Jump to content

Armbian for Amlogic S912

chocho

By chocho
in General Chat

 Share

Recommended Posts

Armbian & Khadas are rewarding contributors

balbes150

balbes150

Posted
Posted
16 hours ago, kingul said:

Can you please modify my original dtb.img to work with one of your firmware made with kernel 4.9.40?
It's working with 3.14.29, gcc 4.9, but not with gcc > 5 or kernel 4.9.40

This requires access to source code and information from the manufacturer. Maybe you need to contact the manufacturer of your TV box.

Link to comment
Share on other sites
samarcanda

samarcanda

Posted
Posted

Please how can i reboot android from armbian without remove usb key?

I install successfull Armbian_5.37_S9xxx_Debian_stretch_3.14.29_server_20171226 image on usb key and boot armbian from android with armlogic tool app, but i don't know how to do the opposite.

I can do it only removing physically the usb key.

Link to comment
Share on other sites
balbes150

balbes150

Posted
Posted
30 minutes ago, samarcanda said:

Please how can i reboot android from armbian without remove usb key?

I install successfull Armbian_5.37_S9xxx_Debian_stretch_3.14.29_server_20171226 image on usb key and boot armbian from android with armlogic tool app, but i don't know how to do the opposite.

I can do it only removing physically the usb key.

Only disconnecting the USB media. Either learn how to do selective start of the system.

 

 

Link to comment
Share on other sites
chiefthunder

chiefthunder

Posted
Posted

Can anyone help me on getting WiFi to work on my H96 pro mini? I managed to get Armbian_5.37_S9xxx_Ubuntu_xenial_3.14.29_server_20171226 running off an SD card. Did not need any DBT file. Only thing I need working is wifi and just cannot work out how to fix it. Any ideas?

Link to comment
Share on other sites
chiefthunder

chiefthunder

Posted
Posted
8 hours ago, chiefthunder said:

Can anyone help me on getting WiFi to work on my H96 pro mini? I managed to get Armbian_5.37_S9xxx_Ubuntu_xenial_3.14.29_server_20171226 running off an SD card. Did not need any DBT file. Only thing I need working is wifi and just cannot work out how to fix it. Any ideas?

I've done it myself. Not to worry, used the wifi_dummy

Link to comment
Share on other sites
etatto

etatto

Posted
Posted

Hi @balbes150

 

Could you please tell me if the latest (on the 27th of January as of this writing) on your github is working on a Khadas VIM2 Max?

I have compiled for Debian Stretch and the boot doesn't pass this stage, with a stack trace: 

