Samsung Galaxy Note II Source Code Released

← Back to Stories (view on slashdot.org)

Samsung Galaxy Note II Source Code Released

Posted by samzenpus on Monday October 8, 2012 @05:35AM from the open-up dept.

An anonymous reader writes "Samsung has released the source code for the Samsung GALAXY Note II. This clears the way for custom ROM's for the smartphone. From the article: 'It's now been posted for the international GT-N7100 model, giving developers a peek at the 5.5-incher's inner workings and allowing them to get to work on new mods.'"

32 comments

Min score:

Reason:

Sort:

Link in summary is (slightly) wrong by admdrew · 2012-10-08 05:38 · Score: 4, Informative

...and currently just links to the main Engadget page. Please fix by taking out that extra L at the end of the link:
http://www.engadget.com/2012/10/08/samsung-releases-galaxy-note-ii-source-code/

--
LegendMUD
1. Re:Link in summary is (slightly) wrong by davester666 · 2012-10-08 06:09 · Score: 5, Informative
  
  Why link to an ad-filled hit-whore website?
  How about straight to the source:
  http://opensource.samsung.com/
  
  --
  Sleep your way to a whiter smile...date a dentist!
2. Re:Link in summary is (slightly) wrong by admdrew · 2012-10-08 08:25 · Score: 1, Informative
  
  Well, all *I* was trying to do was fix the original submission. That said, I do actually like a lot of Engadget writeups, and unlike MANY other ad-filled hit-whore websites, Engadget always links to the actual source.
  
  --
  LegendMUD
Re:Link is Broken by Anonymous Coward · 2012-10-08 05:41 · Score: 1

Link
In case anyone here actually wants to read the article.
Just as I suspected. by 0100010001010011 · 2012-10-08 05:48 · Score: 5, Funny

Copyright (C) 2005, 2006, 2007, 2008, 2009 Apple Inc. All rights reserved.
Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
THIS SOFTWARE IS PROVIDED BY APPLE INC. AND ITS CONTRIBUTORS ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES,.....
1. Re:Just as I suspected. by Flipao · 2012-10-08 05:55 · Score: 4, Funny
  
  I'm just going to leave this here, so I can come back to it a couple years from now, when Apple unveils their "revolutionary" iPhone Ink, with a big display and the ability to accept both capacitive touch and pen input. "This changes everything again... again"
2. Re:Just as I suspected. by jonbryce · 2012-10-08 10:05 · Score: 1
  
  Is that the Webkit source? Some of that is in the 2 clause BSD licence, and Apple are involved in it. The original authors were the KDE project though.
meh by aaron552 · 2012-10-08 05:55 · Score: 4, Insightful

The exynos chipsets are a nightmare for device maintainers. The kernel source certainly helps, but the binary blobs for the device drivers, HAL, RIL, hwcomposer, etc. are still going to have to be hacked around.

--
I had a sig once. It was lost in the great storm of '09.
1. Re:meh by Anonymous Coward · 2012-10-08 06:43 · Score: 1
  
  It's better than nothing, and it's because of them we can be sure to see custom firmwares like CyanogenMod within days, more likely hours of its release. And it's definitely better than most other manufacturers that wait months to release the source (or those that don't at all).
2. Re:meh by Anonymous Coward · 2012-10-08 06:45 · Score: 0
  
  Yeah, Apple are clearly winning on this score! No hacking around their binary blobs!
3. Re:meh by Andy+Dodd · 2012-10-08 07:05 · Score: 5, Insightful
  
  Wrong. If you look at the history of CM bringups, kernel source is a small part of the equation.
  I have no idea why this was able to make Slashdot. Really, since when is an Android manufacturer doing the utterly bare legal minimum of what they are required to do by the GPL newsworthy? Do we even know if they're even complying with the GPL with this release? (See below...)
  As to CyanogenMod on Note 2 - It's not going to happen unless a new maintainer steps up to the plate. None of the current Exynos maintainers have any intention on purchasing any additional non-Nexus Exynos devices. We're tired of Samsung's constant GPL violations (Frequently, their source releases do NOT match that of shipped devices - for example none of the source releases for the Note 10.1 produce a viable BCM4334 driver for wi-fi.) and of their total lack of cooperation with the open source community.
  Take a look at the omapzoom (TI) reference platform source. Take a look at CodeAurora (Qualcomm) reference platform source. Then take a look at the Insignal git repos, or one of Hardkernel's 2GB tarballs. Note that of the two latter examples (both for Exynos), neither has a respository with any git history. They also don't even remotely match anything that is in Samsung's shipped devices in addition to being vastly outdated. If you use the Hardkernel or Insignal hardwarecomposer source code on a device, and then completely delete hwcomposer, you will see ZERO DIFFERENCE in behavior!
  Background: I am the CyanogenMod co-maintainer for the AT&T Galaxy S II (SGH-I777), International Galaxy Note (GT-N7000), and Note 10.1 (GT-N8013). The GT-N8013 is my last non-Nexus Exynos device as I'm tired of working with an undocumented platform with no source code and broken hacked-up binaries. On a regular basis, the quality of CM on Exynos devices lags months behind OMAP and Qualcomm devices due to this. I'd like to, for once, be able to actually maintain a device that's in good shape and start focusing on adding new features, instead of constantly fighting massive bugs due to the lack of documentation of Samsung's blobs. (Of which we have FAR more to deal with than OMAP or Qualcomm devices.)
  
