| T42 | <Papoucifle> Hello ! I fixed my fixup-mountpoints for my motorola Addison, and I don't have "dev-block-platform-soc-7824900.sdhci-by\x2dname-system.device timout" error anymore, but it seems that my partitions are still not mounted. Here is my journactl : https://pastebin.com/rKU8KRxs and dmesg : https://pastebin.com/3XHUx3DU . I still have "failed to start the lipstick ui" and "failed to start droid-hal-init", among many other errors | 10:48 |
|---|---|---|
| mal | you should be using "journalctl -b --no-pager" to get full journal log | 11:12 |
| mal | that is now missing some of the early messages which would explain what is wrong | 11:13 |
| T42 | <elros34> disable audit according to instruction anddo you even have now /usr/lib/systemd/system/*.mount and | 11:24 |
| T42 | <elros34> system.mount*? | 11:24 |
| T42 | <elros34> check what is mounted with 'findmnt' | 11:25 |
| T42 | <Papoucifle> Ok I guess I failed with my fixup-mountpoints. /usr/lib/systemd/system.mount points to /dev/block/platform/soc/7824900.sdhci/by-name/, but in my fixup-mountpoints I've put "-e 's block/bootdevice/by-name/system mmcblk0p53 '", so sed couldn't update the file correctly. I fixed system.mount manually on the device and now /system mounts correctly. | 12:02 |
| T42 | <Papoucifle> | 12:02 |
| T42 | <Papoucifle> The problem is that prior to that, I had the correct "-e 's block/platform/soc/7824900.sdhci/by-name/system mmcblk0p53 '" in my fixup-mountpoints file but the device did not even boot, I had no access to telnet on port 2323 (only 23, which I believe is not the same console). Do you think I should fix all files manually, or do you think I made a mistake with my previous fixup-mountpoints file ? | 12:02 |
| T42 | <elros34> of course you should fix all mount units manually now to move forward | 12:05 |
| T42 | <elros34> checking whether you made some mistake is up to you. Make the changes, build hybris/droid-hal and unpack and check whether you have correct files in droid-hal-device*.rpm | 12:06 |
| T42 | <Papoucifle> Ok i fixed system.mount and vendor.mount, cause none other .mount files had references to /dev/block/platform/soc/7824900.sdhci/by-name/ . Here is the result of findmnt now : https://pastebin.com/AbVwpUPe . And the full journalctl : https://pastebin.com/6u513SrR (I did not deactivate audit yet) | 12:37 |
| T42 | <elros34> Is it system-as-root device? what do you have in /system now? | 12:39 |
| T42 | <Papoucifle> I have this : https://pastebin.com/EVuzV7qi . I don't really understand how to know if it's a "system-as-root" device 😅 | 12:41 |
| T42 | <elros34> btw latest release is 4.5.0.19 | 12:41 |
| T42 | <elros34> yeah it is, see your /system/ looks like / and it contains another system/ and symlinks like product -> /system/product | 14:52 |
| b100dian | mal: I have kernel 5.4 and image header 3 on sake if you need this info by chance https://github.com/mer-hybris/hybris-boot/compare/master...sailfishos-on-sake:hybris-boot:hybris-18.1#diff-8d88eb632967032ab70dc52ece6ec958fb13d818e22d3513130ae177993b2fd7 | 20:21 |
| mal | b100dian: doesn't BOARD_MKBOOTIMG_ARGS contain header version? a bit strange if it doesn't | 20:24 |
| mal | or do you see some variable for it in device boardconfig | 20:24 |
| b100dian | it was not in my case. I am just posting this for reference, I have it working with those hacks | 20:25 |
| b100dian | also, the original firware was shipped with gz images, only the lineage one was using lz4, and I seem to have read that you need to match the compression for boot/vendor_boot | 20:26 |
| b100dian | I stand corrected, BOARD_MKBOOTIMG_ARGS has this, can't recall why I added it | 20:28 |
| mal | why doesn't gz work? you can enable different ramdisk compression formats in kernel configs | 20:35 |
| mal | ah, so it needs to match vendor_boot, I have very little experience with vendor_boot | 20:36 |
| piggz | back to nfcd debugging//// | 20:43 |
| b100dian | mal: probably bootloader concatenates the two images' ramdisks. More devices will come with image header 3 and 4 https://source.android.com/docs/core/architecture/partitions/vendor-boot-partitions#bootloader-support | 20:51 |
| b100dian | piggz: just curious, how do you test nfc in waydroid? | 20:51 |
| piggz | b100dian: i got an nfc app from fdroid | 20:53 |
| b100dian | ok, so no settings change. I installed NFC reader and I have one device working on Waydroid 11 and another one not working on Waydroid 10. Neither have I debugged in sailfishos. | 21:06 |
| piggz | b100dian: try my hack build of nfcd-binder-plugin | 21:24 |
| b100dian | is that armv7hl? | 22:20 |
| T42 | <adampigg> both | 22:22 |
| T42 | <b100dian> yes, copypac'd but I was surprised that the repo was 32-bit. Is it, or is it a too late hour here | 22:24 |
| T42 | <b100dian> I dont' have much success with either package, on the device which worked with Waydroid.., sorry I will debug in the next days | 22:50 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!