This is what makes me crazy about Google and Android (and yes, companies like Samsung) - the uneven distribution of updates (I know, I know). In this case, some devices are already receiving 4.0.4 while many of us are still stuck waiting for the (empty?) promises to update to 4.0.
Just sayin'.
http://androidcommunity.com/android-4-0-4-ics-rolling-out-for-nexus-s-xoom-and-galaxy-nexus-20120328/
Google just develop the OS itself. Whether the device run the system depends on hardware vendor.
The current situation is like MS just released Windows 7 but your PC(part) manufacturer has not released new driver yet.
That is a good way to look at it and completely agree with it. The laptop I have hasn't seen a graphics driver update in 6 months now. Something that should be kept up to date on a monthly basis
If it really comes down to it hopefully Samsung will get their act together and release the full source code for our Tablet. Than at that point we can take matters into our own hands.
C'mon guys. The 4.0.3 had lots of bugs. I updated it on my GF'S Nexus S when just been released and battery was drained in few hours. Devs from all devices (for example Xoom - which I have) had to work out to fix issues that have been fixed by Google after months in 4.0.4
So I'm not so disappointed for having not received the 4.0.3 (even 'cuz we have too few devs). I trust in Samsung and believe there will be avaible on April Let's wait
N.b.
I have a Galaxy Note and the only ROMS for ICS avaibles are leaked from the Chinese version of my device. ICS has not been released yet (and we're talking about NOTE...)
Sent from my Galaxy Note with DarkyROM 3.3
i don't really care what bs version samsung is going to release for tab plus's ICS.
the only thing i care is the source code! the faster they release the source code, the faster someone could work on CM9 for tab7plus. THAT is what matter!
gingerboy92 said:
i don't really care what bs version samsung is going to release for tab plus's ICS.
the only thing i care is the source code! the faster they release the source code, the faster someone could work on CM9 for tab7plus. THAT is what matter!
Click to expand...
Click to collapse
ICS source from Samsung has no real effect on CM9. CyanogenMod is an AOSP ROM which means that it's not built from Samsung's sources, it's built straight from Google's upstream sources. That's how devices like the HP TouchPad, Kindle Fire and Nook Tablet have CM9 builds now - none of these devices ever had ICS sources released.
Right now, the only thing that's really holding back CM9 development on the tab7plus is a lack of tab7plus owners contributing to a CM9 port.
yeah Im waiting for CM9, gave up on Samsung. If they release it great. Id rather CM9 anyday, even with no wifi!!
h2g2 said:
ICS source from Samsung has no real effect on CM9. CyanogenMod is an AOSP ROM which means that it's not built from Samsung's sources, it's built straight from Google's upstream sources. That's how devices like the HP TouchPad, Kindle Fire and Nook Tablet have CM9 builds now - none of these devices ever had ICS sources released.
Right now, the only thing that's really holding back CM9 development on the tab7plus is a lack of tab7plus owners contributing to a CM9 port.
Click to expand...
Click to collapse
not the kernel source, and don't forget the wireless chip/adapter/card(?)'s source code.
but yeah, i do agree. we need more devs to work on our tab!
We're still lacking driver source (i.e. our Atheros wifi driver) so even if we got someone to port CM9 to our tabs, we'd have no wifi functionality. The only way to get drivers (with no proper source code released for them) is to reverse engineer the driver(s) themselves, which is a difficult task to say the least.
The reason why most of us say we're waiting for our tabs "ICS Source" is because we're hoping Samsung actually releases our wifi Driver Source code this time.
But...if I remember correctly...someone had pointed out that although Samsung had not released proper source code for our wifi driver, another Sony device that uses it had released code that we could "possibly" use. No one has really confirmed nor denied anything past that recently. Maybe Gary can comment on this issue soon, could clear up a lot of questions for me and everyone else
Sent from my SPH-D700 using xda premium
the real problem here is that samsung is "promising" ICS "soon". it would be easier if they say they are going to release it next year or even not going to release it at all.
gingerboy92 said:
not the kernel source, and don't forget the wireless chip/adapter/card(?)'s source code.
but yeah, i do agree. we need more devs to work on our tab!
Click to expand...
Click to collapse
Kernel source doesn't mean driver sources. Remember that drivers don't need to be integrated into the kernel - if Samsung (or the manufacturer of the device) chooses to implement the driver as a binary-only lkm, there's no obligation on Samsung's part to release the source code for the driver.
This has been a problem with Linux drivers in general for a while. I could be wrong, but my sense is that if the needed driver isn't already available in source code from the manufacturer directly, it's unlikely that Samsung will include it in their kernel sources.
---------- Post added at 10:35 PM ---------- Previous post was at 10:20 PM ----------
Ceelos09 said:
We're still lacking driver source (i.e. our Atheros wifi driver) so even if we got someone to port CM9 to our tabs, we'd have no wifi functionality. The only way to get drivers (with no proper source code released for them) is to reverse engineer the driver(s) themselves, which is a difficult task to say the least.
The reason why most of us say we're waiting for our tabs "ICS Source" is because we're hoping Samsung actually releases our wifi Driver Source code this time.
But...if I remember correctly...someone had pointed out that although Samsung had not released proper source code for our wifi driver, another Sony device that uses it had released code that we could "possibly" use. No one has really confirmed nor denied anything past that recently. Maybe Gary can comment on this issue soon, could clear up a lot of questions for me and everyone else
Sent from my SPH-D700 using xda premium
Click to expand...
