Re: [PATCH] usb/core/devio.c:tolerate wrong direction flag in control endpoints

2013-09-23 Thread Kurt Garloff
Hi Alan, Alan Stern schrieb: > On Mon, 23 Sep 2013, Kurt Garloff wrote: > > > >> that qualifies as a bug or not. Maybe it should not claim to be a > > >> HID device then? > > > Maybe not. This particular combination of bRequestType and > bRequest > > > values (0x22, 0x09) is not defined in t

Re: [PATCH] usb/core/devio.c:tolerate wrong direction flag in control endpoints

2013-09-23 Thread Alan Stern
On Mon, 23 Sep 2013, Kurt Garloff wrote: > >> that qualifies as a bug or not. Maybe it should not claim to be a > >> HID device then? > > Maybe not. This particular combination of bRequestType and bRequest > > values (0x22, 0x09) is not defined in the HID 1.11 spec. Do you know > > if it is de

Re: [PATCH] usb/core/devio.c:tolerate wrong direction flag in control endpoints

2013-09-23 Thread Kurt Garloff
Hi Alan, On 09/23/2013 04:28 AM, Alan Stern wrote: > On Sun, 22 Sep 2013, Kurt Garloff wrote: > >> Well, this seems to be a question of terminology, no? >> I saw the endpoint byte as consisting of endpoint index plus the direction >> bit. > See the entry for "Endpoint Address" in Chapter 2 (Terms

Re: [PATCH] usb/core/devio.c:tolerate wrong direction flag in control endpoints

2013-09-22 Thread Alan Stern
On Sun, 22 Sep 2013, Kurt Garloff wrote: > Hi Alan, > > thanks for your review and your constructive comments! You're welcome. > >> Hi, > >> > >> USB devio rejects control messages when the index does not have the > >> direction bit set correctly. > > > >I wouldn't describe it that way. It wo

Re: [PATCH] usb/core/devio.c:tolerate wrong direction flag in control endpoints

2013-09-22 Thread Kurt Garloff
Hi Alan, thanks for your review and your constructive comments! Alan Stern schrieb: >On Sun, 22 Sep 2013, Kurt Garloff wrote: > >> Hi, >> >> USB devio rejects control messages when the index does not have the >> direction bit set correctly. > >I wouldn't describe it that way. It would be mor