Dantes

Can't get hdmi output in Armbian/6.11 yet, so I'll have to wait for that to clear up. However these 3rd party OpenWRT builds (https://github.com/mj22226/openwrt/releases) all show the same thing: OpenWRT Linux 6.1.94 # grep RELEASE /etc/os-release OPENWRT_RELEASE="OpenWrt SNAPSHOT r0-4dc539f" # uname -a Linux 6.1.94 #0 SMP Sun Jun 16 23:31:23 2024 aarch64 GNU/Linux # lspci 0003:30:00.0 PCI bridge: Rockchip Electronics Co., Ltd RK3588 (rev 01) 0003:31:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8125 2.5GbE Controller (rev 05) 0004:40:00.0 PCI bridge: Rockchip Electronics Co., Ltd RK3588 (rev 01) # lspci -vvvs 01:00|grep -E "L1,ASPM" # OpenWRT Linux 6.4.2 # grep RELEASE /etc/os-release OPENWRT_RELEASE="OpenWrt SNAPSHOT r0-1a94f52" # uname -a Linux 6.4.2 #0 SMP PREEMPT Fri Jul 7 23:03:17 2023 aarch64 GNU/Linux # lspci 0003:30:00.0 PCI bridge: Rockchip Electronics Co., Ltd Device 3588 (rev 01) 0003:31:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8125 2.5GbE Controller (rev 05) 0004:40:00.0 PCI bridge: Rockchip Electronics Co., Ltd Device 3588 (rev 01) # lspci -vvvs 01:00|grep -E "L1,ASPM" # OpenWRT Linux 6.6.48 # grep RELEASE /etc/os-release OPENWRT_RELEASE="OpenWrt SNAPSHOT r0-5a5368e" # uname -a Linux 6.6.48 #0 SMP Fri Aug 30 11:13:30 2024 aarch64 GNU/Linux # lspci 0003:30:00.0 PCI bridge: Rockchip Electronics Co., Ltd RK3588 (rev 01) 0003:31:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8125 2.5GbE Controller (rev 05) 0004:40:00.0 PCI bridge: Rockchip Electronics Co., Ltd RK3588 (rev 01) # lspci -vvvs 01:00|grep -E "L1,ASPM" # OpenWRT Linux 6.11-rc # grep RELEASE /etc/os-release OPENWRT_RELEASE="OpenWrt SNAPSHOT r0-fe4b9c3" # uname -a Linux 6.11-rc5 #0 SMP Sun Aug 25 07:38:04 2024 aarch64 GNU/Linux # lspci 0003:30:00.0 PCI bridge: Rockchip Electronics Co., Ltd RK3588 (rev 01) 0003:31:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8125 2.5GbE Controller (rev 05) 0004:40:00.0 PCI bridge: Rockchip Electronics Co., Ltd RK3588 (rev 01) # lspci -vvvs 01:00|grep -E "L1,ASPM" #

That might be because its more of an appliance/router than a desktop. It might be a regression. I'd wait for 6.11 to arrive, Sunday we're at 6.11-rc6, so that would mean another 3 weeks give or take. But I have a couple, so I checked them for you: OpenWRT # grep RELEASE /etc/os-release OPENWRT_RELEASE="OpenWrt SNAPSHOT r0-212b6f9" # uname -a Linux 6.10 #0 SMP Mon Jul 15 13:18:21 2024 aarch64 GNU/Linux # lspci 0003:30:00.0 PCI bridge: Rockchip Electronics Co., Ltd RK3588 (rev 01) 0003:31:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8125 2.5GbE Controller (rev 05) 0004:40:00.0 PCI bridge: Rockchip Electronics Co., Ltd RK3588 (rev 01) # lspci -vvvs 01:00|grep -E "L1,ASPM" # Armbian $ grep ^PRETTY /etc/os-release PRETTY_NAME="Armbian 24.11.0-trunk.71 bookworm" $ uname -a Linux 6.10.7-current-rockchip-rk3588 #1 SMP PREEMPT Thu Aug 29 15:36:13 UTC 2024 aarch64 GNU/Linux $ lspci 0003:30:00.0 PCI bridge: Rockchip Electronics Co., Ltd RK3588 (rev 01) 0003:31:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8125 2.5GbE Controller (rev 05) 0004:40:00.0 PCI bridge: Rockchip Electronics Co., Ltd RK3588 (rev 01) $ lspci -vvvs 01:00|grep -E "L1,ASPM" $ Dantes

Dear Community, This might only be with the NanoPI-r6c images, but the SHASUM contains a temporary path which you don't have an thus fails. Since we cannot change the SHASUM because then the ASC would fail, I have a temporary workaround: sed 's,/.*/,,' armbian.img.sha | sha256sum -c In a script you could either 1. CD into the directory of the image 2. Run the SHASUM there 3. Use $OLDPWD to return or correct the SHASUM inline: sha256sum /path/to/file | sed 's,/.*/,,' > armbian.img.sha Dantes

Enable logging and check for hardware keys with $ tail -f ~/.kodi/temp/kodi.log then start pressing keys to see which ones are working. Those are the only ones you can work with.

