[R&D] System Dump, FSTAB, ADB, FASTBOOT, HARDWARE and Screenshots

Search This thread

Childofthehorn

Inactive Recognized Developer
Jun 6, 2010
260
185
DFW, Texaaas
FSTAB:
http://pastebin.com/LN75y6Jj

Build.Prop:
http://pastebin.com/ehgwGL8A

System Dump:
http://www.mediafire.com/download/ux9gg1661ojkao9/sprat-user_4.4W_KMV78N_SYSTEM_DUMP.zip
*Copyright owned by Samsung and Google

Getprop
http://pastebin.com/2mhPZhLV

Hardware Info
https://www.youtube.com/watch?v=NNtBzOiXLPw

IMG_20140701_142732.jpg

ENABLE ADB
- Tap on top right of watchface
- scroll down to Settings
- Scroll down and tap About
- Tap Build Number 5x's
- Go back to settings and Development Options should be available
- Enable ADB debugging
- Plug in the charger dock with a USB connected to the computer
- Authorize on your Phone for USB


14+-+1

RECOVERY MODE


IMG_20140701_151120.jpg

FASTBOOT/BOOTLOADER Mode
 
Last edited:

Childofthehorn

Inactive Recognized Developer
Jun 6, 2010
260
185
DFW, Texaaas
Codename: SPRAT

Product: Gear Live

Key Specs:

- Qualcomm 400 chipset (Quadcore msm8226 - Same exact one as Moto G!)
- Adreno 305
- OpenGL 3.0 support
- 320x320 screen (hdpi)

Sensors / Parts:
- Invensence MPU6515 MEMs Gryo HERE
- AK8963c 3 Axis Compass and Hall effect sensor HERE (TOGETHER 1 +2 = InvenSense MP92M 9-axis Gyro + Accelerometer + Compass)
- Heart Rate Detector ( didn't see what the chipspec was).
- Samsung Driver abstracted "sensors" (just conglomerates info from the existing sensors).
- MAX77836—MUIC
- STMicroelectronics STM32F401 Cortex M4 microcontroller
- Qualcom PM8226 power management IC
- InvenSense ICS-43430 microphone
- Cypress Semiconductor CYTMA54 touchscreen controller


* holding for space *
 
Last edited:

AdamOutler

Retired Senior Recognized Developer
Feb 18, 2011
5,224
9,827
Miami, Fl̨̞̲̟̦̀̈̃͛҃҅͟orida
Fastboot commands extracted from aboot:
Code:
continue
reboot
reboot-bootloader
preflash
flash
boot
oem lock
oem enable-console
oem disable-console
oem device-info
oem enable-charger-screen
oem disable-charger-screen
oem off-mode-charge 0
oem off-mode charge 1
default settings:
Code:
adamoutler@adamoutler-desktop:~/Downloads$ sudo fastboot oem device-info
...
(bootloader) 	Device tampered: false
(bootloader) 	Device unlocked: false
(bootloader) 	Charger screen enabled: true
(bootloader) 	Display panel: �������������������������������������������
You'll want disable-console, enable-charger-screen, and off-mode-charge 1 in order to get stock functionality with best battery life. For debugging, you can switch these options.

UART boot log: http://pastebin.com/M5ctEFtT

It was said that the Gear Live should only be running one processor. You can see in the UART log that there are 4 processors active. Here is /proc/cpuinfo which confirms this.
Code:
shell@sprat:/proc $ cat cpuinfo
processor	: 0
model name	: ARMv7 Processor rev 3 (v7l)
BogoMIPS	: 38.40
Features	: swp half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt 
CPU implementer	: 0x41
CPU architecture: 7
CPU variant	: 0x0
CPU part	: 0xc07
CPU revision	: 3

processor	: 1
model name	: ARMv7 Processor rev 3 (v7l)
BogoMIPS	: 38.40
Features	: swp half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt 
CPU implementer	: 0x41
CPU architecture: 7
CPU variant	: 0x0
CPU part	: 0xc07
CPU revision	: 3

processor	: 2
model name	: ARMv7 Processor rev 3 (v7l)
BogoMIPS	: 38.40
Features	: swp half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt 
CPU implementer	: 0x41
CPU architecture: 7
CPU variant	: 0x0
CPU part	: 0xc07
CPU revision	: 3

processor	: 3
model name	: ARMv7 Processor rev 3 (v7l)
BogoMIPS	: 38.40
Features	: swp half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt 
CPU implementer	: 0x41
CPU architecture: 7
CPU variant	: 0x0
CPU part	: 0xc07
CPU revision	: 3

Hardware	: Qualcomm MSM 8226 (Flattened Device Tree)
Revision	: 0004
Serial		: 0000000000000000
 
