T42 | <edp_17> @edp_17 [Hi guys, Can somebody help me out please? On S โฆ], Okay, I have found this. All services have moved from /lib/systemd to /usr/lib/systemd. | 00:31 |
---|---|---|
mal | @edp_17 no need to downgrade sensorfw for 4.0.1, only 3.4.0 needed the downgrade, for broadcom-bluetooth https://github.com/mer-hybris/broadcom-bluetooth/pull/2 it seems I forgot to merge that | 00:45 |
mal | assuming you want to use bluez5 of course, not sure which kernel your device has | 00:46 |
T42 | <edp_17> I have kernel 3.10.9 (samsung_exynos5433) | 00:47 |
mal | did you have bluez4 or bluez5 before? | 00:48 |
T42 | <edp_17> Thanks for the others. If sensorfw shouldn't be downgraded, I don't know why it didn't work out of the box.. | 00:48 |
mal | what do logs say? | 00:49 |
T42 | <edp_17> I think have bluez5. | 00:49 |
mal | how did you build the bluetooth things before then? | 00:50 |
mal | which android base is that? | 00:50 |
T42 | <edp_17> LOS 14.1. | 00:50 |
T42 | <edp_17> I've manually built broadcom-bluetooth and bluetooth-rfkill-event, then installed them on device and that worked. | 00:51 |
mal | ok | 00:52 |
mal | maybe you then should manually install those to the target also | 00:52 |
T42 | <edp_17> Yeah, but when I installed them on device bt doesn't work. | 00:53 |
T42 | <edp_17> I used this command to rebuld: rpm/dhd/helpers/build_packages.sh -D -o -b hybris/mw/broadcom-bluetooth -s rpm/broadcom-bluetooth.spec | 00:53 |
mal | ok, maybe it needs some more debugging later then | 00:53 |
T42 | <edp_17> Okay. | 00:54 |
T42 | <edp_17> I'll try with this: rpm/dhd/helpers/build_packages.sh --mw=https://github.com/mer-hybris/broadcom-bluetooth -s rpm/broadcom-bluetooth.spec | 00:54 |
T42 | <edp_17> And similarly the bluetooth-rfkill-event. | 00:55 |
mal | in case you don't know you can manually install or removing things using sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -m sdk-install -R zypper | 00:56 |
T42 | <edp_17> Sorry, now I remember what was the problem with BT. I've tried to rebuild broadcom-bluetooth and bluetooth-rfkill-event but got these errors: https://paste.ubuntu.com/p/DZ8BxyfNrQ/ | 00:57 |
T42 | <edp_17> Thank you for mentioning. Although, I used that sb2 -t for unstalling/removing stuff from target. | 00:59 |
meowdude | hello? | 01:35 |
meowdude | Finally I can connect. I've been getting "Error <creator>[02/09 01:32:15] : Command 'modprobe' is not available." - I saw in the logs someone else had this issue but didn't see how they fixed it. has anyone else run into this? I'm proper stumped on this one. | 01:36 |
meowdude | I tried "echo $PATH" and peaked around the directories for a modprobe folder, no luck. In my regular linux terminal modprobe works fine but in the platform sdk it's not recognized | 01:37 |
meowdude | I zypper installed kmod, then ran "alias modprobe=kmod" and now things seem to be rolling. whether mic will actually successfully create an image? no idea. | 02:17 |
meowdude | Its crazy to think i was like 15 when I was filling the IRC with bloat questions. I hope now I will be less of a nuisaince. | 02:19 |
mighty17 | <mal "mighty17: just curious, do you d"> I didn't understand the question mal if build_packages.sh does it yes, I do nothing extra from the hadk | 04:06 |
Mister_Magister | mal: same thing | 07:18 |
