| liux | hi, guys, my usb mode keeps on charging only when plugined in cable, cannot use mtp and developer mode | 04:27 |
|---|---|---|
| liux | log pasted here: https://paste.ubuntu.com/p/6hYN59WNNW/, | 04:36 |
| *** OhYash1 is now known as ohyash | 05:07 | |
| spiiroin | liux: in that paste: line 22 could be = some left over android usb scripts/stte interfere with usb moded; possibly leading to line 71 = mtp device does not get mounted/configured/activated & mtp daemon gives up; leading to line line 42 = usb moded gives up too | 07:21 |
| foobar2018 | hi | 07:22 |
| foobar2018 | does xiaomi Redmi Note 5 works with sailfish? | 07:23 |
| T42 | <birdzhang> If you mean vince, yes. | 07:25 |
| foobar2018 | @T42: ive found one but it has a snapdragon 636, will it run? | 07:29 |
| T42 | <birdzhang> nope, vince is Redmi note 5(India)/ Redmi 5 plus(Global), so your Redmi note 5 is Global version | 07:31 |
| T42 | <birdzhang> it's codename is whyred | 07:32 |
| foobar2018 | T42: which are "vince" specs? | 07:35 |
| T42 | <birdzhang> https://www.gsmarena.com/xiaomi_redmi_5_plus_(redmi_note_5)-8959.php | 07:37 |
| foobar2018 | cant find it in aliexpress | 07:38 |
| foobar2018 | T42: how should I look for it in aliexpress? | 07:39 |
| T42 | <birdzhang> sorry, idk. btw, T42 is the telegram bridge bot | 07:41 |
| foobar2018 | I see | 07:44 |
| r0kk3rz | man xiaomi has a confusing naming standard | 09:03 |
| T42 | <adampigg> It has a standard?? | 09:05 |
| r0kk3rz | maybe thats the problem :) | 09:06 |
| T42 | <adampigg> Monich, around? | 09:33 |
| Mister_Magister | monich: is small m are you typing from phone? | 09:34 |
| Mister_Magister | @adampigg now you are getting ahead in the game | 09:35 |
| T42 | <adampigg> Tablet | 09:36 |
| T42 | <adampigg> It will still highlight :P | 09:36 |
| T42 | <adampigg> Mal, i wondered, in your position inside the hull of the ship, can you influence the enablement of community FP sensor support?.;) | 09:50 |
| Mister_Magister | @adampigg even jolla have trouble using it lol | 09:51 |
| T42 | <adampigg> In that case best to hand it over to the community :) | 09:52 |
| deathmist | I still need help with debugging pulseaudio/ohmd and getting call audio & mic working on 15.1 cheeseburger (OnePlus 5) SFOS 3.0.3.9 :/ collectively I have all repos here https://git.io/fjKv8 pactl list https://pastebin.com/dcsmPYYB logs during call https://pastebin.com/eBbcRgW6 https://pastebin.com/tw5wvCAd and before calling https://pastebin.com/C | 09:57 |
| deathmist | ZPgxQpA https://pastebin.com/fM4DWskY | 09:57 |
| deathmist | why would "msm8974_platform" appear in logcats instead of "msm8998_platform" ? | 10:08 |
| Mister_Magister | snap801 as snap845 | 10:13 |
| Mister_Magister | quite upgrade | 10:13 |
| mal | @adampigg I will ask again if that could be moved to public, at least some parts as long as all are available | 10:16 |
| Mister_Magister | mal: binary blob would be okish just way of adapting it to ports | 10:16 |
| mal | Mister_Magister: that's what the latter part meant, the adaptation side might be better as open but the higher level can in theory be closed if needed | 10:19 |
| Mister_Magister | yeah | 10:22 |
| jusa | deathmist: could you give short reference what is not working and how? | 10:24 |
| T42 | <adampigg> Mal, gr8 | 10:25 |
| Mister_Magister | jusa: hi boi wanna fix some mics? | 10:26 |
| deathmist | jusa: when I place or receive a call I hear nothing from the other side and they only hear very low volume crackling from my side, audio works perfectly outside calls, headset, bt, speaker etc same for microphone(s) | 10:26 |
