I have an ATSAMD51 microcontroller integrated in a project-specific board which has also a xbee3 RF module. This xbee3 module is configured for Transparent Mode and the device type is 'router'. I can check this on the XCTU software.
Now, my goal is to install micropython on ATSAMD51 and then send commands to the xbee3 via UART since unfortunately I am not aware of any micropython library that can interact with the xbee. And just out of confusion, I don't want to use the xbee with micropython.
++++++++++++++ ++++++++++++++
++ ATSAMD51 ++ --UART--> ++ XBEE3 RF ++
++++++++++++++ ++++++++++++++
(running (transparent
micropython) mode)
I successfully installed the micropython on the ATSAMD51 and I managed to send commands to the xbee3 via UART. I know this works since when I do uart.write('1234') I can check on the XCTU console that the coordinator node receives this message.
However, If I try to send xbee3 AT commands, the uart.read() only gets 2 bytes at most.
Example:
screen /dev/tty.usbmodem0000000000001 115200 # enter the python REPL of the ATSAMD51
>>> from machine import UART, Pin
>>> uart = UART(5, rx=Pin('PB02'), tx=Pin('PB03'), baudrate=115200)
>>> uart
UART(5, baudrate=115200, bits=8, parity=None, stop=1, timeout=1, timeout_char=1, rxbuf=256, txbuf=256)
>>> uart.write('1234') # this actually transmits the message to the coordinator node
4
>>> uart.write('+++') # entering in command mode
3
>>> uart.read()
b'OK\r'
>>> uart.write('ATID\r')
5
>>> uart.read()
b'FO\r' # this should be FOO which is the device PAN ID
>>> uart.write('ATXX\r')
5
>>> uart.read()
b'ER\r' # this should be ERROR
Any command I send I always get only 2 characters + the \r (carriage return).
Thank you!
I've tried different baudrates and timeouts.
If I try to connect to the xbee directly from the computer via serial this works as expected. So I guess this is somehow related with the micropython UART configuration in the SAMD51?
The answer for the problem is here. A temporary fix is available here.
Related
We connected L89 GNSS module to processor via UART. I may running gpsmon and gpspipe commands on board but but not getting NMEA logs. But when we were connected L89 to processor via USB, that time I am getting NMEA logs on gpsmon and gpspipe, what is issue with UART?
I edit /etc/default/gpsd file.
# Default settings for gpsd.
# Please do not edit this file directly - use `dpkg-reconfigure gpsd' to
# change the options.
START_DAEMON="true" command to continue
GPSD_OPTIONS=""
DEVICES="/dev/ttyS1"
USBAUTO="false"
GPSD_SOCKET="/var/run/gpsd.sock"
but I am not getting NMEA data using gpsd...
I am running a simple SparkStreaming application, that consists in sending messages through a socket server to the SparkStreaming Context and printing them.
This is my code, which I am running in IntelliJ IDE:
SparkConf sparkConfiguration= new SparkConf().setAppName("DataAnalysis").setMaster("spark://IP:7077");
JavaStreamingContext sparkStrContext=new JavaStreamingContext(sparkConfiguration, Durations.seconds(1));
JavaReceiverInputDStream<String> receiveData=sparkStrContext.socketTextStream("localhost",5554);
I am running this application in a standalone cluster mode, with one worker (an Ubuntu VM) and a master (my Windows host).
This is the problem: When I run the application, I see that it successfully connected to the master, but it doesn't print any lines:
it just stays this way permanently.
If I go to the Spark UI, I find that the SparkStreaming Context is receiving inputs, but they are not being processed:
Can someone help me please? Thank you so much.
You need to perform below.
sparkStrContext.start(); // Start the computation
sparkStrContext.awaitTermination(); // Wait for the computation to terminate
Once you do this , you need to post the messages at port 5554 , for this you will first need to run Netcat (a small utility found in most Unix-like systems) as a data server by using and start pushing the stream.
For example , you need to do like below.
