On Tue, Jul 20, 2010 at 01:22:00PM -0700, Greg KH wrote: > On Tue, Jul 20, 2010 at 12:06:37PM -0700, Sarah Sharp wrote: > > On Tue, Jul 20, 2010 at 10:34:19AM -0700, Greg KH wrote: > > > On Tue, Jul 20, 2010 at 10:58:47AM -0500, Subs wrote: > > > > Dear all, > > > > > > > > We are facing a problem with one of the legacy mouse that we have. > > > > The mouse is expecting the Get_Descriptor, while xHCI is sending the > > > > SET_ADDRESS command and hence, getting rejected. > > > > Is there a way to send the Get_Descriptor ahead of Set_Address > > > > command, in the xHCI driver > > > > > > The xhci driver should not be sending these requests, but the usb core > > > does, right? > > > > The xHCI hardware wants to issue the set address control transfer > > because of virtualization (the guest doesn't need to know anything about > > what USB addresses are currently in use). The host will actually refuse > > a control transfer that it identifies as a set address command. > > > > So the only way to set a device's address is to issue a Set Address > > command to the xHCI host controller's command ring. > > Ah, ok, so I'm guessing that this type of device would also break on > Windows, right? How are they going to handle this? Probably with the BSR bit in the xHCI Address Device command. See my other mail I sent to Subs. Sarah Sharp -- 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
