xref: /openbmc/linux/drivers/net/xen-netfront.c (revision ae213c44)
1 /*
2  * Virtual network driver for conversing with remote driver backends.
3  *
4  * Copyright (c) 2002-2005, K A Fraser
5  * Copyright (c) 2005, XenSource Ltd
6  *
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:
12  *
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:
19  *
20  * The above copyright notice and this permission notice shall be included in
21  * all copies or substantial portions of the Software.
22  *
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
29  * IN THE SOFTWARE.
30  */
31 
32 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
33 
34 #include <linux/module.h>
35 #include <linux/kernel.h>
36 #include <linux/netdevice.h>
37 #include <linux/etherdevice.h>
38 #include <linux/skbuff.h>
39 #include <linux/ethtool.h>
40 #include <linux/if_ether.h>
41 #include <net/tcp.h>
42 #include <linux/udp.h>
43 #include <linux/moduleparam.h>
44 #include <linux/mm.h>
45 #include <linux/slab.h>
46 #include <net/ip.h>
47 
48 #include <xen/xen.h>
49 #include <xen/xenbus.h>
50 #include <xen/events.h>
51 #include <xen/page.h>
52 #include <xen/platform_pci.h>
53 #include <xen/grant_table.h>
54 
55 #include <xen/interface/io/netif.h>
56 #include <xen/interface/memory.h>
57 #include <xen/interface/grant_table.h>
58 
59 /* Module parameters */
60 #define MAX_QUEUES_DEFAULT 8
61 static unsigned int xennet_max_queues;
62 module_param_named(max_queues, xennet_max_queues, uint, 0644);
63 MODULE_PARM_DESC(max_queues,
64 		 "Maximum number of queues per virtual interface");
65 
66 static const struct ethtool_ops xennet_ethtool_ops;
67 
68 struct netfront_cb {
69 	int pull_to;
70 };
71 
72 #define NETFRONT_SKB_CB(skb)	((struct netfront_cb *)((skb)->cb))
73 
74 #define RX_COPY_THRESHOLD 256
75 
76 #define GRANT_INVALID_REF	0
77 
78 #define NET_TX_RING_SIZE __CONST_RING_SIZE(xen_netif_tx, XEN_PAGE_SIZE)
79 #define NET_RX_RING_SIZE __CONST_RING_SIZE(xen_netif_rx, XEN_PAGE_SIZE)
80 
81 /* Minimum number of Rx slots (includes slot for GSO metadata). */
82 #define NET_RX_SLOTS_MIN (XEN_NETIF_NR_SLOTS_MIN + 1)
83 
84 /* Queue name is interface name with "-qNNN" appended */
85 #define QUEUE_NAME_SIZE (IFNAMSIZ + 6)
86 
87 /* IRQ name is queue name with "-tx" or "-rx" appended */
88 #define IRQ_NAME_SIZE (QUEUE_NAME_SIZE + 3)
89 
90 static DECLARE_WAIT_QUEUE_HEAD(module_wq);
91 
92 struct netfront_stats {
93 	u64			packets;
94 	u64			bytes;
95 	struct u64_stats_sync	syncp;
96 };
97 
98 struct netfront_info;
99 
100 struct netfront_queue {
101 	unsigned int id; /* Queue ID, 0-based */
102 	char name[QUEUE_NAME_SIZE]; /* DEVNAME-qN */
103 	struct netfront_info *info;
104 
105 	struct napi_struct napi;
106 
107 	/* Split event channels support, tx_* == rx_* when using
108 	 * single event channel.
109 	 */
110 	unsigned int tx_evtchn, rx_evtchn;
111 	unsigned int tx_irq, rx_irq;
112 	/* Only used when split event channels support is enabled */
113 	char tx_irq_name[IRQ_NAME_SIZE]; /* DEVNAME-qN-tx */
114 	char rx_irq_name[IRQ_NAME_SIZE]; /* DEVNAME-qN-rx */
115 
116 	spinlock_t   tx_lock;
117 	struct xen_netif_tx_front_ring tx;
118 	int tx_ring_ref;
119 
120 	/*
121 	 * {tx,rx}_skbs store outstanding skbuffs. Free tx_skb entries
122 	 * are linked from tx_skb_freelist through skb_entry.link.
123 	 *
124 	 *  NB. Freelist index entries are always going to be less than
125 	 *  PAGE_OFFSET, whereas pointers to skbs will always be equal or
126 	 *  greater than PAGE_OFFSET: we use this property to distinguish
127 	 *  them.
128 	 */
129 	union skb_entry {
130 		struct sk_buff *skb;
131 		unsigned long link;
132 	} tx_skbs[NET_TX_RING_SIZE];
133 	grant_ref_t gref_tx_head;
134 	grant_ref_t grant_tx_ref[NET_TX_RING_SIZE];
135 	struct page *grant_tx_page[NET_TX_RING_SIZE];
136 	unsigned tx_skb_freelist;
137 
138 	spinlock_t   rx_lock ____cacheline_aligned_in_smp;
139 	struct xen_netif_rx_front_ring rx;
140 	int rx_ring_ref;
141 
142 	struct timer_list rx_refill_timer;
143 
144 	struct sk_buff *rx_skbs[NET_RX_RING_SIZE];
145 	grant_ref_t gref_rx_head;
146 	grant_ref_t grant_rx_ref[NET_RX_RING_SIZE];
147 };
148 
149 struct netfront_info {
150 	struct list_head list;
151 	struct net_device *netdev;
152 
153 	struct xenbus_device *xbdev;
154 
155 	/* Multi-queue support */
156 	struct netfront_queue *queues;
157 
158 	/* Statistics */
159 	struct netfront_stats __percpu *rx_stats;
160 	struct netfront_stats __percpu *tx_stats;
161 
162 	atomic_t rx_gso_checksum_fixup;
163 };
164 
165 struct netfront_rx_info {
166 	struct xen_netif_rx_response rx;
167 	struct xen_netif_extra_info extras[XEN_NETIF_EXTRA_TYPE_MAX - 1];
168 };
169 
170 static void skb_entry_set_link(union skb_entry *list, unsigned short id)
171 {
172 	list->link = id;
173 }
174 
175 static int skb_entry_is_link(const union skb_entry *list)
176 {
177 	BUILD_BUG_ON(sizeof(list->skb) != sizeof(list->link));
178 	return (unsigned long)list->skb < PAGE_OFFSET;
179 }
180 
181 /*
182  * Access macros for acquiring freeing slots in tx_skbs[].
183  */
184 
185 static void add_id_to_freelist(unsigned *head, union skb_entry *list,
186 			       unsigned short id)
187 {
188 	skb_entry_set_link(&list[id], *head);
189 	*head = id;
190 }
191 
192 static unsigned short get_id_from_freelist(unsigned *head,
193 					   union skb_entry *list)
194 {
195 	unsigned int id = *head;
196 	*head = list[id].link;
197 	return id;
198 }
199 
200 static int xennet_rxidx(RING_IDX idx)
201 {
202 	return idx & (NET_RX_RING_SIZE - 1);
203 }
204 
205 static struct sk_buff *xennet_get_rx_skb(struct netfront_queue *queue,
206 					 RING_IDX ri)
207 {
208 	int i = xennet_rxidx(ri);
209 	struct sk_buff *skb = queue->rx_skbs[i];
210 	queue->rx_skbs[i] = NULL;
211 	return skb;
212 }
213 
214 static grant_ref_t xennet_get_rx_ref(struct netfront_queue *queue,
215 					    RING_IDX ri)
216 {
217 	int i = xennet_rxidx(ri);
218 	grant_ref_t ref = queue->grant_rx_ref[i];
219 	queue->grant_rx_ref[i] = GRANT_INVALID_REF;
220 	return ref;
221 }
222 
223 #ifdef CONFIG_SYSFS
224 static const struct attribute_group xennet_dev_group;
225 #endif
226 
227 static bool xennet_can_sg(struct net_device *dev)
228 {
229 	return dev->features & NETIF_F_SG;
230 }
231 
232 
233 static void rx_refill_timeout(struct timer_list *t)
234 {
235 	struct netfront_queue *queue = from_timer(queue, t, rx_refill_timer);
236 	napi_schedule(&queue->napi);
237 }
238 
239 static int netfront_tx_slot_available(struct netfront_queue *queue)
240 {
241 	return (queue->tx.req_prod_pvt - queue->tx.rsp_cons) <
242 		(NET_TX_RING_SIZE - XEN_NETIF_NR_SLOTS_MIN - 1);
243 }
244 
245 static void xennet_maybe_wake_tx(struct netfront_queue *queue)
246 {
247 	struct net_device *dev = queue->info->netdev;
248 	struct netdev_queue *dev_queue = netdev_get_tx_queue(dev, queue->id);
249 
250 	if (unlikely(netif_tx_queue_stopped(dev_queue)) &&
251 	    netfront_tx_slot_available(queue) &&
252 	    likely(netif_running(dev)))
253 		netif_tx_wake_queue(netdev_get_tx_queue(dev, queue->id));
254 }
255 
256 
257 static struct sk_buff *xennet_alloc_one_rx_buffer(struct netfront_queue *queue)
258 {
259 	struct sk_buff *skb;
260 	struct page *page;
261 
262 	skb = __netdev_alloc_skb(queue->info->netdev,
263 				 RX_COPY_THRESHOLD + NET_IP_ALIGN,
264 				 GFP_ATOMIC | __GFP_NOWARN);
265 	if (unlikely(!skb))
266 		return NULL;
267 
268 	page = alloc_page(GFP_ATOMIC | __GFP_NOWARN);
269 	if (!page) {
270 		kfree_skb(skb);
271 		return NULL;
272 	}
273 	skb_add_rx_frag(skb, 0, page, 0, 0, PAGE_SIZE);
274 
275 	/* Align ip header to a 16 bytes boundary */
276 	skb_reserve(skb, NET_IP_ALIGN);
277 	skb->dev = queue->info->netdev;
278 
279 	return skb;
280 }
281 
282 
283 static void xennet_alloc_rx_buffers(struct netfront_queue *queue)
284 {
285 	RING_IDX req_prod = queue->rx.req_prod_pvt;
286 	int notify;
287 	int err = 0;
288 
289 	if (unlikely(!netif_carrier_ok(queue->info->netdev)))
290 		return;
291 
292 	for (req_prod = queue->rx.req_prod_pvt;
293 	     req_prod - queue->rx.rsp_cons < NET_RX_RING_SIZE;
294 	     req_prod++) {
295 		struct sk_buff *skb;
296 		unsigned short id;
297 		grant_ref_t ref;
298 		struct page *page;
299 		struct xen_netif_rx_request *req;
300 
301 		skb = xennet_alloc_one_rx_buffer(queue);
302 		if (!skb) {
303 			err = -ENOMEM;
304 			break;
305 		}
306 
307 		id = xennet_rxidx(req_prod);
308 
309 		BUG_ON(queue->rx_skbs[id]);
310 		queue->rx_skbs[id] = skb;
311 
312 		ref = gnttab_claim_grant_reference(&queue->gref_rx_head);
313 		WARN_ON_ONCE(IS_ERR_VALUE((unsigned long)(int)ref));
314 		queue->grant_rx_ref[id] = ref;
315 
316 		page = skb_frag_page(&skb_shinfo(skb)->frags[0]);
317 
318 		req = RING_GET_REQUEST(&queue->rx, req_prod);
319 		gnttab_page_grant_foreign_access_ref_one(ref,
320 							 queue->info->xbdev->otherend_id,
321 							 page,
322 							 0);
323 		req->id = id;
324 		req->gref = ref;
325 	}
326 
327 	queue->rx.req_prod_pvt = req_prod;
328 
329 	/* Try again later if there are not enough requests or skb allocation
330 	 * failed.
