• Introducing XDA Computing: Discussion zones for Hardware, Software, and more!    Check it out!
  • Fill out your device list and let everyone know which phones you have!    Edit Your Device Inventory

[R&D] Unlock Bootloaders

Not open for further replies.
Search This thread


Senior Member
Jun 23, 2010
Chicago - Northwest Suburbs
What about flashing all of the partitions from another phone in place of our own, then copying back our .efs partition (for imei and such).

I just feel like with having the same hardware as other carriers it shouldn't be all too difficult. I heard something about a bootloader from another phone being used but causing bricks. What if the bootloader is dependant upon matching up with another partition.

I know this isn't too technical just trying to think outside of the box.

Sent from my SCH-I535 using Tapatalk 2


Senior Member
Apr 11, 2008
There is a hidden UI:

I found it and many others using Logging Checker by TevE. If I long press on that item I get this:

When I press the big "Qualcomm USB Settings" button I get this:

From what I've read, and I could be very wrong, this is how to access Qualcomm diagnostics over usb. Unfortunately, I don't know much about what to do from there. I know RNDIS stands for Remote Network Driver Interface Specification. And RMNET is Qualcomm's proprietary version of RNDIS for their phones. I assume DM means Direct Media as in DMA direct media access, but I could be wrong.

I haven't had time to have a go at talking to the device using any of the last three ports or combos of ports. Could we possibly disable Qualcomm Secure boot with this? I've seen instructions on how to do it using Jtag, but I don't have a Jtag set up. The reason I bring it up, is because as I understand it Qualcomm Secure boot checks for any changes to the bootloader. If we could disable it, then cracking the bootloader might be a bit easier.


EDIT: I'm working on a text list of all the hidden UI's but I haven't had time to put it all together.

That is the way to access Qualcomm over usb as i just did it last night was able to successfully fully flash the phone to cricket with 3G and no need for an LTE sim card there is a lil more to it but it works very well and DM stands for Diag Mode which is used by QC devices to give access to the Diagnostic functions and reprogramming


Senior Member
Jun 21, 2010
If you think this should be released in a separate thread in Android Development, then thank the post above this one.. I will try to gauge this relatively. Otherwise I will simply post a new version in here when I update.

This works on Linux, Windows and Mac. CASUAL stands for Cross-Platform ADB Scripting, Unified Android Loader. It uses "debug mode" on the device so that must be enabled.. It is an open source project found at android-casual.googlecode.com. It is a packaging system and scripting language. Existing scripts can be converted into CASUAL format to make them work as one-clicks on all platforms. The idea is to simplify all previous development to make things reproducible.

In order to use CASUAL you must have Java. Do you have java?

This CASUAL package contains:
  • Root with DebugFSRoot
  • Install CWM Recovery
  • Reboot into Download/Recovery modes
  • Flash Stock Kernel/Recovery
  • Unroot device
Still to come: Additional tools, Install less secure bootloaders, additional hacks



Download Verizon GS3 CASUAL
Less secure bootloaders you say, does that mean you can actually unlock it and not use kexec, or would the kexec still be in play?


Retired Senior Recognized Developer
Sep 4, 2010
I think I'm about 2 weeks late with this, but like Adam I've also been dealing with the hassles moving.

Here's some info that will get you started looking at the bootloader files in IDA

Each of the bootloader files (sbl1, sbl2, sbl3, tz, rpm, aboot) have a 40 byte header that has info we'll need later

struct sbl_header {
  int load_index;
  int flash_partition_version; 	// 3 = nand
  int image_source_pointer;	// This + 40 is the start of the code in the file
  int image_dest_pointer;	// Where it's loaded in memory
  int image_size;               // code_size + signature_size + cert_chain_size
  int code_size;	        // Only what's loaded to memory
  int signature_addr;
  int signature_size;
  int cert_chain_addr; 		// Max of 3 certs?
  int cert_chain_size;

This is located at the very beginning of the file for sbl2, sbl3, aboot, tz, and rpm, and at 0x280C in the sbl1.

The signature and cert_chain aren't loaded into memory, but they're checked by the previous loader.
For reference, the load order of these bootchain files is:

			appsbl (aboot)

This following was written based on IDA 6.3, but it should be almost exactly the same if you're using the 6.1 leak

First you open the file in IDA
In the "Load a new file" dialog, it'll just show up as "Binary file".
Change the processor type to "ARM processors: ARM" and hit the Set button.
Hit "Processor options", then "Edit ARM architecture options", and change the radiobutton from ARMv6 to Any. Failing to do this will prevent IDA from disassembling some instructions.
Hit Ok, Ok again, and Ok a third time to close the "Load a new file" dialog.

Now you'll be presented with dialog that allows you to customize the loading of the file.
ROM start address: put the value found in the sbl_header.image_dest_pointer (for example, aboot.img is 0x88E00000).
Rom size: put the value from sbl_header.code_size
Loading address: put the image_dest_pointer value
File offset: put 40 (0x2850 for sbl1) since we want to start after the sbl_header structure, and then hit Ok
You may want to add a ram section too, starting at image_dest_pointer+code_size and let it be 0x20000 or so bytes

You may get a popup about ARM/THUMB instruction modes, but generally its detection is good enough that you'll never need to manually adjust the T segment register.
It'll then warn you that it doesn't know where to start analysis, but we do so hit Ok.

Put the cursor at the very first byte (Should be right under where it says CODE32), and hit C to analyze that as code. That's the entry point for the binary

From there you can follow the jumps the program makes looking at the B/BX/BL/BLX instructions, and you can hit C anywhere you think there might be code for it to analyze that area.

If you don't know ARM that well, I'd suggest hitting Options -> General, and tick the "Auto comments" box. That will put a comment out to the side of each instruction explaining what it's doing

Let me know if you have any other questions

Now I've got a few questions

Is there a wiki anywhere I could throw info on? Seems like a waste if I don't share some of the things I find, but they aren't always related to the bootloader so I didn't feel like they were appropriate for this thread.

