2 * Virtual network driver for conversing with remote driver backends.
4 * Copyright (c) 2002-2005, K A Fraser
5 * Copyright (c) 2005, XenSource Ltd
7 * This program is free software; you can redistribute it and/or
8 * modify it under the terms of the GNU General Public License version 2
9 * as published by the Free Software Foundation; or, when distributed
10 * separately from the Linux kernel or incorporated into other
11 * software packages, subject to the following license:
13 * Permission is hereby granted, free of charge, to any person obtaining a copy
14 * of this source file (the "Software"), to deal in the Software without
15 * restriction, including without limitation the rights to use, copy, modify,
16 * merge, publish, distribute, sublicense, and/or sell copies of the Software,
17 * and to permit persons to whom the Software is furnished to do so, subject to
18 * the following conditions:
20 * The above copyright notice and this permission notice shall be included in
21 * all copies or substantial portions of the Software.
23 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
24 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
25 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
26 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
27 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
28 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
32 #include <linux/module.h>
33 #include <linux/kernel.h>
34 #include <linux/netdevice.h>
35 #include <linux/etherdevice.h>
36 #include <linux/skbuff.h>
37 #include <linux/ethtool.h>
38 #include <linux/if_ether.h>
39 #include <linux/tcp.h>
40 #include <linux/udp.h>
41 #include <linux/moduleparam.h>
43 #include <linux/slab.h>
46 #include <asm/xen/page.h>
48 #include <xen/xenbus.h>
49 #include <xen/events.h>
51 #include <xen/platform_pci.h>
52 #include <xen/grant_table.h>
54 #include <xen/interface/io/netif.h>
55 #include <xen/interface/memory.h>
56 #include <xen/interface/grant_table.h>
58 static const struct ethtool_ops xennet_ethtool_ops;
64 #define NETFRONT_SKB_CB(skb) ((struct netfront_cb *)((skb)->cb))
66 #define RX_COPY_THRESHOLD 256
68 #define GRANT_INVALID_REF 0
70 #define NET_TX_RING_SIZE __CONST_RING_SIZE(xen_netif_tx, PAGE_SIZE)
71 #define NET_RX_RING_SIZE __CONST_RING_SIZE(xen_netif_rx, PAGE_SIZE)
72 #define TX_MAX_TARGET min_t(int, NET_TX_RING_SIZE, 256)
74 struct netfront_stats {
79 struct u64_stats_sync syncp;
82 struct netfront_info {
83 struct list_head list;
84 struct net_device *netdev;
86 struct napi_struct napi;
89 struct xenbus_device *xbdev;
92 struct xen_netif_tx_front_ring tx;
96 * {tx,rx}_skbs store outstanding skbuffs. Free tx_skb entries
97 * are linked from tx_skb_freelist through skb_entry.link.
99 * NB. Freelist index entries are always going to be less than
100 * PAGE_OFFSET, whereas pointers to skbs will always be equal or
101 * greater than PAGE_OFFSET: we use this property to distinguish
107 } tx_skbs[NET_TX_RING_SIZE];
108 grant_ref_t gref_tx_head;
109 grant_ref_t grant_tx_ref[NET_TX_RING_SIZE];
110 unsigned tx_skb_freelist;
112 spinlock_t rx_lock ____cacheline_aligned_in_smp;
113 struct xen_netif_rx_front_ring rx;
116 /* Receive-ring batched refills. */
117 #define RX_MIN_TARGET 8
118 #define RX_DFL_MIN_TARGET 64
119 #define RX_MAX_TARGET min_t(int, NET_RX_RING_SIZE, 256)
120 unsigned rx_min_target, rx_max_target, rx_target;
121 struct sk_buff_head rx_batch;
123 struct timer_list rx_refill_timer;
125 struct sk_buff *rx_skbs[NET_RX_RING_SIZE];
126 grant_ref_t gref_rx_head;
127 grant_ref_t grant_rx_ref[NET_RX_RING_SIZE];
129 unsigned long rx_pfn_array[NET_RX_RING_SIZE];
130 struct multicall_entry rx_mcl[NET_RX_RING_SIZE+1];
131 struct mmu_update rx_mmu[NET_RX_RING_SIZE];
134 struct netfront_stats __percpu *stats;
136 unsigned long rx_gso_checksum_fixup;
139 struct netfront_rx_info {
140 struct xen_netif_rx_response rx;
141 struct xen_netif_extra_info extras[XEN_NETIF_EXTRA_TYPE_MAX - 1];
144 static void skb_entry_set_link(union skb_entry *list, unsigned short id)
149 static int skb_entry_is_link(const union skb_entry *list)
151 BUILD_BUG_ON(sizeof(list->skb) != sizeof(list->link));
152 return (unsigned long)list->skb < PAGE_OFFSET;
156 * Access macros for acquiring freeing slots in tx_skbs[].
159 static void add_id_to_freelist(unsigned *head, union skb_entry *list,
162 skb_entry_set_link(&list[id], *head);
166 static unsigned short get_id_from_freelist(unsigned *head,
167 union skb_entry *list)
169 unsigned int id = *head;
170 *head = list[id].link;
174 static int xennet_rxidx(RING_IDX idx)
176 return idx & (NET_RX_RING_SIZE - 1);
179 static struct sk_buff *xennet_get_rx_skb(struct netfront_info *np,
182 int i = xennet_rxidx(ri);
183 struct sk_buff *skb = np->rx_skbs[i];
184 np->rx_skbs[i] = NULL;
188 static grant_ref_t xennet_get_rx_ref(struct netfront_info *np,
191 int i = xennet_rxidx(ri);
192 grant_ref_t ref = np->grant_rx_ref[i];
193 np->grant_rx_ref[i] = GRANT_INVALID_REF;
198 static int xennet_sysfs_addif(struct net_device *netdev);
199 static void xennet_sysfs_delif(struct net_device *netdev);
200 #else /* !CONFIG_SYSFS */
201 #define xennet_sysfs_addif(dev) (0)
202 #define xennet_sysfs_delif(dev) do { } while (0)
205 static bool xennet_can_sg(struct net_device *dev)
207 return dev->features & NETIF_F_SG;
211 static void rx_refill_timeout(unsigned long data)
213 struct net_device *dev = (struct net_device *)data;
214 struct netfront_info *np = netdev_priv(dev);
215 napi_schedule(&np->napi);
218 static int netfront_tx_slot_available(struct netfront_info *np)
220 return (np->tx.req_prod_pvt - np->tx.rsp_cons) <
221 (TX_MAX_TARGET - MAX_SKB_FRAGS - 2);
224 static void xennet_maybe_wake_tx(struct net_device *dev)
226 struct netfront_info *np = netdev_priv(dev);
228 if (unlikely(netif_queue_stopped(dev)) &&
229 netfront_tx_slot_available(np) &&
230 likely(netif_running(dev)))
231 netif_wake_queue(dev);
234 static void xennet_alloc_rx_buffers(struct net_device *dev)
237 struct netfront_info *np = netdev_priv(dev);
240 int i, batch_target, notify;
241 RING_IDX req_prod = np->rx.req_prod_pvt;
245 struct xen_netif_rx_request *req;
247 if (unlikely(!netif_carrier_ok(dev)))
251 * Allocate skbuffs greedily, even though we batch updates to the
252 * receive ring. This creates a less bursty demand on the memory
253 * allocator, so should reduce the chance of failed allocation requests
254 * both for ourself and for other kernel subsystems.