331 	 * Enough requests is quantified as the sum of newly created slots and
332 	 * the unconsumed slots at the backend.
333 	 */
334 	if (req_prod - queue->rx.rsp_cons < NET_RX_SLOTS_MIN ||
335 	    unlikely(err)) {
336 		mod_timer(&queue->rx_refill_timer, jiffies + (HZ/10));
337 		return;
338 	}
339 
340 	RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&queue->rx, notify);
341 	if (notify)
342 		notify_remote_via_irq(queue->rx_irq);
343 }
344 
345 static int xennet_open(struct net_device *dev)
346 {
347 	struct netfront_info *np = netdev_priv(dev);
348 	unsigned int num_queues = dev->real_num_tx_queues;
349 	unsigned int i = 0;
350 	struct netfront_queue *queue = NULL;
351 
352 	if (!np->queues)
353 		return -ENODEV;
354 
355 	for (i = 0; i < num_queues; ++i) {
356 		queue = &np->queues[i];
357 		napi_enable(&queue->napi);
358 
359 		spin_lock_bh(&queue->rx_lock);
360 		if (netif_carrier_ok(dev)) {
361 			xennet_alloc_rx_buffers(queue);
362 			queue->rx.sring->rsp_event = queue->rx.rsp_cons + 1;
363 			if (RING_HAS_UNCONSUMED_RESPONSES(&queue->rx))
364 				napi_schedule(&queue->napi);
365 		}
366 		spin_unlock_bh(&queue->rx_lock);
367 	}
368 
369 	netif_tx_start_all_queues(dev);
370 
371 	return 0;
372 }
373 
374 static void xennet_tx_buf_gc(struct netfront_queue *queue)
375 {
376 	RING_IDX cons, prod;
377 	unsigned short id;
378 	struct sk_buff *skb;
379 	bool more_to_do;
380 
381 	BUG_ON(!netif_carrier_ok(queue->info->netdev));
382 
383 	do {
384 		prod = queue->tx.sring->rsp_prod;
385 		rmb(); /* Ensure we see responses up to 'rp'. */
386 
387 		for (cons = queue->tx.rsp_cons; cons != prod; cons++) {
388 			struct xen_netif_tx_response *txrsp;
389 
390 			txrsp = RING_GET_RESPONSE(&queue->tx, cons);
391 			if (txrsp->status == XEN_NETIF_RSP_NULL)
392 				continue;
393 
394 			id  = txrsp->id;
395 			skb = queue->tx_skbs[id].skb;
396 			if (unlikely(gnttab_query_foreign_access(
397 				queue->grant_tx_ref[id]) != 0)) {
398 				pr_alert("%s: warning -- grant still in use by backend domain\n",
399 					 __func__);
400 				BUG();
401 			}
402 			gnttab_end_foreign_access_ref(
403 				queue->grant_tx_ref[id], GNTMAP_readonly);
404 			gnttab_release_grant_reference(
405 				&queue->gref_tx_head, queue->grant_tx_ref[id]);
406 			queue->grant_tx_ref[id] = GRANT_INVALID_REF;
407 			queue->grant_tx_page[id] = NULL;
408 			add_id_to_freelist(&queue->tx_skb_freelist, queue->tx_skbs, id);
409 			dev_kfree_skb_irq(skb);
410 		}
411 
412 		queue->tx.rsp_cons = prod;
413 
414 		RING_FINAL_CHECK_FOR_RESPONSES(&queue->tx, more_to_do);
415 	} while (more_to_do);
416 
417 	xennet_maybe_wake_tx(queue);
418 }
419 
420 struct xennet_gnttab_make_txreq {
421 	struct netfront_queue *queue;
422 	struct sk_buff *skb;
423 	struct page *page;
424 	struct xen_netif_tx_request *tx; /* Last request */
425 	unsigned int size;
426 };
427 
428 static void xennet_tx_setup_grant(unsigned long gfn, unsigned int offset,
429 				  unsigned int len, void *data)
430 {
431 	struct xennet_gnttab_make_txreq *info = data;
432 	unsigned int id;
433 	struct xen_netif_tx_request *tx;
434 	grant_ref_t ref;
435 	/* convenient aliases */
436 	struct page *page = info->page;
437 	struct netfront_queue *queue = info->queue;
438 	struct sk_buff *skb = info->skb;
439 
440 	id = get_id_from_freelist(&queue->tx_skb_freelist, queue->tx_skbs);
441 	tx = RING_GET_REQUEST(&queue->tx, queue->tx.req_prod_pvt++);
442 	ref = gnttab_claim_grant_reference(&queue->gref_tx_head);
443 	WARN_ON_ONCE(IS_ERR_VALUE((unsigned long)(int)ref));
444 
445 	gnttab_grant_foreign_access_ref(ref, queue->info->xbdev->otherend_id,
446 					gfn, GNTMAP_readonly);
447 
448 	queue->tx_skbs[id].skb = skb;
449 	queue->grant_tx_page[id] = page;
450 	queue->grant_tx_ref[id] = ref;
451 
452 	tx->id = id;
453 	tx->gref = ref;
454 	tx->offset = offset;
455 	tx->size = len;
456 	tx->flags = 0;
457 
458 	info->tx = tx;
459 	info->size += tx->size;
460 }
461 
462 static struct xen_netif_tx_request *xennet_make_first_txreq(
463 	struct netfront_queue *queue, struct sk_buff *skb,
464 	struct page *page, unsigned int offset, unsigned int len)
465 {
466 	struct xennet_gnttab_make_txreq info = {
467 		.queue = queue,
468 		.skb = skb,
469 		.page = page,
470 		.size = 0,
471 	};
472 
473 	gnttab_for_one_grant(page, offset, len, xennet_tx_setup_grant, &info);
474 
475 	return info.tx;
476 }
477 
478 static void xennet_make_one_txreq(unsigned long gfn, unsigned int offset,
479 				  unsigned int len, void *data)
480 {
481 	struct xennet_gnttab_make_txreq *info = data;
482 
483 	info->tx->flags |= XEN_NETTXF_more_data;
484 	skb_get(info->skb);
485 	xennet_tx_setup_grant(gfn, offset, len, data);
486 }
487 
488 static struct xen_netif_tx_request *xennet_make_txreqs(
489 	struct netfront_queue *queue, struct xen_netif_tx_request *tx,
490 	struct sk_buff *skb, struct page *page,
491 	unsigned int offset, unsigned int len)
492 {
493 	struct xennet_gnttab_make_txreq info = {
494 		.queue = queue,
495 		.skb = skb,
496 		.tx = tx,
497 	};
498 
499 	/* Skip unused frames from start of page */
500 	page += offset >> PAGE_SHIFT;
501 	offset &= ~PAGE_MASK;
502 
503 	while (len) {
504 		info.page = page;
505 		info.size = 0;
506 
507 		gnttab_foreach_grant_in_range(page, offset, len,
508 					      xennet_make_one_txreq,
509 					      &info);
510 
511 		page++;
512 		offset = 0;
513 		len -= info.size;
514 	}
515 
516 	return info.tx;
517 }
518 
519 /*
520  * Count how many ring slots are required to send this skb. Each frag
521  * might be a compound page.
522  */
523 static int xennet_count_skb_slots(struct sk_buff *skb)
524 {
525 	int i, frags = skb_shinfo(skb)->nr_frags;
526 	int slots;
527 
528 	slots = gnttab_count_grant(offset_in_page(skb->data),
529 				   skb_headlen(skb));
530 
531 	for (i = 0; i < frags; i++) {
532 		skb_frag_t *frag = skb_shinfo(skb)->frags + i;
533 		unsigned long size = skb_frag_size(frag);
534 		unsigned long offset = frag->page_offset;
535 
536 		/* Skip unused frames from start of page */
537 		offset &= ~PAGE_MASK;
538 
539 		slots += gnttab_count_grant(offset, size);
540 	}
541 
542 	return slots;
543 }
544 
545 static u16 xennet_select_queue(struct net_device *dev, struct sk_buff *skb,
546 			       struct net_device *sb_dev)
547 {
548 	unsigned int num_queues = dev->real_num_tx_queues;
549 	u32 hash;
550 	u16 queue_idx;
551 
552 	/* First, check if there is only one queue */
553 	if (num_queues == 1) {
554 		queue_idx = 0;
555 	} else {
556 		hash = skb_get_hash(skb);
557 		queue_idx = hash % num_queues;
558 	}
559 
560 	return queue_idx;
561 }
562 
563 #define MAX_XEN_SKB_FRAGS (65536 / XEN_PAGE_SIZE + 1)
564 
565 static netdev_tx_t xennet_start_xmit(struct sk_buff *skb, struct net_device *dev)
566 {
567 	struct netfront_info *np = netdev_priv(dev);
568 	struct netfront_stats *tx_stats = this_cpu_ptr(np->tx_stats);
569 	struct xen_netif_tx_request *tx, *first_tx;
570 	unsigned int i;
571 	int notify;
572 	int slots;
573 	struct page *page;
574 	unsigned int offset;
575 	unsigned int len;
576 	unsigned long flags;
577 	struct netfront_queue *queue = NULL;
578 	unsigned int num_queues = dev->real_num_tx_queues;
579 	u16 queue_index;
580 	struct sk_buff *nskb;
581 
582 	/* Drop the packet if no queues are set up */
583 	if (num_queues < 1)
584 		goto drop;
585 	/* Determine which queue to transmit this SKB on */
586 	queue_index = skb_get_queue_mapping(skb);
587 	queue = &np->queues[queue_index];
588 
589 	/* If skb->len is too big for wire format, drop skb and alert
590 	 * user about misconfiguration.
