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 144D7A0093; Tue, 10 May 2022 22:17:29 +0200 (CEST) Received: from [217.70.189.124] (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id A7810410F2; Tue, 10 May 2022 22:17:28 +0200 (CEST) Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by mails.dpdk.org (Postfix) with ESMTP id 3DA36406B4 for ; Tue, 10 May 2022 22:17:27 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1652213846; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=HmsXkGPExsVWeGBd4XGp3ddz+GbOlp/Yfu2oFWNC3Fw=; b=Dihb2BlFs8ps+k7dc704+cu5BFA/hxpB6y2IAe80NWeKDNw/RXXV/Sk+0trvk0wF99dyfm c9HNXg2EJg9BdxAYXRFbEJ3ScmTTZICqPIJvd1h36gaaito7DLXjw4ZmAfn6YyT+3ngWg+ 840XYohWSrN6vsd2XDQfybzX4f9RZYY= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-31-8B4g2QX-P_quwhy0wLZRmA-1; Tue, 10 May 2022 16:17:25 -0400 X-MC-Unique: 8B4g2QX-P_quwhy0wLZRmA-1 Received: from smtp.corp.redhat.com (int-mx01.intmail.prod.int.rdu2.redhat.com [10.11.54.1]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 30EEB8002BF; Tue, 10 May 2022 20:17:25 +0000 (UTC) Received: from max-t490s.redhat.com (unknown [10.39.208.28]) by smtp.corp.redhat.com (Postfix) with ESMTP id 0D7CC40CFD06; Tue, 10 May 2022 20:17:23 +0000 (UTC) From: Maxime Coquelin To: dev@dpdk.org, chenbo.xia@intel.com, david.marchand@redhat.com, i.maximets@ovn.org Cc: Maxime Coquelin Subject: [PATCH 1/5] vhost: add per-virtqueue statistics support Date: Tue, 10 May 2022 22:17:16 +0200 Message-Id: <20220510201720.1262368-2-maxime.coquelin@redhat.com> In-Reply-To: <20220510201720.1262368-1-maxime.coquelin@redhat.com> References: <20220510201720.1262368-1-maxime.coquelin@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.84 on 10.11.54.1 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=maxime.coquelin@redhat.com X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset="US-ASCII"; x-default=true 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 This patch introduces new APIs for the application to query and reset per-virtqueue statistics. The patch also introduces generic counters. Signed-off-by: Maxime Coquelin --- doc/guides/prog_guide/vhost_lib.rst | 24 ++++++ lib/vhost/rte_vhost.h | 99 ++++++++++++++++++++++++ lib/vhost/socket.c | 4 +- lib/vhost/version.map | 4 +- lib/vhost/vhost.c | 112 +++++++++++++++++++++++++++- lib/vhost/vhost.h | 18 ++++- lib/vhost/virtio_net.c | 55 ++++++++++++++ 7 files changed, 311 insertions(+), 5 deletions(-) diff --git a/doc/guides/prog_guide/vhost_lib.rst b/doc/guides/prog_guide/vhost_lib.rst index f287b76ebf..0f91215a90 100644 --- a/doc/guides/prog_guide/vhost_lib.rst +++ b/doc/guides/prog_guide/vhost_lib.rst @@ -130,6 +130,15 @@ The following is an overview of some key Vhost API functions: It is disabled by default. + - ``RTE_VHOST_USER_NET_STATS_ENABLE`` + + Per-virtqueue statistics collection will be enabled when this flag is set. + When enabled, the application may use rte_vhost_stats_get_names() and + rte_vhost_stats_get() to collect statistics, and rte_vhost_stats_reset() to + reset them. + + It is disabled by default + * ``rte_vhost_driver_set_features(path, features)`` This function sets the feature bits the vhost-user driver supports. The @@ -282,6 +291,21 @@ The following is an overview of some key Vhost API functions: Clear inflight packets which are submitted to DMA engine in vhost async data path. Completed packets are returned to applications through ``pkts``. +* ``rte_vhost_vring_stats_get_names(int vid, uint16_t queue_id, struct rte_vhost_stat_name *names, unsigned int size)`` + + This function returns the names of the queue statistics. It requires + statistics