Hello, Another point I would like to make is that all recommended tools to write sdcards are not in Armbian. So I would hereby formally request "usbimager" as an Armbian package since it was the only one I saw with source code available. That being said, I found myself being locked out as an arm64 only user. All packages are either for MAC or WIN, and if they are for Linux they are x86/x64. So what can you do ? Step 1. Install packages: $ apt install pv xz-utils Step 2. Verify checksum $ sha256sum -c armbian.img.xz.sha amrbian.img.xz: OK Step 3. Unmount sdcard: (very important! to avoid corruption) $ umount /dev/sdcard* umount: /dev/sdcard: not mounted. Step 4. Write image to sdcard $ pv armbian.img.xz | xzcat | sudo dd bs=10M oflag=sync of=/dev/sdcard [sudo] password for test: 1,31GiB 0:01:54 [11,6MiB/s] [===========================================>] 100% 0+867065 records in 0+867065 records out 7256145920 bytes (7,3 GB, 6,8 GiB) copied, 113,024 s, 64,2 MB/s Step 5. Verify image from sdcard (--size=bytes copied from step 4.) $ pv --stop-at-size --size=7256145920 /dev/sdcard | sha256sum 1,31GiB 0:01:43 [11,9MiB/s] [===========================================>] 100% 05ea4603da3c5f5517f6e3f83b8bc456d31e299ac97f564e77a49487f93e5f1b - Step 6. Check if the hash matches $ cat armbian.img.sha 05ea4603da3c5f5517f6e3f83b8bc456d31e299ac97f564e77a49487f93e5f1b *armbian.img.xz My 2 cents

Hi fella's Well I think this is worth updating the #how-to-prepare-a-sd-card page with an extra bullet point pointing to: https://www.grc.com/validrive.htm. This freeware tool detects fraudulent USB mass storage devices. The page explains it in more detail.

Maybe there is a sticky somewhere on this forum, (if not there should be) pointing to this url: https://gitlab.collabora.com/hardware-enablement/rockchip-3588/notes-for-rockchip-3588/-/blob/main/mainline-status.md This is the mainline status of the rk3588(s). How far a long they are and what features have been implemented. I'd give it another year or so.. but I'm looking forward to 6.12 that will be released before years end.

Still don't know what it is. The problem persists in Armbian_community_24.11.0-trunk.25_Nanopi-r6s_noble_current_6.10.6_gnome-kisak_desktop.img.xz

Speeeeeeeeeeeeeeeeeeeeeeeeeedy, I love it

Just found this: source: https://blog.qualys.com/vulnerabilities-threat-research/2024/07/01/regresshion-remote-unauthenticated-code-execution-vulnerability-in-openssh-server source: https://arstechnica.com/security/2024/07/regresshion-vulnerability-in-openssh-gives-attackers-root-on-linux/ The vulnerability, which is a signal handler race condition in OpenSSH’s server (sshd), allows unauthenticated remote code execution (RCE) as root on glibc-based Linux systems; that presents a significant security risk. This race condition affects sshd in its default configuration. Are we affected running Armbian ?

It seems a lot of mate-panel layouts do not work and are not visible in mate-tweak. First I downloaded an old ubuntu-mate 18.04 from archive.org. Then I copied the old layouts from the ISO into /usr/share/mate-panel/layouts And they work just fine on Debian Bookworm (go figure ) If you find Debian Mate too stale for your taste, unpack the .tgz below into the root / tar -zxvf mate-panel-layouts.tgz -C / Then run mate-tweak and select one of the layouts. I must say that except for the older kernel, this could almost be a daily driver. mate-panel-layouts.tgz

My audio volume was very very low, and very annoyingly I had to boost to 200% 300% which had the problem that the volume buttons were not working properly. You just cant see anything above 100% unless you go into the control panel. Now when it comes to audio I'm the village idiot, but I found a proper solution I think. After some experimentation I found that the module "module-cli-protocol-unix" ( #20 in my case) can be amplified with pactl set-sink-input-volume 20 10db Now YMMV depending on your attached hardware, so change the 10db to whatever suits your situation. After that I made it permanent by adding it to the /etc/pulse/default.pa like this: # Default volume amplification set-sink-input-volume 20 10db Yay!

