| ||
[WIP] ClockworkMod Recovery/CyanogenMod for DAWA D7/Augen Gentouch78
#41
Posted 27 December 2010 - 10:26 AM
#43
Posted 27 December 2010 - 09:57 PM
http://wenku.baidu.c...2707220829.html
It's for earlier kernel, but in part 3.2.2 and 3.2.3 there is 2 different type of drivers in the kernel and HW mod.
I'll join the development/testing when i receive iMito IM7. Plan to set up it in my car with external GPS/Bluetooth and USB audio card (hope it will be real with working open kernel) ;-)
#44
Posted 28 January 2011 - 06:55 PM
ML
#45
Posted 28 January 2011 - 07:41 PM
Quote
sorry, no progress for Dawa/Augen ;)
ithamar found workaround and/or solution about touchscreen issue. I need to talk with him... but no enough time
#46
Posted 29 January 2011 - 06:08 AM
Quote
ithamar found workaround and/or solution about touchscreen issue. I need to talk with him... but no enough time
I'm working on a Smartbook Surfer tablet, and have gotten a hack working using their kernel. I will try to create the same patch with the Dawa/Augen kernel and see how it goes. The main problem seems to be that the minimum/maximum values that the touchscreen kernel driver reports are way off, and therefore all the touch events don't come through (as they are seen as out of range).
I've played a little with tslib support in fun_'s build but that does not seem to really fix the problem, not with the pointercal file that was supplied with my tablet, at least.
By now, I'm doing builds for my tablet with out of AOSP master, just plain Android builds (2.3 atm) and hope to have a fully working build available soonish.
Just waiting for my ordered tablets to arrive, so I can start building a working kernel for all these tablets, as without kernel source we can only work around using hacks
Ithamar.
#47
Posted 30 January 2011 - 05:44 PM
I have the OLD ( as in July 2010 ) source for the Augen Kernel and part of the system image....this is obviously for 1st Gen Augen's, but if you think it'll help you...I'll dig it up and if you tell me where to post it, I'll put it up as a point of comparison.
OTOH, if I can lend a hand, let me know.
#48
Posted 03 February 2011 - 01:03 AM
Kory
#49
Posted 03 February 2011 - 08:49 AM
Quote
Cool, that would be very helpful. I'm collecting as much telechips tablet kernel trees as possible, so we can (hopefully) merge them into one main source tree in the near future. My tablet with serial has arrived, should be able to start playing with it next week, if all goes well (and work doesn't keep me away from it).
If you could upload those files to rapidshare or anywhere, it would be very much appreciated!
As soon as I get something working for my tablet, I will post it on the working code on my github account, where I am collecting as much sources as possible to get a proper clean Android build for the telechips tablets together.
Who knows, we might get 3.0 running on these ;)
#50
Posted 03 February 2011 - 08:59 AM
Basically, the magic hex values in the patch are taken from running getevents through adb and capturing the min/max values for both X and Y. I hacked android to use these values instead of what the driver returns, which is 0-1000 which is obviously wrong.
The last thing I do is for any event containing an X value, I "mirror" the value so it comes out correct. This was all that was required.
The values used in the patch might not work for the Augen, I don't have one so I couldn't check the values returned.
Hope this helps someone!
#51
Posted 03 February 2011 - 06:50 PM
Kory
#52
Posted 04 February 2011 - 05:24 AM
Quote
Kory
You'll need to do a full android build, preferably from fun_'s repositories @ github. Problem is, his repositories are a little behind the upstream CM repositories so won't build out-of-the-box. You need to merge in the latest CM changes. This is quite a bit of work.
I'll see if I can find some time to do a build again, and upload a system.img with the patch applied...
Ithamar.
#53
Posted 04 February 2011 - 08:36 AM
Quote
I'll see if I can find some time to do a build again, and upload a system.img with the patch applied...
thank you for your great work. I'm in the middle of updating my repo... hopefully it will be done this weekend...
#54
Posted 04 February 2011 - 08:58 AM
Kory
#55
Posted 04 February 2011 - 11:15 AM
Quote
Standing on the shoulders of giants, would never have gotten this far so quickly without your github repository
Anyway, if you're interested in including the patch, be warned that it will seriously disrupt the touchscreen functionality on builds using kernels that return 'proper' values, so we will either have to produce libui.so as a patch for augen-like devices or make this hack optional through properties or so. I haven't tried, as I want to get a kernel build for my Surfer that fixes the problem, so I can continue to do "master" builds out of the generic Android repository.
If you run into any probs with the patch, just ping me and I'll help.
Ithamar.
#56
Posted 04 February 2011 - 07:07 PM
Quote
Could you capture the output of "adb dmesg" for me, and post it here (as attachment, I suspect it'll be quite a bit of text)? Please also report with it if you are running an official Augen build or wether your running a modded build (e.g. the CM froyo builds or so).
Thanks!
#57
Posted 04 February 2011 - 08:12 PM
Note: I'm on Ubuntu. Not sure if that will have any bearing on anything.
Kory
#58
Posted 06 February 2011 - 05:26 AM
Quote
Note: I'm on Ubuntu. Not sure if that will have any bearing on anything.
Thanks! I'm working from Ubuntu too, but the log you captured are the kernel messages from the device, so it should not really matter (except possibly for line-endings in the text files).
Ithamar.
#59
Posted 07 February 2011 - 10:39 AM
Kory
#60
Posted 08 February 2011 - 05:55 AM
I'm sure once I have it working on the X6D unit it should be trivial (even without serial) to get a kernel booting and working on the Gentouch78/Surfer/D7 as well.

Help