| deathmist | the random crackling is pretty much inaudible as I had to turn on the speaker at a high volume to hear it on the receiving end | 10:28 |
| r0kk3rz | nice! | 10:32 |
| Mister_Magister | r0kk3rz: WHATS NICE XD | 10:35 |
| Mister_Magister | oh sry caps | 10:35 |
| jusa | deathmist: could you take full pulseaudio logs from making a call? systemctl --user stop pulseaudio ; pulseaudio -vvvv -n --file=/etc/pulse/arm_droid_default.pa | 10:38 |
| jusa | also I guess you have audioflingerglue in use? | 10:38 |
| jusa | deathmist: and about the platform, msm8974 msm8226 msm8084 msm8992 msm8994 msm8996 msm8998 sdm845 all use msm8974 audio platform | 10:39 |
| jusa | Mister_Magister: provide all logs, state clearly what happens, make sure things work in aosp or whatever you are using as base | 10:41 |
| mal | piggz: Mister_Magister fyi https://github.com/sailfishos/droidmedia/pull/49 https://github.com/mer-hybris/audioflingerglue/pull/19 | 10:44 |
| Mister_Magister | mal: <3 | 10:46 |
| Mister_Magister | i haven't seen sledges here in loooooooooooooooooong time | 10:47 |
| deathmist | jusa: interestingly enough pressing the number pad during a call makes sounds on both phones; pa log https://hastebin.com/yiyorenaju.http journalctl https://hastebin.com/ihaxupiyar.sql logcat https://hastebin.com/kinoyifusi.sql all during ~50 second call | 10:56 |
| jusa | deathmist: yep, dtmf is not transmitted as audio but played locally | 10:57 |
| Mister_Magister | jusa: well you were fixing it already but only redirected mic so i thought maybe you could finnish fixing it. Maybe you will be able to remember i have phone in which in call in loudspeaker mode mics is working but in earspeaker mics aren't working. you redirected loudspeaker mic in mixer_paths but quality is you know how good. I'm wondering if there is something more you can do? | 10:58 |
| jusa | deathmist: do you have audioflingerglue (miniafservice) running? | 11:00 |
| deathmist | yep | 11:01 |
| Mister_Magister | jusa: something rings the bell? | 11:01 |
| jusa | deathmist: hm.. strange that there are no set/get_parameters calls to audioflinger | 11:02 |
| liux | spiiroin, The POWER_SUPPLY_PRESENT of usb power_supply is always 0 whenever usb plugin or not. usb_moded disconnect the connection then usb net and mtp failed | 11:04 |
| liux | I think I found the reason but cannot resolve it. have no idea why POWER_SUPPLY_PRESENT is always zero | 11:04 |
| deathmist | jusa: should I remove the symlinks for audio, mixer & media codecs from /etc to isolate the possibility of them causing issues? | 11:05 |
| mal | those really shouldn't break anything | 11:06 |
| Mister_Magister | jusa is ignoring me ;-; | 11:07 |
| mal | those are either used or not used, in latter case they won't cause issues and if they are used then they are needed | 11:07 |
| liux | who has met similar issue? it's "Fallback since cable detection might not be accurate. Will connect on any voltage on charger." in journal log | 11:12 |
| spiiroin | liux: iirc either present/online will do, but you might want to adjust usb-moded vebosity -> add "-Dl" to ExecStart at /lib/systemd/system/usb-moded.service | 11:13 |
| spiiroin | ... or if you can connect to device via wlan -> systemctl stop usb-moded ; usb_moded -TDl | 11:13 |
| Mister_Magister | now we only miss monich to the party :P | 11:16 |
| liux | spiiroin, yes, i've tried -D option. I'm sure the error comes from usb_moded got the value of POWER_SUPPLY_PRESENT 0 then disconnect. | 11:16 |
| * monich is on vacation | 11:16 | |
