| TAMplus39_ | Hi, Im trying to fix the autobrightness toggle but i dont have $ANDROID_ROOT/hybris/droid-configs/sparse/usr/share/csd/settings.d | 02:25 |
|---|---|---|
| *** zhxt_ is now known as zhxt___ | 02:56 | |
| MeowDude | TAMplus39_ : just create the file yourself and fill in the blanks using device specific configurations | 05:01 |
| MeowDude | just borrow a template file from a nother device's github | 05:01 |
| MeowDude | I reccomend an up to date device such as the OnePlusX or something, one that is up to date Sailfish OS wise | 05:01 |
| MeowDude | also seriously, no one has any tips with RILD not starting? | 05:01 |
| MeowDude | !seen r0kk3rz | 05:07 |
| merbot | MeowDude: r0kk3rz was last seen in #sailfishos-porters 18 hours, 28 minutes, and 27 seconds ago: <r0kk3rz> yeah sensor.qcom is a custom service, so make sure its running and check logcat | 05:07 |
| AndroUser2 | Hi. There are CyanogenMod 13.0 and LineageOS 14.1 for my device (Xiaomi Redmi 3). Which ROM is better for HADK? | 09:01 |
| r0kk3rz | which is better supported for the device? | 09:04 |
| r0kk3rz | either of those should be fine for the HADK | 09:04 |
| AndroUser2 | Lineage is better supported. Thanks. | 09:06 |
| T4 | <abhishek_0> AndroUser2 try 14.1 , but i think sailfish for your device already exists | 09:17 |
| AndroUser2 | I saw Sailfish for Redmi Note 3 (not Redmi 3). On Mer Wiki there are no Redmi 3. | 09:19 |
| r0kk3rz | whats its codename? | 09:20 |
| AndroUser2 | ido | 09:20 |
| r0kk3rz | doesnt sound familiar, but its probably not that different to the other xiaomi devices we have | 09:21 |
| r0kk3rz | most are well working | 09:21 |
| T4 | <abhishek_0> AndroUser2 redmi note 3 is kenzo | 09:22 |
| T4 | <abhishek_0> not ido | 09:22 |
| T4 | <abhishek_0> https://wiki.merproject.org/wiki/Adaptations/libhybris/Install_SailfishOS_for_kenzo | 09:23 |
| AndroUser2 | I said about Redmi 3. It is ido. Redmi Note 3 is kenzo. I know it :) | 09:23 |
| T4 | <abhishek_0> oh man .. these names... grrrrr | 09:25 |
| T4 | <abhishek_0> *facepalm* | 09:26 |
| r0kk3rz | its not as bad as when xiaomi reuses codenames for different devices | 09:28 |
| r0kk3rz | that gets more confusing | 09:28 |
| AndroUser2 | Oh... New Xiaomi devices has a very terrible naming. Redmi 5, Redmi 5A, Redmi Note 5, Redmi Note 5 Pro, Redmi 5 Plus... OMG | 09:29 |
| tbr | they always had a terrible naming convention | 09:32 |
| tbr | one more reason to just go by code names | 09:34 |
| MeowDude | hello? | 11:10 |
| MeowDude | r0kk3rz: hey I need help debugging rild | 11:11 |
| MeowDude | based on dmesg, it starts up, spawns /dev/socket/rild, but then for whatever quits and restarts endlessly | 11:11 |
| MeowDude | also it despawns /dev/socket/rild each time too | 11:12 |
| MeowDude | anybody? | 11:15 |
| mal | MeowDude: most likely something to do with either firmware not being found or some other android side service failing (like some inter-process communitation service) | 11:18 |
| MeowDude | *peeps* | 11:18 |
| MeowDude | *same IP* | 11:18 |
| MeowDude | ? | 11:18 |
| MeowDude | mal: how do I debug or trace it? | 11:18 |
| MeowDude | I tried strace -p PID but that didn't work | 11:18 |
| TuringFish | MeowDude: CDU? | 11:19 |
| MeowDude | TuringFish: yes. huh, that's neat. | 11:19 |
| mal | MeowDude: you need to look in logcat (and maybe dmesg), you can try to start rild manually with strace but make sure it runs with correct user so there are no permission issues | 11:20 |
| MeowDude | mal: open("/dev/block/platform/omap/omap_hsmmc.0/by-name/radio", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory) --- I think this might be it, because that file path is incorrect | 11:23 |
| MeowDude | its /dev/block/platform/omap_hsmmc.0/radio, not /dev/block/platform/omap/omap_hsmmc.0/radio | 11:23 |
| mal | MeowDude: show your /lib/udev/rules.d/998-droid-system.rules | 11:24 |
| MeowDude | mal: I just rebooted after editing fstab.tuna, gimme a sec | 11:25 |
| MeowDude | oof I didn't mean to ping you with that *facepalm* | 11:25 |
| mal | MeowDude: you have fstab on your device? | 11:25 |
| MeowDude | yes | 11:25 |
| MeowDude | I added itmyself... which... I actually forgot that it was extra... | 11:26 |
| mal | sailfish doesn't use fstab, at least it normally shouldn't | 11:26 |
| MeowDude | it's supposedly imported by /init.rc, which is what logcat said, so I just satisifed a dependancy. I figured mer mounted everything without it anyway | 11:27 |
| mal | it was removed for a reason | 11:27 |
| mal | systemd handles mounts | 11:27 |
| MeowDude | mal: https://pastebin.com/wXWfdGWV | 11:28 |
| MeowDude | can I try manually editting systemd? | 11:28 |
| MeowDude | erm, the mount script? | 11:28 |
| mal | krnlyng: some issue with the new udev rules ^ | 11:29 |
| MeowDude | is it a known issue, and if so, follow up question, is there a known fix? | 11:29 |
| mal | MeowDude: pastebin ls -lR /dev/block/ | 11:30 |
