ics, linaro is a better choice - Kindle Fire General

http://releases.linaro.org/11.11/android/images-ics-preview/mainline-panda/
http://www.linaro.org/downloads/
http://android.git.linaro.org/gitweb
panda is also omap4

riverzhou said:
http://releases.linaro.org/11.11/android/images-ics-preview/mainline-panda/
http://www.linaro.org/downloads/
http://android.git.linaro.org/gitweb
panda is also omap4
Click to expand...
Click to collapse
Slight bit of English might not hurt?

What is this?

What's there not to understand? Pandaboard is one of the most popular OMAP4 targets.
Sent from my SAMSUNG-SGH-I777 using xda premium

pyrostic said:
Slight bit of English might not hurt?
Click to expand...
Click to collapse
http://i16.photobucket.com/albums/b12/tehglow/manytank.jpg
that is many tank?
yes?

Linaro is an Alliance of ARM TI IBM Samsung Freescale ST-Ericsson . It's working for Open source Software for Linux on the ARM Cortex-A family of processors , Android on ARM, Unbuntu on ARM , linux/anrdoid toolchains and some libs .
They release their own source distribution Optimizated for Cortex-A8 and Cortex-A9 , and image distribution for PandaBoard(TI OMAP4430) , Origen(Samsung Exynos4210) , Snowball(ST-Ericsson's Nova A9500) , QuickStart(Freescale i.MX53) .
In one word , linaro android distribution will be better for A8 and A9 processors , much more Optimization .

This looks interesting, how do I test it out?

I just took a look at this and I don't get it sort off. It seems to be to be a version of android for like the PandaBoard.etc. However it doesn't come in the form of a zip but an img. I reckon with a bit of porting we could get it to work somehow but would it be worth it?

XperiaPlayer said:
I just took a look at this and I don't get it sort off. It seems to be to be a version of android for like the PandaBoard.etc. However it doesn't come in the form of a zip but an img. I reckon with a bit of porting we could get it to work somehow but would it be worth it?
Click to expand...
Click to collapse
They have a git repo to pull the sources from, so devs could base roms off of this instead of AOSP. I don't see the advantage, but it's another option I guess~

XperiaPlayer said:
I just took a look at this and I don't get it sort off. It seems to be to be a version of android for like the PandaBoard.etc. However it doesn't come in the form of a zip but an img. I reckon with a bit of porting we could get it to work somehow but would it be worth it?
Click to expand...
Click to collapse
You can get source as same as CM or AOSP.
repo init -u git://android.git.linaro.org/platform/manifest.git -b linaro-android-11.12-release
repo sync
Linaro Android distribution will be better for A8 and A9 processors , much more Optimization .
And , you can use gcc4.5.4(linaro version) or gcc4.6.3(linaro version) to compile this Andoid distribution but CM or AOSP only for gcc4.4.3 from google.

shobon said:
They have a git repo to pull the sources from, so devs could base roms off of this instead of AOSP. I don't see the advantage, but it's another option I guess~
Click to expand...
Click to collapse
It's a full optimization version for A8/A9 platform. It's all.

Anyway, linaro gcc is the best compiler for arm.
It can be use to compile android kernel , ever for CM or AOSP .