Click to collapse
My guess is that if Samsung has not released the needed sources previously, they probably won't do so with ICS. This is just pure speculation on my part, it's possible Samsung's failure to release wifi driver source was an oversight on Samsung's part but if that were the case, it seems like they would have fixed that already.
I don't know what chip the GT7P uses but some Googling suggests that it's an Atheros 6000 series. If that's the case, the Linux driver source is available: http://linuxwireless.org/en/users/Drivers/ar6k
ar6k isn't part of the mainline Linux kernel so there's no obligation on Samsung's part to release it as part of their kernel source release. Whether Samsung is obligated to release their ar6k driver source is probably dependent on their agreement with Atheros (ie, just because an open source Linux ar6k exists doesn't preclude the possibility that Samsung licensed a non-GPL version of the driver).
h2g2 said:
Right now, the only thing that's really holding back CM9 development on the tab7plus is a lack of tab7plus owners contributing to a CM9 port.
Click to expand...
Click to collapse
What is the best way for us Tab 7.0+ owners to contribute?
h2g2 said:
Kernel source doesn't mean driver sources. Remember that drivers don't need to be integrated into the kernel - if Samsung (or the manufacturer of the device) chooses to implement the driver as a binary-only lkm, there's no obligation on Samsung's part to release the source code for the driver.
This has been a problem with Linux drivers in general for a while. I could be wrong, but my sense is that if the needed driver isn't already available in source code from the manufacturer directly, it's unlikely that Samsung will include it in their kernel sources.
---------- Post added at 10:35 PM ---------- Previous post was at 10:20 PM ----------
My guess is that if Samsung has not released the needed sources previously, they probably won't do so with ICS. This is just pure speculation on my part, it's possible Samsung's failure to release wifi driver source was an oversight on Samsung's part but if that were the case, it seems like they would have fixed that already.
I don't know what chip the GT7P uses but some Googling suggests that it's an Atheros 6000 series. If that's the case, the Linux driver source is available: http://linuxwireless.org/en/users/Drivers/ar6k
ar6k isn't part of the mainline Linux kernel so there's no obligation on Samsung's part to release it as part of their kernel source release. Whether Samsung is obligated to release their ar6k driver source is probably dependent on their agreement with Atheros (ie, just because an open source Linux ar6k exists doesn't preclude the possibility that Samsung licensed a non-GPL version of the driver).
Click to expand...
Click to collapse
thanks for the clarification.
bittersound said:
What is the best way for us Tab 7.0+ owners to contribute?
Click to expand...
Click to collapse
In a general sense, there is no answer as it depends on what the problem is. If there are willing developers who don't have access to hardware, you can always donate hardware or chip in to help with that. If there aren't enough willing developers, well, you just have to learn to code or chip in for a bounty to get a developer interested. If, on the other hand, it's a lack of vital documentation or source code from Samsung, well, there's not much you can do except perhaps get some sort of letter writing or phone call campaign over to the appropriate corporate offices.
In a specific sense, best bet is to go over to the development sub-forum, find someone with a project you'd like to support and ask them what help they need.
Gary once mentioned he would help with porting CM9, but he did make it clear to us that it would be a while until he would jump in to help. Life must be keeping him busy atm.
For now, I'm working on learning some code, maybe I'll be of "some" help down the road.
Sent from my SPH-D700 using xda premium
Related
Via Google. Who's ready for CM9?
YES
yes yes cant wait!!
it's only for the maguro = Galaxy Nexus
the other devices profiles are not included with this release
AllGamer said:
it's only for the maguro = Galaxy Nexus
the other devices profiles are not included with this release
Click to expand...
Click to collapse
I doubt that'll stop porting being done pretty quickly.
Can't wait for a rom to be available for the amaze!
Hope to see something cool for the Atrix 4G. Probably won't happen any time soon, but a man can dream.
Can not wait till it finds it way to my device.
Surely this will reach the Galaxy S2 before the end of the year? Have an upgrade due next month and as much as I want ICS I can't justify upgrading to the Galaxy Nexus as it's just not good enough, so think I'll wait for S3 and hope that Cyanogen ports this over ASAP! Will definitely donate a nice amount to the cause.
im hoping it reaches my gs2 asap
worth the wait!
Google really seems to be embarrassed by the HC source, but glad to see that they kept to their openness promise.
I want my 2.3.5 update for my S2 and then I want ICS by jan-fep.
Can anyone familiar with the dev-scene comment on whether this will help in porting ICS to GSII ? Last I remember, the problem with SDK port was with the graphic driver (Mail400) and the source-code release for sure does not help on that front!
its about time!
let's get it on!
Yessssssss!!!!!!!!!!!!!!!!!!!!! Finally!!! Xmas today!
Still waiting working on my phone
How does the development for other devices (than Galaxy Nexus) look like from this point:
1) someone said there are no "profiles" for other devices - does it mean that Google has still to release something in addition to this source to make it usable for the developers, so that they'll be able to make custom ROMs like CM for other devices?
or
2) the developers can handle it without that "profiles"?
And of course I understand the devs have to collect all device-specific code like drivers etc.
--
Tom
Nice!
10chars
AllGamer said:
it's only for the maguro = Galaxy Nexus
the other devices profiles are not included with this release
Click to expand...
Click to collapse
Have you read the information on the source page?
News Source Code Available for Android 4.0
The source code for the Android 4.0 platform and software stack has been released! This release allows OEMs to begin preparing Android 4.0 for installation on new and existing devices, and allows hobbyists, enthusiasts, and researchers to develop custom builds. For information on how to obtain the software, visit our Getting the Source page.
Click to expand...
Click to collapse
Sent from my Zio using xda premium
Awesome. What a treat.
Yeah let the roms coming!!
https://groups.google.com/forum/#!msg/android-building/T4XZJCZnqF8/WkWhGUYb4MAJ
Hi! We just released a bit of code we thought this group might be interested in.
Over at our Android Open-Source Project git servers, the source code
for Android version 4.0 (Ice Cream Sandwich) is now available.
Here's how to get it:Follow the instructions at
http://source.android.com/source/downloading.htmlCheck out the
'ics-release' branch:repo init -u
https://android.googlesource.com/platform/manifest -b android-4.0.1_r1
That's it! However since this is a large push, please be aware that it
will take some time to complete. If you sync before it's done, you'll
get an incomplete copy that you won't be able to use, so please wait
for us to give the all-clear before you sync.
This is actually the source code for version 4.0.1 of Android, which
is the specific version that will ship on the Galaxy Nexus, the first
Android 4.0 device. In the source tree, you will find a device build
target named "full_maguro" that you can use to build a system image
for Galaxy Nexus. Build configurations for other devices will come
later.
Unfortunately we still don't have our Gerrit code review servers back
online. That remains our top priority though, and we hope to have them
back soon.
This release includes the full history of the Android source code
tree, which naturally includes all the source code for the Honeycomb
releases. However, since Honeycomb was a little incomplete, we want
everyone to focus on Ice Cream Sandwich. So, we haven't created any
tags that correspond to the Honeycomb releases (even though the
changes are present in the history.)
JBQ, on behalf of the AOSP team.
--
Jean-Baptiste M. "JBQ" Queru
Software Engineer, Android Open-Source Project, Google.
Click to expand...
Click to collapse
It looks like the HoneyComb source was released also. Will this help any active HoneyComb ROMS?
Getting some real support for the Tegra in the Gtab is all I care. Might as well just skip HoneyComb and go straight to ICS.
I thought I had a much longer wait for this, can't wait for the CyanogenMod 9 builds.
Yes, it will help. I hope that people can post the rest of the ICS posts under this topic because with the release of the Honeycomb and Ice Cream Sandwich source codes noobs are gonna go crazy.
Noob here and yes I am going crazy.
Woooooo!!!!
Sent from my Eris using XDA App
HwyXingFrog said:
Getting some real support for the Tegra in the Gtab is all I care. Might as well just skip HoneyComb and go straight to ICS.
Click to expand...
Click to collapse
Ditto on that. Honeycomb really is (was) the Vista of Android.
well
I hope we get a smooth running build of one of the two. I am currently running Vegantab and its pretty smooth but it might freeze here and there or hang on the browser, etc.. Its still my daily runner.... So again, I simplty hope we get a very stable ROM egardless of which. Plus, as HWYxingfrog said, some actual tegra support would be great...
Woot
Sent from my Epic 4G Touch
ICS source released. Will gtab be getting it?
I hope so and will support the devs who are!
please keep topics like this in this thread. http://forum.xda-developers.com/showthread.php?t=1346829
It will help....but keep in mind that there are still no hardware drivers/kernel for the Gtablet beyond Froyo (OS 2.2).
The same issues getting Honeycomb running will be present with ICS though more insight in to the OS will be available with the source. Still no Nvidia "Harmony" kernel or drivers though....
Y'all work on this...
I'll be in Skyrim!!!
I would be surprised if Nvidia released anything for the Harmony board. Considering they already have an HC 3.2 image for Ventana, my guess is this is were all their efforts will go. Afraid we will be stuck with ported versions with no supported drivers.
what happened with the .36 kernel and all the current development for gtab?? are the devs going to have to start all over? why no update for gtabcomb? it ne'er came out of beta? hmmmm...why develop for transformer and galaxy tab if;they already have stable version of os?
jedibbq said:
what happened with the .36 kernel and all the current development for gtab?? are the devs going to have to start all over? why no update for gtabcomb? it ne'er came out of beta? hmmmm...why develop for transformer and galaxy tab if;they already have stable version of os?
Click to expand...
Click to collapse
It's still being worked on. IF you read my posts in this thread I've posted basically all the recent updates the devs have made with the .36 kernel.
http://forum.xda-developers.com/showthread.php?t=1231577&page=10
jedibbq said:
what happened with the .36 kernel and all the current development for gtab?? are the devs going to have to start all over? why no update for gtabcomb? it ne'er came out of beta? hmmmm...why develop for transformer and galaxy tab if;they already have stable version of os?
Click to expand...
Click to collapse
Allow me to disavow this uninformed complaining. Thanks for all the hard work devs.
The HC roms run pretty smoothly for the most part, is the release of the source code going to help the development of ICS roms? I know itll help but I mean will it make it easier then it was trying to get HC on the G-Tab
I remember something about nvidia/vsonic (cant remember which one) saying they wouldnt release something for honeycomb since google didnt release the source code..
So i xpect they at least release driver specs or source this time
Sent from my DROID3 using Tapatalk
brettdwagner said:
Allow me to disavow this uninformed complaining. Thanks for all the hard work devs.
Click to expand...
Click to collapse
not complaining just asking legitimate questions.. supposedly even though google sort of released the code for honeycomb its still up to nvidia to release drivers.without those no working ics rom for gtablet
jedibbq said:
not complaining just asking legitimate questions.. supposedly even though google sort of released the code for honeycomb its still up to nvidia to release drivers.without those no working ics rom for gtablet
Click to expand...