| Mister_Magister | monich: when r u comming back? | 11:17 |
| monich | what happened? | 11:17 |
| Mister_Magister | problem with ril | 11:17 |
| Mister_Magister | gib second | 11:17 |
| spiiroin | liux: and there is no online prop? what device is this? qualcomm, mtk, something else? | 11:17 |
| Mister_Magister | Incompatybile headset when registering network | 11:17 |
| Mister_Magister | monich: ^ | 11:17 |
| liux | spiiroin, qualcomm msm8960 | 11:18 |
| liux | POWER_SUPPLY_ONLINE is also 0 | 11:18 |
| jusa | deathmist: when the call is accepted none of the vsids changes state, the all stay in 1 (inactive), thus the routing is never updated... | 11:18 |
| Mister_Magister | monich: incompatible handset* | 11:18 |
| monich | Mister_Magister: logs? | 11:19 |
| spiiroin | liux: does it charge? properly? with both wall charger and from pc? | 11:19 |
| Mister_Magister | @adampigg u there | 11:19 |
| T42 | <adampigg> Yup | 11:19 |
| Mister_Magister | grab logs for monich | 11:19 |
| *** JustGerd7 is now known as JustGerd | 11:19 | |
| Mister_Magister | ofono logger app :P | 11:19 |
| spiiroin | liux: btw if you still have that -D log, can you pastebin/stte it? | 11:20 |
| deathmist | jusa: just found a bunch of missing mixer and audio related files in my current /vendor partition compared to stock /vendor, should I flash it over and see if the situation changes? | 11:20 |
| T42 | <adampigg> Cant atm, at work, and phone not booting, needs a boot flashed again | 11:20 |
| Mister_Magister | now we have everybody in the party :P | 11:20 |
| liux | spiiroin, yes it charges normal, from pc | 11:20 |
| T42 | <adampigg> Something in my build is broke, and it reboots after dhi | 11:21 |
| Mister_Magister | monich: soo you are on vacation? can ping u tomorrow? we can wait untill your vacation is over if you tell us when its over :P | 11:21 |
| Mister_Magister | also mal friendly reminder | 11:21 |
| T42 | <adampigg> Monich, ^^ | 11:22 |
| T42 | <adampigg> (Photo, 1920x2560) https://irc.thaodan.de/.imgstore/SDu8o01bNf.png | 11:22 |
| T42 | <adampigg> Will get logs later after fixing stuff | 11:22 |
| liux | spiiroin, log pasted here: https://paste.ubuntu.com/p/hYx6fJRnjc/ | 11:24 |
| monich | Mister_Magister: adampigg: T42: that "Incompatible handset" thing comes out of your sim cars | 11:26 |
| monich | via stk (sim toolkit) interface | 11:26 |
| Mister_Magister | so ril is fine just sim card doesn't like us? | 11:26 |
| monich | yes, that's how it looks to me | 11:27 |
| Mister_Magister | hmm maybe it's assigned to one phone or something | 11:27 |
| Mister_Magister | i will try with mine | 11:27 |
| monich | the sim in turn might have received that from the operator | 11:27 |
| Mister_Magister | that's what i'm thinking about | 11:27 |
| T42 | <adampigg> Ive used that sim for many years on different handsets | 11:28 |
| T42 | <adampigg> Odd | 11:28 |
| Mister_Magister | @adampigg do you have imei in settings? | 11:28 |
| T42 | <adampigg> Dunno, ill check ... Maybe they dont like the handset? | 11:28 |
| Mister_Magister | oh u can't check | 11:29 |
| Mister_Magister | @adampigg i will check with my sim and then we will think | 11:29 |
| T42 | <adampigg> Is it the operator saying the phone is incompatble, or the sim being incomlatible i wonder | 11:29 |
| Mister_Magister | the phone | 11:30 |
| monich | the operator might not like your imei/sim combination | 11:30 |
| monich | if it's a special plan for a specific phone | 11:30 |
| Mister_Magister | he said he used it on different phones so it's not then | 11:30 |
