Arvid Brodin wrote:
These patches fixes one known and a couple of suspected bugs:
Thanks for working on this.
1) urb_dequeue() always aborted the active transfer on the urb's
endpoint, even if the transfer in question belonged to a different
urb. (known bug)
2) Since there is no known way to actually terminate a started
transfer with the isp1760, we reinstate the use of the skip maps
and use them as a filter for interrupts and done maps.
Didn't you say once that the transfer is still active? With the use of the
skipmap we don't get interrupts with (qtd == NULL) but the slot might be
re-used right? If this is still the case what about leaving things as it
and assigning qtd = (void*)1 on dequeue. Once we get the interrupt for qtd
== 1 we would set it to NULL. So we won't re-use this slot too early.
Sebastian
--
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