collection to be enabled at registration time. + +* ``rte_vhost_vring_stats_get(int vid, uint16_t queue_id, struct rte_vhost_stat *stats, unsigned int n)`` + + This function returns the queue statistics. It requires statistics + collection to be enabled at registration time. + +* ``rte_vhost_vring_stats_reset(int vid, uint16_t queue_id)`` + + This function resets the queue statistics. It requires statistics + collection to be enabled at registration time. + Vhost-user Implementations -------------------------- diff --git a/lib/vhost/rte_vhost.h b/lib/vhost/rte_vhost.h index c733f857c6..266048f3e5 100644 --- a/lib/vhost/rte_vhost.h +++ b/lib/vhost/rte_vhost.h @@ -39,6 +39,7 @@ extern "C" { #define RTE_VHOST_USER_LINEARBUF_SUPPORT (1ULL << 6) #define RTE_VHOST_USER_ASYNC_COPY (1ULL << 7) #define RTE_VHOST_USER_NET_COMPLIANT_OL_FLAGS (1ULL << 8) +#define RTE_VHOST_USER_NET_STATS_ENABLE (1ULL << 9) /* Features. */ #ifndef VIRTIO_NET_F_GUEST_ANNOUNCE @@ -321,6 +322,32 @@ struct rte_vhost_power_monitor_cond { uint8_t match; }; +/** Maximum name length for the statistics counters */ +#define RTE_VHOST_STATS_NAME_SIZE 64 + +/** + * Vhost virtqueue statistics structure + * + * This structure is used by rte_vhost_vring_stats_get() to provide + * virtqueue statistics to the calling application. + * It maps a name ID, corresponding to an index in the array returned + * by rte_vhost_vring_stats_get_names(), to a statistic value. + */ +struct rte_vhost_stat { + uint64_t id; /**< The index in xstats name array. */ + uint64_t value; /**< The statistic counter value. */ +}; + +/** + * Vhost virtqueue statistic name element + * + * This structure is used by rte_vhost_vring_stats_get_names() to + * provide virtqueue statistics names to the calling application. + */ +struct rte_vhost_stat_name { + char name[RTE_VHOST_STATS_NAME_SIZE]; /**< The statistic name. */ +}; + /** * Convert guest physical address to host virtual address * @@ -1063,6 +1090,78 @@ __rte_experimental int rte_vhost_slave_config_change(int vid, bool need_reply); +/** + * Retrieve names of statistics of a Vhost virtqueue. + * + * There is an assumption that 'stat_names' and 'stats' arrays are matched + * by array index: stats_names[i].name => stats[i].value + * + * @param vid + * vhost device ID + * @param queue_id + * vhost queue index + * @param name + * array of at least size elements to be filled. + * If set to NULL, the function returns the required number of elements. + * @param size + * The number of elements in stats_names array. + * @return + * - Success if greater than 0 and lower or equal to *size*. The return value + * indicates the number of elements filled in the *names* array. + * - Failure if greater than *size*. The return value indicates the number of + * elements the *names* array that should be given to succeed. + * - Failure if lower than 0. The device ID or queue ID is invalid or + + statistics collection is not enabled. + */ +__rte_experimental +int +rte_vhost_vring_stats_get_names(int vid, uint16_t queue_id, + struct rte_vhost_stat_name *name, unsigned int size); + +/** + * Retrieve statistics of a Vhost virtqueue. + * + * There is an assumption that 'stat_names' and 'stats' arrays are matched + * by array index: stats_names[i].name => stats[i].value + * + * @param vid + * vhost device ID + * @param queue_id + * vhost queue index + * @param stats + * A pointer to a table of structure of type rte_vhost_stat to be filled with + * virtqueue statistics ids and values. + * @param n + * The number of elements in stats array. + * @return + * - Success if greater than 0 and lower or equal to *n*. The return