| T42 | <adampigg> It isnt, its off contract for many years | 11:30 |
| T42 | <adampigg> But, this is a pre prod phone | 11:31 |
| Mister_Magister | doubt that's the reason | 11:31 |
| jusa | deathmist: let's try this first, compile pulseaudio-modules-droid from here: (dev branch) https://github.com/jusa/pulseaudio-modules-droid/tree/dev | 11:31 |
| jusa | deathmist: then after installing and restarting pulseaudio, after the call is active call: paclt set-card-profile droid_card.primary voicecall-voicemmode1 | 11:32 |
| jusa | deathmist: and without typo: pactl set-card-profile droid_card.primary voicecall-voicemmode1 | 11:32 |
| deathmist | okay, will do | 11:36 |
| deathmist | jusa: it automatically chooses no on the conflict and doesn't let me build, or am I doing it wrong? https://pastebin.com/3G5MNj2y | 11:54 |
| jusa | deathmist: you shouldn't build pulseaudio-modules-droid-sbj | 11:55 |
| jusa | that's legacy for jolla 1 device | 11:55 |
| spiiroin | liux: it does indeed look like writing zero to /sys/class/android_usb/android0/enable would show up as disconnect (via udev) - which would be odd / make mode switch with existing code fail. but ... | 11:57 |
| deathmist | jusa: the package is not on my phone but it just builds it here, do I just remove the spec file for it? | 11:58 |
| spiiroin | liux: ... have you used the "ask from user" default mode? what happens if you use developer mode / mtp as default mode? (probably makes no difference, but I do not recall for sure) | 11:58 |
| spiiroin | liux: have you checked what the actual changes that are visible via udev are? maybe we're just misinterpreting something? | 12:01 |
| jusa | deathmist: I'm not sure how the build script works but if removing the spec helps then sure | 12:03 |
| jusa | but if the package is built you can also install the rpm by hand | 12:03 |
| deathmist | alright I installed it and audio came from the earpiece and mic worked after doing the command during a call! :D | 12:06 |
| jusa | deathmist: yeps, then it has something to do with audioflingerglue, as ril (iirc) should update the vsid during call | 12:08 |
| jusa | apparently module-droid-card-glue successfully registers as handler to miniafservice but still for some reason audioflinger set_parameters is not called | 12:13 |
| liux | spiiroin, if use developer mode as default, it's same disconnected in usb-moded | 12:23 |
| liux | spiiroin, how can i trace actual changes about udev? | 12:23 |
| deathmist | jusa: could it have something to do with me still using the Xperia XA2 RIL stuff https://git.io/fjKXU | 12:35 |
| jusa | deathmist: could be, what android is the base? | 12:35 |
| deathmist | 15.1 | 12:36 |
| jusa | is that android 8 or what? | 12:36 |
| deathmist | yes, I meant lineage-15.1 (which is 8.1) | 12:36 |
| jusa | yep.. hm.. can't help with ofono ril, mal? | 12:37 |
| spiiroin | liux: udev-monitor perhaps - but it is rather clumsy and verbose. another alternative could be to utilize mce battery/charger tracking ... | 12:40 |
| spiiroin | printf "[BatteryUDevPropertyBlacklist]\nfoo=bar\n" > /etc/mce/99-udev-hacking.ini ; systemctl stop mce ; mce -Tq -l modules/battery-udev.c:udevproperty_set | 12:40 |
| spiiroin | (just remember to remove that hacking.ini when done) | 12:40 |
| mal | jusa: deathmist those shouldn't affect call audio afaik | 12:40 |
| mal | of course toggling those to different values is worth a try | 12:41 |
| mal | deathmist: probably good idea to read what those do https://git.merproject.org/mer-core/ofono/blob/master/ofono/drivers/ril/ril_subscription.conf | 12:42 |
