From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga06.intel.com (mga06.intel.com [134.134.136.31]) by dpdk.org (Postfix) with ESMTP id 3A7045589 for ; Thu, 29 Sep 2016 14:38:10 +0200 (CEST) Received: from orsmga003.jf.intel.com ([10.7.209.27]) by orsmga104.jf.intel.com with ESMTP; 29 Sep 2016 05:38:10 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.30,415,1470726000"; d="scan'208";a="885063294" Received: from dpdk2.bj.intel.com ([172.16.182.65]) by orsmga003.jf.intel.com with ESMTP; 29 Sep 2016 05:38:09 -0700 From: Zhiyong Yang To: dev@dpdk.org Cc: yuanhan.liu@linux.intel.com, ciara.loftus@intel.com, Zhiyong Yang Date: Thu, 29 Sep 2016 20:35:49 +0800 Message-Id: <1475152549-12295-3-git-send-email-zhiyong.yang@intel.com> X-Mailer: git-send-email 2.5.5 In-Reply-To: <1475152549-12295-1-git-send-email-zhiyong.yang@intel.com> References: <1475069208-137698-2-git-send-email-zhiyong.yang@intel.com> <1475152549-12295-1-git-send-email-zhiyong.yang@intel.com> Subject: [dpdk-dev] [PATCH v8 2/2] net/vhost: add pmd xstats X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 29 Sep 2016 12:38:12 -0000 This feature adds vhost pmd extended statistics from per port perspective in order to meet the requirements of the applications such as OVS etc. RX/TX xstats count the bytes without CRC. This is different from physical NIC stats with CRC. The statistics counters are based on RFC 2819 and RFC 2863 as follows: rx/tx_good_packets rx/tx_total_bytes rx/tx_missed_pkts rx/tx_broadcast_packets rx/tx_multicast_packets rx/tx_unicast_packets rx/tx_undersize_errors rx/tx_size_64_packets rx/tx_size_65_to_127_packets; rx/tx_size_128_to_255_packets; rx/tx_size_256_to_511_packets; rx/tx_size_512_to_1023_packets; rx/tx_size_1024_to_1522_packets; rx/tx_1523_to_max_packets; rx/tx_errors rx_fragmented_errors rx_jabber_errors rx_unknown_protos_packets; No API is changed or added. rte_eth_xstats_get_names() to retrieve what kinds of vhost xstats are supported, rte_eth_xstats_get() to retrieve vhost extended statistics, rte_eth_xstats_reset() to reset vhost extended statistics. The usage of vhost pmd xstats is the same as virtio pmd xstats. for example, when test-pmd application is running in interactive mode vhost pmd xstats will support the two following commands: show port xstats all | port_id will show vhost xstats clear port xstats all | port_id will reset vhost xstats net/virtio pmd xstats(the function virtio_update_packet_stats) is used as reference when implementing the feature. Thanks for the patches Zhiyong. I've tested the size stats and they look good to me. Tested-by: Ciara Loftus Signed-off-by: Zhiyong Yang --- Changes in V7: Removed the "_portX" prepended to the xstat names. Keep vhost xstats name consistent with physical NIC i40e, ixgbe, etc. Changes in V6: 1. Change xstats from per queue to per port. Keep vhost consistent with physical NIC i40e, ixgbe, etc. 2. Added the release note. Changes in V5: for vhost_count_multicast_broadcast, passing struct rte_mbuf *buf instead of struct rte_mbuf **buf and remove the 3th parameter uint16_t count;. Changes in v4: 1. add a member VHOST_XSTATS_MAX in enum vhost_xstats_pkts, So, we can define uint64_t xstats[VHOST_XSTATS_MAX]; instead of xstats[16]. 