From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: packet received in a wrong rx-queue? Date: Wed, 18 May 2011 23:36:14 -0400 (EDT) Message-ID: <20110518.233614.1134947771941589398.davem@davemloft.net> References: <4A6A2125329CFD4D8CC40C9E8ABCAB9F250D749114@MILEXCH2.ds.jdsu.net> <20110518.213225.1546509125052154115.davem@davemloft.net> <4A6A2125329CFD4D8CC40C9E8ABCAB9F250D7491A8@MILEXCH2.ds.jdsu.net> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Cc: e1000-devel@lists.sourceforge.net, netdev@vger.kernel.org To: Jon.Zhou@jdsu.com Return-path: In-Reply-To: <4A6A2125329CFD4D8CC40C9E8ABCAB9F250D7491A8@MILEXCH2.ds.jdsu.net> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: e1000-devel-bounces@lists.sourceforge.net List-Id: netdev.vger.kernel.org From: Jon Zhou Date: Wed, 18 May 2011 20:22:07 -0700 > form the 82599 datasheet, the hash algorithm is consist of src/dst > ip, src/dst port,protocol why it got different hash value with same > ip/port pair? The same reason why feeding different sets of discrete 32-bit and 16-bit values to a cryptographic hash results in a different final hash value. Our software RPS/RFS implementation used to have this quality too. ------------------------------------------------------------------------------ What Every C/C++ and Fortran developer Should Know! Read this article and learn how Intel has extended the reach of its next-generation tools to help Windows* and Linux* C/C++ and Fortran developers boost performance applications - including clusters. http://p.sf.net/sfu/intel-dev2devmay _______________________________________________ E1000-devel mailing list E1000-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/e1000-devel To learn more about Intel® Ethernet, visit http://communities.intel.com/community/wired