Related

IDEA : git repository

I find that many developers are working on porting android 2.1 or 1.6 vanila and it result in much duplicated work.
I was thinking of maybe setting up 2 git repositories one for 1.6 AOSP and 2.1 AOSP on which everyone can work on, improve and fix bugs.
lgstoian said:
I find that many developers are working on porting android 2.1 or 1.6 vanila and it result in much duplicated work.
I was thinking of maybe setting up 2 git repositories one for 1.6 AOSP and 2.1 AOSP on which everyone can work on, improve and fix bugs.
Click to expand...
Click to collapse
That's a nice idea, im begging to look after a site to host the repositories...
We can use github for hosting or any other site. I'm interested in looking after the repositories too.
A centralized effort will always be better then a fragmented one.
lgstoian said:
We can use github for hosting or any other site. I'm interested in looking after the repositories too.
A centralized effort will always be better then a fragmented one.
Click to expand...
Click to collapse
If you have alot of money, then Github would be ok... But it costs $50/month for the package that is minimum for two sets of android source code...
Know any places where we can host the code?
Mod. edit: not dev related, moved to general
Gitorious and Github are both free and unlimited for open source projects.
We can use sourcefordge too. The offer svn and git repositories. So what about it folks, who would be interested?
I would be interested in this...
I think we should open a project on sourcefordge for android 2.2 with the first task being to get the new kernel (2.6.32) working on the tattoo.
It could host the long awaited universal froyo upgrade
-------------------------------------
Sent via the XDA Tapatalk App
lgstoian said:
I think we should open a project on sourcefordge for android 2.2 with the first task being to get the new kernel (2.6.32) working on the tattoo.
Click to expand...
Click to collapse
That will be hard becuase the new kernel (2.6.32) is very big so we would need something like GRUB to boot it, from another partition than the boot partition...
If we disable enough modules the size should go down but that would mean it won't be universal any more.
There could be a regular version and a 'lite' version for Tattoo and older phones.
Wait wait wait.
I can play full divx movies smoothly so some seldom used modules could be stored on sd with no prob.
-------------------------------------
Sent via the XDA Tapatalk App

