From patchwork Wed Apr 24 13:21:35 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139646 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 7C5E643EAD; Wed, 24 Apr 2024 15:22:21 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 2F1D343480; Wed, 24 Apr 2024 15:22:16 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id 313E640273; Wed, 24 Apr 2024 15:22:14 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964935; x=1745500935; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=pH5TlTdGXAg9O85La1eq6EPtkLVnpxXM+bYQy6K+eRI=; b=Kl8MRG5lN4gBYQqpAYjRG4ikp/GVg0XD6r5xYPho86Hg8UhtpKQtpJt2 fJSW7UrFi+J6pZla/W/ycCA1dD5MrpcifZxBqPFmjn/Ihy8ePjx+7lCAn FpZKOz0Mllj8zW++AewFWn6s7aqQHbADoe6Ucg8NEl0aJQbMlOfn4YMm8 zeZZ+NhtAukmHz7d0/5MMCJ9A/55VXbPs3WJ48nTkdIbsivwWs3WEg4DW NDtHxSix0tOVIGbFC+Wc45INrMJHgOks+78Ymi+ZMjAO3JRvXZFJwnWDS R17kmkZ/Nk34dTuQ+dLNr8AMbO62eUSS7GVxZ+a4S6lAqR3gWaNnR6BTL Q==; X-CSE-ConnectionGUID: sc5YhMqYRESSBYtJ1XFRfQ== X-CSE-MsgGUID: N60uDQuAROOK/tChSqi7dw== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289235" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289235" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:14 -0700 X-CSE-ConnectionGUID: UkVyEm4RTI6m0sgBW4J9Ew== X-CSE-MsgGUID: ZiDlHWmZSByoEJFJ8pErpA== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749274" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:11 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Piotr Pietruszewski , vladimir.medvedkin@intel.com, bruce.richardson@intel.com, stable@dpdk.org, Skajewski@dpdk.org, PiotrX , Marek Mical , Kirsher@dpdk.org, Jeffrey T , Michael@dpdk.org, Alice Subject: [PATCH v1 01/22] net/ixgbe/base: revert remove default advertising for x550 2.5G/5G Date: Wed, 24 Apr 2024 14:21:35 +0100 Message-ID: <2accdfc6a96ea61acb30ef431e17c380dc538eba.1713964708.git.anatoly.burakov@intel.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Piotr Pietruszewski This reverts the following commit: 13de2444449e ("net/ixgbe/base: remove default advertising for x550 2.5G/5G") The commit removing advertising 2.5G and 5G speeds from ixgbe_get_copper_speeds_supported() was supposed to disable 2.5G and 5G during the autonegotiation with default settings. However, that change prevented OS-specific part of the driver from forcing these speeds at user request. This change reverts the commit and adds 2.5G and 5G as speeds supported by the X550. Cc: stable@dpdk.org Signed-off-by: Piotr Pietruszewski Reviewed-by: Skajewski, PiotrX Reviewed-by: Marek Mical Reviewed-by: Kirsher, Jeffrey T Reviewed-by: Michael, Alice --- drivers/net/ixgbe/base/ixgbe_phy.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/ixgbe/base/ixgbe_phy.c b/drivers/net/ixgbe/base/ixgbe_phy.c index 74c5db16fa..3a8e603472 100644 --- a/drivers/net/ixgbe/base/ixgbe_phy.c +++ b/drivers/net/ixgbe/base/ixgbe_phy.c @@ -915,6 +915,10 @@ static s32 ixgbe_get_copper_speeds_supported(struct ixgbe_hw *hw) hw->phy.speeds_supported |= IXGBE_LINK_SPEED_100_FULL; switch (hw->mac.type) { + case ixgbe_mac_X550: + hw->phy.speeds_supported |= IXGBE_LINK_SPEED_2_5GB_FULL; + hw->phy.speeds_supported |= IXGBE_LINK_SPEED_5GB_FULL; + break; case ixgbe_mac_X550EM_x: case ixgbe_mac_X550EM_a: hw->phy.speeds_supported &= ~IXGBE_LINK_SPEED_100_FULL; From patchwork Wed Apr 24 13:21:36 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139647 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 9187743EAD; Wed, 24 Apr 2024 15:22:29 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 1D4944348B; Wed, 24 Apr 2024 15:22:19 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id 0C74343483; Wed, 24 Apr 2024 15:22:16 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964937; x=1745500937; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=3+1q1rJsnAnHbPex4c0izxH0zCks4iPMXqmPPj95sEI=; b=Qn2YQGDryTdgfgzVdQ3fwe/A491j5T5FJILjhjEB44SdHum4SXHNaMja j0QP3GH7i712GlO4wD8ZV/xFO5w/50QgsAThN5cihGHEYi7LgjMniCYPN UMaGK+CdEcgByvnXGGNwLTwV9elYqm1iox78C33gaiUuwWD87yeTY8mXu yLizM0slV48lzipSO8RbOsxQdhe00NItfUKopwM9GlNKDV5CT5Ym2R3D3 UdPaO4TH0v1qUGUGY3y5rG7x6VHJUkKs4ZVQ6ZtZxEF6k/wrut8yO2kJF w4IGmaIPuy3DbA4WwuHn96N9hjZvNlv56etoONhvsO/a5zuJgUFP7eRzT Q==; X-CSE-ConnectionGUID: QD795TRJSyKK1HnfaPTC5A== X-CSE-MsgGUID: RRlF7KKESbWlnegvRGF63g== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289237" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289237" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:17 -0700 X-CSE-ConnectionGUID: 5w4SC7NkR5Ce+dphUy2UhA== X-CSE-MsgGUID: cvqXuc38SgCC+rgQygH31Q== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749286" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:14 -0700 From: Anatoly Burakov To: dev@dpdk.org, Wei Dai Cc: Piotr Skajewski , vladimir.medvedkin@intel.com, bruce.richardson@intel.com, stable@dpdk.org, Tyl@dpdk.org, RadoslawX , Mrozowicz@dpdk.org, SlawomirX , Michael@dpdk.org, Alice Subject: [PATCH v1 02/22] net/ixgbe/base: fix wrong 5G link speed reported on VF Date: Wed, 24 Apr 2024 14:21:36 +0100 Message-ID: <0331a520ca7dc7f2a384e56d7ae66f27ba9ba731.1713964708.git.anatoly.burakov@intel.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Piotr Skajewski When 5000 Base-T was set on PF the VF reported 100 Base-T. This patch change ixgbe_check_mac_link_vf function where was incorrect conditional which points to PF mac types, now it is pointing correctly to VF mac types. Fixes: 12e20906905c ("net/ixgbe/base: include new speeds in VFLINK interpretation") Cc: stable@dpdk.org Signed-off-by: Piotr Skajewski Reviewed-by: Tyl, RadoslawX Reviewed-by: Mrozowicz, SlawomirX Reviewed-by: Michael, Alice --- drivers/net/ixgbe/base/ixgbe_vf.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/ixgbe/base/ixgbe_vf.c b/drivers/net/ixgbe/base/ixgbe_vf.c index 5e3ae1b519..11dbbe2a86 100644 --- a/drivers/net/ixgbe/base/ixgbe_vf.c +++ b/drivers/net/ixgbe/base/ixgbe_vf.c @@ -585,7 +585,7 @@ s32 ixgbe_check_mac_link_vf(struct ixgbe_hw *hw, ixgbe_link_speed *speed, switch (links_reg & IXGBE_LINKS_SPEED_82599) { case IXGBE_LINKS_SPEED_10G_82599: *speed = IXGBE_LINK_SPEED_10GB_FULL; - if (hw->mac.type >= ixgbe_mac_X550) { + if (hw->mac.type >= ixgbe_mac_X550_vf) { if (links_reg & IXGBE_LINKS_SPEED_NON_STD) *speed = IXGBE_LINK_SPEED_2_5GB_FULL; } @@ -595,7 +595,7 @@ s32 ixgbe_check_mac_link_vf(struct ixgbe_hw *hw, ixgbe_link_speed *speed, break; case IXGBE_LINKS_SPEED_100_82599: *speed = IXGBE_LINK_SPEED_100_FULL; - if (hw->mac.type == ixgbe_mac_X550) { + if (hw->mac.type == ixgbe_mac_X550_vf) { if (links_reg & IXGBE_LINKS_SPEED_NON_STD) *speed = IXGBE_LINK_SPEED_5GB_FULL; } @@ -603,7 +603,7 @@ s32 ixgbe_check_mac_link_vf(struct ixgbe_hw *hw, ixgbe_link_speed *speed, case IXGBE_LINKS_SPEED_10_X550EM_A: *speed = IXGBE_LINK_SPEED_UNKNOWN; /* Since Reserved in older MAC's */ - if (hw->mac.type >= ixgbe_mac_X550) + if (hw->mac.type >= ixgbe_mac_X550_vf) *speed = IXGBE_LINK_SPEED_10_FULL; break; default: From patchwork Wed Apr 24 13:21:37 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139648 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 62DA943EAD; Wed, 24 Apr 2024 15:22:36 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 51B5043483; Wed, 24 Apr 2024 15:22:21 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id 87EA64347F; Wed, 24 Apr 2024 15:22:19 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964940; x=1745500940; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=MiRQVDfu+ENHKZICn9OlVLShw/50Smpfl3hrP7t+rEU=; b=D3tTK3SHPeIpw65S6m+6uBtmwmXAroXYfGivbnqdMgWhQ+EmhXr2jBPK dk1kZjPJ8iEj8WTn9rs51N8yBGSiwcSaklHXzl6EtGQXGwH60LLocPRXi ifmIJLVZajtCCp8onwqJSHs//n59arxEWDUKBYiqIIImo9kLK9gtndgd+ c4j1RxiwmsOirLSWvcCFZzZCWgStt3JW/y9OwIKCQDiBM7sCOVxk0jzmk D04TaIMLAhiZzDj2vayUFq0gHR9vq2yaVo+7h6IBMLak0uh+BaYlrJx5X bz974xk/j338vriU3P2GVDg3RDPKwaPnRx9/uRIl5Aa/xsmE6gB84mAhb Q==; X-CSE-ConnectionGUID: D5ycO6RnQbu2DtcF2n6wxg== X-CSE-MsgGUID: WwM+9BBSSma3Ufqih/fxCQ== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289241" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289241" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:19 -0700 X-CSE-ConnectionGUID: qI1hm4CCSQiOAxWRygMJ4g== X-CSE-MsgGUID: 6J2G9e3XSWKN0KZdbag3zA== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749296" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:17 -0700 From: Anatoly Burakov To: dev@dpdk.org, Wenzhuo Lu , Helin Zhang Cc: Radoslaw Tyl , vladimir.medvedkin@intel.com, bruce.richardson@intel.com, stable@dpdk.org, Skajewski@dpdk.org, PiotrX , Michael@dpdk.org, Alice Subject: [PATCH v1 03/22] net/ixgbe/base: fix PHY ID for X550 Date: Wed, 24 Apr 2024 14:21:37 +0100 Message-ID: X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Radoslaw Tyl Function ixgbe_get_phy_type_from_id() for X550_PHY_ID2 and X550_PHY_ID3 always return ixgbe_phy_unknown instead of ixgbe_phy_aq because phy ID's last 4 bits are always masked, and should not be taken into account when selecting phy type. This patch adds default PHY ID for X550 devices with mask on last 4 bits (0xFFFFFFF0), and fixes the switch statement to use it. Fixes: 58ddc803e412 ("ixgbe/base: add new X550 PHY ids") Cc: stable@dpdk.org Signed-off-by: Radoslaw Tyl Reviewed-by: Skajewski, PiotrX Reviewed-by: Michael, Alice --- drivers/net/ixgbe/base/ixgbe_phy.c | 3 +-- drivers/net/ixgbe/base/ixgbe_type.h | 1 + 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ixgbe/base/ixgbe_phy.c b/drivers/net/ixgbe/base/ixgbe_phy.c index 3a8e603472..56267bb00d 100644 --- a/drivers/net/ixgbe/base/ixgbe_phy.c +++ b/drivers/net/ixgbe/base/ixgbe_phy.c @@ -432,8 +432,7 @@ enum ixgbe_phy_type ixgbe_get_phy_type_from_id(u32 phy_id) case TN1010_PHY_ID: phy_type = ixgbe_phy_tn; break; - case X550_PHY_ID2: - case X550_PHY_ID3: + case X550_PHY_ID: case X540_PHY_ID: phy_type = ixgbe_phy_aq; break; diff --git a/drivers/net/ixgbe/base/ixgbe_type.h b/drivers/net/ixgbe/base/ixgbe_type.h index 35212a561b..f709681df2 100644 --- a/drivers/net/ixgbe/base/ixgbe_type.h +++ b/drivers/net/ixgbe/base/ixgbe_type.h @@ -1664,6 +1664,7 @@ struct ixgbe_dmac_config { #define TN1010_PHY_ID 0x00A19410 #define TNX_FW_REV 0xB #define X540_PHY_ID 0x01540200 +#define X550_PHY_ID 0x01540220 #define X550_PHY_ID2 0x01540223 #define X550_PHY_ID3 0x01540221 #define X557_PHY_ID 0x01540240 From patchwork Wed Apr 24 13:21:38 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139649 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id E9DA643EAD; Wed, 24 Apr 2024 15:22:45 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id EF5F94349E; Wed, 24 Apr 2024 15:22:23 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id 802B443494 for ; Wed, 24 Apr 2024 15:22:22 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964943; x=1745500943; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=QOviK+YSqxxD3YaR9GNVJ8DXPrU9J7AOsoUHakLmOmQ=; b=MsxJEtHXWvUI7EtjD0e041Mmu7SJwlUiVo1wHyIqcN21xkkeAKIH6O66 HZJCTZpuw4dUrR/2Nja4bsr33clJS9bwsXnti299v5a++Cly1twiB5Ylw IapminmdPGBUkAcxSI6nzBIP52UavExuoorpDNuPeXFfw995Lw+y8wG7R WJUOu37JUl0m9c6XlTvjHGWfj8jGc0zpLN7N5HozA3sake23ltUstW9ed AHxGwcbwSwX6D9MmGm56W6QTpzqvd0ZkV/6tmASUdJDRrz7Xpee7rFRBD 5aIw8LCxbg6Kx54rXENbNcO7AQAOSAnR5+kG5P8k1xnk08Wgc4YI110Hj w==; X-CSE-ConnectionGUID: O9JAUE39R3Gna5qsMmn+OA== X-CSE-MsgGUID: 4S5aQWWDT3OH8DASXdLwDg== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289244" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289244" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:22 -0700 X-CSE-ConnectionGUID: QTmJzoNXTheoyrott7mpYA== X-CSE-MsgGUID: OO9fe5nNS+a20Yx7T8JT1Q== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749312" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:19 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Jakub Chylkowski , vladimir.medvedkin@intel.com, bruce.richardson@intel.com, Zalfresso-jundzillo@dpdk.org, MarekX , Michael@dpdk.org, Alice , Skajewski@dpdk.org, PiotrX , Mrozowicz@dpdk.org, SlawomirX Subject: [PATCH v1 04/22] net/ixgbe/base: rename message type macros Date: Wed, 24 Apr 2024 14:21:38 +0100 Message-ID: X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Jakub Chylkowski There is name similarity within IXGBE_VT_MSGTYPE_ACK and PFMAILBOX.ACK / VFMAILBOX.ACK which may cause confusion. Rename MSGTYPE macros to SUCCESS and FAILURE as they are not specified in datasheet and now will be easily distinguishable. Signed-off-by: Jakub Chylkowski Reviewed-by: Zalfresso-jundzillo, MarekX Reviewed-by: Michael, Alice Reviewed-by: Skajewski, PiotrX Tested-by: Skajewski, PiotrX Reviewed-by: Mrozowicz, SlawomirX Tested-by: Michael, Alice --- drivers/net/ixgbe/base/ixgbe_mbx.h | 17 +++++++++------- drivers/net/ixgbe/base/ixgbe_vf.c | 32 ++++++++++++++++-------------- drivers/net/ixgbe/ixgbe_ethdev.c | 4 ++-- drivers/net/ixgbe/ixgbe_pf.c | 12 +++++------ 4 files changed, 35 insertions(+), 30 deletions(-) diff --git a/drivers/net/ixgbe/base/ixgbe_mbx.h b/drivers/net/ixgbe/base/ixgbe_mbx.h index 1a45e49c2f..28a2d94d02 100644 --- a/drivers/net/ixgbe/base/ixgbe_mbx.h +++ b/drivers/net/ixgbe/base/ixgbe_mbx.h @@ -38,14 +38,17 @@ /* If it's a IXGBE_VF_* msg then it originates in the VF and is sent to the * PF. The reverse is true if it is IXGBE_PF_*. - * Message ACK's are the value or'd with 0xF0000000 + * Message results are the value or'd with 0xF0000000 */ -#define IXGBE_VT_MSGTYPE_ACK 0x80000000 /* Messages below or'd with - * this are the ACK */ -#define IXGBE_VT_MSGTYPE_NACK 0x40000000 /* Messages below or'd with - * this are the NACK */ -#define IXGBE_VT_MSGTYPE_CTS 0x20000000 /* Indicates that VF is still - * clear to send requests */ +#define IXGBE_VT_MSGTYPE_SUCCESS 0x80000000 /* Messages or'd with this + * have succeeded + */ +#define IXGBE_VT_MSGTYPE_FAILURE 0x40000000 /* Messages or'd with this + * have failed + */ +#define IXGBE_VT_MSGTYPE_CTS 0x20000000 /* Indicates that VF is still + * clear to send requests + */ #define IXGBE_VT_MSGINFO_SHIFT 16 /* bits 23:16 are used for extra info for certain messages */ #define IXGBE_VT_MSGINFO_MASK (0xFF << IXGBE_VT_MSGINFO_SHIFT) diff --git a/drivers/net/ixgbe/base/ixgbe_vf.c b/drivers/net/ixgbe/base/ixgbe_vf.c index 11dbbe2a86..d390b0d5e0 100644 --- a/drivers/net/ixgbe/base/ixgbe_vf.c +++ b/drivers/net/ixgbe/base/ixgbe_vf.c @@ -194,11 +194,11 @@ s32 ixgbe_reset_hw_vf(struct ixgbe_hw *hw) if (ret_val) return ret_val; - if (msgbuf[0] != (IXGBE_VF_RESET | IXGBE_VT_MSGTYPE_ACK) && - msgbuf[0] != (IXGBE_VF_RESET | IXGBE_VT_MSGTYPE_NACK)) + if (msgbuf[0] != (IXGBE_VF_RESET | IXGBE_VT_MSGTYPE_SUCCESS) && + msgbuf[0] != (IXGBE_VF_RESET | IXGBE_VT_MSGTYPE_FAILURE)) return IXGBE_ERR_INVALID_MAC_ADDR; - if (msgbuf[0] == (IXGBE_VF_RESET | IXGBE_VT_MSGTYPE_ACK)) + if (msgbuf[0] == (IXGBE_VF_RESET | IXGBE_VT_MSGTYPE_SUCCESS)) memcpy(hw->mac.perm_addr, addr, IXGBE_ETH_LENGTH_OF_ADDRESS); hw->mac.mc_filter_type = msgbuf[IXGBE_VF_MC_TYPE_WORD]; @@ -327,9 +327,9 @@ s32 ixgbe_set_rar_vf(struct ixgbe_hw *hw, u32 index, u8 *addr, u32 vmdq, msgbuf[0] &= ~IXGBE_VT_MSGTYPE_CTS; - /* if nacked the address was rejected, use "perm_addr" */ + /* if we had failure, the address was rejected, use "perm_addr" */ if (!ret_val && - (msgbuf[0] == (IXGBE_VF_SET_MAC_ADDR | IXGBE_VT_MSGTYPE_NACK))) { + (msgbuf[0] == (IXGBE_VF_SET_MAC_ADDR | IXGBE_VT_MSGTYPE_FAILURE))) { ixgbe_get_mac_addr_vf(hw, hw->mac.addr); return IXGBE_ERR_MBX; } @@ -418,7 +418,7 @@ s32 ixgbevf_update_xcast_mode(struct ixgbe_hw *hw, int xcast_mode) return err; msgbuf[0] &= ~IXGBE_VT_MSGTYPE_CTS; - if (msgbuf[0] == (IXGBE_VF_UPDATE_XCAST_MODE | IXGBE_VT_MSGTYPE_NACK)) + if (msgbuf[0] == (IXGBE_VF_UPDATE_XCAST_MODE | IXGBE_VT_MSGTYPE_FAILURE)) return IXGBE_ERR_FEATURE_NOT_SUPPORTED; return IXGBE_SUCCESS; } @@ -446,10 +446,10 @@ s32 ixgbe_set_vfta_vf(struct ixgbe_hw *hw, u32 vlan, u32 vind, msgbuf[0] |= vlan_on << IXGBE_VT_MSGINFO_SHIFT; ret_val = ixgbevf_write_msg_read_ack(hw, msgbuf, msgbuf, 2); - if (!ret_val && (msgbuf[0] & IXGBE_VT_MSGTYPE_ACK)) + if (!ret_val && (msgbuf[0] & IXGBE_VT_MSGTYPE_SUCCESS)) return IXGBE_SUCCESS; - return ret_val | (msgbuf[0] & IXGBE_VT_MSGTYPE_NACK); + return ret_val | (msgbuf[0] & IXGBE_VT_MSGTYPE_FAILURE); } /** @@ -514,7 +514,7 @@ s32 ixgbevf_set_uc_addr_vf(struct ixgbe_hw *hw, u32 index, u8 *addr) if (!ret_val) { msgbuf[0] &= ~IXGBE_VT_MSGTYPE_CTS; - if (msgbuf[0] == (msgbuf_chk | IXGBE_VT_MSGTYPE_NACK)) + if (msgbuf[0] == (msgbuf_chk | IXGBE_VT_MSGTYPE_FAILURE)) return IXGBE_ERR_OUT_OF_MEM; } @@ -617,8 +617,10 @@ s32 ixgbe_check_mac_link_vf(struct ixgbe_hw *hw, ixgbe_link_speed *speed, goto out; if (!(in_msg & IXGBE_VT_MSGTYPE_CTS)) { - /* msg is not CTS and is NACK we must have lost CTS status */ - if (in_msg & IXGBE_VT_MSGTYPE_NACK) + /* msg is not CTS and is FAILURE, + * we must have lost CTS status. + */ + if (in_msg & IXGBE_VT_MSGTYPE_FAILURE) ret_val = -1; goto out; } @@ -656,7 +658,7 @@ s32 ixgbevf_rlpml_set_vf(struct ixgbe_hw *hw, u16 max_size) if (retval) return retval; if ((msgbuf[0] & IXGBE_VF_SET_LPE) && - (msgbuf[0] & IXGBE_VT_MSGTYPE_NACK)) + (msgbuf[0] & IXGBE_VT_MSGTYPE_FAILURE)) return IXGBE_ERR_MBX; return 0; @@ -682,7 +684,7 @@ int ixgbevf_negotiate_api_version(struct ixgbe_hw *hw, int api) msg[0] &= ~IXGBE_VT_MSGTYPE_CTS; /* Store value and return 0 on success */ - if (msg[0] == (IXGBE_VF_API_NEGOTIATE | IXGBE_VT_MSGTYPE_ACK)) { + if (msg[0] == (IXGBE_VF_API_NEGOTIATE | IXGBE_VT_MSGTYPE_SUCCESS)) { hw->api_version = api; return 0; } @@ -718,11 +720,11 @@ int ixgbevf_get_queues(struct ixgbe_hw *hw, unsigned int *num_tcs, msg[0] &= ~IXGBE_VT_MSGTYPE_CTS; /* - * if we we didn't get an ACK there must have been + * if we didn't get a SUCCESS there must have been * some sort of mailbox error so we should treat it * as such */ - if (msg[0] != (IXGBE_VF_GET_QUEUES | IXGBE_VT_MSGTYPE_ACK)) + if (msg[0] != (IXGBE_VF_GET_QUEUES | IXGBE_VT_MSGTYPE_SUCCESS)) return IXGBE_ERR_MBX; /* record and validate values from message */ diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c index c61c52b296..8c60351636 100644 --- a/drivers/net/ixgbe/ixgbe_ethdev.c +++ b/drivers/net/ixgbe/ixgbe_ethdev.c @@ -4169,8 +4169,8 @@ ixgbevf_check_link(struct ixgbe_hw *hw, ixgbe_link_speed *speed, goto out; if (!(in_msg & IXGBE_VT_MSGTYPE_CTS)) { - /* msg is not CTS and is NACK we must have lost CTS status */ - if (in_msg & IXGBE_VT_MSGTYPE_NACK) + /* msg is not CTS and is FAILURE we must have lost CTS status */ + if (in_msg & IXGBE_VT_MSGTYPE_FAILURE) mac->get_link_status = false; goto out; } diff --git a/drivers/net/ixgbe/ixgbe_pf.c b/drivers/net/ixgbe/ixgbe_pf.c index 0a0f639e39..71104247da 100644 --- a/drivers/net/ixgbe/ixgbe_pf.c +++ b/drivers/net/ixgbe/ixgbe_pf.c @@ -448,8 +448,8 @@ ixgbe_vf_reset(struct rte_eth_dev *dev, uint16_t vf, uint32_t *msgbuf) /* Disable multicast promiscuous at reset */ ixgbe_disable_vf_mc_promisc(dev, vf); - /* reply to reset with ack and vf mac address */ - msgbuf[0] = IXGBE_VF_RESET | IXGBE_VT_MSGTYPE_ACK; + /* reply to reset with success and vf mac address */ + msgbuf[0] = IXGBE_VF_RESET | IXGBE_VT_MSGTYPE_SUCCESS; rte_memcpy(new_mac, vf_mac, RTE_ETHER_ADDR_LEN); /* * Piggyback the multicast filter type so VF can compute the @@ -840,7 +840,7 @@ ixgbe_rcv_msg_from_vf(struct rte_eth_dev *dev, uint16_t vf) } /* do nothing with the message already been processed */ - if (msgbuf[0] & (IXGBE_VT_MSGTYPE_ACK | IXGBE_VT_MSGTYPE_NACK)) + if (msgbuf[0] & (IXGBE_VT_MSGTYPE_SUCCESS | IXGBE_VT_MSGTYPE_FAILURE)) return retval; /* flush the ack before we write any messages back */ @@ -919,9 +919,9 @@ ixgbe_rcv_msg_from_vf(struct rte_eth_dev *dev, uint16_t vf) /* response the VF according to the message process result */ if (retval) - msgbuf[0] |= IXGBE_VT_MSGTYPE_NACK; + msgbuf[0] |= IXGBE_VT_MSGTYPE_FAILURE; else - msgbuf[0] |= IXGBE_VT_MSGTYPE_ACK; + msgbuf[0] |= IXGBE_VT_MSGTYPE_SUCCESS; msgbuf[0] |= IXGBE_VT_MSGTYPE_CTS; @@ -933,7 +933,7 @@ ixgbe_rcv_msg_from_vf(struct rte_eth_dev *dev, uint16_t vf) static inline void ixgbe_rcv_ack_from_vf(struct rte_eth_dev *dev, uint16_t vf) { - uint32_t msg = IXGBE_VT_MSGTYPE_NACK; + uint32_t msg = IXGBE_VT_MSGTYPE_FAILURE; struct ixgbe_hw *hw = IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private); struct ixgbe_vf_info *vfinfo = From patchwork Wed Apr 24 13:21:39 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139650 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 1B31143EAD; Wed, 24 Apr 2024 15:22:53 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 1D48A43491; Wed, 24 Apr 2024 15:22:27 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id 615FA4348A for ; Wed, 24 Apr 2024 15:22:25 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964946; x=1745500946; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=BaBny1Z8ptqio0BlU3BiQ92uVV0ZoBM8AzX759r7Y3g=; b=OCc3ufJ6PdJ49WTeB8PQxM8mkjfAde/If79fjuHtK6MOl2HjbQMlW+B/ MxLEvxU4kI4uAGZhxVIvBTRFAHCxU0kzCPwX4/WrK8YZjaoI6ihK+XntL ahLdVuWltWt/OFtnSlC/KKFRlYzsb59G0CvO4q1hVNrOKiCi2N9N28YOP 5QeQ6076u3PpJtC5J5Ffri8iVAfrjiGEuR0jx5bh6diN24+JB9CvB2/ll jR7M4YpABzJWvdO5Jyjjck5dZckl0CighMP+x1UKQ/X1y0sfzPg+tawLJ 75EeQJ1iOpIGvguuTBp2dOfRhXVohv6U/DHa+9OQoMxPZc6tDSDx1QQ00 w==; X-CSE-ConnectionGUID: C4+Tkx9IQIuFujuGYHulxw== X-CSE-MsgGUID: HVa3UXgXSAujzPON2VPHKg== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289252" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289252" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:25 -0700 X-CSE-ConnectionGUID: /qHScdwcSNKuI+YTHeqsig== X-CSE-MsgGUID: OOUGbir6SjueG7z+5FQ1yg== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749328" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:22 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Jakub Chylkowski , vladimir.medvedkin@intel.com, bruce.richardson@intel.com, Zalfresso-jundzillo@dpdk.org, MarekX , Michael@dpdk.org, Alice , Skajewski@dpdk.org, PiotrX , Mrozowicz@dpdk.org, SlawomirX Subject: [PATCH v1 05/22] net/ixgbe/base: correct registers names to match datasheet Date: Wed, 24 Apr 2024 14:21:39 +0100 Message-ID: <27bc50509a27b3a9e9a31aa5d25c388c8b14bbd5.1713964708.git.anatoly.burakov@intel.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Jakub Chylkowski Some of mailbox-related registers have different names than it is specified in datasheet. Correct these names to correspond to their datasheet counterparts. Additionally, several calculations are changed to no longer use magic numbers but dedicated macros instead. Signed-off-by: Jakub Chylkowski Reviewed-by: Zalfresso-jundzillo, MarekX Reviewed-by: Michael, Alice Reviewed-by: Skajewski, PiotrX Tested-by: Skajewski, PiotrX Reviewed-by: Mrozowicz, SlawomirX Tested-by: Michael, Alice --- drivers/net/ixgbe/base/ixgbe_mbx.c | 30 ++++++++++++++--------------- drivers/net/ixgbe/base/ixgbe_mbx.h | 8 ++++---- drivers/net/ixgbe/base/ixgbe_type.h | 12 ++++++------ 3 files changed, 25 insertions(+), 25 deletions(-) diff --git a/drivers/net/ixgbe/base/ixgbe_mbx.c b/drivers/net/ixgbe/base/ixgbe_mbx.c index 4dddff2c58..d645dcf827 100644 --- a/drivers/net/ixgbe/base/ixgbe_mbx.c +++ b/drivers/net/ixgbe/base/ixgbe_mbx.c @@ -496,12 +496,12 @@ void ixgbe_init_mbx_params_vf(struct ixgbe_hw *hw) STATIC s32 ixgbe_check_for_bit_pf(struct ixgbe_hw *hw, u32 mask, s32 index) { - u32 mbvficr = IXGBE_READ_REG(hw, IXGBE_MBVFICR(index)); + u32 pfmbicr = IXGBE_READ_REG(hw, IXGBE_PFMBICR(index)); s32 ret_val = IXGBE_ERR_MBX; - if (mbvficr & mask) { + if (pfmbicr & mask) { ret_val = IXGBE_SUCCESS; - IXGBE_WRITE_REG(hw, IXGBE_MBVFICR(index), mask); + IXGBE_WRITE_REG(hw, IXGBE_PFMBICR(index), mask); } return ret_val; @@ -516,13 +516,13 @@ STATIC s32 ixgbe_check_for_bit_pf(struct ixgbe_hw *hw, u32 mask, s32 index) **/ STATIC s32 ixgbe_check_for_msg_pf(struct ixgbe_hw *hw, u16 vf_number) { + u32 vf_shift = IXGBE_PFMBICR_SHIFT(vf_number); + s32 index = IXGBE_PFMBICR_INDEX(vf_number); s32 ret_val = IXGBE_ERR_MBX; - s32 index = IXGBE_MBVFICR_INDEX(vf_number); - u32 vf_bit = vf_number % 16; DEBUGFUNC("ixgbe_check_for_msg_pf"); - if (!ixgbe_check_for_bit_pf(hw, IXGBE_MBVFICR_VFREQ_VF1 << vf_bit, + if (!ixgbe_check_for_bit_pf(hw, IXGBE_PFMBICR_VFREQ_VF1 << vf_shift, index)) { ret_val = IXGBE_SUCCESS; hw->mbx.stats.reqs++; @@ -540,13 +540,13 @@ STATIC s32 ixgbe_check_for_msg_pf(struct ixgbe_hw *hw, u16 vf_number) **/ STATIC s32 ixgbe_check_for_ack_pf(struct ixgbe_hw *hw, u16 vf_number) { + u32 vf_shift = IXGBE_PFMBICR_SHIFT(vf_number); + s32 index = IXGBE_PFMBICR_INDEX(vf_number); s32 ret_val = IXGBE_ERR_MBX; - s32 index = IXGBE_MBVFICR_INDEX(vf_number); - u32 vf_bit = vf_number % 16; DEBUGFUNC("ixgbe_check_for_ack_pf"); - if (!ixgbe_check_for_bit_pf(hw, IXGBE_MBVFICR_VFACK_VF1 << vf_bit, + if (!ixgbe_check_for_bit_pf(hw, IXGBE_PFMBICR_VFACK_VF1 << vf_shift, index)) { ret_val = IXGBE_SUCCESS; hw->mbx.stats.acks++; @@ -564,22 +564,22 @@ STATIC s32 ixgbe_check_for_ack_pf(struct ixgbe_hw *hw, u16 vf_number) **/ STATIC s32 ixgbe_check_for_rst_pf(struct ixgbe_hw *hw, u16 vf_number) { - u32 reg_offset = (vf_number < 32) ? 0 : 1; - u32 vf_shift = vf_number % 32; + u32 vf_shift = IXGBE_PFVFLRE_SHIFT(vf_number); + u32 index = IXGBE_PFVFLRE_INDEX(vf_number); + s32 ret_val = IXGBE_ERR_MBX; u32 vflre = 0; - s32 ret_val = IXGBE_ERR_MBX; DEBUGFUNC("ixgbe_check_for_rst_pf"); switch (hw->mac.type) { case ixgbe_mac_82599EB: - vflre = IXGBE_READ_REG(hw, IXGBE_VFLRE(reg_offset)); + vflre = IXGBE_READ_REG(hw, IXGBE_PFVFLRE(index)); break; case ixgbe_mac_X550: case ixgbe_mac_X550EM_x: case ixgbe_mac_X550EM_a: case ixgbe_mac_X540: - vflre = IXGBE_READ_REG(hw, IXGBE_VFLREC(reg_offset)); + vflre = IXGBE_READ_REG(hw, IXGBE_PFVFLREC(index)); break; default: break; @@ -587,7 +587,7 @@ STATIC s32 ixgbe_check_for_rst_pf(struct ixgbe_hw *hw, u16 vf_number) if (vflre & (1 << vf_shift)) { ret_val = IXGBE_SUCCESS; - IXGBE_WRITE_REG(hw, IXGBE_VFLREC(reg_offset), (1 << vf_shift)); + IXGBE_WRITE_REG(hw, IXGBE_PFVFLREC(index), (1 << vf_shift)); hw->mbx.stats.rsts++; } diff --git a/drivers/net/ixgbe/base/ixgbe_mbx.h b/drivers/net/ixgbe/base/ixgbe_mbx.h index 28a2d94d02..f7861e6bde 100644 --- a/drivers/net/ixgbe/base/ixgbe_mbx.h +++ b/drivers/net/ixgbe/base/ixgbe_mbx.h @@ -30,10 +30,10 @@ #define IXGBE_PFMAILBOX_PFU 0x00000008 /* PF owns the mailbox buffer */ #define IXGBE_PFMAILBOX_RVFU 0x00000010 /* Reset VFU - used when VF stuck */ -#define IXGBE_MBVFICR_VFREQ_MASK 0x0000FFFF /* bits for VF messages */ -#define IXGBE_MBVFICR_VFREQ_VF1 0x00000001 /* bit for VF 1 message */ -#define IXGBE_MBVFICR_VFACK_MASK 0xFFFF0000 /* bits for VF acks */ -#define IXGBE_MBVFICR_VFACK_VF1 0x00010000 /* bit for VF 1 ack */ +#define IXGBE_PFMBICR_VFREQ_MASK 0x0000FFFF /* bits for VF messages */ +#define IXGBE_PFMBICR_VFREQ_VF1 0x00000001 /* bit for VF 1 message */ +#define IXGBE_PFMBICR_VFACK_MASK 0xFFFF0000 /* bits for VF acks */ +#define IXGBE_PFMBICR_VFACK_VF1 0x00010000 /* bit for VF 1 ack */ /* If it's a IXGBE_VF_* msg then it originates in the VF and is sent to the diff --git a/drivers/net/ixgbe/base/ixgbe_type.h b/drivers/net/ixgbe/base/ixgbe_type.h index f709681df2..5036b2a907 100644 --- a/drivers/net/ixgbe/base/ixgbe_type.h +++ b/drivers/net/ixgbe/base/ixgbe_type.h @@ -451,8 +451,14 @@ struct ixgbe_nvm_version { #define IXGBE_PFMAILBOX(_i) (0x04B00 + (4 * (_i))) /* 64 total */ /* 64 Mailboxes, 16 DW each */ #define IXGBE_PFMBMEM(_i) (0x13000 + (64 * (_i))) +#define IXGBE_PFMBICR_INDEX(_i) ((_i) >> 4) +#define IXGBE_PFMBICR_SHIFT(_i) ((_i) % 16) #define IXGBE_PFMBICR(_i) (0x00710 + (4 * (_i))) /* 4 total */ #define IXGBE_PFMBIMR(_i) (0x00720 + (4 * (_i))) /* 4 total */ +#define IXGBE_PFVFLRE(_i) ((((_i) & 1) ? 0x001C0 : 0x00600)) +#define IXGBE_PFVFLREC(_i) (0x00700 + ((_i) * 4)) +#define IXGBE_PFVFLRE_INDEX(_i) ((_i) >> 5) +#define IXGBE_PFVFLRE_SHIFT(_i) ((_i) % 32) #define IXGBE_VFRE(_i) (0x051E0 + ((_i) * 4)) #define IXGBE_VFTE(_i) (0x08110 + ((_i) * 4)) #define IXGBE_VMECM(_i) (0x08790 + ((_i) * 4)) @@ -2866,12 +2872,6 @@ enum { #define IXGBE_RX_DESC_SPECIAL_PRI_MASK 0xE000 /* Priority in upper 3 bits */ #define IXGBE_RX_DESC_SPECIAL_PRI_SHIFT 0x000D /* Priority in upper 3 of 16 */ #define IXGBE_TX_DESC_SPECIAL_PRI_SHIFT IXGBE_RX_DESC_SPECIAL_PRI_SHIFT - -/* SR-IOV specific macros */ -#define IXGBE_MBVFICR_INDEX(vf_number) (vf_number >> 4) -#define IXGBE_MBVFICR(_i) (0x00710 + ((_i) * 4)) -#define IXGBE_VFLRE(_i) (((_i & 1) ? 0x001C0 : 0x00600)) -#define IXGBE_VFLREC(_i) (0x00700 + ((_i) * 4)) /* Translated register #defines */ #define IXGBE_PVFCTRL(P) (0x00300 + (4 * (P))) #define IXGBE_PVFSTATUS(P) (0x00008 + (0 * (P))) From patchwork Wed Apr 24 13:21:40 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139651 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 4E45443EAD; Wed, 24 Apr 2024 15:23:00 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 577C0434A1; Wed, 24 Apr 2024 15:22:30 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id 8555E43498 for ; Wed, 24 Apr 2024 15:22:28 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964949; x=1745500949; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=bRJeAngSfQ7RpA3Zdte91PMUb79otTy2iLecU4nwdYI=; b=W8twoE5ZsyCS4YOSIaEegsh2r35UGQnwmNaxWDgHxNiQd0/6Pzg8T0JM 58+lpbc//2LyTU6B4Brtfzr919xO6zUy6LAO82LqEA5Y/ccN/nW2QVquC 9qluxRVPo0/mjlK9cgE7m4hzD/5XEe/wHVLCnrej04hLp53VM77ngpA1o TmQCt64pBoK2vPsyL52pM6WZx9lbMjBdJm568w1M/7KsUzZG4+JRGsweL 25rp0G536Kj5Twfo7kLHTE0CNbb4XoKk/GcRVbUp3JMWxWvuZYfVuzeHT saQDSGuILkLwNEfTe2UB4xnre0Rm1tNvAdbrtnJ8P4unr19EoR44tkVLJ g==; X-CSE-ConnectionGUID: RCHDl3NVQI+TUkWypjZGJg== X-CSE-MsgGUID: aRSjLPCVQSSQ3HrL9gjbkA== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289259" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289259" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:28 -0700 X-CSE-ConnectionGUID: tY5txZ3MTyuXdr0xKEPOig== X-CSE-MsgGUID: vryan7+gQ+aqp/L4GTjB2g== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749350" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:25 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Jakub Chylkowski , vladimir.medvedkin@intel.com, bruce.richardson@intel.com, Michael@dpdk.org, Alice , Pietruszewski@dpdk.org, Piotr , Skajewski@dpdk.org, PiotrX Subject: [PATCH v1 06/22] net/ixgbe/base: introduce new mailbox API Date: Wed, 24 Apr 2024 14:21:40 +0100 Message-ID: X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Jakub Chylkowski Current mailbox API does not work as described in documentation and is prone to errors (for example, it is doing locks on read). Introduce new mailbox API and provide compatibility functions with old API. New error codes have been introduced: - IXGBE_ERR_CONFIG - ixgbe_mbx_operations is not correctly set - IXGBE_ERR_TIMEOUT - mailbox operation, e.g. poll for message, timedout - IXGBE_ERR_MBX_NOMSG - no message available on read In addition, some refactoring has been done: mailbox structures were defined twice: in ixgbe_type.h and ixgbe_vf.h. Move them into ixgbe_mbx.h as this header is dedicated for mailbox. Signed-off-by: Jakub Chylkowski Reviewed-by: Michael, Alice Reviewed-by: Pietruszewski, Piotr Tested-by: Michael, Alice Tested-by: Skajewski, PiotrX --- drivers/net/ixgbe/base/ixgbe_82599.c | 4 +- drivers/net/ixgbe/base/ixgbe_hv_vf.c | 2 +- drivers/net/ixgbe/base/ixgbe_mbx.c | 843 +++++++++++++++++++-------- drivers/net/ixgbe/base/ixgbe_mbx.h | 59 +- drivers/net/ixgbe/base/ixgbe_type.h | 32 +- drivers/net/ixgbe/base/ixgbe_vf.c | 42 +- drivers/net/ixgbe/base/ixgbe_x540.c | 4 +- drivers/net/ixgbe/ixgbe_ethdev.c | 6 +- drivers/net/ixgbe/ixgbe_pf.c | 4 +- 9 files changed, 702 insertions(+), 294 deletions(-) diff --git a/drivers/net/ixgbe/base/ixgbe_82599.c b/drivers/net/ixgbe/base/ixgbe_82599.c index c6e8b7e976..7d2938b253 100644 --- a/drivers/net/ixgbe/base/ixgbe_82599.c +++ b/drivers/net/ixgbe/base/ixgbe_82599.c @@ -291,6 +291,7 @@ s32 ixgbe_init_ops_82599(struct ixgbe_hw *hw) struct ixgbe_phy_info *phy = &hw->phy; struct ixgbe_eeprom_info *eeprom = &hw->eeprom; s32 ret_val; + u16 i; DEBUGFUNC("ixgbe_init_ops_82599"); @@ -352,7 +353,8 @@ s32 ixgbe_init_ops_82599(struct ixgbe_hw *hw) mac->arc_subsystem_valid = !!(IXGBE_READ_REG(hw, IXGBE_FWSM_BY_MAC(hw)) & IXGBE_FWSM_MODE_MASK); - hw->mbx.ops.init_params = ixgbe_init_mbx_params_pf; + for (i = 0; i < 64; i++) + hw->mbx.ops[i].init_params = ixgbe_init_mbx_params_pf; /* EEPROM */ eeprom->ops.read = ixgbe_read_eeprom_82599; diff --git a/drivers/net/ixgbe/base/ixgbe_hv_vf.c b/drivers/net/ixgbe/base/ixgbe_hv_vf.c index 4572411d39..34add03e8b 100644 --- a/drivers/net/ixgbe/base/ixgbe_hv_vf.c +++ b/drivers/net/ixgbe/base/ixgbe_hv_vf.c @@ -97,7 +97,7 @@ static s32 ixgbevf_hv_check_mac_link_vf(struct ixgbe_hw *hw, UNREFERENCED_1PARAMETER(autoneg_wait_to_complete); /* If we were hit with a reset drop the link */ - if (!mbx->ops.check_for_rst(hw, 0) || !mbx->timeout) + if (!mbx->ops[0].check_for_rst(hw, 0) || !mbx->timeout) mac->get_link_status = true; if (!mac->get_link_status) diff --git a/drivers/net/ixgbe/base/ixgbe_mbx.c b/drivers/net/ixgbe/base/ixgbe_mbx.c index d645dcf827..2dab347396 100644 --- a/drivers/net/ixgbe/base/ixgbe_mbx.c +++ b/drivers/net/ixgbe/base/ixgbe_mbx.c @@ -5,6 +5,9 @@ #include "ixgbe_type.h" #include "ixgbe_mbx.h" +STATIC s32 ixgbe_poll_for_msg(struct ixgbe_hw *hw, u16 mbx_id); +STATIC s32 ixgbe_poll_for_ack(struct ixgbe_hw *hw, u16 mbx_id); + /** * ixgbe_read_mbx - Reads a message from the mailbox * @hw: pointer to the HW structure @@ -17,42 +20,91 @@ s32 ixgbe_read_mbx(struct ixgbe_hw *hw, u32 *msg, u16 size, u16 mbx_id) { struct ixgbe_mbx_info *mbx = &hw->mbx; - s32 ret_val = IXGBE_ERR_MBX; DEBUGFUNC("ixgbe_read_mbx"); /* limit read to size of mailbox */ - if (size > mbx->size) + if (size > mbx->size) { + ERROR_REPORT3(IXGBE_ERROR_ARGUMENT, + "Invalid mailbox message size %u, changing to %u", + size, mbx->size); size = mbx->size; + } - if (mbx->ops.read) - ret_val = mbx->ops.read(hw, msg, size, mbx_id); + if (mbx->ops[mbx_id].read) + return mbx->ops[mbx_id].read(hw, msg, size, mbx_id); + + return IXGBE_ERR_CONFIG; +} + +/** + * ixgbe_poll_mbx - Wait for message and read it from the mailbox + * @hw: pointer to the HW structure + * @msg: The message buffer + * @size: Length of buffer + * @mbx_id: id of mailbox to read + * + * returns SUCCESS if it successfully read message from buffer + **/ +s32 ixgbe_poll_mbx(struct ixgbe_hw *hw, u32 *msg, u16 size, u16 mbx_id) +{ + struct ixgbe_mbx_info *mbx = &hw->mbx; + s32 ret_val; + + DEBUGFUNC("ixgbe_poll_mbx"); + + if (!mbx->ops[mbx_id].read || !mbx->ops[mbx_id].check_for_msg || + !mbx->timeout) + return IXGBE_ERR_CONFIG; + + /* limit read to size of mailbox */ + if (size > mbx->size) { + ERROR_REPORT3(IXGBE_ERROR_ARGUMENT, + "Invalid mailbox message size %u, changing to %u", + size, mbx->size); + size = mbx->size; + } + + ret_val = ixgbe_poll_for_msg(hw, mbx_id); + /* if ack received read message, otherwise we timed out */ + if (!ret_val) + return mbx->ops[mbx_id].read(hw, msg, size, mbx_id); return ret_val; } /** - * ixgbe_write_mbx - Write a message to the mailbox + * ixgbe_write_mbx - Write a message to the mailbox and wait for ACK * @hw: pointer to the HW structure * @msg: The message buffer * @size: Length of buffer * @mbx_id: id of mailbox to write * - * returns SUCCESS if it successfully copied message into the buffer + * returns SUCCESS if it successfully copied message into the buffer and + * received an ACK to that message within specified period **/ s32 ixgbe_write_mbx(struct ixgbe_hw *hw, u32 *msg, u16 size, u16 mbx_id) { struct ixgbe_mbx_info *mbx = &hw->mbx; - s32 ret_val = IXGBE_SUCCESS; + s32 ret_val = IXGBE_ERR_MBX; DEBUGFUNC("ixgbe_write_mbx"); + /* + * exit if either we can't write, release + * or there is no timeout defined + */ + if (!mbx->ops[mbx_id].write || !mbx->ops[mbx_id].check_for_ack || + !mbx->ops[mbx_id].release || !mbx->timeout) + return IXGBE_ERR_CONFIG; + if (size > mbx->size) { - ret_val = IXGBE_ERR_MBX; + ret_val = IXGBE_ERR_PARAM; ERROR_REPORT2(IXGBE_ERROR_ARGUMENT, - "Invalid mailbox message size %d", size); - } else if (mbx->ops.write) - ret_val = mbx->ops.write(hw, msg, size, mbx_id); + "Invalid mailbox message size %u", size); + } else { + ret_val = mbx->ops[mbx_id].write(hw, msg, size, mbx_id); + } return ret_val; } @@ -67,12 +119,12 @@ s32 ixgbe_write_mbx(struct ixgbe_hw *hw, u32 *msg, u16 size, u16 mbx_id) s32 ixgbe_check_for_msg(struct ixgbe_hw *hw, u16 mbx_id) { struct ixgbe_mbx_info *mbx = &hw->mbx; - s32 ret_val = IXGBE_ERR_MBX; + s32 ret_val = IXGBE_ERR_CONFIG; DEBUGFUNC("ixgbe_check_for_msg"); - if (mbx->ops.check_for_msg) - ret_val = mbx->ops.check_for_msg(hw, mbx_id); + if (mbx->ops[mbx_id].check_for_msg) + ret_val = mbx->ops[mbx_id].check_for_msg(hw, mbx_id); return ret_val; } @@ -87,12 +139,12 @@ s32 ixgbe_check_for_msg(struct ixgbe_hw *hw, u16 mbx_id) s32 ixgbe_check_for_ack(struct ixgbe_hw *hw, u16 mbx_id) { struct ixgbe_mbx_info *mbx = &hw->mbx; - s32 ret_val = IXGBE_ERR_MBX; + s32 ret_val = IXGBE_ERR_CONFIG; DEBUGFUNC("ixgbe_check_for_ack"); - if (mbx->ops.check_for_ack) - ret_val = mbx->ops.check_for_ack(hw, mbx_id); + if (mbx->ops[mbx_id].check_for_ack) + ret_val = mbx->ops[mbx_id].check_for_ack(hw, mbx_id); return ret_val; } @@ -107,12 +159,12 @@ s32 ixgbe_check_for_ack(struct ixgbe_hw *hw, u16 mbx_id) s32 ixgbe_check_for_rst(struct ixgbe_hw *hw, u16 mbx_id) { struct ixgbe_mbx_info *mbx = &hw->mbx; - s32 ret_val = IXGBE_ERR_MBX; + s32 ret_val = IXGBE_ERR_CONFIG; DEBUGFUNC("ixgbe_check_for_rst"); - if (mbx->ops.check_for_rst) - ret_val = mbx->ops.check_for_rst(hw, mbx_id); + if (mbx->ops[mbx_id].check_for_rst) + ret_val = mbx->ops[mbx_id].check_for_rst(hw, mbx_id); return ret_val; } @@ -131,22 +183,23 @@ STATIC s32 ixgbe_poll_for_msg(struct ixgbe_hw *hw, u16 mbx_id) DEBUGFUNC("ixgbe_poll_for_msg"); - if (!countdown || !mbx->ops.check_for_msg) - goto out; + if (!countdown || !mbx->ops[mbx_id].check_for_msg) + return IXGBE_ERR_CONFIG; - while (countdown && mbx->ops.check_for_msg(hw, mbx_id)) { + while (countdown && mbx->ops[mbx_id].check_for_msg(hw, mbx_id)) { countdown--; if (!countdown) break; usec_delay(mbx->usec_delay); } - if (countdown == 0) + if (countdown == 0) { ERROR_REPORT2(IXGBE_ERROR_POLLING, - "Polling for VF%d mailbox message timedout", mbx_id); + "Polling for VF%u mailbox message timedout", mbx_id); + return IXGBE_ERR_TIMEOUT; + } -out: - return countdown ? IXGBE_SUCCESS : IXGBE_ERR_MBX; + return IXGBE_SUCCESS; } /** @@ -163,114 +216,71 @@ STATIC s32 ixgbe_poll_for_ack(struct ixgbe_hw *hw, u16 mbx_id) DEBUGFUNC("ixgbe_poll_for_ack"); - if (!countdown || !mbx->ops.check_for_ack) - goto out; + if (!countdown || !mbx->ops[mbx_id].check_for_ack) + return IXGBE_ERR_CONFIG; - while (countdown && mbx->ops.check_for_ack(hw, mbx_id)) { + while (countdown && mbx->ops[mbx_id].check_for_ack(hw, mbx_id)) { countdown--; if (!countdown) break; usec_delay(mbx->usec_delay); } - if (countdown == 0) + if (countdown == 0) { ERROR_REPORT2(IXGBE_ERROR_POLLING, - "Polling for VF%d mailbox ack timedout", mbx_id); + "Polling for VF%u mailbox ack timedout", mbx_id); + return IXGBE_ERR_TIMEOUT; + } -out: - return countdown ? IXGBE_SUCCESS : IXGBE_ERR_MBX; + return IXGBE_SUCCESS; } /** - * ixgbe_read_posted_mbx - Wait for message notification and receive message + * ixgbe_read_mailbox_vf - read VF's mailbox register * @hw: pointer to the HW structure - * @msg: The message buffer - * @size: Length of buffer - * @mbx_id: id of mailbox to write * - * returns SUCCESS if it successfully received a message notification and - * copied it into the receive buffer. + * This function is used to read the mailbox register dedicated for VF without + * losing the read to clear status bits. **/ -s32 ixgbe_read_posted_mbx(struct ixgbe_hw *hw, u32 *msg, u16 size, u16 mbx_id) +STATIC u32 ixgbe_read_mailbox_vf(struct ixgbe_hw *hw) { - struct ixgbe_mbx_info *mbx = &hw->mbx; - s32 ret_val = IXGBE_ERR_MBX; + u32 vf_mailbox = IXGBE_READ_REG(hw, IXGBE_VFMAILBOX); - DEBUGFUNC("ixgbe_read_posted_mbx"); + vf_mailbox |= hw->mbx.vf_mailbox; + hw->mbx.vf_mailbox |= vf_mailbox & IXGBE_VFMAILBOX_R2C_BITS; - if (!mbx->ops.read) - goto out; - - ret_val = ixgbe_poll_for_msg(hw, mbx_id); - - /* if ack received read message, otherwise we timed out */ - if (!ret_val) - ret_val = mbx->ops.read(hw, msg, size, mbx_id); -out: - return ret_val; + return vf_mailbox; } -/** - * ixgbe_write_posted_mbx - Write a message to the mailbox, wait for ack - * @hw: pointer to the HW structure - * @msg: The message buffer - * @size: Length of buffer - * @mbx_id: id of mailbox to write - * - * returns SUCCESS if it successfully copied message into the buffer and - * received an ack to that message within delay * timeout period - **/ -s32 ixgbe_write_posted_mbx(struct ixgbe_hw *hw, u32 *msg, u16 size, - u16 mbx_id) +STATIC void ixgbe_clear_msg_vf(struct ixgbe_hw *hw) { - struct ixgbe_mbx_info *mbx = &hw->mbx; - s32 ret_val = IXGBE_ERR_MBX; - - DEBUGFUNC("ixgbe_write_posted_mbx"); - - /* exit if either we can't write or there isn't a defined timeout */ - if (!mbx->ops.write || !mbx->timeout) - goto out; - - /* send msg */ - ret_val = mbx->ops.write(hw, msg, size, mbx_id); + u32 vf_mailbox = ixgbe_read_mailbox_vf(hw); - /* if msg sent wait until we receive an ack */ - if (!ret_val) - ret_val = ixgbe_poll_for_ack(hw, mbx_id); -out: - return ret_val; + if (vf_mailbox & IXGBE_VFMAILBOX_PFSTS) { + hw->mbx.stats.reqs++; + hw->mbx.vf_mailbox &= ~IXGBE_VFMAILBOX_PFSTS; + } } -/** - * ixgbe_init_mbx_ops_generic - Initialize MB function pointers - * @hw: pointer to the HW structure - * - * Setups up the mailbox read and write message function pointers - **/ -void ixgbe_init_mbx_ops_generic(struct ixgbe_hw *hw) +STATIC void ixgbe_clear_ack_vf(struct ixgbe_hw *hw) { - struct ixgbe_mbx_info *mbx = &hw->mbx; + u32 vf_mailbox = ixgbe_read_mailbox_vf(hw); - mbx->ops.read_posted = ixgbe_read_posted_mbx; - mbx->ops.write_posted = ixgbe_write_posted_mbx; + if (vf_mailbox & IXGBE_VFMAILBOX_PFACK) { + hw->mbx.stats.acks++; + hw->mbx.vf_mailbox &= ~IXGBE_VFMAILBOX_PFACK; + } } -/** - * ixgbe_read_v2p_mailbox - read v2p mailbox - * @hw: pointer to the HW structure - * - * This function is used to read the v2p mailbox without losing the read to - * clear status bits. - **/ -STATIC u32 ixgbe_read_v2p_mailbox(struct ixgbe_hw *hw) +STATIC void ixgbe_clear_rst_vf(struct ixgbe_hw *hw) { - u32 v2p_mailbox = IXGBE_READ_REG(hw, IXGBE_VFMAILBOX); - - v2p_mailbox |= hw->mbx.v2p_mailbox; - hw->mbx.v2p_mailbox |= v2p_mailbox & IXGBE_VFMAILBOX_R2C_BITS; + u32 vf_mailbox = ixgbe_read_mailbox_vf(hw); - return v2p_mailbox; + if (vf_mailbox & (IXGBE_VFMAILBOX_RSTI | IXGBE_VFMAILBOX_RSTD)) { + hw->mbx.stats.rsts++; + hw->mbx.vf_mailbox &= ~(IXGBE_VFMAILBOX_RSTI | + IXGBE_VFMAILBOX_RSTD); + } } /** @@ -283,15 +293,12 @@ STATIC u32 ixgbe_read_v2p_mailbox(struct ixgbe_hw *hw) **/ STATIC s32 ixgbe_check_for_bit_vf(struct ixgbe_hw *hw, u32 mask) { - u32 v2p_mailbox = ixgbe_read_v2p_mailbox(hw); - s32 ret_val = IXGBE_ERR_MBX; + u32 vf_mailbox = ixgbe_read_mailbox_vf(hw); - if (v2p_mailbox & mask) - ret_val = IXGBE_SUCCESS; + if (vf_mailbox & mask) + return IXGBE_SUCCESS; - hw->mbx.v2p_mailbox &= ~mask; - - return ret_val; + return IXGBE_ERR_MBX; } /** @@ -303,17 +310,13 @@ STATIC s32 ixgbe_check_for_bit_vf(struct ixgbe_hw *hw, u32 mask) **/ STATIC s32 ixgbe_check_for_msg_vf(struct ixgbe_hw *hw, u16 mbx_id) { - s32 ret_val = IXGBE_ERR_MBX; - UNREFERENCED_1PARAMETER(mbx_id); DEBUGFUNC("ixgbe_check_for_msg_vf"); - if (!ixgbe_check_for_bit_vf(hw, IXGBE_VFMAILBOX_PFSTS)) { - ret_val = IXGBE_SUCCESS; - hw->mbx.stats.reqs++; - } + if (!ixgbe_check_for_bit_vf(hw, IXGBE_VFMAILBOX_PFSTS)) + return IXGBE_SUCCESS; - return ret_val; + return IXGBE_ERR_MBX; } /** @@ -325,17 +328,16 @@ STATIC s32 ixgbe_check_for_msg_vf(struct ixgbe_hw *hw, u16 mbx_id) **/ STATIC s32 ixgbe_check_for_ack_vf(struct ixgbe_hw *hw, u16 mbx_id) { - s32 ret_val = IXGBE_ERR_MBX; - UNREFERENCED_1PARAMETER(mbx_id); DEBUGFUNC("ixgbe_check_for_ack_vf"); if (!ixgbe_check_for_bit_vf(hw, IXGBE_VFMAILBOX_PFACK)) { - ret_val = IXGBE_SUCCESS; - hw->mbx.stats.acks++; + /* TODO: should this be autocleared? */ + ixgbe_clear_ack_vf(hw); + return IXGBE_SUCCESS; } - return ret_val; + return IXGBE_ERR_MBX; } /** @@ -347,18 +349,17 @@ STATIC s32 ixgbe_check_for_ack_vf(struct ixgbe_hw *hw, u16 mbx_id) **/ STATIC s32 ixgbe_check_for_rst_vf(struct ixgbe_hw *hw, u16 mbx_id) { - s32 ret_val = IXGBE_ERR_MBX; - UNREFERENCED_1PARAMETER(mbx_id); DEBUGFUNC("ixgbe_check_for_rst_vf"); - if (!ixgbe_check_for_bit_vf(hw, (IXGBE_VFMAILBOX_RSTD | - IXGBE_VFMAILBOX_RSTI))) { - ret_val = IXGBE_SUCCESS; - hw->mbx.stats.rsts++; + if (!ixgbe_check_for_bit_vf(hw, IXGBE_VFMAILBOX_RSTI | + IXGBE_VFMAILBOX_RSTD)) { + /* TODO: should this be autocleared? */ + ixgbe_clear_rst_vf(hw); + return IXGBE_SUCCESS; } - return ret_val; + return IXGBE_ERR_MBX; } /** @@ -369,20 +370,114 @@ STATIC s32 ixgbe_check_for_rst_vf(struct ixgbe_hw *hw, u16 mbx_id) **/ STATIC s32 ixgbe_obtain_mbx_lock_vf(struct ixgbe_hw *hw) { + struct ixgbe_mbx_info *mbx = &hw->mbx; + int countdown = mbx->timeout; s32 ret_val = IXGBE_ERR_MBX; + u32 vf_mailbox; DEBUGFUNC("ixgbe_obtain_mbx_lock_vf"); - /* Take ownership of the buffer */ - IXGBE_WRITE_REG(hw, IXGBE_VFMAILBOX, IXGBE_VFMAILBOX_VFU); + if (!mbx->timeout) + return IXGBE_ERR_CONFIG; - /* reserve mailbox for vf use */ - if (ixgbe_read_v2p_mailbox(hw) & IXGBE_VFMAILBOX_VFU) - ret_val = IXGBE_SUCCESS; + while (countdown--) { + /* Reserve mailbox for VF use */ + vf_mailbox = ixgbe_read_mailbox_vf(hw); + vf_mailbox |= IXGBE_VFMAILBOX_VFU; + IXGBE_WRITE_REG(hw, IXGBE_VFMAILBOX, vf_mailbox); + + /* Verify that VF is the owner of the lock */ + if (ixgbe_read_mailbox_vf(hw) & IXGBE_VFMAILBOX_VFU) { + ret_val = IXGBE_SUCCESS; + break; + } + + /* Wait a bit before trying again */ + usec_delay(mbx->usec_delay); + } + + if (ret_val != IXGBE_SUCCESS) { + ERROR_REPORT1(IXGBE_ERROR_INVALID_STATE, + "Failed to obtain mailbox lock"); + ret_val = IXGBE_ERR_TIMEOUT; + } return ret_val; } +/** + * ixgbe_release_mbx_lock_dummy - release mailbox lock + * @hw: pointer to the HW structure + * @mbx_id: id of mailbox to read + **/ +STATIC void ixgbe_release_mbx_lock_dummy(struct ixgbe_hw *hw, u16 mbx_id) +{ + UNREFERENCED_2PARAMETER(hw, mbx_id); + + DEBUGFUNC("ixgbe_release_mbx_lock_dummy"); +} + +/** + * ixgbe_release_mbx_lock_vf - release mailbox lock + * @hw: pointer to the HW structure + * @mbx_id: id of mailbox to read + **/ +STATIC void ixgbe_release_mbx_lock_vf(struct ixgbe_hw *hw, u16 mbx_id) +{ + u32 vf_mailbox; + + UNREFERENCED_1PARAMETER(mbx_id); + + DEBUGFUNC("ixgbe_release_mbx_lock_vf"); + + /* Return ownership of the buffer */ + vf_mailbox = ixgbe_read_mailbox_vf(hw); + vf_mailbox &= ~IXGBE_VFMAILBOX_VFU; + IXGBE_WRITE_REG(hw, IXGBE_VFMAILBOX, vf_mailbox); +} + +/** + * ixgbe_write_mbx_vf_legacy - Write a message to the mailbox + * @hw: pointer to the HW structure + * @msg: The message buffer + * @size: Length of buffer + * @mbx_id: id of mailbox to write + * + * returns SUCCESS if it successfully copied message into the buffer + **/ +STATIC s32 ixgbe_write_mbx_vf_legacy(struct ixgbe_hw *hw, u32 *msg, u16 size, + u16 mbx_id) +{ + s32 ret_val; + u16 i; + + UNREFERENCED_1PARAMETER(mbx_id); + DEBUGFUNC("ixgbe_write_mbx_vf_legacy"); + + /* lock the mailbox to prevent pf/vf race condition */ + ret_val = ixgbe_obtain_mbx_lock_vf(hw); + if (ret_val) + return ret_val; + + /* flush msg and acks as we are overwriting the message buffer */ + ixgbe_check_for_msg_vf(hw, 0); + ixgbe_clear_msg_vf(hw); + ixgbe_check_for_ack_vf(hw, 0); + ixgbe_clear_ack_vf(hw); + + /* copy the caller specified message to the mailbox memory buffer */ + for (i = 0; i < size; i++) + IXGBE_WRITE_REG_ARRAY(hw, IXGBE_VFMBMEM, i, msg[i]); + + /* update stats */ + hw->mbx.stats.msgs_tx++; + + /* interrupt the PF to tell it a message has been sent */ + IXGBE_WRITE_REG(hw, IXGBE_VFMAILBOX, IXGBE_VFMAILBOX_REQ); + + return IXGBE_SUCCESS; +} + /** * ixgbe_write_mbx_vf - Write a message to the mailbox * @hw: pointer to the HW structure @@ -395,6 +490,7 @@ STATIC s32 ixgbe_obtain_mbx_lock_vf(struct ixgbe_hw *hw) STATIC s32 ixgbe_write_mbx_vf(struct ixgbe_hw *hw, u32 *msg, u16 size, u16 mbx_id) { + u32 vf_mailbox; s32 ret_val; u16 i; @@ -405,11 +501,11 @@ STATIC s32 ixgbe_write_mbx_vf(struct ixgbe_hw *hw, u32 *msg, u16 size, /* lock the mailbox to prevent pf/vf race condition */ ret_val = ixgbe_obtain_mbx_lock_vf(hw); if (ret_val) - goto out_no_write; + goto out; /* flush msg and acks as we are overwriting the message buffer */ - ixgbe_check_for_msg_vf(hw, 0); - ixgbe_check_for_ack_vf(hw, 0); + ixgbe_clear_msg_vf(hw); + ixgbe_clear_ack_vf(hw); /* copy the caller specified message to the mailbox memory buffer */ for (i = 0; i < size; i++) @@ -418,13 +514,56 @@ STATIC s32 ixgbe_write_mbx_vf(struct ixgbe_hw *hw, u32 *msg, u16 size, /* update stats */ hw->mbx.stats.msgs_tx++; - /* Drop VFU and interrupt the PF to tell it a message has been sent */ - IXGBE_WRITE_REG(hw, IXGBE_VFMAILBOX, IXGBE_VFMAILBOX_REQ); + /* interrupt the PF to tell it a message has been sent */ + vf_mailbox = ixgbe_read_mailbox_vf(hw); + vf_mailbox |= IXGBE_VFMAILBOX_REQ; + IXGBE_WRITE_REG(hw, IXGBE_VFMAILBOX, vf_mailbox); + + /* if msg sent wait until we receive an ack */ + ixgbe_poll_for_ack(hw, mbx_id); + +out: + hw->mbx.ops[mbx_id].release(hw, mbx_id); -out_no_write: return ret_val; } +/** + * ixgbe_read_mbx_vf_legacy - Reads a message from the inbox intended for vf + * @hw: pointer to the HW structure + * @msg: The message buffer + * @size: Length of buffer + * @mbx_id: id of mailbox to read + * + * returns SUCCESS if it successfully read message from buffer + **/ +STATIC s32 ixgbe_read_mbx_vf_legacy(struct ixgbe_hw *hw, u32 *msg, u16 size, + u16 mbx_id) +{ + s32 ret_val; + u16 i; + + DEBUGFUNC("ixgbe_read_mbx_vf_legacy"); + UNREFERENCED_1PARAMETER(mbx_id); + + /* lock the mailbox to prevent pf/vf race condition */ + ret_val = ixgbe_obtain_mbx_lock_vf(hw); + if (ret_val) + return ret_val; + + /* copy the message from the mailbox memory buffer */ + for (i = 0; i < size; i++) + msg[i] = IXGBE_READ_REG_ARRAY(hw, IXGBE_VFMBMEM, i); + + /* Acknowledge receipt and release mailbox, then we're done */ + IXGBE_WRITE_REG(hw, IXGBE_VFMAILBOX, IXGBE_VFMAILBOX_ACK); + + /* update stats */ + hw->mbx.stats.msgs_rx++; + + return IXGBE_SUCCESS; +} + /** * ixgbe_read_mbx_vf - Reads a message from the inbox intended for vf * @hw: pointer to the HW structure @@ -437,55 +576,58 @@ STATIC s32 ixgbe_write_mbx_vf(struct ixgbe_hw *hw, u32 *msg, u16 size, STATIC s32 ixgbe_read_mbx_vf(struct ixgbe_hw *hw, u32 *msg, u16 size, u16 mbx_id) { - s32 ret_val = IXGBE_SUCCESS; + u32 vf_mailbox; + s32 ret_val; u16 