Re: [PATCH 3 of 3] Add __raw_memcpy_toio32 to each arch

[Bryan O'Sullivan]

On Tue, 2006-01-10 at 21:08 +0100, Andi Kleen wrote:
> On Tuesday 10 January 2006 20:53, Bryan O'Sullivan wrote:
> > Most arches use the generic routine.  x86_64 uses memcpy32 instead;
> > this is substantially faster, even over a bus that is much slower than
> > the CPU.
> 
> So did you run numbers against the C implementation with -funroll-loops ? 
> What were the results?

The C implementation is about 5% slower when copying over
HyperTransport.

	<b

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/