Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753110AbbDAQUT (ORCPT ); Wed, 1 Apr 2015 12:20:19 -0400 Received: from mail-bn1on0111.outbound.protection.outlook.com ([157.56.110.111]:40727 "EHLO na01-bn1-obe.outbound.protection.outlook.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1753069AbbDAQUN (ORCPT ); Wed, 1 Apr 2015 12:20:13 -0400 Authentication-Results: spf=fail (sender IP is 192.88.168.50) smtp.mailfrom=freescale.com; freescale.mail.onmicrosoft.com; dkim=none (message not signed) header.d=none; From: Madalin Bucur To: , CC: , Madalin Bucur Subject: [PATCH RFC 08/10] dpaa_eth: add debugfs counters Date: Wed, 1 Apr 2015 19:19:54 +0300 Message-ID: <1427905196-27778-7-git-send-email-madalin.bucur@freescale.com> X-Mailer: git-send-email 1.5.6.5 In-Reply-To: <1427905196-27778-6-git-send-email-madalin.bucur@freescale.com> References: <1427905196-27778-1-git-send-email-madalin.bucur@freescale.com> <1427905196-27778-2-git-send-email-madalin.bucur@freescale.com> <1427905196-27778-3-git-send-email-madalin.bucur@freescale.com> <1427905196-27778-4-git-send-email-madalin.bucur@freescale.com> <1427905196-27778-5-git-send-email-madalin.bucur@freescale.com> <1427905196-27778-6-git-send-email-madalin.bucur@freescale.com> Reply-To: X-EOPAttributedMessage: 0 X-Forefront-Antispam-Report: CIP:192.88.168.50;CTRY:US;IPV:NLI;EFV:NLI;BMV:1;SFV:NSPM;SFS:(10019020)(6009001)(339900001)(199003)(189002)(105606002)(47776003)(48376002)(33646002)(104016003)(87936001)(46102003)(86362001)(50466002)(36756003)(43066003)(229853001)(106466001)(49486002)(62966003)(6806004)(50226001)(77096005)(93886004)(77156002)(2950100001)(76176999)(53806999)(50986999)(19580395003)(85426001)(19580405001)(217873001);DIR:OUT;SFP:1102;SCL:1;SRVR:BL2PR03MB546;H:tx30smr01.am.freescale.net;FPR:;SPF:Fail;MLV:sfv;A:1;MX:1;LANG:en; MIME-Version: 1.0 Content-Type: text/plain X-Microsoft-Antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BL2PR03MB546; X-Microsoft-Antispam-PRVS: X-Exchange-Antispam-Report-Test: UriScan:; X-Exchange-Antispam-Report-CFA-Test: BCL:0;PCL:0;RULEID:(601004)(5002010)(5005006);SRVR:BL2PR03MB546;BCL:0;PCL:0;RULEID:;SRVR:BL2PR03MB546; X-Forefront-PRVS: 053315510E X-OriginatorOrg: freescale.com X-MS-Exchange-CrossTenant-OriginalArrivalTime: 01 Apr 2015 16:20:09.5812 (UTC) X-MS-Exchange-CrossTenant-Id: 710a03f5-10f6-4d38-9ff4-a80b81da590d X-MS-Exchange-CrossTenant-OriginalAttributedTenantConnectingIp: TenantId=710a03f5-10f6-4d38-9ff4-a80b81da590d;Ip=[192.88.168.50];Helo=[tx30smr01.am.freescale.net] X-MS-Exchange-CrossTenant-FromEntityHeader: HybridOnPrem X-MS-Exchange-Transport-CrossTenantHeadersStamped: BL2PR03MB546 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 7655 Lines: 216 Add a series of counters to be exported through debugfs: - add detailed counters for reception errors; - add detailed counters for QMan enqueue reject events; - count the number of fragmented skbs received from the stack; - count all frames received on the Tx confirmation path; - add congestion group statistics; - count the number of interrupts for each CPU. Signed-off-by: Madalin Bucur --- drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 12 +++++++ drivers/net/ethernet/freescale/dpaa/dpaa_eth.h | 35 ++++++++++++++++++ .../net/ethernet/freescale/dpaa/dpaa_eth_common.c | 41 ++++++++++++++++++++-- .../net/ethernet/freescale/dpaa/dpaa_eth_common.h | 2 ++ drivers/net/ethernet/freescale/dpaa/dpaa_eth_sg.c | 1 + 5 files changed, 89 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c index f50638e7..eba0809 100644 --- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c @@ -99,6 +99,15 @@ static void _dpa_rx_error(struct net_device *net_dev, percpu_priv->stats.rx_errors++; + if (fd->status & FM_PORT_FRM_ERR_DMA) + percpu_priv->rx_errors.dme++; + if (fd->status & FM_PORT_FRM_ERR_PHYSICAL) + percpu_priv->rx_errors.fpe++; + if (fd->status & FM_PORT_FRM_ERR_SIZE) + percpu_priv->rx_errors.fse++; + if (fd->status & FM_PORT_FRM_ERR_PRS_HDR_ERR) + percpu_priv->rx_errors.phe++; + dpa_fd_release(net_dev, fd); } @@ -162,6 +171,8 @@ static void __hot _dpa_tx_conf(struct net_device *net_dev, percpu_priv->stats.tx_errors++; } + percpu_priv->tx_confirm++; + skb = _dpa_cleanup_tx_fd(priv, fd); dev_kfree_skb(skb); @@ -297,6 +308,7 @@ static void priv_ern(struct qman_portal *portal, percpu_priv->stats.tx_dropped++; percpu_priv->stats.tx_fifo_errors++; + count_ern(percpu_priv, msg); /* If we intended this buffer to go into the pool * when the FM was done, we need to put it in diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.h b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.h index 117627b..761063d 100644 --- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.h +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.h @@ -199,6 +199,25 @@ struct dpa_bp { void (*free_buf_cb)(void *addr); }; +struct dpa_rx_errors { + u64 dme; /* DMA Error */ + u64 fpe; /* Frame Physical Error */ + u64 fse; /* Frame Size Error */ + u64 phe; /* Header Error */ +}; + +/* Counters for QMan ERN frames - one counter per rejection code */ +struct dpa_ern_cnt { + u64 cg_tdrop; /* Congestion group taildrop */ + u64 wred; /* WRED congestion */ + u64 err_cond; /* Error condition */ + u64 early_window; /* Order restoration, frame too early */ + u64 late_window; /* Order restoration, frame too late */ + u64 fq_tdrop; /* FQ taildrop */ + u64 fq_retired; /* FQ is retired */ + u64 orp_zero; /* ORP disabled */ +}; + struct dpa_napi_portal { struct napi_struct napi; struct qman_portal *p; @@ -207,7 +226,13 @@ struct dpa_napi_portal { struct dpa_percpu_priv_s { struct net_device *net_dev; struct dpa_napi_portal *np; + u64 in_interrupt; + u64 tx_confirm; + /* fragmented (non-linear) skbuffs received from the stack */ + u64 tx_frag_skbuffs; struct rtnl_link_stats64 stats; + struct dpa_rx_errors rx_errors; + struct dpa_ern_cnt ern_cnt; }; struct dpa_priv_s { @@ -235,6 +260,15 @@ struct dpa_priv_s { * (and the same) congestion group. */ struct qman_cgr cgr; + /* If congested, when it began. Used for performance stats. */ + u32 congestion_start_jiffies; + /* Number of jiffies the Tx port was congested. */ + u32 congested_jiffies; + /** + * Counter for the number of times the CGR + * entered congestion state + */ + u32 cgr_congested_count; } cgr_data; /* Use a per-port CGR for ingress traffic. */ bool use_ingress_cgr; @@ -296,6 +330,7 @@ static inline int dpaa_eth_napi_schedule(struct dpa_percpu_priv_s *percpu_priv, np->p = portal; napi_schedule(&np->napi); + percpu_priv->in_interrupt++; return 1; } } diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth_common.