256 batch_target = np->rx_target - (req_prod - np->rx.rsp_cons);
257 for (i = skb_queue_len(&np->rx_batch); i < batch_target; i++) {
258 skb = __netdev_alloc_skb(dev, RX_COPY_THRESHOLD + NET_IP_ALIGN,
259 GFP_ATOMIC | __GFP_NOWARN);
263 /* Align ip header to a 16 bytes boundary */
264 skb_reserve(skb, NET_IP_ALIGN);
266 page = alloc_page(GFP_ATOMIC | __GFP_NOWARN);
270 /* Any skbuffs queued for refill? Force them out. */
273 /* Could not allocate any skbuffs. Try again later. */
274 mod_timer(&np->rx_refill_timer,
279 __skb_fill_page_desc(skb, 0, page, 0, 0);
280 skb_shinfo(skb)->nr_frags = 1;
281 __skb_queue_tail(&np->rx_batch, skb);
284 /* Is the batch large enough to be worthwhile? */
285 if (i < (np->rx_target/2)) {
286 if (req_prod > np->rx.sring->req_prod)
291 /* Adjust our fill target if we risked running out of buffers. */
292 if (((req_prod - np->rx.sring->rsp_prod) < (np->rx_target / 4)) &&
293 ((np->rx_target *= 2) > np->rx_max_target))
294 np->rx_target = np->rx_max_target;
298 skb = __skb_dequeue(&np->rx_batch);
304 id = xennet_rxidx(req_prod + i);
306 BUG_ON(np->rx_skbs[id]);
307 np->rx_skbs[id] = skb;
309 ref = gnttab_claim_grant_reference(&np->gref_rx_head);
310 BUG_ON((signed short)ref < 0);
311 np->grant_rx_ref[id] = ref;
313 pfn = page_to_pfn(skb_frag_page(&skb_shinfo(skb)->frags[0]));
314 vaddr = page_address(skb_frag_page(&skb_shinfo(skb)->frags[0]));
316 req = RING_GET_REQUEST(&np->rx, req_prod + i);
317 gnttab_grant_foreign_access_ref(ref,
318 np->xbdev->otherend_id,
326 wmb(); /* barrier so backend seens requests */
328 /* Above is a suitable barrier to ensure backend will see requests. */
329 np->rx.req_prod_pvt = req_prod + i;
331 RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&np->rx, notify);
333 notify_remote_via_irq(np->netdev->irq);
336 static int xennet_open(struct net_device *dev)
338 struct netfront_info *np = netdev_priv(dev);
340 napi_enable(&np->napi);
342 spin_lock_bh(&np->rx_lock);
343 if (netif_carrier_ok(dev)) {
344 xennet_alloc_rx_buffers(dev);
345 np->rx.sring->rsp_event = np->rx.rsp_cons + 1;
346 if (RING_HAS_UNCONSUMED_RESPONSES(&np->rx))
347 napi_schedule(&np->napi);
349 spin_unlock_bh(&np->rx_lock);
351 netif_start_queue(dev);
356 static void xennet_tx_buf_gc(struct net_device *dev)
360 struct netfront_info *np = netdev_priv(dev);
363 BUG_ON(!netif_carrier_ok(dev));
366 prod = np->tx.sring->rsp_prod;
367 rmb(); /* Ensure we see responses up to 'rp'. */
369 for (cons = np->tx.rsp_cons; cons != prod; cons++) {
370 struct xen_netif_tx_response *txrsp;
372 txrsp = RING_GET_RESPONSE(&np->tx, cons);
373 if (txrsp->status == XEN_NETIF_RSP_NULL)
377 skb = np->tx_skbs[id].skb;
378 if (unlikely(gnttab_query_foreign_access(
379 np->grant_tx_ref[id]) != 0)) {
380 printk(KERN_ALERT "xennet_tx_buf_gc: warning "
381 "-- grant still in use by backend "
385 gnttab_end_foreign_access_ref(
386 np->grant_tx_ref[id], GNTMAP_readonly);
387 gnttab_release_grant_reference(
388 &np->gref_tx_head, np->grant_tx_ref[id]);
389 np->grant_tx_ref[id] = GRANT_INVALID_REF;
390 add_id_to_freelist(&np->tx_skb_freelist, np->tx_skbs, id);
391 dev_kfree_skb_irq(skb);
394 np->tx.rsp_cons = prod;
397 * Set a new event, then check for race with update of tx_cons.
398 * Note that it is essential to schedule a callback, no matter
399 * how few buffers are pending. Even if there is space in the
400 * transmit ring, higher layers may be blocked because too much
401 * data is outstanding: in such cases notification from Xen is
402 * likely to be the only kick that we'll get.
404 np->tx.sring->rsp_event =
405 prod + ((np->tx.sring->req_prod - prod) >> 1) + 1;
406 mb(); /* update shared area */
407 } while ((cons == prod) && (prod != np->tx.sring->rsp_prod));
409 xennet_maybe_wake_tx(dev);
412 static void xennet_make_frags(struct sk_buff *skb, struct net_device *dev,
413 struct xen_netif_tx_request *tx)
415 struct netfront_info *np = netdev_priv(dev);
416 char *data = skb->data;
418 RING_IDX prod = np->tx.req_prod_pvt;
419 int frags = skb_shinfo(skb)->nr_frags;
420 unsigned int offset = offset_in_page(data);
421 unsigned int len = skb_headlen(skb);
426 /* While the header overlaps a page boundary (including being
427 larger than a page), split it it into page-sized chunks. */
428 while (len > PAGE_SIZE - offset) {
429 tx->size = PAGE_SIZE - offset;
430 tx->flags |= XEN_NETTXF_more_data;
435 id = get_id_from_freelist(&np->tx_skb_freelist, np->tx_skbs);
436 np->tx_skbs[id].skb = skb_get(skb);
437 tx = RING_GET_REQUEST(&np->tx, prod++);
439 ref = gnttab_claim_grant_reference(&np->gref_tx_head);
440 BUG_ON((signed short)ref < 0);
442 mfn = virt_to_mfn(data);
443 gnttab_grant_foreign_access_ref(ref, np->xbdev->otherend_id,
444 mfn, GNTMAP_readonly);
446 tx->gref = np->grant_tx_ref[id] = ref;
452 /* Grant backend access to each skb fragment page. */
453 for (i = 0; i < frags; i++) {
454 skb_frag_t *frag = skb_shinfo(skb)->frags + i;
455 struct page *page = skb_frag_page(frag);
457 len = skb_frag_size(frag);
458 offset = frag->page_offset;
460 /* Data must not cross a page boundary. */
461 BUG_ON(len + offset > PAGE_SIZE<<compound_order(page));
463 /* Skip unused frames from start of page */
464 page += offset >> PAGE_SHIFT;
465 offset &= ~PAGE_MASK;
470 BUG_ON(offset >= PAGE_SIZE);
472 bytes = PAGE_SIZE - offset;
476 tx->flags |= XEN_NETTXF_more_data;
478 id = get_id_from_freelist(&np->tx_skb_freelist,
480 np->tx_skbs[id].skb = skb_get(skb);
481 tx = RING_GET_REQUEST(&np->tx, prod++);
483 ref = gnttab_claim_grant_reference(&np->gref_tx_head);
484 BUG_ON((signed short)ref < 0);
486 mfn = pfn_to_mfn(page_to_pfn(page));
487 gnttab_grant_foreign_access_ref(ref,
488 np->xbdev->otherend_id,
489 mfn, GNTMAP_readonly);
491 tx->gref = np->grant_tx_ref[id] = ref;
500 if (offset == PAGE_SIZE && len) {
501 BUG_ON(!PageCompound(page));
508 np->tx.req_prod_pvt = prod;
512 * Count how many ring slots are required to send the frags of this
513 * skb. Each frag might be a compound page.
