[RFC] [PATCH] usbatm.[ch]: logic changes + error handling
rkagan at mail.ru
Thu Apr 21 03:29:08 EDT 2005
On Mon, Apr 18, 2005 at 05:40:28PM +0200, Duncan Sands wrote:
> > I'm not sure: usbatm_atm_send(), where it's calculated now (well, in my
> > patch the crc on the padding is calculated in usbatm_write_cells, but
> > let's forget about it for the moment), is called by the atm layer
> > normaly when it has produced the data it gives us in the skb, so it's
> > cache-hot (unless the cache is too small to fit the whole skb in).
> Well, maybe. There's only one way to be sure.
Agreed. FWIW in my patch moving the crc computation to
usbatm_write_cells() would be a pretty local change, and would even make
the code a bit more concise. The only reason I didn't do it was that I
thought it would result in the same cache utilization, as I wrote
before, but in increased number of calls to crc32_be(), so the overall
cost would go slightly up.
More information about the Usbatm