On Wed, 23 Jun 2004, Tommy Faasen wrote:

> >That patch does DMA on the stack.
> >Maybe including such a patch is a good idea, following the other OS.
> >But it'll require a kmalloced buffer.
> >  
> >
> Why, the buffer isn't used anyway? It's just used to jumpstart some usb 
> devices?

Wrong.  It's true that the subroutine itself doesn't use the buffer.  But 
it passes the buffer to usb_control_msg(), which does use it -- the 
descriptor data from the device is stored there.

There's a simpler way to get an equivalent effect.  Temporarily set 
dev->devnum to 0 and call usb_get_device_descriptor().  But this won't 
help solve your problem any more than the old patch did.

One thing you might try doing is to change the

        msleep(10);

just before hub_port_init() calls usb_get_device_descriptor() for the
first time, and make it use a longer delay.  Maybe with 100 milliseconds
breathing space instead of 10 the device will work better.

Alan Stern



-------------------------------------------------------
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