On Thu, 2007-02-22 at 14:27, David Brownell wrote:
> On Thursday 22 February 2007 1:35 pm, Jeff Warren wrote:
> 
> > To be honest, I don't know what at91_udc does when it receives a ZLP
> > since I can't easily generate one on the Windows host.  So maybe it does
> > set one of the FIFO bank flags on the csr register and passes through
> > read_fifo(), which in turn then calls done().
> 
> It's certainly _supposed_ to Do The Right Thing and terminate the
> read immediately.  I don't have time to re-learn that hardware, so
> unless you can show that it's got a bug, I'll assume it works right.
That's fair.
> 
> That's a fairly basic mechanism, and I'd be surprised if such a bug
> has gone undetected for over a year and a half ...
> 
> - Dave
It sounds like having the host send a couple of extra bytes in that
specific case is good enough.  The right way would be to have the host
send a ZLP, but that would be more work than just having the host send a
couple extra bytes in that specific case.  I'm going to go with easy.

Thanks Again,

Jeff Warren


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
[email protected]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to