Patch Comment List
List patch comments
GET /api/patches/337/comments/?format=api
[ { "id": 960, "web_url": "http://patches.dpdk.org/comment/960/", "msgid": "<1855491.X6obi8qxOV@xps13>", "list_archive_url": "https://inbox.dpdk.org/dev/1855491.X6obi8qxOV@xps13", "date": "2014-09-24T15:18:09", "subject": "Re: [dpdk-dev] [memnic PATCH 4/7] pmd: use compiler barrier", "submitter": { "id": 1, "url": "http://patches.dpdk.org/api/people/1/?format=api", "name": "Thomas Monjalon", "email": "thomas.monjalon@6wind.com" }, "content": "2014-09-11 07:48, Hiroshi Shimamoto:\n> x86 can keep store ordering with standard operations.\n\nAre we sure it's always the case (including old 32-bit CPU)?\nI would prefer to have a reference here. I know we already discussed\nthis kind of things but having a reference in commit log could help\nfor future discussions.\n\n> Using memory barrier is much expensive in main packet processing loop.\n> Removing this improves xmit/recv packet performance.\n> \n> We can see performance improvements with memnic-tester.\n> Using Xeon E5-2697 v2 @ 2.70GHz, 4 vCPU.\n> size | before | after\n> 64 | 4.18Mpps | 4.59Mpps\n> 128 | 3.85Mpps | 4.87Mpps\n> 256 | 4.01Mpps | 4.72Mpps\n> 512 | 3.52Mpps | 4.41Mpps\n> 1024 | 3.18Mpps | 3.64Mpps\n> 1280 | 2.86Mpps | 3.15Mpps\n> 1518 | 2.59Mpps | 2.87Mpps\n> \n> Note: we have to take care if we use temporal cache.\n\nPlease, could you explain this last sentence?\n\nThanks", "headers": { "Return-Path": "<dev-bounces@dpdk.org>", "X-Original-To": "patchwork@dpdk.org", "Delivered-To": "patchwork@dpdk.org", "Received": [ "from [92.243.14.124] (localhost [IPv6:::1])\n\tby dpdk.org (Postfix) with ESMTP id 2D4ABAFCA;\n\tWed, 24 Sep 2014 17:12:06 +0200 (CEST)", "from mail-wi0-f181.google.com (mail-wi0-f181.google.com\n\t[209.85.212.181]) by dpdk.org (Postfix) with ESMTP id DAF6A6896\n\tfor <dev@dpdk.org>; Wed, 24 Sep 2014 17:12:04 +0200 (CEST)", "by mail-wi0-f181.google.com with SMTP id z2so7514940wiv.8\n\tfor <dev@dpdk.org>; Wed, 24 Sep 2014 08:18:18 -0700 (PDT)", "from xps13.localnet (136-92-190-109.dsl.ovh.fr. [109.190.92.136])\n\tby mx.google.com with ESMTPSA id\n\tec18sm57685wic.0.2014.09.24.08.18.16 for <multiple recipients>\n\t(version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128);\n\tWed, 24 Sep 2014 08:18:17 -0700 (PDT)" ], "X-Google-DKIM-Signature": "v=1; a=rsa-sha256; c=relaxed/relaxed;\n\td=1e100.net; s=20130820;\n\th=x-gm-message-state:from:to:cc:subject:date:message-id:organization\n\t:user-agent:in-reply-to:references:mime-version\n\t:content-transfer-encoding:content-type;\n\tbh=ylZSqgr6HPPpj6ouNtb0BrG2Nb5jcct/iZhA0md9BvI=;\n\tb=XX4lzPNzD1mS7oSzHjfgb7tUHzvF2Fw6moXeNz9yqoFlWuaC7Ml0NslmOOwC0YmqT+\n\tFVt8eqX4Y6SDkwAXpaGP+CSlyUiCr8t7N+tw1fwrWevqqfQmh0pq81REILBmfVItrjQ5\n\twDxm1+LXUJYA920nKrVh3dh8vG+l6pQZ4A6pgEyoUHHRZWNQQy2epKxLqbpvXIAu7BhK\n\tfxEiO8SgVpkcYD0tD0O0xuwcfvrwaRkA0iqUX04Y72B0OvZBPH5STkvzSB/ZJGGxEOJu\n\tm4hfbxUlI6fVAA3czZgwW7j6ij5t/LD9udsEimcvSEwRlZnjrprxwWRuJ0mYx+z5zlsQ\n\tPWQA==", "X-Gm-Message-State": "ALoCoQlf7HXrKxV3O+QobFkAcANdRFgrFGzxUENcev4t9Ms6EZFLI96kWDVa21GyXpNsnPsdaX8K", "X-Received": "by 10.180.83.195 with SMTP