[ROM][UNOFFICIAL] Lineage OS 13 For Obake [22/2/17]

[otmitia]

Hi pbararette, there was a new build lineage-13.0-20170923-nightly-ghost-signed.zip and in it ASB September 2017, version string to 2017-09-01. Make us a new build?))

 

[dDefy]

It really works, after I did all the steps.
Thank you pbararette, made my day !

 

[pbarrette]

All,

The September 2017 security patches have been released and I compiled a new ROM.
I am not going to release it at this time because the merged code completely breaks SSL certificate validations.
This means that virtually all applications and services that rely on internet functions are crashing.

I will keep tracking this and post an update when the issue is corrected.

Hi pbararette, there was a new build lineage-13.0-20170923-nightly-ghost-signed.zip and in it ASB September 2017, version string to 2017-09-01. Make us a new build?))

Hi otmitia,

As you can see from my post above, I delayed a release because of problems introduced with code changes to SSL found in this merge:
https://review.lineageos.org/#/c/181332/

I was able to build a new version yesterday that reverts the code change and it has been working for me, but there is no official merged fix in the Lineage OS github tree. I have uploaded new images with the September security fixes, but it's not a mainline image because of the code revert that has to be manually applied.

Do you know anyone who is running the 20170923 image for ghost (Moto X)? I suspect they will have the same SSL problems I did. Those problems include all Google apps crashing, to include Chrome, crashes of all email clients, etc.

Working ROMs for obake (lineage-13.0-20170922-UNOFFICIAL), both with and without root, can be found at the usual AndroidFileHost location below:
https://www.androidfilehost.com/?w=files&flid=194742

 

[dDefy]

The August updates were finally merged and I built a new release yesterday evening.

The ROM zips are here.

Note that there are some WLAN changes that were merged in late July for our kernel. I haven't noticed any differences in WiFi connectivity, so I assume the updated code is stable:
https://review.lineageos.org/#/q/pr...rola_ghost+branch:cm-13.0+topic:CVE-2015-0571

Saw that you've made a new build, any new things in ?
[ sorry, way to impatient , saw your post above ]

 

[otmitia]

Do you know anyone who is running the 20170923 image for ghost (Moto X)? I suspect they will have the same SSL problems I did. Those problems include all Google apps crashing, to include Chrome, crashes of all email clients, etc.

I do not know, at 4pda.ru everything is quiet, well, probably correct, soon)
Thanks for the release, let's hope for a quick fix!)

 

[smilesinlife]

i wana try it but stuck on kitkat 4.4 su-6-7.7 firmware, any way to root in and unlock its bootloader

 

[mushbert]

The short answer is no. Root can be obtained but you cannot write to the system partition without an unlocked bootloader. The only current way to unlock is using Sunshine (available at http://theroot.ninja), which works with phones running 4.2 through early 4.4. You can try it if you want to satisfy your curiosity though.

 

[pbarrette]

Hi Everyone,

The October security updates have been merged in Lineage and I have compiled a new ROM found here:
https://androidfilehost.com/?w=files&flid=194742

There are a couple items to note:
1] This bumps Android to 6.0.1_r80 which is MOB31Z.
2] The SSL problems are gone after I destroyed and re-created my development VM. So the earlier problem was on me.
3] I changed the start-point of the "obake_thermal" service from "on property:init.svc.bootanim=stopped" to "on post-fs" to get the thermal control in place earlier.
- This should help some with heating after installing a ROM update.
- The github project has also been updated.
4] Immediately after I installed the ROM yesterday, I had to re-select my SIM card and my radio started crashing. A reboot fixed the issue.

 

[pbarrette]

All,

I have purchased a Sony Xperia XZ1 Compact and moved from Verizon to T-Mobile.
It's barely larger than the Droid Mini, has a modern CPU, launched with Oreo and supports T-Mobile US's VoLTE and WiFi calling, so I couldn't justify using the Droid Mini as my daily phone any more.

As such, I can no longer do any "soak" testing of compiled ROMs and plan on pushing a final ROM update with the December 2017 security updates (whenever those are merged into the LOS-13 code tree). So, basically, two more ROMs that I can't really test before releasing to you guys.

I'm also willing to walk anyone through the procedure of setting up a build environment if they wish to continue supporting the obake line.

My build environment is a Ubuntu virtual machine with 128GB of storage running on Hyper-V on a Windows 10 laptop. Not the fastest build system, but it's fairly simple to set up if you've got a little experience with linux.