  • Like
Reactions: Childofthehorn

AdamOutler

Retired Senior Recognized Developer
Feb 18, 2011
5,224
9,827
Miami, Fl̨̞̲̟̦̀̈̃͛҃҅͟orida
Recommendations for developers to increase battery life:
1. Remove useless drivers such as:
Code:
  SCSI subsystem initialized
Code:
 SCSI Media Changer driver v0.25
Code:
 sdhci: Secure Digital Host Controller Interface driver
and many others
2. Disable unused processors, or offer a "power saving" mode to disable those additional processors
3. examine processor scaling
4. reduce use of SELinux. it's worthless.
4a. when device is near phone, phone security is off
4b. turning on ADB and accessing storage is easy as security is disabled on the phone
4c. because of the above reasons the device causes the phone to be insecure
4d. the device is wasting battery on its own security while reducing the security of the phone.

5. enable thermal max frequency
Code:
[    4.700499] msm_thermal:disable_msm_thermal Max frequency reset for CPU1
[    4.700509] msm_thermal:disable_msm_thermal Max frequency reset for CPU2
[    4.700517] msm_thermal:disable_msm_thermal Max frequency reset for CPU3

6. above all: audit your logs.
 

20mmrain

Member
Apr 2, 2013
42
1
Milwaukee
Anyone having this issue? Trying to unlock my Samsung Gear Live. ADB recognized device, Enabled communication through my phone, watch boots just fine using "ADB reboot bootloader" however, when entering "fastboot oem unlock" I get "waiting for device"?

Yes my watch is plugged in to my computer. I have tried several different ports. Any help?
I couldn't find the answer to this using the search function.

Thanks guys

*Edit*

I figured out the issue. If anyone is having this problem visit here:
http://stackoverflow.com/questions/24583354/adb-doesnt-recognize-samsung-gear-live-or-lg-g-watch
Follow the directions accpet one thing. Toward the bottom you will see a post telling you to install a "Samsung Device" instead of a generic "Android Device" Make sure you install the "Samsung device" if using a Gear Live.
 
Last edited:

AdamOutler

Retired Senior Recognized Developer
Feb 18, 2011
5,224
9,827
Miami, Fl̨̞̲̟̦̀̈̃͛҃҅͟orida
This device charges VERY slowly, and on top of that, it's also picky about the charging port. In order to be sure you're always charging, you'll want to use a USB condom. This is important because many aftermarket chargers are unlabeled.

P_20140720_203409_HDR.jpg


Here you can see it won't charge at all on an "Apple" charger
P_20140720_203436_HDR.jpg


And it charges well on an "Android" charger with D+ and D- tied together.
P_20140720_203721_HDR.jpg




The good news is that the charging rate is generally controlled by software, so if samsung would release kernel code, we could likely fix this.
 
Last edited:
  • Like
Reactions: Muo and zuberio

JoeFCaputo113

Senior Member
Feb 2, 2013
6,613
2,850
New York
Recommendations for developers to increase battery life:
1. Remove useless drivers such as:
Code:
  SCSI subsystem initialized
Code:
 SCSI Media Changer driver v0.25
Code:
 sdhci: Secure Digital Host Controller Interface driver
and many others
2. Disable unused processors, or offer a "power saving" mode to disable those additional processors
3. examine processor scaling
4. reduce use of SELinux. it's worthless.
4a. when device is near phone, phone security is off
4b. turning on ADB and accessing storage is easy as security is disabled on the phone
4c. because of the above reasons the device causes the phone to be insecure
4d. the device is wasting battery on its own security while reducing the security of the phone.

5. enable thermal max frequency
Code:
[    4.700499] msm_thermal:disable_msm_thermal Max frequency reset for CPU1
[    4.700509] msm_thermal:disable_msm_thermal Max frequency reset for CPU2
[    4.700517] msm_thermal:disable_msm_thermal Max frequency reset for CPU3

6. above all: audit your logs.

Can I do that thru ADB? I am no developer at all btw!
 

E:V:A

Inactive Recognized Developer
Dec 6, 2011
1,447
2,222
-∇ϕ
Doesn't removing "sdhci" disable OTG functionality and possibly other USB devices, enumerated during DLOAD mode etc?
What is a USB condom?
And what is that nice USB-to-micro-USB thing you've got there? I see some useful pins... on it.
And yes, it is disturbing to see that a wrist watch has twice the number of processor cores than my daily phone!

EDIT: I just found answers here:
http://www.xda-developers.com/android/smartphone-chargers-xdatv/
http://www.amazon.com/Breett-Charger-Doctor-Current-Voltage/dp/B00JO73278/?tag=viglink20238-20
http://int3.cc/products/usbcondoms

Great! Will get at least one of each!
 