Click to collapse
Double question marks "??" and "hmmmmm..." don't typically accompany legitimate questions.
"Why isn't this out of beta? Why are they working on other devices and not mine?"
http://www.androidcentral.com/android-403-source-code-being-pushed-aosp#comments
how does this affect the build of ics for the g2x?
Motorola Xoom was in their. Tegra 2 device... interested for sure.
Android 4.0.3 kernel sources pushed
https://groups.google.com/group/android-building/msg/ff4e97808535b396?pli=1
all this could be old news. the devs might be one step ahead on this. i just want to help out as much as i can. i hate not having any news on the devs work. i still love what you guys done so far. dont give up and dont let it die.
Xoom has a 4.0.3 port.
Xoom hw is not the same. Not the same board
Sent from my LG-P999 using Tapatalk
well if this post has nothing to offer to the build of ics for the g2x can the mod. please delete this post. i dont need this to take up space. thanks
i was under the impression that the source code for the tegra 2 kernel, and not the source code for ics itself was the issue.
Thats not entirely correct. ICS changed the way certain interfaces behaved especially HW accelerated graphics. Under Gingerbread, the system expected round peg and the nVidia blob provided round peg.
The problem is now ICS wants square peg. The blob we have only ever provides round peg.
There were three potential solutions to this problem:
1) nVidia releases source code and developers update as needed. This is how development works on the Qualcomm and TI driven devices which are basically fully open. This is also why you see old as dirt devices like the G1 updated far beyond their last "official" ROM.
2) nVidia releases an ICS compatible blob. This is what we can hopefully pull from the Xoom tree. It leaves us dependent on nVidia for "substantial" updates, like when interfaces change but would probably allow for a number of updates. There are Gingerbread ROMs that use Froyo kernels as an example.
3) Developers rewrite the interfaces in question to accept round peg. In the case of HW acceleration, this would mean back-porting the Gingerbread method of HW acceleration. It would work but would be labor intensive to maintain and may introduce instability or comparability issues in apps that specifically expect the ICS method. Think, for example, Swype talking about having to rewrite the keyboard to accomidate the way the screen is presented to programs.
I am not a ROM developer but I have college level programming experience and a pretty decent understanding of the technical issues here.
nVidia was never going to release their source. They have always maintained their source code as confidential and proprietary and as far as I know they've not released source for anything as far back as the Riva TNT.
In other product lines, the reason for this is that they differentiate some features through software. In essence the HW is capable of more and the software locks it away. Back in the day there was a GeForce to Quadro softmod that drastically increased performance in some programs.
I don't really know if thats the case here, nobody does. nVidia only wants you to access what they sell you not what the HW is capable of.
Personally, I hope the blob for the Xoom helps the CM team. I imagine we will know shortly.
rayman2k1 said:
i was under the impression that the source code for the tegra 2 kernel, and not the source code for ics itself was the issue.
Click to expand...
Click to collapse
Sent from my LG-P999 using Tapatalk
That was the most intelligent and courteous explanation I have ever read here on xda. Now watch all the omniscient people strike it down.
Sent from my LG-P999 using xda premium
TerryMathews said:
Thats not entirely correct. ICS changed the way certain interfaces behaved especially HW accelerated graphics. Under Gingerbread, the system expected round peg and the nVidia blob provided round peg.
The problem is now ICS wants square peg. The blob we have only ever provides round peg.
There were three potential solutions to this problem:
1) nVidia releases source code and developers update as needed. This is how development works on the Qualcomm and TI driven devices which are basically fully open. This is also why you see old as dirt devices like the G1 updated far beyond their last "official" ROM.
2) nVidia releases an ICS compatible blob. This is what we can hopefully pull from the Xoom tree. It leaves us dependent on nVidia for "substantial" updates, like when interfaces change but would probably allow for a number of updates. There are Gingerbread ROMs that use Froyo kernels as an example.
3) Developers rewrite the interfaces in question to accept round peg. In the case of HW acceleration, this would mean back-porting the Gingerbread method of HW acceleration. It would work but would be labor intensive to maintain and may introduce instability or comparability issues in apps that specifically expect the ICS method. Think, for example, Swype talking about having to rewrite the keyboard to accomidate the way the screen is presented to programs.
I am not a ROM developer but I have college level programming experience and a pretty decent understanding of the technical issues here.
nVidia was never going to release their source. They have always maintained their source code as confidential and proprietary and as far as I know they've not released source for anything as far back as the Riva TNT.
In other product lines, the reason for this is that they differentiate some features through software. In essence the HW is capable of more and the software locks it away. Back in the day there was a GeForce to Quadro softmod that drastically increased performance in some programs.
I don't really know if thats the case here, nobody does. nVidia only wants you to access what they sell you not what the HW is capable of.
Personally, I hope the blob for the Xoom helps the CM team. I imagine we will know shortly.
Sent from my LG-P999 using Tapatalk
Click to expand...
Click to collapse
Absolutely right. I don't have any hope at all regarding the Xoom blobs though, as they are not the same board type, and thus likely will not be compatible.