TERMINAL 1:
# Running Netcat
$ nc -lk 5554
hello world
TERMINAL 2: RUNNING Your streaming program
-------------------------------------------
Time: 1357008430000 ms
-------------------------------------------
hello world
...
...
You can check similar example here
I'm trying to use the WinUsb.sys driver that comes with Windows (including Windows 10 IoT of Raspberry Pi 2). Using devcon.exe I can see that the USB I'm trying to use is connected (it is named USB\VID_1234&PID_ABCD\5&3753427A&0&4), but I don't know how to force it to use the WinUsb.sys driver.
I found some instructions on microsoft.com but that seems to be for a standard Windows installation where you have Device Manager available (which I don't have on IoT). The INF file example on this page also refers to a CAT-file, which I assume this is some sort of driver signature, and I don't know how to generate this (or if I even need to). There's also a reference to Windows NT (Signature = "$Windows NT$") and I don't know if that needs to be changed for IoT or not (or if anything else needs to be changed for IoT for that matter).
So, using devcon.exe and some sort of INF file, how can I get Windows IoT to use WinUsb.sys as a driver for the USB device I'm attaching?
After a lot of trial and error I finally got it working. Here's a complete INF-file for future reference:
; WinUSB installation file for USB device
[Version]
Signature = "$Windows NT$"
Class = USBDevice
ClassGUID = {88BAE032-5A81-49f0-BC3D-A4FF138216D6}
Provider = %ManufacturerName%
CatalogFile = WinUSBInstallation.cat
DriverVer=09/04/2012,13.54.20.543
; ========== Manufacturer/Models sections ===========
[Manufacturer]
%ManufacturerName%=Standard,NTarm
[Standard.NTarm]
%DeviceName% =USB_Install, USB\VID_1234&PID_ABCD
; ========== Class definition ===========
[ClassInstall32]
AddReg = ClassInstall_AddReg
[ClassInstall_AddReg]
HKR,,,,%ClassName%
HKR,,NoInstallClass,,1
HKR,,IconPath,%REG_MULTI_SZ%,"%systemroot%\system32\setupapi.dll,-20"
HKR,,LowerLogoVersion,,5.2
; =================== Installation ===================
[USB_Install]
Include = winusb.inf
Needs = WINUSB.NT
[USB_Install.Services]
Include =winusb.inf
Needs = WINUSB.NT.Services
[USB_Install.HW]
AddReg=Dev_AddReg
[Dev_AddReg]
HKR,,DeviceInterfaceGUIDs,0x10000,"{ec55ee47-5758-4378-926b-68ed0aec8170}"
; =================== Strings ===================
[Strings]
ManufacturerName="The name of the company producing your device"
ClassName="Universal Serial Bus devices"
DeviceName="The name of your device"
REG_MULTI_SZ = 0x00010000
Replace the Vendor ID (VID) and Product ID (PID) in [Standard.NTarm] with the corresponding VID and PID of the USB you're adding.
Finally replace the ManufacturerName and DeviceName near the bottom with the correct info for your device.
Put this file somewhere on the Raspberry Pi 2, using either SMB or FTP.
SSH or PowerShell to the Raspberry Pi 2 and go to the folder where you put the INF file.
Run the following command: devcon dp_add .\<name of your INF file>
You should see the following result: Driver package 'oem0.inf' added.
Finally restart the RP2 (shutdown -r -t 0 from SSH/PowerShell).
When the RP2 gets back up your device should be listed under "Connected Devices" on the default startup app, and you should now be able to use functionality from Windows.Devices.Usb to communicate with the USB device.
Summary
My aim is to control the GPIO pins in Peppermint 4 Linux (Kernel version 3.8.0) on an Intel motherboard (NM70 chipset with C1037U processor).
I'm debugging issues I'm having using the sysfs interface and am trying to understand the conditions where /sys/kernel/debug/gpio would be empty?
When attempting to export pins 0 to 255 by
echo XX > /sys/class/gpio/export
for XX from 0 to 255, I get the following error message
echo: write error: No such device
Under what conditions would /sys/kernel/debug/gpio be empty?