mighty17 | <mighty17 "@elros34 ermm we have a problem "> Well @elros34 any ideas what to do? | 07:44 |
T42 | <edp_17> mal: I've updated the sensorfw to the latest and now all sensors do work on the device. | 09:22 |
T42 | Marti %lastname% was added by: Marti %lastname% | 09:30 |
T42 | <elros34> @might17: What error do you get? | 09:31 |
T42 | <edp_17> There are a few. ๐ | 09:31 |
T42 | <edp_17> I've refreshed the mer-kernel-check and it shows a few unset options: CONFIG_QUOTACTL, CONFIG_QFMT_V2, CONFIG_QUOTA and CONFIG_QUOTA_NETLINK_INTERFACE. | 09:33 |
T42 | <edp_17> Do you think any of this can be a reason why the sailjail didn't work? | 09:33 |
T42 | <edp_17> Oh, and because of the dependencies I also needed to add these two: CONFIG_XFS_QUOTA and CONFIG_XFS_FS | 09:44 |
T42 | <edp_17> I am re-building the hybris-boot and then retry the sandboxing on the device with a jolla app. | 09:44 |
mighty17 | <T42 "<elros34> @might17: What error d"> @elros34 i cant find any error related to ones i got before, no dsscomp issue nor -13 or -14 https://pastebin.ubuntu.com/p/Db5RwxsP6w/ | 10:02 |
mighty17 | https://pastebin.ubuntu.com/p/VypxswtJYz/ logcat | 10:02 |
T42 | <elros34> these logs doesn't even have moment when lipstick or anything graphics related starts | 10:04 |
T42 | <elros34> any chance you msaked user@100000 or logs ar just incomplete | 10:05 |
T42 | <elros34> masked* | 10:05 |
T42 | <edp_17> @elros34 : Nope, those flags didn't solve the sailjail issue, I am still getting the error and the apps do not start: https://paste.ubuntu.com/p/xkcj4kSnY3/ | 10:28 |
T42 | <edp_17> Although, the mer-config-check now doesn't return with error. | 10:28 |
mighty17 | <T42 "<elros34> any chance you msaked "> Nope well I will reflash and check maybe a clean flash will fix issues | 10:32 |
mighty17 | Also usb and Bluetooth don't work and touch is broken only in the browser | 10:32 |
T42 | <elros34> @edp_17 I guess you have that patch in kernel otherwise systemd would probably fail but confirm it: https://github.com/mer-hybris/hadk-faq/blob/master/README.rst#kernel-changes-needed-for-updated-systemd-in-sailfish-2-1-1-x | 10:45 |
T42 | <elros34> also stracingv sailjail might provide some point what is wrong | 10:48 |
mighty17 | okay bt might be fixed by hadk-faq, usb seems to be issue | 10:49 |
T42 | <edp_17> @elros34 [@edp_17 I guess you have that patch in kernel โฆ], Well, I do not have this in the kernel. I don't even have tid_fd_revalidate() and *proc_fd_instantiate() in there. | 10:51 |
T42 | <elros34> Probably only needed on older kernel | 10:53 |
mighty17 | you guys working on sfos4.0? | 10:53 |
T42 | <edp_17> @elros34 [Probably only needed on older kernel], I think so as systemd is happy on it. Strace of the jolla-calendar: https://paste.ubuntu.com/p/7TYgkXskJv/ | 10:54 |
T42 | <edp_17> I don't have /etc/sailjail.conf and some other stuff, is it problem? | 10:55 |
T42 | <edp_17> And there are some permission problems. | 10:58 |
T42 | <elros34> looks like strace can't be used that way with sailjail | 11:22 |
T42 | <edp_17> mal mentioned this could be a missing option in the kernel but I don't know which one. I only have warnings now in mer-kernel-check, so I am not sure. | 11:26 |
T42 | <edp_17> I thougth those flags will fix this but seemingly I was wrong. | 11:26 |
T42 | <elros34> quota according to README is for monitoring user's disk usage | 11:27 |
