Levent Erenler

February 3, 2021

hi @FeiJi i dont have a bpi m64, so i dont have a chance to test it but i will try on pine64 with 5.10 kernel. A few months ago, I have a similar problem because of broken camera. Everything were looking normal but i was getting error from driver when i checked dmesg. I change cam and it works. Do you have chance to try different camera.

February 3, 2021

On 11/21/2020 at 4:41 PM, @lex said:

Can you post the format info for your USB camera?

v4l2-ctl -d 4 --list-formats

or

v4l2-ctl -d 5 --list-formats

I have seen some USB camera has a YUY2 format.

Cpu usage ~52% looks good for OpenCV.

Sorry @@lex for late reply, i did not see any notification until last message. Thank you

usb camera formats are;

ioctl: VIDIOC_ENUM_FMT
Type: Video Capture

[0]: 'MJPG' (Motion-JPEG, compressed)
[1]: 'YUYV' (YUYV 4:2:2)

November 16, 2020

On 11/16/2020 at 4:50 PM, @lex said:

I would build opencv with debugging info and try to find the bottleneck. And why there is an image conversion in the DVP case.

Push the limits to 720P/30fps or even 1080P and see what you get.

Thanks for your support, I prefer 640*480 cause opencv face or tag recognition consume lots of cpu power, 640*480@30 is acceptable. I try a few more resolution and its getting worst

[fmt:JPEG_1X8/1280x720] %100 cpu usage limited 15 FPS, could not reach more fps, cpu usage at single core that means not multithreaded
[fmt:UYVY8_2X8/1280x720] // %190 cpu usage @30 fps ,multi threaded
[fmt:UYVY8_2X8/1980x1080] // could not handle 30fps ,%200 cpu usage at 15FPS

UPDATE: my best scenario is fmt:UYVY8_2X8 and opencv CV_CAP_PROP_FOURCC property is CV_FOURCC('Y', 'U', '1', '2') Cpu usage ~%52

November 16, 2020

5 minutes ago, @lex said:

There must be no conversion in the image format to achieve this. It is interesting to find out more about it. Can you share more about your application and your setup and usb camera?

yes, u are right that means there is no conversion. The app is very simple; open camera with opencv videocapturer and grap image with opencv(v3.4.10) and calculate fps (no display). Setup Pine64+ 1GB A64 Kernel 5.9.3 with HBV1318 ov5640 usb camera

cpu usage is %6-8 at each core and total cpu usage %25 -31 ( 640*480*30FPS)

device listed as

doccamera: doccamera (usb-1c1a000.usb-1):
/dev/video4
/dev/video5

November 16, 2020

Thank you @@lex i got my new OV5640 camera and working with 5.9.x Cpu usage around %70 with opencv V4l2 capture fmt:UYVY8_2X8/640x480@30FPS and %60 with JPEG_1X8/640x480@30FPS.

It is higher than i expect. i thought csi camera uses GPU to grab image. Usb ov5640 cpu usage around %25 with similar settings. is it normal? Do you have any idea. Thank you.

November 11, 2020

Update: Analog sound + mic working

I attached overlay for microphone and headphone. Tested with Pine64+ A64 Kernel 5.9.5 and working. You have to add overlay and set alsa configuration. Settings are just for template, you could modify to find best. Thank you.

amixer cset name='Headphone Playback Switch' on
amixer cset name='Headphone Source Playback Route' DAC
amixer cset name='Line In Playback Switch' off
amixer cset name='Line Out Playback Switch' off
amixer cset name='Line Out Source Playback Route' Mono Differential
amixer cset name='Mic1 Playback Switch' off
amixer cset name='Mic2 Playback Switch' off
amixer cset name='AIF1 DA0 Playback Volume' 160
amixer cset name='AIF1 DA0 Stereo Playback Route' Stereo
amixer cset name='AIF2 DAC Playback Volume' 160
amixer cset name='AIF3 ADC Source Capture Route' None
amixer cset name='AIF2 DAC Source Playback Route' None
amixer cset name='DAC Playback Switch' on
amixer cset name='DAC Playback Volume' 160
amixer cset name='ADC Digital DAC Playback Switch' off
amixer cset name='AIF1 Slot 0 Digital DAC Playback Switch' on
amixer cset name='AIF2 Digital DAC Playback Switch' on
amixer cset name='DAC Reversed Playback Switch' off
amixer cset name='Earpiece Playback Switch' off
amixer cset name='Earpiece Source Playback Route' DACL

