@DumTux said in CAN Bus: using MCP2515 with Omega2:
Next, I enabled CAN bus with loopback mode.
ip link set can0 down ip link set can0 up type can bitrate 125000 loopback on cansend can0 01a#112233No CAN packet waveform I can see!
that's the usual behaviour in loopback mode. Packets are not transmitted to the transceiver.
I use the loopback mode for testing purpose. If this works (received the same data as send) everything is fine with the communication to the CAN controller.
@luz said in CAN Bus: using MCP2515 with Omega2:
I made ugly patches to fix that back then, I can dig them out if someone wants to use them, but I can't exactly recommend (it was kind of an emergency hack for an urgent project).
I would be interested in this patch
As far as I know, the issue should be fixed in the 22.03 version of
gpio-mt7621.c, as the entire driver has been reworked to align with other modern GPIO drivers and handles the banks differently.
I switched to 22.03 but the problem is still present
@DumTux
AFAIK the main problem is related to the wrong GPIO/IRQ mapping:
https://forum.openwrt.org/t/can-mcp2515-gpio-irq-problem/92879/3
Maybe luz or Lazar could give a hint ?
Hi Chris,
the used PIC18F2xk80 only handle CAN2.0 and not CAN FD. If you want to use Omega2+ the USB port would be the best way to connect a MCU of your choice .
The fastest way to get a working Linux CAN FD board is to use a SBC/CPU with CAN FD IP or something like BananaPi Zero with MCP2518/TCAN4550.
Regards
Gerd
AFAIK Ramips/Mediateks SPI device is half duplex and it will not work without modifying the SPI driver or the MCP2515 kernel module to control a MCP2515. I've done a dirty hack for a Carambola1 board (RT5350) years ago. The Carambola1 uses non SPI flash (NAND) so the patch doesn't harm the flash access. But the Omega2 uses the SPI flash - bad luck.
The other way would be using bitbanged SPI. It worked better than expected.
As of today I'm using a PIC connected via ttyS1:
http://community.onion.io/topic/1853/connect-omega2-to-canbus
The PIC is programmed inband from the Omega2.
Regards
Gerd
@cory-callcott
AFAIK none of the cheap USB2Serial devices is able to transfer 9bit. You can emulate it with mark/space parity but it takes a lot of time switching between the modes and hurts badly the effective transfer rate.
Hi,
the WiFi issues are being solved with newer versions of OpenwWRT. The Gleisbox is now running fine.
Here some pictures:
Regards
Gerd
@Michael-Bock said in Connect Omega2 to CanBus:
@Gerhard-Bertelsmann Thanks for your answer ! Do I understand it correctly that this board is like a dock for the Omega2+ and through a serial communication lib will read/write the CAN BUS ?
yes, that's right. It's using the SLCAN API which could be used in conjunction with SocketCAN. The firmware was mainly developed by Darron Broad
The PIC firmware is highly optimized so that there are lots of free cycles even under heavy load on the CAN-Bus.
Can you share more details of it and is it possible to buy it ?
Please have a look at:
https://github.com/GBert/misc/tree/master/srse2
It's working but the whole project is on hold due to the WiFi issues ...
Sorry, it isn't sold but I could share a PCB if you want. Please contact me by email
You have different choices for CAN on Omega2. The obvious way is using a MCP2515 connected to SPI. I'm not 100% sure if you can use hardware accelerated SPI on Omega2. The necessary pins seems to be connected to the pin header but Ralink/Mediatek SPI is normally half duplex and the MCP2515 kernel modul expect full duplex.
Another problem is the latency of the SPI Linux Framework. On high CANBus load packets might getting out of order or being lost.
For testing I've written a combined MCP2515-SPI (GPIO-Bitbanged) driver:
https://github.com/GBert/openwrt-misc/tree/master/mcp2515-banged
The results were better than expexcted - no packet loss on a saturated 250kbit CANBus.
As of today I'm prefering using a CAN2Serail converter based on a cheap PIC18F25K80. The Firmware was writen by Darron Broad. The serial to CAN conversion is amazing fast and is only limited by the serial speed. The good thing is the PIC could be flashed in circuit and the software, including assembler is part of an OpenWRT/Lede feed. If you are interested:
https://github.com/GBert/misc/tree/master/srse2
But: The easiest way is to use a USB based CAN Adapter, of course 
@luz Many thx for your updates. Great News. However I didn't get it working so far albeit I hadn't much time to check my setup. I going to take your Settings (network&wireless) and see how it goes ...
