Am Dienstag, 19. Juni 2007 schrieb Pete Zaitcev:
> On Tue, 19 Jun 2007 11:11:18 +0800, "jidong xiao" <[EMAIL PROTECTED]> wrote:
>
> > But I remember, the reason we use sg is that, we can transfer
> > several buffers in one DMA operation, thus we can speed things up
> > accordingly because we turned multiple operations into a single DMA.
> > That's the benefit. However, in above situation, we use sg, but we
> > still perform multiple DMA operations, how to explain this?
>
> The main problem is that the URB does not have an s/g list. If it
> did, the whole library would be unnecessary. Nonetheless, we still
And that from the man who complained about 24 more bytes in struct urb ;-)
If I understand your position correctly, you think that an URB should
either have all bells and whistles regardless of its size or it should be
lean and provide only the bare minimum. Just the current compromise
is bad. Is that correct?
Regards
Oliver
-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
linux-usb-devel@lists.sourceforge.net
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
