diff options
Diffstat (limited to 'drivers/net/xen-netfront.c')
| -rw-r--r-- | drivers/net/xen-netfront.c | 2638 |
1 files changed, 1613 insertions, 1025 deletions
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index ff7f111fffee..7c2220366623 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -44,8 +44,10 @@ #include <linux/mm.h> #include <linux/slab.h> #include <net/ip.h> +#include <linux/bpf.h> +#include <net/page_pool/types.h> +#include <linux/bpf_trace.h> -#include <asm/xen/page.h> #include <xen/xen.h> #include <xen/xenbus.h> #include <xen/events.h> @@ -57,6 +59,19 @@ #include <xen/interface/memory.h> #include <xen/interface/grant_table.h> +/* Module parameters */ +#define MAX_QUEUES_DEFAULT 8 +static unsigned int xennet_max_queues; +module_param_named(max_queues, xennet_max_queues, uint, 0644); +MODULE_PARM_DESC(max_queues, + "Maximum number of queues per virtual interface"); + +static bool __read_mostly xennet_trusted = true; +module_param_named(trusted, xennet_trusted, bool, 0644); +MODULE_PARM_DESC(trusted, "Is the backend trusted"); + +#define XENNET_TIMEOUT (5 * HZ) + static const struct ethtool_ops xennet_ethtool_ops; struct netfront_cb { @@ -67,23 +82,34 @@ struct netfront_cb { #define RX_COPY_THRESHOLD 256 -#define GRANT_INVALID_REF 0 +#define NET_TX_RING_SIZE __CONST_RING_SIZE(xen_netif_tx, XEN_PAGE_SIZE) +#define NET_RX_RING_SIZE __CONST_RING_SIZE(xen_netif_rx, XEN_PAGE_SIZE) + +/* Minimum number of Rx slots (includes slot for GSO metadata). */ +#define NET_RX_SLOTS_MIN (XEN_NETIF_NR_SLOTS_MIN + 1) + +/* Queue name is interface name with "-qNNN" appended */ +#define QUEUE_NAME_SIZE (IFNAMSIZ + 6) + +/* IRQ name is queue name with "-tx" or "-rx" appended */ +#define IRQ_NAME_SIZE (QUEUE_NAME_SIZE + 3) -#define NET_TX_RING_SIZE __CONST_RING_SIZE(xen_netif_tx, PAGE_SIZE) -#define NET_RX_RING_SIZE __CONST_RING_SIZE(xen_netif_rx, PAGE_SIZE) -#define TX_MAX_TARGET min_t(int, NET_TX_RING_SIZE, 256) +static DECLARE_WAIT_QUEUE_HEAD(module_wq); struct netfront_stats { - u64 rx_packets; - u64 tx_packets; - u64 rx_bytes; - u64 tx_bytes; + u64 packets; + u64 bytes; struct u64_stats_sync syncp; }; -struct netfront_info { - struct list_head list; - struct net_device *netdev; +struct netfront_info; + +struct netfront_queue { + unsigned int id; /* Queue ID, 0-based */ + char name[QUEUE_NAME_SIZE]; /* DEVNAME-qN */ + struct netfront_info *info; + + struct bpf_prog __rcu *xdp_prog; struct napi_struct napi; @@ -93,10 +119,8 @@ struct netfront_info { unsigned int tx_evtchn, rx_evtchn; unsigned int tx_irq, rx_irq; /* Only used when split event channels support is enabled */ - char tx_irq_name[IFNAMSIZ+4]; /* DEVNAME-tx */ - char rx_irq_name[IFNAMSIZ+4]; /* DEVNAME-rx */ - - struct xenbus_device *xbdev; + char tx_irq_name[IRQ_NAME_SIZE]; /* DEVNAME-qN-tx */ + char rx_irq_name[IRQ_NAME_SIZE]; /* DEVNAME-qN-rx */ spinlock_t tx_lock; struct xen_netif_tx_front_ring tx; @@ -104,46 +128,59 @@ struct netfront_info { /* * {tx,rx}_skbs store outstanding skbuffs. Free tx_skb entries - * are linked from tx_skb_freelist through skb_entry.link. - * - * NB. Freelist index entries are always going to be less than - * PAGE_OFFSET, whereas pointers to skbs will always be equal or - * greater than PAGE_OFFSET: we use this property to distinguish - * them. + * are linked from tx_skb_freelist through tx_link. */ - union skb_entry { - struct sk_buff *skb; - unsigned long link; - } tx_skbs[NET_TX_RING_SIZE]; + struct sk_buff *tx_skbs[NET_TX_RING_SIZE]; + unsigned short tx_link[NET_TX_RING_SIZE]; +#define TX_LINK_NONE 0xffff +#define TX_PENDING 0xfffe grant_ref_t gref_tx_head; grant_ref_t grant_tx_ref[NET_TX_RING_SIZE]; + struct page *grant_tx_page[NET_TX_RING_SIZE]; unsigned tx_skb_freelist; + unsigned int tx_pend_queue; spinlock_t rx_lock ____cacheline_aligned_in_smp; struct xen_netif_rx_front_ring rx; int rx_ring_ref; - /* Receive-ring batched refills. */ -#define RX_MIN_TARGET 8 -#define RX_DFL_MIN_TARGET 64 -#define RX_MAX_TARGET min_t(int, NET_RX_RING_SIZE, 256) - unsigned rx_min_target, rx_max_target, rx_target; - struct sk_buff_head rx_batch; - struct timer_list rx_refill_timer; struct sk_buff *rx_skbs[NET_RX_RING_SIZE]; grant_ref_t gref_rx_head; grant_ref_t grant_rx_ref[NET_RX_RING_SIZE]; - unsigned long rx_pfn_array[NET_RX_RING_SIZE]; - struct multicall_entry rx_mcl[NET_RX_RING_SIZE+1]; - struct mmu_update rx_mmu[NET_RX_RING_SIZE]; + unsigned int rx_rsp_unconsumed; + spinlock_t rx_cons_lock; + + struct page_pool *page_pool; + struct xdp_rxq_info xdp_rxq; +}; + +struct netfront_info { + struct list_head list; + struct net_device *netdev; + + struct xenbus_device *xbdev; + + /* Multi-queue support */ + struct netfront_queue *queues; /* Statistics */ - struct netfront_stats __percpu *stats; + struct netfront_stats __percpu *rx_stats; + struct netfront_stats __percpu *tx_stats; + + /* XDP state */ + bool netback_has_xdp_headroom; + bool netfront_xdp_enabled; + + /* Is device behaving sane? */ + bool broken; - unsigned long rx_gso_checksum_fixup; + /* Should skbs be bounced into a zeroed buffer? */ + bool bounce; + + atomic_t rx_gso_checksum_fixup; }; struct netfront_rx_info { @@ -151,33 +188,25 @@ struct netfront_rx_info { struct xen_netif_extra_info extras[XEN_NETIF_EXTRA_TYPE_MAX - 1]; }; -static void skb_entry_set_link(union skb_entry *list, unsigned short id) -{ - list->link = id; -} - -static int skb_entry_is_link(const union skb_entry *list) -{ - BUILD_BUG_ON(sizeof(list->skb) != sizeof(list->link)); - return (unsigned long)list->skb < PAGE_OFFSET; -} - /* * Access macros for acquiring freeing slots in tx_skbs[]. */ -static void add_id_to_freelist(unsigned *head, union skb_entry *list, - unsigned short id) +static void add_id_to_list(unsigned *head, unsigned short *list, + unsigned short id) { - skb_entry_set_link(&list[id], *head); + list[id] = *head; *head = id; } -static unsigned short get_id_from_freelist(unsigned *head, - union skb_entry *list) +static unsigned short get_id_from_list(unsigned *head, unsigned short *list) { unsigned int id = *head; - *head = list[id].link; + + if (id != TX_LINK_NONE) { + *head = list[id]; + list[id] = TX_LINK_NONE; + } return id; } @@ -186,30 +215,26 @@ static int xennet_rxidx(RING_IDX idx) return idx & (NET_RX_RING_SIZE - 1); } -static struct sk_buff *xennet_get_rx_skb(struct netfront_info *np, +static struct sk_buff *xennet_get_rx_skb(struct netfront_queue *queue, RING_IDX ri) { int i = xennet_rxidx(ri); - struct sk_buff *skb = np->rx_skbs[i]; - np->rx_skbs[i] = NULL; + struct sk_buff *skb = queue->rx_skbs[i]; + queue->rx_skbs[i] = NULL; return skb; } -static grant_ref_t xennet_get_rx_ref(struct netfront_info *np, +static grant_ref_t xennet_get_rx_ref(struct netfront_queue *queue, RING_IDX ri) { int i = xennet_rxidx(ri); - grant_ref_t ref = np->grant_rx_ref[i]; - np->grant_rx_ref[i] = GRANT_INVALID_REF; + grant_ref_t ref = queue->grant_rx_ref[i]; + queue->grant_rx_ref[i] = INVALID_GRANT_REF; return ref; } #ifdef CONFIG_SYSFS -static int xennet_sysfs_addif(struct net_device *netdev); -static void xennet_sysfs_delif(struct net_device *netdev); -#else /* !CONFIG_SYSFS */ -#define xennet_sysfs_addif(dev) (0) -#define xennet_sysfs_delif(dev) do { } while (0) +static const struct attribute_group xennet_dev_group; #endif static bool xennet_can_sg(struct net_device *dev) @@ -218,343 +243,492 @@ static bool xennet_can_sg(struct net_device *dev) } -static void rx_refill_timeout(unsigned long data) +static void rx_refill_timeout(struct timer_list *t) { - struct net_device *dev = (struct net_device *)data; - struct netfront_info *np = netdev_priv(dev); - napi_schedule(&np->napi); + struct netfront_queue *queue = timer_container_of(queue, t, + rx_refill_timer); + napi_schedule(&queue->napi); } -static int netfront_tx_slot_available(struct netfront_info *np) +static int netfront_tx_slot_available(struct netfront_queue *queue) { - return (np->tx.req_prod_pvt - np->tx.rsp_cons) < - (TX_MAX_TARGET - MAX_SKB_FRAGS - 2); + return (queue->tx.req_prod_pvt - queue->tx.rsp_cons) < + (NET_TX_RING_SIZE - XEN_NETIF_NR_SLOTS_MIN - 1); } -static void xennet_maybe_wake_tx(struct net_device *dev) +static void xennet_maybe_wake_tx(struct netfront_queue *queue) { - struct netfront_info *np = netdev_priv(dev); + struct net_device *dev = queue->info->netdev; + struct netdev_queue *dev_queue = netdev_get_tx_queue(dev, queue->id); - if (unlikely(netif_queue_stopped(dev)) && - netfront_tx_slot_available(np) && + if (unlikely(netif_tx_queue_stopped(dev_queue)) && + netfront_tx_slot_available(queue) && likely(netif_running(dev))) - netif_wake_queue(dev); + netif_tx_wake_queue(netdev_get_tx_queue(dev, queue->id)); } -static void xennet_alloc_rx_buffers(struct net_device *dev) + +static struct sk_buff *xennet_alloc_one_rx_buffer(struct netfront_queue *queue) { - unsigned short id; - struct netfront_info *np = netdev_priv(dev); struct sk_buff *skb; struct page *page; - int i, batch_target, notify; - RING_IDX req_prod = np->rx.req_prod_pvt; - grant_ref_t ref; - unsigned long pfn; - void *vaddr; - struct xen_netif_rx_request *req; - if (unlikely(!netif_carrier_ok(dev))) - return; + skb = __netdev_alloc_skb(queue->info->netdev, + RX_COPY_THRESHOLD + NET_IP_ALIGN, + GFP_ATOMIC | __GFP_NOWARN); + if (unlikely(!skb)) + return NULL; + + page = page_pool_alloc_pages(queue->page_pool, + GFP_ATOMIC | __GFP_NOWARN | __GFP_ZERO); + if (unlikely(!page)) { + kfree_skb(skb); + return NULL; + } + skb_add_rx_frag(skb, 0, page, 0, 0, PAGE_SIZE); + skb_mark_for_recycle(skb); - /* - * Allocate skbuffs greedily, even though we batch updates to the - * receive ring. This creates a less bursty demand on the memory - * allocator, so should reduce the chance of failed allocation requests - * both for ourself and for other kernel subsystems. - */ - batch_target = np->rx_target - (req_prod - np->rx.rsp_cons); - for (i = skb_queue_len(&np->rx_batch); i < batch_target; i++) { - skb = __netdev_alloc_skb(dev, RX_COPY_THRESHOLD + NET_IP_ALIGN, - GFP_ATOMIC | __GFP_NOWARN); - if (unlikely(!skb)) - goto no_skb; - - /* Align ip header to a 16 bytes boundary */ - skb_reserve(skb, NET_IP_ALIGN); - - page = alloc_page(GFP_ATOMIC | __GFP_NOWARN); - if (!page) { - kfree_skb(skb); -no_skb: - /* Any skbuffs queued for refill? Force them out. */ - if (i != 0) - goto refill; - /* Could not allocate any skbuffs. Try again later. */ - mod_timer(&np->rx_refill_timer, - jiffies + (HZ/10)); - break; - } + /* Align ip header to a 16 bytes boundary */ + skb_reserve(skb, NET_IP_ALIGN); + skb->dev = queue->info->netdev; - __skb_fill_page_desc(skb, 0, page, 0, 0); - skb_shinfo(skb)->nr_frags = 1; - __skb_queue_tail(&np->rx_batch, skb); - } + return skb; +} - /* Is the batch large enough to be worthwhile? */ - if (i < (np->rx_target/2)) { - if (req_prod > np->rx.sring->req_prod) - goto push; - return; - } - /* Adjust our fill target if we risked running out of buffers. */ - if (((req_prod - np->rx.sring->rsp_prod) < (np->rx_target / 4)) && - ((np->rx_target *= 2) > np->rx_max_target)) - np->rx_target = np->rx_max_target; +static void xennet_alloc_rx_buffers(struct netfront_queue *queue) +{ + RING_IDX req_prod = queue->rx.req_prod_pvt; + int notify; + int err = 0; - refill: - for (i = 0; ; i++) { - skb = __skb_dequeue(&np->rx_batch); - if (skb == NULL) - break; + if (unlikely(!netif_carrier_ok(queue->info->netdev))) + return; - skb->dev = dev; + for (req_prod = queue->rx.req_prod_pvt; + req_prod - queue->rx.rsp_cons < NET_RX_RING_SIZE; + req_prod++) { + struct sk_buff *skb; + unsigned short id; + grant_ref_t ref; + struct page *page; + struct xen_netif_rx_request *req; - id = xennet_rxidx(req_prod + i); + skb = xennet_alloc_one_rx_buffer(queue); + if (!skb) { + err = -ENOMEM; + break; + } - BUG_ON(np->rx_skbs[id]); - np->rx_skbs[id] = skb; + id = xennet_rxidx(req_prod); - ref = gnttab_claim_grant_reference(&np->gref_rx_head); - BUG_ON((signed short)ref < 0); - np->grant_rx_ref[id] = ref; + BUG_ON(queue->rx_skbs[id]); + queue->rx_skbs[id] = skb; - pfn = page_to_pfn(skb_frag_page(&skb_shinfo(skb)->frags[0])); - vaddr = page_address(skb_frag_page(&skb_shinfo(skb)->frags[0])); + ref = gnttab_claim_grant_reference(&queue->gref_rx_head); + WARN_ON_ONCE(IS_ERR_VALUE((unsigned long)(int)ref)); + queue->grant_rx_ref[id] = ref; - req = RING_GET_REQUEST(&np->rx, req_prod + i); - gnttab_grant_foreign_access_ref(ref, - np->xbdev->otherend_id, - pfn_to_mfn(pfn), - 0); + page = skb_frag_page(&skb_shinfo(skb)->frags[0]); + req = RING_GET_REQUEST(&queue->rx, req_prod); + gnttab_page_grant_foreign_access_ref_one(ref, + queue->info->xbdev->otherend_id, + page, + 0); req->id = id; req->gref = ref; } - wmb(); /* barrier so backend seens requests */ + queue->rx.req_prod_pvt = req_prod; - /* Above is a suitable barrier to ensure backend will see requests. */ - np->rx.req_prod_pvt = req_prod + i; - push: - RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&np->rx, notify); + /* Try again later if there are not enough requests or skb allocation + * failed. + * Enough requests is quantified as the sum of newly created slots and + * the unconsumed slots at the backend. + */ + if (req_prod - queue->rx.rsp_cons < NET_RX_SLOTS_MIN || + unlikely(err)) { + mod_timer(&queue->rx_refill_timer, jiffies + (HZ/10)); + return; + } + + RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&queue->rx, notify); if (notify) - notify_remote_via_irq(np->rx_irq); + notify_remote_via_irq(queue->rx_irq); } static int xennet_open(struct net_device *dev) { struct netfront_info *np = netdev_priv(dev); + unsigned int num_queues = dev->real_num_tx_queues; + unsigned int i = 0; + struct netfront_queue *queue = NULL; + + if (!np->queues || np->broken) + return -ENODEV; - napi_enable(&np->napi); + for (i = 0; i < num_queues; ++i) { + queue = &np->queues[i]; + napi_enable(&queue->napi); - spin_lock_bh(&np->rx_lock); - if (netif_carrier_ok(dev)) { - xennet_alloc_rx_buffers(dev); - np->rx.sring->rsp_event = np->rx.rsp_cons + 1; - if (RING_HAS_UNCONSUMED_RESPONSES(&np->rx)) - napi_schedule(&np->napi); + spin_lock_bh(&queue->rx_lock); + if (netif_carrier_ok(dev)) { + xennet_alloc_rx_buffers(queue); + queue->rx.sring->rsp_event = queue->rx.rsp_cons + 1; + if (RING_HAS_UNCONSUMED_RESPONSES(&queue->rx)) + napi_schedule(&queue->napi); + } + spin_unlock_bh(&queue->rx_lock); } - spin_unlock_bh(&np->rx_lock); - netif_start_queue(dev); + netif_tx_start_all_queues(dev); return 0; } -static void xennet_tx_buf_gc(struct net_device *dev) +static bool xennet_tx_buf_gc(struct netfront_queue *queue) { RING_IDX cons, prod; unsigned short id; - struct netfront_info *np = netdev_priv(dev); struct sk_buff *skb; + bool more_to_do; + bool work_done = false; + const struct device *dev = &queue->info->netdev->dev; - BUG_ON(!netif_carrier_ok(dev)); + BUG_ON(!netif_carrier_ok(queue->info->netdev)); do { - prod = np->tx.sring->rsp_prod; + prod = queue->tx.sring->rsp_prod; + if (RING_RESPONSE_PROD_OVERFLOW(&queue->tx, prod)) { + dev_alert(dev, "Illegal number of responses %u\n", + prod - queue->tx.rsp_cons); + goto err; + } rmb(); /* Ensure we see responses up to 'rp'. */ - for (cons = np->tx.rsp_cons; cons != prod; cons++) { - struct xen_netif_tx_response *txrsp; + for (cons = queue->tx.rsp_cons; cons != prod; cons++) { + struct xen_netif_tx_response txrsp; + + work_done = true; - txrsp = RING_GET_RESPONSE(&np->tx, cons); - if (txrsp->status == XEN_NETIF_RSP_NULL) + RING_COPY_RESPONSE(&queue->tx, cons, &txrsp); + if (txrsp.status == XEN_NETIF_RSP_NULL) continue; - id = txrsp->id; - skb = np->tx_skbs[id].skb; - if (unlikely(gnttab_query_foreign_access( - np->grant_tx_ref[id]) != 0)) { - pr_alert("%s: warning -- grant still in use by backend domain\n", - __func__); - BUG(); + id = txrsp.id; + if (id >= RING_SIZE(&queue->tx)) { + dev_alert(dev, + "Response has incorrect id (%u)\n", + id); + goto err; + } + if (queue->tx_link[id] != TX_PENDING) { + dev_alert(dev, + "Response for inactive request\n"); + goto err; + } + + queue->tx_link[id] = TX_LINK_NONE; + skb = queue->tx_skbs[id]; + queue->tx_skbs[id] = NULL; + if (unlikely(!gnttab_end_foreign_access_ref( + queue->grant_tx_ref[id]))) { + dev_alert(dev, + "Grant still in use by backend domain\n"); + goto err; } - gnttab_end_foreign_access_ref( - np->grant_tx_ref[id], GNTMAP_readonly); gnttab_release_grant_reference( - &np->gref_tx_head, np->grant_tx_ref[id]); - np->grant_tx_ref[id] = GRANT_INVALID_REF; - add_id_to_freelist(&np->tx_skb_freelist, np->tx_skbs, id); + &queue->gref_tx_head, queue->grant_tx_ref[id]); + queue->grant_tx_ref[id] = INVALID_GRANT_REF; + queue->grant_tx_page[id] = NULL; + add_id_to_list(&queue->tx_skb_freelist, queue->tx_link, id); dev_kfree_skb_irq(skb); } - np->tx.rsp_cons = prod; + queue->tx.rsp_cons = prod; - /* - * Set a new event, then check for race with update of tx_cons. - * Note that it is essential to schedule a callback, no matter - * how few buffers are pending. Even if there is space in the - * transmit ring, higher layers may be blocked because too much - * data is outstanding: in such cases notification from Xen is - * likely to be the only kick that we'll get. - */ - np->tx.sring->rsp_event = - prod + ((np->tx.sring->req_prod - prod) >> 1) + 1; - mb(); /* update shared area */ - } while ((cons == prod) && (prod != np->tx.sring->rsp_prod)); + RING_FINAL_CHECK_FOR_RESPONSES(&queue->tx, more_to_do); + } while (more_to_do); - xennet_maybe_wake_tx(dev); + xennet_maybe_wake_tx(queue); + + return work_done; + + err: + queue->info->broken = true; + dev_alert(dev, "Disabled for further use\n"); + + return work_done; } -static void xennet_make_frags(struct sk_buff *skb, struct net_device *dev, - struct xen_netif_tx_request *tx) +struct xennet_gnttab_make_txreq { + struct netfront_queue *queue; + struct sk_buff *skb; + struct page *page; + struct xen_netif_tx_request *tx; /* Last request on ring page */ + struct xen_netif_tx_request tx_local; /* Last request local copy*/ + unsigned int size; +}; + +static void xennet_tx_setup_grant(unsigned long gfn, unsigned int offset, + unsigned int len, void *data) { - struct netfront_info *np = netdev_priv(dev); - char *data = skb->data; - unsigned long mfn; - RING_IDX prod = np->tx.req_prod_pvt; - int frags = skb_shinfo(skb)->nr_frags; - unsigned int offset = offset_in_page(data); - unsigned int len = skb_headlen(skb); + struct xennet_gnttab_make_txreq *info = data; unsigned int id; + struct xen_netif_tx_request *tx; grant_ref_t ref; - int i; + /* convenient aliases */ + struct page *page = info->page; + struct netfront_queue *queue = info->queue; + struct sk_buff *skb = info->skb; - /* While the header overlaps a page boundary (including being - larger than a page), split it it into page-sized chunks. */ - while (len > PAGE_SIZE - offset) { - tx->size = PAGE_SIZE - offset; - tx->flags |= XEN_NETTXF_more_data; - len -= tx->size; - data += tx->size; - offset = 0; + id = get_id_from_list(&queue->tx_skb_freelist, queue->tx_link); + tx = RING_GET_REQUEST(&queue->tx, queue->tx.req_prod_pvt++); + ref = gnttab_claim_grant_reference(&queue->gref_tx_head); + WARN_ON_ONCE(IS_ERR_VALUE((unsigned long)(int)ref)); - id = get_id_from_freelist(&np->tx_skb_freelist, np->tx_skbs); - np->tx_skbs[id].skb = skb_get(skb); - tx = RING_GET_REQUEST(&np->tx, prod++); - tx->id = id; - ref = gnttab_claim_grant_reference(&np->gref_tx_head); - BUG_ON((signed short)ref < 0); + gnttab_grant_foreign_access_ref(ref, queue->info->xbdev->otherend_id, + gfn, GNTMAP_readonly); - mfn = virt_to_mfn(data); - gnttab_grant_foreign_access_ref(ref, np->xbdev->otherend_id, - mfn, GNTMAP_readonly); + queue->tx_skbs[id] = skb; + queue->grant_tx_page[id] = page; + queue->grant_tx_ref[id] = ref; - tx->gref = np->grant_tx_ref[id] = ref; - tx->offset = offset; - tx->size = len; - tx->flags = 0; - } - - /* Grant backend access to each skb fragment page. */ - for (i = 0; i < frags; i++) { - skb_frag_t *frag = skb_shinfo(skb)->frags + i; - struct page *page = skb_frag_page(frag); + info->tx_local.id = id; + info->tx_local.gref = ref; + info->tx_local.offset = offset; + info->tx_local.size = len; + info->tx_local.flags = 0; - len = skb_frag_size(frag); - offset = frag->page_offset; + *tx = info->tx_local; - /* Data must not cross a page boundary. */ - BUG_ON(len + offset > PAGE_SIZE<<compound_order(page)); + /* + * Put the request in the pending queue, it will be set to be pending + * when the producer index is about to be raised. + */ + add_id_to_list(&queue->tx_pend_queue, queue->tx_link, id); - /* Skip unused frames from start of page */ - page += offset >> PAGE_SHIFT; - offset &= ~PAGE_MASK; + info->tx = tx; + info->size += info->tx_local.size; +} - while (len > 0) { - unsigned long bytes; +static struct xen_netif_tx_request *xennet_make_first_txreq( + struct xennet_gnttab_make_txreq *info, + unsigned int offset, unsigned int len) +{ + info->size = 0; - BUG_ON(offset >= PAGE_SIZE); + gnttab_for_one_grant(info->page, offset, len, xennet_tx_setup_grant, info); - bytes = PAGE_SIZE - offset; - if (bytes > len) - bytes = len; + return info->tx; +} - tx->flags |= XEN_NETTXF_more_data; +static void xennet_make_one_txreq(unsigned long gfn, unsigned int offset, + unsigned int len, void *data) +{ + struct xennet_gnttab_make_txreq *info = data; - id = get_id_from_freelist(&np->tx_skb_freelist, - np->tx_skbs); - np->tx_skbs[id].skb = skb_get(skb); - tx = RING_GET_REQUEST(&np->tx, prod++); - tx->id = id; - ref = gnttab_claim_grant_reference(&np->gref_tx_head); - BUG_ON((signed short)ref < 0); + info->tx->flags |= XEN_NETTXF_more_data; + skb_get(info->skb); + xennet_tx_setup_grant(gfn, offset, len, data); +} - mfn = pfn_to_mfn(page_to_pfn(page)); - gnttab_grant_foreign_access_ref(ref, - np->xbdev->otherend_id, - mfn, GNTMAP_readonly); +static void xennet_make_txreqs( + struct xennet_gnttab_make_txreq *info, + struct page *page, + unsigned int offset, unsigned int len) +{ + /* Skip unused frames from start of page */ + page += offset >> PAGE_SHIFT; + offset &= ~PAGE_MASK; - tx->gref = np->grant_tx_ref[id] = ref; - tx->offset = offset; - tx->size = bytes; - tx->flags = 0; + while (len) { + info->page = page; + info->size = 0; - offset += bytes; - len -= bytes; + gnttab_foreach_grant_in_range(page, offset, len, + xennet_make_one_txreq, + info); - /* Next frame */ - if (offset == PAGE_SIZE && len) { - BUG_ON(!PageCompound(page)); - page++; - offset = 0; - } - } + page++; + offset = 0; + len -= info->size; } - - np->tx.req_prod_pvt = prod; } /* - * Count how many ring slots are required to send the frags of this - * skb. Each frag might be a compound page. + * Count how many ring slots are required to send this skb. Each frag + * might be a compound page. */ -static int xennet_count_skb_frag_slots(struct sk_buff *skb) +static int xennet_count_skb_slots(struct sk_buff *skb) { int i, frags = skb_shinfo(skb)->nr_frags; - int pages = 0; + int slots; + + slots = gnttab_count_grant(offset_in_page(skb->data), + skb_headlen(skb)); for (i = 0; i < frags; i++) { skb_frag_t *frag = skb_shinfo(skb)->frags + i; unsigned long size = skb_frag_size(frag); - unsigned long offset = frag->page_offset; + unsigned long offset = skb_frag_off(frag); /* Skip unused frames from start of page */ offset &= ~PAGE_MASK; - pages += PFN_UP(offset + size); + slots += gnttab_count_grant(offset, size); } - return pages; + return slots; } -static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev) +static u16 xennet_select_queue(struct net_device *dev, struct sk_buff *skb, + struct net_device *sb_dev) +{ + unsigned int num_queues = dev->real_num_tx_queues; + u32 hash; + u16 queue_idx; + + /* First, check if there is only one queue */ + if (num_queues == 1) { + queue_idx = 0; + } else { + hash = skb_get_hash(skb); + queue_idx = hash % num_queues; + } + + return queue_idx; +} + +static void xennet_mark_tx_pending(struct netfront_queue *queue) +{ + unsigned int i; + + while ((i = get_id_from_list(&queue->tx_pend_queue, queue->tx_link)) != + TX_LINK_NONE) + queue->tx_link[i] = TX_PENDING; +} + +static int xennet_xdp_xmit_one(struct net_device *dev, + struct netfront_queue *queue, + struct xdp_frame *xdpf) { - unsigned short id; struct netfront_info *np = netdev_priv(dev); - struct netfront_stats *stats = this_cpu_ptr(np->stats); - struct xen_netif_tx_request *tx; - char *data = skb->data; - RING_IDX i; - grant_ref_t ref; - unsigned long mfn; + struct netfront_stats *tx_stats = this_cpu_ptr(np->tx_stats); + struct xennet_gnttab_make_txreq info = { + .queue = queue, + .skb = NULL, + .page = virt_to_page(xdpf->data), + }; + int notify; + + xennet_make_first_txreq(&info, + offset_in_page(xdpf->data), + xdpf->len); + + xennet_mark_tx_pending(queue); + + RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&queue->tx, notify); + if (notify) + notify_remote_via_irq(queue->tx_irq); + + u64_stats_update_begin(&tx_stats->syncp); + tx_stats->bytes += xdpf->len; + tx_stats->packets++; + u64_stats_update_end(&tx_stats->syncp); + + return 0; +} + +static int xennet_xdp_xmit(struct net_device *dev, int n, + struct xdp_frame **frames, u32 flags) +{ + unsigned int num_queues = dev->real_num_tx_queues; + struct netfront_info *np = netdev_priv(dev); + struct netfront_queue *queue = NULL; + unsigned long irq_flags; + int nxmit = 0; + int i; + + if (unlikely(np->broken)) + return -ENODEV; + if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK)) + return -EINVAL; + + queue = &np->queues[smp_processor_id() % num_queues]; + + spin_lock_irqsave(&queue->tx_lock, irq_flags); + for (i = 0; i < n; i++) { + struct xdp_frame *xdpf = frames[i]; + + if (!xdpf) + continue; + if (xennet_xdp_xmit_one(dev, queue, xdpf)) + break; + nxmit++; + } + spin_unlock_irqrestore(&queue->tx_lock, irq_flags); + + return nxmit; +} + +static struct sk_buff *bounce_skb(const struct sk_buff *skb) +{ + unsigned int headerlen = skb_headroom(skb); + /* Align size to allocate full pages and avoid contiguous data leaks */ + unsigned int size = ALIGN(skb_end_offset(skb) + skb->data_len, + XEN_PAGE_SIZE); + struct sk_buff *n = alloc_skb(size, GFP_ATOMIC | __GFP_ZERO); + + if (!n) + return NULL; + + if (!IS_ALIGNED((uintptr_t)n->head, XEN_PAGE_SIZE)) { + WARN_ONCE(1, "misaligned skb allocated\n"); + kfree_skb(n); + return NULL; + } + + /* Set the data pointer */ + skb_reserve(n, headerlen); + /* Set the tail pointer and length */ + skb_put(n, skb->len); + + BUG_ON(skb_copy_bits(skb, -headerlen, n->head, headerlen + skb->len)); + + skb_copy_header(n, skb); + return n; +} + +#define MAX_XEN_SKB_FRAGS (65536 / XEN_PAGE_SIZE + 1) + +static netdev_tx_t xennet_start_xmit(struct sk_buff *skb, struct net_device *dev) +{ + struct netfront_info *np = netdev_priv(dev); + struct netfront_stats *tx_stats = this_cpu_ptr(np->tx_stats); + struct xen_netif_tx_request *first_tx; + unsigned int i; int notify; int slots; - unsigned int offset = offset_in_page(data); - unsigned int len = skb_headlen(skb); + struct page *page; + unsigned int offset; + unsigned int len; unsigned long flags; + struct netfront_queue *queue = NULL; + struct xennet_gnttab_make_txreq info = { }; + unsigned int num_queues = dev->real_num_tx_queues; + u16 queue_index; + struct sk_buff *nskb; + + /* Drop the packet if no queues are set up */ + if (num_queues < 1) + goto drop; + if (unlikely(np->broken)) + goto drop; + /* Determine which queue to transmit this SKB on */ + queue_index = skb_get_queue_mapping(skb); + queue = &np->queues[queue_index]; /* If skb->len is too big for wire format, drop skb and alert * user about misconfiguration. @@ -566,58 +740,78 @@ static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev) goto drop; } - slots = DIV_ROUND_UP(offset + len, PAGE_SIZE) + - xennet_count_skb_frag_slots(skb); - if (unlikely(slots > MAX_SKB_FRAGS + 1)) { - net_alert_ratelimited( - "xennet: skb rides the rocket: %d slots\n", slots); - goto drop; + slots = xennet_count_skb_slots(skb); + if (unlikely(slots > MAX_XEN_SKB_FRAGS + 1)) { + net_dbg_ratelimited("xennet: skb rides the rocket: %d slots, %d bytes\n", + slots, skb->len); + if (skb_linearize(skb)) + goto drop; + } + + page = virt_to_page(skb->data); + offset = offset_in_page(skb->data); + + /* The first req should be at least ETH_HLEN size or the packet will be + * dropped by netback. + * + * If the backend is not trusted bounce all data to zeroed pages to + * avoid exposing contiguous data on the granted page not belonging to + * the skb. + */ + if (np->bounce || unlikely(PAGE_SIZE - offset < ETH_HLEN)) { + nskb = bounce_skb(skb); + if (!nskb) + goto drop; + dev_consume_skb_any(skb); + skb = nskb; + page = virt_to_page(skb->data); + offset = offset_in_page(skb->data); } - spin_lock_irqsave(&np->tx_lock, flags); + len = skb_headlen(skb); + + spin_lock_irqsave(&queue->tx_lock, flags); if (unlikely(!netif_carrier_ok(dev) || (slots > 1 && !xennet_can_sg(dev)) || netif_needs_gso(skb, netif_skb_features(skb)))) { - spin_unlock_irqrestore(&np->tx_lock, flags); + spin_unlock_irqrestore(&queue->tx_lock, flags); goto drop; } - i = np->tx.req_prod_pvt; - - id = get_id_from_freelist(&np->tx_skb_freelist, np->tx_skbs); - np->tx_skbs[id].skb = skb; - - tx = RING_GET_REQUEST(&np->tx, i); - - tx->id = id; - ref = gnttab_claim_grant_reference(&np->gref_tx_head); - BUG_ON((signed short)ref < 0); - mfn = virt_to_mfn(data); - gnttab_grant_foreign_access_ref( - ref, np->xbdev->otherend_id, mfn, GNTMAP_readonly); - tx->gref = np->grant_tx_ref[id] = ref; - tx->offset = offset; - tx->size = len; + /* First request for the linear area. */ + info.queue = queue; + info.skb = skb; + info.page = page; + first_tx = xennet_make_first_txreq(&info, offset, len); + offset += info.tx_local.size; + if (offset == PAGE_SIZE) { + page++; + offset = 0; + } + len -= info.tx_local.size; - tx->flags = 0; if (skb->ip_summed == CHECKSUM_PARTIAL) /* local packet? */ - tx->flags |= XEN_NETTXF_csum_blank | XEN_NETTXF_data_validated; + first_tx->flags |= XEN_NETTXF_csum_blank | + XEN_NETTXF_data_validated; else if (skb->ip_summed == CHECKSUM_UNNECESSARY) /* remote but checksummed. */ - tx->flags |= XEN_NETTXF_data_validated; + first_tx->flags |= XEN_NETTXF_data_validated; + /* Optional extra info after the first request. */ if (skb_shinfo(skb)->gso_size) { struct xen_netif_extra_info *gso; gso = (struct xen_netif_extra_info *) - RING_GET_REQUEST(&np->tx, ++i); + RING_GET_REQUEST(&queue->tx, queue->tx.req_prod_pvt++); - tx->flags |= XEN_NETTXF_extra_info; + first_tx->flags |= XEN_NETTXF_extra_info; gso->u.gso.size = skb_shinfo(skb)->gso_size; - gso->u.gso.type = XEN_NETIF_GSO_TYPE_TCPV4; + gso->u.gso.type = (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6) ? + XEN_NETIF_GSO_TYPE_TCPV6 : + XEN_NETIF_GSO_TYPE_TCPV4; gso->u.gso.pad = 0; gso->u.gso.features = 0; @@ -625,65 +819,117 @@ static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev) gso->flags = 0; } - np->tx.req_prod_pvt = i + 1; + /* Requests for the rest of the linear area. */ + xennet_make_txreqs(&info, page, offset, len); - xennet_make_frags(skb, dev, tx); - tx->size = skb->len; + /* Requests for all the frags. */ + for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { + skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; + xennet_make_txreqs(&info, skb_frag_page(frag), + skb_frag_off(frag), + skb_frag_size(frag)); + } - RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&np->tx, notify); - if (notify) - notify_remote_via_irq(np->tx_irq); + /* First request has the packet length. */ + first_tx->size = skb->len; - u64_stats_update_begin(&stats->syncp); - stats->tx_bytes += skb->len; - stats->tx_packets++; - u64_stats_update_end(&stats->syncp); + /* timestamp packet in software */ + skb_tx_timestamp(skb); - /* Note: It is not safe to access skb after xennet_tx_buf_gc()! */ - xennet_tx_buf_gc(dev); + xennet_mark_tx_pending(queue); - if (!netfront_tx_slot_available(np)) - netif_stop_queue(dev); + RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&queue->tx, notify); + if (notify) + notify_remote_via_irq(queue->tx_irq); + + u64_stats_update_begin(&tx_stats->syncp); + tx_stats->bytes += skb->len; + tx_stats->packets++; + u64_stats_update_end(&tx_stats->syncp); + + if (!netfront_tx_slot_available(queue)) + netif_tx_stop_queue(netdev_get_tx_queue(dev, queue->id)); - spin_unlock_irqrestore(&np->tx_lock, flags); + spin_unlock_irqrestore(&queue->tx_lock, flags); return NETDEV_TX_OK; drop: dev->stats.tx_dropped++; - dev_kfree_skb(skb); + dev_kfree_skb_any(skb); return NETDEV_TX_OK; } static int xennet_close(struct net_device *dev) { struct netfront_info *np = netdev_priv(dev); - netif_stop_queue(np->netdev); - napi_disable(&np->napi); + unsigned int num_queues = np->queues ? dev->real_num_tx_queues : 0; + unsigned int i; + struct netfront_queue *queue; + netif_tx_stop_all_queues(np->netdev); + for (i = 0; i < num_queues; ++i) { + queue = &np->queues[i]; + napi_disable(&queue->napi); + } return 0; } -static void xennet_move_rx_slot(struct netfront_info *np, struct sk_buff *skb, - grant_ref_t ref) +static void xennet_destroy_queues(struct netfront_info *info) +{ + unsigned int i; + + if (!info->queues) + return; + + for (i = 0; i < info->netdev->real_num_tx_queues; i++) { + struct netfront_queue *queue = &info->queues[i]; + + if (netif_running(info->netdev)) + napi_disable(&queue->napi); + netif_napi_del(&queue->napi); + } + + kfree(info->queues); + info->queues = NULL; +} + +static void xennet_uninit(struct net_device *dev) { - int new = xennet_rxidx(np->rx.req_prod_pvt); + struct netfront_info *np = netdev_priv(dev); + xennet_destroy_queues(np); +} + +static void xennet_set_rx_rsp_cons(struct netfront_queue *queue, RING_IDX val) +{ + unsigned long flags; + + spin_lock_irqsave(&queue->rx_cons_lock, flags); + queue->rx.rsp_cons = val; + queue->rx_rsp_unconsumed = XEN_RING_NR_UNCONSUMED_RESPONSES(&queue->rx); + spin_unlock_irqrestore(&queue->rx_cons_lock, flags); +} - BUG_ON(np->rx_skbs[new]); - np->rx_skbs[new] = skb; - np->grant_rx_ref[new] = ref; - RING_GET_REQUEST(&np->rx, np->rx.req_prod_pvt)->id = new; - RING_GET_REQUEST(&np->rx, np->rx.req_prod_pvt)->gref = ref; - np->rx.req_prod_pvt++; +static void xennet_move_rx_slot(struct netfront_queue *queue, struct sk_buff *skb, + grant_ref_t ref) +{ + int new = xennet_rxidx(queue->rx.req_prod_pvt); + + BUG_ON(queue->rx_skbs[new]); + queue->rx_skbs[new] = skb; + queue->grant_rx_ref[new] = ref; + RING_GET_REQUEST(&queue->rx, queue->rx.req_prod_pvt)->id = new; + RING_GET_REQUEST(&queue->rx, queue->rx.req_prod_pvt)->gref = ref; + queue->rx.req_prod_pvt++; } -static int xennet_get_extras(struct netfront_info *np, +static int xennet_get_extras(struct netfront_queue *queue, struct xen_netif_extra_info *extras, RING_IDX rp) { - struct xen_netif_extra_info *extra; - struct device *dev = &np->netdev->dev; - RING_IDX cons = np->rx.rsp_cons; + struct xen_netif_extra_info extra; + struct device *dev = &queue->info->netdev->dev; + RING_IDX cons = queue->rx.rsp_cons; int err = 0; do { @@ -697,66 +943,119 @@ static int xennet_get_extras(struct netfront_info *np, break; } - extra = (struct xen_netif_extra_info *) - RING_GET_RESPONSE(&np->rx, ++cons); + RING_COPY_RESPONSE(&queue->rx, ++cons, &extra); - if (unlikely(!extra->type || - extra->type >= XEN_NETIF_EXTRA_TYPE_MAX)) { + if (unlikely(!extra.type || + extra.type >= XEN_NETIF_EXTRA_TYPE_MAX)) { if (net_ratelimit()) dev_warn(dev, "Invalid extra type: %d\n", - extra->type); + extra.type); err = -EINVAL; } else { - memcpy(&extras[extra->type - 1], extra, - sizeof(*extra)); + extras[extra.type - 1] = extra; } - skb = xennet_get_rx_skb(np, cons); - ref = xennet_get_rx_ref(np, cons); - xennet_move_rx_slot(np, skb, ref); - } while (extra->flags & XEN_NETIF_EXTRA_FLAG_MORE); + skb = xennet_get_rx_skb(queue, cons); + ref = xennet_get_rx_ref(queue, cons); + xennet_move_rx_slot(queue, skb, ref); + } while (extra.flags & XEN_NETIF_EXTRA_FLAG_MORE); - np->rx.rsp_cons = cons; + xennet_set_rx_rsp_cons(queue, cons); return err; } -static int xennet_get_responses(struct netfront_info *np, +static u32 xennet_run_xdp(struct netfront_queue *queue, struct page *pdata, + struct xen_netif_rx_response *rx, struct bpf_prog *prog, + struct xdp_buff *xdp, bool *need_xdp_flush) +{ + struct xdp_frame *xdpf; + u32 len = rx->status; + u32 act; + int