> But it also proves that it's not spec compliant :).
> See posting from Roman Weissgaerber
> http://www.geocrawler.com/lists/3/SourceForge/11993/50/8141454/

I am using the Microtune(transilica) 1760 bluetooth USB module, which is
less famous but qualified.
http://www.transilica.com/tech-qualification.html
OneChipT Bluetooth USB/Serial PortTR076010/20/01Eriksson, Lars

However, I found that the Roman's point is very valid as the zero length
case is only mentioned in both spec USB 1.1(page182) and 2.0 (page 228) (
"If the data structure is an exact multiple of
wMaxPacketSize for the pipe, the function will return a zero-length packet
to indicate the end of the Data stage." ) in the section 8.5.3.2
*****Variable-length Data Stage****** under the 8.5.3 Control Transfers not
in the bulk section. But in Bluetooth the HCI and header and HCI data are
known. So the zero length need not be used here?

Among all these Linux USB driver, only the IRDA drivers use the zero -length
packet and still under experiment. Here is my finding in Universal Serial
Bus IrDA Bridge Device Definition
http://www.usb.org/developers/data/devclass/usbir10.pdf


##################
Since the native transfer size of IrLAP frames may exceed the maximum packet
size of a USB bulk endpoint, it is necessary to delineate the beginning and
ending of IrLAP frames within the data stream delivered by a bulk endpoint.
This positive delineation of IrLAP frames is done using a USB short packet
mechanism. When an IrLAP frame spans N USB packets, the first packet through
packet N-1 shall be the maximum packet size defined for the USB endpoint. If
the Nth packet is less than maximum packet size the USB transfer of this
short packet will identify the end of the IrLAP frame. *****If the Nth
packet is exactly maximum packet size, it shall be followed by a zero-length
packet; the zero length packet identifies the end of the IrLAP frame.
**********
###################

And this should be why only irda-usb.c file uses the zero length packet.
This means we might not need the zero length in the blueZ , just like
Openbt.

I was told that the CRS has problem with the size 52 using l2ping as its
bulk default size is 64.  Can anyone please give me some detail about the
test on the CSR kit . I don't have one with me.

Thank you all.
Regards,
Xiaoyong







_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to