NAND Testing - 05-25 Update: New LK, Recovery.img, Kernel Updates through Recovery

[arrrghhh]

Originally Posted by [ACL]

i think we are done with the CM talk. Not nand related.

ive decided to find the good fight on the panel. Others have been asking for headset but i think panel gets priority.

Agreed, on both fronts. Panel definitely gets priority. 3.5mm headset just works for blackstone, and camro described it as an easy change... Doesn't seem to easy, but we'll see.

[natemcnutty]

Originally Posted by arrrghhh

Agreed, on both fronts. Panel definitely gets priority. 3.5mm headset just works for blackstone, and camro described it as an easy change... Doesn't seem to easy, but we'll see.

Actually, I thought I'd take a look into the 3.5mm headset since camro was saying to just rip the contents from the blackstone board and change the gpio's. The problem is that I can't get haretconsole to log any gpios for me. I'd love to help map them out, but it just doesn't detect anything

Anyway, I'm still learning a ton, and for now, I'm back on WM with Haret until there is something on NAND that needs testing. I still haven't been able to get the latest commits to work on my device. Seems to have been broken since the hw3d change was implemented. Not sure why.

[arrrghhh]

Originally Posted by natemcnutty

Actually, I thought I'd take a look into the 3.5mm headset since camro was saying to just rip the contents from the blackstone board and change the gpio's. The problem is that I can't get haretconsole to log any gpios for me. I'd love to help map them out, but it just doesn't detect anything

Anyway, I'm still learning a ton, and for now, I'm back on WM with Haret until there is something on NAND that needs testing. I still haven't been able to get the latest commits to work on my device. Seems to have been broken since the hw3d change was implemented. Not sure why.

Keep fighting the good fight! We all appreciate your work

[natemcnutty]

Wirelessly posted (Opera/9.80 (Windows Mobile; Opera Mini/5.1.21594/20.2489; U; en) Presto/2.5.25)

Originally Posted by arrrghhh

Keep fighting the good fight! We all appreciate your work

None of it is my work. It's all the devs. I'm just trying to learn and help them in any way I can.

[hceuterpe]

Originally Posted by [ACL]

How is Nesoid actually scaling? i want the actual gpu to do the scaling for us not it be handled by a library or android. Android itself will think its running a completely different resolution.

.35? haha.. new kernel. We are running .27 while everyone else is probably running .32. We would run on cleaner code and about 2.5 years worth of improvements. Not to mention bragging rights. You can actually tell someone "your kernel is dog shit.. .35 bitch !!!"

turns our blackstone has headset working, so im going to see if we can use the same code for rhod. Lets cross our fingers.

Hate to rain on anyone's parade, but I always thought it was weird to say the TP2 was running Froyo, yet had only a .27 kernel. Correct me if I'm wrong, but isn't every Android version built from a particular kernel version? So Cupcake (v1.5) is .27, Donut (v1.6) .29, Eclair (v2.0/2.1) on .29 still, and of course Froyo .32?

I personally would feel irked jumping ahead with .35 just yet, but Gingerbread is based on .35 so it wouldn't hurt to think of at least. It might be the software developer mentality in me to want to stick to official versions at least somewhat in ports. I've been reading some of the specific details of the differences between the different versions--I would think at least some of those changes have to be kernel related, and it would definitely help the TP2?

For instance, our phones have Bluetooth 2.1 EDR, and the Android bluetooth stack is based on both BlueZ which in turn is built into the kernel and that doesn't support BT 2.1 until BlueZ v4.x which starts at Android v2.0 and at least kernel .29 (we have 3.3x something). I wonder if Bluetooth would be easier to get working if we simply moved to .32? I've heard all sorts of problems related to Bluetooth in Android versions below 2.0, and that's only with phones that only have Bluetooth 2.0 (not BT 2.1 like do).

So I don't sound like a total bitch, I can try to contribute by merging the kernel code from .27 to .32? Man, that's such a huge jump!! :O

[[ACL]]

Originally Posted by hceuterpe

Hate to rain on anyone's parade, but I always thought it was weird to say the TP2 was running Froyo, yet had only a .27 kernel. Correct me if I'm wrong, but isn't every Android version built from a particular kernel version? So Cupcake (v1.5) is .27, Donut (v1.6) .29, Eclair (v2.0/2.1) on .29 still, and of course Froyo .32?

I personally would feel irked jumping ahead with .35 just yet, but Gingerbread is based on .35 so it wouldn't hurt to think of at least. It might be the software developer mentality in me to want to stick to official versions at least somewhat in ports. I've been reading some of the specific details of the differences between the different versions--I would think at least some of those changes have to be kernel related, and it would definitely help the TP2?

For instance, our phones have Bluetooth 2.1 EDR, and the Android bluetooth stack is based on both BlueZ which in turn is built into the kernel and that doesn't support BT 2.1 until BlueZ v4.x which starts at Android v2.0 and at least kernel .29 (we have 3.3x something). I wonder if Bluetooth would be easier to get working if we simply moved to .32? I've heard all sorts of problems related to Bluetooth in Android versions below 2.0, and that's only with phones that only have Bluetooth 2.0 (not BT 2.1 like do).

So I don't sound like a total bitch, I can try to contribute by merging the kernel code from .27 to .32? Man, that's such a huge jump!! :O

not really.. i know android does have some minimum requirements and i think for froyo it was .29.. but we later found out that was not the case. But then again our kernel has so much crap backported from higher kernels its hard to even say where we are at anymore. For example in nand we are running the usb gadget framework which is from .32 .. yet we run it happily.

jump to .32 is not worth it. If we are gonna move up might as well go all the way. Also the biggest reason why its so hard to jump is because we support so many devices. So we can move rhod, but the other devices would be left behind. Getting all the devs to agree on stuff isnt alwys easy. You should read some of the logs ..lol

wanna contribute ? joint us on #htc-linux

[[ACL]]

**** so i wanted to commit the panel fix today but im missing something. So now we can power down the lcd and bring it back up .. but when it comes back up .. boom death.. need to check with fellas tomorrow and see what im missing. Once thats done, we can finally power down the damn lcd properly.

[damage_them]

being that there are new kernals out how do us nand users go about updating these with out having to bug you guys all the time ..if its possible??..thanks

[natemcnutty]

Wirelessly posted (Opera/9.80 (Windows Mobile; Opera Mini/5.1.21594/20.2489; U; en) Presto/2.5.25)

Originally Posted by damage_them

being that there are new kernals out how do us nand users go about updating these with out having to bug you guys all the time ..if its possible??..thanks

I am more than happy to generate new packages, and I have been pretty on top of them. The problem is that there is something not quite right with the repo and the builds haven't been working since the hw3d commit.

ACL has been awesome in giving his time to the nand project, and I know there is probably a decent amount of code he has that hasn't been committed. I'm hoping we can figure out what is missing to get the compiles working again.

[[ACL]]

Originally Posted by damage_them

being that there are new kernals out how do us nand users go about updating these with out having to bug you guys all the time ..if its possible??..thanks

the good news is that there arent that many updates.. trust me i commit to the main kernel .. :-p . i try to make sure they are as close as possible. Keep in mind they are different repos.

those kernels floating around are tests. Nothing solid, so when there are big changes that are worth building, i will let yall know. I havent had a chance to hunt down Netripper, i want to ask him to help us do an autobuild for nand. He handles autobuild for HD2 so it shouldnt be that bad. Only other guy is glemsom but he is rarely seen on irc anymore..

I need to work also with natemcnutty since he can also build the nand kernel.