amixer cset name='Line In Capture Switch' off
amixer cset name='Mic1 Capture Switch' off
amixer cset name='Mic1 Boost Volume' 0
amixer cset name='Mic1 Playback Volume' 7
amixer cset name='Mic2 Capture Switch' on
amixer cset name='Mic2 Boost Volume' 0
amixer cset name='Mic2 Playback Volume' 7
amixer cset name='Mixer Capture Switch' off
amixer cset name='Mixer Reversed Capture Switch' off
amixer cset name='ADC Capture Volume' 160
amixer cset name='ADC Gain Capture Volume' 7
amixer cset name='AIF1 AD0 Capture Volume' 160
amixer cset name='AIF1 Data Digital ADC Capture Switch' on
amixer cset name='AIF2 ADC Capture Volume' 160
amixer cset name='AIF2 ADC Mixer ADC Capture Switch' on
amixer cset name='AIF2 ADC Mixer AIF1 DA0 Capture Switch' off
amixer cset name='AIF2 ADC Mixer AIF2 DAC Rev Capture Switch' off
amixer cset name='AIF2 ADC Mixer AIF1 DA0 Capture Switch' off
amixer cset name='AIF2 ADC Mixer AIF1 DA0 Capture Switch' off

you could test speaker with:

speaker-test -twav -c2

you could test microphone with;

arecord -f S16_LE -d 5 -r 16000 -c 2 test.wav

aplay test.wav

sound_overlay.dts

November 9, 2020

Platform : Kernel 5.9.5 Pine64+ A64 buster server

Analog sound working ! After a few days, i figure out the analog sound problem. There is a widget property in sound definition at sun50i-a64-pine64.dts file. This mainly specifies audio off-codec DAPM widgets (Dynamic Audio Power Management for Portable Devices according to devicetree/bindings/sound/widgets.txt) In my case this widgets disable analog sound and i couldnt find a way to enable it I remove widget line and linked routing properties from dts file and rebuild it and it works. If you dont want to rebuild armbian alternate way is using a custom overlay. Dont forget to unmute HeadPhone with "m" key and increase AIF1 DA0. I hope it will help to someone. Unfortunately i could not enable microphone input, any ideas?? Thanks

sound_overlay.dts

November 7, 2020

yes exactly, now it works with custom dts. I write a test dts depending your dts file and it worked. Now i am trying to find why it is now working with original dts than i will create a patch to share with others. Thanks for your effort.

November 7, 2020

yes u are right, i forget unmute while taking photo, unfortunatelly it did not fix, i try every possibility in alsamixer Thank you for pointing that.

November 7, 2020

Thanks @guidol I am testing with pine64+ A64 kernel 5.9.5 i have same warnings in original dtb. I try many things but i could not figure whats wrong! I try to compare my dts file with yours.Do you have source, i mean dts files before compiled(easy to read) or is that https://github.com/megous/linux/tree/orange-pi-5.9 Thank you

i try to gave more information

UPDATE: I found a solution

November 7, 2020

Kernel = 5.9.5 Pine64+ A64

HDMI sound is working, unfortunately Analog sound out not .I try to check dts and try to change mixer settings but no luck. I could not find what am i missing? I need help to fix analog sound issue Thank you.

I attached dtb info and alsa info

current_dtb.txt alsa-info.txt

November 6, 2020

On 11/1/2020 at 5:08 PM, Hannes Worst said:

Great, Guidol!

Thank you very much for this good news. I wil try it out! Thanks for the one that solved the problem.

