Re: PR 60404: Beaglebone Black: nonfunctional usb wifi

Brook Milligan <[email protected]> wrote:
> The attached patch (see below) fixes the non-functional USB controller
> on the Beaglebone Black.  With this patch, mass storage devices seem to
> work fine, and wifi devices are detected by both the run and urtwn
> drivers.
>
> At the moment, this patch is really verbose, because I am reporting
> various bits along the way.  Thus, there are obvious cleanups to be
> made.
>
> More importantly, however, I would very much appreciate feedback on how
> to improve the functional code parts.  For example, it seems that it
> should be possible to access some of the relevant contants from the
> device tree, but I am not seeing how that can be done.  The BBB memory
> map separates control registers from the main device registers, so the
> same mapped memory cannot be used for both, and the device tree seems
> not to have a way to extract the base address of the control module.  I
> may, however, be not understanding or there may be other ways to access
> these registers or to differentiate between the instances of the driver
> attachments.  All code improvements welcome.

I think the expected way to handle this is to create a separate driver
for the PHY, it would then map its own register set as defined in
the device tree.

The USB driver would then call fdtbus_phy_get() with the appropriate
node name then fdtbus_phy_enable().