Is there a more recent Odin3 leak than 3.07? The reason I ask is that the odin handler in aboot seems to be able to handle more than what the 3.07 software can do
Last edited:


Retired Senior Recognized Developer
Feb 18, 2011
Miami, Fl̨̞̲̟̦̀̈̃͛҃҅͟orida
Alright... I seriously need a 1/2 decent processor manual.. We need to figure out how Qualcomm has adapted this BOOT_SCUR pin into this newer chip. This is from the MSM7XXX processor... A previous generation Qualcomm chip.
4.1 Secure boot configuration
The MSM7xxx device can be configured to boot in secure mode using the BOOT_SCURpin or the FORCE_TRUSTED_BOOT Qfuse.
■The BOOT_SCUR pin (N5) determines the boot mode of the MSM7xxx IC. TheMSM7xxx IC can be booted in secure mode and it then follows the boot processdescribed in the previous section. If the device is booted in non-secure mode,authentication of the boot code is not performed


I have a feeling if we can find a labeled processor pinout, busting up the security may be as easy as opening the device and cutting a board trace or popping off a resistor. We seriously need documentation. Hacking consists of Input, Knowledge of what's inside and Output.. We need at least two of these things. Right now we have 1/2 of output. We need a processor manual.


Senior Member
Oct 8, 2010
Just a thought but if we can locate the boatloader signature validation string we can change the data return value to trick the validation process. For example:

(Please don't judge my example code it's for visual concept purpose)

Signature Verified returned 0 = No (abort)

Signature Verified returned 1 = Yes (update)

All we would have to do is change the data return value to equal verified on both options. Then we could load any boot.img we want. This works with several Windows applications for security verification. Editing it would be simple with a hexadecimal editor. Any thoughts?


Inactive Recognized Developer
Dec 6, 2011
So we need to know how to read these Qfuse registers, before we can do much else. Then, if we are lucky,
we could by comparing what we can read, figure out what these registers are.

Here is a more readable summary of what Adam wrote above, as an excerpt from the document above;
"Qualcomm MSM7xxx Application note on Qfuses and security". (BTW. We're using an MSM8960.)

[SIZE=2][I][B]FORCE_TRUSTED_BOOT[/B][/I]:    (Blow to a "1")

Override trusted/non-trusted boot mode input pin (GPIO95/BOOT_SCUR)
0: Do not override
1: Disable non-trusted boot

Blowing this fuse forces the PBL to authenticate the off-chip SBL.

[COLOR=Navy][B]2.3 Mode pin[/B][/COLOR]

The mode pin is used to determine whether the MSM device will boot in secure 
mode from the on-chip PBL. The pin is available behind GPIO and cannot be used 
as GPIO unless the corresponding Qfuse is blown.

[B]BOOT_SCUR[/B] (Input pin):

If this pin (GPIO 95) is enabled (high), it forces the SBL or any subsequent 
code to be authenticated for security. This pin function is available only if 
FORCED_TRUSTED_BOOT fuse is not blown, otherwise it is used as a GPIO.

[B][COLOR=Navy]4.1 Secure boot configuration[/COLOR][/B]

The MSM7xxx device can be configured to boot in secure mode using the BOOT_SCUR 
pin or the FORCE_TRUSTED_BOOT Qfuse.

The BOOT_SCUR pin (N5) determines the boot mode of the MSM7xxx IC. The MSM7xxx 
IC can be booted in secure mode and it then follows the boot process described 
in the previous section. If the device is booted in non-secure mode, 
authentication of the boot code is not performed.

The MSM7xxx IC is booted in non-secure mode when BOOT_SCUR = 0. When the 
BOOT_SCUR pin is pulled to high (BOOT_SCUR = 1) then the MSM IC is booted in 
secure mode and all subsequent code is authenticated. This configuration is only 
valid when the FORCE_TRUSTED_BOOT Qfuse is NOT BLOWN.

The BOOT_SCUR pin functionality is shared with GPIO[95]. When the pin is used as 
BOOT_SCUR, the GPIO functionality is not available.

The MSM7xxx IC can be configured to boot in secure mode by blowing the 
FORCE_TRUSTED_BOOT Qfuse. If this fuse is blown, the SBL and any subsequent code 
are authenticated for security. The BOOT_SCUR pin becomes meaningless when the 
fuse is blown and does not need to be configured for secure boot. Once the 
FORCE_TRUSTED_BOOT fuse is blown, the GPIO[95] is available for use
We may also be interested in the DEBUG_EN_KEY that is a customer specific key that
is used to enable debug features that have been disabled by blowing other Qfuses.

Here is a simple table for specific chip information/generations:

How do we read the Qfuse registers?

Then a small a question and a half.
Is the Verizon SGS3 hardware similar to any other SGS3 "out there"?
Are the boot-loaders on all these the same?
Last edited:


Inactive Recognized Developer
Dec 6, 2011
Just some more info to dump of my mind.

"Usage of system IMEM on AMSS8960"
In System IMEM there is a contiguous 4KB chunk used as shared memory between
multiple subsystems. the offset start address is 0x2A03F000, the below table
lists the detail.

Subsystem        Offset        Size (bytes)
BOOT                    0x0              200  
USB                     0xC8             200  
DDR                     0x190            1024 
EFS                     0x590            200  
HLOS                    0x658            200  
RPM                     0x720            8    
Unused space            0x728            12   
Securemsm               0x734            512  
peripheral debug        0x934            24   
Image version info      0x94C            200  
Unused space            0xA14            1416 
BOOTROM                 0xF9C            100  

"Generate new private key with PHK (Private Hardware Key) in MSM7x30"

"QPST-Emergency Download Support"

I have no idea if any of this is of any use or help. We need to have a more clear approach of how to reach the unlocked bootloader challenge.

What has already been done (including results) and what need to be done?


Senior Member
May 29, 2012
I don't know if this will help at all, but after Googling for documentation , I came across this website from a comment from a post on the qualcomm developer zone thingy...


I just stumbled across this link and thought it would be worth to share with the offset (no pun intended) chance that it would be useful. However, in the case that it is not useful, please remove this post if desired in order to keep the good thoughts flowing.


Inactive Recognized Developer
Dec 6, 2011
Here's some info that will get you started looking at the bootloader files in IDA...
Do you know how to use IDA scripts to trace back, locate and identify various system/library routines/calls?

There's some old scripts around for iPhone reversing (still on ARM), that did this, but I have not seen any for Android... If you know how to use this, please share it. Here is an example of what I'm taking about:

"IDA Pro Signature Files for iPhone Baseband Reversing"


Inactive Recognized Developer
Dec 6, 2011
The only interesting thing that I have found so far is that I can connect via USB while the device is in ODIN Mode, and it shows up as SAMSUNG_MDM in sysinfo. I can then actually connect to that device via serial terminal. Problem is, I don't know any commands to send. ...
(Answer to old post.) You're most likely connected to the modem. Try to send some AT commands and look what happens. Try "ATZ" for example, also make sure you have your terminal set so that you can see what you type...