Hi, do you have a chance to test it? I could not get sound from audio jack. Could you share your dts if the analog audio working? Thank you

November 5, 2020

Armbianmonitor:

http://ix.io/2D9S

Hi, I try to enable sound on headphone jack with pine64+ but i could not Alsa says "no sound found" at dmesg Could anyone help me? Thank you

root@pine64:~# dmesg|grep -i sound
[ 2.686397] No soundcards found.
[ 6.323086] input: sun50i-a64-audio Headset Jack as /devices/platform/sound/sound/card0/input5

root@pine64:~# aplay -l
**** List of PLAYBACK Hardware Devices ****
card 0: sun50ia64audio [sun50i-a64-audio], device 0: 1c22c00.dai-sun8i-codec-aif1 sun8i-codec-aif1-0 [1c22c00.dai-sun8i-codec-aif1 sun8i-codec-aif1-0]
Subdevices: 1/1
Subdevice #0: subdevice #0
card 1: sun50ia64hdmi [sun50i-a64-hdmi], device 0: 1c22800.i2s-i2s-hifi i2s-hifi-0 [1c22800.i2s-i2s-hifi i2s-hifi-0]
Subdevices: 1/1
Subdevice #0: subdevice #0

Alsamixer Settings;

November 3, 2020

3 minutes ago, @lex said:

This error is: /* No such device or address */

You should double-check:

1. Is your sensor OV5640?

2. Check the connector, some are reversed 180º which is the case for BPI and Orange Pi. There is a Thread about it, 3 years old a think.

yes sensor is ov5640 and not reversed maybe there was a hardware issue, i dont know, i ordered a new one. Btw i have a reversed one (BP camera). I double check everything but no luck. May i ask is ov5640 works without any issue with a64 (pine64) or could you advise any camera ? i need 30 fps 640x480 to process with opencv. Thank you

November 3, 2020

1 hour ago, Robert Heller said:

A fragment of my DTS is elsewhere in this thread. Note: it is for a Banana Pi M64, which is probably different from your Pine64. I got my working properly by refering to the schematic diagram I have for the Banana Pi M64. You need to get the schematic diagram for your Pine64, so you know exactly which GPIO pines are wired to the CSI connector, then you can be sure that the DTS file properly sets up the correct GPIO pins to the correct functions.

Thanks for fast reply,First I changed them according to schematic and i checked, gpios are correct, if there was an issue on my camera or not connectted, could i see device wiith "v4l2-ctl --list-devices" ? Maybe there is a problem on camera hardware. Thank you

November 3, 2020

On 10/31/2020 at 7:40 PM, Robert Heller said:

OK, make sure that the ov5640 module is also enabled.

Hi, I am working more than 4 days but no luck, could you share your dts and config file. I dont understand whats wrong. i get "ov5640 1-004c: ov5640_init_slave_id: failed with -6" error

Thank you

root@pine64:~# dmesg |grep ov5640
[ 5.692745] sun6i-csi 1cb0000.csi: creating ov5640 1-004c:0 -> sun6i-csi:0 link
[ 5.966763] ov5640 1-004c: ov5640_init_slave_id: failed with -6
[ 595.120402] ov5640 1-004c: ov5640_init_slave_id: failed with -6

root@pine64:~# v4l2-ctl --list-devices
cedrus (platform:cedrus):
/dev/video0

root@pine64:~# lsmod | grep ov
ov5640 32768 1
v4l2_fwnode 28672 2 ov5640,sun6i_csi
videodev 258048 7 sunxi_cedrus,v4l2_fwnode,ov5640,videobuf2_v4l2,sun6i_csi,videobuf2_common,v4l2_mem2mem
mc 57344 7 sunxi_cedrus,videodev,ov5640,videobuf2_v4l2,sun6i_csi,videobuf2_common,v4l2_mem2mem

October 31, 2020

Actually i am working on pine64 a64+ , i try to enable ov5640. i thought, i could fix with pinetab csi settings as you mention at above.

