[U-Boot] [PATCH] usb: Move determination of TT hub address/port into seperate function

[Hans de Goede]

Hi,

On 19-12-15 18:17, Stefan Bruens wrote:
> On Friday 18 December 2015 11:00:19 Hans de Goede wrote:
>> Hi,
>>
>> On 18-12-15 02:11, Stefan Brüns wrote:
>>> Start split and complete split tokens need the hub address and the
>>> downstream port of the first HS hub (device view).
>>>
>>> The core of the function was duplicated in both host/ehci_hcd and
>>> musb-new/usb-compat.h.
>>>
>>> Signed-off-by: Stefan Brüns <stefan.bruens at rwth-aachen.de>
>>
>> Thanks for working on this, I think I've spotted one small bug though, see
>> comments inline.
>
> Ah, sorry for not mentioning this.
>
> Yes, this changes the musb code, but this was on purpose.

IMHO that is not how this should be dealt with, if what we're currently
doing for musb is wrong (I do not know if it is), then fixing this does
not belong in a patch which is only moving code around. Such a fix
clearly belongs in a separate follow-up patch, and then you can use
everything you've just typed:

 > Rationale:
>
> The ifdef'ed Linux kernel code uses the 1 based port number, whereas U-Boot
> puts a 0 based port number into the register. The reason the 0 based port
> number apparently works can probably be taken from the USB 2.0 spec:
>
> 8.4.2.2 Start-Split Transaction Token
> ... The host must correctly set the port field for single and multiple TT hub
> implementations. A single TT hub implementation *may ignore* the port field.
>
> Actually, as far as I unterstand, a multi TT hub defaults to single TT
> (bAlternateSetting: 0) until switched via SetInterface, so even "port 42"
> would work.
>
> I have somewhat verified this assumption by hardcoding the port number and
> split transactions still work. Used hubs are the RPi onboard SMC9514 and an
> external "05e3:0608 Genesys Logic, Inc. USB-2.0 4-Port HUB". The former is a
> multi TT hub, the latter single TT only.

As a commit msg for such a separate patch. As a general rule of thumb never
make 2 separate / independent changes in a single commit just because they
happen to touch overlapping lines of code.

> I have no board with musb, but I think a 0 based port number is wrong.

I've a board with musb, and I plan to test your patch as soon as a new version
which re-adds the -1 for musb is posted. If you decided to do a second patch to
remove the -1, I can then test that afterwards, with as an added bonus that if
things break I can also tell you if it is the new shared helper which breaks
things, or the removing of the -1 :)

Regards,

Hans