You may have noticed that shortly before the last ROM release, I added a "proprietary_vendor_motorola" repo (forked and cleaned up from TheMuppets) to my github that contains the proprietary motorola binaries and firmware. This was in an effort to streamline the build process and make it as simple as possible. Hopefully, if someone decides to continue support, it will make the process easier for them.

I will update my github with my "local_manifests" repo before I stop support. It basically changes the build sources for "obake" from the out-of-date LineageOS official github repo to mine, and downloads the vendor blobs so you don't have to do an initial "adb pull" from a working phone to get them.

 

[dDefy]

Hi pbarrette,
Good luck with your new phone, it is a decent update from the Droid Mini, altoough I'm missing the wireless charging on.

 

[SnipeR_TT]

pbarrette,
Thank you for your contribution to our devices! I've moved on to another phone a while now, but it's because of you that I still can use my Mini as a backup.

 

[otmitia]

Hi pibarrette, I want to express my gratitude in supporting the device, I also went to another motorola, Moto X Force))

 

[dDefy]

I'm still with my Droid Mini

 

[GeneralSchnook]

Bluetooth Calls on 2015 Jeep Grand Cherokee Uconnect

TL;DR
Bluetooth calls on Jeep GC 2015 causes Lineage with GAPPs to soft reboot phone. No audio goes through bluetooth.

Wow, glad I found this thread. A little bit of background:

1. I got my xt1080m Droid Maxx a few years ago when they were still new, and immediately wished I had an unlocked bootloader. At the time, was locked down and didn't have root or anything. Oh well, phone worked good enough. Froze some bloat apps and installed a custom launcher and all was well.
2. I'm addicted to root, and couldn't resist it. Started following some of the happenings with JCase and Sunshine, plus a few blackhat/defcon talks about the particular exploit being used to bypass the locked down bootloader. Sounded promising. When the OTAs came, I disabled the update app and was determined to sit on my unpatched phone until Sunshine would work (basically just needed a root 'sploit).
3. My impatience led me to seek out and pay some sketch Chinese dude for an unlock code. Didn't work and now the Chinese have my IMEI. At least I got a refund...
4. After a few months of waiting on my old version, my patience payed off!! Rooted the phone, ran sunshine. BAM. Bootloader unlocked. At this point life got busy, and I had a working phone so no further action for a good chunk of time (couple of years at least). Keep in mind I was sitting on the unpatched version of the phone the whole time, and at one point I ran out of space from d/l'ing too many things, so just last week I was sitting on 64+ un-updated android apps too. Not a very good security practice...

Enter October 2017. I finally decided to get my act together with this phone. Figured enough time has passed that there would be a few decent ROMs for this thing. Kind of surprised there aren't more, but I guess with the amount of work it took to unlock the bootloader, not terribly surprising. Glad to see this has been worked on relatively recently.

Alright, here's what I have done so far:

1. Factory wiped my old, un-updated phone from 2013 or however long ago it was. Just used stock recovery for this...
2. Took all OTAs to make sure all the modems and such were up-to-date. To put in perspective how terribly out-of-date my phone was, I think it took 7 OTAs before I was on SU6-7.7, lol
3. Wiped in stock recovery one more time, just for good measure
4. fastboot installed TWRP 2.8.4.0
5. wiped via TWRP (it failed, probably because whatever the stock recovery does to wipe confused it or something)
6. let phone boot to "initialize" the modem or whatever (I am kind of a n00b on how android firmware and modem stuff actually work together)
7. wiped in TWRP (it worked this time)
8. let SU6-7.7 boot one more time. Checked the modem version to make sure everything was good (real reason I wiped was because I forgot to do it the first time, lol). I noticed that even though I had applied 7 OTAs and wiped twice, the root stuff still was there from when I sunshined way back years before this (so I guess I was working with a dirty install at this point).
9. uploaded 8Oct Lineage from this thread to internal storage along with GAPPS stock.
10. wiped system,cache,dalvik,data. Applied lineage zip. Wiped Cache/Dalvik. Applied GAPPS. Wiped Cache/Dalvik. Booted
11. noticed right away GPS did not work. Wiped modemst1 and modemst2. Gps works now.
12. paired bluetooth to my car. Used GAPPs dialer to make a call. No sound through car. Car suddenly is disconnected from phone. Try to end phone call (Red phone hang-up icon). Phone becomes unresponsive. Soft Reboots. Crap, everything else works perfectly. BUT IT IS A PHONE AND IF IT CAN'T MAKE CALLS, DEFEATS THE PURPOSE :laugh:
13. clean install lineage again but with GAPPS pico (maybe it's GAPPs dialer's fault?). Same thing.

Okay, so now that I am here, I decided to RSDLite the thing to the stock CFC SU6-7.7 zip file thingy (btw, where do these zips come from, and how do we know that there isn't a little extra something-something added to, idk say the system.img for nefarious purposes?). Running fully updated stock again. Bluetooth works with my car just fine.

My next step is to repeat the whole process from a clean install (probably going to stick with pico for now, or maybe even try without GAPPs to see what happens). I have already looked at a logcat, but it wasn't terribly helpful (at least to me, but maybe I don't know what I am looking for). If it happens again I will probably post the relevant logcat stuff here.

But for now, any ideas?? If I can't figure this out, probably gonna upgrade. Any recommendations for good, Verizon, root/bootloader unlocked friendly phones?

 

[pbarrette]

TL;DR
Bluetooth calls on Jeep GC 2015 causes Lineage with GAPPs to soft reboot phone. No audio goes through bluetooth.

Okay, so now that I am here, I decided to RSDLite the thing to the stock CFC SU6-7.7 zip file thingy (btw, where do these zips come from, and how do we know that there isn't a little extra something-something added to, idk say the system.img for nefarious purposes?). Running fully updated stock again. Bluetooth works with my car just fine.

My next step is to repeat the whole process from a clean install (probably going to stick with pico for now, or maybe even try without GAPPs to see what happens). I have already looked at a logcat, but it wasn't terribly helpful (at least to me, but maybe I don't know what I am looking for). If it happens again I will probably post the relevant logcat stuff here.

But for now, any ideas?? If I can't figure this out, probably gonna upgrade. Any recommendations for good, Verizon, root/bootloader unlocked friendly phones?

One thing you could try is flashing the modem from the latest Moto-X (2013) model, codename "ghost".

Essentially, the phones have the same processor and modem hardware. Also, the LineageOS kernel and support libraries for our phone (obake) are the same as that for the ghost platform. In fact, to support obake originally, it looks like they had to split off a portion of the android_device_motorola_ghost repo into the android_device_motorola_msm8960dt-common repo and have both ghost and obake point to the common repo. So the whole LOS port started with ghost.

I don't think there's anyone left who really supports obake, but there are a couple who are still working on the Moto-X and sometimes they still contribute upstream to the LOS repositories.

So there's a chance that the modem from "XT1060_GHOST_VERIZON_5.1_LPAS23.12-39.7-1_cid2_CFC" will be more compatible with what LineageOS expects from it.

Essentially, you'd need to flash the NON-HLOS.bin file, clear modemst1 & 2 and flash the fsg.mbn file from the ghost firmware.

The ghost firmware's fsg.mbn file contains modem config binaries for ghost, obake, obakem (the Mini) and obake_maxx (the Maxx). I assume that because it's the same platform, Motorola just compiled it all together and bundled it all together without regard to which phone the stuff was going into. There's also a "sasquatch" modem in there, but I don't know what phone that is.

As far as someone adding nefarious bits to the CFC firmware.. I suppose it's possible, but it would take a lot of work to bypass the Moto signed binaries and tamper checks. If someone could do that, they'd probably be selling a post 4.2 bootloader unlock instead. Or a simple to use root method for locked bootloaders, because these images can be installed on phones that still have the bootloader locked. So you'll just have to take it on faith, basically.

I mean, you installed LineageOS provided by some random idiot on the internet (me) on the faith that I haven't planted anything nefarious in there.
If it is a concern, then you should compile your own LOS build from source. But then you'll need to verify that nobody has planted malicious code in the LOS repos, so it's really just turtles all the way down at some point.

Question to everyone following this thread:
Is anyone else having reboots during calls?

 

[GeneralSchnook]

Thanks for the reply. Based on what you said, are you sure you didn't add a little extra something to your lineage build ? I'm just messing with you

Tried completely restoring to stock via the CFC file for my model, so I know I was starting with a clean phone. Still encountered the same problem, even without flashing GAPPS (only lineage).

I also tried it with a different bluetooth handsfree device (my Bose speaker), and the same error was present. Although sometimes it doesn't reboot the phone it seems, but it sure takes a while for the dialer to end the call (and then usually reboots).

Important to note that I don't have any issues if no handsfree/headset/phone bluetooth devices are connected.

