Re: Connection table hashing

[Roberto Nibali <ratz@xxxxxxxxxxxx>]

Hi Julian,

> http://www.ssi.bg/~ja/lvs/
> test.txt contains short instructions for testing.
> ipvs-1.0.9-hash1.diff - test version for 2.4.21
What do you think about exporting the hash bucket distribution to the 
proc-fs for debug=1? People could send it in easier for us. I also think 
that the multi-cycle latency instructions from mix() might yield 
increased TLB flushes on HT enabled Pentium processors.
Cheers,
ratz
--
echo '[q]sa[ln0=aln256%Pln256/snlbx]sb3135071790101768542287578439snlbxq'|dc