On Mon, 7 Nov 2011, Sebastian Andrzej Siewior wrote:
> >> {
> >> - u32 i;
> >> -
> >> - if (dum->gadget.speed == USB_SPEED_SUPER) {
> >> - for (i = 0; i < DUMMY_ENDPOINTS; i++)
> >> - dum->ep[i].ep.max_streams = 0x10;
> >> + if (dum->gadget.speed == USB_SPEED_SUPER)
> >> dum->ep[0].ep.maxpacket = 9;
> >> - } else {
> >> - for (i = 0; i < DUMMY_ENDPOINTS; i++)
> >> - dum->ep[i].ep.max_streams = 0;
> >
> >It might be a good idea to keep this code -- just to be safe.
> Are you talking about an #if 0 block?
No, I meant the three lines above, where max_streams is set to 0 if the
emulated connection isn't SuperSpeed. Leave it in and let it run.
(Another thing I just noticed: i should be declared as an "int", not a
"u32".)
> >> static int dummy_pullup (struct usb_gadget *_gadget, int value)
> >> @@ -955,6 +948,7 @@ static void init_dummy_udc_hw(struct dummy *dum)
> >> ep->halted = ep->wedged = ep->already_seen =
> >> ep->setup_stage = 0;
> >> ep->ep.maxpacket = ~0;
> >> + ep->ep.max_streams = 0x10;
> >
> >Can we please just use a plain old "16" instead of "0x10"? It's not
> >like the max number of streams has some mystic binary significance. I
> >realize it was done this way in the original code, but that doesn't
> >mean you have to copy it blindly.
>
> Sure. However I'm not sure if we could change it to streams_supported. I
> saw only one real hardware so far so my view is very limited. It
> supported streams on all endpoints and the number of limited streams
> seems not to be limited.
I don't know about USB-3 hardware in general. But 16 streams seems
like a reasonable number to use for testing. You could make it a sysfs
attribute if you want, although that might be overkill.
Alan Stern
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
