From patchwork Tue Mar 5 10:17:47 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Mingjin Ye X-Patchwork-Id: 137986 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 D34D143B86; Tue, 5 Mar 2024 11:36:00 +0100 (CET) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id ACF754026B; Tue, 5 Mar 2024 11:36:00 +0100 (CET) Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.19]) by mails.dpdk.org (Postfix) with ESMTP id 78B6C4014F for ; Tue, 5 Mar 2024 11:35:58 +0100 (CET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1709634958; x=1741170958; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=lcX0UUZSWlbE2HDo9QpWY3WSmI0SR2KTNjY4d/dYhBc=; b=YKTqfCTE0wNxibF2uRnmKFxlY9kZutV8Ej31oD6cltBNf9ZZK22S7Fvz wR4PcJQvSll+R8FjmJd1YpHGWydmLeMRGc5j1k/kwVaLUL8Mbpw5hFXWS xP20ukVTJ8wivTLKL4hMD9QC0lA5yNPOdHRiDzr8jQMwRhKHj6bwStvZz U69VWYEMq5grPGfFg5Nv7vY1AQkOVorh+hj/xuYgEn6ow0h1kJtJlJq2E JcBdxUeMp51Z7R3owyJ92CDh4ugGOpbA/W14SokOORk/4SJJS1XLmXiYA UAHqXY7MqKWMqqvMWLJnviNU5BmDG9rqRCM4GlAvDSQd46KFcleClPftS Q==; X-IronPort-AV: E=McAfee;i="6600,9927,11003"; a="4038821" X-IronPort-AV: E=Sophos;i="6.06,205,1705392000"; d="scan'208";a="4038821" Received: from fmviesa001.fm.intel.com ([10.60.135.141]) by fmvoesa113.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 05 Mar 2024 02:35:57 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.06,205,1705392000"; d="scan'208";a="40326783" Received: from unknown (HELO localhost.localdomain) ([10.239.252.253]) by smtpauth.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 05 Mar 2024 02:35:55 -0800 From: Mingjin Ye To: dev@dpdk.org Cc: Mingjin Ye , Yuying Zhang Subject: [PATCH v7] net/i40e: add diagnostic support in Tx path Date: Tue, 5 Mar 2024 10:17:47 +0000 Message-Id: <20240305101747.769464-1-mingjinx.ye@intel.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20240304093321.592061-1-mingjinx.ye@intel.com> References: <20240304093321.592061-1-mingjinx.ye@intel.com> 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 Implemented a Tx wrapper to perform a thorough check on mbufs, categorizing and counting invalid cases by type for diagnostic purposes. The count of invalid cases is accessible through xstats_get. Also, the devarg option "mbuf_check" was introduced to configure the diagnostic parameters to enable the appropriate diagnostic features. supported cases: mbuf, size, segment, offload. 1. mbuf: Check for corrupted mbuf. 2. size: Check min/max packet length according to HW spec. 3. segment: Check number of mbuf segments not exceed HW limits. 4. offload: Check for use of an unsupported offload flag. parameter format: "mbuf_check=" or "mbuf_check=[,]" eg: dpdk-testpmd -a 0000:87:00.0,mbuf_check=[mbuf,size] -- -i Signed-off-by: Mingjin Ye Acked-by: Bruce Richardson --- v2: remove strict. --- v3: optimised. --- v4: rebase. --- v5: fix ci error. --- v6: Changes the commit log. --- v7: Remove unnecessary changes. --- doc/guides/nics/i40e.rst | 13 +++ drivers/net/i40e/i40e_ethdev.c | 142 ++++++++++++++++++++++++++++++++- drivers/net/i40e/i40e_ethdev.h | 14 ++++ drivers/net/i40e/i40e_rxtx.c | 112 ++++++++++++++++++++++++++ drivers/net/i40e/i40e_rxtx.h | 2 + 5 files changed, 282 insertions(+), 1 deletion(-) diff --git a/doc/guides/nics/i40e.rst b/doc/guides/nics/i40e.rst index 15689ac958..bf1d1e5d60 100644 --- a/doc/guides/nics/i40e.rst +++ b/doc/guides/nics/i40e.rst @@ -275,6 +275,19 @@ Runtime Configuration -a 84:00.0,vf_msg_cfg=80@120:180 +- ``Support TX diagnostics`` (default ``not enabled``) + + Set the ``devargs`` parameter ``mbuf_check`` to enable TX diagnostics. For example, + ``-a 18:01.0,mbuf_check=`` or ``-a 18:01.0,mbuf_check=[,...]