On Fri, 18 Jul 2014, Amit Virdi wrote:
> On 7/17/2014 8:25 PM, Alan Stern wrote:
> > I can't say this is actually wrong, but have you ever encountered a
> > situation where this would be needed? How often does anyone need to do
> > a multi-packet transfer over an interrupt endpoint?
>
> Honestly, I haven't found any such real device yet. I did this change
> while I was going through the code when adding support for interrupt
> transfers in gadget zero. I'm a no expert, but the spec says it should
> be supported so this code should be added.
I just remembered, the HID spec requires that all reports (except the
longest) end with a short packet. And since output reports are allowed
to be sent over an Interrupt-OUT endpoint, we have to accept the
URB_ZERO_PACKET flag for interrupt URBs.
> However, I messed up a little in this patch. It should have been
> ---
> case USB_ENDPOINT_XFER_ISOC:
> allowed |= URB_ISO_ASAP;
> break;
> + case USB_ENDPOINT_XFER_INT:
> + if (is_out)
> + allowed |= URB_ZERO_PACKET;
> + else
> + allowed |= URB_SHORT_NOT_OK;
> + break;
> }
> allowed &= urb->transfer_flags;
>
> ---
> Otherwise, it sets zero packet flag for control out transfers too.
>
> I'll send a V2 with this change if you agree to setting of the zero
> packet flag for interrupt transfers.
How about this change instead?
URB_FREE_BUFFER);
switch (xfertype) {
case USB_ENDPOINT_XFER_BULK:
+ case USB_ENDPOINT_XFER_INT:
if (is_out)
allowed |= URB_ZERO_PACKET;
/* FALLTHROUGH */
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