591 	 */
592 	if (unlikely(skb->len > XEN_NETIF_MAX_TX_SIZE)) {
593 		net_alert_ratelimited(
594 			"xennet: skb->len = %u, too big for wire format\n",
595 			skb->len);
596 		goto drop;
597 	}
598 
599 	slots = xennet_count_skb_slots(skb);
600 	if (unlikely(slots > MAX_XEN_SKB_FRAGS + 1)) {
601 		net_dbg_ratelimited("xennet: skb rides the rocket: %d slots, %d bytes\n",
602 				    slots, skb->len);
603 		if (skb_linearize(skb))
604 			goto drop;
605 	}
606 
607 	page = virt_to_page(skb->data);
608 	offset = offset_in_page(skb->data);
609 
610 	/* The first req should be at least ETH_HLEN size or the packet will be
611 	 * dropped by netback.
612 	 */
613 	if (unlikely(PAGE_SIZE - offset < ETH_HLEN)) {
614 		nskb = skb_copy(skb, GFP_ATOMIC);
615 		if (!nskb)
616 			goto drop;
617 		dev_consume_skb_any(skb);
618 		skb = nskb;
619 		page = virt_to_page(skb->data);
620 		offset = offset_in_page(skb->data);
621 	}
622 
623 	len = skb_headlen(skb);
624 
625 	spin_lock_irqsave(&queue->tx_lock, flags);
626 
627 	if (unlikely(!netif_carrier_ok(dev) ||
628 		     (slots > 1 && !xennet_can_sg(dev)) ||
629 		     netif_needs_gso(skb, netif_skb_features(skb)))) {
630 		spin_unlock_irqrestore(&queue->tx_lock, flags);
631 		goto drop;
632 	}
633 
634 	/* First request for the linear area. */
635 	first_tx = tx = xennet_make_first_txreq(queue, skb,
636 						page, offset, len);
637 	offset += tx->size;
638 	if (offset == PAGE_SIZE) {
639 		page++;
640 		offset = 0;
641 	}
642 	len -= tx->size;
643 
644 	if (skb->ip_summed == CHECKSUM_PARTIAL)
645 		/* local packet? */
646 		tx->flags |= XEN_NETTXF_csum_blank | XEN_NETTXF_data_validated;
647 	else if (skb->ip_summed == CHECKSUM_UNNECESSARY)
648 		/* remote but checksummed. */
649 		tx->flags |= XEN_NETTXF_data_validated;
650 
651 	/* Optional extra info after the first request. */
652 	if (skb_shinfo(skb)->gso_size) {
653 		struct xen_netif_extra_info *gso;
654 
655 		gso = (struct xen_netif_extra_info *)
656 			RING_GET_REQUEST(&queue->tx, queue->tx.req_prod_pvt++);
657 
658 		tx->flags |= XEN_NETTXF_extra_info;
659 
660 		gso->u.gso.size = skb_shinfo(skb)->gso_size;
661 		gso->u.gso.type = (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6) ?
662 			XEN_NETIF_GSO_TYPE_TCPV6 :
663 			XEN_NETIF_GSO_TYPE_TCPV4;
664 		gso->u.gso.pad = 0;
665 		gso->u.gso.features = 0;
666 
667 		gso->type = XEN_NETIF_EXTRA_TYPE_GSO;
668 		gso->flags = 0;
669 	}
670 
671 	/* Requests for the rest of the linear area. */
672 	tx = xennet_make_txreqs(queue, tx, skb, page, offset, len);
673 
674 	/* Requests for all the frags. */
675 	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
676 		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
677 		tx = xennet_make_txreqs(queue, tx, skb,
678 					skb_frag_page(frag), frag->page_offset,
679 					skb_frag_size(frag));
680 	}
681 
682 	/* First request has the packet length. */
683 	first_tx->size = skb->len;
684 
685 	RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&queue->tx, notify);
686 	if (notify)
687 		notify_remote_via_irq(queue->tx_irq);
688 
689 	u64_stats_update_begin(&tx_stats->syncp);
690 	tx_stats->bytes += skb->len;
691 	tx_stats->packets++;
692 	u64_stats_update_end(&tx_stats->syncp);
693 
694 	/* Note: It is not safe to access skb after xennet_tx_buf_gc()! */
695 	xennet_tx_buf_gc(queue);
696 
697 	if (!netfront_tx_slot_available(queue))
698 		netif_tx_stop_queue(netdev_get_tx_queue(dev, queue->id));
699 
700 	spin_unlock_irqrestore(&queue->tx_lock, flags);
701 
702 	return NETDEV_TX_OK;
703 
704  drop:
705 	dev->stats.tx_dropped++;
706 	dev_kfree_skb_any(skb);
707 	return NETDEV_TX_OK;
708 }
709 
710 static int xennet_close(struct net_device *dev)
711 {
712 	struct netfront_info *np = netdev_priv(dev);
713 	unsigned int num_queues = dev->real_num_tx_queues;
714 	unsigned int i;
715 	struct netfront_queue *queue;
716 	netif_tx_stop_all_queues(np->netdev);
717 	for (i = 0; i < num_queues; ++i) {
718 		queue = &np->queues[i];
719 		napi_disable(&queue->napi);
720 	}
721 	return 0;
722 }
723 
724 static void xennet_move_rx_slot(struct netfront_queue *queue, struct sk_buff *skb,
725 				grant_ref_t ref)
726 {
727 	int new = xennet_rxidx(queue->rx.req_prod_pvt);
728 
729 	BUG_ON(queue->rx_skbs[new]);
730 	queue->rx_skbs[new] = skb;
731 	queue->grant_rx_ref[new] = ref;
732 	RING_GET_REQUEST(&queue->rx, queue->rx.req_prod_pvt)->id = new;
733 	RING_GET_REQUEST(&queue->rx, queue->rx.req_prod_pvt)->gref = ref;
734 	queue->rx.req_prod_pvt++;
735 }
736 
737 static int xennet_get_extras(struct netfront_queue *queue,
738 			     struct xen_netif_extra_info *extras,
739 			     RING_IDX rp)
740 
741 {
742 	struct xen_netif_extra_info *extra;
743 	struct device *dev = &queue->info->netdev->dev;
744 	RING_IDX cons = queue->rx.rsp_cons;
745 	int err = 0;
746 
747 	do {
748 		struct sk_buff *skb;
749 		grant_ref_t ref;
750 
751 		if (unlikely(cons + 1 == rp)) {
752 			if (net_ratelimit())
753 				dev_warn(dev, "Missing extra info\n");
754 			err = -EBADR;
755 			break;
756 		}
757 
758 		extra = (struct xen_netif_extra_info *)
759 			RING_GET_RESPONSE(&queue->rx, ++cons);
760 
761 		if (unlikely(!extra->type ||
762 			     extra->type >= XEN_NETIF_EXTRA_TYPE_MAX)) {
763 			if (net_ratelimit())
764 				dev_warn(dev, "Invalid extra type: %d\n",
765 					extra->type);
766 			err = -EINVAL;
767 		} else {
768 			memcpy(&extras[extra->type - 1], extra,
769 			       sizeof(*extra));
770 		}
771 
772 		skb = xennet_get_rx_skb(queue, cons);
773 		ref = xennet_get_rx_ref(queue, cons);
774 		xennet_move_rx_slot(queue, skb, ref);
775 	} while (extra->flags & XEN_NETIF_EXTRA_FLAG_MORE);
776 
777 	queue->rx.rsp_cons = cons;
778 	return err;
779 }
780 
781 static int xennet_get_responses(struct netfront_queue *queue,
782 				struct netfront_rx_info *rinfo, RING_IDX rp,
783 				struct sk_buff_head *list)
784 {
785 	struct xen_netif_rx_response *rx = &rinfo->rx;
786 	struct xen_netif_extra_info *extras = rinfo->extras;
787 	struct device *dev = &queue->info->netdev->dev;
788 	RING_IDX cons = queue->rx.rsp_cons;
789 	struct sk_buff *skb = xennet_get_rx_skb(queue, cons);
790 	grant_ref_t ref = xennet_get_rx_ref(queue, cons);
791 	int max = XEN_NETIF_NR_SLOTS_MIN + (rx->status <= RX_COPY_THRESHOLD);
792 	int slots = 1;
793 	int err = 0;
794 	unsigned long ret;
795 
796 	if (rx->flags & XEN_NETRXF_extra_info) {
797 		err = xennet_get_extras(queue, extras, rp);
798 		cons = queue->rx.rsp_cons;
799 	}
800 
801 	for (;;) {
802 		if (unlikely(rx->status < 0 ||
803 			     rx->offset + rx->status > XEN_PAGE_SIZE)) {
804 			if (net_ratelimit())
805 				dev_warn(dev, "rx->offset: %u, size: %d\n",
806 					 rx->offset, rx->status);
807 			xennet_move_rx_slot(queue, skb, ref);
808 			err = -EINVAL;
809 			goto next;
810 		}
811 
812 		/*
813 		 * This definitely indicates a bug, either in this driver or in
814 		 * the backend driver. In future this should flag the bad
815 		 * situation to the system controller to reboot the backend.
816 		 */
817 		if (ref == GRANT_INVALID_REF) {
818 			if (net_ratelimit())
819 				dev_warn(dev, "Bad rx response id %d.\n",
820 					 rx->id);
821 			err = -EINVAL;
822 			goto next;
823 		}
824 
825 		ret = gnttab_end_foreign_access_ref(ref, 0);
826 		BUG_ON(!ret);
827 
828 		gnttab_release_grant_reference(&queue->gref_rx_head, ref);
829 
830 		__skb_queue_tail(list, skb);
831 
832 next:
833 		if (!(rx->flags & XEN_NETRXF_more_data))
834 			break;
835 
836 		if (cons + slots == rp) {
837 			if (net_ratelimit())
838 				dev_warn(dev, "Need more slots\n");
839 			err = -ENOENT;
840 			break;
841 		}
842 
843 		rx = RING_GET_RESPONSE(&queue->rx, cons + slots);
844 		skb = xennet_get_rx_skb(queue, cons + slots);
845 		ref = xennet_get_rx_ref(queue, cons + slots);
846 		slots++;
847 	}
848 
849 	if (unlikely(slots > max)) {
850 		if (net_ratelimit())
851 			dev_warn(dev, "Too many slots\n");
852 		err = -E2BIG;
853 	}
854 
855 	if (unlikely(err))
856 		queue->rx.rsp_cons = cons + slots;
857 
858 	return err;
859 }
860 
861 static int xennet_set_skb_gso(struct sk_buff *skb,
862 			      struct xen_netif_extra_info *gso)
863 {
864 	if (!gso->u.gso.size) {
865 		if (net_ratelimit())
866 			pr_warn("GSO size must not be zero\n");
867 		return -EINVAL;
868 	}
869 
870 	if (gso->u.gso.type != XEN_NETIF_GSO_TYPE_TCPV4 &&
871 	    gso->u.gso.type != XEN_NETIF_GSO_TYPE_TCPV6) {
872 		if (net_ratelimit())
873 			pr_warn("Bad GSO type %d\n", gso->u.gso.type);
874 		return -EINVAL;
875 	}
876 
877 	skb_shinfo(skb)->gso_size = gso->u.gso.size;
878 	skb_shinfo(skb)->gso_type =
879 		(gso->u.gso.type == XEN_NETIF_GSO_TYPE_TCPV4) ?
