Commit | Line | Data |
---|---|---|
0d160211 JF |
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 | ||
383eda32 JP |
32 | #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt |
33 | ||
0d160211 JF |
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> | |
9ecd1a75 | 41 | #include <net/tcp.h> |
0d160211 JF |
42 | #include <linux/udp.h> |
43 | #include <linux/moduleparam.h> | |
44 | #include <linux/mm.h> | |
5a0e3ad6 | 45 | #include <linux/slab.h> |
0d160211 JF |
46 | #include <net/ip.h> |
47 | ||
1ccbf534 | 48 | #include <xen/xen.h> |
0d160211 JF |
49 | #include <xen/xenbus.h> |
50 | #include <xen/events.h> | |
51 | #include <xen/page.h> | |
b9136d20 | 52 | #include <xen/platform_pci.h> |
0d160211 JF |
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 | ||
50ee6061 AB |
59 | /* Module parameters */ |
60 | static unsigned int xennet_max_queues; | |
61 | module_param_named(max_queues, xennet_max_queues, uint, 0644); | |
62 | MODULE_PARM_DESC(max_queues, | |
63 | "Maximum number of queues per virtual interface"); | |
64 | ||
0fc0b732 | 65 | static const struct ethtool_ops xennet_ethtool_ops; |
0d160211 JF |
66 | |
67 | struct netfront_cb { | |
3683243b | 68 | int pull_to; |
0d160211 JF |
69 | }; |
70 | ||
71 | #define NETFRONT_SKB_CB(skb) ((struct netfront_cb *)((skb)->cb)) | |
72 | ||
73 | #define RX_COPY_THRESHOLD 256 | |
74 | ||
75 | #define GRANT_INVALID_REF 0 | |
76 | ||
30c5d7f0 JG |
77 | #define NET_TX_RING_SIZE __CONST_RING_SIZE(xen_netif_tx, XEN_PAGE_SIZE) |
78 | #define NET_RX_RING_SIZE __CONST_RING_SIZE(xen_netif_rx, XEN_PAGE_SIZE) | |
1f3c2eba DV |
79 | |
80 | /* Minimum number of Rx slots (includes slot for GSO metadata). */ | |
81 | #define NET_RX_SLOTS_MIN (XEN_NETIF_NR_SLOTS_MIN + 1) | |
0d160211 | 82 | |
2688fcb7 AB |
83 | /* Queue name is interface name with "-qNNN" appended */ |
84 | #define QUEUE_NAME_SIZE (IFNAMSIZ + 6) | |
85 | ||
86 | /* IRQ name is queue name with "-tx" or "-rx" appended */ | |
87 | #define IRQ_NAME_SIZE (QUEUE_NAME_SIZE + 3) | |
88 | ||
e00f85be | 89 | struct netfront_stats { |
900e1833 DV |
90 | u64 packets; |
91 | u64 bytes; | |
e00f85be | 92 | struct u64_stats_sync syncp; |
93 | }; | |
94 | ||
2688fcb7 AB |
95 | struct netfront_info; |
96 | ||
97 | struct netfront_queue { | |
98 | unsigned int id; /* Queue ID, 0-based */ | |
99 | char name[QUEUE_NAME_SIZE]; /* DEVNAME-qN */ | |
100 | struct netfront_info *info; | |
0d160211 | 101 | |
bea3348e | 102 | struct napi_struct napi; |
0d160211 | 103 | |
d634bf2c WL |
104 | /* Split event channels support, tx_* == rx_* when using |
105 | * single event channel. | |
106 | */ | |
107 | unsigned int tx_evtchn, rx_evtchn; | |
108 | unsigned int tx_irq, rx_irq; | |
109 | /* Only used when split event channels support is enabled */ | |
2688fcb7 AB |
110 | char tx_irq_name[IRQ_NAME_SIZE]; /* DEVNAME-qN-tx */ |
111 | char rx_irq_name[IRQ_NAME_SIZE]; /* DEVNAME-qN-rx */ | |
0d160211 | 112 | |
84284d3c JF |
113 | spinlock_t tx_lock; |
114 | struct xen_netif_tx_front_ring tx; | |
115 | int tx_ring_ref; | |
0d160211 JF |
116 | |
117 | /* | |
118 | * {tx,rx}_skbs store outstanding skbuffs. Free tx_skb entries | |
119 | * are linked from tx_skb_freelist through skb_entry.link. | |
120 | * | |
121 | * NB. Freelist index entries are always going to be less than | |
122 | * PAGE_OFFSET, whereas pointers to skbs will always be equal or | |
123 | * greater than PAGE_OFFSET: we use this property to distinguish | |
124 | * them. | |
125 | */ | |
126 | union skb_entry { | |
127 | struct sk_buff *skb; | |
1ffb40b8 | 128 | unsigned long link; |
0d160211 JF |
129 | } tx_skbs[NET_TX_RING_SIZE]; |
130 | grant_ref_t gref_tx_head; | |
131 | grant_ref_t grant_tx_ref[NET_TX_RING_SIZE]; | |
cefe0078 | 132 | struct page *grant_tx_page[NET_TX_RING_SIZE]; |
0d160211 JF |
133 | unsigned tx_skb_freelist; |
134 | ||
84284d3c JF |
135 | spinlock_t rx_lock ____cacheline_aligned_in_smp; |
136 | struct xen_netif_rx_front_ring rx; | |
137 | int rx_ring_ref; | |
138 | ||
84284d3c JF |
139 | struct timer_list rx_refill_timer; |
140 | ||
0d160211 JF |
141 | struct sk_buff *rx_skbs[NET_RX_RING_SIZE]; |
142 | grant_ref_t gref_rx_head; | |
143 | grant_ref_t grant_rx_ref[NET_RX_RING_SIZE]; | |
2688fcb7 AB |
144 | }; |
145 | ||
146 | struct netfront_info { | |
147 | struct list_head list; | |
148 | struct net_device *netdev; | |
149 | ||
150 | struct xenbus_device *xbdev; | |
151 | ||
152 | /* Multi-queue support */ | |
153 | struct netfront_queue *queues; | |
e0ce4af9 IC |
154 | |
155 | /* Statistics */ | |
900e1833 DV |
156 | struct netfront_stats __percpu *rx_stats; |
157 | struct netfront_stats __percpu *tx_stats; | |
e00f85be | 158 | |
2688fcb7 | 159 | atomic_t rx_gso_checksum_fixup; |
0d160211 JF |
160 | }; |
161 | ||
162 | struct netfront_rx_info { | |
163 | struct xen_netif_rx_response rx; | |
164 | struct xen_netif_extra_info extras[XEN_NETIF_EXTRA_TYPE_MAX - 1]; | |
165 | }; | |
166 | ||
1ffb40b8 IY |
167 | static void skb_entry_set_link(union skb_entry *list, unsigned short id) |
168 | { | |
169 | list->link = id; | |
170 | } | |
171 | ||
172 | static int skb_entry_is_link(const union skb_entry *list) | |
173 | { | |
174 | BUILD_BUG_ON(sizeof(list->skb) != sizeof(list->link)); | |
807540ba | 175 | return (unsigned long)list->skb < PAGE_OFFSET; |
1ffb40b8 IY |
176 | } |
177 | ||
0d160211 JF |
178 | /* |
179 | * Access macros for acquiring freeing slots in tx_skbs[]. | |
180 | */ | |
181 | ||
182 | static void add_id_to_freelist(unsigned *head, union skb_entry *list, | |
183 | unsigned short id) | |
184 | { | |
1ffb40b8 | 185 | skb_entry_set_link(&list[id], *head); |
0d160211 JF |
186 | *head = id; |
187 | } | |
188 | ||
189 | static unsigned short get_id_from_freelist(unsigned *head, | |
190 | union skb_entry *list) | |
191 | { | |
192 | unsigned int id = *head; | |
193 | *head = list[id].link; | |
194 | return id; | |
195 | } | |
196 | ||
197 | static int xennet_rxidx(RING_IDX idx) | |
198 | { | |
199 | return idx & (NET_RX_RING_SIZE - 1); | |
200 | } | |
201 | ||
2688fcb7 | 202 | static struct sk_buff *xennet_get_rx_skb(struct netfront_queue *queue, |
0d160211 JF |
203 | RING_IDX ri) |
204 | { | |
205 | int i = xennet_rxidx(ri); | |
2688fcb7 AB |
206 | struct sk_buff *skb = queue->rx_skbs[i]; |
207 | queue->rx_skbs[i] = NULL; | |
0d160211 JF |
208 | return skb; |
209 | } | |
210 | ||
2688fcb7 | 211 | static grant_ref_t xennet_get_rx_ref(struct netfront_queue *queue, |
0d160211 JF |
212 | RING_IDX ri) |
213 | { | |
214 | int i = xennet_rxidx(ri); | |
2688fcb7 AB |
215 | grant_ref_t ref = queue->grant_rx_ref[i]; |
216 | queue->grant_rx_ref[i] = GRANT_INVALID_REF; | |
0d160211 JF |
217 | return ref; |
218 | } | |
219 | ||
220 | #ifdef CONFIG_SYSFS | |
27b917e5 | 221 | static const struct attribute_group xennet_dev_group; |
0d160211 JF |
222 | #endif |
223 | ||
3ad9b358 | 224 | static bool xennet_can_sg(struct net_device *dev) |
0d160211 | 225 | { |
3ad9b358 | 226 | return dev->features & NETIF_F_SG; |
0d160211 JF |
227 | } |
228 | ||
229 | ||
230 | static void rx_refill_timeout(unsigned long data) | |
231 | { | |
2688fcb7 AB |
232 | struct netfront_queue *queue = (struct netfront_queue *)data; |