T42 | <edp_17> Hmm, why is it mandatory in the kernel then? Anyway, that's not an issue, so don't waste time on that. | 11:28 |
mal | @edp_17 can you grab the kernel config from the device and pastebin it, "zcat /proc/config.gz" | 11:29 |
T42 | <edp_17> Sure. Here it is: https://paste.ubuntu.com/p/Cd6387BYsg/ | 11:30 |
T42 | <elros34> sailjail have verbose mode enabled with -vv. You could try it. It fails at https://github.com/netblue30/firejail/blob/ce7dd15609d5ab9016c6dbf3fdf8833dcf12d4f2/src/firejail/fs.c#L504 | 11:42 |
T42 | <edp_17> @elros34 : sailjal with verbose log: https://paste.ubuntu.com/p/qdKFT9sXpM/ | 11:50 |
T42 | <elros34> -vv is more verbose than single -v | 11:56 |
T42 | <edp_17> @elros34 : Here it is: https://paste.ubuntu.com/p/53fVC4wdfB/ | 12:03 |
mal | Mister_Magister: did you try the tracing env vars with the new libhybris? | 12:13 |
Mister_Magister | i always use them | 12:13 |
Mister_Magister | HYBRIS_TRACE=1 HYBRIS_TRACE_UNHOOKED=1 HYBRIS_TRACE_HOOKED=1 EGL_PLATFORM=hwcomposer QT_QPA_PLATFORM=hwcomposer test_hwcomposer | 12:13 |
mal | and you got no new output? | 12:18 |
Mister_Magister | nope | 12:23 |
Mister_Magister | just the GRALLOC TEST: 0xab19a0 (nil) | 12:23 |
T42 | <elros34> @edp_17 so if I get it right it for some reason it fails at etc mount: https://github.com/netblue30/firejail/blob/ce7dd15609d5ab9016c6dbf3fdf8833dcf12d4f2/src/firejail/fs.c#L808 | 12:40 |
T42 | <edp_17> Why do you think it fails at the /etc? | 12:44 |
T42 | <edp_17> I don't have /lib64, /lib32 and /libx32 | 12:46 |
T42 | <elros34> dunno why. lack of these dirs is normal | 12:47 |
T42 | <edp_17> Okay. The /etc is a directory so I am not sure why is it trying to mount it. | 12:50 |
mighty17 | umm guys is shutdown bugged in sfos? like i clicked on it and it even said goodbye but my battery drained from 80 to 60 plus device was warm to touch | 12:52 |
T42 | <elros34> @edp_17 I don't know the details but probably to isolate and limit access to filesystem | 12:58 |
T42 | <edp_17> @elros34 : The problem is that if we don't find a solution to this firejail thingy, that means to an end of ports for lots of old devices. As a solution, I amended all .desktop files and removed the firejail from it. This way apps can be started from the app drawer but the internal calls are still use firejail, so those don't start the | 13:28 |
T42 | app. (Like the weather app. It can be launched from drawer but doesn't start when I call it with taping on the weather on events screen.) | 13:28 |
T42 | <edp_17> I'll leave this as is now and move on to the bluetooth issue. | 13:28 |
T42 | <elros34> so far only 1 device:) And if you want to fix it you could check who contributes to sailjail and ping him | 13:31 |
T42 | <edp_17> ๐ ๐ | 13:32 |
T42 | <elros34> @edp_17 I guess weather app started from events screen use dbus. In /usr/share/dbus-1/services/ you can find sailjail too | 13:41 |
T42 | <edp_17> Yes. I think it would be better to find a solution rather than trying to remove. ๐ | 13:43 |
T42 | <linusdan> mighty17: Congratulations on the Galaxy Tab 2 port! | 20:22 |
T42 | <Akatsu39> Gratz | 20:53 |
T42 | เบธ %lastname% was added by: เบธ %lastname% | 21:12 |
Mister_Magister | mal: got nothing new to test? | 21:57 |
mal | not right now | 22:05 |
*** Richie is now known as Guest53053 | 22:49 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!