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth_common.c index 7921b64..e92183a 100644 --- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth_common.c +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth_common.c @@ -782,10 +782,15 @@ static void dpaa_eth_cgscn(struct qman_portal *qm, struct qman_cgr *cgr, struct dpa_priv_s *priv = (struct dpa_priv_s *)container_of(cgr, struct dpa_priv_s, cgr_data.cgr); - if (congested) + if (congested) { + priv->cgr_data.congestion_start_jiffies = jiffies; netif_tx_stop_all_queues(priv->net_dev); - else + priv->cgr_data.cgr_congested_count++; + } else { + priv->cgr_data.congested_jiffies += + (jiffies - priv->cgr_data.congestion_start_jiffies); netif_tx_wake_all_queues(priv->net_dev); + } } int dpaa_eth_cgr_init(struct dpa_priv_s *priv) @@ -1226,6 +1231,38 @@ dpa_fd_release(const struct net_device *net_dev, const struct qm_fd *fd) } EXPORT_SYMBOL(dpa_fd_release); +void count_ern(struct dpa_percpu_priv_s *percpu_priv, + const struct qm_mr_entry *msg) +{ + switch (msg->ern.rc & QM_MR_RC_MASK) { + case QM_MR_RC_CGR_TAILDROP: + percpu_priv->ern_cnt.cg_tdrop++; + break; + case QM_MR_RC_WRED: + percpu_priv->ern_cnt.wred++; + break; + case QM_MR_RC_ERROR: + percpu_priv->ern_cnt.err_cond++; + break; + case QM_MR_RC_ORPWINDOW_EARLY: + percpu_priv->ern_cnt.early_window++; + break; + case QM_MR_RC_ORPWINDOW_LATE: + percpu_priv->ern_cnt.late_window++; + break; + case QM_MR_RC_FQ_TAILDROP: + percpu_priv->ern_cnt.fq_tdrop++; + break; + case QM_MR_RC_ORPWINDOW_RETIRED: + percpu_priv->ern_cnt.fq_retired++; + break; + case QM_MR_RC_ORP_ZERO: + percpu_priv->ern_cnt.orp_zero++; + break; + } +} +EXPORT_SYMBOL(count_ern); + /** * Turn on HW checksum computation for this outgoing frame. * If the current protocol is not something we support in this regard diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth_common.h b/drivers/net/ethernet/freescale/dpaa/dpaa_eth_common.h index 53da367..33acacc 100644 --- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth_common.h +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth_common.h @@ -121,6 +121,8 @@ void dpaa_eth_init_ports(struct mac_device *mac_dev, void dpa_release_sgt(struct qm_sg_entry *sgt); void __attribute__((nonnull)) dpa_fd_release(const struct net_device *net_dev, const struct qm_fd *fd); +void count_ern(struct dpa_percpu_priv_s *percpu_priv, + const struct qm_mr_entry *msg); int dpa_enable_tx_csum(struct dpa_priv_s *priv, struct sk_buff *skb, struct qm_fd *fd, diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth_sg.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth_sg.c index 410effd..74f22eb 100644 --- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth_sg.c +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth_sg.c @@ -656,6 +656,7 @@ int __hot dpa_tx(struct sk_buff *skb, struct net_device *net_dev) likely(skb_shinfo(skb)->nr_frags < DPA_SGT_MAX_ENTRIES)) { /* Just create a S/G fd based on the skb */ err = skb_to_sg_fd(priv, skb, &fd); + percpu_priv->tx_frag_skbuffs++; } else { /* Make sure we have enough headroom to accommodate private * data, parse results, etc. Normally this shouldn't happen if -- 1.7.11.7 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/