October 31, 2020

Yes i did it, unfortunatelly stuck editing dts file. could you give me a start point to use customized dts. I found dtb file under "/build/cache/sources/linux-mainline/orange-pi-5.8/arch/arm64/boot/dts/allwinner/ " but how could i customize dts before compile the kernel, i dont understand. I mean, how to attach customized dts. Thanks for your help.

October 31, 2020

I try to enable CSI Camera for pine64 A64+ Mainline Kernel (5.8x) config "CONFIG_VIDEO_SUN6I_CSI=m" NOT located in the kernel config file. And I try to add this line to /build/userpatches/linux-sunxi64-current.config and recompile. Unfortunately my compiled konfig file doesnt contain CONFIG_VIDEO_SUN6I_CSI, i try to find it in KKonfig menu but i coult not locate in the menu. Could anyone help? thank you.

October 20, 2020

there was an issue about mac address when i was working on it. I dont have a setup to test with current kernel sorry

September 12, 2020

1 hour ago, Tido said:

ArmbianIO API - in C
https://forum.armbian.com/topic/5655-armbianio-api-proposal/

pyGPIO - A 'more general' python GPIO library based on pyA20
https://forum.armbian.com/topic/5662-pygpio-a-more-general-python-gpio-library

Thanks for your answer, i checked ArmbianIO, it is based on sysfs.

pyGPIO is based on dev/mem, this is what i am looking unfortunatelly it does not support interrupts. I could not find a way to get gpio int from memory.

thank you.

September 12, 2020

On 11/2/2018 at 8:43 AM, martinayotte said:
The simplest way is to use gpio sysfs ...
echo 20 > /sys/class/gpio/export
echo out > /sys/class/gpio/gpio20/direction
echo 1 > /sys/class/gpio/gpio20/value

Hi,I am a developing a c++ application with OPI 0+2 (h5 + mainline kernel), controls a motor with an encoder and a driver. I did it , but it consume to much cpu power (%8) . I could access with using sysfs and character device. Both of them works like open->read->close .Actually it works very well with polling. Unfortunately consume to much cpu. Interrupt freq around 6600hz (150 us) . Is there anyway to control interrupt with dev mem? Or could anyone give me a startup point to access (c++) gpio interrupts efficiently. Btw i tyied libgpiod based on character device. Thank you.

April 29, 2020

Bluetooth init success with Orange pi zero plus 2 Mainline kernel 5.4.28

i solve problem with these steps:

1) install bluetooth support from armbian-config

2) install devmem2 and compile if not installed

wget http://free-electrons.com/pub/mirror/devmem2.c
gcc ./devmem2.c
mv ./a.out /usr/local/bin/devmem2

3) add modules bluetooth and hci_uart to /etc/modules

4) remove bluetooth service (we will add new script)

update-rc.d bluetooth remove

5) remove blutooth init script (editing not worked for me)

rm /etc/init.d/bluetooth

6) copy modified bluetooth init script file(ap6212) to /etc/init.d/

7) edit permissions

chmod +x /etc/init.d/ap6212

8) enable service script

systemctl enable ap6212

9) reboot

after reboot probably you will see your service succefully running

systemctl status ap6212

● ap6212.service - LSB: Start bluetooth daemons
Loaded: loaded (/etc/init.d/ap6212; generated)
Active: active (running) since Tue 2020-04-28 23:31:54 UTC; 38s ago
Docs: man:systemd-sysv-generator(8)
Process: 735 ExecStart=/etc/init.d/ap6212 start (code=exited, status=0/SUCCESS)
Tasks: 1 (limit: 371)
Memory: 4.2M
CGroup: /system.slice/ap6212.service
└─1071 /usr/sbin/bluetoothd

