23/Sep: Kernel Source

[Guest rjm2k]

Excellent! B)

And have they responded to that request in a form that should engender hope or despair?

No specific response but they are communicating and they are releasing source/updates so no reason to assume we wont get what we need even if it does take some polite prodding.

Who's gonna work on porting to Gingerbread then?!

[Guest oh!dougal]

No specific response but they are communicating ...

So, since this is the critical time, maybe a polite reminder would not be unreasonable?

Its surely important that they acquire the habit of taking code snapshots (for later publication), rather than retrospectively re-creating.

Edited November 23, 2010 by oh!dougal

[Guest mELIANTE]

If it interests anyone I uploaded it to megaupload. Faster than ZTE's servers.

http://www.megaupload.com/?d=WF1EJV2F

[Guest twatter]

If it interests anyone I uploaded it to megaupload. Faster than ZTE's servers.

http://www.megaupload.com/?d=WF1EJV2F

thanks R

[Guest tom619]

will we able to get froyo from the latest update?

[Guest rjm2k]

will we able to get froyo from the latest update?

Froyo should be coming Mid-december in China etc along with the source, so fingers crossed some of the guys will get it ripped and re-done for the sanfran.

[Guest tom619]

sounds promising thanks!

[Guest kag8]

so what can devs do with the latest source release? can i look forward to more custom roms? B)

[Guest kallt_kaffe]

so what can devs do with the latest source release? can i look forward to more custom roms? B)

I've posted a kernel with the new source here: http://android.modaco.com/index.php?s=&amp...t&p=1486598

I had to manually remove a duplicate #endif in one file and remove an external variable in the new LCD driver to make it compile. Works fine on my Orange San Francisco but I'm unsure if the new LCD driver works as it should on the Hungarian (or other LCD based) blades. I got a feeling the external variable is supposed to be used by other parts of the kernel but I've rgrep:ed the whole three and it's not used anywhere else. It's like they missed to include another updated file somewhere.

[Guest Sebastian404]

I've posted a kernel with the new source here: http://android.modaco.com/index.php?s=&amp...t&p=1486598

I had to manually remove a duplicate #endif in one file and remove an external variable in the new LCD driver to make it compile. Works fine on my Orange San Francisco but I'm unsure if the new LCD driver works as it should on the Hungarian (or other LCD based) blades. I got a feeling the external variable is supposed to be used by other parts of the kernel but I've rgrep:ed the whole three and it's not used anywhere else. It's like they missed to include another updated file somewhere.

<snip>

any ideas where I went wrong?

apart from getting out of bed... ignore me, stupidity strikes :P

Edited November 30, 2010 by Sebastian404

[Guest hecatae]

http://support.zte.com.cn/support/news/New...?newsId=1000322 zte-blade?

http://support.zte.com.cn/support/news/New...?newsId=1000304 san francisco?

http://support.zte.com.cn/support/news/New...?newsId=1000382 vodafone 945 from 1st november 2010

can someone check if the 945 is a zte blade with a 5mp cam?

Edited December 6, 2010 by hecatae

[Guest Sebastian404]

http://support.zte.com.cn/support/news/New...?newsId=1000322 zte-blade?

http://support.zte.com.cn/support/news/New...?newsId=1000304 san francisco?

http://support.zte.com.cn/support/news/New...?newsId=1000382 vodafone 945 from 1st november 2010

can someone check if the 945 is a zte blade with a 5mp cam?

The first two links point to the same thing, and that's been posted a few posts up,

The third one points to original (and more broken) version of the source.

[Guest reallordx]

...

can someone check if the 945 is a zte blade with a 5mp cam?

If information @ gsmarena is correct, then no.

According to gsmarena (and other sources):

Vodafone 945 has lower resolution (400x320), smaller screen size (3,2) and a LED Flash.

[Guest rjm2k]

Not Kernel source specific but I thought I'd try get something out of ZTE about recent developments, no definate answers and lots of reading between the lines needed.....

"we have noticed that there are new versions of the Blade in the UK with a TFT screen instead of OLED, there also seems to be some new software to support it (OUK_P729BV1.0.0B08) which seems to work for both OLED and TFT models, can you confirm that it was designed for both? Is the move to TFT a permanent thing or just whilst there is an OLED shortage?

Also, it’s been posted that the Greek blade will get an upgrade to Froyo, do you know if this is the case?

We have been working on P729CUV1.0.0B04 which was made available by someone in China, can you confirm if this is the final version for Unicom or just another Beta? "

Response

". The newest code we released had made a compatible between OLED and TFT. So the new software can support both OLED and TFT models blade.

The issue that blade moving to TFT implicate the ZTE phone strategy matter, so please forgive me that i can not give you any informations.

ZTE do not have any plan to update the Greed Blade to Froyo. If we decide to upgrade the phone, i think you can get the information from an open media.

As i know, the P729CU for China Unicom havn't appeared on the market. So i think the P729CUV1.0.0B04 is not the final version for Unicom. "

Edited December 6, 2010 by rjm2k

[Guest oh!dougal]

Not Kernel source specific but I thought I'd try get something out of ZTE about recent developments, no definate answers and lots of reading between the lines needed.....