515 static int xennet_count_skb_frag_slots(struct sk_buff *skb)
517 int i, frags = skb_shinfo(skb)->nr_frags;
520 for (i = 0; i < frags; i++) {
521 skb_frag_t *frag = skb_shinfo(skb)->frags + i;
522 unsigned long size = skb_frag_size(frag);
523 unsigned long offset = frag->page_offset;
525 /* Skip unused frames from start of page */
526 offset &= ~PAGE_MASK;
528 pages += PFN_UP(offset + size);
534 static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev)
537 struct netfront_info *np = netdev_priv(dev);
538 struct netfront_stats *stats = this_cpu_ptr(np->stats);
539 struct xen_netif_tx_request *tx;
540 char *data = skb->data;
546 unsigned int offset = offset_in_page(data);
547 unsigned int len = skb_headlen(skb);
550 slots = DIV_ROUND_UP(offset + len, PAGE_SIZE) +
551 xennet_count_skb_frag_slots(skb);
552 if (unlikely(slots > MAX_SKB_FRAGS + 1)) {
553 net_alert_ratelimited(
554 "xennet: skb rides the rocket: %d slots\n", slots);
558 spin_lock_irqsave(&np->tx_lock, flags);
560 if (unlikely(!netif_carrier_ok(dev) ||
561 (slots > 1 && !xennet_can_sg(dev)) ||
562 netif_needs_gso(skb, netif_skb_features(skb)))) {
563 spin_unlock_irqrestore(&np->tx_lock, flags);
567 i = np->tx.req_prod_pvt;
569 id = get_id_from_freelist(&np->tx_skb_freelist, np->tx_skbs);
570 np->tx_skbs[id].skb = skb;
572 tx = RING_GET_REQUEST(&np->tx, i);
575 ref = gnttab_claim_grant_reference(&np->gref_tx_head);
576 BUG_ON((signed short)ref < 0);
577 mfn = virt_to_mfn(data);
578 gnttab_grant_foreign_access_ref(
579 ref, np->xbdev->otherend_id, mfn, GNTMAP_readonly);
580 tx->gref = np->grant_tx_ref[id] = ref;
585 if (skb->ip_summed == CHECKSUM_PARTIAL)
587 tx->flags |= XEN_NETTXF_csum_blank | XEN_NETTXF_data_validated;
588 else if (skb->ip_summed == CHECKSUM_UNNECESSARY)
589 /* remote but checksummed. */
590 tx->flags |= XEN_NETTXF_data_validated;
592 if (skb_shinfo(skb)->gso_size) {
593 struct xen_netif_extra_info *gso;
595 gso = (struct xen_netif_extra_info *)
596 RING_GET_REQUEST(&np->tx, ++i);
598 tx->flags |= XEN_NETTXF_extra_info;
600 gso->u.gso.size = skb_shinfo(skb)->gso_size;
601 gso->u.gso.type = XEN_NETIF_GSO_TYPE_TCPV4;
603 gso->u.gso.features = 0;
605 gso->type = XEN_NETIF_EXTRA_TYPE_GSO;
609 np->tx.req_prod_pvt = i + 1;
611 xennet_make_frags(skb, dev, tx);
614 RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&np->tx, notify);
616 notify_remote_via_irq(np->netdev->irq);
618 u64_stats_update_begin(&stats->syncp);
619 stats->tx_bytes += skb->len;
621 u64_stats_update_end(&stats->syncp);
623 /* Note: It is not safe to access skb after xennet_tx_buf_gc()! */
624 xennet_tx_buf_gc(dev);
626 if (!netfront_tx_slot_available(np))
627 netif_stop_queue(dev);
629 spin_unlock_irqrestore(&np->tx_lock, flags);
634 dev->stats.tx_dropped++;
639 static int xennet_close(struct net_device *dev)
641 struct netfront_info *np = netdev_priv(dev);
642 netif_stop_queue(np->netdev);
643 napi_disable(&np->napi);
647 static void xennet_move_rx_slot(struct netfront_info *np, struct sk_buff *skb,
650 int new = xennet_rxidx(np->rx.req_prod_pvt);
652 BUG_ON(np->rx_skbs[new]);
653 np->rx_skbs[new] = skb;
654 np->grant_rx_ref[new] = ref;
655 RING_GET_REQUEST(&np->rx, np->rx.req_prod_pvt)->id = new;
656 RING_GET_REQUEST(&np->rx, np->rx.req_prod_pvt)->gref = ref;
657 np->rx.req_prod_pvt++;
660 static int xennet_get_extras(struct netfront_info *np,
661 struct xen_netif_extra_info *extras,
665 struct xen_netif_extra_info *extra;
666 struct device *dev = &np->netdev->dev;
667 RING_IDX cons = np->rx.rsp_cons;
674 if (unlikely(cons + 1 == rp)) {
676 dev_warn(dev, "Missing extra info\n");
681 extra = (struct xen_netif_extra_info *)
682 RING_GET_RESPONSE(&np->rx, ++cons);
684 if (unlikely(!extra->type ||
685 extra->type >= XEN_NETIF_EXTRA_TYPE_MAX)) {
687 dev_warn(dev, "Invalid extra type: %d\n",
691 memcpy(&extras[extra->type - 1], extra,
695 skb = xennet_get_rx_skb(np, cons);
696 ref = xennet_get_rx_ref(np, cons);
697 xennet_move_rx_slot(np, skb, ref);
698 } while (extra->flags & XEN_NETIF_EXTRA_FLAG_MORE);
700 np->rx.rsp_cons = cons;
704 static int xennet_get_responses(struct netfront_info *np,
705 struct netfront_rx_info *rinfo, RING_IDX rp,
706 struct sk_buff_head *list)
708 struct xen_netif_rx_response *rx = &rinfo->rx;
709 struct xen_netif_extra_info *extras = rinfo->extras;
710 struct device *dev = &np->netdev->dev;
711 RING_IDX cons = np->rx.rsp_cons;
712 struct sk_buff *skb = xennet_get_rx_skb(np, cons);
713 grant_ref_t ref = xennet_get_rx_ref(np, cons);
714 int max = MAX_SKB_FRAGS + (rx->status <= RX_COPY_THRESHOLD);
719 if (rx->flags & XEN_NETRXF_extra_info) {
720 err = xennet_get_extras(np, extras, rp);
721 cons = np->rx.rsp_cons;
725 if (unlikely(rx->status < 0 ||
726 rx->offset + rx->status > PAGE_SIZE)) {
728 dev_warn(dev, "rx->offset: %x, size: %u\n",
729 rx->offset, rx->status);
730 xennet_move_rx_slot(np, skb, ref);
736 * This definitely indicates a bug, either in this driver or in
737 * the backend driver. In future this should flag the bad
738 * situation to the system controller to reboot the backed.