Senior Member
Jun 1, 2012
I can confirm that a device in ODIN or DM mode creates an /dev/ttyACM file.

Would it be helpful to establish a bounty for a service manual?

Do we need an I535 one or any LTE S3 (including TMo?).


Senior Recognized Developer
May 13, 2009
Do you know how to use IDA scripts to trace back, locate and identify various system/library routines/calls?

There's some old scripts around for iPhone reversing (still on ARM), that did this, but I have not seen any for Android... If you know how to use this, please share it. Here is an example of what I'm taking about:

"IDA Pro Signature Files for iPhone Baseband Reversing"

These are signatures, not scripts. It helps IDA to automatically identify functions from libc and libc++ in binaries compiled with RVCT. Qualcomm tend to use RVCT for their firmwares. Might be helpful in AMSS analyse. I don't think it'll allow IDA to match anything inside bootloaders, but not sure.
Thanks anyway. ;)


Dec 11, 2008
Gilbert, AZ
I can confirm that a device in ODIN or DM mode creates an /dev/ttyACM file.

Would it be helpful to establish a bounty for a service manual?

Do we need an I535 one or any LTE S3 (including TMo?).

I was able to get the Verizon Galaxy Nexus service manuals off of our work Samsung account a while back, I'll check later to see if the Galaxy S3 manuals are on there.. if anyone is interested. (not promising anything yet.. lol)
Not open for further replies.

Top Liked Posts

  • There are no posts matching your filters.
  • 561
    Verizon GS3 is now Bootloader UNLOCKED.
    We now have access to an unsecure bootloader. This was leaked by an African-Canadian Sock Monkey.

    Let me make this clear. If Samsung updates your device's bootloaders, using this tool could potentially brick your device. Once you apply this, never accept a factory update without first flashing the Odin Packages in the Original Post of this thread. As a general rule, you want to be the last guy to apply any Samsung update. Run custom.

    As of the date of this posting, this works great on Linux and it should work wonderfully on Mac too. NOTE: this may work on windows, but please, windows users.. learn to use your computer before you ask questions on XDA-Developers. This is one-click on Linux and Mac every darn time. If you're using Windows, I recommend downloading Windows Ubuntu Installer(WUBI) to install Ubuntu from within Windows.


    1. Open this file
    2. Select Root with DebugFSRoot and Do It
    3. Select Flash Unsecure Aboot and Do It
    4. Use Odin or CWM to flash kernels to your device


    To flash from device without the above tool:
    • root your device
    • Download this link to your /sdcard/Downloads/ folder: http://d-h.st/Piq
    • Type this in the terminal emulator
      su -c dd if=/sdcard/Downloads/aboot.img of=/dev/block/mmcblk0p5

    This was tested with a Sprint kernel flashed via Odin. Although the Sprint kernel caused the device to have a blank screen due to hardware incompatibility, it's more than enough for a proof-of-concept. Stock bootloaders will not let you flash improper kernels with Odin and will cause the device not to boot. This corrects the problem. I'll leave implementation to other developers. If you feel uncomfortable flashing this on your own, wait for your favorite kernel developer to release something.

    Note to developers: This CASUAL package contains everything you need. A jar can be opened as a zip file. CASUAL format sticks all scripts in the /SCRIPTS/ folder. You can obtain all files needed from within this package, then repackage them into CWM format. In order to avoid a mass brick fest, please apply an assert to your CWM scripts to verify ro.build.version.incremental and do not allow updates past what has been tested. As of the time of this writing I535VRALG7B is safe.

    With the unlock of the GS3, this thread is locked. There will be no victory dancing in here. Move along to General or something. This thread will lie dormant until it is needed again in the future. Ralekdev will be releasing another exploit in the future as soon as this one stops working. Feel free to review what was learned until then.

    P.S. Sorry to those who I have offended by having posts removed. I'm also sorry to those who had their intelligence insulted before I had both of our posts removed. I hope you understand that in 6 months from now when everyone forgets about this thread but needs to catch back up, the information will still be right here in condensed format.
    Do not post in here unless you have something constructive to say. "Thanks", "Hey this is wonderful", and any other comments like that are not wanted. They take up space and make it more difficult to find information. I'm requesting that this thread be heavily moderated. In order to work efficiently, information density must be kept high. We are all guilty of adding in a few off-topic sentances from time-to-time, but this thread is strictly business and I expect the moderators to moderate me as well.

    What is this?
    This is the place where we can research and develop a method to unlock the bootloader of the Verizon Galaxy SIII. Hopefully, this will be development at its finest.

    Why not just buy a developer edition
    GTFO! Not a single person got started developing by buying a developer phone. They started developing because they were unhappy with the features of their device and wanted something better. They wanted something more. This developer phone is a tax on developer innovation. We do not stand for that. We will break the security and we will enable XDA-Developers to do what they do best.

    Until security is broken and available for everyone, this device will get updates last, users will be unhappy because there are no additional features and Samsung violates the spirit of Open Source and copyright laws. Take a look at the bottom line of GPL-Violations.org FAQ located here: http://gpl-violations.org/faq/sourcecode-faq.html

    What are the goals?
    • Attain a bootloader recovery - 75% JTAG (the extra 25% will be for a user-friendly method)
      The Galaxy S3 is bootable from SDCard. In case of emergency this is needed. We need to verify that this works on the Verizon GS3 to bring up Odin. This will set up infrastructure for research.
    • Attain a full stock restoration via Odin or Heimdall - 90%
      For use with Odin3.
      Bootloader - BOOTLOADER_I535VRALF2_618049_REV09_user_low_ship.tar.md5 - 1.97 MB - Thanks nbsdx
      PDA - SCH-I535_VZW_1_20120705143513_fti2qg2lmf.zip
      NEED CSC PACKAGE (MODEM, PARAMS and Other Miscellaneous partitions). This is enough to recover a device though.
      To include bootloaders and recovery to a working and stock condition with the EMMC wiped entirely. Heimdall is a work in progress for this device. This will complete the infrastructure needed for research.
    • Collect information
      This will be the longest and most difficult part of this development. The information provided by Qualcomm is not readily available. Samsung is notoriously secretive about their bootloaders. Mainly we, as a community, will generate information. Please post any relevant datasheets, theory-of-operation, or manuals which you can find.
    • Provide a way to remove security checks from Odin3.] 100% - insecure aboot.img which may break in the future
      By removing security checks from Odin3 on the computer or the Loki daemon on the device we can flash anything through Odin or Heimdall.
    • Provide a way to bypass security checks within bootloaders. 200% we have two exploits, only one has been released.
      This is the ultimate goal. Once we can bypass the security checks, kernels can be flashed giving us the control required to develop

