Hi Nikolai,

quoted

quoted

The first two patches are bug fixes found in the process. I am hoping
for some test feedback on those. The third patch is the main one.

I've tested the last patch and it seems to work as required. I did a series
of generic<->specific driver rebindings both ways in a loop and it works.

So,
Tested-by: Nikolai Kondrashov <redacted>

Great, thanks!

I'm not sure, though, if one thing works as it should: the quirks are not
reset when changing the driver. So, quirks set by one driver may affect
behavior of another. I don't understand full quirks usage and ownership yet,
so can't argue if this is right or wrong.

The common quirks are a device property, so this is fine. There are
some drivers with quirks too, but they handle those themselves.

Regarding the code, could it be the time to fix naming of report
descriptor-handling functions to improve distinction from report-handling
ones? I'm attaching a patch applying on top of Henrik's changes which should
illustrate the proposal.

The changes you propose force changes to out-of-tree-drivers, without
any obvious benefit. Also, there is already a large set of patches on
review based on this one. So although I agree that some renames and
movements could clarify things, the content and timing will need some
more thought.

Thanks,
Henrik