[U-Boot] At91 mass storage device

[E Robertson]

On Thu, Mar 12, 2009 at 5:50 PM, Remy Bohmer <linux at bohmer.net> wrote:
> Hello,
>
> Please don't top post.
>
> A: Because it messes up the order in which people normally read text.
> Q: Why is top-posting such a bad thing?
> A: Top-posting.
> Q: What is the most annoying thing in e-mail?
> (see http://www.netmeister.org/news/learn2quote.html)

I'm aware of the top post irritant, nonetheless I employ some
discretion on short emails.
Don't worry, I wouldn't repeat it :).

> 2009/3/12 E Robertson <e.robertson.svg at gmail.com>:
>> I'm using git source from a few weeks back, sometime early part of last month.
>> Have those changes merged or are they somewhere else?
>
> Okay, you are quite up-to-date, so you are not using an ancient
> version :-), it worked in versions like 2008.10 (if PLLB is configured
> properly, bootstrap code did it wrong, see also
> cpu/arm926ejs/at91/usb.c of git), maybe there is some regression here.
> Unfortunately I do not have the hardware so I cannot verify it, or
> debug it...
>
> Besides, there are still USB sticks that do not work at all, have you
> tried a different stick already?
>
> Strange, in the logging I see that it only seems to communicate with
> the roothub only, it never communicates to the device at all, so my
> first guess would be the PLLB not running, but that was repaired
> somewhere early this year in a recent git...
> Maybe you can verify if it is running and all peripheral clocks are available?

I revisit these and it looked like my UHP power wasn't enabled.
I thought I did this in the Bootstrap code but apparently not:


    at91_sys_write(AT91_PMC_PCER, (1 << AT91SAM9263_ID_PIOA)
                                  | (1<<AT91SAM9263_ID_UHP));
    /* Enable VBus on UHP ports */
    at91_set_gpio_output(AT91_PIN_PA21, 1);
    at91_set_gpio_output(AT91_PIN_PA24, 1);


One thing I never understood is that after a power cycle with a usb
memory stick plugged in,
I always have to start and restart the usb probe to recognize the
storage device.
This happened in a previous version of u-boot (like 1.2.0) and it's
also the case with the current source.
Is their a reason for that? Perhaps an inadequate peripheral reset?