Jump to content

Recommended Posts

Guest luca020400
Posted

Why you have removed the latest stable kernel ??

Can you reupload it ??

Posted (edited)

Do any one of you really know how much phone socs their is with totally fuli open sourced code without property blobs?

Well their is totally NONE!

I whose searching for RIL or at least WiFi driver (fully functional) in source code & haven't find any.

This whose place I looked for:

http://redmine.replicant.us/projects/replicant/wiki/ReplicantStatus

It's not up to Huawei at all to publish them nor they can do that as those represent other companies (Qualicomms) IP rights. If it whose a HiSylicon SoC you could try to preserved them to do some work about it, this way you are only barking on a wrong tree. I hope this will put an end to any future discussion about this.

Edited by ZolaIII
Posted

Works in the stable kernel (26.07.) Paired Bluetooth hands free in the car, or do I have to stay with the stock?

Posted (edited)

I can't test Bluetooth too much, because I don't have any devices to pair with.

@Zola: I am aware of the conflict with licenses, especially when we talk about code of Qualcom. However for almost every device there are OpenSource versions of the driver available and many SoC manufacturers are making use of them. There are enough developers for custom android kernels not having to face such issues in getting WiFi working properly.

For our ATH6KL exists OpenSource drivers as well, which are working, too. The problem with WiFi tethering is actually the user-space application. The version present on our ROM is incompatibel to those driver versions. Dazzozo compiled the wpa_supplicant for CM on his own and thus it works with the open coded kernel driver.

Edited by Guest
Posted

Another reason, why other companies don't release any WiFi sources is: they don't have to, if those are already included into Linux kernel sources ;)

Guest luca020400
Posted

Why you have removed the latest stable kernel ??

Can you reupload it ??

Up ;)

My first up XD

Posted (edited)

I removed it, because it wasn't stable according to several user reports. Regarding Romagnolo1973's feedback I've re-upload the build of 26.07.2014 into the stable folder.

Edited by Guest
Posted

On 26/07 build the second core doesn't gets activated if i disable MP-Decision !

Someone confirm?

Posted

Since that kernel version doesn't have auto-hotplug, that's an absolutely expected behavior ;)

Posted (edited)

On 26/07 build the second core doesn't gets activated if i disable MP-Decision !

Someone confirm?

It will if you after that change governor or frequency parameters. At least it should.

It just need a little push as it stays remarked in off power state when you turn MP decision off...

Edited by ZolaIII
Posted

Logcat sometimes got some crazy output:

I/DEBUG   ( 9040): debuggerd committing suicide to free the zombie!

Aaaaaalright. Let's free the zombies? huh.png

Posted

Logcat sometimes got some crazy output:

I/DEBUG   ( 9040): debuggerd committing suicide to free the zombie!

Aaaaaalright. Let's free the zombies? huh.png

 

I got that on stock too  :D

Posted (edited)

Logcat sometimes got some crazy output:

I/DEBUG   ( 9040): debuggerd committing suicide to free the zombie!
should switch off the phone at night then,can't say when the zombie gets free :wacko:

:lol:

Edited by SH3H1
Guest Romagnolo1973
Posted
Posted

Thx Roma, although I know about that thread already happy.png

 

Overclocking our CPU might be possible on "PLL_2" level. But since this would increase bus-frequencies as well, I dropped playing with it.

Posted

It's hard to get real technically informations from Qualcomm and ARM. They keep a lot of stuff I could need non-public for manufactures only. So I am forced to go the trial-and-error-way most of the time sad.png

Guest luca020400
Posted (edited)

So is a lot of usless work :(

How the kernel development is going ??

Which is the kernel version ( 3.4.X ) ??

Edited by luca020400
Posted

The version behind "3.4" isn't really important, because it only reflects the patch-state in terms of mainline patches.

 

Going with mainline patches is another problem amongst many. Since we got a custom Linux kernel by Qualcomm, not all of the mainline patches are suitable unfortunately. It would be different if it was an "vanilla 3.4.0 kernel".

 

Some patches are incompatible to our sources and I need to be more selective what to add and what not. It's a lot less work on Nexus-series phones for example mellow.png

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.