  --
  retrorocket.o not found, launch anyway?
4. Re:meh by rodrigoandrade · 2012-10-08 07:17 · Score: 1, Funny
  
  Mod this up!!!!
  If you have no idea what he just said, his 3-digit UID and Cornell e-mail should give you a hint that he knows his shit.
5. Re:meh by confusedwiseman · 2012-10-08 08:19 · Score: 1
  
  Odd correlation based on the 3-digit UID and Cornell e-mail, however, it is an insight to CyanogenMod to which I was not aware. I have great respect for the CyanogenMod team, as they have done some fantastic work on a number of Android devices.
6. Re:meh by the_humeister · 2012-10-08 08:41 · Score: 1
  
  I'm rather surprised by this. Didn't cyanogen get hired by Samsung? Why aren't Samsung's open source releases and documentation better?
7. Re:meh by Andy+Dodd · 2012-10-08 08:51 · Score: 4, Interesting
  
  He works for Samsung USA. Samsung Korea is the one controlling nearly anything Exynos-related. The situation for Qualcomm and OMAP-based Samsungs is quite a bit better - While it seems he doesn't have the authority to use any source code outside of what is available from CAF, the fact is he has a high degree of familiarity with these devices and hence knows how to get the CAF stuff to work VERY well. (The USA guys have, historically, primarily worked with Qualcomm-based devices.)
  The CM Exynos maintainers do have a contact within Samsung Korea, but nearly all of the time when our contact forwards requests to the relevant department, the answer is either "no" or it is a blatant lie. (See the above comment about the Note 10.1 GPL compliance issues - the OSRC guys actually had the balls to claim that the UEALGB build, which was preinstalled on every Note 10.1 sold in the USA for at least one month, was a "leak" and hence they didn't have to provide source that matched it.) Our contact DOES care and wants to make a difference, but their management and the other departments within Samsung Mobile are completely noncooperative.
  
  --
  retrorocket.o not found, launch anyway?
8. Re:meh by the_humeister · 2012-10-08 09:05 · Score: 1
  
  BTW as an SGH-1777 owner, thanks! Running a CM10 without issues.
9. Re:meh by interkin3tic · 2012-10-08 10:40 · Score: 1
  
  What if any motivation to lie is there? Does samsung somehow lose money when one installs CM? Or is it just a case of "You're annoying me with your requests, fuck you."
10. Re:meh by Anonymous Coward · 2012-10-08 14:52 · Score: 0
  
  Judging from this video, you probably don't want to disagree with him anyway.
11. Re:meh by Raenex · 2012-10-08 17:25 · Score: 1
  
  Probably a little bit of both. "Mine, mine, mine!" and "Go away, you smelly nerd."
12. Re:meh by Anonymous Coward · 2012-10-08 21:23 · Score: 0
  
  [quote]We're tired of Samsung's constant GPL violations (Frequently, their source releases do NOT match that of shipped devices - for example none of the source releases for the Note 10.1 produce a viable BCM4334 driver for wi-fi.) and of their total lack of cooperation with the open source community.[/quote]
  The driver is module at Android, not developed by Samsung under GPLv2 (what Linux OS is licensed) so they do not need to ship it with the Linux source code what GPLv2 dictates.
  You should know that....
  But the fact still remains, having a change (legal and technical) to make binary blob drivers, can stop open source community from extending further bought devices lifetime so simply.
  But same time allowing them, many pussy company engineer and chief are willing to even consider and make the binary blob drivers in first place.
13. Re:meh by Andy+Dodd · 2012-10-09 01:22 · Score: 1
  
  WRONG. The bcmdhd driver IS GPLv2.
  See https://github.com/CyanogenMod/android_kernel_samsung_smdk4412/blob/jellybean/drivers/net/wireless/bcmdhd/bcmsdh_linux.c as one example.
  The only place in which your statement is valid is the AR6k fiasco on the Tab 7 Plus and Tab 7.7 - These DO have a driver that is dual-license GPL and BSD from Atheros, Samsung chooses BSD. While it is insanely frustrating, I can't throw Samsung under the bus for GPL noncompliance with this one. However, I can for bcmdhd.
  