740 if (ref == GRANT_INVALID_REF) {
742 dev_warn(dev, "Bad rx response id %d.\n",
748 ret = gnttab_end_foreign_access_ref(ref, 0);
751 gnttab_release_grant_reference(&np->gref_rx_head, ref);
753 __skb_queue_tail(list, skb);
756 if (!(rx->flags & XEN_NETRXF_more_data))
759 if (cons + slots == rp) {
761 dev_warn(dev, "Need more slots\n");
766 rx = RING_GET_RESPONSE(&np->rx, cons + slots);
767 skb = xennet_get_rx_skb(np, cons + slots);
768 ref = xennet_get_rx_ref(np, cons + slots);
772 if (unlikely(slots > max)) {
774 dev_warn(dev, "Too many frags\n");
779 np->rx.rsp_cons = cons + slots;
784 static int xennet_set_skb_gso(struct sk_buff *skb,
785 struct xen_netif_extra_info *gso)
787 if (!gso->u.gso.size) {
789 printk(KERN_WARNING "GSO size must not be zero.\n");
793 /* Currently only TCPv4 S.O. is supported. */
794 if (gso->u.gso.type != XEN_NETIF_GSO_TYPE_TCPV4) {
796 printk(KERN_WARNING "Bad GSO type %d.\n", gso->u.gso.type);
800 skb_shinfo(skb)->gso_size = gso->u.gso.size;
801 skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4;
803 /* Header must be checked, and gso_segs computed. */
804 skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY;
805 skb_shinfo(skb)->gso_segs = 0;
810 static RING_IDX xennet_fill_frags(struct netfront_info *np,
812 struct sk_buff_head *list)
814 struct skb_shared_info *shinfo = skb_shinfo(skb);
815 int nr_frags = shinfo->nr_frags;
816 RING_IDX cons = np->rx.rsp_cons;
817 struct sk_buff *nskb;
819 while ((nskb = __skb_dequeue(list))) {
820 struct xen_netif_rx_response *rx =
821 RING_GET_RESPONSE(&np->rx, ++cons);
822 skb_frag_t *nfrag = &skb_shinfo(nskb)->frags[0];
824 __skb_fill_page_desc(skb, nr_frags,
825 skb_frag_page(nfrag),
826 rx->offset, rx->status);
828 skb->data_len += rx->status;
830 skb_shinfo(nskb)->nr_frags = 0;
836 shinfo->nr_frags = nr_frags;
840 static int checksum_setup(struct net_device *dev, struct sk_buff *skb)
845 int recalculate_partial_csum = 0;
848 * A GSO SKB must be CHECKSUM_PARTIAL. However some buggy
849 * peers can fail to set NETRXF_csum_blank when sending a GSO
850 * frame. In this case force the SKB to CHECKSUM_PARTIAL and
851 * recalculate the partial checksum.
853 if (skb->ip_summed != CHECKSUM_PARTIAL && skb_is_gso(skb)) {
854 struct netfront_info *np = netdev_priv(dev);
855 np->rx_gso_checksum_fixup++;
856 skb->ip_summed = CHECKSUM_PARTIAL;
857 recalculate_partial_csum = 1;
860 /* A non-CHECKSUM_PARTIAL SKB does not require setup. */
861 if (skb->ip_summed != CHECKSUM_PARTIAL)
864 if (skb->protocol != htons(ETH_P_IP))
867 iph = (void *)skb->data;
868 th = skb->data + 4 * iph->ihl;
869 if (th >= skb_tail_pointer(skb))
872 skb->csum_start = th - skb->head;
873 switch (iph->protocol) {
875 skb->csum_offset = offsetof(struct tcphdr, check);
877 if (recalculate_partial_csum) {
878 struct tcphdr *tcph = (struct tcphdr *)th;
879 tcph->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr,
880 skb->len - iph->ihl*4,
885 skb->csum_offset = offsetof(struct udphdr, check);
887 if (recalculate_partial_csum) {
888 struct udphdr *udph = (struct udphdr *)th;
889 udph->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr,
890 skb->len - iph->ihl*4,
896 printk(KERN_ERR "Attempting to checksum a non-"
897 "TCP/UDP packet, dropping a protocol"
898 " %d packet", iph->protocol);
902 if ((th + skb->csum_offset + 2) > skb_tail_pointer(skb))
911 static int handle_incoming_queue(struct net_device *dev,
912 struct sk_buff_head *rxq)
914 struct netfront_info *np = netdev_priv(dev);
915 struct netfront_stats *stats = this_cpu_ptr(np->stats);
916 int packets_dropped = 0;
919 while ((skb = __skb_dequeue(rxq)) != NULL) {
920 int pull_to = NETFRONT_SKB_CB(skb)->pull_to;
922 __pskb_pull_tail(skb, pull_to - skb_headlen(skb));
924 /* Ethernet work: Delayed to here as it peeks the header. */
925 skb->protocol = eth_type_trans(skb, dev);
927 if (checksum_setup(dev, skb)) {
930 dev->stats.rx_errors++;
934 u64_stats_update_begin(&stats->syncp);
936 stats->rx_bytes += skb->len;
937 u64_stats_update_end(&stats->syncp);
940 netif_receive_skb(skb);
943 return packets_dropped;
946 static int xennet_poll(struct napi_struct *napi, int budget)
948 struct netfront_info *np = container_of(napi, struct netfront_info, napi);
949 struct net_device *dev = np->netdev;
951 struct netfront_rx_info rinfo;
952 struct xen_netif_rx_response *rx = &rinfo.rx;
953 struct xen_netif_extra_info *extras = rinfo.extras;
956 struct sk_buff_head rxq;
957 struct sk_buff_head errq;
958 struct sk_buff_head tmpq;
962 spin_lock(&np->rx_lock);
964 skb_queue_head_init(&rxq);
965 skb_queue_head_init(&errq);
966 skb_queue_head_init(&tmpq);
968 rp = np->rx.sring->rsp_prod;
969 rmb(); /* Ensure we see queued responses up to 'rp'. */
973 while ((i != rp) && (work_done < budget)) {
974 memcpy(rx, RING_GET_RESPONSE(&np->rx, i), sizeof(*rx));
975 memset(extras, 0, sizeof(rinfo.extras));
977 err = xennet_get_responses(np, &rinfo, rp, &tmpq);
981 while ((skb = __skb_dequeue(&tmpq)))
982 __skb_queue_tail(&errq, skb);
983 dev->stats.rx_errors++;
988 skb = __skb_dequeue(&tmpq);
990 if (extras[XEN_NETIF_EXTRA_TYPE_GSO - 1].type) {
991 struct xen_netif_extra_info *gso;
992 gso = &extras[XEN_NETIF_EXTRA_TYPE_GSO - 1];
994 if (unlikely(xennet_set_skb_gso(skb, gso))) {
995 __skb_queue_head(&tmpq, skb);
996 np->rx.rsp_cons += skb_queue_len(&tmpq);
1001 NETFRONT_SKB_CB(skb)->pull_to = rx->status;
1002 if (NETFRONT_SKB_CB(skb)->pull_to > RX_COPY_THRESHOLD)
1003 NETFRONT_SKB_CB(skb)->pull_to = RX_COPY_THRESHOLD;
1005 skb_shinfo(skb)->frags[0].page_offset = rx->offset;
1006 skb_frag_size_set(&skb_shinfo(skb)->frags[0], rx->status);
1007 skb->data_len = rx->status;
1009 i = xennet_fill_frags(np, skb, &tmpq);
1012 * Truesize is the actual allocation size, even if the
1013 * allocation is only partially used.