233 | napi_schedule(&queue->napi); | |
0d160211 JF |
234 | } |
235 | ||
2688fcb7 | 236 | static int netfront_tx_slot_available(struct netfront_queue *queue) |
0d160211 | 237 | { |
2688fcb7 | 238 | return (queue->tx.req_prod_pvt - queue->tx.rsp_cons) < |
1f3c2eba | 239 | (NET_TX_RING_SIZE - MAX_SKB_FRAGS - 2); |
0d160211 JF |
240 | } |
241 | ||
2688fcb7 | 242 | static void xennet_maybe_wake_tx(struct netfront_queue *queue) |
0d160211 | 243 | { |
2688fcb7 AB |
244 | struct net_device *dev = queue->info->netdev; |
245 | struct netdev_queue *dev_queue = netdev_get_tx_queue(dev, queue->id); | |
0d160211 | 246 | |
2688fcb7 AB |
247 | if (unlikely(netif_tx_queue_stopped(dev_queue)) && |
248 | netfront_tx_slot_available(queue) && | |
0d160211 | 249 | likely(netif_running(dev))) |
2688fcb7 | 250 | netif_tx_wake_queue(netdev_get_tx_queue(dev, queue->id)); |
0d160211 JF |
251 | } |
252 | ||
1f3c2eba DV |
253 | |
254 | static struct sk_buff *xennet_alloc_one_rx_buffer(struct netfront_queue *queue) | |
0d160211 | 255 | { |
0d160211 JF |
256 | struct sk_buff *skb; |
257 | struct page *page; | |
0d160211 | 258 | |
1f3c2eba DV |
259 | skb = __netdev_alloc_skb(queue->info->netdev, |
260 | RX_COPY_THRESHOLD + NET_IP_ALIGN, | |
261 | GFP_ATOMIC | __GFP_NOWARN); | |
262 | if (unlikely(!skb)) | |
263 | return NULL; | |
0d160211 | 264 | |
1f3c2eba DV |
265 | page = alloc_page(GFP_ATOMIC | __GFP_NOWARN); |
266 | if (!page) { | |
267 | kfree_skb(skb); | |
268 | return NULL; | |
0d160211 | 269 | } |
1f3c2eba DV |
270 | skb_add_rx_frag(skb, 0, page, 0, 0, PAGE_SIZE); |
271 | ||
272 | /* Align ip header to a 16 bytes boundary */ | |
273 | skb_reserve(skb, NET_IP_ALIGN); | |
274 | skb->dev = queue->info->netdev; | |
275 | ||
276 | return skb; | |
277 | } | |
0d160211 | 278 | |
1f3c2eba DV |
279 | |
280 | static void xennet_alloc_rx_buffers(struct netfront_queue *queue) | |
281 | { | |
282 | RING_IDX req_prod = queue->rx.req_prod_pvt; | |
283 | int notify; | |
284 | ||
285 | if (unlikely(!netif_carrier_ok(queue->info->netdev))) | |
0d160211 | 286 | return; |
0d160211 | 287 | |
1f3c2eba DV |
288 | for (req_prod = queue->rx.req_prod_pvt; |
289 | req_prod - queue->rx.rsp_cons < NET_RX_RING_SIZE; | |
290 | req_prod++) { | |
291 | struct sk_buff *skb; | |
292 | unsigned short id; | |
293 | grant_ref_t ref; | |
30c5d7f0 | 294 | struct page *page; |
1f3c2eba | 295 | struct xen_netif_rx_request *req; |
0d160211 | 296 | |
1f3c2eba DV |
297 | skb = xennet_alloc_one_rx_buffer(queue); |
298 | if (!skb) | |
0d160211 JF |
299 | break; |
300 | ||
1f3c2eba | 301 | id = xennet_rxidx(req_prod); |
0d160211 | 302 | |
2688fcb7 AB |
303 | BUG_ON(queue->rx_skbs[id]); |
304 | queue->rx_skbs[id] = skb; | |
0d160211 | 305 | |
2688fcb7 | 306 | ref = gnttab_claim_grant_reference(&queue->gref_rx_head); |
0d160211 | 307 | BUG_ON((signed short)ref < 0); |
2688fcb7 | 308 | queue->grant_rx_ref[id] = ref; |
0d160211 | 309 | |
30c5d7f0 | 310 | page = skb_frag_page(&skb_shinfo(skb)->frags[0]); |
0d160211 | 311 | |
1f3c2eba | 312 | req = RING_GET_REQUEST(&queue->rx, req_prod); |
30c5d7f0 JG |
313 | gnttab_page_grant_foreign_access_ref_one(ref, |
314 | queue->info->xbdev->otherend_id, | |
315 | page, | |
316 | 0); | |
0d160211 JF |
317 | req->id = id; |
318 | req->gref = ref; | |
319 | } | |
320 | ||
1f3c2eba DV |
321 | queue->rx.req_prod_pvt = req_prod; |
322 | ||
323 | /* Not enough requests? Try again later. */ | |
324 | if (req_prod - queue->rx.rsp_cons < NET_RX_SLOTS_MIN) { | |
325 | mod_timer(&queue->rx_refill_timer, jiffies + (HZ/10)); | |
326 | return; | |
327 | } | |
328 | ||
5dcddfae | 329 | wmb(); /* barrier so backend seens requests */ |
0d160211 | 330 | |
2688fcb7 | 331 | RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&queue->rx, notify); |
0d160211 | 332 | if (notify) |
2688fcb7 | 333 | notify_remote_via_irq(queue->rx_irq); |
0d160211 JF |
334 | } |
335 | ||
336 | static int xennet_open(struct net_device *dev) | |
337 | { | |
338 | struct netfront_info *np = netdev_priv(dev); | |
2688fcb7 AB |
339 | unsigned int num_queues = dev->real_num_tx_queues; |
340 | unsigned int i = 0; | |
341 | struct netfront_queue *queue = NULL; | |
342 | ||
343 | for (i = 0; i < num_queues; ++i) { | |
344 | queue = &np->queues[i]; | |
345 | napi_enable(&queue->napi); | |
346 | ||
347 | spin_lock_bh(&queue->rx_lock); | |
348 | if (netif_carrier_ok(dev)) { | |
349 | xennet_alloc_rx_buffers(queue); | |
350 | queue->rx.sring->rsp_event = queue->rx.rsp_cons + 1; | |
351 | if (RING_HAS_UNCONSUMED_RESPONSES(&queue->rx)) | |
352 | napi_schedule(&queue->napi); | |
353 | } | |
354 | spin_unlock_bh(&queue->rx_lock); | |
0d160211 | 355 | } |
0d160211 | 356 | |
2688fcb7 | 357 | netif_tx_start_all_queues(dev); |
0d160211 JF |
358 | |
359 | return 0; | |
360 | } | |
361 | ||
2688fcb7 | 362 | static void xennet_tx_buf_gc(struct netfront_queue *queue) |
0d160211 JF |
363 | { |
364 | RING_IDX cons, prod; | |
365 | unsigned short id; | |
0d160211 JF |
366 | struct sk_buff *skb; |
367 | ||
2688fcb7 | 368 | BUG_ON(!netif_carrier_ok(queue->info->netdev)); |
0d160211 JF |
369 | |
370 | do { | |
2688fcb7 | 371 | prod = queue->tx.sring->rsp_prod; |
0d160211 JF |
372 | rmb(); /* Ensure we see responses up to 'rp'. */ |
373 | ||
2688fcb7 | 374 | for (cons = queue->tx.rsp_cons; cons != prod; cons++) { |
0d160211 JF |
375 | struct xen_netif_tx_response *txrsp; |
376 | ||
2688fcb7 | 377 | txrsp = RING_GET_RESPONSE(&queue->tx, cons); |
f942dc25 | 378 | if (txrsp->status == XEN_NETIF_RSP_NULL) |
0d160211 JF |
379 | continue; |
380 | ||
381 | id = txrsp->id; | |
2688fcb7 | 382 | skb = queue->tx_skbs[id].skb; |
0d160211 | 383 | if (unlikely(gnttab_query_foreign_access( |
2688fcb7 | 384 | queue->grant_tx_ref[id]) != 0)) { |
383eda32 JP |
385 | pr_alert("%s: warning -- grant still in use by backend domain\n", |
386 | __func__); | |
0d160211 JF |
387 | BUG(); |
388 | } | |
389 | gnttab_end_foreign_access_ref( | |
2688fcb7 | 390 | queue->grant_tx_ref[id], GNTMAP_readonly); |
0d160211 | 391 | gnttab_release_grant_reference( |
2688fcb7 AB |
392 | &queue->gref_tx_head, queue->grant_tx_ref[id]); |
393 | queue->grant_tx_ref[id] = GRANT_INVALID_REF; | |
394 | queue->grant_tx_page[id] = NULL; | |
395 | add_id_to_freelist(&queue->tx_skb_freelist, queue->tx_skbs, id); | |
0d160211 JF |
396 | dev_kfree_skb_irq(skb); |
397 | } | |
398 | ||
2688fcb7 | 399 | queue->tx.rsp_cons = prod; |
0d160211 JF |
400 | |
401 | /* | |
402 | * Set a new event, then check for race with update of tx_cons. | |
403 | * Note that it is essential to schedule a callback, no matter | |
404 | * how few buffers are pending. Even if there is space in the | |
405 | * transmit ring, higher layers may be blocked because too much | |
406 | * data is outstanding: in such cases notification from Xen is | |
407 | * likely to be the only kick that we'll get. | |
408 | */ | |
2688fcb7 AB |
409 | queue->tx.sring->rsp_event = |
410 | prod + ((queue->tx.sring->req_prod - prod) >> 1) + 1; | |
0d160211 | 411 | mb(); /* update shared area */ |
2688fcb7 | 412 | } while ((cons == prod) && (prod != queue->tx.sring->rsp_prod)); |
0d160211 | 413 | |
2688fcb7 | 414 | xennet_maybe_wake_tx(queue); |
0d160211 JF |
415 | } |
416 | ||