Background
Motherboard: Intel with NM70 chipset
Processor: C1037U processor
OS: Peppermint 4 Linux
Kernel version: 3.8.0
GPIO interface: sysfs
I'm attempting to use the sysfs interface, which allows GPIO pins to be accessed from userspace through the filesystem.
I’ve successfully followed the "Alternate Build Method: The Old-Fashioned Debian Way" section of https://help.ubuntu.com/community/Kernel/Compile to recompile the kernel in order to expose GPIO access in user space and to turn on debug mode for GPIO:
Once the new kernel was compiled, I was able to see the GPIO folder in /sys/class/gpio for the first time. Then, in theory, it should be a case of being able to turn GPIO ports ON/OFF by writing to the filesystem. This approach is outlined at http://falsinsoft.blogspot.co.uk/2012/11/access-gpio-from-linux-user-space.html.
When attempting to export pins 0 to 255 by
echo XX > /sys/class/gpio/export
for XX from 0 to 255, I get the following error message
echo: write error: No such device
When attempting to export pins outside the range 0 to 255 by
echo XX > /sys/class/gpio/export
I get the following error message
echo: write error: Invalid argument
The tutorial suggests this could be because the GPIO ports are reserved for another program and that, if so, the debug file (/sys/kernel/debug/gpio) would be able to show where they are reserved.
However, /sys/kernel/debug/gpio is empty.
I can see and control the GPIO pins in the BIOS (change pins to be input or output HIGH/LOW).
Related questions
writing to /sys/class/gpio/export failing
Enable pullup GPIO
/sys/kernel/debug/gpio will be empty if there no GPIO device registered (warning: when I say GPIO device, I don't mean the piece of hardware, but rather the kernel representation of it).
So, these GPIO devices are registered at runtime by the kernel and associated to a specific GPIO device driver.
In turn, the GPIO device driver is selected and associated to a given device, because it is the one declaring compatibility with said GPIO device.
E.g., the kernel would match on PCI vendor and product ID, and probe a GPIO driver that claims support for that PCI vendor/product. When the GPIO driver is probed, it typically registers the GPIO device instance.
Finally, that registered GPIO device is what provides GPIOs shown in /sys/kernel/debug/gpio.
The above is part of the so-called "Device Driver model" in Linux. Although it's a bit outdated, you can read [1].
Now, let's see what GPIO driver you need to select for your NM70 chipset. Wikipedia says that the chipset codename is "Panther Point-M" [2]. With some luck, the lpc_ich driver could support it. You would have to build your kernel with CONFIG_LPC_ICH=y.
Alternatively, if your GPIOs are provided by a PCI device, you could use lspci to get the IDs, and then grep in the kernel sources for those IDs.
[1] http://www.oreilly.com/openbook/linuxdrive3/book/ch14.pdf
[2] https://en.wikipedia.org/wiki/List_of_Intel_chipsets
I'm trying to program (upload new messages to) a scrolling LED badge via python on Linux instead of via the windows exe. I've got a capture of a successful upload via the windows app that I'll use as a template, but at the moment I'm having trouble writing anything at all to the device.
When I run my script, I get an assertion error, which I assume means that no OUT endpoints have been found. lsusb seems to corroborate this, but I could be mistaken. This is basically straight from the pyusb tutorial with alternate_setting removed from intf because it was causing a pipe error and is apparently not necessary.
intf = usb.util.find_descriptor(cfg, bInterfaceNumber = interface_number,)
ep = usb.util.find_descriptor(
intf,
# match the first OUT endpoint
custom_match = \
lambda e: \
usb.util.endpoint_direction(e.bEndpointAddress) == \
usb.util.ENDPOINT_OUT
)
assert ep is not None # < assertion error
In case it helps, lsusb -vv for the LED badge: http://bpaste.net/show/biixy6e38A71WlCw7euU/
Any guidance would be much appreciated - I'm entirely new to USB protocol, and not great at python either.