1015 skb->truesize += PAGE_SIZE * skb_shinfo(skb)->nr_frags;
1016 skb->len += skb->data_len;
1018 if (rx->flags & XEN_NETRXF_csum_blank)
1019 skb->ip_summed = CHECKSUM_PARTIAL;
1020 else if (rx->flags & XEN_NETRXF_data_validated)
1021 skb->ip_summed = CHECKSUM_UNNECESSARY;
1023 __skb_queue_tail(&rxq, skb);
1025 np->rx.rsp_cons = ++i;
1029 __skb_queue_purge(&errq);
1031 work_done -= handle_incoming_queue(dev, &rxq);
1033 /* If we get a callback with very few responses, reduce fill target. */
1034 /* NB. Note exponential increase, linear decrease. */
1035 if (((np->rx.req_prod_pvt - np->rx.sring->rsp_prod) >
1036 ((3*np->rx_target) / 4)) &&
1037 (--np->rx_target < np->rx_min_target))
1038 np->rx_target = np->rx_min_target;
1040 xennet_alloc_rx_buffers(dev);
1042 if (work_done < budget) {
1045 local_irq_save(flags);
1047 RING_FINAL_CHECK_FOR_RESPONSES(&np->rx, more_to_do);
1049 __napi_complete(napi);
1051 local_irq_restore(flags);
1054 spin_unlock(&np->rx_lock);
1059 static int xennet_change_mtu(struct net_device *dev, int mtu)
1061 int max = xennet_can_sg(dev) ? 65535 - ETH_HLEN : ETH_DATA_LEN;
1069 static struct rtnl_link_stats64 *xennet_get_stats64(struct net_device *dev,
1070 struct rtnl_link_stats64 *tot)
1072 struct netfront_info *np = netdev_priv(dev);
1075 for_each_possible_cpu(cpu) {
1076 struct netfront_stats *stats = per_cpu_ptr(np->stats, cpu);
1077 u64 rx_packets, rx_bytes, tx_packets, tx_bytes;
1081 start = u64_stats_fetch_begin_bh(&stats->syncp);
1083 rx_packets = stats->rx_packets;
1084 tx_packets = stats->tx_packets;
1085 rx_bytes = stats->rx_bytes;
1086 tx_bytes = stats->tx_bytes;
1087 } while (u64_stats_fetch_retry_bh(&stats->syncp, start));
1089 tot->rx_packets += rx_packets;
1090 tot->tx_packets += tx_packets;
1091 tot->rx_bytes += rx_bytes;
1092 tot->tx_bytes += tx_bytes;
1095 tot->rx_errors = dev->stats.rx_errors;
1096 tot->tx_dropped = dev->stats.tx_dropped;
1101 static void xennet_release_tx_bufs(struct netfront_info *np)
1103 struct sk_buff *skb;
1106 for (i = 0; i < NET_TX_RING_SIZE; i++) {
1107 /* Skip over entries which are actually freelist references */
1108 if (skb_entry_is_link(&np->tx_skbs[i]))
1111 skb = np->tx_skbs[i].skb;
1112 gnttab_end_foreign_access_ref(np->grant_tx_ref[i],
1114 gnttab_release_grant_reference(&np->gref_tx_head,
1115 np->grant_tx_ref[i]);
1116 np->grant_tx_ref[i] = GRANT_INVALID_REF;
1117 add_id_to_freelist(&np->tx_skb_freelist, np->tx_skbs, i);
1118 dev_kfree_skb_irq(skb);
1122 static void xennet_release_rx_bufs(struct netfront_info *np)
1124 struct mmu_update *mmu = np->rx_mmu;
1125 struct multicall_entry *mcl = np->rx_mcl;
1126 struct sk_buff_head free_list;
1127 struct sk_buff *skb;
1129 int xfer = 0, noxfer = 0, unused = 0;
1132 dev_warn(&np->netdev->dev, "%s: fix me for copying receiver.\n",
1136 skb_queue_head_init(&free_list);
1138 spin_lock_bh(&np->rx_lock);
1140 for (id = 0; id < NET_RX_RING_SIZE; id++) {
1141 ref = np->grant_rx_ref[id];
1142 if (ref == GRANT_INVALID_REF) {
1147 skb = np->rx_skbs[id];
1148 mfn = gnttab_end_foreign_transfer_ref(ref);
1149 gnttab_release_grant_reference(&np->gref_rx_head, ref);
1150 np->grant_rx_ref[id] = GRANT_INVALID_REF;
1153 skb_shinfo(skb)->nr_frags = 0;
1159 if (!xen_feature(XENFEAT_auto_translated_physmap)) {
1160 /* Remap the page. */
1161 const struct page *page =
1162 skb_frag_page(&skb_shinfo(skb)->frags[0]);
1163 unsigned long pfn = page_to_pfn(page);
1164 void *vaddr = page_address(page);
1166 MULTI_update_va_mapping(mcl, (unsigned long)vaddr,
1167 mfn_pte(mfn, PAGE_KERNEL),
1170 mmu->ptr = ((u64)mfn << PAGE_SHIFT)
1171 | MMU_MACHPHYS_UPDATE;
1175 set_phys_to_machine(pfn, mfn);
1177 __skb_queue_tail(&free_list, skb);
1181 dev_info(&np->netdev->dev, "%s: %d xfer, %d noxfer, %d unused\n",
1182 __func__, xfer, noxfer, unused);
1185 if (!xen_feature(XENFEAT_auto_translated_physmap)) {
1186 /* Do all the remapping work and M2P updates. */
1187 MULTI_mmu_update(mcl, np->rx_mmu, mmu - np->rx_mmu,
1190 HYPERVISOR_multicall(np->rx_mcl, mcl - np->rx_mcl);
1194 __skb_queue_purge(&free_list);
1196 spin_unlock_bh(&np->rx_lock);
1199 static void xennet_uninit(struct net_device *dev)
1201 struct netfront_info *np = netdev_priv(dev);
1202 xennet_release_tx_bufs(np);
1203 xennet_release_rx_bufs(np);
1204 gnttab_free_grant_references(np->gref_tx_head);
1205 gnttab_free_grant_references(np->gref_rx_head);
1208 static netdev_features_t xennet_fix_features(struct net_device *dev,
1209 netdev_features_t features)
1211 struct netfront_info *np = netdev_priv(dev);
1214 if (features & NETIF_F_SG) {
1215 if (xenbus_scanf(XBT_NIL, np->xbdev->otherend, "feature-sg",
1220 features &= ~NETIF_F_SG;
1223 if (features & NETIF_F_TSO) {
1224 if (xenbus_scanf(XBT_NIL, np->xbdev->otherend,
1225 "feature-gso-tcpv4", "%d", &val) < 0)
1229 features &= ~NETIF_F_TSO;
1235 static int xennet_set_features(struct net_device *dev,
1236 netdev_features_t features)
1238 if (!(features & NETIF_F_SG) && dev->mtu > ETH_DATA_LEN) {
1239 netdev_info(dev, "Reducing MTU because no SG offload");
1240 dev->mtu = ETH_DATA_LEN;
1246 static irqreturn_t xennet_interrupt(int irq, void *dev_id)
1248 struct net_device *dev = dev_id;
1249 struct netfront_info *np = netdev_priv(dev);
1250 unsigned long flags;
1252 spin_lock_irqsave(&np->tx_lock, flags);
1254 if (likely(netif_carrier_ok(dev))) {
1255 xennet_tx_buf_gc(dev);
