tmp64 | What does blinking orange (green in logs?) LED mean? | 05:59 |
---|---|---|
*** pketo_ is now known as pketo | 07:05 | |
mal | tmp64: when does that happen? | 08:20 |
tmp64 | mal: Sometimes when booting up. Other times it just panics with no log in /sys/fs/pstore. | 08:32 |
tmp64 | I was able to get systemd unit list, if that helps: https://pastebin.com/56bSHpkX | 08:35 |
mal | tmp64: journalctl output would be more useful, and dmesg | 08:36 |
tmp64 | mal: Ok, it's going to take a few minutes | 08:38 |
mal | tmp64: also I suggest getting those in separate telnet connections using dmesg -w and journalctl -f -n10000 so you will get even the last messages | 08:42 |
tmp64 | mal: I only have 10-15 sec window, so I piped everything into two files. I'll post them here and try again | 08:44 |
mal | at least dmesg you should take with that to see where it might fail | 08:46 |
tmp64 | mal: journalctl: https://pastebin.com/eLgBvc94 dmesg: https://pastebin.com/8FkcDnpJ | 08:47 |
mal | tmp64: so it reboots after that? | 08:49 |
mal | did you do the things I mentioned yesterday, also are you sure you correctly disabled selinux? | 08:50 |
tmp64 | mal: It goes into "panic mode": screen becomes blue for a frame and then completely black, after that the MMC partitions are available as USB Mass Storage. | 08:51 |
tmp64 | mal: I didn't disable it in the kernel, but I added "selinux=0 audit=0" to kernel params. | 08:52 |
mal | maybe show your modified defconfig | 08:52 |
tmp64 | Last thing in new dmesg is "android_usb: already enabled", nothing else useful | 08:52 |
tmp64 | mal: https://pastebin.com/MecNydhj | 08:53 |
mal | tmp64: nothing /proc/last_kmsg? | 08:54 |
tmp64 | mal: Nope | 08:54 |
mal | try next dmesg -w to see if you get the final messages before it fails | 08:55 |
tmp64 | I got journalctl -f https://pastebin.com/cpLNHEuh | 08:56 |
tmp64 | mal: Looks like I have to tap into UART | 08:56 |
mal | something about tmp files | 08:57 |
tmp64 | I don't think I enabled any memory CGroups | 08:58 |
*** jusa_ is now known as jusa | 08:58 | |
mal | tmp64: the ones mentioned in kernel checker are for 3.4 kernel, 3.10 has CONFIG_MEMCG=y, CONFIG_MEMCG_SWAP=y and CONFIG_MEMCG_KMEM=y | 08:59 |
tmp64 | mal: Ok, I'll try enabling them. Last time I enabled everything LXC requested and the kernel failed to compile. It was probably user namespaces, but next time I only selected required options. | 09:07 |
mal | the lxc configs are optional | 09:08 |
tmp64 | mal: Nope, still panics | 09:34 |
mal | tmp64: run ln -s /dev/null /etc/systemd/system/droid-hal-init.service and maybe also ln -s /dev/null /etc/systemd/system/user@100000.service | 09:37 |
tmp64 | mal: Didn't help. I'm going to tear down the phone and connect to serial console. I've enabled it in the kernel. | 09:42 |
jusa | fyi: pulseaudio-modules-droid-glue 11.1.69 supports parsing of new XML audio policy configuration. if you have port that has this new XML style config please report if you have any issues. | 10:13 |
jusa | and of course I meant package pulseaudio-modules-droid | 10:14 |
spiiroin | review appreciated: https://git.merproject.org/mer-core/mce/merge_requests/83 | 10:39 |
spiiroin | ^ some devices had visible brightness bumping on unblank | 10:39 |
spiiroin | ^ elros | 10:39 |
Yardanico | By the way, why does HADK use GCC 4.9 (and not 6+) for building the kernel? | 11:03 |
r0kk3rz | because thats the version of gcc we use for everything | 11:04 |
Yardanico | r0kk3rz, ok :) | 11:04 |
Yardanico | just a lot of custom ROM developers on android build their kernels with everything like GCC 7/8, even clang sometimes | 11:05 |
r0kk3rz | yeah 'why we use old gcc' is an involved question | 11:17 |
tmp64 | What may cause udhcpd to not start after switch_root? Because I have to manually set IP via ifconfig. | 11:35 |
tmp64 | Init.log says it's running | 11:35 |
tmp64 | Wait | 11:35 |
tmp64 | init-debug: line 285: /etc/udhcpd.conf: No such file or directory | 11:35 |
mal | that has happened to me on some devices | 11:36 |
elros1 | spiiroin: I use elros-als-bug branch without any brightness issues since you created it | 11:39 |
spiiroin | elros1: yeah, just that the fix was done just prior to my vacation ... and then forgotten | 11:42 |
spiiroin | elros1: merged to master now (sans extra debug logging that was in that original wip branch) | 11:43 |
tmp64 | I think I found something | 11:46 |
elros1 | spiiroin: great, thank you. Will it be in next sfos release? | 11:47 |
tmp64 | Device starts with USB serial "Mer", then it changes to device's serial and then to (error) when it goes into "panic mode". | 11:47 |
spiiroin | elros1: basically yes (though I'm not sure about the next / the one after that atm) | 12:34 |
taixzo | What's the channel for sailfish app development? | 12:58 |
kimmoli | #sailfishos | 12:59 |
hpag | hi, can i build sailfish os using a Mac OS device? | 23:09 |
hpag | can sailfishos built using Mac OS | 23:27 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!