i; DEBUGFUNC("ixgbe_read_mbx_vf"); UNREFERENCED_1PARAMETER(mbx_id); - /* lock the mailbox to prevent pf/vf race condition */ - ret_val = ixgbe_obtain_mbx_lock_vf(hw); - if (ret_val) - goto out_no_read; + /* check if there is a message from PF */ + ret_val = ixgbe_check_for_msg_vf(hw, 0); + if (ret_val != IXGBE_SUCCESS) + return IXGBE_ERR_MBX_NOMSG; + + ixgbe_clear_msg_vf(hw); /* copy the message from the mailbox memory buffer */ for (i = 0; i < size; i++) msg[i] = IXGBE_READ_REG_ARRAY(hw, IXGBE_VFMBMEM, i); - /* Acknowledge receipt and release mailbox, then we're done */ - IXGBE_WRITE_REG(hw, IXGBE_VFMAILBOX, IXGBE_VFMAILBOX_ACK); + /* Acknowledge receipt */ + vf_mailbox = ixgbe_read_mailbox_vf(hw); + vf_mailbox |= IXGBE_VFMAILBOX_ACK; + IXGBE_WRITE_REG(hw, IXGBE_VFMAILBOX, vf_mailbox); /* update stats */ hw->mbx.stats.msgs_rx++; -out_no_read: - return ret_val; + return IXGBE_SUCCESS; } /** * ixgbe_init_mbx_params_vf - set initial values for vf mailbox * @hw: pointer to the HW structure * - * Initializes the hw->mbx struct to correct values for vf mailbox + * Initializes single set the hw->mbx struct to correct values for vf mailbox + * Set of legacy functions is being used here */ void ixgbe_init_mbx_params_vf(struct ixgbe_hw *hw) { struct ixgbe_mbx_info *mbx = &hw->mbx; - /* start mailbox as timed out and let the reset_hw call set the timeout - * value to begin communications */ - mbx->timeout = 0; + mbx->timeout = IXGBE_VF_MBX_INIT_TIMEOUT; mbx->usec_delay = IXGBE_VF_MBX_INIT_DELAY; mbx->size = IXGBE_VFMAILBOX_SIZE; - mbx->ops.read = ixgbe_read_mbx_vf; - mbx->ops.write = ixgbe_write_mbx_vf; - mbx->ops.read_posted = ixgbe_read_posted_mbx; - mbx->ops.write_posted = ixgbe_write_posted_mbx; - mbx->ops.check_for_msg = ixgbe_check_for_msg_vf; - mbx->ops.check_for_ack = ixgbe_check_for_ack_vf; - mbx->ops.check_for_rst = ixgbe_check_for_rst_vf; + /* VF has only one mailbox connection, no need for more IDs */ + mbx->ops[0].release = ixgbe_release_mbx_lock_dummy; + mbx->ops[0].read = ixgbe_read_mbx_vf_legacy; + mbx->ops[0].write = ixgbe_write_mbx_vf_legacy; + mbx->ops[0].check_for_msg = ixgbe_check_for_msg_vf; + mbx->ops[0].check_for_ack = ixgbe_check_for_ack_vf; + mbx->ops[0].check_for_rst = ixgbe_check_for_rst_vf; mbx->stats.msgs_tx = 0; mbx->stats.msgs_rx = 0; @@ -494,54 +636,109 @@ void ixgbe_init_mbx_params_vf(struct ixgbe_hw *hw) mbx->stats.rsts = 0; } +/** + * ixgbe_upgrade_mbx_params_vf - set initial values for vf mailbox + * @hw: pointer to the HW structure + * + * Initializes the hw->mbx struct to correct values for vf mailbox + */ +void ixgbe_upgrade_mbx_params_vf(struct ixgbe_hw *hw) +{ + struct ixgbe_mbx_info *mbx = &hw->mbx; + + mbx->timeout = IXGBE_VF_MBX_INIT_TIMEOUT; + mbx->usec_delay = IXGBE_VF_MBX_INIT_DELAY; + + mbx->size = IXGBE_VFMAILBOX_SIZE; + + /* VF has only one mailbox connection, no need for more IDs */ + mbx->ops[0].release = ixgbe_release_mbx_lock_vf; + mbx->ops[0].read = ixgbe_read_mbx_vf; + mbx->ops[0].write = ixgbe_write_mbx_vf; + mbx->ops[0].check_for_msg = ixgbe_check_for_msg_vf; + mbx->ops[0].check_for_ack = ixgbe_check_for_ack_vf; + mbx->ops[0].check_for_rst = ixgbe_check_for_rst_vf; + mbx->ops[0].clear = NULL; + + mbx->stats.msgs_tx = 0; + mbx->stats.msgs_rx = 0; + mbx->stats.reqs = 0; + mbx->stats.acks = 0; + mbx->stats.rsts = 0; +} + +STATIC void ixgbe_clear_msg_pf(struct ixgbe_hw *hw, u16 vf_id) +{ + u32 vf_shift = IXGBE_PFMBICR_SHIFT(vf_id); + s32 index = IXGBE_PFMBICR_INDEX(vf_id); + u32 pfmbicr; + + pfmbicr = IXGBE_READ_REG(hw, IXGBE_PFMBICR(index)); + + if (pfmbicr & (IXGBE_PFMBICR_VFREQ_VF1 << vf_shift)) + hw->mbx.stats.reqs++; + + IXGBE_WRITE_REG(hw, IXGBE_PFMBICR(index), + IXGBE_PFMBICR_VFREQ_VF1 << vf_shift); +} + +STATIC void ixgbe_clear_ack_pf(struct ixgbe_hw *hw, u16 vf_id) +{ + u32 vf_shift = IXGBE_PFMBICR_SHIFT(vf_id); + s32 index = IXGBE_PFMBICR_INDEX(vf_id); + u32 pfmbicr; + + pfmbicr = IXGBE_READ_REG(hw, IXGBE_PFMBICR(index)); + + if (pfmbicr & (IXGBE_PFMBICR_VFACK_VF1 << vf_shift)) + hw->mbx.stats.acks++; + + IXGBE_WRITE_REG(hw, IXGBE_PFMBICR(index), + IXGBE_PFMBICR_VFACK_VF1 << vf_shift); +} + STATIC s32 ixgbe_check_for_bit_pf(struct ixgbe_hw *hw, u32 mask, s32 index) { u32 pfmbicr = IXGBE_READ_REG(hw, IXGBE_PFMBICR(index)); - s32 ret_val = IXGBE_ERR_MBX; - if (pfmbicr & mask) { - ret_val = IXGBE_SUCCESS; - IXGBE_WRITE_REG(hw, IXGBE_PFMBICR(index), mask); - } + if (pfmbicr & mask) + return IXGBE_SUCCESS; - return ret_val; + return IXGBE_ERR_MBX; } /** * ixgbe_check_for_msg_pf - checks to see if the VF has sent mail * @hw: pointer to the HW structure - * @vf_number: the VF index + * @vf_id: the VF index * * returns SUCCESS if the VF has set the Status bit or else ERR_MBX **/ -STATIC s32 ixgbe_check_for_msg_pf(struct ixgbe_hw *hw, u16 vf_number) +STATIC s32 ixgbe_check_for_msg_pf(struct ixgbe_hw *hw, u16 vf_id) { - u32 vf_shift = IXGBE_PFMBICR_SHIFT(vf_number); - s32 index = IXGBE_PFMBICR_INDEX(vf_number); - s32 ret_val = IXGBE_ERR_MBX; + u32 vf_shift = IXGBE_PFMBICR_SHIFT(vf_id); + s32 index = IXGBE_PFMBICR_INDEX(vf_id); DEBUGFUNC("ixgbe_check_for_msg_pf"); if (!ixgbe_check_for_bit_pf(hw, IXGBE_PFMBICR_VFREQ_VF1 << vf_shift, - index)) { - ret_val = IXGBE_SUCCESS; - hw->mbx.stats.reqs++; - } + index)) + return IXGBE_SUCCESS; - return ret_val; + return IXGBE_ERR_MBX; } /** * ixgbe_check_for_ack_pf - checks to see if the VF has ACKed * @hw: pointer to the HW structure - * @vf_number: the VF index + * @vf_id: the VF index * * returns SUCCESS if the VF has set the Status bit or else ERR_MBX **/ -STATIC s32 ixgbe_check_for_ack_pf(struct ixgbe_hw *hw, u16 vf_number) +STATIC s32 ixgbe_check_for_ack_pf(struct ixgbe_hw *hw, u16 vf_id) { - u32 vf_shift = IXGBE_PFMBICR_SHIFT(vf_number); - s32 index = IXGBE_PFMBICR_INDEX(vf_number); + u32 vf_shift = IXGBE_PFMBICR_SHIFT(vf_id); + s32 index = IXGBE_PFMBICR_INDEX(vf_id); s32 ret_val = IXGBE_ERR_MBX; DEBUGFUNC("ixgbe_check_for_ack_pf"); @@ -549,7 +746,8 @@ STATIC s32 ixgbe_check_for_ack_pf(struct ixgbe_hw *hw, u16 vf_number) if (!ixgbe_check_for_bit_pf(hw, IXGBE_PFMBICR_VFACK_VF1 << vf_shift, index)) { ret_val = IXGBE_SUCCESS; - hw->mbx.stats.acks++; + /* TODO: should this be autocleared? */ + ixgbe_clear_ack_pf(hw, vf_id); } return ret_val; @@ -558,14 +756,14 @@ STATIC s32 ixgbe_check_for_ack_pf(struct ixgbe_hw *hw, u16 vf_number) /** * ixgbe_check_for_rst_pf - checks to see if the VF has reset * @hw: pointer to the HW structure - * @vf_number: the VF index + * @vf_id: the VF index * * returns SUCCESS if the VF has set the Status bit or else ERR_MBX **/ -STATIC s32 ixgbe_check_for_rst_pf(struct ixgbe_hw *hw, u16 vf_number) +STATIC s32 ixgbe_check_for_rst_pf(struct ixgbe_hw *hw, u16 vf_id) { - u32 vf_shift = IXGBE_PFVFLRE_SHIFT(vf_number); - u32 index = IXGBE_PFVFLRE_INDEX(vf_number); + u32 vf_shift = IXGBE_PFVFLRE_SHIFT(vf_id); + u32 index = IXGBE_PFVFLRE_INDEX(vf_id); s32 ret_val = IXGBE_ERR_MBX; u32 vflre = 0; @@ -597,121 +795,268 @@ STATIC s32 ixgbe_check_for_rst_pf(struct ixgbe_hw *hw, u16 vf_number) /** * ixgbe_obtain_mbx_lock_pf - obtain mailbox lock * @hw: pointer to the HW structure - * @vf_number: the VF index + * @vf_id: the VF index * * return SUCCESS if we obtained the mailbox lock **/ -STATIC s32 ixgbe_obtain_mbx_lock_pf(struct ixgbe_hw *hw, u16 vf_number) +STATIC s32 ixgbe_obtain_mbx_lock_pf(struct ixgbe_hw *hw, u16 vf_id) { + struct ixgbe_mbx_info *mbx = &hw->mbx; + int countdown = mbx->timeout; s32 ret_val = IXGBE_ERR_MBX; - u32 p2v_mailbox; + u32 pf_mailbox; DEBUGFUNC("ixgbe_obtain_mbx_lock_pf"); - /* Take ownership of the buffer */ - IXGBE_WRITE_REG(hw, IXGBE_PFMAILBOX(vf_number), IXGBE_PFMAILBOX_PFU); + if (!mbx->timeout) + return IXGBE_ERR_CONFIG; - /* reserve mailbox for vf use */ - p2v_mailbox = IXGBE_READ_REG(hw, IXGBE_PFMAILBOX(vf_number)); - if (p2v_mailbox & IXGBE_PFMAILBOX_PFU) - ret_val = IXGBE_SUCCESS; - else - ERROR_REPORT2(IXGBE_ERROR_POLLING, - "Failed to obtain mailbox lock for VF%d", vf_number); + while (countdown--) { + /* Reserve mailbox for PF use */ + pf_mailbox = IXGBE_READ_REG(hw, IXGBE_PFMAILBOX(vf_id)); + pf_mailbox |= IXGBE_PFMAILBOX_PFU; + IXGBE_WRITE_REG(hw, IXGBE_PFMAILBOX(vf_id), pf_mailbox); + /* Verify that PF is the owner of the lock */ + pf_mailbox = IXGBE_READ_REG(hw, IXGBE_PFMAILBOX(vf_id)); + if (pf_mailbox & IXGBE_PFMAILBOX_PFU) { + ret_val = IXGBE_SUCCESS; + break; + } + + /* Wait a bit before trying again */ + usec_delay(mbx->usec_delay); + } + + if (ret_val != IXGBE_SUCCESS) { + ERROR_REPORT1(IXGBE_ERROR_INVALID_STATE, + "Failed to obtain mailbox lock"); + ret_val = IXGBE_ERR_TIMEOUT; + } return ret_val; } +/** + * ixgbe_release_mbx_lock_pf - release mailbox lock + * @hw: pointer to the HW structure + * @vf_id: the VF index + **/ +STATIC void ixgbe_release_mbx_lock_pf(struct ixgbe_hw *hw, u16 vf_id) +{ + u32 pf_mailbox; + + DEBUGFUNC("ixgbe_release_mbx_lock_pf"); + + /* Return ownership of the buffer */ + pf_mailbox = IXGBE_READ_REG(hw, IXGBE_PFMAILBOX(vf_id)); + pf_mailbox &= ~IXGBE_PFMAILBOX_PFU; + IXGBE_WRITE_REG(hw, IXGBE_PFMAILBOX(vf_id), pf_mailbox); +} + +/** + * ixgbe_write_mbx_pf_legacy - Places a message in the mailbox + * @hw: pointer to the HW structure + * @msg: The message buffer + * @size: Length of buffer + * @vf_id: the VF index + * + * returns SUCCESS if it successfully copied message into the buffer + **/ +STATIC s32 ixgbe_write_mbx_pf_legacy(struct ixgbe_hw *hw, u32 *msg, u16 size, + u16 vf_id) +{ + s32 ret_val; + u16 i; + + DEBUGFUNC("ixgbe_write_mbx_pf_legacy"); + + /* lock the mailbox to prevent pf/vf race condition */ + ret_val = ixgbe_obtain_mbx_lock_pf(hw, vf_id); + if (ret_val) + return ret_val; + + /* flush msg and acks as we are overwriting the message buffer */ + ixgbe_check_for_msg_pf(hw, vf_id); + ixgbe_clear_msg_pf(hw, vf_id); + ixgbe_check_for_ack_pf(hw, vf_id); + ixgbe_clear_ack_pf(hw, vf_id); + + /* copy the caller specified message to the mailbox memory buffer */ + for (i = 0; i < size; i++) + IXGBE_WRITE_REG_ARRAY(hw, IXGBE_PFMBMEM(vf_id), i, msg[i]); + + /* Interrupt VF to tell it a message has been sent and release buffer*/ + IXGBE_WRITE_REG(hw, IXGBE_PFMAILBOX(vf_id), IXGBE_PFMAILBOX_STS); + + /* update stats */ + hw->mbx.stats.msgs_tx++; + + return IXGBE_SUCCESS; +} + /** * ixgbe_write_mbx_pf - Places a message in the mailbox * @hw: pointer to the HW structure * @msg: The message buffer * @size: Length of buffer - * @vf_number: the VF index + * @vf_id: the VF index * * returns SUCCESS if it successfully copied message into the buffer **/ STATIC s32 ixgbe_write_mbx_pf(struct ixgbe_hw *hw, u32 *msg, u16 size, - u16 vf_number) + u16 vf_id) { + u32 pf_mailbox; s32 ret_val; u16 i; DEBUGFUNC("ixgbe_write_mbx_pf"); /* lock the mailbox to prevent pf/vf race condition */ - ret_val = ixgbe_obtain_mbx_lock_pf(hw, vf_number); + ret_val = ixgbe_obtain_mbx_lock_pf(hw, vf_id); if (ret_val) - goto out_no_write; + goto out; /* flush msg and acks as we are overwriting the message buffer */ - ixgbe_check_for_msg_pf(hw, vf_number); - ixgbe_check_for_ack_pf(hw, vf_number); + ixgbe_clear_msg_pf(hw, vf_id); + ixgbe_clear_ack_pf(hw, vf_id); /* copy the caller specified message to the mailbox memory buffer */ for (i = 0; i < size; i++) - IXGBE_WRITE_REG_ARRAY(hw, IXGBE_PFMBMEM(vf_number), i, msg[i]); + IXGBE_WRITE_REG_ARRAY(hw, IXGBE_PFMBMEM(vf_id), i, msg[i]); - /* Interrupt VF to tell it a message has been sent and release buffer*/ - IXGBE_WRITE_REG(hw, IXGBE_PFMAILBOX(vf_number), IXGBE_PFMAILBOX_STS); + /* Interrupt VF to tell it a message has been sent */ + pf_mailbox = IXGBE_READ_REG(hw, IXGBE_PFMAILBOX(vf_id)); + pf_mailbox |= IXGBE_PFMAILBOX_STS; + IXGBE_WRITE_REG(hw, IXGBE_PFMAILBOX(vf_id), pf_mailbox); + + /* if msg sent wait until we receive an ack */ + ixgbe_poll_for_ack(hw, vf_id); /* update stats */ hw->mbx.stats.msgs_tx++; -out_no_write: +out: + hw->mbx.ops[vf_id].release(hw, vf_id); + return ret_val; } +/** + * ixgbe_read_mbx_pf_legacy - Read a message from the mailbox + * @hw: pointer to the HW structure + * @msg: The message buffer + * @size: Length of buffer + * @vf_id: the VF index + * + * This function copies a message from the mailbox buffer to the caller's + * memory buffer. The presumption is that the caller knows that there was + * a message due to a VF request so no polling for message is needed. + **/ +STATIC s32 ixgbe_read_mbx_pf_legacy(struct ixgbe_hw *hw, u32 *msg, u16 size, + u16 vf_id) +{ + s32 ret_val; + u16 i; + + DEBUGFUNC("ixgbe_read_mbx_pf_legacy"); + + /* lock the mailbox to prevent pf/vf race condition */ + ret_val = ixgbe_obtain_mbx_lock_pf(hw, vf_id); + if (ret_val != IXGBE_SUCCESS) + return ret_val; + + /* copy the message to the mailbox memory buffer */ + for (i = 0; i < size; i++) + msg[i] = IXGBE_READ_REG_ARRAY(hw, IXGBE_PFMBMEM(vf_id), i); + + /* Acknowledge the message and release buffer */ + IXGBE_WRITE_REG(hw, IXGBE_PFMAILBOX(vf_id), IXGBE_PFMAILBOX_ACK); + + /* update stats */ + hw->mbx.stats.msgs_rx++; + + return IXGBE_SUCCESS; +} + /** * ixgbe_read_mbx_pf - Read a message from the mailbox * @hw: pointer to the HW structure * @msg: The message buffer * @size: Length of buffer - * @vf_number: the VF index + * @vf_id: the VF index * * This function copies a message from the mailbox buffer to the caller's * memory buffer. The presumption is that the caller knows that there was * a message due to a VF request so no polling for message is needed. **/ STATIC s32 ixgbe_read_mbx_pf(struct ixgbe_hw *hw, u32 *msg, u16 size, - u16 vf_number) + u16 vf_id) { + u32 pf_mailbox; s32 ret_val; u16 i; DEBUGFUNC("ixgbe_read_mbx_pf"); - /* lock the mailbox to prevent pf/vf race condition */ - ret_val = ixgbe_obtain_mbx_lock_pf(hw, vf_number); - if (ret_val) - goto out_no_read; + /* check if there is a message from VF */ + ret_val = ixgbe_check_for_msg_pf(hw, vf_id); + if (ret_val != IXGBE_SUCCESS) + return IXGBE_ERR_MBX_NOMSG; + + ixgbe_clear_msg_pf(hw, vf_id); /* copy the message to the mailbox memory buffer */ for (i = 0; i < size; i++) - msg[i] = IXGBE_READ_REG_ARRAY(hw, IXGBE_PFMBMEM(vf_number), i); + msg[i] = IXGBE_READ_REG_ARRAY(hw, IXGBE_PFMBMEM(vf_id), i); /* Acknowledge the message and release buffer */ - IXGBE_WRITE_REG(hw, IXGBE_PFMAILBOX(vf_number), IXGBE_PFMAILBOX_ACK); + pf_mailbox = IXGBE_READ_REG(hw, IXGBE_PFMAILBOX(vf_id)); + pf_mailbox |= IXGBE_PFMAILBOX_ACK; + IXGBE_WRITE_REG(hw, IXGBE_PFMAILBOX(vf_id), pf_mailbox); /* update stats */ hw->mbx.stats.msgs_rx++; -out_no_read: - return ret_val; + return IXGBE_SUCCESS; +} + +/** + * ixgbe_init_mbx_params_pf_id - set initial values for pf mailbox + * @hw: pointer to the HW structure + * @vf_id: the VF index + * + * Initializes single set of the hw->mbx struct to correct values for pf mailbox + * Set of legacy functions is being used here + */ +void ixgbe_init_mbx_params_pf_id(struct ixgbe_hw *hw, u16 vf_id) +{ + struct ixgbe_mbx_info *mbx = &hw->mbx; + + mbx->ops[vf_id].release = ixgbe_release_mbx_lock_dummy; + mbx->ops[vf_id].read = ixgbe_read_mbx_pf_legacy; + mbx->ops[vf_id].write = ixgbe_write_mbx_pf_legacy; + mbx->ops[vf_id].check_for_msg = ixgbe_check_for_msg_pf; + mbx->ops[vf_id].check_for_ack = ixgbe_check_for_ack_pf; + mbx->ops[vf_id].check_for_rst = ixgbe_check_for_rst_pf; } /** * ixgbe_init_mbx_params_pf - set initial values for pf mailbox * @hw: pointer to the HW structure * - * Initializes the hw->mbx struct to correct values for pf mailbox + * Initializes all sets of the hw->mbx struct to correct values for pf + * mailbox. One set corresponds to single VF. It also initializes counters + * and general variables. A set of legacy functions is used by default. */ void ixgbe_init_mbx_params_pf(struct ixgbe_hw *hw) { + u16 i; struct ixgbe_mbx_info *mbx = &hw->mbx; + /* Ensure we are not calling this function from VF */ if (hw->mac.type != ixgbe_mac_82599EB && hw->mac.type != ixgbe_mac_X550 && hw->mac.type != ixgbe_mac_X550EM_x && @@ -719,18 +1064,58 @@ void ixgbe_init_mbx_params_pf(struct ixgbe_hw *hw) hw->mac.type != ixgbe_mac_X540) return; - mbx->timeout = 0; - mbx->usec_delay = 0; + /* Initialize common mailbox settings */ + mbx->timeout = IXGBE_VF_MBX_INIT_TIMEOUT; + mbx->usec_delay = IXGBE_VF_MBX_INIT_DELAY; + mbx->size = IXGBE_VFMAILBOX_SIZE; + + /* Initialize counters with zeroes */ + mbx->stats.msgs_tx = 0; + mbx->stats.msgs_rx = 0; + mbx->stats.reqs = 0; + mbx->stats.acks = 0; + mbx->stats.rsts = 0; + + /* No matter of VF number, we initialize params for all 64 VFs. */ + /* TODO: 1. Add a define for max VF and refactor SHARED to get rid + * of magic number for that (63 or 64 depending on use case.) + * 2. rewrite the code to dynamically allocate mbx->ops[vf_id] for + * certain number of VFs instead of default maximum value of 64 (0..63) + */ + for (i = 0; i < 64; i++) + ixgbe_init_mbx_params_pf_id(hw, i); +} + +/** + * ixgbe_upgrade_mbx_params_pf - Upgrade initial values for pf mailbox + * @hw: pointer to the HW structure + * @vf_id: the VF index + * + * Initializes the hw->mbx struct to new function set for improved + * stability and handling of messages. + */ +void ixgbe_upgrade_mbx_params_pf(struct ixgbe_hw *hw, u16 vf_id) +{ + struct ixgbe_mbx_info *mbx = &hw->mbx; + + /* Ensure we are not calling this function from VF */ + if (hw->mac.type != ixgbe_mac_82599EB && + hw->mac.type != ixgbe_mac_X550 && + hw->mac.type != ixgbe_mac_X550EM_x && + hw->mac.type != ixgbe_mac_X550EM_a && + hw->mac.type != ixgbe_mac_X540) + return; + mbx->timeout = IXGBE_VF_MBX_INIT_TIMEOUT; + mbx->usec_delay = IXGBE_VF_MBX_INIT_DELAY; mbx->size = IXGBE_VFMAILBOX_SIZE; - mbx->ops.read = ixgbe_read_mbx_pf; - mbx->ops.write = ixgbe_write_mbx_pf; - mbx->ops.read_posted = ixgbe_read_posted_mbx; - mbx->ops.write_posted = ixgbe_write_posted_mbx; - mbx->ops.check_for_msg = ixgbe_check_for_msg_pf; - mbx->ops.check_for_ack = ixgbe_check_for_ack_pf; - mbx->ops.check_for_rst = ixgbe_check_for_rst_pf; + mbx->ops[vf_id].release = ixgbe_release_mbx_lock_pf; + mbx->ops[vf_id].read = ixgbe_read_mbx_pf; + mbx->ops[vf_id].write = ixgbe_write_mbx_pf; + mbx->ops[vf_id].check_for_msg = ixgbe_check_for_msg_pf; + mbx->ops[vf_id].check_for_ack = ixgbe_check_for_ack_pf; + mbx->ops[vf_id].check_for_rst = ixgbe_check_for_rst_pf; mbx->stats.msgs_tx = 0; mbx->stats.msgs_rx = 0; diff --git a/drivers/net/ixgbe/base/ixgbe_mbx.h b/drivers/net/ixgbe/base/ixgbe_mbx.h index f7861e6bde..fadfccaabb 100644 --- a/drivers/net/ixgbe/base/ixgbe_mbx.h +++ b/drivers/net/ixgbe/base/ixgbe_mbx.h @@ -7,8 +7,41 @@ #include "ixgbe_type.h" +struct ixgbe_mbx_operations { + void (*init_params)(struct ixgbe_hw *hw); + void (*release)(struct ixgbe_hw *hw, u16 mbx_id); + s32 (*read)(struct ixgbe_hw *hw, u32 *msg, u16 size, u16 mbx_id); + s32 (*write)(struct ixgbe_hw *hw, u32 *msg, u16 size, u16 mbx_id); + s32 (*check_for_msg)(struct ixgbe_hw *hw, u16 vf_number); + s32 (*check_for_ack)(struct ixgbe_hw *hw, u16 vf_number); + s32 (*check_for_rst)(struct ixgbe_hw *hw, u16 vf_number); + s32 (*clear)(struct ixgbe_hw *hw, u16 vf_number); +}; + +struct ixgbe_mbx_stats { + u32 msgs_tx; + u32 msgs_rx; + + u32 acks; + u32 reqs; + u32 rsts; +}; + +struct ixgbe_mbx_info { + /* + * PF: One set of operations for each VF to handle various API versions + * at the same time + * VF: Only the very first (0) set should be used + */ + struct ixgbe_mbx_operations ops[64]; + struct ixgbe_mbx_stats stats; + u32 timeout; + u32 usec_delay; + u32 vf_mailbox; + u16 size; +}; + #define IXGBE_VFMAILBOX_SIZE 16 /* 16 32 bit words - 64 bytes */ -#define IXGBE_ERR_MBX -100 #define IXGBE_VFMAILBOX 0x002FC #define IXGBE_VFMBMEM 0x00200 @@ -65,6 +98,9 @@ enum ixgbe_pfvf_api_rev { ixgbe_mbox_api_11, /* API version 1.1, linux/freebsd VF driver */ ixgbe_mbox_api_12, /* API version 1.2, linux/freebsd VF driver */ ixgbe_mbox_api_13, /* API version 1.3, linux/freebsd VF driver */ + /* API 1.4 is being used in the upstream for IPsec */ + ixgbe_mbox_api_14, /* API version 1.4, linux/freebsd VF driver */ + ixgbe_mbox_api_15, /* API version 1.5, linux/freebsd VF driver */ /* This value should always be last */ ixgbe_mbox_api_unknown, /* indicates that API version is not known */ }; @@ -125,15 +161,16 @@ enum ixgbevf_xcast_modes { #define IXGBE_VF_MBX_INIT_TIMEOUT 2000 /* number of retries on mailbox */ #define IXGBE_VF_MBX_INIT_DELAY 500 /* microseconds between retries */ -s32 ixgbe_read_mbx(struct ixgbe_hw *, u32 *, u16, u16); -s32 ixgbe_write_mbx(struct ixgbe_hw *, u32 *, u16, u16); -s32 ixgbe_read_posted_mbx(struct ixgbe_hw *, u32 *, u16, u16); -s32 ixgbe_write_posted_mbx(struct ixgbe_hw *, u32 *, u16, u16); -s32 ixgbe_check_for_msg(struct ixgbe_hw *, u16); -s32 ixgbe_check_for_ack(struct ixgbe_hw *, u16); -s32 ixgbe_check_for_rst(struct ixgbe_hw *, u16); -void ixgbe_init_mbx_ops_generic(struct ixgbe_hw *hw); -void ixgbe_init_mbx_params_vf(struct ixgbe_hw *); -void ixgbe_init_mbx_params_pf(struct ixgbe_hw *); +s32 ixgbe_read_mbx(struct ixgbe_hw *hw, u32 *msg, u16 size, u16 mbx_id); +s32 ixgbe_poll_mbx(struct ixgbe_hw *hw, u32 *msg, u16 size, u16 mbx_id); +s32 ixgbe_write_mbx(struct ixgbe_hw *hw, u32 *msg, u16 size, u16 mbx_id); +s32 ixgbe_check_for_msg(struct ixgbe_hw *hw, u16 mbx_id); +s32 ixgbe_check_for_ack(struct ixgbe_hw *hw, u16 mbx_id); +s32 ixgbe_check_for_rst(struct ixgbe_hw *hw, u16 mbx_id); +void ixgbe_init_mbx_params_vf(struct ixgbe_hw *hw); +void ixgbe_upgrade_mbx_params_vf(struct ixgbe_hw *hw); +void ixgbe_init_mbx_params_pf(struct ixgbe_hw *hw); +void ixgbe_init_mbx_params_pf_id(struct ixgbe_hw *hw, u16 vf_id); +void ixgbe_upgrade_mbx_params_pf(struct ixgbe_hw *hw, u16 vf_id); #endif /* _IXGBE_MBX_H_ */ diff --git a/drivers/net/ixgbe/base/ixgbe_type.h b/drivers/net/ixgbe/base/ixgbe_type.h index 5036b2a907..10164e274c 100644 --- a/drivers/net/ixgbe/base/ixgbe_type.h +++ b/drivers/net/ixgbe/base/ixgbe_type.h @@ -4141,35 +4141,6 @@ struct ixgbe_phy_info { #include "ixgbe_mbx.h" -struct ixgbe_mbx_operations { - void (*init_params)(struct ixgbe_hw *hw); - s32 (*read)(struct ixgbe_hw *, u32 *, u16, u16); - s32 (*write)(struct ixgbe_hw *, u32 *, u16, u16); - s32 (*read_posted)(struct ixgbe_hw *, u32 *, u16, u16); - s32 (*write_posted)(struct ixgbe_hw *, u32 *, u16, u16); - s32 (*check_for_msg)(struct ixgbe_hw *, u16); - s32 (*check_for_ack)(struct ixgbe_hw *, u16); - s32 (*check_for_rst)(struct ixgbe_hw *, u16); -}; - -struct ixgbe_mbx_stats { - u32 msgs_tx; - u32 msgs_rx; - - u32 acks; - u32 reqs; - u32 rsts; -}; - -struct ixgbe_mbx_info { - struct ixgbe_mbx_operations ops; - struct ixgbe_mbx_stats stats; - u32 timeout; - u32 usec_delay; - u32 v2p_mailbox; - u16 size; -}; - struct ixgbe_hw { u8 IOMEM *hw_addr; void *back; @@ -4239,6 +4210,9 @@ struct ixgbe_hw { #define IXGBE_ERR_FDIR_CMD_INCOMPLETE -38 #define IXGBE_ERR_FW_RESP_INVALID -39 #define IXGBE_ERR_TOKEN_RETRY -40 +#define IXGBE_ERR_MBX -41 +#define IXGBE_ERR_MBX_NOMSG -42 +#define IXGBE_ERR_TIMEOUT -43 #define IXGBE_NOT_IMPLEMENTED 0x7FFFFFFF diff --git a/drivers/net/ixgbe/base/ixgbe_vf.c b/drivers/net/ixgbe/base/ixgbe_vf.c index d390b0d5e0..603c24b653 100644 --- a/drivers/net/ixgbe/base/ixgbe_vf.c +++ b/drivers/net/ixgbe/base/ixgbe_vf.c @@ -21,6 +21,8 @@ **/ s32 ixgbe_init_ops_vf(struct ixgbe_hw *hw) { + u16 i; + /* MAC */ hw->mac.ops.init_hw = ixgbe_init_hw_vf; hw->mac.ops.reset_hw = ixgbe_reset_hw_vf; @@ -53,7 +55,8 @@ s32 ixgbe_init_ops_vf(struct ixgbe_hw *hw) hw->mac.max_tx_queues = 1; hw->mac.max_rx_queues = 1; - hw->mbx.ops.init_params = ixgbe_init_mbx_params_vf; + for (i = 0; i < 64; i++) + hw->mbx.ops[i].init_params = ixgbe_init_mbx_params_vf; return IXGBE_SUCCESS; } @@ -156,6 +159,7 @@ s32 ixgbe_reset_hw_vf(struct ixgbe_hw *hw) /* reset the api version */ hw->api_version = ixgbe_mbox_api_10; + ixgbe_init_mbx_params_vf(hw); DEBUGOUT("Issuing a function level reset to MAC\n"); @@ -165,7 +169,7 @@ s32 ixgbe_reset_hw_vf(struct ixgbe_hw *hw) msec_delay(50); /* we cannot reset while the RSTI / RSTD bits are asserted */ - while (!mbx->ops.check_for_rst(hw, 0) && timeout) { + while (!mbx->ops[0].check_for_rst(hw, 0) && timeout) { timeout--; usec_delay(5); } @@ -180,7 +184,7 @@ s32 ixgbe_reset_hw_vf(struct ixgbe_hw *hw) mbx->timeout = IXGBE_VF_MBX_INIT_TIMEOUT; msgbuf[0] = IXGBE_VF_RESET; - mbx->ops.write_posted(hw, msgbuf, 1, 0); + ixgbe_write_mbx(hw, msgbuf, 1, 0); msec_delay(10); @@ -189,8 +193,8 @@ s32 ixgbe_reset_hw_vf(struct ixgbe_hw *hw) * also set up the mc_filter_type which is piggy backed * on the mac address in word 3 */ - ret_val = mbx->ops.read_posted(hw, msgbuf, - IXGBE_VF_PERMADDR_MSG_LEN, 0); + ret_val = ixgbe_poll_mbx(hw, msgbuf, + IXGBE_VF_PERMADDR_MSG_LEN, 0); if (ret_val) return ret_val; @@ -295,13 +299,12 @@ STATIC s32 ixgbe_mta_vector(struct ixgbe_hw *hw, u8 *mc_addr) STATIC s32 ixgbevf_write_msg_read_ack(struct ixgbe_hw *hw, u32 *msg, u32 *retmsg, u16 size) { - struct ixgbe_mbx_info *mbx = &hw->mbx; - s32 retval = mbx->ops.write_posted(hw, msg, size, 0); + s32 retval = ixgbe_write_mbx(hw, msg, size, 0); if (retval) return retval; - return mbx->ops.read_posted(hw, retmsg, size, 0); + return ixgbe_poll_mbx(hw, retmsg, size, 0); } /** @@ -351,7 +354,6 @@ s32 ixgbe_update_mc_addr_list_vf(struct ixgbe_hw *hw, u8 *mc_addr_list, u32 mc_addr_count, ixgbe_mc_addr_itr next, bool clear) { - struct ixgbe_mbx_info *mbx = &hw->mbx; u32 msgbuf[IXGBE_VFMAILBOX_SIZE]; u16 *vector_list = (u16 *)&msgbuf[1]; u32 vector; @@ -383,7 +385,7 @@ s32 ixgbe_update_mc_addr_list_vf(struct ixgbe_hw *hw, u8 *mc_addr_list, vector_list[i] = (u16)vector; } - return mbx->ops.write_posted(hw, msgbuf, IXGBE_VFMAILBOX_SIZE, 0); + return ixgbevf_write_msg_read_ack(hw, msgbuf, msgbuf, IXGBE_VFMAILBOX_SIZE); } /** @@ -405,6 +407,7 @@ s32 ixgbevf_update_xcast_mode(struct ixgbe_hw *hw, int xcast_mode) return IXGBE_ERR_FEATURE_NOT_SUPPORTED; /* Fall through */ case ixgbe_mbox_api_13: + case ixgbe_mbox_api_15: break; default: return IXGBE_ERR_FEATURE_NOT_SUPPORTED; @@ -551,12 +554,13 @@ s32 ixgbe_check_mac_link_vf(struct ixgbe_hw *hw, ixgbe_link_speed *speed, struct ixgbe_mbx_info *mbx = &hw->mbx; struct ixgbe_mac_info *mac = &hw->mac; s32 ret_val = IXGBE_SUCCESS; - u32 links_reg; u32 in_msg = 0; + u32 links_reg; + UNREFERENCED_1PARAMETER(autoneg_wait_to_complete); /* If we were hit with a reset drop the link */ - if (!mbx->ops.check_for_rst(hw, 0) || !mbx->timeout) + if (!mbx->ops[0].check_for_rst(hw, 0) || !mbx->timeout) mac->get_link_status = true; if (!mac->get_link_status) @@ -613,21 +617,22 @@ s32 ixgbe_check_mac_link_vf(struct ixgbe_hw *hw, ixgbe_link_speed *speed, /* if the read failed it could just be a mailbox collision, best wait * until we are called again and don't report an error */ - if (mbx->ops.read(hw, &in_msg, 1, 0)) + if (ixgbe_read_mbx(hw, &in_msg, 1, 0)) { + if (hw->api_version >= ixgbe_mbox_api_15) + mac->get_link_status = false; goto out; + } if (!(in_msg & IXGBE_VT_MSGTYPE_CTS)) { - /* msg is not CTS and is FAILURE, - * we must have lost CTS status. - */ + /* msg is not CTS and is FAILURE we must have lost CTS status */ if (in_msg & IXGBE_VT_MSGTYPE_FAILURE) - ret_val = -1; + ret_val = IXGBE_ERR_MBX; goto out; } /* the pf is talking, if we timed out in the past we reinit */ if (!mbx->timeout) { - ret_val = -1; + ret_val = IXGBE_ERR_TIMEOUT; goto out; } @@ -706,6 +711,7 @@ int ixgbevf_get_queues(struct ixgbe_hw *hw, unsigned int *num_tcs, case ixgbe_mbox_api_11: case ixgbe_mbox_api_12: case ixgbe_mbox_api_13: + case ixgbe_mbox_api_15: break; default: return 0; diff --git a/drivers/net/ixgbe/base/ixgbe_x540.c b/drivers/net/ixgbe/base/ixgbe_x540.c index 8efde4645a..60649f1de8 100644 --- a/drivers/net/ixgbe/base/ixgbe_x540.c +++ b/drivers/net/ixgbe/base/ixgbe_x540.c @@ -32,6 +32,7 @@ s32 ixgbe_init_ops_X540(struct ixgbe_hw *hw) struct ixgbe_phy_info *phy = &hw->phy; struct ixgbe_eeprom_info *eeprom = &hw->eeprom; s32 ret_val; + u16 i; DEBUGFUNC("ixgbe_init_ops_X540"); @@ -111,7 +112,8 @@ s32 ixgbe_init_ops_X540(struct ixgbe_hw *hw) mac->arc_subsystem_valid = !!(IXGBE_READ_REG(hw, IXGBE_FWSM_BY_MAC(hw)) & IXGBE_FWSM_MODE_MASK); - hw->mbx.ops.init_params = ixgbe_init_mbx_params_pf; + for (i = 0; i < 64; i++) + hw->mbx.ops[i].init_params = ixgbe_init_mbx_params_pf; /* LEDs */ mac->ops.blink_led_start = ixgbe_blink_led_start_X540; diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c index 8c60351636..f2a397a451 100644 --- a/drivers/net/ixgbe/ixgbe_ethdev.c +++ b/drivers/net/ixgbe/ixgbe_ethdev.c @@ -1663,7 +1663,7 @@ eth_ixgbevf_dev_init(struct rte_eth_dev *eth_dev) } /* init_mailbox_params */ - hw->mbx.ops.init_params(hw); + hw->mbx.ops[0].init_params(hw); /* Reset the hw statistics */ ixgbevf_dev_stats_reset(eth_dev); @@ -4099,7 +4099,7 @@ ixgbevf_check_link(struct ixgbe_hw *hw, ixgbe_link_speed *speed, int ret_val = 0; /* If we were hit with a reset drop the link */ - if (!mbx->ops.check_for_rst(hw, 0) || !mbx->timeout) + if (!mbx->ops[0].check_for_rst(hw, 0) || !mbx->timeout) mac->get_link_status = true; if (!mac->get_link_status) @@ -4165,7 +4165,7 @@ ixgbevf_check_link(struct ixgbe_hw *hw, ixgbe_link_speed *speed, /* if the read failed it could just be a mailbox collision, best wait * until we are called again and don't report an error */ - if (mbx->ops.read(hw, &in_msg, 1, 0)) + if (mbx->ops[0].read(hw, &in_msg, 1, 0)) goto out; if (!(in_msg & IXGBE_VT_MSGTYPE_CTS)) { diff --git a/drivers/net/ixgbe/ixgbe_pf.c b/drivers/net/ixgbe/ixgbe_pf.c index 71104247da..91ba395ac3 100644 --- a/drivers/net/ixgbe/ixgbe_pf.c +++ b/drivers/net/ixgbe/ixgbe_pf.c @@ -77,6 +77,7 @@ int ixgbe_pf_host_init(struct rte_eth_dev *eth_dev) uint16_t vf_num; uint8_t nb_queue; int ret = 0; + size_t i; PMD_INIT_FUNC_TRACE(); @@ -122,7 +123,8 @@ int ixgbe_pf_host_init(struct rte_eth_dev *eth_dev) ixgbe_vf_perm_addr_gen(eth_dev, vf_num); /* init_mailbox_params */ - hw->mbx.ops.init_params(hw); + for (i = 0; i < vf_num; i++) + hw->mbx.ops[i].init_params(hw); /* set mb interrupt mask */ ixgbe_mb_intr_setup(eth_dev); From patchwork Wed Apr 24 13:21:41 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139652 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 20DA443EAD; Wed, 24 Apr 2024 15:23:10 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id E6016434AE; Wed, 24 Apr 2024 15:22:32 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id EEBF6434A7 for ; Wed, 24 Apr 2024 15:22:30 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964951; x=1745500951; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=KNwVgMK74daK+8GeOdSPC/0KjPrYJG4OqNRSZpdF94c=; b=HrrXLzDUuK0KGoRM+wlkrz0zhAou+/KEgU8hNg+zV/MzpqEs+bgwYnS1 RsYnlHZQ8q6AMu3OiM3JGYqRVfAG2Qgx4EOj/vp8q4pW8y4mMCxuGQ4fn DF9k02OwYmXEPLqBf6ZxJA64W84yYmEf725fhIaoE276ME+z1SGA3sJb7 kI2Vb9S6fbkm59sEEOFGGEbaiE/KSOD8BHxCQPJkfywgXm5igaTaOWhso 2fLy3flQZnKkKQSGsb90/GAwAizAhQwNG6m1nGj/gCvx/93ibNAk3enqS bujRwpLPVHFM+dTelqYCLuNY57NLoHAb/a7KxOTpTnVnDnPRod6gC7SFw w==; X-CSE-ConnectionGUID: /T9TdnZRTyyHMHyy/Rv9GA== X-CSE-MsgGUID: NmVUJnrzRVyUSxxJntHdag== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289260" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289260" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:31 -0700 X-CSE-ConnectionGUID: 1e3LJsC0SuuDWi1WFpgtMQ== X-CSE-MsgGUID: 7BGo6OuTSwmjqNfQ3RAAXw== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749361" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:28 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Radoslaw Tyl , vladimir.medvedkin@intel.com, bruce.richardson@intel.com, Skajewski@dpdk.org, PiotrX , Mrozowicz@dpdk.org, SlawomirX , Michael@dpdk.org, Alice Subject: [PATCH v1 07/22] net/ixgbe/base: increase DCB BW calculation for MTU from 4088 to 9128 Date: Wed, 24 Apr 2024 14:21:41 +0100 Message-ID: X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Radoslaw Tyl Change max credit and credit refill to a maximum possible value. Too small values cause the incorrect calculation of the bandwidth limits to each traffic class for frames larger than 4088 bytes. Signed-off-by: Radoslaw Tyl Reviewed-by: Skajewski, PiotrX Tested-by: Skajewski, PiotrX Reviewed-by: Mrozowicz, SlawomirX Reviewed-by: Michael, Alice --- drivers/net/ixgbe/base/ixgbe_dcb.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ixgbe/base/ixgbe_dcb.h b/drivers/net/ixgbe/base/ixgbe_dcb.h index c2a1013ac0..110ad8e550 100644 --- a/drivers/net/ixgbe/base/ixgbe_dcb.h +++ b/drivers/net/ixgbe/base/ixgbe_dcb.h @@ -10,9 +10,9 @@ /* DCB defines */ /* DCB credit calculation defines */ #define IXGBE_DCB_CREDIT_QUANTUM 64 -#define IXGBE_DCB_MAX_CREDIT_REFILL 200 /* 200 * 64B = 12800B */ +#define IXGBE_DCB_MAX_CREDIT_REFILL 511 /* 0x1FF * 64B = 32704B */ #define IXGBE_DCB_MAX_TSO_SIZE (32 * 1024) /* Max TSO pkt size in DCB*/ -#define IXGBE_DCB_MAX_CREDIT (2 * IXGBE_DCB_MAX_CREDIT_REFILL) +#define IXGBE_DCB_MAX_CREDIT 4095 /* Maximum credit supported: 256KB * 1024 / 64B */ /* 513 for 32KB TSO packet */ #define IXGBE_DCB_MIN_TSO_CREDIT \ From patchwork Wed Apr 24 13:21:42 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139653 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 09C5243EAD; Wed, 24 Apr 2024 15:23:17 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 3A343434AA; Wed, 24 Apr 2024 15:22:35 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id A18EC434A6 for ; Wed, 24 Apr 2024 15:22:33 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964954; x=1745500954; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=l9+PUKAsDi2uqwCxaMcNclP+NWYYilGNGmA3IVK50O0=; b=X5Wz5UFaLALxFTGvSvag9wQ4bPKgogA94zKHh2jgxWB0pgQAL9OJjo3x z33nnqHJaHMpKpTEpD3Jm+mlEf25R0vIdEROiqxTVPA7yx/Q4vILFM9/B dxh9Kb3Nf2im+xXN3uYE1kAMC6BFiSmx6qY9gJW2pk6q/6juzy1AifbNP GFCZftcUHo1CJkUDvYN6RI+uYsQIH3/SM2DSM8GRgn4OnCSUmwRgs/R+y 72WloWA7w1ySkBZmv8IknIiMmWd7A6wmAJxEAfHoGUh7rmCL1wZGkpdQR X9dQSt0elJpnJlv+mSt/Acg+UcsE8xayzM2XXknp9fMtmcSvRd5usi9jZ A==; X-CSE-ConnectionGUID: dN5nfcHgRxe0gE2K+WS+Hg== X-CSE-MsgGUID: 0DB4yhKHSwOXK35thEozpQ== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289261" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289261" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:33 -0700 X-CSE-ConnectionGUID: 64nJ7+rzRNWFkKqgU1zsvQ== X-CSE-MsgGUID: aXCbgSAAQCGItgsW2J7xJw== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749375" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:31 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Slawomir Mrozowicz , vladimir.medvedkin@intel.com, bruce.richardson@intel.com, Tyl@dpdk.org, RadoslawX , Skajewski@dpdk.org, PiotrX , nosbuild , Michael@dpdk.org, Alice Subject: [PATCH v1 08/22] net/ixgbe/base: fix crash while loading driver Date: Wed, 24 Apr 2024 14:21:42 +0100 Message-ID: <577c1cb1250710da2839079d85a8e9e83fa34c7e.1713964708.git.anatoly.burakov@intel.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Slawomir Mrozowicz Issue observed: system crash While loading the driver when debug macros were defined. The solution removes the debug macro call from ixgbe_set_mac_type() function which causes the crash. So far, the ixgbe_set_mac_type() function has called in the ixgbe driver ixgbe_probe() function before the struct ixgbe_hw is properly defined. In this case, the field back of the structure that must contain pointer to an adapter has not yet been defined and should not be used to obtain driver information which is needed in debug macros. Moreover bearing in mind the recommendations to not print out excessive amount of information it was decided to remove the debug information. Signed-off-by: Slawomir Mrozowicz Reviewed-by: Tyl, RadoslawX Reviewed-by: Skajewski, PiotrX Tested-by: nosbuild Reviewed-by: Michael, Alice --- Notes: This fix seems to be kernel-specific, so Cc:stable was intentionally omitted. drivers/net/ixgbe/base/ixgbe_api.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/net/ixgbe/base/ixgbe_api.c b/drivers/net/ixgbe/base/ixgbe_api.c index b322c181b4..89e1650d63 100644 --- a/drivers/net/ixgbe/base/ixgbe_api.c +++ b/drivers/net/ixgbe/base/ixgbe_api.c @@ -216,8 +216,6 @@ s32 ixgbe_set_mac_type(struct ixgbe_hw *hw) break; } - DEBUGOUT2("ixgbe_set_mac_type found mac: %d, returns: %d\n", - hw->mac.type, ret_val); return ret_val; } From patchwork Wed Apr 24 13:21:43 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139654 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id BBBC843EAD; Wed, 24 Apr 2024 15:23:24 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 82D71434B1; Wed, 24 Apr 2024 15:22:37 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id 406AC434A7 for ; Wed, 24 Apr 2024 15:22:36 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964957; x=1745500957; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=z2qOzHIS+Ph3SY6OPRWvvgBpdN3jZhr0V5K0tQEAx5o=; b=nRPYKfZg+qg2HXbgQYUTV5EZ09qicKIxCYOjOUzcIBwywPQVnl4gSe9c iu7U90PfAS185rYZ4gx/mRN7TBQO/gvQS78ZU2/4iO7qm5fSYm/cvBRIT VKqSdxUefaaFONEUDnLuuYwWtnyTNZKxI2nfOW6VRvzh1s+XZaAAd5k4w EY9REezAQdA65OrpjZ0+O0/uypqOiWWTSrRnNJiVFC/NG24npjRMJCMeV W/DKmsU+vanLWjXOZMWBmOjk/kAt0NyCvS96Rt8ZpNilWQkyHE6w3ekMM aRNt1dOa8C7DUZVZQB5y+sj4Va0fKSkdjELSwNtRWZ7imC6aJFStU4kTg g==; X-CSE-ConnectionGUID: SE84P/65Q8GRCC2f7Dg2nA== X-CSE-MsgGUID: k3NfJ/lWRhysfx3+DLgRfw== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289264" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289264" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:36 -0700 X-CSE-ConnectionGUID: BYeKa6I1SE+8duZaT9rLEQ== X-CSE-MsgGUID: o1gV0lmwR9KSC+mn4DEfkA== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749388" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:33 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Radoslaw Tyl , vladimir.medvedkin@intel.com, bruce.richardson@intel.com, Skajewski@dpdk.org, PiotrX , Mrozowicz@dpdk.org, SlawomirX , Michael@dpdk.org, Alice Subject: [PATCH v1 09/22] net/ixgbe/base: improve function comments Date: Wed, 24 Apr 2024 14:21:43 +0100 Message-ID: <6b2eaa1a91a7cfe3bb2baa74e87c25a6ff3ba31a.1713964708.git.anatoly.burakov@intel.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Radoslaw Tyl Some function comments have mismatches between actual function names and function name in comments, which causes warnings with kernel-doc. Fix comments to match function names. Signed-off-by: Radoslaw Tyl Reviewed-by: Skajewski, PiotrX Reviewed-by: Mrozowicz, SlawomirX Reviewed-by: Michael, Alice --- drivers/net/ixgbe/base/ixgbe_82599.c | 2 +- drivers/net/ixgbe/base/ixgbe_api.c | 2 +- drivers/net/ixgbe/base/ixgbe_common.c | 11 ++++++----- drivers/net/ixgbe/base/ixgbe_dcb.c | 2 +- drivers/net/ixgbe/base/ixgbe_hv_vf.c | 10 +++++----- drivers/net/ixgbe/base/ixgbe_phy.c | 2 +- drivers/net/ixgbe/base/ixgbe_x550.c | 12 ++++++------ 7 files changed, 21 insertions(+), 20 deletions(-) diff --git a/drivers/net/ixgbe/base/ixgbe_82599.c b/drivers/net/ixgbe/base/ixgbe_82599.c index 7d2938b253..73c7766ecf 100644 --- a/drivers/net/ixgbe/base/ixgbe_82599.c +++ b/drivers/net/ixgbe/base/ixgbe_82599.c @@ -1497,7 +1497,7 @@ u32 ixgbe_atr_compute_sig_hash_82599(union ixgbe_atr_hash_dword input, } /** - * ixgbe_atr_add_signature_filter_82599 - Adds a signature hash filter + * ixgbe_fdir_add_signature_filter_82599 - Adds a signature hash filter * @hw: pointer to hardware structure * @input: unique input dword * @common: compressed common input dword diff --git a/drivers/net/ixgbe/base/ixgbe_api.c b/drivers/net/ixgbe/base/ixgbe_api.c index 89e1650d63..d2b74cdffc 100644 --- a/drivers/net/ixgbe/base/ixgbe_api.c +++ b/drivers/net/ixgbe/base/ixgbe_api.c @@ -870,7 +870,7 @@ s32 ixgbe_validate_eeprom_checksum(struct ixgbe_hw *hw, u16 *checksum_val) } /** - * ixgbe_eeprom_update_checksum - Updates the EEPROM checksum + * ixgbe_update_eeprom_checksum - Updates the EEPROM checksum * @hw: pointer to hardware structure **/ s32 ixgbe_update_eeprom_checksum(struct ixgbe_hw *hw) diff --git a/drivers/net/ixgbe/base/ixgbe_common.c b/drivers/net/ixgbe/base/ixgbe_common.c index aa843bd5c4..dd8310748b 100644 --- a/drivers/net/ixgbe/base/ixgbe_common.c +++ b/drivers/net/ixgbe/base/ixgbe_common.c @@ -3827,14 +3827,15 @@ s32 ixgbe_set_vmdq_generic(struct ixgbe_hw *hw, u32 rar, u32 vmdq) } /** + * ixgbe_set_vmdq_san_mac_generic - Associate default VMDq pool index with + * a rx address + * @hw: pointer to hardware struct + * @vmdq: VMDq pool index + * * This function should only be involved in the IOV mode. * In IOV mode, Default pool is next pool after the number of * VFs advertized and not 0. * MPSAR table needs to be updated for SAN_MAC RAR [hw->mac.san_mac_rar_index] - * - * ixgbe_set_vmdq_san_mac - Associate default VMDq pool index with a rx address - * @hw: pointer to hardware struct - * @vmdq: VMDq pool index **/ s32 ixgbe_set_vmdq_san_mac_generic(struct ixgbe_hw *hw, u32 vmdq) { @@ -4825,7 +4826,7 @@ STATIC const u8 ixgbe_emc_therm_limit[4] = { }; /** - * ixgbe_get_thermal_sensor_data - Gathers thermal sensor data + * ixgbe_get_thermal_sensor_data_generic - Gathers thermal sensor data * @hw: pointer to hardware structure * * Returns the thermal sensor data structure diff --git a/drivers/net/ixgbe/base/ixgbe_dcb.c b/drivers/net/ixgbe/base/ixgbe_dcb.c index 41469c10db..55402b230f 100644 --- a/drivers/net/ixgbe/base/ixgbe_dcb.c +++ b/drivers/net/ixgbe/base/ixgbe_dcb.c @@ -263,7 +263,7 @@ void ixgbe_dcb_unpack_map_cee(struct ixgbe_dcb_config *cfg, int direction, } /** - * ixgbe_dcb_config - Struct containing DCB settings. + * ixgbe_dcb_check_config_cee - Struct containing DCB settings. * @dcb_config: Pointer to DCB config structure * * This function checks DCB rules for DCB settings. diff --git a/drivers/net/ixgbe/base/ixgbe_hv_vf.c b/drivers/net/ixgbe/base/ixgbe_hv_vf.c index 34add03e8b..1279659926 100644 --- a/drivers/net/ixgbe/base/ixgbe_hv_vf.c +++ b/drivers/net/ixgbe/base/ixgbe_hv_vf.c @@ -7,7 +7,7 @@ #include "ixgbe_osdep.h" /** - * Hyper-V variant - just a stub. + * ixgbevf_hv_update_mc_addr_list_vf - Hyper-V variant - just a stub. * @hw: unused * @mc_addr_list: unused * @mc_addr_count: unused @@ -24,7 +24,7 @@ static s32 ixgbevf_hv_update_mc_addr_list_vf(struct ixgbe_hw *hw, u8 *mc_addr_li } /** - * Hyper-V variant - just a stub. + * ixgbevf_hv_update_xcast_mode - Hyper-V variant - just a stub. * @hw: unused * @xcast_mode: unused */ @@ -36,7 +36,7 @@ static s32 ixgbevf_hv_update_xcast_mode(struct ixgbe_hw *hw, int xcast_mode) } /** - * Hyper-V variant - just a stub. + * ixgbevf_hv_set_vfta_vf - Hyper-V variant - just a stub. * @hw: unused * @vlan: unused * @vind: unused @@ -79,12 +79,12 @@ static s32 ixgbevf_hv_set_rar_vf(struct ixgbe_hw *hw, u32 index, u8 *addr, u32 v } /** - * Hyper-V variant; there is no mailbox communication. + * ixgbevf_hv_check_mac_link_vf - Hyper-V variant; there is no mailbox + * communication. * @hw: pointer to hardware structure * @speed: pointer to link speed * @link_up: true is link is up, false otherwise * @autoneg_wait_to_complete: unused - * */ static s32 ixgbevf_hv_check_mac_link_vf(struct ixgbe_hw *hw, ixgbe_link_speed *speed, diff --git a/drivers/net/ixgbe/base/ixgbe_phy.c b/drivers/net/ixgbe/base/ixgbe_phy.c index 56267bb00d..8a7712c8f2 100644 --- a/drivers/net/ixgbe/base/ixgbe_phy.c +++ b/drivers/net/ixgbe/base/ixgbe_phy.c @@ -557,7 +557,7 @@ void ixgbe_restart_auto_neg(struct ixgbe_hw *hw) } /** - * ixgbe_read_phy_mdi - Reads a value from a specified PHY register without + * ixgbe_read_phy_reg_mdi - Reads a value from a specified PHY register without * the SWFW lock * @hw: pointer to hardware structure * @reg_addr: 32 bit address of PHY register to read diff --git a/drivers/net/ixgbe/base/ixgbe_x550.c b/drivers/net/ixgbe/base/ixgbe_x550.c index 8810d1658e..844c06cc08 100644 --- a/drivers/net/ixgbe/base/ixgbe_x550.c +++ b/drivers/net/ixgbe/base/ixgbe_x550.c @@ -715,7 +715,7 @@ static s32 ixgbe_setup_fw_link(struct ixgbe_hw *hw) } /** - * ixgbe_fc_autoneg_fw _ Set up flow control for FW-controlled PHYs + * ixgbe_fc_autoneg_fw - Set up flow control for FW-controlled PHYs * @hw: pointer to hardware structure * * Called at init time to set up flow control. @@ -1843,7 +1843,7 @@ void ixgbe_init_mac_link_ops_X550em(struct ixgbe_hw *hw) } /** - * ixgbe_get_link_capabilities_x550em - Determines link capabilities + * ixgbe_get_link_capabilities_X550em - Determines link capabilities * @hw: pointer to hardware structure * @speed: pointer to link speed * @autoneg: true when autoneg or autotry is enabled @@ -3205,7 +3205,7 @@ s32 ixgbe_read_ee_hostif_buffer_X550(struct ixgbe_hw *hw, } /** - * ixgbe_write_ee_hostif_X550 - Write EEPROM word using hostif + * ixgbe_write_ee_hostif_data_X550 - Write EEPROM word using hostif * @hw: pointer to hardware structure * @offset: offset of word in the EEPROM to write * @data: word write to the EEPROM @@ -3673,7 +3673,7 @@ u64 ixgbe_get_supported_physical_layer_X550em(struct ixgbe_hw *hw) } /** - * ixgbe_get_bus_info_x550em - Set PCI bus info + * ixgbe_get_bus_info_X550em - Set PCI bus info * @hw: pointer to hardware structure * * Sets bus link width and speed to unknown because X550em is @@ -3738,7 +3738,7 @@ void ixgbe_disable_rx_x550(struct ixgbe_hw *hw) } /** - * ixgbe_enter_lplu_x550em - Transition to low power states + * ixgbe_enter_lplu_t_x550em - Transition to low power states * @hw: pointer to hardware structure * * Configures Low Power Link Up on transition to low power states @@ -3846,7 +3846,7 @@ s32 ixgbe_enter_lplu_t_x550em(struct ixgbe_hw *hw) } /** - * ixgbe_get_lcd_x550em - Determine lowest common denominator + * ixgbe_get_lcd_t_x550em - Determine lowest common denominator * @hw: pointer to hardware structure * @lcd_speed: pointer to lowest common link speed * From patchwork Wed Apr 24 13:21:44 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139655 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id CE68E43EAD; Wed, 24 Apr 2024 15:23:30 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id CD718434B7; Wed, 24 Apr 2024 15:22:40 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id C8B94434B9 for ; Wed, 24 Apr 2024 15:22:38 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964959; x=1745500959; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=yNtyhPlqaKQpxLROs19p+u0RRVqEx/1Vo+G7SwqKUGc=; b=FOzJrDlicOZzzljDmfv9+5YhsGq8yIVPHMZHdoSaLms9/egR8w0eWuSZ tz/Tzn+kUN9ptbcvmq+MIXL85ev1okEMRQnB0CpE6fMPYouEtItMGY2Il 8UK4TB7iRz8oT0UQ8OwcqHudJzkMnLhXnFqovC2vkiz9p1tvM+3v+a58R dvrQEVbKHA0dSGTInt2MO1xEY4PJ0ZZQDOOJW/csOlABQgZTe6swUV+iR BvBB9GpcMkbhMX19B9eH0NvfRb80fpOMdsXn6Qy7RhE2i3hSzYAp/XPXM yi8V9nTj8CswEDgFG5EdFA+l8cBLqNEVW19m9THv35A3hrhNt7HOYpYV4 w==; X-CSE-ConnectionGUID: xPcVVr8jT02vM7qL5Bf1gg== X-CSE-MsgGUID: liCG2Ny0TvGf9WC/k50Oww== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289265" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289265" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:38 -0700 X-CSE-ConnectionGUID: xoMK06rVS8eemqqk2FMQeA== X-CSE-MsgGUID: trYjOabsRb6OnOTXR/aUmw== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749417" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:36 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Radoslaw Tyl , vladimir.medvedkin@intel.com, bruce.richardson@intel.com, Skajewski@dpdk.org, PiotrX , Mrozowicz@dpdk.org, SlawomirX , Michael@dpdk.org, Alice Subject: [PATCH v1 10/22] net/ixgbe/base: add fw_rst_cnt field to ixgbe_hw struct Date: Wed, 24 Apr 2024 14:21:44 +0100 Message-ID: <8ea00c51bd6e92b59757fdb861a28ecf5fc5f7cb.1713964708.git.anatoly.burakov@intel.