880 		SKB_GSO_TCPV4 :
881 		SKB_GSO_TCPV6;
882 
883 	/* Header must be checked, and gso_segs computed. */
884 	skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY;
885 	skb_shinfo(skb)->gso_segs = 0;
886 
887 	return 0;
888 }
889 
890 static RING_IDX xennet_fill_frags(struct netfront_queue *queue,
891 				  struct sk_buff *skb,
892 				  struct sk_buff_head *list)
893 {
894 	RING_IDX cons = queue->rx.rsp_cons;
895 	struct sk_buff *nskb;
896 
897 	while ((nskb = __skb_dequeue(list))) {
898 		struct xen_netif_rx_response *rx =
899 			RING_GET_RESPONSE(&queue->rx, ++cons);
900 		skb_frag_t *nfrag = &skb_shinfo(nskb)->frags[0];
901 
902 		if (skb_shinfo(skb)->nr_frags == MAX_SKB_FRAGS) {
903 			unsigned int pull_to = NETFRONT_SKB_CB(skb)->pull_to;
904 
905 			BUG_ON(pull_to < skb_headlen(skb));
906 			__pskb_pull_tail(skb, pull_to - skb_headlen(skb));
907 		}
908 		if (unlikely(skb_shinfo(skb)->nr_frags >= MAX_SKB_FRAGS)) {
909 			queue->rx.rsp_cons = ++cons;
910 			kfree_skb(nskb);
911 			return ~0U;
912 		}
913 
914 		skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags,
915 				skb_frag_page(nfrag),
916 				rx->offset, rx->status, PAGE_SIZE);
917 
918 		skb_shinfo(nskb)->nr_frags = 0;
919 		kfree_skb(nskb);
920 	}
921 
922 	return cons;
923 }
924 
925 static int checksum_setup(struct net_device *dev, struct sk_buff *skb)
926 {
927 	bool recalculate_partial_csum = false;
928 
929 	/*
930 	 * A GSO SKB must be CHECKSUM_PARTIAL. However some buggy
931 	 * peers can fail to set NETRXF_csum_blank when sending a GSO
932 	 * frame. In this case force the SKB to CHECKSUM_PARTIAL and
933 	 * recalculate the partial checksum.
934 	 */
935 	if (skb->ip_summed != CHECKSUM_PARTIAL && skb_is_gso(skb)) {
936 		struct netfront_info *np = netdev_priv(dev);
937 		atomic_inc(&np->rx_gso_checksum_fixup);
938 		skb->ip_summed = CHECKSUM_PARTIAL;
939 		recalculate_partial_csum = true;
940 	}
941 
942 	/* A non-CHECKSUM_PARTIAL SKB does not require setup. */
943 	if (skb->ip_summed != CHECKSUM_PARTIAL)
944 		return 0;
945 
946 	return skb_checksum_setup(skb, recalculate_partial_csum);
947 }
948 
949 static int handle_incoming_queue(struct netfront_queue *queue,
950 				 struct sk_buff_head *rxq)
951 {
952 	struct netfront_stats *rx_stats = this_cpu_ptr(queue->info->rx_stats);
953 	int packets_dropped = 0;
954 	struct sk_buff *skb;
955 
956 	while ((skb = __skb_dequeue(rxq)) != NULL) {
957 		int pull_to = NETFRONT_SKB_CB(skb)->pull_to;
958 
959 		if (pull_to > skb_headlen(skb))
960 			__pskb_pull_tail(skb, pull_to - skb_headlen(skb));
961 
962 		/* Ethernet work: Delayed to here as it peeks the header. */
963 		skb->protocol = eth_type_trans(skb, queue->info->netdev);
964 		skb_reset_network_header(skb);
965 
966 		if (checksum_setup(queue->info->netdev, skb)) {
967 			kfree_skb(skb);
968 			packets_dropped++;
969 			queue->info->netdev->stats.rx_errors++;
970 			continue;
971 		}
972 
973 		u64_stats_update_begin(&rx_stats->syncp);
974 		rx_stats->packets++;
975 		rx_stats->bytes += skb->len;
976 		u64_stats_update_end(&rx_stats->syncp);
977 
978 		/* Pass it up. */
979 		napi_gro_receive(&queue->napi, skb);
980 	}
981 
982 	return packets_dropped;
983 }
984 
985 static int xennet_poll(struct napi_struct *napi, int budget)
986 {
987 	struct netfront_queue *queue = container_of(napi, struct netfront_queue, napi);
988 	struct net_device *dev = queue->info->netdev;
989 	struct sk_buff *skb;
990 	struct netfront_rx_info rinfo;
991 	struct xen_netif_rx_response *rx = &rinfo.rx;
992 	struct xen_netif_extra_info *extras = rinfo.extras;
993 	RING_IDX i, rp;
994 	int work_done;
995 	struct sk_buff_head rxq;
996 	struct sk_buff_head errq;
997 	struct sk_buff_head tmpq;
998 	int err;
999 
1000 	spin_lock(&queue->rx_lock);
1001 
1002 	skb_queue_head_init(&rxq);
1003 	skb_queue_head_init(&errq);
1004 	skb_queue_head_init(&tmpq);
1005 
1006 	rp = queue->rx.sring->rsp_prod;
1007 	rmb(); /* Ensure we see queued responses up to 'rp'. */
1008 
1009 	i = queue->rx.rsp_cons;
1010 	work_done = 0;
1011 	while ((i != rp) && (work_done < budget)) {
1012 		memcpy(rx, RING_GET_RESPONSE(&queue->rx, i), sizeof(*rx));
1013 		memset(extras, 0, sizeof(rinfo.extras));
1014 
1015 		err = xennet_get_responses(queue, &rinfo, rp, &tmpq);
1016 
1017 		if (unlikely(err)) {
1018 err:
1019 			while ((skb = __skb_dequeue(&tmpq)))
1020 				__skb_queue_tail(&errq, skb);
1021 			dev->stats.rx_errors++;
1022 			i = queue->rx.rsp_cons;
1023 			continue;
1024 		}
1025 
1026 		skb = __skb_dequeue(&tmpq);
1027 
1028 		if (extras[XEN_NETIF_EXTRA_TYPE_GSO - 1].type) {
1029 			struct xen_netif_extra_info *gso;
1030 			gso = &extras[XEN_NETIF_EXTRA_TYPE_GSO - 1];
1031 
1032 			if (unlikely(xennet_set_skb_gso(skb, gso))) {
1033 				__skb_queue_head(&tmpq, skb);
1034 				queue->rx.rsp_cons += skb_queue_len(&tmpq);
1035 				goto err;
1036 			}
1037 		}
1038 
1039 		NETFRONT_SKB_CB(skb)->pull_to = rx->status;
1040 		if (NETFRONT_SKB_CB(skb)->pull_to > RX_COPY_THRESHOLD)
1041 			NETFRONT_SKB_CB(skb)->pull_to = RX_COPY_THRESHOLD;
1042 
1043 		skb_shinfo(skb)->frags[0].page_offset = rx->offset;
1044 		skb_frag_size_set(&skb_shinfo(skb)->frags[0], rx->status);
1045 		skb->data_len = rx->status;
1046 		skb->len += rx->status;
1047 
1048 		i = xennet_fill_frags(queue, skb, &tmpq);
1049 		if (unlikely(i == ~0U))
1050 			goto err;
1051 
1052 		if (rx->flags & XEN_NETRXF_csum_blank)
1053 			skb->ip_summed = CHECKSUM_PARTIAL;
1054 		else if (rx->flags & XEN_NETRXF_data_validated)
1055 			skb->ip_summed = CHECKSUM_UNNECESSARY;
1056 
1057 		__skb_queue_tail(&rxq, skb);
1058 
1059 		queue->rx.rsp_cons = ++i;
1060 		work_done++;
1061 	}
1062 
1063 	__skb_queue_purge(&errq);
1064 
1065 	work_done -= handle_incoming_queue(queue, &rxq);
1066 
1067 	xennet_alloc_rx_buffers(queue);
1068 
1069 	if (work_done < budget) {
1070 		int more_to_do = 0;
1071 
1072 		napi_complete_done(napi, work_done);
1073 
1074 		RING_FINAL_CHECK_FOR_RESPONSES(&queue->rx, more_to_do);
1075 		if (more_to_do)
1076 			napi_schedule(napi);
1077 	}
1078 
1079 	spin_unlock(&queue->rx_lock);
1080 
1081 	return work_done;
1082 }
1083 
1084 static int xennet_change_mtu(struct net_device *dev, int mtu)
1085 {
1086 	int max = xennet_can_sg(dev) ? XEN_NETIF_MAX_TX_SIZE : ETH_DATA_LEN;
1087 
1088 	if (mtu > max)
1089 		return -EINVAL;
1090 	dev->mtu = mtu;
1091 	return 0;
1092 }
1093 
1094 static void xennet_get_stats64(struct net_device *dev,
1095 			       struct rtnl_link_stats64 *tot)
1096 {
1097 	struct netfront_info *np = netdev_priv(dev);
1098 	int cpu;
1099 
1100 	for_each_possible_cpu(cpu) {
1101 		struct netfront_stats *rx_stats = per_cpu_ptr(np->rx_stats, cpu);
1102 		struct netfront_stats *tx_stats = per_cpu_ptr(np->tx_stats, cpu);
1103 		u64 rx_packets, rx_bytes, tx_packets, tx_bytes;
1104 		unsigned int start;
1105 
1106 		do {
1107 			start = u64_stats_fetch_begin_irq(&tx_stats->syncp);
1108 			tx_packets = tx_stats->packets;
1109 			tx_bytes = tx_stats->bytes;
1110 		} while (u64_stats_fetch_retry_irq(&tx_stats->syncp, start));
1111 
1112 		do {
1113 			start = u64_stats_fetch_begin_irq(&rx_stats->syncp);
1114 			rx_packets = rx_stats->packets;
1115 			rx_bytes = rx_stats->bytes;
1116 		} while (u64_stats_fetch_retry_irq(&rx_stats->syncp, start));
1117 
1118 		tot->rx_packets += rx_packets;
1119 		tot->tx_packets += tx_packets;
1120 		tot->rx_bytes   += rx_bytes;
1121 		tot->tx_bytes   += tx_bytes;
1122 	}
1123 
1124 	tot->rx_errors  = dev->stats.rx_errors;
1125 	tot->tx_dropped = dev->stats.tx_dropped;
1126 }
1127 
1128 static void xennet_release_tx_bufs(struct netfront_queue *queue)
1129 {
1130 	struct sk_buff *skb;
1131 	int i;
1132 
1133 	for (i = 0; i < NET_TX_RING_SIZE; i++) {
1134 		/* Skip over entries which are actually freelist references */
1135 		if (skb_entry_is_link(&queue->tx_skbs[i]))
1136 			continue;
1137 
1138 		skb = queue->tx_skbs[i].skb;
1139 		get_page(queue->grant_tx_page[i]);
1140 		gnttab_end_foreign_access(queue->grant_tx_ref[i],
1141 					  GNTMAP_readonly,
1142 					  (unsigned long)page_address(queue->grant_tx_page[i]));
1143 		queue->grant_tx_page[i] = NULL;
1144 		queue->grant_tx_ref[i] = GRANT_INVALID_REF;
1145 		add_id_to_freelist(&queue->tx_skb_freelist, queue->tx_skbs, i);
1146 		dev_kfree_skb_irq(skb);
1147 	}
1148 }
1149 
1150 static void xennet_release_rx_bufs(struct netfront_queue *queue)
1151 {
1152 	int id, ref;
1153 
1154 	spin_lock_bh(&queue->rx_lock);
1155 
1156 	for (id = 0; id < NET_RX_RING_SIZE; id++) {
1157 		struct sk_buff *skb;
1158 		struct page *page;
1159 
1160 		skb = queue->rx_skbs[id];
1161 		if (!skb)
1162 			continue;
1163 
1164 		ref = queue->grant_rx_ref[id];
1165 		if (ref == GRANT_INVALID_REF)
1166 			continue;
1167 
1168 		page = skb_frag_page(&skb_shinfo(skb)->frags[0]);
1169 
1170 		/* gnttab_end_foreign_access() needs a page ref until
1171 		 * foreign access is ended (which may be deferred).
