From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from dpdk.org (dpdk.org [92.243.14.124]) by inbox.dpdk.org (Postfix) with ESMTP id C5945A04A5; Wed, 17 Jun 2020 18:45:46 +0200 (CEST) Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id 487F21252; Wed, 17 Jun 2020 18:45:46 +0200 (CEST) Received: from mga01.intel.com (mga01.intel.com [192.55.52.88]) by dpdk.org (Postfix) with ESMTP id 0D1921023 for ; Wed, 17 Jun 2020 18:45:43 +0200 (CEST) IronPort-SDR: QKnzga7qHv0hsUQEYdmQ11l8KkUU2FRbgtX0PbbOi2zknmp1XBKOhLe3Jy8ITPN487Gfms4M1n sWvvUsDkgg/A== X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga005.jf.intel.com ([10.7.209.41]) by fmsmga101.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 17 Jun 2020 09:45:38 -0700 IronPort-SDR: QF1KHWSV0Thvb3PRLjTAyyywKjlrDUFU8rbMtuXFpj6meD0Pys4msxFL8vlPNfirqOXzzWJAqG wrb3ZL9NNhjg== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.73,523,1583222400"; d="scan'208";a="450313730" Received: from fyigit-mobl.ger.corp.intel.com (HELO [10.213.242.251]) ([10.213.242.251]) by orsmga005.jf.intel.com with ESMTP; 17 Jun 2020 09:45:36 -0700 To: i.dyukov@samsung.com, dev@dpdk.org, v.kuramshin@samsung.com, thomas@monjalon.net, david.marchand@redhat.com, arybchenko@solarflare.com, wei.zhao1@intel.com, jia.guo@intel.com, beilei.xing@intel.com, qiming.yang@intel.com, wenzhuo.lu@intel.com References: <20200427095737.11082-1-i.dyukov@samsung.com> <20200615090158.18912-1-i.dyukov@samsung.com> <20200615090158.18912-3-i.dyukov@samsung.com> From: Ferruh Yigit Autocrypt: addr=ferruh.yigit@intel.com; prefer-encrypt=mutual; keydata= mQINBFXZCFABEADCujshBOAaqPZpwShdkzkyGpJ15lmxiSr3jVMqOtQS/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++O7QARAQABtCVGZXJydWggWWln aXQgPGZlcnJ1aC55aWdpdEBpbnRlbC5jb20+iQJsBBMBCgBWAhsDAh4BAheABQsJCAcDBRUK CQgLBRYCAwEABQkKqZZ8FiEE0jZTh0IuwoTjmYHH+TPrQ98TYR8FAl6ha3sXGHZrczovL2tl eXMub3BlbnBncC5vcmcACgkQ+TPrQ98TYR8uLA//QwltuFliUWe60xwmu9sY38c1DXvX67wk UryQ1WijVdIoj4H8cf/s2KtyIBjc89R254KMEfJDao/LrXqJ69KyGKXFhFPlF3VmFLsN4XiT PSfxkx8s6kHVaB3O183p4xAqnnl/ql8nJ5ph9HuwdL8CyO5/7dC/MjZ/mc4NGq5O9zk3YRGO lvdZAp5HW9VKW4iynvy7rl3tKyEqaAE62MbGyfJDH3C/nV/4+mPc8Av5rRH2hV+DBQourwuC ci6noiDP6GCNQqTh1FHYvXaN4GPMHD9DX6LtT8Fc5mL/V9i9kEVikPohlI0WJqhE+vQHFzR2 1q5nznE+pweYsBi3LXIMYpmha9oJh03dJOdKAEhkfBr6n8BWkWQMMiwfdzg20JX0o7a/iF8H 4dshBs+dXdIKzPfJhMjHxLDFNPNH8zRQkB02JceY9ESEah3wAbzTwz+e/9qQ5OyDTQjKkVOo cxC2U7CqeNt0JZi0tmuzIWrfxjAUulVhBmnceqyMOzGpSCQIkvalb6+eXsC9V1DZ4zsHZ2Mx Hi+7pCksdraXUhKdg5bOVCt8XFmx1MX4AoV3GWy6mZ4eMMvJN2hjXcrreQgG25BdCdcxKgqp e9cMbCtF+RZax8U6LkAWueJJ1QXrav1Jk5SnG8/5xANQoBQKGz+yFiWcgEs9Tpxth15o2v59 gXK5Ag0EV9ZMvgEQAKc0Db17xNqtSwEvmfp4tkddwW9XA0tWWKtY4KUdd/jijYqc3fDD54ES YpV8QWj0xK4YM0dLxnDU2IYxjEshSB1TqAatVWz9WtBYvzalsyTqMKP3w34FciuL7orXP4Ai bPtrHuIXWQOBECcVZTTOdZYGAzaYzxiAONzF9eTiwIqe9/oaOjTwTLnOarHt16QApTYQSnxD UQljeNvKYt1lZE/gAUUxNLWsYyTT+22/vU0GDUahsJxs1+f1yEr+OGrFiEAmqrzpF0lCS3f/ 3HVTU6rS9cK3glVUeaTF4+1SK5ZNO35piVQCwphmxa+dwTG/DvvHYCtgOZorTJ+OHfvCnSVj sM4kcXGjJPy3JZmUtyL9UxEbYlrffGPQI3gLXIGD5AN5XdAXFCjjaID/KR1c9RHd7Oaw0Pdc q9UtMLgM1vdX8RlDuMGPrj5sQrRVbgYHfVU/TQCk1C9KhzOwg4Ap2T3tE1umY/DqrXQgsgH7 1PXFucVjOyHMYXXugLT8YQ0gcBPHy9mZqw5mgOI5lCl6d4uCcUT0l/OEtPG/rA1lxz8ctdFB VOQOxCvwRG2QCgcJ/UTn5vlivul+cThi6ERPvjqjblLncQtRg8izj2qgmwQkvfj+h7Ex88bI 8iWtu5+I3K3LmNz/UxHBSWEmUnkg4fJlRr7oItHsZ0ia6wWQ8lQnABEBAAGJAjwEGAEKACYC GwwWIQTSNlOHQi7ChOOZgcf5M+tD3xNhHwUCXqFrngUJCKxSYAAKCRD5M+tD3xNhH3YWD/9b cUiWaHJasX+OpiuZ1Li5GG3m9aw4lR/k2lET0UPRer2Jy1JsL+uqzdkxGvPqzFTBXgx/6Byz EMa2mt6R9BCyR286s3lxVS5Bgr5JGB3EkpPcoJT3A7QOYMV95jBiiJTy78Qdzi5LrIu4tW6H o0MWUjpjdbR01cnj6EagKrDx9kAsqQTfvz4ff5JIFyKSKEHQMaz1YGHyCWhsTwqONhs0G7V2 0taQS1bGiaWND0dIBJ/u0pU998XZhmMzn765H+/MqXsyDXwoHv1rcaX/kcZIcN3sLUVcbdxA