Re: Arm drivers: are device tree descriptions required for every device?

[Brook Milligan <brook%nmsu.edu@localhost>]

> On Apr 15, 2023, at 11:21 AM, Brian Buhrow <buhrow%nfbcal.org@localhost> wrote:
> 
> 	hello.  As a very simple starting point, I'd watch the output of vmstat -i to see if your
> USB chips are generating interupts when you add or remove devices from the USB bus.  The
> behavior you describe, combined with the dmesg output, makes me think interrupt routing for the
> USB chips isnt working, so the interupts generated when you add and remove devices from the USB
> bus aren't getting delivered.  Or, it's possible the interrupts aren't getting enabled by the
> USB chip driver (motg?)

Thanks for the idea.

I have done a variety of tests and there is a pending email with information, but for some reason it does not seem to have hit the mailing list yet.

As that post will describe when it eventually emerges is that (i) the USB controllers (motg) seem to be on irq18 and irq19; the latter should be the USB-A connector I am worrying about, (ii) the interrupt count reported by vmstat -i is only 3 for irq18 and does not change; irq19 is never listed at all.  This seems to support the idea that interrupts are to blame.

I have also booted the version of Debian flashed on the board, which talks fine to a wireless USB keyboard; the hardware appears to work fine.  

I have also booted NetBSD with all the Debian dtb device tree files; that seems to behave the same as. NetBSD with its own dtb files.

Does anyone have a BBB with NetBSD?  Can you share your dimesg output?  Does the USB-A port work for you?

If interrupts are the issue, how can this be debugged?

Thanks again for ideas and help.

Cheers,
Brook