From mboxrd@z Thu Jan 1 00:00:00 1970 From: Olivier MATZ Subject: Re: [PATCH v8 10/10] app/testpmd:test VxLAN Tx checksum offload Date: Wed, 12 Nov 2014 15:56:33 +0100 Message-ID: <54637521.9000709@6wind.com> References: <1414376006-31402-1-git-send-email-jijiang.liu@intel.com> <1414376006-31402-11-git-send-email-jijiang.liu@intel.com> <54588BF7.309@6wind.com> <1ED644BD7E0A5F4091CF203DAFB8E4CC01D8510E@SHSMSX101.ccr.corp.intel.com>, <5459FBB2.1040408@6wind.com> <0c654d2c0d304b45a40af6ca38b70adf@EX13-MBX-026.vmware.com> <545CFE56.60605@6wind.com> <2601191342CEEE43887BDE71AB977258213A38D2@IRSMSX105.ger.corp.intel.com> <5460E07F.6060303@6wind.com> <2601191342CEEE43887BDE71AB977258213A3F5F@IRSMSX105.ger.corp.intel.com> <54635B2B.5040603@6wind.com> <2601191342CEEE43887BDE71AB977258213A40EB@IRSMSX105.ger.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: "dev-VfR2kkLFssw@public.gmane.org" To: "Ananyev, Konstantin" , Yong Wang , "Liu, Jijiang" Return-path: In-Reply-To: <2601191342CEEE43887BDE71AB977258213A40EB-kPTMFJFq+rEu0RiL9chJVbfspsVTdybXVpNB7YpNyf8@public.gmane.org> 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" Hi Konstantin, On 11/12/2014 03:39 PM, Ananyev, Konstantin wrote: >> I'm not sure having get_ipv4_udptcp_checksum() in librte_net would >> help. The value we have to set in the TCP checksum field depends on the >> PMD (altought only ixgbe is supported now). So, it would require >> another parameter and a new PMD eth_ops... which looks very >> similar to dev_prep_tx() (except that dev_prep_tx() can be bulked). >> I think a stack will not be able to call get_udptcp_checksum(m ,port) >> because it does not know the physical port at the time the packet is >> built. Moreover, calling a function through a pointer is more efficient >> when bulked. So I think the dev_prep_tx() you initially describe is >> a better answer to the problem. > > Yes I understand that it might not be applicable for non-Intel NICs. > Though I thought it is ok as a temporary measure as right now we > support these offloads for Intel NICs only. > Basically my thought was what you proposed as option 3 below. > Why common function in librte_net? > So people don't need to write their own each time. > Plus as I remember all 3 Intel NIC types (ixgbe/igb/i40e) we support have similar > requirements for what need to be set/calculated for these TX offloads. > So my thought was that having a common function might help to avoid code duplication in future, > If/when will implement dev_prep_tx(). OK, now I understand better what you suggest. I'll try to implement the option 3 (below) with a common checksum function in librte_net in the next version of the TSO patchset. Regards, Olivier > >> >> I don't know what is the exact timeframe for 1.8, maybe Thomas can help >> on this? Depending on it, we have several options: >> >> - implement dev_prep_tx() for 1.8 in the TSO series: this implies that >> the community agrees on this new API. We need to check that it will >> be faster in a pipeline model (I think this is obvious) but also that >> it does not penalize the run-to-completion model: introducing another >> function dev_prep_tx() can result in duplicated tests in the driver >> (ex: test the offload flag values). >> >> - postpone dev_prep_tx() or similar to next version and push the current >> TSO patchset (including the comments done on the list). It does not >> modify the current offload API, it provides the TSO feature on ixgbe >> based on a similar API concept (set the TCP phdr cksum). The drawback >> is a potential performance loss when using a pipeline model. >> >> - another option that you may prefer is to bind the API behavior to >> ixgbe (for 1.8): we can ask the application to set the pseudo-header >> checksum without the IP len when doing TSO, as required by the ixgbe >> driver. Then, for next release, we can think about dev_prep_tx(). The >> drawback of this solution is that we may go back on this choice if the >> dev_prep_tx() approach is not validated by the community.