Re: MMX bzero of a page...

Paul Koning (pkoning@xedia.com)
Wed, 11 Mar 1998 09:38:14 -0500

On many (most?) high speed machines, memory-intensive functions like
bzero or memcpy are memory bandwidth limited by a large margin. You
may want to benchmark what you described against the "trivial"
implementation; it may be that they have identical performance limited
by the memory bus.

For a related example, on many processors it's possible to do a TCP
checksum + memcpy combined in the same time as just a memcpy, because
the checksum arithmetic simply consumes cycles that otherwise would be
memory stall cycles. (And someone the other day mentioned
decompression at 0.5 times memcpy speed -- that sounds plausible for
the same sort of reasons.)

paul

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu