[Info] ICS KERNEL SOURCE OPENED!!! * Added MIRROR

Search This thread

tsk1979

Senior Member
Jan 24, 2011
154
8
So what does it mean for users.
Does it mean that ROMs like CM9 will start supporting everything like S-Pen, FM radio etc.,?
Or does it mean we will see slightly modified versions of stock roms. For example touchwiz removed, but swype there.
 

jaiiscool

Senior Member
Jan 30, 2008
1,011
360
Mumbai
jaiparekh.com
So what does it mean for users.
Does it mean that ROMs like CM9 will start supporting everything like S-Pen, FM radio etc.,?
Or does it mean we will see slightly modified versions of stock roms. For example touchwiz removed, but swype there.

All of above +

Awesome custom kernels from devs.

Not sure about FM Radio though....

Sent from my GT-N7000 using xda premium
 

chasmodo

Senior Member
Dec 28, 2011
12,406
41,141
Novi Sad
So what does it mean for users.
Does it mean that ROMs like CM9 will start supporting everything like S-Pen, FM radio etc.,?
Or does it mean we will see slightly modified versions of stock roms. For example touchwiz removed, but swype there.

You are going to see a lot of different Roms once we have several safe custom kerenels.
CM9 will always remain lean and Samsung-free, but there'll be lots of CM9-based hybrid Roms catering for everybody's needs. Also, a lot of TW-based Roms.
I suspect that our Dev will soon start to look like SGS2 one:

http://forum.xda-developers.com/forumdisplay.php?f=1058
http://forum.xda-developers.com/forumdisplay.php?f=1204
 

Arobase40

Senior Member
  • Nov 22, 2010
    1,439
    259
    As suspected, the N7000 ICS kernel is far from different from any i9100 iCS kernels !!!

    I made comparison from all i91000 ICS kernels, and there were only slight diffs from "European" (called Open source : GT-I9100_ICS and GT-I9100P_ICS), with Korean Update 3 and Update 4.

    The major changes came from i9100 Korean ICS update 5 !!!

    This first N7000 ICS kernel is completely different from any previously quoted i9100 ICS kernels, including Korean ICS update 5 !!! :eek:

    Haven't got time to figure out if eMMC "bug" is still there or not, tho. ;)

    Need more time to study all that stuffs... ^^

    ---------- Post added at 04:08 PM ---------- Previous post was at 04:03 PM ----------

    So what does it mean for users.
    Does it mean that ROMs like CM9 will start supporting everything like S-Pen, FM radio etc.,?
    Or does it mean we will see slightly modified versions of stock roms. For example touchwiz removed, but swype there.

    What do all these have anything to do with Kernels ?
     

    Top Liked Posts

    • There are no posts matching your filters.
    • 8
      http://opensource.samsung.com/

      YEAH!! Go go go Custom Kernel!!

      EDIT :

      MIRROR!

      http://www.mediafire.com/?y14t9y9j661jmt3

      is alright?
      6
      Kernels built from the GT-I9100 Update4 source code release were supposed to be safe, but if this is related to firmware, how can they be safe unless they doesn't make use of the firmware ?
      Try reading ALL of the research, instead of picking and choosing what you want to hear.

      It's stated very clearly WHAT the bug in fwrev 0x19 is - it doesn't handle eMMC command 38 (ERASE) well.

      SO - if you don't fire ERASE commands at it, it's safe.

      It may be possible, if you fire ERASE commands at it in the right order, to be safe (this is one of two possibilities for why Gingerbread kernels are safe, the other being that even though MMC_CAP_ERASE is set in the driver, something else is blocking the ERASE command from being sent.)

      What do we know about I9100 Update4? The driver does not set the MMC_CAP_ERASE flag. So the kernel will never send an ERASE command to the kernel.

      And I NEVER stated that N7000 sources would be a direct fix. In fact, I was sure that they would be just as dangerous as the kernels.

      But that's the point of source code - when you have source code for a kernel that is safe but doesn't quite support your hardware correctly (see DAFUQ - safe but broken wifi, no video hwaccel, busted speakerphone), and a kernel that is dangerous but supports all your hardware features - with time you can mix and match and get a kernel that is both safe and supports everything.

      N7000 source isn't a miracle magic bullet that fixes everything immediately. However it's the prerequisite to even having a hope of creating a kernel that is safe AND fully functional for N7000.

      As to Snapdragon devices - they're unknown. However it was confirmed that the SGH-I717 (AT&T Galaxy Note) appears to have eMMC fwrev 0x25, which is safe. It seems that only Exynos devices were ever shipped with fwrev 0x19. Of course - this could change, a new discovery could be made, but I suspect if any of the other Snapdragon-based Samsungs were affected, we'd have heard something about it by now.
      3
      You have a point there. It is very shady/dishonest. But, let's hope they make things right. Maybe they'll give us all Note 10.1s? Haha, jk, that won't happen. But, what good phone does Sony have??

      Sony's hardware isn't so hot, but their developer support and software quality lately has been unparalleled.

      Kind of the opposite of Samsung. Great hardware, utterly **** software quality.
      3
      Cool. I am new to source code, did some googling and searching but was unable to find versioning in the source. Anyone know if its version 25 and not 19, which has the bug?

      Anyone know of a good tutorial on jow to work with the source? All I found are pretty technical, plus in this day and age cannot fathom why there is so much 1970 command line crap to it lmao serious respect to the devs, just the hours they put in to find stuff out. Its amazing

      Sent from my GT-N7000 using Tapatalk 2
      The eMMC fwrev has NOTHING to do with the kernel itself.

      You need TWO things for Superbrick:
      buggy fwrev 0x19 AND a kernel that triggers the bug
      The eMMC firmware currently cannot be changed - what you have is what your device shipped with, and that is 0x19 for every Exynos device anyone has checked so far
      The kernel can be changed to not trigger the bug - but this source definitely still has MMC_CAP_ERASE enabled

      It's not the first kernel/source that Samsung has let loose with the bug trigger - SHW-M250S/K/L devices are also affected.

      So even tho Samsung or whoever was aware of the eMMC bug they didnt fix for source ?


      Sent from my HTC One X using Tapatalk 2
      Nope. Seriously - they've been aware of this since before they started ICS development for this device, as it was discovered initially with Galaxy Nexus PROTOTYPES. As in - prior to November 2011.
      2
      Find the original reason for Entroy's reply, the post he was replying to was asking which kernel version this source release is and if it contained the bug.
      The bug isn't in the kernel, its in the EMMc Firmware that we all have. The Kernel contains a call that triggers the bug.

      New kernel wont erase the bug, but it *might* if we're lucky and guys like Entropy are clever enough, find a way to not have the trigger in it to fire the bug off.
      It looks like we'll have the bug forever...

      There is an emmc firmware fix samsung is working with internally from what I have gathered from the PSA stickies, but the issue is that it is difficult to apply for end users. It would involve a low level flash on the device, known as JTAG i believe, which requires hardware and know-how that is typically only available at service centers.

      I am positive that as costly as the prospect of a recall would be on the how many millions of devices that are affected now, it is most likely the only way to truly fix the problem officially. But due to that cost it seems samsung would be more likely to just issue kernels that work around the problem by simply avoiding the problematic function.

      This means that the GS3 should NOT have this bug at all.

      Sent from my GT-N7000 using XDA