2. restore unicast_packets and update it in the function vhost_dev_xstats_get 3. move the loop out of function vhost_count_multicast_broadcast in order to reduce the computation. Changes in v3: 1. rework the vhost_update_packet_xstats and separate it into two parts. One function deals with the generic packets update, another one deals with increasing the broadcast and multicast with failure packets sent according to RFC2863 page42 ifHCOutMulticastPkts ifHCOutBroadcastPkts. 2. define enum vhost_stat_pkts to replace the magic numbers and enhance the code readability. 3. remove some unnecessary type casts and fix one format issue. Changes in v2: 1. remove the compiling switch. 2. fix two code bugs. doc/guides/rel_notes/release_16_11.rst | 3 + drivers/net/vhost/rte_eth_vhost.c | 270 +++++++++++++++++++++++++++++++++ 2 files changed, 273 insertions(+) diff --git a/doc/guides/rel_notes/release_16_11.rst b/doc/guides/rel_notes/release_16_11.rst index 1e37e48..5fc93e3 100644 --- a/doc/guides/rel_notes/release_16_11.rst +++ b/doc/guides/rel_notes/release_16_11.rst @@ -50,6 +50,9 @@ New Features * **Added virtio NEON support for ARM.** + * **Added vhost pmd xstats support.** + + Added vhost pmd extended statistics from per port perspective. Resolved Issues --------------- diff --git a/drivers/net/vhost/rte_eth_vhost.c b/drivers/net/vhost/rte_eth_vhost.c index c4af0f7..6943ac9 100644 --- a/drivers/net/vhost/rte_eth_vhost.c +++ b/drivers/net/vhost/rte_eth_vhost.c @@ -72,10 +72,30 @@ static struct ether_addr base_eth_addr = { } }; +enum vhost_xstats_pkts { + VHOST_UNDERSIZE_PKT = 0, + VHOST_64_PKT, + VHOST_65_TO_127_PKT, + VHOST_128_TO_255_PKT, + VHOST_256_TO_511_PKT, + VHOST_512_TO_1023_PKT, + VHOST_1024_TO_1522_PKT, + VHOST_1523_TO_MAX_PKT, + VHOST_BROADCAST_PKT, + VHOST_MULTICAST_PKT, + VHOST_UNICAST_PKT, + VHOST_ERRORS_PKT, + VHOST_ERRORS_FRAGMENTED, + VHOST_ERRORS_JABBER, + VHOST_UNKNOWN_PROTOCOL, + VHOST_XSTATS_MAX, +}; + struct vhost_stats { uint64_t pkts; uint64_t bytes; uint64_t missed_pkts; + uint64_t xstats[VHOST_XSTATS_MAX]; }; struct vhost_queue { @@ -129,6 +149,242 @@ struct rte_vhost_vring_state { static struct rte_vhost_vring_state *vring_states[RTE_MAX_ETHPORTS]; +#define VHOST_XSTATS_NAME_SIZE 64 + +struct vhost_xstats_name_off { + char name[VHOST_XSTATS_NAME_SIZE]; + uint64_t offset; +}; + +/* [rx]_is prepended to the name string here */ +static const struct vhost_xstats_name_off vhost_rxport_stat_strings[] = { + {"good_packets", + offsetof(struct vhost_queue, stats.pkts)}, + {"total_bytes", + offsetof(struct vhost_queue, stats.bytes)}, + {"missed_pkts", + offsetof(struct vhost_queue, stats.missed_pkts)}, + {"broadcast_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_BROADCAST_PKT])}, + {"multicast_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_MULTICAST_PKT])}, + {"unicast_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_UNICAST_PKT])}, + {"undersize_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_UNDERSIZE_PKT])}, + {"size_64_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_64_PKT])}, + {"size_65_to_127_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_65_TO_127_PKT])}, + {"size_128_to_255_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_128_TO_255_PKT])}, + {"size_256_to_511_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_256_TO_511_PKT])}, + {"size_512_to_1023_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_512_TO_1023_PKT])}, + {"size_1024_to_1522_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_1024_TO_1522_PKT])}, + {"size_1523_to_max_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_1523_TO_MAX_PKT])}, + {"errors_with_bad_CRC", + offsetof(struct