I also followed your suggestion and tried with the MOTO X modem and fsg, but it didn't change anything.

Took a closer look at the logcat, (and even compared the output to a normal non-bluetooth call), but it basically just told me that bluetooth messed up and then the phone rebooted. Wasn't very helpful.

 

[pbarrette]

Thanks for the reply. Based on what you said, are you sure you didn't add a little extra something to your lineage build ? I'm just messing with you

Tried completely restoring to stock via the CFC file for my model, so I know I was starting with a clean phone. Still encountered the same problem, even without flashing GAPPS (only lineage).

I also tried it with a different bluetooth handsfree device (my Bose speaker), and the same error was present. Although sometimes it doesn't reboot the phone it seems, but it sure takes a while for the dialer to end the call (and then usually reboots).

Important to note that I don't have any issues if no handsfree/headset/phone bluetooth devices are connected.

I also followed your suggestion and tried with the MOTO X modem and fsg, but it didn't change anything.

Took a closer look at the logcat, (and even compared the output to a normal non-bluetooth call), but it basically just told me that bluetooth messed up and then the phone rebooted. Wasn't very helpful.

The reboot reason would probably be caught in the kernel messages.
You could try something like this while starting a call:
adb shell cat /proc/kmesg

That should dump the kernel messages continuously, so you'll catch the problem as it occurs.

Another one is:
adb shell logcat -b all

Logcat only reports the "main", "system" and "crash" buffers by default, but there are also "radio" and "events" buffers that can be reported.

 

[GeneralSchnook]

logcat -b all is pretty neat, able to see a lot more messages.

In particular, a whole bunch of underflow messages inside a bluetooth module:

10-22 17:11:43.125 13452 13488 E bt_btif : ## Audio Congestion (iterations:4 > max (3))
10-22 17:11:43.135 13452 13488 W bt_btif : ### UNDERFLOW :: ONLY READ 0 BYTES OUT OF 512 ###
10-22 17:11:43.135 13452 13488 W bt_btif : btif_media_aa_prep_sbc_2_send underflow 9, 0
10-22 17:11:43.149 13452 13488 W bt_btif : ### UNDERFLOW :: ONLY READ 0 BYTES OUT OF 512 ###
10-22 17:11:43.149 13452 13488 W bt_btif : btif_media_aa_prep_sbc_2_send underflow 9, 0
10-22 17:11:43.159 13452 13488 W bt_btif : ### UNDERFLOW :: ONLY READ 0 BYTES OUT OF 512 ###
10-22 17:11:43.159 13452 13488 W bt_btif : btif_media_aa_prep_sbc_2_send underflow 9, 0

These messages repeat a fair amount, then it looks like something tries to suspend BT:

10-22 17:11:43.472 13452 13477 D bt_btif : AV Sevent(0x41)=0x1223(STR_SUSPEND_CFM) state=3(OPEN)
10-22 17:11:43.472 13452 13477 E bt_btif : bta_av_suspend_cfm: suspend failed, closing connection
10-22 17:11:43.473 13452 13477 D bt_btif : AV Sevent(0x41)=0x120a(API_CLOSE) state=3(OPEN)
10-22 17:11:43.473 13452 13477 W bt_avp  : avdt_scb_snd_stream_close c:0, off:0
10-22 17:11:43.473 13452 13477 W bt_btif : bta_dm_rm_cback:1, status:6
10-22 17:11:43.474 13452 13472 D bt_btif : btif_av_state_started_handler event:BTA_AV_SUSPEND_EVT flags 1  index =0
10-22 17:11:43.474 13452 13472 D bt_btif : ## ON A2DP SUSPENDED ##
10-22 17:11:43.474 13452 13472 D bt_btif : ## a2dp ack : A2DP_CTRL_CMD_SUSPEND, status 1 ##
10-22 17:11:43.474  4134 14349 I bt_a2dp_hw: a2dp_command: A2DP COMMAND A2DP_CTRL_CMD_SUSPEND DONE STATUS 1

After that, looks like something is trying to close bluetooth, followed by a whole bunch of these messages:

10-22 17:11:43.532 13452 13488 E bt_btif : ERROR Media task Scheduled after Suspend

The reboot seems to happen because of something making the watchdog timer angry:

10-22 17:12:47.539  4450  5055 I Watchdog_N: dumpKernelStacks
10-22 17:12:47.652  4450  5055 E Watchdog: Triggering SysRq for system_server watchdog
...
10-22 17:12:47.705  4450  5055 W Watchdog: *** WATCHDOG KILLING SYSTEM PROCESS: Blocked in handler on main thread (main)

---------- Post added at 05:49 PM ---------- Previous post was at 05:47 PM ----------

Well, I actually need to use my phone without this nonsense, so I am going back to stock for now...