WHXOktGTQemQfEpCNuf2jeeJlp8sHmAQmV3dLS1R49h0q7hH4qOPEIvXjQebJGs5W7s2vxbA 5u5nLujmMkkfg1XHsds0u7Zdp2n200VC4GQf8vsUp6CSMgjedHeF9zKv1W4lYXpHp576ZV7T GgsEsvveAE1xvHnpV9d7ZehPuZfYlP4qgo2iutA1c0AXZLn5LPcDBgZ+KQZTzm05RU1gkx7n gL9CdTzVrYFy7Y5R+TrE9HFUnsaXaGsJwOB/emByGPQEKrupz8CZFi9pkqPuAPwjN6Wonokv ChAewHXPUadcJmCTj78Oeg9uXR6yjpxyFjx3vdijQIYgi5TEGpeTQBymLANOYxYWYOjXk+ae dYuOYKR9nbPv+2zK9pwwQ2NXbUBystaGyQ== Message-ID: <8d238174-0419-1524-95ad-a1434f848e0d@intel.com> Date: Wed, 17 Jun 2020 17:45:35 +0100 MIME-Version: 1.0 In-Reply-To: <20200615090158.18912-3-i.dyukov@samsung.com> Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 8bit Subject: Re: [dpdk-dev] [PATCH v3 2/7] ethdev: add a link status text representation 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: , Errors-To: dev-bounces@dpdk.org Sender: "dev" On 6/15/2020 10:01 AM, Ivan Dyukov wrote: > This commit add function which treat link status structure > and format it to text representation. If I am following correctly, the initial need was to escape from speed checks everytime loging link information caused by this new 'unknown' speed. And later suggestion was to have a pre-formatted text for link logging. This patch brings additional link status printing/formatting capability with custom format string support and with new format specifiers for link (like, '%D' link duplex state), although this is nice work and thanks for it, I am not sure this complexity and two new APIs are really needed. For me only 'rte_eth_link_format()' without custom format support looks good enough but I won't object if the concensus is to have them. I am aware there are multiple applications you are updating logging slightly different which requires this flexibility but what happens if they use same pre-formatted text, is that difference really required or happened by time based on developers taste? I will put some comments below in any case. > > Signed-off-by: Ivan Dyukov <...> > @@ -249,6 +249,9 @@ SRCS-$(CONFIG_RTE_LIBRTE_SECURITY) += test_security.c > > SRCS-$(CONFIG_RTE_LIBRTE_IPSEC) += test_ipsec.c test_ipsec_perf.c > SRCS-$(CONFIG_RTE_LIBRTE_IPSEC) += test_ipsec_sad.c > + > +SRCS-$(CONFIG_RTE_LIBRTE_ETHER) += test_ethdev_link.c +1 to unit test. <...> > +int > +rte_eth_link_printf(const char *const fmt, > + struct rte_eth_link *link) > +{ > + char text[200]; > + int ret; > + ret = rte_eth_link_format(text, 200, fmt, link); Will it be paranoid to add "text[199] = 0" to be sure any custom 'fmt' won't cause any harm? > + printf("%s", text); Not sure if the error still should be printed on error case? Like for example what the code does when fmt="%X"? > + return ret; > +} > + > +int > +rte_eth_link_format(char *str, int32_t len, const char *const fmt, > + struct rte_eth_link *link) Why not have the 'len' type 'size_t'? > + int offset = 0; > + int32_t clen = len; > + const char *fmt_cur = fmt; > + double gbits = (double)link->link_speed / 1000.; > + /* TBD: make it international? */ > + static const char LINK_DOWN_STR[] = "Link down"; > + static const char LINK_UP_STR[] = "Link up at "; > + static const char UNKNOWN_SPEED_STR[] = "Unknown speed"; > + static const char MBITS_STR[] = "Mbit/s"; > + static const char GBITS_STR[] = "Gbit/s"; > + static const char AUTONEG_STR[] = "Autoneg"; > + static const char FIXED_STR[] = "Fixed"; > + static const char FDX_STR[] = "FDX"; > + static const char HDX_STR[] = "HDX"; > + static const char UNKNOWN_STR[] = "Unknown"; > + static const char UP_STR[] = "Up"; > + static const char DOWN_STR[] = "Down"; > + if (str == NULL || len == 0) > + return -1; > + /* default format string, if no fmt is specified */ > + if (fmt == NULL) { > + if (link->link_status == ETH_LINK_DOWN) > + return snprintf(str, (size_t)clen, "%s", LINK_DOWN_STR); > + > + offset = snprintf(str, (size_t)clen, "%s", LINK_UP_STR); > + if (offset < 0 || (clen - offset) <= 0) > + return -1; > + clen -= offset; > + str += offset; > + if (link->link_speed == ETH_SPEED_NUM_UNKNOWN) { > + offset = snprintf(str, clen, "%s", > + UNKNOWN_SPEED_STR); > + if (offset < 0 || (clen - offset) <= 0) > + return -1; better to use 'strlcpy' & 'strlcat', they are easier to use for these kind of checks. <...