zgoda_j

You have to build kernel driver from provided sources. In my opinion not worth the hassle, I dumped such dongle and bought another.

LeMaker fex configuration for 5in display works for me. The problem I had was from typo I made manually copying values. There are 4 sections that have to be overwritten, then you have to load ft5x_ts module to get touch working.

Thank you Thomas, you're the boss. I use network manager on my desktop and didn't even consider it for headless machine.

Back on this again. This is what I have in dmesg. If I'm reading this correctly, interface connects to AP and authenticates succesfully. Yet still it does not seem available? [ 9.088127] ADDRCONF(NETDEV_UP): wlan0: link is not ready [ 10.939428] wlan0: authenticate with 04:8d:38:c3:5a:24 [ 10.985502] wlan0: send auth to 04:8d:38:c3:5a:24 (try 1/3) [ 10.987038] wlan0: authenticated [ 11.020072] wlan0: associate with 04:8d:38:c3:5a:24 (try 1/3) [ 11.023916] wlan0: RX AssocResp from 04:8d:38:c3:5a:24 (capab=0x411 status=0 aid=4) [ 11.057886] ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready [ 11.058163] wlan0: associated [ 12.770284] PHY: gmac0-0:00 - Link is Up - 100/Full [ 18.830038] eth0: no IPv6 routers present [ 21.730038] wlan0: no IPv6 routers present [ 31.873128] ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready And $ iwconfig wlan0 wlan0 IEEE 802.11bgn ESSID:"destroyer9001" Mode:Managed Frequency:2.462 GHz Access Point: 04:8D:38:C3:5A:24 Bit Rate=30 Mb/s Tx-Power=20 dBm Retry long limit:7 RTS thr:off Fragment thr:off Power Management:on Link Quality=69/70 Signal level=-41 dBm Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0 Tx excessive retries:0 Invalid misc:253 Missed beacon:0 WAT?!

This is not related to hardware, result is exactly the same on 5.20 on H3 device with legacy kernel.

Here's the diff. I have no idea which options play any role in the problem. http://pastebin.com/rPYs5ugn

Works fine on Bananian with 3.4.112 kernel so it might be related to Armbian build. Or is related to systemd-udevd because Bananian does not use systemd, usbmuxd is started from plain udevd. # uname -a Linux bananapi 3.4.112-bananian #6 SMP PREEMPT Fri May 13 18:21:01 UTC 2016 armv7l GNU/Linux

Any hints where should I look for clues?

Dongle works almost oob on fresh 5.20 install so this must be related to my broken instance.

Unfortunately that did not change anything, will look for a dongle with another chip.

Tried to install firmware-ralink but: $ sudo apt install firmware-ralink Reading package lists... Done Building dependency tree Reading state information... Done The following NEW packages will be installed: firmware-ralink 0 upgraded, 1 newly installed, 0 to remove and 2 not upgraded. Need to get 22,2 kB of archives. After this operation, 103 kB of additional disk space will be used. Get:1 http://httpredir.debian.org/debian/ jessie/non-free firmware-ralink all 0.43 [22,2 kB] Fetched 22,2 kB in 0s (31,0 kB/s) Selecting previously unselected package firmware-ralink. (Reading database ... 155356 files and directories currently installed.) Preparing to unpack .../firmware-ralink_0.43_all.deb ... Unpacking firmware-ralink (0.43) ... dpkg: error processing archive /var/cache/apt/archives/firmware-ralink_0.43_all.deb (--unpack): trying to overwrite '/lib/firmware/rt3290.bin', which is also in package armbian-firmware 5.20 Errors were encountered while processing: /var/cache/apt/archives/firmware-ralink_0.43_all.deb E: Sub-process /usr/bin/dpkg returned an error code (1)