Sent from my LG-P999 using Tapatalk
You are totally right but nvidia is planning to release the drivers for tegra 2 platform in Q1 of 2012, if you haven't heard it yet. Here is the link:
http://forums.developer.nvidia.com/devforum/discussion/comment/3236?Sort=popular
Regards
Sent from my Optimus 2X using Tapatalk
Missing Dateline
ironman159 said:
You are totally right but nvidia is planning to release the drivers for tegra 2 platform in Q1 of 2012, if you haven't heard it yet. Here is the link:
http://forums.developer.nvidia.com/devforum/discussion/comment/3236?Sort=popular
Regards
Sent from my Optimus 2X using Tapatalk
Click to expand...
Click to collapse
I dont see any dateline regarding nVidia will release tegra2 drivers in Q1 of 2012 here (http://forums.developer.nvidia.com/devforum/discussion/comment/3236?Sort=popular). Am I missing sth?
It says early next year.
Sent from my Optimus 2X using Tapatalk
ironman159 said:
It says early next year.
Sent from my Optimus 2X using Tapatalk
Click to expand...
Click to collapse
+1 early could mean the first half of 2012.
NVIDIA NEVER AGAIN!
By Summer.
We will have cm9 and other ICS roms probaly by the end of January-mid February. LG and Nvidia could learn much from our talented devs that do all this for free and the joy of it.
Drivers Release?
IF Nvidia is going to release the necessary drivers, does this mean we might see our camera features back again on future ICS roms?
Chinese281 said:
IF Nvidia is going to release the necessary drivers, does this mean we might see our camera features back again on future ICS roms?
Click to expand...
Click to collapse
This might mean nothing unless lg releases there stuff too. Lg changed some things for this device. I'm not sure but I think lg did something weird to get the tegra 2 to work with the board they designed. nvidia said lg needs to release there changes too for what we are trying to do.
Sent from my CM 7.1.0 G2x using XDA app
...of the year 2015.
moderators: I'm very sorry (not really) for the sarcastic spam, but I've grown more and more unhappy with Samsung in regards to them keeping their promises. At this point, other than my P6210 tablet, I've sold every single one of my android devices that are controlled by samsung.
I still have a galaxy nexus, but thankfully the source for that is controlled by google - not samsung.
Q1 has come and gone. No ICS. Hell, they STILL won't release the source for the damn wifi driver! The same source that qualcomm (who owns atheros) released under the GPL, but samsung says that they (samsung) get to choose between GPL and BSD and they are choosing BSD. Why? What good does it do Samsung to NOT release the source? If Apple.. er.. samsung wants to treat their customers this way, I can take my future business elsewhere.
If I wanted a closed platform, I'd buy an iphone or ipad.
Gary
I agree
And I feel that with the increasingly number of similar tablets that Samsung puts in the market, the hopes for further updates are extremely low. Maybe we will have a first revision of ICS, but better it is good, because I don't think they will fix whatever is wrong.
I will agree with Gary in terms of how closed Samsung is making their "open" device. Though that is coming from a developer. From a consumer perspective having or not having ICS makes no difference to them. Unfortunately as our world becomes more tech illiterate the more these devices will become more and more locked down and cause stagnation in innovation.
Sorry to hear you leaving though you were a great help here.
I'm not leaving the p6210... its the only samsung device I'm keeping. However, until and unless Samsung gives me something more than incomplete and outdated source, there's nothing else for me to do. I can't fix the wifi bugs, because Apple..er..Samsung won't release the source.
If they ever get around to pushing out ICS (big "if" there) AND they release the source, I'll play with that. I might even find the time to just port AOSP (or even better - AOKP) over.
After using a galaxy nexus for a few days, I don't miss touchwiz at all.
Well that's good to hear... It sounded as if you were leaving us there. Though supposedly someone is making headway with wifi with the cm9 ics build. You might want to check in there.
What's with the driver? Does the GPL driver not support our cards, is it just missing the pci id? I'm new to this slate, so I don't have much exposure to the current issues like this.
Sent from my GT-P6210 using Tapatalk 2
Even Huawei Springboard (my 2nd tablet) already received ICS, I also disappointed a bigger company like Samsung get update slower then Huawei
Sent from my GT-P6200 using XDA App
fewt said:
What's with the driver? Does the GPL driver not support our cards, is it just missing the pci id? I'm new to this slate, so I don't have much exposure to the current issues like this.
Sent from my GT-P6210 using Tapatalk 2
Click to expand...
Click to collapse
GPL is not a driver, GPL is a open source licence.
Yes, I know the difference between GPL & BSD (both are OSS licenses.) I was asking if the GPL version would work if it was updated with the p6210 WIFI PCI ID.
Sent from my GT-P6210 using Tapatalk 2
fewt said:
What's with the driver? Does the GPL driver not support our cards, is it just missing the pci id? I'm new to this slate, so I don't have much exposure to the current issues like this.
Click to expand...
Click to collapse
For doing an AOSP type implementation where 100% of the source is available, yes - it might.
For trying to do a kernel that would work with the rest of the samsung firmware, no - it won't. There are 3 parts that have to work together here: the actual driver (ar6003.ko in samsung's firmware), wpa_supplicant (which, on stock p6210 implementations has a ar6003 specific interface compiled in) and the rest of the firmware (settings page, etc.)
The "rest of the firmware" calls wpa_supplicant to do the dirty work of the wifi driver, and wpa_supplicant makes calls directly into the driver.
I'm sure I typed all this before, in another thread months ago, and went into great detail. The short version is this:
I need the source for the ar6003 driver and the source for wpa_supplicant (for the ar6003 interface.) Both of these are released in "GPL/BSD" dual licenses by their authors, meaning that samsung can, in theory, choose which license model to use when they include that code. Samsung has told me that they are claiming BSD for both and therefore will refuse to release the source to any modifications that might be included.