| liux | spiiroin, sorry i'm not familiar with these tools. let me try them.... | 12:43 |
| spiiroin | liux: to be specific, it is not a "tool". mce is a system daemon that a) happens to track udev power_supply devices (bit like what usb-moded does) and b) can be configured to do diagnostic loggin on all udev property changes -> can be used "as tool" for this purpose | 12:46 |
| spiiroin | getting back to normal state of affairs then requires: removing that custom ini-file and doing: systemctl start mce | 12:47 |
| KALUBE | I'm having issues with droid hal init, my device won't accept custom kernel cmdline paramaters so I have disabled selinux in the kernel config, I have masked droid-hal-init and stracing it produces the following output https://pastebin.com/uJRZ5X2f | 12:53 |
| KALUBE | Here is my kernel config https://github.com/calebccff/android_kernel_samsung_sharkls/blob/hybris-14.1/arch/arm/configs/j3xnlte_defconfig#L3444 | 12:53 |
| KALUBE | It seems to try and open /sys/fs/selinux which doesn't exist | 12:54 |
| liux | spiiroin, thanks for the instruction. perhaps i need some time to get it understand. btw, my colleague told me his qcom kernel has same issue. perhaps it's bug of kernel charger module? | 12:55 |
| mal | KALUBE: comment out this line https://github.com/mer-hybris/android_system_core/blob/hybris-15.1/init/property_service.cpp#L75 | 12:57 |
| mal | from system/core/init/ | 12:57 |
| mal | then run make hybris-hal again and after that "build_packages.sh -d" | 12:57 |
| KALUBE | Awesome, progress! Building now | 12:58 |
| spiiroin | liux: could be, or... in some devices charging/usb behavior depends on whether some android binary blob is running or not (say, fuelgauged / mtk) | 13:02 |
| liux | spiiroin, I've checked other qcom kernel on github there "power_supply_set_present" which does not exist in my aries msm8960 kernel, nor in other msm8960 kernel for other devices. but if it is indeed issue of binaries I can do nothing then | 13:09 |
| spiiroin | liux: I'm not saying that it is, just that there are other possibilities in addition to bugs. and if it is that some binary needs to be watching over stuff / something, you probably have that binary already on android side. then again - if it charges without issues, this would be unlikely | 13:17 |
| * spiiroin needs to go | 13:20 | |
| T42 | <elros34> liux: Last time I checked usb-moded needed POWER_SUPPLY_TYPE to correctly detect usb connection. I had to add dirty patch to my msm8960's pm8921-charger driver to avoid patching usb-moded on every release. use udevadm monitor -u -p to confirm that your kernel reports this property. | 13:25 |
| liux | elros34, my kernel does not report the property... may I take a look at your patch? I'm patching to charger driver but it seems not easy for me | 13:33 |
| T42 | <elros34> liux: https://github.com/elros34/android_kernel_motorola_msm8960-common/commit/d166d4acd96bf395d35f8400f13f1db772b9a44b. Also here is patch for usb-moded which I use before fixing kernel. It may not work with latest usb-moded: https://github.com/elros34/sfos-moto_msm8960_jbbl-patches/blob/b434456088a60e800b8bd706533a81800cda64d8/usb-m | 13:37 |
| T42 | oded.patch | 13:37 |
| KALUBE | droid-hal-init now starts! But I can't su nemo and other systemctl commands hang, heres systemctl status https://pastebin.com/6sERQE0K. I think it's the wait-for-file /run/user-session/user-init-done.flag | 13:42 |
| liux | elros34: Thank you very much. I think i can cherry-pick your patch to my kernel and make usb-moded work then | 13:44 |