[    6.477332@3] hdmitx: update physcial size: 16 9
[    6.481921@3] config: hdmitx: unknown cmd: 0x14000000
[   27.491744@3] INFO: rcu_preempt detected stalls on CPUs/tasks:
[   27.495829@1]        3-...: (0 ticks this GP) idle=0f7/140000000000000/0 softirq=392/393 fqs=2626
[   27.502733@1]        (detected by 1, t=5255 jiffies, g=-98, c=-99, q=48)
[   27.508858@1] Task dump for CPU 3:
[   27.513010@1] hdmi_init.sh    R  running task        0  2370      1 0x00000002
[   27.520873@1] Call trace:
[   27.525077@1] [<ffffff8009086434>] __switch_to+0x94/0xa8
[   27.529331@1] [<000000000000000a>] 0xa
[   62.475758@1] fb: mem_free_work, free memory: addr:6af000
[   90.511743@1] INFO: rcu_preempt detected stalls on CPUs/tasks:
[   90.516096@1]        3-...: (0 ticks this GP) idle=0f7/140000000000000/0 softirq=392/393 fqs=10502
[   90.523339@1]        (detected by 1, t=21010 jiffies, g=-98, c=-99, q=70)
[   90.529551@1] Task dump for CPU 3:
[   90.533926@1] hdmi_init.sh    R  running task        0  2370      1 0x00000002
[   90.541768@1] Call trace:
[   90.546145@1] [<ffffff8009086434>] __switch_to+0x94/0xa8
[   90.550546@1] [<000000000000000a>] 0xa
[  153.531742@1] INFO: rcu_preempt detected stalls on CPUs/tasks:
[  153.536221@1]        3-...: (0 ticks this GP) idle=0f7/140000000000000/0 softirq=392/393 fqs=18377
[  153.543592@1]        (detected by 1, t=36765 jiffies, g=-98, c=-99, q=70)
[  153.549803@1] Task dump for CPU 3:
[  153.554288@1] hdmi_init.sh    R  running task        0  2370      1 0x00000002
[  153.562099@1] Call trace:
[  153.566551@1] [<ffffff8009086434>] __switch_to+0x94/0xa8
[  153.571047@1] [<000000000000000a>] 0xa
[  160.067789@0] random: crng init done
[  216.551741@0] INFO: rcu_preempt detected stalls on CPUs/tasks:
[  216.556227@1]        3-...: (0 ticks this GP) idle=0f7/140000000000000/0 softirq=392/393 fqs=26252
[  216.563637@1]        (detected by 1, t=52520 jiffies, g=-98, c=-99, q=70)
[  216.569849@1] Task dump for CPU 3:
[  216.574384@1] hdmi_init.sh    R  running task        0  2370      1 0x00000002
[  216.582187@1] Call trace:
[  216.586653@1] [<ffffff8009086434>] __switch_to+0x94/0xa8
[  216.591156@1] [<000000000000000a>] 0xa
[  242.695808@0] INFO: task kworker/1:1:1237 blocked for more than 120 seconds.
[  242.701706@0]       Not tainted 4.9.40 #1
[  242.706184@0] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[  242.713600@0] kworker/1:1     D    0  1237      2 0x00000000
[  242.721111@0] Workqueue: events aml_tvout_mode_work
[  242.728395@0] Call trace:
[  242.732988@0] [<ffffff8009086434>] __switch_to+0x94/0xa8
[  242.737651@0] [<ffffff8009be2864>] __schedule+0x21c/0x748
[  242.742638@0] [<ffffff8009be2dd0>] schedule+0x40/0xa8
[  242.747644@0] [<ffffff8009be32c4>] schedule_preempt_disabled+0x1c/0x30
[  242.754111@0] [<ffffff8009be4b94>] __mutex_lock_slowpath+0xac/0x188
[  242.760663@0] [<ffffff8009be4cd8>] mutex_lock+0x68/0x80
[  242.765499@0] [<ffffff80098139e0>] validate_vmode+0x30/0x90
[  242.771017@0] [<ffffff8009814688>] refresh_tvout_mode+0x58/0xf0
[  242.776880@0] [<ffffff80098149c0>] aml_tvout_mode_work+0x30/0xa8
[  242.782836@0] [<ffffff80090bbf58>] process_one_work+0x1e0/0x498
[  242.788700@0] [<ffffff80090bc260>] worker_thread+0x50/0x4a8
[  242.794219@0] [<ffffff80090c2cd8>] kthread+0xf8/0x110
[  242.799221@0] [<ffffff8009083680>] ret_from_fork+0x10/0x50
[  279.571742@0] INFO: rcu_preempt detected stalls on CPUs/tasks:
[  279.576483@1]        3-...: (0 ticks this GP) idle=0f7/140000000000000/0 softirq=392/393 fqs=34127
[  279.584149@1]        (detected by 1, t=68275 jiffies, g=-98, c=-99, q=70)
[  279.590466@1] Task dump for CPU 3:
[  279.595164@1] hdmi_init.sh    R  running task        0  2370      1 0x00000002
[  279.602956@1] Call trace:
[  279.607721@1] [<ffffff8009086434>] __switch_to+0x94/0xa8
[  279.612556@1] [<000000000000000a>] 0xa

 

Do you have an idea? Tks.

Link to comment
Share on other sites
etatto

etatto

Posted
Posted
1 hour ago, balbes150 said:

Wait for the nearest update GIT. It will be added a new version with the kernel 4.9.40 for the VIM2. I will write the steps that you will need to perform to build such an option.

Thanks, I wait :thumbup:

Link to comment
Share on other sites
icusu

icusu

Posted
Posted

Hello all!  I have successfully gotten @balbes150's debian server image installed onto a few H96's that I am setting up for compiling.  I am hoping to reclaim some of the memory that systemd is using.  Specifically, I intend to completely disable hdmi output as well as the reserved memory for the mali gpu.  Does anyone have a guide I could read for uboot or something else that would allow me to accomplish this?  I'm not looking for someone to just fix the problem for me, but a pointer in the right direction would be quite nice. :)  Either way, thanks go out to balbes150 and others in these tv box threads which have gotten me to this point.

