libera/#maemo-leste/ Saturday, 2024-03-30

sicelohttps://maemomotors.co.za/08:00
sicelo :p08:00
dsc_its strange seeing all kinds of towns in south-africa having dutch names ;P08:12
dsc_'rustenburg'08:12
siceloof course the afrikaans people are of dutch origin :-D08:15
sicelothe language is different, obviously, but they can somewhat hear each other, from what i've heard08:16
* sicelo isn't afrikaans, so08:16
dsc_im dutch, and I can understand some afrikaans automatically hehe08:19
sicelotmlind: just reminding you about https://lore.kernel.org/linux-devicetree/ZeBB1GYb31UAW886@tp440p.steeds.sam/T/#t   :-)13:01
sicelothis xz stuff is crazy :-/13:29
inkyyep13:31
inkyhttps://gist.github.com/thesamesam/223949d5a074ebc3dce9ee78baad9e2713:32
inkyi changed the battery of d4. it's not the battery. it doesn't take charge.16:29
inkyand even doesn't boot to hildon.16:29
inkyi saw some usb0 related error and then screen went off.16:29
inkyhmm i connected d4 to pinecharger's 5.3 v port and it looks charging. but i don't know.16:43
inkybut it booted to hildon now.16:43
inkyok let's see. sorry for botherign you with it.16:43
freemangordonsicelo: do I get it right that someone injected a backdoor in xz-utils?16:46
freemangordonlike, in the release tarball?16:48
sicelofreemangordon: yes, and no just someone .. upstream itself :-P16:50
freemangordonwell, upstream...16:50
sicelowell, a sneaky fellow who worked their way up to maintainership16:50
freemangordonis that proved already?16:50
sicelowhat?16:51
sicelothe actions of the person in question? they're right there in git and mailing list history16:51
freemangordonare we sure that new maintainer is the one who put the backdoor?16:51
freemangordonah, ok16:51
sicelohttps://tukaani.org/xz-backdoor16:52
freemangordonI see16:52
siceloi16:53
sicelosux really.16:53
freemangordonyeah16:53
inkybtw we didnt get update for xz from devuan. probably it is not affected.21:34
Wizzupit was only unstable/experimental22:48
arno11sicelo: last updates (devel) broke battery applet23:04
arno11no battery animation, no charging indication, no led23:05
arno11:(23:05
sicelono LED at all? not even charging?23:08
arno11yes23:08
sicelolet me see how that could happen23:09
arno11ok23:09
sicelobut yes, the kernel should have been the first change to go in. hope uvos shows up soon23:09
arno11ok23:10
arno11charging seems to not work at all23:10
sicelommm, that should be completely unrelated23:11
arno11yes23:11
arno11let me check my charger23:11
sicelobecause charging is not controlled by userspace at all ... so it should be charging23:12
siceloand, the kernel change is not on your device as well :-)23:12
arno11yed23:13
arno11yes23:13
sicelothat is ... assuming it's a broken change, which it isn't23:13
arno11(the charger was working few min before the update btw)23:14
sicelowhat kernel version you have?23:14
arno11when i connect it, i see a white led a couple of msec23:14
siceloalso was kernel updated in the first place?23:14
sicelowhite? mmm. this is on running device, or switched off?23:14
arno11still devel 6.1.7623:15
sicelothis is my kernel change, https://github.com/maemo-leste/droid4-linux/pull/9/files ... and it can't break charging in any way. we're only adding an ONLINE property to the charger.23:16
siceloin any case, this change is not on your device, so it isn't the cause :-)23:17
sicelolet's troubleshoot ... you say charging isn't working23:17
sicelolet's see what `upower -d` says. please send full output to a pastebin somewhere23:17
arno11sorry i can't: only have my n900 and low battery...23:18
sicelook. power off the N900 and charge while it's switched off. whatever is broken in Leste won't affect it when switched off23:19
arno11yes i know that, i'll have a look quickly at upower output anyway23:19
arno11btw which device is relevant now ?23:22
Wizzupdo you need help with kernel?23:22
siceloideally you'd show us the full `upower -d` .. so we can see everything that upower sees23:22
