Ihor Dutchak
Ihor Dutchak
How long does it take to perform the write on platforms where it doesn't fail? (Or with 0.8.0 implementation?) If it takes longer than 1sec, it means to properly fix...
> Do you have the link to the line in the kernel? https://github.com/libusb/hidapi/issues/493#issuecomment-1369927578 > I would vote for adding hid_write_timeout to the api and implementing it for all platforms... Again,...
> The libusb backend has the same issue, can we add a solution for that too? That is my plan, yes. > documented as a "best effort" call and no...
To be resolved when libusb part is done too
> Thanks for @Youw your review:). > > Then, What is your opinion about this? > > > Currently, some stuff like the Report Descriptor parser and error registry routine...
> Oops. I forget we have udev-devd stuff. Maybe we should use udev also? I lost context here. What for? Seem like you have all the functionality implemented already. Aren't...
> > > Oops. I forget we have udev-devd stuff. Maybe we should use udev also? > > > > > > I lost context here. What for? Seem like...
I think one of the easiet ways is to switch to hidraw as a default when HIDAPI bumps its version to 1.0. But we could also consider some alternatives.
> > And I extended the functionality of the libusb method to be able to calculate the maximum output and feature report sizes as well. This has no functional use...
I lost track on this one a bit. Can someone remind where do we stand right now? Except for failing CI - are there active change requests/fixes?