1172 		 */
1173 		get_page(page);
1174 		gnttab_end_foreign_access(ref, 0,
1175 					  (unsigned long)page_address(page));
1176 		queue->grant_rx_ref[id] = GRANT_INVALID_REF;
1177 
1178 		kfree_skb(skb);
1179 	}
1180 
1181 	spin_unlock_bh(&queue->rx_lock);
1182 }
1183 
1184 static netdev_features_t xennet_fix_features(struct net_device *dev,
1185 	netdev_features_t features)
1186 {
1187 	struct netfront_info *np = netdev_priv(dev);
1188 
1189 	if (features & NETIF_F_SG &&
1190 	    !xenbus_read_unsigned(np->xbdev->otherend, "feature-sg", 0))
1191 		features &= ~NETIF_F_SG;
1192 
1193 	if (features & NETIF_F_IPV6_CSUM &&
1194 	    !xenbus_read_unsigned(np->xbdev->otherend,
1195 				  "feature-ipv6-csum-offload", 0))
1196 		features &= ~NETIF_F_IPV6_CSUM;
1197 
1198 	if (features & NETIF_F_TSO &&
1199 	    !xenbus_read_unsigned(np->xbdev->otherend, "feature-gso-tcpv4", 0))
1200 		features &= ~NETIF_F_TSO;
1201 
1202 	if (features & NETIF_F_TSO6 &&
1203 	    !xenbus_read_unsigned(np->xbdev->otherend, "feature-gso-tcpv6", 0))
1204 		features &= ~NETIF_F_TSO6;
1205 
1206 	return features;
1207 }
1208 
1209 static int xennet_set_features(struct net_device *dev,
1210 	netdev_features_t features)
1211 {
1212 	if (!(features & NETIF_F_SG) && dev->mtu > ETH_DATA_LEN) {
1213 		netdev_info(dev, "Reducing MTU because no SG offload");
1214 		dev->mtu = ETH_DATA_LEN;
1215 	}
1216 
1217 	return 0;
1218 }
1219 
1220 static irqreturn_t xennet_tx_interrupt(int irq, void *dev_id)
1221 {
1222 	struct netfront_queue *queue = dev_id;
1223 	unsigned long flags;
1224 
1225 	spin_lock_irqsave(&queue->tx_lock, flags);
1226 	xennet_tx_buf_gc(queue);
1227 	spin_unlock_irqrestore(&queue->tx_lock, flags);
1228 
1229 	return IRQ_HANDLED;
1230 }
1231 
1232 static irqreturn_t xennet_rx_interrupt(int irq, void *dev_id)
1233 {
1234 	struct netfront_queue *queue = dev_id;
1235 	struct net_device *dev = queue->info->netdev;
1236 
1237 	if (likely(netif_carrier_ok(dev) &&
1238 		   RING_HAS_UNCONSUMED_RESPONSES(&queue->rx)))
1239 		napi_schedule(&queue->napi);
1240 
1241 	return IRQ_HANDLED;
1242 }
1243 
1244 static irqreturn_t xennet_interrupt(int irq, void *dev_id)
1245 {
1246 	xennet_tx_interrupt(irq, dev_id);
1247 	xennet_rx_interrupt(irq, dev_id);
1248 	return IRQ_HANDLED;
1249 }
1250 
1251 #ifdef CONFIG_NET_POLL_CONTROLLER
1252 static void xennet_poll_controller(struct net_device *dev)
1253 {
1254 	/* Poll each queue */
1255 	struct netfront_info *info = netdev_priv(dev);
1256 	unsigned int num_queues = dev->real_num_tx_queues;
1257 	unsigned int i;
1258 	for (i = 0; i < num_queues; ++i)
1259 		xennet_interrupt(0, &info->queues[i]);
1260 }
1261 #endif
1262 
1263 static const struct net_device_ops xennet_netdev_ops = {
1264 	.ndo_open            = xennet_open,
1265 	.ndo_stop            = xennet_close,
1266 	.ndo_start_xmit      = xennet_start_xmit,
1267 	.ndo_change_mtu	     = xennet_change_mtu,
1268 	.ndo_get_stats64     = xennet_get_stats64,
1269 	.ndo_set_mac_address = eth_mac_addr,
1270 	.ndo_validate_addr   = eth_validate_addr,
1271 	.ndo_fix_features    = xennet_fix_features,
1272 	.ndo_set_features    = xennet_set_features,
1273 	.ndo_select_queue    = xennet_select_queue,
1274 #ifdef CONFIG_NET_POLL_CONTROLLER
1275 	.ndo_poll_controller = xennet_poll_controller,
1276 #endif
1277 };
1278 
1279 static void xennet_free_netdev(struct net_device *netdev)
1280 {
1281 	struct netfront_info *np = netdev_priv(netdev);
1282 
1283 	free_percpu(np->rx_stats);
1284 	free_percpu(np->tx_stats);
1285 	free_netdev(netdev);
1286 }
1287 
1288 static struct net_device *xennet_create_dev(struct xenbus_device *dev)
1289 {
1290 	int err;
1291 	struct net_device *netdev;
1292 	struct netfront_info *np;
1293 
1294 	netdev = alloc_etherdev_mq(sizeof(struct netfront_info), xennet_max_queues);
1295 	if (!netdev)
1296 		return ERR_PTR(-ENOMEM);
1297 
1298 	np                   = netdev_priv(netdev);
1299 	np->xbdev            = dev;
1300 
1301 	np->queues = NULL;
1302 
1303 	err = -ENOMEM;
1304 	np->rx_stats = netdev_alloc_pcpu_stats(struct netfront_stats);
1305 	if (np->rx_stats == NULL)
1306 		goto exit;
1307 	np->tx_stats = netdev_alloc_pcpu_stats(struct netfront_stats);
1308 	if (np->tx_stats == NULL)
1309 		goto exit;
1310 
1311 	netdev->netdev_ops	= &xennet_netdev_ops;
1312 
1313 	netdev->features        = NETIF_F_IP_CSUM | NETIF_F_RXCSUM |
1314 				  NETIF_F_GSO_ROBUST;
1315 	netdev->hw_features	= NETIF_F_SG |
1316 				  NETIF_F_IPV6_CSUM |
1317 				  NETIF_F_TSO | NETIF_F_TSO6;
1318 
1319 	/*
1320          * Assume that all hw features are available for now. This set
1321          * will be adjusted by the call to netdev_update_features() in
1322          * xennet_connect() which is the earliest point where we can
1323          * negotiate with the backend regarding supported features.
1324          */
1325 	netdev->features |= netdev->hw_features;
1326 
1327 	netdev->ethtool_ops = &xennet_ethtool_ops;
1328 	netdev->min_mtu = ETH_MIN_MTU;
1329 	netdev->max_mtu = XEN_NETIF_MAX_TX_SIZE;
1330 	SET_NETDEV_DEV(netdev, &dev->dev);
1331 
1332 	np->netdev = netdev;
1333 
1334 	netif_carrier_off(netdev);
1335 
1336 	xenbus_switch_state(dev, XenbusStateInitialising);
1337 	wait_event(module_wq,
1338 		   xenbus_read_driver_state(dev->otherend) !=
1339 		   XenbusStateClosed &&
1340 		   xenbus_read_driver_state(dev->otherend) !=
1341 		   XenbusStateUnknown);
1342 	return netdev;
1343 
1344  exit:
1345 	xennet_free_netdev(netdev);
1346 	return ERR_PTR(err);
1347 }
1348 
1349 /**
1350  * Entry point to this code when a new device is created.  Allocate the basic
1351  * structures and the ring buffers for communication with the backend, and
1352  * inform the backend of the appropriate details for those.