 

[pbarrette]

logcat -b all is pretty neat, able to see a lot more messages.

In particular, a whole bunch of underflow messages inside a bluetooth module:

10-22 17:11:43.125 13452 13488 E bt_btif : ## Audio Congestion (iterations:4 > max (3))
10-22 17:11:43.135 13452 13488 W bt_btif : ### UNDERFLOW :: ONLY READ 0 BYTES OUT OF 512 ###
10-22 17:11:43.135 13452 13488 W bt_btif : btif_media_aa_prep_sbc_2_send underflow 9, 0
10-22 17:11:43.149 13452 13488 W bt_btif : ### UNDERFLOW :: ONLY READ 0 BYTES OUT OF 512 ###
10-22 17:11:43.149 13452 13488 W bt_btif : btif_media_aa_prep_sbc_2_send underflow 9, 0
10-22 17:11:43.159 13452 13488 W bt_btif : ### UNDERFLOW :: ONLY READ 0 BYTES OUT OF 512 ###
10-22 17:11:43.159 13452 13488 W bt_btif : btif_media_aa_prep_sbc_2_send underflow 9, 0

These messages repeat a fair amount, then it looks like something tries to suspend BT:

10-22 17:11:43.472 13452 13477 D bt_btif : AV Sevent(0x41)=0x1223(STR_SUSPEND_CFM) state=3(OPEN)
10-22 17:11:43.472 13452 13477 E bt_btif : bta_av_suspend_cfm: suspend failed, closing connection
10-22 17:11:43.473 13452 13477 D bt_btif : AV Sevent(0x41)=0x120a(API_CLOSE) state=3(OPEN)
10-22 17:11:43.473 13452 13477 W bt_avp  : avdt_scb_snd_stream_close c:0, off:0
10-22 17:11:43.473 13452 13477 W bt_btif : bta_dm_rm_cback:1, status:6
10-22 17:11:43.474 13452 13472 D bt_btif : btif_av_state_started_handler event:BTA_AV_SUSPEND_EVT flags 1  index =0
10-22 17:11:43.474 13452 13472 D bt_btif : ## ON A2DP SUSPENDED ##
10-22 17:11:43.474 13452 13472 D bt_btif : ## a2dp ack : A2DP_CTRL_CMD_SUSPEND, status 1 ##
10-22 17:11:43.474  4134 14349 I bt_a2dp_hw: a2dp_command: A2DP COMMAND A2DP_CTRL_CMD_SUSPEND DONE STATUS 1

After that, looks like something is trying to close bluetooth, followed by a whole bunch of these messages:

10-22 17:11:43.532 13452 13488 E bt_btif : ERROR Media task Scheduled after Suspend

The reboot seems to happen because of something making the watchdog timer angry:

10-22 17:12:47.539  4450  5055 I Watchdog_N: dumpKernelStacks
10-22 17:12:47.652  4450  5055 E Watchdog: Triggering SysRq for system_server watchdog
...
10-22 17:12:47.705  4450  5055 W Watchdog: *** WATCHDOG KILLING SYSTEM PROCESS: Blocked in handler on main thread (main)

---------- Post added at 05:49 PM ---------- Previous post was at 05:47 PM ----------

Well, I actually need to use my phone without this nonsense, so I am going back to stock for now...

Have you seen otmitia's post here:
https://forum.xda-developers.com/showpost.php?p=72892345&postcount=283

You may have to disable A2DP for your bluetooth device and use the headset profile instead.

 

[dDefy]

....
I mean, you installed LineageOS provided by some random idiot on the internet (me) on the faith that I haven't planted anything nefarious in there.
If it is a concern, then you should compile your own LOS build from source. But then you'll need to verify that nobody has planted malicious code in the LOS repos, so it's really just turtles all the way down at some point.

Question to everyone following this thread:
Is anyone else having reboots during calls?

Nope, all working great here, after I've followed your steps. Also don't have any problem using the A2DP bluetooth protocol.

I know pbarrette have offered help to complile my own LOS based ROM, so when you'll have a some time please make a tutorial/help on that.
It is not urgent as I know you've moved over another phone, but I still like my Droid Mini