On Sun, Apr 12, 2009 at 4:32 AM, Tilman Schmidt <tilman@xxxxxxx> wrote: > On Sun, 05 Apr 2009 16:02 -0700, "David Brownell" <david-b@xxxxxxxxxxx> > wrote: >> Why is it assuming that altsetting has endpoints? >> If that interface uses ISO transfers, then the >> first altsetting must not have any endpoints... > > I'm not sure I understand you correctly. the way I read the spec, the > first altsetting may well have endpoints, just no isochronous ones. > Namely, I have a device which uses isochronous transfers and which in > its fully functional altsetting 3 has (in addition to EP0 of course) > two interrupt endpoints (one in, one out) and four isochronous > endpoints (two in, two out). In altsetting 0, the isochronous > endpoints are absent, in accordance with the standard. The interrupt > endpoints, however, do exist in every altsetting. > Actually a device can (and I have seen a MS webcam that does) have alt0 with isoc endpoints, but the isoc endpoints must have zero maxpacket size. I thought (before I just looked it up) that a interface/alt 0 couldn't have any periodic endpoints (iso and int), but the usb2.0 spec "5.7.3 Interrupt Transfer Packet Size Constraints" says: <quote>All high-speed device default interface settings must not include any interrupt endpoints with a data payload size (specified via wMaxPacketSize in the endpoint descriptor) greater than 64 bytes. Alternate interface settings may specify larger data payload sizes for interrupt endpoints. If the interrupt endpoints have a large data payload size, it is recommended that additional configurations or alternate interface settings be used to specify a range of data payload sizes.</quote> Regards, Steve -- 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