1353  */
1354 static int netfront_probe(struct xenbus_device *dev,
1355 			  const struct xenbus_device_id *id)
1356 {
1357 	int err;
1358 	struct net_device *netdev;
1359 	struct netfront_info *info;
1360 
1361 	netdev = xennet_create_dev(dev);
1362 	if (IS_ERR(netdev)) {
1363 		err = PTR_ERR(netdev);
1364 		xenbus_dev_fatal(dev, err, "creating netdev");
1365 		return err;
1366 	}
1367 
1368 	info = netdev_priv(netdev);
1369 	dev_set_drvdata(&dev->dev, info);
1370 #ifdef CONFIG_SYSFS
1371 	info->netdev->sysfs_groups[0] = &xennet_dev_group;
1372 #endif
1373 
1374 	return 0;
1375 }
1376 
1377 static void xennet_end_access(int ref, void *page)
1378 {
1379 	/* This frees the page as a side-effect */
1380 	if (ref != GRANT_INVALID_REF)
1381 		gnttab_end_foreign_access(ref, 0, (unsigned long)page);
1382 }
1383 
1384 static void xennet_disconnect_backend(struct netfront_info *info)
1385 {
1386 	unsigned int i = 0;
1387 	unsigned int num_queues = info->netdev->real_num_tx_queues;
1388 
1389 	netif_carrier_off(info->netdev);
1390 
1391 	for (i = 0; i < num_queues && info->queues; ++i) {
1392 		struct netfront_queue *queue = &info->queues[i];
1393 
1394 		del_timer_sync(&queue->rx_refill_timer);
1395 
1396 		if (queue->tx_irq && (queue->tx_irq == queue->rx_irq))
1397 			unbind_from_irqhandler(queue->tx_irq, queue);
1398 		if (queue->tx_irq && (queue->tx_irq != queue->rx_irq)) {
1399 			unbind_from_irqhandler(queue->tx_irq, queue);
1400 			unbind_from_irqhandler(queue->rx_irq, queue);
1401 		}
1402 		queue->tx_evtchn = queue->rx_evtchn = 0;
1403 		queue->tx_irq = queue->rx_irq = 0;
1404 
1405 		if (netif_running(info->netdev))
1406 			napi_synchronize(&queue->napi);
1407 
1408 		xennet_release_tx_bufs(queue);
1409 		xennet_release_rx_bufs(queue);
1410 		gnttab_free_grant_references(queue->gref_tx_head);
1411 		gnttab_free_grant_references(queue->gref_rx_head);
1412 
1413 		/* End access and free the pages */
1414 		xennet_end_access(queue->tx_ring_ref, queue->tx.sring);
1415 		xennet_end_access(queue->rx_ring_ref, queue->rx.sring);
1416 
1417 		queue->tx_ring_ref = GRANT_INVALID_REF;
1418 		queue->rx_ring_ref = GRANT_INVALID_REF;
1419 		queue->tx.sring = NULL;
1420 		queue->rx.sring = NULL;
1421 	}
1422 }
1423 
1424 /**
1425  * We are reconnecting to the backend, due to a suspend/resume, or a backend
1426  * driver restart.  We tear down our netif structure and recreate it, but
1427  * leave the device-layer structures intact so that this is transparent to the
1428  * rest of the kernel.
1429  */
1430 static int netfront_resume(struct xenbus_device *dev)
1431 {
1432 	struct netfront_info *info = dev_get_drvdata(&dev->dev);
1433 
1434 	dev_dbg(&dev->dev, "%s\n", dev->nodename);
1435 
1436 	xennet_disconnect_backend(info);
1437 	return 0;
1438 }
1439 
1440 static int xen_net_read_mac(struct xenbus_device *dev, u8 mac[])
1441 {
1442 	char *s, *e, *macstr;
1443 	int i;
1444 
1445 	macstr = s = xenbus_read(XBT_NIL, dev->nodename, "mac", NULL);
1446 	if (IS_ERR(macstr))
1447 		return PTR_ERR(macstr);
1448 
1449 	for (i = 0; i < ETH_ALEN; i++) {
1450 		mac[i] = simple_strtoul(s, &e, 16);
1451 		if ((s == e) || (*e != ((i == ETH_ALEN-1) ? '\0' : ':'))) {
1452 			kfree(macstr);
1453 			return -ENOENT;
1454 		}
1455 		s = e+1;
1456 	}
1457 
1458 	kfree(macstr);
1459 	return 0;
1460 }
1461 
1462 static int setup_netfront_single(struct netfront_queue *queue)
1463 {
1464 	int err;
1465 
1466 	err = xenbus_alloc_evtchn(queue->info->xbdev, &queue->tx_evtchn);
1467 	if (err < 0)
1468 		goto fail;
1469 
1470 	err = bind_evtchn_to_irqhandler(queue->tx_evtchn,
1471 					xennet_interrupt,
1472 					0, queue->info->netdev->name, queue);
1473 	if (err < 0)
1474 		goto bind_fail;
1475 	queue->rx_evtchn = queue->tx_evtchn;
1476 	queue->rx_irq = queue->tx_irq = err;
1477 
1478 	return 0;
1479 
1480 bind_fail:
1481 	xenbus_free_evtchn(queue->info->xbdev, queue->tx_evtchn);
1482 	queue->tx_evtchn = 0;
1483 fail:
1484 	return err;
1485 }
1486 
1487 static int setup_netfront_split(struct netfront_queue *queue)
1488 {
1489 	int err;
1490 
1491 	err = xenbus_alloc_evtchn(queue->info->xbdev, &queue->tx_evtchn);
1492 	if (err < 0)
1493 		goto fail;
1494 	err = xenbus_alloc_evtchn(queue->info->xbdev, &queue->rx_evtchn);
1495 	if (err < 0)
1496 		goto alloc_rx_evtchn_fail;
1497 
1498 	snprintf(queue->tx_irq_name, sizeof(queue->tx_irq_name),
1499 		 "%s-tx", queue->name);
1500 	err = bind_evtchn_to_irqhandler(queue->tx_evtchn,
1501 					xennet_tx_interrupt,
1502 					0, queue->tx_irq_name, queue);
1503 	if (err < 0)
1504 		goto bind_tx_fail;
1505 	queue->tx_irq = err;
1506 
1507 	snprintf(queue->rx_irq_name, sizeof(queue->rx_irq_name),
1508 		 "%s-rx", queue->name);
1509 	err = bind_evtchn_to_irqhandler(queue->rx_evtchn,
1510 					xennet_rx_interrupt,
1511 					0, queue->rx_irq_name, queue);
1512 	if (err < 0)
1513 		goto bind_rx_fail;
1514 	queue->rx_irq = err;
1515 
1516 	return 0;
1517 
1518 bind_rx_fail:
1519 	unbind_from_irqhandler(queue->tx_irq, queue);
1520 	queue->tx_irq = 0;
1521 bind_tx_fail:
1522 	xenbus_free_evtchn(queue->info->xbdev, queue->rx_evtchn);
1523 	queue->rx_evtchn = 0;
1524 alloc_rx_evtchn_fail:
1525 	xenbus_free_evtchn(queue->info->xbdev, queue->tx_evtchn);
1526 	queue->tx_evtchn = 0;
1527 fail:
1528 	return err;
1529 }
1530 
1531 static int setup_netfront(struct xenbus_device *dev,
1532 			struct netfront_queue *queue, unsigned int feature_split_evtchn)
1533 {
1534 	struct xen_netif_tx_sring *txs;
1535 	struct xen_netif_rx_sring *rxs;
1536 	grant_ref_t gref;
1537 	int err;
1538 
1539 	queue->tx_ring_ref = GRANT_INVALID_REF;
1540 	queue->rx_ring_ref = GRANT_INVALID_REF;
1541 	queue->rx.sring = NULL;
1542 	queue->tx.sring = NULL;
1543 
1544 	txs = (struct xen_netif_tx_sring *)get_zeroed_page(GFP_NOIO | __GFP_HIGH);
1545 	if (!txs) {
1546 		err = -ENOMEM;
1547 		xenbus_dev_fatal(dev, err, "allocating tx ring page");
1548 		goto fail;
1549 	}
1550 	SHARED_RING_INIT(txs);
1551 	FRONT_RING_INIT(&queue->tx, txs, XEN_PAGE_SIZE);
1552 
1553 	err = xenbus_grant_ring(dev, txs, 1, &gref);
1554 	if (err < 0)
1555 		goto grant_tx_ring_fail;
1556 	queue->tx_ring_ref = gref;
1557 
1558 	rxs = (struct xen_netif_rx_sring *)get_zeroed_page(GFP_NOIO | __GFP_HIGH);
1559 	if (!rxs) {
1560 		err = -ENOMEM;
1561 		xenbus_dev_fatal(dev, err, "allocating rx ring page");
1562 		goto alloc_rx_ring_fail;
1563 	}
1564 	SHARED_RING_INIT(rxs);
1565 	FRONT_RING_INIT(&queue->rx, rxs, XEN_PAGE_SIZE);
1566 
1567 	err = xenbus_grant_ring(dev, rxs, 1, &gref);
1568 	if (err < 0)
1569 		goto grant_rx_ring_fail;
1570 	queue->rx_ring_ref = gref;
1571 
1572 	if (feature_split_evtchn)
1573 		err = setup_netfront_split(queue);
1574 	/* setup single event channel if
1575 	 *  a) feature-split-event-channels == 0
1576 	 *  b) feature-split-event-channels == 1 but failed to setup
1577 	 */
1578 	if (!feature_split_evtchn || (feature_split_evtchn && err))
1579 		err = setup_netfront_single(queue);
1580 
1581 	if (err)
1582 		goto alloc_evtchn_fail;
1583 
1584 	return 0;
1585 
1586 	/* If we fail to setup netfront, it is safe to just revoke access to
1587 	 * granted pages because backend is not accessing it at this point.
1588 	 */
1589 alloc_evtchn_fail:
1590 	gnttab_end_foreign_access_ref(queue->rx_ring_ref, 0);
1591 grant_rx_ring_fail:
1592 	free_page((unsigned long)rxs);
1593 alloc_rx_ring_fail:
1594 	gnttab_end_foreign_access_ref(queue->tx_ring_ref, 0);
1595 grant_tx_ring_fail:
1596 	free_page((unsigned long)txs);
1597 fail:
1598 	return err;
1599 }
1600 
1601 /* Queue-specific initialisation
1602  * This used to be done in xennet_create_dev() but must now
1603  * be run per-queue.
1604  */
1605 static int xennet_init_queue(struct netfront_queue *queue)
1606 {
1607 	unsigned short i;
1608 	int err = 0;
1609 	char *devid;
1610 
1611 	spin_lock_init(&queue->tx_lock);
1612 	spin_lock_init(&queue->rx_lock);
1613 
1614 	timer_setup(&queue->rx_refill_timer, rx_refill_timeout, 0);
1615 
1616 	devid = strrchr(queue->info->xbdev->nodename, '/') + 1;
1617 	snprintf(queue->name, sizeof(queue->name), "vif%s-q%u",
1618 		 devid, queue->id);
1619 
1620 	/* Initialise tx_skbs as a free chain containing every entry. */
1621 	queue->tx_skb_freelist = 0;
1622 	for (i = 0; i < NET_TX_RING_SIZE; i++) {
1623 		skb_entry_set_link(&queue->tx_skbs[i], i+1);
1624 		queue->grant_tx_ref[i] = GRANT_INVALID_REF;
1625 		queue->grant_tx_page[i] = NULL;
1626 	}
1627 
1628 	/* Clear out rx_skbs */
1629 	for (i = 0; i < NET_RX_RING_SIZE; i++) {
1630 		queue->rx_skbs[i] = NULL;
1631 		queue->grant_rx_ref[i] = GRANT_INVALID_REF;
1632 	}
1633 
1634 	/* A grant for every tx ring slot */
1635 	if (gnttab_alloc_grant_references(NET_TX_RING_SIZE,
1636 					  &queue->gref_tx_head) < 0) {
1637 		pr_alert("can't alloc tx grant refs\n");
1638 		err = -ENOMEM;
1639 		goto exit;
1640 	}
1641 
1642 	/* A grant for every rx ring slot */
1643 	if (gnttab_alloc_grant_references(NET_RX_RING_SIZE,
1644 					  &queue->gref_rx_head) < 0) {
1645 		pr_alert("can't alloc rx grant refs\n");
1646 		err = -ENOMEM;
1647 		goto exit_free_tx;
1648 	}
1649 
1650 	return 0;
1651 
1652  exit_free_tx:
1653 	gnttab_free_grant_references(queue->gref_tx_head);
1654  exit:
1655 	return err;
1656 }
1657 
1658 static int write_queue_xenstore_keys(struct netfront_queue *queue,
1659 			   struct xenbus_transaction *xbt, int write_hierarchical)
1660 {
1661 	/* Write the queue-specific keys into XenStore in the traditional
1662 	 * way for a single queue, or in a queue subkeys for multiple
1663 	 * queues.