1256 /* Under tx_lock: protects access to rx shared-ring indexes. */
1257 if (RING_HAS_UNCONSUMED_RESPONSES(&np->rx))
1258 napi_schedule(&np->napi);
1261 spin_unlock_irqrestore(&np->tx_lock, flags);
1266 #ifdef CONFIG_NET_POLL_CONTROLLER
1267 static void xennet_poll_controller(struct net_device *dev)
1269 xennet_interrupt(0, dev);
1273 static const struct net_device_ops xennet_netdev_ops = {
1274 .ndo_open = xennet_open,
1275 .ndo_uninit = xennet_uninit,
1276 .ndo_stop = xennet_close,
1277 .ndo_start_xmit = xennet_start_xmit,
1278 .ndo_change_mtu = xennet_change_mtu,
1279 .ndo_get_stats64 = xennet_get_stats64,
1280 .ndo_set_mac_address = eth_mac_addr,
1281 .ndo_validate_addr = eth_validate_addr,
1282 .ndo_fix_features = xennet_fix_features,
1283 .ndo_set_features = xennet_set_features,
1284 #ifdef CONFIG_NET_POLL_CONTROLLER
1285 .ndo_poll_controller = xennet_poll_controller,
1289 static struct net_device *xennet_create_dev(struct xenbus_device *dev)
1292 struct net_device *netdev;
1293 struct netfront_info *np;
1295 netdev = alloc_etherdev(sizeof(struct netfront_info));
1297 return ERR_PTR(-ENOMEM);
1299 np = netdev_priv(netdev);
1302 spin_lock_init(&np->tx_lock);
1303 spin_lock_init(&np->rx_lock);
1305 skb_queue_head_init(&np->rx_batch);
1306 np->rx_target = RX_DFL_MIN_TARGET;
1307 np->rx_min_target = RX_DFL_MIN_TARGET;
1308 np->rx_max_target = RX_MAX_TARGET;
1310 init_timer(&np->rx_refill_timer);
1311 np->rx_refill_timer.data = (unsigned long)netdev;
1312 np->rx_refill_timer.function = rx_refill_timeout;
1315 np->stats = alloc_percpu(struct netfront_stats);
1316 if (np->stats == NULL)
1319 /* Initialise tx_skbs as a free chain containing every entry. */
1320 np->tx_skb_freelist = 0;
1321 for (i = 0; i < NET_TX_RING_SIZE; i++) {
1322 skb_entry_set_link(&np->tx_skbs[i], i+1);
1323 np->grant_tx_ref[i] = GRANT_INVALID_REF;
1326 /* Clear out rx_skbs */
1327 for (i = 0; i < NET_RX_RING_SIZE; i++) {
1328 np->rx_skbs[i] = NULL;
1329 np->grant_rx_ref[i] = GRANT_INVALID_REF;
1332 /* A grant for every tx ring slot */
1333 if (gnttab_alloc_grant_references(TX_MAX_TARGET,
1334 &np->gref_tx_head) < 0) {
1335 printk(KERN_ALERT "#### netfront can't alloc tx grant refs\n");
1337 goto exit_free_stats;
1339 /* A grant for every rx ring slot */
1340 if (gnttab_alloc_grant_references(RX_MAX_TARGET,
1341 &np->gref_rx_head) < 0) {
1342 printk(KERN_ALERT "#### netfront can't alloc rx grant refs\n");
1347 netdev->netdev_ops = &xennet_netdev_ops;
1349 netif_napi_add(netdev, &np->napi, xennet_poll, 64);
1350 netdev->features = NETIF_F_IP_CSUM | NETIF_F_RXCSUM |
1352 netdev->hw_features = NETIF_F_IP_CSUM | NETIF_F_SG | NETIF_F_TSO;
1355 * Assume that all hw features are available for now. This set
1356 * will be adjusted by the call to netdev_update_features() in
1357 * xennet_connect() which is the earliest point where we can
1358 * negotiate with the backend regarding supported features.
1360 netdev->features |= netdev->hw_features;
1362 SET_ETHTOOL_OPS(netdev, &xennet_ethtool_ops);
1363 SET_NETDEV_DEV(netdev, &dev->dev);
1365 np->netdev = netdev;
1367 netif_carrier_off(netdev);
1372 gnttab_free_grant_references(np->gref_tx_head);
1374 free_percpu(np->stats);
1376 free_netdev(netdev);
1377 return ERR_PTR(err);
1381 * Entry point to this code when a new device is created. Allocate the basic
1382 * structures and the ring buffers for communication with the backend, and
1383 * inform the backend of the appropriate details for those.
1385 static int netfront_probe(struct xenbus_device *dev,
1386 const struct xenbus_device_id *id)
1389 struct net_device *netdev;
1390 struct netfront_info *info;
1392 netdev = xennet_create_dev(dev);
1393 if (IS_ERR(netdev)) {
1394 err = PTR_ERR(netdev);
1395 xenbus_dev_fatal(dev, err, "creating netdev");
1399 info = netdev_priv(netdev);
1400 dev_set_drvdata(&dev->dev, info);
1402 err = register_netdev(info->netdev);
1404 printk(KERN_WARNING "%s: register_netdev err=%d\n",
1409 err = xennet_sysfs_addif(info->netdev);
1411 unregister_netdev(info->netdev);
1412 printk(KERN_WARNING "%s: add sysfs failed err=%d\n",
1420 free_netdev(netdev);
1421 dev_set_drvdata(&dev->dev, NULL);
1425 static void xennet_end_access(int ref, void *page)
1427 /* This frees the page as a side-effect */
1428 if (ref != GRANT_INVALID_REF)
1429 gnttab_end_foreign_access(ref, 0, (unsigned long)page);
1432 static void xennet_disconnect_backend(struct netfront_info *info)
1434 /* Stop old i/f to prevent errors whilst we rebuild the state. */
1435 spin_lock_bh(&info->rx_lock);
1436 spin_lock_irq(&info->tx_lock);
1437 netif_carrier_off(info->netdev);
1438 spin_unlock_irq(&info->tx_lock);
1439 spin_unlock_bh(&info->rx_lock);
1441 if (info->netdev->irq)
1442 unbind_from_irqhandler(info->netdev->irq, info->netdev);
1443 info->evtchn = info->netdev->irq = 0;
1445 /* End access and free the pages */
1446 xennet_end_access(info->tx_ring_ref, info->tx.sring);
1447 xennet_end_access(info->rx_ring_ref, info->rx.sring);
1449 info->tx_ring_ref = GRANT_INVALID_REF;
1450 info->rx_ring_ref = GRANT_INVALID_REF;
1451 info->tx.sring = NULL;
1452 info->rx.sring = NULL;
1456 * We are reconnecting to the backend, due to a suspend/resume, or a backend
1457 * driver restart. We tear down our netif structure and recreate it, but
1458 * leave the device-layer structures intact so that this is transparent to the
1459 * rest of the kernel.
