Re: [PATCH v4 1/4] xhci: Add host support for eUSB2 double isochronous bandwidth devices

[Sakari Ailus <sakari.ailus@linux.intel.com>]

On Wed, Aug 20, 2025 at 10:43:04AM +0200, Michał Pecio wrote:
> On Wed, 20 Aug 2025 11:24:24 +0300, Sakari Ailus wrote:
> > Hi Michał,
> > 
> > Thanks for the review.
> > 
> > On Mon, Aug 18, 2025 at 11:50:16AM +0200, Michał Pecio wrote:
> > > > @@ -1351,8 +1369,18 @@ static u32 xhci_get_endpoint_max_burst(struct usb_device *udev,
> > > >  
> > > >  	if (udev->speed == USB_SPEED_HIGH &&
> > > >  	    (usb_endpoint_xfer_isoc(&ep->desc) ||
> > > > -	     usb_endpoint_xfer_int(&ep->desc)))
> > > > +	     usb_endpoint_xfer_int(&ep->desc))) {
> > > > +		/*
> > > > +		 * eUSB2 double isoc bw endpoints max packet field service
> > > > +		 * opportunity bits 12:11 are not valid, so set the ctx burst to
> > > > +		 * max service opportunity "2" as these eps support transferring
> > > > +		 * over 3072 bytes per interval
> > > > +		 */  
> > > 
> > > I think a shorter comment would suffice: eUSB2 BWD uses fixed burst
> > > size and max packets bits 12:11 are invalid.  
> > 
> > I'll use this:
> > 
> > +                * eUSB2 double isochronous BW ECN uses fixed burst size and max
> > +                * packets bits 12:11 are invalid.
> >
> 
> Fine. And by the way, it may look like my comment was overly pedantic,
> but I though that mentioning "more than 3072 bytes per interval" not
> only isn't necessary here, but it adds confusion. It was something that
> would be more relevant to 'mult' than 'burst.

Further I corrected the ECN title, the result being:

+                * USB 2 Isochronous Double IN Bandwidth ECN uses fixed burst
+                * size and max packets bits 12:11 are invalid.

There are now quite a few other changes for v5 as well, I'll post it
soonish.

-- 
Sakari Ailus