    Initial information
    [BOOTLOADER] Locked bootloader research and news: http://forum.xda-developers.com/showthread.php?t=1756919

    My own research

    SBL1 is the first booting partition. Qualcomm provides the Modem partition so it comes first on the EMMC. SBL1 is the first bootloader and that is specified by Qualcomm standards. Qualcom mmake sthe primitive bootloader and allows their customers (Samsung) to make a Secondary bootloader. Samsung chose to use three secondary bootloaders.

    The following 0p* are located in /dev/block/mmcblk*

    0p1 = modem
    Built by se.infra
    I take this to mean this Qualcomm modem was built in Hudson Georgia.
    I was not able to find signatures on this block :). This does NOT mean that there are no signatures on this block. The file is 33 megs. The file is unencrypted.
    The modem uses the BLAST Kernerl ver : Unfortunately we need someone who speaks French(???) to understand how this works http://blast.darkphpbb.com/faq.php
    Judging by the contents of this file, it is an operating system of it's own including keyboard, mouse and a lot of debugging information. We need to find out more about the BLAST Kernel and this partition.

    Samsung Proprietary partitions SBL1,2,3
    Overall I'm not entirely familiar with this new 3 SBL setup. If someone could help me out, that would be great. This 3 SBL setup looks like they tried to adapt (slopily) their IBL+PBL+SBL setup to the Qualcomm and added overhead.

    This block is signed by Samsung, we will not be able to modify it.
    Some Strings we expect to see on UART are:

    This block is signed by Samsung, we will not be able to modify it.

    Some of the strings we may see over UART are:
    RPM loading is successful.
    cancel RPM loading!
    SBL2, End
    SBL2, Delta
    sbl2_hw_init, Start
    sbl2_hw_init, Delta
    sbl2_hw_init_secondary, Start
    h/w version : %d
    sbl2_hw_init_secondary, Delta
    .SBL2, Start
    scatterload_region & ram_init, Start
    .scatterload_region & ram_init, Delta
    sbl2_retrieve_shared_info_from_sbl1, Start
    .sbl2_retrieve_shared_info_from_sbl1, Delta

    This block is signed by Samsung, we will not be able to modify it.

    Possibly useful information:
    SVC: R1-R14

    This block appears to be a full OS of its own. I'm not sure of its purpose.

    op5= aboot
    This block is signed by Samsung, we will not be able to modify it

    This block contains HTML information. It would appear that it is possible to put the device into a mode where it will provide a webserver which displays state information.

    This block appears to be a complete operating system

    This block contains the Loke Daemon which communicates with Odin3.

    0p6= rpm
    This block is signed by Samsung we will not be able to modify it

    0p7= boot
    This is the kernel. There are several things we can do here... I belive this package itself is not signed, but the zImage itself is... here is the bootimg.cfg file

    [email protected]:~/Desktop/VZWGS3$ cat ./bootimg.cfg 
    bootsize = 0xa00000
    pagesize = 0x800
    kerneladdr = 0x80208000
    ramdiskaddr = 0x81500000
    secondaddr = 0x81100000
    tagsaddr = 0x80200100
    name = 
    cmdline = console=null androidboot.hardware=qcom user_debug=31

    It may be possible to use that cmdline variable as an exploit.

    0p8= tzTrust Zone
    0p9= pad
    0p10= param -boot mode parameters - this could be a potential exploitation point.
    0p11= efs -serial numbers
    I've honestly got no clue about most of the following partitions.
    0p12= modemst1
    0p13= modemst2
    0p14= system - Android stuff
    0p15= userdata - App Stuff
    0p16= persist
    0p17= cache - Storage for updates
    0p18= recovery - recovery partition
    0p19= fota
    0p20= backup
    0p21= fsg
    0p22= ssd
    0p23= grow

