mal | usually via recovery but sometimes twrp for example has issue using tar because of some memory issues or something | 00:06 |
---|---|---|
Mister_Magister | or twrp doesn't have bzip2 binary | 00:07 |
Mister_Magister | lol $PLATFORM_SDK_ROOT is empty | 00:50 |
Mister_Magister | ahh i know why | 00:50 |
Mister_Magister | someone dumb didn't add it to bashrc | 00:51 |
Mister_Magister | mal: do you have old ubu-chroot images lying around on jolla's servers? | 01:56 |
mal | Mister_Magister: isn't the link in section 4.3.3 in latest hadk pdf working? | 10:03 |
mal | deathmist: just mentioning that I managed to boot hybris-18.1 to UI yesterday, it still has a terrible bug which prevents direct boot to UI which I need to solve | 11:49 |
mal | all other features seemed to be working fine, I didn't test fingerprint yet | 11:49 |
Mister_Magister | mal: fook i missed it | 15:54 |
Mister_Magister | i just hacked in openjdk 1.7 xd | 15:54 |
deathmist | mal: ooo nice! any rough estimates on when this should be usable from git? I'd like to give it a go before going back to military stuff after xmas :p | 18:02 |
Mister_Magister | mal: do you have patches for the rendering on surfaceflinger? can't find them anywhere | 18:12 |
piggz_ | Mister_Magister: why do you want to do a silly thing like that? | 18:14 |
Mister_Magister | piggz_: compare performance | 18:14 |
Mister_Magister | just to test tings | 18:14 |
Mister_Magister | i had some positive experience when while rendering on surfaceflinger ui was milion times smoother | 18:15 |
Mister_Magister | real 60fps | 18:15 |
piggz_ | weird | 18:15 |
Mister_Magister | i have never seen smooth sfos ui beside huawei p8lite on surfaceflinger | 18:16 |
mal | deathmist: depends on how I manage to fix that one bug | 18:39 |
deathmist | mal: what is the bug out of curiosity? | 18:52 |
mal | deathmist: one bind mount in not done the way it should be which causes android linker to not find linker config | 18:59 |
deathmist | mal: sounds fun, but also not terribly difficult to fix :p | 19:09 |
mal | deathmist: it's always fun to go through android init code to see why it's not done | 19:12 |
T42 | <b100dian> mal: have you used gdbserver and then gdb + symbols from where the build was done? (I tried and failed with a /vendor binary, maybe init works) | 19:15 |
mal | I haven't used that | 19:33 |
T42 | <b100dian> something like this https://wladimir-tm4pda.github.io/porting/debugging_gdb.html | 19:39 |
T42 | <foxclore> Hey hello. I was building for hybris-18.1, and I got into bootloop, it doesn't even boot to boot animation. In wiki its said it might selinux, but as in faq I applied all patches and set selinux to permissive via "androidboot.selinux=permissive" | 21:23 |
T42 | <foxclore> So here are all necessary (i hope) files to help to resolve the issue: https://gist.github.com/foxclore/4cd45846db9ddec9df0e4cc78cbffe74 | 21:23 |
T42 | <foxclore> Any ideas whats going on? | 21:23 |
mal | hybris-18.1 is not really ready yet, at least it depends on what kind of configuration is used in the device | 21:51 |
mal | one possible reason is that you don't have a working kernel init in the hybris-boot.img, it fails because cpio is not allowed, I have a patch for it but it's not in hybris-patches yet | 21:53 |
Thaodan | or build hybris-boot.img outside your self | 21:54 |
mal | build/soong/ui/build/paths/config.go has a list of allowed binaries and it needs cpio in there | 21:56 |
T42 | <foxclore> well xd very unfortunate because lineage 18.1 is the minimum version available for my device | 21:59 |
T42 | <foxclore> So, apply fixes and then rebuild which parts? | 22:00 |
mal | hybris-hal | 22:01 |
mal | but it's possible you will have more issues later, but that really depends on which type of apex setup the device uses | 22:02 |
T42 | <b100dian> is gst-omx.conf something still in used today? (lineage 17.1 base) | 22:29 |
T42 | <adampigg> iirc all the omx stuff was abandoned? | 22:32 |
T42 | <b100dian> I don't know, that's why i ask - I stumbled upon one in a 2018 discussion - I still don't have video recording working so that's why the question | 22:34 |
T42 | <b100dian> so there was some kind of specific plugin. oh well. now I got errors from the omx service on recording, thought I found gold, not this time:) | 22:46 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!