I can't get this dongle to work with 5.20. $ lsusb Bus 002 Device 002: ID 148f:5370 Ralink Technology, Corp. RT5370 Wireless Adapter Seems to be known one. $ lsmod Module Size Used by rt2800usb 13858 0 rt2800lib 42835 1 rt2800usb rt2x00usb 8227 1 rt2800usb rt2x00lib 34544 3 rt2x00usb,rt2800lib,rt2800usb mac80211 363732 3 rt2x00lib,rt2x00usb,rt2800lib mali 185786 5 ump 30408 1 mali mali_drm 2728 0 drm 178585 1 mali_drm Nothing wrong here? I guess so, wpa_supplicant is properly configured (ssid & psk), appropriate entry in /etc/network/interfaces but I still get wlan0 Link encap:Ethernet HWaddr 24:3c:20:0d:7a:46 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:16 errors:0 dropped:0 overruns:0 frame:0 TX packets:20 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:2932 (2.8 KiB) TX bytes:2820 (2.7 KiB) So the interface is up but not configured. /etc/network/interfaces: # Wireless adapter #1 allow-hotplug wlan0 iface wlan0 inet dhcp wpa-conf /etc/wpa_supplicant/wpa_supplicant.conf /etc/wpa_supplicant/wpa_supplicant.conf: ap_scan=1 ctrl_interface=DIR=/var/run/wpa_supplicant GROUP=netdev update_config=1 network={ ssid="myssid" psk=keyhash key_mgmt=WPA-PSK } syslog is flooded with messages like: Sep 29 10:07:59 localhost NetworkManager[602]: <error> [1475136479.267512] [supplicant-manager/nm-supplicant-interface.c:856] interface_add_cb(): (wlan0): error adding interface: wpa_supplicant couldn't grab this interface. Sep 29 10:07:59 localhost NetworkManager[602]: <info> (wlan0): supplicant interface state: starting -> down Sep 29 10:07:59 localhost wpa_supplicant[783]: ctrl_iface exists and seems to be in use - cannot override it Sep 29 10:07:59 localhost wpa_supplicant[783]: Delete '/run/wpa_supplicant/wlan0' manually if it is not used anymore Sep 29 10:07:59 localhost wpa_supplicant[783]: Failed to initialize control interface '/run/wpa_supplicant'.#012You may have another wpa_supplicant process already running or the file was#012left by an unclean termination of wpa_supplicant in which case you will need#012to manually remove this file before starting wpa_supplicant again. I must be missing something obvious, any hints? EDIT: it's 5.20.

Fortunately old bin subdirectory has been preserved as bin.old and I did not lost my precious script.bin

Tried on both 5.17 and 5.20, Debian and Ubuntu with legacy kernel. Debian has systemd 215, Ubuntu has 225 iirc. If I plug an iphone usbmuxd is started by systemd-udevd as directed by udev rule (stock one) and gets shut down when I unplug it, again how it's supposed to be. The problem is, it does not start if I plug any Apple device again. Restarting systemd-udevd service does not help at all, and when I reboot the board usbmuxd gets started when I plug iphone but again only once. 1st plug in/plug off in syslog: Sep 19 09:26:57 localhost kernel: [ 90.005422] ehci_irq: port change detect Sep 19 09:26:57 localhost kernel: [ 90.279798] usb 2-1: new high-speed USB device number 2 using sw-ehci Sep 19 09:26:57 localhost systemd[1]: Started Socket daemon for the usbmux protocol used by Apple devices. Sep 19 09:26:57 localhost usbmuxd[1070]: [09:26:57.922][3] usbmuxd v1.1.0 starting up Sep 19 09:26:57 localhost usbmuxd[1070]: [09:26:57.928][3] Successfully dropped privileges to 'usbmux' Sep 19 09:26:57 localhost usbmuxd[1070]: [09:26:57.945][3] Could not get old configuration descriptor for device 2-2: -5 Sep 19 09:26:57 localhost usbmuxd[1070]: [09:26:57.949][3] Connecting to new device on location 0x20002 as ID 1 Sep 19 09:26:57 localhost usbmuxd[1070]: [09:26:57.950][3] Initialization complete Sep 19 09:26:57 localhost usbmuxd[1070]: [09:26:57.950][3] Enabled exit on SIGUSR1 if no devices are attached. Start a new instance with "--exit" to trigger. Sep 19 09:26:57 localhost usbmuxd[1070]: [09:26:57.950][3] Connected to v2.0 device 1 on location 0x20002 with serial number 4d97a270aba9ba766a13989d9a90caa806d31b6b Sep 19 09:26:58 localhost kernel: [ 90.768031] ipheth 2-1:4.2: Apple iPhone USB Ethernet device attached Sep 19 09:26:58 localhost kernel: [ 90.775448] usbcore: registered new interface driver ipheth Sep 19 09:28:27 localhost kernel: [ 180.016143] ehci_irq: port change detect Sep 19 09:28:27 localhost kernel: [ 180.020243] usb 2-1: USB disconnect, device number 2 Sep 19 09:28:27 localhost usbmuxd[1070]: [09:28:27.212][3] Removed device 1 on location 0x20002 Sep 19 09:28:27 localhost kernel: [ 180.077995] ipheth 2-1:4.2: Apple iPhone USB Ethernet now disconnected Sep 19 09:28:27 localhost usbmuxd[1070]: [09:28:27.275][3] usbmuxd shutting down Sep 19 09:28:27 localhost usbmuxd[1070]: [09:28:27.376][3] Shutdown complete Then the same device plugged in: Sep 19 09:29:31 localhost kernel: [ 244.610407] ehci_irq: port change detect Sep 19 09:29:32 localhost kernel: [ 244.878100] usb 2-1: new high-speed USB device number 3 using sw-ehci And that's all. This works on amd64 Ubuntu 16.04 as expected. EDIT: usbmuxd is properly started by systemd-udevd on Jessie vanilla, so this seems related to kernel. EDIT2: board is Banana Pro.