> > + /* Formated status */ > + } else { > + char c = *fmt_cur; > + while (c) { > + if (clen <= 0) > + return -1; > + if (c == '%') { > + c = *++fmt_cur; > + switch (c) { > + /* Speed in Mbits/s */ > + case 'M': > + if (link->link_speed == > + ETH_SPEED_NUM_UNKNOWN) > + offset = snprintf(str, > + clen, "%s", > + UNKNOWN_STR); > + else > + offset = snprintf(str, > + clen, "%u", > + link->link_speed); Code readiblity is not great here because you hit the 80char limit, this is a sign that something is wrong like function is already too long. Can you please try to fix this, like extracting some part of the code to its own function or return after end of the 'if' statement which can save one more level indentation etc... <...> > diff --git a/lib/librte_ethdev/rte_ethdev.h b/lib/librte_ethdev/rte_ethdev.h > index 2090af501..83291e656 100644 > --- a/lib/librte_ethdev/rte_ethdev.h > +++ b/lib/librte_ethdev/rte_ethdev.h > @@ -2295,6 +2295,58 @@ int rte_eth_link_get(uint16_t port_id, struct rte_eth_link *link); > */ > int rte_eth_link_get_nowait(uint16_t port_id, struct rte_eth_link *link); > > + > +/** > + * print formated link status to stdout. This function threats all > + * special values like ETH_SPEED_NUM_UNKNOWN, ETH_LINK_DOWN etc. and convert > + * them to textual representation. > + * > + * @param fmt > + * Format string which allow to format link status. If NULL is provided > + * , default formating will be applied. > + * Following specifiers are available: > + * - '%M' link speed in Mbits/s > + * - '%G' link speed in Gbits/s > + * - '%S' link status. e.g. Up or Down > + * - '%A' link autonegotiation state > + * - '%D' link duplex state > + * @param link > + * Link status provided by rte_eth_link_get function > + * @return > + * - Number of bytes written to stdout. In case of error, -1 is returned. Does it worth to mention the log still will be printed on error? > + * > + */ > +int rte_eth_link_printf(const char *const fmt, > + struct rte_eth_link *link); > + > +/** > + * Format link status to textual representation. This function threats all > + * special values like ETH_SPEED_NUM_UNKNOWN, ETH_LINK_DOWN etc. and convert > + * them to textual representation. > + * > + * @param str > + * A pointer to a string to be filled with textual representation of > + * device status. > + * @param len > + * Length of available memory at 'str' string. > + * @param fmt > + * Format string which allow to format link status. If NULL is provided > + * , default formating will be applied. > + * Following specifiers are available: > + * - '%M' link speed in Mbits/s > + * - '%G' link speed in Gbits/s > + * - '%S' link status. e.g. Up or Down > + * - '%A' link autonegotiation state > + * - '%D' link duplex state > + * @param link > + * Link status provided by rte_eth_link_get function > + * @return > + * - Number of bytes written to str array. In case of error, -1 is returned. > + * > + */ > +int rte_eth_link_format(char *str, int32_t len, const char *const fmt, > + struct rte_eth_link *eth_link); > + These new APIs needs to be experimental by process (__rte_experimental). Need the add these APIs to the .map file (rte_ethdev_version.map), so that they will be exported in the dynamic library (.so).