tizen phone...23:03
alinblimey... i did not know they can dumb down proper linux to android23:04
alini wonder if we can put sailfish on it23:07
klopsi-u3alin: what for example, is dumbed down?23:14
alinno ssh... some adb clone... no software management app... the interface is ugly.... and believe I have very low aesthetical expectections from software23:18
alinall the interface looks like a cheap android clone23:19
*** vakkov <vakkov!> has joined #sailfishos-porters23:24
alinklopsi-u3: ok apparently there is a way to downlaod other stuff on it23:49
alinand put a better tizen... if supports it...23:49
klopsi-u3it is the samsung z1?23:50
alinklopsi-u3: yap23:50
klopsi-u3so jolla has the better UI?23:51
alinklopsi-u3: yap23:52
alinklopsi-u3: ubuntu one is not bad23:52
alinklopsi-u3: anyhow will be a good experience to put the proper tizen on it to see23:52
klopsi-u3yes i suppose23:53
klopsi-u3do you work with phones for a job alin?>23:54
alinklopsi-u3: nope23:56
alinklopsi-u3: just curious23:57
*** vakkov <vakkov!> has quit IRC (Ping timeout: 250 seconds)23:57

Generated by 2.17.1 by Marius Gedminas - find it at!