I've tried pulling in the generic ar600x code from mainline linux, but it wasn't working for me with the rest of the samsung stuff... and I simply don't have the time to mess with that for endless hours just because samsung is trying to be apple-like and make android a closed platform.
garyd9 said:
For doing an AOSP type implementation where 100% of the source is available, yes - it might.
For trying to do a kernel that would work with the rest of the samsung firmware, no - it won't. There are 3 parts that have to work together here: the actual driver (ar6003.ko in samsung's firmware), wpa_supplicant (which, on stock p6210 implementations has a ar6003 specific interface compiled in) and the rest of the firmware (settings page, etc.)
The "rest of the firmware" calls wpa_supplicant to do the dirty work of the wifi driver, and wpa_supplicant makes calls directly into the driver.
I'm sure I typed all this before, in another thread months ago, and went into great detail. The short version is this:
I need the source for the ar6003 driver and the source for wpa_supplicant (for the ar6003 interface.) Both of these are released in "GPL/BSD" dual licenses by their authors, meaning that samsung can, in theory, choose which license model to use when they include that code. Samsung has told me that they are claiming BSD for both and therefore will refuse to release the source to any modifications that might be included.
I've tried pulling in the generic ar600x code from mainline linux, but it wasn't working for me with the rest of the samsung stuff... and I simply don't have the time to mess with that for endless hours just because samsung is trying to be apple-like and make android a closed platform.
Click to expand...
Click to collapse
wpa_supplicant is what's probably killing you if it has been forked, more so than the driver itself. Makes perfect sense why it is such a pain.
Do you know of a thread somewhere that describes how to get started building kernels for these things? I'll play around with it, I don't have much experience with Android outside of some hacking with adb but I know my way around Linux as well as most.
I'll start poking around more, but thanks for the short version it is appreciated.
Zadeis said:
I will agree with Gary in terms of how closed Samsung is making their "open" device. Though that is coming from a developer. From a consumer perspective having or not having ICS makes no difference to them. Unfortunately as our world becomes more tech illiterate the more these devices will become more and more locked down and cause stagnation in innovation.
Sorry to hear you leaving though you were a great help here.
Click to expand...
Click to collapse
I disagree that consumers don't care. Call me naive that I didn't scour the internet deep enough to find good, honest feedback, but the only reason I chose the more expensive Samsung 7.0 was its IR blaster and its advertised ability to control the home theater components with it. That feature is bolded and blaster all over Samsungs feature list and descriptions.
So I get me GT7+ 2 months ago and how does it work? It doesn't do what I want. Peel, the only IR app available for the tablet force closes every time, and updates have proven fruitless. I email the developers of Peel and what do they tell me? They won't fix (or can't fix) the problem until they get updated ICS drivers for the SG7+. So for now I am **** out of luck until Samsung updates. I try to get a hold of Samsung tech support and get nowhere. The best I got was from a "Live Chat" bot that said ICS will be available in the future. No more specifics could be given.
And to add insult to injury, the screen on my Tab is fritzing out and needs to be sent back for service already. On paper the SG7+ looks great but for me it's been nothing but a hassle.
I want ICS so I can have everything work as advertised. It probably never will so I'll chalk this one up to experience and sell the POS.
Just remember that samsung never actually promised ICS for this device. There were "leaks" (completely unofficial) and "targets", but never any legally binding promises.
Why should samsung spend the money developing ICS for people who own a device when that device is ALREADY 6 months old? At the rate samsung is coming out with new devices, they need that money developing for newer devices.
In a sick kind of way, this makes sense to me. (Samsung seems to be forgetting, however, that I buy a new tablet every 6-12 months and my next one will NOT be a samsung device due to the experience I'm having with them over this one.)
What doesn't make sense to me is the way that they are withholding source code. It just doesn't profit them or even save them money. There isn't even anything proprietary in the ar6003 drivers and wpa_supplicant code. No trade secrets. It's almost as if they are deliberately chasing AWAY technical people... but that just doesn't make sense when your primary OS is an open one that depends on geeks.
The only thing I can figure out is that they are, in fact, hiding something. Perhaps they are embarassed about the programming? Perhaps they cut corners and don't want it to become public knowledge? Maybe they just have a nasty streak and are deliberately trying to prevent any repeat customers. I'm just taking wild guesses here - I honestly don't know.
The flip side to that is that you are aware of the fact ICS will have an impact to functionality on a current situation. Most people won't, as you put it, scour the internet to find the solution or future solution, to a problem and actually understand that an OS update will have a great impact on their experience with their current product. I had a discussion with someone the other day on this mindset and how it has been driving me nuts. (This person also considers them self an average user and not like most of us on this forum) You know that android 3.2 is on your your Tablet. YOU know what android 4.0 ICS is and how it impacts you. Most don't. That is what I was trying to conveying.
Though I am surprised that you are having issues with it on account I have had no problems with it (even though I haven't used it much).
P.S. Sorry about the irritated tone I do not mean any harm it's just a sore subject with me :/
Zadeis said:
The flip side to that is that you are aware of the fact ICS will have an impact to functionality on a current situation. Most people won't, as you put it, scour the internet to find the solution or future solution, to a problem and actually understand that an OS update will have a great impact on their experience with their current product. I had a discussion with someone the other day on this mindset and how it has been driving me nuts. (This person also considers them self an average user and not like most of us on this forum) You know that android 3.2 is on your your Tablet. YOU know what android 4.0 ICS is and how it impacts you. Most don't. That is what I was trying to conveying.