| mal | fix is to figure out what rule breaks | 11:30 |
| krnlyng | MeowDude, did this appear now or was it always like this on this device? | 11:30 |
| krnlyng | i.e. did you have success with that before the new udev rules? | 11:30 |
| MeowDude | mal: https://pastebin.com/aRk5qmXF | 11:31 |
| MeowDude | krnlyng: this is a new port device, the first version it has ever ran is 2.2.0.29 | 11:31 |
| krnlyng | i think i know what the issue is. but i need to know if it has always been an issue or if it is new | 11:31 |
| krnlyng | MeowDude, ok so probably always been like that | 11:32 |
| MeowDude | what do I do assuming it is new? | 11:32 |
| MeowDude | is there any use trying out a random patch you know? | 11:32 |
| mal | MeowDude: it's just a matter of fixing that file I asked, shouldn't be too difficult | 11:33 |
| MeowDude | oh great! | 11:34 |
| MeowDude | let me peep through it to see if I can personally find anything wrong with it | 11:34 |
| mal | the file looks quite complex if you are not used to udev rules | 11:35 |
| MeowDude | yeah, basic C/C++ I can handle, but this? nah this is mostly gibberish. I didn't see any actual radio/modem related lines though, only one radio line and it was for logs not the actual radio | 11:35 |
| krnlyng | MeowDude, this has nothing to do with radio, it's partition symlinks | 11:36 |
| mal | MeowDude: it doesn't have anything that specific, it's just generic rules for handling everything | 11:36 |
| MeowDude | oh. that makes more sense. | 11:36 |
| MeowDude | well then the symlinks for the partitions must be goofed. oh wait we already know that sorry... | 11:37 |
| krnlyng | MeowDude, ls -l /sys/devices/platform/omap | 11:37 |
| krnlyng | MeowDude, ls -l /sys/devices/platform/ | 11:37 |
| krnlyng | both please | 11:37 |
| mal | MeowDude: these kinds of issue happen occasionally, you are the first one after the new udev rules though | 11:38 |
| MeowDude | krnlyng: https://pastebin.com/5GZtefGR | 11:38 |
| MeowDude | both in the same paste seperated by a line | 11:38 |
| MeowDude | odd, when I run "ls" inside /dev/block/platform, the "omap" folder doesn't show up | 11:40 |
| krnlyng | MeowDude, "udevadm info -a /sys/devices/omap" "udevadm info -a /sys/devices/omap/omap_hsmmc.0" | 11:42 |
| MeowDude | krnlyng: https://pastebin.com/irdVXuM5 | 11:43 |
| MeowDude | didn't really work | 11:43 |
| krnlyng | MeowDude, same with platform | 11:44 |
| MeowDude | okay | 11:44 |
| MeowDude | krnlyng: https://pastebin.com/p9DCAavZ | 11:45 |
| MeowDude | same exact result | 11:45 |
| krnlyng | MeowDude, "udevadm info -a /sys/devices/platform/omap" "udevadm info -a /sys/devices/platform/omap/omap_hsmmc.0" | 11:45 |
| MeowDude | ha! now we are getting somwhere, give me a second | 11:46 |
| MeowDude | krnlyng: https://pastebin.com/rHzbFQrk | 11:46 |
| krnlyng | MeowDude, you can use that info to check what's wrong with the udev rules | 11:47 |
| MeowDude | krnlyng: so just fill in the missing information? | 11:51 |
| MeowDude | hmm it didn't work. | 11:55 |
| MeowDude | mal: I am trying to make sense of these udev rules and I think i am KIND of getting them, basically they allow you to universally deal with multiple device file systems and blocks without needing to make specific device files | 12:03 |
| MeowDude | so it would seem that by default this udev points to /dev/block/platform/FOLDER/DEVICE/* but mine doesn't do that, it's more like /dev/block/platform/DEVICE/* | 12:03 |
| r0kk3rz | udev is well documented, i suggest you read some of them | 12:03 |
| MeowDude | alright | 12:04 |
| r0kk3rz | but yes, its an abstraction layer for dealing with many different devices in a generic way | 12:04 |
| T4 | <abhishek_0> is lbt up ? | 12:05 |
| krnlyng | the rules for these partitions are a bit complicated, but if you have questions about a specific one ask | 12:05 |
| MeowDude | I just need to figure out how to make it stop pointing at /dev/block/platform/omap/omap_hsmmc.0 and start pointing at /dev/block/omap_hsmmc.0 | 12:05 |
| krnlyng | no, you need to change the udev rules to create the /dev/block/platform/omap/omap_hsmmc.0/by-name/radio symlink | 12:06 |
| MeowDude | I don't even know how to begin... so basically symlink /dev/block/platform/omap_hsmmc.0/by-name/radio to /dev/block/platform/omap/omap_hsmmc.0/by-name/radio? | 12:07 |
| mal | MeowDude: you don't want to specifically symlink only one partition, but all partitions | 12:10 |
| MeowDude | *scratches head* How do I do that? I started writing a subsystem line at the top... | 12:11 |
| MeowDude | I assume I am going in the wrong direction | 12:11 |
| lbt | nemo:devel:hw:wingtech:wt88047 is done abhishek_0 | 12:12 |
| MeowDude | mal: I added this line: SUBSYSTEM=="linkmain", KERNEL!="", RUN+="/bin/ln -s /dev/block/platform/$env{PLATFORM_FOLDER}/$env{PLATFORM_DEVICE} /dev/block/platform/omap_hsmmc.0" | 12:15 |
| mal | MeowDude: you misunderstood completely what to do | 12:16 |
| MeowDude | symlink the two directories | 12:17 |
