Hi,

On Thu, Jan 09, 2003 at 12:21:56PM -0800, David Brownell wrote:
> I've never exactly been clear on when usb_resetep() could be used
> correctly ... it just sets the toggle to DATA0, even if the endpoint
> is busy.  So there's basically no way to know if it's even safe to
> try changing the toggle at that point; and there's no way to know if
> the device expects DATA0, so it can easily lead to dropped packets
> (and then to reads that hang, as your case).

I see. I'll better remove this function from SANE, even if no problem
occured until know with other devices.

> Near as I can tell, "resetep" is a crappy function that will, at
> least half the time you use it, cause misbehavior ... and the rest
> of the time, will be a NOP (because the toggle is already DATA0,
> as it should be).  I've suspected that the correct implementation
> in usbfs would be just to always fail with some errno.
> 
> You might _also_ have crappy hardware (or firmware in it).

I think so, see the other answer.

Thanks,
  Henning


-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to