Yay! I have it working But I'm not an expert in compiling so ymmv Feel free to comment or point out any mistakes # when libshaderc-dev is installed the build succeeds, but has a run-time failure sudo apt remove \ libshaderc-dev # install compiling tools sudo apt install \ autogen automake cmake libtool meson pkg-config # install development libraries sudo apt install \ libalsaplayer-dev libaom-dev libarchive-dev libass-dev libavcodec-dev libavfilter-dev \ libcaca-dev libd3dadapter9-mesa-dev libegl1-mesa-dev libfdk-aac-dev libfontconfig-dev \ libfreetype-dev libfribidi-dev libharfbuzz-dev libjack-dev libjpeg-dev libmp3lame-dev \ libmujs-dev libopenal-dev libopus-dev libpipewire-0.3-dev libplacebo-dev libsdl2-dev \ libsixel-dev libsoxr-dev libunwind-dev libva-dev libvdpau-dev libvkd3d-dev libvorbis-dev \ libvpx-dev libx264-dev libx265-dev libxpresent-dev libxxhash-dev libzimg-dev # clone mpv build script repository git clone https://github.com/mpv-player/mpv-build # configure ffmpeg options by creating a file called mpv-build/ffmpeg_options with content: --enable-static --disable-shared --enable-gpl --enable-libaom --enable-libfdk-aac --enable-libfribidi --enable-libmp3lame --enable-libopus --enable-libsmbclient --enable-libsoxr --enable-libvorbis --enable-libvpx --enable-libx264 --enable-libx265 --enable-nonfree --enable-version3 # configure mpv options by creating a file called mpv-build/mpv_options with content: -Dlibmpv=true # start building cd mpv-build ./rebuild -j$(nproc) # strip executable cp -p mpv/build/mpv /tmp/mpv strip --strip-unneeded /tmp/mpv touch -r mpv/build/mpv /tmp/mpv # test mpv /tmp/mpv smb://myserver/myshare/myfile.mkv Yay!

@TRS-80 Tried to compile it, but apparently my skills suck in this department. I'm ending up with errors every time. So this is best suited for someone who actually knows what he's doing. What I did: git clone https://github.com/mpv-player/mpv sudo apt install \ libsmbclient-dev \ meson \ cmake \ pkg-config \ libshaderc-dev \ autogen \ libunwind-dev \ libplacebo-dev \ autoconf \ automake \ libtool \ libfreetype-dev \ libfribidi-dev \ libharfbuzz-dev \ libfontconfig-dev cd mpv ./rebuild -j4

It seems a package is missing from the Armbian 23.8 Bookworm CLI that results in this error. I encountered an add-apt-repository: traceback error. To fix it just run: sudo apt install python3-launchpadlib

Awesome

As you can see the WAN-port/NIC registers fine with a label from the nanopi-r6c device tree $ grep fe1c0000 /boot/dtb/rockchip/rk3588s-nanopi-r6c.dtb grep: /boot/dtb/rockchip/rk3588s-nanopi-r6c.dtb: binary file matches dmesg: dmesg | grep fe1c0000 [ 7.287277] rk_gmac-dwmac fe1c0000.ethernet: init for RGMII_RXID [ 7.287373] rk_gmac-dwmac fe1c0000.ethernet: User ID: 0x30, Synopsys ID: 0x51 [ 7.287377] rk_gmac-dwmac fe1c0000.ethernet: DWMAC4/5 [ 7.287381] rk_gmac-dwmac fe1c0000.ethernet: DMA HW capability register supported [ 7.287385] rk_gmac-dwmac fe1c0000.ethernet: RX Checksum Offload Engine supported [ 7.287388] rk_gmac-dwmac fe1c0000.ethernet: TX Checksum insertion supported [ 7.287391] rk_gmac-dwmac fe1c0000.ethernet: Wake-Up On Lan supported [ 7.287423] rk_gmac-dwmac fe1c0000.ethernet: TSO supported [ 7.287426] rk_gmac-dwmac fe1c0000.ethernet: Enable RX Mitigation via HW Watchdog Timer [ 7.287430] rk_gmac-dwmac fe1c0000.ethernet: Enabled Flow TC (entries=2) [ 7.287433] rk_gmac-dwmac fe1c0000.ethernet: TSO feature enabled [ 7.287437] rk_gmac-dwmac fe1c0000.ethernet: Using 32 bits DMA width However the LAN-port does not have such a label in dmesg output. [ 7.932846] pci 0003:31:00.0: [10ec:8125] type 00 class 0x020000 [ 7.932902] pci 0003:31:00.0: reg 0x10: initial BAR value 0x00000000 invalid [ 7.932905] pci 0003:31:00.0: reg 0x10: [io size 0x0100] [ 7.932958] pci 0003:31:00.0: reg 0x18: [mem 0x00000000-0x0000ffff 64bit] [ 7.932993] pci 0003:31:00.0: reg 0x20: [mem 0x00000000-0x00003fff 64bit] [ 7.933301] pci 0003:31:00.0: supports D1 D2 [ 7.933303] pci 0003:31:00.0: PME# supported from D0 D1 D2 D3hot D3cold [ 7.948845] pci 0003:31:00.0: BAR 2: assigned [mem 0xf3200000-0xf320ffff 64bit] [ 7.948873] pci 0003:31:00.0: BAR 4: assigned [mem 0xf3210000-0xf3213fff 64bit] [ 7.948899] pci 0003:31:00.0: BAR 0: assigned [io 0x1000-0x10ff] [ 7.950294] r8125 0003:31:00.0: enabling device (0000 -> 0003) [ 7.967108] r8125 0003:31:00.0 (unnamed net_device) (uninitialized): Invalid ether addr 00:00:00:00:00:00 [ 7.967112] r8125 0003:31:00.0 (unnamed net_device) (uninitialized): Random ether addr a2:83:9c:9c:cc:e8

