On Fri, 11 May 2012, Andiry Xu wrote: > > Okay. Why do you care about short packets on isochronous transfers? I > > wouldn't think they make any difference. > > > > If we do not set ISP for ISOC IN transfers, there is no event produced > when a short packet occurs and the driver cannot update the status and > length field of iso_packet_descriptor accordingly. You are begging the question. Why do you care if there is no event produced when a short packet occurs? After all, there is no event when a full-sized packet occurs; what's special about short packets? As for the iso_packet_descriptor status field... Surely that is already being updated. Otherwise it would always contain -EINPROGRESS, because that's what it gets set to when URBs are submitted. Much the same is true for the length field. In fact, you must really mean the actual_length field, because the length field does not get updated. Besides, doesn't Sarah's most recent patch set the actual_length field correctly when a short packet occurs? Alan Stern -- 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
