Hi List!
I'm currently using gbridge / tcpip with the IID1-simple-gpio-module.mnfs example to hook up a fake module that just responds to socket I/O.
A couple of small changes were required, but everything seems to be parsed, bundles, cports are all created.
However, for some reason, greybus does not seem to automatically bind the device to the gb-gpio driver, nor does it probe the device. I only see pings after the svc has inserted my fake module and handshake is complete.
Is anyone aware if a separate step that is required to bind the device to the driver? I've tried
sudo sh -c 'echo -n 1-2.2.1 > /sys/bus/gbphy/drivers/gpio/bind'
but that gives me -ENODEV
Any pointers would be great!
Thanks,
C
On Thu, Aug 15, 2019 at 12:49 PM Christopher Friedt chrisfriedt@gmail.com wrote:
I'm currently using gbridge / tcpip with the IID1-simple-gpio-module.mnfs example to hook up a fake module that
I see what the problem is - in manifest/examples/IID1-simple-gpio-module.mnfs, the bundle class should have been 0x0a (bridged phy) not 0x02 (reserved).
On Thu, Aug 15, 2019 at 12:49:51PM -0400, Christopher Friedt wrote:
Hi List!
I'm currently using gbridge / tcpip with the IID1-simple-gpio-module.mnfs example to hook up a fake module that just responds to socket I/O.
A couple of small changes were required, but everything seems to be parsed, bundles, cports are all created.
However, for some reason, greybus does not seem to automatically bind the device to the gb-gpio driver, nor does it probe the device. I only see pings after the svc has inserted my fake module and handshake is complete.
Is anyone aware if a separate step that is required to bind the device to the driver? I've tried
sudo sh -c 'echo -n 1-2.2.1 > /sys/bus/gbphy/drivers/gpio/bind'
but that gives me -ENODEV
It should "just work" if the manifest signatures match the what the greybus gpio driver expects.
Do you get any logging messages in your kernel log when you do this?
thanks,
greg k-h
On Thu, Aug 15, 2019 at 3:39 PM Greg KH greg@kroah.com wrote:
On Thu, Aug 15, 2019 at 12:49:51PM -0400, Christopher Friedt wrote:
Is anyone aware if a separate step that is required to bind the device to the driver? I've tried
sudo sh -c 'echo -n 1-2.2.1 > /sys/bus/gbphy/drivers/gpio/bind'
but that gives me -ENODEV
It should "just work" if the manifest signatures match the what the greybus gpio driver expects.
Do you get any logging messages in your kernel log when you do this?
The bundle class in the manifest was incorrect. Should have been 0x0a (bridged phy) - GPIO protocol is a go ;-)
Hi Christopher, Just getting back from some time off and catching up with the greybus movements. I was the guy that have implemented some of the protocols and maintain/used the gbsim more.
On Thu 15 Aug 2019 at 20:46, Christopher Friedt wrote:
On Thu, Aug 15, 2019 at 3:39 PM Greg KH greg@kroah.com wrote:
On Thu, Aug 15, 2019 at 12:49:51PM -0400, Christopher Friedt wrote:
Is anyone aware if a separate step that is required to bind the device to the driver? I've tried
sudo sh -c 'echo -n 1-2.2.1 > /sys/bus/gbphy/drivers/gpio/bind'
but that gives me -ENODEV
It should "just work" if the manifest signatures match the what the greybus gpio driver expects.
Do you get any logging messages in your kernel log when you do this?
The bundle class in the manifest was incorrect. Should have been 0x0a (bridged phy) - GPIO protocol is a go ;-)
Great to see someone trying to use it, let me know if you need anything more and I will try to have more attention to this list.
Cheers, Rui
-
Christopher Friedt -
Greg KH -
Rui Miguel Silva