30c5d7f0 JG |
417 | struct xennet_gnttab_make_txreq { |
418 | struct netfront_queue *queue; | |
419 | struct sk_buff *skb; | |
420 | struct page *page; | |
421 | struct xen_netif_tx_request *tx; /* Last request */ | |
422 | unsigned int size; | |
423 | }; | |
424 | ||
425 | static void xennet_tx_setup_grant(unsigned long gfn, unsigned int offset, | |
426 | unsigned int len, void *data) | |
0d160211 | 427 | { |
30c5d7f0 | 428 | struct xennet_gnttab_make_txreq *info = data; |
0d160211 | 429 | unsigned int id; |
a55e8bb8 | 430 | struct xen_netif_tx_request *tx; |
0d160211 | 431 | grant_ref_t ref; |
30c5d7f0 JG |
432 | /* convenient aliases */ |
433 | struct page *page = info->page; | |
434 | struct netfront_queue *queue = info->queue; | |
435 | struct sk_buff *skb = info->skb; | |
0d160211 | 436 | |
a55e8bb8 DV |
437 | id = get_id_from_freelist(&queue->tx_skb_freelist, queue->tx_skbs); |
438 | tx = RING_GET_REQUEST(&queue->tx, queue->tx.req_prod_pvt++); | |
439 | ref = gnttab_claim_grant_reference(&queue->gref_tx_head); | |
440 | BUG_ON((signed short)ref < 0); | |
0d160211 | 441 | |
30c5d7f0 JG |
442 | gnttab_grant_foreign_access_ref(ref, queue->info->xbdev->otherend_id, |
443 | gfn, GNTMAP_readonly); | |
0d160211 | 444 | |
a55e8bb8 DV |
445 | queue->tx_skbs[id].skb = skb; |
446 | queue->grant_tx_page[id] = page; | |
447 | queue->grant_tx_ref[id] = ref; | |
0d160211 | 448 | |
a55e8bb8 DV |
449 | tx->id = id; |
450 | tx->gref = ref; | |
451 | tx->offset = offset; | |
452 | tx->size = len; | |
453 | tx->flags = 0; | |
0d160211 | 454 | |
30c5d7f0 JG |
455 | info->tx = tx; |
456 | info->size += tx->size; | |
457 | } | |
458 | ||
459 | static struct xen_netif_tx_request *xennet_make_first_txreq( | |
460 | struct netfront_queue *queue, struct sk_buff *skb, | |
461 | struct page *page, unsigned int offset, unsigned int len) | |
462 | { | |
463 | struct xennet_gnttab_make_txreq info = { | |
464 | .queue = queue, | |
465 | .skb = skb, | |
466 | .page = page, | |
467 | .size = 0, | |
468 | }; | |
469 | ||
470 | gnttab_for_one_grant(page, offset, len, xennet_tx_setup_grant, &info); | |
471 | ||
472 | return info.tx; | |
473 | } | |
474 | ||
475 | static void xennet_make_one_txreq(unsigned long gfn, unsigned int offset, | |
476 | unsigned int len, void *data) | |
477 | { | |
478 | struct xennet_gnttab_make_txreq *info = data; | |
479 | ||
480 | info->tx->flags |= XEN_NETTXF_more_data; | |
481 | skb_get(info->skb); | |
482 | xennet_tx_setup_grant(gfn, offset, len, data); | |
a55e8bb8 | 483 | } |
0d160211 | 484 | |
a55e8bb8 DV |
485 | static struct xen_netif_tx_request *xennet_make_txreqs( |
486 | struct netfront_queue *queue, struct xen_netif_tx_request *tx, | |
487 | struct sk_buff *skb, struct page *page, | |
488 | unsigned int offset, unsigned int len) | |
489 | { | |
30c5d7f0 JG |
490 | struct xennet_gnttab_make_txreq info = { |
491 | .queue = queue, | |
492 | .skb = skb, | |
493 | .tx = tx, | |
494 | }; | |
495 | ||
a55e8bb8 DV |
496 | /* Skip unused frames from start of page */ |
497 | page += offset >> PAGE_SHIFT; | |
498 | offset &= ~PAGE_MASK; | |
0d160211 | 499 | |
a55e8bb8 | 500 | while (len) { |
30c5d7f0 JG |
501 | info.page = page; |
502 | info.size = 0; | |
503 | ||
504 | gnttab_foreach_grant_in_range(page, offset, len, | |
505 | xennet_make_one_txreq, | |
506 | &info); | |
507 | ||
a55e8bb8 DV |
508 | page++; |
509 | offset = 0; | |
30c5d7f0 | 510 | len -= info.size; |
0d160211 JF |
511 | } |
512 | ||
30c5d7f0 | 513 | return info.tx; |
0d160211 JF |
514 | } |
515 | ||
f36c3747 | 516 | /* |
e84448d5 DV |
517 | * Count how many ring slots are required to send this skb. Each frag |
518 | * might be a compound page. | |
f36c3747 | 519 | */ |
e84448d5 | 520 | static int xennet_count_skb_slots(struct sk_buff *skb) |
f36c3747 IC |
521 | { |
522 | int i, frags = skb_shinfo(skb)->nr_frags; | |
30c5d7f0 | 523 | int slots; |
e84448d5 | 524 | |
30c5d7f0 JG |
525 | slots = gnttab_count_grant(offset_in_page(skb->data), |
526 | skb_headlen(skb)); | |
f36c3747 IC |
527 | |
528 | for (i = 0; i < frags; i++) { | |
529 | skb_frag_t *frag = skb_shinfo(skb)->frags + i; | |
530 | unsigned long size = skb_frag_size(frag); | |
531 | unsigned long offset = frag->page_offset; | |
532 | ||
533 | /* Skip unused frames from start of page */ | |
534 | offset &= ~PAGE_MASK; | |
535 | ||
30c5d7f0 | 536 | slots += gnttab_count_grant(offset, size); |
f36c3747 IC |
537 | } |
538 | ||
30c5d7f0 | 539 | return slots; |
f36c3747 IC |
540 | } |
541 | ||
50ee6061 AB |
542 | static u16 xennet_select_queue(struct net_device *dev, struct sk_buff *skb, |
543 | void *accel_priv, select_queue_fallback_t fallback) | |
2688fcb7 | 544 | { |
50ee6061 AB |
545 | unsigned int num_queues = dev->real_num_tx_queues; |
546 | u32 hash; | |
547 | u16 queue_idx; | |
548 | ||
549 | /* First, check if there is only one queue */ | |
550 | if (num_queues == 1) { | |
551 | queue_idx = 0; | |
552 | } else { | |
553 | hash = skb_get_hash(skb); | |
554 | queue_idx = hash % num_queues; | |
555 | } | |
556 | ||
557 | return queue_idx; | |
2688fcb7 AB |
558 | } |
559 | ||
30c5d7f0 JG |
560 | #define MAX_XEN_SKB_FRAGS (65536 / XEN_PAGE_SIZE + 1) |
561 | ||
0d160211 JF |
562 | static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev) |
563 | { | |
0d160211 | 564 | struct netfront_info *np = netdev_priv(dev); |
900e1833 | 565 | struct netfront_stats *tx_stats = this_cpu_ptr(np->tx_stats); |
a55e8bb8 DV |
566 | struct xen_netif_tx_request *tx, *first_tx; |
567 | unsigned int i; | |
0d160211 | 568 | int notify; |
f36c3747 | 569 | int slots; |
a55e8bb8 DV |
570 | struct page *page; |
571 | unsigned int offset; | |
572 | unsigned int len; | |
cf66f9d4 | 573 | unsigned long flags; |
2688fcb7 AB |
574 | struct netfront_queue *queue = NULL; |
575 | unsigned int num_queues = dev->real_num_tx_queues; | |
576 | u16 queue_index; | |
577 | ||
578 | /* Drop the packet if no queues are set up */ | |
579 | if (num_queues < 1) | |
580 | goto drop; | |
581 | /* Determine which queue to transmit this SKB on */ | |
582 | queue_index = skb_get_queue_mapping(skb); | |
583 | queue = &np->queues[queue_index]; | |
0d160211 | 584 | |
9ecd1a75 WL |
585 | /* If skb->len is too big for wire format, drop skb and alert |
586 | * user about misconfiguration. | |
587 | */ | |
588 | if (unlikely(skb->len > XEN_NETIF_MAX_TX_SIZE)) { | |
589 | net_alert_ratelimited( | |
590 | "xennet: skb->len = %u, too big for wire format\n", | |
591 | skb->len); | |
592 | goto drop; | |
593 | } | |
594 | ||
e84448d5 | 595 | slots = xennet_count_skb_slots(skb); |
30c5d7f0 | 596 | if (unlikely(slots > MAX_XEN_SKB_FRAGS + 1)) { |
97a6d1bb ZK |
597 | net_dbg_ratelimited("xennet: skb rides the rocket: %d slots, %d bytes\n", |
598 | slots, skb->len); | |
599 | if (skb_linearize(skb)) | |
600 | goto drop; | |
0d160211 JF |
601 | } |
602 | ||
a55e8bb8 DV |
603 | page = virt_to_page(skb->data); |
604 | offset = offset_in_page(skb->data); | |
605 | len = skb_headlen(skb); | |
606 | ||
2688fcb7 | 607 | spin_lock_irqsave(&queue->tx_lock, flags); |
0d160211 JF |
608 | |
609 | if (unlikely(!netif_carrier_ok(dev) || | |
f36c3747 | 610 | (slots > 1 && !xennet_can_sg(dev)) || |
