On Fri, Sep 24, 2010 at 03:55:23AM -0500, Ming Lei wrote:
2010/9/24 Felipe Balbi <balbi@xxxxxx>:
Hi,
On Fri, Sep 24, 2010 at 03:43:09AM -0500, Ming Lei wrote:
Seems there is no such problem of sending _two_ ZLPs in the case.
The request will be giveback immediately after the 1st ZLP is sent
since request.length == request.actual == 0 will trigger giveback.
you never know, some protocols running on top of USB might get confused
if you send two End-Of-Transfer packets, maybe OBEX would be one of
those ?!?
You mean queuing the request in .complete will cause two consecutive ZLPs?
If so, seems it is not caused by gadget driver.
Could we remove request.zero flag and only keep request.length == 0 to
trigger ZLP sending? which may simplify such things.
nope, I think mass storage always sets request->zero and expects
controller driver to handle the ZLP. Anyway, let's keep as is, we can
revisit this later.
--
balbi
--
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