| MeowDude | so it reroutes any activity to the wrong one to the right one | 12:17 |
| mal | MeowDude: that's not how it's done, check the last few lines in that rules file | 12:17 |
| mal | the "Create the partition symlinks" part | 12:17 |
| T4 | <abhishek_0> sadly no third party app work on 2.2.0.29 :( | 12:20 |
| r0kk3rz | ? | 12:20 |
| MeowDude | mal: how does this look: https://pastebin.com/HJEX7Z3z | 12:20 |
| T4 | <abhishek_0> like telegram desktop, ytplayer.. jollagram.. and others | 12:21 |
| mal | MeowDude: what line did you edit? | 12:22 |
| r0kk3rz | all my apps work at least, maybe you're just unlucky | 12:22 |
| MeowDude | mal: the last ones, I removed $env{PLATFORM_FOLDER} from each one since that is the one that added "omap" to it, which was plain incorrect. sorry I should have specified | 12:23 |
| mal | MeowDude: am I missing something, I thought you needed to add something not remove? | 12:24 |
| MeowDude | I just fixed the existing symlinks | 12:24 |
| mal | I don't understand the fix, your path were missing omap from the middle and how did you add it there? | 12:25 |
| T4 | <abhishek_0> r0kk3rz being that dank is injurious to health XD | 12:25 |
| MeowDude | they weren't missing it | 12:25 |
| mal | 14:23 < MeowDude> mal: open("/dev/block/platform/omap/omap_hsmmc.0/by-name/radio", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory) | 12:26 |
| r0kk3rz | abhishek: fite me :P | 12:26 |
| mal | your paths were /dev/block/platform/omap_hsmmc.0/by-name which means it was missing omap | 12:26 |
| MeowDude | /platform/omap/omap_hsmmc.0/* is the WRONG path, it was trying to use this one. the "omap" in the path is wrong it leads to a different directory | 12:26 |
| MeowDude | so i just told it to look in the right one instead of symlinking the right one to the wrong one (not that I'd know how to do that anyway) | 12:27 |
| mal | MeowDude: I don't understand, wtf is doing that open commmand? | 12:27 |
| T4 | <abhishek_0> r0kk3rz qt5 dependencies i have while installing them from storeman | 12:27 |
| MeowDude | strace of /system/bin/rild | 12:27 |
| mal | MeowDude: so you need to have that path on your system then, which means the paths are missing omap | 12:28 |
| mal | I give up | 12:28 |
| MeowDude | I think you are just misunderstanding my issue | 12:28 |
| MeowDude | but thanks for all the help mal really | 12:28 |
| MeowDude | I do appreciate it | 12:28 |
| T4 | <abhishek_0> mal am i dreaming ? | 12:28 |
| mal | MeowDude: you are the one misunderstanding | 12:28 |
| MeowDude | probabbly | 12:28 |
| MeowDude | but I just can't wrap my head around udev rules and how to symlink them | 12:29 |
| mal | MeowDude: rild tries to open /dev/block/platform/omap/omap_hsmmc.0/by-name/radio but you have /dev/block/platform/omap_hsmmc.0/by-name/radio | 12:29 |
| MeowDude | usually I would just run ln -s /dev/block/platform/omap/omap_hsmmc.0/by-name/ /dev/block/platform/omap_hsmmc.0/by-name/ | 12:29 |
| MeowDude | yeah that's it! | 12:29 |
| MeowDude | by removing that PLATFORM_FOLDER part from the sought out file, it removes the "omap" in the path | 12:29 |
| mal | MeowDude: how in the hell can it be so difficult to understand the udev rules | 12:29 |
| MeowDude | they look incredibly verbose and I don't understand the signage. | 12:30 |
| mal | MeowDude: just add the damn omap to the symlink rules in the rule file | 12:30 |
| mal | how can you not understand that | 12:30 |
| mal | MeowDude: removes what | 12:30 |
| MeowDude | hmm it's like teaching a guy what "cout" and "int" and all that means without explaining why they have to add "<<" and "()" and "{" symbols and stuff, I only understand about 40% of the contents of the file | 12:30 |
| mal | MeowDude: I'm totally lost what you mean | 12:30 |
| MeowDude | I am a visual learner mal, I can't see this and I don't know where to add the "omap" | 12:31 |
| mal | I assume rild has hardcoded /dev/block/platform/omap/omap_hsmmc.0/by-name/radio which means you need to make sure udev rules create that | 12:31 |
| MeowDude | okay so i know what "PLATFORM_FOLDER" and stuff mean and how to work with those, great, but stuff like "why does it use += and {}, where and why do I use these" is something that is bothereing me | 12:31 |
| MeowDude | okay so tell udev rules to symlink the real directory with the one rild points too | 12:32 |
| mal | MeowDude: if your current udev rules create /dev/block/platform/omap_hsmmc.0/by-name/ but you need /dev/block/platform/omap/omap_hsmmc.0/by-name/, how can you not see where to add it? | 12:32 |
| MeowDude | that's what you want me to do right? | 12:32 |
| MeowDude | the current ones make /dev/block/omap/omap_hsmmc.0/by-name, I want to REMOVE the omap | 12:32 |
| MeowDude | not add it | 12:32 |
| mal | just add omap to the lines 52-54 | 12:32 |
| MeowDude | I'll try that | 12:32 |
| mal | but that doesn't match what rild expects | 12:33 |
| mal | do I really have to give you the rules? | 12:33 |