8b86a61d | 611 | netif_needs_gso(skb, netif_skb_features(skb)))) { |
2688fcb7 | 612 | spin_unlock_irqrestore(&queue->tx_lock, flags); |
0d160211 JF |
613 | goto drop; |
614 | } | |
615 | ||
a55e8bb8 | 616 | /* First request for the linear area. */ |
30c5d7f0 JG |
617 | first_tx = tx = xennet_make_first_txreq(queue, skb, |
618 | page, offset, len); | |
619 | offset += tx->size; | |
620 | if (offset == PAGE_SIZE) { | |
621 | page++; | |
622 | offset = 0; | |
623 | } | |
a55e8bb8 | 624 | len -= tx->size; |
0d160211 | 625 | |
0d160211 JF |
626 | if (skb->ip_summed == CHECKSUM_PARTIAL) |
627 | /* local packet? */ | |
f942dc25 | 628 | tx->flags |= XEN_NETTXF_csum_blank | XEN_NETTXF_data_validated; |
0d160211 JF |
629 | else if (skb->ip_summed == CHECKSUM_UNNECESSARY) |
630 | /* remote but checksummed. */ | |
f942dc25 | 631 | tx->flags |= XEN_NETTXF_data_validated; |
0d160211 | 632 | |
a55e8bb8 | 633 | /* Optional extra info after the first request. */ |
0d160211 JF |
634 | if (skb_shinfo(skb)->gso_size) { |
635 | struct xen_netif_extra_info *gso; | |
636 | ||
637 | gso = (struct xen_netif_extra_info *) | |
a55e8bb8 | 638 | RING_GET_REQUEST(&queue->tx, queue->tx.req_prod_pvt++); |
0d160211 | 639 | |
e2d617c0 | 640 | tx->flags |= XEN_NETTXF_extra_info; |
0d160211 JF |
641 | |
642 | gso->u.gso.size = skb_shinfo(skb)->gso_size; | |
2c0057de PD |
643 | gso->u.gso.type = (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6) ? |
644 | XEN_NETIF_GSO_TYPE_TCPV6 : | |
645 | XEN_NETIF_GSO_TYPE_TCPV4; | |
0d160211 JF |
646 | gso->u.gso.pad = 0; |
647 | gso->u.gso.features = 0; | |
648 | ||
649 | gso->type = XEN_NETIF_EXTRA_TYPE_GSO; | |
650 | gso->flags = 0; | |
0d160211 JF |
651 | } |
652 | ||
a55e8bb8 DV |
653 | /* Requests for the rest of the linear area. */ |
654 | tx = xennet_make_txreqs(queue, tx, skb, page, offset, len); | |
655 | ||
656 | /* Requests for all the frags. */ | |
657 | for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { | |
658 | skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; | |
659 | tx = xennet_make_txreqs(queue, tx, skb, | |
660 | skb_frag_page(frag), frag->page_offset, | |
661 | skb_frag_size(frag)); | |
662 | } | |
0d160211 | 663 | |
a55e8bb8 DV |
664 | /* First request has the packet length. */ |
665 | first_tx->size = skb->len; | |
0d160211 | 666 | |
2688fcb7 | 667 | RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&queue->tx, notify); |
0d160211 | 668 | if (notify) |
2688fcb7 | 669 | notify_remote_via_irq(queue->tx_irq); |
0d160211 | 670 | |
900e1833 DV |
671 | u64_stats_update_begin(&tx_stats->syncp); |
672 | tx_stats->bytes += skb->len; | |
673 | tx_stats->packets++; | |
674 | u64_stats_update_end(&tx_stats->syncp); | |
10a273a6 JF |
675 | |
676 | /* Note: It is not safe to access skb after xennet_tx_buf_gc()! */ | |
2688fcb7 | 677 | xennet_tx_buf_gc(queue); |
0d160211 | 678 | |
2688fcb7 AB |
679 | if (!netfront_tx_slot_available(queue)) |
680 | netif_tx_stop_queue(netdev_get_tx_queue(dev, queue->id)); | |
0d160211 | 681 | |
2688fcb7 | 682 | spin_unlock_irqrestore(&queue->tx_lock, flags); |
0d160211 | 683 | |
6ed10654 | 684 | return NETDEV_TX_OK; |
0d160211 JF |
685 | |
686 | drop: | |
09f75cd7 | 687 | dev->stats.tx_dropped++; |
979de8a0 | 688 | dev_kfree_skb_any(skb); |
6ed10654 | 689 | return NETDEV_TX_OK; |
0d160211 JF |
690 | } |
691 | ||
692 | static int xennet_close(struct net_device *dev) | |
693 | { | |
694 | struct netfront_info *np = netdev_priv(dev); | |
2688fcb7 AB |
695 | unsigned int num_queues = dev->real_num_tx_queues; |
696 | unsigned int i; | |
697 | struct netfront_queue *queue; | |
698 | netif_tx_stop_all_queues(np->netdev); | |
699 | for (i = 0; i < num_queues; ++i) { | |
700 | queue = &np->queues[i]; | |
701 | napi_disable(&queue->napi); | |
702 | } | |
0d160211 JF |
703 | return 0; |
704 | } | |
705 | ||
2688fcb7 | 706 | static void xennet_move_rx_slot(struct netfront_queue *queue, struct sk_buff *skb, |
0d160211 JF |
707 | grant_ref_t ref) |
708 | { | |
2688fcb7 AB |
709 | int new = xennet_rxidx(queue->rx.req_prod_pvt); |
710 | ||
711 | BUG_ON(queue->rx_skbs[new]); | |
712 | queue->rx_skbs[new] = skb; | |
713 | queue->grant_rx_ref[new] = ref; | |
714 | RING_GET_REQUEST(&queue->rx, queue->rx.req_prod_pvt)->id = new; | |
715 | RING_GET_REQUEST(&queue->rx, queue->rx.req_prod_pvt)->gref = ref; | |
716 | queue->rx.req_prod_pvt++; | |
0d160211 JF |
717 | } |
718 | ||
2688fcb7 | 719 | static int xennet_get_extras(struct netfront_queue *queue, |
0d160211 JF |
720 | struct xen_netif_extra_info *extras, |
721 | RING_IDX rp) | |
722 | ||
723 | { | |
724 | struct xen_netif_extra_info *extra; | |
2688fcb7 AB |
725 | struct device *dev = &queue->info->netdev->dev; |
726 | RING_IDX cons = queue->rx.rsp_cons; | |
0d160211 JF |
727 | int err = 0; |
728 | ||
729 | do { | |
730 | struct sk_buff *skb; | |
731 | grant_ref_t ref; | |
732 | ||
733 | if (unlikely(cons + 1 == rp)) { | |
734 | if (net_ratelimit()) | |
735 | dev_warn(dev, "Missing extra info\n"); | |
736 | err = -EBADR; | |
737 | break; | |
738 | } | |
739 | ||
740 | extra = (struct xen_netif_extra_info *) | |
2688fcb7 | 741 | RING_GET_RESPONSE(&queue->rx, ++cons); |
0d160211 JF |
742 | |
743 | if (unlikely(!extra->type || | |
744 | extra->type >= XEN_NETIF_EXTRA_TYPE_MAX)) { | |
745 | if (net_ratelimit()) | |
746 | dev_warn(dev, "Invalid extra type: %d\n", | |
747 | extra->type); | |
748 | err = -EINVAL; | |
749 | } else { | |
750 | memcpy(&extras[extra->type - 1], extra, | |
751 | sizeof(*extra)); | |
752 | } | |
753 | ||
2688fcb7 AB |
754 | skb = xennet_get_rx_skb(queue, cons); |
755 | ref = xennet_get_rx_ref(queue, cons); | |
756 | xennet_move_rx_slot(queue, skb, ref); | |
0d160211 JF |
757 | } while (extra->flags & XEN_NETIF_EXTRA_FLAG_MORE); |
758 | ||
2688fcb7 | 759 | queue->rx.rsp_cons = cons; |
0d160211 JF |
760 | return err; |
761 | } | |
762 | ||
2688fcb7 | 763 | static int xennet_get_responses(struct netfront_queue *queue, |
0d160211 JF |
764 | struct netfront_rx_info *rinfo, RING_IDX rp, |
765 | struct sk_buff_head *list) | |
766 | { | |
767 | struct xen_netif_rx_response *rx = &rinfo->rx; | |
768 | struct xen_netif_extra_info *extras = rinfo->extras; | |
2688fcb7 AB |
769 | struct device *dev = &queue->info->netdev->dev; |
770 | RING_IDX cons = queue->rx.rsp_cons; | |
771 | struct sk_buff *skb = xennet_get_rx_skb(queue, cons); | |
772 | grant_ref_t ref = xennet_get_rx_ref(queue, cons); | |
0d160211 | 773 | int max = MAX_SKB_FRAGS + (rx->status <= RX_COPY_THRESHOLD); |
7158ff6d | 774 | int slots = 1; |
0d160211 JF |
775 | int err = 0; |
776 | unsigned long ret; | |
777 | ||
f942dc25 | 778 | if (rx->flags & XEN_NETRXF_extra_info) { |
2688fcb7 AB |
779 | err = xennet_get_extras(queue, extras, rp); |
780 | cons = queue->rx.rsp_cons; | |
0d160211 JF |
781 | } |
782 | ||
783 | for (;;) { | |
784 | if (unlikely(rx->status < 0 || | |
30c5d7f0 | 785 | rx->offset + rx->status > XEN_PAGE_SIZE)) { |
0d160211 | 786 | if (net_ratelimit()) |
6c10127d | 787 | dev_warn(dev, "rx->offset: %u, size: %d\n", |
0d160211 | 788 | rx->offset, rx->status); |
2688fcb7 | 789 | xennet_move_rx_slot(queue, skb, ref); |
0d160211 JF |
790 | err = -EINVAL; |
791 | goto next; | |
792 | } | |
793 | ||