``. Also, + ``xstats_get`` can be used to get the error counts, which are collected in + ``tx_mbuf_error_packets`` xstats. For example, ``testpmd> show port xstats all``. + Supported cases: + + * mbuf: Check for corrupted mbuf. + * size: Check min/max packet length according to hw spec. + * segment: Check number of mbuf segments not exceed hw limitation. + * offload: Check any unsupported offload flag. + Vector RX Pre-conditions ~~~~~~~~~~~~~~~~~~~~~~~~ For Vector RX it is assumed that the number of descriptor rings will be a power diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c index 4d21341382..84fefcb1f9 100644 --- a/drivers/net/i40e/i40e_ethdev.c +++ b/drivers/net/i40e/i40e_ethdev.c @@ -48,6 +48,7 @@ #define ETH_I40E_SUPPORT_MULTI_DRIVER "support-multi-driver" #define ETH_I40E_QUEUE_NUM_PER_VF_ARG "queue-num-per-vf" #define ETH_I40E_VF_MSG_CFG "vf_msg_cfg" +#define ETH_I40E_MBUF_CHECK_ARG "mbuf_check" #define I40E_CLEAR_PXE_WAIT_MS 200 #define I40E_VSI_TSR_QINQ_STRIP 0x4010 @@ -412,6 +413,7 @@ static const char *const valid_keys[] = { ETH_I40E_SUPPORT_MULTI_DRIVER, ETH_I40E_QUEUE_NUM_PER_VF_ARG, ETH_I40E_VF_MSG_CFG, + ETH_I40E_MBUF_CHECK_ARG, NULL}; static const struct rte_pci_id pci_id_i40e_map[] = { @@ -545,6 +547,14 @@ static const struct rte_i40e_xstats_name_off rte_i40e_stats_strings[] = { #define I40E_NB_ETH_XSTATS (sizeof(rte_i40e_stats_strings) / \ sizeof(rte_i40e_stats_strings[0])) +static const struct rte_i40e_xstats_name_off i40e_mbuf_strings[] = { + {"tx_mbuf_error_packets", offsetof(struct i40e_mbuf_stats, + tx_pkt_errors)}, +}; + +#define I40E_NB_MBUF_XSTATS (sizeof(i40e_mbuf_strings) / \ + sizeof(i40e_mbuf_strings[0])) + static const struct rte_i40e_xstats_name_off rte_i40e_hw_port_strings[] = { {"tx_link_down_dropped", offsetof(struct i40e_hw_port_stats, tx_dropped_link_down)}, @@ -1373,6 +1383,94 @@ read_vf_msg_config(__rte_unused const char *key, return 0; } +static int +read_mbuf_check_config(__rte_unused const char *key, const char *value, void *args) +{ + char *cur; + char *tmp; + int str_len; + int valid_len; + + int ret = 0; + uint64_t *mc_flags = args; + char *str2 = strdup(value); + if (str2 == NULL) + return -1; + + str_len = strlen(str2); + if (str_len == 0) { + ret = -1; + goto err_end; + } + + /* Try stripping the outer square brackets of the parameter string. */ + str_len = strlen(str2); + if (str2[0] == '[' && str2[str_len - 1] == ']') { + if (str_len < 3) { + ret = -1; + goto err_end; + } + valid_len = str_len - 2; + memmove(str2, str2 + 1, valid_len); + memset(str2 + valid_len, '\0', 2); + } + + cur = strtok_r(str2, ",", &tmp); + while (cur != NULL) { + if (!strcmp(cur, "mbuf")) + *mc_flags |= I40E_MBUF_CHECK_F_TX_MBUF; + else if (!strcmp(cur, "size")) + *mc_flags |= I40E_MBUF_CHECK_F_TX_SIZE; + else if (!strcmp(cur, "segment")) + *mc_flags |= I40E_MBUF_CHECK_F_TX_SEGMENT; + else if (!strcmp(cur, "offload")) + *mc_flags |= I40E_MBUF_CHECK_F_TX_OFFLOAD; + else + PMD_DRV_LOG(ERR, "Unsupported diagnostic type: %s", cur); + cur = strtok_r(NULL, ",", &tmp); + } + +err_end: + free(str2); + return ret; +} + +static int +i40e_parse_mbuf_check(struct