@Werner Maybe I'm wrong but I think that's the old notation. However this worked before (release before may 27th IIRC) and works again after I install a desktop. So I think that its missing a library or other dependency. I will be scouting for it.

I'm trying to setup disk encryption, but for some reason its not opening devices after installing cryptsetup-bin on the installation image. After typing the password it just hangs and does nothing $ cryptsetup open /dev/sda sda_crypt Anyone?

I think I found the problem So I compared logs from a working x86-64 system with the R6C aarch64 and as you can see mpv is compiled without --enable-libsmbclient $ grep Configuration aarch64_mpv.log | $ grep Configuration x86-64_mpv.log [cplayer] [cplayer] Configuration: Configuration: ./waf ./waf configure configure --prefix=/usr --prefix=/usr --libdir=/usr/lib/aarch64-linux-gnu | --libdir=/usr/lib/x86_64-linux-gnu --confdir=/etc/mpv --confdir=/etc/mpv --zshdir=/usr/share/zsh/vendor-completions --zshdir=/usr/share/zsh/vendor-completions --enable-cdda --enable-cdda --enable-dvdnav --enable-dvdnav --enable-libmpv-shared --enable-libmpv-shared > --enable-libsmbclient --enable-sdl2 --enable-sdl2 > --enable-sndio --disable-build-date --disable-build-date --enable-dvbin --enable-dvbin $ $ Just to be clear: celluloid depends on mpv which depends on ffmpeg On x86-64 (Ubuntu 22.04 LTS) where it works: * ffmpeg (4.2.7) is compiled without --enable-libsmbclient * mpv (0.32.0) with --enable-libsmbclient On aarch64 (Armbian/Ubuntu 22.04 LTS) where it does not: * ffmpeg (4.4.2) is compiled with --enable-libsmbclient * mpv (0.34.1) without --enable-libsmbclient I hereby humbly request someone with more skills to verify my findings and hopefully we can all enjoy playing files over smb:// in the future

I hate to sound like a broken record, but that 3rd party NanoPI OpenWRT image I suggested in the other thread comes with SMB4 preinstalled. Its real easy to setup (5 mins). If the problem is the same, it could be a driver, dtb-file or something else both have in common. If not, then its something to do with Armbian itself.

There is a thread on OpenWRT, where it show it can be installed on sdcard or emmc. You could try installing those and measuring the power usage between sdcard and emmc. If the problem is the same, I'll bet its somehow related to the emmc driver. link: https://forum.openwrt.org/t/nanopi-r6s-linux-6-3-arm-soc-updates/153072/34 link to images for nanopi: https://github.com/mj22226/openwrt/releases/tag/linux-6.4

Yes, that's how it is done , but "conv=notrunc" only make sense if the output is a file , not if it is a block device. I made the matrix so I could get a clearer picture of what you where asking, but this adds and extra dimension for sure. Did you try disabling CEC, I reckon you are not using a TV with a remote control ? That might be as you said the cause (HDMI-CEC).

I am unable to play files over SMB, but I can't seem to figure out what the problem is exactly. From what I can tell it *seems* ffmpeg is compiled without networking support, at least this is what smplayer was complaining about: [ffmpeg] Protocol not found. Make sure ffmpeg/Libav is compiled with networking support. Failed to open smb://192.168.1.1/transmission/myfile.mkv. Exiting... (Errors when loading file) But when I look at the ffmpeg switches it was compiled with: --enable-libsmbclient and seems to be present. Now I'm a complete linux n00b, does anyone have an idea ?

@Jiff Just curious, do you have the original case or did you make one yourself ?