Link to comment
Share on other sites
balbes150

balbes150

Posted
Posted
1 hour ago, icusu said:

Hello all!  I have successfully gotten @balbes150's debian server image installed onto a few H96's that I am setting up for compiling.  I am hoping to reclaim some of the memory that systemd is using.  Specifically, I intend to completely disable hdmi output as well as the reserved memory for the mali gpu.  Does anyone have a guide I could read for uboot or something else that would allow me to accomplish this?  I'm not looking for someone to just fix the problem for me, but a pointer in the right direction would be quite nice. :)  Either way, thanks go out to balbes150 and others in these tv box threads which have gotten me to this point.

Here was discussed a similar question (to increase the memory by turning off the redundant memory).

 

http://forum.khadas.com/

 

Link to comment
Share on other sites
etatto

etatto

Posted
Posted 
[    6.531439@3] hdmitx: update physcial size: 16 9
[    6.536003@3] config: hdmitx: unknown cmd: 0x14000000
[   27.556171@3] INFO: rcu_preempt detected stalls on CPUs/tasks:
[   27.577311@6]        3-...: (1 GPs behind) idle=4f9/140000000000000/0 softirq=341/342 fqs=2593
[   27.585638@6]        (detected by 6, t=5259 jiffies, g=-78, c=-79, q=51)
[   27.593029@6] Task dump for CPU 3:
[   27.598885@6] hdmi_init.sh    R  running task        0  2371      1 0x00000002
[   27.606987@6] Call trace:
[   27.612905@6] [<ffffff8009086434>] __switch_to+0x94/0xa8
[   27.618911@6] [<0000000000000001>] 0x1
[   62.476130@1] fb: mem_free_work, free memory: addr:6af000
[   90.576173@1] INFO: rcu_preempt detected stalls on CPUs/tasks:
[   90.592272@5]        3-...: (1 GPs behind) idle=4f9/140000000000000/0 softirq=341/342 fqs=10404
[   90.601310@5]        (detected by 5, t=21014 jiffies, g=-78, c=-79, q=66)
[   90.609069@5] Task dump for CPU 3:
[   90.615263@5] hdmi_init.sh    R  running task        0  2371      1 0x00000002
[   90.623679@5] Call trace:
[   90.629876@5] [<ffffff8009086434>] __switch_to+0x94/0xa8
[   90.636119@5] [<0000000000000001>] 0x1
[  153.600178@5] INFO: rcu_preempt detected stalls on CPUs/tasks:
[  153.616626@7]        3-...: (1 GPs behind) idle=4f9/140000000000000/0 softirq=341/342 fqs=18194
[  153.625694@7]        (detected by 7, t=36768 jiffies, g=-78, c=-79, q=66)
[  153.633622@7] Task dump for CPU 3:
[  153.639999@7] hdmi_init.sh    R  running task        0  2371      1 0x00000002
[  153.648565@7] Call trace:
[  153.654867@7] [<ffffff8009086434>] __switch_to+0x94/0xa8
[  153.661227@7] [<0000000000000001>] 0x1
[  167.944416@0] random: crng init done
[  216.624164@0] INFO: rcu_preempt detected stalls on CPUs/tasks:
[  216.645877@6]        3-...: (1 GPs behind) idle=4f9/140000000000000/0 softirq=341/342 fqs=25995
[  216.654917@6]        (detected by 6, t=52528 jiffies, g=-78, c=-79, q=66)
[  216.662897@6] Task dump for CPU 3:
[  216.669308@6] hdmi_init.sh    R  running task        0  2371      1 0x00000002
[  216.677892@6] Call trace:
[  216.684205@6] [<ffffff8009086434>] __switch_to+0x94/0xa8
[  216.690568@6] [<0000000000000001>] 0x1
[  242.696550@4] INFO: task kworker/1:1:1238 blocked for more than 120 seconds.
[  242.715772@4]       Not tainted 4.9.40 #1
[  242.722104@4] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[  242.730956@4] kworker/1:1     D    0  1238      2 0x00000000
[  242.740331@4] Workqueue: events aml_tvout_mode_work
[  242.748435@4] Call trace:
[  242.754944@4] [<ffffff8009086434>] __switch_to+0x94/0xa8
[  242.761530@4] [<ffffff8009bf2cc4>] __schedule+0x21c/0x748
[  242.768131@4] [<ffffff8009bf3230>] schedule+0x40/0xa8
[  242.774691@4] [<ffffff8009bf3724>] schedule_preempt_disabled+0x1c/0x30
[  242.782158@4] [<ffffff8009bf4ff4>] __mutex_lock_slowpath+0xac/0x188
[  242.789341@4] [<ffffff8009bf5138>] mutex_lock+0x68/0x80
[  242.795992@4] [<ffffff8009821e40>] validate_vmode+0x30/0x90
[  242.802614@4] [<ffffff8009822ae8>] refresh_tvout_mode+0x58/0xf0
[  242.809496@4] [<ffffff8009822e20>] aml_tvout_mode_work+0x30/0xa8
[  242.816486@4] [<ffffff80090bbf58>] process_one_work+0x1e0/0x498
[  242.823435@4] [<ffffff80090bc260>] worker_thread+0x50/0x4a8
[  242.830124@4] [<ffffff80090c2cd8>] kthread+0xf8/0x110
[  242.836821@4] [<ffffff8009083680>] ret_from_fork+0x10/0x50
[  279.648172@4] INFO: rcu_preempt detected stalls on CPUs/tasks:
[  279.670232@6]        3-...: (1 GPs behind) idle=4f9/140000000000000/0 softirq=341/342 fqs=33795
[  279.679620@6]        (detected by 6, t=68284 jiffies, g=-78, c=-79, q=66)
[  279.687897@6] Task dump for CPU 3:
[  279.694527@6] hdmi_init.sh    R  running task        0  2371      1 0x00000002
[  279.703470@6] Call trace:
[  279.710197@6] [<ffffff8009086434>] __switch_to+0x94/0xa8
[  279.717022@6] [<0000000000000001>] 0x1

