From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga07.intel.com (mga07.intel.com [134.134.136.100]) by dpdk.org (Postfix) with ESMTP id 834172BD8 for ; Fri, 16 Nov 2018 18:20:16 +0100 (CET) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga005.jf.intel.com ([10.7.209.41]) by orsmga105.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 16 Nov 2018 09:20:15 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.56,240,1539673200"; d="scan'208";a="274587229" Received: from fyigit-mobl.ger.corp.intel.com (HELO [10.237.221.79]) ([10.237.221.79]) by orsmga005.jf.intel.com with ESMTP; 16 Nov 2018 09:20:14 -0800 To: "Lam, Tiago" , dev@dpdk.org Cc: linville@tuxdriver.com, Bruce Richardson References: <24eb3e7e-17f9-222f-aab1-5acfb86823c7@intel.com> From: Ferruh Yigit Openpgp: preference=signencrypt Autocrypt: addr=ferruh.yigit@intel.com; prefer-encrypt=mutual; keydata= xsFNBFXZCFABEADCujshBOAaqPZpwShdkzkyGpJ15lmxiSr3jVMqOtQS/sB3FYLT0/d3+bvy qbL9YnlbPyRvZfnP3pXiKwkRoR1RJwEo2BOf6hxdzTmLRtGtwWzI9MwrUPj6n/ldiD58VAGQ +iR1I/z9UBUN/ZMksElA2D7Jgg7vZ78iKwNnd+vLBD6I61kVrZ45Vjo3r+pPOByUBXOUlxp9 GWEKKIrJ4eogqkVNSixN16VYK7xR+5OUkBYUO+sE6etSxCr7BahMPKxH+XPlZZjKrxciaWQb +dElz3Ab4Opl+ZT/bK2huX+W+NJBEBVzjTkhjSTjcyRdxvS1gwWRuXqAml/sh+KQjPV1PPHF YK5LcqLkle+OKTCa82OvUb7cr+ALxATIZXQkgmn+zFT8UzSS3aiBBohg3BtbTIWy51jNlYdy ezUZ4UxKSsFuUTPt+JjHQBvF7WKbmNGS3fCid5Iag4tWOfZoqiCNzxApkVugltxoc6rG2TyX CmI2rP0mQ0GOsGXA3+3c1MCdQFzdIn/5tLBZyKy4F54UFo35eOX8/g7OaE+xrgY/4bZjpxC1 1pd66AAtKb3aNXpHvIfkVV6NYloo52H+FUE5ZDPNCGD0/btFGPWmWRmkPybzColTy7fmPaGz cBcEEqHK4T0aY4UJmE7Ylvg255Kz7s6wGZe6IR3N0cKNv++O7QARAQABzSVGZXJydWggWWln aXQgPGZlcnJ1aC55aWdpdEBpbnRlbC5jb20+wsGVBBMBAgA/AhsDBgsJCAcDAgYVCAIJCgsE FgIDAQIeAQIXgBYhBNI2U4dCLsKE45mBx/kz60PfE2EfBQJbughWBQkHwjOGAAoJEPkz60Pf E2Eft84QAIbKWqhgqRfoiw/BbXbA1+qm2o4UgkCRQ0yJgt9QsnbpOmPKydHH0ixCliNz1J8e mRXCkMini1bTpnzp7spOjQGLeAFkNFz6BMq8YF2mVWbGEDE9WgnAxZdi0eLY7ZQnHbE6AxKL SXmpe9INb6z3ztseFt7mqje/W/6DWYIMnH3Yz9KzxujFWDcq8UCAvPkxVQXLTMpauhFgYeEx Nub5HbvhxTfUkapLwRQsSd/HbywzqZ3s/bbYMjj5JO3tgMiM9g9HOjv1G2f1dQjHi5YQiTZl 1eIIqQ3pTic6ROaiZqNmQFXPsoOOFfXF8nN2zg8kl/sSdoXWHhama5hbwwtl1vdaygQYlmdK H2ueiFh/UvT3WG3waNv2eZiEbHV8Rk52Xyn2w1G90lV0fYC6Ket1Xjoch7kjwbx793Kz/RfQ rmBY8/S4DTGn3oq3dMdQY+b6+7VMUeLMMh2CXYO9ErkOq+qNTD1IY+cBAkXnaDbQfz0zbste ZGWH74FAZ9nCpDOqbRTrBL42aMGhfOWEyeA1x7+hl6JZfabBWAuf4nnCXuorKHzBXTrf7u7p fXsKQClWRW77PF1VmzrtKNVSytQAmlCWApQIw20AarFipXmVdIjHmJPU611WoyxZPb4JTOxx 5cv9B+nr/RIB+v5dcStyHCCwO1be7nBDdCgd4F6kTQPLzsFNBFfWTL4BEACnNA29e8TarUsB L5n6eLZHXcFvVwNLVlirWOClHXf44o2KnN3ww+eBEmKVfEFo9MSuGDNHS8Zw1NiGMYxLIUgd U6gGrVVs/VrQWL82pbMk6jCj98N+BXIri+6K1z+AImz7ax7iF1kDgRAnFWU0znWWBgM2mM8Y gDjcxfXk4sCKnvf6Gjo08Ey5zmqx7dekAKU2EEp8Q1EJY3jbymLdZWRP4AFFMTS1rGMk0/tt v71NBg1GobCcbNfn9chK/jhqxYhAJqq86RdJQkt3/9x1U1Oq0vXCt4JVVHmkxePtUiuWTTt+ aYlUAsKYZsWvncExvw77x2ArYDmaK0yfjh37wp0lY7DOJHFxoyT8tyWZlLci/VMRG2Ja33xj 0CN4C1yBg+QDeV3QFxQo42iA/ykdXPUR3ezmsND3XKvVLTC4DNb3V/EZQ7jBj64+bEK0VW4G B31VP00ApNQvSoczsIOAKdk97RNbpmPw6q10ILIB+9T1xbnFYzshzGF17oC0/GENIHATx8vZ masOZoDiOZQpeneLgnFE9JfzhLTxv6wNZcc/HLXRQVTkDsQr8ERtkAoHCf1E5+b5Yr7pfnE4 YuhET746o25S53ELUYPIs49qoJsEJL34/oexMfPGyPIlrbufiNyty5jc/1MRwUlhJlJ5IOHy ZUa+6CLR7GdImusFkPJUJwARAQABwsF8BBgBAgAmAhsMFiEE0jZTh0IuwoTjmYHH+TPrQ98T YR8FAlu6CHAFCQXE7zIACgkQ+TPrQ98TYR9nXxAAqNBgkYNyGuWUuy0GwDQCbu3iiMyH1+D7 llafPcK4NYy1Z4AYuVwC9nmLaoj+ozdqS3ncRo57ncRsKEJC46nDJJZYZ5LSJVn63Y3NBF86 lxQAgjj2oyZEwaLKtKbAFsXL43jv1pUGgSvWwYtDwHITXXFQto9rZEuUDRFSx4sg9OR+Q6/6 LY+nQQ3OdHlBkflzYMPcWgDcvcTAO6yasLEUf7UcYoSWTyMYjLB4QuNlXzTswzGVMssJF/vo V8lD1eqqaSUWG3STF6GVLQOr1NLvN5+kUBiEStHFxBpgSCvYY9sNV8FS6N24CAWMBl+10W+D 2h1yiiP5dOdPcBDYKsgqDD91/sP0WdyMJkwdQJtD49f9f+lYloxHnSAxMleOpyscg1pldw+i mPaUY1bmIknLhhkqfMmjywQOXpac5LRMibAAYkcB8v7y3kwELnt8mhqqZy6LUsqcWygNbH/W K3GGt5tRpeIXeJ25x8gg5EBQ0Jnvp/IbBYQfPLtXH0Myq2QuAhk/1q2yEIbVjS+7iowEZNyE 56K63WBJxsJPB2mvmLgn98GqB4G6GufP1ndS0XDti/2K0o8rep9xoY/JDGi0n0L0tk9BHyoP Y7kaEpu7UyY3nVdRLe5H1/MnFG8hdJ97WqnPS0buYZlrbTV0nRFL/NI2VABl18vEEXvNQiO+ vM8= Message-ID: <507e0be8-f58b-9d8b-873f-9e1d45f14e93@intel.com> Date: Fri, 16 Nov 2018 17:20:13 +0000 User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:60.0) Gecko/20100101 Thunderbird/60.3.1 MIME-Version: 1.0 In-Reply-To: <24eb3e7e-17f9-222f-aab1-5acfb86823c7@intel.com> Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 8bit Subject: Re: [dpdk-dev] af_packet dev default "framesz" of 2048B X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 16 Nov 2018 17:20:17 -0000 On 11/15/2018 7:02 PM, Lam, Tiago wrote: > Hi guys, > > OvS-DPDK has recently had small a change that changed the data room > available in an mbuf (commit dfaf00e in OvS). This seems to have had the > consequence of breaking the initialisation of eth_af_packets interfaces, > when using default values ("options:dpdk- > devargs=eth_af_packet0,iface=enp61s0f3"). > > After investigating, what seems to be happening is that the > eth_af_packet dev expects an available space of "2048B - TPACKET2_HDRLEN > + sizeof(struct sockaddr_ll) = 2016B" to be available in the data room > of each mbuf. Previous to the above commit, OvS would allocate some > extra space, and this would mean there would be enough room for the > checks performed in eth_rx_queue_setup() and eth_dev_mtu_set() in > rte_eth_af_packet.c. However, with the recent commit that isn't the case > anymore, and without that extra space the first check in > eth_rx_queue_setup() will now be hit and setup of a eth_af_packet > interface fails. > > What I'm trying to understand here is, the logic behind setting a > default 'framesz' of 2048B and it being hardcoded (instead of being > based on the underlying MTU of the interface, or the mbuf data room > directly). The documentation in [1] for mmap() and setting up buffer > rings mentions the exact same values > (tp_block_size=4096,tp_frame_size=2048), which seem to have been > introduced on the first commit, back in 2014. The only constraint > for the framesize, it seems, its that it fits inside the blocksize (i.e. > doesn't span multiple blocksizes), and is aligned to TPACKET_ALIGNMENT. Independent