...

Response

". The newest code we released had made a compatible between OLED and TFT. So the new software can support both OLED and TFT models blade.

The issue that blade moving to TFT implicate the ZTE phone strategy matter, so please forgive me that i can not give you any informations.

...

I would not expect any information to be forthcoming as to their future plans.

It would seem to be counter-productive to ask for that.

On the other hand, I do feel that they ought to be pressed on their obligation to release accurate source for EVERY open-source product that they ship.

It would seem very bad manners (at least) to ask them ANY questions about any 'leaks' - and unlikely to assist in developing a productive relationship with a manufacturer who's commercial sensitivity is already causing them to withhold code that they are obliged to release.

And, I don't recall them confirming that they actually understood that the notification led does not operate in (real) sleep. Wasn't there a claimed problem with finding a phone that even had the led fitted?

I don't have any hope for any issues being resolved unless they are understood!

Also, the last I heard regarding wifi was that 1/ they (utterly wrongly) believed that the wifi-sleep-never policy setting actually worked and 2/ that they were only interested in router compatibility with Chinese domestic market products - which seemed so extraordinary that I hope it wasn't allowed to pass without comment.

Edited December 7, 2010 by oh!dougal

[Guest rjm2k]

ZTE have updated their source....

http://support.zte.com.cn/support/news/New...ndroid%E2%80%A6

"Hi guys, now i have upgraded our code that incorrectly mapped power button issue has been fixed.

And also i use a new msm_battery.c instead of the incorrect one.

There may be help for you.

When ZTE make public this news, I will publish the froyo source code.

"

[Guest Sebastian404]

ZTE have updated their source....

http://support.zte.com.cn/support/news/New...ndroid%E2%80%A6

"Hi guys, now i have upgraded our code that incorrectly mapped power button issue has been fixed.

And also i use a new msm_battery.c instead of the incorrect one.

There may be help for you.

When ZTE make public this news, I will publish the froyo source code.

"

I wonder what else they broke fixing this version :)

Is it worth telling them that if they are distributing busybox with the devices (its in most of the ROMS we have seen) they also have other requirments....

http://www.busybox.net/FAQ.html#commercial_use

[Guest rjm2k]

I wonder what else they broke fixing this version :)

Is it worth telling them that if they are distributing busybox with the devices (its in most of the ROMS we have seen) they also have other requirments....

http://www.busybox.net/FAQ.html#commercial_use

I'll do this after we get froyo source, tbh I don't think they have a clue about their license obligations and I don't want to p*ss them off until we can just about get by without them ie when we've had some froyo source and maybe a little bit of "support" with it.

[Guest hero protagonist]

... a little bit of "support" with it.

I am way OT, but I just gotta say that this thread is quite entertaining :)

[Guest JolyonS]

Interestingly

* linux/drivers/mmc/host/mxcmmc.c - Freescale i.MX MMCI driver

*

* This is a driver for the SDHC controller found in Freescale MX2/MX3

* SoCs. It is basically the same hardware as found on MX1 (imxmmc.c).

* Unlike the hardware found on MX1, this hardware just works and does

* not need all the quirks found in imxmmc.c, hence the seperate driver.

*

imxmmc.c no longer included - wonder if this will have any bearing on the SD card problems people have been having.

[Guest bagge]

I doubt there are any plans whatsoever for ZTE to ever release gingerbread for the Blade, and right now all the focus should be on getting a froyo source ofc. But I just read the LG Optimus One, which is a budget phone similar to the Blade will be receiving it officially, so maybe maybe there would be a slim chance of bothering them for a source for it at some later stage, if nothing else maybe some unofficial code that hasnt been thoroughly tested if its not THAT much work for them... One can always dream, right? :)

[Guest HAL18000]

I am a complete newbie in Android development and I have a question:

I guess that these kernel sources contains the Android kernel + device-specific drivers.

What do we need to port these sources to Froyo or Gingerbread ?

My understanding is that each new Android version is based on a new kernel version. Is this kernel version available to anyone ?

If so, in theory what is needed is to port old drivers to the new kernel, am I right ?

Edited December 10, 2010 by HAL18000

[Guest mELIANTE]

Here's another link, just for backup and for anyone that's still interested in download it.

http://www.megaupload.com/?d=K0DN7SEB

Edited December 10, 2010 by mELIANTE

[Guest gameSTICKER]

I am a complete newbie in Android development and I have a question:

I guess that these kernel sources contains the Android kernel + device-specific drivers.

What do we need to port these sources to Froyo or Gingerbread ?

My understanding is that each new Android version is based on a new kernel version. Is this kernel version available to anyone ?

If so, in theory what is needed is to port old drivers to the new kernel, am I right ?

I'd be glad if someone can answer this for me too...

[Guest rjm2k]

A lot of drivers are closed source so the kernel only has stubs to call throught to the binary driver. Since we cant compile drivers for a new kernel there is no guarantee old drivers will work with it.