Last edited:
  • Like
Reactions: AdamOutler

Childofthehorn

Inactive Recognized Developer
Jun 6, 2010
260
185
DFW, Texaaas
This device charges VERY slowly, and on top of that, it's also picky about the charging port. In order to be sure you're always charging, you'll want to use a USB condom. This is important because many aftermarket chargers are unlabeled.

The good news is that the charging rate is generally controlled by software, so if samsung would release kernel code, we could likely fix this.

Adam, the issue with the charger is one which I filed early on the bug tracker and got to samsung.

Its a problem with the MAX77836- MUIC charger recognition code. Its a problem that I have encountered before with Samsung's custom kernel drivers for the MAX chipsets. You can get more in-depth charging info by just plugging a charger and running

"dmesg | grep max77 " , but we don't have root yet ( i haven't really spent any time on it).

I bet that if you change the ADC value on the ID pin of the MicroUSB connector you will get better results.

The MUIC general code is fantastic and then the individual function code classes are absolute junk.

Here is the values for the ADC and functions for most every Samsung Device:

ADC_DOCK_PREV_KEY = 0x04,
ADC_DOCK_NEXT_KEY = 0x07,
ADC_INCOMPATIBLE1_CHG = 0x08, /* 0x01000 10.0K ohm*/
ADC_DOCK_VOL_DN = 0x0a, /* 0x01010 14.46K ohm */
ADC_DOCK_VOL_UP = 0x0b, /* 0x01011 17.26K ohm */
ADC_DOCK_PLAY_PAUSE_KEY = 0x0d,
ADC_VZW_USB_DOCK = 0x0e, /* 0x01110 28.7K ohm VZW Dock */
ADC_SMARTDOCK = 0x10, /* 0x10000 40.2K ohm */
ADC_HMT = 0x11, /* 0x10001 49.9K ohm */
ADC_AUDIODOCK = 0x12, /* 0x10010 64.9K ohm */
ADC_CEA936ATYPE1_CHG = 0x17, /* 0x10111 200K ohm */
ADC_JIG_USB_OFF = 0x18, /* 0x11000 255K ohm */
ADC_JIG_USB_ON = 0x19, /* 0x11001 301K ohm */
ADC_DESKDOCK = 0x1a, /* 0x11010 365K ohm */
ADC_CEA936ATYPE2_CHG = 0x1b, /* 0x11011 442K ohm */
ADC_JIG_UART_OFF = 0x1c, /* 0x11100 523K ohm */
ADC_JIG_UART_ON = 0x1d, /* 0x11101 619K ohm */
ADC_CARDOCK = 0x1d, /* 0x11101 619K ohm */

As you can see, there is the UART deeper enable as well.
 

AdamOutler

Retired Senior Recognized Developer
Feb 18, 2011
5,224
9,827
Miami, Fl̨̞̲̟̦̀̈̃͛҃҅͟orida
Adam, the issue with the charger is one which I filed early on the bug tracker and got to samsung.

Its a problem with the MAX77836- MUIC charger recognition code. Its a problem that I have encountered before with Samsung's custom kernel drivers for the MAX chipsets. You can get more in-depth charging info by just plugging a charger and running

"dmesg | grep max77 " , but we don't have root yet ( i haven't really spent any time on it).

I bet that if you change the ADC value on the ID pin of the MicroUSB connector you will get better results.

The MUIC general code is fantastic and then the individual function code classes are absolute junk.

Here is the values for the ADC and functions for most every Samsung Device:

ADC_DOCK_PREV_KEY = 0x04,
ADC_DOCK_NEXT_KEY = 0x07,
ADC_INCOMPATIBLE1_CHG = 0x08, /* 0x01000 10.0K ohm*/
ADC_DOCK_VOL_DN = 0x0a, /* 0x01010 14.46K ohm */
ADC_DOCK_VOL_UP = 0x0b, /* 0x01011 17.26K ohm */
ADC_DOCK_PLAY_PAUSE_KEY = 0x0d,
ADC_VZW_USB_DOCK = 0x0e, /* 0x01110 28.7K ohm VZW Dock */
ADC_SMARTDOCK = 0x10, /* 0x10000 40.2K ohm */
ADC_HMT = 0x11, /* 0x10001 49.9K ohm */
ADC_AUDIODOCK = 0x12, /* 0x10010 64.9K ohm */
ADC_CEA936ATYPE1_CHG = 0x17, /* 0x10111 200K ohm */
ADC_JIG_USB_OFF = 0x18, /* 0x11000 255K ohm */
ADC_JIG_USB_ON = 0x19, /* 0x11001 301K ohm */
ADC_DESKDOCK = 0x1a, /* 0x11010 365K ohm */
ADC_CEA936ATYPE2_CHG = 0x1b, /* 0x11011 442K ohm */
ADC_JIG_UART_OFF = 0x1c, /* 0x11100 523K ohm */
ADC_JIG_UART_ON = 0x1d, /* 0x11101 619K ohm */
ADC_CARDOCK = 0x1d, /* 0x11101 619K ohm */

As you can see, there is the UART deeper enable as well.
Those resistor values come from the Fairchild Semiconductors chip in the Samsung devices. I didn't find that chip in this device.