    External UART log from initial power up:
    [1630] AST_POWERON
    [    0.000000] heap->name mm, mb->start c0000000
    [    0.000000] Reserving memory at address ea000000 size: 100000
    [    0.000000] sec_dbg_setup: [email protected]
    [    0.000000] sec_dbg_setup: secdbg_paddr = 0x88d90004
    [    0.000000] sec_dbg_setup: secdbg_size = 0x40000
    [    0.000000] etb_buf_setup: [email protected]
    [    0.000000] etb_buf_setup: secdbg_paddr = 0x8fffb9c0
    [    0.000000] etb_buf_setup: secdbg_size = 0x4000
    [    0.174515] rdev_init_debugfs: Error-Bad Function Input
    [    0.174881] AXI: msm_bus_fabric_init_driver(): msm_bus_fabric_init_driver
    [    0.176957] sec_debug_init: enable=0
    [    0.177475] ec_debug_nit: restrt_reason: 0xdf0085c
    [    .216358] msm8960_iit_cam:292]settingdone!!
    [    0.25006] i2c 2c-14: Inalid 7-bi I2C addrss 0x00
        0.25237] i2c ic-14: Can' create evice at x00
    [   0.252220]i2c i2c-1: Failed o registeri2c clien cmc624 t 0x38 (-6)
    [    .252250] 2c i2c-19:Can't crete deviceat 0x38
        0.25433] rdevinit_debufs: Error-ad Functin Input
        0.25222] max892 19-006: DVS mode disabledbecause VD0 and VI1 do not ave prope control.
    [    0.79536] ms_etm msm_tm: ETM tacing is ot enable beacaussec_debug s not enaled!
    [   0.284449 smd_chanel_probe_orker: alocation tble not iitialized
                                                                      [    0.38766] pm_untime: fil to wak up
    [   0.362032]hdmi_msm dmi_msm.1 externalcommon_stte_create sysfs grup de39e68                                                                   
    [    0362673] Iside writback_drivr_init                                                                                                         
    [   0.36275] Insidewritebackprobe                                                                                                               
    [    1.244803] TZCOM: unable to get bus clk                                                                                                     
    [    1.431680] cm36651_setup_reg: initial proximity value = 3                                                                                   
    [    1.549671] msm_otg msm_otg: request irq succeed for otg_power                                                                               
    [    1.566702] mms_ts 3-0048: [TSP] ISC Ver [0xbb] [0x20] [0x20]                                                                                
    [    1.571341] mms_ts 3-0048: [TSP] fw is latest. Do not update.                                                                                
    [    1.583488] [__s5c73m3_probe:3818] S5C73M3 probe                                                                                             
    [    1.587089] [s5c73m3_sensor_probe_cb:3793] Entered                                                                                           
    [    1.591942] [s5c73m3_i2c_probe:3675] Entered                                                                                                 
    [    1.596123] [s5c73m3_init_client:3381] Entered                                                                                               
    [    1.600579] [s5c73m3_i2c_probe:3695] Exit                                                                                                    
    [    1.604608] [s5c73m3_sensor_probe:3726] Entered                                                                                              
    [    1.609095] [s5c73m3_spi_init:226] Entered                                                                                                   
    [    1.613154] [s5c73m3_spi_probe:191] Entered                                                                                                  
    [    1.617335] [s5c73m3_spi_probe:201] s5c73m3_spi successfully probed                                                                          
    [    1.623561] [s5c73m3_sensor_probe :  3749] Probe_done!!                                                                                      
    [    1.672638] mmc0: No card detect facilities available                                                                                        
    [    1.682984] aat1290a_led_probe : Probe                                                                                                       
    [    1.693850] msm_soc_platform_init                                                                                                            
    [    1.697298] msm_afe_afe_probe                                                                                                                
    [    1.843064] msm_asoc_pcm_new                                                                                                                 
    [    1.849748] msm_asoc_pcm_new                                                                                                                 
    [    2.023134] set_dload_mode <1> ( c00176d4 )                                                                                                  
    [    2.052220] cypress_touchkey 16-0020: Touchkey FW Version: 0x06                                                                              
    [    2.123851] init: /init.qcom.rc: 466: invalid command '/system/bin/log'                                                                      
    [    2.129620] init: /init.qcom.rc: 573: ignored duplicate definition of service 'sdcard'                                                       
    [    2.137402] init: /init.qcom.rc: 586: ignored duplicate definition of service 'ftm_ptt'                                                      
    [    2.145490] init: /init.target.rc: 73: ignored duplicate definition of service 'thermald'                                                    
    [    2.154677] init: could not open /dev/keychord                                                                                               
    [    2.239951] init: Device Encryption status is (0)!!                                                                                          
    [    2.243705] init: [disk_config] :::: fsck -> /dev/block/mmcblk0p15 (ext4):::::                                                               
    [    2.251823] init: [disk_config] ext_check -> /system/bin/e2fsck -v -y /dev/block/mmcblk0p15                                                  
    [    2.588921] init: [disk_config] ext_check ->ok                                                                                               
    [    2.611597] init: [disk_config] :::: fsck -> /dev/block/mmcblk0p17 (ext4):::::                                                               
    [    2.617762] init: [disk_config] ext_check -> /system/bin/e2fsck -v -y /dev/block/mmcblk0p17                                                  
    [    2.655333] init: [disk_config] ext_check -> ok                                                                                              
    [    2.664947] init: [disk_config] :::: fsck -> /dev/block/mmcblk0p11 (ext4):::::                                                               
    [    2.671081] init: [disk_config] ext_check -> /system/bin/e2fsck -v -y /dev/block/mmcblk0p11                                                  
    [    2.704532] init: [disk_config] ext_check -> ok                                                                                              
    [    3.259056] init: cannot find '/system/etc/install-recovery.sh', disabling 'flash_recovery'                                                  
    [    3.270471] init: cannot find '/system/bin/dmbserver', disabling 'dmb'

    External UART log from battery-pull and reinsert
    [1630] AST_POWERON
    [    0.000000] heap->name mm, mb->start c0000000
    [    0.000000] Reserving memory at address ea000000 size: 100000
    [    0.000000] sec_dbg_setup: [email protected]
    [    0.000000] sec_dbg_setup: secdbg_paddr = 0x88d90004
    [    0.000000] sec_dbg_setup: secdbg_size = 0x40000
    [    0.000000] etb_buf_setup: [email protected]
    [    0.000000] etb_buf_setup: secdbg_paddr = 0x8fffb9c0
    [    0.000000] etb_buf_setup: secdbg_size = 0x4000
    [    0.174484] rdev_init_debugfs: Error-Bad Function Input
    [    0.174851] AXI: msm_bus_fabric_init_driver(): msm_bus_fabric_init_driver
    [    0.176926] sec_debug_init: enable=0
    [    0.177445] sc_debug_iit: restat_reason  0xdf0086c
    [    0216206] [sm8960_int_cam:299]setting one!!
    [   0.217915 select_req_plan:ACPU PVS:Nominal
        0.25206] i2c ic-14: Invaid 7-bit 2C addres 0x00
    [   0.25207] i2c i2-14: Can'tcreate deice at 0x0
    [    0252250] 2c i2c-19 Failed t register 2c clientcmc624 at0x38 (-16
    [    0252250] ic i2c-19: an't creae device t 0x38
    [   0.25243] rdev_iit_debugs: Error-Bd Functio Input
    [   0.25292] max895 19-0060:DVS modesdisabled ecause VI0 and VID do not hve propercontrols.
                                                                                               [    0.29536] msmetm msm_em: ETM trcing is nt enable!
    [    0.35797] pm_rntime: fal to wakeupllcation tale not intialized
    [    .362093] dmi_msm hmi_msm.1:external_ommon_stae_create:sysfs grop de39e60                                                                   
    [    0.62734] Inide writeack_driverinit                                                                                                         
    [   0.36285] Inside riteback_robe                                                                                                               
    [    1.244803] TZCOM: unable to get bus clk

    possible exploitations
    Possible entry point MODEM - Someone with a JTAG setup test viability of modifying a single byte on /dev/block/mmcblk0p1
    Possible entry point PARAMS - Samsung stores their boot parameters in PARAMS partition. It may be possible to modify PARAMS for insecure boot
    Possible entry point BOOT - Modify CMDLINE parameter to load information from another location.
    Possible entry point BOOT - We may be able to shove an insecure bootloader into memory, boot into that, and then use the recovery partition as our kernel partition. Bauwks 2nd U-Boot. U-Boot is available for the Exynos 4412, we need to find one for Qualcomm.
    Possible entry point SYSTEM - It may be possible to use a 2nd init hack from this partition to load custom kernels into memory and reboot the kernel.