Apr 28 23:31:51 orangepizeroplus2-h5 systemd[1]: Starting LSB: Start bluetooth daemons...
Apr 28 23:31:54 orangepizeroplus2-h5 ap6212[735]: Starting Patch ap6212:/dev/mem opened.
Apr 28 23:31:54 orangepizeroplus2-h5 ap6212[735]: Memory mapped at address 0xffff87107000.
Apr 28 23:31:54 orangepizeroplus2-h5 ap6212[735]: Value at address 0x1F00060 (0xffff87107060): 0x0
Apr 28 23:31:54 orangepizeroplus2-h5 ap6212[735]: Written 0x1; readback 0x1
Apr 28 23:31:54 orangepizeroplus2-h5 ap6212[735]: Starting ap6212: bluetoothd.
Apr 28 23:31:54 orangepizeroplus2-h5 systemd[1]: Started LSB: Start bluetooth daemons.
Apr 28 23:31:54 orangepizeroplus2-h5 bluetoothd[1071]: Bluetooth daemon 5.50
Apr 28 23:31:54 orangepizeroplus2-h5 bluetoothd[1071]: Starting SDP server
Apr 28 23:31:54 orangepizeroplus2-h5 bluetoothd[1071]: Bluetooth management interface 1.14 initialized

Q: what is difference between ap6212 and bluetooth service init?

A: added gpio controls

Now you can attach uart with sucess

hciattach /dev/ttyS1 any 115200 flow bdaddr 00:90:4c:c5:12:38
hciconfig hci0 up

i could not change mac adress device always setup with AA:AA:AA:AA:AA:AA

if you can please let me know.

Thanks

Save as ap6212 and copy to /etc/init.d/

#! /bin/sh
### BEGIN INIT INFO
# Provides: ap6212
# Required-Start:    $local_fs $syslog $remote_fs dbus
# Required-Stop:     $local_fs $syslog $remote_fs
# Default-Start:     2 3 4 5
# Default-Stop:      0 1 6
# Short-Description: Start bluetooth daemons
### END INIT INFO
#
# bluez    Bluetooth subsystem starting and stopping
#
# originally from bluez's scripts/bluetooth.init
#
# Edd Dumbill <ejad@debian.org>
# LSB 3.0 compilance and enhancements by Filippo Giunchedi <filippo@debian.org>
#
# Updated for bluez 4.7 by Mario Limonciello <mario_limonciello@dell.com>
# Updated for bluez 5.5 by Nobuhiro Iwamatsu <iwamatsu@debian.org>
#
# Note: older daemons like dund pand hidd are now shipped inside the
# bluez-compat package

PATH=/sbin:/bin:/usr/sbin:/usr/bin
DESC=ap6212

DAEMON=/usr/sbin/bluetoothd
HCIATTACH=/usr/bin/hciattach

HID2HCI_ENABLED=1
HID2HCI_UNDO=1

SDPTOOL=/usr/bin/sdptool

# If you want to be ignore error of "org.freedesktop.hostname1",
# please enable NOPLUGIN_OPTION.
# NOPLUGIN_OPTION="--noplugin=hostname"
NOPLUGIN_OPTION=""
SSD_OPTIONS="--oknodo --quiet --exec $DAEMON -- $NOPLUGIN_OPTION"

test -f $DAEMON || exit 0

# FIXME: any of the sourced files may fail if/with syntax errors
test -f /etc/default/bluetooth && . /etc/default/bluetooth
test -f /etc/default/rcS && . /etc/default/rcS

. /lib/lsb/init-functions

set -e

# FIXME: this function is possibly a no-op
run_sdptool()
{
	# declaring IFS local in this function, removes the need to
	# save/restore it
	local IFS o

	test -x $SDPTOOL || return 1

# FIXME: where does SDPTOOL_OPTIONS come from?
	if ! test -z "$SDPTOOL_OPTIONS" ; then
		IFS=";"
		for o in $SDPTOOL_OPTIONS ; do
			#echo "execing $SDPTOOL $o"
			IFS=" "
			if [ "$VERBOSE" != no ]; then
				$SDPTOOL $o
			else
				$SDPTOOL $o >/dev/null 2>&1
			fi
		done
	fi
}

