[EMAIL PROTECTED] wrote:

Why is an open bulk IN point used at all? Over the interrupt endpoint the
need of an bulk IN transfer could be signalled and then the bulk IN
transfer could be initiated for that specific transfer only. Thats the
common way interrupts are used for. This technique would result in less
bandwidth consumption regarding PCI and memory accesses. In my point of
view the vendors of USB devices should overwork their protocol designs
regarding this aspect.

Because that sort of "wait for interrupt transfer, THEN issue request" model increases per-request latency by quite a lot (8msec polling period increases latency of IN transfers by close to 10msec), which can kill throughput ... which matters a lot for network style links.

It also requires an extra interrupt-capable endpoint, which isn't
free even if it's available.  Interrupt endpoints consume bandwidth
that's "reserved", and which could run out.

And surely other reasons too, but the basic point is that these are
all tradeoffs.

- Dave



-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training.
Attend Black Hat Briefings & Training, Las Vegas July 24-29 - digital self defense, top technical experts, no vendor pitches, unmatched networking opportunities. Visit www.blackhat.com
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to