**** BEGIN LOGGING AT Mon Jan 15 03:00:01 2018
Jan 15 07:52:27 <Herrie|Laptop> Morning
Jan 15 07:52:31 <Herrie|Laptop> Tofe: Result seems the same :S
Jan 15 07:52:35 <Herrie|Laptop> Any suggestions?
Jan 15 07:52:42 <Herrie|Laptop> On what I could try?
Jan 15 07:52:53 <Herrie|Laptop> lvm in general doesn't seem to be in a very happy place
Jan 15 08:30:10 <Tofe> Morning
Jan 15 08:30:40 <Tofe> Herrie: what did you try ? dm verity ?
Jan 15 08:54:56 <Herrie|Pre3> Tofe: Tried the .zip from miui forum to disable dm-verity via fstab and formatted data in twrp
Jan 15 08:56:06 <Herrie|Pre3> I didn't try to simply mount 049 though as of yet
Jan 15 08:56:09 <Herrie|Pre3> Let me try that
Jan 15 08:57:29 <Tofe> Herrie|Pre3: if you reformatted, it should be a simple mount now, yes
Jan 15 09:06:43 <Herrie|Pre3> Tofe: Seems it works to simply mount it now :)
Jan 15 09:07:02 <Herrie|Pre3> My init.sh is a mess now so need to clean that up a bit
Jan 15 09:07:12 <Herrie|Pre3> I tried a hybrid to get telnetd working as well
Jan 15 09:07:23 <Herrie|Pre3> So I need to figure out what I need and what I don't :P
Jan 15 09:09:09 <Tofe> :) good
Jan 15 09:15:37 <Herrie|Pre3> Because initially adb didn't work, now it does. Could be that some of the halium init has worked on just the android-tools upgrade etc
Jan 15 09:15:49 <Herrie|Pre3> Need to figure that out now
Jan 15 09:42:01 <Tofe> The android-tools upgrade seems like a good candidate
Jan 15 09:42:29 <Tofe> We don't use anything of halium in our boot process
Jan 15 09:44:03 <Herrie|Pre3> Tofe: I'll do some trial and error
Jan 15 09:44:26 <Herrie|Pre3> For preparing the Android USB stuff I used both AsteroidOS and Halium bits
Jan 15 09:44:35 <Herrie|Pre3> Need to see what actually was sticking :P
Jan 15 09:44:47 <Herrie|Pre3> AsteroidOS uses Yocto like us :)
Jan 15 09:44:53 <Herrie|Pre3> So that's helpful :)
Jan 15 09:45:05 <Herrie|Pre3> Can reuse some of their stuff at times
Jan 15 09:50:04 <Tofe> yup
Jan 15 10:36:24 <Herrie|Laptop> Tofe: You have any idea what https://github.com/AsteroidOS/meta-asteroid/blob/master/recipes-core/initrdscripts/initramfs-boot-android/init.sh#L30 is used for?
Jan 15 10:48:13 <Herrie|Laptop> Tofe: Seems I need to mount /system as well as per https://github.com/openembedded/meta-openembedded/blob/pyro/meta-oe/recipes-devtools/android-tools/android-tools-conf/android-gadget-setup#L9
Jan 15 10:48:20 <Herrie|Laptop> Otherwise we don't have build.prop ;)
Jan 15 10:52:43 <Herrie|Laptop> Seems we do later in the script, but seems we need it earlier ;)
Jan 15 11:06:19 <Herrie|Laptop> I guess I could temporarily mount it and umount it after android-gadget-setup
Jan 15 11:13:59 <Herrie|Laptop> Tofe: I think I know why ADB didn't work for you... Seems that there's no enable_adb in /proc/cmdline ;)
Jan 15 11:18:37 <Herrie|Laptop> Seeing that AsteroidOS uses "debug-ramdisk" as per https://github.com/AsteroidOS/meta-asteroid/blob/master/recipes-core/initrdscripts/initramfs-boot-android/init.sh#L42 which I could trace back to their documentation: https://asteroidos.org/wiki/boot-process/
Jan 15 11:18:52 <Herrie|Laptop> So seems we need to add it to kernel cmdline in the linux recipe for each target
Jan 15 11:19:36 <Herrie|Laptop> I.e.: probably add a CMDLINE += "enable_adb" or something after https://github.com/shr-distribution/meta-smartphone/blob/pyro/meta-xiaomi/recipes-kernel/linux/linux-xiaomi-mido_git.bb#L11
Jan 15 11:19:57 <Herrie|Laptop> JaMa: Thnx for merge!
Jan 15 11:30:56 <Herrie|Laptop> Tofe: And we should do that for all targets
Jan 15 11:31:02 <Herrie|Laptop> So we could name it differently as well :P
Jan 15 11:50:00 <Tofe> Herrie|Laptop: oh, wait wait! putting enable_adb *must* be temporary
Jan 15 11:50:11 <Herrie|Laptop> Tofe: Ah OK
Jan 15 11:50:33 <Tofe> one adb is started from init.sh, the boot is finished and it will never exit
Jan 15 11:50:55 <Tofe> (iirc...)