Is it perhaps because I'm running headless with a screen?

Link to comment
Share on other sites
balbes150

balbes150

Posted
Posted
4 hours ago, etatto said:

Is it perhaps because I'm running headless with a screen?

Gathered from the latest commit image stretch 4.9.40. Recorded on the USB flash drive, added the dtb (kvim2_multios.dtb). Has launched on the VIM2 Max. Connected only to a wired network, a USB flash drive with the system and USB power. Joined VIM2 via SSH from the PC and performed the initial configuration. After a reboot everything works. Try to run this image. Maybe when you build something made not so.

 

https://yadi.sk/d/5zSm31vQ3S3ijn

Link to comment
Share on other sites
etatto

etatto

Posted
Posted
On 2/2/2018 at 9:49 PM, balbes150 said:

Gathered from the latest commit image stretch 4.9.40. Recorded on the USB flash drive, added the dtb (kvim2_multios.dtb). Has launched on the VIM2 Max. Connected only to a wired network, a USB flash drive with the system and USB power. Joined VIM2 via SSH from the PC and performed the initial configuration. After a reboot everything works. Try to run this image. Maybe when you build something made not so.

 

https://yadi.sk/d/5zSm31vQ3S3ijn

 

Well I'm puzzled, but maybe I'm doing it the wrong way coz' your image @balbes150 also doesn't boot with the same hang/error...

 

I'm doing the same as I have done when installing Armbian on my Khadas VIM Pros:

- Android on the emmc,

- Flashing aml_autoscript.zip with System update application under android,

- Burning the image on sdcard with Etcher,

- Sdcard in sd-slot,

- Power on

=> Boot Armbian from the sdcard

 

So have these steps changed for the VIM2 (or perhaps for all S905x/S912) ???

Thanks.

Link to comment
Share on other sites
balbes150

balbes150

Posted
Posted

I have this image running on VIM1 (s905x) and VIM2 (s912). What u-boot is installed ? To run the kernel 4.9.40 on VIM1 (s905x) need a new u-boot. I use my version of u-boot with built-in multi-boot.

To run from external media, you can use any option of the dtb. The correct option dtb important to install the system in eMMC. Then you need to use strictly correct dtb.