Linux on A500: The Future

We have had native Linux on our Iconias for quite a while now but the builds and kernels are getting outdated, Unity is here and there is no real direction as to what needs to be accomplished.
This thread has been started principally for two kinds of people:
Kernel and Linux developers whom can offer advice or indeed actual code
Iconia enthusiasts for suggestions / ideas as well as any logic they may have
This thread does not aim to make Linux on our Iconias poof out of nowhere. It requires a huge amount of work to get going although with proper information exchange it should be much easier to get the project going.
The current ToDo list is as follows (feel free to post me more):
RootFS on /data/linux
Kernel initrd which will chroot to /data/linux
Linux Kernel will boot off mmcblk0p7
Upgrade to Ubuntu 12.04 (although RootStock is now deprecated, we can just download a pre-built bundle for the arm-eabi FS, use live-build or get a pre-made RootFS. There's lots of options for this)
Build the kernel from scratch (this is kind of obvious anyway). Building for armhf sounds like a good idea
Full graphics acceleration from the Tegra 2 for Unity
Perhaps a build of Backtrack (this shouldn't be too difficult - they offer an image for arm. If it can share the kernel then we should be okay)
A501 3G Modem
USB Huawei Modems (although initial Googling seems to imply the compatibility of usbserial.ko and hence OOTB compatibility so long as a udev rule is added)
Camera, sensors and non-patchy wifi (issue on 2.6.38 although the nvidia drivers below should fix this)
Latest Linux kernel (ie 3.1.x)
We have plenty of resources at our disposal such as:
Dual Boot Bootloader: http://forum.xda-developers.com/showthread.php?t=1619913
ICS Kernel source: http://global-download.acer.com/GDF...AB&Step3=A500&OS=a08&LC=en&BC=Acer&SC=EMEA_27
A 2.6.x chromeos-based kernel with quite a bit of functionality implemented: https://github.com/astarasikov/iconia-gnu-kernel
Linux for Tegra which provides many drivers for porting Linux to the Tegra chips: http://developer.nvidia.com/linux-tegra
Kexecboot - a bootloader on a kernel image system which could be placed on mmcblk0p2 and can give us huge flexibility in terms of launching a kernel (http://kexecboot.org)
I've looked inside the Linux for Tegra package and a HUGE amount is in there including x configurations, hardware drivers, Broadcom wifi firmware and the list goes on.......
There's also a codec package so we can have native video acceleration in loads of formats.
The bootloader will, at least initially, probably be the default Acer one. We could utilise kexecboot in the future as noted above
What we need:
A system for organising the code. CI & trac?
Perhaps a wiki.... or at least somewhere to organise ideas and reversed data
Either role assignment or a huge number of 'issues' with the code on a system such as trac and anybody can fix any issue
What works without kernel hacking:
Sound: The Wolfson WM8903 sound chip has driver provided by Wolfson which have been integrated directly into the official Linux kernel source (source: http://opensource.wolfsonmicro.com/content/linux-drivers-wolfson-devices)
Wi-Fi (almost) - the broadcom drivers are available with a simple apt-get. They are also available through Linux for Tegra
Graphics Acceleration (Hardware) - the drivers and codecs come from Linux for Tegra
armhf - Support for the hard float capabilities is built right into the Tegra2 and both Ubuntu and Linux for Tegra provide this support alongside the (seemingly legacy!) armel
The RootFS can now be built with qemu-debootstrap
Please let me know if I forgot anything else (which I undoubtedly did!)
I'm all in. I can do whatever is needed. Looking For Backtrack more than Ubuntu but i'd imagine once we get the Kernels sorted any distro will be easy to make work. Looking forward to the FUN.
First, we have 3.0.0 kernel by sp3dev (having issues), I'll take look on that one. Eventhough kernels 3.0.y and 3.2.y (not 3.1.10 - that's EOL!) are good enough.
Secondly, kernel does not chroot, a script in initrd image does chroot (this isn't any magic). As in the previous question about cmdline, it's just say "dynamic configuration" for the kernel (as where to find rootfs etc.). In the linux kernel, it's hardcoded right now (needs to be undefined).
First, it's better to get points 1) - 3) working for now.
blackthund3r said:
There is a strong focus on the preliminary uboot bootloader, however. I'm not sure if this would replace the Acer bootloader or be chainloaded? Ideally we just want a boot img on mmcblk0p7. Can somebody confirm this for me?
Click to expand...
Click to collapse
No that's not possible. U-boot will be standalone bootloader (i.e if u-boot, then no Acer BL). Right now, my patch in Acer BL gives us a "sub-bootloader" (don't want to call it second-stage bootloader, because it always returns back to Acer BL, eventhough we submit the image as a byte array (and not a partition)).
I think its a good idea to start from 3.1.10 that's on nv git and has linux variant too...
thor2002ro said:
I think its a good idea to start from 3.1.10 that's on nv git and has linux variant too...
Click to expand...
Click to collapse
I think its a good idea someone cares to share his android kernel source. I tried to rebase my cros patches on top of nvidia android branch and most stuff seems to boot, but fb is not working.
I think I'll make a new version of uboot that will be able to boot custom kernel from a file/script in data partition and stock kernel from recovery partition. I have made a good port of uboot to sgs2 and now I know how to handle boot mode detection via gpio. Sadly I'm not using my iconia at all lately.
P.S. Everything works on 2.6.38 except camera and sensors. Wifi is unstable. What else could anyone probably want
Add to todo. 3G modem for A501
While the dual-boot bootloader is nice, it's not the best solution, in my opinion. Kexecboot would offer a nice multi-boot solution with a GUI. We'll have the ability to compile kernels on the native Linux environment without flashing every time. By simply passing arguments to the kernel (the one kexecboot loads), we can have it boot off of any partition Linux can see. This would eliminate any restrictions that the dual-boot bootloader may have. This would also be beneficial to users who prefer booting from non mmcblk0p7, non boot-img systems. (I personally prefer something along the lines of mmcblk1p1, with no boot img). Kexecboot already works for the Tegra 2, and has touchscreen abilities, so porting wouldn't be too terribly difficult. See here and here.