rte_eth_dev *dev) +{ + struct i40e_adapter *ad = + I40E_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private); + struct rte_kvargs *kvlist; + int kvargs_count; + int ret = 0; + + if (!dev->device->devargs) + return ret; + + kvlist = rte_kvargs_parse(dev->device->devargs->args, valid_keys); + if (!kvlist) + return -EINVAL; + + kvargs_count = rte_kvargs_count(kvlist, ETH_I40E_MBUF_CHECK_ARG); + if (!kvargs_count) + goto free_end; + + if (kvargs_count > 1) { + PMD_DRV_LOG(ERR, "More than one argument \"%s\"!", + ETH_I40E_MBUF_CHECK_ARG); + ret = -EINVAL; + goto free_end; + } + + if (rte_kvargs_process(kvlist, ETH_I40E_MBUF_CHECK_ARG, + read_mbuf_check_config, &ad->mbuf_check) < 0) + ret = -EINVAL; + +free_end: + rte_kvargs_free(kvlist); + return ret; +} + static int i40e_parse_vf_msg_config(struct rte_eth_dev *dev, struct i40e_vf_msg_cfg *msg_cfg) @@ -1488,6 +1586,7 @@ eth_i40e_dev_init(struct rte_eth_dev *dev, void *init_params __rte_unused) } i40e_parse_vf_msg_config(dev, &pf->vf_msg_cfg); + i40e_parse_mbuf_check(dev); /* Check if need to support multi-driver */ i40e_support_multi_driver(dev); @@ -2324,6 +2423,8 @@ i40e_dev_start(struct rte_eth_dev *dev) struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private); struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private); struct i40e_vsi *main_vsi = pf->main_vsi; + struct i40e_adapter *ad = + I40E_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private); int ret, i; struct rte_pci_device *pci_dev = RTE_ETH_DEV_TO_PCI(dev); struct rte_intr_handle *intr_handle = pci_dev->intr_handle; @@ -2483,6 +2584,7 @@ i40e_dev_start(struct rte_eth_dev *dev) max_frame_size = dev->data->mtu ? dev->data->mtu + I40E_ETH_OVERHEAD : I40E_FRAME_SIZE_MAX; + ad->max_pkt_len = max_frame_size; /* Set the max frame size to HW*/ i40e_aq_set_mac_config(hw, max_frame_size, TRUE, false, 0, NULL); @@ -3502,13 +3604,16 @@ i40e_dev_stats_reset(struct rte_eth_dev *dev) /* read the stats, reading current register values into offset */ i40e_read_stats_registers(pf, hw); + memset(&pf->mbuf_stats, 0, sizeof(struct i40e_mbuf_stats)); + return 0; } static uint32_t i40e_xstats_calc_num(void) { - return I40E_NB_ETH_XSTATS + I40E_NB_HW_PORT_XSTATS + + return I40E_NB_ETH_XSTATS + I40E_NB_MBUF_XSTATS + + I40E_NB_HW_PORT_XSTATS + (I40E_NB_RXQ_PRIO_XSTATS * 8) + (I40E_NB_TXQ_PRIO_XSTATS * 8); } @@ -3533,6 +3638,14 @@ static int i40e_dev_xstats_get_names(__rte_unused struct rte_eth_dev *dev, count++; } + /* Get stats from i40e_mbuf_stats struct */ + for (i = 0; i < I40E_NB_MBUF_XSTATS; i++) { + strlcpy(xstats_names[count].name, + i40e_mbuf_strings[i].name, + sizeof(xstats_names[count].name)); + count++; + } + /* Get individual stats from i40e_hw_port struct */ for (i = 0; i < I40E_NB_HW_PORT_XSTATS; i++) { strlcpy(xstats_names[count].name, @@ -3563,12 +3676,28 @@ static int i40e_dev_xstats_get_names(__rte_unused struct rte_eth_dev *dev, return count; } +static void +i40e_dev_update_mbuf_stats(struct rte_eth_dev *ethdev, + struct i40e_mbuf_stats *mbuf_stats) +{ + uint16_t idx; + struct i40e_tx_queue *txq; + + for (idx = 0; idx < ethdev->data->nb_tx_queues; idx++) { + txq = ethdev->data->tx_queues[idx]; + mbuf_stats->tx_pkt_errors += txq->mbuf_errors; + } +} + static int i40e_dev_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats, unsigned n) { struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private); struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private); + struct i40e_adapter *adapter = + I40E_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private); + struct i40e_mbuf_stats mbuf_stats = {0}; unsigned i, count, prio; struct i40e_hw_port_stats *hw_stats = &pf->stats; @@ -3583,6 +3712,9 @@ i40e_dev_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats, count = 0; + if (adapter->mbuf_check) + i40e_dev_update_mbuf_stats(dev, &mbuf_stats); + /* Get stats from i40e_eth_stats struct */ for (i = 0; i < I40E_NB_ETH_XSTATS; i++) { xstats[count].value = *(uint64_t *)(((char *)&hw_stats->eth) + @@ -3591,6 +3723,14 @@ i40e_dev_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats, count++; } + /* Get stats from i40e_mbuf_stats struct */ + for (i = 0; i < I40E_NB_MBUF_XSTATS; i++) { + xstats[count].value = *(uint64_t *)((char *)&mbuf_stats + + i40e_mbuf_strings[i].offset); + xstats[count].id = count; + count++; + } + /* Get individual stats from i40e_hw_port struct */ for (i = 0; i < I40E_NB_HW_PORT_XSTATS; i++) { xstats[count].value = *(uint64_t *)(((char *)hw_stats) + diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h index 445e1c0b38..b9628ca158 100644 --- a/drivers/net/i40e/i40e_ethdev.h +++ b/drivers/net/i40e/i40e_ethdev.h @@ -1109,6 +1109,10 @@ struct i40e_vf_msg_cfg { uint32_t ignore_second; }; +struct i40e_mbuf_stats { + uint64_t tx_pkt_errors; +}; + /* * Structure to store private data specific for PF instance. */ @@ -1123,6 +1127,7 @@ struct i40e_pf { struct i40e_hw_port_stats stats_offset; struct i40e_hw_port_stats stats; + struct i40e_mbuf_stats mbuf_stats; u64 rx_err1; /* rxerr1 */ u64 rx_err1_offset; @@ -1225,6 +1230,11 @@ struct i40e_vsi_vlan_pvid_info { #define I40E_MAX_PKT_TYPE 256 #define I40E_FLOW_TYPE_MAX 64 +#define I40E_MBUF_CHECK_F_TX_MBUF (1ULL << 0) +#define I40E_MBUF_CHECK_F_TX_SIZE (1ULL << 1) +#define I40E_MBUF_CHECK_F_TX_SEGMENT (1ULL << 2) +#define I40E_MBUF_CHECK_F_TX_OFFLOAD (1ULL << 3) + /* * Structure to store private data for each PF/VF instance. */ @@ -1241,6 +1251,10 @@ struct i40e_adapter { bool tx_simple_allowed; bool tx_vec_allowed; + uint64_t mbuf_check; /* mbuf check flags. */ + uint16_t max_pkt_len; /* Maximum packet length */ + eth_tx_burst_t tx_pkt_burst; + /* For PTP */ struct rte_timecounter systime_tc; struct rte_timecounter rx_tstamp_tc; diff --git a/drivers/net/i40e/i40e_rxtx.c b/drivers/net/i40e/i40e_rxtx.c index a0bc30d45b..a6f864574b 100644 --- a/drivers/net/i40e/i40e_rxtx.c +++ b/drivers/net/i40e/i40e_rxtx.c @@ -1536,6 +1536,112 @@ i40e_xmit_pkts_vec(void *tx_queue, struct rte_mbuf **tx_pkts, return nb_tx; } +/* Tx mbuf check */ +static uint16_t +i40e_xmit_pkts_check(void *tx_queue, struct rte_mbuf **tx_pkts, uint16_t nb_pkts) +{ + struct i40e_tx_queue *txq = tx_queue; + uint16_t idx; + uint64_t ol_flags; + struct rte_mbuf *mb; + bool pkt_error = false; + const char *reason = NULL; + uint16_t good_pkts = nb_pkts; + struct i40e_adapter *adapter = txq->vsi->adapter; + + for (idx = 0; idx < nb_pkts; idx++) { + mb = tx_pkts[idx]; + ol_flags = mb->ol_flags; + + if ((adapter->mbuf_check & I40E_MBUF_CHECK_F_TX_MBUF) && + (rte_mbuf_check(mb, 0, &reason) != 0)) { + PMD_TX_LOG(ERR, "INVALID mbuf: %s\n", reason); + pkt_error = true; + break; + } + + if ((adapter->mbuf_check & I40E_MBUF_CHECK_F_TX_SIZE) && + (mb->data_len > mb->pkt_len || + mb->data_len < I40E_TX_MIN_PKT_LEN || + mb->data_len > adapter->max_pkt_len)) { + PMD_TX_LOG(ERR, "INVALID mbuf: data_len (%u) is out of range, reasonable range (%d - %u)\n", + mb->data_len, I40E_TX_MIN_PKT_LEN, adapter->max_pkt_len); + pkt_error = true; + break; + } + + if (adapter->mbuf_check & I40E_MBUF_CHECK_F_TX_SEGMENT) { + if (!