794 | /* | |
795 | * This definitely indicates a bug, either in this driver or in | |
796 | * the backend driver. In future this should flag the bad | |
697089dc | 797 | * situation to the system controller to reboot the backend. |
0d160211 JF |
798 | */ |
799 | if (ref == GRANT_INVALID_REF) { | |
800 | if (net_ratelimit()) | |
801 | dev_warn(dev, "Bad rx response id %d.\n", | |
802 | rx->id); | |
803 | err = -EINVAL; | |
804 | goto next; | |
805 | } | |
806 | ||
807 | ret = gnttab_end_foreign_access_ref(ref, 0); | |
808 | BUG_ON(!ret); | |
809 | ||
2688fcb7 | 810 | gnttab_release_grant_reference(&queue->gref_rx_head, ref); |
0d160211 JF |
811 | |
812 | __skb_queue_tail(list, skb); | |
813 | ||
814 | next: | |
f942dc25 | 815 | if (!(rx->flags & XEN_NETRXF_more_data)) |
0d160211 JF |
816 | break; |
817 | ||
7158ff6d | 818 | if (cons + slots == rp) { |
0d160211 | 819 | if (net_ratelimit()) |
7158ff6d | 820 | dev_warn(dev, "Need more slots\n"); |
0d160211 JF |
821 | err = -ENOENT; |
822 | break; | |
823 | } | |
824 | ||
2688fcb7 AB |
825 | rx = RING_GET_RESPONSE(&queue->rx, cons + slots); |
826 | skb = xennet_get_rx_skb(queue, cons + slots); | |
827 | ref = xennet_get_rx_ref(queue, cons + slots); | |
7158ff6d | 828 | slots++; |
0d160211 JF |
829 | } |
830 | ||
7158ff6d | 831 | if (unlikely(slots > max)) { |
0d160211 | 832 | if (net_ratelimit()) |
697089dc | 833 | dev_warn(dev, "Too many slots\n"); |
0d160211 JF |
834 | err = -E2BIG; |
835 | } | |
836 | ||
837 | if (unlikely(err)) | |
2688fcb7 | 838 | queue->rx.rsp_cons = cons + slots; |
0d160211 JF |
839 | |
840 | return err; | |
841 | } | |
842 | ||
843 | static int xennet_set_skb_gso(struct sk_buff *skb, | |
844 | struct xen_netif_extra_info *gso) | |
845 | { | |
846 | if (!gso->u.gso.size) { | |
847 | if (net_ratelimit()) | |
383eda32 | 848 | pr_warn("GSO size must not be zero\n"); |
0d160211 JF |
849 | return -EINVAL; |
850 | } | |
851 | ||
2c0057de PD |
852 | if (gso->u.gso.type != XEN_NETIF_GSO_TYPE_TCPV4 && |
853 | gso->u.gso.type != XEN_NETIF_GSO_TYPE_TCPV6) { | |
0d160211 | 854 | if (net_ratelimit()) |
383eda32 | 855 | pr_warn("Bad GSO type %d\n", gso->u.gso.type); |
0d160211 JF |
856 | return -EINVAL; |
857 | } | |
858 | ||
859 | skb_shinfo(skb)->gso_size = gso->u.gso.size; | |
2c0057de PD |
860 | skb_shinfo(skb)->gso_type = |
861 | (gso->u.gso.type == XEN_NETIF_GSO_TYPE_TCPV4) ? | |
862 | SKB_GSO_TCPV4 : | |
863 | SKB_GSO_TCPV6; | |
0d160211 JF |
864 | |
865 | /* Header must be checked, and gso_segs computed. */ | |
866 | skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY; | |
867 | skb_shinfo(skb)->gso_segs = 0; | |
868 | ||
869 | return 0; | |
870 | } | |
871 | ||
2688fcb7 | 872 | static RING_IDX xennet_fill_frags(struct netfront_queue *queue, |
0d160211 JF |
873 | struct sk_buff *skb, |
874 | struct sk_buff_head *list) | |
875 | { | |
876 | struct skb_shared_info *shinfo = skb_shinfo(skb); | |
2688fcb7 | 877 | RING_IDX cons = queue->rx.rsp_cons; |
0d160211 JF |
878 | struct sk_buff *nskb; |
879 | ||
880 | while ((nskb = __skb_dequeue(list))) { | |
881 | struct xen_netif_rx_response *rx = | |
2688fcb7 | 882 | RING_GET_RESPONSE(&queue->rx, ++cons); |
01c68026 | 883 | skb_frag_t *nfrag = &skb_shinfo(nskb)->frags[0]; |
0d160211 | 884 | |
093b9c71 JB |
885 | if (shinfo->nr_frags == MAX_SKB_FRAGS) { |
886 | unsigned int pull_to = NETFRONT_SKB_CB(skb)->pull_to; | |
0d160211 | 887 | |
093b9c71 JB |
888 | BUG_ON(pull_to <= skb_headlen(skb)); |
889 | __pskb_pull_tail(skb, pull_to - skb_headlen(skb)); | |
890 | } | |
891 | BUG_ON(shinfo->nr_frags >= MAX_SKB_FRAGS); | |
892 | ||
893 | skb_add_rx_frag(skb, shinfo->nr_frags, skb_frag_page(nfrag), | |
894 | rx->offset, rx->status, PAGE_SIZE); | |
0d160211 JF |
895 | |
896 | skb_shinfo(nskb)->nr_frags = 0; | |
897 | kfree_skb(nskb); | |
0d160211 JF |
898 | } |
899 | ||
0d160211 JF |
900 | return cons; |
901 | } | |
902 | ||
e0ce4af9 | 903 | static int checksum_setup(struct net_device *dev, struct sk_buff *skb) |
0d160211 | 904 | { |
b5cf66cd | 905 | bool recalculate_partial_csum = false; |
e0ce4af9 IC |
906 | |
907 | /* | |
908 | * A GSO SKB must be CHECKSUM_PARTIAL. However some buggy | |
909 | * peers can fail to set NETRXF_csum_blank when sending a GSO | |
910 | * frame. In this case force the SKB to CHECKSUM_PARTIAL and | |
911 | * recalculate the partial checksum. | |
912 | */ | |
913 | if (skb->ip_summed != CHECKSUM_PARTIAL && skb_is_gso(skb)) { | |
914 | struct netfront_info *np = netdev_priv(dev); | |
2688fcb7 | 915 | atomic_inc(&np->rx_gso_checksum_fixup); |
e0ce4af9 | 916 | skb->ip_summed = CHECKSUM_PARTIAL; |
b5cf66cd | 917 | recalculate_partial_csum = true; |
e0ce4af9 IC |
918 | } |
919 | ||
920 | /* A non-CHECKSUM_PARTIAL SKB does not require setup. */ | |
921 | if (skb->ip_summed != CHECKSUM_PARTIAL) | |
922 | return 0; | |
0d160211 | 923 | |
b5cf66cd | 924 | return skb_checksum_setup(skb, recalculate_partial_csum); |
0d160211 JF |
925 | } |
926 | ||
2688fcb7 | 927 | static int handle_incoming_queue(struct netfront_queue *queue, |
09f75cd7 | 928 | struct sk_buff_head *rxq) |
0d160211 | 929 | { |
900e1833 | 930 | struct netfront_stats *rx_stats = this_cpu_ptr(queue->info->rx_stats); |
0d160211 JF |
931 | int packets_dropped = 0; |
932 | struct sk_buff *skb; | |
933 | ||
934 | while ((skb = __skb_dequeue(rxq)) != NULL) { | |
3683243b | 935 | int pull_to = NETFRONT_SKB_CB(skb)->pull_to; |
0d160211 | 936 | |
093b9c71 JB |
937 | if (pull_to > skb_headlen(skb)) |
938 | __pskb_pull_tail(skb, pull_to - skb_headlen(skb)); | |
0d160211 JF |
939 | |
940 | /* Ethernet work: Delayed to here as it peeks the header. */ | |
2688fcb7 | 941 | skb->protocol = eth_type_trans(skb, queue->info->netdev); |
d554f73d | 942 | skb_reset_network_header(skb); |
0d160211 | 943 | |
2688fcb7 | 944 | if (checksum_setup(queue->info->netdev, skb)) { |
e0ce4af9 IC |
945 | kfree_skb(skb); |
946 | packets_dropped++; | |
2688fcb7 | 947 | queue->info->netdev->stats.rx_errors++; |
e0ce4af9 | 948 | continue; |
0d160211 JF |
949 | } |
950 | ||
900e1833 DV |
951 | u64_stats_update_begin(&rx_stats->syncp); |
952 | rx_stats->packets++; | |
953 | rx_stats->bytes += skb->len; | |
954 | u64_stats_update_end(&rx_stats->syncp); | |
0d160211 JF |
955 | |
956 | /* Pass it up. */ | |
2688fcb7 | 957 | napi_gro_receive(&queue->napi, skb); |
0d160211 JF |
958 | } |
959 | ||
960 | return packets_dropped; | |
961 | } | |
962 | ||
bea3348e | 963 | static int xennet_poll(struct napi_struct *napi, int budget) |
0d160211 | 964 | { |
2688fcb7 AB |
965 | struct netfront_queue *queue = container_of(napi, struct netfront_queue, napi); |
966 | struct net_device *dev = queue->info->netdev; | |
0d160211 JF |
967 | struct sk_buff *skb; |
968 | struct netfront_rx_info rinfo; | |
969 | struct xen_netif_rx_response *rx = &rinfo.rx; | |
970 | struct xen_netif_extra_info *extras = rinfo.extras; | |
971 | RING_IDX i, rp; | |
bea3348e | 972 | int work_done; |
0d160211 JF |
973 | struct sk_buff_head rxq; |
974 | struct sk_buff_head errq; | |
975 | struct sk_buff_head tmpq; | |
0d160211 JF |
976 | int err; |
977 | ||
2688fcb7 | 978 | spin_lock(&queue->rx_lock); |
0d160211 | 979 | |
0d160211 JF |
980 | skb_queue_head_init(&rxq); |
