[docker] 일반유저로 도커 실행하기

sudo usermod -aG docker $USER

sudo su - $USER

groups $USER

[개발상식] Classless Inter Domain Routing ==> CIDR 표기법

마스크가 포함된 네트워크 표기법?

192.168.0.32/24 인 ip가 속해있는 네트워크는?

192.168.0.0~255까지임.

192.168.0.31/26 인 ip가 속해있는 네트워크는?

1100 0000. 1010 1000.0000 0000.0001 1111

(32-26=)6비트표현가능숫자 ==> 64개, ==> 1~63 range.

https://ko.wikipedia.org/wiki/CIDR

https://www.subnet-calculator.com/cidr.php

disable systemd service in yocto?

in file meta-xxx/recipes-core/systemd/systemd_%.bbappend

do_configure_append() {

  #disabling autostart of systemd-timesyncd      sed -i -e "s/enable systemd-timesyncd.service/disable systemd-timesyncd.service/g" ${S}/presets/90-systemd.preset    #disabling autostart of systemd-resolved      sed -i -e "s/enable systemd-resolved.service/disable systemd-resolved.service/g" ${S}/presets/90-systemd.preset  }

userspace 에서의 led class 를 이용하는 방법으로, /dev/uleds 열어서 핸들을 얻어온후.

그 핸들을 이용해서 시스템에 등록된 모든 led 기능을 사용한다.

커널 디바이스 드라이버 --> LED --> userspace led 선택.

https://www.kernel.org/doc/html/latest/leds/uleds.html

리눅스 노트북 빌드가 느릴때 확인사항.

sudo add-apt-repository ppa:linrunner/tlp  sudo apt update

sudo apt install tlp tlp-rdw

sudo apt install acpi-call-dkms

sudo apt install auto-cpufreq

cpufreq

watch 'grep "cpu MHz" /proc/cpuinfo'

Errno=No space left on device (ENOSPC) 해결법

sudo sysctl fs.inotify.max_user_watches=1000000

sudo vi /etc/sysctl.d/99-custom.conf

fs.inotify.max_user_watches=1000000 라고 써

[개발노트] sd카드 만들기.

#!/bin/sh

DEVSDCARD=/dev/sdb
DEFAULT_SDCARD_PARTUUID=e91c4e10-16e6-4c0e-bd0e-77becf4a3582
sudo partx -d ${DEVSDCARD}
sudo sgdisk -o ${DEVSDCARD}
sudo sgdisk --resize-table=128 -a 1  \
        -n 1:34:545 -c 1:fsbl1 \
        -n 2:546:1057 -c 2:fsbl2 \
        -n 3:1058:5153 -c 3:fip \
        -n 4:5154:136225 -c 4:bootfs -A 4:set:2 \
        -n 5:136226:2233377 -c 5:rootfs -u 5:${DEFAULT_SDCARD_PARTUUID} \
        -n 6:2233378:4330529 -c 6:userfs \
        -n 7:4330530:8524833 -c 7:raw \
        -p ${DEVSDCARD}
sudo dd if=arm-trusted-firmware/tf-a-stm32mp153d-ssonic-sdcard.stm32 of=${DEVSDCARD}1 conv=fdatasync  bs=8M
sudo dd if=arm-trusted-firmware/tf-a-stm32mp153d-ssonic-sdcard.stm32 of=${DEVSDCARD}2 conv=fdatasync  bs=8M
sudo dd if=fip/fip-stm32mp153d-ssonic-trusted.bin of=${DEVSDCARD}3 conv=fdatasync  bs=8M
sudo dd if=st-image-bootfs-openstlinux-weston-stm32mp153d-ssonic.ext4 of=${DEVSDCARD}4 conv=fdatasync  bs=8M
sudo dd if=st-image-core-openstlinux-weston-stm32mp153d-ssonic.ext4 of=${DEVSDCARD}5 conv=fdatasync  bs=8M
sudo dd if=st-image-userfs-openstlinux-weston-stm32mp153d-ssonic.ext4 of=${DEVSDCARD}6 conv=fdatasync  bs=8M
sudo dd if=/dev/null of=${DEVSDCARD}7 conv=fdatasync  bs=8M count=4

[개발노트] eeprom에 mac address 넣기 참고용.

^^

 macb1: ethernet@f802c000 {

    compatible = "atmel,sama5d3-macb", "cdns,at91sam9260-macb", "cdns,macb";

    status = "okay";

    phy-mode = "rmii";

    #address-cells = <1>;

    #size-cells = <0>;

    nvmem-cells = <&eth0_addr>;

    nvmem-cell-names = "mac-address";

    phy0: ethernet-phy@1 {

        interrupt-parent = <&pioE>;

        interrupts = <30 IRQ_TYPE_EDGE_FALLING>;

        reg = <1>;

    };

    /*ethernet-phy@1 {

        reg = <0x1>;

    };*/

};

/* Bit banging internal I2C to manage the AT24MAC402 chip */

i2c3@ {

    compatible = "i2c-gpio";

    gpios = <&pioE 1 0 /* SDA */

             &pioE 2 0 /* SCK */

            >;

    i2c-gpio,delay-us = <4>;    /* ~178 kHz */

    #address-cells = <1>;

    #size-cells = <0>;

    /* EEPROM contains the eth0 MAC address */

    eeprom@58 {

        compatible = "atmel,24mac402";

        pagesize = <256>;

        read-only;

        reg = <0x58>;

        #address-cells = <1>;