| MeowDude | wait I see what you mean | 12:33 |
| MeowDude | I can see the lines and how omap is missing from them | 12:33 |
| mal | MeowDude: where did you see /dev/block/omap/omap_hsmmc.0/by-name ? that's not in any of the pastes you gave | 12:34 |
| MeowDude | that was the folder rild was trying to access, the only issue is that that folder does not exist | 12:35 |
| mal | MeowDude: the issue is probably that $env{PLATFORM_FOLDER} is empty | 12:35 |
| mal | MeowDude: no, you said /dev/block/omap/omap_hsmmc.0/by-name but rild needs /dev/block/platform/omap/omap_hsmmc.0/by-name/radio | 12:35 |
| MeowDude | ENV{PLATFORM_FOLDER}="$kernel", which I assume it thinks means omap. | 12:36 |
| MeowDude | I ought to just manually fill in the value | 12:36 |
| mal | MeowDude: just replace the damn $env{PLATFORM_FOLDER} on line 52-54 with omap | 12:36 |
| MeowDude | fuck I replaced the wrong lines | 12:37 |
| mal | that's what I have been trying to tell you all the time | 12:37 |
| MeowDude | mal: https://pastebin.com/mLFhxMtQ | 12:39 |
| MeowDude | there | 12:39 |
| mal | well you didn't replace one of those | 12:43 |
| MeowDude | dmesg shows functioning ril_daemon | 12:43 |
| mal | could you please use your brain a little | 12:43 |
| MeowDude | hopefully it reflects this at boot up | 12:43 |
| MeowDude | awsome rild is running but ofono is being a b**** | 12:44 |
| MeowDude | 1000 thanks mal, I couldn't have done this without you! (or krnlyng) | 12:45 |
| krnlyng | MeowDude, that's not the proper approach. | 12:45 |
| MeowDude | well rild is running isn't it? | 12:45 |
| MeowDude | huh now everytime I turn off the device it shows the loading ring until i swipe out like I am closing an app | 12:46 |
| krnlyng | MeowDude, put that fix on any other device and it will fail. | 12:46 |
| MeowDude | well yeah I know that, because not all devices have an omap processor | 12:46 |
| MeowDude | also now the lockscreen is broken | 12:46 |
| MeowDude | : ( | 12:46 |
| krnlyng | MeowDude, if you can come up with a proper fix (which you should) then others can benefit from your work too | 12:47 |
| MeowDude | also ofono gives this error: Jun 18 22:17:15 Sailfish ofonod[11559]: [mce] ERROR! Failed to attach to system bus: Exhausted all available authentication mechanisms (tried: EXTERNAL, DBUS_COOKIE_SHA1, ANONYMOUS) (available: EXTERNAL, DBUS_COOKIE_SHA1, ANONYMOUS) | 12:47 |
| krnlyng | MeowDude, and may be even more welcome to helping you in turn :) | 12:48 |
| MeowDude | I'll try and figure out how to better define ENV{PLATFORM_FOLDER} | 12:48 |
| krnlyng | thanks :) | 12:48 |
| MeowDude | for now though, I need to debug ofono.. maybe later... it's late and I have school tommorow | 12:48 |
| monich | MeowDude: that system bus thing sounds like a missing dbus policy | 12:52 |
| mal | MeowDude: which release did you build? | 12:53 |
| T4 | <abhishek_0> how do i fix qt5-qtimageformats-plugin-dds dependency | 13:02 |
| elros | spiiroin: I have issue with qtsensors. In messwerk there is wrong lux value when I wake up phone in dark room | 13:06 |
| elros | in this example it was showing 97 instead 2 until als sensor get 3 or more | 13:06 |
| elros | https://pastebin.com/4jcvHjQ6 | 13:06 |
| elros | I don't care that much about what qt messwer shows but I have notice that brightness regulation is also affected | 13:14 |
| lolek | mgrover: hi, I see that there's still one failure for the xperia build, any estimate for fixing this so I can get the image and do the tests? | 13:26 |
| lolek | monich: how's the wideband audio thing going? | 14:08 |
| monich | meh | 14:09 |
| duncan^ | In my chroot, createrepo and android-tools aren't in the repositories. An equivalent createrepo_c and android-tools-hadk are, but I'm not sure if these are suitable (I imagine the latter is, but the former perhaps not) | 14:52 |
| duncan^ | I mean... it build successfully so I guess it's fine... | 14:55 |
| m4r0v3r | lolek, that one isnt required | 15:01 |
| lolek | m4r0v3r: so, what's needed to to make new sfos image for X? | 15:02 |
| lolek | or how to continue from this? | 15:02 |
| r0kk3rz | give ota another go | 15:02 |
| lolek | ok, let's try then ;) | 15:03 |
| lolek | will have info very soon | 15:03 |
| vknecht | hello | 15:07 |
| lolek | r0kk3rz and it's stuck again | 15:15 |
| lolek | r0kk3rz this time the installation took way longer than before, but in the end it started to boot, displayed for 2 s sony logo, then did reboot, to display the logo again and it's stuck | 15:15 |
| r0kk3rz | yeah the short reboot would've been it flashing the kernel | 15:16 |
| T4 | <abhishek_0> when lbt will be back .. if i did everything right with obs | 15:16 |
| lolek | well I don't know, the system is in freezed state, screen display sony logo on white bg, nothing else | 15:17 |
| lolek | I can only turn it off | 15:17 |
| T4 | <abhishek_0> telnet into it | 15:18 |
