Sebastien Roy wrote:
> Mobile IPv4 is being removed from OpenSolaris as detailed in PSARC
> 2007/311
> (http://www.opensolaris.org/os/community/arc/caselog/2007/311/), and
> you can help that effort by participating in the code-review.
>
> To give you an idea of the scope of this review, the work entails the
> removal of two packages (SUNWmipr and SUNWmipu), the removal of 81
> source files, and the modification of 27 remaining files. Within
> those modified files, 2157 lines of code were removed, mostly within
> the ip kernel module.
>
> Please provide comments by Friday July 13th. Also please notify me as
> soon as possible if you plan on participating in the review so that I
> can account for appropriate review coverage.
>
> The webrev is located here:
>
> http://cr.opensolaris.org/~seb/rm_mobileip_webrev/
>
> For those within SWAN, the workspace (including cscope databases in
> usr/src and usr/src/uts) is located here:
>
> /net/zhadum.east/export/ws/seb/rm_mobileip_cr/
>
> Regression tests run have included the TCP, NFSv4, Connectathon NFS,
> NFSv2, and IPv6 basic API tests. All pass. In addition, I've run
> netperf and ttcp tests showing that the change does not affect
> performance (performance improvements were in the noise). I'm still
> waiting on a working CGTP test suite to run CGTP tests. I plan on
> doing this prior to integration.
I'd bet, with a high degree of confidence, that if you tried doing an IP
forwarding test with small packets, you'd find that performance
improvements are _not_ in the noise, unless your noise filter is set too
low.
Another way to test would be to try doing performance runs with a 10g
card using something _other_ than TCP (UDP rx would be good). Look at
the improvements to the packets-per-second count rather than the thruput
numbers. :-)
If you're not CPU bound, you won't see the benefit.
-- Garrett
>
> Thanks,
> -Seb
> _______________________________________________
> networking-discuss mailing list
> networking-discuss at opensolaris.org
