From: Sarah Sharp > On Wed, Jan 22, 2014 at 03:22:53PM -0800, walt wrote: > > On 01/22/2014 12:56 PM, Sarah Sharp wrote: > > > Walt, can you turn on xHCI debugging and look for whether the NEC host > > > that worked with David's patch is a 1.0 host? You'll see a line like: > > > > > > // @%p = 0x%x (CAPLENGTH AND HCIVERSION) > > > > Hi Sarah. This is from the NEC host controller, *not* the ASMedia: > > > > [ 9.089178] xhci_hcd 0000:04:00.0: // @ffffc90004f88000 = 0x960020 (CAPLENGTH AND HCIVERSION) > > Ok, so that's also a 0.96 host, but it's not impacted by the no-op TRBs. > Still, I think disabling the recently added code for everything but 1.0 > hosts makes sense. Actually it might be best to use a urb flag to indicate that the request isn't 'aligned', and only add the NOP for unaligned transfers. There is already a usb device flag usb_device_no_sg_constraint() to indicate that scatter gather transfers may be misaligned. The only code that looks at it is the ax88179_178a ethernet driver. There is, of course, no definition of what the 'constraint' is, but I think all the other requests are in 4k+ blocks. The only problem is that this requires simultaneous updates of the xhci and usbnet code. David -- 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