Jan 15 11:51:13 <Herrie|Laptop> Tofe: Ah that's why halium does telnetd on 23 and 2323
Jan 15 11:54:08 <Herrie|Laptop> Tofe: Then the question is how you enable it temporarily
Jan 15 11:54:33 <Herrie|Laptop> Ah that's where fastboot -c comes in
Jan 15 11:55:45 <Herrie|Laptop> Tofe: But you could stop adbd before switching to rootfs, no ?
Jan 15 11:56:03 <Herrie|Laptop> It would disconnect then briefly but you should be able to pick up again when new one is active?
Jan 15 11:56:09 <Herrie|Laptop> Or am I thinking wrong here?
Jan 15 12:09:39 <Tofe> well the thing is, once initrd is working it rarely breaks
Jan 15 12:10:20 <Tofe> that's why we mainly need adb when degguging initrd when doing a new port, but then it's just not really needed anymore
Jan 15 12:11:01 <Tofe> Personnally, when I want to activate adb in initrd, I directly modify init.sh and rebuild initrd :)
Jan 15 12:40:28 <Herrie|Laptop> Tofe: OK
Jan 15 12:44:57 <Herrie|Laptop> That makes sense... I guess I could add it to the porting guide or something
Jan 15 12:46:59 <Herrie|Laptop> Tofe: You have any idea why AsteroidOS would use FunctionFS? https://android.googlesource.com/kernel/common/+/bcmdhd-3.10/Documentation/usb/functionfs.txt
Jan 15 12:47:06 <Herrie|Laptop> We don't have it in our init
Jan 15 12:56:27 <Herrie|Laptop> Tofe: Ah seems FunctionFS is something "new" that was introduced to replace some legacy stuff: https://android-review.googlesource.com/c/platform/system/core/+/31640
Jan 15 14:48:34 <Tofe> I didn't know about FunctionFS
Jan 15 14:50:01 <Tofe> anyway adb is more useful for shell purposes
Jan 15 14:50:24 <Tofe> otherwise there are plenty of solutions (mtp, sshfs, ftpfs...)
Jan 15 15:02:13 <Herrie|Laptop> Tofe: Well apparantly AsteroidFS requires FunctionFS for adb
Jan 15 15:02:19 <Herrie|Laptop> It replaced legacy stuff it seems
Jan 15 15:02:32 <Herrie|Laptop> So I guess for us it would be the case to make sure it works with all our targets
Jan 15 15:02:48 <Herrie|Laptop> AsteroidOS uses 3.10+ kernels
Jan 15 15:02:58 <Herrie|Laptop> So could be 3.4 might be an issue, but we'll see
Jan 15 15:03:05 <Herrie|Laptop> Shouldn't be hard to test
Jan 15 15:03:19 <Herrie|Laptop> If I'm anyway looking into it, I can test it
Jan 15 15:03:56 <Herrie|Laptop> Just as usual documentation could be a lot better :P
Jan 15 15:06:13 <Herrie|Laptop> Some more info seems to be at https://wiki.linaro.org/LMG/Kernel/AndroidConfigFSGadgets
Jan 15 15:09:42 <Herrie|Laptop> Seems it might be 3.8+
Jan 15 15:09:47 <Herrie|Laptop> So I guess we'd need to double check
Jan 15 15:18:25 <Tofe> ok
Jan 15 15:19:10 <Herrie|Laptop> Should be easy enough to check I guess
Jan 15 15:19:47 <Herrie|Laptop> I.e. generate a simplified initramfs that doesn't do the rfs switch and see if adb works on target?
Jan 15 15:33:45 <Tofe> If it goes into adb, it shouldn't execute the following lines anyway
Jan 15 15:34:58 <Herrie|Laptop> Tofe: Ehm true ;)
Jan 15 18:14:11 <Herrie> Tofe: You have a Mako laying around?
Jan 15 18:14:22 <Herrie> To test the .fastboot?
Jan 15 18:50:52 <Herrie> Tofe: Seems OK for Hammerhead
Jan 15 18:51:22 <Herrie> https://bpaste.net/show/082432b6b4a4
Jan 15 19:18:49 <Herrie> tOFE: NEW
Jan 15 19:19:00 <Herrie> Tofe: New mido log: https://bpaste.net/show/09513719998a
Jan 15 19:39:11 <Herrie> So seems it's getting a bit further ;)
Jan 15 19:39:13 <Herrie> And no boot loop
Jan 15 19:39:17 <Herrie> So that's already something
Jan 15 20:55:05 <Herrie> Hmmz seems now it bootloops
Jan 15 20:55:14 <Herrie> Only thing I updated wsa the persist partition in fstab
Jan 15 20:55:23 <Herrie> So seems a bit random
Jan 15 20:56:01 <Herrie> Tofe: This is the log when it bootloops: https://bpaste.net/show/6d5ca8662860
Jan 15 20:56:47 <Herrie> Tofe: The mako fastboot file: http://vve-goudenleeuw.nl/zImage-mako.fastboot
Jan 15 21:52:15 <Tofe> Herrie: "Failed to find valid rootfs" for mido
Jan 15 21:52:35 <Tofe> for mako I'll try tomorrow, I'm completely tired tonight :p
Jan 15 22:20:17 <Herrie> Tofe: Oops must have overwritten the init from another pc :P
Jan 15 22:42:29 <Herrie> Check this one: https://bpaste.net/show/fac0c867e8e8
**** ENDING LOGGING AT Tue Jan 16 03:00:03 2018