Link to comment
Share on other sites
etatto

etatto

Posted
Posted

Well I've tried your image and mine on one of my VIM Pro: It boots without any issue (with the right dtb of course ^_^). I might say the images are not the cause of the hang on my VIM2 Max

Perhaps a stupid question and not related to my issue: Which revision of the VIM2 Max do you have? Mine is a V1.2 201707. I have spotted on Google V1.0 and V1.1 for the VIM2 and I don't know what might be the differences...

Link to comment
Share on other sites
balbes150

balbes150

Posted
Posted

1. Try updating the multi-boot using only the image Armbian. To capture the image, copy and rename the dtb file, to run Android, to run multi-boot activation.

2. What firmware version is installed in the eMMC of your VIM2 ?

Link to comment
Share on other sites
guidol

guidol

Posted
Posted
On 21/12/2016 at 3:30 PM, lvmc said:
WiFi is now working on GT1.
 
1. Download & Copy AP6255 WiFi drivers extracted from Beelink's Android image:

 

Thanks for the instrutions - they did help me very much with the AP6330 onboard WiFi of the Sunvell T95K Pro
(S912 device = q201_6255):

Manually:
  

modprobe cfg80211
modprobe dhd firmware_path=/lib/firmware/brcm/fw_bcm40183b2.bin nvram_path=/lib/firmware/brcm/nvram_ap6330.txt

On boot:
  

# /etc/modules
cfg80211
dhd

# /etc/modprobe.d/options.conf
options dhd firmware_path=/lib/firmware/brcm/fw_bcm40183b2.bin nvram_path=/lib/firmware/brcm/nvram_ap6330.txt

lsmod: 

root@sunvell-t95kpro:~# lsmod
Module                  Size  Used by
brcmfmac              171742  0
brcmutil                5091  1 brcmfmac
joydev                 10336  0
dhd                   817295  0
cfg80211              415017  2 dhd,brcmfmac
aml_nftl_dev           95108  0

 

Now the WiFi does run very stable and without any big latency :)


Before I had used an external USB-WiFi Dongle from Edimax where I had a high latency :(
(and I did disable power-management in the /etc/network/interfaces) 

 

1
Link to comment
Share on other sites
alanoliver

alanoliver

Posted
Posted
On 13/02/2018 at 10:44 AM, guidol said:

 

Thanks for the instrutions - they did help me very much with the AP6330 onboard WiFi of the Sunvell T95K Pro
(S912 device = q201_6255):

Manually:
  


modprobe cfg80211
modprobe dhd firmware_path=/lib/firmware/brcm/fw_bcm40183b2.bin nvram_path=/lib/firmware/brcm/nvram_ap6330.txt

On boot:
  


# /etc/modules
cfg80211
dhd

# /etc/modprobe.d/options.conf
options dhd firmware_path=/lib/firmware/brcm/fw_bcm40183b2.bin nvram_path=/lib/firmware/brcm/nvram_ap6330.txt

lsmod: 


root@sunvell-t95kpro:~# lsmod
Module                  Size  Used by
brcmfmac              171742  0
brcmutil                5091  1 brcmfmac
joydev                 10336  0
dhd                   817295  0
cfg80211              415017  2 dhd,brcmfmac
aml_nftl_dev           95108  0

 

Now the WiFi does run very stable and without any big latency :)


Before I had used an external USB-WiFi Dongle from Edimax where I had a high latency :(
(and I did disable power-management in the /etc/network/interfaces) 

 

 

 

Link off file wifi!

Link to comment
Share on other sites
alanoliver

alanoliver

Posted
Posted
On 12/02/2017 at 1:08 PM, Quintus23M said:

Has anybody tried to build and run a mainline Linux kernel on a S912 device?

I do have a Beelink GT1 and would like to use a 4.9.x kernel, no graphics required for my headless server, just ethernet and maybe wifi for later. Currently I was able to boot the GT1 with an Armbian image from @balbes150, but this image has only a 3.14.29 kernel.

 

Any advice and pointers are really appreciated.

 

My friend, did you run the 4.9.x kernel on Beelink GT1? I'm trying to run more than not boot.

Link to comment
Share on other sites
Guest
This topic is now closed to further replies.
 Share
Go to topic listing
×
×
  • Create New...

Important Information

Terms of Use - Privacy Policy - Guidelines

  I accept