Date: Fri, 18 Jul 2014 11:15:24 -0700 From: Adrian Chadd <adrian@freebsd.org> To: Peter Grehan <grehan@freebsd.org> Cc: "svn-src-head@freebsd.org" <svn-src-head@freebsd.org> Subject: Re: svn commit: r268837 - head/sys/netinet Message-ID: <CAJ-VmonRRZPi%2BRkGQk25xrXGf9BOhHXvAVhRWse%2BfXKduoJ19g@mail.gmail.com> In-Reply-To: <53C94F61.5020706@freebsd.org> References: <201407180822.s6I8MD5a023838@svn.freebsd.org> <53C94F61.5020706@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 18 July 2014 09:46, Peter Grehan <grehan@freebsd.org> wrote: >> Update the default RSS hash to the Chelsio T5 firmware one - it >> provides >> markedly better distribution of IPv6 address/ports than the previous >> key. > > > That's actually the key that's used for verification in the Microsoft spec, > though it looks like you have the bytes arranged backwards, which appears to > be an artifact of the strange way the key is programmed into the Chelsio. Hah! I wonder why the key from the T3 NIC that Robert committed was so .. terrible then. Are you saying that the actual byte order is totally reversed, or just endian? > My suggestion is: put the bytes in the default key in the order they are in > the spec. This allows independent verification of the hash function in > hardware by injecting frames with the values from the spec and looking at > what h/w provides. Then, fix how it is programmed in the Chelsio. From > memory, the Intel registers were a byte array. I also recall that the Qlogic > FCOE adapters were 32-bits at a time, but big-endian. Cool. > On the contents of the has, so long as there are a enough random 1-bits in > the key, the distribution is as good as the Jenkins hash. Easy to verify > with a test harness and ministat etc, but also verified at Netapp in the > mega performance lab with a number of different key values. I think that was Roberts main concern with using a randomised key at startup. It sounds Netapp have already done a lot of the work that I was thinking about in the back of my mind - I'll add that to my TODO list so we can write a random key generation function at boot time. Thanks1 -a
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAJ-VmonRRZPi%2BRkGQk25xrXGf9BOhHXvAVhRWse%2BfXKduoJ19g>