| KALUBE | So start-user-session.service never gets started, because graphical.target never gets started, because multi-user.target never gets started which only requires basic.target which DOES get started, what could cause multi-user.target to never get started? | 15:59 |
| vknecht | KALUBE, maybe lib/systemd/system/multi-user.target.wants/initial-bootstate.service | 17:13 |
| KALUBE | vknecht: https://pastebin.com/h2uHQSBJ is multi-user.target being reached then? | 17:49 |
| KALUBE | It was started from system.slice so it's not the cause | 17:49 |
| vknecht | try "systemctl --failed" to see if some other failed | 17:51 |
| KALUBE | I think it could be systemd-logind https://pastebin.com/yGJ7YsD4 | 17:52 |
| KALUBE | Also oneshot-root.service is stuck on "activating" and doesn't ever start | 17:53 |
| KALUBE | Could it be causing stuff like systemctl reboot to hang? | 17:53 |
| vknecht | maybe check each failed with status, and also paste "journalctl -alb --no-pager" | 17:55 |
| piggz | mal: can I request an devel:idealte:t5 for myself, Mister_Magister and TheKit pls | 18:01 |
| Mister_Magister | piggz: i think it should be fxtec:t5 | 18:02 |
| piggz | not according to $DEVICE? | 18:02 |
| Mister_Magister | also mal don't forget bout devel:asus:5z | 18:02 |
| Mister_Magister | piggz: according to device | 18:02 |
| Mister_Magister | https://github.com/sailfish-on-fxtecpro1/droid-hal-t5/blob/master/droid-hal-t5.spec | 18:03 |
| piggz | well, whatever, codename or realy name, im not worried | 18:03 |
| Mister_Magister | i'm talking about vendor | 18:04 |
| KALUBE | vknecht: https://pastebin.com/UVQmVFPa, https://paste.opensuse.org/view//76801356 I've removed a bunch of spam to do with the battery in journalctl. | 18:11 |
| mal | piggz: Mister_Magister the vendor can be different in sfos and android sides, like I have for fp2 https://github.com/mlehtima/droid-hal-fp2-sibon/blob/master/droid-hal-fp2-sibon.spec | 18:13 |
| Mister_Magister | i am perfectly aware of that | 18:14 |
| Mister_Magister | and because phone is from company fxtec like jolla is from jolla | 18:14 |
| Mister_Magister | vendor should be fxtec | 18:14 |
| Mister_Magister | mal: so devel:fxtec:t5 please :) | 18:15 |
| mal | well sony nile has vendor=sony and rpm_vendor=qualcomm | 18:15 |
| T42 | <meierrom> Mister_Magister: Your 5z is booting again? | 18:37 |
| Mister_Magister | what do you mean again? | 18:38 |
| Mister_Magister | @meierrom | 18:42 |
| vknecht | KALUBE, not sure it will help but there's a typo persistant <-> persistent for journal storage setting | 18:43 |
| KALUBE | Yeah I know, journalctl defaults to persistent so it's fine. | 18:44 |
| KALUBE | I've tracked it down to oneshot-root.service calling ssu binary which hangs (for some reason??) which prevents boot from continuing, commenting out the call to ssu (as it's not useful) means user@100000.service starts, but it doesn't activate | 18:44 |
| Mister_Magister | mal: so fxtec please :P and asus aswell | 18:58 |
| T42 | <meierrom> Mister_Magister: I thought you bricked it yesterday or so. My mistake I guess... | 18:59 |
| Mister_Magister | it wasn't bricked at all lol | 19:00 |
| Mister_Magister | i mean i have no rndis but that ain't "bricked" | 19:00 |
| T42 | <meierrom> Mister_Magister: Cool then. Keep it up. :) | 19:02 |
| Mister_Magister | @meierrom i even forgot who donated it but i disappointed them ;-; | 19:02 |
| Mister_Magister | but notchfish will come | 19:03 |
| Mister_Magister | @meierrom are you interested in 5z? | 19:06 |
| piggz | mal: what triggers shutdown after startiing DHI? | 19:31 |
| Mister_Magister | selinux | 19:31 |
| piggz | Mister_Magister: i thought so, but in what way? | 19:33 |
| Mister_Magister | just disable it | 19:34 |
| Mister_Magister | i already disabled it iirc | 19:34 |
| piggz | it is | 19:34 |
| Mister_Magister | then dunno? | 19:34 |
| piggz | argh, what have i broken! | 19:37 |
| piggz | mal: expert tips? | 20:05 |
| piggz | if i mask dhi its fine | 20:05 |
| Mister_Magister | debug | 20:05 |
| Mister_Magister | mask dhi and strace it | 20:05 |
| piggz | ok | 20:05 |
| Mister_Magister | or not | 20:05 |
| Mister_Magister | get pstore | 20:05 |
| Mister_Magister | it should be there | 20:06 |
| KALUBE | Can anyone help with my systemd troubles? Any command that seems to require dbus/systemd (e.g ssu, systemctl reboot) seem to hang, user@100000.service is stuck on "activating", running ExecStartPre=/usr/lib/startup/validate-user %i. Kernel config: https://git.io/fjK7p | 20:07 |
| KALUBE | Systemctl status: https://pastebin.com/MJu82gia | 20:08 |
| KALUBE | Failed units: https://pastebin.com/pnMz6NH3 | 20:10 |
| mal | piggz: strace | 20:15 |
| Mister_Magister | HA! told ya! | 20:15 |
| Mister_Magister | i am expert now | 20:15 |
| piggz | mal: i have, but it doesnt look right | 20:15 |
| mal | KALUBE: are you sure kernel configs are ok? | 20:15 |
| KALUBE | Not 100%, currently flicking through but nothing stands out | 20:16 |
| piggz | mal: https://bpaste.net/show/okq3 but, thats not right as ti doesnt do anything ... some environment needed? | 20:16 |
| piggz | when i run it as a shellscrpt, it runs for a minue or so, then reboots | 20:16 |
| mal | piggz: maybe you need 64-bit strace | 20:17 |
| piggz | ah | 20:17 |
| mal | KALUBE: have you verifies the /proc/config.gz on device with mer kernel checker | 20:18 |
| mal | just to be sure it's really correct kernel | 20:19 |
| KALUBE | ma: Definitely correct kernel, of interest I have CONFIG_SECURITY_SELINUX_BOOTPARAM unset, CONFIG_AUDIT unset and CONFIG_SECURITY_YAMA_STACKED unset, no other relevant warnings | 20:22 |
| piggz | mal: well, that sucks, its not rebutting under strace | 20:26 |
| mal | piggz: it still reboots if you just run droid-hal-init from terminal? | 20:28 |
| piggz | mal: yes | 20:29 |
| mal | KALUBE: show dmesg | 20:29 |
| mal | and journalctl -b | 20:30 |
| Mister_Magister | mal: well that experimental audioflingerglue ain't even building ninja: error: '/home/build/android/out/target/common/obj/JAVA_LIBRARIES/WfdCommon_intermediates/classes.jar', needed by '/home/build/android/out/target/common/obj/PACKAGING/boot-jars-package-check_intermediates/stamp', missing and no known rule to make it | 20:35 |
| piggz | mal: when i change the .sh command to exec nohup (strace64) droid-hal-init ... it also reboots | 20:38 |
| piggz | after running from the command line | 20:38 |
| Mister_Magister | just run it manually | 20:39 |
| mal | Mister_Magister: did you minimize you build in any way? | 20:39 |
| Mister_Magister | strace /sbin/droid-hal-init | 20:39 |
| Mister_Magister | mal: nope | 20:39 |
| Mister_Magister | wait wrong device tree | 20:39 |
| KALUBE | Mister_Magister: Does it reboot to recovery? | 20:40 |
| Mister_Magister | reboot what | 20:40 |
| Mister_Magister | nothing reboots | 20:40 |
| KALUBE | sorry, meant piggz, my droid-hal-init was causing reboots and selinux was the cuprit | 20:41 |
| Mister_Magister | selinux is disabled tho probably | 20:41 |
| Mister_Magister | piggz: can't you just reflash it | 20:41 |
| piggz | mal: the whole rootfs? | 20:41 |
| piggz | i suppose | 20:42 |