As far as distros go, we should look at something compiled for armhf. That'll help increase speeds. The tegra2 drivers already have a hard float pre-release available to play with.
If I think of anything else, I'll edit this post.
Added all your changes
What exactly is the aim of this thread?
Is it for the improved quality of different Linux distros in general of the Tab?
Is it for the creation of a single distro which will run decently?
Is it for just for spitballing ideas about the above ideas?
Stating a single goal will help us accomplish said goal with greater ease.
apapousek said:
What exactly is the aim of this thread?
Is it for the improved quality of different Linux distros in general of the Tab?
Is it for the creation of a single distro which will run decently?
Is it for just for spitballing ideas about the above ideas?
Stating a single goal will help us accomplish said goal with greater ease.
Click to expand...
Click to collapse
Threefold:
1) to collaborate what we have
2) to port and produce a fully working Ubuntu distro (complete including all patches, drivers etc)
3) if possible, make a backtrack port too
Sent from my Iconia A500 using Tapatalk 2
blackthund3r said:
Threefold:
1) to collaborate what we have
2) to port and produce a fully working Ubuntu distro (complete including all patches, drivers etc)
3) if possible, make a backtrack port too
Sent from my Iconia A500 using Tapatalk 2
Click to expand...
Click to collapse
1. A wiki would be nice. I have server space & bandwidth available.
2. That's not too difficult, especially with Ubuntu on the AC100 being tegra2
3. I've had one working for a while. It's just not too slick.
apapousek said:
1. A wiki would be nice. I have server space & bandwidth available.
2. That's not too difficult, especially with Ubuntu on the AC100 being tegra2
3. I've had one working for a while. It's just not too slick.
Click to expand...
Click to collapse
1. I have server space too - it powers my flash tool, ICS Root etc. I can install media wiki overnight - not an issue
2. In theory. It still has to be done though and there'll be loads of bugs. We want perfection
3. That's what we aim to change
Sent from my Iconia A500 using Tapatalk 2
Im all in particularly looking forward to backtrack but Ubuntu will be awesome if we can get it running without many if any bugs....
should we be focusing on ubuntu i mean there alot more lightweight os's such as DSL, Slitaz, unity linux, and puppy linux that would be alot less of a resource hog ;-)
gears177 said:
Im all in particularly looking forward to backtrack but Ubuntu will be awesome if we can get it running without many if any bugs....
should we be focusing on ubuntu i mean there alot more lightweight os's such as DSL, Slitaz, unity linux, and puppy linux that would be alot less of a resource hog ;-)
Click to expand...
Click to collapse
The Tegra 2 + 1GB RAM puts the A500 up there as one of the most powerful dual core tabs on the market. Ubuntu will be fantastic.
Sent from my Iconia A500 using Tapatalk 2
blackthund3r said:
The Tegra 2 + 1GB RAM puts the A500 up there as one of the most powerful dual core tabs on the market. Ubuntu will be fantastic.
Sent from my Iconia A500 using Tapatalk 2
Click to expand...
Click to collapse
kk looking forward to it pm me if i can help
gears177 said:
kk looking forward to it pm me if i can help
Click to expand...
Click to collapse
Sure. What skills do you have?
Sent from my Iconia A500 using Tapatalk 2
gears177 said:
Im all in particularly looking forward to backtrack but Ubuntu will be awesome if we can get it running without many if any bugs....
should we be focusing on ubuntu i mean there alot more lightweight os's such as DSL, Slitaz, unity linux, and puppy linux that would be alot less of a resource hog ;-)
Click to expand...
Click to collapse
gears177 said:
kk looking forward to it pm me if i can help
Click to expand...
Click to collapse
Okay, so none of the distros you named could be ran on our tablet. They're all Intel/AMD only. We have an ARM device. It's a beautiful, low-power architecture, that only a few distros support. Arch Linux runs fairly smooth on it, even when running from a Class 4 external microsd card.
If you want to help, learn how to compile a kernel, the differences between x86 and ARM, and about some basic user-space tools which interact directly or almost directly with the kernel.
Honestly, I'm not trying to be an a-hole, but having a plethora of volunteers does nothing if the volunteers don't know how it works.
apapousek said:
Okay, so none of the distros you named could be ran on our tablet. They're all Intel/AMD only. We have an ARM device. It's a beautiful, low-power architecture, that only a few distros support. Arch Linux runs fairly smooth on it, even when running from a Class 4 external microsd card.
If you want to help, learn how to compile a kernel, the differences between x86 and ARM, and about some basic user-space tools which interact directly or almost directly with the kernel.
Honestly, I'm not trying to be an a-hole, but having a plethora of volunteers does nothing if the volunteers don't know how it works.
Click to expand...
Click to collapse
I suppose that's a good point.
Anyhow, now we have loads of volunteers, should we assign roles?
Sent from my Iconia A500 using Tapatalk 2
blackthund3r said:
I suppose that's a good point.
Anyhow, now we have loads of volunteers, should we assign roles?
Sent from my Iconia A500 using Tapatalk 2
Click to expand...
Click to collapse
oh sorry :'( i can build from source and compile kernels but not very good have to learn and refine more sorry alot of my scripting is sloppy :-\
gears177 said:
oh sorry :'( i can build from source and compile kernels but not very good have to learn and refine more sorry alot of my scripting is sloppy :-\
Click to expand...
Click to collapse
I meant no offense by my comment at all. You're inquisitive, which is very very necessary for this project.
I would like to work a bit of the kernel side with gears.
Kernel side question. Do we want it to be compiled for ARM HF? It would give us speed benefits, above all else.