    Current tasks
    What do all of these partitions do?
    Do we have a SDCard based recovery?
    Where can we find an Odin3 CSC Flash?
    Testing methods above is required
    I have heard, but do not know, that there may be plans to get one of the developer phones into Adam's hands to extract from. That may provide insight into how to disable Qualcomm Secure Boot no? Anyone care to shed some light on if this is still planned or not? Thanks

    I don't need another device. I want all of the partitions from a developer device and I'd like to work with someone who has one. Remote access via "WirelessADB" and the device set to be in the "DMZ" of a router would be sufficient for all tests I would need to do.

    Just as an update, I'm slowly getting back to work. For those who were wondering, I packed up everything and moved. I have my stuff 90% set up. I'm just getting back on it. I'm working on compiling all of the Verizon GS3 exploits into a single CASUAL one-click package. Root, recovery, Busybox, Basic Hacking Tools.

    Once I've got a CASUAL package put together I'll go through and read this thread again from start to finish and figure out what needs work... my mind is totally off-topic right now after a move. Time to get back to work. I hope to have some big news at the end of next week.
    It's been a few days so I wanted to give an update on the signature check on boot.img

    As has been previously guessed, everything important in boot.img is included in the signature check

    page_size is always 0x800 since we're using emmc boot

    hash_size = 0x800 (read the first page with the boot_img_header)
    hash_size += page_size * ((page_size + ramdisk_size - 1) / page_size)
    hash_size += page_size * ((page_size + kernel_size - 1) / page_size)
    hash_size += page_size * ((page_size + second_size - 1) / page_size)

    For the stock boot.img, this should come out to be 0x573000, so the first 0x573000 bytes in boot.img are checked.

    These bytes are then SHA1 hashed and passed to the verification function

    After hash_size bytes is a series of 0x100 byte blocks that will be passed to the verification function (img_sig_data parameter below)

    The verification function uses the following structure

    struct sig_ctx_t {
    	int count;
    	int seed[65];
    	int subcheck_seed[64]; // possibly a modulus

    This sig_ctx is located in aboot.img at file offset 0x12642C in VRALF2 and VRALG1 (It'll start with bytes 0x40, 0x00, 0x00, 0x00)

    I've cleaned up the first function a bit from what IDA/Hex-Rays spit out, but the second function I haven't simplified as much

    int signature_check_data(sig_ctx_t *sig_ctx, char *img_sig_data, signed int signature_len, char *sha1_of_contents) {
    	int* img_ofs_0x100 = (int*)(img_sig_data + 0x100);
    	int* img_ofs_0x200 = (int*)(img_sig_data + 0x200);
    	int* img_ofs_0x300 = (int*)(img_sig_data + 0x300);
    	int* img_ofs_0x400 = (int*)(img_sig_data + 0x400); // Temporary storage
    	// Copy 0x0 block to 0x100
    	memcpy(img_ofs_0x100, &img_sig_data[0], signature_len);
    	// ofs_0x200 is filled with byte-swapped ints from img_ofs_0x100
    	for (int i = 0; i < sig_ctx->count; i++) {
    		img_ofs_0x200[i] =  htonl(img_ofs_0x100[sig_ctx->count - 1 - i]);
    	// subcheck(sig_block *block, int *output, int *input1, int *input2)
    	// multiplication maybe?
    	signature_subcheck(sig_ctx, img_ofs_0x300, img_ofs_0x200, sig_ctx->subcheck_seed);
    	signature_subcheck(sig_ctx, img_ofs_0x400, img_ofs_0x300, img_ofs_0x300);
    	signature_subcheck(sig_ctx, img_ofs_0x300, img_ofs_0x400, img_ofs_0x200);
    	if ( sig_ctx->count )
    		count_minus_1 = sig_ctx->count - 1;
    		v18 = img_ofs_0x300[sig_ctx->count - 1];
    		v19 = sig_ctx->seed[sig_ctx->count]; // seed[64]
    		// v19 = *(&sig_ctx->count + sig_ctx->count + 1);
    		if ( v18 >= v19 )
    			if ( v18 == v19 )
    				for (int i = 0; i < sig_ctx->count; i++) {
    					int v22 = img_ofs_0x300[sig_ctx->count - 1 - i];
    					int v23 = sig_ctx->seed[sig_ctx->count - 1 - i];
    					if (v22 < v23) {
    						goto LABEL_18
    			if ( sig_ctx->count > 0 )
    				int carry = 0;
    				for (int i = 0; i < sig_ctx->count; i++) {
    					uint64 temp = img_ofs_0x300[i] - (uint64)sig_ctx->seed[i + 1];
    					img_ofs_0x300[i] = img_ofs_0x300[i] - sig_ctx->seed[i + 1] + carry;
    					carry = (int)(temp >> 32); // get high 32 bits
    		// Store the calculation back into img_ofs_0x100
    		for (int i = 0; i < sig_ctx->count; i++) {
    			int val = img_ofs_0x300[sig_ctx->count - 1 - i];
    			char* dest = &img_ofs_0x100[i];
    			dest[0] = (val & 0xFF000000) >> 24;
    			dest[1] = ((val & 0x00FF0000) >> 16) & 0xFF;
    			dest[2] = ((val & 0x0000FF00) >> 8) & 0xFF;
    			dest[3] = (val & 0xFF);
    		if (memcmp(img_ofs_0x100, sig_check_compare_result, 236)) // sig_check_compare_result is a char[236] with the first 2 bytes 0x00, 0x01, and the rest 0xFF
    			return 0;
    		if (signature_len > 236) {
    			if (memcmp(&img_ofs_0x100[236], sha1_of_contents, signature_len - 236)) // 256-236 = 20
    				return 0;
    			// Signature passed
    			return 1;
    	return 0;