hci_input()
{
    log_progress_msg "switching to HID/HCI no longer done in init script, see /usr/share/doc/bluez/NEWS.Debian.gz" || :
}


fixstart()
{
  log_daemon_msg "Starting Patch $DESC"
  /usr/local/bin/devmem2 0x1f00060 b 1
  echo 10 > /sys/class/gpio/export
  echo out > /sys/class/gpio/gpio10/direction
  echo 0 > /sys/class/gpio/gpio10/value
  echo 1 > /sys/class/gpio/gpio10/value
  sleep 0.1
}

fixstop()
{
	# disable gpio pin, if not wifi not active on soft restart / reboot
  log_daemon_msg "Disable Patch $DESC" 

  /usr/local/bin/devmem2 0x1f00060 b 0
  echo 10 > /sys/class/gpio/export
  echo out > /sys/class/gpio/gpio10/direction
  echo 0 > /sys/class/gpio/gpio10/value
  sleep 0.1
  echo 10 > /sys/class/gpio/unexport
  sleep 0.1
}

alias enable_hci_input=hci_input
alias disable_hci_input=hci_input
alias enable_fix=fixstart
alias disable_fix=fixstop

case $1 in
  start)
  	enable_fix
	
	log_daemon_msg "Starting $DESC"

	if test "$BLUETOOTH_ENABLED" = 0; then
		log_progress_msg "disabled. see /etc/default/bluetooth"
		log_end_msg 0
		exit 0
	fi   

	start-stop-daemon --start --background $SSD_OPTIONS
	log_progress_msg "${DAEMON##*/}"

	run_sdptool || :

	if test "$HID2HCI_ENABLED" = 1; then
		enable_hci_input
	fi

	log_end_msg 0
  ;;
  stop)
  	disable_fix
          
	log_daemon_msg "Stopping $DESC"
	if test "$BLUETOOTH_ENABLED" = 0; then
		log_progress_msg "disabled."
		log_end_msg 0
		exit 0
	fi
	if test "$HID2HCI_UNDO" = 1; then
		disable_hci_input
	fi
	start-stop-daemon --stop $SSD_OPTIONS
	log_progress_msg "${DAEMON}"
	log_end_msg 0
  ;;
  restart|force-reload)
	$0 stop
	sleep 1
	$0 start
  ;;
  status)
	status_of_proc "$DAEMON" "$DESC" && exit 0 || exit $?
  ;;
  *)
	N=/etc/init.d/ap6212
	echo "Usage: $N {start|stop|restart|force-reload|status}" >&2
	exit 1
	;;
esac

exit 0

# vim:noet

April 27, 2020

On 3/2/2020 at 10:49 PM, Spider said:
Thx, for reply. But I finaly launch BT device.

Next steps:

1. Disable uart1 overlay

2. copyed firmware for ap6212 from last broadcom package into /lib/firmware/bcm....

3. modified ap6212-bluetooth

- at start add:
        devmem2 0x1f00060 b 1
        echo 10 > /sys/class/gpio/export
        echo out > /sys/class/gpio/gpio10/direction
        echo 0 > /sys/class/gpio/gpio10/value
        echo 1 > /sys/class/gpio/gpio10/value
        sleep 0.1
- at end add: (else after soft reboot wifi will not work)
        devmem2 0x1f00060 b 0
        echo 10 > /sys/class/gpio/export
        echo out > /sys/class/gpio/gpio10/direction
        echo 0 > /sys/class/gpio/gpio10/value

        sleep 0.1

        echo 10 > /sys/class/gpio/unexport
        sleep 0.1

Hi , i have a same problem with opiz+2 Mainline kernel 5.4.2x i could not find ap6212-bluetooth file (i did it before on legacy kernel), now i have only bluetooth file. I try to edit this file but nothing change. Could you a bit clear how to success with bluetooth. Thanks

Sign In

Forums

Store

Crowdfunding

Applications

Events

Raffles

Community Map

Posts posted by Levent Erenler

Important Information