From patchwork Mon Feb 22 06:31:55 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Zhang, Helin" X-Patchwork-Id: 10727 X-Patchwork-Delegate: thomas@monjalon.net Return-Path: X-Original-To: patchwork@dpdk.org Delivered-To: patchwork@dpdk.org Received: from [92.243.14.124] (localhost [IPv6:::1]) by dpdk.org (Postfix) with ESMTP id 8A6CE2C36; Mon, 22 Feb 2016 14:10:47 +0100 (CET) Received: from mga14.intel.com (mga14.intel.com [192.55.52.115]) by dpdk.org (Postfix) with ESMTP id 8BBCB58D6 for ; Mon, 22 Feb 2016 07:32:08 +0100 (CET) Received: from orsmga003.jf.intel.com ([10.7.209.27]) by fmsmga103.fm.intel.com with ESMTP; 21 Feb 2016 22:32:07 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.22,483,1449561600"; d="scan'208";a="750735648" Received: from shvmail01.sh.intel.com ([10.239.29.42]) by orsmga003.jf.intel.com with ESMTP; 21 Feb 2016 22:32:06 -0800 Received: from shecgisg004.sh.intel.com (shecgisg004.sh.intel.com [10.239.29.89]) by shvmail01.sh.intel.com with ESMTP id u1M6W4tx023060; Mon, 22 Feb 2016 14:32:04 +0800 Received: from shecgisg004.sh.intel.com (localhost [127.0.0.1]) by shecgisg004.sh.intel.com (8.13.6/8.13.6/SuSE Linux 0.8) with ESMTP id u1M6W01V019140; Mon, 22 Feb 2016 14:32:02 +0800 Received: (from hzhan75@localhost) by shecgisg004.sh.intel.com (8.13.6/8.13.6/Submit) id u1M6W0aO019136; Mon, 22 Feb 2016 14:32:00 +0800 From: Helin Zhang To: dev@dpdk.org Date: Mon, 22 Feb 2016 14:31:55 +0800 Message-Id: <1456122717-19099-2-git-send-email-helin.zhang@intel.com> X-Mailer: git-send-email 1.7.4.1 In-Reply-To: <1456122717-19099-1-git-send-email-helin.zhang@intel.com> References: <1450665486-8335-1-git-send-email-helin.zhang@intel.com> <1456122717-19099-1-git-send-email-helin.zhang@intel.com> Subject: [dpdk-dev] [PATCH v2 1/3] i40e: enable extended tag 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: , Errors-To: dev-bounces@dpdk.org Sender: "dev" PCIe feature of 'Extended Tag' is important for 40G performance. It adds its enabling during each port initialization, to ensure the high performance. Signed-off-by: Helin Zhang Acked-by: Jingjing Wu --- doc/guides/rel_notes/release_16_04.rst | 6 ++++ drivers/net/i40e/i40e_ethdev.c | 65 ++++++++++++++++++++++++++++++++-- 2 files changed, 68 insertions(+), 3 deletions(-) v2: - Changed the type of return value of i40e_enable_extended_tag() to 'void'. diff --git a/doc/guides/rel_notes/release_16_04.rst b/doc/guides/rel_notes/release_16_04.rst index 5786f74..bed5779 100644 --- a/doc/guides/rel_notes/release_16_04.rst +++ b/doc/guides/rel_notes/release_16_04.rst @@ -46,6 +46,12 @@ This section should contain new features added in this release. Sample format: * **Added vhost-user live migration support.** +* **i40e: Enabled extended tag.