err; + + xdp_init_buff(xdp, XEN_PAGE_SIZE - XDP_PACKET_HEADROOM, + &queue->xdp_rxq); + xdp_prepare_buff(xdp, page_address(pdata), XDP_PACKET_HEADROOM, + len, false); + + act = bpf_prog_run_xdp(prog, xdp); + switch (act) { + case XDP_TX: + xdpf = xdp_convert_buff_to_frame(xdp); + if (unlikely(!xdpf)) { + trace_xdp_exception(queue->info->netdev, prog, act); + break; + } + get_page(pdata); + err = xennet_xdp_xmit(queue->info->netdev, 1, &xdpf, 0); + if (unlikely(err <= 0)) { + if (err < 0) + trace_xdp_exception(queue->info->netdev, prog, act); + xdp_return_frame_rx_napi(xdpf); + } + break; + case XDP_REDIRECT: + get_page(pdata); + err = xdp_do_redirect(queue->info->netdev, xdp, prog); + *need_xdp_flush = true; + if (unlikely(err)) { + trace_xdp_exception(queue->info->netdev, prog, act); + xdp_return_buff(xdp); + } + break; + case XDP_PASS: + case XDP_DROP: + break; + + case XDP_ABORTED: + trace_xdp_exception(queue->info->netdev, prog, act); + break; + + default: + bpf_warn_invalid_xdp_action(queue->info->netdev, prog, act); + } + + return act; +} + +static int xennet_get_responses(struct netfront_queue *queue, struct netfront_rx_info *rinfo, RING_IDX rp, - struct sk_buff_head *list) + struct sk_buff_head *list, + bool *need_xdp_flush) { - struct xen_netif_rx_response *rx = &rinfo->rx; + struct xen_netif_rx_response *rx = &rinfo->rx, rx_local; + int max = XEN_NETIF_NR_SLOTS_MIN + (rx->status <= RX_COPY_THRESHOLD); + RING_IDX cons = queue->rx.rsp_cons; + struct sk_buff *skb = xennet_get_rx_skb(queue, cons); struct xen_netif_extra_info *extras = rinfo->extras; - struct device *dev = &np->netdev->dev; - RING_IDX cons = np->rx.rsp_cons; - struct sk_buff *skb = xennet_get_rx_skb(np, cons); - grant_ref_t ref = xennet_get_rx_ref(np, cons); - int max = MAX_SKB_FRAGS + (rx->status <= RX_COPY_THRESHOLD); + grant_ref_t ref = xennet_get_rx_ref(queue, cons); + struct device *dev = &queue->info->netdev->dev; + struct bpf_prog *xdp_prog; + struct xdp_buff xdp; int slots = 1; int err = 0; - unsigned long ret; + u32 verdict; if (rx->flags & XEN_NETRXF_extra_info) { - err = xennet_get_extras(np, extras, rp); - cons = np->rx.rsp_cons; - } + err = xennet_get_extras(queue, extras, rp); + if (!err) { + if (extras[XEN_NETIF_EXTRA_TYPE_XDP - 1].type) { + struct xen_netif_extra_info *xdp; - for (;;) { - if (unlikely(rx->status < 0 || - rx->offset + rx->status > PAGE_SIZE)) { - if (net_ratelimit()) - dev_warn(dev, "rx->offset: %x, size: %u\n", - rx->offset, rx->status); - xennet_move_rx_slot(np, skb, ref); - err = -EINVAL; - goto next; + xdp = &extras[XEN_NETIF_EXTRA_TYPE_XDP - 1]; + rx->offset = xdp->u.xdp.headroom; + } } + cons = queue->rx.rsp_cons; + } + for (;;) { /* * This definitely indicates a bug, either in this driver or in * the backend driver. In future this should flag the bad * situation to the system controller to reboot the backend. */ - if (ref == GRANT_INVALID_REF) { + if (ref == INVALID_GRANT_REF) { if (net_ratelimit()) dev_warn(dev, "Bad rx response id %d.\n", rx->id); @@ -764,10 +1063,42 @@ static int xennet_get_responses(struct netfront_info *np, goto next; } - ret = gnttab_end_foreign_access_ref(ref, 0); - BUG_ON(!ret); + if (unlikely(rx->status < 0 || + rx->offset + rx->status > XEN_PAGE_SIZE)) { + if (net_ratelimit()) + dev_warn(dev, "rx->offset: %u, size: %d\n", + rx->offset, rx->status); + xennet_move_rx_slot(queue, skb, ref); + err = -EINVAL; + goto next; + } - gnttab_release_grant_reference(&np->gref_rx_head, ref); + if (!gnttab_end_foreign_access_ref(ref)) { + dev_alert(dev, + "Grant still in use by backend domain\n"); + queue->info->broken = true; + dev_alert(dev, "Disabled for further use\n"); + return -EINVAL; + } + + gnttab_release_grant_reference(&queue->gref_rx_head, ref); + + rcu_read_lock(); + xdp_prog = rcu_dereference(queue->xdp_prog); + if (xdp_prog) { + if (!(rx->flags & XEN_NETRXF_more_data)) { + /* currently only a single page contains data */ + verdict = xennet_run_xdp(queue, + skb_frag_page(&skb_shinfo(skb)->frags[0]), + rx, xdp_prog, &xdp, need_xdp_flush); + if (verdict != XDP_PASS) + err = -EINVAL; + } else { + /* drop the frame */ + err = -EINVAL; + } + } + rcu_read_unlock(); __skb_queue_tail(list, skb); @@ -782,9 +1113,10 @@ next: break; } - rx = RING_GET_RESPONSE(&np->rx, cons + slots); - skb = xennet_get_rx_skb(np, cons + slots); - ref = xennet_get_rx_ref(np, cons + slots); + RING_COPY_RESPONSE(&queue->rx, cons + slots, &rx_local); + rx = &rx_local; + skb = xennet_get_rx_skb(queue, cons + slots); + ref = xennet_get_rx_ref(queue, cons + slots); slots++; } @@ -795,7 +1127,7 @@ next: } if (unlikely(err)) - np->rx.rsp_cons = cons + slots; + xennet_set_rx_rsp_cons(queue, cons + slots); return err; } @@ -809,15 +1141,18 @@ static int xennet_set_skb_gso(struct sk_buff *skb, return -EINVAL; } - /* Currently only TCPv4 S.O. is supported. */ - if (gso->u.gso.type != XEN_NETIF_GSO_TYPE_TCPV4) { + if (gso->u.gso.type != XEN_NETIF_GSO_TYPE_TCPV4 && + gso->u.gso.type != XEN_NETIF_GSO_TYPE_TCPV6) { if (net_ratelimit()) pr_warn("Bad GSO type %d\n", gso->u.gso.type); return -EINVAL; } skb_shinfo(skb)->gso_size = gso->u.gso.size; - skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4; + skb_shinfo(skb)->gso_type = + (gso->u.gso.type == XEN_NETIF_GSO_TYPE_TCPV4) ? + SKB_GSO_TCPV4 : + SKB_GSO_TCPV6; /* Header must be checked, and gso_segs computed. */ skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY; @@ -826,41 +1161,48 @@ static int xennet_set_skb_gso(struct sk_buff *skb, return 0; } -static RING_IDX xennet_fill_frags(struct netfront_info *np, - struct sk_buff *skb, - struct sk_buff_head *list) +static int xennet_fill_frags(struct netfront_queue *queue, + struct sk_buff *skb, + struct sk_buff_head *list) { - struct skb_shared_info *shinfo = skb_shinfo(skb); - int nr_frags = shinfo->nr_frags; - RING_IDX cons = np->rx.rsp_cons; + RING_IDX cons = queue->rx.rsp_cons; struct sk_buff *nskb; while ((nskb = __skb_dequeue(list))) { - struct xen_netif_rx_response *rx = - RING_GET_RESPONSE(&np->rx, ++cons); + struct xen_netif_rx_response rx; skb_frag_t *nfrag = &skb_shinfo(nskb)->frags[0]; - __skb_fill_page_desc(skb, nr_frags, - skb_frag_page(nfrag), - rx->offset, rx->status); + RING_COPY_RESPONSE(&queue->rx, ++cons, &rx); - skb->data_len += rx->status; + if (skb_shinfo(skb)->nr_frags == MAX_SKB_FRAGS) { + unsigned int pull_to = NETFRONT_SKB_CB(skb)->pull_to; + + BUG_ON(pull_to < skb_headlen(skb)); + __pskb_pull_tail(skb, pull_to - skb_headlen(skb)); + } + if (unlikely(skb_shinfo(skb)->nr_frags >= MAX_SKB_FRAGS)) { + xennet_set_rx_rsp_cons(queue, + ++cons + skb_queue_len(list)); + kfree_skb(nskb); + return -ENOENT; + } + + skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, + skb_frag_page(nfrag), + rx.offset, rx.status, PAGE_SIZE); skb_shinfo(nskb)->nr_frags = 0; kfree_skb(nskb); - - nr_frags++; } - shinfo->nr_frags = nr_frags; - return cons; + xennet_set_rx_rsp_cons(queue, cons); + + return 0; } static int checksum_setup(struct net_device *dev, struct sk_buff *skb) { - struct iphdr *iph; - int err = -EPROTO; - int recalculate_partial_csum = 0; + bool recalculate_partial_csum = false; /* * A GSO SKB must be CHECKSUM_PARTIAL. However some buggy @@ -870,88 +1212,49 @@ static int checksum_setup(struct net_device *dev, struct sk_buff *skb) */ if (skb->ip_summed != CHECKSUM_PARTIAL && skb_is_gso(skb)) { struct netfront_info *np = netdev_priv(dev); - np->rx_gso_checksum_fixup++; + atomic_inc(&np->rx_gso_checksum_fixup); skb->ip_summed = CHECKSUM_PARTIAL; - recalculate_partial_csum = 1; + recalculate_partial_csum = true; } /* A non-CHECKSUM_PARTIAL SKB does not require setup. */ if (skb->ip_summed != CHECKSUM_PARTIAL) return 0; - if (skb->protocol != htons(ETH_P_IP)) - goto out; - - iph = (void *)skb->data; - - switch (iph->protocol) { - case IPPROTO_TCP: - if (!skb_partial_csum_set(skb, 4 * iph->ihl, - offsetof(struct tcphdr, check))) - goto out; - - if (recalculate_partial_csum) { - struct tcphdr *tcph = tcp_hdr(skb); - tcph->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr, - skb->len - iph->ihl*4, - IPPROTO_TCP, 0); - } - break; - case IPPROTO_UDP: - if (!skb_partial_csum_set(skb, 4 * iph->ihl, - offsetof(struct udphdr, check))) - goto out; - - if (recalculate_partial_csum) { - struct udphdr *udph = udp_hdr(skb); - udph->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr, - skb->len - iph->ihl*4, - IPPROTO_UDP, 0); - } - break; - default: - if (net_ratelimit()) - pr_err("Attempting to checksum a non-TCP/UDP packet, dropping a protocol %d packet\n", - iph->protocol); - goto out; - } - - err = 0; - -out: - return err; + return skb_checksum_setup(skb, recalculate_partial_csum); } -static int handle_incoming_queue(struct net_device *dev, +static int handle_incoming_queue(struct netfront_queue *queue, struct sk_buff_head *rxq) { - struct netfront_info *np = netdev_priv(dev); - struct netfront_stats *stats = this_cpu_ptr(np->stats); + struct netfront_stats *rx_stats = this_cpu_ptr(queue->info->rx_stats); int packets_dropped = 0; struct sk_buff *skb; while ((skb = __skb_dequeue(rxq)) != NULL) { int pull_to = NETFRONT_SKB_CB(skb)->pull_to; - __pskb_pull_tail(skb, pull_to - skb_headlen(skb)); + if (pull_to > skb_headlen(skb)) + __pskb_pull_tail(skb, pull_to - skb_headlen(skb)); /* Ethernet work: Delayed to here as it peeks the header. */ - skb->protocol = eth_type_trans(skb, dev); + skb->protocol = eth_type_trans(skb, queue->info->netdev); + skb_reset_network_header(skb); - if (checksum_setup(dev, skb)) { + if (checksum_setup(queue->info->netdev, skb)) { kfree_skb(skb); packets_dropped++; - dev->stats.rx_errors++; + queue->info->netdev->stats.rx_errors++; continue; } - u64_stats_update_begin(&stats->syncp); - stats->rx_packets++; - stats->rx_bytes += skb->len; - u64_stats_update_end(&stats->syncp); + u64_stats_update_begin(&rx_stats->syncp); + rx_stats->packets++; + rx_stats->bytes += skb->len; + u64_stats_update_end(&rx_stats->syncp); /* Pass it up. */ - netif_receive_skb(skb); + napi_gro_receive(&queue->napi, skb); } return packets_dropped; @@ -959,8 +1262,8 @@ static int handle_incoming_queue(struct net_device *dev, static int xennet_poll(struct napi_struct *napi, int budget) { - struct netfront_info *np = container_of(napi, struct netfront_info, napi); - struct net_device *dev = np->netdev; + struct netfront_queue *queue = container_of(napi, struct netfront_queue, napi); + struct net_device *dev = queue->info->netdev; struct sk_buff *skb; struct netfront_rx_info rinfo; struct xen_netif_rx_response *rx = &rinfo.rx; @@ -970,32 +1273,44 @@ static int xennet_poll(struct napi_struct *napi, int budget) struct sk_buff_head rxq; struct sk_buff_head errq; struct sk_buff_head tmpq; - unsigned long flags; int err; + bool need_xdp_flush = false; - spin_lock(&np->rx_lock); + spin_lock(&queue->rx_lock); skb_queue_head_init(&rxq); skb_queue_head_init(&errq); skb_queue_head_init(&tmpq); - rp = np->rx.sring->rsp_prod; + rp = queue->rx.sring->rsp_prod; + if (RING_RESPONSE_PROD_OVERFLOW(&queue->rx, rp)) { + dev_alert(&dev->dev, "Illegal number of responses %u\n", + rp - queue->rx.rsp_cons); + queue->info->broken = true; + spin_unlock(&queue->rx_lock); + return 0; + } rmb(); /* Ensure we see queued responses up to 'rp'. */ - i = np->rx.rsp_cons; + i = queue->rx.rsp_cons; work_done = 0; while ((i != rp) && (work_done < budget)) { - memcpy(rx, RING_GET_RESPONSE(&np->rx, i), sizeof(*rx)); + RING_COPY_RESPONSE(&queue->rx, i, rx); memset(extras, 0, sizeof(rinfo.extras)); - err = xennet_get_responses(np, &rinfo, rp, &tmpq); + err = xennet_get_responses(queue, &rinfo, rp, &tmpq, + &need_xdp_flush); if (unlikely(err)) { + if (queue->info->broken) { + spin_unlock(&queue->rx_lock); + return 0; + } err: while ((skb = __skb_dequeue(&tmpq))) __skb_queue_tail(&errq, skb); dev->stats.rx_errors++; - i = np->rx.rsp_cons; + i = queue->rx.rsp_cons; continue; } @@ -1007,7 +1322,9 @@ err: if (unlikely(xennet_set_skb_gso(skb, gso))) { __skb_queue_head(&tmpq, skb); - np->rx.rsp_cons += skb_queue_len(&tmpq); + xennet_set_rx_rsp_cons(queue, + queue->rx.rsp_cons + + skb_queue_len(&tmpq)); goto