Ronciak, John a écrit :
In this combination of hardware and in this forwarding test
copybreak is bad but prefetching helps.
e1000 vanilla 1150 kpps
e1000 6.2.15 1084
e1000 6.2.15 copybreak disabled 1216
e1000 6.2.15 copybreak disabled and no rx prefetch 1097
This is what we are seeing as well. The copybreak performance makes
sense since every packet in this test would be copied increasing the CPU
to probably where it's pegged (that's what we see).
Note that on a router (ie most packets are not locally delivered), copybreak
is useless and expensive.
But if most packets are locally delivered (on local TCP or UDP queues), then
copybreak is a win because less memory is taken by not yet read packets in queues.
Allocating a 256 bytes block instead of a full 4096 page if a 16 factor.
For a machine doing some p2p trafic (ie receiving lot of small UDP frames or
handling a lot of TCP sockets), copybreak is definitly a good feature.
So maybe an ethtool tune should be added to set the copybreak limit on each
nic. AFAIK tg3 uses a fixed size copybreak limit.
Eric
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