** + + It enabled extended tag by checking and writing corresponding PCI config + space bytes, to boost the performance. In the meanwhile, it deprecated the + legacy way via reading/writing sysfile supported by kernel module of igb_uio. + Resolved Issues --------------- diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c index ef24122..7e68c61 100644 --- a/drivers/net/i40e/i40e_ethdev.c +++ b/drivers/net/i40e/i40e_ethdev.c @@ -273,6 +273,17 @@ #define I40E_INSET_IPV6_TC_MASK 0x0009F00FUL #define I40E_INSET_IPV6_NEXT_HDR_MASK 0x000C00FFUL +/* PCI offset for querying capability */ +#define PCI_DEV_CAP_REG 0xA4 +/* PCI offset for enabling/disabling Extended Tag */ +#define PCI_DEV_CTRL_REG 0xA8 +/* Bit mask of Extended Tag capability */ +#define PCI_DEV_CAP_EXT_TAG_MASK 0x20 +/* Bit shift of Extended Tag enable/disable */ +#define PCI_DEV_CTRL_EXT_TAG_SHIFT 8 +/* Bit mask of Extended Tag enable/disable */ +#define PCI_DEV_CTRL_EXT_TAG_MASK (1 << PCI_DEV_CTRL_EXT_TAG_SHIFT) + static int eth_i40e_dev_init(struct rte_eth_dev *eth_dev); static int eth_i40e_dev_uninit(struct rte_eth_dev *eth_dev); static int i40e_dev_configure(struct rte_eth_dev *dev); @@ -386,7 +397,7 @@ static int i40e_dev_filter_ctrl(struct rte_eth_dev *dev, static int i40e_dev_get_dcb_info(struct rte_eth_dev *dev, struct rte_eth_dcb_info *dcb_info); static void i40e_configure_registers(struct i40e_hw *hw); -static void i40e_hw_init(struct i40e_hw *hw); +static void i40e_hw_init(struct rte_eth_dev *dev); static int i40e_config_qinq(struct i40e_hw *hw, struct i40e_vsi *vsi); static int i40e_mirror_rule_set(struct rte_eth_dev *dev, struct rte_eth_mirror_conf *mirror_conf, @@ -765,7 +776,7 @@ eth_i40e_dev_init(struct rte_eth_dev *dev) i40e_clear_hw(hw); /* Initialize the hardware */ - i40e_hw_init(hw); + i40e_hw_init(dev); /* Reset here to make sure all is clean for each PF */ ret = i40e_pf_reset(hw); @@ -7262,13 +7273,61 @@ i40e_dev_filter_ctrl(struct rte_eth_dev *dev, } /* + * Check and enable Extended Tag. + * Enabling Extended Tag is important for 40G performance. + */ +static void +i40e_enable_extended_tag(struct rte_eth_dev *dev) +{ + uint32_t buf = 0; + int ret; + + ret = rte_eal_pci_read_config(dev->pci_dev, &buf, sizeof(buf), + PCI_DEV_CAP_REG); + if (ret < 0) { + PMD_DRV_LOG(ERR, "Failed to read PCI offset 0x%x", + PCI_DEV_CAP_REG); + return; + } + if (!(buf & PCI_DEV_CAP_EXT_TAG_MASK)) { + PMD_DRV_LOG(ERR, "Does not support Extended Tag"); + return; + } + + buf = 0; + ret = rte_eal_pci_read_config(dev->pci_dev, &buf, sizeof(buf), + PCI_DEV_CTRL_REG); + if (ret < 0) { + PMD_DRV_LOG(ERR, "Failed to read PCI offset 0x%x", + PCI_DEV_CTRL_REG); + return; + } + if (buf & PCI_DEV_CTRL_EXT_TAG_MASK) { + PMD_DRV_LOG(DEBUG, "Extended Tag has already been enabled"); + return; + } + buf |= PCI_DEV_CTRL_EXT_TAG_MASK; + ret = rte_eal_pci_write_config(dev->pci_dev, &buf, sizeof(buf), + PCI_DEV_CTRL_REG); + if (ret < 0) { + PMD_DRV_LOG(ERR, "Failed to write PCI offset 0x%x", + PCI_DEV_CTRL_REG); + return; + } +} + +/* * As some registers wouldn't be reset unless a global hardware reset, * hardware initialization is needed to put those registers into an * expected initial state. */ static void -i40e_hw_init(struct i40e_hw *hw) +i40e_hw_init(struct rte_eth_dev *dev) { + struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private); + + i40e_enable_extended_tag(dev); + /* clear the PF Queue Filter control register */ I40E_WRITE_REG(hw, I40E_PFQF_CTL_0, 0);