    Here's the subcheck function, it looks like arbitrary-precision math, possibly mulmod

    void __fastcall signature_subcheck(sig_ctx_t *sig_data, int *output, int *input1, int *input2)
      int v5; // [email protected]
      int count; // [email protected]
      unsigned __int64 v7; // [email protected]
      unsigned __int64 v8; // [email protected]
      int inner_index; // [email protected]
      int block1_pos; // [email protected]
      int v11; // [email protected]
      __int64 v12; // [email protected]
      int v13; // [email protected]
      unsigned __int64 v14; // [email protected]
      int v15; // [email protected]
      int v16; // [sp+18h] [bp-48h]@6
      unsigned int v17; // [sp+1Ch] [bp-44h]@6
      int outer_index; // [sp+2Ch] [bp-34h]@5
      if ( sig_data->count > 0 )
        v5 = 0;
          output[v5++] = 0;                         // this do while is just memset(output, 0, 4 * sig_data->count)
          count = sig_data->count;
        while ( sig_data->count > v5 );
        if ( count > 0 )
          outer_index = 0;
            v16 = input1[outer_index];
            v7 = (unsigned int)v16 * (unsigned __int64)(unsigned int)*input2 + (unsigned int)*output;// v7 = input1[outer_index] * (uint64)input2[0] + output[0]
            v17 = sig_data->seed[0] * v7;
            v8 = sig_data->seed[1] * (unsigned __int64)v17 + (unsigned int)v7;
            if ( count <= 1 )
              block1_pos = 1;
              inner_index = 0;
              block1_pos = 1;
                v7 = (unsigned int)v16 * (unsigned __int64)(unsigned int)input2[block1_pos]
                   + (unsigned int)output[block1_pos]
                   + HIDWORD(v7);
                v8 = sig_data->seed[inner_index + 2] * (unsigned __int64)v17 + HIDWORD(v8) + (unsigned int)v7;
                output[inner_index] = v8;
              while ( block1_pos < sig_data->count );
            output[block1_pos - 1] = HIDWORD(v8) + HIDWORD(v7);
            if ( (HIDWORD(v8) + (unsigned __int64)HIDWORD(v7)) >> 32 )
              if ( sig_data->count <= 0 )
              v11 = 0;
              v12 = 0LL;
              v13 = 0;
                v14 = (unsigned int)output[v11] - (unsigned __int64)sig_data->seed[v11 + 1];
                v15 = output[v11] - sig_data->seed[v11 + 1];
                output[v11] = output[v11] - sig_data->seed[v11 + 1] + v12;
                count = sig_data->count;
                v12 = (signed int)((__PAIR__(HIDWORD(v14), v15) + v12) >> 32);
              while ( v13 < sig_data->count );
              count = sig_data->count;
          while ( outer_index < count );

    The goal is to make it so that after all the calculations the 256 byte block located at img_sig_data+0x100 has the contents 0x00, 0x01, 0xFF * 236, and then the sha1 of our boot.img

    I'm in the middle of moving at the moment, so I don't have as much time as I would like to look at this right now, but that should clear up in a few days.

    Also, if there's any interest I can post a guide on how to get the bootloader files loaded into IDA for analysis. Some knowledge of ARM assembly would be required though.


    In other news, I found what keeps resetting the 16 byte encrypted romtype in param.img. It's libcordon.so, which is from /system/app/SysScope.apk (it'll also be copied to /system/lib/libcordon.so). It's using quite a few checks to see if you've modified your system.

    There's an adb scanner, checking to see if you've changed the ro.secure or ro.debuggable props.

    The root process scanner checks running processes and returns true if any are found running as root that are not one of:
    "debuggerd", "init", "installd", "servicemanager", "vold", "zygote", "netd", "ueventd", "dock_kbd_attach", "pppd", "pppd_runner", "mpdecision", "thermald", "hdmid", "sec_keyboard", "seccmmond", "mfsc", "mfdp"

    There's also a partition check, kernel checker, su scanner, and a file scanning mechanism using data from a sqlite db

    So to completely remove the Samsung custom screen on bootup and 5 second delay you'd need to disable the SysScope.apk, then encrypt and write the 16 bytes yourself using 0xFF000000 as the first int to mark yourself as official
    A gentleman named Lee contacted me via email. He said he has 0 posts so he could not post in here. This post contains his email to me. I am not wrapping it in quotes because quotes are destroyed in future posts. This is literally the best development we've had in this thread.

    ------email from Lee------
    I've been looking at the bootloader in aboot.img the past day or so and wanted to contribute what I know about the param.img partition and how it's used. I've been following the thread at xda, but since my account has 0 posts I can't actually post this in that thread.

    Please note these are a little rough around the edges, just things I jotted down while reverse engineering.

    param.img Structure

    At offset 0 there's an 88 byte structure I've called the header

    struct param_header {
    int status; // need to investigate more. some relationships between this and boot modes. 4 == firmware error int unk_04; // haven't seen this used anywhere int unk_08; // haven't seen this used anywhere int emmc_checksum_attempted; int emmc_checksum_ok; int nvdata_backup; // says whether we have a backup of modemst1 in "fsg" partition and a backup of modemst2 in "backup" partition?
    int unk_18[16]; // haven't seen this used anywhere };

    status (NEEDS WORK):
    1 = ?
    2 = boot_mode 3?
    3 = recovery?
    4 = boot_mode 1 - fastboot. displays "firmware update issue" image
    5 = boot_mode 4?

    at offset 0x900000 there's a structure controlling some debug variables

    struct param_debug {
    int debug_level;
    int unk_04; // 4 in dumps. haven't seen this used anywhere int unk_08; // 0 in dumps. haven't seen this used anywhere int emmc_checksum_attempted; // mirror of param_header.emmc_checksum_attempted
    int emmc_checksum_ok; // mirror of param_header.emmc_checksum_ok };

    About param_debug.debug_level:
    It has 3 possible values, and it changes some flags are passed to the kernel.
    DLOW is the default, but some features like ramdump mode only work on DMID or DHIG

    1. 0x574F4C44 (DLOW) - Low debug setting strcat(boot_img_hdr->cmdline, " androidboot.debug_level=0x4f4c");// OL strcat(boot_img_hdr->cmdline, " sec_debug.enable=0"); strcat(boot_img_hdr->cmdline, " sec_debug.enable_user=0");

    2. 0x44494D44 (DMID) - Mid-level debugging strcat(boot_img_hdr->cmdline, " androidboot.debug_level=0x494d");// IM strcat(boot_img_hdr->cmdline, " sec_debug.enable=1"); strcat(boot_img_hdr->cmdline, " sec_debug.enable_user=0");

    3. 0x47494844 (DHIG) - Full debugging
    strcat(boot_img_hdr->cmdline, " androidboot.debug_level=0x4948");// IH strcat(boot_img_hdr->cmdline, " sec_debug.enable=1"); strcat(boot_img_hdr->cmdline, " sec_debug.enable_user=1"); strcat(boot_img_hdr->cmdline, " slub_debug=FPUZ");

    Check drivers/misc/sec_misc.c for what these values do for the kernel

    At offset 0x9FFC00 (sizeof(param.img) - 0x400 is how the offset is calculated by the BL):
    Here are 16 bytes unique to each device, and they are part of what determines whether or not you have a custom rom.

    It's AES128 encrypted using a key made from the emmc's psn and some static data

    Key generation:
    First, the 4byte psn is expanded to 8 bytes

    char first_half[14];
    snprintf(first_half, 13, "%08x", mmc_get_psn()); memcpy(aes_initial_key, first_half, 8);

    The second half is calculated based on all static data

    char custom_check_index_shuf_table[] = { 1, 3, 2, 4, 5, 1, 0, 4, 4, 5, 4, 0 }; char custom_check_table[] = { 0x40, 0x74, 0x25, 0x61, 0x21, 0x74, 0x70, 0x62, 0x62, 0x24, 0x33, 0x5E }; char romtype_enc_key_buf[32];

    char* custom_check_shuffle_calc(signed int always_199, int count) { int out_index; // [email protected] int last_index; // [email protected] int odd_index; // [email protected] int table_index; // [email protected] char table_value;

    if ( count <= 0 )
    out_index = 0;
    out_index = 0;
    last_index = 0;
    odd_index = always_199 & 1;
    always_199 >>= 1;
    table_index = odd_index + 2 * last_index; table_value = custom_check_table[table_index]; last_index = custom_check_index_shuf_table[table_index];
    romtype_enc_key_buf[out_index++] = table_value; } while ( out_index != count ); } romtype_enc_key_buf[out_index] = 0; return romtype_enc_key_buf; }

    This function is used like this (the parameters are always 199 and 8 in the vzw aboot):
    char* second_half = custom_check_shuffle_calc(199, 8); memcpy(&aes_initial_key[8], second_half, 8);

    Now we have 16 bytes in aes_initial_key, but it's shuffled again with the following function

    char custom_check_final_index_table[] = { 0, 4, 5, 0xD, 3, 8, 0xE, 9, 0xA, 2, 1, 7, 0xB, 6, 0xC, 0xF }; void custom_check_shuffle_final_key(char *iv, char *final) { int v2; // [email protected] int v5; // [email protected]

    v2 = 0;
    final[custom_check_final_index_table[v2]] = iv[v2];
    while ( v2 != 16 );
    v5 = 0;
    final[custom_check_final_index_table[v5]] = iv[v5] ^ final[v5];
    while ( v5 != 16 );

    char aes_final_key[16];
    custom_check_shuffle_final_key(aes_initial_key, aes_final_key);

    This final key should be able to decrypt the 16 bytes

    The first 4 decrypted bytes cast to an int will be 0xFF000000 if you're running an official rom, or 0xEE000000 if you've flashed something custom If it's 0xEE000000 then you will be shown the "Custom" boot screen with the padlock on it, and it also causes a call to mdelay(5000) before actually booting the kernel.
    I've also seen 0xCC000000 mentioned in debug prints, causing it to print the device status as "Scanning" instead of "Official" or "Custom"

    Unfortunately this doesn't seem to help much with the boot.img check, but I've found where that is and am reversing it now.


    DDI Data
    Here's where the values like the flash count are stored (sometimes this might be called triangle state?) It's stored at 0x3FFE00 on the mmc

    struct ddi_data {
    int magic; // must be 0x12340012
    int custom_flash_count;
    int odin_count;
    int binary_type; // 0 = samsung official, 1 = custom, 2 = "Unknown"
    char model_name[16];
    int rom_type; // this is the first 4 bytes of the decrypted 16 bytes in the param partition. 0xFF000000 = samsung, 0xEE000000 = custom }


    Reboot Reason

    Values and effects for the reboot reason stored at 0x2A03F65C

    0x12345671 - ?
    0x12345678 - Normal mode

    0x77665500 - FASTBOOT_MODE. displays "downloading" boot image
    0x77665501 - ? seen checked but haven't found it used anywhere
    0x77665502 - RECOVERY_MODE. sets param_header.state to 3
    0x77665503 - sets param_header.state to 4. haven't seen it actually used

    0x77665507 - display the "not authorized" picture

    if ((reason & ~0xF) == 0x77665510) then they're commands for manipulating the nvdata I wouldn't play around with these unless you really know what you're doing All of them reboot the device into the normal mode except 0x77665515

    0x77665511 - copy modemst1 to fsg partition and copy modemst2 to backup partition. sets param_header.nvdata_backup to 1
    0x77665512 - copy fsg to modemst1 and copy backup to modemst2. checks to ensure param_header.nvdata_backup=1 first
    0x77665514 - erase fsg and backup partitions. clears param_header.nvdata_backup
    0x77665515 - same as 0x77665511 but then reboots the device into RECOVERY_MODE

    0x776655EE - RAMDUMP_MODE (only valid if param_debug.debug_level is DMID/DHIG)

    0xABCD4F4C - set param_debug.debug_level to DLOW 0xABCD494D - set param_debug.debug_level to DMID
    0xABCD4948 - set param_debug.debug_level to DHIG


    boot_type INCOMPLETE
    1 = fastboot
    2 = ramdump mode
    3 = recovery. resets param_debug
    4 = ?



    0xF00 - jig mask
    0x100 - put the device into factory mode
    0x400 - change "console" boot parameter to "console=ttyHSL0,115200,n8%s" where %s is replaced by whatever was originally after "console="



    In addition to the ODIN/LOKE handshake sequence I saw in heimdall, there are 2 more in the S3.
    Send "FPGM" and you should get a response of "OK". It functions exactly as the ODIN/LOKE sequence.
    Send "ROOTING" and it responds with the current DDI data and terminates.