Monday, 2024-11-04

rinigus@b100dian: very nice!04:15
T42<elros34> @bruuuh121 why don't you follow instruction from https://github.com/sailfish-on-ginkgo/main? Looks like it has all needed informations05:50
T42<neelssh> I installed waydroid but as soon as i open it, it shows android session finished, how can i fix this ?12:02
*** phlixi is now known as Guest843716:31
*** phlixi_ is now known as phlixi16:31
malrinigus: a14 r67 built without issues from latest sony repos atfer they merged all of the changes19:37
rinigusmal: yes, I have just rebuilt myself as well19:37
rinigusif you have r67 working, I wonder if you have aosp keyboard working (assuming that you flash it)19:38
rinigusif you don't that's ok - I opened an issue regarding it already and let them test it too19:38
malwell not sure if I will even boot android but just directly test in sailfish19:39
malmaybe I could test android first19:39
malI did user build though, not sure if they fixed that, user builds had issues with selinux or something19:40
rinigusno, I don't think they did. there are lots of selinux messages :(19:44
rinigus@b100dian: hacked to make keyboard work by moving corresponding lib into correct location. that allowed me to test wifi and ... it crashed on aosp14 too. will open an issue regarding it19:45
T42<b100dian> Hopefully userland crash? Or even vendor services, we should not have that in sfos for wifi19:58
T42<b100dian> rinigus: I'm going to reduce the changes to nagara hybris boot which will result in a forced-pushed commit, since I had unfinished changes "published" while debugging.20:14
T42<b100dian> Normally we'd want to avoid that:)20:15
rinigus@b100dian: no idea whether its user or not. device just reboots :(20:16
T42<b100dian> o_O you mean "crashed" -> kernel crash ?20:17
rinigusre force push - sure. and we will avoid it when we get into a more stable state20:17
rinigus@b100dian: yes, total proper crash20:17
T42<b100dian> rinigus: does this kernel has anything in /sys/fs/pstore upon reboot?20:17
riniguslet's see20:17
T42<b100dian> another thing you could try is running dmesg over adb shell while doing this.20:18
T42<b100dian> I had 1 out of 2 chances of seeing a message before reboot in dmesg over telnet previously, hopefully with adb it is possible too20:18
rinigus@b100dian: it reboots into recovery and there /sys/fs/pstore is empty :(20:22
rinigusbut there are some messages in logcat, maybe that would help. I'll open an issue at sony20:23
T42<b100dian> sure, but logcat might be too slow to win a race with a kernel thread bringing all down. dmesg has a higher chance of outputing the last error, and this is what you want for a kernel oops20:25
T42<b100dian> gosh, only know I observe this setting commented in boardconfig `#BOARD_KERNEL_CMDLINE += earlycon=simplefb,0xb8000000,1080,2520`  with the comment `# FB console`. Wonder if it really works, that would've helped yesterday ;)20:32
T42<b100dian> rinigus: all pstore configs seem to be set. I would dare to say that there could be no dtsi config for the memory to be kept between reboots20:39
T42<b100dian> it seems that voidanix fixed this for kumano https://github.com/sonyxperiadev/kernel/pull/2579/files20:39
rinigus@b100dian: here is my config description and the log: https://github.com/sonyxperiadev/bug_tracker/issues/85220:41
T42<b100dian> oh, you got [kworke] messages alright.. so, .. firmware? :snif:20:43
rinigus@b100dian: are you still on stock? With all these bugs I am considering whether porting to lineage via stock would be easier. maybe even not full lineage...20:43
T42<b100dian> As I mentioned somewhere, someone just unpacked rootfs from one of my ports, including the `usr/libexec/droid-hybris/system` which was lineage based, and ran that with a vendor system image (so not lineage)20:45
T42<b100dian> It is not impossible to think that if we build on the AOSP system, that can be used on top of Lineage GSI20:45
rinigusso, I wonder whether you could boot stock using aosp14/sfos kernel and maybe dtbo. or is it very bad idea? can it damage/erase something20:46
rinigus?20:46
rinigusyes, I remember that with your port. and the fact that we have gsi which almost works with stock...20:46
T42<b100dian> ah, kernel difference - forgot about it - there was 5.10 on stock20:46
T42<b100dian> hmm.. will need also modules I guess20:47
T42<b100dian> so kernel 5.15 from AOSP + dtbo + modules.. when you factor modules in.. need to think about it.20:48
T42<b100dian> I am not anymore on stock because recovery adb root access.. just flashed not working AOSP14 r5520:48
rinigusanyway, its late for me and maybe we should ponder about it. modules and kernel dump are available from sony. we can also grab dtb data either from device or from sony's dump. links are in wiki under lineageos20:49
rinigusno problem, maybe I get to test stock tomorrow20:49
rinigusgn20:49
T42<b100dian> 'nite20:51
T42<b100dian> rinigus: kernel pushed https://github.com/sailfishos-sony-nagara/android_kernel_sony_msm and cleaned up https://github.com/sailfishos-sony-nagara/hybris-boot/22:38
T42<b100dian> mal: small PR for hybris boot: https://github.com/mer-hybris/hybris-boot/pull/22222:38
maldid you verify that you need to use rndis.rndis?23:04
T42<b100dian> you mean to try rndis.usb0 ? Hmm... no23:19
T42<b100dian> mal: works without changing rndis.usb0/rndis_bam lines too23:25
T42<b100dian> of course, bam still doesn't work, as "mkdir: can't create directory '/config/usb_gadget/g1/functions/rndis_bam.rndis'"23:25
T42<b100dian> so I could revert even those changes. But its nighty-night time ;)23:30
T42<arcyfix> Try `waydroid logcat > ~/debug.txt` (re @neelssh: I installed waydroid...)23:39

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