sicelobut if you want to show only a bit, check what `/sys/class/power_supply/bq24150a-0/uevent` shows, and `/sys/class/power_supply/bq27200-0/uevent`23:23
siceloWizzup: it'd be great if kernel 6.1.80 was built for repos ... although i think arno11's issue is not related23:24
arno11i'll try using links23:24
sicelowith my changes being out of sync as they are, i.e. with kernel & mce changes missing, the worst that should happen, as far as i can see/understand is - animation will go on forever, which it already did anyway23:25
siceloa regression would be ... it would show incorrect percentage too, because the charger will look like a 2nd battery and confuse the calculations, which is not harmful, but annoying UX of course23:26
Wizzupon my n900 the led is green now btw, but it only appears after a long time, and I still have to update23:28
arno11sicelo: sorry, i can't pastebin23:29
siceloit's alright :-)23:30
arno11i can paste upower -d here btw23:30
sicelommm, maybe paste in pm23:30
arno11ok23:30
siceloit's a lot of lines, might suck to post in chan23:30
siceloyou're doing that `upower -d` while connected to your charger?23:31
arno11yes23:36
arno11is display device relevant ?23:36
arno11cause it recognizes charger plug/unplug23:36
arno11i mean charging/discharging23:37
sicelono, leste doesn't use it23:37
sicelofrom your output, some observations - 1. you're not running the latest leste-config ... otherwise you wouldn't have those twl* items23:38
arno11same with bq24150, but not bq2720023:38
sicelobut the more concerning issue is .. where is isp1704/ isp1707 ...23:39
siceloanyway, your battery is not that low yet23:39
siceloso where is isp170x ... can you `ls /sys/class/power_supply` and see if you have something like that?23:39
arno11indeed23:39
arno11100% sure to run last leste-config23:40
siceloisp170x is the one that actually detects the charger, then informs bq24150a, which is the actual charger23:40
Wizzupsicelo: oh..23:41
WizzupI think gne-displace wasn't run for leste-config23:41
Wizzuplet me see23:41
siceloregarding leste-config, do you have `/etc/modprobe.d/blacklist-supply.conf` ?23:41
Wizzupyeah he won't have it23:41
siceloWizzup: ah :-p23:41
arno11lol23:42
siceloanyway, arno11's problem is deeper than that ... our changes, even with these 'accidents' ... are not touching the charger in the way observed now23:42
siceloarno11: please do `lsmod | grep isp17`23:43
arno11ok23:43
Wizzupbuilding new leste-config23:43
arno11it shows isp1704 charger23:44
sicelook, and this is also available in `/sys/class/power_supply`? can you show us output of its uevent file23:44
arno11there is no isp17 in /sys/class/power_supply23:45
siceloso that's why it's not charging23:46
sicelocheck dmesg ... should have some clues23:46
sicelohttps://paste.debian.net/1312558/  ... arno11's upower output for the benefit of anyone following along23:46
arno11ok (again charger worked few min before the upgrade)23:47
siceloyeah, something else happened, which i hope dmesg may answer23:48
arno11isp1704_charger: probe of isp1707 failed with error -11023:48
siceloisp1704 not loading successfully will break charging and usb networking23:49
arno11ok23:49
sicelodid you try a reboot btw? :-)23:50
arno11yes i tried23:51
siceloyou could also try `apt-get install --reinstall linux-image-omap` in case something happened to your kernel. no one has touched isp170x in ages23:51
arno11ok but ii didn't noticed any error, and everything else works fine23:52
sicelo:-)23:52
siceloexcept the charger ... which is controlled only by the kernel23:52
arno11ok weird stuff23:53
arno11currently reinstalling linux image omap...hope i have really 49% of battery lol23:56
siceloyou do23:57
siceloanyway what really matters is the voltage. it'll shutdown after you've reached 3.248V23:57
arno11yes i know :)23:59

Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!