981 | skb_queue_head_init(&errq); | |
982 | skb_queue_head_init(&tmpq); | |
983 | ||
2688fcb7 | 984 | rp = queue->rx.sring->rsp_prod; |
0d160211 JF |
985 | rmb(); /* Ensure we see queued responses up to 'rp'. */ |
986 | ||
2688fcb7 | 987 | i = queue->rx.rsp_cons; |
0d160211 JF |
988 | work_done = 0; |
989 | while ((i != rp) && (work_done < budget)) { | |
2688fcb7 | 990 | memcpy(rx, RING_GET_RESPONSE(&queue->rx, i), sizeof(*rx)); |
0d160211 JF |
991 | memset(extras, 0, sizeof(rinfo.extras)); |
992 | ||
2688fcb7 | 993 | err = xennet_get_responses(queue, &rinfo, rp, &tmpq); |
0d160211 JF |
994 | |
995 | if (unlikely(err)) { | |
996 | err: | |
997 | while ((skb = __skb_dequeue(&tmpq))) | |
998 | __skb_queue_tail(&errq, skb); | |
09f75cd7 | 999 | dev->stats.rx_errors++; |
2688fcb7 | 1000 | i = queue->rx.rsp_cons; |
0d160211 JF |
1001 | continue; |
1002 | } | |
1003 | ||
1004 | skb = __skb_dequeue(&tmpq); | |
1005 | ||
1006 | if (extras[XEN_NETIF_EXTRA_TYPE_GSO - 1].type) { | |
1007 | struct xen_netif_extra_info *gso; | |
1008 | gso = &extras[XEN_NETIF_EXTRA_TYPE_GSO - 1]; | |
1009 | ||
1010 | if (unlikely(xennet_set_skb_gso(skb, gso))) { | |
1011 | __skb_queue_head(&tmpq, skb); | |
2688fcb7 | 1012 | queue->rx.rsp_cons += skb_queue_len(&tmpq); |
0d160211 JF |
1013 | goto err; |
1014 | } | |
1015 | } | |
1016 | ||
3683243b IC |
1017 | NETFRONT_SKB_CB(skb)->pull_to = rx->status; |
1018 | if (NETFRONT_SKB_CB(skb)->pull_to > RX_COPY_THRESHOLD) | |
1019 | NETFRONT_SKB_CB(skb)->pull_to = RX_COPY_THRESHOLD; | |
0d160211 | 1020 | |
3683243b IC |
1021 | skb_shinfo(skb)->frags[0].page_offset = rx->offset; |
1022 | skb_frag_size_set(&skb_shinfo(skb)->frags[0], rx->status); | |
1023 | skb->data_len = rx->status; | |
093b9c71 | 1024 | skb->len += rx->status; |
0d160211 | 1025 | |
2688fcb7 | 1026 | i = xennet_fill_frags(queue, skb, &tmpq); |
0d160211 | 1027 | |
f942dc25 | 1028 | if (rx->flags & XEN_NETRXF_csum_blank) |
0d160211 | 1029 | skb->ip_summed = CHECKSUM_PARTIAL; |
f942dc25 | 1030 | else if (rx->flags & XEN_NETRXF_data_validated) |
0d160211 JF |
1031 | skb->ip_summed = CHECKSUM_UNNECESSARY; |
1032 | ||
1033 | __skb_queue_tail(&rxq, skb); | |
1034 | ||
2688fcb7 | 1035 | queue->rx.rsp_cons = ++i; |
0d160211 JF |
1036 | work_done++; |
1037 | } | |
1038 | ||
56cfe5d0 | 1039 | __skb_queue_purge(&errq); |
0d160211 | 1040 | |
2688fcb7 | 1041 | work_done -= handle_incoming_queue(queue, &rxq); |
0d160211 | 1042 | |
2688fcb7 | 1043 | xennet_alloc_rx_buffers(queue); |
0d160211 | 1044 | |
0d160211 | 1045 | if (work_done < budget) { |
bea3348e SH |
1046 | int more_to_do = 0; |
1047 | ||
6a6dc08f | 1048 | napi_complete(napi); |
0d160211 | 1049 | |
2688fcb7 | 1050 | RING_FINAL_CHECK_FOR_RESPONSES(&queue->rx, more_to_do); |
6a6dc08f DV |
1051 | if (more_to_do) |
1052 | napi_schedule(napi); | |
0d160211 JF |
1053 | } |
1054 | ||
2688fcb7 | 1055 | spin_unlock(&queue->rx_lock); |
0d160211 | 1056 | |
bea3348e | 1057 | return work_done; |
0d160211 JF |
1058 | } |
1059 | ||
1060 | static int xennet_change_mtu(struct net_device *dev, int mtu) | |
1061 | { | |
0c36820e | 1062 | int max = xennet_can_sg(dev) ? XEN_NETIF_MAX_TX_SIZE : ETH_DATA_LEN; |
0d160211 JF |
1063 | |
1064 | if (mtu > max) | |
1065 | return -EINVAL; | |
1066 | dev->mtu = mtu; | |
1067 | return 0; | |
1068 | } | |
1069 | ||
e00f85be | 1070 | static struct rtnl_link_stats64 *xennet_get_stats64(struct net_device *dev, |
1071 | struct rtnl_link_stats64 *tot) | |
1072 | { | |
1073 | struct netfront_info *np = netdev_priv(dev); | |
1074 | int cpu; | |
1075 | ||
1076 | for_each_possible_cpu(cpu) { | |
900e1833 DV |
1077 | struct netfront_stats *rx_stats = per_cpu_ptr(np->rx_stats, cpu); |
1078 | struct netfront_stats *tx_stats = per_cpu_ptr(np->tx_stats, cpu); | |
e00f85be | 1079 | u64 rx_packets, rx_bytes, tx_packets, tx_bytes; |
1080 | unsigned int start; | |
1081 | ||
1082 | do { | |
900e1833 DV |
1083 | start = u64_stats_fetch_begin_irq(&tx_stats->syncp); |
1084 | tx_packets = tx_stats->packets; | |
1085 | tx_bytes = tx_stats->bytes; | |
1086 | } while (u64_stats_fetch_retry_irq(&tx_stats->syncp, start)); | |
e00f85be | 1087 | |
900e1833 DV |
1088 | do { |
1089 | start = u64_stats_fetch_begin_irq(&rx_stats->syncp); | |
1090 | rx_packets = rx_stats->packets; | |
1091 | rx_bytes = rx_stats->bytes; | |
1092 | } while (u64_stats_fetch_retry_irq(&rx_stats->syncp, start)); | |
e00f85be | 1093 | |
1094 | tot->rx_packets += rx_packets; | |
1095 | tot->tx_packets += tx_packets; | |
1096 | tot->rx_bytes += rx_bytes; | |
1097 | tot->tx_bytes += tx_bytes; | |
1098 | } | |
1099 | ||
1100 | tot->rx_errors = dev->stats.rx_errors; | |
1101 | tot->tx_dropped = dev->stats.tx_dropped; | |
1102 | ||
1103 | return tot; | |
1104 | } | |
1105 | ||
2688fcb7 | 1106 | static void xennet_release_tx_bufs(struct netfront_queue *queue) |
0d160211 JF |
1107 | { |
1108 | struct sk_buff *skb; | |
1109 | int i; | |
1110 | ||
1111 | for (i = 0; i < NET_TX_RING_SIZE; i++) { | |
1112 | /* Skip over entries which are actually freelist references */ | |
2688fcb7 | 1113 | if (skb_entry_is_link(&queue->tx_skbs[i])) |
0d160211 JF |
1114 | continue; |
1115 | ||
2688fcb7 AB |
1116 | skb = queue->tx_skbs[i].skb; |
1117 | get_page(queue->grant_tx_page[i]); | |
1118 | gnttab_end_foreign_access(queue->grant_tx_ref[i], | |
cefe0078 | 1119 | GNTMAP_readonly, |
2688fcb7 AB |
1120 | (unsigned long)page_address(queue->grant_tx_page[i])); |
1121 | queue->grant_tx_page[i] = NULL; | |
1122 | queue->grant_tx_ref[i] = GRANT_INVALID_REF; | |
1123 | add_id_to_freelist(&queue->tx_skb_freelist, queue->tx_skbs, i); | |
0d160211 JF |
1124 | dev_kfree_skb_irq(skb); |
1125 | } | |
1126 | } | |
1127 | ||
2688fcb7 | 1128 | static void xennet_release_rx_bufs(struct netfront_queue *queue) |
0d160211 | 1129 | { |
0d160211 JF |
1130 | int id, ref; |
1131 | ||
2688fcb7 | 1132 | spin_lock_bh(&queue->rx_lock); |
0d160211 JF |
1133 | |
1134 | for (id = 0; id < NET_RX_RING_SIZE; id++) { | |
cefe0078 AL |
1135 | struct sk_buff *skb; |
1136 | struct page *page; | |
0d160211 | 1137 | |
2688fcb7 | 1138 | skb = queue->rx_skbs[id]; |
cefe0078 | 1139 | if (!skb) |
0d160211 | 1140 | continue; |
0d160211 | 1141 | |
2688fcb7 | 1142 | ref = queue->grant_rx_ref[id]; |
cefe0078 AL |
1143 | if (ref == GRANT_INVALID_REF) |
1144 | continue; | |
0d160211 | 1145 | |
cefe0078 | 1146 | page = skb_frag_page(&skb_shinfo(skb)->frags[0]); |
0d160211 | 1147 | |
cefe0078 AL |
1148 | /* gnttab_end_foreign_access() needs a page ref until |
1149 | * foreign access is ended (which may be deferred). | |
1150 | */ | |
1151 | get_page(page); | |
1152 | gnttab_end_foreign_access(ref, 0, | |
1153 | (unsigned long)page_address(page)); | |
2688fcb7 | 1154 | queue->grant_rx_ref[id] = GRANT_INVALID_REF; |
0d160211 | 1155 | |
cefe0078 | 1156 | kfree_skb(skb); |
0d160211 JF |
1157 | } |
1158 | ||
2688fcb7 | 1159 | spin_unlock_bh(&queue->rx_lock); |
0d160211 JF |
1160 | } |
1161 | ||
c8f44aff MM |
1162 | static netdev_features_t xennet_fix_features(struct net_device *dev, |
1163 | netdev_features_t features) | |
8f7b01a1 ED |
1164 | { |
1165 | struct netfront_info *np = netdev_priv(dev); | |
1166 | int val; | |
1167 | ||
1168 | if (features & NETIF_F_SG) { | |