Though I am surprised that you are having issues with it on account I have had no problems with it (even though I haven't used it much).
P.S. Sorry about the irritated tone I do not mean any harm it's just a sore subject with me :/
Click to expand...
Click to collapse
The latest release od Peel finally has it working on such a basic level that's it is no more useful that the remoate that shipped with my television. It can tuen my TV on & off, it can change channels, and it can adjust the volume of the television. When I want to add another device such as a cable box or AV receiver? No go. It will power the device in setup but it will not save it to the application for future use. I blame buggy peel software. It can obviously fire the codes it needs but they're going to blame Samsung and say it's a driver issue. The only way I could believe that possible is if the radio frequencies vthe two devices used were too close together to be discerened by the GT7+. Regardless, it's a feature which does not work as it explicitly advertises and, as Gary points out multiple times, they refuse to release the source code so crafty & eager developers (which I am not) can make their own functional software.
Peel doesn't use RF... it fires the infrared emitter on the device. I actually played with the "peel" software once. For about 10 minutes. I found it a complete joke and froze the software. I never really cared much about that aspect of the tablet. To me, its for reading ebooks, playing games, and "tinkering." Okay, more for tinkering - but don't tell my wife that. She already yells at me about how expensive my toys are.
garyd9 said:
Peel doesn't use RF... it fires the infrared emitter on the device. I actually played with the "peel" software once. For about 10 minutes. I found it a complete joke and froze the software. I never really cared much about that aspect of the tablet. To me, its for reading ebooks, playing games, and "tinkering." Okay, more for tinkering - but don't tell my wife that. She already yells at me about how expensive my toys are.
Click to expand...
Click to collapse
I bought this Tab for the exact same reasons, I was actually going to get the Kindle Fire as all I really wanted was an e-reader, but the IR Blaster changed my mind. I set up Peel for my home theater and used it once than decided my universal remote for my dish actually works better and haven't used it since. I came in knowing from reading teh threads that Samsung wasn't the greatest at support or putting out updates but all I wanted really was the ability to root it to remove bloatware. IMO HTC devices are much better and easier to customize as HTC is more open about sharing the source code and also provide their own program to unlock their devices. But they at one time were as tight fisted as Samsung and getting updates out of them is still excruciatingly slow. Hopefully Verizon will be getting a One X device soon as I'm up for an upgrade in July and right now the best HTC device they have is the Rezound.
fcorona76 said:
The latest release od Peel finally has it working on such a basic level that's it is no more useful that the remoate that shipped with my television. It can tuen my TV on & off, it can change channels, and it can adjust the volume of the television. When I want to add another device such as a cable box or AV receiver? No go. It will power the device in setup but it will not save it to the application for future use. I blame buggy peel software. It can obviously fire the codes it needs but they're going to blame Samsung and say it's a driver issue. The only way I could believe that possible is if the radio frequencies vthe two devices used were too close together to be discerened by the GT7+. Regardless, it's a feature which does not work as it explicitly advertises and, as Gary points out multiple times, they refuse to release the source code so crafty & eager developers (which I am not) can make their own functional software.
Click to expand...
Click to collapse
Why would Peel blame Samsung? From what you described, the hardware is working properly but the functionality to save multiple devices is missing in software.
I think this is what Zadeis is trying to get at with regards to expectations - Peel sounds like it's either broken or not designed to meet your expectations. Either way, it's not something that's going to be addressed by an ICS update.
I think you hit the nail on the head when you say that the GT7+ doesn't work as advertised. But the proper course of action here is to seek remedy with Samsung, not wait for a software update. When you get a bad meal at a restaurant, you send it back. You don't eat it, then hope dessert will be better.
Apologies if my tone comes off as harsh or unsympathetic, I do not intend to be either, but pinning too many hopes on to an OS update is just setting yourself up for more frustration down the road when it doesn't match your expectations.
---------- Post added at 06:21 PM ---------- Previous post was at 06:13 PM ----------
garyd9 said:
What doesn't make sense to me is the way that they are withholding source code. It just doesn't profit them or even save them money. There isn't even anything proprietary in the ar6003 drivers and wpa_supplicant code. No trade secrets. It's almost as if they are deliberately chasing AWAY technical people... but that just doesn't make sense when your primary OS is an open one that depends on geeks.
The only thing I can figure out is that they are, in fact, hiding something. Perhaps they are embarassed about the programming? Perhaps they cut corners and don't want it to become public knowledge? Maybe they just have a nasty streak and are deliberately trying to prevent any repeat customers. I'm just taking wild guesses here - I honestly don't know.
Click to expand...
Click to collapse
I think you already answered this question in what you meant as a joke earlier... Apple.
Samsung makes Apple's SoCs. GT7+ uses Samsung's own Exynos SoC. I think chances are good that Exynos and the A5 share IP and therefore, Samsung might be withholding the source code because the modifications they've made include IP that is covered by whatever NDA exists between Apple and Samsung.
so what is the latest news on ics update for the 6210 ? , guessing samsung said no ics love for us ? , or .... just wondering what latest news is ..
h2g2 said:
Samsung makes Apple's SoCs. GT7+ uses Samsung's own Exynos SoC. I think chances are good that Exynos and the A5 share IP and therefore, Samsung might be withholding the source code because the modifications they've made include IP that is covered by whatever NDA exists between Apple and Samsung.
Click to expand...