id s3mr32728934wiy.12.1411571897907; \n\tWed, 24 Sep 2014 08:18:17 -0700 (PDT)", "From": "Thomas Monjalon <thomas.monjalon@6wind.com>", "To": "dev@dpdk.org, Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>", "Date": "Wed, 24 Sep 2014 17:18:09 +0200", "Message-ID": "<1855491.X6obi8qxOV@xps13>", "Organization": "6WIND", "User-Agent": "KMail/4.13.3 (Linux/3.15.8-1-ARCH; KDE/4.13.3; x86_64; ; )", "In-Reply-To": "<7F861DC0615E0C47A872E6F3C5FCDDBD011A9934@BPXM14GP.gisp.nec.co.jp>", "References": "<7F861DC0615E0C47A872E6F3C5FCDDBD011A9934@BPXM14GP.gisp.nec.co.jp>", "MIME-Version": "1.0", "Content-Transfer-Encoding": "7Bit", "Content-Type": "text/plain; charset=\"us-ascii\"", "Cc": "Hayato Momma <h-momma@ce.jp.nec.com>", "Subject": "Re: [dpdk-dev] [memnic PATCH 4/7] pmd: use compiler barrier", "X-BeenThere": "dev@dpdk.org", "X-Mailman-Version": "2.1.15", "Precedence": "list", "List-Id": "patches and discussions about DPDK <dev.dpdk.org>", "List-Unsubscribe": "<http://dpdk.org/ml/options/dev>,\n\t<mailto:dev-request@dpdk.org?subject=unsubscribe>", "List-Archive": "<http://dpdk.org/ml/archives/dev/>", "List-Post": "<mailto:dev@dpdk.org>", "List-Help": "<mailto:dev-request@dpdk.org?subject=help>", "List-Subscribe": "<http://dpdk.org/ml/listinfo/dev>,\n\t<mailto:dev-request@dpdk.org?subject=subscribe>", "Errors-To": "dev-bounces@dpdk.org", "Sender": "\"dev\" <dev-bounces@dpdk.org>" }, "addressed": null }, { "id": 978, "web_url": "http://patches.dpdk.org/comment/978/", "msgid": "<7F861DC0615E0C47A872E6F3C5FCDDBD02AD9F40@BPXM14GP.gisp.nec.co.jp>", "list_archive_url": "https://inbox.dpdk.org/dev/7F861DC0615E0C47A872E6F3C5FCDDBD02AD9F40@BPXM14GP.gisp.nec.co.jp", "date": "2014-09-25T00:35:51", "subject": "Re: [dpdk-dev] [memnic PATCH 4/7] pmd: use compiler barrier", "submitter": { "id": 38, "url": "http://patches.dpdk.org/api/people/38/?format=api", "name": "Hiroshi Shimamoto", "email": "h-shimamoto@ct.jp.nec.com" }, "content": "> Subject: Re: [dpdk-dev] [memnic PATCH 4/7] pmd: use compiler barrier\n> \n> 2014-09-11 07:48, Hiroshi Shimamoto:\n> > x86 can keep store ordering with standard operations.\n> \n> Are we sure it's always the case (including old 32-bit CPU)?\n> I would prefer to have a reference here. I know we already discussed\n> this kind of things but having a reference in commit log could help\n> for future discussions.\n> \n> > Using memory barrier is much expensive in main packet processing loop.\n> > Removing this improves xmit/recv packet performance.\n> >\n> > We can see performance improvements with memnic-tester.\n> > Using Xeon E5-2697 v2 @ 2.70GHz, 4 vCPU.\n> > size | before | after\n> > 64 | 4.18Mpps | 4.59Mpps\n> > 128 | 3.85Mpps | 4.87Mpps\n> > 256 | 4.01Mpps | 4.72Mpps\n> > 512 | 3.52Mpps | 4.41Mpps\n> > 1024 | 3.18Mpps | 3.64Mpps\n> > 1280 | 2.86Mpps | 3.15Mpps\n> > 1518 | 2.59Mpps | 2.87Mpps\n> >\n> > Note: we have to take care if we use temporal cache.\n> \n> Please, could you explain this last sentence?\n\nOops, I have mistaken the word, \"temporal\" should be \"non-temporal\".\n\nBy the way, there are some instructions which use non-temporal\ncache liek MOVNTx series.