value + * indicates the number of elements filled in the *stats* array. + * - Failure if greater than *n*. The return value indicates the number of + * elements the *stats* array that should be given to succeed. + * - Failure if lower than 0. The device ID or queue ID is invalid, or + * statistics collection is not enabled. + */ +__rte_experimental +int +rte_vhost_vring_stats_get(int vid, uint16_t queue_id, + struct rte_vhost_stat *stats, unsigned int n); + +/** + * Reset statistics of a Vhost virtqueue. + * + * @param vid + * vhost device ID + * @param queue_id + * vhost queue index + * @return + * - Success if 0. Statistics have been reset. + * - Failure if lower than 0. The device ID or queue ID is invalid, or + * statistics collection is not enabled. + */ +__rte_experimental +int +rte_vhost_vring_stats_reset(int vid, uint16_t queue_id); + #ifdef __cplusplus } #endif diff --git a/lib/vhost/socket.c b/lib/vhost/socket.c index b304339de9..baf6c338d3 100644 --- a/lib/vhost/socket.c +++ b/lib/vhost/socket.c @@ -42,6 +42,7 @@ struct vhost_user_socket { bool linearbuf; bool async_copy; bool net_compliant_ol_flags; + bool stats_enabled; /* * The "supported_features" indicates the feature bits the @@ -227,7 +228,7 @@ vhost_user_add_connection(int fd, struct vhost_user_socket *vsocket) vhost_set_ifname(vid, vsocket->path, size); vhost_setup_virtio_net(vid, vsocket->use_builtin_virtio_net, - vsocket->net_compliant_ol_flags); + vsocket->net_compliant_ol_flags, vsocket->stats_enabled); vhost_attach_vdpa_device(vid, vsocket->vdpa_dev); @@ -863,6 +864,7 @@ rte_vhost_driver_register(const char *path, uint64_t flags) vsocket->linearbuf = flags & RTE_VHOST_USER_LINEARBUF_SUPPORT; vsocket->async_copy = flags & RTE_VHOST_USER_ASYNC_COPY; vsocket->net_compliant_ol_flags = flags & RTE_VHOST_USER_NET_COMPLIANT_OL_FLAGS; + vsocket->stats_enabled = flags & RTE_VHOST_USER_NET_STATS_ENABLE; if (vsocket->async_copy && (flags & (RTE_VHOST_USER_IOMMU_SUPPORT | diff --git a/lib/vhost/version.map b/lib/vhost/version.map index 5841315386..34df41556d 100644 --- a/lib/vhost/version.map +++ b/lib/vhost/version.map @@ -90,7 +90,9 @@ EXPERIMENTAL { # added in 22.07 rte_vhost_async_get_inflight_thread_unsafe; - + rte_vhost_vring_stats_get_names; + rte_vhost_vring_stats_get; + rte_vhost_vring_stats_reset; }; INTERNAL { diff --git a/lib/vhost/vhost.c b/lib/vhost/vhost.c index df0bb9d043..b22f10de72 100644 --- a/lib/vhost/vhost.c +++ b/lib/vhost/vhost.c @@ -24,6 +24,28 @@ struct virtio_net *vhost_devices[RTE_MAX_VHOST_DEVICE]; pthread_mutex_t vhost_dev_lock = PTHREAD_MUTEX_INITIALIZER; +struct vhost_vq_stats_name_off { + char name[RTE_VHOST_STATS_NAME_SIZE]; + unsigned int offset; +}; + +static const struct vhost_vq_stats_name_off vhost_vq_stat_strings[] = { + {"good_packets", offsetof(struct vhost_virtqueue, stats.packets)}, + {"good_bytes", offsetof(struct vhost_virtqueue, stats.bytes)}, + {"multicast_packets", offsetof(struct vhost_virtqueue, stats.multicast)}, + {"broadcast_packets", offsetof(struct vhost_virtqueue, stats.broadcast)}, + {"undersize_packets", offsetof(struct vhost_virtqueue, stats.size_bins[0])}, + {"size_64_packets", offsetof(struct vhost_virtqueue, stats.size_bins[1])}, + {"size_65_127_packets", offsetof(struct vhost_virtqueue, stats.size_bins[2])}, + {"size_128_255_packets", offsetof(struct vhost_virtqueue, stats.size_bins[3])}, + {"size_256_511_packets", offsetof(struct vhost_virtqueue, stats.size_bins[4])}, + {"size_512_1023_packets", offsetof(struct vhost_virtqueue, stats.size_bins[5])}, + {"size_1024_1518_packets", offsetof(struct vhost_virtqueue, stats.size_bins[6])}, + {"size_1519_max_packets", offsetof(struct