err; } } @@ -1016,18 +1333,13 @@ err: if (NETFRONT_SKB_CB(skb)->pull_to > RX_COPY_THRESHOLD) NETFRONT_SKB_CB(skb)->pull_to = RX_COPY_THRESHOLD; - skb_shinfo(skb)->frags[0].page_offset = rx->offset; + skb_frag_off_set(&skb_shinfo(skb)->frags[0], rx->offset); skb_frag_size_set(&skb_shinfo(skb)->frags[0], rx->status); skb->data_len = rx->status; + skb->len += rx->status; - i = xennet_fill_frags(np, skb, &tmpq); - - /* - * Truesize is the actual allocation size, even if the - * allocation is only partially used. - */ - skb->truesize += PAGE_SIZE * skb_shinfo(skb)->nr_frags; - skb->len += skb->data_len; + if (unlikely(xennet_fill_frags(queue, skb, &tmpq))) + goto err; if (rx->flags & XEN_NETRXF_csum_blank) skb->ip_summed = CHECKSUM_PARTIAL; @@ -1036,70 +1348,67 @@ err: __skb_queue_tail(&rxq, skb); - np->rx.rsp_cons = ++i; + i = queue->rx.rsp_cons + 1; + xennet_set_rx_rsp_cons(queue, i); work_done++; } + if (need_xdp_flush) + xdp_do_flush(); __skb_queue_purge(&errq); - work_done -= handle_incoming_queue(dev, &rxq); - - /* If we get a callback with very few responses, reduce fill target. */ - /* NB. Note exponential increase, linear decrease. */ - if (((np->rx.req_prod_pvt - np->rx.sring->rsp_prod) > - ((3*np->rx_target) / 4)) && - (--np->rx_target < np->rx_min_target)) - np->rx_target = np->rx_min_target; + work_done -= handle_incoming_queue(queue, &rxq); - xennet_alloc_rx_buffers(dev); + xennet_alloc_rx_buffers(queue); if (work_done < budget) { int more_to_do = 0; - local_irq_save(flags); - - RING_FINAL_CHECK_FOR_RESPONSES(&np->rx, more_to_do); - if (!more_to_do) - __napi_complete(napi); + napi_complete_done(napi, work_done); - local_irq_restore(flags); + RING_FINAL_CHECK_FOR_RESPONSES(&queue->rx, more_to_do); + if (more_to_do) + napi_schedule(napi); } - spin_unlock(&np->rx_lock); + spin_unlock(&queue->rx_lock); return work_done; } static int xennet_change_mtu(struct net_device *dev, int mtu) { - int max = xennet_can_sg(dev) ? - XEN_NETIF_MAX_TX_SIZE - MAX_TCP_HEADER : ETH_DATA_LEN; + int max = xennet_can_sg(dev) ? XEN_NETIF_MAX_TX_SIZE : ETH_DATA_LEN; if (mtu > max) return -EINVAL; - dev->mtu = mtu; + WRITE_ONCE(dev->mtu, mtu); return 0; } -static struct rtnl_link_stats64 *xennet_get_stats64(struct net_device *dev, - struct rtnl_link_stats64 *tot) +static void xennet_get_stats64(struct net_device *dev, + struct rtnl_link_stats64 *tot) { struct netfront_info *np = netdev_priv(dev); int cpu; for_each_possible_cpu(cpu) { - struct netfront_stats *stats = per_cpu_ptr(np->stats, cpu); + struct netfront_stats *rx_stats = per_cpu_ptr(np->rx_stats, cpu); + struct netfront_stats *tx_stats = per_cpu_ptr(np->tx_stats, cpu); u64 rx_packets, rx_bytes, tx_packets, tx_bytes; unsigned int start; do { - start = u64_stats_fetch_begin_bh(&stats->syncp); + start = u64_stats_fetch_begin(&tx_stats->syncp); + tx_packets = tx_stats->packets; + tx_bytes = tx_stats->bytes; + } while (u64_stats_fetch_retry(&tx_stats->syncp, start)); - rx_packets = stats->rx_packets; - tx_packets = stats->tx_packets; - rx_bytes = stats->rx_bytes; - tx_bytes = stats->tx_bytes; - } while (u64_stats_fetch_retry_bh(&stats->syncp, start)); + do { + start = u64_stats_fetch_begin(&rx_stats->syncp); + rx_packets = rx_stats->packets; + rx_bytes = rx_stats->bytes; + } while (u64_stats_fetch_retry(&rx_stats->syncp, start)); tot->rx_packets += rx_packets; tot->tx_packets += tx_packets; @@ -1109,140 +1418,84 @@ static struct rtnl_link_stats64 *xennet_get_stats64(struct net_device *dev, tot->rx_errors = dev->stats.rx_errors; tot->tx_dropped = dev->stats.tx_dropped; - - return tot; } -static void xennet_release_tx_bufs(struct netfront_info *np) +static void xennet_release_tx_bufs(struct netfront_queue *queue) { struct sk_buff *skb; int i; for (i = 0; i < NET_TX_RING_SIZE; i++) { /* Skip over entries which are actually freelist references */ - if (skb_entry_is_link(&np->tx_skbs[i])) + if (!queue->tx_skbs[i]) continue; - skb = np->tx_skbs[i].skb; - gnttab_end_foreign_access_ref(np->grant_tx_ref[i], - GNTMAP_readonly); - gnttab_release_grant_reference(&np->gref_tx_head, - np->grant_tx_ref[i]); - np->grant_tx_ref[i] = GRANT_INVALID_REF; - add_id_to_freelist(&np->tx_skb_freelist, np->tx_skbs, i); + skb = queue->tx_skbs[i]; + queue->tx_skbs[i] = NULL; + get_page(queue->grant_tx_page[i]); + gnttab_end_foreign_access(queue->grant_tx_ref[i], + queue->grant_tx_page[i]); + queue->grant_tx_page[i] = NULL; + queue->grant_tx_ref[i] = INVALID_GRANT_REF; + add_id_to_list(&queue->tx_skb_freelist, queue->tx_link, i); dev_kfree_skb_irq(skb); } } -static void xennet_release_rx_bufs(struct netfront_info *np) +static void xennet_release_rx_bufs(struct netfront_queue *queue) { - struct mmu_update *mmu = np->rx_mmu; - struct multicall_entry *mcl = np->rx_mcl; - struct sk_buff_head free_list; - struct sk_buff *skb; - unsigned long mfn; - int xfer = 0, noxfer = 0, unused = 0; int id, ref; - dev_warn(&np->netdev->dev, "%s: fix me for copying receiver.\n", - __func__); - return; - - skb_queue_head_init(&free_list); - - spin_lock_bh(&np->rx_lock); + spin_lock_bh(&queue->rx_lock); for (id = 0; id < NET_RX_RING_SIZE; id++) { - ref = np->grant_rx_ref[id]; - if (ref == GRANT_INVALID_REF) { - unused++; - continue; - } + struct sk_buff *skb; + struct page *page; - skb = np->rx_skbs[id]; - mfn = gnttab_end_foreign_transfer_ref(ref); - gnttab_release_grant_reference(&np->gref_rx_head, ref); - np->grant_rx_ref[id] = GRANT_INVALID_REF; + skb = queue->rx_skbs[id]; + if (!skb) + continue; - if (0 == mfn) { - skb_shinfo(skb)->nr_frags = 0; - dev_kfree_skb(skb); - noxfer++; + ref = queue->grant_rx_ref[id]; + if (ref == INVALID_GRANT_REF) continue; - } - if (!xen_feature(XENFEAT_auto_translated_physmap)) { - /* Remap the page. */ - const struct page *page = - skb_frag_page(&skb_shinfo(skb)->frags[0]); - unsigned long pfn = page_to_pfn(page); - void *vaddr = page_address(page); - - MULTI_update_va_mapping(mcl, (unsigned long)vaddr, - mfn_pte(mfn, PAGE_KERNEL), - 0); - mcl++; - mmu->ptr = ((u64)mfn << PAGE_SHIFT) - | MMU_MACHPHYS_UPDATE; - mmu->val = pfn; - mmu++; - - set_phys_to_machine(pfn, mfn); - } - __skb_queue_tail(&free_list, skb); - xfer++; - } + page = skb_frag_page(&skb_shinfo(skb)->frags[0]); - dev_info(&np->netdev->dev, "%s: %d xfer, %d noxfer, %d unused\n", - __func__, xfer, noxfer, unused); + /* gnttab_end_foreign_access() needs a page ref until + * foreign access is ended (which may be deferred). + */ + get_page(page); + gnttab_end_foreign_access(ref, page); + queue->grant_rx_ref[id] = INVALID_GRANT_REF; - if (xfer) { - if (!xen_feature(XENFEAT_auto_translated_physmap)) { - /* Do all the remapping work and M2P updates. */ - MULTI_mmu_update(mcl, np->rx_mmu, mmu - np->rx_mmu, - NULL, DOMID_SELF); - mcl++; - HYPERVISOR_multicall(np->rx_mcl, mcl - np->rx_mcl); - } + kfree_skb(skb); } - __skb_queue_purge(&free_list); - - spin_unlock_bh(&np->rx_lock); -} - -static void xennet_uninit(struct net_device *dev) -{ - struct netfront_info *np = netdev_priv(dev); - xennet_release_tx_bufs(np); - xennet_release_rx_bufs(np); - gnttab_free_grant_references(np->gref_tx_head); - gnttab_free_grant_references(np->gref_rx_head); + spin_unlock_bh(&queue->rx_lock); } static netdev_features_t xennet_fix_features(struct net_device *dev, netdev_features_t features) { struct netfront_info *np = netdev_priv(dev); - int val; - if (features & NETIF_F_SG) { - if (xenbus_scanf(XBT_NIL, np->xbdev->otherend, "feature-sg", - "%d", &val) < 0) - val = 0; + if (features & NETIF_F_SG && + !xenbus_read_unsigned(np->xbdev->otherend, "feature-sg", 0)) + features &= ~NETIF_F_SG; - if (!val) - features &= ~NETIF_F_SG; - } + if (features & NETIF_F_IPV6_CSUM && + !xenbus_read_unsigned(np->xbdev->otherend, + "feature-ipv6-csum-offload", 0)) + features &= ~NETIF_F_IPV6_CSUM; - if (features & NETIF_F_TSO) { - if (xenbus_scanf(XBT_NIL, np->xbdev->otherend, - "feature-gso-tcpv4", "%d", &val) < 0) - val = 0; + if (features & NETIF_F_TSO && + !xenbus_read_unsigned(np->xbdev->otherend, "feature-gso-tcpv4", 0)) + features &= ~NETIF_F_TSO; - if (!val) - features &= ~NETIF_F_TSO; - } + if (features & NETIF_F_TSO6 && + !xenbus_read_unsigned(np->xbdev->otherend, "feature-gso-tcpv6", 0)) + features &= ~NETIF_F_TSO6; return features; } @@ -1258,48 +1511,180 @@ static int xennet_set_features(struct net_device *dev, return 0; } -static irqreturn_t xennet_tx_interrupt(int irq, void *dev_id) +static bool xennet_handle_tx(struct netfront_queue *queue, unsigned int *eoi) { - struct netfront_info *np = dev_id; - struct net_device *dev = np->netdev; unsigned long flags; - spin_lock_irqsave(&np->tx_lock, flags); - xennet_tx_buf_gc(dev); - spin_unlock_irqrestore(&np->tx_lock, flags); + if (unlikely(queue->info->broken)) + return false; + + spin_lock_irqsave(&queue->tx_lock, flags); + if (xennet_tx_buf_gc(queue)) + *eoi = 0; + spin_unlock_irqrestore(&queue->tx_lock, flags); + + return true; +} + +static irqreturn_t xennet_tx_interrupt(int irq, void *dev_id) +{ + unsigned int eoiflag = XEN_EOI_FLAG_SPURIOUS; + + if (likely(xennet_handle_tx(dev_id, &eoiflag))) + xen_irq_lateeoi(irq, eoiflag); return IRQ_HANDLED; } +static bool xennet_handle_rx(struct netfront_queue *queue, unsigned int *eoi) +{ + unsigned int work_queued; + unsigned long flags; + + if (unlikely(queue->info->broken)) + return false; + + spin_lock_irqsave(&queue->rx_cons_lock, flags); + work_queued = XEN_RING_NR_UNCONSUMED_RESPONSES(&queue->rx); + if (work_queued > queue->rx_rsp_unconsumed) { + queue->rx_rsp_unconsumed = work_queued; + *eoi = 0; + } else if (unlikely(work_queued < queue->rx_rsp_unconsumed)) { + const struct device *dev = &queue->info->netdev->dev; + + spin_unlock_irqrestore(&queue->rx_cons_lock, flags); + dev_alert(dev, "RX producer index going backwards\n"); + dev_alert(dev, "Disabled for further use\n"); + queue->info->broken = true; + return false; + } + spin_unlock_irqrestore(&queue->rx_cons_lock, flags); + + if (likely(netif_carrier_ok(queue->info->netdev) && work_queued)) + napi_schedule(&queue->napi); + + return true; +} + static irqreturn_t xennet_rx_interrupt(int irq, void *dev_id) { - struct netfront_info *np = dev_id; - struct net_device *dev = np->netdev; + unsigned int eoiflag = XEN_EOI_FLAG_SPURIOUS; - if (likely(netif_carrier_ok(dev) && - RING_HAS_UNCONSUMED_RESPONSES(&np->rx))) - napi_schedule(&np->napi); + if (likely(xennet_handle_rx(dev_id, &eoiflag))) + xen_irq_lateeoi(irq, eoiflag); return IRQ_HANDLED; } static irqreturn_t xennet_interrupt(int irq, void *dev_id) { - xennet_tx_interrupt(irq, dev_id); - xennet_rx_interrupt(irq, dev_id); + unsigned int eoiflag = XEN_EOI_FLAG_SPURIOUS; + + if (xennet_handle_tx(dev_id, &eoiflag) && + xennet_handle_rx(dev_id, &eoiflag)) + xen_irq_lateeoi(irq, eoiflag); + return IRQ_HANDLED; } #ifdef CONFIG_NET_POLL_CONTROLLER static void xennet_poll_controller(struct net_device *dev) { - xennet_interrupt(0, dev); + /* Poll each queue */ + struct netfront_info *info = netdev_priv(dev); + unsigned int num_queues = dev->real_num_tx_queues; + unsigned int i; + + if (info->broken) + return; + + for (i = 0; i < num_queues; ++i) + xennet_interrupt(0, &info->queues[i]); } #endif +#define NETBACK_XDP_HEADROOM_DISABLE 0 +#define NETBACK_XDP_HEADROOM_ENABLE 1 + +static int talk_to_netback_xdp(struct netfront_info *np, int xdp) +{ + int err; + unsigned short headroom; + + headroom = xdp ? XDP_PACKET_HEADROOM : 0; + err = xenbus_printf(XBT_NIL, np->xbdev->nodename, + "xdp-headroom", "%hu", + headroom); + if (err) + pr_warn("Error writing xdp-headroom\n"); + + return err; +} + +static int xennet_xdp_set(struct net_device *dev, struct bpf_prog *prog, + struct netlink_ext_ack *extack) +{ + unsigned long max_mtu = XEN_PAGE_SIZE - XDP_PACKET_HEADROOM; + struct netfront_info *np = netdev_priv(dev); + struct bpf_prog *old_prog; + unsigned int i, err; + + if (dev->mtu > max_mtu) { + netdev_warn(dev, "XDP requires MTU less than %lu\n", max_mtu); + return -EINVAL; + } + + if (!np->netback_has_xdp_headroom) + return 0; + + xenbus_switch_state(np->xbdev, XenbusStateReconfiguring); + + err = talk_to_netback_xdp(np, prog ? NETBACK_XDP_HEADROOM_ENABLE : + NETBACK_XDP_HEADROOM_DISABLE); + if (err) + return err; + + /* avoid the race with XDP headroom adjustment */ + wait_event(module_wq, + xenbus_read_driver_state(np->xbdev->otherend) == + XenbusStateReconfigured); + np->netfront_xdp_enabled = true; + + old_prog = rtnl_dereference(np->queues[0].xdp_prog); + + if (prog) + bpf_prog_add(prog, dev->real_num_tx_queues); + + for (i = 0; i < dev->real_num_tx_queues; ++i) + rcu_assign_pointer(np->queues[i].xdp_prog, prog); + + if (old_prog) + for (i = 0; i < dev->real_num_tx_queues; ++i) + bpf_prog_put(old_prog); + + xenbus_switch_state(np->xbdev, XenbusStateConnected); + + return 0; +} + +static int xennet_xdp(struct net_device *dev, struct netdev_bpf *xdp) +{ + struct netfront_info *np = netdev_priv(dev); + + if (np->broken) + return -ENODEV; + + switch (xdp->command) { + case XDP_SETUP_PROG: + return xennet_xdp_set(dev, xdp->prog, xdp->extack); + default: + return -EINVAL; + } +} + static const struct net_device_ops xennet_netdev_ops = { - .ndo_open = xennet_open, .ndo_uninit = xennet_uninit, + .ndo_open = xennet_open, .ndo_stop = xennet_close, .ndo_start_xmit = xennet_start_xmit, .ndo_change_mtu = xennet_change_mtu, @@ -1308,75 +1693,53 @@ static const struct net_device_ops xennet_netdev_ops = { .ndo_validate_addr = eth_validate_addr, .ndo_fix_features = xennet_fix_features, .ndo_set_features = xennet_set_features, + .ndo_select_queue = xennet_select_queue, + .ndo_bpf = xennet_xdp, + .ndo_xdp_xmit = xennet_xdp_xmit, #ifdef CONFIG_NET_POLL_CONTROLLER .ndo_poll_controller = xennet_poll_controller, #endif }; +static void xennet_free_netdev(struct net_device *netdev) +{ + struct netfront_info *np = netdev_priv(netdev); + + free_percpu(np->rx_stats); + free_percpu(np->tx_stats); + free_netdev(netdev); +} + static struct net_device *xennet_create_dev(struct xenbus_device *dev) { - int i, err; + int err; struct net_device *netdev; struct netfront_info *np; - netdev = alloc_etherdev(sizeof(struct netfront_info)); + netdev = alloc_etherdev_mq(sizeof(struct netfront_info), xennet_max_queues); if (!netdev) return ERR_PTR(-ENOMEM); np = netdev_priv(netdev); np->xbdev = dev; - spin_lock_init(&np->tx_lock); - spin_lock_init(&np->rx_lock); - - skb_queue_head_init(&np->rx_batch); - np->rx_target = RX_DFL_MIN_TARGET; - np->rx_min_target = RX_DFL_MIN_TARGET; - np->rx_max_target = RX_MAX_TARGET; - - init_timer(&np->rx_refill_timer); - np->rx_refill_timer.data = (unsigned long)netdev; - np->rx_refill_timer.function = rx_refill_timeout; + np->queues = NULL; err = -ENOMEM; - np->stats = alloc_percpu(struct netfront_stats); - if (np->stats == NULL) + np->rx_stats = netdev_alloc_pcpu_stats(struct netfront_stats); + if (np->rx_stats == NULL) + goto exit; + np->tx_stats = netdev_alloc_pcpu_stats(struct netfront_stats); + if (np->tx_stats == NULL) goto exit; - - /* Initialise tx_skbs as a free chain containing every entry. */ - np->tx_skb_freelist = 0; - for (i = 0; i < NET_TX_RING_SIZE; i++) { - skb_entry_set_link(&np->tx_skbs[i], i+1); - np->grant_tx_ref[i] = GRANT_INVALID_REF; - } - - /* Clear out rx_skbs */ - for (i = 0; i < NET_RX_RING_SIZE; i++) { - np->rx_skbs[i] = NULL; - np->grant_rx_ref[i] = GRANT_INVALID_REF; - } - - /* A grant for every tx ring slot */ - if (gnttab_alloc_grant_references(TX_MAX_TARGET, - &np->gref_tx_head) < 0) { - pr_alert("can't alloc tx grant refs\n"); - err = -ENOMEM; - goto exit_free_stats; - } - /* A grant for every rx ring slot */ - if (gnttab_alloc_grant_references(RX_MAX_TARGET, - &np->gref_rx_head) < 0) { - pr_alert("can't alloc rx grant refs\n"); - err = -ENOMEM; - goto exit_free_tx; - } netdev->netdev_ops = &xennet_netdev_ops; - netif_napi_add(netdev, &np->napi, xennet_poll, 64); netdev->features = NETIF_F_IP_CSUM | NETIF_F_RXCSUM | NETIF_F_GSO_ROBUST; - netdev->hw_features = NETIF_F_IP_CSUM | NETIF_F_SG | NETIF_F_TSO; + netdev->hw_features = NETIF_F_SG | + NETIF_F_IPV6_CSUM | + NETIF_F_TSO | NETIF_F_TSO6; /* * Assume that all hw features are available for now. This set @@ -1385,28 +1748,36 @@ static struct net_device *xennet_create_dev(struct xenbus_device *dev) * negotiate with the backend regarding supported features. */ netdev->features |= netdev->hw_features; + netdev->xdp_features = NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT | + NETDEV_XDP_ACT_NDO_XMIT; - SET_ETHTOOL_OPS(netdev, &xennet_ethtool_ops); + netdev->ethtool_ops = &xennet_ethtool_ops; + netdev->min_mtu = ETH_MIN_MTU; + netdev->max_mtu = XEN_NETIF_MAX_TX_SIZE; SET_NETDEV_DEV(netdev, &dev->dev); - netif_set_gso_max_size(netdev, XEN_NETIF_MAX_TX_SIZE - MAX_TCP_HEADER); - np->netdev = netdev; + np->netfront_xdp_enabled = false; netif_carrier_off(netdev); + do { + xenbus_switch_state(dev, XenbusStateInitialising); + err = wait_event_timeout(module_wq, + xenbus_read_driver_state(dev->otherend) != + XenbusStateClosed && + xenbus_read_driver_state(dev->otherend) != + XenbusStateUnknown, XENNET_TIMEOUT); + } while (!err); + return netdev; - exit_free_tx: - gnttab_free_grant_references(np->gref_tx_head); - exit_free_stats: - free_percpu(np->stats); exit: - free_netdev(netdev); + xennet_free_netdev(netdev); return ERR_PTR(err); } -/** +/* * Entry point to this code when a new device is created. Allocate the basic * structures and the ring buffers for communication with the backend, and * inform the backend of the appropriate details for those. @@ -1427,64 +1798,63 @@ static int netfront_probe(struct xenbus_device *dev, info = netdev_priv(netdev); dev_set_drvdata(&dev->dev, info); - - err = register_netdev(info->netdev); - if (err) { - pr_warn("%s: register_netdev err=%d\n", __func__, err); - goto fail; - } - - err = xennet_sysfs_addif(info->netdev); - if (err) { - unregister_netdev(info->netdev); - pr_warn("%s: add sysfs failed err=%d\n", __func__, err); - goto fail; - } +#ifdef CONFIG_SYSFS + info->netdev->sysfs_groups[0] = &xennet_dev_group; +#endif return 0; - - fail: - free_netdev(netdev); - dev_set_drvdata(&dev->dev, NULL); - return err; } static void xennet_end_access(int ref, void *page) { /* This frees the page as a side-effect */ - if (ref != GRANT_INVALID_REF) - gnttab_end_foreign_access(ref, 0, (unsigned long)page); + if (ref != INVALID_GRANT_REF) + gnttab_end_foreign_access(ref, virt_to_page(page)); } static void xennet_disconnect_backend(struct netfront_info *info) { - /* Stop old i/f to prevent errors whilst we rebuild the state. */ - spin_lock_bh(&info->rx_lock); - spin_lock_irq(&info->tx_lock); + unsigned int i = 0; + unsigned int num_queues = info->netdev->real_num_tx_queues; + netif_carrier_off(info->netdev); - spin_unlock_irq(&info->tx_lock); - spin_unlock_bh(&info->rx_lock); - if (info->tx_irq && (info->tx_irq == info->rx_irq)) - unbind_from_irqhandler(info->tx_irq, info); - if (info->tx_irq && (info->tx_irq != info->rx_irq)) { - unbind_from_irqhandler(info->tx_irq, info); - unbind_from_irqhandler(info->rx_irq, info); - } - info->tx_evtchn = info->rx_evtchn = 0; - info->tx_irq = info->rx_irq = 0; + for (i = 0; i < num_queues && info->queues; ++i) { + struct netfront_queue *queue = &info->queues[i]; + + timer_delete_sync(&queue->rx_refill_timer); + + if (queue->tx_irq && (queue->tx_irq == queue->rx_irq)) + unbind_from_irqhandler(queue->tx_irq, queue); + if (queue->tx_irq && (queue->tx_irq != queue->rx_irq)) { + unbind_from_irqhandler(queue->tx_irq, queue); + unbind_from_irqhandler(queue->rx_irq, queue); + } + queue->tx_evtchn = queue->rx_evtchn = 0; + queue->tx_irq = queue->rx_irq = 0; - /* End access and free the pages */ - xennet_end_access(info->tx_ring_ref, info->tx.sring); - xennet_end_access(info->rx_ring_ref, info->rx.sring); + if (netif_running(info->netdev)) + napi_synchronize(&queue->napi); - info->tx_ring_ref = GRANT_INVALID_REF; - info->rx_ring_ref = GRANT_INVALID_REF; - info->tx.sring = NULL; - info->rx.sring = NULL; + xennet_release_tx_bufs(queue); + xennet_release_rx_bufs(queue); + gnttab_free_grant_references(queue->gref_tx_head); + gnttab_free_grant_references(queue->gref_rx_head); + + /* End access and free the pages */ + xennet_end_access(queue->tx_ring_ref, queue->tx.sring); + xennet_end_access(queue->rx_ring_ref, queue->rx.sring); + + queue->tx_ring_ref = INVALID_GRANT_REF; + queue->rx_ring_ref = INVALID_GRANT_REF; + queue->tx.sring = NULL; + queue->rx.sring = NULL; + + page_pool_destroy(queue->page_pool); + } } -/** +/* * We are reconnecting to the backend, due to a suspend/resume, or a backend * driver restart. We tear down our netif structure and recreate it, but * leave the device-layer structures intact so that this is transparent to the @@ -1496,7 +1866,17 @@ static int netfront_resume(struct xenbus_device *dev) dev_dbg(&dev->dev, "%s\n", dev->nodename); + netif_tx_lock_bh(info->netdev); + netif_device_detach(info->netdev); + netif_tx_unlock_bh(info->netdev); + xennet_disconnect_backend(info); + + rtnl_lock(); + if (info->queues) + xennet_destroy_queues(info); + rtnl_unlock(); + return 0; } @@ -1522,158 +1902,357 @@ static int xen_net_read_mac(struct xenbus_device *dev, u8 mac[]) return 0; } -static int setup_netfront_single(struct netfront_info *info) +static int setup_netfront_single(struct netfront_queue *queue) { int err; - err = xenbus_alloc_evtchn(info->xbdev, &info->tx_evtchn); + err = xenbus_alloc_evtchn(queue->info->xbdev, &queue->tx_evtchn); if (err < 0) goto fail; - err = bind_evtchn_to_irqhandler(info->tx_evtchn, - xennet_interrupt, - 0, info->netdev->name, info); + err = bind_evtchn_to_irqhandler_lateeoi(queue->tx_evtchn, + xennet_interrupt, 0, + queue->info->netdev->name, + queue); if (err < 0) goto bind_fail; - info->rx_evtchn = info->tx_evtchn; - info->rx_irq = info->tx_irq = err; + queue->rx_evtchn = queue->tx_evtchn; + queue->rx_irq = queue->tx_irq = err; return 0; bind_fail: - xenbus_free_evtchn(info->xbdev, info->tx_evtchn); - info->tx_evtchn = 0; + xenbus_free_evtchn(queue->info->xbdev, queue->tx_evtchn); + queue->tx_evtchn = 0; fail: return err; } -static int setup_netfront_split(struct netfront_info *info) +static int setup_netfront_split(struct netfront_queue *queue) { int err; - err = xenbus_alloc_evtchn(info->xbdev, &info->tx_evtchn); + err = xenbus_alloc_evtchn(queue->info->xbdev, &queue->tx_evtchn); if (err < 0) goto fail; - err = xenbus_alloc_evtchn(info->xbdev, &info->rx_evtchn); + err = xenbus_alloc_evtchn(queue->info->xbdev, &queue->rx_evtchn); if (err < 0) goto alloc_rx_evtchn_fail; - snprintf(info->tx_irq_name, sizeof(info->tx_irq_name), - "%s-tx", info->netdev->name); - err = bind_evtchn_to_irqhandler(info->tx_evtchn, - xennet_tx_interrupt, - 0, info->tx_irq_name, info); + snprintf(queue->tx_irq_name, sizeof(queue->tx_irq_name), + "%s-tx", queue->name); + err = bind_evtchn_to_irqhandler_lateeoi(queue->tx_evtchn, + xennet_tx_interrupt, 0, + queue->tx_irq_name, queue); if (err < 0) goto bind_tx_fail; - info->tx_irq = err; + queue->tx_irq = err; - snprintf(info->rx_irq_name, sizeof(info->rx_irq_name), - "%s-rx", info->netdev->name); - err = bind_evtchn_to_irqhandler(info->rx_evtchn, - xennet_rx_interrupt, - 0, info->rx_irq_name, info); + snprintf(queue->rx_irq_name, sizeof(queue->rx_irq_name), + "%s-rx", queue->name); + err = bind_evtchn_to_irqhandler_lateeoi(queue->rx_evtchn, + xennet_rx_interrupt, 0, + queue->rx_irq_name, queue); if (err < 0) goto bind_rx_fail; - info->rx_irq = err; + queue->rx_irq = err; return 0; bind_rx_fail: - unbind_from_irqhandler(info->tx_irq, info); - info->tx_irq = 0; + unbind_from_irqhandler(queue->tx_irq, queue); + queue->tx_irq = 0; bind_tx_fail: - xenbus_free_evtchn(info->xbdev, info->rx_evtchn); - info->rx_evtchn = 0; + xenbus_free_evtchn(queue->info->xbdev, queue->rx_evtchn); + queue->rx_evtchn = 0; alloc_rx_evtchn_fail: - xenbus_free_evtchn(info->xbdev, info->tx_evtchn); - info->tx_evtchn = 0; + xenbus_free_evtchn(queue->info->xbdev, queue->tx_evtchn); + queue->tx_evtchn = 0; fail: return err; } -static int setup_netfront(struct xenbus_device *dev, struct netfront_info *info) +static int setup_netfront(struct xenbus_device *dev, + struct netfront_queue *queue, unsigned int feature_split_evtchn) { struct