1664 	 */
1665 	struct xenbus_device *dev = queue->info->xbdev;
1666 	int err;
1667 	const char *message;
1668 	char *path;
1669 	size_t pathsize;
1670 
1671 	/* Choose the correct place to write the keys */
1672 	if (write_hierarchical) {
1673 		pathsize = strlen(dev->nodename) + 10;
1674 		path = kzalloc(pathsize, GFP_KERNEL);
1675 		if (!path) {
1676 			err = -ENOMEM;
1677 			message = "out of memory while writing ring references";
1678 			goto error;
1679 		}
1680 		snprintf(path, pathsize, "%s/queue-%u",
1681 				dev->nodename, queue->id);
1682 	} else {
1683 		path = (char *)dev->nodename;
1684 	}
1685 
1686 	/* Write ring references */
1687 	err = xenbus_printf(*xbt, path, "tx-ring-ref", "%u",
1688 			queue->tx_ring_ref);
1689 	if (err) {
1690 		message = "writing tx-ring-ref";
1691 		goto error;
1692 	}
1693 
1694 	err = xenbus_printf(*xbt, path, "rx-ring-ref", "%u",
1695 			queue->rx_ring_ref);
1696 	if (err) {
1697 		message = "writing rx-ring-ref";
1698 		goto error;
1699 	}
1700 
1701 	/* Write event channels; taking into account both shared
1702 	 * and split event channel scenarios.
1703 	 */
1704 	if (queue->tx_evtchn == queue->rx_evtchn) {
1705 		/* Shared event channel */
1706 		err = xenbus_printf(*xbt, path,
1707 				"event-channel", "%u", queue->tx_evtchn);
1708 		if (err) {
1709 			message = "writing event-channel";
1710 			goto error;
1711 		}
1712 	} else {
1713 		/* Split event channels */
1714 		err = xenbus_printf(*xbt, path,
1715 				"event-channel-tx", "%u", queue->tx_evtchn);
1716 		if (err) {
1717 			message = "writing event-channel-tx";
1718 			goto error;
1719 		}
1720 
1721 		err = xenbus_printf(*xbt, path,
1722 				"event-channel-rx", "%u", queue->rx_evtchn);
1723 		if (err) {
1724 			message = "writing event-channel-rx";
1725 			goto error;
1726 		}
1727 	}
1728 
1729 	if (write_hierarchical)
1730 		kfree(path);
1731 	return 0;
1732 
1733 error:
1734 	if (write_hierarchical)
1735 		kfree(path);
1736 	xenbus_dev_fatal(dev, err, "%s", message);
1737 	return err;
1738 }
1739 
1740 static void xennet_destroy_queues(struct netfront_info *info)
1741 {
1742 	unsigned int i;
1743 
1744 	for (i = 0; i < info->netdev->real_num_tx_queues; i++) {
1745 		struct netfront_queue *queue = &info->queues[i];
1746 
1747 		if (netif_running(info->netdev))
1748 			napi_disable(&queue->napi);
1749 		netif_napi_del(&queue->napi);
1750 	}
1751 
1752 	kfree(info->queues);
1753 	info->queues = NULL;
1754 }
1755 
1756 static int xennet_create_queues(struct netfront_info *info,
1757 				unsigned int *num_queues)
1758 {
1759 	unsigned int i;
1760 	int ret;
1761 
1762 	info->queues = kcalloc(*num_queues, sizeof(struct netfront_queue),
1763 			       GFP_KERNEL);
1764 	if (!info->queues)
1765 		return -ENOMEM;
1766 
1767 	for (i = 0; i < *num_queues; i++) {
1768 		struct netfront_queue *queue = &info->queues[i];
1769 
1770 		queue->id = i;
1771 		queue->info = info;
1772 
1773 		ret = xennet_init_queue(queue);
1774 		if (ret < 0) {
1775 			dev_warn(&info->xbdev->dev,
1776 				 "only created %d queues\n", i);
1777 			*num_queues = i;
1778 			break;
1779 		}
1780 
1781 		netif_napi_add(queue->info->netdev, &queue->napi,
1782 			       xennet_poll, 64);
1783 		if (netif_running(info->netdev))
1784 			napi_enable(&queue->napi);
1785 	}
1786 
1787 	netif_set_real_num_tx_queues(info->netdev, *num_queues);
1788 
1789 	if (*num_queues == 0) {
1790 		dev_err(&info->xbdev->dev, "no queues\n");
1791 		return -EINVAL;
1792 	}
1793 	return 0;
1794 }
1795 
1796 /* Common code used when first setting up, and when resuming. */
1797 static int talk_to_netback(struct xenbus_device *dev,
1798 			   struct netfront_info *info)
1799 {
1800 	const char *message;
1801 	struct xenbus_transaction xbt;
1802 	int err;
1803 	unsigned int feature_split_evtchn;
1804 	unsigned int i = 0;
1805 	unsigned int max_queues = 0;
1806 	struct netfront_queue *queue = NULL;
1807 	unsigned int num_queues = 1;
1808 
1809 	info->netdev->irq = 0;
1810 
1811 	/* Check if backend supports multiple queues */
1812 	max_queues = xenbus_read_unsigned(info->xbdev->otherend,
1813 					  "multi-queue-max-queues", 1);
1814 	num_queues = min(max_queues, xennet_max_queues);
1815 
1816 	/* Check feature-split-event-channels */
1817 	feature_split_evtchn = xenbus_read_unsigned(info->xbdev->otherend,
1818 					"feature-split-event-channels", 0);
1819 
1820 	/* Read mac addr. */
1821 	err = xen_net_read_mac(dev, info->netdev->dev_addr);
1822 	if (err) {
1823 		xenbus_dev_fatal(dev, err, "parsing %s/mac", dev->nodename);
1824 		goto out_unlocked;
1825 	}
1826 
1827 	rtnl_lock();
1828 	if (info->queues)
1829 		xennet_destroy_queues(info);
1830 
1831 	err = xennet_create_queues(info, &num_queues);
1832 	if (err < 0) {
1833 		xenbus_dev_fatal(dev, err, "creating queues");
1834 		kfree(info->queues);
1835 		info->queues = NULL;
1836 		goto out;
1837 	}
1838 	rtnl_unlock();
1839 
1840 	/* Create shared ring, alloc event channel -- for each queue */
1841 	for (i = 0; i < num_queues; ++i) {
1842 		queue = &info->queues[i];
1843 		err = setup_netfront(dev, queue, feature_split_evtchn);
1844 		if (err)
1845 			goto destroy_ring;
1846 	}
1847 
1848 again:
1849 	err = xenbus_transaction_start(&xbt);
1850 	if (err) {
1851 		xenbus_dev_fatal(dev, err, "starting transaction");
1852 		goto destroy_ring;
1853 	}
1854 
1855 	if (xenbus_exists(XBT_NIL,
1856 			  info->xbdev->otherend, "multi-queue-max-queues")) {
1857 		/* Write the number of queues */
1858 		err = xenbus_printf(xbt, dev->nodename,
1859 				    "multi-queue-num-queues", "%u", num_queues);
1860 		if (err) {
1861 			message = "writing multi-queue-num-queues";
1862 			goto abort_transaction_no_dev_fatal;
1863 		}
1864 	}
1865 
1866 	if (num_queues == 1) {
1867 		err = write_queue_xenstore_keys(&info->queues[0], &xbt, 0); /* flat */
1868 		if (err)
1869 			goto abort_transaction_no_dev_fatal;
1870 	} else {
1871 		/* Write the keys for each queue */
1872 		for (i = 0; i < num_queues; ++i) {
1873 			queue = &info->queues[i];
1874 			err = write_queue_xenstore_keys(queue, &xbt, 1); /* hierarchical */
1875 			if (err)
1876 				goto abort_transaction_no_dev_fatal;
1877 		}
1878 	}
1879 
1880 	/* The remaining keys are not queue-specific */
1881 	err = xenbus_printf(xbt, dev->nodename, "request-rx-copy", "%u",
1882 			    1);
1883 	if (err) {
1884 		message = "writing request-rx-copy";
1885 		goto abort_transaction;
1886 	}
1887 
1888 	err = xenbus_printf(xbt, dev->nodename, "feature-rx-notify", "%d", 1);
1889 	if (err) {
1890 		message = "writing feature-rx-notify";
1891 		goto abort_transaction;
1892 	}
1893 
1894 	err = xenbus_printf(xbt, dev->nodename, "feature-sg", "%d", 1);
1895 	if (err) {
1896 		message = "writing feature-sg";
1897 		goto abort_transaction;
1898 	}
1899 
1900 	err = xenbus_printf(xbt, dev->nodename, "feature-gso-tcpv4", "%d", 1);
1901 	if (err) {
1902 		message = "writing feature-gso-tcpv4";
1903 		goto abort_transaction;
1904 	}
1905 
1906 	err = xenbus_write(xbt, dev->nodename, "feature-gso-tcpv6", "1");
1907 	if (err) {
1908 		message = "writing feature-gso-tcpv6";
1909 		goto abort_transaction;
1910 	}
1911 
1912 	err = xenbus_write(xbt, dev->nodename, "feature-ipv6-csum-offload",
1913 			   "1");
1914 	if (err) {
1915 		message = "writing feature-ipv6-csum-offload";
1916 		goto abort_transaction;
1917 	}
1918 
1919 	err = xenbus_transaction_end(xbt, 0);
1920 	if (err) {
1921 		if (err == -EAGAIN)
1922 			goto again;
1923 		xenbus_dev_fatal(dev, err, "completing transaction");
1924 		goto destroy_ring;
1925 	}
1926 
1927 	return 0;
1928 
1929  abort_transaction:
1930 	xenbus_dev_fatal(dev, err, "%s", message);
1931 abort_transaction_no_dev_fatal:
1932 	xenbus_transaction_end(xbt, 1);
1933  destroy_ring:
1934 	xennet_disconnect_backend(info);
1935 	rtnl_lock();
1936 	xennet_destroy_queues(info);
1937  out:
1938 	rtnl_unlock();
1939 out_unlocked:
1940 	device_unregister(&dev->dev);
1941 	return err;
1942 }
1943 
1944 static int xennet_connect(struct net_device *dev)
1945 {
1946 	struct netfront_info *np = netdev_priv(dev);
1947 	unsigned int num_queues = 0;
1948 	int err;
1949 	unsigned int j = 0;
1950 	struct netfront_queue *queue = NULL;
1951 
1952 	if (!xenbus_read_unsigned(np->xbdev->otherend, "feature-rx-copy", 0)) {
1953 		dev_info(&dev->dev,
1954 			 "backend does not support copying receive path\n");
1955 		return -ENODEV;
1956 	}
1957 
1958 	err = talk_to_netback(np->xbdev, np);
1959 	if (err)
1960 		return err;
1961 
1962 	/* talk_to_netback() sets the correct number of queues */
1963 	num_queues = dev->real_num_tx_queues;
1964 
1965 	if (dev->reg_state == NETREG_UNINITIALIZED) {
1966 		err = register_netdev(dev);
1967 		if (err) {
1968 			pr_warn("%s: register_netdev err=%d\n", __func__, err);
1969 			device_unregister(&np->xbdev->dev);
1970 			return err;
1971 		}
1972 	}
1973 
1974 	rtnl_lock();
1975 	netdev_update_features(dev);
1976 	rtnl_unlock();
1977 
1978 	/*
1979 	 * All public and private state should now be sane.  Get
1980 	 * ready to start sending and receiving packets and give the driver
1981 	 * domain a kick because we've probably just requeued some
1982 	 * packets.
