Patch Comment List
List patch comments
GET /api/patches/308/comments/?format=api
[ { "id": 677, "web_url": "https://patches.dpdk.org/comment/677/", "msgid": "<540D7C5F.8000406@6wind.com>", "list_archive_url": "https://inbox.dpdk.org/dev/540D7C5F.8000406@6wind.com", "date": "2014-09-08T09:52:31", "subject": "Re: [dpdk-dev] [PATCH 01/13] mbuf: replace data pointer by an offset", "submitter": { "id": 8, "url": "https://patches.dpdk.org/api/people/8/?format=api", "name": "Olivier Matz", "email": "olivier.matz@6wind.com" }, "content": "Hi Bruce,\n\nOn 09/03/2014 05:49 PM, Bruce Richardson wrote:\n> From: Olivier Matz <olivier.matz@6wind.com>\n> \n> Original patch:\n> The mbuf structure already contains a pointer to the beginning of the\n> buffer (m->buf_addr). It is not needed to use 8 bytes again to store\n> another pointer to the beginning of the data.\n> \n> Using a 16 bits unsigned integer is enough as we know that a mbuf is\n> never longer than 64KB. We gain 6 bytes in the structure thanks to\n> this modification.\n> \n> Signed-off-by: Olivier Matz <olivier.matz@6wind.com>\n> \n> This version:\n> * Updated original patch to apply to latest on mainline.\n> * Disabled vector PMD in config as it relies heavily on the mbuf layout\n> This will be re-enabled in a subsequent commit once vPMD has been\n> reworked to take account of mbuf changes.\n> \n> Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>\n>\n> [...]\n>\n> diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h\n> index 32e8474..669e7f5 100644\n> --- a/lib/librte_mbuf/rte_mbuf.h\n> +++ b/lib/librte_mbuf/rte_mbuf.h\n> @@ -119,6 +119,13 @@ struct rte_mbuf {\n> \tvoid *buf_addr; /**< Virtual address of segment buffer. */\n> \tphys_addr_t buf_physaddr; /**< Physical address of segment buffer. */\n> \tuint16_t buf_len; /**< Length of segment buffer. */\n> +\n> +\t/* valid for any segment */\n> +\tstruct rte_mbuf *next; /**< Next segment of scattered packet. */\n> +\tuint16_t data_off;\n> +\tuint16_t data_len; /**< Amount of data in segment buffer. */\n> +\tuint32_t pkt_len; /**< Total pkt len: sum of all segments. */\n> +\n\nHere, the compiler will add some padding between \"buf_len\" and \"next\"\n(6 bytes in 64bits, 2 in 32bits). Shouldn't we add \"reserved\" fields\nor at least a comment ? Another idea would be to reorganize the fields\nto avoid padding, if possible.\n\nExcept this remarq, Acked-by: Olivier Matz <olivier.matz@6wind.com>", "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 B3DCC68A8;\n\tMon, 8 Sep 2014 11:47:51 +0200 (CEST)", "from mail.droids-corp.org (zoll.droids-corp.org [94.23.50.67])\n\tby dpdk.org (Postfix) with ESMTP id 86A6338EB\n\tfor <dev@dpdk.org>; Mon, 8 Sep 2014 11:47:42 +0200 (CEST)", "from was59-1-82-226-113-214.fbx.proxad.net ([82.226.113.214]\n\thelo=[192.168.0.10]) by mail.droids-corp.org with esmtpsa\n\t(TLS1.0:DHE_RSA_AES_128_CBC_SHA1:128)\n\t(Exim 4.80) (envelope-from <olivier.matz@6wind.com>)\n\tid 1XQvfF-0007gA-Cg; Mon, 08 Sep 2014 11:55:17 +0200" ], "Message-ID": "<540D7C5F.8000406@6wind.com>", "Date": "Mon, 08 Sep 2014 11:52:31 +0200", "From": "Olivier MATZ <olivier.matz@6wind.com>", "User-Agent": "Mozilla/5.0 (X11; Linux x86_64;\n\trv:24.0) Gecko/20100101 Icedove/24.5.0", "MIME-Version": "1.0", "To": "Bruce Richardson <bruce.richardson@intel.com>, dev@dpdk.org", "References": "<1409759378-10113-1-git-send-email-bruce.richardson@intel.com>\n\t<1409759378-10113-2-git-send-email-bruce.richardson@intel.com>", "In-Reply-To": "<1409759378-10113-2-git-send-email-bruce.richardson@intel.com>", "Content-Type": "text/plain; charset=ISO-8859-1", "Content-Transfer-Encoding": "7bit", "Subject": "Re: [dpdk-dev] [PATCH 01/13] mbuf: replace data pointer by an offset", "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": 678, "web_url": "https://patches.dpdk.org/comment/678/", "msgid": "<540D7D1E.1020500@6wind.com>", "list_archive_url": "https://inbox.dpdk.org/dev/540D7D1E.1020500@6wind.com", "date": "2014-09-08T09:55:42", "subject": "Re: [dpdk-dev] [PATCH 01/13] mbuf: replace data pointer by an offset", "submitter": { "id": 8, "url": "https://patches.dpdk.org/api/people/8/?format=api", "name": "Olivier Matz", "email": "olivier.matz@6wind.com" }, "content": ">> diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h\n>> index 32e8474..669e7f5 100644\n>> --- a/lib/librte_mbuf/rte_mbuf.h\n>> +++ b/lib/librte_mbuf/rte_mbuf.h\n>> @@ -119,6 +119,13 @@ struct rte_mbuf {\n>> \tvoid *buf_addr; /**< Virtual address of segment buffer. */\n>> \tphys_addr_t buf_physaddr; /**< Physical address of segment buffer. */\n>> \tuint16_t buf_len; /**< Length of segment buffer. */\n>> +\n>> +\t/* valid for any segment */\n>> +\tstruct rte_mbuf *next; /**< Next segment of scattered packet. */\n>> +\tuint16_t data_off;\n>> +\tuint16_t data_len; /**< Amount of data in segment buffer. */\n>> +\tuint32_t pkt_len; /**< Total pkt len: sum of all segments. */\n>> +\n> \n> Here, the compiler will add some padding between \"buf_len\" and \"next\"\n> (6 bytes in 64bits, 2 in 32bits). Shouldn't we add \"reserved\" fields\n> or at least a comment ? Another idea would be to reorganize the fields\n> to avoid padding, if possible.\n\nI've just seen that patch 02/13 reorders the mbuf fields. So maybe this\nremarq is useless.\n\n\nOlivier", "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 C247BB392;\n\tMon, 8 Sep 2014 11:50:55 +0200 (CEST)", "from mail.droids-corp.org (zoll.droids-corp.org [94.23.50.67])\n\tby dpdk.org (Postfix) with ESMTP id B508868A8\n\tfor <dev@dpdk.org>; Mon, 8 Sep 2014 11:50:53 +0200 (CEST)", "from was59-1-82-226-113-214.fbx.proxad.net ([82.226.113.214]\n\thelo=[192.168.0.10]) by mail.droids-corp.org with esmtpsa\n\t(TLS1.0:DHE_RSA_AES_128_CBC_SHA1:128)\n\t(Exim 4.80) (envelope-from <olivier.matz@6wind.com>)\n\tid 1XQviK-0007gP-KA; Mon, 08 Sep 2014 11:58:27 +0200" ], "Message-ID": "<540D7D1E.1020500@6wind.com>", "Date": "Mon, 08 Sep 2014 11:55:42 +0200", "From": "Olivier MATZ <olivier.matz@6wind.com>", "User-Agent": "Mozilla/5.0 (X11; Linux x86_64;\n\trv:24.0) Gecko/20100101 Icedove/24.5.0", "MIME-Version": "1.0", "To": "Bruce Richardson <bruce.richardson@intel.com>, dev@dpdk.org", "References": "<1409759378-10113-1-git-send-email-bruce.richardson@intel.com>\n\t<1409759378-10113-2-git-send-email-bruce.richardson@intel.com>\n\t<540D7C5F.8000406@6wind.com>", "In-Reply-To": "<540D7C5F.8000406@6wind.com>", "Content-Type": "text/plain; charset=ISO-8859-1", "Content-Transfer-Encoding": "7bit", "Subject": "Re: [dpdk-dev] [PATCH 01/13] mbuf: replace data pointer by an offset", "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 } ]