\nThe store ordering of these instructions is not kept.\n\nRef. Intel Software Developer Manual\n Vol.1 10.4.6.2 Caching of Temporal vs. Non-Temporal Data\n Vol.3 8.2 Memory Ordering\n\nthanks,\nHiroshi\n\n> \n> Thanks\n> --\n> Thomas", "headers": { "Return-Path": "<dev-bounces@dpdk.org>", "X-Original-To": "patchwork@dpdk.org", "Delivered-To": "patchwork@dpdk.org", "Received": [ "from [92.243.14.124] (localhost [IPv6:::1])\n\tby dpdk.org (Postfix) with ESMTP id 9BCD45922;\n\tThu, 25 Sep 2014 02:30:01 +0200 (CEST)", "from tyo202.gate.nec.co.jp (TYO202.gate.nec.co.jp [210.143.35.52])\n\tby dpdk.org (Postfix) with ESMTP id 815E01F5\n\tfor <dev@dpdk.org>; Thu, 25 Sep 2014 02:29:59 +0200 (CEST)", "from mailgate3.nec.co.jp ([10.7.69.197])\n\tby tyo202.gate.nec.co.jp (8.13.8/8.13.4) with ESMTP id s8P0aA9E005985;\n\tThu, 25 Sep 2014 09:36:10 +0900 (JST)", "from mailsv4.nec.co.jp (imss61.nec.co.jp [10.7.69.156]) by\n\tmailgate3.nec.co.jp (8.11.7/3.7W-MAILGATE-NEC) with ESMTP\n\tid s8P0aAR11550; Thu, 25 Sep 2014 09:36:10 +0900 (JST)", "from mail01b.kamome.nec.co.jp (mail01b.kamome.nec.co.jp\n\t[10.25.43.2])\n\tby mailsv4.nec.co.jp (8.13.8/8.13.4) with ESMTP id s8P0ZrrU021977;\n\tThu, 25 Sep 2014 09:36:10 +0900 (JST)", "from bpxc99gp.gisp.nec.co.jp ([10.38.151.145] [10.38.151.145]) by\n\tmail01b.kamome.nec.co.jp with ESMTP id BT-MMP-2085628;\n\tThu, 25 Sep 2014 09:35:52 +0900", "from BPXM14GP.gisp.nec.co.jp ([169.254.1.136]) by\n\tBPXC17GP.gisp.nec.co.jp ([10.38.151.145]) with mapi id 14.03.0174.002;\n\tThu, 25 Sep 2014 09:35:51 +0900" ], "From": "Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>", "To": "Thomas Monjalon <thomas.monjalon@6wind.com>,\n\t\"dev@dpdk.org\" <dev@dpdk.org>", "Thread-Topic": "[dpdk-dev] [memnic PATCH 4/7] pmd: use compiler barrier", "Thread-Index": "AQHP2ArG85sYq/F0GkOG2kfCJWfWLpwQ+Zlg", "Date": "Thu, 25 Sep 2014 00:35:51 +0000", "Message-ID": "<7F861DC0615E0C47A872E6F3C5FCDDBD02AD9F40@BPXM14GP.gisp.nec.co.jp>", "References": "<7F861DC0615E0C47A872E6F3C5FCDDBD011A9934@BPXM14GP.gisp.nec.co.jp>\n\t<1855491.X6obi8qxOV@xps13>", "In-Reply-To": "<1855491.X6obi8qxOV@xps13>", "Accept-Language": "ja-JP, en-US", "Content-Language": "ja-JP", "X-MS-Has-Attach": "", "X-MS-TNEF-Correlator": "", "x-originating-ip": "[10.205.5.123]", "Content-Type": "text/plain; charset=\"iso-2022-jp\"", "Content-Transfer-Encoding": "quoted-printable", "MIME-Version": "1.0", "Cc": "Hayato Momma <h-momma@ce.jp.nec.com>", "Subject": "Re: [dpdk-dev] [memnic PATCH 4/7] pmd: use compiler barrier", "X-BeenThere": "dev@dpdk.org", "X-Mailman-Version": "2.1.15", "Precedence": "list", "List-Id": "patches and discussions about DPDK <dev.dpdk.org>", "List-Unsubscribe": "<http://dpdk.org/ml/options/dev>,\n\t<mailto:dev-request@dpdk.org?subject=unsubscribe>", "List-Archive": "<http://dpdk.org/ml/archives/dev/>", "List-Post": "<mailto:dev@dpdk.org>", "List-Help": "<mailto:dev-request@dpdk.org?subject=help>", "List-Subscribe": "<http://dpdk.org/ml/listinfo/dev>,\n\t<mailto:dev-request@dpdk.org?subject=subscribe>", "Errors-To": "dev-bounces@dpdk.org", "Sender": "\"dev\" <dev-bounces@dpdk.org>" }, "addressed": null } ]