Firefox OS

Mozilla have launched Firefox OS based on Linux kernel ... Its open source ... I would like any of our good developers to look forward towards developing this OS for our phones
More detail here: http://www.mozilla.org/en-US/b2g/
would be hard if there's no propietary driver kernel source..
It's possible.
Here's manifest: https://github.com/mozilla-b2g/b2g-manifest
I'm too busy to port it for now. Maybe someone else can do that.
i will give it a try to port it
This isn't android based... it's not a ROM it's an other OS... i think if it's possible it will be very hard
i have knowledge of unbranded os types
It's not android. But it is linux and android is also linux, so I think drivers can be ported from android to this, no (though no api stuff and such will be the same)?
I assume it'll be hard, much harder than ics and jelly bean, but still possible.
But I also think that it won't run as good/smooth as ics and in the future jb will run.
So it might run, but if it's useful is another question...
Nevertheless it would be awesome to see it running on the gio.
Also does it need to run on it's own or is it possible to run it on top of the android kernel? If this is the case then it becomes rather easy as we have enough stable android kernels for the gio.
Marcin, is that the purpose of the link you posted?
marcin1147 said:
It's possible.
Here's manifest: https://github.com/mozilla-b2g/b2g-manifest
I'm too busy to port it for now. Maybe someone else can do that.
Click to expand...
Click to collapse
I think it's Android based, it uses AOSP repo besided Linaro repo and Mozzila repo!
Code:
<remote name="aosp"
fetch="https://android.googlesource.com/" />
bgm92 said:
I think it's Android based, it uses AOSP repo besided Linaro repo and Mozzila repo!
Code:
<remote name="aosp"
fetch="https://android.googlesource.com/" />
Click to expand...
Click to collapse
but i see no apps in the manifest
tioammar said:
but i see no apps in the manifest
Click to expand...
Click to collapse
Of course not. It's based on android, so the core shares a lot with the core of android. This doesn't make it android.
The whole os is (except for the core stuff) web based, which probably means all apps need to be written in html 5 or maybe other web standards.
So android apk apps don't work, and so it makes no sence including them.
Stannieman said:
Of course not. It's based on android, so the core shares a lot with the core of android. This doesn't make it android.
The whole os is (except for the core stuff) web based, which probably means all apps need to be written in html 5 or maybe other web standards.
So android apk apps don't work, and so it makes no sence including them.
Click to expand...
Click to collapse
simple and great explanation.. thanks
currently trying to port give me some days
warface123 said:
currently trying to port give me some days
Click to expand...
Click to collapse
https://github.com/MaclawStudios
Here are sources for android ICS on GIO. Since the manifest uses android 4.0.4 I think you may need to use some things from this (don't forget to mention Maclaw and Marcin if you do).
Soon this will be updated to JB I think, I don't know if FFOS works with JB as well or it specifically needs ICS though.
About firefox OS: http://en.wikipedia.org/wiki/Firefox_OS
So it's not entirely android based. It only uses some parts for some of the hardware.
thanks for the sourches
It's based android. Because s2 users can install it.
its dam hard but i am making progress what do you want guys odin or cwm just wanna now
warface123 said:
its dam hard but i am making progress what do you want guys odin or cwm just wanna now
Click to expand...
Click to collapse
odin
alright it will be odin then now i gonna sleep but will continu tommorow
How much android is really in it? Mozilla made it so that it can run on devices that run android. So they probably just mean ARM with a certain ammount of ram etc...
Wikipedia says it's a linux kernel, some components of android for hardware communication, mozilla's own api's for app -> hardware communication such as phone, cam, speakers etc, and last but not least the gecko engine.
So if the kernel used is the same as the one used for android then you could use maclaw (and team)'s ICS kernel + the android stuff needed, merged with the b2g sources.
I'm not really a dev so it's probably a lot harder then what I'm saying here. But I think it's easier then a fully fledged android...
Were you able to boot something already?