| Mister_Magister | yeah | 20:42 |
| Mister_Magister | y do u guys ping wrong people | 20:42 |
| mal | piggz: ? | 20:42 |
| Mister_Magister | mal: he wanted to ping me | 20:43 |
| Mister_Magister | but it's late and everybody ping wrong people | 20:43 |
| piggz | https://bpaste.net/show/QE7v | 20:43 |
| Mister_Magister | just in case no CONFIG_SECURITY_SELINUX_BOOTPARAM_VALUE=1 doesn't matter | 20:44 |
| piggz | mal: i ran strace droid-hal-init from cmdline, and it didnt reboot .... i edited the droid-hal-startup.sh to add an strace, and it still rebooted | 20:44 |
| Mister_Magister | piggz: but u masked dhi right | 20:44 |
| Mister_Magister | when i mask dhi on pro1 i have rndis but on 5z i don't ;-; | 20:44 |
| piggz | Mister_Magister: yes, i ran it using droid-hal-startup.sh | 20:44 |
| Mister_Magister | check pstore fnally please | 20:45 |
| Mister_Magister | cat /sys/fs/pstore/console-ramoops | 20:45 |
| piggz | we dont have pstore atm | 20:45 |
| Mister_Magister | we do | 20:45 |
| piggz | well, its empty | 20:45 |
| Mister_Magister | it's enabled in kernel | 20:46 |
| piggz | sh-3.2# ls -lh /sys/fs/pstore/ | 20:46 |
| piggz | total 0 | 20:46 |
| Mister_Magister | and there is systemd service | 20:46 |
| Mister_Magister | hm | 20:46 |
| KALUBE | mal: dmesg: https://pastebin.com/edsJtCGJ and journalctl -b: http://paste.opensuse.org/view//18743646. I've removed spam from sm5701 and power supply, any way to prevent them from spamming? | 20:46 |
| Mister_Magister | THE HELL building audioflingerglue builds 120k objects | 20:59 |
| mal | KALUBE: that cannot be journalctl -b, reboot and get new log | 21:05 |
| mal | piggz: is that the same device you already had running at some point? | 21:07 |
| mal | piggz: Mister_Magister pstore is only filled if kernel crashes afaik, not if droid-hal-init reboots the device | 21:07 |
| Mister_Magister | okay | 21:08 |
| Mister_Magister | mal: yes it is | 21:08 |
| mal | really? I thought it's just a crash log | 21:09 |
| Mister_Magister | i mean | 21:10 |
| Mister_Magister | it's same device | 21:10 |
| mal | ok | 21:10 |
| KALUBE | mal: this log is as soon as possible after I get telnet http://paste.opensuse.org/view//35602616 | 21:12 |
| piggz | mal: yes, same device | 21:17 |
| piggz | flashed rootfs and same bahaviour | 21:17 |
| piggz | lets see what logcat says | 21:18 |
| piggz | hmmmm | 21:21 |
| T42 | <meierrom> Mister_Magister: I don't think you disappointed anybody so far. Unless you decided to give up on porting 5z? | 21:21 |
| mal | KALUBE: I see nothing really bad, except some probably unrelated chmod errors | 21:25 |
| piggz | rats, thought I had it.... | 21:27 |
| KALUBE | I really don't know what could be causing it. I've heard that if a process doesn't fork that can make systemctl unresponsive, or if a process that doesn't fork doesn't exit for some reason | 21:28 |
| KALUBE | I'm not sure how to track it down, if that is the case | 21:28 |
| piggz | mal: Mister_Magister: thought i had it, failing to load selinux_stubs due to ordering service problem ... but even after fixing htat it still reboots | 21:37 |
| piggz | im calling it anight! | 21:38 |
| mal | KALUBE: try running "journalctl -b -f > log.txt" to grab the log for longer time, not very long but maybe a minute or two | 21:38 |
| mal | that way the log file should hopefully have all | 21:39 |
| liux | I've set dataCallFormat=11 in ril_subscription.conf. the sms sending/receving does not work sometimes. Should I set to other value? | 23:59 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
