From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga01.intel.com (mga01.intel.com [192.55.52.88]) by dpdk.org (Postfix) with ESMTP id 53B0D6C94 for ; Mon, 29 Feb 2016 12:30:31 +0100 (CET) Received: from orsmga001.jf.intel.com ([10.7.209.18]) by fmsmga101.fm.intel.com with ESMTP; 29 Feb 2016 03:30:30 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.22,520,1449561600"; d="scan'208";a="897357232" Received: from sie-lab-212-120.ir.intel.com (HELO silpixa00394367.ir.intel.com) ([10.237.212.120]) by orsmga001.jf.intel.com with ESMTP; 29 Feb 2016 03:30:29 -0800 From: Harry van Haaren To: john.mcnamara@intel.com Date: Mon, 29 Feb 2016 11:29:58 +0000 Message-Id: <1456745398-26996-1-git-send-email-harry.van.haaren@intel.com> X-Mailer: git-send-email 2.5.0 Cc: dev@dpdk.org Subject: [dpdk-dev] [PATCH] doc/nic: add ixgbe statistics on read frequency X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 29 Feb 2016 11:30:31 -0000 This patch adds a note to the ixgbe PMD guide, stating the minimum time that statistics must be polled from the hardware in order to avoid register values becoming saturated and "sticking" to the max value. Signed-off-by: Harry van Haaren --- doc/guides/nics/ixgbe.rst | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) diff --git a/doc/guides/nics/ixgbe.rst b/doc/guides/nics/ixgbe.rst index 8cae299..c8085a8 100644 --- a/doc/guides/nics/ixgbe.rst +++ b/doc/guides/nics/ixgbe.rst @@ -178,3 +178,27 @@ load_balancer As in the case of l3fwd, set configure port_conf.rxmode.hw_ip_checksum=0 to enable vPMD. In addition, for improved performance, use -bsz "(32,32),(64,64),(32,32)" in load_balancer to avoid using the default burst size of 144. + +Statistics +---------- + +The statistics of ixgbe hardware must be polled regularly in order for it to +remain consistent. Running a DPDK application without polling the statistcs will +cause registers on hardware to count to thier maxiumum value, and "stick" at +that value. + +In order to avoid statistic registers every reaching thier maxiumum value, +read the statistics from the hardware using ``rte_eth_stats_get()`` or +``rte_eth_xstats_get()``. + +The maxiumum time between statistics polls that ensures consistent results can +be calculated as follows: + +.. code-block:: c + + max_read_interval = UINT_MAX / max_packets_per_second + max_read_interval = 4294967295 / 14880952 + max_read_interval = 288.6218096127183 (seconds) + max_read_interval = ~4 mins 48 sec. + +In order to ensure valid results, it is recommended to poll every 4 minutes. -- 2.5.0