vhost_virtqueue, stats.size_bins[7])}, +}; + +#define VHOST_NB_VQ_STATS RTE_DIM(vhost_vq_stat_strings) + /* Called with iotlb_lock read-locked */ uint64_t __vhost_iova_to_vva(struct virtio_net *dev, struct vhost_virtqueue *vq, @@ -755,7 +777,7 @@ vhost_set_ifname(int vid, const char *if_name, unsigned int if_len) } void -vhost_setup_virtio_net(int vid, bool enable, bool compliant_ol_flags) +vhost_setup_virtio_net(int vid, bool enable, bool compliant_ol_flags, bool stats_enabled) { struct virtio_net *dev = get_device(vid); @@ -770,6 +792,10 @@ vhost_setup_virtio_net(int vid, bool enable, bool compliant_ol_flags) dev->flags |= VIRTIO_DEV_LEGACY_OL_FLAGS; else dev->flags &= ~VIRTIO_DEV_LEGACY_OL_FLAGS; + if (stats_enabled) + dev->flags |= VIRTIO_DEV_STATS_ENABLED; + else + dev->flags &= ~VIRTIO_DEV_STATS_ENABLED; } void @@ -1971,5 +1997,89 @@ rte_vhost_get_monitor_addr(int vid, uint16_t queue_id, return 0; } + +int +rte_vhost_vring_stats_get_names(int vid, uint16_t queue_id, + struct rte_vhost_stat_name *name, unsigned int size) +{ + struct virtio_net *dev = get_device(vid); + unsigned int i; + + if (dev == NULL) + return -1; + + if (queue_id >= dev->nr_vring) + return -1; + + if (!(dev->flags & VIRTIO_DEV_STATS_ENABLED)) + return -1; + + if (name == NULL || size < VHOST_NB_VQ_STATS) + return VHOST_NB_VQ_STATS; + + for (i = 0; i < VHOST_NB_VQ_STATS; i++) + snprintf(name[i].name, sizeof(name[i].name), "%s_q%u_%s", + (queue_id & 1) ? "rx" : "tx", + queue_id / 2, vhost_vq_stat_strings[i].name); + + return VHOST_NB_VQ_STATS; +} + +int +rte_vhost_vring_stats_get(int vid, uint16_t queue_id, + struct rte_vhost_stat *stats, unsigned int n) +{ + struct virtio_net *dev = get_device(vid); + struct vhost_virtqueue *vq; + unsigned int i; + + if (dev == NULL) + return -1; + + if (queue_id >= dev->nr_vring) + return -1; + + if (!(dev->flags & VIRTIO_DEV_STATS_ENABLED)) + return -1; + + if (stats == NULL || n < VHOST_NB_VQ_STATS) + return VHOST_NB_VQ_STATS; + + vq = dev->virtqueue[queue_id]; + + rte_spinlock_lock(&vq->access_lock); + for (i = 0; i < VHOST_NB_VQ_STATS; i++) { + stats[i].value = + *(uint64_t *)(((char *)vq) + vhost_vq_stat_strings[i].offset); + stats[i].id = i; + } + rte_spinlock_unlock(&vq->access_lock); + + return VHOST_NB_VQ_STATS; +} + +int rte_vhost_vring_stats_reset(int vid, uint16_t queue_id) +{ + struct virtio_net *dev = get_device(vid); + struct vhost_virtqueue *vq; + + if (dev == NULL) + return -1; + + if (queue_id >= dev->nr_vring) + return -1; + + if (!(dev->flags & VIRTIO_DEV_STATS_ENABLED)) + return -1; + + vq = dev->virtqueue[queue_id]; + + rte_spinlock_lock(&vq->access_lock); + memset(&vq->stats, 0, sizeof(vq->stats)); + rte_spinlock_unlock(&vq->access_lock); + + return 0; +} + RTE_LOG_REGISTER_SUFFIX(vhost_config_log_level, config, INFO); RTE_LOG_REGISTER_SUFFIX(vhost_data_log_level, data, WARNING); diff --git a/lib/vhost/vhost.h b/lib/vhost/vhost.h index a9edc271aa..01b97011aa 100644 --- a/lib/vhost/vhost.h +++ b/lib/vhost/vhost.h @@ -37,6 +37,8 @@ #define VIRTIO_DEV_FEATURES_FAILED ((uint32_t)1 << 4) /* Used to indicate that the virtio_net tx code should fill TX ol_flags */ #define VIRTIO_DEV_LEGACY_OL_FLAGS ((uint32_t)1 << 5) +/* Used to indicate the application has requested statistics collection */ +#define VIRTIO_DEV_STATS_ENABLED ((uint32_t)1 << 6) /* Backend value set by guest. */ #define VIRTIO_DEV_STOPPED -1 @@ -121,6 +123,18 @@ struct vring_used_elem_packed { uint32_t count; }; +/** + * Virtqueue statistics + */ +struct virtqueue_stats { + uint64_t packets; + uint64_t bytes; + uint64_t multicast; + uint64_t