4.3 kernel sources released! grab them here

http://www.sammobile.com/2013/11/05/samsung-releases-android-4-3-kernel-source-for-galaxy-s3/
Wow man latest news isn t it?? Using Internet Explorer??
And? Could you explain a bit in how it help us?
Thanks.
Enviado desde mi GT-I9300 usando Tapatalk
It means we can finally have custom kernels woooooo
Sent from my GT-I9300 using XDA Premium 4 mobile app
Well, I downloaded the source kernel and I need a clear way how to compile it smoothly. I wonder whether I can flash the kernel via Odin or CWM to a rooted stock I9300XXUGMJ9 firmware or not. Will it work? Any walkthroughs?
ahmedmelm said:
Well, I downloaded the source kernel and I need a clear way how to compile it smoothly. I wonder whether I can flash the kernel via Odin or CWM to a rooted stock I9300XXUGMJ9 firmware or not. Will it work? Any walkthroughs?
Click to expand...
Click to collapse
http://source.android.com/source/building.html
Read everything on left hand side
-----------------------
Sent via tapatalk.
I do NOT reply to support queries over PM. Please keep support queries to the Q&A section, so that others may benefit
rootSU said:
http://source.android.com/source/building.html
Read everything on left hand side
Click to expand...
Click to collapse
That's way cool. But any matter to accelerate the process? Would really compiling this kernel and pushing it to existing ROM work?
What do you think source code is exactly? What is it you think you want?
-----------------------
Sent via tapatalk.
I do NOT reply to support queries over PM. Please keep support queries to the Q&A section, so that others may benefit
I know this is not a Q&A thread but could I ask? Does this release of source kernel code mean, that Samsung finally released the source code for the dreaded mali GPU driver and the FM radio blobs that we so need to make CM roms fully functional and smooth?
granets said:
I know this is not a Q&A thread but could I ask? Does this release of source kernel code mean, that Samsung finally released the source code for the dreaded mali GPU driver and the FM radio blobs that we so need to make CM roms fully functional and smooth?
Click to expand...
Click to collapse
No. Read following for explanation.
dhiru1602 said:
Blobs are userspace components which correspond to the HAL (Hardware Abstraction Layer). It is this layer which is responsible for communication between the kernel and the operating system. GPU kernel driver is opensource, but the GPU HAL is closed source. This is the case with all the devices, but on our device, much more components of the HAL are closed source like FIMC, FIMG, HWJPEG, TVOUT. Samsung doesn't release the source code for this components because it doesn't necessarily have to. Without the source code, we are forced to use the compiled binary which may or may not work properly. Since we are left without options, we often mess around with the Android framework, making dirty hacks to get them to work. We will not be able to futureproof our device with newer versions of Android unless we have the source code since making edits on compiled binaries is not possible and hence we have to wait for official releases so that we could get newer binaries (blobs). Here is the list of all the blobs that we use on our device. Having proprietary code is not uncommon, but Samsung doesn't release the source code for essential parts of the OS like Graphics subsystem, OMX and DSP, which makes things more complex provided there is no documentation for the SOC, nor they are wiling to help developers.
Click to expand...
Click to collapse
dhirend_6d said:
No. Read following for explanation.
Click to expand...
Click to collapse
Thanks for the explanation. So basicaly those kernel sources don't mean much. And I was so excited
granets said:
Thanks for the explanation. So basicaly those kernel sources don't mean much. And I was so excited
Click to expand...
Click to collapse
They mean as much as every other kernel source release. We still dont have everything we need to make AOSP work properly, but we can now get custom kernels for 4.3 touchwiz roms.
rootSU said:
They mean as much as every other kernel source release. We still dont have everything we need to make AOSP work properly, but we can now get custom kernels for 4.3 touchwiz roms.
Click to expand...
Click to collapse
These include the latest mali r3p2. So we have now userspace binaries + kernel source drivers. It makes this possible, which is awesome. It's coming for all exynos4 devices.

