From mboxrd@z Thu Jan 1 00:00:00 1970 From: Yerden Zhumabekov Subject: Re: Load-balancing position field in DPDK load_balancer sample app vs. Hash table Date: Fri, 14 Nov 2014 23:20:42 +0600 Message-ID: <546639EA.3040404@sts.kz> References: <2601191342CEEE43887BDE71AB977258213ADDFA@IRSMSX105.ger.corp.intel.com> <54662C64.9040500@sts.kz> <2601191342CEEE43887BDE71AB977258213ADEA6@IRSMSX105.ger.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Cc: Yuanzhang Hu To: "Chilikin, Andrey" , "Ananyev, Konstantin" , Kamraan Nasim , "dev-VfR2kkLFssw@public.gmane.org" Return-path: In-Reply-To: List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces-VfR2kkLFssw@public.gmane.org Sender: "dev" Thank you. And one more thing, does Fortville (or Niantic) support various L2 headers when calculating RSS hash? I mean MPLS, QinQ, etc.? 14.11.2014 22:57, Chilikin, Andrey =D0=BF=D0=B8=D1=88=D0=B5=D1=82: > Fortville supports symmetrical hashing on HW level, a patch for i40e PM= D was submitted a couple of weeks ago. For Niantic you can use symmetrica= l rss key recommended by Konstantin. > > Regards, > Andrey > > -----Original Message----- > From: dev [mailto:dev-bounces-VfR2kkLFssw@public.gmane.org] On Behalf Of Ananyev, Konstanti= n > Sent: Friday, November 14, 2014 4:50 PM > To: Yerden Zhumabekov; Kamraan Nasim; dev-VfR2kkLFssw@public.gmane.org > Cc: Yuanzhang Hu > Subject: Re: [dpdk-dev] Load-balancing position field in DPDK load_bala= ncer sample app vs. Hash table > >> -----Original Message----- >> From: Yerden Zhumabekov [mailto:e_zhumabekov-8EHiFRVJVgQ@public.gmane.org] >> Sent: Friday, November 14, 2014 4:23 PM >> To: Ananyev, Konstantin; Kamraan Nasim; dev-VfR2kkLFssw@public.gmane.org >> Cc: Yuanzhang Hu >> Subject: Re: [dpdk-dev] Load-balancing position field in DPDK=20 >> load_balancer sample app vs. Hash table >> >> I'd like to interject a question here. >> >> In case of flow classification, one might possibly prefer for packets=20 >> from the same flow to fall on the same logical core. With this '%'=20 >> load balancing, it would require to get the same RSS hash value for=20 >> packets with direct (src to dst) and swapped (dst to src) IPs and=20 >> ports. Am I correct that hardware RSS calculation cannot provide this = symmetry? > As I remember, it is possible but you have to tweak rss key values. > Here is a paper describing how to do that: > http://www.ndsl.kaist.edu/~shinae/papers/TR-symRSS.pdf > > Konstantin > --=20 Sincerely, Yerden Zhumabekov State Technical Service Astana, KZ