broadcast; + /* Size bins in array as RFC 2819, undersized [0], 64 [1], etc */ + uint64_t size_bins[8]; +}; + /** * iovec */ @@ -305,6 +319,7 @@ struct vhost_virtqueue { #define VIRTIO_UNINITIALIZED_NOTIF (-1) struct vhost_vring_addr ring_addrs; + struct virtqueue_stats stats; } __rte_cache_aligned; /* Virtio device status as per Virtio specification */ @@ -780,7 +795,7 @@ int alloc_vring_queue(struct virtio_net *dev, uint32_t vring_idx); void vhost_attach_vdpa_device(int vid, struct rte_vdpa_device *dev); void vhost_set_ifname(int, const char *if_name, unsigned int if_len); -void vhost_setup_virtio_net(int vid, bool enable, bool legacy_ol_flags); +void vhost_setup_virtio_net(int vid, bool enable, bool legacy_ol_flags, bool stats_enabled); void vhost_enable_extbuf(int vid); void vhost_enable_linearbuf(int vid); int vhost_enable_guest_notification(struct virtio_net *dev, @@ -957,5 +972,4 @@ mbuf_is_consumed(struct rte_mbuf *m) return true; } - #endif /* _VHOST_NET_CDEV_H_ */ diff --git a/lib/vhost/virtio_net.c b/lib/vhost/virtio_net.c index 5f432b0d77..b1ea9fa4a5 100644 --- a/lib/vhost/virtio_net.c +++ b/lib/vhost/virtio_net.c @@ -47,6 +47,54 @@ is_valid_virt_queue_idx(uint32_t idx, int is_tx, uint32_t nr_vring) return (is_tx ^ (idx & 1)) == 0 && idx < nr_vring; } +/* + * This function must be called with virtqueue's access_lock taken. + */ +static inline void +vhost_queue_stats_update(struct virtio_net *dev, struct vhost_virtqueue *vq, + struct rte_mbuf **pkts, uint16_t count) +{ + struct virtqueue_stats *stats = &vq->stats; + int i; + + if (!(dev->flags & VIRTIO_DEV_STATS_ENABLED)) + return; + + for (i = 0; i < count; i++) { + struct rte_ether_addr *ea; + struct rte_mbuf *pkt = pkts[i]; + uint32_t pkt_len = rte_pktmbuf_pkt_len(pkt); + + stats->packets++; + stats->bytes += pkt_len; + + if (pkt_len == 64) { + stats->size_bins[1]++; + } else if (pkt_len > 64 && pkt_len < 1024) { + uint32_t bin; + + /* count zeros, and offset into correct bin */ + bin = (sizeof(pkt_len) * 8) - __builtin_clz(pkt_len) - 5; + stats->size_bins[bin]++; + } else { + if (pkt_len < 64) + stats->size_bins[0]++; + else if (pkt_len < 1519) + stats->size_bins[6]++; + else + stats->size_bins[7]++; + } + + ea = rte_pktmbuf_mtod(pkt, struct rte_ether_addr *); + if (rte_is_multicast_ether_addr(ea)) { + if (rte_is_broadcast_ether_addr(ea)) + stats->broadcast++; + else + stats->multicast++; + } + } +} + static __rte_always_inline int64_t vhost_async_dma_transfer_one(struct virtio_net *dev, struct vhost_virtqueue *vq, int16_t dma_id, uint16_t vchan_id, uint16_t flag_idx, @@ -1509,6 +1557,8 @@ virtio_dev_rx(struct virtio_net *dev, uint16_t queue_id, else nb_tx = virtio_dev_rx_split(dev, vq, pkts, count); + vhost_queue_stats_update(dev, vq, pkts, nb_tx); + out: if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM)) vhost_user_iotlb_rd_unlock(vq); @@ -2064,6 +2114,8 @@ rte_vhost_poll_enqueue_completed(int vid, uint16_t queue_id, n_pkts_cpl = vhost_poll_enqueue_completed(dev, queue_id, pkts, count, dma_id, vchan_id); + vhost_queue_stats_update(dev, vq, pkts, n_pkts_cpl); + out: rte_spinlock_unlock(&vq->access_lock); @@ -3113,6 +3165,7 @@ rte_vhost_dequeue_burst(int vid, uint16_t queue_id, * learning table will get updated first. */ pkts[0] = rarp_mbuf; + vhost_queue_stats_update(dev, vq, pkts, 1); pkts++; count -= 1; } @@ -3129,6 +3182,8 @@ rte_vhost_dequeue_burst(int vid, uint16_t queue_id, count = virtio_dev_tx_split_compliant(dev, vq, mbuf_pool, pkts, count); } + vhost_queue_stats_update(dev, vq, pkts, count); + out: if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM)) vhost_user_iotlb_rd_unlock(vq); -- 2.35.1