vhost_queue, stats.xstats[VHOST_ERRORS_PKT])}, + {"fragmented_errors", + offsetof(struct vhost_queue, stats.xstats[VHOST_ERRORS_FRAGMENTED])}, + {"jabber_errors", + offsetof(struct vhost_queue, stats.xstats[VHOST_ERRORS_JABBER])}, + {"unknown_protos_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_UNKNOWN_PROTOCOL])}, +}; + +/* [tx]_ is prepended to the name string here */ +static const struct vhost_xstats_name_off vhost_txport_stat_strings[] = { + {"good_packets", + offsetof(struct vhost_queue, stats.pkts)}, + {"total_bytes", + offsetof(struct vhost_queue, stats.bytes)}, + {"missed_pkts", + offsetof(struct vhost_queue, stats.missed_pkts)}, + {"broadcast_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_BROADCAST_PKT])}, + {"multicast_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_MULTICAST_PKT])}, + {"unicast_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_UNICAST_PKT])}, + {"undersize_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_UNDERSIZE_PKT])}, + {"size_64_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_64_PKT])}, + {"size_65_to_127_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_65_TO_127_PKT])}, + {"size_128_to_255_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_128_TO_255_PKT])}, + {"size_256_to_511_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_256_TO_511_PKT])}, + {"size_512_to_1023_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_512_TO_1023_PKT])}, + {"size_1024_to_1522_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_1024_TO_1522_PKT])}, + {"size_1523_to_max_packets", + offsetof(struct vhost_queue, stats.xstats[VHOST_1523_TO_MAX_PKT])}, + {"errors_with_bad_CRC", + offsetof(struct vhost_queue, stats.xstats[VHOST_ERRORS_PKT])}, +}; + +#define VHOST_NB_XSTATS_RXPORT (sizeof(vhost_rxport_stat_strings) / \ + sizeof(vhost_rxport_stat_strings[0])) + +#define VHOST_NB_XSTATS_TXPORT (sizeof(vhost_txport_stat_strings) / \ + sizeof(vhost_txport_stat_strings[0])) + +static void +vhost_dev_xstats_reset(struct rte_eth_dev *dev) +{ + struct vhost_queue *vq = NULL; + unsigned int i = 0; + + for (i = 0; i < dev->data->nb_rx_queues; i++) { + vq = dev->data->rx_queues[i]; + if (!vq) + continue; + memset(&vq->stats, 0, sizeof(vq->stats)); + } + for (i = 0; i < dev->data->nb_tx_queues; i++) { + vq = dev->data->tx_queues[i]; + if (!vq) + continue; + memset(&vq->stats, 0, sizeof(vq->stats)); + } +} + +static int +vhost_dev_xstats_get_names(struct rte_eth_dev *dev __rte_unused, + struct rte_eth_xstat_name *xstats_names, + unsigned int limit __rte_unused) +{ + unsigned int t = 0; + int count = 0; + int nstats = VHOST_NB_XSTATS_RXPORT + VHOST_NB_XSTATS_TXPORT; + + if (!xstats_names) + return nstats; + for (t = 0; t < VHOST_NB_XSTATS_RXPORT; t++) { + snprintf(xstats_names[count].name, + sizeof(xstats_names[count].name), + "rx_%s", vhost_rxport_stat_strings[t].name); + count++; + } + for (t = 0; t < VHOST_NB_XSTATS_TXPORT; t++) { + snprintf(xstats_names[count].name, + sizeof(xstats_names[count].name), + "tx_%s", vhost_txport_stat_strings[t].name); + count++; + } + return count; +} + +static int +vhost_dev_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats, + unsigned int n) +{ + unsigned int i; + unsigned int t; + unsigned int count = 0; + struct vhost_queue *vq = NULL; + unsigned int nxstats = VHOST_NB_XSTATS_RXPORT + VHOST_NB_XSTATS_TXPORT; + + if (n < nxstats) + return nxstats; + + for (i = 