1169 | if (xenbus_scanf(XBT_NIL, np->xbdev->otherend, "feature-sg", | |
1170 | "%d", &val) < 0) | |
1171 | val = 0; | |
1172 | ||
1173 | if (!val) | |
1174 | features &= ~NETIF_F_SG; | |
1175 | } | |
1176 | ||
2c0057de PD |
1177 | if (features & NETIF_F_IPV6_CSUM) { |
1178 | if (xenbus_scanf(XBT_NIL, np->xbdev->otherend, | |
1179 | "feature-ipv6-csum-offload", "%d", &val) < 0) | |
1180 | val = 0; | |
1181 | ||
1182 | if (!val) | |
1183 | features &= ~NETIF_F_IPV6_CSUM; | |
1184 | } | |
1185 | ||
8f7b01a1 ED |
1186 | if (features & NETIF_F_TSO) { |
1187 | if (xenbus_scanf(XBT_NIL, np->xbdev->otherend, | |
1188 | "feature-gso-tcpv4", "%d", &val) < 0) | |
1189 | val = 0; | |
1190 | ||
1191 | if (!val) | |
1192 | features &= ~NETIF_F_TSO; | |
1193 | } | |
1194 | ||
2c0057de PD |
1195 | if (features & NETIF_F_TSO6) { |
1196 | if (xenbus_scanf(XBT_NIL, np->xbdev->otherend, | |
1197 | "feature-gso-tcpv6", "%d", &val) < 0) | |
1198 | val = 0; | |
1199 | ||
1200 | if (!val) | |
1201 | features &= ~NETIF_F_TSO6; | |
1202 | } | |
1203 | ||
8f7b01a1 ED |
1204 | return features; |
1205 | } | |
1206 | ||
c8f44aff MM |
1207 | static int xennet_set_features(struct net_device *dev, |
1208 | netdev_features_t features) | |
8f7b01a1 ED |
1209 | { |
1210 | if (!(features & NETIF_F_SG) && dev->mtu > ETH_DATA_LEN) { | |
1211 | netdev_info(dev, "Reducing MTU because no SG offload"); | |
1212 | dev->mtu = ETH_DATA_LEN; | |
1213 | } | |
1214 | ||
1215 | return 0; | |
1216 | } | |
1217 | ||
d634bf2c | 1218 | static irqreturn_t xennet_tx_interrupt(int irq, void *dev_id) |
cf66f9d4 | 1219 | { |
2688fcb7 | 1220 | struct netfront_queue *queue = dev_id; |
cf66f9d4 KRW |
1221 | unsigned long flags; |
1222 | ||
2688fcb7 AB |
1223 | spin_lock_irqsave(&queue->tx_lock, flags); |
1224 | xennet_tx_buf_gc(queue); | |
1225 | spin_unlock_irqrestore(&queue->tx_lock, flags); | |
cf66f9d4 | 1226 | |
d634bf2c WL |
1227 | return IRQ_HANDLED; |
1228 | } | |
1229 | ||
1230 | static irqreturn_t xennet_rx_interrupt(int irq, void *dev_id) | |
1231 | { | |
2688fcb7 AB |
1232 | struct netfront_queue *queue = dev_id; |
1233 | struct net_device *dev = queue->info->netdev; | |
d634bf2c WL |
1234 | |
1235 | if (likely(netif_carrier_ok(dev) && | |
2688fcb7 | 1236 | RING_HAS_UNCONSUMED_RESPONSES(&queue->rx))) |
76541869 | 1237 | napi_schedule(&queue->napi); |
cf66f9d4 | 1238 | |
d634bf2c WL |
1239 | return IRQ_HANDLED; |
1240 | } | |
cf66f9d4 | 1241 | |
d634bf2c WL |
1242 | static irqreturn_t xennet_interrupt(int irq, void *dev_id) |
1243 | { | |
1244 | xennet_tx_interrupt(irq, dev_id); | |
1245 | xennet_rx_interrupt(irq, dev_id); | |
cf66f9d4 KRW |
1246 | return IRQ_HANDLED; |
1247 | } | |
1248 | ||
1249 | #ifdef CONFIG_NET_POLL_CONTROLLER | |
1250 | static void xennet_poll_controller(struct net_device *dev) | |
1251 | { | |
2688fcb7 AB |
1252 | /* Poll each queue */ |
1253 | struct netfront_info *info = netdev_priv(dev); | |
1254 | unsigned int num_queues = dev->real_num_tx_queues; | |
1255 | unsigned int i; | |
1256 | for (i = 0; i < num_queues; ++i) | |
1257 | xennet_interrupt(0, &info->queues[i]); | |
cf66f9d4 KRW |
1258 | } |
1259 | #endif | |
1260 | ||
0a0b9d2e SH |
1261 | static const struct net_device_ops xennet_netdev_ops = { |
1262 | .ndo_open = xennet_open, | |
0a0b9d2e SH |
1263 | .ndo_stop = xennet_close, |
1264 | .ndo_start_xmit = xennet_start_xmit, | |
1265 | .ndo_change_mtu = xennet_change_mtu, | |
e00f85be | 1266 | .ndo_get_stats64 = xennet_get_stats64, |
0a0b9d2e SH |
1267 | .ndo_set_mac_address = eth_mac_addr, |
1268 | .ndo_validate_addr = eth_validate_addr, | |
fb507934 MM |
1269 | .ndo_fix_features = xennet_fix_features, |
1270 | .ndo_set_features = xennet_set_features, | |
2688fcb7 | 1271 | .ndo_select_queue = xennet_select_queue, |
cf66f9d4 KRW |
1272 | #ifdef CONFIG_NET_POLL_CONTROLLER |
1273 | .ndo_poll_controller = xennet_poll_controller, | |
1274 | #endif | |
0a0b9d2e SH |
1275 | }; |
1276 | ||
900e1833 DV |
1277 | static void xennet_free_netdev(struct net_device *netdev) |
1278 | { | |
1279 | struct netfront_info *np = netdev_priv(netdev); | |
1280 | ||
1281 | free_percpu(np->rx_stats); | |
1282 | free_percpu(np->tx_stats); | |
1283 | free_netdev(netdev); | |
1284 | } | |
1285 | ||
8e0e46bb | 1286 | static struct net_device *xennet_create_dev(struct xenbus_device *dev) |
0d160211 | 1287 | { |
2688fcb7 | 1288 | int err; |
0d160211 JF |
1289 | struct net_device *netdev; |
1290 | struct netfront_info *np; | |
1291 | ||
50ee6061 | 1292 | netdev = alloc_etherdev_mq(sizeof(struct netfront_info), xennet_max_queues); |
41de8d4c | 1293 | if (!netdev) |
0d160211 | 1294 | return ERR_PTR(-ENOMEM); |
0d160211 JF |
1295 | |
1296 | np = netdev_priv(netdev); | |
1297 | np->xbdev = dev; | |
1298 | ||
2688fcb7 | 1299 | np->queues = NULL; |
0d160211 | 1300 | |
e00f85be | 1301 | err = -ENOMEM; |
900e1833 DV |
1302 | np->rx_stats = netdev_alloc_pcpu_stats(struct netfront_stats); |
1303 | if (np->rx_stats == NULL) | |
1304 | goto exit; | |
1305 | np->tx_stats = netdev_alloc_pcpu_stats(struct netfront_stats); | |
1306 | if (np->tx_stats == NULL) | |
e00f85be | 1307 | goto exit; |
1308 | ||
0a0b9d2e SH |
1309 | netdev->netdev_ops = &xennet_netdev_ops; |
1310 | ||
fb507934 MM |
1311 | netdev->features = NETIF_F_IP_CSUM | NETIF_F_RXCSUM | |
1312 | NETIF_F_GSO_ROBUST; | |
2c0057de PD |
1313 | netdev->hw_features = NETIF_F_SG | |
1314 | NETIF_F_IPV6_CSUM | | |
1315 | NETIF_F_TSO | NETIF_F_TSO6; | |
0d160211 | 1316 | |
fc3e5941 IC |
1317 | /* |
1318 | * Assume that all hw features are available for now. This set | |
1319 | * will be adjusted by the call to netdev_update_features() in | |
1320 | * xennet_connect() which is the earliest point where we can | |
1321 | * negotiate with the backend regarding supported features. | |
1322 | */ | |
1323 | netdev->features |= netdev->hw_features; | |
1324 | ||
7ad24ea4 | 1325 | netdev->ethtool_ops = &xennet_ethtool_ops; |
0d160211 JF |
1326 | SET_NETDEV_DEV(netdev, &dev->dev); |
1327 | ||
1328 | np->netdev = netdev; | |
1329 | ||
1330 | netif_carrier_off(netdev); | |
1331 | ||
1332 | return netdev; | |
1333 | ||
0d160211 | 1334 | exit: |
900e1833 | 1335 | xennet_free_netdev(netdev); |
0d160211 JF |
1336 | return ERR_PTR(err); |
1337 | } | |
1338 | ||
1339 | /** | |
1340 | * Entry point to this code when a new device is created. Allocate the basic | |
1341 | * structures and the ring buffers for communication with the backend, and | |
1342 | * inform the backend of the appropriate details for those. | |
1343 | */ | |
8e0e46bb | 1344 | static int netfront_probe(struct xenbus_device *dev, |
1dd06ae8 | 1345 | const struct xenbus_device_id *id) |
0d160211 JF |
1346 | { |
1347 | int err; | |
1348 | struct net_device *netdev; | |
1349 | struct netfront_info *info; | |
1350 | ||
1351 | netdev = xennet_create_dev(dev); | |
1352 | if (IS_ERR(netdev)) { | |
1353 | err = PTR_ERR(netdev); | |
1354 | xenbus_dev_fatal(dev, err, "creating netdev"); | |
1355 | return err; | |
1356 | } | |
1357 | ||
1358 | info = netdev_priv(netdev); | |
1b713e00 | 1359 | dev_set_drvdata(&dev->dev, info); |
27b917e5 TI |
1360 | #ifdef CONFIG_SYSFS |
1361 | info->netdev->sysfs_groups[0] = &xennet_dev_group; | |