1983 	 */
1984 	netif_carrier_on(np->netdev);
1985 	for (j = 0; j < num_queues; ++j) {
1986 		queue = &np->queues[j];
1987 
1988 		notify_remote_via_irq(queue->tx_irq);
1989 		if (queue->tx_irq != queue->rx_irq)
1990 			notify_remote_via_irq(queue->rx_irq);
1991 
1992 		spin_lock_irq(&queue->tx_lock);
1993 		xennet_tx_buf_gc(queue);
1994 		spin_unlock_irq(&queue->tx_lock);
1995 
1996 		spin_lock_bh(&queue->rx_lock);
1997 		xennet_alloc_rx_buffers(queue);
1998 		spin_unlock_bh(&queue->rx_lock);
1999 	}
2000 
2001 	return 0;
2002 }
2003 
2004 /**
2005  * Callback received when the backend's state changes.
2006  */
2007 static void netback_changed(struct xenbus_device *dev,
2008 			    enum xenbus_state backend_state)
2009 {
2010 	struct netfront_info *np = dev_get_drvdata(&dev->dev);
2011 	struct net_device *netdev = np->netdev;
2012 
2013 	dev_dbg(&dev->dev, "%s\n", xenbus_strstate(backend_state));
2014 
2015 	wake_up_all(&module_wq);
2016 
2017 	switch (backend_state) {
2018 	case XenbusStateInitialising:
2019 	case XenbusStateInitialised:
2020 	case XenbusStateReconfiguring:
2021 	case XenbusStateReconfigured:
2022 	case XenbusStateUnknown:
2023 		break;
2024 
2025 	case XenbusStateInitWait:
2026 		if (dev->state != XenbusStateInitialising)
2027 			break;
2028 		if (xennet_connect(netdev) != 0)
2029 			break;
2030 		xenbus_switch_state(dev, XenbusStateConnected);
2031 		break;
2032 
2033 	case XenbusStateConnected:
2034 		netdev_notify_peers(netdev);
2035 		break;
2036 
2037 	case XenbusStateClosed:
2038 		if (dev->state == XenbusStateClosed)
2039 			break;
2040 		/* Fall through - Missed the backend's CLOSING state. */
2041 	case XenbusStateClosing:
2042 		xenbus_frontend_closed(dev);
2043 		break;
2044 	}
2045 }
2046 
2047 static const struct xennet_stat {
2048 	char name[ETH_GSTRING_LEN];
2049 	u16 offset;
2050 } xennet_stats[] = {
2051 	{
2052 		"rx_gso_checksum_fixup",
2053 		offsetof(struct netfront_info, rx_gso_checksum_fixup)
2054 	},
2055 };
2056 
2057 static int xennet_get_sset_count(struct net_device *dev, int string_set)
2058 {
2059 	switch (string_set) {
2060 	case ETH_SS_STATS:
2061 		return ARRAY_SIZE(xennet_stats);
2062 	default:
2063 		return -EINVAL;
2064 	}
2065 }
2066 
2067 static void xennet_get_ethtool_stats(struct net_device *dev,
2068 				     struct ethtool_stats *stats, u64 * data)
2069 {
2070 	void *np = netdev_priv(dev);
2071 	int i;
2072 
2073 	for (i = 0; i < ARRAY_SIZE(xennet_stats); i++)
2074 		data[i] = atomic_read((atomic_t *)(np + xennet_stats[i].offset));
2075 }
2076 
2077 static void xennet_get_strings(struct net_device *dev, u32 stringset, u8 * data)
2078 {
2079 	int i;
2080 
2081 	switch (stringset) {
2082 	case ETH_SS_STATS:
2083 		for (i = 0; i < ARRAY_SIZE(xennet_stats); i++)
2084 			memcpy(data + i * ETH_GSTRING_LEN,
2085 			       xennet_stats[i].name, ETH_GSTRING_LEN);
2086 		break;
2087 	}
2088 }
2089 
2090 static const struct ethtool_ops xennet_ethtool_ops =
2091 {
2092 	.get_link = ethtool_op_get_link,
2093 
2094 	.get_sset_count = xennet_get_sset_count,
2095 	.get_ethtool_stats = xennet_get_ethtool_stats,
2096 	.get_strings = xennet_get_strings,
2097 };
2098 
2099 #ifdef CONFIG_SYSFS
2100 static ssize_t show_rxbuf(struct device *dev,
2101 			  struct device_attribute *attr, char *buf)
2102 {
2103 	return sprintf(buf, "%lu\n", NET_RX_RING_SIZE);
2104 }
2105 
2106 static ssize_t store_rxbuf(struct device *dev,
2107 			   struct device_attribute *attr,
2108 			   const char *buf, size_t len)
2109 {
2110 	char *endp;
2111 	unsigned long target;
2112 
2113 	if (!capable(CAP_NET_ADMIN))
2114 		return -EPERM;
2115 
2116 	target = simple_strtoul(buf, &endp, 0);
2117 	if (endp == buf)
2118 		return -EBADMSG;
2119 
2120 	/* rxbuf_min and rxbuf_max are no longer configurable. */
2121 
2122 	return len;
2123 }
2124 
2125 static DEVICE_ATTR(rxbuf_min, 0644, show_rxbuf, store_rxbuf);
2126 static DEVICE_ATTR(rxbuf_max, 0644, show_rxbuf, store_rxbuf);
2127 static DEVICE_ATTR(rxbuf_cur, 0444, show_rxbuf, NULL);
2128 
2129 static struct attribute *xennet_dev_attrs[] = {
2130 	&dev_attr_rxbuf_min.attr,
2131 	&dev_attr_rxbuf_max.attr,
2132 	&dev_attr_rxbuf_cur.attr,
2133 	NULL
2134 };
2135 
2136 static const struct attribute_group xennet_dev_group = {
2137 	.attrs = xennet_dev_attrs
2138 };
2139 #endif /* CONFIG_SYSFS */
2140 
2141 static int xennet_remove(struct xenbus_device *dev)
2142 {
2143 	struct netfront_info *info = dev_get_drvdata(&dev->dev);
2144 
2145 	dev_dbg(&dev->dev, "%s\n", dev->nodename);
2146 
2147 	if (xenbus_read_driver_state(dev->otherend) != XenbusStateClosed) {
2148 		xenbus_switch_state(dev, XenbusStateClosing);
2149 		wait_event(module_wq,
2150 			   xenbus_read_driver_state(dev->otherend) ==
2151 			   XenbusStateClosing ||
2152 			   xenbus_read_driver_state(dev->otherend) ==
2153 			   XenbusStateUnknown);
2154 
2155 		xenbus_switch_state(dev, XenbusStateClosed);
2156 		wait_event(module_wq,
2157 			   xenbus_read_driver_state(dev->otherend) ==
2158 			   XenbusStateClosed ||
2159 			   xenbus_read_driver_state(dev->otherend) ==
2160 			   XenbusStateUnknown);
2161 	}
2162 
2163 	xennet_disconnect_backend(info);
2164 
2165 	if (info->netdev->reg_state == NETREG_REGISTERED)
2166 		unregister_netdev(info->netdev);
2167 
2168 	if (info->queues) {
2169 		rtnl_lock();
2170 		xennet_destroy_queues(info);
2171 		rtnl_unlock();
2172 	}
2173 	xennet_free_netdev(info->netdev);
2174 
2175 	return 0;
2176 }
2177 
2178 static const struct xenbus_device_id netfront_ids[] = {
2179 	{ "vif" },
2180 	{ "" }
2181 };
2182 
2183 static struct xenbus_driver netfront_driver = {
2184 	.ids = netfront_ids,
2185 	.probe = netfront_probe,
2186 	.remove = xennet_remove,
2187 	.resume = netfront_resume,
2188 	.otherend_changed = netback_changed,
2189 };
2190 
2191 static int __init netif_init(void)
2192 {
2193 	if (!xen_domain())
2194 		return -ENODEV;
2195 
2196 	if (!xen_has_pv_nic_devices())
2197 		return -ENODEV;
2198 
2199 	pr_info("Initialising Xen virtual ethernet driver\n");
2200 
2201 	/* Allow as many queues as there are CPUs inut max. 8 if user has not
2202 	 * specified a value.
2203 	 */
2204 	if (xennet_max_queues == 0)
2205 		xennet_max_queues = min_t(unsigned int, MAX_QUEUES_DEFAULT,
2206 					  num_online_cpus());
2207 
2208 	return xenbus_register_frontend(&netfront_driver);
2209 }
2210 module_init(netif_init);
2211 
2212 
2213 static void __exit netif_exit(void)
2214 {
2215 	xenbus_unregister_driver(&netfront_driver);
2216 }
2217 module_exit(netif_exit);
2218 
2219 MODULE_DESCRIPTION("Xen virtual network device frontend");
2220 MODULE_LICENSE("GPL");
2221 MODULE_ALIAS("xen:vif");
2222 MODULE_ALIAS("xennet");
2223