ElementalX Kernel

Hi there people,
I've been hearing good things about ElementalX kernel, but from what I see there isn't this kernel for our phone, what do you think?
evronetwork said:
Hi there people,
I've been hearing good things about ElementalX kernel, but from what I see there isn't this kernel for our phone, what do you think?
Click to expand...
Click to collapse
Why don't just choose from available kernels instead? The dev of elementalX kernel is not supporting bacon because he does not have this device.
Mr.Ak said:
Why don't just choose from available kernels instead? The dev of elementalX kernel is not supporting bacon because he does not have this device.
Click to expand...
Click to collapse
Already using Franco kernel but why not try this one if it was possible(or if someone ported this to our device)? ...thread was created about this kernel and not talk about alternatives
P.S. I heared quite good stuff about this kernel that is all
evronetwork said:
Already using Franco kernel but why not try this one if it was possible(or if someone ported this to our device)? ...thread was created about this kernel and not talk about alternatives
P.S. I heared quite good stuff about this kernel that is all
Click to expand...
Click to collapse
But you need permission from the kernel developer first.
Mr.Ak said:
But you need permission from the kernel developer first.
Click to expand...
Click to collapse
If he says you need then ok if he doesn't and gives his source like that then no ...just port it and say it's his - problems solved. You get into technical problems while I comment on what you think about the kernel and/or if someone is willing to do a port(asking dev if needed OC)
evronetwork said:
If he says you need then ok if he doesn't and gives his source like that then no ...just port it and say it's his - problems solved. You get into technical problems while I comment on what you think about the kernel and/or if someone is willing to do a port(asking dev if needed OC)
Click to expand...
Click to collapse
Tbh I don't think anyone will.
Mr.Ak said:
Tbh I don't think anyone will.
Click to expand...
Click to collapse
tbh I'd rather watch devs working on 3.10 kernel(there's one for our phone) or newer e.g. 3.18 but that's a different matter
evronetwork said:
tbh I'd rather watch devs working on 3.10 kernel(there's one for our phone) or newer e.g. 3.18 but that's a different matter
Click to expand...
Click to collapse
Since erorcun's OPO's screen is broken he probably won't work on 3.18.xx.
Mr.Ak said:
Since erorcun's OPO's screen is broken he probably won't work on 3.18.xx.
Click to expand...
Click to collapse
Gather 10 skilled devs to work around 3.18.xx for a few months and you'll have a pretty decent stable core with quite a few benefits ...but kernel isn't like roms, needs fine tuning and when we're talking about upped versions there are some restrictions, you need to create workarounds for some stuff and some devs don't like to do that and since we've got no actual open source drivers workarounds are inevitable
evronetwork said:
Gather 10 skilled devs to work around 3.18.xx for a few months and you'll have a pretty decent stable core with quite a few benefits ...but kernel isn't like roms, needs fine tuning and when we're talking about upped versions there are some restrictions, you need to create workarounds for some stuff and some devs don't like to do that and since we've got no actual open source drivers workarounds are inevitable
Click to expand...
Click to collapse
Most of the devs think upstreaming kernel is just a waste of time and efforts.
Mr.Ak said:
Most of the devs think upstreaming kernel is just a waste of time and efforts.
Click to expand...
Click to collapse
Yeap true still it's the future and there won't be one if we stuck in 3.4.x (3.4.x already reached EOL - https://www.kernel.org/category/releases.html - linux kernels EOL )
P.S. Sultan tried to make a kernel but saw it would require too much time and effort and stopped, if he has time and passion about the kernel this would have a much better future. It's obvious that a proper optimized kernel does have benefits(of course to match the maturity of 3.4.x is not an easy thing ...still this would make our phone last 5+ years ....if a proper stable 3.18.x kernel was around) - not sure if the electronics of my 1+1 would live by then but just saying