I posted the uart output which contains the dmesg output. I'm at work so i can't check the max77 in there.
 

AdamOutler

Retired Senior Recognized Developer
Feb 18, 2011
5,224
9,827
Miami, Fl̨̞̲̟̦̀̈̃͛҃҅͟orida
[Thread]
Scratch that end of thread tag. I was overjoyed. There's still a lot of Linuxy stuff that needs to be done. Things like LED Brightness on the kernel and other standard hardware references which haven't been properly implemented. Standard Linux Device Tree stuff for command-line access is still needed.
 
  • Like
Reactions: Krunk_Kracker

AdamOutler

Retired Senior Recognized Developer
Feb 18, 2011
5,224
9,827
Miami, Fl̨̞̲̟̦̀̈̃͛҃҅͟orida
The device is configured BT software to charge at .14A. In a stock kernel without the standard ramdisk I was able to pull a max of .45A. Which theoretically could charge the device in about 45minutes. Bumping up the charging rate should be examined as this device's 300mAh battery is far smaller than that of larger phones with the same battery materials, with 2000mAh batteries and charges in the same period of time.
 
  • Like
Reactions: Krunk_Kracker

Top Liked Posts

  • There are no posts matching your filters.
  • 9
    FSTAB:
    http://pastebin.com/LN75y6Jj

    Build.Prop:
    http://pastebin.com/ehgwGL8A

    System Dump:
    http://www.mediafire.com/download/ux9gg1661ojkao9/sprat-user_4.4W_KMV78N_SYSTEM_DUMP.zip
    *Copyright owned by Samsung and Google

    Getprop
    http://pastebin.com/2mhPZhLV

    Hardware Info
    https://www.youtube.com/watch?v=NNtBzOiXLPw

    IMG_20140701_142732.jpg

    ENABLE ADB
    - Tap on top right of watchface
    - scroll down to Settings
    - Scroll down and tap About
    - Tap Build Number 5x's
    - Go back to settings and Development Options should be available
    - Enable ADB debugging
    - Plug in the charger dock with a USB connected to the computer
    - Authorize on your Phone for USB


    14+-+1

    RECOVERY MODE


    IMG_20140701_151120.jpg

    FASTBOOT/BOOTLOADER Mode
    2
    Recommendations for developers to increase battery life:
    1. Remove useless drivers such as:
    Code:
      SCSI subsystem initialized
    Code:
     SCSI Media Changer driver v0.25
    Code:
     sdhci: Secure Digital Host Controller Interface driver
    and many others
    2. Disable unused processors, or offer a "power saving" mode to disable those additional processors
    3. examine processor scaling
    4. reduce use of SELinux. it's worthless.
    4a. when device is near phone, phone security is off
    4b. turning on ADB and accessing storage is easy as security is disabled on the phone
    4c. because of the above reasons the device causes the phone to be insecure
    4d. the device is wasting battery on its own security while reducing the security of the phone.

    5. enable thermal max frequency
    Code:
    [    4.700499] msm_thermal:disable_msm_thermal Max frequency reset for CPU1
    [    4.700509] msm_thermal:disable_msm_thermal Max frequency reset for CPU2
    [    4.700517] msm_thermal:disable_msm_thermal Max frequency reset for CPU3

    6. above all: audit your logs.
    2
    We need a boot.img or at least a buildable kernel source to work with in order to go any further.
    2
    Can I do that thru ADB? I am no developer at all btw!

    You probly don't want to be in this thread then.
    2
    This device charges VERY slowly, and on top of that, it's also picky about the charging port. In order to be sure you're always charging, you'll want to use a USB condom. This is important because many aftermarket chargers are unlabeled.

    P_20140720_203409_HDR.jpg


    Here you can see it won't charge at all on an "Apple" charger
    P_20140720_203436_HDR.jpg


    And it charges well on an "Android" charger with D+ and D- tied together.
    P_20140720_203721_HDR.jpg




    The good news is that the charging rate is generally controlled by software, so if samsung would release kernel code, we could likely fix this.