PPCGeeks

PPCGeeks (http://forum.ppcgeeks.com/index.php)
-   Android On TP2 Development (http://forum.ppcgeeks.com/forumdisplay.php?f=319)
-   -   NAND Testing - 05-25 Update: New LK, Recovery.img, Kernel Updates through Recovery (http://forum.ppcgeeks.com/showthread.php?t=134598)

tpdjr 12-12-2010 09:59 PM

Re: NAND Boot Testing - 12-11: Panel power off and on now working!
 
hey nate is there anyway i can use f22 rootfs so i can have the end key be recent apps on nand or do i have to use the custom rootfs?

[ACL] 12-12-2010 11:57 PM

Re: NAND Boot Testing - 11-29: New Framebuffer, gralloc, and libGLES
 
Quote:

Originally Posted by mpannen (Post 2011860)
I ran a trace copying what Bzo used for his run of haret. The first trace was the action of turning my screen on, the second trace was the action of turning the screen off:

HaRET(8)# wirq 10 irq:8002b418@0xa03443f8=810d1

Not sure if this is any different than Bzo's trace.

Yep this is a completely different panel.. i'm still trying to figure out how to check which init to use. Looks like there is about 4 types of panels to look out for.

[ACL] 12-13-2010 12:01 AM

Re: NAND Boot Testing - 11-29: New Framebuffer, gralloc, and libGLES
 
Quote:

Originally Posted by guruvince (Post 2012057)
Nate, you should mirror this thread over to XDA. Nand is making great progress, and you could at least get a few more testers over there. I bet there are a bunch of people who dont even realize there is a thread on this topic.

Reason i didnt mention this on xda is because of the noobery out there. I'm trying to avoid the floods like reef has when people try to use the build. This is why i try to keep a nice tight team of testers here who can help with the progress. I'm maintaining the source but its nice to know Nate can build the code at anytime and help deploy. ALso the rest of the gang is very savy and can help dump haret info so they have really helped out with development.

And of course the last reason is that i dont have a gsm phone and since xda is mostly gsm we will simply get tons of "it does not work fix it" kind of replies..

[ACL] 12-13-2010 12:03 AM

Re: NAND Boot Testing - 12-11: Panel power off and on now working!
 
Quote:

Originally Posted by mpannen (Post 2012309)
Im trying to figure out how ACL got the init code from Bzo's trace--There must be a lot of work to go from that trace to his initialization code, LOL!!

I cheated... i just matched some of the numbers then disassembled wince libs to get the actual init that wince uses.

pays to know a little asm

[ACL] 12-13-2010 12:05 AM

Re: NAND Boot Testing - 12-11: Panel power off and on now working!
 
Quote:

Originally Posted by pts69666 (Post 2012341)
Ok, RHOD400 here. I seem to be having a little problem. Flashing the NBH file goes perfectly according to plan, however when I boot up the device (after already having the modules, system.ext2, rootfs.img, rhod400.kb, and gralloc.msm7k.so files in my andboot folder) It starts to load, the text scrolls, I get the two vibrates and then the screen goes blank. I know that the device is still on, because it doesn't reboot when I hit the power button. It does nothing. I would just wait for it to finish booting and then just power on, however the screen calibration is supposed to happen at some point during the initial boot. The screen doesn't come back to allow me to do that.

Any Ideas?

EDIT:

ok, i decided to let it sit for a while and then i hit the power key and it randomly turned on. However it appeared to be frozen it said
mdp irq already on 4 4
ndp_dma: busy

and if i hit enter those two lines just keep repeating

Yep.. this happens to me as well. Our code calls the panel resume during boot and once it does that the panel simply goes blank. Its not till android takes over till it seems to come back to life.

Ignore the mdp irp .. those will eventually be commented out. They arent related to the panel .. hopefully next patch i will make sure the panel is always on

[ACL] 12-13-2010 12:07 AM

Re: NAND Boot Testing - 12-11: Panel power off and on now working!
 
Quote:

Originally Posted by tpdjr (Post 2012358)
hey so basically nand is just as good as going haret now right?

nope.. nand has .35 frame buffer :-) .. so its slightly ahead but behind in some areas. We dont have auto brightness since that seems to just kill our panel.

pts69666 12-13-2010 12:15 AM

Re: NAND Boot Testing - 12-11: Panel power off and on now working!
 
Quote:

Originally Posted by [ACL] (Post 2012472)
Yep.. this happens to me as well. Our code calls the panel resume during boot and once it does that the panel simply goes blank. Its not till android takes over till it seems to come back to life.

Ignore the mdp irp .. those will eventually be commented out. They arent related to the panel .. hopefully next patch i will make sure the panel is always on

How long should the initial boot take? For me, the screen doesn't come back on until I hit the power button, and it doesn't let me do that for a good 15 minuets or so. From there, it just sits and doesn't do anything...

I'll let it sit longer, to see what happens.

tpdjr 12-13-2010 12:38 AM

Re: NAND Boot Testing - 12-11: Panel power off and on now working!
 
Quote:

Originally Posted by [ACL] (Post 2012473)
nope.. nand has .35 frame buffer :-) .. so its slightly ahead but behind in some areas. We dont have auto brightness since that seems to just kill our panel.

Well thats seems like nand is definitely ahead of haret to me because auto brightness doesn't seem to be critical for most people. Plus i have a rhod 400 so nand seems perfect because with nand you don't have to worry about messing up other devices with updates because it is only for rhods am i correct? hey is battery life just as good as haret though cuz im going to go nand now i was just curious thanks again for the quick response ACL and all the hard work.

coolsilver 12-13-2010 12:48 AM

I retried from scratch with Autiobuild not renaming modules file. Boots faster again. GPS working quickly with GPS status and agps data.

I have a good stripe at the bottom of the phone that is corrupt video. My phone was also only half initiallizing the back light. It would come on but had a shade of red. I am glad it wasn't my lcd failing. I manually set in settings all the way to the left and powers brightly again... though not staying on as before (auto bl code)

Thanks.

pts69666 12-13-2010 01:35 AM

Re: NAND Boot Testing - 12-11: Panel power off and on now working!
 
Ok, after an hour and 15 min of trying to boot, it still isn't working. Any other ideas as to what I'm missing?


All times are GMT -4. The time now is 06:38 PM.

Powered by vBulletin® ©2000 - 2025, Jelsoft Enterprises Ltd.
©2012 - PPCGeeks.com


Content Relevant URLs by vBSEO 3.6.0