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Radoslaw Tyl Add fw_rst_cnt to store the number of resets after fw update. This value is required to detect if the EICR.MNG event occurred after firmware update reset. Signed-off-by: Radoslaw Tyl Reviewed-by: Skajewski, PiotrX Reviewed-by: Mrozowicz, SlawomirX Reviewed-by: Michael, Alice --- drivers/net/ixgbe/base/ixgbe_type.h | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ixgbe/base/ixgbe_type.h b/drivers/net/ixgbe/base/ixgbe_type.h index 10164e274c..51b9ef274b 100644 --- a/drivers/net/ixgbe/base/ixgbe_type.h +++ b/drivers/net/ixgbe/base/ixgbe_type.h @@ -4164,6 +4164,7 @@ struct ixgbe_hw { bool allow_unsupported_sfp; bool wol_enabled; bool need_crosstalk_fix; + u32 fw_rst_cnt; }; #define ixgbe_call_func(hw, func, params, error) \ From patchwork Wed Apr 24 13:21:45 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139656 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id E1A5643EAD; Wed, 24 Apr 2024 15:23:36 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 20DED4349D; Wed, 24 Apr 2024 15:22:44 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id A1BFB434BC for ; Wed, 24 Apr 2024 15:22:41 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964962; x=1745500962; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=TtwpR2fF2mh/hl/L3hZxCjAqYlr8MD5kx5h7Q1JadU4=; b=DygZymZo96Rv0azxyIJyrFA9RW+UTlXfpaoOXikYVtRK1fqQEIoJF/gV xZTb/KbFx6kRp8wreExIkqO5axZeEJ2qNEH4M+3e2EdYg/0JUIn9pTO3x bfUdBCxZBEN/i0SHPZdu+sFRoSCwTud7LEuSD2/WL3jxRH2Z323ag66xi yo9w0VfjVKBkfHsHNMx9PbCYkRXvrPZnYNe9ysBbUidfFwBWEOjBnxufn k7Ypiv8o43mAMeorftCy40+UR5J1FiRR/A9sefbF7SZwfr0H+zWoLmmzd RZjwNq4j2IegaainnTRkTb3VF2VUMVNy9Y46Ts+5/Rn6fFwvToDhQNovt g==; X-CSE-ConnectionGUID: VTs4l2MQQKCfoJ3BmFraIQ== X-CSE-MsgGUID: lkk7Hy2mTWy6OU1RTqwg1Q== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289267" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289267" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:41 -0700 X-CSE-ConnectionGUID: S9W+d+nHToGTk0DN6GD8GQ== X-CSE-MsgGUID: xTlSz+wWTVemNeHLIHcY/w== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749428" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:39 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Mical MarekX , vladimir.medvedkin@intel.com, bruce.richardson@intel.com, Galazka@dpdk.org, Krzysztof , Eryk Rybak , Racicot@dpdk.org, Francis , Michael@dpdk.org, Alice Subject: [PATCH v1 11/22] net/ixgbe/base: replace HIC with direct register access Date: Wed, 24 Apr 2024 14:21:45 +0100 Message-ID: X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Mical MarekX Unifying FW access method to direct register read/writes across all Atom(R) C3000 products. Atom(R) C3000 fiber exhibited an issue with the Host Interface Command execution being locked when another LAN function attempted to acquire the SWFW sync on Manageability Host. This resulted in HIC atomicity break and bogus data being read since the other LAN function cleared all semaphores on timeout whereas HIC execution continued after unlock. Direct register IOSF access showed higher stability and reliability. Signed-off-by: Mical MarekX Reviewed-by: Galazka, Krzysztof Reviewed-by: Eryk Rybak Reviewed-by: Racicot, Francis Reviewed-by: Michael, Alice --- drivers/net/ixgbe/base/ixgbe_x550.c | 76 +---------------------------- drivers/net/ixgbe/base/ixgbe_x550.h | 4 -- 2 files changed, 2 insertions(+), 78 deletions(-) diff --git a/drivers/net/ixgbe/base/ixgbe_x550.c b/drivers/net/ixgbe/base/ixgbe_x550.c index 844c06cc08..ebfc78235e 100644 --- a/drivers/net/ixgbe/base/ixgbe_x550.c +++ b/drivers/net/ixgbe/base/ixgbe_x550.c @@ -764,14 +764,8 @@ s32 ixgbe_init_ops_X550EM_a(struct ixgbe_hw *hw) /* Start with generic X550EM init */ ret_val = ixgbe_init_ops_X550EM(hw); - if (hw->device_id == IXGBE_DEV_ID_X550EM_A_SGMII || - hw->device_id == IXGBE_DEV_ID_X550EM_A_SGMII_L) { - mac->ops.read_iosf_sb_reg = ixgbe_read_iosf_sb_reg_x550; - mac->ops.write_iosf_sb_reg = ixgbe_write_iosf_sb_reg_x550; - } else { - mac->ops.read_iosf_sb_reg = ixgbe_read_iosf_sb_reg_x550a; - mac->ops.write_iosf_sb_reg = ixgbe_write_iosf_sb_reg_x550a; - } + mac->ops.read_iosf_sb_reg = ixgbe_read_iosf_sb_reg_x550; + mac->ops.write_iosf_sb_reg = ixgbe_write_iosf_sb_reg_x550; mac->ops.acquire_swfw_sync = ixgbe_acquire_swfw_sync_X550a; mac->ops.release_swfw_sync = ixgbe_release_swfw_sync_X550a; @@ -1252,72 +1246,6 @@ s32 ixgbe_put_phy_token(struct ixgbe_hw *hw) return IXGBE_ERR_FW_RESP_INVALID; } -/** - * ixgbe_write_iosf_sb_reg_x550a - Writes a value to specified register - * of the IOSF device - * @hw: pointer to hardware structure - * @reg_addr: 32 bit PHY register to write - * @device_type: 3 bit device type - * @data: Data to write to the register - **/ -s32 ixgbe_write_iosf_sb_reg_x550a(struct ixgbe_hw *hw, u32 reg_addr, - u32 device_type, u32 data) -{ - struct ixgbe_hic_internal_phy_req write_cmd; - s32 status; - UNREFERENCED_1PARAMETER(device_type); - - memset(&write_cmd, 0, sizeof(write_cmd)); - write_cmd.hdr.cmd = FW_INT_PHY_REQ_CMD; - write_cmd.hdr.buf_len = FW_INT_PHY_REQ_LEN; - write_cmd.hdr.checksum = FW_DEFAULT_CHECKSUM; - write_cmd.port_number = hw->bus.lan_id; - write_cmd.command_type = FW_INT_PHY_REQ_WRITE; - write_cmd.address = IXGBE_CPU_TO_BE16(reg_addr); - write_cmd.write_data = IXGBE_CPU_TO_BE32(data); - - status = ixgbe_host_interface_command(hw, (u32 *)&write_cmd, - sizeof(write_cmd), - IXGBE_HI_COMMAND_TIMEOUT, false); - - return status; -} - -/** - * ixgbe_read_iosf_sb_reg_x550a - Reads specified register of the IOSF device - * @hw: pointer to hardware structure - * @reg_addr: 32 bit PHY register to write - * @device_type: 3 bit device type - * @data: Pointer to read data from the register - **/ -s32 ixgbe_read_iosf_sb_reg_x550a(struct ixgbe_hw *hw, u32 reg_addr, - u32 device_type, u32 *data) -{ - union { - struct ixgbe_hic_internal_phy_req cmd; - struct ixgbe_hic_internal_phy_resp rsp; - } hic; - s32 status; - UNREFERENCED_1PARAMETER(device_type); - - memset(&hic, 0, sizeof(hic)); - hic.cmd.hdr.cmd = FW_INT_PHY_REQ_CMD; - hic.cmd.hdr.buf_len = FW_INT_PHY_REQ_LEN; - hic.cmd.hdr.checksum = FW_DEFAULT_CHECKSUM; - hic.cmd.port_number = hw->bus.lan_id; - hic.cmd.command_type = FW_INT_PHY_REQ_READ; - hic.cmd.address = IXGBE_CPU_TO_BE16(reg_addr); - - status = ixgbe_host_interface_command(hw, (u32 *)&hic.cmd, - sizeof(hic.cmd), - IXGBE_HI_COMMAND_TIMEOUT, true); - - /* Extract the register value from the response. */ - *data = IXGBE_BE32_TO_CPU(hic.rsp.read_data); - - return status; -} - /** * ixgbe_disable_mdd_X550 * @hw: pointer to hardware structure diff --git a/drivers/net/ixgbe/base/ixgbe_x550.h b/drivers/net/ixgbe/base/ixgbe_x550.h index 10086ab423..cff107e2ab 100644 --- a/drivers/net/ixgbe/base/ixgbe_x550.h +++ b/drivers/net/ixgbe/base/ixgbe_x550.h @@ -40,10 +40,6 @@ s32 ixgbe_set_fw_drv_ver_x550(struct ixgbe_hw *hw, u8 maj, u8 min, u8 build, u8 ver, u16 len, const char *str); s32 ixgbe_get_phy_token(struct ixgbe_hw *); s32 ixgbe_put_phy_token(struct ixgbe_hw *); -s32 ixgbe_write_iosf_sb_reg_x550a(struct ixgbe_hw *hw, u32 reg_addr, - u32 device_type, u32 data); -s32 ixgbe_read_iosf_sb_reg_x550a(struct ixgbe_hw *hw, u32 reg_addr, - u32 device_type, u32 *data); void ixgbe_disable_mdd_X550(struct ixgbe_hw *hw); void ixgbe_enable_mdd_X550(struct ixgbe_hw *hw); void ixgbe_mdd_event_X550(struct ixgbe_hw *hw, u32 *vf_bitmap); From patchwork Wed Apr 24 13:21:46 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139657 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 95AB443EAD; Wed, 24 Apr 2024 15:23:45 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 69F44434BE; Wed, 24 Apr 2024 15:22:47 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id 92A7F434BC for ; Wed, 24 Apr 2024 15:22:44 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964965; x=1745500965; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=Az6ekstKTYrf2jGrZ3RcLmxxo/FXdjG1D0NxeKvHWnY=; b=nCUzP/J47WlcI3uFckqos82Qp7Yj5C4A2uQAFngPbOzfivSH4FUFWYz/ FVpjbAeHKmkAqPoO6k4LmAKrbuFxZPJ93nO1i8K5wbYIGgCfKwnICEkdk DtS+y0K3w1tI0kNfYr+Vm92VqcdypKvQ5WquuRaMasrvof7GpdceH5Jf6 00XZ6TKsYLE3fzNBo8AHZTXhVlgrTOsToSD/YdYTz/e/CWb4ZW/VU8vH0 E4SOBfTimf6QKvQ3/jRkFhiTti6sIG3Qc8jQ8prti15/njK2vbRL1/A3b H9BUJhYb2Oq9GBXEOmEN44IskmNWXJo3db9hzmIzcRd52prFHa2ZE0eAe w==; X-CSE-ConnectionGUID: rvjI3Sx0TFKTfImn6qVxHg== X-CSE-MsgGUID: +H8N1E/XSkSYttlDuoEEbA== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289269" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289269" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:44 -0700 X-CSE-ConnectionGUID: B9U/Wi+dQtGBOjLQ4lWyGA== X-CSE-MsgGUID: 4u0uM8tBTRClZKd/JmdEeg== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749446" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:42 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Slawomir Mrozowicz , vladimir.medvedkin@intel.com, bruce.richardson@intel.com, Tyl@dpdk.org, RadoslawX , Skajewski@dpdk.org, PiotrX , Michael@dpdk.org, Alice Subject: [PATCH v1 12/22] net/ixgbe/base: added link state handling Date: Wed, 24 Apr 2024 14:21:46 +0100 Message-ID: <299db742da84ed0767017d267694745e406b9c54.1713964708.git.anatoly.burakov@intel.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Slawomir Mrozowicz Adding a functionality that retrieves information from PF regarding the link state of a given VF. It is part of the general functionality that allows the PF driver to control the state of the virtual link VF devices. Signed-off-by: Slawomir Mrozowicz Reviewed-by: Tyl, RadoslawX Reviewed-by: Skajewski, PiotrX Reviewed-by: Michael, Alice --- drivers/net/ixgbe/base/ixgbe_mbx.h | 1 + drivers/net/ixgbe/base/ixgbe_type.h | 1 + drivers/net/ixgbe/base/ixgbe_vf.c | 29 +++++++++++++++++++++++++++++ drivers/net/ixgbe/base/ixgbe_vf.h | 1 + 4 files changed, 32 insertions(+) diff --git a/drivers/net/ixgbe/base/ixgbe_mbx.h b/drivers/net/ixgbe/base/ixgbe_mbx.h index fadfccaabb..f368b1d745 100644 --- a/drivers/net/ixgbe/base/ixgbe_mbx.h +++ b/drivers/net/ixgbe/base/ixgbe_mbx.h @@ -123,6 +123,7 @@ enum ixgbe_pfvf_api_rev { #define IXGBE_VF_GET_RETA 0x0a /* VF request for RETA */ #define IXGBE_VF_GET_RSS_KEY 0x0b /* get RSS key */ #define IXGBE_VF_UPDATE_XCAST_MODE 0x0c +#define IXGBE_VF_GET_LINK_STATE 0x10 /* get vf link state */ /* mode choices for IXGBE_VF_UPDATE_XCAST_MODE */ enum ixgbevf_xcast_modes { diff --git a/drivers/net/ixgbe/base/ixgbe_type.h b/drivers/net/ixgbe/base/ixgbe_type.h index 51b9ef274b..ec832fb1b0 100644 --- a/drivers/net/ixgbe/base/ixgbe_type.h +++ b/drivers/net/ixgbe/base/ixgbe_type.h @@ -3983,6 +3983,7 @@ struct ixgbe_mac_operations { ixgbe_mc_addr_itr); s32 (*update_mc_addr_list)(struct ixgbe_hw *, u8 *, u32, ixgbe_mc_addr_itr, bool clear); + s32 (*get_link_state)(struct ixgbe_hw *, bool *); s32 (*enable_mc)(struct ixgbe_hw *); s32 (*disable_mc)(struct ixgbe_hw *); s32 (*clear_vfta)(struct ixgbe_hw *); diff --git a/drivers/net/ixgbe/base/ixgbe_vf.c b/drivers/net/ixgbe/base/ixgbe_vf.c index 603c24b653..0d5b29ba50 100644 --- a/drivers/net/ixgbe/base/ixgbe_vf.c +++ b/drivers/net/ixgbe/base/ixgbe_vf.c @@ -46,6 +46,7 @@ s32 ixgbe_init_ops_vf(struct ixgbe_hw *hw) hw->mac.ops.init_rx_addrs = NULL; hw->mac.ops.update_mc_addr_list = ixgbe_update_mc_addr_list_vf; hw->mac.ops.update_xcast_mode = ixgbevf_update_xcast_mode; + hw->mac.ops.get_link_state = ixgbe_get_link_state_vf; hw->mac.ops.enable_mc = NULL; hw->mac.ops.disable_mc = NULL; hw->mac.ops.clear_vfta = NULL; @@ -426,6 +427,34 @@ s32 ixgbevf_update_xcast_mode(struct ixgbe_hw *hw, int xcast_mode) return IXGBE_SUCCESS; } +/** + * ixgbe_get_link_state_vf - Get VF link state from PF + * @hw: pointer to the HW structure + * @link_state: link state storage + * + * Returns state of the operation error or success. + **/ +s32 ixgbe_get_link_state_vf(struct ixgbe_hw *hw, bool *link_state) +{ + u32 msgbuf[2]; + s32 err; + s32 ret_val; + + msgbuf[0] = IXGBE_VF_GET_LINK_STATE; + msgbuf[1] = 0x0; + + err = ixgbevf_write_msg_read_ack(hw, msgbuf, msgbuf, 2); + + if (err || (msgbuf[0] & IXGBE_VT_MSGTYPE_FAILURE)) { + ret_val = IXGBE_ERR_MBX; + } else { + ret_val = IXGBE_SUCCESS; + *link_state = msgbuf[1]; + } + + return ret_val; +} + /** * ixgbe_set_vfta_vf - Set/Unset vlan filter table address * @hw: pointer to the HW structure diff --git a/drivers/net/ixgbe/base/ixgbe_vf.h b/drivers/net/ixgbe/base/ixgbe_vf.h index be58b4f76e..bd10865d57 100644 --- a/drivers/net/ixgbe/base/ixgbe_vf.h +++ b/drivers/net/ixgbe/base/ixgbe_vf.h @@ -106,6 +106,7 @@ s32 ixgbe_update_mc_addr_list_vf(struct ixgbe_hw *hw, u8 *mc_addr_list, u32 mc_addr_count, ixgbe_mc_addr_itr, bool clear); s32 ixgbevf_update_xcast_mode(struct ixgbe_hw *hw, int xcast_mode); +s32 ixgbe_get_link_state_vf(struct ixgbe_hw *hw, bool *link_state); s32 ixgbe_set_vfta_vf(struct ixgbe_hw *hw, u32 vlan, u32 vind, bool vlan_on, bool vlvf_bypass); s32 ixgbevf_rlpml_set_vf(struct ixgbe_hw *hw, u16 max_size); From patchwork Wed Apr 24 13:21:47 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139658 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 974F843EAD; Wed, 24 Apr 2024 15:23:51 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id B44D4434D0; Wed, 24 Apr 2024 15:22:48 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id A6ACC434C4 for ; Wed, 24 Apr 2024 15:22:46 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964967; x=1745500967; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=dePQDzWCecIc5qVMytR5k7/yBC+DFM9dedoM3jAXfY8=; b=B7M1Xw082P/q58QDA8WGIe9rGkchR4274U4oHq98ZFd+Rccu3b5BaJKp hksVRggu4Km/kKu8ByOWNnTYR5aPaBWoRFutEPPC0e7Hkaf0DGctxut5O OQF0E3tlVmZbikJsWskDfkReWRo9LwFxDZQfMWFzsS8gn7hzeBjqnc9I/ wQmdo2/cGehpGQTjPri8wZ6wwBAyvCizFHYP4LRLFJEDLSRBCndzkRaCM UxLGbMYDWthCO9FtYHQu/pkh7/1H95E5qqmlrczxV+XbF6dgpy8ruvGFd 44tn6CDdahEZ+mWznruNhieI+BS/hvLSSbRGLpoFg0+7DK64MlbU6RfFl A==; X-CSE-ConnectionGUID: jN/F7exxTteqGiyDHpHf8Q== X-CSE-MsgGUID: iV27lvEyTFitppM/TRqerQ== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289271" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289271" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:46 -0700 X-CSE-ConnectionGUID: wu0XAbs8QTyPYzY82FnnZg== X-CSE-MsgGUID: SdMjq9r1QgmyAPkIibM5XA== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749457" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:44 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Piotr Skajewski , vladimir.medvedkin@intel.com, bruce.richardson@intel.com, Michael@dpdk.org, Alice Subject: [PATCH v1 13/22] net/ixgbe/base: handle -Wimplicit-fallthrough Date: Wed, 24 Apr 2024 14:21:47 +0100 Message-ID: X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Piotr Skajewski As there have been a long list of flaws due to missing "break" statement kernel no longer allow implict fall-through and reports warning during compilation. Convert all "fall-through" comments to actual code. Signed-off-by: Piotr Skajewski Reviewed-by: Michael, Alice --- drivers/net/ixgbe/base/ixgbe_82599.c | 7 +++++-- drivers/net/ixgbe/base/ixgbe_common.c | 8 ++++++-- drivers/net/ixgbe/base/ixgbe_x550.c | 11 +++++++---- 3 files changed, 18 insertions(+), 8 deletions(-) diff --git a/drivers/net/ixgbe/base/ixgbe_82599.c b/drivers/net/ixgbe/base/ixgbe_82599.c index 73c7766ecf..562034b242 100644 --- a/drivers/net/ixgbe/base/ixgbe_82599.c +++ b/drivers/net/ixgbe/base/ixgbe_82599.c @@ -1719,7 +1719,8 @@ s32 ixgbe_fdir_set_input_mask_82599(struct ixgbe_hw *hw, case 0x0000: /* mask VLAN ID */ fdirm |= IXGBE_FDIRM_VLANID; - /* fall through */ + fdirm |= IXGBE_FDIRM_VLANP; + break; case 0x0FFF: /* mask VLAN priority */ fdirm |= IXGBE_FDIRM_VLANP; @@ -2009,7 +2010,9 @@ s32 ixgbe_fdir_add_perfect_filter_82599(struct ixgbe_hw *hw, DEBUGOUT(" Error on src/dst port\n"); return IXGBE_ERR_CONFIG; } - /* fall through */ + input_mask->formatted.flow_type = IXGBE_ATR_L4TYPE_IPV6_MASK | + IXGBE_ATR_L4TYPE_MASK; + break; case IXGBE_ATR_FLOW_TYPE_TCPV4: case IXGBE_ATR_FLOW_TYPE_TUNNELED_TCPV4: case IXGBE_ATR_FLOW_TYPE_UDPV4: diff --git a/drivers/net/ixgbe/base/ixgbe_common.c b/drivers/net/ixgbe/base/ixgbe_common.c index dd8310748b..27f633bc2f 100644 --- a/drivers/net/ixgbe/base/ixgbe_common.c +++ b/drivers/net/ixgbe/base/ixgbe_common.c @@ -231,7 +231,8 @@ s32 ixgbe_setup_fc_generic(struct ixgbe_hw *hw) if (ret_val != IXGBE_SUCCESS) goto out; - /* fall through - only backplane uses autoc */ + reg = IXGBE_READ_REG(hw, IXGBE_PCS1GANA); + break; case ixgbe_media_type_fiber_qsfp: case ixgbe_media_type_fiber: reg = IXGBE_READ_REG(hw, IXGBE_PCS1GANA); @@ -4725,7 +4726,10 @@ void ixgbe_set_rxpba_generic(struct ixgbe_hw *hw, int num_pb, u32 headroom, rxpktsize <<= IXGBE_RXPBSIZE_SHIFT; for (; i < (num_pb / 2); i++) IXGBE_WRITE_REG(hw, IXGBE_RXPBSIZE(i), rxpktsize); - /* fall through - configure remaining packet buffers */ + rxpktsize = (pbsize / (num_pb - i)) << IXGBE_RXPBSIZE_SHIFT; + for (; i < num_pb; i++) + IXGBE_WRITE_REG(hw, IXGBE_RXPBSIZE(i), rxpktsize); + break; case PBA_STRATEGY_EQUAL: rxpktsize = (pbsize / (num_pb - i)) << IXGBE_RXPBSIZE_SHIFT; for (; i < num_pb; i++) diff --git a/drivers/net/ixgbe/base/ixgbe_x550.c b/drivers/net/ixgbe/base/ixgbe_x550.c index ebfc78235e..29055a818a 100644 --- a/drivers/net/ixgbe/base/ixgbe_x550.c +++ b/drivers/net/ixgbe/base/ixgbe_x550.c @@ -326,8 +326,7 @@ STATIC s32 ixgbe_identify_phy_x550em(struct ixgbe_hw *hw) /* set up for CS4227 usage */ ixgbe_setup_mux_ctl(hw); ixgbe_check_cs4227(hw); - /* Fallthrough */ - + return ixgbe_identify_sfp_module_X550em(hw); case IXGBE_DEV_ID_X550EM_A_SFP_N: return ixgbe_identify_sfp_module_X550em(hw); break; @@ -1838,7 +1837,9 @@ s32 ixgbe_get_link_capabilities_X550em(struct ixgbe_hw *hw, break; } } - /* fall through */ + *speed = IXGBE_LINK_SPEED_10GB_FULL | + IXGBE_LINK_SPEED_1GB_FULL; + break; default: *speed = IXGBE_LINK_SPEED_10GB_FULL | IXGBE_LINK_SPEED_1GB_FULL; @@ -3558,7 +3559,9 @@ u64 ixgbe_get_supported_physical_layer_X550em(struct ixgbe_hw *hw) break; } } - /* fall through */ + physical_layer = IXGBE_PHYSICAL_LAYER_10GBASE_KR | + IXGBE_PHYSICAL_LAYER_1000BASE_KX; + break; case ixgbe_phy_x550em_xfi: physical_layer = IXGBE_PHYSICAL_LAYER_10GBASE_KR | IXGBE_PHYSICAL_LAYER_1000BASE_KX; From patchwork Wed Apr 24 13:21:48 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139659 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id B707643EAD; Wed, 24 Apr 2024 15:23:57 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 1A93A434C5; Wed, 24 Apr 2024 15:22:50 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id 899A4434CB for ; Wed, 24 Apr 2024 15:22:48 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964969; x=1745500969; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=A0z/eidMKpgXrDBHtUuns29/n9YUomGkQ09fbx9dlTw=; b=LtwU5Fz2eIgf0fXJCSzD50c9JiRqqCNwWXQrfSRaWOzRPZwgIca77XoO Cxqw7Ocue7BlMziDlPrviadKc43I5QddRyvOuGI4cBYiAh4I5rRiLExHX GpfYtf9bxD4tbxnOVsDOpWhn1k2Qac/59XbMciW4QAqQLoRnM/aNIFs+b uFdgVF+jmg6CsG+NAgco8U7iFjR1ehcgNLUWmG1c+H+zBn4s+kByr6Nll yONnGu3tCmsz8mSKdxy/tlbX679MLeHjF5LoGC6fmSQiewTXwpeFASWEY ikRgKpoLT0eL8Ft0VmFUrEgqfAp8GUC9V3ZS6sji2r54n4vpRrLNLN8CV w==; X-CSE-ConnectionGUID: qbkVwdUARXiCnfNWN19rPg== X-CSE-MsgGUID: 0lFOY/0WQ/KhR1ioN4DXFg== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289273" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289273" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:48 -0700 X-CSE-ConnectionGUID: FmVmTQ2dTwaUV5IVkQd/9g== X-CSE-MsgGUID: RthjyqMcT5yytLIJfPCSZA== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749466" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:46 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Marcin Jurczak , vladimir.medvedkin@intel.com, bruce.richardson@intel.com, Michael@dpdk.org, Alice Subject: [PATCH v1 14/22] net/ixgbe/base: remove non-inclusive language Date: Wed, 24 Apr 2024 14:21:48 +0100 Message-ID: <30677871bd5aece7288bb6e77178e8cf0286fe63.1713964708.git.anatoly.burakov@intel.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Marcin Jurczak This patch removes non-inclusive language from code, user interface and comments. Signed-off-by: Marcin Jurczak Reviewed-by: Michael, Alice --- drivers/net/ixgbe/base/ixgbe_common.c | 34 +++++++++++++-------------- drivers/net/ixgbe/base/ixgbe_common.h | 2 +- drivers/net/ixgbe/base/ixgbe_type.h | 10 ++++---- drivers/net/ixgbe/ixgbe_ethdev.c | 2 +- 4 files changed, 24 insertions(+), 24 deletions(-) diff --git a/drivers/net/ixgbe/base/ixgbe_common.c b/drivers/net/ixgbe/base/ixgbe_common.c index 27f633bc2f..bb6e72fbe6 100644 --- a/drivers/net/ixgbe/base/ixgbe_common.c +++ b/drivers/net/ixgbe/base/ixgbe_common.c @@ -140,7 +140,7 @@ bool ixgbe_device_supports_autoneg_fc(struct ixgbe_hw *hw) switch (hw->phy.media_type) { case ixgbe_media_type_fiber_qsfp: case ixgbe_media_type_fiber: - /* flow control autoneg black list */ + /* flow control autoneg block list */ switch (hw->device_id) { case IXGBE_DEV_ID_X550EM_A_SFP: case IXGBE_DEV_ID_X550EM_A_SFP_N: @@ -1108,10 +1108,10 @@ s32 ixgbe_stop_adapter_generic(struct ixgbe_hw *hw) msec_delay(2); /* - * Prevent the PCI-E bus from hanging by disabling PCI-E master + * Prevent the PCI-E bus from hanging by disabling PCI-E primary * access and verify no pending requests */ - return ixgbe_disable_pcie_master(hw); + return ixgbe_disable_pcie_primary(hw); } /** @@ -3169,32 +3169,32 @@ STATIC u32 ixgbe_pcie_timeout_poll(struct ixgbe_hw *hw) } /** - * ixgbe_disable_pcie_master - Disable PCI-express master access + * ixgbe_disable_pcie_primary - Disable PCI-express primary access * @hw: pointer to hardware structure * - * Disables PCI-Express master access and verifies there are no pending - * requests. IXGBE_ERR_MASTER_REQUESTS_PENDING is returned if master disable - * bit hasn't caused the master requests to be disabled, else IXGBE_SUCCESS - * is returned signifying master requests disabled. + * Disables PCI-Express primary access and verifies there are no pending + * requests. IXGBE_ERR_PRIMARY_REQUESTS_PENDING is returned if primary disable + * bit hasn't caused the primary requests to be disabled, else IXGBE_SUCCESS + * is returned signifying primary requests disabled. **/ -s32 ixgbe_disable_pcie_master(struct ixgbe_hw *hw) +s32 ixgbe_disable_pcie_primary(struct ixgbe_hw *hw) { s32 status = IXGBE_SUCCESS; u32 i, poll; u16 value; - DEBUGFUNC("ixgbe_disable_pcie_master"); + DEBUGFUNC("ixgbe_disable_pcie_primary"); /* Always set this bit to ensure any future transactions are blocked */ IXGBE_WRITE_REG(hw, IXGBE_CTRL, IXGBE_CTRL_GIO_DIS); - /* Exit if master requests are blocked */ + /* Exit if primary requests are blocked */ if (!(IXGBE_READ_REG(hw, IXGBE_STATUS) & IXGBE_STATUS_GIO) || IXGBE_REMOVED(hw->hw_addr)) goto out; - /* Poll for master request bit to clear */ - for (i = 0; i < IXGBE_PCI_MASTER_DISABLE_TIMEOUT; i++) { + /* Poll for primary request bit to clear */ + for (i = 0; i < IXGBE_PCI_PRIMARY_DISABLE_TIMEOUT; i++) { usec_delay(100); if (!(IXGBE_READ_REG(hw, IXGBE_STATUS) & IXGBE_STATUS_GIO)) goto out; @@ -3202,13 +3202,13 @@ s32 ixgbe_disable_pcie_master(struct ixgbe_hw *hw) /* * Two consecutive resets are required via CTRL.RST per datasheet - * 5.2.5.3.2 Master Disable. We set a flag to inform the reset routine - * of this need. The first reset prevents new master requests from + * 5.2.5.3.2 Primary Disable. We set a flag to inform the reset routine + * of this need. The first reset prevents new primary requests from * being issued by our device. We then must wait 1usec or more for any * remaining completions from the PCIe bus to trickle in, and then reset * again to clear out any effects they may have had on our device. */ - DEBUGOUT("GIO Master Disable bit didn't clear - requesting resets\n"); + DEBUGOUT("GIO Primary Disable bit didn't clear - requesting resets\n"); hw->mac.flags |= IXGBE_FLAGS_DOUBLE_RESET_REQUIRED; if (hw->mac.type >= ixgbe_mac_X550) @@ -3230,7 +3230,7 @@ s32 ixgbe_disable_pcie_master(struct ixgbe_hw *hw) ERROR_REPORT1(IXGBE_ERROR_POLLING, "PCIe transaction pending bit also did not clear.\n"); - status = IXGBE_ERR_MASTER_REQUESTS_PENDING; + status = IXGBE_ERR_PRIMARY_REQUESTS_PENDING; out: return status; diff --git a/drivers/net/ixgbe/base/ixgbe_common.h b/drivers/net/ixgbe/base/ixgbe_common.h index 5bdb484407..bc4466ddf3 100644 --- a/drivers/net/ixgbe/base/ixgbe_common.h +++ b/drivers/net/ixgbe/base/ixgbe_common.h @@ -88,7 +88,7 @@ s32 ixgbe_setup_fc_generic(struct ixgbe_hw *hw); s32 ixgbe_validate_mac_addr(u8 *mac_addr); s32 ixgbe_acquire_swfw_sync(struct ixgbe_hw *hw, u32 mask); void ixgbe_release_swfw_sync(struct ixgbe_hw *hw, u32 mask); -s32 ixgbe_disable_pcie_master(struct ixgbe_hw *hw); +s32 ixgbe_disable_pcie_primary(struct ixgbe_hw *hw); s32 prot_autoc_read_generic(struct ixgbe_hw *hw, bool *, u32 *reg_val); s32 prot_autoc_write_generic(struct ixgbe_hw *hw, u32 reg_val, bool locked); diff --git a/drivers/net/ixgbe/base/ixgbe_type.h b/drivers/net/ixgbe/base/ixgbe_type.h index ec832fb1b0..5bf03a1f62 100644 --- a/drivers/net/ixgbe/base/ixgbe_type.h +++ b/drivers/net/ixgbe/base/ixgbe_type.h @@ -1456,7 +1456,7 @@ struct ixgbe_dmac_config { #define IXGBE_PSRTYPE_RQPL_SHIFT 29 /* CTRL Bit Masks */ -#define IXGBE_CTRL_GIO_DIS 0x00000004 /* Global IO Master Disable bit */ +#define IXGBE_CTRL_GIO_DIS 0x00000004 /* Global IO Primary Disable bit */ #define IXGBE_CTRL_LNK_RST 0x00000008 /* Link Reset. Resets everything. */ #define IXGBE_CTRL_RST 0x04000000 /* Reset (SW) */ #define IXGBE_CTRL_RST_MASK (IXGBE_CTRL_LNK_RST | IXGBE_CTRL_RST) @@ -2104,7 +2104,7 @@ enum { /* STATUS Bit Masks */ #define IXGBE_STATUS_LAN_ID 0x0000000C /* LAN ID */ #define IXGBE_STATUS_LAN_ID_SHIFT 2 /* LAN ID Shift*/ -#define IXGBE_STATUS_GIO 0x00080000 /* GIO Master Ena Status */ +#define IXGBE_STATUS_GIO 0x00080000 /* GIO Primary Ena Status */ #define IXGBE_STATUS_LAN_ID_0 0x00000000 /* LAN ID 0 */ #define IXGBE_STATUS_LAN_ID_1 0x00000004 /* LAN ID 1 */ @@ -2514,8 +2514,8 @@ enum { #define IXGBE_PCIDEVCTRL2_4_8s 0xd #define IXGBE_PCIDEVCTRL2_17_34s 0xe -/* Number of 100 microseconds we wait for PCI Express master disable */ -#define IXGBE_PCI_MASTER_DISABLE_TIMEOUT 800 +/* Number of 100 microseconds we wait for PCI Express primary disable */ +#define IXGBE_PCI_PRIMARY_DISABLE_TIMEOUT 800 /* Check whether address is multicast. This is little-endian specific check.*/ #define IXGBE_IS_MULTICAST(Address) \ @@ -4185,7 +4185,7 @@ struct ixgbe_hw { #define IXGBE_ERR_ADAPTER_STOPPED -9 #define IXGBE_ERR_INVALID_MAC_ADDR -10 #define IXGBE_ERR_DEVICE_NOT_SUPPORTED -11 -#define IXGBE_ERR_MASTER_REQUESTS_PENDING -12 +#define IXGBE_ERR_PRIMARY_REQUESTS_PENDING -12 #define IXGBE_ERR_INVALID_LINK_SETTINGS -13 #define IXGBE_ERR_AUTONEG_NOT_COMPLETE -14 #define IXGBE_ERR_RESET_FAILED -15 diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c index f2a397a451..0a0cfb9674 100644 --- a/drivers/net/ixgbe/ixgbe_ethdev.c +++ b/drivers/net/ixgbe/ixgbe_ethdev.c @@ -3053,7 +3053,7 @@ ixgbe_dev_close(struct rte_eth_dev *dev) ixgbe_dev_free_queues(dev); - ixgbe_disable_pcie_master(hw); + ixgbe_disable_pcie_primary(hw); /* reprogram the RAR[0] in case user changed it. */ ixgbe_set_rar(hw, 0, hw->mac.addr, 0, IXGBE_RAH_AV); From patchwork Wed Apr 24 13:21:49 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139660 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 5605B43EAD; Wed, 24 Apr 2024 15:24:03 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 6307E434AD; Wed, 24 Apr 2024 15:22:52 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id 1D4E4434C7 for ; Wed, 24 Apr 2024 15:22:49 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964970; x=1745500970; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=pGIAgsVSnCsBr0uCmf5KfzmGfU5QPV0DkdvoQ51t2M8=; b=IWLxOpHYiYKSomhWpslDGLgpFy5VlfxaeLYLEr7mLqzmzJvv+1k4GwZM ULYOTl4xK1vuQDdPXTHu6vRY4GiPuyew3RpbUd4RbEaVwS0KrdQ93PY9W vqeVPU0+rObyBeqTaxjJZ4BIFQCusl/jK00npv6jHV2jS8xCAvqCdv1Yr fbDI+3Iq59G7lFXkE8SsR2ZxKszTNvgYczGMW/cCh39qcYeaC3mwbWIjx GYtPNXxCOFXbL2UQMwJKmIxWlrmE+FwB+G6TZm7DWlMNwEcf/nqWrndzz DHiKOG4QA0Ju4bkOZkSjPizfUrwKhnPTRxyFZ+GJRBvQfw1TipZ55gXqw g==; X-CSE-ConnectionGUID: GrGBfYjMTDqZL2VtvQNT0g== X-CSE-MsgGUID: SnCbcEkFSH+3i6HYy4V7CA== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289275" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289275" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:50 -0700 X-CSE-ConnectionGUID: knhvYFegSqu7G0kkRQQMoQ== X-CSE-MsgGUID: HH7LqRGMRm61uzflsVzn0Q== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749472" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:48 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Krzysztof Galazka , vladimir.medvedkin@intel.com, bruce.richardson@intel.com, Jan Sokolowski Subject: [PATCH v1 15/22] net/ixgbe/base: filter out spurious link up indication Date: Wed, 24 Apr 2024 14:21:49 +0100 Message-ID: X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Krzysztof Galazka Extend SFP+ cage crosstalk fix by re-checking link state after 5ms delay to filter out spurious link up indication by transceiver with no fibre cable connected. Signed-off-by: Krzysztof Galazka Reviewed-by: Jan Sokolowski --- drivers/net/ixgbe/base/ixgbe_common.c | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-) diff --git a/drivers/net/ixgbe/base/ixgbe_common.c b/drivers/net/ixgbe/base/ixgbe_common.c index bb6e72fbe6..a19f4d715c 100644 --- a/drivers/net/ixgbe/base/ixgbe_common.c +++ b/drivers/net/ixgbe/base/ixgbe_common.c @@ -4194,10 +4194,25 @@ s32 ixgbe_check_mac_link_generic(struct ixgbe_hw *hw, ixgbe_link_speed *speed, links_reg = IXGBE_READ_REG(hw, IXGBE_LINKS); } } else { - if (links_reg & IXGBE_LINKS_UP) + if (links_reg & IXGBE_LINKS_UP) { + if (ixgbe_need_crosstalk_fix(hw)) { + /* Check the link state again after a delay + * to filter out spurious link up + * notifications. + */ + msec_delay(5); + links_reg = IXGBE_READ_REG(hw, IXGBE_LINKS); + if (!(links_reg & IXGBE_LINKS_UP)) { + *link_up = false; + *speed = IXGBE_LINK_SPEED_UNKNOWN; + return IXGBE_SUCCESS; + } + + } *link_up = true; - else + } else { *link_up = false; + } } switch (links_reg & IXGBE_LINKS_SPEED_82599) { From patchwork Wed Apr 24 13:21:50 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139661 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id DD96743EAD; Wed, 24 Apr 2024 15:24:09 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id A65E1434D8; Wed, 24 Apr 2024 15:22:53 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id A84BB434A6 for ; Wed, 24 Apr 2024 15:22:51 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964972; x=1745500972; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=LxoC1/O3OLBPIcvjUPSpq7QKGPDzQgjZHaA0HCdQUtM=; b=cwWbFJ20Vw4LnM9I00FpNWpzeuXeWG4msqSa6tucawgClvdZ6UDojDsq y783TLpNNC4noilI1/ZpC+WFaqo5IoQMaUZs/gO4r+5+u2t+C3GuGQFnv E3HasXHpH1Td/c/m8DKE+R1ZoX8Gw/R9SxAvEZGHnvHcBIoHHABtB7ICF fH2+4gm3vEn0cYpYnhQX5CSVPEwAF3y6EMchCeluY7/003wpp3GtrQaUK BJWMPEU1fv+ors6h7tdACn5F6NodoXtYERYuljgy5FFnO0S1osqFXNU/8 i4MnU9p6y9LbYUCXIFxF4nPjFOTj5M7nMU9QAjSsJv1umuVP+RlFKcXRR g==; X-CSE-ConnectionGUID: ZfERqNV4Q62+AM5Vh6Wh9A== X-CSE-MsgGUID: n9YZrl5uQYeJdhJiG/8G0Q== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289279" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289279" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:51 -0700 X-CSE-ConnectionGUID: nn8LN18lSjq+1TTgNWiwLQ== X-CSE-MsgGUID: 5m1DHYUYQZKNpAwR2R7joQ== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749480" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:50 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Barbara Skobiej , vladimir.medvedkin@intel.com, bruce.richardson@intel.com Subject: [PATCH v1 16/22] net/ixgbe/base: remove circular header dependency Date: Wed, 24 Apr 2024 14:21:50 +0100 Message-ID: <26a8437e36cf6ce07ed35f693b59bb14155c44b8.1713964708.git.anatoly.burakov@intel.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Barbara Skobiej Including one header file in second header file should be avoided, so fix it by forward declaring the struct instead. Signed-off-by: Barbara Skobiej barbara.skobiej@intel.com --- drivers/net/ixgbe/base/ixgbe_mbx.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ixgbe/base/ixgbe_mbx.h b/drivers/net/ixgbe/base/ixgbe_mbx.h index f368b1d745..56ab435286 100644 --- a/drivers/net/ixgbe/base/ixgbe_mbx.h +++ b/drivers/net/ixgbe/base/ixgbe_mbx.h @@ -5,7 +5,7 @@ #ifndef _IXGBE_MBX_H_ #define _IXGBE_MBX_H_ -#include "ixgbe_type.h" +struct ixgbe_hw; struct ixgbe_mbx_operations { void (*init_params)(struct ixgbe_hw *hw); From patchwork Wed Apr 24 13:21:51 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139662 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 298B443EAD; Wed, 24 Apr 2024 15:24:17 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 6C575434DC; Wed, 24 Apr 2024 15:22:56 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id 596B9434BC for ; Wed, 24 Apr 2024 15:22:53 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964974; x=1745500974; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=nk4D9SUBDAfWvFzTKQ2Sx3ixpH/Y8PPIkUKqRhAn3zo=; b=Q9TYdKJ3akc2HqD816tQEPdkarMXJ0fCwabMRcfdbOSPQVbtb8dei26T zJ5AehFo5I0Vur9PwzAbcofGfsLOukf1fIgT00f8rwTlxxA9Q6B4gEZXw 6vZ0HeNS8v7DPIojA7JWmCNTckLtL0GybkFjqe9gDrQugBT0FS4J43XK3 k1uYUo3RJmZi28peHQhErbXleiNE5iAfVCSPEgtg0jXGewW8Nt1QPhYkM ESrj9FGZpI2DzI8CLXzU0Mhw5GDMgzAqzl0XBfO0oUx8iBnd9ASG4sRXP 2Jm0iAz3LHXKwUaRTq4rgDGHXyKpQ4SqSwaMOtyIQAogCdebjoRD6vG2/ w==; X-CSE-ConnectionGUID: Oml1t941RzyIMSrBzPMQ4w== X-CSE-MsgGUID: HNRo+fE4S7OiZMGm2RYwMg== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289283" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289283" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:53 -0700 X-CSE-ConnectionGUID: dJMX7Qw/TGO6k75cBVGHjA== X-CSE-MsgGUID: fgLKlfPaSlKa7V4qhEpWPw== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749484" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:51 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Barbara Skobiej , vladimir.medvedkin@intel.com, bruce.richardson@intel.com, Jan Sokolowski Subject: [PATCH v1 17/22] net/ixgbe/base: add missing QV defines Date: Wed, 24 Apr 2024 14:21:51 +0100 Message-ID: <1cceb0254215479dea2d8b7eb0ecbc0e499967ef.1713964708.git.anatoly.burakov@intel.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Barbara Skobiej This patch adds missing QV defines: - offset of ANVM data - Immediate Field module pointer offset - 2.5GBASE-T and 5GBASE-T physical layer types for X550 Signed-off-by: Barbara Skobiej Signed-off-by: Jan Sokolowski --- drivers/net/ixgbe/base/ixgbe_type.h | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/ixgbe/base/ixgbe_type.h b/drivers/net/ixgbe/base/ixgbe_type.h index 5bf03a1f62..9fed8b005c 100644 --- a/drivers/net/ixgbe/base/ixgbe_type.h +++ b/drivers/net/ixgbe/base/ixgbe_type.h @@ -3072,6 +3072,7 @@ enum ixgbe_fdir_pballoc_type { #define FW_SHADOW_RAM_DUMP_LEN 0 #define FW_DEFAULT_CHECKSUM 0xFF /* checksum always 0xFF */ #define FW_NVM_DATA_OFFSET 3 +#define FW_ANVM_DATA_OFFSET 3 #define FW_MAX_READ_BUFFER_SIZE 1024 #define FW_DISABLE_RXEN_CMD 0xDE #define FW_DISABLE_RXEN_LEN 0x1 @@ -3143,6 +3144,8 @@ enum ixgbe_fdir_pballoc_type { #define FW_PHY_INFO_ID_HI_MASK 0xFFFF0000u #define FW_PHY_INFO_ID_LO_MASK 0x0000FFFFu +#define IXGBE_SR_IMMEDIATE_VALUES_PTR 0x4E + /* Host Interface Command Structures */ #pragma pack(push, 1) @@ -3446,6 +3449,8 @@ typedef u64 ixgbe_physical_layer; #define IXGBE_PHYSICAL_LAYER_1000BASE_SX 0x04000 #define IXGBE_PHYSICAL_LAYER_10BASE_T 0x08000 #define IXGBE_PHYSICAL_LAYER_2500BASE_KX 0x10000 +#define IXGBE_PHYSICAL_LAYER_2500BASE_T 0x20000 +#define IXGBE_PHYSICAL_LAYER_5000BASE_T 0x40000 /* Flow Control Data Sheet defined values * Calculation and defines taken from 802.1bb Annex O From patchwork Wed Apr 24 13:21:52 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139663 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 53A2743EAD; Wed, 24 Apr 2024 15:24:22 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id B0786434E9; Wed, 24 Apr 2024 15:22:57 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id C9286434DC for ; Wed, 24 Apr 2024 15:22:54 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964975; x=1745500975; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=LXqj1lecBQynbyFGcuN9RQ1OZuHQzV0mtwblDvPYCpE=; b=IZQUUQ2cOBjLT056ioKe00cVnf+4dbuq/2EXyisG0jn/iTgiYW0GIWCm 175efxMJjtcG0APc5xyXH5K4MvEyZhH42CP/KXchfgYyQFLmsFUvlYYh7 tydfj3C9ANjIEeJTcOcCpMPvDKkT9M/PYhdDnjA4KIyQCQohuQWyox6/8 5/iwtOgkWKoqy8WG8BoULIpUi58mSew+Wc0Vgo8DjKdKXtrncDXb+Kq+A mrPrQSCnVIO7h/wMEan/JaTVtjO6lDrDHAqlBpoIW6eHcW1Ix9MJ1Qin3 pfr4smrpTLNWCevFyLtV2+rCzSPjXk5Z6wga6kGaiM+0O5yHjxq2yyzIm g==; X-CSE-ConnectionGUID: gOZSxNmKQw+Uq5/fNKsD5Q== X-CSE-MsgGUID: pBAgjuLOT/KztAZ82AsKfg== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289287" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289287" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:54 -0700 X-CSE-ConnectionGUID: yHcE0kmJSDG1rOUvysScVQ== X-CSE-MsgGUID: GUgh7SVMQ9y8o7W/xTh+vg== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749492" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:53 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Barbara Skobiej , vladimir.medvedkin@intel.com, bruce.richardson@intel.com Subject: [PATCH v1 18/22] net/ixgbe/base: improve SWFW semaphore acquisition Date: Wed, 24 Apr 2024 14:21:52 +0100 Message-ID: <379365e8c57822439cc77a6937c9ce315f5bc5ff.1713964708.git.anatoly.burakov@intel.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Barbara Skobiej HWSW semaphore acquisition in Atom C3000 NIC is a two stage process. Each time two semaphore acquisitions are required. Each second semaphore failure require re-acquisition of first semaphore. This patch decouples the two acquisitions preventing potentially hundreds of thousands of unnecessary loop iterations. Signed-off-by: Barbara Skobiej --- drivers/net/ixgbe/base/ixgbe_x550.c | 41 ++++++++++++++++------------- 1 file changed, 22 insertions(+), 19 deletions(-) diff --git a/drivers/net/ixgbe/base/ixgbe_x550.c b/drivers/net/ixgbe/base/ixgbe_x550.c index 29055a818a..74c2563dd5 100644 --- a/drivers/net/ixgbe/base/ixgbe_x550.c +++ b/drivers/net/ixgbe/base/ixgbe_x550.c @@ -4211,36 +4211,39 @@ STATIC s32 ixgbe_acquire_swfw_sync_X550a(struct ixgbe_hw *hw, u32 mask) DEBUGFUNC("ixgbe_acquire_swfw_sync_X550a"); + status = IXGBE_SUCCESS; + if (hmask) + status = ixgbe_acquire_swfw_sync_X540(hw, hmask); + + if (status) { + DEBUGOUT1("Could not acquire SWFW semaphore, Status = %d\n", status); + return status; + } + + if (!(mask & IXGBE_GSSR_TOKEN_SM)) + return IXGBE_SUCCESS; + while (--retries) { - status = IXGBE_SUCCESS; - if (hmask) - status = ixgbe_acquire_swfw_sync_X540(hw, hmask); - if (status) { - DEBUGOUT1("Could not acquire SWFW semaphore, Status = %d\n", - status); - return status; - } - if (!(mask & IXGBE_GSSR_TOKEN_SM)) - return IXGBE_SUCCESS; - status = ixgbe_get_phy_token(hw); - if (status == IXGBE_ERR_TOKEN_RETRY) - DEBUGOUT1("Could not acquire PHY token, Status = %d\n", - status); if (status == IXGBE_SUCCESS) return IXGBE_SUCCESS; - if (hmask) - ixgbe_release_swfw_sync_X540(hw, hmask); - if (status != IXGBE_ERR_TOKEN_RETRY) { - DEBUGOUT1("Unable to retry acquiring the PHY token, Status = %d\n", - status); + DEBUGOUT1("Retry acquiring the PHY token failed, Status = %d\n", status); + if (hmask) + ixgbe_release_swfw_sync_X540(hw, hmask); return status; } + + if (status == IXGBE_ERR_TOKEN_RETRY) + DEBUGOUT1("Could not acquire PHY token, Status = %d\n", + status); } + if (hmask) + ixgbe_release_swfw_sync_X540(hw, hmask); + DEBUGOUT1("Semaphore acquisition retries failed!: PHY ID = 0x%08X\n", hw->phy.id); return status; From patchwork Wed Apr 24 13:21:53 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139664 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 941E643EAD; Wed, 24 Apr 2024 15:24:28 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 0EFED434E8; Wed, 24 Apr 2024 15:22:59 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id 4533143478 for ; Wed, 24 Apr 2024 15:22:56 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964977; x=1745500977; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=mZl2m7YUtXSxvkJUd0oPtka1e4NgTjBDvKH+O8GquyY=; b=ZFeoHRZptIC6eZNGoFqwlJ0pJ8ISMGLHTYb6+vtRPeqDcc5l3K9NRJzq 357NEslP2uhJKs1F24mpbWZSj84lP87GDomjwMJ68iTfIEFEbKXd2yP// cosEWq13Pt/Jj3Mmc6oGPkMwoV69GWoYvEie9uj67+qIFwzYknXlsB/0i t3gW52vceVFxK2G7PFYOoWVCwd79u9DOvjl0iKljTmhWkqY1lG0xJAnjh tiVv3oeFKaD0/6pkQ5lfv67+WK1wfYnFVb88+IzEqCjAhcEerS47B2vSe SbgfK5rYJyp3MzjyVnvY9T7Pxv2J1uWfLbYF6vXcZsZKGZpSvjTyjFj5e A==; X-CSE-ConnectionGUID: A0gz6FHQQoiMcnETcBBtzA== X-CSE-MsgGUID: U/i73VxtSeuD8jIffUE3uA== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289292" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289292" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:56 -0700 X-CSE-ConnectionGUID: 721FCzWYTWKz3SsA8Aie7Q== X-CSE-MsgGUID: 8ssdqBW5Q/+sW0YNwWTASA== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749499" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:55 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Dawid Zielinski , vladimir.medvedkin@intel.com, bruce.richardson@intel.com Subject: [PATCH v1 19/22] net/ixgbe/base: prevent untrusted loop bound Date: Wed, 24 Apr 2024 14:21:53 +0100 Message-ID: <666102a18ae39c8e1f0beb08640224aaff0169fc.1713964708.git.anatoly.burakov@intel.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Dawid Zielinski Added length check against EEPROM size in words to prevent untrusted loop bound reported by static code analysis. Signed-off-by: Dawid Zielinski --- drivers/net/ixgbe/base/ixgbe_common.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ixgbe/base/ixgbe_common.c b/drivers/net/ixgbe/base/ixgbe_common.c index a19f4d715c..73b5935d88 100644 --- a/drivers/net/ixgbe/base/ixgbe_common.c +++ b/drivers/net/ixgbe/base/ixgbe_common.c @@ -675,7 +675,7 @@ s32 ixgbe_read_pba_string_generic(struct ixgbe_hw *hw, u8 *pba_num, return ret_val; } - if (length == 0xFFFF || length == 0) { + if (length == 0xFFFF || length == 0 || length > hw->eeprom.word_size) { DEBUGOUT("NVM PBA number section invalid length\n"); return IXGBE_ERR_PBA_SECTION; } From patchwork Wed Apr 24 13:21:54 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139665 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 0EC2743EAD; Wed, 24 Apr 2024 15:24:34 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 521A0434F0; Wed, 24 Apr 2024 15:23:01 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id AFE67434E8 for ; Wed, 24 Apr 2024 15:22:57 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964978; x=1745500978; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=suWypDnU0rXFZsLYbTupnvZLj3ppJWv6PtqZtYGi+Xo=; b=nykCJxr0qBWI5952N9fisPX27IDPExZ1BuuUQI5PhmdATRKvSht85V3o qm/R9EeOo+omqsAXWaAGjuA3TlA768NP/FzlvPYg9vHy+fMOIdUujZhY6 Qdkc0ablM0bq2h5QEpBxRi/rmkLqo9O6TWSTnplh3Fa+5IcmhHOu0iLCF gRmcT17AFwAmJQTNWD/mAZKPFLnNrvO7I8iBhpcrzKvYFnvzAkBq5tc/4 N4M7HhhA4tvqXTT/Ehv+GIHNXUP91KNc8D1TVHW+MHmZnvUt0KoBkIJvz qi0MX+3b4vJwp0S/TORVL/YFPGVhZKJo9pyH7/kYLuJ7p0v8i2inaezYf w==; X-CSE-ConnectionGUID: oKxrUK9VRpWNbf8eLrD3Rg== X-CSE-MsgGUID: 9iMJ28RRR+imB6QKC2OEXQ== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289301" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289301" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:57 -0700 X-CSE-ConnectionGUID: PibyXSDvStaqlUAkvQaniw== X-CSE-MsgGUID: YMFpq+WJTYigxItGU+ZSKA== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749503" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:56 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Milosz Szymonek , vladimir.medvedkin@intel.com, bruce.richardson@intel.com Subject: [PATCH v1 20/22] net/ixgbe/base: add IXGBE_ADVTXD_MACLEN_MASK macro Date: Wed, 24 Apr 2024 14:21:54 +0100 Message-ID: X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Milosz Szymonek Add mask for MACLEN indexes in advanced context descriptors (bits 9:15). IXGBE_ADVTXD_MACLEN_MASK will be used in Windows driver code. Signed-off-by: Milosz Szymonek --- drivers/net/ixgbe/base/ixgbe_type.h | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ixgbe/base/ixgbe_type.h b/drivers/net/ixgbe/base/ixgbe_type.h index 9fed8b005c..5db9e03b4d 100644 --- a/drivers/net/ixgbe/base/ixgbe_type.h +++ b/drivers/net/ixgbe/base/ixgbe_type.h @@ -3381,6 +3381,7 @@ struct ixgbe_adv_tx_context_desc { #define IXGBE_ADVTXD_POPTS_RSV 0x00002000 /* POPTS Reserved */ #define IXGBE_ADVTXD_PAYLEN_SHIFT 14 /* Adv desc PAYLEN shift */ #define IXGBE_ADVTXD_MACLEN_SHIFT 9 /* Adv ctxt desc mac len shift */ +#define IXGBE_ADVTXD_MACLEN_MASK (0x7F << IXGBE_ADVTXD_MACLEN_SHIFT) /* Adv ctxt desc mac len mask */ #define IXGBE_ADVTXD_VLAN_SHIFT 16 /* Adv ctxt vlan tag shift */ #define IXGBE_ADVTXD_TUCMD_IPV4 0x00000400 /* IP Packet Type: 1=IPv4 */ #define IXGBE_ADVTXD_TUCMD_IPV6 0x00000000 /* IP Packet Type: 0=IPv6 */ From patchwork Wed Apr 24 13:21:55 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139666 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id C442243EAD; Wed, 24 Apr 2024 15:24:39 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 812AB434F6; Wed, 24 Apr 2024 15:23:02 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id 334E1434EC for ; Wed, 24 Apr 2024 15:22:59 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964979; x=1745500979; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=75pIUvaolgyknQu7XLD/O6YIctRlfC247VmqM/jZLQQ=; b=hxwkGL8UklfZV9VxXpvltRstb8gJaCgm44BAYGy4GVQxh25VwYkFbnbD ENN4kAmsgOFsnlrv3dwOOJ1Zzul9d8Ftp2SWZFRVNou9amwh6zLJ7TUk6 XMDI+KPtSmkrtgkafawFmTx1lgSmhWJ0VUuzhn89JcsCmpF9F2URzAvl+ 3FhmGfTfR+K8i/4WrR8b+B6nkfCYi3Zmzg59ia94HJm99NvPZ2Cuo0RtC NM80GGfT5ssMe+zEA5hPU9C28Orw4HEpV4eEx/8odQReq2GqKJ+Sh11h8 bUMxpAWk88Dq9NKBvPq61OAf4v+Febx0nQ9QXIv/p230v0eO9inVOds4l g==; X-CSE-ConnectionGUID: gDoT4nWtQduRlPueyHcZ4w== X-CSE-MsgGUID: 4G5Li9+iR8SwOoyWHN1rRQ== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289306" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289306" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:22:59 -0700 X-CSE-ConnectionGUID: WoBTq7I3SzK6InuxS4VEXA== X-CSE-MsgGUID: 6tCO84sBT5GIOCYki0IhyA== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749510" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:58 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Chinh Cao , vladimir.medvedkin@intel.com, bruce.richardson@intel.com Subject: [PATCH v1 21/22] net/ixgbe/base: remove prototypes of unimplemented functions Date: Wed, 24 Apr 2024 14:21:55 +0100 Message-ID: X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Chinh Cao There are some function prototypes that were introduced at some point but were never implemented, so remove them. Signed-off-by: Chinh Cao --- drivers/net/ixgbe/base/ixgbe_api.h | 1 - drivers/net/ixgbe/base/ixgbe_x550.h | 2 -- 2 files changed, 3 deletions(-) diff --git a/drivers/net/ixgbe/base/ixgbe_api.h b/drivers/net/ixgbe/base/ixgbe_api.h index 33e7c3c215..51decc5fae 100644 --- a/drivers/net/ixgbe/base/ixgbe_api.h +++ b/drivers/net/ixgbe/base/ixgbe_api.h @@ -118,7 +118,6 @@ u64 ixgbe_get_supported_physical_layer(struct ixgbe_hw *hw); s32 ixgbe_enable_rx_dma(struct ixgbe_hw *hw, u32 regval); s32 ixgbe_disable_sec_rx_path(struct ixgbe_hw *hw); s32 ixgbe_enable_sec_rx_path(struct ixgbe_hw *hw); -s32 ixgbe_mng_fw_enabled(struct ixgbe_hw *hw); s32 ixgbe_reinit_fdir_tables_82599(struct ixgbe_hw *hw); s32 ixgbe_init_fdir_signature_82599(struct ixgbe_hw *hw, u32 fdirctrl); s32 ixgbe_init_fdir_perfect_82599(struct ixgbe_hw *hw, u32 fdirctrl, diff --git a/drivers/net/ixgbe/base/ixgbe_x550.h b/drivers/net/ixgbe/base/ixgbe_x550.h index cff107e2ab..3e3ca8e66a 100644 --- a/drivers/net/ixgbe/base/ixgbe_x550.h +++ b/drivers/net/ixgbe/base/ixgbe_x550.h @@ -72,9 +72,7 @@ s32 ixgbe_read_phy_reg_x550a(struct ixgbe_hw *hw, u32 reg_addr, u32 device_type, u16 *phy_data); s32 ixgbe_write_phy_reg_x550a(struct ixgbe_hw *hw, u32 reg_addr, u32 device_type, u16 phy_data); -s32 ixgbe_setup_fc_fiber_x550em_a(struct ixgbe_hw *hw); s32 ixgbe_setup_fc_backplane_x550em_a(struct ixgbe_hw *hw); -s32 ixgbe_setup_fc_sgmii_x550em_a(struct ixgbe_hw *hw); void ixgbe_fc_autoneg_fiber_x550em_a(struct ixgbe_hw *hw); void ixgbe_fc_autoneg_backplane_x550em_a(struct ixgbe_hw *hw); void ixgbe_fc_autoneg_sgmii_x550em_a(struct ixgbe_hw *hw); From patchwork Wed Apr 24 13:21:56 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Anatoly Burakov X-Patchwork-Id: 139667 X-Patchwork-Delegate: bruce.richardson@intel.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 9994743EAD; Wed, 24 Apr 2024 15:24:45 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id D76D743488; Wed, 24 Apr 2024 15:23:12 +0200 (CEST) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by mails.dpdk.org (Postfix) with ESMTP id BC6784348B for ; Wed, 24 Apr 2024 15:23:09 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1713964990; x=1745500990; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=/mtym0NBuDhV99QuhMlIZxZnwDphERoetG5bQOLPFoo=; b=gb1b8tyjSUjyHC57xkjDA/TmU0yAEBSH3h3GQAnZuX/meAkp8EzcNXmp s+75BphVVxkMUT9N8Lykik0AXvv2U0cGphFr7+TTP9vy8QcsOAOadvo4I AK8f+SgJkck0jJq94omVpkWuODbq0AzSCYl1oelVXa43ZBJKraLiUvFoh znSSF0GVQ87hJ2WtDyFNqqnq6nAEnaGvqvru3vHrPkeX1v/3I7lHyU8un jAN647RH9Q5XubKwRUu67h4bZhtgOVDtr/OXaHChXWA66/L6fYUJQflE9 mcEj3mpYjxu31RRJqQ5pEu/1qzhflLoLxNdkElZDp+gaPJprJwoOAe2lR A==; X-CSE-ConnectionGUID: S6sES1m0TVelWMVVOJabYg== X-CSE-MsgGUID: FkBSp/ECQG+skoPUubcjxA== X-IronPort-AV: E=McAfee;i="6600,9927,11054"; a="20289342" X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="20289342" Received: from fmviesa006.fm.intel.com ([10.60.135.146]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 24 Apr 2024 06:23:09 -0700 X-CSE-ConnectionGUID: nzQnEzCqS72MNgTh+krTug== X-CSE-MsgGUID: hmO95BSvSyiIngpPcBYpLw== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,226,1708416000"; d="scan'208";a="24749529" Received: from silpixa00401119.ir.intel.com ([10.55.129.167]) by fmviesa006.fm.intel.com with ESMTP; 24 Apr 2024 06:22:59 -0700 From: Anatoly Burakov To: dev@dpdk.org Cc: Stefan Wegrzyn , vladimir.medvedkin@intel.com, bruce.richardson@intel.com, Andrii Staikov , Andrzej Komor , Artur Tyminski , Bartosz Jakub Rosadzinski , Brzezinski@dpdk.org, Filip , Carolyn Wyborny , Chinh Cao , Dawid Zielinski , Eryk Rybak , Fabio Pricoco , Jedrzej Jagielski , Julian Grajkowski , Krzysztof Galazka , Leszek Zygo , Mical MarekX , Milosz Szymonek , Pawel Malinowski , Piotr Kubaj , Piotr Skajewski , Slawomir Mrozowicz , Yogesh Bhosale Subject: [PATCH v1 22/22] net/ixgbe/base: add support for E610 device Date: Wed, 24 Apr 2024 14:21:56 +0100 Message-ID: <9dc7b80e7c3542323ab1d0d22ab96a882abb7fff.1713964708.git.anatoly.burakov@intel.