From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id A2EBCA0C53; Wed, 3 Nov 2021 11:06:28 +0100 (CET) Received: from [217.70.189.124] (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id E4EDB411A4; Wed, 3 Nov 2021 11:06:24 +0100 (CET) Received: from mga05.intel.com (mga05.intel.com [192.55.52.43]) by mails.dpdk.org (Postfix) with ESMTP id C99E0411A4; Wed, 3 Nov 2021 11:06:22 +0100 (CET) X-IronPort-AV: E=McAfee;i="6200,9189,10156"; a="317669104" X-IronPort-AV: E=Sophos;i="5.87,205,1631602800"; d="scan'208";a="317669104" Received: from orsmga007.jf.intel.com ([10.7.209.58]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Nov 2021 03:06:22 -0700 X-IronPort-AV: E=Sophos;i="5.87,205,1631602800"; d="scan'208";a="489482623" Received: from unknown (HELO localhost.localdomain) ([10.240.183.93]) by orsmga007-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Nov 2021 03:06:20 -0700 From: dapengx.yu@intel.com To: Qiming Yang , Qi Zhang Cc: dev@dpdk.org, haiyue.wang@intel.com, Dapeng Yu , stable@dpdk.org Date: Wed, 3 Nov 2021 18:05:27 +0800 Message-Id: <20211103100527.1175159-2-dapengx.yu@intel.com> X-Mailer: git-send-email 2.27.0 In-Reply-To: <20211103100527.1175159-1-dapengx.yu@intel.com> References: <20211101084506.513407-1-dapengx.yu@intel.com> <20211103100527.1175159-1-dapengx.yu@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Subject: [dpdk-dev] [PATCH v2 2/2] net/ice: fix flow redirect failure 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 Sender: "dev" From: Dapeng Yu It's possible that a switch rule can't be redirect successfully due to kernel driver is busy to handle an ongoing VF reset, so the redirect action need to be deferred into next redirect request which is promised by kernel driver after VF reset done. This patch uses the saved flow rule's meta to replay switch rule remove/add during next flow redirect. Fixes: 397b4b3c5095 ("net/ice: enable flow redirect on switch") Cc: stable@dpdk.org Signed-off-by: Dapeng Yu --- V2: * Add more filter status and VSI number --- drivers/net/ice/ice_switch_filter.c | 47 +++++++++++++++++++++++++++-- 1 file changed, 45 insertions(+), 2 deletions(-) diff --git a/drivers/net/ice/ice_switch_filter.c b/drivers/net/ice/ice_switch_filter.c index d5add64c53..5d7cdd7583 100644 --- a/drivers/net/ice/ice_switch_filter.c +++ b/drivers/net/ice/ice_switch_filter.c @@ -1926,7 +1926,10 @@ ice_switch_redirect(struct ice_adapter *ad, struct rte_flow *flow, struct ice_flow_redirect *rd) { - struct ice_rule_query_data *rdata = flow->rule; + struct ice_rule_query_data *rdata; + struct ice_switch_filter_conf *filter_conf_ptr = + (struct ice_switch_filter_conf *)flow->rule; + struct ice_rule_query_data added_rdata = { 0 }; struct ice_adv_fltr_mgmt_list_entry *list_itr; struct ice_adv_lkup_elem *lkups_dp = NULL; struct LIST_HEAD_TYPE *list_head; @@ -1936,6 +1939,8 @@ ice_switch_redirect(struct ice_adapter *ad, uint16_t lkups_cnt; int ret; + rdata = &filter_conf_ptr->sw_query_data; + if (rdata->vsi_handle != rd->vsi_handle) return 0; @@ -1946,6 +1951,33 @@ ice_switch_redirect(struct ice_adapter *ad, if (rd->type != ICE_FLOW_REDIRECT_VSI) return -ENOTSUP; + if (filter_conf_ptr->fltr_status == ICE_SW_FLTR_ADD_FAILED_ON_RIDRECT) { + /* Save VSI number for failure recover */ + filter_conf_ptr->vsi_num = rd->new_vsi_num; + + /* Update VSI context */ + hw->vsi_ctx[rd->vsi_handle]->vsi_num = rd->new_vsi_num; + + ret = ice_add_adv_rule(hw, filter_conf_ptr->lkups, + filter_conf_ptr->lkups_num, + &filter_conf_ptr->rule_info, + &added_rdata); + + if (ret) { + PMD_DRV_LOG(ERR, "Failed to replay the rule again"); + return -EINVAL; + } + + filter_conf_ptr->sw_query_data = added_rdata; + filter_conf_ptr->fltr_status = ICE_SW_FLTR_ADDED; + return 0; + } + + if (filter_conf_ptr->fltr_status == ICE_SW_FLTR_RMV_FAILED_ON_RIDRECT) { + /* Recover VSI context */ + hw->vsi_ctx[rd->vsi_handle]->vsi_num = filter_conf_ptr->vsi_num; + } + list_head = &sw->recp_list[rdata->rid].filt_rules; LIST_FOR_EACH_ENTRY(list_itr, list_head, ice_adv_fltr_mgmt_list_entry, list_entry) { @@ -1977,12 +2009,17 @@ ice_switch_redirect(struct ice_adapter *ad, if (!lkups_dp) return -EINVAL; + /* Save VSI number for failure recover */ + filter_conf_ptr->vsi_num = rd->new_vsi_num; + /* Remove the old rule */ ret = ice_rem_adv_rule(hw, list_itr->lkups, lkups_cnt, &rinfo); if (ret) { PMD_DRV_LOG(ERR, "Failed to delete the old rule %d", rdata->rule_id); + filter_conf_ptr->fltr_status = + ICE_SW_FLTR_RMV_FAILED_ON_RIDRECT; ret = -EINVAL; goto out; } @@ -1992,10 +2029,16 @@ ice_switch_redirect(struct ice_adapter *ad, /* Replay the rule */ ret = ice_add_adv_rule(hw, lkups_dp, lkups_cnt, - &rinfo, rdata); + &rinfo, &added_rdata); if (ret) { PMD_DRV_LOG(ERR, "Failed to replay the rule"); + filter_conf_ptr->fltr_status = + ICE_SW_FLTR_ADD_FAILED_ON_RIDRECT; ret = -EINVAL; + } else { + filter_conf_ptr->sw_query_data = added_rdata; + filter_conf_ptr->fltr_status = + ICE_SW_FLTR_ADDED; } out: -- 2.27.0