evronetwork said:
Yeap true still it's the future and there won't be one if we stuck in 3.4.x (3.4.x already reached EOL - https://www.kernel.org/category/releases.html - linux kernels EOL )
P.S. Sultan tried to make a kernel but saw it would require too much time and effort and stopped, if he has time and passion about the kernel this would have a much better future. It's obvious that a proper optimized kernel does have benefits(of course to match the maturity of 3.4.x is not an easy thing ...still this would make our phone last 5+ years ....if a proper stable 3.18.x kernel was around) - not sure if the electronics of my 1+1 would live by then but just saying
Click to expand...
Click to collapse
I think 3.10.xx will reach EOL soon.
Mr.Ak said:
I think 3.10.xx will reach EOL soon.
Click to expand...
Click to collapse
Oct 2017
Mr.Ak said:
I think 3.10.xx will reach EOL soon.
Click to expand...
Click to collapse
arjun said:
Oct 2017
Click to expand...
Click to collapse
Yeap, wish we had some devs working on:
3.18.x for our device
bring treble for our device(not sure if this can be done) - not just have android O(without treble)
This would ensure that the phone would live until all the electronics die from natural cause(age)
I think @erorcun was working on 3.10 and then 3.18 before his screen broke.
I don't know how common it is for devices to have merged or official kernel jumps, but I remember the S3 starting with 3.0 and ending at 3.4. CM builds got it first and then miraculously followed suit in the final OTA, which nobody who cared about their bootloader ever took.
LLStarks said:
I think @erorcun was working on 3.10 and then 3.18 before his screen broke.
I don't know how common it is for devices to have merged or official kernel jumps, but I remember the S3 starting with 3.0 and ending at 3.4. CM builds got it first and then miraculously followed suit in the final OTA, which nobody who cared about their bootloader ever took.
Click to expand...
Click to collapse
Well oneplus won't give you anything ..that's a sure thing. Usually they don't up the kernel version. You get a phone with 3.10 it will die with 3.10.
The real problem comes with the drivers of the device. In order to get a real fully functional kernel without workarounds we need the help of qualcomm, would they provide drivers so we can have a real 3.18 kernel with proper support from android o? ...of course not, why do that when they can sell you a new phone with better SoC(they have a company to run and to do that they want to sell SoC ...so you should buy a new phone )
evronetwork said:
Well oneplus won't give you anything ..that's a sure thing. Usually they don't up the kernel version. You get a phone with 3.10 it will die with 3.10.
The real problem comes with the drivers of the device. In order to get a real fully functional kernel without workarounds we need the help of qualcomm, would they provide drivers so we can have a real 3.18 kernel with proper support from android o? ...of course not, why do that when they can sell you a new phone with better SoC(they have a company to run and to do that they want to sell SoC ...so you should buy a new phone )
Click to expand...
Click to collapse
If my OPO dies,my next device will be Redmi Note 4 which has 3.18.xx
Mr.Ak said:
If my OPO dies,my next device will be Redmi Note 4 which has 3.18.xx
Click to expand...
Click to collapse
Mediatek SoC ...good luck with that
Mediatek is worse than Qualcomm when it comes to open source roms
..but at least you'll have 3.18 which isn't bad at all heh
evronetwork said:
Mediatek SoC ...good luck with that
Mediatek is worse than Qualcomm when it comes to open source roms
..but at least you'll have 3.18 which isn't bad at all heh
Click to expand...
Click to collapse
Just get the version with the snapdragon 625
evronetwork said:
Mediatek SoC ...good luck with that
Mediatek is worse than Qualcomm when it comes to open source roms
..but at least you'll have 3.18 which isn't bad at all heh
Click to expand...
Click to collapse
Sur,I iz puar but not that puar that I'll buy mediatek(It wasn't bad when Qualcomm didn't produce budget chips,but it is way worse now in front of Qualcomm budget chips).
Btw,afaik,mediatek variant only available in China which is exactly why I don't live there.

Categories

Resources