        #size-cells = <1>;

        eth0_addr: eth-mac-addr@9A {

            reg = <0x0 0x06>;

        };

    };

};

[삽질메모] nmcli

#ipv6 off
nmcli con modify br0 ipv6.method "disabled"

systemctl stop NetworkManager
systemctl restart NetworkManager


#static ip and bridge mode.
nmcli con add type bridge ifname br0 con-name br0
nmcli con add type bridge-slave ifname lan1 master br0 con-name lan1
nmcli con add type bridge-slave ifname lan2 master br0 con-name lan2

nmcli con modify br0 ipv4.addresses '192.168.222.77/24'
nmcli con modify br0 ipv4.gateway '192.168.222.1'
nmcli con modify br0 ipv4.dns '8.8.8.8 8.8.4.4'
nmcli con modify br0 ipv6.method disabled
nmcli con modify br0 ipv4.method manual

nmcli con modify br0 bridge.stp no

nmcli con up id br0

[   17.156194] br0: port 1(lan1) entered blocking state

[   17.159734] br0: port 1(lan1) entered listening state

[   32.472068] br0: port 1(lan1) entered learning state

[   47.832064] br0: port 1(lan1) entered forwarding state

[   47.835759] br0: topology change detected, propagating

systemd-networkd만 사용할때는 파란부분 blocking->forwarding 으로 빠르게 진행되지만.

NetworkManager만 사용할때는 빨간부분까지 4단계를 거치면서 빨간단계당 5초이상의 시간을 소요한다. 이것은 stp를 끔으로서 해결되었다.(빨간단계를 수행지않고 파란거만 빨리 진행된다는말)

stp : 스페팅트리프로토콜로써 스위치루프를 감지하여 죽이는 알고리즘이란다. 나는필요없...

systemctl daemon-reload
systemctl restart NetworkManager

#delete connection.
nmcli -f bridge con delete br0
nmcli -f bridge-slave con delete lan1
nmcli -f bridge-slave con delete lan2

nmcli -f ethernet con delete lan1


?? default gateway setup.

#gateway mode.
nmcli con add type ethernet ifname lan1 con-name lan1
nmcli con modify lan1 ipv4.addresses '192.168.222.77/24'
nmcli con modify lan1 ipv4.gateway '192.168.222.1'
nmcli con modify lan1 ipv4.dns '8.8.8.8 8.8.4.4'
nmcli con modify lan1 ipv4.method manual && \
nmcli con up id lan1 && \
nmcli con down id lan1 && \
nmcli con up id lan1


# nmcli connection modify bridge0 ipv4.addresses '192.0.2.1/24'
# nmcli connection modify bridge0 ipv4.gateway '192.0.2.254'
# nmcli connection modify bridge0 ipv4.dns '192.0.2.253'
# nmcli connection modify bridge0 ipv4.dns-search 'example.com'
# nmcli connection modify bridge0 ipv4.method manual // auto

nmcli connection modify bridge-br0 ipv4.method manual
nmcli con down id bridge-br0
nmcli con up id bridge-br0

route delete default gw 192.1678.225.1 lte0


nmcli con down id bridge-br0
nmcli con up id bridge-br0
nmcli con mod bridge-br0 ipv4.route-metric 1

nmcli con add con-name lan1 type ethernet ifname lan1 ipv4.method auto
nmcli con up id lan1
nmcli con add con-name lan2 type ethernet ifname lan2 ipv4.method shared
nmcli con up id lan2

nmcli -f ethernet con delete lan1
nmcli -f ethernet con delete lan2

nmcli connection add type ethernet ifname lan2 ipv4.method shared con-name local

nmcli c add con-name lan2 type ethernet ifname lan2 ipv4.method shared ipv6.method ignore
nmcli c up local

esp32-test softap_start_args happycpu p@ss1234 4

nmcli con add type ethernet ifname ethap0 ipv4.method shared con-name ethap0 ipv4.addresses 192.168.100.1/24

nmcli con up id ethap0

ip route add default via 192.168.225.1 dev lte0

ip route replace default via 192.0.2.1 dev eth0

ip route del default via 1.2.3.4 dev eth0 proto static metric 100
ip route add default via 1.2.3.4 dev eth0 proto static metric 90

[개발노트] overlay.txt란 이런것.

https://developer.toradex.com/linux-bsp/how-to/device-tree/device-tree-overlays-linux/#preenabled-device-tree-overlays 발췌~

Enable/Disable a Device Tree Overlay

To enable or disable applying a pre-built overlay on a Toradex SoM just add/remove it to/from the file overlays.txt located on the boot partition, and reboot the system.

For example, one wants to disable a parallel RGB interface and display on Colibri-iMX8X and enable an HDMI interface (using DSI-to-HDMI adapter). The initial overlays.txt file looks like:

# cat /boot/overlays.txt 

fdt_overlays=colibri-imx8x_parallel-rgb_overlay.dtbo colibri-imx8x_ad7879_overlay.dtbo display-edt7_overlay.dtbo

Steps to disable a parallel RGB and enable an HDMI:

Open the overlays.txt file with any text editor, e.g. vi:

# vi /boot/overlays.txt 

Remove overlays, that is needed for Resistive Touch Display 7" Parallel and add one for DSI-to-HDMI adapter. The file should look like:

# cat /boot/overlays.txt 

fdt_overlays=colibri-imx8x_dsihdmi_overlay.dtbo

Synchronize them filesystems:

# sync

Reboot the system:

# reboot