I’m building TWIN on my DevTerm + R-01. It makes use of a mouse, so to make it work in console mode I need
gpm running. It says it can’t find
/dev/input/mice when it tries to start. I’ve never really used
gpm before. But when I’ve seen it launch in the past its been painless. I suspect this has to do with the way the trackball is wired in.
I thought I’d reach out and see if anyone has this working before I go for the deep-dive.
A follow up for anyone looking or caring. I found that Xorg is using
/dev/input/event5 for the mouse device. It doesn’t mention what protocol its using. So I haven’t been able to get
gpm to behave yet. Perhaps the DevTerm keyboard+mouse is doing its own thing, since
gpm fails to understand the input with the most common mice types.
Hopefully I’ll get some time to nail it down…
I must be the only one trying to run the DevTerm outside of X…
I’ve found the kernel does not have the
input mouse (CONFIG_INPUT_MOUSE) driver enabled.
gpm is not apparently compatible with the raw event interface. Or I just haven’t found the documentation for it. I’ll probably have to look into the source or figure out how to rebuild the kernel…
I’ve been back and forth on this - like you, super frustrated by the lack of
gpm on the R01, but it works on the CM4, which is a lot more promising.
Good sleuthing on what could be behind the problem on the R01 - I need to try another kernel build or have another go at building a new OS image from scratch.
Have you applied the alternative keyboard firmware(s), by the way? I’m using @yatli firmware from somewhere in this thread.
Have you applied the alternative keyboard firmware(s)
No, I have not. Sooo many things… Sooo little time. I have to figure out how to get access to the physical I2C bus so I can put in a DS3231 to keep it on track with time w/o iNet. That will get me the minimum hardware functionality I need to be able to use this on a daily basis. Then I have to rewrite a few programs in C++ since FreePascal (FPC) hasn’t got on board with RISC-V yet and I originally wrote them in FPC for my PocketCHIP.
Much to do.
Is there some compelling reason I should try to upgrade the firmware in my keyboard?
I think when I solve my break out issue I might take a whack at patching the kernel sources together and enabling the mouse support. That should be another large time-suck. It will be nice if the driver can be supplied as a module. But I need to see the menu to know if that’s an option. I wonder how many days it will take to compile the whole kernel…
For the record, I just installed
gpm and it run fine with the default settings here (A06), with
Thanks for this. This helps to confirm the issue is the RISC-V kernel is missing the appropriate driver. Maybe I can get some time this weekend to see about building a module…