See http://forum.armbian.com/index.php/topic/841-bananapro-lemaker-5in-lcd-legacy-kernel/?p=6418

You can mount BPro local storage using sshfs or any other FUSE based proto. I was able to run QtQuick based applications written in Python/PyQt but this required compiling "stuff" (MALI binary drivers and some X11 parts). Just follow https://linux-sunxi.org/Xorgand https://linux-sunxi.org/Mali_binary_driver

Thanks, that was the missing part.

In my /boot/config for 3.4.112 I have CONFIG_SUNXI_PWM=y yet there is no /sys/class/pwm-sunxi directory. Do I need to do anythig special to control onboard PWM? This is stock Armbian 5.16 with legacy kernel.

What is required & missing to support this display with vanilla kernel?

I found Armbian more contemporary and more friendly out of the box: systemd (yeah, I hate it as everybody else but this is just convenience if you had many other systems) bash as a default shell (zsh is not my beast) network configurations are easier to switch (symlinks in /etc/network) boot partition already mounted in filesystem more SBCs supported, makes hardware a commodity, the same distribution on many boards makes management easier (I have ~10 running 24/7 and another 3 - 4 occasionally) Other than that I did not found many reasons to switch from Bananian to Armbian, YMMV.

Of other good news, the all "compiling stuff" gave me working OpenGL ES as X11 OpenGL backend and I have QtQuick for the 1st time on ARM.

I'm sorry guys, I had one wrong value in FEX in lcd0_para section, lcd_if = 8 (should read 0). I don't know what I was thinking about. Now everything works OK, including touch interface with ft5x_ts. Thank you for your support.

Currently: U-Boot SPL 2015.10 (Dec 21 2015 - 09:41:49) and it works. It was working with 2015.07 as well.

With Bananian 15.08 and procedure from LeMaker wiki I got 5in LCD working, FEX changes is all that is required to have working display (without touch feature, of course). I took the whole sections from https://github.com/LeMaker/fex_configuration/blob/master/fex/banana_pro_5lcd.fex- ctp_para, ctp_list_para, disp_init and lcd0_para. Will debug more later on.

$ lsmod Module Size Used by cpufreq_userspace 3524 0 spidev 6366 0 ft5x_ts 44223 0 ap6211 624553 0 But I still see nothing on LCD. The difference between 7in and 5in is that 7in is LVDS and 5in is RGB but it's purely internal. I see no changes in display related sections in FEX config and boot args between my working device and the one I'm trying to set up. I'm going to give Bananian a try, maybe this way I get any working configuration. I see LeMaker site has 15.04, will try both 15.04 and 15.08.