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: References: MIME-Version: 1.0 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Stefan Wegrzyn This patch adds support for E610 device to the ixgbe driver. This is a squashed commit from internal development tree. Signed-off-by: Andrii Staikov Signed-off-by: Andrzej Komor Signed-off-by: Artur Tyminski Signed-off-by: Bartosz Jakub Rosadzinski Signed-off-by: Brzezinski, Filip Signed-off-by: Carolyn Wyborny Signed-off-by: Chinh Cao Signed-off-by: Dawid Zielinski Signed-off-by: Eryk Rybak Signed-off-by: Fabio Pricoco Signed-off-by: Jedrzej Jagielski Signed-off-by: Julian Grajkowski Signed-off-by: Krzysztof Galazka Signed-off-by: Leszek Zygo Signed-off-by: Mical MarekX Signed-off-by: Milosz Szymonek Signed-off-by: Pawel Malinowski Signed-off-by: Piotr Kubaj Signed-off-by: Piotr Skajewski Signed-off-by: Slawomir Mrozowicz Signed-off-by: Stefan Wegrzyn Signed-off-by: Yogesh Bhosale Signed-off-by: Anatoly Burakov --- drivers/net/ixgbe/base/README | 6 +- drivers/net/ixgbe/base/ixgbe_82599.c | 5 +- drivers/net/ixgbe/base/ixgbe_api.c | 58 +- drivers/net/ixgbe/base/ixgbe_api.h | 5 +- drivers/net/ixgbe/base/ixgbe_common.c | 29 +- drivers/net/ixgbe/base/ixgbe_e610.c | 4982 ++++++++++++++++++++++ drivers/net/ixgbe/base/ixgbe_e610.h | 163 + drivers/net/ixgbe/base/ixgbe_hv_vf.c | 3 +- drivers/net/ixgbe/base/ixgbe_mbx.c | 3 + drivers/net/ixgbe/base/ixgbe_osdep.c | 43 + drivers/net/ixgbe/base/ixgbe_osdep.h | 17 + drivers/net/ixgbe/base/ixgbe_phy.c | 4 +- drivers/net/ixgbe/base/ixgbe_type.h | 65 +- drivers/net/ixgbe/base/ixgbe_type_e610.h | 2181 ++++++++++ drivers/net/ixgbe/base/meson.build | 2 + 15 files changed, 7547 insertions(+), 19 deletions(-) create mode 100644 drivers/net/ixgbe/base/ixgbe_e610.c create mode 100644 drivers/net/ixgbe/base/ixgbe_e610.h create mode 100644 drivers/net/ixgbe/base/ixgbe_osdep.c create mode 100644 drivers/net/ixgbe/base/ixgbe_type_e610.h diff --git a/drivers/net/ixgbe/base/README b/drivers/net/ixgbe/base/README index 2c74693924..98353ba26f 100644 --- a/drivers/net/ixgbe/base/README +++ b/drivers/net/ixgbe/base/README @@ -1,12 +1,12 @@ /* SPDX-License-Identifier: BSD-3-Clause - * Copyright(c) 2010-2020 Intel Corporation + * Copyright(c) 2010-2024 Intel Corporation */ Intel® IXGBE driver =================== This directory contains source code of FreeBSD ixgbe driver of version -not-released-cid-ixgbe.2020.06.09.tar.gz released by the team which develop +not-released-cid-ixgbe.2024.04.24.tar.gz released by the team which develop basic drivers for any ixgbe NIC. The sub-directory of base/ contains the original source package. This driver is valid for the product(s) listed below @@ -24,6 +24,7 @@ This driver is valid for the product(s) listed below * Intel® Ethernet Server Adapter X520 Series * Intel® Ethernet Server Adapter X520-T2 * Intel® Ethernet Controller X550 Series +* Intel® Ethernet Controller E610 Series Updating the driver =================== @@ -32,3 +33,4 @@ NOTE: The source code in this directory should not be modified apart from the following file(s): ixgbe_osdep.h + ixgbe_osdep.c diff --git a/drivers/net/ixgbe/base/ixgbe_82599.c b/drivers/net/ixgbe/base/ixgbe_82599.c index 562034b242..419fec689e 100644 --- a/drivers/net/ixgbe/base/ixgbe_82599.c +++ b/drivers/net/ixgbe/base/ixgbe_82599.c @@ -1389,7 +1389,8 @@ void ixgbe_set_fdir_drop_queue_82599(struct ixgbe_hw *hw, u8 dropqueue) fdirctrl |= (dropqueue << IXGBE_FDIRCTRL_DROP_Q_SHIFT); if ((hw->mac.type == ixgbe_mac_X550) || (hw->mac.type == ixgbe_mac_X550EM_x) || - (hw->mac.type == ixgbe_mac_X550EM_a)) + (hw->mac.type == ixgbe_mac_X550EM_a) || + (hw->mac.type == ixgbe_mac_E610)) fdirctrl |= IXGBE_FDIRCTRL_DROP_NO_MATCH; IXGBE_WRITE_REG(hw, IXGBE_FDIRCMD, @@ -1804,6 +1805,7 @@ s32 ixgbe_fdir_set_input_mask_82599(struct ixgbe_hw *hw, case ixgbe_mac_X550: case ixgbe_mac_X550EM_x: case ixgbe_mac_X550EM_a: + case ixgbe_mac_E610: IXGBE_WRITE_REG(hw, IXGBE_FDIRSCTPM, 0xFFFFFFFF); break; default: @@ -1827,6 +1829,7 @@ s32 ixgbe_fdir_set_input_mask_82599(struct ixgbe_hw *hw, case ixgbe_mac_X550: case ixgbe_mac_X550EM_x: case ixgbe_mac_X550EM_a: + case ixgbe_mac_E610: IXGBE_WRITE_REG(hw, IXGBE_FDIRSCTPM, ~fdirtcpm); break; default: diff --git a/drivers/net/ixgbe/base/ixgbe_api.c b/drivers/net/ixgbe/base/ixgbe_api.c index d2b74cdffc..d348a81a88 100644 --- a/drivers/net/ixgbe/base/ixgbe_api.c +++ b/drivers/net/ixgbe/base/ixgbe_api.c @@ -89,6 +89,9 @@ s32 ixgbe_init_shared_code(struct ixgbe_hw *hw) case ixgbe_mac_X550EM_a_vf: status = ixgbe_init_ops_vf(hw); break; + case ixgbe_mac_E610: + status = ixgbe_init_ops_E610(hw); + break; default: status = IXGBE_ERR_DEVICE_NOT_SUPPORTED; break; @@ -208,6 +211,14 @@ s32 ixgbe_set_mac_type(struct ixgbe_hw *hw) hw->mac.type = ixgbe_mac_X550EM_a_vf; hw->mvals = ixgbe_mvals_X550EM_a; break; + case IXGBE_DEV_ID_E610_BACKPLANE: + case IXGBE_DEV_ID_E610_SFP: + case IXGBE_DEV_ID_E610_10G_T: + case IXGBE_DEV_ID_E610_2_5G_T: + case IXGBE_DEV_ID_E610_SGMII: + hw->mac.type = ixgbe_mac_E610; + hw->mvals = ixgbe_mvals_X550EM_a; + break; default: ret_val = IXGBE_ERR_DEVICE_NOT_SUPPORTED; ERROR_REPORT2(IXGBE_ERROR_UNSUPPORTED, @@ -445,7 +456,8 @@ s32 ixgbe_stop_adapter(struct ixgbe_hw *hw) **/ s32 ixgbe_read_pba_string(struct ixgbe_hw *hw, u8 *pba_num, u32 pba_num_size) { - return ixgbe_read_pba_string_generic(hw, pba_num, pba_num_size); + return ixgbe_call_func(hw, hw->eeprom.ops.read_pba_string, (hw, pba_num, + pba_num_size), IXGBE_NOT_IMPLEMENTED); } /** @@ -1141,6 +1153,19 @@ s32 ixgbe_set_fw_drv_ver(struct ixgbe_hw *hw, u8 maj, u8 min, u8 build, IXGBE_NOT_IMPLEMENTED); } +/** + * ixgbe_get_fw_tsam_mode - Returns information whether TSAM is enabled + * @hw: pointer to hardware structure + * + * Checks Thermal Sensor Autonomous Mode by reading the value of the + * dedicated register. + * Returns True if TSAM is enabled, False if TSAM is disabled. + */ +bool ixgbe_get_fw_tsam_mode(struct ixgbe_hw *hw) +{ + return ixgbe_call_func(hw, hw->mac.ops.get_fw_tsam_mode, (hw), + IXGBE_NOT_IMPLEMENTED); +} /** * ixgbe_get_thermal_sensor_data - Gathers thermal sensor data @@ -1684,3 +1709,34 @@ void ixgbe_set_rate_select_speed(struct ixgbe_hw *hw, ixgbe_link_speed speed) if (hw->mac.ops.set_rate_select_speed) hw->mac.ops.set_rate_select_speed(hw, speed); } + +/** + * ixgbe_get_fw_version - get FW version + * @hw: pointer to hardware structure + * + * Get the current FW version. + * + * Return: the exit code of the operation or IXGBE_NOT_IMPLEMENTED + * if the function is not implemented. + */ +s32 ixgbe_get_fw_version(struct ixgbe_hw *hw) +{ + return ixgbe_call_func(hw, hw->mac.ops.get_fw_version, + (hw), IXGBE_NOT_IMPLEMENTED); +} + +/** + * ixgbe_get_nvm_ver - get NVM version + * @hw: pointer to hardware structure + * @nvm: pointer to NVM info structure + * + * Get the current NVM version. + * + * Return: the exit code of the operation or IXGBE_NOT_IMPLEMENTED + * if the function is not implemented. + */ +s32 ixgbe_get_nvm_ver(struct ixgbe_hw* hw, struct ixgbe_nvm_info *nvm) +{ + return ixgbe_call_func(hw, hw->mac.ops.get_nvm_version, + (hw, nvm), IXGBE_NOT_IMPLEMENTED); +} diff --git a/drivers/net/ixgbe/base/ixgbe_api.h b/drivers/net/ixgbe/base/ixgbe_api.h index 51decc5fae..cb572a337d 100644 --- a/drivers/net/ixgbe/base/ixgbe_api.h +++ b/drivers/net/ixgbe/base/ixgbe_api.h @@ -18,6 +18,7 @@ extern s32 ixgbe_init_ops_X550(struct ixgbe_hw *hw); extern s32 ixgbe_init_ops_X550EM(struct ixgbe_hw *hw); extern s32 ixgbe_init_ops_X550EM_x(struct ixgbe_hw *hw); extern s32 ixgbe_init_ops_X550EM_a(struct ixgbe_hw *hw); +extern s32 ixgbe_init_ops_E610(struct ixgbe_hw *hw); extern s32 ixgbe_init_ops_vf(struct ixgbe_hw *hw); s32 ixgbe_set_mac_type(struct ixgbe_hw *hw); @@ -105,6 +106,7 @@ s32 ixgbe_fc_enable(struct ixgbe_hw *hw); s32 ixgbe_setup_fc(struct ixgbe_hw *hw); s32 ixgbe_set_fw_drv_ver(struct ixgbe_hw *hw, u8 maj, u8 min, u8 build, u8 ver, u16 len, char *driver_ver); +bool ixgbe_get_fw_tsam_mode(struct ixgbe_hw *hw); s32 ixgbe_get_thermal_sensor_data(struct ixgbe_hw *hw); s32 ixgbe_init_thermal_sensor_thresh(struct ixgbe_hw *hw); void ixgbe_set_mta(struct ixgbe_hw *hw, u8 *mc_addr); @@ -192,5 +194,6 @@ void ixgbe_disable_rx(struct ixgbe_hw *hw); void ixgbe_enable_rx(struct ixgbe_hw *hw); s32 ixgbe_negotiate_fc(struct ixgbe_hw *hw, u32 adv_reg, u32 lp_reg, u32 adv_sym, u32 adv_asm, u32 lp_sym, u32 lp_asm); - +s32 ixgbe_get_fw_version(struct ixgbe_hw *hw); +s32 ixgbe_get_nvm_ver(struct ixgbe_hw *hw, struct ixgbe_nvm_info *nvm); #endif /* _IXGBE_API_H_ */ diff --git a/drivers/net/ixgbe/base/ixgbe_common.c b/drivers/net/ixgbe/base/ixgbe_common.c index 73b5935d88..51fb4050c6 100644 --- a/drivers/net/ixgbe/base/ixgbe_common.c +++ b/drivers/net/ixgbe/base/ixgbe_common.c @@ -61,6 +61,7 @@ s32 ixgbe_init_ops_generic(struct ixgbe_hw *hw) ixgbe_validate_eeprom_checksum_generic; eeprom->ops.update_checksum = ixgbe_update_eeprom_checksum_generic; eeprom->ops.calc_checksum = ixgbe_calc_eeprom_checksum_generic; + eeprom->ops.read_pba_string = ixgbe_read_pba_string_generic; /* MAC */ mac->ops.init_hw = ixgbe_init_hw_generic; @@ -146,6 +147,7 @@ bool ixgbe_device_supports_autoneg_fc(struct ixgbe_hw *hw) case IXGBE_DEV_ID_X550EM_A_SFP_N: case IXGBE_DEV_ID_X550EM_A_QSFP: case IXGBE_DEV_ID_X550EM_A_QSFP_N: + case IXGBE_DEV_ID_E610_SFP: supported = false; break; default: @@ -177,6 +179,8 @@ bool ixgbe_device_supports_autoneg_fc(struct ixgbe_hw *hw) case IXGBE_DEV_ID_X550EM_A_10G_T: case IXGBE_DEV_ID_X550EM_A_1G_T: case IXGBE_DEV_ID_X550EM_A_1G_T_L: + case IXGBE_DEV_ID_E610_10G_T: + case IXGBE_DEV_ID_E610_2_5G_T: supported = true; break; default: @@ -577,17 +581,11 @@ s32 ixgbe_clear_hw_cntrs_generic(struct ixgbe_hw *hw) } } - if (hw->mac.type == ixgbe_mac_X550 || hw->mac.type == ixgbe_mac_X540) { + if (hw->mac.type == ixgbe_mac_X540 || + hw->mac.type == ixgbe_mac_X550 || + hw->mac.type == ixgbe_mac_E610) { if (hw->phy.id == 0) ixgbe_identify_phy(hw); - hw->phy.ops.read_reg(hw, IXGBE_PCRC8ECL, - IXGBE_MDIO_PCS_DEV_TYPE, &i); - hw->phy.ops.read_reg(hw, IXGBE_PCRC8ECH, - IXGBE_MDIO_PCS_DEV_TYPE, &i); - hw->phy.ops.read_reg(hw, IXGBE_LDPCECL, - IXGBE_MDIO_PCS_DEV_TYPE, &i); - hw->phy.ops.read_reg(hw, IXGBE_LDPCECH, - IXGBE_MDIO_PCS_DEV_TYPE, &i); } return IXGBE_SUCCESS; @@ -998,6 +996,9 @@ void ixgbe_set_pci_config_data_generic(struct ixgbe_hw *hw, u16 link_status) case IXGBE_PCI_LINK_SPEED_8000: hw->bus.speed = ixgbe_bus_speed_8000; break; + case IXGBE_PCI_LINK_SPEED_16000: + hw->bus.speed = ixgbe_bus_speed_16000; + break; default: hw->bus.speed = ixgbe_bus_speed_unknown; break; @@ -1020,7 +1021,9 @@ s32 ixgbe_get_bus_info_generic(struct ixgbe_hw *hw) DEBUGFUNC("ixgbe_get_bus_info_generic"); /* Get the negotiated link width and speed from PCI config space */ - link_status = IXGBE_READ_PCIE_WORD(hw, IXGBE_PCI_LINK_STATUS); + link_status = IXGBE_READ_PCIE_WORD(hw, hw->mac.type == ixgbe_mac_E610 ? + IXGBE_PCI_LINK_STATUS_E610 : + IXGBE_PCI_LINK_STATUS); ixgbe_set_pci_config_data_generic(hw, link_status); @@ -3653,6 +3656,10 @@ u16 ixgbe_get_pcie_msix_count_generic(struct ixgbe_hw *hw) pcie_offset = IXGBE_PCIE_MSIX_82599_CAPS; max_msix_count = IXGBE_MAX_MSIX_VECTORS_82599; break; + case ixgbe_mac_E610: + pcie_offset = IXGBE_PCIE_MSIX_LKV_CAPS; + max_msix_count = IXGBE_MAX_MSIX_VECTORS_82599; + break; default: return msix_count; } @@ -4228,7 +4235,7 @@ s32 ixgbe_check_mac_link_generic(struct ixgbe_hw *hw, ixgbe_link_speed *speed, break; case IXGBE_LINKS_SPEED_100_82599: *speed = IXGBE_LINK_SPEED_100_FULL; - if (hw->mac.type == ixgbe_mac_X550) { + if (hw->mac.type == ixgbe_mac_X550 || hw->mac.type == ixgbe_mac_E610) { if (links_reg & IXGBE_LINKS_SPEED_NON_STD) *speed = IXGBE_LINK_SPEED_5GB_FULL; } diff --git a/drivers/net/ixgbe/base/ixgbe_e610.c b/drivers/net/ixgbe/base/ixgbe_e610.c new file mode 100644 index 0000000000..37e0acf81c --- /dev/null +++ b/drivers/net/ixgbe/base/ixgbe_e610.c @@ -0,0 +1,4982 @@ +#include "ixgbe_type.h" +#include "ixgbe_e610.h" +#include "ixgbe_x550.h" +#include "ixgbe_common.h" +#include "ixgbe_phy.h" +#include "ixgbe_api.h" + +/** + * ixgbe_init_aci - initialization routine for Admin Command Interface + * @hw: pointer to the hardware structure + * + * Initialize the ACI lock. + */ +void ixgbe_init_aci(struct ixgbe_hw *hw) +{ + ixgbe_init_lock(&hw->aci.lock); +} + +/** + * ixgbe_shutdown_aci - shutdown routine for Admin Command Interface + * @hw: pointer to the hardware structure + * + * Destroy the ACI lock. + */ +void ixgbe_shutdown_aci(struct ixgbe_hw *hw) +{ + ixgbe_destroy_lock(&hw->aci.lock); +} + +/** + * ixgbe_should_retry_aci_send_cmd_execute - decide if ACI command should + * be resent + * @opcode: ACI opcode + * + * Check if ACI command should be sent again depending on the provided opcode. + * + * Return: true if the sending command routine should be repeated, + * otherwise false. + */ +STATIC bool ixgbe_should_retry_aci_send_cmd_execute(u16 opcode) +{ + + switch (opcode) { + case ixgbe_aci_opc_disable_rxen: + case ixgbe_aci_opc_get_phy_caps: + case ixgbe_aci_opc_get_link_status: + case ixgbe_aci_opc_get_link_topo: + return true; + } + + return false; +} + +/** + * ixgbe_aci_send_cmd_execute - execute sending FW Admin Command to FW Admin + * Command Interface + * @hw: pointer to the HW struct + * @desc: descriptor describing the command + * @buf: buffer to use for indirect commands (NULL for direct commands) + * @buf_size: size of buffer for indirect commands (0 for direct commands) + * + * Admin Command is sent using CSR by setting descriptor and buffer in specific + * registers. + * + * Return: the exit code of the operation. + * * - IXGBE_SUCCESS - success. + * * - IXGBE_ERR_ACI_DISABLED - CSR mechanism is not enabled. + * * - IXGBE_ERR_ACI_BUSY - CSR mechanism is busy. + * * - IXGBE_ERR_PARAM - buf_size is too big or + * invalid argument buf or buf_size. + * * - IXGBE_ERR_ACI_TIMEOUT - Admin Command X command timeout. + * * - IXGBE_ERR_ACI_ERROR - Admin Command X invalid state of HICR register or + * Admin Command failed because of bad opcode was returned or + * Admin Command failed with error Y. + */ +STATIC s32 +ixgbe_aci_send_cmd_execute(struct ixgbe_hw *hw, struct ixgbe_aci_desc *desc, + void *buf, u16 buf_size) +{ + u32 hicr = 0, tmp_buf_size = 0, i = 0; + u32 *raw_desc = (u32 *)desc; + s32 status = IXGBE_SUCCESS; + bool valid_buf = false; + u32 *tmp_buf = NULL; + u16 opcode = 0; + + do { + hw->aci.last_status = IXGBE_ACI_RC_OK; + + /* It's necessary to check if mechanism is enabled */ + hicr = IXGBE_READ_REG(hw, PF_HICR); + if (!(hicr & PF_HICR_EN)) { + status = IXGBE_ERR_ACI_DISABLED; + break; + } + if (hicr & PF_HICR_C) { + hw->aci.last_status = IXGBE_ACI_RC_EBUSY; + status = IXGBE_ERR_ACI_BUSY; + break; + } + opcode = desc->opcode; + + if (buf_size > IXGBE_ACI_MAX_BUFFER_SIZE) { + status = IXGBE_ERR_PARAM; + break; + } + + if (buf) + desc->flags |= IXGBE_CPU_TO_LE16(IXGBE_ACI_FLAG_BUF); + + /* Check if buf and buf_size are proper params */ + if (desc->flags & IXGBE_CPU_TO_LE16(IXGBE_ACI_FLAG_BUF)) { + if ((buf && buf_size == 0) || + (buf == NULL && buf_size)) { + status = IXGBE_ERR_PARAM; + break; + } + if (buf && buf_size) + valid_buf = true; + } + + if (valid_buf == true) { + if (buf_size % 4 == 0) + tmp_buf_size = buf_size; + else + tmp_buf_size = (buf_size & (u16)(~0x03)) + 4; + + tmp_buf = (u32*)ixgbe_malloc(hw, tmp_buf_size); + if (!tmp_buf) + return IXGBE_ERR_OUT_OF_MEM; + + /* tmp_buf will be firstly filled with 0xFF and after + * that the content of buf will be written into it. + * This approach lets us use valid buf_size and + * prevents us from reading past buf area + * when buf_size mod 4 not equal to 0. + */ + memset(tmp_buf, 0xFF, tmp_buf_size); + memcpy(tmp_buf, buf, buf_size); + + if (tmp_buf_size > IXGBE_ACI_LG_BUF) + desc->flags |= + IXGBE_CPU_TO_LE16(IXGBE_ACI_FLAG_LB); + + desc->datalen = IXGBE_CPU_TO_LE16(buf_size); + + if (desc->flags & IXGBE_CPU_TO_LE16(IXGBE_ACI_FLAG_RD)) { + for (i = 0; i < tmp_buf_size / 4; i++) { + IXGBE_WRITE_REG(hw, PF_HIBA(i), + IXGBE_LE32_TO_CPU(tmp_buf[i])); + } + } + } + + /* Descriptor is written to specific registers */ + for (i = 0; i < IXGBE_ACI_DESC_SIZE_IN_DWORDS; i++) + IXGBE_WRITE_REG(hw, PF_HIDA(i), + IXGBE_LE32_TO_CPU(raw_desc[i])); + + /* SW has to set PF_HICR.C bit and clear PF_HICR.SV and + * PF_HICR_EV + */ + hicr = IXGBE_READ_REG(hw, PF_HICR); + hicr = (hicr | PF_HICR_C) & ~(PF_HICR_SV | PF_HICR_EV); + IXGBE_WRITE_REG(hw, PF_HICR, hicr); + + /* Wait for sync Admin Command response */ + for (i = 0; i < IXGBE_ACI_SYNC_RESPONSE_TIMEOUT; i += 1) { + hicr = IXGBE_READ_REG(hw, PF_HICR); + if ((hicr & PF_HICR_SV) || !(hicr & PF_HICR_C)) + break; + + msec_delay(1); + } + + /* Wait for async Admin Command response */ + if ((hicr & PF_HICR_SV) && (hicr & PF_HICR_C)) { + for (i = 0; i < IXGBE_ACI_ASYNC_RESPONSE_TIMEOUT; + i += 1) { + hicr = IXGBE_READ_REG(hw, PF_HICR); + if ((hicr & PF_HICR_EV) || !(hicr & PF_HICR_C)) + break; + + msec_delay(1); + } + } + + /* Read sync Admin Command response */ + if ((hicr & PF_HICR_SV)) { + for (i = 0; i < IXGBE_ACI_DESC_SIZE_IN_DWORDS; i++) { + raw_desc[i] = IXGBE_READ_REG(hw, PF_HIDA(i)); + raw_desc[i] = IXGBE_CPU_TO_LE32(raw_desc[i]); + } + } + + /* Read async Admin Command response */ + if ((hicr & PF_HICR_EV) && !(hicr & PF_HICR_C)) { + for (i = 0; i < IXGBE_ACI_DESC_SIZE_IN_DWORDS; i++) { + raw_desc[i] = IXGBE_READ_REG(hw, PF_HIDA_2(i)); + raw_desc[i] = IXGBE_CPU_TO_LE32(raw_desc[i]); + } + } + + /* Handle timeout and invalid state of HICR register */ + if (hicr & PF_HICR_C) { + status = IXGBE_ERR_ACI_TIMEOUT; + break; + } else if (!(hicr & PF_HICR_SV) && !(hicr & PF_HICR_EV)) { + status = IXGBE_ERR_ACI_ERROR; + break; + } + + /* For every command other than 0x0014 treat opcode mismatch + * as an error. Response to 0x0014 command read from HIDA_2 + * is a descriptor of an event which is expected to contain + * different opcode than the command. + */ + if (desc->opcode != opcode && + opcode != IXGBE_CPU_TO_LE16(ixgbe_aci_opc_get_fw_event)) { + status = IXGBE_ERR_ACI_ERROR; + break; + } + + if (desc->retval != IXGBE_ACI_RC_OK) { + hw->aci.last_status = (enum ixgbe_aci_err)desc->retval; + status = IXGBE_ERR_ACI_ERROR; + break; + } + + /* Write a response values to a buf */ + if (valid_buf && (desc->flags & + IXGBE_CPU_TO_LE16(IXGBE_ACI_FLAG_BUF))) { + for (i = 0; i < tmp_buf_size / 4; i++) { + tmp_buf[i] = IXGBE_READ_REG(hw, PF_HIBA(i)); + tmp_buf[i] = IXGBE_CPU_TO_LE32(tmp_buf[i]); + } + memcpy(buf, tmp_buf, buf_size); + } + } while (0); + + if (tmp_buf) + ixgbe_free(hw, tmp_buf); + + return status; +} + +/** + * ixgbe_aci_send_cmd - send FW Admin Command to FW Admin Command Interface + * @hw: pointer to the HW struct + * @desc: descriptor describing the command + * @buf: buffer to use for indirect commands (NULL for direct commands) + * @buf_size: size of buffer for indirect commands (0 for direct commands) + * + * Helper function to send FW Admin Commands to the FW Admin Command Interface. + * + * Retry sending the FW Admin Command multiple times to the FW ACI + * if the EBUSY Admin Command error is returned. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_send_cmd(struct ixgbe_hw *hw, struct ixgbe_aci_desc *desc, + void *buf, u16 buf_size) +{ + struct ixgbe_aci_desc desc_cpy; + enum ixgbe_aci_err last_status; + bool is_cmd_for_retry; + u8 *buf_cpy = NULL; + s32 status; + u16 opcode; + u8 idx = 0; + + opcode = IXGBE_LE16_TO_CPU(desc->opcode); + is_cmd_for_retry = ixgbe_should_retry_aci_send_cmd_execute(opcode); + memset(&desc_cpy, 0, sizeof(desc_cpy)); + + if (is_cmd_for_retry) { + if (buf) { + buf_cpy = (u8 *)ixgbe_malloc(hw, buf_size); + if (!buf_cpy) + return IXGBE_ERR_OUT_OF_MEM; + } + memcpy(&desc_cpy, desc, sizeof(desc_cpy)); + } + + do { + ixgbe_acquire_lock(&hw->aci.lock); + status = ixgbe_aci_send_cmd_execute(hw, desc, buf, buf_size); + last_status = hw->aci.last_status; + ixgbe_release_lock(&hw->aci.lock); + + if (!is_cmd_for_retry || status == IXGBE_SUCCESS || + last_status != IXGBE_ACI_RC_EBUSY) + break; + + if (buf) + memcpy(buf, buf_cpy, buf_size); + memcpy(desc, &desc_cpy, sizeof(desc_cpy)); + + msec_delay(IXGBE_ACI_SEND_DELAY_TIME_MS); + } while (++idx < IXGBE_ACI_SEND_MAX_EXECUTE); + + if (buf_cpy) + ixgbe_free(hw, buf_cpy); + + return status; +} + +/** + * ixgbe_aci_check_event_pending - check if there are any pending events + * @hw: pointer to the HW struct + * + * Determine if there are any pending events. + * + * Return: true if there are any currently pending events + * otherwise false. + */ +bool ixgbe_aci_check_event_pending(struct ixgbe_hw *hw) +{ + u32 ep_bit_mask; + u32 fwsts; + + ep_bit_mask = hw->bus.func ? GL_FWSTS_EP_PF1 : GL_FWSTS_EP_PF0; + + /* Check state of Event Pending (EP) bit */ + fwsts = IXGBE_READ_REG(hw, GL_FWSTS); + return (fwsts & ep_bit_mask) ? true : false; +} + +/** + * ixgbe_aci_get_event - get an event from ACI + * @hw: pointer to the HW struct + * @e: event information structure + * @pending: optional flag signaling that there are more pending events + * + * Obtain an event from ACI and return its content + * through 'e' using ACI command (0x0014). + * Provide information if there are more events + * to retrieve through 'pending'. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_get_event(struct ixgbe_hw *hw, struct ixgbe_aci_event *e, + bool *pending) +{ + struct ixgbe_aci_desc desc; + s32 status; + + if (!e || (!e->msg_buf && e->buf_len) || (e->msg_buf && !e->buf_len)) + return IXGBE_ERR_PARAM; + + ixgbe_acquire_lock(&hw->aci.lock); + + /* Check if there are any events pending */ + if (!ixgbe_aci_check_event_pending(hw)) { + status = IXGBE_ERR_ACI_NO_EVENTS; + goto aci_get_event_exit; + } + + /* Obtain pending event */ + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_get_fw_event); + status = ixgbe_aci_send_cmd_execute(hw, &desc, e->msg_buf, e->buf_len); + if (status) + goto aci_get_event_exit; + + /* Returned 0x0014 opcode indicates that no event was obtained */ + if (desc.opcode == IXGBE_CPU_TO_LE16(ixgbe_aci_opc_get_fw_event)) { + status = IXGBE_ERR_ACI_NO_EVENTS; + goto aci_get_event_exit; + } + + /* Determine size of event data */ + e->msg_len = MIN_T(u16, IXGBE_LE16_TO_CPU(desc.datalen), e->buf_len); + /* Write event descriptor to event info structure */ + memcpy(&e->desc, &desc, sizeof(e->desc)); + + /* Check if there are any further events pending */ + if (pending) { + *pending = ixgbe_aci_check_event_pending(hw); + } + +aci_get_event_exit: + ixgbe_release_lock(&hw->aci.lock); + + return status; +} + +/** + * ixgbe_fill_dflt_direct_cmd_desc - fill ACI descriptor with default values. + * @desc: pointer to the temp descriptor (non DMA mem) + * @opcode: the opcode can be used to decide which flags to turn off or on + * + * Helper function to fill the descriptor desc with default values + * and the provided opcode. + */ +void ixgbe_fill_dflt_direct_cmd_desc(struct ixgbe_aci_desc *desc, u16 opcode) +{ + /* zero out the desc */ + memset(desc, 0, sizeof(*desc)); + desc->opcode = IXGBE_CPU_TO_LE16(opcode); + desc->flags = IXGBE_CPU_TO_LE16(IXGBE_ACI_FLAG_SI); +} + +/** + * ixgbe_aci_get_fw_ver - get the firmware version + * @hw: pointer to the HW struct + * + * Get the firmware version using ACI command (0x0001). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_get_fw_ver(struct ixgbe_hw *hw) +{ + struct ixgbe_aci_cmd_get_ver *resp; + struct ixgbe_aci_desc desc; + s32 status; + + resp = &desc.params.get_ver; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_get_ver); + + status = ixgbe_aci_send_cmd(hw, &desc, NULL, 0); + + if (!status) { + hw->fw_branch = resp->fw_branch; + hw->fw_maj_ver = resp->fw_major; + hw->fw_min_ver = resp->fw_minor; + hw->fw_patch = resp->fw_patch; + hw->fw_build = IXGBE_LE32_TO_CPU(resp->fw_build); + hw->api_branch = resp->api_branch; + hw->api_maj_ver = resp->api_major; + hw->api_min_ver = resp->api_minor; + hw->api_patch = resp->api_patch; + } + + return status; +} + +/** + * ixgbe_aci_send_driver_ver - send the driver version to firmware + * @hw: pointer to the HW struct + * @dv: driver's major, minor version + * + * Send the driver version to the firmware + * using the ACI command (0x0002). + * + * Return: the exit code of the operation. + * Returns IXGBE_ERR_PARAM, if dv is NULL. + */ +s32 ixgbe_aci_send_driver_ver(struct ixgbe_hw *hw, struct ixgbe_driver_ver *dv) +{ + struct ixgbe_aci_cmd_driver_ver *cmd; + struct ixgbe_aci_desc desc; + u16 len; + + cmd = &desc.params.driver_ver; + + if (!dv) + return IXGBE_ERR_PARAM; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_driver_ver); + + desc.flags |= IXGBE_CPU_TO_LE16(IXGBE_ACI_FLAG_RD); + cmd->major_ver = dv->major_ver; + cmd->minor_ver = dv->minor_ver; + cmd->build_ver = dv->build_ver; + cmd->subbuild_ver = dv->subbuild_ver; + + len = 0; + while (len < sizeof(dv->driver_string) && + IS_ASCII(dv->driver_string[len]) && dv->driver_string[len]) + len++; + + return ixgbe_aci_send_cmd(hw, &desc, dv->driver_string, len); +} + +/** + * ixgbe_aci_req_res - request a common resource + * @hw: pointer to the HW struct + * @res: resource ID + * @access: access type + * @sdp_number: resource number + * @timeout: the maximum time in ms that the driver may hold the resource + * + * Requests a common resource using the ACI command (0x0008). + * Specifies the maximum time the driver may hold the resource. + * If the requested resource is currently occupied by some other driver, + * a busy return value is returned and the timeout field value indicates the + * maximum time the current owner has to free it. + * + * Return: the exit code of the operation. + */ +static s32 +ixgbe_aci_req_res(struct ixgbe_hw *hw, enum ixgbe_aci_res_ids res, + enum ixgbe_aci_res_access_type access, u8 sdp_number, + u32 *timeout) +{ + struct ixgbe_aci_cmd_req_res *cmd_resp; + struct ixgbe_aci_desc desc; + s32 status; + + cmd_resp = &desc.params.res_owner; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_req_res); + + cmd_resp->res_id = IXGBE_CPU_TO_LE16(res); + cmd_resp->access_type = IXGBE_CPU_TO_LE16(access); + cmd_resp->res_number = IXGBE_CPU_TO_LE32(sdp_number); + cmd_resp->timeout = IXGBE_CPU_TO_LE32(*timeout); + *timeout = 0; + + status = ixgbe_aci_send_cmd(hw, &desc, NULL, 0); + + /* The completion specifies the maximum time in ms that the driver + * may hold the resource in the Timeout field. + */ + + /* If the resource is held by some other driver, the command completes + * with a busy return value and the timeout field indicates the maximum + * time the current owner of the resource has to free it. + */ + if (!status || hw->aci.last_status == IXGBE_ACI_RC_EBUSY) + *timeout = IXGBE_LE32_TO_CPU(cmd_resp->timeout); + + return status; +} + +/** + * ixgbe_aci_release_res - release a common resource using ACI + * @hw: pointer to the HW struct + * @res: resource ID + * @sdp_number: resource number + * + * Release a common resource using ACI command (0x0009). + * + * Return: the exit code of the operation. + */ +static s32 +ixgbe_aci_release_res(struct ixgbe_hw *hw, enum ixgbe_aci_res_ids res, + u8 sdp_number) +{ + struct ixgbe_aci_cmd_req_res *cmd; + struct ixgbe_aci_desc desc; + + cmd = &desc.params.res_owner; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_release_res); + + cmd->res_id = IXGBE_CPU_TO_LE16(res); + cmd->res_number = IXGBE_CPU_TO_LE32(sdp_number); + + return ixgbe_aci_send_cmd(hw, &desc, NULL, 0); +} + +/** + * ixgbe_acquire_res - acquire the ownership of a resource + * @hw: pointer to the HW structure + * @res: resource ID + * @access: access type (read or write) + * @timeout: timeout in milliseconds + * + * Make an attempt to acquire the ownership of a resource using + * the ixgbe_aci_req_res to utilize ACI. + * In case if some other driver has previously acquired the resource and + * performed any necessary updates, the IXGBE_ERR_ACI_NO_WORK is returned, + * and the caller does not obtain the resource and has no further work to do. + * If needed, the function will poll until the current lock owner timeouts. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_acquire_res(struct ixgbe_hw *hw, enum ixgbe_aci_res_ids res, + enum ixgbe_aci_res_access_type access, u32 timeout) +{ +#define IXGBE_RES_POLLING_DELAY_MS 10 + u32 delay = IXGBE_RES_POLLING_DELAY_MS; + u32 res_timeout = timeout; + u32 retry_timeout = 0; + s32 status; + + status = ixgbe_aci_req_res(hw, res, access, 0, &res_timeout); + + /* A return code of IXGBE_ERR_ACI_NO_WORK means that another driver has + * previously acquired the resource and performed any necessary updates; + * in this case the caller does not obtain the resource and has no + * further work to do. + */ + if (status == IXGBE_ERR_ACI_NO_WORK) + goto ixgbe_acquire_res_exit; + + /* If necessary, poll until the current lock owner timeouts. + * Set retry_timeout to the timeout value reported by the FW in the + * response to the "Request Resource Ownership" (0x0008) Admin Command + * as it indicates the maximum time the current owner of the resource + * is allowed to hold it. + */ + retry_timeout = res_timeout; + while (status && retry_timeout && res_timeout) { + msec_delay(delay); + retry_timeout = (retry_timeout > delay) ? + retry_timeout - delay : 0; + status = ixgbe_aci_req_res(hw, res, access, 0, &res_timeout); + + if (status == IXGBE_ERR_ACI_NO_WORK) + /* lock free, but no work to do */ + break; + + if (!status) + /* lock acquired */ + break; + } + +ixgbe_acquire_res_exit: + return status; +} + +/** + * ixgbe_release_res - release a common resource + * @hw: pointer to the HW structure + * @res: resource ID + * + * Release a common resource using ixgbe_aci_release_res. + */ +void ixgbe_release_res(struct ixgbe_hw *hw, enum ixgbe_aci_res_ids res) +{ + u32 total_delay = 0; + s32 status; + + status = ixgbe_aci_release_res(hw, res, 0); + + /* There are some rare cases when trying to release the resource + * results in an admin command timeout, so handle them correctly. + */ + while ((status == IXGBE_ERR_ACI_TIMEOUT) && + (total_delay < IXGBE_ACI_RELEASE_RES_TIMEOUT)) { + msec_delay(1); + status = ixgbe_aci_release_res(hw, res, 0); + total_delay++; + } +} + +/** + * ixgbe_parse_common_caps - Parse common device/function capabilities + * @hw: pointer to the HW struct + * @caps: pointer to common capabilities structure + * @elem: the capability element to parse + * @prefix: message prefix for tracing capabilities + * + * Given a capability element, extract relevant details into the common + * capability structure. + * + * Return: true if the capability matches one of the common capability ids, + * false otherwise. + */ +static bool +ixgbe_parse_common_caps(struct ixgbe_hw *hw, struct ixgbe_hw_common_caps *caps, + struct ixgbe_aci_cmd_list_caps_elem *elem, + const char *prefix) +{ + u32 logical_id = IXGBE_LE32_TO_CPU(elem->logical_id); + u32 phys_id = IXGBE_LE32_TO_CPU(elem->phys_id); + u32 number = IXGBE_LE32_TO_CPU(elem->number); + u16 cap = IXGBE_LE16_TO_CPU(elem->cap); + bool found = true; + + UNREFERENCED_1PARAMETER(hw); + + switch (cap) { + case IXGBE_ACI_CAPS_VALID_FUNCTIONS: + caps->valid_functions = number; + break; + case IXGBE_ACI_CAPS_VMDQ: + caps->vmdq = (number == 1); + break; + case IXGBE_ACI_CAPS_DCB: + caps->dcb = (number == 1); + caps->active_tc_bitmap = logical_id; + caps->maxtc = phys_id; + break; + case IXGBE_ACI_CAPS_RSS: + caps->rss_table_size = number; + caps->rss_table_entry_width = logical_id; + break; + case IXGBE_ACI_CAPS_RXQS: + caps->num_rxq = number; + caps->rxq_first_id = phys_id; + break; + case IXGBE_ACI_CAPS_TXQS: + caps->num_txq = number; + caps->txq_first_id = phys_id; + break; + case IXGBE_ACI_CAPS_MSIX: + caps->num_msix_vectors = number; + caps->msix_vector_first_id = phys_id; + break; + case IXGBE_ACI_CAPS_NVM_VER: + break; + case IXGBE_ACI_CAPS_NVM_MGMT: + caps->sec_rev_disabled = + (number & IXGBE_NVM_MGMT_SEC_REV_DISABLED) ? + true : false; + caps->update_disabled = + (number & IXGBE_NVM_MGMT_UPDATE_DISABLED) ? + true : false; + caps->nvm_unified_update = + (number & IXGBE_NVM_MGMT_UNIFIED_UPD_SUPPORT) ? + true : false; + caps->netlist_auth = + (number & IXGBE_NVM_MGMT_NETLIST_AUTH_SUPPORT) ? + true : false; + break; + case IXGBE_ACI_CAPS_MAX_MTU: + caps->max_mtu = number; + break; + case IXGBE_ACI_CAPS_PCIE_RESET_AVOIDANCE: + caps->pcie_reset_avoidance = (number > 0); + break; + case IXGBE_ACI_CAPS_POST_UPDATE_RESET_RESTRICT: + caps->reset_restrict_support = (number == 1); + break; + case IXGBE_ACI_CAPS_EXT_TOPO_DEV_IMG0: + case IXGBE_ACI_CAPS_EXT_TOPO_DEV_IMG1: + case IXGBE_ACI_CAPS_EXT_TOPO_DEV_IMG2: + case IXGBE_ACI_CAPS_EXT_TOPO_DEV_IMG3: + { + u8 index = cap - IXGBE_ACI_CAPS_EXT_TOPO_DEV_IMG0; + + caps->ext_topo_dev_img_ver_high[index] = number; + caps->ext_topo_dev_img_ver_low[index] = logical_id; + caps->ext_topo_dev_img_part_num[index] = + (phys_id & IXGBE_EXT_TOPO_DEV_IMG_PART_NUM_M) >> + IXGBE_EXT_TOPO_DEV_IMG_PART_NUM_S; + caps->ext_topo_dev_img_load_en[index] = + (phys_id & IXGBE_EXT_TOPO_DEV_IMG_LOAD_EN) != 0; + caps->ext_topo_dev_img_prog_en[index] = + (phys_id & IXGBE_EXT_TOPO_DEV_IMG_PROG_EN) != 0; + break; + } + + case IXGBE_ACI_CAPS_NEXT_CLUSTER_ID: + caps->next_cluster_id_support = (number == 1); + DEBUGOUT2("%s: next_cluster_id_support = %d\n", + prefix, caps->next_cluster_id_support); + break; + default: + /* Not one of the recognized common capabilities */ + found = false; + } + + return found; +} + +/** + * ixgbe_hweight8 - count set bits among the 8 lowest bits + * @w: variable storing set bits to count + * + * Return: the number of set bits among the 8 lowest bits in the provided value. + */ +static u8 ixgbe_hweight8(u32 w) +{ + u8 hweight = 0, i; + + for (i = 0; i < 8; i++) + if (w & (1 << i)) + hweight++; + + return hweight; +} + +/** + * ixgbe_hweight32 - count set bits among the 32 lowest bits + * @w: variable storing set bits to count + * + * Return: the number of set bits among the 32 lowest bits in the + * provided value. + */ +static u8 ixgbe_hweight32(u32 w) +{ + u32 bitMask = 0x1, i; + u8 bitCnt = 0; + + for (i = 0; i < 32; i++) + { + if (w & bitMask) + bitCnt++; + + bitMask = bitMask << 0x1; + } + + return bitCnt; +} + +/** + * ixgbe_func_id_to_logical_id - map from function id to logical pf id + * @active_function_bitmap: active function bitmap + * @pf_id: function number of device + * + * Return: the logical id of a function mapped by the provided pf_id. + */ +static int ixgbe_func_id_to_logical_id(u32 active_function_bitmap, u8 pf_id) +{ + u8 logical_id = 0; + u8 i; + + for (i = 0; i < pf_id; i++) + if (active_function_bitmap & BIT(i)) + logical_id++; + + return logical_id; +} + +/** + * ixgbe_parse_valid_functions_cap - Parse IXGBE_ACI_CAPS_VALID_FUNCTIONS caps + * @hw: pointer to the HW struct + * @dev_p: pointer to device capabilities structure + * @cap: capability element to parse + * + * Parse IXGBE_ACI_CAPS_VALID_FUNCTIONS for device capabilities. + */ +static void +ixgbe_parse_valid_functions_cap(struct ixgbe_hw *hw, + struct ixgbe_hw_dev_caps *dev_p, + struct ixgbe_aci_cmd_list_caps_elem *cap) +{ + u32 number = IXGBE_LE32_TO_CPU(cap->number); + + UNREFERENCED_1PARAMETER(hw); + + dev_p->num_funcs = ixgbe_hweight32(number); + + hw->logical_pf_id = ixgbe_func_id_to_logical_id(number, hw->pf_id); +} + +/** + * ixgbe_parse_vsi_dev_caps - Parse IXGBE_ACI_CAPS_VSI device caps + * @hw: pointer to the HW struct + * @dev_p: pointer to device capabilities structure + * @cap: capability element to parse + * + * Parse IXGBE_ACI_CAPS_VSI for device capabilities. + */ +static void ixgbe_parse_vsi_dev_caps(struct ixgbe_hw *hw, + struct ixgbe_hw_dev_caps *dev_p, + struct ixgbe_aci_cmd_list_caps_elem *cap) +{ + u32 number = IXGBE_LE32_TO_CPU(cap->number); + + UNREFERENCED_1PARAMETER(hw); + + dev_p->num_vsi_allocd_to_host = number; +} + +/** + * ixgbe_parse_1588_dev_caps - Parse IXGBE_ACI_CAPS_1588 device caps + * @hw: pointer to the HW struct + * @dev_p: pointer to device capabilities structure + * @cap: capability element to parse + * + * Parse IXGBE_ACI_CAPS_1588 for device capabilities. + */ +static void ixgbe_parse_1588_dev_caps(struct ixgbe_hw *hw, + struct ixgbe_hw_dev_caps *dev_p, + struct ixgbe_aci_cmd_list_caps_elem *cap) +{ + struct ixgbe_ts_dev_info *info = &dev_p->ts_dev_info; + u32 logical_id = IXGBE_LE32_TO_CPU(cap->logical_id); + u32 phys_id = IXGBE_LE32_TO_CPU(cap->phys_id); + u32 number = IXGBE_LE32_TO_CPU(cap->number); + + UNREFERENCED_1PARAMETER(hw); + + info->ena = ((number & IXGBE_TS_DEV_ENA_M) != 0); + dev_p->common_cap.ieee_1588 = info->ena; + + info->tmr0_owner = number & IXGBE_TS_TMR0_OWNR_M; + info->tmr0_owned = ((number & IXGBE_TS_TMR0_OWND_M) != 0); + info->tmr0_ena = ((number & IXGBE_TS_TMR0_ENA_M) != 0); + + info->tmr1_owner = (number & IXGBE_TS_TMR1_OWNR_M) >> + IXGBE_TS_TMR1_OWNR_S; + info->tmr1_owned = ((number & IXGBE_TS_TMR1_OWND_M) != 0); + info->tmr1_ena = ((number & IXGBE_TS_TMR1_ENA_M) != 0); + + info->ena_ports = logical_id; + info->tmr_own_map = phys_id; + +} + +/** + * ixgbe_parse_fdir_dev_caps - Parse IXGBE_ACI_CAPS_FD device caps + * @hw: pointer to the HW struct + * @dev_p: pointer to device capabilities structure + * @cap: capability element to parse + * + * Parse IXGBE_ACI_CAPS_FD for device capabilities. + */ +static void ixgbe_parse_fdir_dev_caps(struct ixgbe_hw *hw, + struct ixgbe_hw_dev_caps *dev_p, + struct ixgbe_aci_cmd_list_caps_elem *cap) +{ + u32 number = IXGBE_LE32_TO_CPU(cap->number); + + UNREFERENCED_1PARAMETER(hw); + + dev_p->num_flow_director_fltr = number; +} + +/** + * ixgbe_parse_dev_caps - Parse device capabilities + * @hw: pointer to the HW struct + * @dev_p: pointer to device capabilities structure + * @buf: buffer containing the device capability records + * @cap_count: the number of capabilities + * + * Helper device to parse device (0x000B) capabilities list. For + * capabilities shared between device and function, this relies on + * ixgbe_parse_common_caps. + * + * Loop through the list of provided capabilities and extract the relevant + * data into the device capabilities structured. + */ +static void ixgbe_parse_dev_caps(struct ixgbe_hw *hw, + struct ixgbe_hw_dev_caps *dev_p, + void *buf, u32 cap_count) +{ + struct ixgbe_aci_cmd_list_caps_elem *cap_resp; + u32 i; + + cap_resp = (struct ixgbe_aci_cmd_list_caps_elem *)buf; + + memset(dev_p, 0, sizeof(*dev_p)); + + for (i = 0; i < cap_count; i++) { + u16 cap = IXGBE_LE16_TO_CPU(cap_resp[i].cap); + bool found; + + found = ixgbe_parse_common_caps(hw, &dev_p->common_cap, + &cap_resp[i], "dev caps"); + + switch (cap) { + case IXGBE_ACI_CAPS_VALID_FUNCTIONS: + ixgbe_parse_valid_functions_cap(hw, dev_p, + &cap_resp[i]); + break; + case IXGBE_ACI_CAPS_VSI: + ixgbe_parse_vsi_dev_caps(hw, dev_p, &cap_resp[i]); + break; + case IXGBE_ACI_CAPS_1588: + ixgbe_parse_1588_dev_caps(hw, dev_p, &cap_resp[i]); + break; + case IXGBE_ACI_CAPS_FD: + ixgbe_parse_fdir_dev_caps(hw, dev_p, &cap_resp[i]); + break; + default: + /* Don't list common capabilities as unknown */ + if (!found) + break; + } + } + +} + +/** + * ixgbe_get_num_per_func - determine number of resources per PF + * @hw: pointer to the HW structure + * @max: value to be evenly split between each PF + * + * Determine the number of valid functions by going through the bitmap returned + * from parsing capabilities and use this to calculate the number of resources + * per PF based on the max value passed in. + * + * Return: the number of resources per PF or 0, if no PH are available. + */ +static u32 ixgbe_get_num_per_func(struct ixgbe_hw *hw, u32 max) +{ + u8 funcs; + +#define IXGBE_CAPS_VALID_FUNCS_M 0xFF + funcs = ixgbe_hweight8(hw->dev_caps.common_cap.valid_functions & + IXGBE_CAPS_VALID_FUNCS_M); + + if (!funcs) + return 0; + + return max / funcs; +} + +/** + * ixgbe_parse_vsi_func_caps - Parse IXGBE_ACI_CAPS_VSI function caps + * @hw: pointer to the HW struct + * @func_p: pointer to function capabilities structure + * @cap: pointer to the capability element to parse + * + * Extract function capabilities for IXGBE_ACI_CAPS_VSI. + */ +static void ixgbe_parse_vsi_func_caps(struct ixgbe_hw *hw, + struct ixgbe_hw_func_caps *func_p, + struct ixgbe_aci_cmd_list_caps_elem *cap) +{ + func_p->guar_num_vsi = ixgbe_get_num_per_func(hw, IXGBE_MAX_VSI); +} + +/** + * ixgbe_parse_1588_func_caps - Parse IXGBE_ACI_CAPS_1588 function caps + * @hw: pointer to the HW struct + * @func_p: pointer to function capabilities structure + * @cap: pointer to the capability element to parse + * + * Extract function capabilities for IXGBE_ACI_CAPS_1588. + */ +static void ixgbe_parse_1588_func_caps(struct ixgbe_hw *hw, + struct ixgbe_hw_func_caps *func_p, + struct ixgbe_aci_cmd_list_caps_elem *cap) +{ + struct ixgbe_ts_func_info *info = &func_p->ts_func_info; + u32 number = IXGBE_LE32_TO_CPU(cap->number); + + UNREFERENCED_1PARAMETER(hw); + + info->ena = ((number & IXGBE_TS_FUNC_ENA_M) != 0); + func_p->common_cap.ieee_1588 = info->ena; + + info->src_tmr_owned = ((number & IXGBE_TS_SRC_TMR_OWND_M) != 0); + info->tmr_ena = ((number & IXGBE_TS_TMR_ENA_M) != 0); + info->tmr_index_owned = ((number & IXGBE_TS_TMR_IDX_OWND_M) != 0); + info->tmr_index_assoc = ((number & IXGBE_TS_TMR_IDX_ASSOC_M) != 0); + + info->clk_freq = (number & IXGBE_TS_CLK_FREQ_M) >> IXGBE_TS_CLK_FREQ_S; + info->clk_src = ((number & IXGBE_TS_CLK_SRC_M) != 0); + + if (info->clk_freq < NUM_IXGBE_TIME_REF_FREQ) { + info->time_ref = (enum ixgbe_time_ref_freq)info->clk_freq; + } else { + /* Unknown clock frequency, so assume a (probably incorrect) + * default to avoid out-of-bounds look ups of frequency + * related information. + */ + info->time_ref = IXGBE_TIME_REF_FREQ_25_000; + } + +} +/** + * ixgbe_parse_func_caps - Parse function capabilities + * @hw: pointer to the HW struct + * @func_p: pointer to function capabilities structure + * @buf: buffer containing the function capability records + * @cap_count: the number of capabilities + * + * Helper function to parse function (0x000A) capabilities list. For + * capabilities shared between device and function, this relies on + * ixgbe_parse_common_caps. + * + * Loop through the list of provided capabilities and extract the relevant + * data into the function capabilities structured. + */ +static void ixgbe_parse_func_caps(struct ixgbe_hw *hw, + struct ixgbe_hw_func_caps *func_p, + void *buf, u32 cap_count) +{ + struct ixgbe_aci_cmd_list_caps_elem *cap_resp; + u32 i; + + cap_resp = (struct ixgbe_aci_cmd_list_caps_elem *)buf; + + memset(func_p, 0, sizeof(*func_p)); + + for (i = 0; i < cap_count; i++) { + u16 cap = IXGBE_LE16_TO_CPU(cap_resp[i].cap); + + ixgbe_parse_common_caps(hw, &func_p->common_cap, + &cap_resp[i], "func caps"); + + switch (cap) { + case IXGBE_ACI_CAPS_VSI: + ixgbe_parse_vsi_func_caps(hw, func_p, &cap_resp[i]); + break; + case IXGBE_ACI_CAPS_1588: + ixgbe_parse_1588_func_caps(hw, func_p, &cap_resp[i]); + break; + default: + /* Don't list common capabilities as unknown */ + break; + } + } + +} + +/** + * ixgbe_aci_list_caps - query function/device capabilities + * @hw: pointer to the HW struct + * @buf: a buffer to hold the capabilities + * @buf_size: size of the buffer + * @cap_count: if not NULL, set to the number of capabilities reported + * @opc: capabilities type to discover, device or function + * + * Get the function (0x000A) or device (0x000B) capabilities description from + * firmware and store it in the buffer. + * + * If the cap_count pointer is not NULL, then it is set to the number of + * capabilities firmware will report. Note that if the buffer size is too + * small, it is possible the command will return IXGBE_ERR_OUT_OF_MEM. The + * cap_count will still be updated in this case. It is recommended that the + * buffer size be set to IXGBE_ACI_MAX_BUFFER_SIZE (the largest possible + * buffer that firmware could return) to avoid this. + * + * Return: the exit code of the operation. + * Exit code of IXGBE_ERR_OUT_OF_MEM means the buffer size is too small. + */ +s32 ixgbe_aci_list_caps(struct ixgbe_hw *hw, void *buf, u16 buf_size, + u32 *cap_count, enum ixgbe_aci_opc opc) +{ + struct ixgbe_aci_cmd_list_caps *cmd; + struct ixgbe_aci_desc desc; + s32 status; + + cmd = &desc.params.get_cap; + + if (opc != ixgbe_aci_opc_list_func_caps && + opc != ixgbe_aci_opc_list_dev_caps) + return IXGBE_ERR_PARAM; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, opc); + status = ixgbe_aci_send_cmd(hw, &desc, buf, buf_size); + + if (cap_count) + *cap_count = IXGBE_LE32_TO_CPU(cmd->count); + + return status; +} + +/** + * ixgbe_discover_dev_caps - Read and extract device capabilities + * @hw: pointer to the hardware structure + * @dev_caps: pointer to device capabilities structure + * + * Read the device capabilities and extract them into the dev_caps structure + * for later use. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_discover_dev_caps(struct ixgbe_hw *hw, + struct ixgbe_hw_dev_caps *dev_caps) +{ + u32 status, cap_count = 0; + u8 *cbuf = NULL; + + cbuf = (u8*)ixgbe_malloc(hw, IXGBE_ACI_MAX_BUFFER_SIZE); + if (!cbuf) + return IXGBE_ERR_OUT_OF_MEM; + /* Although the driver doesn't know the number of capabilities the + * device will return, we can simply send a 4KB buffer, the maximum + * possible size that firmware can return. + */ + cap_count = IXGBE_ACI_MAX_BUFFER_SIZE / + sizeof(struct ixgbe_aci_cmd_list_caps_elem); + + status = ixgbe_aci_list_caps(hw, cbuf, IXGBE_ACI_MAX_BUFFER_SIZE, + &cap_count, + ixgbe_aci_opc_list_dev_caps); + if (!status) + ixgbe_parse_dev_caps(hw, dev_caps, cbuf, cap_count); + + if (cbuf) + ixgbe_free(hw, cbuf); + + return status; +} + +/** + * ixgbe_discover_func_caps - Read and extract function capabilities + * @hw: pointer to the hardware structure + * @func_caps: pointer to function capabilities structure + * + * Read the function capabilities and extract them into the func_caps structure + * for later use. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_discover_func_caps(struct ixgbe_hw *hw, + struct ixgbe_hw_func_caps *func_caps) +{ + u32 cap_count = 0; + u8 *cbuf = NULL; + s32 status; + + cbuf = (u8*)ixgbe_malloc(hw, IXGBE_ACI_MAX_BUFFER_SIZE); + if(!cbuf) + return IXGBE_ERR_OUT_OF_MEM; + /* Although the driver doesn't know the number of capabilities the + * device will return, we can simply send a 4KB buffer, the maximum + * possible size that firmware can return. + */ + cap_count = IXGBE_ACI_MAX_BUFFER_SIZE / + sizeof(struct ixgbe_aci_cmd_list_caps_elem); + + status = ixgbe_aci_list_caps(hw, cbuf, IXGBE_ACI_MAX_BUFFER_SIZE, + &cap_count, + ixgbe_aci_opc_list_func_caps); + if (!status) + ixgbe_parse_func_caps(hw, func_caps, cbuf, cap_count); + + if (cbuf) + ixgbe_free(hw, cbuf); + + return status; +} + +/** + * ixgbe_get_caps - get info about the HW + * @hw: pointer to the hardware structure + * + * Retrieve both device and function capabilities. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_get_caps(struct ixgbe_hw *hw) +{ + s32 status; + + status = ixgbe_discover_dev_caps(hw, &hw->dev_caps); + if (status) + return status; + + return ixgbe_discover_func_caps(hw, &hw->func_caps); +} + +/** + * ixgbe_aci_disable_rxen - disable RX + * @hw: pointer to the HW struct + * + * Request a safe disable of Receive Enable using ACI command (0x000C). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_disable_rxen(struct ixgbe_hw *hw) +{ + struct ixgbe_aci_cmd_disable_rxen *cmd; + struct ixgbe_aci_desc desc; + + UNREFERENCED_1PARAMETER(hw); + + cmd = &desc.params.disable_rxen; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_disable_rxen); + + cmd->lport_num = (u8)hw->bus.func; + + return ixgbe_aci_send_cmd(hw, &desc, NULL, 0); +} + +/** + * ixgbe_aci_get_phy_caps - returns PHY capabilities + * @hw: pointer to the HW struct + * @qual_mods: report qualified modules + * @report_mode: report mode capabilities + * @pcaps: structure for PHY capabilities to be filled + * + * Returns the various PHY capabilities supported on the Port + * using ACI command (0x0600). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_get_phy_caps(struct ixgbe_hw *hw, bool qual_mods, u8 report_mode, + struct ixgbe_aci_cmd_get_phy_caps_data *pcaps) +{ + struct ixgbe_aci_cmd_get_phy_caps *cmd; + u16 pcaps_size = sizeof(*pcaps); + struct ixgbe_aci_desc desc; + s32 status; + + cmd = &desc.params.get_phy; + + if (!pcaps || (report_mode & ~IXGBE_ACI_REPORT_MODE_M)) + return IXGBE_ERR_PARAM; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_get_phy_caps); + + if (qual_mods) + cmd->param0 |= IXGBE_CPU_TO_LE16(IXGBE_ACI_GET_PHY_RQM); + + cmd->param0 |= IXGBE_CPU_TO_LE16(report_mode); + status = ixgbe_aci_send_cmd(hw, &desc, pcaps, pcaps_size); + + if (status == IXGBE_SUCCESS && + report_mode == IXGBE_ACI_REPORT_TOPO_CAP_MEDIA) { + hw->phy.phy_type_low = IXGBE_LE64_TO_CPU(pcaps->phy_type_low); + hw->phy.phy_type_high = IXGBE_LE64_TO_CPU(pcaps->phy_type_high); + memcpy(hw->link.link_info.module_type, &pcaps->module_type, + sizeof(hw->link.link_info.module_type)); + } + + return status; +} + +/** + * ixgbe_phy_caps_equals_cfg - check if capabilities match the PHY config + * @phy_caps: PHY capabilities + * @phy_cfg: PHY configuration + * + * Helper function to determine if PHY capabilities match PHY + * configuration + * + * Return: true if PHY capabilities match PHY configuration. + */ +bool +ixgbe_phy_caps_equals_cfg(struct ixgbe_aci_cmd_get_phy_caps_data *phy_caps, + struct ixgbe_aci_cmd_set_phy_cfg_data *phy_cfg) +{ + u8 caps_mask, cfg_mask; + + if (!phy_caps || !phy_cfg) + return false; + + /* These bits are not common between capabilities and configuration. + * Do not use them to determine equality. + */ + caps_mask = IXGBE_ACI_PHY_CAPS_MASK & ~(IXGBE_ACI_PHY_AN_MODE | + IXGBE_ACI_PHY_EN_MOD_QUAL); + cfg_mask = IXGBE_ACI_PHY_ENA_VALID_MASK & + ~IXGBE_ACI_PHY_ENA_AUTO_LINK_UPDT; + + if (phy_caps->phy_type_low != phy_cfg->phy_type_low || + phy_caps->phy_type_high != phy_cfg->phy_type_high || + ((phy_caps->caps & caps_mask) != (phy_cfg->caps & cfg_mask)) || + phy_caps->low_power_ctrl_an != phy_cfg->low_power_ctrl_an || + phy_caps->eee_cap != phy_cfg->eee_cap || + phy_caps->eeer_value != phy_cfg->eeer_value || + phy_caps->link_fec_options != phy_cfg->link_fec_opt) + return false; + + return true; +} + +/** + * ixgbe_copy_phy_caps_to_cfg - Copy PHY ability data to configuration data + * @caps: PHY ability structure to copy data from + * @cfg: PHY configuration structure to copy data to + * + * Helper function to copy data from PHY capabilities data structure + * to PHY configuration data structure + */ +void ixgbe_copy_phy_caps_to_cfg(struct ixgbe_aci_cmd_get_phy_caps_data *caps, + struct ixgbe_aci_cmd_set_phy_cfg_data *cfg) +{ + if (!caps || !cfg) + return; + + memset(cfg, 0, sizeof(*cfg)); + cfg->phy_type_low = caps->phy_type_low; + cfg->phy_type_high = caps->phy_type_high; + cfg->caps = caps->caps; + cfg->low_power_ctrl_an = caps->low_power_ctrl_an; + cfg->eee_cap = caps->eee_cap; + cfg->eeer_value = caps->eeer_value; + cfg->link_fec_opt = caps->link_fec_options; + cfg->module_compliance_enforcement = + caps->module_compliance_enforcement; +} + +/** + * ixgbe_aci_set_phy_cfg - set PHY configuration + * @hw: pointer to the HW struct + * @cfg: structure with PHY configuration data to be set + * + * Set the various PHY configuration parameters supported on the Port + * using ACI command (0x0601). + * One or more of the Set PHY config parameters may be ignored in an MFP + * mode as the PF may not have the privilege to set some of the PHY Config + * parameters. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_set_phy_cfg(struct ixgbe_hw *hw, + struct ixgbe_aci_cmd_set_phy_cfg_data *cfg) +{ + struct ixgbe_aci_desc desc; + s32 status; + + if (!cfg) + return IXGBE_ERR_PARAM; + + /* Ensure that only valid bits of cfg->caps can be turned on. */ + if (cfg->caps & ~IXGBE_ACI_PHY_ENA_VALID_MASK) { + cfg->caps &= IXGBE_ACI_PHY_ENA_VALID_MASK; + } + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_set_phy_cfg); + desc.flags |= IXGBE_CPU_TO_LE16(IXGBE_ACI_FLAG_RD); + + status = ixgbe_aci_send_cmd(hw, &desc, cfg, sizeof(*cfg)); + + if (!status) + hw->phy.curr_user_phy_cfg = *cfg; + + return status; +} + +/** + * ixgbe_aci_set_link_restart_an - set up link and restart AN + * @hw: pointer to the HW struct + * @ena_link: if true: enable link, if false: disable link + * + * Function sets up the link and restarts the Auto-Negotiation over the link. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_set_link_restart_an(struct ixgbe_hw *hw, bool ena_link) +{ + struct ixgbe_aci_cmd_restart_an *cmd; + struct ixgbe_aci_desc desc; + + cmd = &desc.params.restart_an; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_restart_an); + + cmd->cmd_flags = IXGBE_ACI_RESTART_AN_LINK_RESTART; + if (ena_link) + cmd->cmd_flags |= IXGBE_ACI_RESTART_AN_LINK_ENABLE; + else + cmd->cmd_flags &= ~IXGBE_ACI_RESTART_AN_LINK_ENABLE; + + return ixgbe_aci_send_cmd(hw, &desc, NULL, 0); +} + +/** + * ixgbe_is_media_cage_present - check if media cage is present + * @hw: pointer to the HW struct + * + * Identify presence of media cage using the ACI command (0x06E0). + * + * Return: true if media cage is present, else false. If no cage, then + * media type is backplane or BASE-T. + */ +static bool ixgbe_is_media_cage_present(struct ixgbe_hw *hw) +{ + struct ixgbe_aci_cmd_get_link_topo *cmd; + struct ixgbe_aci_desc desc; + + cmd = &desc.params.get_link_topo; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_get_link_topo); + + cmd->addr.topo_params.node_type_ctx = + (IXGBE_ACI_LINK_TOPO_NODE_CTX_PORT << + IXGBE_ACI_LINK_TOPO_NODE_CTX_S); + + /* set node type */ + cmd->addr.topo_params.node_type_ctx |= + (IXGBE_ACI_LINK_TOPO_NODE_TYPE_M & + IXGBE_ACI_LINK_TOPO_NODE_TYPE_CAGE); + + /* Node type cage can be used to determine if cage is present. If AQC + * returns error (ENOENT), then no cage present. If no cage present then + * connection type is backplane or BASE-T. + */ + return ixgbe_aci_get_netlist_node(hw, cmd, NULL, NULL); +} + +/** + * ixgbe_get_media_type_from_phy_type - Gets media type based on phy type + * @hw: pointer to the HW struct + * + * Try to identify the media type based on the phy type. + * If more than one media type, the ixgbe_media_type_unknown is returned. + * First, phy_type_low is checked, then phy_type_high. + * If none are identified, the ixgbe_media_type_unknown is returned + * + * Return: type of a media based on phy type in form of enum. + */ +static enum ixgbe_media_type +ixgbe_get_media_type_from_phy_type(struct ixgbe_hw *hw) +{ + struct ixgbe_link_status *hw_link_info; + + if (!hw) + return ixgbe_media_type_unknown; + + hw_link_info = &hw->link.link_info; + if (hw_link_info->phy_type_low && hw_link_info->phy_type_high) + /* If more than one media type is selected, report unknown */ + return ixgbe_media_type_unknown; + + if (hw_link_info->phy_type_low) { + /* 1G SGMII is a special case where some DA cable PHYs + * may show this as an option when it really shouldn't + * be since SGMII is meant to be between a MAC and a PHY + * in a backplane. Try to detect this case and handle it + */ + if (hw_link_info->phy_type_low == IXGBE_PHY_TYPE_LOW_1G_SGMII && + (hw_link_info->module_type[IXGBE_ACI_MOD_TYPE_IDENT] == + IXGBE_ACI_MOD_TYPE_BYTE1_SFP_PLUS_CU_ACTIVE || + hw_link_info->module_type[IXGBE_ACI_MOD_TYPE_IDENT] == + IXGBE_ACI_MOD_TYPE_BYTE1_SFP_PLUS_CU_PASSIVE)) + return ixgbe_media_type_da; + + switch (hw_link_info->phy_type_low) { + case IXGBE_PHY_TYPE_LOW_1000BASE_SX: + case IXGBE_PHY_TYPE_LOW_1000BASE_LX: + case IXGBE_PHY_TYPE_LOW_10GBASE_SR: + case IXGBE_PHY_TYPE_LOW_10GBASE_LR: + case IXGBE_PHY_TYPE_LOW_25GBASE_SR: + case IXGBE_PHY_TYPE_LOW_25GBASE_LR: + return ixgbe_media_type_fiber; + case IXGBE_PHY_TYPE_LOW_10G_SFI_AOC_ACC: + case IXGBE_PHY_TYPE_LOW_25G_AUI_AOC_ACC: + return ixgbe_media_type_fiber; + case IXGBE_PHY_TYPE_LOW_100BASE_TX: + case IXGBE_PHY_TYPE_LOW_1000BASE_T: + case IXGBE_PHY_TYPE_LOW_2500BASE_T: + case IXGBE_PHY_TYPE_LOW_5GBASE_T: + case IXGBE_PHY_TYPE_LOW_10GBASE_T: + case IXGBE_PHY_TYPE_LOW_25GBASE_T: + return ixgbe_media_type_copper; + case IXGBE_PHY_TYPE_LOW_10G_SFI_DA: + case IXGBE_PHY_TYPE_LOW_25GBASE_CR: + case IXGBE_PHY_TYPE_LOW_25GBASE_CR_S: + case IXGBE_PHY_TYPE_LOW_25GBASE_CR1: + return ixgbe_media_type_da; + case IXGBE_PHY_TYPE_LOW_25G_AUI_C2C: + if (ixgbe_is_media_cage_present(hw)) + return ixgbe_media_type_aui; + return ixgbe_media_type_backplane; + case IXGBE_PHY_TYPE_LOW_1000BASE_KX: + case IXGBE_PHY_TYPE_LOW_2500BASE_KX: + case IXGBE_PHY_TYPE_LOW_2500BASE_X: + case IXGBE_PHY_TYPE_LOW_5GBASE_KR: + case IXGBE_PHY_TYPE_LOW_10GBASE_KR_CR1: + case IXGBE_PHY_TYPE_LOW_10G_SFI_C2C: + case IXGBE_PHY_TYPE_LOW_25GBASE_KR: + case IXGBE_PHY_TYPE_LOW_25GBASE_KR1: + case IXGBE_PHY_TYPE_LOW_25GBASE_KR_S: + return ixgbe_media_type_backplane; + } + } else { + switch (hw_link_info->phy_type_high) { + case IXGBE_PHY_TYPE_HIGH_10BASE_T: + return ixgbe_media_type_copper; + } + } + return ixgbe_media_type_unknown; +} + +/** + * ixgbe_update_link_info - update status of the HW network link + * @hw: pointer to the HW struct + * + * Update the status of the HW network link. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_update_link_info(struct ixgbe_hw *hw) +{ + struct ixgbe_aci_cmd_get_phy_caps_data *pcaps; + struct ixgbe_link_status *li; + s32 status; + + if (!hw) + return IXGBE_ERR_PARAM; + + li = &hw->link.link_info; + + status = ixgbe_aci_get_link_info(hw, true, NULL); + if (status) + return status; + + if (li->link_info & IXGBE_ACI_MEDIA_AVAILABLE) { + pcaps = (struct ixgbe_aci_cmd_get_phy_caps_data *) + ixgbe_malloc(hw, sizeof(*pcaps)); + if (!pcaps) + return IXGBE_ERR_OUT_OF_MEM; + + status = ixgbe_aci_get_phy_caps(hw, false, + IXGBE_ACI_REPORT_TOPO_CAP_MEDIA, + pcaps); + + if (status == IXGBE_SUCCESS) + memcpy(li->module_type, &pcaps->module_type, + sizeof(li->module_type)); + + ixgbe_free(hw, pcaps); + } + + return status; +} + +/** + * ixgbe_get_link_status - get status of the HW network link + * @hw: pointer to the HW struct + * @link_up: pointer to bool (true/false = linkup/linkdown) + * + * Variable link_up is true if link is up, false if link is down. + * The variable link_up is invalid if status is non zero. As a + * result of this call, link status reporting becomes enabled + * + * Return: the exit code of the operation. + */ +s32 ixgbe_get_link_status(struct ixgbe_hw *hw, bool *link_up) +{ + s32 status = IXGBE_SUCCESS; + + if (!hw || !link_up) + return IXGBE_ERR_PARAM; + + if (hw->link.get_link_info) { + status = ixgbe_update_link_info(hw); + if (status) { + return status; + } + } + + *link_up = hw->link.link_info.link_info & IXGBE_ACI_LINK_UP; + + return status; +} + +/** + * ixgbe_aci_get_link_info - get the link status + * @hw: pointer to the HW struct + * @ena_lse: enable/disable LinkStatusEvent reporting + * @link: pointer to link status structure - optional + * + * Get the current Link Status using ACI command (0x607). + * The current link can be optionally provided to update + * the status. + * + * Return: the link status of the adapter. + */ +s32 ixgbe_aci_get_link_info(struct ixgbe_hw *hw, bool ena_lse, + struct ixgbe_link_status *link) +{ + struct ixgbe_aci_cmd_get_link_status_data link_data = { 0 }; + struct ixgbe_aci_cmd_get_link_status *resp; + struct ixgbe_link_status *li_old, *li; + struct ixgbe_fc_info *hw_fc_info; + enum ixgbe_media_type *hw_media_type; + struct ixgbe_aci_desc desc; + bool tx_pause, rx_pause; + u8 cmd_flags; + s32 status; + + if (!hw) + return IXGBE_ERR_PARAM; + + li_old = &hw->link.link_info_old; + hw_media_type = &hw->phy.media_type; + li = &hw->link.link_info; + hw_fc_info = &hw->fc; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_get_link_status); + cmd_flags = (ena_lse) ? IXGBE_ACI_LSE_ENA : IXGBE_ACI_LSE_DIS; + resp = &desc.params.get_link_status; + resp->cmd_flags = cmd_flags; + + status = ixgbe_aci_send_cmd(hw, &desc, &link_data, sizeof(link_data)); + + if (status != IXGBE_SUCCESS) + return status; + + /* save off old link status information */ + *li_old = *li; + + /* update current link status information */ + li->link_speed = IXGBE_LE16_TO_CPU(link_data.link_speed); + li->phy_type_low = IXGBE_LE64_TO_CPU(link_data.phy_type_low); + li->phy_type_high = IXGBE_LE64_TO_CPU(link_data.phy_type_high); + *hw_media_type = ixgbe_get_media_type_from_phy_type(hw); + li->link_info = link_data.link_info; + li->link_cfg_err = link_data.link_cfg_err; + li->an_info = link_data.an_info; + li->ext_info = link_data.ext_info; + li->max_frame_size = IXGBE_LE16_TO_CPU(link_data.max_frame_size); + li->fec_info = link_data.cfg & IXGBE_ACI_FEC_MASK; + li->topo_media_conflict = link_data.topo_media_conflict; + li->pacing = link_data.cfg & (IXGBE_ACI_CFG_PACING_M | + IXGBE_ACI_CFG_PACING_TYPE_M); + + /* update fc info */ + tx_pause = !!(link_data.an_info & IXGBE_ACI_LINK_PAUSE_TX); + rx_pause = !!(link_data.an_info & IXGBE_ACI_LINK_PAUSE_RX); + if (tx_pause && rx_pause) + hw_fc_info->current_mode = ixgbe_fc_full; + else if (tx_pause) + hw_fc_info->current_mode = ixgbe_fc_tx_pause; + else if (rx_pause) + hw_fc_info->current_mode = ixgbe_fc_rx_pause; + else + hw_fc_info->current_mode = ixgbe_fc_none; + + li->lse_ena = !!(resp->cmd_flags & IXGBE_ACI_LSE_IS_ENABLED); + + /* save link status information */ + if (link) + *link = *li; + + /* flag cleared so calling functions don't call AQ again */ + hw->link.get_link_info = false; + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_aci_set_event_mask - set event mask + * @hw: pointer to the HW struct + * @port_num: port number of the physical function + * @mask: event mask to be set + * + * Set the event mask using ACI command (0x0613). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_set_event_mask(struct ixgbe_hw *hw, u8 port_num, u16 mask) +{ + struct ixgbe_aci_cmd_set_event_mask *cmd; + struct ixgbe_aci_desc desc; + + cmd = &desc.params.set_event_mask; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_set_event_mask); + + cmd->event_mask = IXGBE_CPU_TO_LE16(mask); + return ixgbe_aci_send_cmd(hw, &desc, NULL, 0); +} + +/** + * ixgbe_configure_lse - enable/disable link status events + * @hw: pointer to the HW struct + * @activate: bool value deciding if lse should be enabled nor disabled + * @mask: event mask to be set; a set bit means deactivation of the + * corresponding event + * + * Set the event mask and then enable or disable link status events + * + * Return: the exit code of the operation. + */ +s32 ixgbe_configure_lse(struct ixgbe_hw *hw, bool activate, u16 mask) +{ + s32 rc; + + rc = ixgbe_aci_set_event_mask(hw, (u8)hw->bus.func, mask); + if (rc) { + return rc; + } + + /* Enabling link status events generation by fw */ + rc = ixgbe_aci_get_link_info(hw, activate, NULL); + if (rc) { + return rc; + } + return IXGBE_SUCCESS; +} + +/** + * ixgbe_aci_get_netlist_node - get a node handle + * @hw: pointer to the hw struct + * @cmd: get_link_topo AQ structure + * @node_part_number: output node part number if node found + * @node_handle: output node handle parameter if node found + * + * Get the netlist node and assigns it to + * the provided handle using ACI command (0x06E0). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_get_netlist_node(struct ixgbe_hw *hw, + struct ixgbe_aci_cmd_get_link_topo *cmd, + u8 *node_part_number, u16 *node_handle) +{ + struct ixgbe_aci_desc desc; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_get_link_topo); + desc.params.get_link_topo = *cmd; + + if (ixgbe_aci_send_cmd(hw, &desc, NULL, 0)) + return IXGBE_ERR_NOT_SUPPORTED; + + if (node_handle) + *node_handle = + IXGBE_LE16_TO_CPU(desc.params.get_link_topo.addr.handle); + if (node_part_number) + *node_part_number = desc.params.get_link_topo.node_part_num; + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_aci_get_netlist_node_pin - get a node pin handle + * @hw: pointer to the hw struct + * @cmd: get_link_topo_pin AQ structure + * @node_handle: output node handle parameter if node found + * + * Get the netlist node pin and assign it to + * the provided handle using ACI command (0x06E1). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_get_netlist_node_pin(struct ixgbe_hw *hw, + struct ixgbe_aci_cmd_get_link_topo_pin *cmd, + u16 *node_handle) +{ + struct ixgbe_aci_desc desc; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_get_link_topo_pin); + desc.params.get_link_topo_pin = *cmd; + + if (ixgbe_aci_send_cmd(hw, &desc, NULL, 0)) + return IXGBE_ERR_NOT_SUPPORTED; + + if (node_handle) + *node_handle = + IXGBE_LE16_TO_CPU(desc.params.get_link_topo_pin.addr.handle); + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_find_netlist_node - find a node handle + * @hw: pointer to the hw struct + * @node_type_ctx: type of netlist node to look for + * @node_part_number: node part number to look for + * @node_handle: output parameter if node found - optional + * + * Find and return the node handle for a given node type and part number in the + * netlist. When found IXGBE_SUCCESS is returned, IXGBE_ERR_NOT_SUPPORTED + * otherwise. If @node_handle provided, it would be set to found node handle. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_find_netlist_node(struct ixgbe_hw *hw, u8 node_type_ctx, + u8 node_part_number, u16 *node_handle) +{ + struct ixgbe_aci_cmd_get_link_topo cmd; + u8 rec_node_part_number; + u16 rec_node_handle; + s32 status; + u8 idx; + + for (idx = 0; idx < IXGBE_MAX_NETLIST_SIZE; idx++) { + memset(&cmd, 0, sizeof(cmd)); + + cmd.addr.topo_params.node_type_ctx = + (node_type_ctx << IXGBE_ACI_LINK_TOPO_NODE_TYPE_S); + cmd.addr.topo_params.index = idx; + + status = ixgbe_aci_get_netlist_node(hw, &cmd, + &rec_node_part_number, + &rec_node_handle); + if (status) + return status; + + if (rec_node_part_number == node_part_number) { + if (node_handle) + *node_handle = rec_node_handle; + return IXGBE_SUCCESS; + } + } + + return IXGBE_ERR_NOT_SUPPORTED; +} + +/** + * ixgbe_aci_read_i2c - read I2C register value + * @hw: pointer to the hw struct + * @topo_addr: topology address for a device to communicate with + * @bus_addr: 7-bit I2C bus address + * @addr: I2C memory address (I2C offset) with up to 16 bits + * @params: I2C parameters: bit [7] - Repeated start, + * bits [6:5] data offset size, + * bit [4] - I2C address type, bits [3:0] - data size + * to read (0-16 bytes) + * @data: pointer to data (0 to 16 bytes) to be read from the I2C device + * + * Read the value of the I2C pin register using ACI command (0x06E2). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_read_i2c(struct ixgbe_hw *hw, + struct ixgbe_aci_cmd_link_topo_addr topo_addr, + u16 bus_addr, __le16 addr, u8 params, u8 *data) +{ + struct ixgbe_aci_desc desc = { 0 }; + struct ixgbe_aci_cmd_i2c *cmd; + u8 data_size; + s32 status; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_read_i2c); + cmd = &desc.params.read_write_i2c; + + if (!data) + return IXGBE_ERR_PARAM; + + data_size = (params & IXGBE_ACI_I2C_DATA_SIZE_M) >> + IXGBE_ACI_I2C_DATA_SIZE_S; + + cmd->i2c_bus_addr = IXGBE_CPU_TO_LE16(bus_addr); + cmd->topo_addr = topo_addr; + cmd->i2c_params = params; + cmd->i2c_addr = addr; + + status = ixgbe_aci_send_cmd(hw, &desc, NULL, 0); + if (!status) { + struct ixgbe_aci_cmd_read_i2c_resp *resp; + u8 i; + + resp = &desc.params.read_i2c_resp; + for (i = 0; i < data_size; i++) { + *data = resp->i2c_data[i]; + data++; + } + } + + return status; +} + +/** + * ixgbe_aci_write_i2c - write a value to I2C register + * @hw: pointer to the hw struct + * @topo_addr: topology address for a device to communicate with + * @bus_addr: 7-bit I2C bus address + * @addr: I2C memory address (I2C offset) with up to 16 bits + * @params: I2C parameters: bit [4] - I2C address type, bits [3:0] - data size + * to write (0-7 bytes) + * @data: pointer to data (0 to 4 bytes) to be written to the I2C device + * + * Write a value to the I2C pin register using ACI command (0x06E3). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_write_i2c(struct ixgbe_hw *hw, + struct ixgbe_aci_cmd_link_topo_addr topo_addr, + u16 bus_addr, __le16 addr, u8 params, u8 *data) +{ + struct ixgbe_aci_desc desc = { 0 }; + struct ixgbe_aci_cmd_i2c *cmd; + u8 i, data_size; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_write_i2c); + cmd = &desc.params.read_write_i2c; + + data_size = (params & IXGBE_ACI_I2C_DATA_SIZE_M) >> + IXGBE_ACI_I2C_DATA_SIZE_S; + + /* data_size limited to 4 */ + if (data_size > 4) + return IXGBE_ERR_PARAM; + + cmd->i2c_bus_addr = IXGBE_CPU_TO_LE16(bus_addr); + cmd->topo_addr = topo_addr; + cmd->i2c_params = params; + cmd->i2c_addr = addr; + + for (i = 0; i < data_size; i++) { + cmd->i2c_data[i] = *data; + data++; + } + + return ixgbe_aci_send_cmd(hw, &desc, NULL, 0); +} + +/** + * ixgbe_aci_set_gpio - set GPIO pin state + * @hw: pointer to the hw struct + * @gpio_ctrl_handle: GPIO controller node handle + * @pin_idx: IO Number of the GPIO that needs to be set + * @value: SW provide IO value to set in the LSB + * + * Set the GPIO pin state that is a part of the topology + * using ACI command (0x06EC). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_set_gpio(struct ixgbe_hw *hw, u16 gpio_ctrl_handle, u8 pin_idx, + bool value) +{ + struct ixgbe_aci_cmd_gpio *cmd; + struct ixgbe_aci_desc desc; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_set_gpio); + cmd = &desc.params.read_write_gpio; + cmd->gpio_ctrl_handle = IXGBE_CPU_TO_LE16(gpio_ctrl_handle); + cmd->gpio_num = pin_idx; + cmd->gpio_val = value ? 1 : 0; + + return ixgbe_aci_send_cmd(hw, &desc, NULL, 0); +} + +/** + * ixgbe_aci_get_gpio - get GPIO pin state + * @hw: pointer to the hw struct + * @gpio_ctrl_handle: GPIO controller node handle + * @pin_idx: IO Number of the GPIO that needs to be set + * @value: IO value read + * + * Get the value of a GPIO signal which is part of the topology + * using ACI command (0x06ED). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_get_gpio(struct ixgbe_hw *hw, u16 gpio_ctrl_handle, u8 pin_idx, + bool *value) +{ + struct ixgbe_aci_cmd_gpio *cmd; + struct ixgbe_aci_desc desc; + s32 status; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_get_gpio); + cmd = &desc.params.read_write_gpio; + cmd->gpio_ctrl_handle = IXGBE_CPU_TO_LE16(gpio_ctrl_handle); + cmd->gpio_num = pin_idx; + + status = ixgbe_aci_send_cmd(hw, &desc, NULL, 0); + if (status) + return status; + + *value = !!cmd->gpio_val; + return IXGBE_SUCCESS; +} + +/** + * ixgbe_aci_sff_eeprom - read/write SFF EEPROM + * @hw: pointer to the HW struct + * @lport: bits [7:0] = logical port, bit [8] = logical port valid + * @bus_addr: I2C bus address of the eeprom (typically 0xA0, 0=topo default) + * @mem_addr: I2C offset. lower 8 bits for address, 8 upper bits zero padding. + * @page: QSFP page + * @page_bank_ctrl: configuration of SFF/CMIS paging and banking control + * @data: pointer to data buffer to be read/written to the I2C device. + * @length: 1-16 for read, 1 for write. + * @write: 0 read, 1 for write. + * + * Read/write SFF EEPROM using ACI command (0x06EE). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_sff_eeprom(struct ixgbe_hw *hw, u16 lport, u8 bus_addr, + u16 mem_addr, u8 page, u8 page_bank_ctrl, u8 *data, + u8 length, bool write) +{ + struct ixgbe_aci_cmd_sff_eeprom *cmd; + struct ixgbe_aci_desc desc; + s32 status; + + if (!data || (mem_addr & 0xff00)) + return IXGBE_ERR_PARAM; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_sff_eeprom); + cmd = &desc.params.read_write_sff_param; + desc.flags = IXGBE_CPU_TO_LE16(IXGBE_ACI_FLAG_RD); + cmd->lport_num = (u8)(lport & 0xff); + cmd->lport_num_valid = (u8)((lport >> 8) & 0x01); + cmd->i2c_bus_addr = IXGBE_CPU_TO_LE16(((bus_addr >> 1) & + IXGBE_ACI_SFF_I2CBUS_7BIT_M) | + ((page_bank_ctrl << + IXGBE_ACI_SFF_PAGE_BANK_CTRL_S) & + IXGBE_ACI_SFF_PAGE_BANK_CTRL_M)); + cmd->i2c_offset = IXGBE_CPU_TO_LE16(mem_addr & 0xff); + cmd->module_page = page; + if (write) + cmd->i2c_bus_addr |= IXGBE_CPU_TO_LE16(IXGBE_ACI_SFF_IS_WRITE); + + status = ixgbe_aci_send_cmd(hw, &desc, data, length); + return status; +} + +/** + * ixgbe_aci_prog_topo_dev_nvm - program Topology Device NVM + * @hw: pointer to the hardware structure + * @topo_params: pointer to structure storing topology parameters for a device + * + * Program Topology Device NVM using ACI command (0x06F2). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_prog_topo_dev_nvm(struct ixgbe_hw *hw, + struct ixgbe_aci_cmd_link_topo_params *topo_params) +{ + struct ixgbe_aci_cmd_prog_topo_dev_nvm *cmd; + struct ixgbe_aci_desc desc; + + cmd = &desc.params.prog_topo_dev_nvm; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_prog_topo_dev_nvm); + + memcpy(&cmd->topo_params, topo_params, sizeof(*topo_params)); + + return ixgbe_aci_send_cmd(hw, &desc, NULL, 0); +} + +/** + * ixgbe_aci_read_topo_dev_nvm - read Topology Device NVM + * @hw: pointer to the hardware structure + * @topo_params: pointer to structure storing topology parameters for a device + * @start_address: byte offset in the topology device NVM + * @data: pointer to data buffer + * @data_size: number of bytes to be read from the topology device NVM + * Read Topology Device NVM (0x06F3) + * + * Read Topology of Device NVM using ACI command (0x06F3). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_read_topo_dev_nvm(struct ixgbe_hw *hw, + struct ixgbe_aci_cmd_link_topo_params *topo_params, + u32 start_address, u8 *data, u8 data_size) +{ + struct ixgbe_aci_cmd_read_topo_dev_nvm *cmd; + struct ixgbe_aci_desc desc; + s32 status; + + if (!data || data_size == 0 || + data_size > IXGBE_ACI_READ_TOPO_DEV_NVM_DATA_READ_SIZE) + return IXGBE_ERR_PARAM; + + cmd = &desc.params.read_topo_dev_nvm; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_read_topo_dev_nvm); + + desc.datalen = IXGBE_CPU_TO_LE16(data_size); + memcpy(&cmd->topo_params, topo_params, sizeof(*topo_params)); + cmd->start_address = IXGBE_CPU_TO_LE32(start_address); + + status = ixgbe_aci_send_cmd(hw, &desc, NULL, 0); + if (status) + return status; + + memcpy(data, cmd->data_read, data_size); + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_acquire_nvm - Generic request for acquiring the NVM ownership + * @hw: pointer to the HW structure + * @access: NVM access type (read or write) + * + * Request NVM ownership. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_acquire_nvm(struct ixgbe_hw *hw, + enum ixgbe_aci_res_access_type access) +{ + u32 fla; + + /* Skip if we are in blank NVM programming mode */ + fla = IXGBE_READ_REG(hw, GLNVM_FLA); + if ((fla & GLNVM_FLA_LOCKED_M) == 0) + return IXGBE_SUCCESS; + + return ixgbe_acquire_res(hw, IXGBE_NVM_RES_ID, access, + IXGBE_NVM_TIMEOUT); +} + +/** + * ixgbe_release_nvm - Generic request for releasing the NVM ownership + * @hw: pointer to the HW structure + * + * Release NVM ownership. + */ +void ixgbe_release_nvm(struct ixgbe_hw *hw) +{ + u32 fla; + + /* Skip if we are in blank NVM programming mode */ + fla = IXGBE_READ_REG(hw, GLNVM_FLA); + if ((fla & GLNVM_FLA_LOCKED_M) == 0) + return; + + ixgbe_release_res(hw, IXGBE_NVM_RES_ID); +} + + +/** + * ixgbe_aci_read_nvm - read NVM + * @hw: pointer to the HW struct + * @module_typeid: module pointer location in words from the NVM beginning + * @offset: byte offset from the module beginning + * @length: length of the section to be read (in bytes from the offset) + * @data: command buffer (size [bytes] = length) + * @last_command: tells if this is the last command in a series + * @read_shadow_ram: tell if this is a shadow RAM read + * + * Read the NVM using ACI command (0x0701). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_read_nvm(struct ixgbe_hw *hw, u16 module_typeid, u32 offset, + u16 length, void *data, bool last_command, + bool read_shadow_ram) +{ + struct ixgbe_aci_desc desc; + struct ixgbe_aci_cmd_nvm *cmd; + + cmd = &desc.params.nvm; + + if (offset > IXGBE_ACI_NVM_MAX_OFFSET) + return IXGBE_ERR_PARAM; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_nvm_read); + + if (!read_shadow_ram && module_typeid == IXGBE_ACI_NVM_START_POINT) + cmd->cmd_flags |= IXGBE_ACI_NVM_FLASH_ONLY; + + /* If this is the last command in a series, set the proper flag. */ + if (last_command) + cmd->cmd_flags |= IXGBE_ACI_NVM_LAST_CMD; + cmd->module_typeid = IXGBE_CPU_TO_LE16(module_typeid); + cmd->offset_low = IXGBE_CPU_TO_LE16(offset & 0xFFFF); + cmd->offset_high = (offset >> 16) & 0xFF; + cmd->length = IXGBE_CPU_TO_LE16(length); + + return ixgbe_aci_send_cmd(hw, &desc, data, length); +} + +/** + * ixgbe_nvm_validate_checksum - validate checksum + * @hw: pointer to the HW struct + * + * Verify NVM PFA checksum validity using ACI command (0x0706). + * If the checksum verification failed, IXGBE_ERR_NVM_CHECKSUM is returned. + * The function acquires and then releases the NVM ownership. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_nvm_validate_checksum(struct ixgbe_hw *hw) +{ + struct ixgbe_aci_cmd_nvm_checksum *cmd; + struct ixgbe_aci_desc desc; + s32 status; + + status = ixgbe_acquire_nvm(hw, IXGBE_RES_READ); + if (status) + return status; + + cmd = &desc.params.nvm_checksum; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_nvm_checksum); + cmd->flags = IXGBE_ACI_NVM_CHECKSUM_VERIFY; + + status = ixgbe_aci_send_cmd(hw, &desc, NULL, 0); + + ixgbe_release_nvm(hw); + + if (!status) + if (IXGBE_LE16_TO_CPU(cmd->checksum) != + IXGBE_ACI_NVM_CHECKSUM_CORRECT) { + ERROR_REPORT1(IXGBE_ERROR_INVALID_STATE, + "Invalid Shadow Ram checksum"); + status = IXGBE_ERR_NVM_CHECKSUM; + } + + return status; +} + +/** + * ixgbe_nvm_recalculate_checksum - recalculate checksum + * @hw: pointer to the HW struct + * + * Recalculate NVM PFA checksum using ACI command (0x0706). + * The function acquires and then releases the NVM ownership. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_nvm_recalculate_checksum(struct ixgbe_hw *hw) +{ + struct ixgbe_aci_cmd_nvm_checksum *cmd; + struct ixgbe_aci_desc desc; + s32 status; + + status = ixgbe_acquire_nvm(hw, IXGBE_RES_READ); + if (status) + return status; + + cmd = &desc.params.nvm_checksum; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_nvm_checksum); + cmd->flags = IXGBE_ACI_NVM_CHECKSUM_RECALC; + + status = ixgbe_aci_send_cmd(hw, &desc, NULL, 0); + + ixgbe_release_nvm(hw); + + return status; +} + +/** + * ixgbe_get_flash_bank_offset - Get offset into requested flash bank + * @hw: pointer to the HW structure + * @bank: whether to read from the active or inactive flash bank + * @module: the module to read from + * + * Based on the module, lookup the module offset from the beginning of the + * flash. + * + * Return: the flash offset. Note that a value of zero is invalid and must be + * treated as an error. + */ +static u32 ixgbe_get_flash_bank_offset(struct ixgbe_hw *hw, + enum ixgbe_bank_select bank, + u16 module) +{ + struct ixgbe_bank_info *banks = &hw->flash.banks; + enum ixgbe_flash_bank active_bank; + bool second_bank_active; + u32 offset, size; + + switch (module) { + case E610_SR_1ST_NVM_BANK_PTR: + offset = banks->nvm_ptr; + size = banks->nvm_size; + active_bank = banks->nvm_bank; + break; + case E610_SR_1ST_OROM_BANK_PTR: + offset = banks->orom_ptr; + size = banks->orom_size; + active_bank = banks->orom_bank; + break; + case E610_SR_NETLIST_BANK_PTR: + offset = banks->netlist_ptr; + size = banks->netlist_size; + active_bank = banks->netlist_bank; + break; + default: + return 0; + } + + switch (active_bank) { + case IXGBE_1ST_FLASH_BANK: + second_bank_active = false; + break; + case IXGBE_2ND_FLASH_BANK: + second_bank_active = true; + break; + default: + return 0; + } + + /* The second flash bank is stored immediately following the first + * bank. Based on whether the 1st or 2nd bank is active, and whether + * we want the active or inactive bank, calculate the desired offset. + */ + switch (bank) { + case IXGBE_ACTIVE_FLASH_BANK: + return offset + (second_bank_active ? size : 0); + case IXGBE_INACTIVE_FLASH_BANK: + return offset + (second_bank_active ? 0 : size); + } + + return 0; +} + +/** + * ixgbe_read_flash_module - Read a word from one of the main NVM modules + * @hw: pointer to the HW structure + * @bank: which bank of the module to read + * @module: the module to read + * @offset: the offset into the module in bytes + * @data: storage for the word read from the flash + * @length: bytes of data to read + * + * Read data from the specified flash module. The bank parameter indicates + * whether or not to read from the active bank or the inactive bank of that + * module. + * + * The word will be read using flat NVM access, and relies on the + * hw->flash.banks data being setup by ixgbe_determine_active_flash_banks() + * during initialization. + * + * Return: the exit code of the operation. + */ +static s32 ixgbe_read_flash_module(struct ixgbe_hw *hw, + enum ixgbe_bank_select bank, + u16 module, u32 offset, u8 *data, u32 length) +{ + s32 status; + u32 start; + + start = ixgbe_get_flash_bank_offset(hw, bank, module); + if (!start) { + return IXGBE_ERR_PARAM; + } + + status = ixgbe_acquire_nvm(hw, IXGBE_RES_READ); + if (status) + return status; + + status = ixgbe_read_flat_nvm(hw, start + offset, &length, data, false); + + ixgbe_release_nvm(hw); + + return status; +} + +/** + * ixgbe_read_nvm_module - Read from the active main NVM module + * @hw: pointer to the HW structure + * @bank: whether to read from active or inactive NVM module + * @offset: offset into the NVM module to read, in words + * @data: storage for returned word value + * + * Read the specified word from the active NVM module. This includes the CSS + * header at the start of the NVM module. + * + * Return: the exit code of the operation. + */ +static s32 ixgbe_read_nvm_module(struct ixgbe_hw *hw, + enum ixgbe_bank_select bank, + u32 offset, u16 *data) +{ + __le16 data_local; + s32 status; + + status = ixgbe_read_flash_module(hw, bank, E610_SR_1ST_NVM_BANK_PTR, + offset * sizeof(u16), + (u8 *)&data_local, + sizeof(u16)); + if (!status) + *data = IXGBE_LE16_TO_CPU(data_local); + + return status; +} + +/** + * ixgbe_get_nvm_css_hdr_len - Read the CSS header length from the + * NVM CSS header + * @hw: pointer to the HW struct + * @bank: whether to read from the active or inactive flash bank + * @hdr_len: storage for header length in words + * + * Read the CSS header length from the NVM CSS header and add the + * Authentication header size, and then convert to words. + * + * Return: the exit code of the operation. + */ +static s32 ixgbe_get_nvm_css_hdr_len(struct ixgbe_hw *hw, + enum ixgbe_bank_select bank, + u32 *hdr_len) +{ + u16 hdr_len_l, hdr_len_h; + u32 hdr_len_dword; + s32 status; + + status = ixgbe_read_nvm_module(hw, bank, IXGBE_NVM_CSS_HDR_LEN_L, + &hdr_len_l); + if (status) + return status; + + status = ixgbe_read_nvm_module(hw, bank, IXGBE_NVM_CSS_HDR_LEN_H, + &hdr_len_h); + if (status) + return status; + + /* CSS header length is in DWORD, so convert to words and add + * authentication header size + */ + hdr_len_dword = hdr_len_h << 16 | hdr_len_l; + *hdr_len = (hdr_len_dword * 2) + IXGBE_NVM_AUTH_HEADER_LEN; + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_read_nvm_sr_copy - Read a word from the Shadow RAM copy in the NVM bank + * @hw: pointer to the HW structure + * @bank: whether to read from the active or inactive NVM module + * @offset: offset into the Shadow RAM copy to read, in words + * @data: storage for returned word value + * + * Read the specified word from the copy of the Shadow RAM found in the + * specified NVM module. + * + * Return: the exit code of the operation. + */ +static s32 ixgbe_read_nvm_sr_copy(struct ixgbe_hw *hw, + enum ixgbe_bank_select bank, + u32 offset, u16 *data) +{ + u32 hdr_len; + s32 status; + + status = ixgbe_get_nvm_css_hdr_len(hw, bank, &hdr_len); + if (status) + return status; + + hdr_len = ROUND_UP(hdr_len, 32); + + return ixgbe_read_nvm_module(hw, bank, hdr_len + offset, data); +} + +/** + * ixgbe_get_nvm_minsrevs - Get the minsrevs values from flash + * @hw: pointer to the HW struct + * @minsrevs: structure to store NVM and OROM minsrev values + * + * Read the Minimum Security Revision TLV and extract + * the revision values from the flash image + * into a readable structure for processing. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_get_nvm_minsrevs(struct ixgbe_hw *hw, + struct ixgbe_minsrev_info *minsrevs) +{ + struct ixgbe_aci_cmd_nvm_minsrev data; + s32 status; + u16 valid; + + status = ixgbe_acquire_nvm(hw, IXGBE_RES_READ); + if (status) + return status; + + status = ixgbe_aci_read_nvm(hw, IXGBE_ACI_NVM_MINSREV_MOD_ID, + 0, sizeof(data), &data, + true, false); + + ixgbe_release_nvm(hw); + + if (status) + return status; + + valid = IXGBE_LE16_TO_CPU(data.validity); + + /* Extract NVM minimum security revision */ + if (valid & IXGBE_ACI_NVM_MINSREV_NVM_VALID) { + u16 minsrev_l = IXGBE_LE16_TO_CPU(data.nvm_minsrev_l); + u16 minsrev_h = IXGBE_LE16_TO_CPU(data.nvm_minsrev_h); + + minsrevs->nvm = minsrev_h << 16 | minsrev_l; + minsrevs->nvm_valid = true; + } + + /* Extract the OROM minimum security revision */ + if (valid & IXGBE_ACI_NVM_MINSREV_OROM_VALID) { + u16 minsrev_l = IXGBE_LE16_TO_CPU(data.orom_minsrev_l); + u16 minsrev_h = IXGBE_LE16_TO_CPU(data.orom_minsrev_h); + + minsrevs->orom = minsrev_h << 16 | minsrev_l; + minsrevs->orom_valid = true; + } + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_get_nvm_srev - Read the security revision from the NVM CSS header + * @hw: pointer to the HW struct + * @bank: whether to read from the active or inactive flash bank + * @srev: storage for security revision + * + * Read the security revision out of the CSS header of the active NVM module + * bank. + * + * Return: the exit code of the operation. + */ +static s32 ixgbe_get_nvm_srev(struct ixgbe_hw *hw, + enum ixgbe_bank_select bank, u32 *srev) +{ + u16 srev_l, srev_h; + s32 status; + + status = ixgbe_read_nvm_module(hw, bank, IXGBE_NVM_CSS_SREV_L, &srev_l); + if (status) + return status; + + status = ixgbe_read_nvm_module(hw, bank, IXGBE_NVM_CSS_SREV_H, &srev_h); + if (status) + return status; + + *srev = srev_h << 16 | srev_l; + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_get_nvm_ver_info - Read NVM version information + * @hw: pointer to the HW struct + * @bank: whether to read from the active or inactive flash bank + * @nvm: pointer to NVM info structure + * + * Read the NVM EETRACK ID and map version of the main NVM image bank, filling + * in the nvm info structure. + * + * Return: the exit code of the operation. + */ +static s32 ixgbe_get_nvm_ver_info(struct ixgbe_hw *hw, + enum ixgbe_bank_select bank, + struct ixgbe_nvm_info *nvm) +{ + u16 eetrack_lo, eetrack_hi, ver; + s32 status; + + status = ixgbe_read_nvm_sr_copy(hw, bank, + E610_SR_NVM_DEV_STARTER_VER, &ver); + if (status) { + return status; + } + + nvm->major = (ver & E610_NVM_VER_HI_MASK) >> E610_NVM_VER_HI_SHIFT; + nvm->minor = (ver & E610_NVM_VER_LO_MASK) >> E610_NVM_VER_LO_SHIFT; + + status = ixgbe_read_nvm_sr_copy(hw, bank, E610_SR_NVM_EETRACK_LO, + &eetrack_lo); + if (status) { + return status; + } + status = ixgbe_read_nvm_sr_copy(hw, bank, E610_SR_NVM_EETRACK_HI, + &eetrack_hi); + if (status) { + return status; + } + + nvm->eetrack = (eetrack_hi << 16) | eetrack_lo; + + status = ixgbe_get_nvm_srev(hw, bank, &nvm->srev); + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_get_inactive_nvm_ver - Read Option ROM version from the inactive bank + * @hw: pointer to the HW structure + * @nvm: storage for Option ROM version information + * + * Reads the NVM EETRACK ID, Map version, and security revision of the + * inactive NVM bank. Used to access version data for a pending update that + * has not yet been activated. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_get_inactive_nvm_ver(struct ixgbe_hw *hw, struct ixgbe_nvm_info *nvm) +{ + return ixgbe_get_nvm_ver_info(hw, IXGBE_INACTIVE_FLASH_BANK, nvm); +} + +/** + * ixgbe_get_active_nvm_ver - Read Option ROM version from the active bank + * @hw: pointer to the HW structure + * @nvm: storage for Option ROM version information + * + * Reads the NVM EETRACK ID, Map version, and security revision of the + * active NVM bank. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_get_active_nvm_ver(struct ixgbe_hw *hw, struct ixgbe_nvm_info *nvm) +{ + return ixgbe_get_nvm_ver_info(hw, IXGBE_ACTIVE_FLASH_BANK, nvm); +} + +/** + * ixgbe_read_sr_pointer - Read the value of a Shadow RAM pointer word + * @hw: pointer to the HW structure + * @offset: the word offset of the Shadow RAM word to read + * @pointer: pointer value read from Shadow RAM + * + * Read the given Shadow RAM word, and convert it to a pointer value specified + * in bytes. This function assumes the specified offset is a valid pointer + * word. + * + * Each pointer word specifies whether it is stored in word size or 4KB + * sector size by using the highest bit. The reported pointer value will be in + * bytes, intended for flat NVM reads. + * + * Return: the exit code of the operation. + */ +static s32 ixgbe_read_sr_pointer(struct ixgbe_hw *hw, u16 offset, u32 *pointer) +{ + s32 status; + u16 value; + + status = ixgbe_read_ee_aci_E610(hw, offset, &value); + if (status) + return status; + + /* Determine if the pointer is in 4KB or word units */ + if (value & IXGBE_SR_NVM_PTR_4KB_UNITS) + *pointer = (value & ~IXGBE_SR_NVM_PTR_4KB_UNITS) * 4 * 1024; + else + *pointer = value * 2; + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_read_sr_area_size - Read an area size from a Shadow RAM word + * @hw: pointer to the HW structure + * @offset: the word offset of the Shadow RAM to read + * @size: size value read from the Shadow RAM + * + * Read the given Shadow RAM word, and convert it to an area size value + * specified in bytes. This function assumes the specified offset is a valid + * area size word. + * + * Each area size word is specified in 4KB sector units. This function reports + * the size in bytes, intended for flat NVM reads. + * + * Return: the exit code of the operation. + */ +static s32 ixgbe_read_sr_area_size(struct ixgbe_hw *hw, u16 offset, u32 *size) +{ + s32 status; + u16 value; + + status = ixgbe_read_ee_aci_E610(hw, offset, &value); + if (status) + return status; + + /* Area sizes are always specified in 4KB units */ + *size = value * 4 * 1024; + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_discover_flash_size - Discover the available flash size. + * @hw: pointer to the HW struct + * + * The device flash could be up to 16MB in size. However, it is possible that + * the actual size is smaller. Use bisection to determine the accessible size + * of flash memory. + * + * Return: the exit code of the operation. + */ +static s32 ixgbe_discover_flash_size(struct ixgbe_hw *hw) +{ + u32 min_size = 0, max_size = IXGBE_ACI_NVM_MAX_OFFSET + 1; + s32 status; + + status = ixgbe_acquire_nvm(hw, IXGBE_RES_READ); + if (status) + return status; + + while ((max_size - min_size) > 1) { + u32 offset = (max_size + min_size) / 2; + u32 len = 1; + u8 data; + + status = ixgbe_read_flat_nvm(hw, offset, &len, &data, false); + if (status == IXGBE_ERR_ACI_ERROR && + hw->aci.last_status == IXGBE_ACI_RC_EINVAL) { + status = IXGBE_SUCCESS; + max_size = offset; + } else if (!status) { + min_size = offset; + } else { + /* an unexpected error occurred */ + goto err_read_flat_nvm; + } + } + + hw->flash.flash_size = max_size; + +err_read_flat_nvm: + ixgbe_release_nvm(hw); + + return status; +} + +/** + * ixgbe_determine_active_flash_banks - Discover active bank for each module + * @hw: pointer to the HW struct + * + * Read the Shadow RAM control word and determine which banks are active for + * the NVM, OROM, and Netlist modules. Also read and calculate the associated + * pointer and size. These values are then cached into the ixgbe_flash_info + * structure for later use in order to calculate the correct offset to read + * from the active module. + * + * Return: the exit code of the operation. + */ +static s32 ixgbe_determine_active_flash_banks(struct ixgbe_hw *hw) +{ + struct ixgbe_bank_info *banks = &hw->flash.banks; + u16 ctrl_word; + s32 status; + + status = ixgbe_read_ee_aci_E610(hw, E610_SR_NVM_CTRL_WORD, &ctrl_word); + if (status) { + return status; + } + + /* Check that the control word indicates validity */ + if ((ctrl_word & IXGBE_SR_CTRL_WORD_1_M) >> IXGBE_SR_CTRL_WORD_1_S != + IXGBE_SR_CTRL_WORD_VALID) { + return IXGBE_ERR_CONFIG; + } + + if (!(ctrl_word & IXGBE_SR_CTRL_WORD_NVM_BANK)) + banks->nvm_bank = IXGBE_1ST_FLASH_BANK; + else + banks->nvm_bank = IXGBE_2ND_FLASH_BANK; + + if (!(ctrl_word & IXGBE_SR_CTRL_WORD_OROM_BANK)) + banks->orom_bank = IXGBE_1ST_FLASH_BANK; + else + banks->orom_bank = IXGBE_2ND_FLASH_BANK; + + if (!(ctrl_word & IXGBE_SR_CTRL_WORD_NETLIST_BANK)) + banks->netlist_bank = IXGBE_1ST_FLASH_BANK; + else + banks->netlist_bank = IXGBE_2ND_FLASH_BANK; + + status = ixgbe_read_sr_pointer(hw, E610_SR_1ST_NVM_BANK_PTR, + &banks->nvm_ptr); + if (status) { + return status; + } + + status = ixgbe_read_sr_area_size(hw, E610_SR_NVM_BANK_SIZE, + &banks->nvm_size); + if (status) { + return status; + } + + status = ixgbe_read_sr_pointer(hw, E610_SR_1ST_OROM_BANK_PTR, + &banks->orom_ptr); + if (status) { + return status; + } + + status = ixgbe_read_sr_area_size(hw, E610_SR_OROM_BANK_SIZE, + &banks->orom_size); + if (status) { + return status; + } + + status = ixgbe_read_sr_pointer(hw, E610_SR_NETLIST_BANK_PTR, + &banks->netlist_ptr); + if (status) { + return status; + } + + status = ixgbe_read_sr_area_size(hw, E610_SR_NETLIST_BANK_SIZE, + &banks->netlist_size); + if (status) { + return status; + } + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_init_nvm - initializes NVM setting + * @hw: pointer to the HW struct + * + * Read and populate NVM settings such as Shadow RAM size, + * max_timeout, and blank_nvm_mode + * + * Return: the exit code of the operation. + */ +s32 ixgbe_init_nvm(struct ixgbe_hw *hw) +{ + struct ixgbe_flash_info *flash = &hw->flash; + u32 fla, gens_stat, status; + u8 sr_size; + + /* The SR size is stored regardless of the NVM programming mode + * as the blank mode may be used in the factory line. + */ + gens_stat = IXGBE_READ_REG(hw, GLNVM_GENS); + sr_size = (gens_stat & GLNVM_GENS_SR_SIZE_M) >> GLNVM_GENS_SR_SIZE_S; + + /* Switching to words (sr_size contains power of 2) */ + flash->sr_words = BIT(sr_size) * IXGBE_SR_WORDS_IN_1KB; + + /* Check if we are in the normal or blank NVM programming mode */ + fla = IXGBE_READ_REG(hw, GLNVM_FLA); + if (fla & GLNVM_FLA_LOCKED_M) { /* Normal programming mode */ + flash->blank_nvm_mode = false; + } else { + /* Blank programming mode */ + flash->blank_nvm_mode = true; + return IXGBE_ERR_NVM_BLANK_MODE; + } + + status = ixgbe_discover_flash_size(hw); + if (status) { + return status; + } + + status = ixgbe_determine_active_flash_banks(hw); + if (status) { + return status; + } + + status = ixgbe_get_nvm_ver_info(hw, IXGBE_ACTIVE_FLASH_BANK, + &flash->nvm); + if (status) { + return status; + } + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_sanitize_operate - Clear the user data + * @hw: pointer to the HW struct + * + * Clear user data from NVM using ACI command (0x070C). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_sanitize_operate(struct ixgbe_hw *hw) +{ + s32 status; + u8 values; + + u8 cmd_flags = IXGBE_ACI_SANITIZE_REQ_OPERATE | + IXGBE_ACI_SANITIZE_OPERATE_SUBJECT_CLEAR; + + status = ixgbe_sanitize_nvm(hw, cmd_flags, &values); + if (status) + return status; + if ((!(values & IXGBE_ACI_SANITIZE_OPERATE_HOST_CLEAN_DONE) && + !(values & IXGBE_ACI_SANITIZE_OPERATE_BMC_CLEAN_DONE)) || + ((values & IXGBE_ACI_SANITIZE_OPERATE_HOST_CLEAN_DONE) && + !(values & IXGBE_ACI_SANITIZE_OPERATE_HOST_CLEAN_SUCCESS)) || + ((values & IXGBE_ACI_SANITIZE_OPERATE_BMC_CLEAN_DONE) && + !(values & IXGBE_ACI_SANITIZE_OPERATE_BMC_CLEAN_SUCCESS))) + return IXGBE_ERR_ACI_ERROR; + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_sanitize_nvm - Sanitize NVM + * @hw: pointer to the HW struct + * @cmd_flags: flag to the ACI command + * @values: values returned from the command + * + * Sanitize NVM using ACI command (0x070C). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_sanitize_nvm(struct ixgbe_hw *hw, u8 cmd_flags, u8 *values) +{ + struct ixgbe_aci_desc desc; + struct ixgbe_aci_cmd_nvm_sanitization *cmd; + s32 status; + + cmd = &desc.params.nvm_sanitization; + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_nvm_sanitization); + cmd->cmd_flags = cmd_flags; + + status = ixgbe_aci_send_cmd(hw, &desc, NULL, 0); + if (values) + *values = cmd->values; + + return status; +} + +/** + * ixgbe_read_sr_word_aci - Reads Shadow RAM via ACI + * @hw: pointer to the HW structure + * @offset: offset of the Shadow RAM word to read (0x000000 - 0x001FFF) + * @data: word read from the Shadow RAM + * + * Reads one 16 bit word from the Shadow RAM using ixgbe_read_flat_nvm. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_read_sr_word_aci(struct ixgbe_hw *hw, u16 offset, u16 *data) +{ + u32 bytes = sizeof(u16); + __le16 data_local; + s32 status; + + status = ixgbe_read_flat_nvm(hw, offset * sizeof(u16), &bytes, + (u8 *)&data_local, true); + if (status) + return status; + + *data = IXGBE_LE16_TO_CPU(data_local); + return IXGBE_SUCCESS; +} + +/** + * ixgbe_read_sr_buf_aci - Reads Shadow RAM buf via ACI + * @hw: pointer to the HW structure + * @offset: offset of the Shadow RAM word to read (0x000000 - 0x001FFF) + * @words: (in) number of words to read; (out) number of words actually read + * @data: words read from the Shadow RAM + * + * Reads 16 bit words (data buf) from the Shadow RAM. Ownership of the NVM is + * taken before reading the buffer and later released. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_read_sr_buf_aci(struct ixgbe_hw *hw, u16 offset, u16 *words, + u16 *data) +{ + u32 bytes = *words * 2, i; + s32 status; + + status = ixgbe_read_flat_nvm(hw, offset * 2, &bytes, (u8 *)data, true); + + *words = bytes / 2; + + for (i = 0; i < *words; i++) + data[i] = IXGBE_LE16_TO_CPU(((__le16 *)data)[i]); + + return status; +} + +/** + * ixgbe_read_flat_nvm - Read portion of NVM by flat offset + * @hw: pointer to the HW struct + * @offset: offset from beginning of NVM + * @length: (in) number of bytes to read; (out) number of bytes actually read + * @data: buffer to return data in (sized to fit the specified length) + * @read_shadow_ram: if true, read from shadow RAM instead of NVM + * + * Reads a portion of the NVM, as a flat memory space. This function correctly + * breaks read requests across Shadow RAM sectors, prevents Shadow RAM size + * from being exceeded in case of Shadow RAM read requests and ensures that no + * single read request exceeds the maximum 4KB read for a single admin command. + * + * Returns a status code on failure. Note that the data pointer may be + * partially updated if some reads succeed before a failure. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_read_flat_nvm(struct ixgbe_hw *hw, u32 offset, u32 *length, + u8 *data, bool read_shadow_ram) +{ + u32 inlen = *length; + u32 bytes_read = 0; + bool last_cmd; + s32 status; + + *length = 0; + + /* Verify the length of the read if this is for the Shadow RAM */ + if (read_shadow_ram && ((offset + inlen) > + (hw->eeprom.word_size * 2u))) { + return IXGBE_ERR_PARAM; + } + + do { + u32 read_size, sector_offset; + + /* ixgbe_aci_read_nvm cannot read more than 4KB at a time. + * Additionally, a read from the Shadow RAM may not cross over + * a sector boundary. Conveniently, the sector size is also 4KB. + */ + sector_offset = offset % IXGBE_ACI_MAX_BUFFER_SIZE; + read_size = MIN_T(u32, + IXGBE_ACI_MAX_BUFFER_SIZE - sector_offset, + inlen - bytes_read); + + last_cmd = !(bytes_read + read_size < inlen); + + /* ixgbe_aci_read_nvm takes the length as a u16. Our read_size + * is calculated using a u32, but the IXGBE_ACI_MAX_BUFFER_SIZE + * maximum size guarantees that it will fit within the 2 bytes. + */ + status = ixgbe_aci_read_nvm(hw, IXGBE_ACI_NVM_START_POINT, + offset, (u16)read_size, + data + bytes_read, last_cmd, + read_shadow_ram); + if (status) + break; + + bytes_read += read_size; + offset += read_size; + } while (!last_cmd); + + *length = bytes_read; + return status; +} + +/** + * ixgbe_aci_alternate_write - write to alternate structure + * @hw: pointer to the hardware structure + * @reg_addr0: address of first dword to be written + * @reg_val0: value to be written under 'reg_addr0' + * @reg_addr1: address of second dword to be written + * @reg_val1: value to be written under 'reg_addr1' + * + * Write one or two dwords to alternate structure using ACI command (0x0900). + * Fields are indicated by 'reg_addr0' and 'reg_addr1' register numbers. + * + * Return: 0 on success and error code on failure. + */ +s32 ixgbe_aci_alternate_write(struct ixgbe_hw *hw, u32 reg_addr0, + u32 reg_val0, u32 reg_addr1, u32 reg_val1) +{ + struct ixgbe_aci_cmd_read_write_alt_direct *cmd; + struct ixgbe_aci_desc desc; + s32 status; + + cmd = &desc.params.read_write_alt_direct; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_write_alt_direct); + cmd->dword0_addr = IXGBE_CPU_TO_LE32(reg_addr0); + cmd->dword1_addr = IXGBE_CPU_TO_LE32(reg_addr1); + cmd->dword0_value = IXGBE_CPU_TO_LE32(reg_val0); + cmd->dword1_value = IXGBE_CPU_TO_LE32(reg_val1); + + status = ixgbe_aci_send_cmd(hw, &desc, NULL, 0); + + return status; +} + +/** + * ixgbe_aci_alternate_read - read from alternate structure + * @hw: pointer to the hardware structure + * @reg_addr0: address of first dword to be read + * @reg_val0: pointer for data read from 'reg_addr0' + * @reg_addr1: address of second dword to be read + * @reg_val1: pointer for data read from 'reg_addr1' + * + * Read one or two dwords from alternate structure using ACI command (0x0902). + * Fields are indicated by 'reg_addr0' and 'reg_addr1' register numbers. + * If 'reg_val1' pointer is not passed then only register at 'reg_addr0' + * is read. + * + * Return: 0 on success and error code on failure. + */ +s32 ixgbe_aci_alternate_read(struct ixgbe_hw *hw, u32 reg_addr0, + u32 *reg_val0, u32 reg_addr1, u32 *reg_val1) +{ + struct ixgbe_aci_cmd_read_write_alt_direct *cmd; + struct ixgbe_aci_desc desc; + s32 status; + + cmd = &desc.params.read_write_alt_direct; + + if (!reg_val0) + return IXGBE_ERR_PARAM; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_read_alt_direct); + cmd->dword0_addr = IXGBE_CPU_TO_LE32(reg_addr0); + cmd->dword1_addr = IXGBE_CPU_TO_LE32(reg_addr1); + + status = ixgbe_aci_send_cmd(hw, &desc, NULL, 0); + + if (status == IXGBE_SUCCESS) { + *reg_val0 = IXGBE_LE32_TO_CPU(cmd->dword0_value); + + if (reg_val1) + *reg_val1 = IXGBE_LE32_TO_CPU(cmd->dword1_value); + } + + return status; +} + +/** + * ixgbe_aci_alternate_write_done - check if writing to alternate structure + * is done + * @hw: pointer to the HW structure. + * @bios_mode: indicates whether the command is executed by UEFI or legacy BIOS + * @reset_needed: indicates the SW should trigger GLOBAL reset + * + * Indicates to the FW that alternate structures have been changed. + * + * Return: 0 on success and error code on failure. + */ +s32 ixgbe_aci_alternate_write_done(struct ixgbe_hw *hw, u8 bios_mode, + bool *reset_needed) +{ + struct ixgbe_aci_cmd_done_alt_write *cmd; + struct ixgbe_aci_desc desc; + s32 status; + + cmd = &desc.params.done_alt_write; + + if (!reset_needed) + return IXGBE_ERR_PARAM; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_done_alt_write); + cmd->flags = bios_mode; + + status = ixgbe_aci_send_cmd(hw, &desc, NULL, 0); + if (!status) + *reset_needed = (IXGBE_LE16_TO_CPU(cmd->flags) & + IXGBE_ACI_RESP_RESET_NEEDED) != 0; + + return status; +} + +/** + * ixgbe_aci_alternate_clear - clear alternate structure + * @hw: pointer to the HW structure. + * + * Clear the alternate structures of the port from which the function + * is called. + * + * Return: 0 on success and error code on failure. + */ +s32 ixgbe_aci_alternate_clear(struct ixgbe_hw *hw) +{ + struct ixgbe_aci_desc desc; + s32 status; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, + ixgbe_aci_opc_clear_port_alt_write); + + status = ixgbe_aci_send_cmd(hw, &desc, NULL, 0); + + return status; +} + +/** + * ixgbe_aci_get_internal_data - get internal FW/HW data + * @hw: pointer to the hardware structure + * @cluster_id: specific cluster to dump + * @table_id: table ID within cluster + * @start: index of line in the block to read + * @buf: dump buffer + * @buf_size: dump buffer size + * @ret_buf_size: return buffer size (returned by FW) + * @ret_next_cluster: next cluster to read (returned by FW) + * @ret_next_table: next block to read (returned by FW) + * @ret_next_index: next index to read (returned by FW) + * + * Get internal FW/HW data using ACI command (0xFF08) for debug purposes. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_aci_get_internal_data(struct ixgbe_hw *hw, u16 cluster_id, + u16 table_id, u32 start, void *buf, + u16 buf_size, u16 *ret_buf_size, + u16 *ret_next_cluster, u16 *ret_next_table, + u32 *ret_next_index) +{ + struct ixgbe_aci_cmd_debug_dump_internals *cmd; + struct ixgbe_aci_desc desc; + s32 status; + + cmd = &desc.params.debug_dump; + + if (buf_size == 0 || !buf) + return IXGBE_ERR_PARAM; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, + ixgbe_aci_opc_debug_dump_internals); + + cmd->cluster_id = IXGBE_CPU_TO_LE16(cluster_id); + cmd->table_id = IXGBE_CPU_TO_LE16(table_id); + cmd->idx = IXGBE_CPU_TO_LE32(start); + + status = ixgbe_aci_send_cmd(hw, &desc, buf, buf_size); + + if (!status) { + if (ret_buf_size) + *ret_buf_size = IXGBE_LE16_TO_CPU(desc.datalen); + if (ret_next_cluster) + *ret_next_cluster = IXGBE_LE16_TO_CPU(cmd->cluster_id); + if (ret_next_table) + *ret_next_table = IXGBE_LE16_TO_CPU(cmd->table_id); + if (ret_next_index) + *ret_next_index = IXGBE_LE32_TO_CPU(cmd->idx); + } + + return status; +} + +/** + * ixgbe_validate_nvm_rw_reg - Check that an NVM access request is valid + * @cmd: NVM access command structure + * + * Validates that an NVM access structure is request to read or write a valid + * register offset. First validates that the module and flags are correct, and + * then ensures that the register offset is one of the accepted registers. + * + * Return: 0 if the register access is valid, out of range error code otherwise. + */ +static s32 +ixgbe_validate_nvm_rw_reg(struct ixgbe_nvm_access_cmd *cmd) +{ + u16 i; + + switch (cmd->offset) { + case GL_HICR: + case GL_HICR_EN: /* Note, this register is read only */ + case GL_FWSTS: + case GL_MNG_FWSM: + case GLNVM_GENS: + case GLNVM_FLA: + case GL_FWRESETCNT: + return 0; + default: + break; + } + + for (i = 0; i <= GL_HIDA_MAX_INDEX; i++) + if (cmd->offset == (u32)GL_HIDA(i)) + return 0; + + for (i = 0; i <= GL_HIBA_MAX_INDEX; i++) + if (cmd->offset == (u32)GL_HIBA(i)) + return 0; + + /* All other register offsets are not valid */ + return IXGBE_ERR_OUT_OF_RANGE; +} + +/** + * ixgbe_nvm_access_read - Handle an NVM read request + * @hw: pointer to the HW struct + * @cmd: NVM access command to process + * @data: storage for the register value read + * + * Process an NVM access request to read a register. + * + * Return: 0 if the register read is valid and successful, + * out of range error code otherwise. + */ +static s32 ixgbe_nvm_access_read(struct ixgbe_hw *hw, + struct ixgbe_nvm_access_cmd *cmd, + struct ixgbe_nvm_access_data *data) +{ + s32 status; + + /* Always initialize the output data, even on failure */ + memset(&data->regval, 0, cmd->data_size); + + /* Make sure this is a valid read/write access request */ + status = ixgbe_validate_nvm_rw_reg(cmd); + if (status) + return status; + + DEBUGOUT1("NVM access: reading register %08x\n", cmd->offset); + + /* Read the register and store the contents in the data field */ + data->regval = IXGBE_READ_REG(hw, cmd->offset); + + return 0; +} + +/** + * ixgbe_nvm_access_write - Handle an NVM write request + * @hw: pointer to the HW struct + * @cmd: NVM access command to process + * @data: NVM access data to write + * + * Process an NVM access request to write a register. + * + * Return: 0 if the register write is valid and successful, + * out of range error code otherwise. + */ +static s32 ixgbe_nvm_access_write(struct ixgbe_hw *hw, + struct ixgbe_nvm_access_cmd *cmd, + struct ixgbe_nvm_access_data *data) +{ + s32 status; + + /* Make sure this is a valid read/write access request */ + status = ixgbe_validate_nvm_rw_reg(cmd); + if (status) + return status; + + /* Reject requests to write to read-only registers */ + switch (cmd->offset) { + case GL_HICR_EN: + return IXGBE_ERR_OUT_OF_RANGE; + default: + break; + } + + DEBUGOUT2("NVM access: writing register %08x with value %08x\n", + cmd->offset, data->regval); + + /* Write the data field to the specified register */ + IXGBE_WRITE_REG(hw, cmd->offset, data->regval); + + return 0; +} + +/** + * ixgbe_handle_nvm_access - Handle an NVM access request + * @hw: pointer to the HW struct + * @cmd: NVM access command info + * @data: pointer to read or return data + * + * Process an NVM access request. Read the command structure information and + * determine if it is valid. If not, report an error indicating the command + * was invalid. + * + * For valid commands, perform the necessary function, copying the data into + * the provided data buffer. + * + * Return: 0 if the nvm access request is valid and successful, + * error code otherwise. + */ +s32 ixgbe_handle_nvm_access(struct ixgbe_hw *hw, + struct ixgbe_nvm_access_cmd *cmd, + struct ixgbe_nvm_access_data *data) +{ + switch (cmd->command) { + case IXGBE_NVM_CMD_READ: + return ixgbe_nvm_access_read(hw, cmd, data); + case IXGBE_NVM_CMD_WRITE: + return ixgbe_nvm_access_write(hw, cmd, data); + default: + return IXGBE_ERR_PARAM; + } +} + +/** + * ixgbe_init_ops_E610 - Inits func ptrs and MAC type + * @hw: pointer to hardware structure + * + * Initialize the function pointers and assign the MAC type for E610. + * Does not touch the hardware. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_init_ops_E610(struct ixgbe_hw *hw) +{ + struct ixgbe_eeprom_info *eeprom = &hw->eeprom; + struct ixgbe_mac_info *mac = &hw->mac; + struct ixgbe_phy_info *phy = &hw->phy; + s32 ret_val; + + ret_val = ixgbe_init_ops_X550(hw); + /* TODO Additional ops overrides for e610 to go here */ + + /* MAC */ + mac->ops.reset_hw = ixgbe_reset_hw_E610; + mac->ops.start_hw = ixgbe_start_hw_E610; + mac->ops.get_media_type = ixgbe_get_media_type_E610; + mac->ops.get_supported_physical_layer = + ixgbe_get_supported_physical_layer_E610; + mac->ops.get_san_mac_addr = NULL; + mac->ops.set_san_mac_addr = NULL; + mac->ops.get_wwn_prefix = NULL; + mac->ops.setup_link = ixgbe_setup_link_E610; + mac->ops.check_link = ixgbe_check_link_E610; + mac->ops.get_link_capabilities = ixgbe_get_link_capabilities_E610; + mac->ops.setup_fc = ixgbe_setup_fc_E610; + mac->ops.fc_autoneg = ixgbe_fc_autoneg_E610; + mac->ops.set_fw_drv_ver = ixgbe_set_fw_drv_ver_E610; + mac->ops.disable_rx = ixgbe_disable_rx_E610; + mac->ops.setup_eee = ixgbe_setup_eee_E610; + mac->ops.fw_recovery_mode = ixgbe_fw_recovery_mode_E610; + mac->ops.get_fw_tsam_mode = ixgbe_get_fw_tsam_mode_E610; + mac->ops.get_fw_version = ixgbe_aci_get_fw_ver; + mac->ops.get_nvm_version = ixgbe_get_active_nvm_ver; + + /* PHY */ + phy->ops.init = ixgbe_init_phy_ops_E610; + phy->ops.identify = ixgbe_identify_phy_E610; + phy->eee_speeds_supported = IXGBE_LINK_SPEED_10_FULL | + IXGBE_LINK_SPEED_100_FULL | + IXGBE_LINK_SPEED_1GB_FULL; + phy->eee_speeds_advertised = phy->eee_speeds_supported; + + /* Additional ops overrides for e610 to go here */ + eeprom->ops.init_params = ixgbe_init_eeprom_params_E610; + eeprom->ops.read = ixgbe_read_ee_aci_E610; + eeprom->ops.read_buffer = ixgbe_read_ee_aci_buffer_E610; + eeprom->ops.write = NULL; + eeprom->ops.write_buffer = NULL; + eeprom->ops.calc_checksum = ixgbe_calc_eeprom_checksum_E610; + eeprom->ops.update_checksum = NULL; + eeprom->ops.validate_checksum = ixgbe_validate_eeprom_checksum_E610; + eeprom->ops.read_pba_string = ixgbe_read_pba_string_E610; + + /* Initialize bus function number */ + hw->mac.ops.set_lan_id(hw); + + return ret_val; +} + +/** + * ixgbe_reset_hw_E610 - Perform hardware reset + * @hw: pointer to hardware structure + * + * Resets the hardware by resetting the transmit and receive units, masks + * and clears all interrupts, and perform a reset. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_reset_hw_E610(struct ixgbe_hw *hw) +{ + u32 swfw_mask = hw->phy.phy_semaphore_mask; + u32 ctrl, i; + s32 status; + + DEBUGFUNC("ixgbe_reset_hw_E610"); + + /* Call adapter stop to disable tx/rx and clear interrupts */ + status = hw->mac.ops.stop_adapter(hw); + if (status != IXGBE_SUCCESS) + goto reset_hw_out; + + /* flush pending Tx transactions */ + ixgbe_clear_tx_pending(hw); + + status = hw->phy.ops.init(hw); + if (status != IXGBE_SUCCESS) + DEBUGOUT1("Failed to initialize PHY ops, STATUS = %d\n", + status); +mac_reset_top: + status = hw->mac.ops.acquire_swfw_sync(hw, swfw_mask); + if (status != IXGBE_SUCCESS) { + ERROR_REPORT2(IXGBE_ERROR_CAUTION, + "semaphore failed with %d", status); + return IXGBE_ERR_SWFW_SYNC; + } + ctrl = IXGBE_CTRL_RST; + ctrl |= IXGBE_READ_REG(hw, IXGBE_CTRL); + IXGBE_WRITE_REG(hw, IXGBE_CTRL, ctrl); + IXGBE_WRITE_FLUSH(hw); + hw->mac.ops.release_swfw_sync(hw, swfw_mask); + + /* Poll for reset bit to self-clear indicating reset is complete */ + for (i = 0; i < 10; i++) { + usec_delay(1); + ctrl = IXGBE_READ_REG(hw, IXGBE_CTRL); + if (!(ctrl & IXGBE_CTRL_RST_MASK)) + break; + } + + if (ctrl & IXGBE_CTRL_RST_MASK) { + status = IXGBE_ERR_RESET_FAILED; + ERROR_REPORT1(IXGBE_ERROR_POLLING, + "Reset polling failed to complete.\n"); + } + msec_delay(100); + + /* + * Double resets are required for recovery from certain error + * conditions. Between resets, it is necessary to stall to allow time + * for any pending HW events to complete. + */ + if (hw->mac.flags & IXGBE_FLAGS_DOUBLE_RESET_REQUIRED) { + hw->mac.flags &= ~IXGBE_FLAGS_DOUBLE_RESET_REQUIRED; + goto mac_reset_top; + } + + /* Set the Rx packet buffer size. */ + IXGBE_WRITE_REG(hw, IXGBE_RXPBSIZE(0), 384 << IXGBE_RXPBSIZE_SHIFT); + + /* Store the permanent mac address */ + hw->mac.ops.get_mac_addr(hw, hw->mac.perm_addr); + + /* + * Store MAC address from RAR0, clear receive address registers, and + * clear the multicast table. Also reset num_rar_entries to 128, + * since we modify this value when programming the SAN MAC address. + */ + hw->mac.num_rar_entries = 128; + hw->mac.ops.init_rx_addrs(hw); + +reset_hw_out: + return status; +} +/** + * ixgbe_fw_ver_check - Check the reported FW API version + * @hw: pointer to the hardware structure + * + * Checks if the driver should load on a given FW API version. + * + * Return: 'true' if the driver should attempt to load. 'false' otherwise. + */ +static bool ixgbe_fw_ver_check(struct ixgbe_hw *hw) +{ + if (hw->api_maj_ver > IXGBE_FW_API_VER_MAJOR) { + ERROR_REPORT1(IXGBE_ERROR_UNSUPPORTED, "The driver for the device stopped because the NVM image is newer than expected. You must install the most recent version of the network driver.\n"); + return false; + } else if (hw->api_maj_ver == IXGBE_FW_API_VER_MAJOR) { + if (hw->api_min_ver > + (IXGBE_FW_API_VER_MINOR + IXGBE_FW_API_VER_DIFF_ALLOWED)) { + ERROR_REPORT1(IXGBE_ERROR_CAUTION, "The driver for the device detected a newer version of the NVM image than expected. Please install the most recent version of the network driver.\n"); + } else if ((hw->api_min_ver + IXGBE_FW_API_VER_DIFF_ALLOWED) < + IXGBE_FW_API_VER_MINOR) { + ERROR_REPORT1(IXGBE_ERROR_CAUTION, "The driver for the device detected an older version of the NVM image than expected. Please update the NVM image.\n"); + } + } else { + ERROR_REPORT1(IXGBE_ERROR_CAUTION, "The driver for the device detected an older version of the NVM image than expected. Please update the NVM image.\n"); + } + return true; +} +/** + * ixgbe_start_hw_E610 - Prepare hardware for Tx/Rx + * @hw: pointer to hardware structure + * + * Gets firmware version and if API version matches it + * starts the hardware using the generic start_hw function + * and the generation start_hw function. + * Then performs revision-specific operations, if any. + **/ +s32 ixgbe_start_hw_E610(struct ixgbe_hw *hw) +{ + s32 ret_val = IXGBE_SUCCESS; + + ret_val = hw->mac.ops.get_fw_version(hw); + if (ret_val) + goto out; + + if (!ixgbe_fw_ver_check(hw)) { + ret_val = IXGBE_ERR_FW_API_VER; + goto out; + } + ret_val = ixgbe_start_hw_generic(hw); + if (ret_val != IXGBE_SUCCESS) + goto out; + + ixgbe_start_hw_gen2(hw); + +out: + return ret_val; +} + +/** + * ixgbe_get_media_type_E610 - Gets media type + * @hw: pointer to the HW struct + * + * In order to get the media type, the function gets PHY + * capabilities and later on use them to identify the PHY type + * checking phy_type_high and phy_type_low. + * + * Return: the type of media in form of ixgbe_media_type enum + * or ixgbe_media_type_unknown in case of an error. + */ +enum ixgbe_media_type ixgbe_get_media_type_E610(struct ixgbe_hw *hw) +{ + struct ixgbe_aci_cmd_get_phy_caps_data pcaps; + u64 phy_mask = 0; + s32 rc; + u8 i; + + rc = ixgbe_update_link_info(hw); + if (rc) { + return ixgbe_media_type_unknown; + } + + /* If there is no link but PHY (dongle) is available SW should use + * Get PHY Caps admin command instead of Get Link Status, find most + * significant bit that is set in PHY types reported by the command + * and use it to discover media type. + */ + if (!(hw->link.link_info.link_info & IXGBE_ACI_LINK_UP) && + (hw->link.link_info.link_info & IXGBE_ACI_MEDIA_AVAILABLE)) { + /* Get PHY Capabilities */ + rc = ixgbe_aci_get_phy_caps(hw, false, + IXGBE_ACI_REPORT_TOPO_CAP_MEDIA, + &pcaps); + if (rc) { + return ixgbe_media_type_unknown; + } + + /* Check if there is some bit set in phy_type_high */ + for (i = 64; i > 0; i--) { + phy_mask = (u64)((u64)1 << (i - 1)); + if ((pcaps.phy_type_high & phy_mask) != 0) { + /* If any bit is set treat it as PHY type */ + hw->link.link_info.phy_type_high = phy_mask; + hw->link.link_info.phy_type_low = 0; + break; + } + phy_mask = 0; + } + + /* If nothing found in phy_type_high search in phy_type_low */ + if (phy_mask == 0) { + for (i = 64; i > 0; i--) { + phy_mask = (u64)((u64)1 << (i - 1)); + if ((pcaps.phy_type_low & phy_mask) != 0) { + /* If any bit is set treat it as PHY type */ + hw->link.link_info.phy_type_high = 0; + hw->link.link_info.phy_type_low = phy_mask; + break; + } + } + } + + /* Based on search above try to discover media type */ + hw->phy.media_type = ixgbe_get_media_type_from_phy_type(hw); + } + + return hw->phy.media_type; +} + +/** + * ixgbe_get_supported_physical_layer_E610 - Returns physical layer type + * @hw: pointer to hardware structure + * + * Determines physical layer capabilities of the current configuration. + * + * Return: the exit code of the operation. + **/ +u64 ixgbe_get_supported_physical_layer_E610(struct ixgbe_hw *hw) +{ + u64 physical_layer = IXGBE_PHYSICAL_LAYER_UNKNOWN; + struct ixgbe_aci_cmd_get_phy_caps_data pcaps; + u64 phy_type; + s32 rc; + + rc = ixgbe_aci_get_phy_caps(hw, false, IXGBE_ACI_REPORT_TOPO_CAP_MEDIA, + &pcaps); + if (rc) + return IXGBE_PHYSICAL_LAYER_UNKNOWN; + + phy_type = IXGBE_LE64_TO_CPU(pcaps.phy_type_low); + if(phy_type & IXGBE_PHY_TYPE_LOW_10GBASE_T) + physical_layer |= IXGBE_PHYSICAL_LAYER_10GBASE_T; + if(phy_type & IXGBE_PHY_TYPE_LOW_1000BASE_T) + physical_layer |= IXGBE_PHYSICAL_LAYER_1000BASE_T; + if(phy_type & IXGBE_PHY_TYPE_LOW_100BASE_TX) + physical_layer |= IXGBE_PHYSICAL_LAYER_100BASE_TX; + if(phy_type & IXGBE_PHY_TYPE_LOW_10GBASE_LR) + physical_layer |= IXGBE_PHYSICAL_LAYER_10GBASE_LR; + if(phy_type & IXGBE_PHY_TYPE_LOW_10GBASE_SR) + physical_layer |= IXGBE_PHYSICAL_LAYER_10GBASE_SR; + if(phy_type & IXGBE_PHY_TYPE_LOW_1000BASE_KX) + physical_layer |= IXGBE_PHYSICAL_LAYER_1000BASE_KX; + if(phy_type & IXGBE_PHY_TYPE_LOW_10GBASE_KR_CR1) + physical_layer |= IXGBE_PHYSICAL_LAYER_10GBASE_KR; + if(phy_type & IXGBE_PHY_TYPE_LOW_1000BASE_SX) + physical_layer |= IXGBE_PHYSICAL_LAYER_1000BASE_SX; + if(phy_type & IXGBE_PHY_TYPE_LOW_2500BASE_KX) + physical_layer |= IXGBE_PHYSICAL_LAYER_2500BASE_KX; + if(phy_type & IXGBE_PHY_TYPE_LOW_2500BASE_T) + physical_layer |= IXGBE_PHYSICAL_LAYER_2500BASE_T; + if(phy_type & IXGBE_PHY_TYPE_LOW_5GBASE_T) + physical_layer |= IXGBE_PHYSICAL_LAYER_5000BASE_T; + + phy_type = IXGBE_LE64_TO_CPU(pcaps.phy_type_high); + if(phy_type & IXGBE_PHY_TYPE_HIGH_10BASE_T) + physical_layer |= IXGBE_PHYSICAL_LAYER_10BASE_T; + + return physical_layer; +} + +/** + * ixgbe_setup_link_E610 - Set up link + * @hw: pointer to hardware structure + * @speed: new link speed + * @autoneg_wait: true when waiting for completion is needed + * + * Set up the link with the specified speed. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_setup_link_E610(struct ixgbe_hw *hw, ixgbe_link_speed speed, + bool autoneg_wait) +{ + + /* Simply request FW to perform proper PHY setup */ + return hw->phy.ops.setup_link_speed(hw, speed, autoneg_wait); +} + +/** + * ixgbe_check_link_E610 - Determine link and speed status + * @hw: pointer to hardware structure + * @speed: pointer to link speed + * @link_up: true when link is up + * @link_up_wait_to_complete: bool used to wait for link up or not + * + * Determine if the link is up and the current link speed + * using ACI command (0x0607). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_check_link_E610(struct ixgbe_hw *hw, ixgbe_link_speed *speed, + bool *link_up, bool link_up_wait_to_complete) +{ + s32 rc; + u32 i; + + if (!speed || !link_up) + return IXGBE_ERR_PARAM; + + /* Set get_link_info flag to ensure that fresh + * link information will be obtained from FW + * by sending Get Link Status admin command. */ + hw->link.get_link_info = true; + + /* Update link information in adapter context. */ + rc = ixgbe_get_link_status(hw, link_up); + if (rc) + return rc; + + /* Wait for link up if it was requested. */ + if (link_up_wait_to_complete && *link_up == false) { + for (i = 0; i < hw->mac.max_link_up_time; i++) { + msec_delay(100); + hw->link.get_link_info = true; + rc = ixgbe_get_link_status(hw, link_up); + if (rc) + return rc; + if (*link_up) + break; + } + } + + /* Use link information in adapter context updated by the call + * to ixgbe_get_link_status() to determine current link speed. + * Link speed information is valid only when link up was + * reported by FW. */ + if (*link_up) { + switch (hw->link.link_info.link_speed) { + case IXGBE_ACI_LINK_SPEED_10MB: + *speed = IXGBE_LINK_SPEED_10_FULL; + break; + case IXGBE_ACI_LINK_SPEED_100MB: + *speed = IXGBE_LINK_SPEED_100_FULL; + break; + case IXGBE_ACI_LINK_SPEED_1000MB: + *speed = IXGBE_LINK_SPEED_1GB_FULL; + break; + case IXGBE_ACI_LINK_SPEED_2500MB: + *speed = IXGBE_LINK_SPEED_2_5GB_FULL; + break; + case IXGBE_ACI_LINK_SPEED_5GB: + *speed = IXGBE_LINK_SPEED_5GB_FULL; + break; + case IXGBE_ACI_LINK_SPEED_10GB: + *speed = IXGBE_LINK_SPEED_10GB_FULL; + break; + default: + *speed = IXGBE_LINK_SPEED_UNKNOWN; + break; + } + } else { + *speed = IXGBE_LINK_SPEED_UNKNOWN; + } + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_get_link_capabilities_E610 - Determine link capabilities + * @hw: pointer to hardware structure + * @speed: pointer to link speed + * @autoneg: true when autoneg or autotry is enabled + * + * Determine speed and AN parameters of a link. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_get_link_capabilities_E610(struct ixgbe_hw *hw, + ixgbe_link_speed *speed, + bool *autoneg) +{ + + if (!speed || !autoneg) + return IXGBE_ERR_PARAM; + + *autoneg = true; + *speed = hw->phy.speeds_supported; + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_cfg_phy_fc - Configure PHY Flow Control (FC) data based on FC mode + * @hw: pointer to hardware structure + * @cfg: PHY configuration data to set FC mode + * @req_mode: FC mode to configure + * + * Configures PHY Flow Control according to the provided configuration. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_cfg_phy_fc(struct ixgbe_hw *hw, + struct ixgbe_aci_cmd_set_phy_cfg_data *cfg, + enum ixgbe_fc_mode req_mode) +{ + struct ixgbe_aci_cmd_get_phy_caps_data* pcaps = NULL; + s32 status = IXGBE_SUCCESS; + u8 pause_mask = 0x0; + + if (!cfg) + return IXGBE_ERR_PARAM; + + switch (req_mode) { + case ixgbe_fc_auto: + { + pcaps = (struct ixgbe_aci_cmd_get_phy_caps_data *) + ixgbe_malloc(hw, sizeof(*pcaps)); + if (!pcaps) { + status = IXGBE_ERR_OUT_OF_MEM; + goto out; + } + + /* Query the value of FC that both the NIC and the attached + * media can do. */ + status = ixgbe_aci_get_phy_caps(hw, false, + IXGBE_ACI_REPORT_TOPO_CAP_MEDIA, pcaps); + if (status) + goto out; + + pause_mask |= pcaps->caps & IXGBE_ACI_PHY_EN_TX_LINK_PAUSE; + pause_mask |= pcaps->caps & IXGBE_ACI_PHY_EN_RX_LINK_PAUSE; + + break; + } + case ixgbe_fc_full: + pause_mask |= IXGBE_ACI_PHY_EN_TX_LINK_PAUSE; + pause_mask |= IXGBE_ACI_PHY_EN_RX_LINK_PAUSE; + break; + case ixgbe_fc_rx_pause: + pause_mask |= IXGBE_ACI_PHY_EN_RX_LINK_PAUSE; + break; + case ixgbe_fc_tx_pause: + pause_mask |= IXGBE_ACI_PHY_EN_TX_LINK_PAUSE; + break; + default: + break; + } + + /* clear the old pause settings */ + cfg->caps &= ~(IXGBE_ACI_PHY_EN_TX_LINK_PAUSE | + IXGBE_ACI_PHY_EN_RX_LINK_PAUSE); + + /* set the new capabilities */ + cfg->caps |= pause_mask; + +out: + if (pcaps) + ixgbe_free(hw, pcaps); + return status; +} + +/** + * ixgbe_setup_fc_E610 - Set up flow control + * @hw: pointer to hardware structure + * + * Set up flow control. This has to be done during init time. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_setup_fc_E610(struct ixgbe_hw *hw) +{ + struct ixgbe_aci_cmd_get_phy_caps_data pcaps = { 0 }; + struct ixgbe_aci_cmd_set_phy_cfg_data cfg = { 0 }; + s32 status; + + /* Get the current PHY config */ + status = ixgbe_aci_get_phy_caps(hw, false, + IXGBE_ACI_REPORT_ACTIVE_CFG, &pcaps); + if (status) + return status; + + ixgbe_copy_phy_caps_to_cfg(&pcaps, &cfg); + + /* Configure the set PHY data */ + status = ixgbe_cfg_phy_fc(hw, &cfg, hw->fc.requested_mode); + if (status) + return status; + + /* If the capabilities have changed, then set the new config */ + if (cfg.caps != pcaps.caps) { + cfg.caps |= IXGBE_ACI_PHY_ENA_AUTO_LINK_UPDT; + + status = ixgbe_aci_set_phy_cfg(hw, &cfg); + if (status) + return status; + } + + return status; +} + +/** + * ixgbe_fc_autoneg_E610 - Configure flow control + * @hw: pointer to hardware structure + * + * Configure Flow Control. + */ +void ixgbe_fc_autoneg_E610(struct ixgbe_hw *hw) +{ + s32 status; + + /* Get current link status. + * Current FC mode will be stored in the hw context. */ + status = ixgbe_aci_get_link_info(hw, false, NULL); + if (status) { + goto out; + } + + /* Check if the link is up */ + if (!(hw->link.link_info.link_info & IXGBE_ACI_LINK_UP)) { + status = IXGBE_ERR_FC_NOT_NEGOTIATED; + goto out; + } + + /* Check if auto-negotiation has completed */ + if (!(hw->link.link_info.an_info & IXGBE_ACI_AN_COMPLETED)) { + status = IXGBE_ERR_FC_NOT_NEGOTIATED; + goto out; + } + +out: + if (status == IXGBE_SUCCESS) { + hw->fc.fc_was_autonegged = true; + } else { + hw->fc.fc_was_autonegged = false; + hw->fc.current_mode = hw->fc.requested_mode; + } +} + +/** + * ixgbe_set_fw_drv_ver_E610 - Send driver version to FW + * @hw: pointer to the HW structure + * @maj: driver version major number + * @minor: driver version minor number + * @build: driver version build number + * @sub: driver version sub build number + * @len: length of driver_ver string + * @driver_ver: driver string + * + * Send driver version number to Firmware using ACI command (0x0002). + * + * Return: the exit code of the operation. + * IXGBE_SUCCESS - OK + * IXGBE_ERR_PARAM - incorrect parameters were given + * IXGBE_ERR_ACI_ERROR - encountered an error during sending the command + * IXGBE_ERR_ACI_TIMEOUT - a timeout occurred + * IXGBE_ERR_OUT_OF_MEM - ran out of memory + */ +s32 ixgbe_set_fw_drv_ver_E610(struct ixgbe_hw *hw, u8 maj, u8 minor, u8 build, + u8 sub, u16 len, const char *driver_ver) +{ + size_t limited_len = min(len, (u16)IXGBE_DRV_VER_STR_LEN_E610); + struct ixgbe_driver_ver dv; + + DEBUGFUNC("ixgbe_set_fw_drv_ver_E610"); + + if (!len || !driver_ver) + return IXGBE_ERR_PARAM; + + dv.major_ver = maj; + dv.minor_ver = minor; + dv.build_ver = build; + dv.subbuild_ver = sub; + + memset(dv.driver_string, 0, IXGBE_DRV_VER_STR_LEN_E610); + memcpy(dv.driver_string, driver_ver, limited_len); + + return ixgbe_aci_send_driver_ver(hw, &dv); +} + +/** + * ixgbe_disable_rx_E610 - Disable RX unit + * @hw: pointer to hardware structure + * + * Disable RX DMA unit on E610 with use of ACI command (0x000C). + * + * Return: the exit code of the operation. + */ +void ixgbe_disable_rx_E610(struct ixgbe_hw *hw) +{ + u32 rxctrl; + + DEBUGFUNC("ixgbe_disable_rx_E610"); + + rxctrl = IXGBE_READ_REG(hw, IXGBE_RXCTRL); + if (rxctrl & IXGBE_RXCTRL_RXEN) { + u32 pfdtxgswc; + s32 status; + + pfdtxgswc = IXGBE_READ_REG(hw, IXGBE_PFDTXGSWC); + if (pfdtxgswc & IXGBE_PFDTXGSWC_VT_LBEN) { + pfdtxgswc &= ~IXGBE_PFDTXGSWC_VT_LBEN; + IXGBE_WRITE_REG(hw, IXGBE_PFDTXGSWC, pfdtxgswc); + hw->mac.set_lben = true; + } else { + hw->mac.set_lben = false; + } + + status = ixgbe_aci_disable_rxen(hw); + + /* If we fail - disable RX using register write */ + if (status) { + rxctrl = IXGBE_READ_REG(hw, IXGBE_RXCTRL); + if (rxctrl & IXGBE_RXCTRL_RXEN) { + rxctrl &= ~IXGBE_RXCTRL_RXEN; + IXGBE_WRITE_REG(hw, IXGBE_RXCTRL, rxctrl); + } + } + } +} + +/** + * ixgbe_setup_eee_E610 - Enable/disable EEE support + * @hw: pointer to the HW structure + * @enable_eee: boolean flag to enable EEE + * + * Enables/disable EEE based on enable_eee flag. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_setup_eee_E610(struct ixgbe_hw *hw, bool enable_eee) +{ + struct ixgbe_aci_cmd_get_phy_caps_data phy_caps = { 0 }; + struct ixgbe_aci_cmd_set_phy_cfg_data phy_cfg = { 0 }; + u16 eee_cap = 0; + s32 status; + + status = ixgbe_aci_get_phy_caps(hw, false, + IXGBE_ACI_REPORT_ACTIVE_CFG, &phy_caps); + if (status != IXGBE_SUCCESS) + return status; + + ixgbe_copy_phy_caps_to_cfg(&phy_caps, &phy_cfg); + + phy_cfg.caps |= IXGBE_ACI_PHY_ENA_LINK; + phy_cfg.caps |= IXGBE_ACI_PHY_ENA_AUTO_LINK_UPDT; + + if (enable_eee) { + if (phy_caps.phy_type_low & IXGBE_PHY_TYPE_LOW_100BASE_TX) + eee_cap |= IXGBE_ACI_PHY_EEE_EN_100BASE_TX; + if (phy_caps.phy_type_low & IXGBE_PHY_TYPE_LOW_1000BASE_T) + eee_cap |= IXGBE_ACI_PHY_EEE_EN_1000BASE_T; + if (phy_caps.phy_type_low & IXGBE_PHY_TYPE_LOW_1000BASE_KX) + eee_cap |= IXGBE_ACI_PHY_EEE_EN_1000BASE_KX; + if (phy_caps.phy_type_low & IXGBE_PHY_TYPE_LOW_10GBASE_T) + eee_cap |= IXGBE_ACI_PHY_EEE_EN_10GBASE_T; + if (phy_caps.phy_type_low & IXGBE_PHY_TYPE_LOW_10GBASE_KR_CR1) + eee_cap |= IXGBE_ACI_PHY_EEE_EN_10GBASE_KR; + if (phy_caps.phy_type_low & IXGBE_PHY_TYPE_LOW_25GBASE_KR || + phy_caps.phy_type_low & IXGBE_PHY_TYPE_LOW_25GBASE_KR_S || + phy_caps.phy_type_low & IXGBE_PHY_TYPE_LOW_25GBASE_KR1) + eee_cap |= IXGBE_ACI_PHY_EEE_EN_25GBASE_KR; + + if (phy_caps.phy_type_high & IXGBE_PHY_TYPE_HIGH_10BASE_T) + eee_cap |= IXGBE_ACI_PHY_EEE_EN_10BASE_T; + } + + /* Set EEE capability for particular PHY types */ + phy_cfg.eee_cap = IXGBE_CPU_TO_LE16(eee_cap); + + status = ixgbe_aci_set_phy_cfg(hw, &phy_cfg); + + return status; +} + +/** + * ixgbe_fw_recovery_mode_E610 - Check FW NVM recovery mode + * @hw: pointer to hardware structure + * + * Checks FW NVM recovery mode by + * reading the value of the dedicated register. + * + * Return: true if FW is in recovery mode, otherwise false. + */ +bool ixgbe_fw_recovery_mode_E610(struct ixgbe_hw *hw) +{ + u32 fwsm = IXGBE_READ_REG(hw, GL_MNG_FWSM); + + return !!(fwsm & GL_MNG_FWSM_FW_MODES_RECOVERY_M); +} + +/** + * ixgbe_get_fw_tsam_mode_E610 - Check FW NVM Thermal Sensor Autonomous Mode + * @hw: pointer to hardware structure + * + * Checks Thermal Sensor Autonomous Mode by reading the + * value of the dedicated register. + * + * Return: true if FW is in TSAM, otherwise false. + */ +bool ixgbe_get_fw_tsam_mode_E610(struct ixgbe_hw *hw) +{ + u32 fwsm = IXGBE_READ_REG(hw, IXGBE_FWSM_X550EM_a); + + return !!(fwsm & IXGBE_FWSM_TS_ENABLED); +} + +/** + * ixgbe_init_phy_ops_E610 - PHY specific init + * @hw: pointer to hardware structure + * + * Initialize any function pointers that were not able to be + * set during init_shared_code because the PHY type was not known. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_init_phy_ops_E610(struct ixgbe_hw *hw) +{ + struct ixgbe_mac_info *mac = &hw->mac; + struct ixgbe_phy_info *phy = &hw->phy; + s32 ret_val; + + phy->ops.identify_sfp = ixgbe_identify_module_E610; + phy->ops.read_reg = NULL; /* PHY reg access is not required */ + phy->ops.write_reg = NULL; + phy->ops.read_reg_mdi = NULL; + phy->ops.write_reg_mdi = NULL; + phy->ops.setup_link = ixgbe_setup_phy_link_E610; + phy->ops.get_firmware_version = ixgbe_get_phy_firmware_version_E610; + phy->ops.read_i2c_byte = NULL; /* disabled for E610 */ + phy->ops.write_i2c_byte = NULL; /* disabled for E610 */ + phy->ops.read_i2c_sff8472 = ixgbe_read_i2c_sff8472_E610; + phy->ops.read_i2c_eeprom = ixgbe_read_i2c_eeprom_E610; + phy->ops.write_i2c_eeprom = ixgbe_write_i2c_eeprom_E610; + phy->ops.i2c_bus_clear = NULL; /* do not use generic implementation */ + phy->ops.check_overtemp = ixgbe_check_overtemp_E610; + if (mac->ops.get_media_type(hw) == ixgbe_media_type_copper) + phy->ops.set_phy_power = ixgbe_set_phy_power_E610; + else + phy->ops.set_phy_power = NULL; + phy->ops.enter_lplu = ixgbe_enter_lplu_E610; + phy->ops.handle_lasi = NULL; /* no implementation for E610 */ + phy->ops.read_i2c_byte_unlocked = NULL; /* disabled for E610 */ + phy->ops.write_i2c_byte_unlocked = NULL; /* disabled for E610 */ + + /* TODO: Set functions pointers based on device ID */ + + /* Identify the PHY */ + ret_val = phy->ops.identify(hw); + if (ret_val != IXGBE_SUCCESS) + return ret_val; + + /* TODO: Set functions pointers based on PHY type */ + + return ret_val; +} + +/** + * ixgbe_identify_phy_E610 - Identify PHY + * @hw: pointer to hardware structure + * + * Determine PHY type, supported speeds and PHY ID. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_identify_phy_E610(struct ixgbe_hw *hw) +{ + struct ixgbe_aci_cmd_get_phy_caps_data pcaps; + s32 rc; + + /* Set PHY type */ + hw->phy.type = ixgbe_phy_fw; + + rc = ixgbe_aci_get_phy_caps(hw, false, IXGBE_ACI_REPORT_TOPO_CAP_MEDIA, + &pcaps); + if (rc) + return rc; + + if (!(pcaps.module_compliance_enforcement & + IXGBE_ACI_MOD_ENFORCE_STRICT_MODE)) { + /* Handle lenient mode */ + rc = ixgbe_aci_get_phy_caps(hw, false, + IXGBE_ACI_REPORT_TOPO_CAP_NO_MEDIA, + &pcaps); + if (rc) + return rc; + } + + /* Determine supported speeds */ + hw->phy.speeds_supported = IXGBE_LINK_SPEED_UNKNOWN; + + if (pcaps.phy_type_high & IXGBE_PHY_TYPE_HIGH_10BASE_T || + pcaps.phy_type_high & IXGBE_PHY_TYPE_HIGH_10M_SGMII) + hw->phy.speeds_supported |= IXGBE_LINK_SPEED_10_FULL; + if (pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_100BASE_TX || + pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_100M_SGMII || + pcaps.phy_type_high & IXGBE_PHY_TYPE_HIGH_100M_USXGMII) + hw->phy.speeds_supported |= IXGBE_LINK_SPEED_100_FULL; + if (pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_1000BASE_T || + pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_1000BASE_SX || + pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_1000BASE_LX || + pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_1000BASE_KX || + pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_1G_SGMII || + pcaps.phy_type_high & IXGBE_PHY_TYPE_HIGH_1G_USXGMII) + hw->phy.speeds_supported |= IXGBE_LINK_SPEED_1GB_FULL; + if (pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_2500BASE_T || + pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_2500BASE_X || + pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_2500BASE_KX || + pcaps.phy_type_high & IXGBE_PHY_TYPE_HIGH_2500M_SGMII || + pcaps.phy_type_high & IXGBE_PHY_TYPE_HIGH_2500M_USXGMII) + hw->phy.speeds_supported |= IXGBE_LINK_SPEED_2_5GB_FULL; + if (pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_5GBASE_T || + pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_5GBASE_KR || + pcaps.phy_type_high & IXGBE_PHY_TYPE_HIGH_5G_USXGMII) + hw->phy.speeds_supported |= IXGBE_LINK_SPEED_5GB_FULL; + if (pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_10GBASE_T || + pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_10G_SFI_DA || + pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_10GBASE_SR || + pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_10GBASE_LR || + pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_10GBASE_KR_CR1 || + pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_10G_SFI_AOC_ACC || + pcaps.phy_type_low & IXGBE_PHY_TYPE_LOW_10G_SFI_C2C || + pcaps.phy_type_high & IXGBE_PHY_TYPE_HIGH_10G_USXGMII) + hw->phy.speeds_supported |= IXGBE_LINK_SPEED_10GB_FULL; + + /* Initialize autoneg speeds */ + if (!hw->phy.autoneg_advertised) + hw->phy.autoneg_advertised = hw->phy.speeds_supported; + + /* Set PHY ID */ + memcpy(&hw->phy.id, pcaps.phy_id_oui, sizeof(u32)); + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_identify_module_E610 - Identify SFP module type + * @hw: pointer to hardware structure + * + * Identify the SFP module type. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_identify_module_E610(struct ixgbe_hw *hw) +{ + bool media_available; + u8 module_type; + s32 rc; + + rc = ixgbe_update_link_info(hw); + if (rc) + goto err; + + media_available = + (hw->link.link_info.link_info & + IXGBE_ACI_MEDIA_AVAILABLE) ? true : false; + + if (media_available) { + hw->phy.sfp_type = ixgbe_sfp_type_unknown; + + /* Get module type from hw context updated by ixgbe_update_link_info() */ + module_type = hw->link.link_info.module_type[IXGBE_ACI_MOD_TYPE_IDENT]; + + if ((module_type & IXGBE_ACI_MOD_TYPE_BYTE1_SFP_PLUS_CU_PASSIVE) || + (module_type & IXGBE_ACI_MOD_TYPE_BYTE1_SFP_PLUS_CU_ACTIVE)) { + hw->phy.sfp_type = ixgbe_sfp_type_da_cu; + } else if (module_type & IXGBE_ACI_MOD_TYPE_BYTE1_10G_BASE_SR) { + hw->phy.sfp_type = ixgbe_sfp_type_sr; + } else if ((module_type & IXGBE_ACI_MOD_TYPE_BYTE1_10G_BASE_LR) || + (module_type & IXGBE_ACI_MOD_TYPE_BYTE1_10G_BASE_LRM)) { + hw->phy.sfp_type = ixgbe_sfp_type_lr; + } + rc = IXGBE_SUCCESS; + } else { + hw->phy.sfp_type = ixgbe_sfp_type_not_present; + rc = IXGBE_ERR_SFP_NOT_PRESENT; + } +err: + return rc; +} + +/** + * ixgbe_setup_phy_link_E610 - Sets up firmware-controlled PHYs + * @hw: pointer to hardware structure + * + * Set the parameters for the firmware-controlled PHYs. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_setup_phy_link_E610(struct ixgbe_hw *hw) +{ + struct ixgbe_aci_cmd_get_phy_caps_data pcaps; + struct ixgbe_aci_cmd_set_phy_cfg_data pcfg; + u8 rmode = IXGBE_ACI_REPORT_ACTIVE_CFG; + s32 rc; + + rc = ixgbe_aci_get_link_info(hw, false, NULL); + if (rc) { + goto err; + } + + /* If media is not available get default config */ + if (!(hw->link.link_info.link_info & IXGBE_ACI_MEDIA_AVAILABLE)) + rmode = IXGBE_ACI_REPORT_DFLT_CFG; + + rc = ixgbe_aci_get_phy_caps(hw, false, rmode, &pcaps); + if (rc) { + goto err; + } + + ixgbe_copy_phy_caps_to_cfg(&pcaps, &pcfg); + + /* Set default PHY types for a given speed */ + pcfg.phy_type_low = 0; + pcfg.phy_type_high = 0; + + if (hw->phy.autoneg_advertised & IXGBE_LINK_SPEED_10_FULL) { + pcfg.phy_type_high |= IXGBE_PHY_TYPE_HIGH_10BASE_T; + pcfg.phy_type_high |= IXGBE_PHY_TYPE_HIGH_10M_SGMII; + } + if (hw->phy.autoneg_advertised & IXGBE_LINK_SPEED_100_FULL) { + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_100BASE_TX; + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_100M_SGMII; + pcfg.phy_type_high |= IXGBE_PHY_TYPE_HIGH_100M_USXGMII; + } + if (hw->phy.autoneg_advertised & IXGBE_LINK_SPEED_1GB_FULL) { + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_1000BASE_T; + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_1000BASE_SX; + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_1000BASE_LX; + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_1000BASE_KX; + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_1G_SGMII; + pcfg.phy_type_high |= IXGBE_PHY_TYPE_HIGH_1G_USXGMII; + } + if (hw->phy.autoneg_advertised & IXGBE_LINK_SPEED_2_5GB_FULL) { + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_2500BASE_T; + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_2500BASE_X; + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_2500BASE_KX; + pcfg.phy_type_high |= IXGBE_PHY_TYPE_HIGH_2500M_SGMII; + pcfg.phy_type_high |= IXGBE_PHY_TYPE_HIGH_2500M_USXGMII; + } + if (hw->phy.autoneg_advertised & IXGBE_LINK_SPEED_5GB_FULL) { + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_5GBASE_T; + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_5GBASE_KR; + pcfg.phy_type_high |= IXGBE_PHY_TYPE_HIGH_5G_USXGMII; + } + if (hw->phy.autoneg_advertised & IXGBE_LINK_SPEED_10GB_FULL) { + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_10GBASE_T; + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_10G_SFI_DA; + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_10GBASE_SR; + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_10GBASE_LR; + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_10GBASE_KR_CR1; + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_10G_SFI_AOC_ACC; + pcfg.phy_type_low |= IXGBE_PHY_TYPE_LOW_10G_SFI_C2C; + pcfg.phy_type_high |= IXGBE_PHY_TYPE_HIGH_10G_USXGMII; + } + + /* Mask the set values to avoid requesting unsupported link types */ + pcfg.phy_type_low &= pcaps.phy_type_low; + pcfg.phy_type_high &= pcaps.phy_type_high; + + if (pcfg.phy_type_high != pcaps.phy_type_high || + pcfg.phy_type_low != pcaps.phy_type_low || + pcfg.caps != pcaps.caps) { + pcfg.caps |= IXGBE_ACI_PHY_ENA_LINK; + pcfg.caps |= IXGBE_ACI_PHY_ENA_AUTO_LINK_UPDT; + + rc = ixgbe_aci_set_phy_cfg(hw, &pcfg); + } + +err: + return rc; +} + +/** + * ixgbe_get_phy_firmware_version_E610 - Gets the PHY Firmware Version + * @hw: pointer to hardware structure + * @firmware_version: pointer to the PHY Firmware Version + * + * Determines PHY FW version based on response to Get PHY Capabilities + * admin command (0x0600). + * + * Return: the exit code of the operation. + */ +s32 ixgbe_get_phy_firmware_version_E610(struct ixgbe_hw *hw, + u16 *firmware_version) +{ + struct ixgbe_aci_cmd_get_phy_caps_data pcaps; + s32 status; + + if (!firmware_version) + return IXGBE_ERR_PARAM; + + status = ixgbe_aci_get_phy_caps(hw, false, + IXGBE_ACI_REPORT_ACTIVE_CFG, + &pcaps); + if (status) + return status; + + /* TODO: determine which bytes of the 8-byte phy_fw_ver + * field should be written to the 2-byte firmware_version + * output argument. */ + memcpy(firmware_version, pcaps.phy_fw_ver, sizeof(u16)); + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_read_i2c_sff8472_E610 - Reads 8 bit word over I2C interface + * @hw: pointer to hardware structure + * @byte_offset: byte offset at address 0xA2 + * @sff8472_data: value read + * + * Performs byte read operation from SFP module's SFF-8472 data over I2C. + * + * Return: the exit code of the operation. + **/ +s32 ixgbe_read_i2c_sff8472_E610(struct ixgbe_hw *hw, u8 byte_offset, + u8 *sff8472_data) +{ + return ixgbe_aci_sff_eeprom(hw, 0, IXGBE_I2C_EEPROM_DEV_ADDR2, + byte_offset, 0, + IXGBE_ACI_SFF_NO_PAGE_BANK_UPDATE, + sff8472_data, 1, false); +} + +/** + * ixgbe_read_i2c_eeprom_E610 - Reads 8 bit EEPROM word over I2C interface + * @hw: pointer to hardware structure + * @byte_offset: EEPROM byte offset to read + * @eeprom_data: value read + * + * Performs byte read operation from SFP module's EEPROM over I2C interface. + * + * Return: the exit code of the operation. + **/ +s32 ixgbe_read_i2c_eeprom_E610(struct ixgbe_hw *hw, u8 byte_offset, + u8 *eeprom_data) +{ + return ixgbe_aci_sff_eeprom(hw, 0, IXGBE_I2C_EEPROM_DEV_ADDR, + byte_offset, 0, + IXGBE_ACI_SFF_NO_PAGE_BANK_UPDATE, + eeprom_data, 1, false); +} + +/** + * ixgbe_write_i2c_eeprom_E610 - Writes 8 bit EEPROM word over I2C interface + * @hw: pointer to hardware structure + * @byte_offset: EEPROM byte offset to write + * @eeprom_data: value to write + * + * Performs byte write operation to SFP module's EEPROM over I2C interface. + * + * Return: the exit code of the operation. + **/ +s32 ixgbe_write_i2c_eeprom_E610(struct ixgbe_hw *hw, u8 byte_offset, + u8 eeprom_data) +{ + return ixgbe_aci_sff_eeprom(hw, 0, IXGBE_I2C_EEPROM_DEV_ADDR, + byte_offset, 0, + IXGBE_ACI_SFF_NO_PAGE_BANK_UPDATE, + &eeprom_data, 1, true); +} + +/** + * ixgbe_check_overtemp_E610 - Check firmware-controlled PHYs for overtemp + * @hw: pointer to hardware structure + * + * Get the link status and check if the PHY temperature alarm detected. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_check_overtemp_E610(struct ixgbe_hw *hw) +{ + struct ixgbe_aci_cmd_get_link_status_data link_data = { 0 }; + struct ixgbe_aci_cmd_get_link_status *resp; + struct ixgbe_aci_desc desc; + s32 status = IXGBE_SUCCESS; + + if (!hw) + return IXGBE_ERR_PARAM; + + ixgbe_fill_dflt_direct_cmd_desc(&desc, ixgbe_aci_opc_get_link_status); + resp = &desc.params.get_link_status; + resp->cmd_flags = IXGBE_CPU_TO_LE16(IXGBE_ACI_LSE_NOP); + + status = ixgbe_aci_send_cmd(hw, &desc, &link_data, sizeof(link_data)); + if (status != IXGBE_SUCCESS) + return status; + + if (link_data.ext_info & IXGBE_ACI_LINK_PHY_TEMP_ALARM) { + ERROR_REPORT1(IXGBE_ERROR_CAUTION, + "PHY Temperature Alarm detected"); + status = IXGBE_ERR_OVERTEMP; + } + + return status; +} + +/** + * ixgbe_set_phy_power_E610 - Control power for copper PHY + * @hw: pointer to hardware structure + * @on: true for on, false for off + * + * Set the power on/off of the PHY + * by getting its capabilities and setting the appropriate + * configuration parameters. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_set_phy_power_E610(struct ixgbe_hw *hw, bool on) +{ + struct ixgbe_aci_cmd_get_phy_caps_data phy_caps = { 0 }; + struct ixgbe_aci_cmd_set_phy_cfg_data phy_cfg = { 0 }; + s32 status; + + status = ixgbe_aci_get_phy_caps(hw, false, + IXGBE_ACI_REPORT_ACTIVE_CFG, &phy_caps); + if (status != IXGBE_SUCCESS) + return status; + + ixgbe_copy_phy_caps_to_cfg(&phy_caps, &phy_cfg); + + if (on) { + phy_cfg.caps &= ~IXGBE_ACI_PHY_ENA_LOW_POWER; + } else { + phy_cfg.caps |= IXGBE_ACI_PHY_ENA_LOW_POWER; + } + + /* PHY is already in requested power mode */ + if (phy_caps.caps == phy_cfg.caps) + return IXGBE_SUCCESS; + + phy_cfg.caps |= IXGBE_ACI_PHY_ENA_LINK; + phy_cfg.caps |= IXGBE_ACI_PHY_ENA_AUTO_LINK_UPDT; + + status = ixgbe_aci_set_phy_cfg(hw, &phy_cfg); + + return status; +} + +/** + * ixgbe_enter_lplu_E610 - Transition to low power states + * @hw: pointer to hardware structure + * + * Configures Low Power Link Up on transition to low power states + * (from D0 to non-D0). Link is required to enter LPLU so avoid resetting the + * X557 PHY immediately prior to entering LPLU. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_enter_lplu_E610(struct ixgbe_hw *hw) +{ + struct ixgbe_aci_cmd_get_phy_caps_data phy_caps = { 0 }; + struct ixgbe_aci_cmd_set_phy_cfg_data phy_cfg = { 0 }; + s32 status; + + status = ixgbe_aci_get_phy_caps(hw, false, + IXGBE_ACI_REPORT_ACTIVE_CFG, &phy_caps); + if (status != IXGBE_SUCCESS) + return status; + + ixgbe_copy_phy_caps_to_cfg(&phy_caps, &phy_cfg); + + phy_cfg.low_power_ctrl_an |= IXGBE_ACI_PHY_EN_D3COLD_LOW_POWER_AUTONEG; + + status = ixgbe_aci_set_phy_cfg(hw, &phy_cfg); + + return status; +} + +/** + * ixgbe_init_eeprom_params_E610 - Initialize EEPROM params + * @hw: pointer to hardware structure + * + * Initializes the EEPROM parameters ixgbe_eeprom_info within the + * ixgbe_hw struct in order to set up EEPROM access. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_init_eeprom_params_E610(struct ixgbe_hw *hw) +{ + struct ixgbe_eeprom_info *eeprom = &hw->eeprom; + u32 gens_stat; + u8 sr_size; + + if (eeprom->type == ixgbe_eeprom_uninitialized) { + eeprom->type = ixgbe_flash; + + gens_stat = IXGBE_READ_REG(hw, GLNVM_GENS); + sr_size = (gens_stat & GLNVM_GENS_SR_SIZE_M) >> + GLNVM_GENS_SR_SIZE_S; + + /* Switching to words (sr_size contains power of 2) */ + eeprom->word_size = BIT(sr_size) * IXGBE_SR_WORDS_IN_1KB; + + DEBUGOUT2("Eeprom params: type = %d, size = %d\n", + eeprom->type, eeprom->word_size); + } + + return IXGBE_SUCCESS; +} + +/** + * ixgbe_read_ee_aci_E610 - Read EEPROM word using the admin command. + * @hw: pointer to hardware structure + * @offset: offset of word in the EEPROM to read + * @data: word read from the EEPROM + * + * Reads a 16 bit word from the EEPROM using the ACI. + * If the EEPROM params are not initialized, the function + * initialize them before proceeding with reading. + * The function acquires and then releases the NVM ownership. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_read_ee_aci_E610(struct ixgbe_hw *hw, u16 offset, u16 *data) +{ + s32 status; + + if (hw->eeprom.type == ixgbe_eeprom_uninitialized) { + status = ixgbe_init_eeprom_params(hw); + if (status) + return status; + } + + status = ixgbe_acquire_nvm(hw, IXGBE_RES_READ); + if (status) + return status; + + status = ixgbe_read_sr_word_aci(hw, offset, data); + ixgbe_release_nvm(hw); + + return status; +} + +/** + * ixgbe_read_ee_aci_buffer_E610- Read EEPROM word(s) using admin commands. + * @hw: pointer to hardware structure + * @offset: offset of word in the EEPROM to read + * @words: number of words + * @data: word(s) read from the EEPROM + * + * Reads a 16 bit word(s) from the EEPROM using the ACI. + * If the EEPROM params are not initialized, the function + * initialize them before proceeding with reading. + * The function acquires and then releases the NVM ownership. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_read_ee_aci_buffer_E610(struct ixgbe_hw *hw, u16 offset, + u16 words, u16 *data) +{ + s32 status; + + if (hw->eeprom.type == ixgbe_eeprom_uninitialized) { + status = ixgbe_init_eeprom_params(hw); + if (status) + return status; + } + + status = ixgbe_acquire_nvm(hw, IXGBE_RES_READ); + if (status) + return status; + + status = ixgbe_read_sr_buf_aci(hw, offset, &words, data); + ixgbe_release_nvm(hw); + + return status; +} + +/** + * ixgbe_calc_eeprom_checksum_E610 - Calculates and returns the checksum + * @hw: pointer to hardware structure + * + * Calculate SW Checksum that covers the whole 64kB shadow RAM + * except the VPD and PCIe ALT Auto-load modules. The structure and size of VPD + * is customer specific and unknown. Therefore, this function skips all maximum + * possible size of VPD (1kB). + * If the EEPROM params are not initialized, the function + * initializes them before proceeding. + * The function acquires and then releases the NVM ownership. + * + * Return: the negative error code on error, or the 16-bit checksum + */ +s32 ixgbe_calc_eeprom_checksum_E610(struct ixgbe_hw *hw) +{ + bool nvm_acquired = false; + u16 pcie_alt_module = 0; + u16 checksum_local = 0; + u16 checksum = 0; + u16 vpd_module; + void *vmem; + s32 status; + u16 *data; + u16 i; + + if (hw->eeprom.type == ixgbe_eeprom_uninitialized) { + status = ixgbe_init_eeprom_params(hw); + if (status) + return status; + } + + vmem = ixgbe_calloc(hw, IXGBE_SR_SECTOR_SIZE_IN_WORDS, sizeof(u16)); + if (!vmem) + return IXGBE_ERR_OUT_OF_MEM; + data = (u16 *)vmem; + status = ixgbe_acquire_nvm(hw, IXGBE_RES_READ); + if (status) + goto ixgbe_calc_sr_checksum_exit; + nvm_acquired = true; + + /* read pointer to VPD area */ + status = ixgbe_read_sr_word_aci(hw, E610_SR_VPD_PTR, &vpd_module); + if (status) + goto ixgbe_calc_sr_checksum_exit; + + /* read pointer to PCIe Alt Auto-load module */ + status = ixgbe_read_sr_word_aci(hw, E610_SR_PCIE_ALT_AUTO_LOAD_PTR, + &pcie_alt_module); + if (status) + goto ixgbe_calc_sr_checksum_exit; + + /* Calculate SW checksum that covers the whole 64kB shadow RAM + * except the VPD and PCIe ALT Auto-load modules + */ + for (i = 0; i < hw->eeprom.word_size; i++) { + /* Read SR page */ + if ((i % IXGBE_SR_SECTOR_SIZE_IN_WORDS) == 0) { + u16 words = IXGBE_SR_SECTOR_SIZE_IN_WORDS; + + status = ixgbe_read_sr_buf_aci(hw, i, &words, data); + if (status != IXGBE_SUCCESS) + goto ixgbe_calc_sr_checksum_exit; + } + + /* Skip Checksum word */ + if (i == E610_SR_SW_CHECKSUM_WORD) + continue; + /* Skip VPD module (convert byte size to word count) */ + if (i >= (u32)vpd_module && + i < ((u32)vpd_module + E610_SR_VPD_SIZE_WORDS)) + continue; + /* Skip PCIe ALT module (convert byte size to word count) */ + if (i >= (u32)pcie_alt_module && + i < ((u32)pcie_alt_module + E610_SR_PCIE_ALT_SIZE_WORDS)) + continue; + + checksum_local += data[i % IXGBE_SR_SECTOR_SIZE_IN_WORDS]; + } + + checksum = (u16)IXGBE_SR_SW_CHECKSUM_BASE - checksum_local; + +ixgbe_calc_sr_checksum_exit: + if(nvm_acquired) + ixgbe_release_nvm(hw); + ixgbe_free(hw, vmem); + + if(!status) + return (s32)checksum; + else + return status; +} + +/** + * ixgbe_validate_eeprom_checksum_E610 - Validate EEPROM checksum + * @hw: pointer to hardware structure + * @checksum_val: calculated checksum + * + * Performs checksum calculation and validates the EEPROM checksum. If the + * caller does not need checksum_val, the value can be NULL. + * If the EEPROM params are not initialized, the function + * initialize them before proceeding. + * The function acquires and then releases the NVM ownership. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_validate_eeprom_checksum_E610(struct ixgbe_hw *hw, u16 *checksum_val) +{ + u32 status; + + if (hw->eeprom.type == ixgbe_eeprom_uninitialized) { + status = ixgbe_init_eeprom_params(hw); + if (status) + return status; + } + + status = ixgbe_nvm_validate_checksum(hw); + + if (status) + return status; + + if (checksum_val) { + u16 tmp_checksum; + status = ixgbe_acquire_nvm(hw, IXGBE_RES_READ); + if (status) + return status; + + status = ixgbe_read_sr_word_aci(hw, E610_SR_SW_CHECKSUM_WORD, + &tmp_checksum); + ixgbe_release_nvm(hw); + + if (!status) + *checksum_val = tmp_checksum; + } + + return status; +} + +/** + * ixgbe_get_pfa_module_tlv - Reads sub module TLV from NVM PFA + * @hw: pointer to hardware structure + * @module_tlv: pointer to module TLV to return + * @module_tlv_len: pointer to module TLV length to return + * @module_type: module type requested + * + * Finds the requested sub module TLV type from the Preserved Field + * Area (PFA) and returns the TLV pointer and length. The caller can + * use these to read the variable length TLV value. + * + * Return: the exit code of the operation. + */ +STATIC s32 ixgbe_get_pfa_module_tlv(struct ixgbe_hw *hw, u16 *module_tlv, + u16 *module_tlv_len, u16 module_type) +{ + u16 pfa_len, pfa_ptr, pfa_end_ptr; + u16 next_tlv; + s32 status; + + status = ixgbe_read_ee_aci_E610(hw, E610_SR_PFA_PTR, &pfa_ptr); + if (status != IXGBE_SUCCESS) { + return status; + } + status = ixgbe_read_ee_aci_E610(hw, pfa_ptr, &pfa_len); + if (status != IXGBE_SUCCESS) { + return status; + } + /* Starting with first TLV after PFA length, iterate through the list + * of TLVs to find the requested one. + */ + next_tlv = pfa_ptr + 1; + pfa_end_ptr = pfa_ptr + pfa_len; + while (next_tlv < pfa_end_ptr) { + u16 tlv_sub_module_type, tlv_len; + + /* Read TLV type */ + status = ixgbe_read_ee_aci_E610(hw, next_tlv, + &tlv_sub_module_type); + if (status != IXGBE_SUCCESS) { + break; + } + /* Read TLV length */ + status = ixgbe_read_ee_aci_E610(hw, next_tlv + 1, &tlv_len); + if (status != IXGBE_SUCCESS) { + break; + } + if (tlv_sub_module_type == module_type) { + if (tlv_len) { + *module_tlv = next_tlv; + *module_tlv_len = tlv_len; + return IXGBE_SUCCESS; + } + return IXGBE_ERR_INVAL_SIZE; + } + /* Check next TLV, i.e. current TLV pointer + length + 2 words + * (for current TLV's type and length) + */ + next_tlv = next_tlv + tlv_len + 2; + } + /* Module does not exist */ + return IXGBE_ERR_DOES_NOT_EXIST; +} + +/** + * ixgbe_read_pba_string_E610 - Reads part number string from NVM + * @hw: pointer to hardware structure + * @pba_num: stores the part number string from the NVM + * @pba_num_size: part number string buffer length + * + * Reads the part number string from the NVM. + * + * Return: the exit code of the operation. + */ +s32 ixgbe_read_pba_string_E610(struct ixgbe_hw *hw, u8 *pba_num, + u32 pba_num_size) +{ + u16 pba_tlv, pba_tlv_len; + u16 pba_word, pba_size; + s32 status; + u16 i; + + status = ixgbe_get_pfa_module_tlv(hw, &pba_tlv, &pba_tlv_len, + E610_SR_PBA_BLOCK_PTR); + if (status != IXGBE_SUCCESS) { + return status; + } + + /* pba_size is the next word */ + status = ixgbe_read_ee_aci_E610(hw, (pba_tlv + 2), &pba_size); + if (status != IXGBE_SUCCESS) { + return status; + } + + if (pba_tlv_len < pba_size) { + return IXGBE_ERR_INVAL_SIZE; + } + + /* Subtract one to get PBA word count (PBA Size word is included in + * total size) + */ + pba_size--; + if (pba_num_size < (((u32)pba_size * 2) + 1)) { + return IXGBE_ERR_PARAM; + } + + for (i = 0; i < pba_size; i++) { + status = ixgbe_read_ee_aci_E610(hw, (pba_tlv + 2 + 1) + i, + &pba_word); + if (status != IXGBE_SUCCESS) { + return status; + } + + pba_num[(i * 2)] = (pba_word >> 8) & 0xFF; + pba_num[(i * 2) + 1] = pba_word & 0xFF; + } + pba_num[(pba_size * 2)] = '\0'; + + return status; +} diff --git a/drivers/net/ixgbe/base/ixgbe_e610.h b/drivers/net/ixgbe/base/ixgbe_e610.h new file mode 100644 index 0000000000..f241955ada --- /dev/null +++ b/drivers/net/ixgbe/base/ixgbe_e610.h @@ -0,0 +1,163 @@ +#ifndef _IXGBE_E610_H_ +#define _IXGBE_E610_H_ + +#include "ixgbe_type.h" + +void ixgbe_init_aci(struct ixgbe_hw *hw); +void ixgbe_shutdown_aci(struct ixgbe_hw *hw); + +s32 ixgbe_aci_send_cmd(struct ixgbe_hw *hw, struct ixgbe_aci_desc *desc, + void *buf, u16 buf_size); +bool ixgbe_aci_check_event_pending(struct ixgbe_hw *hw); +s32 ixgbe_aci_get_event(struct ixgbe_hw *hw, struct ixgbe_aci_event *e, + bool *pending); + +void ixgbe_fill_dflt_direct_cmd_desc(struct ixgbe_aci_desc *desc, u16 opcode); + +s32 ixgbe_aci_get_fw_ver(struct ixgbe_hw *hw); +s32 ixgbe_aci_send_driver_ver(struct ixgbe_hw *hw, struct ixgbe_driver_ver *dv); +s32 ixgbe_aci_set_pf_context(struct ixgbe_hw *hw, u8 pf_id); + +s32 ixgbe_acquire_res(struct ixgbe_hw *hw, enum ixgbe_aci_res_ids res, + enum ixgbe_aci_res_access_type access, u32 timeout); +void ixgbe_release_res(struct ixgbe_hw *hw, enum ixgbe_aci_res_ids res); +s32 ixgbe_aci_list_caps(struct ixgbe_hw *hw, void *buf, u16 buf_size, + u32 *cap_count, enum ixgbe_aci_opc opc); +s32 ixgbe_discover_dev_caps(struct ixgbe_hw *hw, + struct ixgbe_hw_dev_caps *dev_caps); +s32 ixgbe_discover_func_caps(struct ixgbe_hw* hw, + struct ixgbe_hw_func_caps* func_caps); +s32 ixgbe_get_caps(struct ixgbe_hw *hw); +s32 ixgbe_aci_disable_rxen(struct ixgbe_hw *hw); +s32 ixgbe_aci_get_phy_caps(struct ixgbe_hw *hw, bool qual_mods, u8 report_mode, + struct ixgbe_aci_cmd_get_phy_caps_data *pcaps); +bool ixgbe_phy_caps_equals_cfg(struct ixgbe_aci_cmd_get_phy_caps_data *caps, + struct ixgbe_aci_cmd_set_phy_cfg_data *cfg); +void ixgbe_copy_phy_caps_to_cfg(struct ixgbe_aci_cmd_get_phy_caps_data *caps, + struct ixgbe_aci_cmd_set_phy_cfg_data *cfg); +s32 ixgbe_aci_set_phy_cfg(struct ixgbe_hw *hw, + struct ixgbe_aci_cmd_set_phy_cfg_data *cfg); +s32 ixgbe_aci_set_link_restart_an(struct ixgbe_hw *hw, bool ena_link); +s32 ixgbe_update_link_info(struct ixgbe_hw *hw); +s32 ixgbe_get_link_status(struct ixgbe_hw *hw, bool *link_up); +s32 ixgbe_aci_get_link_info(struct ixgbe_hw *hw, bool ena_lse, + struct ixgbe_link_status *link); +s32 ixgbe_aci_set_event_mask(struct ixgbe_hw *hw, u8 port_num, u16 mask); +s32 ixgbe_configure_lse(struct ixgbe_hw *hw, bool activate, u16 mask); + +s32 ixgbe_aci_get_netlist_node(struct ixgbe_hw *hw, + struct ixgbe_aci_cmd_get_link_topo *cmd, + u8 *node_part_number, u16 *node_handle); +s32 ixgbe_aci_get_netlist_node_pin(struct ixgbe_hw *hw, + struct ixgbe_aci_cmd_get_link_topo_pin *cmd, + u16 *node_handle); +s32 ixgbe_find_netlist_node(struct ixgbe_hw *hw, u8 node_type_ctx, + u8 node_part_number, u16 *node_handle); +s32 ixgbe_aci_read_i2c(struct ixgbe_hw *hw, + struct ixgbe_aci_cmd_link_topo_addr topo_addr, + u16 bus_addr, __le16 addr, u8 params, u8 *data); +s32 ixgbe_aci_write_i2c(struct ixgbe_hw *hw, + struct ixgbe_aci_cmd_link_topo_addr topo_addr, + u16 bus_addr, __le16 addr, u8 params, u8 *data); +s32 ixgbe_aci_set_gpio(struct ixgbe_hw *hw, u16 gpio_ctrl_handle, u8 pin_idx, + bool value); +s32 ixgbe_aci_get_gpio(struct ixgbe_hw *hw, u16 gpio_ctrl_handle, u8 pin_idx, + bool *value); +s32 ixgbe_aci_sff_eeprom(struct ixgbe_hw *hw, u16 lport, u8 bus_addr, + u16 mem_addr, u8 page, u8 page_bank_ctrl, u8 *data, + u8 length, bool write); +s32 ixgbe_aci_prog_topo_dev_nvm(struct ixgbe_hw *hw, + struct ixgbe_aci_cmd_link_topo_params *topo_params); +s32 ixgbe_aci_read_topo_dev_nvm(struct ixgbe_hw *hw, + struct ixgbe_aci_cmd_link_topo_params *topo_params, + u32 start_address, u8 *data, u8 data_size); + +s32 ixgbe_acquire_nvm(struct ixgbe_hw *hw, + enum ixgbe_aci_res_access_type access); +void ixgbe_release_nvm(struct ixgbe_hw *hw); + +s32 ixgbe_aci_read_nvm(struct ixgbe_hw *hw, u16 module_typeid, u32 offset, + u16 length, void *data, bool last_command, + bool read_shadow_ram); + +s32 ixgbe_nvm_validate_checksum(struct ixgbe_hw *hw); +s32 ixgbe_nvm_recalculate_checksum(struct ixgbe_hw *hw); + +s32 ixgbe_get_nvm_minsrevs(struct ixgbe_hw *hw, struct ixgbe_minsrev_info *minsrevs); +s32 ixgbe_get_inactive_nvm_ver(struct ixgbe_hw *hw, struct ixgbe_nvm_info *nvm); +s32 ixgbe_get_active_nvm_ver(struct ixgbe_hw *hw, struct ixgbe_nvm_info *nvm); +s32 ixgbe_init_nvm(struct ixgbe_hw *hw); + +s32 ixgbe_sanitize_operate(struct ixgbe_hw *hw); +s32 ixgbe_sanitize_nvm(struct ixgbe_hw *hw, u8 cmd_flags, u8 *values); + +s32 ixgbe_read_sr_word_aci(struct ixgbe_hw *hw, u16 offset, u16 *data); +s32 ixgbe_read_sr_buf_aci(struct ixgbe_hw *hw, u16 offset, u16 *words, u16 *data); +s32 ixgbe_read_flat_nvm(struct ixgbe_hw *hw, u32 offset, u32 *length, + u8 *data, bool read_shadow_ram); + +s32 ixgbe_aci_alternate_write(struct ixgbe_hw *hw, u32 reg_addr0, + u32 reg_val0, u32 reg_addr1, u32 reg_val1); +s32 ixgbe_aci_alternate_read(struct ixgbe_hw *hw, u32 reg_addr0, + u32 *reg_val0, u32 reg_addr1, u32 *reg_val1); +s32 ixgbe_aci_alternate_write_done(struct ixgbe_hw *hw, u8 bios_mode, + bool *reset_needed); +s32 ixgbe_aci_alternate_clear(struct ixgbe_hw *hw); + +s32 ixgbe_aci_get_internal_data(struct ixgbe_hw *hw, u16 cluster_id, + u16 table_id, u32 start, void *buf, + u16 buf_size, u16 *ret_buf_size, + u16 *ret_next_cluster, u16 *ret_next_table, + u32 *ret_next_index); + +s32 ixgbe_handle_nvm_access(struct ixgbe_hw *hw, + struct ixgbe_nvm_access_cmd *cmd, + struct ixgbe_nvm_access_data *data); + +/* E610 operations */ +s32 ixgbe_reset_hw_E610(struct ixgbe_hw *hw); +s32 ixgbe_start_hw_E610(struct ixgbe_hw *hw); +enum ixgbe_media_type ixgbe_get_media_type_E610(struct ixgbe_hw *hw); +u64 ixgbe_get_supported_physical_layer_E610(struct ixgbe_hw *hw); +s32 ixgbe_setup_link_E610(struct ixgbe_hw *hw, ixgbe_link_speed speed, + bool autoneg_wait); +s32 ixgbe_check_link_E610(struct ixgbe_hw *hw, ixgbe_link_speed *speed, + bool *link_up, bool link_up_wait_to_complete); +s32 ixgbe_get_link_capabilities_E610(struct ixgbe_hw *hw, + ixgbe_link_speed *speed, + bool *autoneg); +s32 ixgbe_cfg_phy_fc(struct ixgbe_hw *hw, + struct ixgbe_aci_cmd_set_phy_cfg_data *cfg, + enum ixgbe_fc_mode req_mode); +s32 ixgbe_setup_fc_E610(struct ixgbe_hw *hw); +void ixgbe_fc_autoneg_E610(struct ixgbe_hw *hw); +s32 ixgbe_set_fw_drv_ver_E610(struct ixgbe_hw *hw, u8 maj, u8 min, u8 build, + u8 sub, u16 len, const char *driver_ver); +void ixgbe_disable_rx_E610(struct ixgbe_hw *hw); +s32 ixgbe_setup_eee_E610(struct ixgbe_hw *hw, bool enable_eee); +bool ixgbe_fw_recovery_mode_E610(struct ixgbe_hw *hw); +bool ixgbe_get_fw_tsam_mode_E610(struct ixgbe_hw *hw); +s32 ixgbe_init_phy_ops_E610(struct ixgbe_hw *hw); +s32 ixgbe_identify_phy_E610(struct ixgbe_hw *hw); +s32 ixgbe_identify_module_E610(struct ixgbe_hw *hw); +s32 ixgbe_setup_phy_link_E610(struct ixgbe_hw *hw); +s32 ixgbe_get_phy_firmware_version_E610(struct ixgbe_hw *hw, + u16 *firmware_version); +s32 ixgbe_read_i2c_sff8472_E610(struct ixgbe_hw *hw, u8 byte_offset, + u8 *sff8472_data); +s32 ixgbe_read_i2c_eeprom_E610(struct ixgbe_hw *hw, u8 byte_offset, + u8 *eeprom_data); +s32 ixgbe_write_i2c_eeprom_E610(struct ixgbe_hw *hw, u8 byte_offset, + u8 eeprom_data); +s32 ixgbe_check_overtemp_E610(struct ixgbe_hw *hw); +s32 ixgbe_set_phy_power_E610(struct ixgbe_hw *hw, bool on); +s32 ixgbe_enter_lplu_E610(struct ixgbe_hw *hw); +s32 ixgbe_init_eeprom_params_E610(struct ixgbe_hw *hw); +s32 ixgbe_read_ee_aci_E610(struct ixgbe_hw *hw, u16 offset, u16 *data); +s32 ixgbe_read_ee_aci_buffer_E610(struct ixgbe_hw *hw, u16 offset, + u16 words, u16 *data); +s32 ixgbe_calc_eeprom_checksum_E610(struct ixgbe_hw *hw); +s32 ixgbe_validate_eeprom_checksum_E610(struct ixgbe_hw *hw, u16 *checksum_val); +s32 ixgbe_read_pba_string_E610(struct ixgbe_hw *hw, u8 *pba_num, u32 pba_num_size); + +#endif /* _IXGBE_E610_H_ */ diff --git a/drivers/net/ixgbe/base/ixgbe_hv_vf.c b/drivers/net/ixgbe/base/ixgbe_hv_vf.c index 1279659926..246dafcdf1 100644 --- a/drivers/net/ixgbe/base/ixgbe_hv_vf.c +++ b/drivers/net/ixgbe/base/ixgbe_hv_vf.c @@ -136,7 +136,8 @@ static s32 ixgbevf_hv_check_mac_link_vf(struct ixgbe_hw *hw, break; case IXGBE_LINKS_SPEED_100_82599: *speed = IXGBE_LINK_SPEED_100_FULL; - if (hw->mac.type == ixgbe_mac_X550) { + if (hw->mac.type == ixgbe_mac_X550 || + hw->mac.type == ixgbe_mac_E610) { if (links_reg & IXGBE_LINKS_SPEED_NON_STD) *speed = IXGBE_LINK_SPEED_5GB_FULL; } diff --git a/drivers/net/ixgbe/base/ixgbe_mbx.c b/drivers/net/ixgbe/base/ixgbe_mbx.c index 2dab347396..9ad45fbfcc 100644 --- a/drivers/net/ixgbe/base/ixgbe_mbx.c +++ b/drivers/net/ixgbe/base/ixgbe_mbx.c @@ -777,6 +777,7 @@ STATIC s32 ixgbe_check_for_rst_pf(struct ixgbe_hw *hw, u16 vf_id) case ixgbe_mac_X550EM_x: case ixgbe_mac_X550EM_a: case ixgbe_mac_X540: + case ixgbe_mac_E610: vflre = IXGBE_READ_REG(hw, IXGBE_PFVFLREC(index)); break; default: @@ -1061,6 +1062,7 @@ void ixgbe_init_mbx_params_pf(struct ixgbe_hw *hw) hw->mac.type != ixgbe_mac_X550 && hw->mac.type != ixgbe_mac_X550EM_x && hw->mac.type != ixgbe_mac_X550EM_a && + hw->mac.type != ixgbe_mac_E610 && hw->mac.type != ixgbe_mac_X540) return; @@ -1103,6 +1105,7 @@ void ixgbe_upgrade_mbx_params_pf(struct ixgbe_hw *hw, u16 vf_id) hw->mac.type != ixgbe_mac_X550 && hw->mac.type != ixgbe_mac_X550EM_x && hw->mac.type != ixgbe_mac_X550EM_a && + hw->mac.type != ixgbe_mac_E610 && hw->mac.type != ixgbe_mac_X540) return; diff --git a/drivers/net/ixgbe/base/ixgbe_osdep.c b/drivers/net/ixgbe/base/ixgbe_osdep.c new file mode 100644 index 0000000000..fe4db60a8d --- /dev/null +++ b/drivers/net/ixgbe/base/ixgbe_osdep.c @@ -0,0 +1,43 @@ +#include + +#include + +#include "ixgbe_osdep.h" + +void * +ixgbe_calloc(struct ixgbe_hw __rte_unused *hw, size_t count, size_t size) +{ + return malloc(count * size); +} + +void * +ixgbe_malloc(struct ixgbe_hw __rte_unused *hw, size_t size) +{ + return malloc(size); +} + +void +ixgbe_free(struct ixgbe_hw __rte_unused *hw, void *addr) +{ + free(addr); +} + +void ixgbe_init_lock(struct ixgbe_lock *lock) +{ + pthread_mutex_init(&lock->mutex, NULL); +} + +void ixgbe_destroy_lock(struct ixgbe_lock *lock) +{ + pthread_mutex_destroy(&lock->mutex); +} + +void ixgbe_acquire_lock(struct ixgbe_lock *lock) +{ + pthread_mutex_lock(&lock->mutex); +} + +void ixgbe_release_lock(struct ixgbe_lock *lock) +{ + pthread_mutex_unlock(&lock->mutex); +} diff --git a/drivers/net/ixgbe/base/ixgbe_osdep.h b/drivers/net/ixgbe/base/ixgbe_osdep.h index 6c25f608b1..721043fb2e 100644 --- a/drivers/net/ixgbe/base/ixgbe_osdep.h +++ b/drivers/net/ixgbe/base/ixgbe_osdep.h @@ -5,6 +5,7 @@ #ifndef _IXGBE_OS_H_ #define _IXGBE_OS_H_ +#include #include #include #include @@ -79,7 +80,9 @@ enum { #define IXGBE_NTOHS(_i) rte_be_to_cpu_16(_i) #define IXGBE_CPU_TO_LE16(_i) rte_cpu_to_le_16(_i) #define IXGBE_CPU_TO_LE32(_i) rte_cpu_to_le_32(_i) +#define IXGBE_LE16_TO_CPU(_i) rte_le_to_cpu_16(_i) #define IXGBE_LE32_TO_CPU(_i) rte_le_to_cpu_32(_i) +#define IXGBE_LE64_TO_CPU(_i) rte_le_to_cpu_64(_i) #define IXGBE_LE32_TO_CPUS(_i) rte_le_to_cpu_32(_i) #define IXGBE_CPU_TO_BE16(_i) rte_cpu_to_be_16(_i) #define IXGBE_CPU_TO_BE32(_i) rte_cpu_to_be_32(_i) @@ -152,4 +155,18 @@ do { \ rte_delay_ms(1); \ } while (0) +struct ixgbe_hw; +struct ixgbe_lock { + pthread_mutex_t mutex; +}; + +void *ixgbe_calloc(struct ixgbe_hw *hw, size_t count, size_t size); +void *ixgbe_malloc(struct ixgbe_hw *hw, size_t size); +void ixgbe_free(struct ixgbe_hw *hw, void *addr); + +void ixgbe_init_lock(struct ixgbe_lock *lock); +void ixgbe_destroy_lock(struct ixgbe_lock *lock); +void ixgbe_acquire_lock(struct ixgbe_lock *lock); +void ixgbe_release_lock(struct ixgbe_lock *lock); + #endif /* _IXGBE_OS_H_ */ diff --git a/drivers/net/ixgbe/base/ixgbe_phy.c b/drivers/net/ixgbe/base/ixgbe_phy.c index 8a7712c8f2..47dab74b3c 100644 --- a/drivers/net/ixgbe/base/ixgbe_phy.c +++ b/drivers/net/ixgbe/base/ixgbe_phy.c @@ -800,7 +800,8 @@ s32 ixgbe_setup_phy_link_generic(struct ixgbe_hw *hw) IXGBE_MDIO_AUTO_NEG_DEV_TYPE, &autoneg_reg); - if (hw->mac.type == ixgbe_mac_X550) { + if ((hw->mac.type == ixgbe_mac_X550) || + (hw->mac.type == ixgbe_mac_E610)) { /* Set or unset auto-negotiation 5G advertisement */ autoneg_reg &= ~IXGBE_MII_5GBASE_T_ADVERTISE; if ((hw->phy.autoneg_advertised & IXGBE_LINK_SPEED_5GB_FULL) && @@ -915,6 +916,7 @@ static s32 ixgbe_get_copper_speeds_supported(struct ixgbe_hw *hw) switch (hw->mac.type) { case ixgbe_mac_X550: + case ixgbe_mac_E610: hw->phy.speeds_supported |= IXGBE_LINK_SPEED_2_5GB_FULL; hw->phy.speeds_supported |= IXGBE_LINK_SPEED_5GB_FULL; break; diff --git a/drivers/net/ixgbe/base/ixgbe_type.h b/drivers/net/ixgbe/base/ixgbe_type.h index 5db9e03b4d..1070c281a2 100644 --- a/drivers/net/ixgbe/base/ixgbe_type.h +++ b/drivers/net/ixgbe/base/ixgbe_type.h @@ -44,6 +44,7 @@ */ #include "ixgbe_osdep.h" +#include "ixgbe_type_e610.h" /* Override this by setting IOMEM in your ixgbe_osdep.h header */ @@ -124,6 +125,11 @@ #define IXGBE_DEV_ID_X550EM_A_VF_HV 0x15B4 #define IXGBE_DEV_ID_X550EM_X_VF 0x15A8 #define IXGBE_DEV_ID_X550EM_X_VF_HV 0x15A9 +#define IXGBE_DEV_ID_E610_BACKPLANE 0x57AE +#define IXGBE_DEV_ID_E610_SFP 0x57AF +#define IXGBE_DEV_ID_E610_10G_T 0x57B0 +#define IXGBE_DEV_ID_E610_2_5G_T 0x57B1 +#define IXGBE_DEV_ID_E610_SGMII 0x57B2 #define IXGBE_CAT(r, m) IXGBE_##r##m @@ -1887,6 +1893,7 @@ enum { #define IXGBE_EICR_MAILBOX 0x00080000 /* VF to PF Mailbox Interrupt */ #define IXGBE_EICR_LSC 0x00100000 /* Link Status Change */ #define IXGBE_EICR_LINKSEC 0x00200000 /* PN Threshold */ +#define IXGBE_EICR_FW_EVENT 0x00200000 /* Async FW event */ #define IXGBE_EICR_MNG 0x00400000 /* Manageability Event Interrupt */ #define IXGBE_EICR_TS 0x00800000 /* Thermal Sensor Event */ #define IXGBE_EICR_TIMESYNC 0x01000000 /* Timesync Event */ @@ -1922,6 +1929,7 @@ enum { #define IXGBE_EICS_PCI IXGBE_EICR_PCI /* PCI Exception */ #define IXGBE_EICS_MAILBOX IXGBE_EICR_MAILBOX /* VF to PF Mailbox Int */ #define IXGBE_EICS_LSC IXGBE_EICR_LSC /* Link Status Change */ +#define IXGBE_EICS_FW_EVENT IXGBE_EICR_FW_EVENT /* Async FW event */ #define IXGBE_EICS_MNG IXGBE_EICR_MNG /* MNG Event Interrupt */ #define IXGBE_EICS_TIMESYNC IXGBE_EICR_TIMESYNC /* Timesync Event */ #define IXGBE_EICS_GPI_SDP0 IXGBE_EICR_GPI_SDP0 /* SDP0 Gen Purpose Int */ @@ -1943,6 +1951,7 @@ enum { #define IXGBE_EIMS_PCI IXGBE_EICR_PCI /* PCI Exception */ #define IXGBE_EIMS_MAILBOX IXGBE_EICR_MAILBOX /* VF to PF Mailbox Int */ #define IXGBE_EIMS_LSC IXGBE_EICR_LSC /* Link Status Change */ +#define IXGBE_EIMS_FW_EVENT IXGBE_EICR_FW_EVENT /* Async FW event */ #define IXGBE_EIMS_MNG IXGBE_EICR_MNG /* MNG Event Interrupt */ #define IXGBE_EIMS_TS IXGBE_EICR_TS /* Thermal Sensor Event */ #define IXGBE_EIMS_TIMESYNC IXGBE_EICR_TIMESYNC /* Timesync Event */ @@ -1965,6 +1974,7 @@ enum { #define IXGBE_EIMC_PCI IXGBE_EICR_PCI /* PCI Exception */ #define IXGBE_EIMC_MAILBOX IXGBE_EICR_MAILBOX /* VF to PF Mailbox Int */ #define IXGBE_EIMC_LSC IXGBE_EICR_LSC /* Link Status Change */ +#define IXGBE_EIMC_FW_EVENT IXGBE_EICR_FW_EVENT /* Async FW event */ #define IXGBE_EIMC_MNG IXGBE_EICR_MNG /* MNG Event Interrupt */ #define IXGBE_EIMC_TIMESYNC IXGBE_EICR_TIMESYNC /* Timesync Event */ #define IXGBE_EIMC_GPI_SDP0 IXGBE_EICR_GPI_SDP0 /* SDP0 Gen Purpose Int */ @@ -2372,6 +2382,7 @@ enum { #define IXGBE_82599_SERIAL_NUMBER_MAC_ADDR 0x11 #define IXGBE_X550_SERIAL_NUMBER_MAC_ADDR 0x04 +#define IXGBE_PCIE_MSIX_LKV_CAPS 0xB2 #define IXGBE_PCIE_MSIX_82599_CAPS 0x72 #define IXGBE_MAX_MSIX_VECTORS_82599 0x40 #define IXGBE_PCIE_MSIX_82598_CAPS 0x62 @@ -2489,6 +2500,7 @@ enum { #define IXGBE_PCI_DEVICE_STATUS 0xAA #define IXGBE_PCI_DEVICE_STATUS_TRANSACTION_PENDING 0x0020 #define IXGBE_PCI_LINK_STATUS 0xB2 +#define IXGBE_PCI_LINK_STATUS_E610 0x82 #define IXGBE_PCI_DEVICE_CONTROL2 0xC8 #define IXGBE_PCI_LINK_WIDTH 0x3F0 #define IXGBE_PCI_LINK_WIDTH_1 0x10 @@ -2499,6 +2511,7 @@ enum { #define IXGBE_PCI_LINK_SPEED_2500 0x1 #define IXGBE_PCI_LINK_SPEED_5000 0x2 #define IXGBE_PCI_LINK_SPEED_8000 0x3 +#define IXGBE_PCI_LINK_SPEED_16000 0x4 #define IXGBE_PCI_HEADER_TYPE_REGISTER 0x0E #define IXGBE_PCI_HEADER_TYPE_MULTIFUNC 0x80 #define IXGBE_PCI_DEVICE_CONTROL2_16ms 0x0005 @@ -2626,6 +2639,7 @@ enum { #define IXGBE_RXMTRL_V2_MGMT_MSG 0x0D00 #define IXGBE_FCTRL_SBP 0x00000002 /* Store Bad Packet */ +#define IXGBE_FCTRL_TPE 0x00000080 /* Tag Promiscuous Ena*/ #define IXGBE_FCTRL_MPE 0x00000100 /* Multicast Promiscuous Ena*/ #define IXGBE_FCTRL_UPE 0x00000200 /* Unicast Promiscuous Ena */ #define IXGBE_FCTRL_BAM 0x00000400 /* Broadcast Accept Mode */ @@ -2693,6 +2707,7 @@ enum { /* Multiple Transmit Queue Command Register */ #define IXGBE_MTQC_RT_ENA 0x1 /* DCB Enable */ #define IXGBE_MTQC_VT_ENA 0x2 /* VMDQ2 Enable */ +#define IXGBE_MTQC_NUM_TC_OR_Q 0xC /* Numer of TCs or TxQs per pool */ #define IXGBE_MTQC_64Q_1PB 0x0 /* 64 queues 1 pack buffer */ #define IXGBE_MTQC_32VF 0x8 /* 4 TX Queues per pool w/32VF's */ #define IXGBE_MTQC_64VF 0x4 /* 2 TX Queues per pool w/64VF's */ @@ -3660,6 +3675,7 @@ enum ixgbe_mac_type { ixgbe_mac_X550_vf, ixgbe_mac_X550EM_x_vf, ixgbe_mac_X550EM_a_vf, + ixgbe_mac_E610, ixgbe_num_macs }; @@ -3738,7 +3754,9 @@ enum ixgbe_media_type { ixgbe_media_type_copper, ixgbe_media_type_backplane, ixgbe_media_type_cx4, - ixgbe_media_type_virtual + ixgbe_media_type_virtual, + ixgbe_media_type_da, + ixgbe_media_type_aui }; /* Flow Control Settings */ @@ -3747,6 +3765,8 @@ enum ixgbe_fc_mode { ixgbe_fc_rx_pause, ixgbe_fc_tx_pause, ixgbe_fc_full, + ixgbe_fc_auto, + ixgbe_fc_pfc, ixgbe_fc_default }; @@ -3779,6 +3799,7 @@ enum ixgbe_bus_speed { ixgbe_bus_speed_2500 = 2500, ixgbe_bus_speed_5000 = 5000, ixgbe_bus_speed_8000 = 8000, + ixgbe_bus_speed_16000 = 16000, ixgbe_bus_speed_reserved }; @@ -3923,6 +3944,7 @@ struct ixgbe_eeprom_operations { s32 (*validate_checksum)(struct ixgbe_hw *, u16 *); s32 (*update_checksum)(struct ixgbe_hw *); s32 (*calc_checksum)(struct ixgbe_hw *); + s32 (*read_pba_string)(struct ixgbe_hw *, u8 *, u32); }; struct ixgbe_mac_operations { @@ -4029,6 +4051,9 @@ struct ixgbe_mac_operations { void (*mdd_event)(struct ixgbe_hw *hw, u32 *vf_bitmap); void (*restore_mdd_vf)(struct ixgbe_hw *hw, u32 vf); bool (*fw_recovery_mode)(struct ixgbe_hw *hw); + bool (*get_fw_tsam_mode)(struct ixgbe_hw *hw); + s32 (*get_fw_version)(struct ixgbe_hw *hw); + s32 (*get_nvm_version)(struct ixgbe_hw *hw, struct ixgbe_nvm_info *nvm); }; struct ixgbe_phy_operations { @@ -4073,6 +4098,9 @@ struct ixgbe_link_operations { struct ixgbe_link_info { struct ixgbe_link_operations ops; u8 addr; + struct ixgbe_link_status link_info; + struct ixgbe_link_status link_info_old; + u8 get_link_info; }; struct ixgbe_eeprom_info { @@ -4144,6 +4172,9 @@ struct ixgbe_phy_info { bool reset_if_overtemp; bool qsfp_shared_i2c_bus; u32 nw_mng_if_sel; + u64 phy_type_low; + u64 phy_type_high; + struct ixgbe_aci_cmd_set_phy_cfg_data curr_user_phy_cfg; }; #include "ixgbe_mbx.h" @@ -4165,6 +4196,8 @@ struct ixgbe_hw { u16 subsystem_device_id; u16 subsystem_vendor_id; u8 revision_id; + u8 pf_id; + u8 logical_pf_id; bool adapter_stopped; int api_version; bool force_full_reset; @@ -4172,6 +4205,19 @@ struct ixgbe_hw { bool wol_enabled; bool need_crosstalk_fix; u32 fw_rst_cnt; + u8 api_branch; + u8 api_maj_ver; + u8 api_min_ver; + u8 api_patch; + u8 fw_branch; + u8 fw_maj_ver; + u8 fw_min_ver; + u8 fw_patch; + u32 fw_build; + struct ixgbe_aci_info aci; + struct ixgbe_flash_info flash; + struct ixgbe_hw_dev_caps dev_caps; + struct ixgbe_hw_func_caps func_caps; }; #define ixgbe_call_func(hw, func, params, error) \ @@ -4221,6 +4267,23 @@ struct ixgbe_hw { #define IXGBE_ERR_MBX -41 #define IXGBE_ERR_MBX_NOMSG -42 #define IXGBE_ERR_TIMEOUT -43 +#define IXGBE_ERR_NOT_SUPPORTED -45 +#define IXGBE_ERR_OUT_OF_RANGE -46 + +#define IXGBE_ERR_NVM -50 +#define IXGBE_ERR_NVM_CHECKSUM -51 +#define IXGBE_ERR_BUF_TOO_SHORT -52 +#define IXGBE_ERR_NVM_BLANK_MODE -53 +#define IXGBE_ERR_INVAL_SIZE -54 +#define IXGBE_ERR_DOES_NOT_EXIST -55 + +#define IXGBE_ERR_ACI_ERROR -100 +#define IXGBE_ERR_ACI_DISABLED -101 +#define IXGBE_ERR_ACI_TIMEOUT -102 +#define IXGBE_ERR_ACI_BUSY -103 +#define IXGBE_ERR_ACI_NO_WORK -104 +#define IXGBE_ERR_ACI_NO_EVENTS -105 +#define IXGBE_ERR_FW_API_VER -106 #define IXGBE_NOT_IMPLEMENTED 0x7FFFFFFF diff --git a/drivers/net/ixgbe/base/ixgbe_type_e610.h b/drivers/net/ixgbe/base/ixgbe_type_e610.h new file mode 100644 index 0000000000..0b734d5af7 --- /dev/null +++ b/drivers/net/ixgbe/base/ixgbe_type_e610.h @@ -0,0 +1,2181 @@ +#ifndef _IXGBE_TYPE_E610_H_ +#define _IXGBE_TYPE_E610_H_ + +/* Little Endian defines */ +#ifndef __le16 +#define __le16 u16 +#endif +#ifndef __le32 +#define __le32 u32 +#endif +#ifndef __le64 +#define __le64 u64 +#endif + +/* Generic defines */ +#ifndef BIT +#define BIT(a) (1UL << (a)) +#endif /* !BIT */ +#ifndef BIT_ULL +#define BIT_ULL(a) (1ULL << (a)) +#endif /* !BIT_ULL */ +#ifndef BITS_PER_BYTE +#define BITS_PER_BYTE 8 +#endif /* !BITS_PER_BYTE */ +#ifndef DIVIDE_AND_ROUND_UP +#define DIVIDE_AND_ROUND_UP(a, b) (((a) + (b) - 1) / (b)) +#endif /* !DIVIDE_AND_ROUND_UP */ + +#ifndef ROUND_UP +/** + * ROUND_UP - round up to next arbitrary multiple (not a power of 2) + * @a: value to round up + * @b: arbitrary multiple + * + * Round up to the next multiple of the arbitrary b. + */ +#define ROUND_UP(a, b) ((b) * DIVIDE_AND_ROUND_UP((a), (b))) +#endif /* !ROUND_UP */ + +#define MAKEMASK(mask, shift) (mask << shift) + +#define BYTES_PER_WORD 2 +#define BYTES_PER_DWORD 4 + +#ifndef BITS_PER_LONG +#define BITS_PER_LONG 64 +#endif /* !BITS_PER_LONG */ +#ifndef BITS_PER_LONG_LONG +#define BITS_PER_LONG_LONG 64 +#endif /* !BITS_PER_LONG_LONG */ +#undef GENMASK +#define GENMASK(h, l) \ + (((~0UL) << (l)) & (~0UL >> (BITS_PER_LONG - 1 - (h)))) +#undef GENMASK_ULL +#define GENMASK_ULL(h, l) \ + (((~0ULL) << (l)) & (~0ULL >> (BITS_PER_LONG_LONG - 1 - (h)))) + +/* Data type manipulation macros. */ +#define HI_DWORD(x) ((u32)((((x) >> 16) >> 16) & 0xFFFFFFFF)) +#define LO_DWORD(x) ((u32)((x) & 0xFFFFFFFF)) +#define HI_WORD(x) ((u16)(((x) >> 16) & 0xFFFF)) +#define LO_WORD(x) ((u16)((x) & 0xFFFF)) +#define HI_BYTE(x) ((u8)(((x) >> 8) & 0xFF)) +#define LO_BYTE(x) ((u8)((x) & 0xFF)) + +#define MIN_T(_t, _a, _b) min((_t)(_a), (_t)(_b)) + +#define IS_ASCII(_ch) ((_ch) < 0x80) + +#define STRUCT_HACK_VAR_LEN +/** + * ixgbe_struct_size - size of struct with C99 flexible array member + * @ptr: pointer to structure + * @field: flexible array member (last member of the structure) + * @num: number of elements of that flexible array member + */ +#define ixgbe_struct_size(ptr, field, num) \ + (sizeof(*(ptr)) + sizeof(*(ptr)->field) * (num)) + +/* General E610 defines */ +#define IXGBE_MAX_VSI 768 + +/* Auxiliary field, mask and shift definition for Shadow RAM and NVM Flash */ +#define E610_SR_VPD_SIZE_WORDS 512 +#define E610_SR_PCIE_ALT_SIZE_WORDS 512 + +/* Checksum and Shadow RAM pointers */ +#define E610_SR_NVM_DEV_STARTER_VER 0x18 +#define E610_NVM_VER_LO_SHIFT 0 +#define E610_NVM_VER_LO_MASK (0xff << E610_NVM_VER_LO_SHIFT) +#define E610_NVM_VER_HI_SHIFT 12 +#define E610_NVM_VER_HI_MASK (0xf << E610_NVM_VER_HI_SHIFT) +#define E610_SR_NVM_MAP_VER 0x29 +#define E610_SR_NVM_EETRACK_LO 0x2D +#define E610_SR_NVM_EETRACK_HI 0x2E +#define E610_SR_VPD_PTR 0x2F +#define E610_SR_PCIE_ALT_AUTO_LOAD_PTR 0x3E +#define E610_SR_SW_CHECKSUM_WORD 0x3F +#define E610_SR_PFA_PTR 0x40 +#define E610_SR_1ST_NVM_BANK_PTR 0x42 +#define E610_SR_NVM_BANK_SIZE 0x43 +#define E610_SR_1ST_OROM_BANK_PTR 0x44 +#define E610_SR_OROM_BANK_SIZE 0x45 +#define E610_SR_NETLIST_BANK_PTR 0x46 +#define E610_SR_NETLIST_BANK_SIZE 0x47 +#define E610_SR_POINTER_TYPE_BIT BIT(15) +#define E610_SR_POINTER_MASK 0x7fff +#define E610_SR_HALF_4KB_SECTOR_UNITS 2048 +#define E610_GET_PFA_POINTER_IN_WORDS(offset) \ + ((offset & E610_SR_POINTER_TYPE_BIT) == E610_SR_POINTER_TYPE_BIT) ? \ + ((offset & E610_SR_POINTER_MASK) * E610_SR_HALF_4KB_SECTOR_UNITS) : \ + (offset & E610_SR_POINTER_MASK) + +/* Checksum and Shadow RAM pointers */ +#define E610_SR_NVM_CTRL_WORD 0x00 +#define E610_SR_PBA_BLOCK_PTR 0x16 + +/* The Orom version topology */ +#define IXGBE_OROM_VER_PATCH_SHIFT 0 +#define IXGBE_OROM_VER_PATCH_MASK (0xff << IXGBE_OROM_VER_PATCH_SHIFT) +#define IXGBE_OROM_VER_BUILD_SHIFT 8 +#define IXGBE_OROM_VER_BUILD_MASK (0xffff << IXGBE_OROM_VER_BUILD_SHIFT) +#define IXGBE_OROM_VER_SHIFT 24 +#define IXGBE_OROM_VER_MASK (0xff << IXGBE_OROM_VER_SHIFT) + +/* CSS Header words */ +#define IXGBE_NVM_CSS_HDR_LEN_L 0x02 +#define IXGBE_NVM_CSS_HDR_LEN_H 0x03 +#define IXGBE_NVM_CSS_SREV_L 0x14 +#define IXGBE_NVM_CSS_SREV_H 0x15 + +/* Length of Authentication header section in words */ +#define IXGBE_NVM_AUTH_HEADER_LEN 0x08 + +/* The Netlist ID Block is located after all of the Link Topology nodes. */ +#define IXGBE_NETLIST_ID_BLK_SIZE 0x30 +#define IXGBE_NETLIST_ID_BLK_OFFSET(n) IXGBE_NETLIST_LINK_TOPO_OFFSET(0x0004 + 2 * (n)) + +/* netlist ID block field offsets (word offsets) */ +#define IXGBE_NETLIST_ID_BLK_MAJOR_VER_LOW 0x02 +#define IXGBE_NETLIST_ID_BLK_MAJOR_VER_HIGH 0x03 +#define IXGBE_NETLIST_ID_BLK_MINOR_VER_LOW 0x04 +#define IXGBE_NETLIST_ID_BLK_MINOR_VER_HIGH 0x05 +#define IXGBE_NETLIST_ID_BLK_TYPE_LOW 0x06 +#define IXGBE_NETLIST_ID_BLK_TYPE_HIGH 0x07 +#define IXGBE_NETLIST_ID_BLK_REV_LOW 0x08 +#define IXGBE_NETLIST_ID_BLK_REV_HIGH 0x09 +#define IXGBE_NETLIST_ID_BLK_SHA_HASH_WORD(n) (0x0A + (n)) +#define IXGBE_NETLIST_ID_BLK_CUST_VER 0x2F + +/* The Link Topology Netlist section is stored as a series of words. It is + * stored in the NVM as a TLV, with the first two words containing the type + * and length. + */ +#define IXGBE_NETLIST_LINK_TOPO_MOD_ID 0x011B +#define IXGBE_NETLIST_TYPE_OFFSET 0x0000 +#define IXGBE_NETLIST_LEN_OFFSET 0x0001 + +/* The Link Topology section follows the TLV header. When reading the netlist + * using ixgbe_read_netlist_module, we need to account for the 2-word TLV + * header. + */ +#define IXGBE_NETLIST_LINK_TOPO_OFFSET(n) ((n) + 2) +#define IXGBE_LINK_TOPO_MODULE_LEN IXGBE_NETLIST_LINK_TOPO_OFFSET(0x0000) +#define IXGBE_LINK_TOPO_NODE_COUNT IXGBE_NETLIST_LINK_TOPO_OFFSET(0x0001) +#define IXGBE_LINK_TOPO_NODE_COUNT_M MAKEMASK(0x3FF, 0) + +/* Auxiliary field, mask and shift definition for Shadow RAM and NVM Flash */ +#define IXGBE_SR_CTRL_WORD_1_S 0x06 +#define IXGBE_SR_CTRL_WORD_1_M (0x03 << IXGBE_SR_CTRL_WORD_1_S) +#define IXGBE_SR_CTRL_WORD_VALID 0x1 +#define IXGBE_SR_CTRL_WORD_OROM_BANK BIT(3) +#define IXGBE_SR_CTRL_WORD_NETLIST_BANK BIT(4) +#define IXGBE_SR_CTRL_WORD_NVM_BANK BIT(5) +#define IXGBE_SR_NVM_PTR_4KB_UNITS BIT(15) + +/* These macros strip from NVM Image Revision the particular part of NVM ver: + major ver, minor ver and image id */ +#define E610_NVM_MAJOR_VER(x) ((x & 0xF000) >> 12) +#define E610_NVM_MINOR_VER(x) (x & 0x00FF) + +/* Minimal Security Revision */ + +/* Shadow RAM related */ +#define IXGBE_SR_SECTOR_SIZE_IN_WORDS 0x800 +#define IXGBE_SR_WORDS_IN_1KB 512 +/* Checksum should be calculated such that after adding all the words, + * including the checksum word itself, the sum should be 0xBABA. + */ +#define IXGBE_SR_SW_CHECKSUM_BASE 0xBABA + +/* Netlist */ +#define IXGBE_MAX_NETLIST_SIZE 10 + +/* General registers */ + +/* Firmware Status Register (GL_FWSTS) */ +#define GL_FWSTS 0x00083048 /* Reset Source: POR */ +#define GL_FWSTS_FWS0B_S 0 +#define GL_FWSTS_FWS0B_M MAKEMASK(0xFF, 0) +#define GL_FWSTS_FWROWD_S 8 +#define GL_FWSTS_FWROWD_M BIT(8) +#define GL_FWSTS_FWRI_S 9 +#define GL_FWSTS_FWRI_M BIT(9) +#define GL_FWSTS_FWS1B_S 16 +#define GL_FWSTS_FWS1B_M MAKEMASK(0xFF, 16) +#define GL_FWSTS_EP_PF0 BIT(24) +#define GL_FWSTS_EP_PF1 BIT(25) + +/* Recovery mode values of Firmware Status 1 Byte (FWS1B) bitfield */ +#define GL_FWSTS_FWS1B_RECOVERY_MODE_CORER_LEGACY 0x0B +#define GL_FWSTS_FWS1B_RECOVERY_MODE_GLOBR_LEGACY 0x0C +#define GL_FWSTS_FWS1B_RECOVERY_MODE_CORER 0x30 +#define GL_FWSTS_FWS1B_RECOVERY_MODE_GLOBR 0x31 +#define GL_FWSTS_FWS1B_RECOVERY_MODE_TRANSITION 0x32 +#define GL_FWSTS_FWS1B_RECOVERY_MODE_NVM 0x33 + +/* Firmware Status (GL_MNG_FWSM) */ +#define GL_MNG_FWSM 0x000B6134 /* Reset Source: POR */ +#define GL_MNG_FWSM_FW_MODES_S 0 +#define GL_MNG_FWSM_FW_MODES_M MAKEMASK(0x7, 0) +#define GL_MNG_FWSM_RSV0_S 2 +#define GL_MNG_FWSM_RSV0_M MAKEMASK(0xFF, 2) +#define GL_MNG_FWSM_EEP_RELOAD_IND_S 10 +#define GL_MNG_FWSM_EEP_RELOAD_IND_M BIT(10) +#define GL_MNG_FWSM_RSV1_S 11 +#define GL_MNG_FWSM_RSV1_M MAKEMASK(0xF, 11) +#define GL_MNG_FWSM_RSV2_S 15 +#define GL_MNG_FWSM_RSV2_M BIT(15) +#define GL_MNG_FWSM_PCIR_AL_FAILURE_S 16 +#define GL_MNG_FWSM_PCIR_AL_FAILURE_M BIT(16) +#define GL_MNG_FWSM_POR_AL_FAILURE_S 17 +#define GL_MNG_FWSM_POR_AL_FAILURE_M BIT(17) +#define GL_MNG_FWSM_RSV3_S 18 +#define GL_MNG_FWSM_RSV3_M BIT(18) +#define GL_MNG_FWSM_EXT_ERR_IND_S 19 +#define GL_MNG_FWSM_EXT_ERR_IND_M MAKEMASK(0x3F, 19) +#define GL_MNG_FWSM_RSV4_S 25 +#define GL_MNG_FWSM_RSV4_M BIT(25) +#define GL_MNG_FWSM_RESERVED_11_S 26 +#define GL_MNG_FWSM_RESERVED_11_M MAKEMASK(0xF, 26) +#define GL_MNG_FWSM_RSV5_S 30 +#define GL_MNG_FWSM_RSV5_M MAKEMASK(0x3, 30) + +/* FW mode indications */ +#define GL_MNG_FWSM_FW_MODES_DEBUG_M BIT(0) +#define GL_MNG_FWSM_FW_MODES_RECOVERY_M BIT(1) +#define GL_MNG_FWSM_FW_MODES_ROLLBACK_M BIT(2) + +/* PF - Manageability Registers */ + +/* Global NVM General Status Register */ +#define GLNVM_GENS 0x000B6100 /* Reset Source: POR */ +#define GLNVM_GENS_NVM_PRES_S 0 +#define GLNVM_GENS_NVM_PRES_M BIT(0) +#define GLNVM_GENS_SR_SIZE_S 5 +#define GLNVM_GENS_SR_SIZE_M MAKEMASK(0x7, 5) +#define GLNVM_GENS_BANK1VAL_S 8 +#define GLNVM_GENS_BANK1VAL_M BIT(8) +#define GLNVM_GENS_ALT_PRST_S 23 +#define GLNVM_GENS_ALT_PRST_M BIT(23) +#define GLNVM_GENS_FL_AUTO_RD_S 25 +#define GLNVM_GENS_FL_AUTO_RD_M BIT(25) + +/* Flash Access Register */ +#define GLNVM_FLA 0x000B6108 /* Reset Source: POR */ +#define GLNVM_FLA_LOCKED_S 6 +#define GLNVM_FLA_LOCKED_M BIT(6) + +/* Bit Bang registers */ +#define RDASB_MSGCTL 0x000B6820 +#define RDASB_MSGCTL_HDR_DWS_S 0 +#define RDASB_MSGCTL_EXP_RDW_S 8 +#define RDASB_MSGCTL_CMDV_M BIT(31) +#define RDASB_RSPCTL 0x000B6824 +#define RDASB_RSPCTL_BAD_LENGTH_M BIT(30) +#define RDASB_RSPCTL_NOT_SUCCESS_M BIT(31) +#define RDASB_WHDR0 0x000B68F4 +#define RDASB_WHDR1 0x000B68F8 +#define RDASB_WHDR2 0x000B68FC +#define RDASB_WHDR3 0x000B6900 +#define RDASB_WHDR4 0x000B6904 +#define RDASB_RHDR0 0x000B6AFC +#define RDASB_RHDR0_RESPONSE_S 27 +#define RDASB_RHDR0_RESPONSE_M MAKEMASK(0x7, 27) +#define RDASB_RDATA0 0x000B6B00 +#define RDASB_RDATA1 0x000B6B04 + +/* SPI Registers */ +#define SPISB_MSGCTL 0x000B7020 +#define SPISB_MSGCTL_HDR_DWS_S 0 +#define SPISB_MSGCTL_EXP_RDW_S 8 +#define SPISB_MSGCTL_MSG_MODE_S 26 +#define SPISB_MSGCTL_TOKEN_MODE_S 28 +#define SPISB_MSGCTL_BARCLR_S 30 +#define SPISB_MSGCTL_CMDV_S 31 +#define SPISB_MSGCTL_CMDV_M BIT(31) +#define SPISB_RSPCTL 0x000B7024 +#define SPISB_RSPCTL_BAD_LENGTH_M BIT(30) +#define SPISB_RSPCTL_NOT_SUCCESS_M BIT(31) +#define SPISB_WHDR0 0x000B70F4 +#define SPISB_WHDR0_DEST_SEL_S 12 +#define SPISB_WHDR0_OPCODE_SEL_S 16 +#define SPISB_WHDR0_TAG_S 24 +#define SPISB_WHDR1 0x000B70F8 +#define SPISB_WHDR2 0x000B70FC +#define SPISB_RDATA 0x000B7300 +#define SPISB_WDATA 0x000B7100 + +/* Firmware Reset Count register */ +#define GL_FWRESETCNT 0x00083100 /* Reset Source: POR */ +#define GL_FWRESETCNT_FWRESETCNT_S 0 +#define GL_FWRESETCNT_FWRESETCNT_M MAKEMASK(0xFFFFFFFF, 0) + +/* Admin Command Interface (ACI) registers */ +#define PF_HIDA(_i) (0x00085000 + ((_i) * 4)) +#define PF_HIDA_2(_i) (0x00085020 + ((_i) * 4)) +#define PF_HIBA(_i) (0x00084000 + ((_i) * 4)) +#define PF_HICR 0x00082048 + +#define PF_HIDA_MAX_INDEX 15 +#define PF_HIBA_MAX_INDEX 1023 + +#define PF_HICR_EN BIT(0) +#define PF_HICR_C BIT(1) +#define PF_HICR_SV BIT(2) +#define PF_HICR_EV BIT(3) + +#define GL_HIDA(_i) (0x00082000 + ((_i) * 4)) +#define GL_HIDA_2(_i) (0x00082020 + ((_i) * 4)) +#define GL_HIBA(_i) (0x00081000 + ((_i) * 4)) +#define GL_HICR 0x00082040 + +#define GL_HIDA_MAX_INDEX 15 +#define GL_HIBA_MAX_INDEX 1023 + +#define GL_HICR_C BIT(1) +#define GL_HICR_SV BIT(2) +#define GL_HICR_EV BIT(3) + +#define GL_HICR_EN 0x00082044 + +#define GL_HICR_EN_CHECK BIT(0) + +/* Admin Command Interface (ACI) defines */ +/* Defines that help manage the driver vs FW API checks. + */ +#define IXGBE_FW_API_VER_BRANCH 0x00 +#define IXGBE_FW_API_VER_MAJOR 0x01 +#define IXGBE_FW_API_VER_MINOR 0x05 +#define IXGBE_FW_API_VER_DIFF_ALLOWED 0x02 + +#define IXGBE_ACI_DESC_SIZE 32 +#define IXGBE_ACI_DESC_SIZE_IN_DWORDS IXGBE_ACI_DESC_SIZE / BYTES_PER_DWORD + +#define IXGBE_ACI_MAX_BUFFER_SIZE 4096 /* Size in bytes */ +#define IXGBE_ACI_DESC_COOKIE_L_DWORD_OFFSET 3 +#define IXGBE_ACI_SEND_DELAY_TIME_MS 10 +#define IXGBE_ACI_SEND_MAX_EXECUTE 3 +/* [ms] timeout of waiting for sync response */ +#define IXGBE_ACI_SYNC_RESPONSE_TIMEOUT 100000 +/* [ms] timeout of waiting for async response */ +#define IXGBE_ACI_ASYNC_RESPONSE_TIMEOUT 150000 +/* [ms] timeout of waiting for resource release */ +#define IXGBE_ACI_RELEASE_RES_TIMEOUT 10000 + +/* Timestamp spacing for Tools ACI: queue is active if spacing is within the range [LO..HI] */ +#define IXGBE_TOOLS_ACI_ACTIVE_STAMP_SPACING_LO 0 +#define IXGBE_TOOLS_ACI_ACTIVE_STAMP_SPACING_HI 200 + +/* Timestamp spacing for Tools ACI: queue is expired if spacing is outside the range [LO..HI] */ +#define IXGBE_TOOLS_ACI_EXPIRED_STAMP_SPACING_LO -5 +#define IXGBE_TOOLS_ACI_EXPIRED_STAMP_SPACING_HI 205 + +/* FW defined boundary for a large buffer, 4k >= Large buffer > 512 bytes */ +#define IXGBE_ACI_LG_BUF 512 + +/* Flags sub-structure + * |0 |1 |2 |3 |4 |5 |6 |7 |8 |9 |10 |11 |12 |13 |14 |15 | + * |DD |CMP|ERR|VFE| * * RESERVED * * |LB |RD |VFC|BUF|SI |EI |FE | + */ + +/* command flags and offsets */ +#define IXGBE_ACI_FLAG_DD_S 0 +#define IXGBE_ACI_FLAG_CMP_S 1 +#define IXGBE_ACI_FLAG_ERR_S 2 +#define IXGBE_ACI_FLAG_VFE_S 3 +#define IXGBE_ACI_FLAG_LB_S 9 +#define IXGBE_ACI_FLAG_RD_S 10 +#define IXGBE_ACI_FLAG_VFC_S 11 +#define IXGBE_ACI_FLAG_BUF_S 12 +#define IXGBE_ACI_FLAG_SI_S 13 +#define IXGBE_ACI_FLAG_EI_S 14 +#define IXGBE_ACI_FLAG_FE_S 15 + +#define IXGBE_ACI_FLAG_DD BIT(IXGBE_ACI_FLAG_DD_S) /* 0x1 */ +#define IXGBE_ACI_FLAG_CMP BIT(IXGBE_ACI_FLAG_CMP_S) /* 0x2 */ +#define IXGBE_ACI_FLAG_ERR BIT(IXGBE_ACI_FLAG_ERR_S) /* 0x4 */ +#define IXGBE_ACI_FLAG_VFE BIT(IXGBE_ACI_FLAG_VFE_S) /* 0x8 */ +#define IXGBE_ACI_FLAG_LB BIT(IXGBE_ACI_FLAG_LB_S) /* 0x200 */ +#define IXGBE_ACI_FLAG_RD BIT(IXGBE_ACI_FLAG_RD_S) /* 0x400 */ +#define IXGBE_ACI_FLAG_VFC BIT(IXGBE_ACI_FLAG_VFC_S) /* 0x800 */ +#define IXGBE_ACI_FLAG_BUF BIT(IXGBE_ACI_FLAG_BUF_S) /* 0x1000 */ +#define IXGBE_ACI_FLAG_SI BIT(IXGBE_ACI_FLAG_SI_S) /* 0x2000 */ +#define IXGBE_ACI_FLAG_EI BIT(IXGBE_ACI_FLAG_EI_S) /* 0x4000 */ +#define IXGBE_ACI_FLAG_FE BIT(IXGBE_ACI_FLAG_FE_S) /* 0x8000 */ + +/* Admin Command Interface (ACI) error codes */ +enum ixgbe_aci_err { + IXGBE_ACI_RC_OK = 0, /* Success */ + IXGBE_ACI_RC_EPERM = 1, /* Operation not permitted */ + IXGBE_ACI_RC_ENOENT = 2, /* No such element */ + IXGBE_ACI_RC_ESRCH = 3, /* Bad opcode */ + IXGBE_ACI_RC_EINTR = 4, /* Operation interrupted */ + IXGBE_ACI_RC_EIO = 5, /* I/O error */ + IXGBE_ACI_RC_ENXIO = 6, /* No such resource */ + IXGBE_ACI_RC_E2BIG = 7, /* Arg too long */ + IXGBE_ACI_RC_EAGAIN = 8, /* Try again */ + IXGBE_ACI_RC_ENOMEM = 9, /* Out of memory */ + IXGBE_ACI_RC_EACCES = 10, /* Permission denied */ + IXGBE_ACI_RC_EFAULT = 11, /* Bad address */ + IXGBE_ACI_RC_EBUSY = 12, /* Device or resource busy */ + IXGBE_ACI_RC_EEXIST = 13, /* Object already exists */ + IXGBE_ACI_RC_EINVAL = 14, /* Invalid argument */ + IXGBE_ACI_RC_ENOTTY = 15, /* Not a typewriter */ + IXGBE_ACI_RC_ENOSPC = 16, /* No space left or allocation failure */ + IXGBE_ACI_RC_ENOSYS = 17, /* Function not implemented */ + IXGBE_ACI_RC_ERANGE = 18, /* Parameter out of range */ + IXGBE_ACI_RC_EFLUSHED = 19, /* Cmd flushed due to prev cmd error */ + IXGBE_ACI_RC_BAD_ADDR = 20, /* Descriptor contains a bad pointer */ + IXGBE_ACI_RC_EMODE = 21, /* Op not allowed in current dev mode */ + IXGBE_ACI_RC_EFBIG = 22, /* File too big */ + IXGBE_ACI_RC_ESBCOMP = 23, /* SB-IOSF completion unsuccessful */ + IXGBE_ACI_RC_ENOSEC = 24, /* Missing security manifest */ + IXGBE_ACI_RC_EBADSIG = 25, /* Bad RSA signature */ + IXGBE_ACI_RC_ESVN = 26, /* SVN number prohibits this package */ + IXGBE_ACI_RC_EBADMAN = 27, /* Manifest hash mismatch */ + IXGBE_ACI_RC_EBADBUF = 28, /* Buffer hash mismatches manifest */ + IXGBE_ACI_RC_EACCES_BMCU = 29, /* BMC Update in progress */ +}; + +/* Admin Command Interface (ACI) opcodes */ +enum ixgbe_aci_opc { + ixgbe_aci_opc_get_ver = 0x0001, + ixgbe_aci_opc_driver_ver = 0x0002, + ixgbe_aci_opc_get_exp_err = 0x0005, + + /* resource ownership */ + ixgbe_aci_opc_req_res = 0x0008, + ixgbe_aci_opc_release_res = 0x0009, + + /* device/function capabilities */ + ixgbe_aci_opc_list_func_caps = 0x000A, + ixgbe_aci_opc_list_dev_caps = 0x000B, + + /* safe disable of RXEN */ + ixgbe_aci_opc_disable_rxen = 0x000C, + + /* FW events */ + ixgbe_aci_opc_get_fw_event = 0x0014, + + /* PHY commands */ + ixgbe_aci_opc_get_phy_caps = 0x0600, + ixgbe_aci_opc_set_phy_cfg = 0x0601, + ixgbe_aci_opc_restart_an = 0x0605, + ixgbe_aci_opc_get_link_status = 0x0607, + ixgbe_aci_opc_set_event_mask = 0x0613, + ixgbe_aci_opc_get_link_topo = 0x06E0, + ixgbe_aci_opc_get_link_topo_pin = 0x06E1, + ixgbe_aci_opc_read_i2c = 0x06E2, + ixgbe_aci_opc_write_i2c = 0x06E3, + ixgbe_aci_opc_read_mdio = 0x06E4, + ixgbe_aci_opc_write_mdio = 0x06E5, + ixgbe_aci_opc_set_gpio_by_func = 0x06E6, + ixgbe_aci_opc_get_gpio_by_func = 0x06E7, + ixgbe_aci_opc_set_gpio = 0x06EC, + ixgbe_aci_opc_get_gpio = 0x06ED, + ixgbe_aci_opc_sff_eeprom = 0x06EE, + ixgbe_aci_opc_prog_topo_dev_nvm = 0x06F2, + ixgbe_aci_opc_read_topo_dev_nvm = 0x06F3, + + /* NVM commands */ + ixgbe_aci_opc_nvm_read = 0x0701, + ixgbe_aci_opc_nvm_erase = 0x0702, + ixgbe_aci_opc_nvm_write = 0x0703, + ixgbe_aci_opc_nvm_cfg_read = 0x0704, + ixgbe_aci_opc_nvm_cfg_write = 0x0705, + ixgbe_aci_opc_nvm_checksum = 0x0706, + ixgbe_aci_opc_nvm_write_activate = 0x0707, + ixgbe_aci_opc_nvm_sr_dump = 0x0707, + ixgbe_aci_opc_nvm_save_factory_settings = 0x0708, + ixgbe_aci_opc_nvm_update_empr = 0x0709, + ixgbe_aci_opc_nvm_pkg_data = 0x070A, + ixgbe_aci_opc_nvm_pass_component_tbl = 0x070B, + ixgbe_aci_opc_nvm_sanitization = 0x070C, + + /* Alternate Structure Commands */ + ixgbe_aci_opc_write_alt_direct = 0x0900, + ixgbe_aci_opc_write_alt_indirect = 0x0901, + ixgbe_aci_opc_read_alt_direct = 0x0902, + ixgbe_aci_opc_read_alt_indirect = 0x0903, + ixgbe_aci_opc_done_alt_write = 0x0904, + ixgbe_aci_opc_clear_port_alt_write = 0x0906, + + /* debug commands */ + ixgbe_aci_opc_debug_dump_internals = 0xFF08, + + /* SystemDiagnostic commands */ + ixgbe_aci_opc_set_health_status_config = 0xFF20, + ixgbe_aci_opc_get_supported_health_status_codes = 0xFF21, + ixgbe_aci_opc_get_health_status = 0xFF22, + ixgbe_aci_opc_clear_health_status = 0xFF23, + +}; + +/* This macro is used to generate a compilation error if a structure + * is not exactly the correct length. It gives a divide by zero error if the + * structure is not of the correct size, otherwise it creates an enum that is + * never used. + */ +#define IXGBE_CHECK_STRUCT_LEN(n, X) enum ixgbe_static_assert_enum_##X \ + { ixgbe_static_assert_##X = (n) / ((sizeof(struct X) == (n)) ? 1 : 0) } + +/* This macro is used to generate a compilation error if a variable-length + * structure is not exactly the correct length assuming a single element of + * the variable-length object as the last element of the structure. It gives + * a divide by zero error if the structure is not of the correct size, + * otherwise it creates an enum that is never used. + */ +#define IXGBE_CHECK_VAR_LEN_STRUCT_LEN(n, X, T) enum ixgbe_static_assert_enum_##X \ + { ixgbe_static_assert_##X = (n) / \ + (((sizeof(struct X) + sizeof(T)) == (n)) ? 1 : 0) } + +/* This macro is used to ensure that parameter structures (i.e. structures + * in the params union member of struct ixgbe_aci_desc) are 16 bytes in length. + * + * NOT intended to be used to check the size of an indirect command/response + * additional data buffer (e.g. struct foo) which should just happen to be 16 + * bytes (instead, use IXGBE_CHECK_STRUCT_LEN(16, foo) for that). + */ +#define IXGBE_CHECK_PARAM_LEN(X) IXGBE_CHECK_STRUCT_LEN(16, X) + +struct ixgbe_aci_cmd_generic { + __le32 param0; + __le32 param1; + __le32 addr_high; + __le32 addr_low; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_generic); + +/* Get version (direct 0x0001) */ +struct ixgbe_aci_cmd_get_ver { + __le32 rom_ver; + __le32 fw_build; + u8 fw_branch; + u8 fw_major; + u8 fw_minor; + u8 fw_patch; + u8 api_branch; + u8 api_major; + u8 api_minor; + u8 api_patch; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_get_ver); + +#define IXGBE_DRV_VER_STR_LEN_E610 32 + +struct ixgbe_driver_ver { + u8 major_ver; + u8 minor_ver; + u8 build_ver; + u8 subbuild_ver; + u8 driver_string[IXGBE_DRV_VER_STR_LEN_E610]; +}; + +/* Send driver version (indirect 0x0002) */ +struct ixgbe_aci_cmd_driver_ver { + u8 major_ver; + u8 minor_ver; + u8 build_ver; + u8 subbuild_ver; + u8 reserved[4]; + __le32 addr_high; + __le32 addr_low; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_driver_ver); + +/* Get Expanded Error Code (0x0005, direct) */ +struct ixgbe_aci_cmd_get_exp_err { + __le32 reason; +#define IXGBE_ACI_EXPANDED_ERROR_NOT_PROVIDED 0xFFFFFFFF + __le32 identifier; + u8 rsvd[8]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_get_exp_err); + +/* FW update timeout definitions are in milliseconds */ +#define IXGBE_NVM_TIMEOUT 180000 +#define IXGBE_CHANGE_LOCK_TIMEOUT 1000 +#define IXGBE_GLOBAL_CFG_LOCK_TIMEOUT 3000 + +enum ixgbe_aci_res_access_type { + IXGBE_RES_READ = 1, + IXGBE_RES_WRITE +}; + +enum ixgbe_aci_res_ids { + IXGBE_NVM_RES_ID = 1, + IXGBE_SPD_RES_ID, + IXGBE_CHANGE_LOCK_RES_ID, + IXGBE_GLOBAL_CFG_LOCK_RES_ID +}; + +/* Request resource ownership (direct 0x0008) + * Release resource ownership (direct 0x0009) + */ +struct ixgbe_aci_cmd_req_res { + __le16 res_id; +#define IXGBE_ACI_RES_ID_NVM 1 +#define IXGBE_ACI_RES_ID_SDP 2 +#define IXGBE_ACI_RES_ID_CHNG_LOCK 3 +#define IXGBE_ACI_RES_ID_GLBL_LOCK 4 + __le16 access_type; +#define IXGBE_ACI_RES_ACCESS_READ 1 +#define IXGBE_ACI_RES_ACCESS_WRITE 2 + + /* Upon successful completion, FW writes this value and driver is + * expected to release resource before timeout. This value is provided + * in milliseconds. + */ + __le32 timeout; +#define IXGBE_ACI_RES_NVM_READ_DFLT_TIMEOUT_MS 3000 +#define IXGBE_ACI_RES_NVM_WRITE_DFLT_TIMEOUT_MS 180000 +#define IXGBE_ACI_RES_CHNG_LOCK_DFLT_TIMEOUT_MS 1000 +#define IXGBE_ACI_RES_GLBL_LOCK_DFLT_TIMEOUT_MS 3000 + /* For SDP: pin ID of the SDP */ + __le32 res_number; + /* Status is only used for IXGBE_ACI_RES_ID_GLBL_LOCK */ + __le16 status; +#define IXGBE_ACI_RES_GLBL_SUCCESS 0 +#define IXGBE_ACI_RES_GLBL_IN_PROG 1 +#define IXGBE_ACI_RES_GLBL_DONE 2 + u8 reserved[2]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_req_res); + +/* Get function capabilities (indirect 0x000A) + * Get device capabilities (indirect 0x000B) + */ +struct ixgbe_aci_cmd_list_caps { + u8 cmd_flags; + u8 pf_index; + u8 reserved[2]; + __le32 count; + __le32 addr_high; + __le32 addr_low; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_list_caps); + +/* Device/Function buffer entry, repeated per reported capability */ +struct ixgbe_aci_cmd_list_caps_elem { + __le16 cap; +#define IXGBE_ACI_CAPS_VALID_FUNCTIONS 0x0005 +#define IXGBE_ACI_MAX_VALID_FUNCTIONS 0x8 +#define IXGBE_ACI_CAPS_VMDQ 0x0014 +#define IXGBE_ACI_CAPS_VSI 0x0017 +#define IXGBE_ACI_CAPS_DCB 0x0018 +#define IXGBE_ACI_CAPS_RSS 0x0040 +#define IXGBE_ACI_CAPS_RXQS 0x0041 +#define IXGBE_ACI_CAPS_TXQS 0x0042 +#define IXGBE_ACI_CAPS_MSIX 0x0043 +#define IXGBE_ACI_CAPS_FD 0x0045 +#define IXGBE_ACI_CAPS_1588 0x0046 +#define IXGBE_ACI_CAPS_MAX_MTU 0x0047 +#define IXGBE_ACI_CAPS_NVM_VER 0x0048 +#define IXGBE_ACI_CAPS_OROM_VER 0x004A +#define IXGBE_ACI_CAPS_INLINE_IPSEC 0x0070 +#define IXGBE_ACI_CAPS_NUM_ENABLED_PORTS 0x0072 +#define IXGBE_ACI_CAPS_PCIE_RESET_AVOIDANCE 0x0076 +#define IXGBE_ACI_CAPS_POST_UPDATE_RESET_RESTRICT 0x0077 +#define IXGBE_ACI_CAPS_NVM_MGMT 0x0080 +#define IXGBE_ACI_CAPS_EXT_TOPO_DEV_IMG0 0x0081 +#define IXGBE_ACI_CAPS_EXT_TOPO_DEV_IMG1 0x0082 +#define IXGBE_ACI_CAPS_EXT_TOPO_DEV_IMG2 0x0083 +#define IXGBE_ACI_CAPS_EXT_TOPO_DEV_IMG3 0x0084 +#define IXGBE_ACI_CAPS_NEXT_CLUSTER_ID 0x0096 + u8 major_ver; + u8 minor_ver; + /* Number of resources described by this capability */ + __le32 number; + /* Only meaningful for some types of resources */ + __le32 logical_id; + /* Only meaningful for some types of resources */ + __le32 phys_id; + __le64 rsvd1; + __le64 rsvd2; +}; + +IXGBE_CHECK_STRUCT_LEN(32, ixgbe_aci_cmd_list_caps_elem); + +/* Disable RXEN (direct 0x000C) */ +struct ixgbe_aci_cmd_disable_rxen { + u8 lport_num; + u8 reserved[15]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_disable_rxen); + +/* Get FW Event (indirect 0x0014) */ +struct ixgbe_aci_cmd_get_fw_event { + __le16 fw_buf_status; +#define IXGBE_ACI_GET_FW_EVENT_STATUS_OBTAINED BIT(0) +#define IXGBE_ACI_GET_FW_EVENT_STATUS_PENDING BIT(1) + u8 rsvd[14]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_get_fw_event); + +/* Get PHY capabilities (indirect 0x0600) */ +struct ixgbe_aci_cmd_get_phy_caps { + u8 lport_num; + u8 reserved; + __le16 param0; + /* 18.0 - Report qualified modules */ +#define IXGBE_ACI_GET_PHY_RQM BIT(0) + /* 18.1 - 18.3 : Report mode + * 000b - Report topology capabilities, without media + * 001b - Report topology capabilities, with media + * 010b - Report Active configuration + * 011b - Report PHY Type and FEC mode capabilities + * 100b - Report Default capabilities + */ +#define IXGBE_ACI_REPORT_MODE_S 1 +#define IXGBE_ACI_REPORT_MODE_M (7 << IXGBE_ACI_REPORT_MODE_S) +#define IXGBE_ACI_REPORT_TOPO_CAP_NO_MEDIA 0 +#define IXGBE_ACI_REPORT_TOPO_CAP_MEDIA BIT(1) +#define IXGBE_ACI_REPORT_ACTIVE_CFG BIT(2) +#define IXGBE_ACI_REPORT_DFLT_CFG BIT(3) + __le32 reserved1; + __le32 addr_high; + __le32 addr_low; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_get_phy_caps); + +/* This is #define of PHY type (Extended): + * The first set of defines is for phy_type_low. + */ +#define IXGBE_PHY_TYPE_LOW_100BASE_TX BIT_ULL(0) +#define IXGBE_PHY_TYPE_LOW_100M_SGMII BIT_ULL(1) +#define IXGBE_PHY_TYPE_LOW_1000BASE_T BIT_ULL(2) +#define IXGBE_PHY_TYPE_LOW_1000BASE_SX BIT_ULL(3) +#define IXGBE_PHY_TYPE_LOW_1000BASE_LX BIT_ULL(4) +#define IXGBE_PHY_TYPE_LOW_1000BASE_KX BIT_ULL(5) +#define IXGBE_PHY_TYPE_LOW_1G_SGMII BIT_ULL(6) +#define IXGBE_PHY_TYPE_LOW_2500BASE_T BIT_ULL(7) +#define IXGBE_PHY_TYPE_LOW_2500BASE_X BIT_ULL(8) +#define IXGBE_PHY_TYPE_LOW_2500BASE_KX BIT_ULL(9) +#define IXGBE_PHY_TYPE_LOW_5GBASE_T BIT_ULL(10) +#define IXGBE_PHY_TYPE_LOW_5GBASE_KR BIT_ULL(11) +#define IXGBE_PHY_TYPE_LOW_10GBASE_T BIT_ULL(12) +#define IXGBE_PHY_TYPE_LOW_10G_SFI_DA BIT_ULL(13) +#define IXGBE_PHY_TYPE_LOW_10GBASE_SR BIT_ULL(14) +#define IXGBE_PHY_TYPE_LOW_10GBASE_LR BIT_ULL(15) +#define IXGBE_PHY_TYPE_LOW_10GBASE_KR_CR1 BIT_ULL(16) +#define IXGBE_PHY_TYPE_LOW_10G_SFI_AOC_ACC BIT_ULL(17) +#define IXGBE_PHY_TYPE_LOW_10G_SFI_C2C BIT_ULL(18) +#define IXGBE_PHY_TYPE_LOW_25GBASE_T BIT_ULL(19) +#define IXGBE_PHY_TYPE_LOW_25GBASE_CR BIT_ULL(20) +#define IXGBE_PHY_TYPE_LOW_25GBASE_CR_S BIT_ULL(21) +#define IXGBE_PHY_TYPE_LOW_25GBASE_CR1 BIT_ULL(22) +#define IXGBE_PHY_TYPE_LOW_25GBASE_SR BIT_ULL(23) +#define IXGBE_PHY_TYPE_LOW_25GBASE_LR BIT_ULL(24) +#define IXGBE_PHY_TYPE_LOW_25GBASE_KR BIT_ULL(25) +#define IXGBE_PHY_TYPE_LOW_25GBASE_KR_S BIT_ULL(26) +#define IXGBE_PHY_TYPE_LOW_25GBASE_KR1 BIT_ULL(27) +#define IXGBE_PHY_TYPE_LOW_25G_AUI_AOC_ACC BIT_ULL(28) +#define IXGBE_PHY_TYPE_LOW_25G_AUI_C2C BIT_ULL(29) +#define IXGBE_PHY_TYPE_LOW_MAX_INDEX 29 +/* The second set of defines is for phy_type_high. */ +#define IXGBE_PHY_TYPE_HIGH_10BASE_T BIT_ULL(1) +#define IXGBE_PHY_TYPE_HIGH_10M_SGMII BIT_ULL(2) +#define IXGBE_PHY_TYPE_HIGH_2500M_SGMII BIT_ULL(56) +#define IXGBE_PHY_TYPE_HIGH_100M_USXGMII BIT_ULL(57) +#define IXGBE_PHY_TYPE_HIGH_1G_USXGMII BIT_ULL(58) +#define