0; i < dev->data->nb_rx_queues; i++) { + vq = dev->data->rx_queues[i]; + if (!vq) + continue; + vq->stats.xstats[VHOST_UNICAST_PKT] = vq->stats.pkts + - (vq->stats.xstats[VHOST_BROADCAST_PKT] + + vq->stats.xstats[VHOST_MULTICAST_PKT]); + } + for (i = 0; i < dev->data->nb_tx_queues; i++) { + vq = dev->data->tx_queues[i]; + if (!vq) + continue; + vq->stats.xstats[VHOST_UNICAST_PKT] = vq->stats.pkts + + vq->stats.missed_pkts + - (vq->stats.xstats[VHOST_BROADCAST_PKT] + + vq->stats.xstats[VHOST_MULTICAST_PKT]); + } + for (t = 0; t < VHOST_NB_XSTATS_RXPORT; t++) { + xstats[count].value = 0; + for (i = 0; i < dev->data->nb_rx_queues; i++) { + vq = dev->data->rx_queues[i]; + if (!vq) + continue; + xstats[count].value += + *(uint64_t *)(((char *)vq) + + vhost_rxport_stat_strings[t].offset); + } + count++; + } + for (t = 0; t < VHOST_NB_XSTATS_TXPORT; t++) { + xstats[count].value = 0; + for (i = 0; i < dev->data->nb_tx_queues; i++) { + vq = dev->data->tx_queues[i]; + if (!vq) + continue; + xstats[count].value += + *(uint64_t *)(((char *)vq) + + vhost_txport_stat_strings[t].offset); + } + count++; + } + return count; +} + +static inline void +vhost_count_multicast_broadcast(struct vhost_queue *vq, + struct rte_mbuf *mbuf) +{ + struct ether_addr *ea = NULL; + struct vhost_stats *pstats = &vq->stats; + + ea = rte_pktmbuf_mtod(mbuf, struct ether_addr *); + if (is_multicast_ether_addr(ea)) { + if (is_broadcast_ether_addr(ea)) + pstats->xstats[VHOST_BROADCAST_PKT]++; + else + pstats->xstats[VHOST_MULTICAST_PKT]++; + } +} + +static void +vhost_update_packet_xstats(struct vhost_queue *vq, + struct rte_mbuf **bufs, + uint16_t count) +{ + uint32_t pkt_len = 0; + uint64_t i = 0; + uint64_t index; + struct vhost_stats *pstats = &vq->stats; + + for (i = 0; i < count ; i++) { + pkt_len = bufs[i]->pkt_len; + if (pkt_len == 64) { + pstats->xstats[VHOST_64_PKT]++; + } else if (pkt_len > 64 && pkt_len < 1024) { + index = (sizeof(pkt_len) * 8) + - __builtin_clz(pkt_len) - 5; + pstats->xstats[index]++; + } else { + if (pkt_len < 64) + pstats->xstats[VHOST_UNDERSIZE_PKT]++; + else if (pkt_len <= 1522) + pstats->xstats[VHOST_1024_TO_1522_PKT]++; + else if (pkt_len > 1522) + pstats->xstats[VHOST_1523_TO_MAX_PKT]++; + } + vhost_count_multicast_broadcast(vq, bufs[i]); + } +} + static uint16_t eth_vhost_rx(void *q, struct rte_mbuf **bufs, uint16_t nb_bufs) { @@ -154,6 +410,8 @@ eth_vhost_rx(void *q, struct rte_mbuf **bufs, uint16_t nb_bufs) r->stats.bytes += bufs[i]->pkt_len; } + vhost_update_packet_xstats(r, bufs, nb_rx); + out: rte_atomic32_set(&r->while_queuing, 0); @@ -184,6 +442,15 @@ eth_vhost_tx(void *q, struct rte_mbuf **bufs, uint16_t nb_bufs) for (i = 0; likely(i < nb_tx); i++) r->stats.bytes += bufs[i]->pkt_len; + vhost_update_packet_xstats(r, bufs, nb_tx); + + /* According to RFC2863 page42 section ifHCOutMulticastPkts and + * ifHCOutBroadcastPkts, the counters "multicast" and "broadcast" + * are increased when packets are not transmitted successfully. + */ + for (i = nb_tx; i < nb_bufs; i++) + vhost_count_multicast_broadcast(r, bufs[i]); + for (i = 0; likely(i < nb_tx); i++) rte_pktmbuf_free(bufs[i]); out: @@ -684,6 +951,9 @@ static const struct eth_dev_ops ops = { .link_update = eth_link_update, .stats_get = eth_stats_get, .stats_reset = eth_stats_reset, + .xstats_reset = vhost_dev_xstats_reset, + .xstats_get = vhost_dev_xstats_get, + .xstats_get_names = vhost_dev_xstats_get_names, }; static int -- 2.5.5