1461 static int netfront_resume(struct xenbus_device *dev)
1463 struct netfront_info *info = dev_get_drvdata(&dev->dev);
1465 dev_dbg(&dev->dev, "%s\n", dev->nodename);
1467 xennet_disconnect_backend(info);
1471 static int xen_net_read_mac(struct xenbus_device *dev, u8 mac[])
1473 char *s, *e, *macstr;
1476 macstr = s = xenbus_read(XBT_NIL, dev->nodename, "mac", NULL);
1478 return PTR_ERR(macstr);
1480 for (i = 0; i < ETH_ALEN; i++) {
1481 mac[i] = simple_strtoul(s, &e, 16);
1482 if ((s == e) || (*e != ((i == ETH_ALEN-1) ? '\0' : ':'))) {
1493 static int setup_netfront(struct xenbus_device *dev, struct netfront_info *info)
1495 struct xen_netif_tx_sring *txs;
1496 struct xen_netif_rx_sring *rxs;
1498 struct net_device *netdev = info->netdev;
1500 info->tx_ring_ref = GRANT_INVALID_REF;
1501 info->rx_ring_ref = GRANT_INVALID_REF;
1502 info->rx.sring = NULL;
1503 info->tx.sring = NULL;
1506 err = xen_net_read_mac(dev, netdev->dev_addr);
1508 xenbus_dev_fatal(dev, err, "parsing %s/mac", dev->nodename);
1512 txs = (struct xen_netif_tx_sring *)get_zeroed_page(GFP_NOIO | __GFP_HIGH);
1515 xenbus_dev_fatal(dev, err, "allocating tx ring page");
1518 SHARED_RING_INIT(txs);
1519 FRONT_RING_INIT(&info->tx, txs, PAGE_SIZE);
1521 err = xenbus_grant_ring(dev, virt_to_mfn(txs));
1523 free_page((unsigned long)txs);
1527 info->tx_ring_ref = err;
1528 rxs = (struct xen_netif_rx_sring *)get_zeroed_page(GFP_NOIO | __GFP_HIGH);
1531 xenbus_dev_fatal(dev, err, "allocating rx ring page");
1534 SHARED_RING_INIT(rxs);
1535 FRONT_RING_INIT(&info->rx, rxs, PAGE_SIZE);
1537 err = xenbus_grant_ring(dev, virt_to_mfn(rxs));
1539 free_page((unsigned long)rxs);
1542 info->rx_ring_ref = err;
1544 err = xenbus_alloc_evtchn(dev, &info->evtchn);
1548 err = bind_evtchn_to_irqhandler(info->evtchn, xennet_interrupt,
1549 0, netdev->name, netdev);
1559 /* Common code used when first setting up, and when resuming. */
1560 static int talk_to_netback(struct xenbus_device *dev,
1561 struct netfront_info *info)
1563 const char *message;
1564 struct xenbus_transaction xbt;
1567 /* Create shared ring, alloc event channel. */
1568 err = setup_netfront(dev, info);
1573 err = xenbus_transaction_start(&xbt);
1575 xenbus_dev_fatal(dev, err, "starting transaction");
1579 err = xenbus_printf(xbt, dev->nodename, "tx-ring-ref", "%u",
1582 message = "writing tx ring-ref";
1583 goto abort_transaction;
1585 err = xenbus_printf(xbt, dev->nodename, "rx-ring-ref", "%u",
1588 message = "writing rx ring-ref";
1589 goto abort_transaction;
1591 err = xenbus_printf(xbt, dev->nodename,
1592 "event-channel", "%u", info->evtchn);
1594 message = "writing event-channel";
1595 goto abort_transaction;
1598 err = xenbus_printf(xbt, dev->nodename, "request-rx-copy", "%u",
1601 message = "writing request-rx-copy";
1602 goto abort_transaction;
1605 err = xenbus_printf(xbt, dev->nodename, "feature-rx-notify", "%d", 1);
1607 message = "writing feature-rx-notify";
1608 goto abort_transaction;
1611 err = xenbus_printf(xbt, dev->nodename, "feature-sg", "%d", 1);
1613 message = "writing feature-sg";
1614 goto abort_transaction;
1617 err = xenbus_printf(xbt, dev->nodename, "feature-gso-tcpv4", "%d", 1);
1619 message = "writing feature-gso-tcpv4";
1620 goto abort_transaction;
1623 err = xenbus_transaction_end(xbt, 0);
1627 xenbus_dev_fatal(dev, err, "completing transaction");
1634 xenbus_transaction_end(xbt, 1);
1635 xenbus_dev_fatal(dev, err, "%s", message);
1637 xennet_disconnect_backend(info);
1642 static int xennet_connect(struct net_device *dev)
1644 struct netfront_info *np = netdev_priv(dev);
1645 int i, requeue_idx, err;
1646 struct sk_buff *skb;
1648 struct xen_netif_rx_request *req;
1649 unsigned int feature_rx_copy;
1651 err = xenbus_scanf(XBT_NIL, np->xbdev->otherend,
1652 "feature-rx-copy", "%u", &feature_rx_copy);
1654 feature_rx_copy = 0;
1656 if (!feature_rx_copy) {
1658 "backend does not support copying receive path\n");
1662 err = talk_to_netback(np->xbdev, np);
1667 netdev_update_features(dev);
1670 spin_lock_bh(&np->rx_lock);
1671 spin_lock_irq(&np->tx_lock);
1673 /* Step 1: Discard all pending TX packet fragments. */
1674 xennet_release_tx_bufs(np);
1676 /* Step 2: Rebuild the RX buffer freelist and the RX ring itself. */
1677 for (requeue_idx = 0, i = 0; i < NET_RX_RING_SIZE; i++) {
1679 const struct page *page;
1680 if (!np->rx_skbs[i])
1683 skb = np->rx_skbs[requeue_idx] = xennet_get_rx_skb(np, i);
1684 ref = np->grant_rx_ref[requeue_idx] = xennet_get_rx_ref(np, i);
1685 req = RING_GET_REQUEST(&np->rx, requeue_idx);
1687 frag = &skb_shinfo(skb)->frags[0];
1688 page = skb_frag_page(frag);
1689 gnttab_grant_foreign_access_ref(
1690 ref, np->xbdev->otherend_id,
1691 pfn_to_mfn(page_to_pfn(page)),
1694 req->id = requeue_idx;
1699 np->rx.req_prod_pvt = requeue_idx;
1702 * Step 3: All public and private state should now be sane. Get
1703 * ready to start sending and receiving packets and give the driver
1704 * domain a kick because we've probably just requeued some
1707 netif_carrier_on(np->netdev);
1708 notify_remote_via_irq(np->netdev->irq);
1709 xennet_tx_buf_gc(dev);
1710 xennet_alloc_rx_buffers(dev);
1712 spin_unlock_irq(&np->tx_lock);
1713 spin_unlock_bh(&np->rx_lock);
1719 * Callback received when the backend's state changes.
