Lineage based port for Xperia 10 IV (murray)

From the screenshot and desc it looks like that one main camera lens is working

The main camera and selfie camera both work fine. Same as the 10 iii.

It is possible that a toggle like that which @nephros developed for the 10iii could be made to work here in order to get the other two lenses going, but I haven’t had the chance to try it yet. I am doubtful that the exact same solution used to enable three cameras on the 10iii would work given that we a) don’t have a r/w /vendor partition and b) don’t seem to have the /vendor/etc/camera/ directory. Best case scenario it can be done with an overlay file.

Alternatively, the Camera2 API would get all four lenses working. But I don’t think video recording works on Camera2 yet. Again, haven’t tried it.

4 Likes

Out of curiosity, how did you decide to spend over 1 month for a 10iv? I mean the phone is already quite old, so the port might not be useful for very long.

Is it:

  1. You don’t care, you had a 10iv and too much free time?
  2. The 10iv it’s an easy target due to be close to the 10iii?
  3. You wanted to gain experience and later you’ll start a more difficult port?
  4. …
1 Like

The reason is, there are a lot of people who have a 10iv in the cupboard waiting for SFOS, as Jolla planned to support this phone a while ago.

2 Likes

So my 10IV doesnt boot. It shows the Sony-Logo for a long time but thats it.

This phone has flashed with various lineageOS and half-ready SFOS-ports. So I flashed it back to android, flashed LineageOS 22.2 like suggested. Android was working, Lineage too. But after flashing your SFOS-port it hangs on Sony-Logo (no error occured while flashing).
Telnet is not answering, but perhaps thats to expect, as dmesg shows this:


[...]
[108769.212587] [ T250200] usb 1-9: new high-speed USB device number 48 using xhci_hcd
[108769.394554] [ T250200] usb 1-9: New USB device found, idVendor=0fce, idProduct=0dde, bcdDevice= 1.00
[108769.394567] [ T250200] usb 1-9: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[108769.394572] [ T250200] usb 1-9: Product: Android
[108769.394577] [ T250200] usb 1-9: Manufacturer: Google
[108769.394581] [ T250200] usb 1-9: SerialNumber: HQ632A1FB7
[108787.640186] [ T250200] usb 1-9: USB disconnect, device number 48


Any idea how to investigate further?

Because:

  • It is a very nice phone that I spent a lot of money on a few years ago, and it was in my pocket already
  • I know it works on Australian telco networks (which is no longer a guarantee for phones not sold in this country)
  • There was no significant monetary expense involved (ie. I did not have to drop upwards of $700AUD on an Xperia 10 VII or similar, I just bought an $80 dumbphone to answer calls on for a month while I was porting. Remember Australia is very big and a long way away from everyone else. Everything is more expensive here because it has to sit on a boat for three months to get here)
  • I knew there were other people on this forum who owned a 10 IV (or potentially even 10 V) who would benefit from it in some way
  • If I bought some other new phone (ie. not an expensive, grey-market Sony phone) and ported it instead, I’d inevitably end up with a device significantly bigger / wider, as all new phones are. In my view, even the 10IV is big enough.
10 Likes

Bugger - I’ve sent you a DM, we’ll try to debug it. Maybe I got my instructions wrong, sorry. It could also be you’re in the wrong slot.

2 Likes

All sorted - I have updated the instructions, there shouldn’t be any trouble now. Sorry about that!

12 Likes

You should be sorry. How dare you make the world a better place and be so helpful.

7 Likes

Followed the new instructions and booted SFOS 5.1 successful and happily!

EDIT1: Mine is a XQ-CC54

EDIT2: calling works on german o2 on 5G (no calling sound tho), 4G and 2G!

5 Likes

Hi @mettska, so glad to hear it!

Interesting to learn about 5G calls. I don’t have a 5G service in my area so that was untested. To be honest, I did not realise there was a difference between 5G and 4G in terms of VoLTE. I will look into the problem but I don’t know how much I’ll be able to solve seeing as I don’t have it available to debug with.

At any rate, I’m glad the port works on the XQ-CC54 as well. Great news!

Edit: reading the SFOS 5.2 release notes, perhaps this isn’t a problem specific to this port and it may be improved by changes coming in 5.2.

3 Likes

i actually made a port for the 10V based on your port and pushed the changes to github. i’m on holiday with this device though so i will test it when I’m back home

19 Likes

This is great news @smatkovi. Thanks for doing this. Look forward to seeing the port after your holiday.

Small update - I made some changes to the kernel configs tonight to satisfy Waydroid’s requirements. I have never used Waydroid before so am still wrapping my head around how to set it up, but it looks like it might work if I manage to get it configured right. Giving it a rest for the evening now, but watch this space - we may have Android Apps running again one day.

10 Likes

Did you try to install the official AAS rpms? It even works on the 5 IV and V ports as far as I understand.

2 Likes

No, I didn’t save a copy of them. But if they did work it would be neat. Obviously it is not documented what is required to make them work because you’re not ā€œsupposedā€ to make them work. That said, part of the motivation for attempting this port was that Jolla said AppSupport might actually be allowed on it - so if that were to happen I’d be more than willing to entertain the possibility.

In the meantime, as I say, I’ll attempt getting waydroid running.

3 Likes

Great! I had in plan to try this to adapt for 10V but im noob in porting so was thinking that i must have more free time… i looking forward how it will work :wink:

I tried to get the official AAS-rpms working on the Community-Port. It doesnt work. I dont understand why - and ChatGPT was turning circles for hours but didnt figure out, why AAS is not starting.
The problem could lay in the difference of HIDL on official port and the usage of AIDL in some parts on the Community-Port.

2 Likes

Interesting, thanks for the feedback @mettska.

Indeed, the Lineage base seems to make more use of AIDL hardware interfaces than the SODP base. I’m not completely surprised that things don’t map over 1:1. From what I have read, big changes happened in Android 15 in this regard.

Even within HIDL interfaces there are significant differences. For example, the nagara port has android.hardware.audio@7.0, and we have android.hardware.audio@6.0 - despite our Android base being one version newer than theirs.

It’s possibly a similar reason as to why the official fingerprint software doesn’t talk properly. I suspect both that and the AAS solution are very device-dependent and heavily rely on having the right hardware interfaces available from the Android base.

I am slowly learning more about how these sorts of questions can be answered, but for now unless someone more well-versed on the topic can weigh in, I’ve got a bit of work in front of me. At any rate, I will try to get the kernel changes I worked on yesterday published tonight, so if you are running this port and particularly bored, have a go at updating with zypper ref && zypper dup in twelve hours time. (I was unsuccessful, come back again tomorrow). You may then be able to start reading up on debugging Waydroid.

5 Likes

Required kernel configs for Waydroid should be there if you update now - Thanks again @mettska for being my beta tester! These added kernel configs may also help AAS but given the HIDL / AIDL troubles I am not sure.

Similarly, getting Waydroid running is now just a case of getting the software talking to the hardware which I’m still working on. It appears that the gap between Android 13/14 and 15 is more significant than I’d hoped.

6 Likes