| T4 | <abhishek_0> device will detected as network interface on host | 15:19 |
| lolek | r0kk3rz so, any idea? | 15:22 |
| lolek | or any option how can I debug it? | 15:22 |
| r0kk3rz | maybe you need to use cli ota method | 15:24 |
| duncan^ | According to the hardware adaption document, the build_packages helper script creates a kickstart file. However, it doesn't seem to create one for me. How necessary is this, and how should I create it? | 15:40 |
| r0kk3rz | very, and im sure it says in the hadk | 15:41 |
| krnlyng | duncan^, droid-configs is built? | 15:41 |
| duncan^ | krnlyng: Yes | 15:42 |
| krnlyng | duncan^, can you check if $ANDROID_ROOT/hybris/droid-configs/installroot exists? | 15:43 |
| duncan^ | According to the hardware adaption document, it is in hybris/droid-configs/installroot/usr/share/kickstarts/, but this directory doesn't exist | 15:43 |
| krnlyng | duncan^, how recently did you install/update your sdk | 15:44 |
| duncan^ | I pulled it in yesterday, as far as I know, it's the latest | 15:45 |
| krnlyng | ok, then you might need such a thing: https://bpaste.net/show/02aa4ba3150a | 15:45 |
| duncan^ | In hybris/droid-configs/, all I see are the following directories: RPMS documentation.list droid-configs-device patterns rpm sparse tmp | 15:45 |
| duncan^ | OK, I'll try that. Thanks for the help. | 15:46 |
| krnlyng | duncan^, yes the new sdk does that. put the script i sent you into $ANDROID_ROOT and name it get_kickstart.sh or something like that, then execute it with "source get_kickstart.sh" | 15:46 |
| krnlyng | from $ANDROID_ROOT | 15:46 |
| lolek | r0kk3rz which is? | 15:48 |
| duncan^ | krnlyng: I ran the script, so that means it's sufficient to skip the step in the hardware adaption document involving creating the kickstart file? | 15:56 |
| duncan^ | (which involves this installroot/ directory in droid-configs) | 15:56 |
| krnlyng | duncan^, you have to still run the replacement stuff on it, this only copies the plain file | 15:56 |
| duncan^ | Thanks | 16:02 |
| T4 | <toxip> Hey guys, how are Xiaomi phones in terms of portability? | 16:25 |
| T4 | <toxip> easy to port? | 16:26 |
| T4 | <NotKit> if you have kernel source, generally no different from other devices | 16:26 |
| T4 | <NotKit> [Edit] if you have kernel source, generally not different from other devices | 16:26 |
| beidl | hello hello! I'm back to porting sfos to the gnex again and I'm wondering where the ubu-chroot script went. | 16:27 |
| T4 | <toxip> @NotKit [if you have kernel source, generally not diffe …], all available https://www.xda-developers.com/xiaomi-redmi-4x-kernel-sources/ | 16:27 |
| T4 | <toxip> this seems like a nice budget phone to port to :) | 16:28 |
| duncan^ | I see now mic is failing with the error Unable to find pattern: Jolla Configuration $DEVICE | 16:32 |
| duncan^ | In hybris/droid-configs/patterns/, I definitely have a pattern which specifies that description. | 16:33 |
| lolek | r0kk3rz did update using cmd line, no visible errors, the same issue persist | 16:35 |
| lolek | so if you have any other ideas what I can test/do/check pls let me know | 16:36 |
| mal | duncan^: did you process patterns as instructed? | 16:39 |
| duncan^ | Yes, I ran rpm/dhd/helpers/build_packages.sh --configs as instructed in the section "8.4: Modifying a pattern" | 16:42 |
| duncan^ | Is that the correct way to do it? | 16:42 |
| duncan^ | Do I need to recompile? | 16:43 |
| r0kk3rz | theres a process patterns script too | 16:43 |
| mal | duncan^: I mean the line hybris/droid-configs/droid-configs-device/helpers/process_patterns.sh in chapter 8.5 | 16:44 |
| duncan^ | I ran that (again?) and it is still saying that it can't find the pattern | 16:45 |
| duncan^ | when I ran the process_patterns script, it printed a warning "C_CREATEREPOLIB: Warning: Record with type "patterns" already exists in repomd.xml | 16:45 |
| duncan^ | but I'm not sure if that's relevant | 16:45 |
| duncan^ | Error <creator>: Unable to find pattern: Jolla Configuration $DEVICE | 16:47 |
| duncan^ | (if that's useful) | 16:47 |
| mal | duncan^: how did you solve the createrepo issue you had? | 16:48 |
| duncan^ | I appended _c to all instances of createrepo, because the programs in the package (and the package itself) appeared to have been renamed as such | 16:49 |
| duncan^ | same with modifyrepo | 16:50 |
| duncan^ | android-tools had been renamed android-tools-hadk so I merely renamed them to that | 16:50 |
| duncan^ | (but the programs in android-tools-hadk were not renamed) | 16:51 |
| mal | maybe you need to run sudo ssu ar mer-tools http://www.merproject.org/logs/%23sailfishos-porters/%23sailfishos-porters.2018-06-16.log.html#t2018-06-16T12:47:21 | 16:55 |
| T4 | <abhishek_0> mal how do i improve smoothness ? | 16:59 |
| T4 | <abhishek_0> everything is a bit jittery | 17:00 |
| mal | it depends on what the reason for not being smooth is | 17:00 |
| T4 | <abhishek_0> from where do i start debugging | 17:00 |