xen_netif_tx_sring *txs; struct xen_netif_rx_sring *rxs; int err; - struct net_device *netdev = info->netdev; - unsigned int feature_split_evtchn; - - info->tx_ring_ref = GRANT_INVALID_REF; - info->rx_ring_ref = GRANT_INVALID_REF; - info->rx.sring = NULL; - info->tx.sring = NULL; - netdev->irq = 0; - err = xenbus_scanf(XBT_NIL, info->xbdev->otherend, - "feature-split-event-channels", "%u", - &feature_split_evtchn); - if (err < 0) - feature_split_evtchn = 0; + queue->tx_ring_ref = INVALID_GRANT_REF; + queue->rx_ring_ref = INVALID_GRANT_REF; + queue->rx.sring = NULL; + queue->tx.sring = NULL; - err = xen_net_read_mac(dev, netdev->dev_addr); - if (err) { - xenbus_dev_fatal(dev, err, "parsing %s/mac", dev->nodename); + err = xenbus_setup_ring(dev, GFP_NOIO | __GFP_HIGH, (void **)&txs, + 1, &queue->tx_ring_ref); + if (err) goto fail; - } - txs = (struct xen_netif_tx_sring *)get_zeroed_page(GFP_NOIO | __GFP_HIGH); - if (!txs) { - err = -ENOMEM; - xenbus_dev_fatal(dev, err, "allocating tx ring page"); - goto fail; - } - SHARED_RING_INIT(txs); - FRONT_RING_INIT(&info->tx, txs, PAGE_SIZE); + XEN_FRONT_RING_INIT(&queue->tx, txs, XEN_PAGE_SIZE); - err = xenbus_grant_ring(dev, virt_to_mfn(txs)); - if (err < 0) - goto grant_tx_ring_fail; - - info->tx_ring_ref = err; - rxs = (struct xen_netif_rx_sring *)get_zeroed_page(GFP_NOIO | __GFP_HIGH); - if (!rxs) { - err = -ENOMEM; - xenbus_dev_fatal(dev, err, "allocating rx ring page"); - goto alloc_rx_ring_fail; - } - SHARED_RING_INIT(rxs); - FRONT_RING_INIT(&info->rx, rxs, PAGE_SIZE); + err = xenbus_setup_ring(dev, GFP_NOIO | __GFP_HIGH, (void **)&rxs, + 1, &queue->rx_ring_ref); + if (err) + goto fail; - err = xenbus_grant_ring(dev, virt_to_mfn(rxs)); - if (err < 0) - goto grant_rx_ring_fail; - info->rx_ring_ref = err; + XEN_FRONT_RING_INIT(&queue->rx, rxs, XEN_PAGE_SIZE); if (feature_split_evtchn) - err = setup_netfront_split(info); + err = setup_netfront_split(queue); /* setup single event channel if * a) feature-split-event-channels == 0 * b) feature-split-event-channels == 1 but failed to setup */ - if (!feature_split_evtchn || (feature_split_evtchn && err)) - err = setup_netfront_single(info); + if (!feature_split_evtchn || err) + err = setup_netfront_single(queue); if (err) - goto alloc_evtchn_fail; + goto fail; + + return 0; + + fail: + xenbus_teardown_ring((void **)&queue->rx.sring, 1, &queue->rx_ring_ref); + xenbus_teardown_ring((void **)&queue->tx.sring, 1, &queue->tx_ring_ref); + + return err; +} + +/* Queue-specific initialisation + * This used to be done in xennet_create_dev() but must now + * be run per-queue. + */ +static int xennet_init_queue(struct netfront_queue *queue) +{ + unsigned short i; + int err = 0; + char *devid; + + spin_lock_init(&queue->tx_lock); + spin_lock_init(&queue->rx_lock); + spin_lock_init(&queue->rx_cons_lock); + + timer_setup(&queue->rx_refill_timer, rx_refill_timeout, 0); + + devid = strrchr(queue->info->xbdev->nodename, '/') + 1; + snprintf(queue->name, sizeof(queue->name), "vif%s-q%u", + devid, queue->id); + + /* Initialise tx_skb_freelist as a free chain containing every entry. */ + queue->tx_skb_freelist = 0; + queue->tx_pend_queue = TX_LINK_NONE; + for (i = 0; i < NET_TX_RING_SIZE; i++) { + queue->tx_link[i] = i + 1; + queue->grant_tx_ref[i] = INVALID_GRANT_REF; + queue->grant_tx_page[i] = NULL; + } + queue->tx_link[NET_TX_RING_SIZE - 1] = TX_LINK_NONE; + + /* Clear out rx_skbs */ + for (i = 0; i < NET_RX_RING_SIZE; i++) { + queue->rx_skbs[i] = NULL; + queue->grant_rx_ref[i] = INVALID_GRANT_REF; + } + + /* A grant for every tx ring slot */ + if (gnttab_alloc_grant_references(NET_TX_RING_SIZE, + &queue->gref_tx_head) < 0) { + pr_alert("can't alloc tx grant refs\n"); + err = -ENOMEM; + goto exit; + } + + /* A grant for every rx ring slot */ + if (gnttab_alloc_grant_references(NET_RX_RING_SIZE, + &queue->gref_rx_head) < 0) { + pr_alert("can't alloc rx grant refs\n"); + err = -ENOMEM; + goto exit_free_tx; + } return 0; - /* If we fail to setup netfront, it is safe to just revoke access to - * granted pages because backend is not accessing it at this point. + exit_free_tx: + gnttab_free_grant_references(queue->gref_tx_head); + exit: + return err; +} + +static int write_queue_xenstore_keys(struct netfront_queue *queue, + struct xenbus_transaction *xbt, int write_hierarchical) +{ + /* Write the queue-specific keys into XenStore in the traditional + * way for a single queue, or in a queue subkeys for multiple + * queues. */ -alloc_evtchn_fail: - gnttab_end_foreign_access_ref(info->rx_ring_ref, 0); -grant_rx_ring_fail: - free_page((unsigned long)rxs); -alloc_rx_ring_fail: - gnttab_end_foreign_access_ref(info->tx_ring_ref, 0); -grant_tx_ring_fail: - free_page((unsigned long)txs); -fail: + struct xenbus_device *dev = queue->info->xbdev; + int err; + const char *message; + char *path; + size_t pathsize; + + /* Choose the correct place to write the keys */ + if (write_hierarchical) { + pathsize = strlen(dev->nodename) + 10; + path = kzalloc(pathsize, GFP_KERNEL); + if (!path) { + err = -ENOMEM; + message = "out of memory while writing ring references"; + goto error; + } + snprintf(path, pathsize, "%s/queue-%u", + dev->nodename, queue->id); + } else { + path = (char *)dev->nodename; + } + + /* Write ring references */ + err = xenbus_printf(*xbt, path, "tx-ring-ref", "%u", + queue->tx_ring_ref); + if (err) { + message = "writing tx-ring-ref"; + goto error; + } + + err = xenbus_printf(*xbt, path, "rx-ring-ref", "%u", + queue->rx_ring_ref); + if (err) { + message = "writing rx-ring-ref"; + goto error; + } + + /* Write event channels; taking into account both shared + * and split event channel scenarios. + */ + if (queue->tx_evtchn == queue->rx_evtchn) { + /* Shared event channel */ + err = xenbus_printf(*xbt, path, + "event-channel", "%u", queue->tx_evtchn); + if (err) { + message = "writing event-channel"; + goto error; + } + } else { + /* Split event channels */ + err = xenbus_printf(*xbt, path, + "event-channel-tx", "%u", queue->tx_evtchn); + if (err) { + message = "writing event-channel-tx"; + goto error; + } + + err = xenbus_printf(*xbt, path, + "event-channel-rx", "%u", queue->rx_evtchn); + if (err) { + message = "writing event-channel-rx"; + goto error; + } + } + + if (write_hierarchical) + kfree(path); + return 0; + +error: + if (write_hierarchical) + kfree(path); + xenbus_dev_fatal(dev, err, "%s", message); + return err; +} + + + +static int xennet_create_page_pool(struct netfront_queue *queue) +{ + int err; + struct page_pool_params pp_params = { + .order = 0, + .flags = 0, + .pool_size = NET_RX_RING_SIZE, + .nid = NUMA_NO_NODE, + .dev = &queue->info->netdev->dev, + .offset = XDP_PACKET_HEADROOM, + .max_len = XEN_PAGE_SIZE - XDP_PACKET_HEADROOM, + }; + + queue->page_pool = page_pool_create(&pp_params); + if (IS_ERR(queue->page_pool)) { + err = PTR_ERR(queue->page_pool); + queue->page_pool = NULL; + return err; + } + + err = xdp_rxq_info_reg(&queue->xdp_rxq, queue->info->netdev, + queue->id, 0); + if (err) { + netdev_err(queue->info->netdev, "xdp_rxq_info_reg failed\n"); + goto err_free_pp; + } + + err = xdp_rxq_info_reg_mem_model(&queue->xdp_rxq, + MEM_TYPE_PAGE_POOL, queue->page_pool); + if (err) { + netdev_err(queue->info->netdev, "xdp_rxq_info_reg_mem_model failed\n"); + goto err_unregister_rxq; + } + return 0; + +err_unregister_rxq: + xdp_rxq_info_unreg(&queue->xdp_rxq); +err_free_pp: + page_pool_destroy(queue->page_pool); + queue->page_pool = NULL; return err; } +static int xennet_create_queues(struct netfront_info *info, + unsigned int *num_queues) +{ + unsigned int i; + int ret; + + info->queues = kcalloc(*num_queues, sizeof(struct netfront_queue), + GFP_KERNEL); + if (!info->queues) + return -ENOMEM; + + for (i = 0; i < *num_queues; i++) { + struct netfront_queue *queue = &info->queues[i]; + + queue->id = i; + queue->info = info; + + ret = xennet_init_queue(queue); + if (ret < 0) { + dev_warn(&info->xbdev->dev, + "only created %d queues\n", i); + *num_queues = i; + break; + } + + /* use page pool recycling instead of buddy allocator */ + ret = xennet_create_page_pool(queue); + if (ret < 0) { + dev_err(&info->xbdev->dev, "can't allocate page pool\n"); + *num_queues = i; + return ret; + } + + netif_napi_add(queue->info->netdev, &queue->napi, xennet_poll); + if (netif_running(info->netdev)) + napi_enable(&queue->napi); + } + + netif_set_real_num_tx_queues(info->netdev, *num_queues); + + if (*num_queues == 0) { + dev_err(&info->xbdev->dev, "no queues\n"); + return -EINVAL; + } + return 0; +} + /* Common code used when first setting up, and when resuming. */ static int talk_to_netback(struct xenbus_device *dev, struct netfront_info *info) @@ -1681,11 +2260,70 @@ static int talk_to_netback(struct xenbus_device *dev, const char *message; struct xenbus_transaction xbt; int err; + unsigned int feature_split_evtchn; + unsigned int i = 0; + unsigned int max_queues = 0; + struct netfront_queue *queue = NULL; + unsigned int num_queues = 1; + u8 addr[ETH_ALEN]; - /* Create shared ring, alloc event channel. */ - err = setup_netfront(dev, info); - if (err) + info->netdev->irq = 0; + + /* Check if backend is trusted. */ + info->bounce = !xennet_trusted || + !xenbus_read_unsigned(dev->nodename, "trusted", 1); + + /* Check if backend supports multiple queues */ + max_queues = xenbus_read_unsigned(info->xbdev->otherend, + "multi-queue-max-queues", 1); + num_queues = min(max_queues, xennet_max_queues); + + /* Check feature-split-event-channels */ + feature_split_evtchn = xenbus_read_unsigned(info->xbdev->otherend, + "feature-split-event-channels", 0); + + /* Read mac addr. */ + err = xen_net_read_mac(dev, addr); + if (err) { + xenbus_dev_fatal(dev, err, "parsing %s/mac", dev->nodename); + goto out_unlocked; + } + eth_hw_addr_set(info->netdev, addr); + + info->netback_has_xdp_headroom = xenbus_read_unsigned(info->xbdev->otherend, + "feature-xdp-headroom", 0); + if (info->netback_has_xdp_headroom) { + /* set the current xen-netfront xdp state */ + err = talk_to_netback_xdp(info, info->netfront_xdp_enabled ? + NETBACK_XDP_HEADROOM_ENABLE : + NETBACK_XDP_HEADROOM_DISABLE); + if (err) + goto out_unlocked; + } + + rtnl_lock(); + if (info->queues) + xennet_destroy_queues(info); + + /* For the case of a reconnect reset the "broken" indicator. */ + info->broken = false; + + err = xennet_create_queues(info, &num_queues); + if (err < 0) { + xenbus_dev_fatal(dev, err, "creating queues"); + kfree(info->queues); + info->queues = NULL; goto out; + } + rtnl_unlock(); + + /* Create shared ring, alloc event channel -- for each queue */ + for (i = 0; i < num_queues; ++i) { + queue = &info->queues[i]; + err = setup_netfront(dev, queue, feature_split_evtchn); + if (err) + goto destroy_ring; + } again: err = xenbus_transaction_start(&xbt); @@ -1694,41 +2332,32 @@ again: goto destroy_ring; } - err = xenbus_printf(xbt, dev->nodename, "tx-ring-ref", "%u", - info->tx_ring_ref); - if (err) { - message = "writing tx ring-ref"; - goto abort_transaction; - } - err = xenbus_printf(xbt, dev->nodename, "rx-ring-ref", "%u", - info->rx_ring_ref); - if (err) { - message = "writing rx ring-ref"; - goto abort_transaction; - } - - if (info->tx_evtchn == info->rx_evtchn) { + if (xenbus_exists(XBT_NIL, + info->xbdev->otherend, "multi-queue-max-queues")) { + /* Write the number of queues */ err = xenbus_printf(xbt, dev->nodename, - "event-channel", "%u", info->tx_evtchn); + "multi-queue-num-queues", "%u", num_queues); if (err) { - message = "writing event-channel"; - goto abort_transaction; + message = "writing multi-queue-num-queues"; + goto abort_transaction_no_dev_fatal; } + } + + if (num_queues == 1) { + err = write_queue_xenstore_keys(&info->queues[0], &xbt, 0); /* flat */ + if (err) + goto abort_transaction_no_dev_fatal; } else { - err = xenbus_printf(xbt, dev->nodename, - "event-channel-tx", "%u", info->tx_evtchn); - if (err) { - message = "writing event-channel-tx"; - goto abort_transaction; - } - err = xenbus_printf(xbt, dev->nodename, - "event-channel-rx", "%u", info->rx_evtchn); - if (err) { - message = "writing