  --
  retrorocket.o not found, launch anyway?
14. Re:meh by Anonymous Coward · 2012-10-09 07:22 · Score: 0
  
  For reference the "unofficial cm" thread is here: http://forum.xda-developers.com/showthread.php?t=1926564&page=3
  Seems Samsung are willing to resolve the issue, let's wait and see. US release is still two weeks away. If it's not fixed by then there's definitely gonna be an influx of fresh blood to improve the situation on the firmware front. Some reverse whimpering can do wonders if the right people get into it.
15. Re:meh by Anonymous Coward · 2012-10-09 07:28 · Score: 0
  
  Aargh – reverse engineering, not reverse whimpering.. Stupid Swype
16. Re:meh by Andy+Dodd · 2012-10-10 05:41 · Score: 1
  
  It's an unofficial winzip kang... Winzip kangs NEVER manage to make it to the finish line, it's simply impossible.
  You know, it's the perfect opportunity for a new maintainer to create a device try and to try and patch support into the kernel... But if no one has even bothered to do that it's not good.
  Bringing up CM10 on the device won't be difficult - Getting N7100 working as well as I9300 wouldn't be difficult at all. It's just that none of the current maintainers with Exynos experience have any desire to do it again. We don't want to support Samsung by making their devices more attractive to a given market when they just jerk us around on a constant basis.
  However, prospects for CM11 and further are extremely poor unless someone completely new takes up the task of being a proper maintainer.
  As to reverse engineering - we're sick and fucking tired of reverse engineering things to find single-line code differences when Samsung could have just given us a goddamned header file - https://github.com/CyanogenMod/android_device_samsung_galaxys2-common/blob/jellybean/overlay/include/hardware/gps.h#L273 - that one fucking field took most of a week to figure out.
  
  --
  retrorocket.o not found, launch anyway?
Re:good, freedom by AliasMarlowe · 2012-10-08 06:18 · Score: 2

This is much better than the competitors walled garden(s). I'd much rather have control of my own landscape, weeds and all, than have a perfectly designed and maintained landscape that somehow still feels cold and sterile.
But Apple gives you a fully hoe'd garden! All they want is your money (and all of it).
With Android (S3 user here, my daughter got HTC), you get the garden you want.

--
Those who can make you believe absurdities can make you commit atrocities. - Voltaire
Re:good, freedom by Anonymous Coward · 2012-10-08 06:27 · Score: 1

With Android (S3 user here, my daughter got HTC), you get the garden you want
So long as you don't want to change the configuration of the garden. That would require you to find a way to dupe one of the garden staff into giving you Head Gardner privileges.
However that same exploit is also what bad guys use when they want to trash the garden, but we overlook that in Android World for some reason.
Re:good, freedom by Anonymous Coward · 2012-10-08 06:35 · Score: 1

Hey now, we only allow extended car metaphors here. We'll have none of your wild garden talk!
Re:good, freedom by Anonymous Coward · 2012-10-08 06:58 · Score: 0

That's what you have to do if you like apple orchards, but want a bit of freedom.
In Android gardens you can often just walk to gardener's house and he'll pass you the keys to the toolshed. Sometimes - like in TFA - he'll even show you the plans and tell you about his secret fertilizer.
Re:good, freedom by Anonymous Coward · 2012-10-08 07:37 · Score: 0

In Android gardens you can often just walk to gardener's house and he'll pass you the keys to the toolshed.

Uh... where exactly is the Grant Root Privileges switch in my Galaxy S2? Must have overlooked it.
Re:good, freedom by Andy+Dodd · 2012-10-08 08:53 · Score: 3, Informative

The complete lack of any signature checking of /system when flashing images with Odin makes rooting Samsungs child's play.
I have to give Samsung credit - while working with the Exynos platform is a nightmare for a Cyanogenmod device maintainer, their lenient bootloader locking policies are unsurpassed.

--
retrorocket.o not found, launch anyway?