| mal | was there anything odd in top output? what is using cpu when that happens | 17:00 |
| T4 | <abhishek_0> lipstick | 17:06 |
| mal | but how much | 17:06 |
| mal | and some quite high cpu usage from lipstick is normal | 17:06 |
| T4 | <abhishek_0> 31% | 17:06 |
| mal | does that device have one of those cpus with two types of cores? | 17:08 |
| T4 | <abhishek_0> no, i guess it has 4 cores all are cortex a53 | 17:09 |
| T4 | <abhishek_0> snapdragon 410 | 17:09 |
| T4 | <abhishek_0> here is the logcat doing stuff like opening camera , swiping through ui https://hastebin.com/favewuhiri.sql | 17:15 |
| T4 | <abhishek_0> even a 720p video lags | 17:18 |
| mal | wondering if it somehow doesn't use hardware acceleration correctly | 17:19 |
| T4 | <abhishek_0> mal where i should start looking ? | 17:21 |
| mal | not sure | 17:22 |
| duncan^ | OK, I ran the configuration script again for the patterns, and it produced the attributeerror as noted in the document. | 17:23 |
| duncan^ | However, I still can't get mic to recognize the pattern | 17:23 |
| T4 | <abhishek_0> but still any idea .. guesses ? | 17:24 |
| saidinesh5 | @abhishek_0 write a little sample qt application to display opengl info.. | 17:24 |
| saidinesh5 | lipstick shouldn't be eating 30% in idle | 17:24 |
| mal | duncan^: that happened after installing the proper createrepo? | 17:24 |
| r0kk3rz | duncan^: do process patterns > build packages configs > mic and post the whole thing to pastebin | 17:25 |
| mal | saidinesh5: not on idle but when doing something | 17:25 |
| saidinesh5 | oh | 17:25 |
| saidinesh5 | 06-18 17:04:27.101 10152 10152 I Adreno-EGL: OpenGL ES Shader Compiler Version: XE031.09.00.03 | 17:28 |
| T4 | <abhishek_0> any app which uses opengl ? | 17:29 |
| saidinesh5 | doesn't look like a problem from opengl | 17:29 |
| saidinesh5 | @abhishek_0 all of your UI uses opengl , as it is built on QtQuick | 17:29 |
| saidinesh5 | well opengles | 17:29 |
| saidinesh5 | libEGL : failed to load libgui: dlopen failed: | 17:30 |
| saidinesh5 | that one seems dodgy... | 17:30 |
| saidinesh5 | it keeps repeating over and over | 17:30 |
| duncan^ | https://paste.debian.net/hidden/31f5fc89/ | 17:30 |
| saidinesh5 | so maybe something is wrong | 17:30 |
| duncan^ | here we are | 17:30 |
| mal | duncan^: your .ks file is missing local repo? | 17:32 |
| T4 | <abhishek_0> saidinesh5 yes it does .. i think this issue is pretty common on here | 17:32 |
| mal | duncan^: droid-local-repo that is | 17:32 |
| saidinesh5 | @abhishek_0 maybe worth it to look why it isn't loading. take a simple opengles app and try to run it | 17:33 |
| mal | duncan^: something like this: repo --name=adaptation-community-zoom2-@RELEASE@ --baseurl=file:///home/duncan/build/sfos/mer-hybris/droid-local-repo/zoom2 | 17:34 |
| saidinesh5 | https://android.googlesource.com/platform/frameworks/native/+/1966cf6%5E%21/ came up when googling for libegl and libgui | 17:34 |
| Mister_Magister | saidinesh5: hello :) | 17:35 |
| saidinesh5 | hey Mister_Magister | 17:35 |
| Mister_Magister | whatsup? | 17:35 |
| saidinesh5 | oh not much.. fixing my schedule again, and finishing off work forthe day | 17:35 |
| Mister_Magister | kay :P i just sit down trying to fix account management in morsender :P | 17:36 |
| saidinesh5 | Ahh | 17:36 |
| mal | duncan^: btw, you don't have to regenerate .ks file very often, I always use the same one | 17:39 |
| duncan^ | mal, thanks, that worked. It's now able to find my patterns. | 17:44 |
| duncan^ | OK, now I've got a different problem. mic produced a flashable zip called sailfishos-$DEVICE-release-.zip | 19:41 |
| duncan^ | I attempted to flash this image over a known good cyanogenmod image, but it sais "Failed to extract filesystem! E: Error executing binary in zip '/data/media/sailfishos-$DEVICE-release-.zip' | 19:43 |
| duncan^ | (should be noted it actually says the device name in place of $DEVICE, I just write $DEVICE here because it's more appropriate) | 19:43 |
| duncan^ | I suspect the image is broken somehow, and fixing the underlying issue causing awry naming might fix the build of the image. | 19:44 |
| mal | duncan^: the filename looks odd, are you sure you remembered to define the needed things like RELEASE and EXTRA_NAME before running mic | 19:45 |
| mal | the filename should look something like this sailfishos-fp2-sibon-release-2.1.4.14-alpha6.zip | 19:46 |
| mal | with values matching your build of course | 19:47 |
| duncan^ | Looks like I defined them the first time I ran mic, then didn't redefine them after fixing the issue with local repo | 19:49 |
| duncan^ | sorry :/ thanks for your patience | 19:49 |
| mal | try building again and see if the zip looks correct after that | 19:50 |
| r0kk3rz | duncan^: i put those env vars into hadk.env | 19:52 |
| duncan^ | It looks like it's getting the filenames right now | 19:53 |
| mal | the size of the zip should be about 300 MB | 19:55 |
| duncan^ | Yes, it was 267M | 20:03 |