event-channel-rx"; - goto abort_transaction; + /* Write the keys for each queue */ + for (i = 0; i < num_queues; ++i) { + queue = &info->queues[i]; + err = write_queue_xenstore_keys(queue, &xbt, 1); /* hierarchical */ + if (err) + goto abort_transaction_no_dev_fatal; } } + /* The remaining keys are not queue-specific */ err = xenbus_printf(xbt, dev->nodename, "request-rx-copy", "%u", 1); if (err) { @@ -1754,6 +2383,19 @@ again: goto abort_transaction; } + err = xenbus_write(xbt, dev->nodename, "feature-gso-tcpv6", "1"); + if (err) { + message = "writing feature-gso-tcpv6"; + goto abort_transaction; + } + + err = xenbus_write(xbt, dev->nodename, "feature-ipv6-csum-offload", + "1"); + if (err) { + message = "writing feature-ipv6-csum-offload"; + goto abort_transaction; + } + err = xenbus_transaction_end(xbt, 0); if (err) { if (err == -EAGAIN) @@ -1765,29 +2407,29 @@ again: return 0; abort_transaction: - xenbus_transaction_end(xbt, 1); xenbus_dev_fatal(dev, err, "%s", message); +abort_transaction_no_dev_fatal: + xenbus_transaction_end(xbt, 1); destroy_ring: xennet_disconnect_backend(info); + rtnl_lock(); + xennet_destroy_queues(info); out: + rtnl_unlock(); +out_unlocked: + device_unregister(&dev->dev); return err; } static int xennet_connect(struct net_device *dev) { struct netfront_info *np = netdev_priv(dev); - int i, requeue_idx, err; - struct sk_buff *skb; - grant_ref_t ref; - struct xen_netif_rx_request *req; - unsigned int feature_rx_copy; - - err = xenbus_scanf(XBT_NIL, np->xbdev->otherend, - "feature-rx-copy", "%u", &feature_rx_copy); - if (err != 1) - feature_rx_copy = 0; + unsigned int num_queues = 0; + int err; + unsigned int j = 0; + struct netfront_queue *queue = NULL; - if (!feature_rx_copy) { + if (!xenbus_read_unsigned(np->xbdev->otherend, "feature-rx-copy", 0)) { dev_info(&dev->dev, "backend does not support copying receive path\n"); return -ENODEV; @@ -1796,62 +2438,55 @@ static int xennet_connect(struct net_device *dev) err = talk_to_netback(np->xbdev, np); if (err) return err; + if (np->netback_has_xdp_headroom) + pr_info("backend supports XDP headroom\n"); + if (np->bounce) + dev_info(&np->xbdev->dev, + "bouncing transmitted data to zeroed pages\n"); - rtnl_lock(); - netdev_update_features(dev); - rtnl_unlock(); - - spin_lock_bh(&np->rx_lock); - spin_lock_irq(&np->tx_lock); - - /* Step 1: Discard all pending TX packet fragments. */ - xennet_release_tx_bufs(np); - - /* Step 2: Rebuild the RX buffer freelist and the RX ring itself. */ - for (requeue_idx = 0, i = 0; i < NET_RX_RING_SIZE; i++) { - skb_frag_t *frag; - const struct page *page; - if (!np->rx_skbs[i]) - continue; - - skb = np->rx_skbs[requeue_idx] = xennet_get_rx_skb(np, i); - ref = np->grant_rx_ref[requeue_idx] = xennet_get_rx_ref(np, i); - req = RING_GET_REQUEST(&np->rx, requeue_idx); + /* talk_to_netback() sets the correct number of queues */ + num_queues = dev->real_num_tx_queues; - frag = &skb_shinfo(skb)->frags[0]; - page = skb_frag_page(frag); - gnttab_grant_foreign_access_ref( - ref, np->xbdev->otherend_id, - pfn_to_mfn(page_to_pfn(page)), - 0); - req->gref = ref; - req->id = requeue_idx; - - requeue_idx++; + if (dev->reg_state == NETREG_UNINITIALIZED) { + err = register_netdev(dev); + if (err) { + pr_warn("%s: register_netdev err=%d\n", __func__, err); + device_unregister(&np->xbdev->dev); + return err; + } } - np->rx.req_prod_pvt = requeue_idx; + rtnl_lock(); + netdev_update_features(dev); + rtnl_unlock(); /* - * Step 3: All public and private state should now be sane. Get + * All public and private state should now be sane. Get * ready to start sending and receiving packets and give the driver * domain a kick because we've probably just requeued some * packets. */ + netif_tx_lock_bh(np->netdev); + netif_device_attach(np->netdev); + netif_tx_unlock_bh(np->netdev); + netif_carrier_on(np->netdev); - notify_remote_via_irq(np->tx_irq); - if (np->tx_irq != np->rx_irq) - notify_remote_via_irq(np->rx_irq); - xennet_tx_buf_gc(dev); - xennet_alloc_rx_buffers(dev); + for (j = 0; j < num_queues; ++j) { + queue = &np->queues[j]; - spin_unlock_irq(&np->tx_lock); - spin_unlock_bh(&np->rx_lock); + notify_remote_via_irq(queue->tx_irq); + if (queue->tx_irq != queue->rx_irq) + notify_remote_via_irq(queue->rx_irq); + + spin_lock_bh(&queue->rx_lock); + xennet_alloc_rx_buffers(queue); + spin_unlock_bh(&queue->rx_lock); + } return 0; } -/** +/* * Callback received when the backend's state changes. */ static void netback_changed(struct xenbus_device *dev, @@ -1862,13 +2497,14 @@ static void netback_changed(struct xenbus_device *dev, dev_dbg(&dev->dev, "%s\n", xenbus_strstate(backend_state)); + wake_up_all(&module_wq); + switch (backend_state) { case XenbusStateInitialising: case XenbusStateInitialised: case XenbusStateReconfiguring: case XenbusStateReconfigured: case XenbusStateUnknown: - case XenbusStateClosed: break; case XenbusStateInitWait: @@ -1883,6 +2519,10 @@ static void netback_changed(struct xenbus_device *dev, netdev_notify_peers(netdev); break; + case XenbusStateClosed: + if (dev->state == XenbusStateClosed) + break; + fallthrough; /* Missed the backend's CLOSING state */ case XenbusStateClosing: xenbus_frontend_closed(dev); break; @@ -1916,7 +2556,7 @@ static void xennet_get_ethtool_stats(struct net_device *dev, int i; for (i = 0; i < ARRAY_SIZE(xennet_stats); i++) - data[i] = *(unsigned long *)(np + xennet_stats[i].offset); + data[i] = atomic_read((atomic_t *)(np + xennet_stats[i].offset)); } static void xennet_get_strings(struct net_device *dev, u32 stringset, u8 * data) @@ -1939,183 +2579,131 @@ static const struct ethtool_ops xennet_ethtool_ops = .get_sset_count = xennet_get_sset_count, .get_ethtool_stats = xennet_get_ethtool_stats, .get_strings = xennet_get_strings, + .get_ts_info = ethtool_op_get_ts_info, }; #ifdef CONFIG_SYSFS -static ssize_t show_rxbuf_min(struct device *dev, - struct device_attribute *attr, char *buf) +static ssize_t show_rxbuf(struct device *dev, + struct device_attribute *attr, char *buf) { - struct net_device *netdev = to_net_dev(dev); - struct netfront_info *info = netdev_priv(netdev); - - return sprintf(buf, "%u\n", info->rx_min_target); + return sprintf(buf, "%lu\n", NET_RX_RING_SIZE); } -static ssize_t store_rxbuf_min(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t len) +static ssize_t store_rxbuf(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t len) { - struct net_device *netdev = to_net_dev(dev); - struct netfront_info *np = netdev_priv(netdev); char *endp; - unsigned long target; if (!capable(CAP_NET_ADMIN)) return -EPERM; - target = simple_strtoul(buf, &endp, 0); + simple_strtoul(buf, &endp, 0); if (endp == buf) return -EBADMSG; - if (target < RX_MIN_TARGET) - target = RX_MIN_TARGET; - if (target > RX_MAX_TARGET) - target = RX_MAX_TARGET; - - spin_lock_bh(&np->rx_lock); - if (target > np->rx_max_target) - np->rx_max_target = target; - np->rx_min_target = target; - if (target > np->rx_target) - np->rx_target = target; - - xennet_alloc_rx_buffers(netdev); + /* rxbuf_min and rxbuf_max are no longer configurable. */ - spin_unlock_bh(&np->rx_lock); return len; } -static ssize_t show_rxbuf_max(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct net_device *netdev = to_net_dev(dev); - struct netfront_info *info = netdev_priv(netdev); - - return sprintf(buf, "%u\n", info->rx_max_target); -} +static DEVICE_ATTR(rxbuf_min, 0644, show_rxbuf, store_rxbuf); +static DEVICE_ATTR(rxbuf_max, 0644, show_rxbuf, store_rxbuf); +static DEVICE_ATTR(rxbuf_cur, 0444, show_rxbuf, NULL); -static ssize_t store_rxbuf_max(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t len) -{ - struct net_device *netdev = to_net_dev(dev); - struct netfront_info *np = netdev_priv(netdev); - char *endp; - unsigned long target; - - if (!capable(CAP_NET_ADMIN)) - return -EPERM; - - target = simple_strtoul(buf, &endp, 0); - if (endp == buf) - return -EBADMSG; +static struct attribute *xennet_dev_attrs[] = { + &dev_attr_rxbuf_min.attr, + &dev_attr_rxbuf_max.attr, + &dev_attr_rxbuf_cur.attr, + NULL +}; - if (target < RX_MIN_TARGET) - target = RX_MIN_TARGET; - if (target > RX_MAX_TARGET) - target = RX_MAX_TARGET; +static const struct attribute_group xennet_dev_group = { + .attrs = xennet_dev_attrs +}; +#endif /* CONFIG_SYSFS */ - spin_lock_bh(&np->rx_lock); - if (target < np->rx_min_target) - np->rx_min_target = target; - np->rx_max_target = target; - if (target < np->rx_target) - np->rx_target = target; +static void xennet_bus_close(struct xenbus_device *dev) +{ + int ret; - xennet_alloc_rx_buffers(netdev); + if (xenbus_read_driver_state(dev->otherend) == XenbusStateClosed) + return; + do { + xenbus_switch_state(dev, XenbusStateClosing); + ret = wait_event_timeout(module_wq, + xenbus_read_driver_state(dev->otherend) == + XenbusStateClosing || + xenbus_read_driver_state(dev->otherend) == + XenbusStateClosed || + xenbus_read_driver_state(dev->otherend) == + XenbusStateUnknown, + XENNET_TIMEOUT); + } while (!ret); + + if (xenbus_read_driver_state(dev->otherend) == XenbusStateClosed) + return; - spin_unlock_bh(&np->rx_lock); - return len; + do { + xenbus_switch_state(dev, XenbusStateClosed); + ret = wait_event_timeout(module_wq, + xenbus_read_driver_state(dev->otherend) == + XenbusStateClosed || + xenbus_read_driver_state(dev->otherend) == + XenbusStateUnknown, + XENNET_TIMEOUT); + } while (!ret); } -static ssize_t show_rxbuf_cur(struct device *dev, - struct device_attribute *attr, char *buf) +static void xennet_remove(struct xenbus_device *dev) { - struct net_device *netdev = to_net_dev(dev); - struct netfront_info *info = netdev_priv(netdev); - - return sprintf(buf, "%u\n", info->rx_target); -} + struct netfront_info *info = dev_get_drvdata(&dev->dev); -static struct device_attribute xennet_attrs[] = { - __ATTR(rxbuf_min, S_IRUGO|S_IWUSR, show_rxbuf_min, store_rxbuf_min), - __ATTR(rxbuf_max, S_IRUGO|S_IWUSR, show_rxbuf_max, store_rxbuf_max), - __ATTR(rxbuf_cur, S_IRUGO, show_rxbuf_cur, NULL), -}; + xennet_bus_close(dev); + xennet_disconnect_backend(info); -static int xennet_sysfs_addif(struct net_device *netdev) -{ - int i; - int err; + if (info->netdev->reg_state == NETREG_REGISTERED) + unregister_netdev(info->netdev); - for (i = 0; i < ARRAY_SIZE(xennet_attrs); i++) { - err = device_create_file(&netdev->dev, - &xennet_attrs[i]); - if (err) - goto fail; + if (info->queues) { + rtnl_lock(); + xennet_destroy_queues(info); + rtnl_unlock(); } - return 0; - - fail: - while (--i >= 0) - device_remove_file(&netdev->dev, &xennet_attrs[i]); - return err; -} - -static void xennet_sysfs_delif(struct net_device *netdev) -{ - int i; - - for (i = 0; i < ARRAY_SIZE(xennet_attrs); i++) - device_remove_file(&netdev->dev, &xennet_attrs[i]); + xennet_free_netdev(info->netdev); } -#endif /* CONFIG_SYSFS */ - static const struct xenbus_device_id netfront_ids[] = { { "vif" }, { "" } }; - -static int xennet_remove(struct xenbus_device *dev) -{ - struct netfront_info *info = dev_get_drvdata(&dev->dev); - - dev_dbg(&dev->dev, "%s\n", dev->nodename); - - xennet_disconnect_backend(info); - - xennet_sysfs_delif(info->netdev); - - unregister_netdev(info->netdev); - - del_timer_sync(&info->rx_refill_timer); - - free_percpu(info->stats); - - free_netdev(info->netdev); - - return 0; -} - -static DEFINE_XENBUS_DRIVER(netfront, , +static struct xenbus_driver netfront_driver = { + .ids = netfront_ids, .probe = netfront_probe, .remove = xennet_remove, .resume = netfront_resume, .otherend_changed = netback_changed, -); +}; static int __init netif_init(void) { if (!xen_domain()) return -ENODEV; - if (xen_hvm_domain() && !xen_platform_pci_unplug) + if (!xen_has_pv_nic_devices()) return -ENODEV; pr_info("Initialising Xen virtual ethernet driver\n"); + /* Allow the number of queues to match the number of CPUs, but not exceed + * the maximum limit. If the user has not specified a value, the default + * maximum limit is 8. + */ + if (xennet_max_queues == 0) + xennet_max_queues = min_t(unsigned int, MAX_QUEUES_DEFAULT, + num_online_cpus()); + return xenbus_register_frontend(&netfront_driver); } module_init(netif_init); |