I measured a raw "toogle speed" of 7.69MHz:
https://github.com/GBert/openwrt-misc/tree/master/gpio-test
For your reference:
CPU | toggle speed
----------------|-------------
RT3052@320MHz | 6.67 MHz
AR9331@400MHz | 7.69 MHz
AR9341@533MHz | 7.69 MHz
MT7688@580MHz | 7.69 MHz
The toggle speed seems to be independent from the CPU speed.
@luz , @james-lee
thanks for your replies.
It's really sad that only the blob wifi driver is stable. I've seen a 4.4.50 mtk-wifi in the wild (Widora). Did somebody testetd it ?
I would like to use the OpenWRT/Lede SDK because most of my work is based on it. I could live with a blob for a while (and with a Kernel 4.4.50) if the blob driver could be cleanly integrated in Lede build environment ...
Hi,
I have a problem setting up a stock Lede Image (Reboot (SNAPSHOT, r3866-322a722)) as WiFi client. I getting follwoing log (partly copied):
[ 73.169182] wlan0: authenticate with c0:25:06:a2:ab:58
[ 73.193115] wlan0: send auth to c0:25:06:a2:ab:58 (try 1/3)
[ 73.208673] wlan0: authenticated
[ 73.218475] wlan0: associate with c0:25:06:a2:ab:58 (try 1/3)
[ 73.285887] wlan0: RX AssocResp from c0:25:06:a2:ab:58 (capab=0x431 status=0 aid=3)
[ 73.296354] wlan0: associated
[ 79.388765] wlan0: disassociated from c0:25:06:a2:ab:58 (Reason: 2)
[ 80.429614] wlan0: authenticate with c0:25:06:a2:ab:58
[ 80.453474] wlan0: send auth to c0:25:06:a2:ab:58 (try 1/3)
[ 81.068250] wlan0: send auth to c0:25:06:a2:ab:58 (try 2/3)
[ 82.068227] wlan0: send auth to c0:25:06:a2:ab:58 (try 3/3)
[ 83.068230] wlan0: authentication with c0:25:06:a2:ab:58 timed out
[ 88.089179] wlan0: authenticate with c0:25:06:a2:ab:58
[ 88.113038] wlan0: send auth to c0:25:06:a2:ab:58 (try 1/3)
[ 88.127131] wlan0: authenticated
[ 88.138342] wlan0: associate with c0:25:06:a2:ab:58 (try 1/3)
[ 88.207698] wlan0: RX AssocResp from c0:25:06:a2:ab:58 (capab=0x431 status=0 aid=3)
[ 88.218149] wlan0: associated
[ 94.069667] wlan0: authenticate with c0:25:06:a2:ab:58
[ 94.089164] wlan0: send auth to c0:25:06:a2:ab:58 (try 1/3)
[ 94.102877] wlan0: authenticated
[ 94.108454] wlan0: associate with c0:25:06:a2:ab:58 (try 1/3)
[ 94.191309] wlan0: RX AssocResp from c0:25:06:a2:ab:58 (capab=0x431 status=0 aid=3)
[ 94.201786] wlan0: associated
[ 98.196571] wlan0: deauthenticating from c0:25:06:a2:ab:58 by local choice (Reason: 3=DEAUTH_LEAVING)
The Wifi associates with the base station and disconnects again. With a Omega2 Image it does work. It there a difference in the WiFi driver ?
Here my config:
root@LEDE:/etc/config# cat wireless
config wifi-device 'radio0'
option type 'mac80211'
option channel '11'
option hwmode '11g'
option path 'platform/10300000.wmac'
option htmode 'HT20'
option country 'DE'
config wifi-iface 'default_radio0'
option device 'radio0'
option network 'wlan'
option ssid 'Fischernetz'
option mode 'sta'
option encryption 'psk2'
option key 'geheim'
root@LEDE:/etc/config# cat network
config interface 'loopback'
option ifname 'lo'
option proto 'static'
option ipaddr '127.0.0.1'
option netmask '255.0.0.0'
config globals 'globals'
option ula_prefix 'fd8e:2845:f08a::/48'
config interface 'lan'
option type 'bridge'
option ifname 'eth0'
option proto 'dhcp'
option ip6assign '60'
config device 'lan_dev'
option name 'eth0'
option macaddr '40:a3:6b:c0:75:aa'
config interface 'wlan'
option ifname 'wlan0'
option proto 'dhcp'
config switch
option name 'switch0'
option reset '1'
option enable_vlan '0'
root@LEDE:/# dmesg | grep mt76_wmac
[ 10.243057] mt76_wmac 10300000.wmac: ASIC revision: 76280001
[ 12.292223] mt76_wmac 10300000.wmac: Firmware Version: _e2_mp
[ 12.298076] mt76_wmac 10300000.wmac: Build Time: 20150211175503
Regards
Gerd