| duncan^ | Nonetheless, I'm getting the same error. | 20:04 |
| duncan^ | It says it's not finding the updater binary in the image. | 20:04 |
| r0kk3rz | post output | 20:05 |
| r0kk3rz | basically always post outputs :) | 20:06 |
| duncan^ | I thought I couldn't, but there is obscure option in recovery to copy loog... one second | 20:07 |
| r0kk3rz | mic output is a good idea too | 20:08 |
| duncan^ | https://paste.debian.net/hidden/c3d1d89a/ | 20:08 |
| r0kk3rz | what device is this? | 20:12 |
| duncan^ | Line 835 is the start of relevant parts | 20:12 |
| duncan^ | r0kk3rz: Nook Simple Touch | 20:12 |
| r0kk3rz | 1400 looks like the most relevant line to me | 20:12 |
| duncan^ | OK, that's fine - I guess I just need to expand the partitions? | 20:13 |
| duncan^ | It's an SD card, so that's not so hard | 20:13 |
| duncan^ | they're set to about ~500M each | 20:13 |
| r0kk3rz | yeah thats probably a little small | 20:15 |
| mal | duncan^: on one old device with limited space I used sdcard for rootfs | 20:17 |
| mal | duncan^: ah, so you also use sdcard | 20:17 |
| duncan^ | So it's got 2G of internal storage, it will boot from SD if you provide it with U-boot + MLO, and a valid kernel+ramdisk | 20:18 |
| duncan^ | which is nice for recovery | 20:18 |
| mal | but anyway maybe at least 1 GB for sailfish rootfs is needed so there is some space for apps etc, preferrably more | 20:19 |
| piggz | mal: nothing in faq about new pulse configs .... what do i need for headphones to work again in 11? | 20:21 |
| mal | piggz: link to your config repo | 20:22 |
| piggz | mal: https://github.com/piggz/droid-config-mido | 20:22 |
| mal | piggz: only headphones fail? | 20:23 |
| piggz | mal: so far | 20:23 |
| mal | piggz: is this still accurate https://github.com/piggz/droid-config-mido/blob/master/sparse/etc/ohm/plugins.d/accessories.ini | 20:24 |
| mal | I mean check on device that name and device are ok | 20:24 |
| piggz | mal: apparently not | 20:25 |
| piggz | how odd | 20:25 |
| piggz | ----====( /dev/input/event7 )====---- | 20:25 |
| piggz | Name: "msm8953-snd-card-mtp Headset Jack" | 20:25 |
| mal | piggz: have you done other changes that could cause the device to change, other option is to see if having only jack-match is enough | 20:26 |
| piggz | mal: i did a kernel upgrade | 20:26 |
| mal | piggz: that's probably the reason then | 20:26 |
| piggz | mal: likely ... this kernel maybe adds s2w | 20:27 |
| piggz | as event2 | 20:27 |
| r0kk3rz | maybe some input/event roulette is going on | 20:30 |
| piggz | s2w doesnt seem to work anyway | 20:31 |
| piggz | mal: radio also stopped , but i think i remember reading about config changes for that | 20:34 |
| mal | piggz: your configs look reasonable | 20:37 |
| mal | piggz: just to be sure, were you using a patched fm radio middleware? | 20:38 |
| mal | piggz: if that is the case you might have wrong version in use, I updated the one in common | 20:38 |
| piggz | mal: hmm, i was... | 20:40 |
| T4 | <abhishek_0> piggz do you have that libgui.so or libEGL error (unrelated just asking) | 20:42 |
| piggz | when doing what? | 20:42 |
| piggz | mal: have no other ports had the issue of init/de-initing the radio ? | 20:45 |
| mal | piggz: not sure, if you check the new code it does init and deinit every time, but not via /sys | 20:47 |
| mal | but probably not related to your issue | 20:48 |
| mal | piggz: wouldn't it just be easier to patch the kernel | 20:49 |
| piggz | mal: what would the patch look like? | 20:49 |
| piggz | ( i mean roughly ... not exactly!) | 20:50 |
| mal | piggz: most likely reverting this https://github.com/piggz/android_kernel_xiaomi_msm8953/commit/fb5596bf718d6fbd85e291ec6c4f757d72478693#diff-540f47e54d82292e7f106daa845ee4e8 | 20:54 |
| piggz | hmmmm, dejavu | 20:54 |
| mal | but might need other changes also | 20:54 |
| T4 | <abhishek_0> piggz doing stuff .. browsing, media, swiping through the apps | 20:56 |
| T4 | <abhishek_0> in the logcat | 20:56 |
| abranson | vknecht: I think you can see in there mention of the 0x7fa30c04 format that we do support. what's interesting is those C2DColorConvert errors starting around 358. it's being asked to transform the colour format internally - we don't use that class in droidmedia - and failing | 20:56 |
| abranson | if you want to see more of what's going on, find ACodec.cpp in your android tree - it'll be under frameworks/av, and find the #define LOG_NDEBUG line near the top that's commented out. | 20:57 |
| duncan^ | Hmm, it is taking rather a while to unpack | 20:58 |
| abranson | uncomment it and build your android libs again, and all the verbose ALOGVs will be printed out, which will tell you more about the formats set for the codecs. if for some reason there's a strange transformation going on, you might see it in there. | 20:58 |
| abranson | other source files near ACodec might tell you things too. each one needs to have its logging enabled individually, | 20:59 |