1721 static void netback_changed(struct xenbus_device *dev,
1722 enum xenbus_state backend_state)
1724 struct netfront_info *np = dev_get_drvdata(&dev->dev);
1725 struct net_device *netdev = np->netdev;
1727 dev_dbg(&dev->dev, "%s\n", xenbus_strstate(backend_state));
1729 switch (backend_state) {
1730 case XenbusStateInitialising:
1731 case XenbusStateInitialised:
1732 case XenbusStateReconfiguring:
1733 case XenbusStateReconfigured:
1734 case XenbusStateUnknown:
1735 case XenbusStateClosed:
1738 case XenbusStateInitWait:
1739 if (dev->state != XenbusStateInitialising)
1741 if (xennet_connect(netdev) != 0)
1743 xenbus_switch_state(dev, XenbusStateConnected);
1746 case XenbusStateConnected:
1747 netdev_notify_peers(netdev);
1750 case XenbusStateClosing:
1751 xenbus_frontend_closed(dev);
1756 static const struct xennet_stat {
1757 char name[ETH_GSTRING_LEN];
1759 } xennet_stats[] = {
1761 "rx_gso_checksum_fixup",
1762 offsetof(struct netfront_info, rx_gso_checksum_fixup)
1766 static int xennet_get_sset_count(struct net_device *dev, int string_set)
1768 switch (string_set) {
1770 return ARRAY_SIZE(xennet_stats);
1776 static void xennet_get_ethtool_stats(struct net_device *dev,
1777 struct ethtool_stats *stats, u64 * data)
1779 void *np = netdev_priv(dev);
1782 for (i = 0; i < ARRAY_SIZE(xennet_stats); i++)
1783 data[i] = *(unsigned long *)(np + xennet_stats[i].offset);
1786 static void xennet_get_strings(struct net_device *dev, u32 stringset, u8 * data)
1790 switch (stringset) {
1792 for (i = 0; i < ARRAY_SIZE(xennet_stats); i++)
1793 memcpy(data + i * ETH_GSTRING_LEN,
1794 xennet_stats[i].name, ETH_GSTRING_LEN);
1799 static const struct ethtool_ops xennet_ethtool_ops =
1801 .get_link = ethtool_op_get_link,
1803 .get_sset_count = xennet_get_sset_count,
1804 .get_ethtool_stats = xennet_get_ethtool_stats,
1805 .get_strings = xennet_get_strings,
1809 static ssize_t show_rxbuf_min(struct device *dev,
1810 struct device_attribute *attr, char *buf)
1812 struct net_device *netdev = to_net_dev(dev);
1813 struct netfront_info *info = netdev_priv(netdev);
1815 return sprintf(buf, "%u\n", info->rx_min_target);
1818 static ssize_t store_rxbuf_min(struct device *dev,
1819 struct device_attribute *attr,
1820 const char *buf, size_t len)
1822 struct net_device *netdev = to_net_dev(dev);
1823 struct netfront_info *np = netdev_priv(netdev);
1825 unsigned long target;
1827 if (!capable(CAP_NET_ADMIN))
1830 target = simple_strtoul(buf, &endp, 0);
1834 if (target < RX_MIN_TARGET)
1835 target = RX_MIN_TARGET;
1836 if (target > RX_MAX_TARGET)
1837 target = RX_MAX_TARGET;
1839 spin_lock_bh(&np->rx_lock);
1840 if (target > np->rx_max_target)
1841 np->rx_max_target = target;
1842 np->rx_min_target = target;
1843 if (target > np->rx_target)
1844 np->rx_target = target;
1846 xennet_alloc_rx_buffers(netdev);
1848 spin_unlock_bh(&np->rx_lock);
1852 static ssize_t show_rxbuf_max(struct device *dev,
1853 struct device_attribute *attr, char *buf)
1855 struct net_device *netdev = to_net_dev(dev);
1856 struct netfront_info *info = netdev_priv(netdev);
1858 return sprintf(buf, "%u\n", info->rx_max_target);
1861 static ssize_t store_rxbuf_max(struct device *dev,
1862 struct device_attribute *attr,
1863 const char *buf, size_t len)
1865 struct net_device *netdev = to_net_dev(dev);
1866 struct netfront_info *np = netdev_priv(netdev);
1868 unsigned long target;
1870 if (!capable(CAP_NET_ADMIN))
1873 target = simple_strtoul(buf, &endp, 0);
1877 if (target < RX_MIN_TARGET)
1878 target = RX_MIN_TARGET;
1879 if (target > RX_MAX_TARGET)
1880 target = RX_MAX_TARGET;
1882 spin_lock_bh(&np->rx_lock);
1883 if (target < np->rx_min_target)
1884 np->rx_min_target = target;
1885 np->rx_max_target = target;
1886 if (target < np->rx_target)
1887 np->rx_target = target;
1889 xennet_alloc_rx_buffers(netdev);
1891 spin_unlock_bh(&np->rx_lock);
1895 static ssize_t show_rxbuf_cur(struct device *dev,
1896 struct device_attribute *attr, char *buf)
1898 struct net_device *netdev = to_net_dev(dev);
1899 struct netfront_info *info = netdev_priv(netdev);
1901 return sprintf(buf, "%u\n", info->rx_target);
1904 static struct device_attribute xennet_attrs[] = {
1905 __ATTR(rxbuf_min, S_IRUGO|S_IWUSR, show_rxbuf_min, store_rxbuf_min),
1906 __ATTR(rxbuf_max, S_IRUGO|S_IWUSR, show_rxbuf_max, store_rxbuf_max),
1907 __ATTR(rxbuf_cur, S_IRUGO, show_rxbuf_cur, NULL),
1910 static int xennet_sysfs_addif(struct net_device *netdev)
1915 for (i = 0; i < ARRAY_SIZE(xennet_attrs); i++) {
1916 err = device_create_file(&netdev->dev,
1925 device_remove_file(&netdev->dev, &xennet_attrs[i]);
1929 static void xennet_sysfs_delif(struct net_device *netdev)
1933 for (i = 0; i < ARRAY_SIZE(xennet_attrs); i++)
1934 device_remove_file(&netdev->dev, &xennet_attrs[i]);
1937 #endif /* CONFIG_SYSFS */
1939 static const struct xenbus_device_id netfront_ids[] = {
1945 static int xennet_remove(struct xenbus_device *dev)
1947 struct netfront_info *info = dev_get_drvdata(&dev->dev);
1949 dev_dbg(&dev->dev, "%s\n", dev->nodename);
1951 xennet_disconnect_backend(info);
1953 xennet_sysfs_delif(info->netdev);
1955 unregister_netdev(info->netdev);
1957 del_timer_sync(&info->rx_refill_timer);
1959 free_percpu(info->stats);
1961 free_netdev(info->netdev);
1966 static DEFINE_XENBUS_DRIVER(netfront, ,
1967 .probe = netfront_probe,
1968 .remove = xennet_remove,
1969 .resume = netfront_resume,
1970 .otherend_changed = netback_changed,
1973 static int __init netif_init(void)
1978 if (xen_hvm_domain() && !xen_platform_pci_unplug)
1981 printk(KERN_INFO "Initialising Xen virtual ethernet driver.\n");
1983 return xenbus_register_frontend(&netfront_driver);
1985 module_init(netif_init);
1988 static void __exit netif_exit(void)
1990 xenbus_unregister_driver(&netfront_driver);
1992 module_exit(netif_exit);
1994 MODULE_DESCRIPTION("Xen virtual network device frontend");
1995 MODULE_LICENSE("GPL");
1996 MODULE_ALIAS("xen:vif");
1997 MODULE_ALIAS("xennet");