IXGBE_PHY_TYPE_HIGH_2500M_USXGMII BIT_ULL(59) +#define IXGBE_PHY_TYPE_HIGH_5G_USXGMII BIT_ULL(60) +#define IXGBE_PHY_TYPE_HIGH_10G_USXGMII BIT_ULL(61) +#define IXGBE_PHY_TYPE_HIGH_MAX_INDEX 61 + +struct ixgbe_aci_cmd_get_phy_caps_data { + __le64 phy_type_low; /* Use values from IXGBE_PHY_TYPE_LOW_* */ + __le64 phy_type_high; /* Use values from IXGBE_PHY_TYPE_HIGH_* */ + u8 caps; +#define IXGBE_ACI_PHY_EN_TX_LINK_PAUSE BIT(0) +#define IXGBE_ACI_PHY_EN_RX_LINK_PAUSE BIT(1) +#define IXGBE_ACI_PHY_LOW_POWER_MODE BIT(2) +#define IXGBE_ACI_PHY_EN_LINK BIT(3) +#define IXGBE_ACI_PHY_AN_MODE BIT(4) +#define IXGBE_ACI_PHY_EN_MOD_QUAL BIT(5) +#define IXGBE_ACI_PHY_EN_LESM BIT(6) +#define IXGBE_ACI_PHY_EN_AUTO_FEC BIT(7) +#define IXGBE_ACI_PHY_CAPS_MASK MAKEMASK(0xff, 0) + u8 low_power_ctrl_an; +#define IXGBE_ACI_PHY_EN_D3COLD_LOW_POWER_AUTONEG BIT(0) +#define IXGBE_ACI_PHY_AN_EN_CLAUSE28 BIT(1) +#define IXGBE_ACI_PHY_AN_EN_CLAUSE73 BIT(2) +#define IXGBE_ACI_PHY_AN_EN_CLAUSE37 BIT(3) + __le16 eee_cap; +#define IXGBE_ACI_PHY_EEE_EN_100BASE_TX BIT(0) +#define IXGBE_ACI_PHY_EEE_EN_1000BASE_T BIT(1) +#define IXGBE_ACI_PHY_EEE_EN_10GBASE_T BIT(2) +#define IXGBE_ACI_PHY_EEE_EN_1000BASE_KX BIT(3) +#define IXGBE_ACI_PHY_EEE_EN_10GBASE_KR BIT(4) +#define IXGBE_ACI_PHY_EEE_EN_25GBASE_KR BIT(5) +#define IXGBE_ACI_PHY_EEE_EN_10BASE_T BIT(11) + __le16 eeer_value; + u8 phy_id_oui[4]; /* PHY/Module ID connected on the port */ + u8 phy_fw_ver[8]; + u8 link_fec_options; +#define IXGBE_ACI_PHY_FEC_10G_KR_40G_KR4_EN BIT(0) +#define IXGBE_ACI_PHY_FEC_10G_KR_40G_KR4_REQ BIT(1) +#define IXGBE_ACI_PHY_FEC_25G_RS_528_REQ BIT(2) +#define IXGBE_ACI_PHY_FEC_25G_KR_REQ BIT(3) +#define IXGBE_ACI_PHY_FEC_25G_RS_544_REQ BIT(4) +#define IXGBE_ACI_PHY_FEC_25G_RS_CLAUSE91_EN BIT(6) +#define IXGBE_ACI_PHY_FEC_25G_KR_CLAUSE74_EN BIT(7) +#define IXGBE_ACI_PHY_FEC_MASK MAKEMASK(0xdf, 0) + u8 module_compliance_enforcement; +#define IXGBE_ACI_MOD_ENFORCE_STRICT_MODE BIT(0) + u8 extended_compliance_code; +#define IXGBE_ACI_MODULE_TYPE_TOTAL_BYTE 3 + u8 module_type[IXGBE_ACI_MODULE_TYPE_TOTAL_BYTE]; +#define IXGBE_ACI_MOD_TYPE_BYTE0_SFP_PLUS 0xA0 +#define IXGBE_ACI_MOD_TYPE_BYTE0_QSFP_PLUS 0x80 +#define IXGBE_ACI_MOD_TYPE_IDENT 1 +#define IXGBE_ACI_MOD_TYPE_BYTE1_SFP_PLUS_CU_PASSIVE BIT(0) +#define IXGBE_ACI_MOD_TYPE_BYTE1_SFP_PLUS_CU_ACTIVE BIT(1) +#define IXGBE_ACI_MOD_TYPE_BYTE1_10G_BASE_SR BIT(4) +#define IXGBE_ACI_MOD_TYPE_BYTE1_10G_BASE_LR BIT(5) +#define IXGBE_ACI_MOD_TYPE_BYTE1_10G_BASE_LRM BIT(6) +#define IXGBE_ACI_MOD_TYPE_BYTE1_10G_BASE_ER BIT(7) +#define IXGBE_ACI_MOD_TYPE_BYTE2_SFP_PLUS 0xA0 +#define IXGBE_ACI_MOD_TYPE_BYTE2_QSFP_PLUS 0x86 + u8 qualified_module_count; + u8 rsvd2[7]; /* Bytes 47:41 reserved */ +#define IXGBE_ACI_QUAL_MOD_COUNT_MAX 16 + struct { + u8 v_oui[3]; + u8 rsvd3; + u8 v_part[16]; + __le32 v_rev; + __le64 rsvd4; + } qual_modules[IXGBE_ACI_QUAL_MOD_COUNT_MAX]; +}; + +IXGBE_CHECK_STRUCT_LEN(560, ixgbe_aci_cmd_get_phy_caps_data); + +/* Set PHY capabilities (direct 0x0601) + * NOTE: This command must be followed by setup link and restart auto-neg + */ +struct ixgbe_aci_cmd_set_phy_cfg { + u8 reserved[8]; + __le32 addr_high; + __le32 addr_low; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_set_phy_cfg); + +/* Set PHY config command data structure */ +struct ixgbe_aci_cmd_set_phy_cfg_data { + __le64 phy_type_low; /* Use values from IXGBE_PHY_TYPE_LOW_* */ + __le64 phy_type_high; /* Use values from IXGBE_PHY_TYPE_HIGH_* */ + u8 caps; +#define IXGBE_ACI_PHY_ENA_VALID_MASK MAKEMASK(0xef, 0) +#define IXGBE_ACI_PHY_ENA_TX_PAUSE_ABILITY BIT(0) +#define IXGBE_ACI_PHY_ENA_RX_PAUSE_ABILITY BIT(1) +#define IXGBE_ACI_PHY_ENA_LOW_POWER BIT(2) +#define IXGBE_ACI_PHY_ENA_LINK BIT(3) +#define IXGBE_ACI_PHY_ENA_AUTO_LINK_UPDT BIT(5) +#define IXGBE_ACI_PHY_ENA_LESM BIT(6) +#define IXGBE_ACI_PHY_ENA_AUTO_FEC BIT(7) + u8 low_power_ctrl_an; + __le16 eee_cap; /* Value from ixgbe_aci_get_phy_caps */ + __le16 eeer_value; /* Use defines from ixgbe_aci_get_phy_caps */ + u8 link_fec_opt; /* Use defines from ixgbe_aci_get_phy_caps */ + u8 module_compliance_enforcement; +}; + +IXGBE_CHECK_STRUCT_LEN(24, ixgbe_aci_cmd_set_phy_cfg_data); + +/* Restart AN command data structure (direct 0x0605) + * Also used for response, with only the lport_num field present. + */ +struct ixgbe_aci_cmd_restart_an { + u8 reserved[2]; + u8 cmd_flags; +#define IXGBE_ACI_RESTART_AN_LINK_RESTART BIT(1) +#define IXGBE_ACI_RESTART_AN_LINK_ENABLE BIT(2) + u8 reserved2[13]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_restart_an); + +#pragma pack(1) +/* Get link status (indirect 0x0607), also used for Link Status Event */ +struct ixgbe_aci_cmd_get_link_status { + u8 reserved[2]; + u8 cmd_flags; +#define IXGBE_ACI_LSE_M 0x3 +#define IXGBE_ACI_LSE_NOP 0x0 +#define IXGBE_ACI_LSE_DIS 0x2 +#define IXGBE_ACI_LSE_ENA 0x3 + /* only response uses this flag */ +#define IXGBE_ACI_LSE_IS_ENABLED 0x1 + u8 reserved2[5]; + __le32 addr_high; + __le32 addr_low; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_get_link_status); + +/* Get link status response data structure, also used for Link Status Event */ +struct ixgbe_aci_cmd_get_link_status_data { + u8 topo_media_conflict; +#define IXGBE_ACI_LINK_TOPO_CONFLICT BIT(0) +#define IXGBE_ACI_LINK_MEDIA_CONFLICT BIT(1) +#define IXGBE_ACI_LINK_TOPO_CORRUPT BIT(2) +#define IXGBE_ACI_LINK_TOPO_UNREACH_PRT BIT(4) +#define IXGBE_ACI_LINK_TOPO_UNDRUTIL_PRT BIT(5) +#define IXGBE_ACI_LINK_TOPO_UNDRUTIL_MEDIA BIT(6) +#define IXGBE_ACI_LINK_TOPO_UNSUPP_MEDIA BIT(7) + u8 link_cfg_err; +#define IXGBE_ACI_LINK_CFG_ERR BIT(0) +#define IXGBE_ACI_LINK_CFG_COMPLETED BIT(1) +#define IXGBE_ACI_LINK_ACT_PORT_OPT_INVAL BIT(2) +#define IXGBE_ACI_LINK_FEAT_ID_OR_CONFIG_ID_INVAL BIT(3) +#define IXGBE_ACI_LINK_TOPO_CRITICAL_SDP_ERR BIT(4) +#define IXGBE_ACI_LINK_MODULE_POWER_UNSUPPORTED BIT(5) +#define IXGBE_ACI_LINK_EXTERNAL_PHY_LOAD_FAILURE BIT(6) +#define IXGBE_ACI_LINK_INVAL_MAX_POWER_LIMIT BIT(7) + u8 link_info; +#define IXGBE_ACI_LINK_UP BIT(0) /* Link Status */ +#define IXGBE_ACI_LINK_FAULT BIT(1) +#define IXGBE_ACI_LINK_FAULT_TX BIT(2) +#define IXGBE_ACI_LINK_FAULT_RX BIT(3) +#define IXGBE_ACI_LINK_FAULT_REMOTE BIT(4) +#define IXGBE_ACI_LINK_UP_PORT BIT(5) /* External Port Link Status */ +#define IXGBE_ACI_MEDIA_AVAILABLE BIT(6) +#define IXGBE_ACI_SIGNAL_DETECT BIT(7) + u8 an_info; +#define IXGBE_ACI_AN_COMPLETED BIT(0) +#define IXGBE_ACI_LP_AN_ABILITY BIT(1) +#define IXGBE_ACI_PD_FAULT BIT(2) /* Parallel Detection Fault */ +#define IXGBE_ACI_FEC_EN BIT(3) +#define IXGBE_ACI_PHY_LOW_POWER BIT(4) /* Low Power State */ +#define IXGBE_ACI_LINK_PAUSE_TX BIT(5) +#define IXGBE_ACI_LINK_PAUSE_RX BIT(6) +#define IXGBE_ACI_QUALIFIED_MODULE BIT(7) + u8 ext_info; +#define IXGBE_ACI_LINK_PHY_TEMP_ALARM BIT(0) +#define IXGBE_ACI_LINK_EXCESSIVE_ERRORS BIT(1) /* Excessive Link Errors */ + /* Port Tx Suspended */ +#define IXGBE_ACI_LINK_TX_S 2 +#define IXGBE_ACI_LINK_TX_M (0x03 << IXGBE_ACI_LINK_TX_S) +#define IXGBE_ACI_LINK_TX_ACTIVE 0 +#define IXGBE_ACI_LINK_TX_DRAINED 1 +#define IXGBE_ACI_LINK_TX_FLUSHED 3 + u8 lb_status; +#define IXGBE_ACI_LINK_LB_PHY_LCL BIT(0) +#define IXGBE_ACI_LINK_LB_PHY_RMT BIT(1) +#define IXGBE_ACI_LINK_LB_MAC_LCL BIT(2) +#define IXGBE_ACI_LINK_LB_PHY_IDX_S 3 +#define IXGBE_ACI_LINK_LB_PHY_IDX_M (0x7 << IXGBE_ACI_LB_PHY_IDX_S) + __le16 max_frame_size; + u8 cfg; +#define IXGBE_ACI_LINK_25G_KR_FEC_EN BIT(0) +#define IXGBE_ACI_LINK_25G_RS_528_FEC_EN BIT(1) +#define IXGBE_ACI_LINK_25G_RS_544_FEC_EN BIT(2) +#define IXGBE_ACI_FEC_MASK MAKEMASK(0x7, 0) + /* Pacing Config */ +#define IXGBE_ACI_CFG_PACING_S 3 +#define IXGBE_ACI_CFG_PACING_M (0xF << IXGBE_ACI_CFG_PACING_S) +#define IXGBE_ACI_CFG_PACING_TYPE_M BIT(7) +#define IXGBE_ACI_CFG_PACING_TYPE_AVG 0 +#define IXGBE_ACI_CFG_PACING_TYPE_FIXED IXGBE_ACI_CFG_PACING_TYPE_M + /* External Device Power Ability */ + u8 power_desc; +#define IXGBE_ACI_PWR_CLASS_M 0x3F +#define IXGBE_ACI_LINK_PWR_BASET_LOW_HIGH 0 +#define IXGBE_ACI_LINK_PWR_BASET_HIGH 1 +#define IXGBE_ACI_LINK_PWR_QSFP_CLASS_1 0 +#define IXGBE_ACI_LINK_PWR_QSFP_CLASS_2 1 +#define IXGBE_ACI_LINK_PWR_QSFP_CLASS_3 2 +#define IXGBE_ACI_LINK_PWR_QSFP_CLASS_4 3 + __le16 link_speed; +#define IXGBE_ACI_LINK_SPEED_M 0x7FF +#define IXGBE_ACI_LINK_SPEED_10MB BIT(0) +#define IXGBE_ACI_LINK_SPEED_100MB BIT(1) +#define IXGBE_ACI_LINK_SPEED_1000MB BIT(2) +#define IXGBE_ACI_LINK_SPEED_2500MB BIT(3) +#define IXGBE_ACI_LINK_SPEED_5GB BIT(4) +#define IXGBE_ACI_LINK_SPEED_10GB BIT(5) +#define IXGBE_ACI_LINK_SPEED_20GB BIT(6) +#define IXGBE_ACI_LINK_SPEED_25GB BIT(7) +#define IXGBE_ACI_LINK_SPEED_40GB BIT(8) +#define IXGBE_ACI_LINK_SPEED_50GB BIT(9) +#define IXGBE_ACI_LINK_SPEED_100GB BIT(10) +#define IXGBE_ACI_LINK_SPEED_200GB BIT(11) +#define IXGBE_ACI_LINK_SPEED_UNKNOWN BIT(15) + __le16 reserved3; /* Aligns next field to 8-byte boundary */ + u8 ext_fec_status; +#define IXGBE_ACI_LINK_RS_272_FEC_EN BIT(0) /* RS 272 FEC enabled */ + u8 reserved4; + __le64 phy_type_low; /* Use values from IXGBE_PHY_TYPE_LOW_* */ + __le64 phy_type_high; /* Use values from IXGBE_PHY_TYPE_HIGH_* */ + /* Get link status version 2 link partner data */ + __le64 lp_phy_type_low; /* Use values from IXGBE_PHY_TYPE_LOW_* */ + __le64 lp_phy_type_high; /* Use values from IXGBE_PHY_TYPE_HIGH_* */ + u8 lp_fec_adv; +#define IXGBE_ACI_LINK_LP_10G_KR_FEC_CAP BIT(0) +#define IXGBE_ACI_LINK_LP_25G_KR_FEC_CAP BIT(1) +#define IXGBE_ACI_LINK_LP_RS_528_FEC_CAP BIT(2) +#define IXGBE_ACI_LINK_LP_50G_KR_272_FEC_CAP BIT(3) +#define IXGBE_ACI_LINK_LP_100G_KR_272_FEC_CAP BIT(4) +#define IXGBE_ACI_LINK_LP_200G_KR_272_FEC_CAP BIT(5) + u8 lp_fec_req; +#define IXGBE_ACI_LINK_LP_10G_KR_FEC_REQ BIT(0) +#define IXGBE_ACI_LINK_LP_25G_KR_FEC_REQ BIT(1) +#define IXGBE_ACI_LINK_LP_RS_528_FEC_REQ BIT(2) +#define IXGBE_ACI_LINK_LP_KR_272_FEC_REQ BIT(3) + u8 lp_flowcontrol; +#define IXGBE_ACI_LINK_LP_PAUSE_ADV BIT(0) +#define IXGBE_ACI_LINK_LP_ASM_DIR_ADV BIT(1) + u8 reserved5[5]; +}; +#pragma pack() + +IXGBE_CHECK_STRUCT_LEN(56, ixgbe_aci_cmd_get_link_status_data); + +/* Set event mask command (direct 0x0613) */ +struct ixgbe_aci_cmd_set_event_mask { + u8 reserved[8]; + __le16 event_mask; +#define IXGBE_ACI_LINK_EVENT_UPDOWN BIT(1) +#define IXGBE_ACI_LINK_EVENT_MEDIA_NA BIT(2) +#define IXGBE_ACI_LINK_EVENT_LINK_FAULT BIT(3) +#define IXGBE_ACI_LINK_EVENT_PHY_TEMP_ALARM BIT(4) +#define IXGBE_ACI_LINK_EVENT_EXCESSIVE_ERRORS BIT(5) +#define IXGBE_ACI_LINK_EVENT_SIGNAL_DETECT BIT(6) +#define IXGBE_ACI_LINK_EVENT_AN_COMPLETED BIT(7) +#define IXGBE_ACI_LINK_EVENT_MODULE_QUAL_FAIL BIT(8) +#define IXGBE_ACI_LINK_EVENT_PORT_TX_SUSPENDED BIT(9) +#define IXGBE_ACI_LINK_EVENT_TOPO_CONFLICT BIT(10) +#define IXGBE_ACI_LINK_EVENT_MEDIA_CONFLICT BIT(11) +#define IXGBE_ACI_LINK_EVENT_PHY_FW_LOAD_FAIL BIT(12) + u8 reserved1[6]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_set_event_mask); + +struct ixgbe_aci_cmd_link_topo_params { + u8 lport_num; + u8 lport_num_valid; +#define IXGBE_ACI_LINK_TOPO_PORT_NUM_VALID BIT(0) + u8 node_type_ctx; +#define IXGBE_ACI_LINK_TOPO_NODE_TYPE_S 0 +#define IXGBE_ACI_LINK_TOPO_NODE_TYPE_M (0xF << IXGBE_ACI_LINK_TOPO_NODE_TYPE_S) +#define IXGBE_ACI_LINK_TOPO_NODE_TYPE_PHY 0 +#define IXGBE_ACI_LINK_TOPO_NODE_TYPE_GPIO_CTRL 1 +#define IXGBE_ACI_LINK_TOPO_NODE_TYPE_MUX_CTRL 2 +#define IXGBE_ACI_LINK_TOPO_NODE_TYPE_LED_CTRL 3 +#define IXGBE_ACI_LINK_TOPO_NODE_TYPE_LED 4 +#define IXGBE_ACI_LINK_TOPO_NODE_TYPE_THERMAL 5 +#define IXGBE_ACI_LINK_TOPO_NODE_TYPE_CAGE 6 +#define IXGBE_ACI_LINK_TOPO_NODE_TYPE_MEZZ 7 +#define IXGBE_ACI_LINK_TOPO_NODE_TYPE_ID_EEPROM 8 +#define IXGBE_ACI_LINK_TOPO_NODE_TYPE_GPS 11 +#define IXGBE_ACI_LINK_TOPO_NODE_CTX_S 4 +#define IXGBE_ACI_LINK_TOPO_NODE_CTX_M \ + (0xF << IXGBE_ACI_LINK_TOPO_NODE_CTX_S) +#define IXGBE_ACI_LINK_TOPO_NODE_CTX_GLOBAL 0 +#define IXGBE_ACI_LINK_TOPO_NODE_CTX_BOARD 1 +#define IXGBE_ACI_LINK_TOPO_NODE_CTX_PORT 2 +#define IXGBE_ACI_LINK_TOPO_NODE_CTX_NODE 3 +#define IXGBE_ACI_LINK_TOPO_NODE_CTX_NODE_HANDLE 4 +#define IXGBE_ACI_LINK_TOPO_NODE_CTX_DIRECT_BUS_ACCESS 5 +#define IXGBE_ACI_LINK_TOPO_NODE_CTX_NODE_HANDLE_BUS_ADDRESS 6 + u8 index; +}; + +IXGBE_CHECK_STRUCT_LEN(4, ixgbe_aci_cmd_link_topo_params); + +struct ixgbe_aci_cmd_link_topo_addr { + struct ixgbe_aci_cmd_link_topo_params topo_params; + __le16 handle; +#define IXGBE_ACI_LINK_TOPO_HANDLE_S 0 +#define IXGBE_ACI_LINK_TOPO_HANDLE_M (0x3FF << IXGBE_ACI_LINK_TOPO_HANDLE_S) +/* Used to decode the handle field */ +#define IXGBE_ACI_LINK_TOPO_HANDLE_BRD_TYPE_M BIT(9) +#define IXGBE_ACI_LINK_TOPO_HANDLE_BRD_TYPE_LOM BIT(9) +#define IXGBE_ACI_LINK_TOPO_HANDLE_BRD_TYPE_MEZZ 0 +#define IXGBE_ACI_LINK_TOPO_HANDLE_NODE_S 0 +/* In case of a Mezzanine type */ +#define IXGBE_ACI_LINK_TOPO_HANDLE_MEZZ_NODE_M \ + (0x3F << IXGBE_ACI_LINK_TOPO_HANDLE_NODE_S) +#define IXGBE_ACI_LINK_TOPO_HANDLE_MEZZ_S 6 +#define IXGBE_ACI_LINK_TOPO_HANDLE_MEZZ_M \ + (0x7 << IXGBE_ACI_LINK_TOPO_HANDLE_MEZZ_S) +/* In case of a LOM type */ +#define IXGBE_ACI_LINK_TOPO_HANDLE_LOM_NODE_M \ + (0x1FF << IXGBE_ACI_LINK_TOPO_HANDLE_NODE_S) +}; + +IXGBE_CHECK_STRUCT_LEN(6, ixgbe_aci_cmd_link_topo_addr); + +/* Get Link Topology Handle (direct, 0x06E0) */ +struct ixgbe_aci_cmd_get_link_topo { + struct ixgbe_aci_cmd_link_topo_addr addr; + u8 node_part_num; +#define IXGBE_ACI_GET_LINK_TOPO_NODE_NR_PCA9575 0x21 +#define IXGBE_ACI_GET_LINK_TOPO_NODE_NR_GEN_GPS 0x48 +#define IXGBE_ACI_GET_LINK_TOPO_NODE_NR_E610_PTC 0x49 + u8 rsvd[9]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_get_link_topo); + +/* Get Link Topology Pin (direct, 0x06E1) */ +struct ixgbe_aci_cmd_get_link_topo_pin { + struct ixgbe_aci_cmd_link_topo_addr addr; + u8 input_io_params; +#define IXGBE_ACI_LINK_TOPO_INPUT_IO_FUNC_S 0 +#define IXGBE_ACI_LINK_TOPO_INPUT_IO_FUNC_M \ + (0x1F << IXGBE_ACI_LINK_TOPO_INPUT_IO_FUNC_S) +#define IXGBE_ACI_LINK_TOPO_IO_FUNC_GPIO 0 +#define IXGBE_ACI_LINK_TOPO_IO_FUNC_RESET_N 1 +#define IXGBE_ACI_LINK_TOPO_IO_FUNC_INT_N 2 +#define IXGBE_ACI_LINK_TOPO_IO_FUNC_PRESENT_N 3 +#define IXGBE_ACI_LINK_TOPO_IO_FUNC_TX_DIS 4 +#define IXGBE_ACI_LINK_TOPO_IO_FUNC_MODSEL_N 5 +#define IXGBE_ACI_LINK_TOPO_IO_FUNC_LPMODE 6 +#define IXGBE_ACI_LINK_TOPO_IO_FUNC_TX_FAULT 7 +#define IXGBE_ACI_LINK_TOPO_IO_FUNC_RX_LOSS 8 +#define IXGBE_ACI_LINK_TOPO_IO_FUNC_RS0 9 +#define IXGBE_ACI_LINK_TOPO_IO_FUNC_RS1 10 +#define IXGBE_ACI_LINK_TOPO_IO_FUNC_EEPROM_WP 11 +/* 12 repeats intentionally due to two different uses depending on context */ +#define IXGBE_ACI_LINK_TOPO_IO_FUNC_LED 12 +#define IXGBE_ACI_LINK_TOPO_IO_FUNC_RED_LED 12 +#define IXGBE_ACI_LINK_TOPO_IO_FUNC_GREEN_LED 13 +#define IXGBE_ACI_LINK_TOPO_IO_FUNC_BLUE_LED 14 +#define IXGBE_ACI_LINK_TOPO_INPUT_IO_TYPE_S 5 +#define IXGBE_ACI_LINK_TOPO_INPUT_IO_TYPE_M \ + (0x7 << IXGBE_ACI_LINK_TOPO_INPUT_IO_TYPE_S) +#define IXGBE_ACI_LINK_TOPO_INPUT_IO_TYPE_GPIO 3 +/* Use IXGBE_ACI_LINK_TOPO_NODE_TYPE_* for the type values */ + u8 output_io_params; +#define IXGBE_ACI_LINK_TOPO_OUTPUT_IO_FUNC_S 0 +#define IXGBE_ACI_LINK_TOPO_OUTPUT_IO_FUNC_M \ + (0x1F << \ IXGBE_ACI_LINK_TOPO_INPUT_IO_FUNC_NUM_S) +/* Use IXGBE_ACI_LINK_TOPO_IO_FUNC_* for the non-numerical options */ +#define IXGBE_ACI_LINK_TOPO_OUTPUT_IO_TYPE_S 5 +#define IXGBE_ACI_LINK_TOPO_OUTPUT_IO_TYPE_M \ + (0x7 << IXGBE_ACI_LINK_TOPO_INPUT_IO_TYPE_S) +/* Use IXGBE_ACI_LINK_TOPO_NODE_TYPE_* for the type values */ + u8 output_io_flags; +#define IXGBE_ACI_LINK_TOPO_OUTPUT_SPEED_S 0 +#define IXGBE_ACI_LINK_TOPO_OUTPUT_SPEED_M \ + (0x7 << IXGBE_ACI_LINK_TOPO_OUTPUT_SPEED_S) +#define IXGBE_ACI_LINK_TOPO_OUTPUT_INT_S 3 +#define IXGBE_ACI_LINK_TOPO_OUTPUT_INT_M \ + (0x3 << IXGBE_ACI_LINK_TOPO_OUTPUT_INT_S) +#define IXGBE_ACI_LINK_TOPO_OUTPUT_POLARITY BIT(5) +#define IXGBE_ACI_LINK_TOPO_OUTPUT_VALUE BIT(6) +#define IXGBE_ACI_LINK_TOPO_OUTPUT_DRIVEN BIT(7) + u8 rsvd[7]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_get_link_topo_pin); +/* Read/Write I2C (direct, 0x06E2/0x06E3) */ +struct ixgbe_aci_cmd_i2c { + struct ixgbe_aci_cmd_link_topo_addr topo_addr; + __le16 i2c_addr; + u8 i2c_params; +#define IXGBE_ACI_I2C_DATA_SIZE_S 0 +#define IXGBE_ACI_I2C_DATA_SIZE_M (0xF << IXGBE_ACI_I2C_DATA_SIZE_S) +#define IXGBE_ACI_I2C_ADDR_TYPE_M BIT(4) +#define IXGBE_ACI_I2C_ADDR_TYPE_7BIT 0 +#define IXGBE_ACI_I2C_ADDR_TYPE_10BIT IXGBE_ACI_I2C_ADDR_TYPE_M +#define IXGBE_ACI_I2C_DATA_OFFSET_S 5 +#define IXGBE_ACI_I2C_DATA_OFFSET_M (0x3 << IXGBE_ACI_I2C_DATA_OFFSET_S) +#define IXGBE_ACI_I2C_USE_REPEATED_START BIT(7) + u8 rsvd; + __le16 i2c_bus_addr; +#define IXGBE_ACI_I2C_ADDR_7BIT_MASK 0x7F +#define IXGBE_ACI_I2C_ADDR_10BIT_MASK 0x3FF + u8 i2c_data[4]; /* Used only by write command, reserved in read. */ +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_i2c); + +/* Read I2C Response (direct, 0x06E2) */ +struct ixgbe_aci_cmd_read_i2c_resp { + u8 i2c_data[16]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_read_i2c_resp); + +/* Read/Write MDIO (direct, 0x06E4/0x06E5) */ +struct ixgbe_aci_cmd_mdio { + struct ixgbe_aci_cmd_link_topo_addr topo_addr; + u8 mdio_device_addr; +#define IXGBE_ACI_MDIO_DEV_S 0 +#define IXGBE_ACI_MDIO_DEV_M (0x1F << IXGBE_ACI_MDIO_DEV_S) +#define IXGBE_ACI_MDIO_CLAUSE_22 BIT(5) +#define IXGBE_ACI_MDIO_CLAUSE_45 BIT(6) + u8 mdio_bus_address; +#define IXGBE_ACI_MDIO_BUS_ADDR_S 0 +#define IXGBE_ACI_MDIO_BUS_ADDR_M (0x1F << IXGBE_ACI_MDIO_BUS_ADDR_S) + __le16 offset; + __le16 data; /* Input in write cmd, output in read cmd. */ + u8 rsvd1[4]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_mdio); + +/* Set/Get GPIO By Function (direct, 0x06E6/0x06E7) */ +struct ixgbe_aci_cmd_gpio_by_func { + struct ixgbe_aci_cmd_link_topo_addr topo_addr; + u8 io_func_num; +#define IXGBE_ACI_GPIO_FUNC_S 0 +#define IXGBE_ACI_GPIO_FUNC_M (0x1F << IXGBE_ACI_GPIO_IO_FUNC_NUM_S) + u8 io_value; /* Input in write cmd, output in read cmd. */ +#define IXGBE_ACI_GPIO_ON BIT(0) +#define IXGBE_ACI_GPIO_OFF 0 + u8 rsvd[8]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_gpio_by_func); + +/* Set/Get GPIO (direct, 0x06EC/0x06ED) */ +struct ixgbe_aci_cmd_gpio { + __le16 gpio_ctrl_handle; +#define IXGBE_ACI_GPIO_HANDLE_S 0 +#define IXGBE_ACI_GPIO_HANDLE_M (0x3FF << IXGBE_ACI_GPIO_HANDLE_S) + u8 gpio_num; + u8 gpio_val; + u8 rsvd[12]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_gpio); + +/* Read/Write SFF EEPROM command (indirect 0x06EE) */ +struct ixgbe_aci_cmd_sff_eeprom { + u8 lport_num; + u8 lport_num_valid; +#define IXGBE_ACI_SFF_PORT_NUM_VALID BIT(0) + __le16 i2c_bus_addr; +#define IXGBE_ACI_SFF_I2CBUS_7BIT_M 0x7F +#define IXGBE_ACI_SFF_I2CBUS_10BIT_M 0x3FF +#define IXGBE_ACI_SFF_I2CBUS_TYPE_M BIT(10) +#define IXGBE_ACI_SFF_I2CBUS_TYPE_7BIT 0 +#define IXGBE_ACI_SFF_I2CBUS_TYPE_10BIT IXGBE_ACI_SFF_I2CBUS_TYPE_M +#define IXGBE_ACI_SFF_PAGE_BANK_CTRL_S 11 +#define IXGBE_ACI_SFF_PAGE_BANK_CTRL_M (0x3 << IXGBE_ACI_SFF_PAGE_BANK_CTRL_S) +#define IXGBE_ACI_SFF_NO_PAGE_BANK_UPDATE 0 +#define IXGBE_ACI_SFF_UPDATE_PAGE 1 +#define IXGBE_ACI_SFF_UPDATE_BANK 2 +#define IXGBE_ACI_SFF_UPDATE_PAGE_BANK 3 +#define IXGBE_ACI_SFF_IS_WRITE BIT(15) + __le16 i2c_offset; + u8 module_bank; + u8 module_page; + __le32 addr_high; + __le32 addr_low; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_sff_eeprom); + +/* Program Topology Device NVM (direct, 0x06F2) */ +struct ixgbe_aci_cmd_prog_topo_dev_nvm { + struct ixgbe_aci_cmd_link_topo_params topo_params; + u8 rsvd[12]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_prog_topo_dev_nvm); + +/* Read Topology Device NVM (direct, 0x06F3) */ +struct ixgbe_aci_cmd_read_topo_dev_nvm { + struct ixgbe_aci_cmd_link_topo_params topo_params; + __le32 start_address; +#define IXGBE_ACI_READ_TOPO_DEV_NVM_DATA_READ_SIZE 8 + u8 data_read[IXGBE_ACI_READ_TOPO_DEV_NVM_DATA_READ_SIZE]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_read_topo_dev_nvm); + +/* NVM Read command (indirect 0x0701) + * NVM Erase commands (direct 0x0702) + * NVM Write commands (indirect 0x0703) + * NVM Write Activate commands (direct 0x0707) + * NVM Shadow RAM Dump commands (direct 0x0707) + */ +struct ixgbe_aci_cmd_nvm { +#define IXGBE_ACI_NVM_MAX_OFFSET 0xFFFFFF + __le16 offset_low; + u8 offset_high; /* For Write Activate offset_high is used as flags2 */ + u8 cmd_flags; +#define IXGBE_ACI_NVM_LAST_CMD BIT(0) +#define IXGBE_ACI_NVM_PCIR_REQ BIT(0) /* Used by NVM Write reply */ +#define IXGBE_ACI_NVM_PRESERVATION_S 1 /* Used by NVM Write Activate only */ +#define IXGBE_ACI_NVM_PRESERVATION_M (3 << IXGBE_ACI_NVM_PRESERVATION_S) +#define IXGBE_ACI_NVM_NO_PRESERVATION (0 << IXGBE_ACI_NVM_PRESERVATION_S) +#define IXGBE_ACI_NVM_PRESERVE_ALL BIT(1) +#define IXGBE_ACI_NVM_FACTORY_DEFAULT (2 << IXGBE_ACI_NVM_PRESERVATION_S) +#define IXGBE_ACI_NVM_PRESERVE_SELECTED (3 << IXGBE_ACI_NVM_PRESERVATION_S) +#define IXGBE_ACI_NVM_ACTIV_SEL_NVM BIT(3) /* Write Activate/SR Dump only */ +#define IXGBE_ACI_NVM_ACTIV_SEL_OROM BIT(4) +#define IXGBE_ACI_NVM_ACTIV_SEL_NETLIST BIT(5) +#define IXGBE_ACI_NVM_SPECIAL_UPDATE BIT(6) +#define IXGBE_ACI_NVM_REVERT_LAST_ACTIV BIT(6) /* Write Activate only */ +#define IXGBE_ACI_NVM_ACTIV_SEL_MASK MAKEMASK(0x7, 3) +#define IXGBE_ACI_NVM_FLASH_ONLY BIT(7) +#define IXGBE_ACI_NVM_RESET_LVL_M MAKEMASK(0x3, 0) /* Write reply only */ +#define IXGBE_ACI_NVM_POR_FLAG 0 +#define IXGBE_ACI_NVM_PERST_FLAG 1 +#define IXGBE_ACI_NVM_EMPR_FLAG 2 +#define IXGBE_ACI_NVM_EMPR_ENA BIT(0) /* Write Activate reply only */ + /* For Write Activate, several flags are sent as part of a separate + * flags2 field using a separate byte. For simplicity of the software + * interface, we pass the flags as a 16 bit value so these flags are + * all offset by 8 bits + */ +#define IXGBE_ACI_NVM_ACTIV_REQ_EMPR BIT(8) /* NVM Write Activate only */ + __le16 module_typeid; + __le16 length; +#define IXGBE_ACI_NVM_ERASE_LEN 0xFFFF + __le32 addr_high; + __le32 addr_low; +}; + +/* NVM Module_Type ID, needed offset and read_len for struct ixgbe_aci_cmd_nvm. */ +#define IXGBE_ACI_NVM_SECTOR_UNIT 4096 /* In Bytes */ +#define IXGBE_ACI_NVM_WORD_UNIT 2 /* In Bytes */ + +#define IXGBE_ACI_NVM_START_POINT 0 +#define IXGBE_ACI_NVM_EMP_SR_PTR_OFFSET 0x90 +#define IXGBE_ACI_NVM_EMP_SR_PTR_RD_LEN 2 /* In Bytes */ +#define IXGBE_ACI_NVM_EMP_SR_PTR_M MAKEMASK(0x7FFF, 0) +#define IXGBE_ACI_NVM_EMP_SR_PTR_TYPE_S 15 +#define IXGBE_ACI_NVM_EMP_SR_PTR_TYPE_M BIT(15) +#define IXGBE_ACI_NVM_EMP_SR_PTR_TYPE_SECTOR 1 + +#define IXGBE_ACI_NVM_LLDP_CFG_PTR_OFFSET 0x46 +#define IXGBE_ACI_NVM_LLDP_CFG_HEADER_LEN 2 /* In Bytes */ +#define IXGBE_ACI_NVM_LLDP_CFG_PTR_RD_LEN 2 /* In Bytes */ + +#define IXGBE_ACI_NVM_LLDP_PRESERVED_MOD_ID 0x129 +#define IXGBE_ACI_NVM_CUR_LLDP_PERSIST_RD_OFFSET 2 /* In Bytes */ +#define IXGBE_ACI_NVM_LLDP_STATUS_M MAKEMASK(0xF, 0) +#define IXGBE_ACI_NVM_LLDP_STATUS_M_LEN 4 /* In Bits */ +#define IXGBE_ACI_NVM_LLDP_STATUS_RD_LEN 4 /* In Bytes */ + +#define IXGBE_ACI_NVM_MINSREV_MOD_ID 0x130 + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_nvm); + +/* Used for reading and writing MinSRev using 0x0701 and 0x0703. Note that the + * type field is excluded from the section when reading and writing from + * a module using the module_typeid field with these AQ commands. + */ +struct ixgbe_aci_cmd_nvm_minsrev { + __le16 length; + __le16 validity; +#define IXGBE_ACI_NVM_MINSREV_NVM_VALID BIT(0) +#define IXGBE_ACI_NVM_MINSREV_OROM_VALID BIT(1) + __le16 nvm_minsrev_l; + __le16 nvm_minsrev_h; + __le16 orom_minsrev_l; + __le16 orom_minsrev_h; +}; + +IXGBE_CHECK_STRUCT_LEN(12, ixgbe_aci_cmd_nvm_minsrev); + +/* Used for 0x0704 as well as for 0x0705 commands */ +struct ixgbe_aci_cmd_nvm_cfg { + u8 cmd_flags; +#define IXGBE_ACI_ANVM_MULTIPLE_ELEMS BIT(0) +#define IXGBE_ACI_ANVM_IMMEDIATE_FIELD BIT(1) +#define IXGBE_ACI_ANVM_NEW_CFG BIT(2) + u8 reserved; + __le16 count; + __le16 id; + u8 reserved1[2]; + __le32 addr_high; + __le32 addr_low; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_nvm_cfg); + +struct ixgbe_aci_cmd_nvm_cfg_data { + __le16 field_id; + __le16 field_options; + __le16 field_value; +}; + +IXGBE_CHECK_STRUCT_LEN(6, ixgbe_aci_cmd_nvm_cfg_data); + +/* NVM Checksum Command (direct, 0x0706) */ +struct ixgbe_aci_cmd_nvm_checksum { + u8 flags; +#define IXGBE_ACI_NVM_CHECKSUM_VERIFY BIT(0) +#define IXGBE_ACI_NVM_CHECKSUM_RECALC BIT(1) + u8 rsvd; + __le16 checksum; /* Used only by response */ +#define IXGBE_ACI_NVM_CHECKSUM_CORRECT 0xBABA + u8 rsvd2[12]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_nvm_checksum); + +/* Used for NVM Sanitization command - 0x070C */ +struct ixgbe_aci_cmd_nvm_sanitization { + u8 cmd_flags; +#define IXGBE_ACI_SANITIZE_REQ_READ 0 +#define IXGBE_ACI_SANITIZE_REQ_OPERATE BIT(0) + +#define IXGBE_ACI_SANITIZE_READ_SUBJECT_NVM_BITS 0 +#define IXGBE_ACI_SANITIZE_READ_SUBJECT_NVM_STATE BIT(1) +#define IXGBE_ACI_SANITIZE_OPERATE_SUBJECT_CLEAR 0 + u8 values; +#define IXGBE_ACI_SANITIZE_NVM_BITS_HOST_CLEAN_SUPPORT BIT(0) +#define IXGBE_ACI_SANITIZE_NVM_BITS_BMC_CLEAN_SUPPORT BIT(2) +#define IXGBE_ACI_SANITIZE_NVM_STATE_HOST_CLEAN_DONE BIT(0) +#define IXGBE_ACI_SANITIZE_NVM_STATE_HOST_CLEAN_SUCCESS BIT(1) +#define IXGBE_ACI_SANITIZE_NVM_STATE_BMC_CLEAN_DONE BIT(2) +#define IXGBE_ACI_SANITIZE_NVM_STATE_BMC_CLEAN_SUCCESS BIT(3) +#define IXGBE_ACI_SANITIZE_OPERATE_HOST_CLEAN_DONE BIT(0) +#define IXGBE_ACI_SANITIZE_OPERATE_HOST_CLEAN_SUCCESS BIT(1) +#define IXGBE_ACI_SANITIZE_OPERATE_BMC_CLEAN_DONE BIT(2) +#define IXGBE_ACI_SANITIZE_OPERATE_BMC_CLEAN_SUCCESS BIT(3) + u8 reserved[14]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_nvm_sanitization); + +/* Write/Read Alternate - Direct (direct 0x0900/0x0902) */ +struct ixgbe_aci_cmd_read_write_alt_direct { + __le32 dword0_addr; + __le32 dword0_value; + __le32 dword1_addr; + __le32 dword1_value; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_read_write_alt_direct); + +/* Write/Read Alternate - Indirect (indirect 0x0901/0x0903) */ +struct ixgbe_aci_cmd_read_write_alt_indirect { + __le32 base_dword_addr; + __le32 num_dwords; + __le32 addr_high; + __le32 addr_low; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_read_write_alt_indirect); + +/* Done Alternate Write (direct 0x0904) */ +struct ixgbe_aci_cmd_done_alt_write { + u8 flags; +#define IXGBE_ACI_CMD_UEFI_BIOS_MODE BIT(0) +#define IXGBE_ACI_RESP_RESET_NEEDED BIT(1) + u8 reserved[15]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_done_alt_write); + +/* Clear Port Alternate Write (direct 0x0906) */ +struct ixgbe_aci_cmd_clear_port_alt_write { + u8 reserved[16]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_clear_port_alt_write); + +/* Get CGU abilities command response data structure (indirect 0x0C61) */ +struct ixgbe_aci_cmd_get_cgu_abilities { + u8 num_inputs; + u8 num_outputs; + u8 pps_dpll_idx; + u8 synce_dpll_idx; + __le32 max_in_freq; + __le32 max_in_phase_adj; + __le32 max_out_freq; + __le32 max_out_phase_adj; + u8 cgu_part_num; + u8 rsvd[3]; +}; + +IXGBE_CHECK_STRUCT_LEN(24, ixgbe_aci_cmd_get_cgu_abilities); + +#define IXGBE_ACI_NODE_HANDLE_VALID BIT(10) +#define IXGBE_ACI_NODE_HANDLE MAKEMASK(0x3FF, 0) +#define IXGBE_ACI_DRIVING_CLK_NUM_SHIFT 10 +#define IXGBE_ACI_DRIVING_CLK_NUM MAKEMASK(0x3F, IXGBE_ACI_DRIVING_CLK_NUM_SHIFT) + +/* Set CGU input config (direct 0x0C62) */ +struct ixgbe_aci_cmd_set_cgu_input_config { + u8 input_idx; + u8 flags1; +#define IXGBE_ACI_SET_CGU_IN_CFG_FLG1_UPDATE_FREQ BIT(6) +#define IXGBE_ACI_SET_CGU_IN_CFG_FLG1_UPDATE_DELAY BIT(7) + u8 flags2; +#define IXGBE_ACI_SET_CGU_IN_CFG_FLG2_INPUT_EN BIT(5) +#define IXGBE_ACI_SET_CGU_IN_CFG_FLG2_ESYNC_EN BIT(6) + u8 rsvd; + __le32 freq; + __le32 phase_delay; + u8 rsvd2[2]; + __le16 node_handle; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_set_cgu_input_config); + +/* Get CGU input config response descriptor structure (direct 0x0C63) */ +struct ixgbe_aci_cmd_get_cgu_input_config { + u8 input_idx; + u8 status; +#define IXGBE_ACI_GET_CGU_IN_CFG_STATUS_LOS BIT(0) +#define IXGBE_ACI_GET_CGU_IN_CFG_STATUS_SCM_FAIL BIT(1) +#define IXGBE_ACI_GET_CGU_IN_CFG_STATUS_CFM_FAIL BIT(2) +#define IXGBE_ACI_GET_CGU_IN_CFG_STATUS_GST_FAIL BIT(3) +#define IXGBE_ACI_GET_CGU_IN_CFG_STATUS_PFM_FAIL BIT(4) +#define IXGBE_ACI_GET_CGU_IN_CFG_STATUS_ESYNC_FAIL BIT(6) +#define IXGBE_ACI_GET_CGU_IN_CFG_STATUS_ESYNC_CAP BIT(7) + u8 type; +#define IXGBE_ACI_GET_CGU_IN_CFG_TYPE_READ_ONLY BIT(0) +#define IXGBE_ACI_GET_CGU_IN_CFG_TYPE_GPS BIT(4) +#define IXGBE_ACI_GET_CGU_IN_CFG_TYPE_EXTERNAL BIT(5) +#define IXGBE_ACI_GET_CGU_IN_CFG_TYPE_PHY BIT(6) + u8 flags1; +#define IXGBE_ACI_GET_CGU_IN_CFG_FLG1_PHASE_DELAY_SUPP BIT(0) +#define IXGBE_ACI_GET_CGU_IN_CFG_FLG1_1PPS_SUPP BIT(2) +#define IXGBE_ACI_GET_CGU_IN_CFG_FLG1_10MHZ_SUPP BIT(3) +#define IXGBE_ACI_GET_CGU_IN_CFG_FLG1_ANYFREQ BIT(7) + __le32 freq; + __le32 phase_delay; + u8 flags2; +#define IXGBE_ACI_GET_CGU_IN_CFG_FLG2_INPUT_EN BIT(5) +#define IXGBE_ACI_GET_CGU_IN_CFG_FLG2_ESYNC_EN BIT(6) + u8 rsvd[1]; + __le16 node_handle; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_get_cgu_input_config); + +/* Set CGU output config (direct 0x0C64) */ +struct ixgbe_aci_cmd_set_cgu_output_config { + u8 output_idx; + u8 flags; +#define IXGBE_ACI_SET_CGU_OUT_CFG_OUT_EN BIT(0) +#define IXGBE_ACI_SET_CGU_OUT_CFG_ESYNC_EN BIT(1) +#define IXGBE_ACI_SET_CGU_OUT_CFG_UPDATE_FREQ BIT(2) +#define IXGBE_ACI_SET_CGU_OUT_CFG_UPDATE_PHASE BIT(3) +#define IXGBE_ACI_SET_CGU_OUT_CFG_UPDATE_SRC_SEL BIT(4) + u8 src_sel; +#define IXGBE_ACI_SET_CGU_OUT_CFG_DPLL_SRC_SEL MAKEMASK(0x1F, 0) + u8 rsvd; + __le32 freq; + __le32 phase_delay; + u8 rsvd2[2]; + __le16 node_handle; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_set_cgu_output_config); + +/* Get CGU output config (direct 0x0C65) */ +struct ixgbe_aci_cmd_get_cgu_output_config { + u8 output_idx; + u8 flags; +#define IXGBE_ACI_GET_CGU_OUT_CFG_OUT_EN BIT(0) +#define IXGBE_ACI_GET_CGU_OUT_CFG_ESYNC_EN BIT(1) +#define IXGBE_ACI_GET_CGU_OUT_CFG_ESYNC_ABILITY BIT(2) + u8 src_sel; +#define IXGBE_ACI_GET_CGU_OUT_CFG_DPLL_SRC_SEL_SHIFT 0 +#define IXGBE_ACI_GET_CGU_OUT_CFG_DPLL_SRC_SEL \ + MAKEMASK(0x1F, IXGBE_ACI_GET_CGU_OUT_CFG_DPLL_SRC_SEL_SHIFT) +#define IXGBE_ACI_GET_CGU_OUT_CFG_DPLL_MODE_SHIFT 5 +#define IXGBE_ACI_GET_CGU_OUT_CFG_DPLL_MODE \ + MAKEMASK(0x7, IXGBE_ACI_GET_CGU_OUT_CFG_DPLL_MODE_SHIFT) + u8 rsvd; + __le32 freq; + __le32 src_freq; + u8 rsvd2[2]; + __le16 node_handle; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_get_cgu_output_config); + +/* Get CGU DPLL status (direct 0x0C66) */ +struct ixgbe_aci_cmd_get_cgu_dpll_status { + u8 dpll_num; + u8 ref_state; +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_REF_SW_LOS BIT(0) +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_REF_SW_SCM BIT(1) +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_REF_SW_CFM BIT(2) +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_REF_SW_GST BIT(3) +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_REF_SW_PFM BIT(4) +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_FAST_LOCK_EN BIT(5) +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_REF_SW_ESYNC BIT(6) + __le16 dpll_state; +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_STATE_LOCK BIT(0) +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_STATE_HO BIT(1) +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_STATE_HO_READY BIT(2) +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_STATE_FLHIT BIT(5) +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_STATE_PSLHIT BIT(7) +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_STATE_CLK_REF_SHIFT 8 +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_STATE_CLK_REF_SEL \ + MAKEMASK(0x1F, IXGBE_ACI_GET_CGU_DPLL_STATUS_STATE_CLK_REF_SHIFT) +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_STATE_MODE_SHIFT 13 +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_STATE_MODE \ + MAKEMASK(0x7, IXGBE_ACI_GET_CGU_DPLL_STATUS_STATE_MODE_SHIFT) + __le32 phase_offset_h; + __le32 phase_offset_l; + u8 eec_mode; +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_EEC_MODE_1 0xA +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_EEC_MODE_2 0xB +#define IXGBE_ACI_GET_CGU_DPLL_STATUS_EEC_MODE_UNKNOWN 0xF + u8 rsvd[1]; + __le16 node_handle; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_get_cgu_dpll_status); + +/* Set CGU DPLL config (direct 0x0C67) */ +struct ixgbe_aci_cmd_set_cgu_dpll_config { + u8 dpll_num; + u8 ref_state; +#define IXGBE_ACI_SET_CGU_DPLL_CONFIG_REF_SW_LOS BIT(0) +#define IXGBE_ACI_SET_CGU_DPLL_CONFIG_REF_SW_SCM BIT(1) +#define IXGBE_ACI_SET_CGU_DPLL_CONFIG_REF_SW_CFM BIT(2) +#define IXGBE_ACI_SET_CGU_DPLL_CONFIG_REF_SW_GST BIT(3) +#define IXGBE_ACI_SET_CGU_DPLL_CONFIG_REF_SW_PFM BIT(4) +#define IXGBE_ACI_SET_CGU_DPLL_CONFIG_REF_FLOCK_EN BIT(5) +#define IXGBE_ACI_SET_CGU_DPLL_CONFIG_REF_SW_ESYNC BIT(6) + u8 rsvd; + u8 config; +#define IXGBE_ACI_SET_CGU_DPLL_CONFIG_CLK_REF_SEL MAKEMASK(0x1F, 0) +#define IXGBE_ACI_SET_CGU_DPLL_CONFIG_MODE MAKEMASK(0x7, 5) + u8 rsvd2[8]; + u8 eec_mode; + u8 rsvd3[1]; + __le16 node_handle; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_set_cgu_dpll_config); + +/* Set CGU reference priority (direct 0x0C68) */ +struct ixgbe_aci_cmd_set_cgu_ref_prio { + u8 dpll_num; + u8 ref_idx; + u8 ref_priority; + u8 rsvd[11]; + __le16 node_handle; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_set_cgu_ref_prio); + +/* Get CGU reference priority (direct 0x0C69) */ +struct ixgbe_aci_cmd_get_cgu_ref_prio { + u8 dpll_num; + u8 ref_idx; + u8 ref_priority; /* Valid only in response */ + u8 rsvd[13]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_get_cgu_ref_prio); + +/* Get CGU info (direct 0x0C6A) */ +struct ixgbe_aci_cmd_get_cgu_info { + __le32 cgu_id; + __le32 cgu_cfg_ver; + __le32 cgu_fw_ver; + u8 node_part_num; + u8 dev_rev; + __le16 node_handle; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_get_cgu_info); + +/* Debug Dump Internal Data (indirect 0xFF08) */ +struct ixgbe_aci_cmd_debug_dump_internals { + __le16 cluster_id; /* Expresses next cluster ID in response */ +#define IXGBE_ACI_DBG_DUMP_CLUSTER_ID_LINK 0 +#define IXGBE_ACI_DBG_DUMP_CLUSTER_ID_FULL_CSR_SPACE 1 + __le16 table_id; /* Used only for non-memory clusters */ + __le32 idx; /* In table entries for tables, in bytes for memory */ + __le32 addr_high; + __le32 addr_low; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_debug_dump_internals); + +/* Set Health Status (direct 0xFF20) */ +struct ixgbe_aci_cmd_set_health_status_config { + u8 event_source; +#define IXGBE_ACI_HEALTH_STATUS_SET_PF_SPECIFIC_MASK BIT(0) +#define IXGBE_ACI_HEALTH_STATUS_SET_ALL_PF_MASK BIT(1) +#define IXGBE_ACI_HEALTH_STATUS_SET_GLOBAL_MASK BIT(2) + u8 reserved[15]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_set_health_status_config); + +#define IXGBE_ACI_HEALTH_STATUS_ERR_UNKNOWN_MOD_STRICT 0x101 +#define IXGBE_ACI_HEALTH_STATUS_ERR_MOD_TYPE 0x102 +#define IXGBE_ACI_HEALTH_STATUS_ERR_MOD_QUAL 0x103 +#define IXGBE_ACI_HEALTH_STATUS_ERR_MOD_COMM 0x104 +#define IXGBE_ACI_HEALTH_STATUS_ERR_MOD_CONFLICT 0x105 +#define IXGBE_ACI_HEALTH_STATUS_ERR_MOD_NOT_PRESENT 0x106 +#define IXGBE_ACI_HEALTH_STATUS_INFO_MOD_UNDERUTILIZED 0x107 +#define IXGBE_ACI_HEALTH_STATUS_ERR_UNKNOWN_MOD_LENIENT 0x108 +#define IXGBE_ACI_HEALTH_STATUS_ERR_MOD_DIAGNOSTIC_FEATURE 0x109 +#define IXGBE_ACI_HEALTH_STATUS_ERR_INVALID_LINK_CFG 0x10B +#define IXGBE_ACI_HEALTH_STATUS_ERR_PORT_ACCESS 0x10C +#define IXGBE_ACI_HEALTH_STATUS_ERR_PORT_UNREACHABLE 0x10D +#define IXGBE_ACI_HEALTH_STATUS_INFO_PORT_SPEED_MOD_LIMITED 0x10F +#define IXGBE_ACI_HEALTH_STATUS_ERR_PARALLEL_FAULT 0x110 +#define IXGBE_ACI_HEALTH_STATUS_INFO_PORT_SPEED_PHY_LIMITED 0x111 +#define IXGBE_ACI_HEALTH_STATUS_ERR_NETLIST_TOPO 0x112 +#define IXGBE_ACI_HEALTH_STATUS_ERR_NETLIST 0x113 +#define IXGBE_ACI_HEALTH_STATUS_ERR_TOPO_CONFLICT 0x114 +#define IXGBE_ACI_HEALTH_STATUS_ERR_LINK_HW_ACCESS 0x115 +#define IXGBE_ACI_HEALTH_STATUS_ERR_LINK_RUNTIME 0x116 +#define IXGBE_ACI_HEALTH_STATUS_ERR_DNL_INIT 0x117 +#define IXGBE_ACI_HEALTH_STATUS_ERR_PHY_NVM_PROG 0x120 +#define IXGBE_ACI_HEALTH_STATUS_ERR_PHY_FW_LOAD 0x121 +#define IXGBE_ACI_HEALTH_STATUS_INFO_RECOVERY 0x500 +#define IXGBE_ACI_HEALTH_STATUS_ERR_FLASH_ACCESS 0x501 +#define IXGBE_ACI_HEALTH_STATUS_ERR_NVM_AUTH 0x502 +#define IXGBE_ACI_HEALTH_STATUS_ERR_OROM_AUTH 0x503 +#define IXGBE_ACI_HEALTH_STATUS_ERR_DDP_AUTH 0x504 +#define IXGBE_ACI_HEALTH_STATUS_ERR_NVM_COMPAT 0x505 +#define IXGBE_ACI_HEALTH_STATUS_ERR_OROM_COMPAT 0x506 +#define IXGBE_ACI_HEALTH_STATUS_ERR_NVM_SEC_VIOLATION 0x507 +#define IXGBE_ACI_HEALTH_STATUS_ERR_OROM_SEC_VIOLATION 0x508 +#define IXGBE_ACI_HEALTH_STATUS_ERR_DCB_MIB 0x509 +#define IXGBE_ACI_HEALTH_STATUS_ERR_MNG_TIMEOUT 0x50A +#define IXGBE_ACI_HEALTH_STATUS_ERR_BMC_RESET 0x50B +#define IXGBE_ACI_HEALTH_STATUS_ERR_LAST_MNG_FAIL 0x50C +#define IXGBE_ACI_HEALTH_STATUS_ERR_RESOURCE_ALLOC_FAIL 0x50D +#define IXGBE_ACI_HEALTH_STATUS_ERR_FW_LOOP 0x1000 +#define IXGBE_ACI_HEALTH_STATUS_ERR_FW_PFR_FAIL 0x1001 +#define IXGBE_ACI_HEALTH_STATUS_ERR_LAST_FAIL_AQ 0x1002 + +/* Get Health Status codes (indirect 0xFF21) */ +struct ixgbe_aci_cmd_get_supported_health_status_codes { + __le16 health_code_count; + u8 reserved[6]; + __le32 addr_high; + __le32 addr_low; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_get_supported_health_status_codes); + +/* Get Health Status (indirect 0xFF22) */ +struct ixgbe_aci_cmd_get_health_status { + __le16 health_status_count; + u8 reserved[6]; + __le32 addr_high; + __le32 addr_low; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_get_health_status); + +/* Get Health Status event buffer entry, (0xFF22) + * repeated per reported health status + */ +struct ixgbe_aci_cmd_health_status_elem { + __le16 health_status_code; + __le16 event_source; +#define IXGBE_ACI_HEALTH_STATUS_PF (0x1) +#define IXGBE_ACI_HEALTH_STATUS_PORT (0x2) +#define IXGBE_ACI_HEALTH_STATUS_GLOBAL (0x3) + __le32 internal_data1; +#define IXGBE_ACI_HEALTH_STATUS_UNDEFINED_DATA (0xDEADBEEF) + __le32 internal_data2; +}; + +IXGBE_CHECK_STRUCT_LEN(12, ixgbe_aci_cmd_health_status_elem); + +/* Clear Health Status (direct 0xFF23) */ +struct ixgbe_aci_cmd_clear_health_status { + __le32 reserved[4]; +}; + +IXGBE_CHECK_PARAM_LEN(ixgbe_aci_cmd_clear_health_status); + +/** + * struct ixgbe_aq_desc - Admin Command (AC) descriptor + * @flags: IXGBE_ACI_FLAG_* flags + * @opcode: Admin command opcode + * @datalen: length in bytes of indirect/external data buffer + * @retval: return value from firmware + * @cookie_high: opaque data high-half + * @cookie_low: opaque data low-half + * @params: command-specific parameters + * + * Descriptor format for commands the driver posts via the Admin Command Interface + * (ACI). The firmware writes back onto the command descriptor and returns + * the result of the command. Asynchronous events that are not an immediate + * result of the command are written to the Admin Command Interface (ACI) using + * the same descriptor format. Descriptors are in little-endian notation with + * 32-bit words. + */ +struct ixgbe_aci_desc { + __le16 flags; + __le16 opcode; + __le16 datalen; + __le16 retval; + __le32 cookie_high; + __le32 cookie_low; + union { + u8 raw[16]; + struct ixgbe_aci_cmd_generic generic; + struct ixgbe_aci_cmd_get_ver get_ver; + struct ixgbe_aci_cmd_driver_ver driver_ver; + struct ixgbe_aci_cmd_get_exp_err exp_err; + struct ixgbe_aci_cmd_req_res res_owner; + struct ixgbe_aci_cmd_list_caps get_cap; + struct ixgbe_aci_cmd_disable_rxen disable_rxen; + struct ixgbe_aci_cmd_get_fw_event get_fw_event; + struct ixgbe_aci_cmd_get_phy_caps get_phy; + struct ixgbe_aci_cmd_set_phy_cfg set_phy; + struct ixgbe_aci_cmd_restart_an restart_an; + struct ixgbe_aci_cmd_get_link_status get_link_status; + struct ixgbe_aci_cmd_set_event_mask set_event_mask; + struct ixgbe_aci_cmd_get_link_topo get_link_topo; + struct ixgbe_aci_cmd_get_link_topo_pin get_link_topo_pin; + struct ixgbe_aci_cmd_i2c read_write_i2c; + struct ixgbe_aci_cmd_read_i2c_resp read_i2c_resp; + struct ixgbe_aci_cmd_mdio read_write_mdio; + struct ixgbe_aci_cmd_mdio read_mdio; + struct ixgbe_aci_cmd_mdio write_mdio; + struct ixgbe_aci_cmd_gpio_by_func read_write_gpio_by_func; + struct ixgbe_aci_cmd_gpio read_write_gpio; + struct ixgbe_aci_cmd_sff_eeprom read_write_sff_param; + struct ixgbe_aci_cmd_prog_topo_dev_nvm prog_topo_dev_nvm; + struct ixgbe_aci_cmd_read_topo_dev_nvm read_topo_dev_nvm; + struct ixgbe_aci_cmd_nvm nvm; + struct ixgbe_aci_cmd_nvm_cfg nvm_cfg; + struct ixgbe_aci_cmd_nvm_checksum nvm_checksum; + struct ixgbe_aci_cmd_read_write_alt_direct read_write_alt_direct; + struct ixgbe_aci_cmd_read_write_alt_indirect read_write_alt_indirect; + struct ixgbe_aci_cmd_done_alt_write done_alt_write; + struct ixgbe_aci_cmd_clear_port_alt_write clear_port_alt_write; + struct ixgbe_aci_cmd_debug_dump_internals debug_dump; + struct ixgbe_aci_cmd_set_health_status_config + set_health_status_config; + struct ixgbe_aci_cmd_get_supported_health_status_codes + get_supported_health_status_codes; + struct ixgbe_aci_cmd_get_health_status get_health_status; + struct ixgbe_aci_cmd_clear_health_status clear_health_status; + struct ixgbe_aci_cmd_nvm_sanitization nvm_sanitization; + } params; +}; + +/* LKV-specific adapter context structures */ + +struct ixgbe_link_status { + /* Refer to ixgbe_aci_phy_type for bits definition */ + u64 phy_type_low; + u64 phy_type_high; + u8 topo_media_conflict; + u16 max_frame_size; + u16 link_speed; + u16 req_speeds; + u8 link_cfg_err; + u8 lse_ena; /* Link Status Event notification */ + u8 link_info; + u8 an_info; + u8 ext_info; + u8 fec_info; + u8 pacing; + /* Refer to #define from module_type[IXGBE_ACI_MODULE_TYPE_TOTAL_BYTE] of + * ixgbe_aci_get_phy_caps structure + */ + u8 module_type[IXGBE_ACI_MODULE_TYPE_TOTAL_BYTE]; +}; + +/* Common HW capabilities for SW use */ +struct ixgbe_hw_common_caps { + /* Write CSR protection */ + u64 wr_csr_prot; + u32 switching_mode; + /* switching mode supported - EVB switching (including cloud) */ +#define IXGBE_NVM_IMAGE_TYPE_EVB 0x0 + + /* Manageability mode & supported protocols over MCTP */ + u32 mgmt_mode; +#define IXGBE_MGMT_MODE_PASS_THRU_MODE_M 0xF +#define IXGBE_MGMT_MODE_CTL_INTERFACE_M 0xF0 +#define IXGBE_MGMT_MODE_REDIR_SB_INTERFACE_M 0xF00 + + u32 mgmt_protocols_mctp; +#define IXGBE_MGMT_MODE_PROTO_RSVD BIT(0) +#define IXGBE_MGMT_MODE_PROTO_PLDM BIT(1) +#define IXGBE_MGMT_MODE_PROTO_OEM BIT(2) +#define IXGBE_MGMT_MODE_PROTO_NC_SI BIT(3) + + u32 os2bmc; + u32 valid_functions; + /* DCB capabilities */ + u32 active_tc_bitmap; + u32 maxtc; + + /* RSS related capabilities */ + u32 rss_table_size; /* 512 for PFs and 64 for VFs */ + u32 rss_table_entry_width; /* RSS Entry width in bits */ + + /* Tx/Rx queues */ + u32 num_rxq; /* Number/Total Rx queues */ + u32 rxq_first_id; /* First queue ID for Rx queues */ + u32 num_txq; /* Number/Total Tx queues */ + u32 txq_first_id; /* First queue ID for Tx queues */ + + /* MSI-X vectors */ + u32 num_msix_vectors; + u32 msix_vector_first_id; + + /* Max MTU for function or device */ + u32 max_mtu; + + /* WOL related */ + u32 num_wol_proxy_fltr; + u32 wol_proxy_vsi_seid; + + /* LED/SDP pin count */ + u32 led_pin_num; + u32 sdp_pin_num; + + /* LED/SDP - Supports up to 12 LED pins and 8 SDP signals */ +#define IXGBE_MAX_SUPPORTED_GPIO_LED 12 +#define IXGBE_MAX_SUPPORTED_GPIO_SDP 8 + u8 led[IXGBE_MAX_SUPPORTED_GPIO_LED]; + u8 sdp[IXGBE_MAX_SUPPORTED_GPIO_SDP]; + /* VMDQ */ + u8 vmdq; /* VMDQ supported */ + + /* EVB capabilities */ + u8 evb_802_1_qbg; /* Edge Virtual Bridging */ + u8 evb_802_1_qbh; /* Bridge Port Extension */ + + u8 dcb; + u8 iscsi; + u8 ieee_1588; + u8 mgmt_cem; + + /* WoL and APM support */ +#define IXGBE_WOL_SUPPORT_M BIT(0) +#define IXGBE_ACPI_PROG_MTHD_M BIT(1) +#define IXGBE_PROXY_SUPPORT_M BIT(2) + u8 apm_wol_support; + u8 acpi_prog_mthd; + u8 proxy_support; + bool sec_rev_disabled; + bool update_disabled; + bool nvm_unified_update; + bool netlist_auth; +#define IXGBE_NVM_MGMT_SEC_REV_DISABLED BIT(0) +#define IXGBE_NVM_MGMT_UPDATE_DISABLED BIT(1) +#define IXGBE_NVM_MGMT_UNIFIED_UPD_SUPPORT BIT(3) +#define IXGBE_NVM_MGMT_NETLIST_AUTH_SUPPORT BIT(5) + bool no_drop_policy_support; + /* PCIe reset avoidance */ + bool pcie_reset_avoidance; /* false: not supported, true: supported */ + /* Post update reset restriction */ + bool reset_restrict_support; /* false: not supported, true: supported */ + + /* External topology device images within the NVM */ +#define IXGBE_EXT_TOPO_DEV_IMG_COUNT 4 + u32 ext_topo_dev_img_ver_high[IXGBE_EXT_TOPO_DEV_IMG_COUNT]; + u32 ext_topo_dev_img_ver_low[IXGBE_EXT_TOPO_DEV_IMG_COUNT]; + u8 ext_topo_dev_img_part_num[IXGBE_EXT_TOPO_DEV_IMG_COUNT]; +#define IXGBE_EXT_TOPO_DEV_IMG_PART_NUM_S 8 +#define IXGBE_EXT_TOPO_DEV_IMG_PART_NUM_M \ + MAKEMASK(0xFF, IXGBE_EXT_TOPO_DEV_IMG_PART_NUM_S) + bool ext_topo_dev_img_load_en[IXGBE_EXT_TOPO_DEV_IMG_COUNT]; +#define IXGBE_EXT_TOPO_DEV_IMG_LOAD_EN BIT(0) + bool ext_topo_dev_img_prog_en[IXGBE_EXT_TOPO_DEV_IMG_COUNT]; +#define IXGBE_EXT_TOPO_DEV_IMG_PROG_EN BIT(1) + bool next_cluster_id_support; +}; + +/* IEEE 1588 TIME_SYNC specific info */ +/* Function specific definitions */ +#define IXGBE_TS_FUNC_ENA_M BIT(0) +#define IXGBE_TS_SRC_TMR_OWND_M BIT(1) +#define IXGBE_TS_TMR_ENA_M BIT(2) +#define IXGBE_TS_TMR_IDX_OWND_S 4 +#define IXGBE_TS_TMR_IDX_OWND_M BIT(4) +#define IXGBE_TS_CLK_FREQ_S 16 +#define IXGBE_TS_CLK_FREQ_M MAKEMASK(0x7, IXGBE_TS_CLK_FREQ_S) +#define IXGBE_TS_CLK_SRC_S 20 +#define IXGBE_TS_CLK_SRC_M BIT(20) +#define IXGBE_TS_TMR_IDX_ASSOC_S 24 +#define IXGBE_TS_TMR_IDX_ASSOC_M BIT(24) + +/* TIME_REF clock rate specification */ +enum ixgbe_time_ref_freq { + IXGBE_TIME_REF_FREQ_25_000 = 0, + IXGBE_TIME_REF_FREQ_122_880 = 1, + IXGBE_TIME_REF_FREQ_125_000 = 2, + IXGBE_TIME_REF_FREQ_153_600 = 3, + IXGBE_TIME_REF_FREQ_156_250 = 4, + IXGBE_TIME_REF_FREQ_245_760 = 5, + + NUM_IXGBE_TIME_REF_FREQ +}; + +struct ixgbe_ts_func_info { + /* Function specific info */ + enum ixgbe_time_ref_freq time_ref; + u8 clk_freq; + u8 clk_src; + u8 tmr_index_assoc; + u8 ena; + u8 tmr_index_owned; + u8 src_tmr_owned; + u8 tmr_ena; +}; + +/* Device specific definitions */ +#define IXGBE_TS_TMR0_OWNR_M 0x7 +#define IXGBE_TS_TMR0_OWND_M BIT(3) +#define IXGBE_TS_TMR1_OWNR_S 4 +#define IXGBE_TS_TMR1_OWNR_M MAKEMASK(0x7, IXGBE_TS_TMR1_OWNR_S) +#define IXGBE_TS_TMR1_OWND_M BIT(7) +#define IXGBE_TS_DEV_ENA_M BIT(24) +#define IXGBE_TS_TMR0_ENA_M BIT(25) +#define IXGBE_TS_TMR1_ENA_M BIT(26) + +struct ixgbe_ts_dev_info { + /* Device specific info */ + u32 ena_ports; + u32 tmr_own_map; + u32 tmr0_owner; + u32 tmr1_owner; + u8 tmr0_owned; + u8 tmr1_owned; + u8 ena; + u8 tmr0_ena; + u8 tmr1_ena; +}; + +#pragma pack(1) +struct ixgbe_orom_civd_info { + u8 signature[4]; /* Must match ASCII '$CIV' characters */ + u8 checksum; /* Simple modulo 256 sum of all structure bytes must equal 0 */ + __le32 combo_ver; /* Combo Image Version number */ + u8 combo_name_len; /* Length of the unicode combo image version string, max of 32 */ + __le16 combo_name[32]; /* Unicode string representing the Combo Image version */ +}; +#pragma pack() + +/* Function specific capabilities */ +struct ixgbe_hw_func_caps { + struct ixgbe_hw_common_caps common_cap; + u32 guar_num_vsi; + struct ixgbe_ts_func_info ts_func_info; + bool no_drop_policy_ena; +}; + +/* Device wide capabilities */ +struct ixgbe_hw_dev_caps { + struct ixgbe_hw_common_caps common_cap; + u32 num_vsi_allocd_to_host; /* Excluding EMP VSI */ + u32 num_flow_director_fltr; /* Number of FD filters available */ + struct ixgbe_ts_dev_info ts_dev_info; + u32 num_funcs; +}; + +/* ACI event information */ +struct ixgbe_aci_event { + struct ixgbe_aci_desc desc; + u16 msg_len; + u16 buf_len; + u8 *msg_buf; +}; + +struct ixgbe_aci_info { + enum ixgbe_aci_err last_status; /* last status of sent admin command */ + struct ixgbe_lock lock; /* admin command interface lock */ +}; + +/* Minimum Security Revision information */ +struct ixgbe_minsrev_info { + u32 nvm; + u32 orom; + u8 nvm_valid : 1; + u8 orom_valid : 1; +}; + +/* Enumeration of which flash bank is desired to read from, either the active + * bank or the inactive bank. Used to abstract 1st and 2nd bank notion from + * code which just wants to read the active or inactive flash bank. + */ +enum ixgbe_bank_select { + IXGBE_ACTIVE_FLASH_BANK, + IXGBE_INACTIVE_FLASH_BANK, +}; + +/* Option ROM version information */ +struct ixgbe_orom_info { + u8 major; /* Major version of OROM */ + u8 patch; /* Patch version of OROM */ + u16 build; /* Build version of OROM */ + u32 srev; /* Security revision */ +}; + +/* NVM version information */ +struct ixgbe_nvm_info { + u32 eetrack; + u32 srev; + u8 major; + u8 minor; +}; + +/* Enumeration of possible flash banks for the NVM, OROM, and Netlist modules + * of the flash image. + */ +enum ixgbe_flash_bank { + IXGBE_INVALID_FLASH_BANK, + IXGBE_1ST_FLASH_BANK, + IXGBE_2ND_FLASH_BANK, +}; + +/* information for accessing NVM, OROM, and Netlist flash banks */ +struct ixgbe_bank_info { + u32 nvm_ptr; /* Pointer to 1st NVM bank */ + u32 nvm_size; /* Size of NVM bank */ + u32 orom_ptr; /* Pointer to 1st OROM bank */ + u32 orom_size; /* Size of OROM bank */ + u32 netlist_ptr; /* Pointer to 1st Netlist bank */ + u32 netlist_size; /* Size of Netlist bank */ + enum ixgbe_flash_bank nvm_bank; /* Active NVM bank */ + enum ixgbe_flash_bank orom_bank; /* Active OROM bank */ + enum ixgbe_flash_bank netlist_bank; /* Active Netlist bank */ +}; + +/* Flash Chip Information */ +struct ixgbe_flash_info { + struct ixgbe_orom_info orom; /* Option ROM version info */ + struct ixgbe_nvm_info nvm; /* NVM version information */ + struct ixgbe_bank_info banks; /* Flash Bank information */ + u16 sr_words; /* Shadow RAM size in words */ + u32 flash_size; /* Size of available flash in bytes */ + u8 blank_nvm_mode; /* is NVM empty (no FW present) */ +}; + +#define IXGBE_NVM_CMD_READ 0x0000000B +#define IXGBE_NVM_CMD_WRITE 0x0000000C + +/* NVM Access command */ +struct ixgbe_nvm_access_cmd { + u32 command; /* NVM command: READ or WRITE */ + u32 offset; /* Offset to read/write, in bytes */ + u32 data_size; /* Size of data field, in bytes */ +}; + +/* NVM Access data */ +struct ixgbe_nvm_access_data { + u32 regval; /* Storage for register value */ +}; + +#endif /* _IXGBE_TYPE_E610_H_ */ diff --git a/drivers/net/ixgbe/base/meson.build b/drivers/net/ixgbe/base/meson.build index f6497014da..6d72c11504 100644 --- a/drivers/net/ixgbe/base/meson.build +++ b/drivers/net/ixgbe/base/meson.build @@ -9,8 +9,10 @@ sources = [ 'ixgbe_dcb_82598.c', 'ixgbe_dcb_82599.c', 'ixgbe_dcb.c', + 'ixgbe_e610.c', 'ixgbe_hv_vf.c', 'ixgbe_mbx.c', + 'ixgbe_osdep.c', 'ixgbe_phy.c', 'ixgbe_vf.c', 'ixgbe_x540.c',