from Bruce's comment to not use smaller mbuf size, I believe af_packet can be updated to be more flexible. Related to 'framesize', it has default value 2048 but can be updated via 'framesz=' devarg, so not exactly hardcoded. Your change looks good there, use devarg value if exist, get from underlying device and set default if it fails. A few comment below. > > Thus, given those constraints, could we instead base the setting of the > framesize like on the below patch? It basically tries to set the > framesize to the MTU of the underlying interface + TPACKET2_HDRLEN + > sizeof(structsockaddr_ll), aligned to TPACKET_ALIGNMENT. This would > allow applications such as OvS to use this interface by default, based > on a framesize set dynamically, instead of relying on a default of > 2048B; If one is setting up an eth_af_packet interface with MTU 9000B > and the underlying MTU is 1500B, for example, this will still fail, and > that's fine as they can always supply the "framesz" argument > ("options:dpdk-devargs=eth_af_packet0,iface=enp61s0f3,framesz=9000"). > However, on the default case, or when one has configured manually the > underlying interface with the expected MTU already, this would work out > of the box. > > Any thoughts on this? > > Thanks, > Tiago. > > diff --git a/drivers/net/af_packet/rte_eth_af_packet.c b/drivers/net/af_packet/rte_eth_af_packet.c > index 95a98c6..451cec3 100644 > --- a/drivers/net/af_packet/rte_eth_af_packet.c > +++ b/drivers/net/af_packet/rte_eth_af_packet.c > @@ -24,6 +24,8 @@ > #include > #include > > +#define RTE_ROUNDUP(x, y) ((((x) + ((y) - 1)) / (y)) * (y)) This can go into more generic header <...> > @@ -622,6 +617,38 @@ rte_pmd_init_internals(struct rte_vdev_device *dev, > return -1; > } > memcpy(&(*internals)->eth_addr, ifr.ifr_hwaddr.sa_data, ETH_ALEN); > + if (!framesize) { > + if (ioctl(sockfd, SIOCGIFMTU, &ifr) == -1) { > + RTE_LOG(ERR, PMD, > + "%s: ioctl failed (SIOCGIFMTU)", > + name); > + framesize = DFLT_FRAME_SIZE; > + } else { > + framesize = ifr.ifr_mtu; > + framesize += TPACKET2_HDRLEN + sizeof(struct sockaddr_ll); > + framesize = RTE_ROUNDUP(framesize, TPACKET_ALIGNMENT); > + } > + } > + > + blockcnt = framecnt / (blocksize / framesize); If returned framesize > blocksize, result will be 0 since both are uint, which will end framecnt/0 and crash. <...> > @@ -887,21 +913,8 @@ rte_eth_from_packet(struct rte_vdev_device *dev, > return -1; > } > > - blockcount = framecount / (blocksize / framesize); > - if (!blockcount) { > - PMD_LOG(ERR, > - "%s: invalid AF_PACKET MMAP parameters", name); > - return -1; > - } > - > - PMD_LOG(INFO, "%s: AF_PACKET MMAP parameters:", name); > - PMD_LOG(INFO, "%s:\tblock size %d", name, blocksize); > - PMD_LOG(INFO, "%s:\tblock count %d", name, blockcount); > - PMD_LOG(INFO, "%s:\tframe size %d", name, framesize); > - PMD_LOG(INFO, "%s:\tframe count %d", name, framecount); > - Why move this block or change rte_pmd_init_internals(), adding SIOCGIFMTU above this block looks simpler.