(ol_flags & RTE_MBUF_F_TX_TCP_SEG)) { + /** + * No TSO case: nb->segs, pkt_len to not exceed + * the limites. + */ + if (mb->nb_segs > I40E_TX_MAX_MTU_SEG) { + PMD_TX_LOG(ERR, "INVALID mbuf: nb_segs (%d) exceeds HW limit, maximum allowed value is %d\n", + mb->nb_segs, I40E_TX_MAX_MTU_SEG); + pkt_error = true; + break; + } + if (mb->pkt_len > I40E_FRAME_SIZE_MAX) { + PMD_TX_LOG(ERR, "INVALID mbuf: pkt_len (%d) exceeds HW limit, maximum allowed value is %d\n", + mb->nb_segs, I40E_FRAME_SIZE_MAX); + pkt_error = true; + break; + } + } else if (ol_flags & RTE_MBUF_F_TX_TCP_SEG) { + /** TSO case: tso_segsz, nb_segs, pkt_len not exceed + * the limits. + */ + if (mb->tso_segsz < I40E_MIN_TSO_MSS || + mb->tso_segsz > I40E_MAX_TSO_MSS) { + /** + * MSS outside the range are considered malicious + */ + PMD_TX_LOG(ERR, "INVALID mbuf: tso_segsz (%u) is out of range, reasonable range (%d - %u)\n", + mb->tso_segsz, I40E_MIN_TSO_MSS, I40E_MAX_TSO_MSS); + pkt_error = true; + break; + } + if (mb->nb_segs > + ((struct i40e_tx_queue *)tx_queue)->nb_tx_desc) { + PMD_TX_LOG(ERR, "INVALID mbuf: nb_segs out of ring length\n"); + pkt_error = true; + break; + } + if (mb->pkt_len > I40E_TSO_FRAME_SIZE_MAX) { + PMD_TX_LOG(ERR, "INVALID mbuf: pkt_len (%d) exceeds HW limit, maximum allowed value is %d\n", + mb->nb_segs, I40E_TSO_FRAME_SIZE_MAX); + pkt_error = true; + break; + } + } + } + + if (adapter->mbuf_check & I40E_MBUF_CHECK_F_TX_OFFLOAD) { + if (ol_flags & I40E_TX_OFFLOAD_NOTSUP_MASK) { + PMD_TX_LOG(ERR, "INVALID mbuf: TX offload is not supported\n"); + pkt_error = true; + break; + } + + if (!rte_validate_tx_offload(mb)) { + PMD_TX_LOG(ERR, "INVALID mbuf: TX offload setup error\n"); + pkt_error = true; + break; + } + } + } + + if (pkt_error) { + txq->mbuf_errors++; + good_pkts = idx; + if (good_pkts == 0) + return 0; + } + + return adapter->tx_pkt_burst(tx_queue, tx_pkts, good_pkts); +} + /********************************************************************* * * TX simple prep functions @@ -3468,6 +3574,7 @@ i40e_set_tx_function(struct rte_eth_dev *dev) { struct i40e_adapter *ad = I40E_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private); + uint64_t mbuf_check = ad->mbuf_check; int i; if (rte_eal_process_type() == RTE_PROC_PRIMARY) { @@ -3530,6 +3637,11 @@ i40e_set_tx_function(struct rte_eth_dev *dev) dev->tx_pkt_burst = i40e_xmit_pkts; dev->tx_pkt_prepare = i40e_prep_pkts; } + + if (mbuf_check) { + ad->tx_pkt_burst = dev->tx_pkt_burst; + dev->tx_pkt_burst = i40e_xmit_pkts_check; + } } static const struct { diff --git a/drivers/net/i40e/i40e_rxtx.h b/drivers/net/i40e/i40e_rxtx.h index 31dd947222..70320cf25e 100644 --- a/drivers/net/i40e/i40e_rxtx.h +++ b/drivers/net/i40e/i40e_rxtx.h @@ -162,6 +162,8 @@ struct i40e_tx_queue { uint16_t tx_next_dd; uint16_t tx_next_rs; bool q_set; /**< indicate if tx queue has been configured */ + uint64_t mbuf_errors; + bool tx_deferred_start; /**< don't start this queue in dev start */ uint8_t dcb_tc; /**< Traffic class of tx queue */ uint64_t offloads; /**< Tx offload flags of RTE_ETH_TX_OFFLOAD_* */