Click to collapse
You are a bit confused, I think. The exynos isn't the problem. While samsung hasn't exactly given us full technical documents on it, they added support for it in the mainline linux kernel - publishing enough source. They pretty much HAD to do that, as CPU support in linux can't be done as a module, and therefore must be opensourced in order to run linux at all.
The issue, at least in the case of the 7+, is the wifi chip/driver. It's an atheros 6003 chip. Qualcomm (who owns atheros) released the driver for that under the GPL. Samsung has claimed to me that they (samsung) are licensing it from Atheros under terms that allow them to choose to re-release the driver under either GPL or BSD terms and that they (samsung) are choosing the BSD model (which doesn't require the release of source.) In theory, Qualcomm could FORCE samsung to release the source, but I doubt qualcomm really cares too much. In fact, the module itself as released in binary form in the 7+ firmware (at least up to LA1) actually claims GPL licensing. (Can be verified by using modinfo on ar6000.ko) However, only the copyright holder can enforce the licensing.
A nearly identical situation exists with wpa_supplicant and the intergration between ar6003 and wpa_supplicant. In that case, I know for a fact that the author allows either GPL or BSD terms so wouldn't force the issue.
Those are the only two things I, personally, care about right now. If I had the source for those two parts, I'd be able to not only FIX the wifi issues on the p6210, but also enchance the functionality. I'm also unable to make certain unrelated changes to the kernel, as doing so without being able to recompile the ar6003 driver will render the existing ar6003 non-functional.
I've used a few other androids before with custom ROMs, and a major obstacle to stability seems to often be the fact that manufacturers typically don't include open source drivers, which leads to reverse engineered open source drivers being developed on xda, which often aren't as good as the binaries that can only be used with stock based ROMs.
Since the OPP ships with cm11, does this mean that it uses open source drivers that will work well with any open source ROM?
I'm not dev, but my guess is not entirely. Qualcomm is extremely protective of it's IP, which is why the Nexus 7 2013 source got pulled momentarily I think. This device run their SOC do I assume it'll use their drivers.
But as you can see there are a good number of roms for it already, so it's still a dev friendly device. They released their kernel source so it's useful for making AOSP roms and regular CM11, but certain features from CM11S is not available. CM isn't fully open source either from what I understand.
If the device was running pure AOSP with no Google services and using Texus Instruments OMAP chips, then it'll probably be more open.
But probably don't take what I say seriously.
eksasol said:
I'm not dev, but my guess is not entirely. Qualcomm is extremely protective of it's IP, which is why the Nexus 7 2013 source got pulled momentarily I think. This device run their SOC do I assume it'll use their drivers.
But as you can see there are a good number of roms for it already, so it's still a dev friendly device. They released their kernel source so it's useful for making AOSP roms and regular CM11, but certain features from CM11S is not available. CM isn't fully open source either from what I understand.
If the device was running pure AOSP with no Google services and using Texus Instruments OMAP chips, then it'll probably be more open.
But probably don't take what I say seriously.
Click to expand...
Click to collapse
Well I appreciate your honesty lol you've given me some interesting things to research. In the future I may be avoiding Qualcomm devices, I'm sick of having to decide between stability and crappy ui and features, vs the awesome of cyanogenmod or other community Roms with unstable reverse engineered drivers.
Hardware manufacturers have got to quit messing with android so much, Samsung makes great hardware but pretty crappy software, I'm tired of having to choose.
My next phone will either be an OPP or a Nexus 5 I think. On that magical day that I have cash to blow....
eksasol said:
I'm not dev, but my guess is not entirely. Qualcomm is extremely protective of it's IP, which is why the Nexus 7 2013 source got pulled momentarily I think. This device run their SOC do I assume it'll use their drivers.
But as you can see there are a good number of roms for it already, so it's still a dev friendly device. They released their kernel source so it's useful for making AOSP roms and regular CM11, but certain features from CM11S is not available. CM isn't fully open source either from what I understand.
If the device was running pure AOSP with no Google services and using Texus Instruments OMAP chips, then it'll probably be more open.
But probably don't take what I say seriously.
Click to expand...
Click to collapse
TI OMAP left the mobile market for SoC.
Well I appreciate your honesty lol you've given me some interesting things to research. In the future I may be avoiding Qualcomm devices, I'm sick of having to decide between stability and crappy ui and features, vs the awesome of cyanogenmod or other community Roms with unstable reverse engineered drivers.
Hardware manufacturers have got to quit messing with android so much, Samsung makes great hardware but pretty crappy software, I'm tired of having to choose.
My next phone will either be an OPP or a Nexus 5 I think. On that magical day that I have cash to blow....
Click to expand...
Click to collapse
You can't avoid Qualcomm, they pretty much own the market share when it comes to SoC. They're not bad people, their SoC are undoubtedly the best on the market. And there aren't many phones at all that offer other system on chips. The only others being NVIDIA Tegra but they're not really on phones anymore, Samsung Exynos is on international devices... if you buy a international Samsung Note/S5 then chances are it won't support U.S. LTE bands (unless you live outside of the U.S./NA).
At the end of the day, the factory images are there. And yes the N7 were pulled but its back and has been there for awhile.
The real difference is that aosp is built and tested on Nexus devices and CM uses aosp (google) to build there platform from.
So any other device is for lack of better words is a port from android nexus development aka android built for and test on nexus.
So in a way the one plus is just like any other non nexus device. You can't just repo sync a Google aosp repo and build for this phone. Without changes being made to make it work.
Not knocking the One it's a really nice piece of hardware.