| *** Nokius_ is now known as Nokius | 21:03 | |
| mal | piggz: a hint, please rebase, don't merge | 21:03 |
| piggz | mal: radio not working with patched version anyway | 21:06 |
| duncan^ | It appeared to break /boot somehow. The partition has corrupted itself nicely. | 21:07 |
| mal | piggz: odd | 21:07 |
| mal | piggz: I assume it used to work | 21:07 |
| piggz | mal: it did...tho ive added a kernel upgrade tooo..... | 21:07 |
| T4 | <abhishek_0> is lbt lbt- online ? | 21:07 |
| piggz | elts debug briefly | 21:07 |
| T4 | <abhishek_0> !seen lbt | 21:08 |
| T4 | <abhishek_0> oh wait i am a bot | 21:08 |
| mal | piggz: does it tune to channels? | 21:08 |
| piggz | mal: its working ..... | 21:09 |
| piggz | when you 'su nemo' it ignores the env variable | 21:09 |
| Mr_Doge | !seen T4 | 21:09 |
| merbot | Mr_Doge: T4 was last seen in #sailfishos-porters 59 seconds ago: <T4> <abhishek_0> oh wait i am a bot | 21:09 |
| Mr_Doge | !seen lbt | 21:10 |
| merbot | Mr_Doge: lbt was last seen in #sailfishos-porters 8 hours, 57 minutes, and 56 seconds ago: <lbt> nemo:devel:hw:wingtech:wt88047 is done abhishek_0 | 21:10 |
| mal | piggz: ok, good | 21:10 |
| piggz | mal: panic over, headphones and radio working again in my devel: | 21:10 |
| piggz | sorry for sloppy git practices | 21:10 |
| piggz | ! | 21:10 |
| mal | piggz: always a good idea to ask me :P | 21:10 |
| Mr_Doge | oh its done,,, sorry for the ping | 21:11 |
| piggz | i still say my small patch is easier than hacking the kernel driver and is safe to upstream ;) | 21:11 |
| mal | piggz: it might be easier but whether it's better is another question | 21:12 |
| piggz | true, but frigging around in the kernel driver is a lot tougher | 21:13 |
| piggz | and dangerous | 21:13 |
| Nokius | r0kk3rz: let's see if we can do something about the cam as next | 21:13 |
| piggz | mal: anyone looked at FP stuff yet? | 21:13 |
| mal | piggz: not that I know of | 21:14 |
| piggz | mal: does 2.2 add better BT for your watch? | 21:14 |
| mal | I haven't tested yet | 21:14 |
| r0kk3rz | Nokius: recording could be tricky | 21:15 |
| r0kk3rz | it seems to be on android side | 21:15 |
| mal | piggz: probably not | 21:16 |
| Nokius | r0kk3rz: yeap android is bity here | 21:17 |
| T4 | <abhishek_0> mal it it says it is excluded https://build.merproject.org/project/monitor/nemo:devel:hw:wingtech:wt88047?arch_armv8el=1&defaults=0&excluded=1&repo_sailfish_latest_armv7hl=1 | 21:18 |
| T4 | <abhishek_0> not sure why | 21:18 |
| mal | @abhishek_0 it should be like that | 21:19 |
| mal | @abhishek_0 on #mer-boss 00:19 < merbot> Updating patterns from nemo:devel:hw:wingtech:wt88047/droid-config-wt88047 | 21:19 |
| mal | it's in the process of updating the patterns | 21:20 |
| duncan^ | So the zip installer tries to unpack hybris-boot.img onto /dev/by-name/boot. However, this isn't desirable for me, since I need boot to have just uImage+uRamdisk. So, I'm wondering what to do - should I repack hybris-boot.img into a kernel+ramdisk? | 21:21 |
| mal | you probably need to edit the updater script in hybris/hybris-boot | 21:22 |
| T4 | <abhishek_0> after that is done, how will i know i did everything right ? maybe i am asking too much questions :P | 21:22 |
| mal | assuming the repos are using correct versions it looks ok | 21:28 |
| duncan^ | abootimg is claiming that it's not a valid Android bootimage, which is odd. | 21:33 |
| duncan^ | It says: "sizes mismatches in boot image" | 21:34 |
| T4 | <abhishek_0> sledges please create nemo:testing:hw:wingtech:wt88047, if it is already there then provide me the maintainership username abhishek9650 | 22:02 |
| vknecht | abranson, ack, thx, 'll try s∞n ;-) | 22:05 |
| MeowDude | okay so rild is running (again, thanks guys), and ofono is running but I am getting an error about connection refused, and there is no empty sim icon on my phone. I grabbed a pastebin of systemctl status ofono and ps ax | grep rild: https://pastebin.com/bZnPRBP8 | 23:26 |
| MeowDude | if I restart ofono manually there are no errors in systemctl status ofono, but there is still no empty sim icon of internet sharing option in settings | 23:27 |
| MeowDude | running /system/bin/logcat -b radio gives: Unable to open log device '/dev/log/radio': Not a directory | 23:32 |
| MeowDude | upon manually reading /dev/alog/radio I get this: https://pastebin.com/xcbjPy6Q | 23:36 |
| MeowDude | arg I can't symlink an empty /dev/log directory with /dev/alog because there is already something in /dev names log | 23:45 |
| MeowDude | FINALLY I grabbed a log using a dirty work around: https://pastebin.com/PmQe5rXk | 23:52 |
| MeowDude | this seems to be the only error in the log: E/use-Rlog/RLOG-RIL(s)( 480): loadOpnameFile: Can't open the Opname file: reason No such file or directory. | 23:53 |
| MeowDude | https://pastebin.com/ScJk05Fz | 23:57 |
| MeowDude | ofono seems happy now... | 23:57 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!