1362 | #endif | |
0d160211 JF |
1363 | err = register_netdev(info->netdev); |
1364 | if (err) { | |
383eda32 | 1365 | pr_warn("%s: register_netdev err=%d\n", __func__, err); |
0d160211 JF |
1366 | goto fail; |
1367 | } | |
1368 | ||
0d160211 JF |
1369 | return 0; |
1370 | ||
1371 | fail: | |
900e1833 | 1372 | xennet_free_netdev(netdev); |
1b713e00 | 1373 | dev_set_drvdata(&dev->dev, NULL); |
0d160211 JF |
1374 | return err; |
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 | { | |
2688fcb7 | 1386 | unsigned int i = 0; |
2688fcb7 AB |
1387 | unsigned int num_queues = info->netdev->real_num_tx_queues; |
1388 | ||
f9feb1e6 DV |
1389 | netif_carrier_off(info->netdev); |
1390 | ||
9a873c71 | 1391 | for (i = 0; i < num_queues && info->queues; ++i) { |
76541869 DV |
1392 | struct netfront_queue *queue = &info->queues[i]; |
1393 | ||
2688fcb7 AB |
1394 | if (queue->tx_irq && (queue->tx_irq == queue->rx_irq)) |
1395 | unbind_from_irqhandler(queue->tx_irq, queue); | |
1396 | if (queue->tx_irq && (queue->tx_irq != queue->rx_irq)) { | |
1397 | unbind_from_irqhandler(queue->tx_irq, queue); | |
1398 | unbind_from_irqhandler(queue->rx_irq, queue); | |
1399 | } | |
1400 | queue->tx_evtchn = queue->rx_evtchn = 0; | |
1401 | queue->tx_irq = queue->rx_irq = 0; | |
0d160211 | 1402 | |
274b0455 CW |
1403 | if (netif_running(info->netdev)) |
1404 | napi_synchronize(&queue->napi); | |
f9feb1e6 | 1405 | |
a5b5dc3c DV |
1406 | xennet_release_tx_bufs(queue); |
1407 | xennet_release_rx_bufs(queue); | |
1408 | gnttab_free_grant_references(queue->gref_tx_head); | |
1409 | gnttab_free_grant_references(queue->gref_rx_head); | |
1410 | ||
2688fcb7 AB |
1411 | /* End access and free the pages */ |
1412 | xennet_end_access(queue->tx_ring_ref, queue->tx.sring); | |
1413 | xennet_end_access(queue->rx_ring_ref, queue->rx.sring); | |
0d160211 | 1414 | |
2688fcb7 AB |
1415 | queue->tx_ring_ref = GRANT_INVALID_REF; |
1416 | queue->rx_ring_ref = GRANT_INVALID_REF; | |
1417 | queue->tx.sring = NULL; | |
1418 | queue->rx.sring = NULL; | |
1419 | } | |
0d160211 JF |
1420 | } |
1421 | ||
1422 | /** | |
1423 | * We are reconnecting to the backend, due to a suspend/resume, or a backend | |
1424 | * driver restart. We tear down our netif structure and recreate it, but | |
1425 | * leave the device-layer structures intact so that this is transparent to the | |
1426 | * rest of the kernel. | |
1427 | */ | |
1428 | static int netfront_resume(struct xenbus_device *dev) | |
1429 | { | |
1b713e00 | 1430 | struct netfront_info *info = dev_get_drvdata(&dev->dev); |
0d160211 JF |
1431 | |
1432 | dev_dbg(&dev->dev, "%s\n", dev->nodename); | |
1433 | ||
1434 | xennet_disconnect_backend(info); | |
1435 | return 0; | |
1436 | } | |
1437 | ||
1438 | static int xen_net_read_mac(struct xenbus_device *dev, u8 mac[]) | |
1439 | { | |
1440 | char *s, *e, *macstr; | |
1441 | int i; | |
1442 | ||
1443 | macstr = s = xenbus_read(XBT_NIL, dev->nodename, "mac", NULL); | |
1444 | if (IS_ERR(macstr)) | |
1445 | return PTR_ERR(macstr); | |
1446 | ||
1447 | for (i = 0; i < ETH_ALEN; i++) { | |
1448 | mac[i] = simple_strtoul(s, &e, 16); | |
1449 | if ((s == e) || (*e != ((i == ETH_ALEN-1) ? '\0' : ':'))) { | |
1450 | kfree(macstr); | |
1451 | return -ENOENT; | |
1452 | } | |
1453 | s = e+1; | |
1454 | } | |
1455 | ||
1456 | kfree(macstr); | |
1457 | return 0; | |
1458 | } | |
1459 | ||
2688fcb7 | 1460 | static int setup_netfront_single(struct netfront_queue *queue) |
d634bf2c WL |
1461 | { |
1462 | int err; | |
1463 | ||
2688fcb7 | 1464 | err = xenbus_alloc_evtchn(queue->info->xbdev, &queue->tx_evtchn); |
d634bf2c WL |
1465 | if (err < 0) |
1466 | goto fail; | |
1467 | ||
2688fcb7 | 1468 | err = bind_evtchn_to_irqhandler(queue->tx_evtchn, |
d634bf2c | 1469 | xennet_interrupt, |
2688fcb7 | 1470 | 0, queue->info->netdev->name, queue); |
d634bf2c WL |
1471 | if (err < 0) |
1472 | goto bind_fail; | |
2688fcb7 AB |
1473 | queue->rx_evtchn = queue->tx_evtchn; |
1474 | queue->rx_irq = queue->tx_irq = err; | |
d634bf2c WL |
1475 | |
1476 | return 0; | |
1477 | ||
1478 | bind_fail: | |
2688fcb7 AB |
1479 | xenbus_free_evtchn(queue->info->xbdev, queue->tx_evtchn); |
1480 | queue->tx_evtchn = 0; | |
d634bf2c WL |
1481 | fail: |
1482 | return err; | |
1483 | } | |
1484 | ||
2688fcb7 | 1485 | static int setup_netfront_split(struct netfront_queue *queue) |
d634bf2c WL |
1486 | { |
1487 | int err; | |
1488 | ||
2688fcb7 | 1489 | err = xenbus_alloc_evtchn(queue->info->xbdev, &queue->tx_evtchn); |
d634bf2c WL |
1490 | if (err < 0) |
1491 | goto fail; | |
2688fcb7 | 1492 | err = xenbus_alloc_evtchn(queue->info->xbdev, &queue->rx_evtchn); |
d634bf2c WL |
1493 | if (err < 0) |
1494 | goto alloc_rx_evtchn_fail; | |
1495 | ||
2688fcb7 AB |
1496 | snprintf(queue->tx_irq_name, sizeof(queue->tx_irq_name), |
1497 | "%s-tx", queue->name); | |
1498 | err = bind_evtchn_to_irqhandler(queue->tx_evtchn, | |
d634bf2c | 1499 | xennet_tx_interrupt, |
2688fcb7 | 1500 | 0, queue->tx_irq_name, queue); |
d634bf2c WL |
1501 | if (err < 0) |
1502 | goto bind_tx_fail; | |
2688fcb7 | 1503 | queue->tx_irq = err; |
d634bf2c | 1504 | |
2688fcb7 AB |
1505 | snprintf(queue->rx_irq_name, sizeof(queue->rx_irq_name), |
1506 | "%s-rx", queue->name); | |
1507 | err = bind_evtchn_to_irqhandler(queue->rx_evtchn, | |
d634bf2c | 1508 | xennet_rx_interrupt, |
2688fcb7 | 1509 | 0, queue->rx_irq_name, queue); |
d634bf2c WL |
1510 | if (err < 0) |
1511 | goto bind_rx_fail; | |
2688fcb7 | 1512 | queue->rx_irq = err; |
d634bf2c WL |
1513 | |
1514 | return 0; | |
1515 | ||
1516 | bind_rx_fail: | |
2688fcb7 AB |
1517 | unbind_from_irqhandler(queue->tx_irq, queue); |
1518 | queue->tx_irq = 0; | |
d634bf2c | 1519 | bind_tx_fail: |
2688fcb7 AB |
1520 | xenbus_free_evtchn(queue->info->xbdev, queue->rx_evtchn); |
1521 | queue->rx_evtchn = 0; | |
d634bf2c | 1522 | alloc_rx_evtchn_fail: |
2688fcb7 AB |
1523 | xenbus_free_evtchn(queue->info->xbdev, queue->tx_evtchn); |
1524 | queue->tx_evtchn = 0; | |
d634bf2c WL |
1525 | fail: |
1526 | return err; | |
1527 | } | |
1528 | ||
2688fcb7 AB |
1529 | static int setup_netfront(struct xenbus_device *dev, |
1530 | struct netfront_queue *queue, unsigned int feature_split_evtchn) | |
0d160211 JF |
1531 | { |
1532 | struct xen_netif_tx_sring *txs; | |
1533 | struct xen_netif_rx_sring *rxs; | |
ccc9d90a | 1534 | grant_ref_t gref; |
0d160211 | 1535 | int err; |
0d160211 | 1536 | |
2688fcb7 AB |
1537 | queue->tx_ring_ref = GRANT_INVALID_REF; |
1538 | queue->rx_ring_ref = GRANT_INVALID_REF; | |
1539 | queue->rx.sring = NULL; | |
1540 | queue->tx.sring = NULL; | |
0d160211 | 1541 | |
a144ff09 | 1542 | txs = (struct xen_netif_tx_sring *)get_zeroed_page(GFP_NOIO | __GFP_HIGH); |
0d160211 JF |
1543 | if (!txs) { |
1544 | err = -ENOMEM; | |
1545 | xenbus_dev_fatal(dev, err, "allocating tx ring page"); | |
1546 | goto fail; | |
1547 | } | |
1548 | SHARED_RING_INIT(txs); | |
30c5d7f0 | 1549 | FRONT_RING_INIT(&queue->tx, txs, XEN_PAGE_SIZE); |
0d160211 | 1550 | |
ccc9d90a | 1551 | err = xenbus_grant_ring(dev, txs, 1, &gref); |
1ca2983a WL |
1552 | if (err < 0) |
1553 | goto grant_tx_ring_fail; | |
ccc9d90a | 1554 | queue->tx_ring_ref = gref; |
0d160211 | 1555 | |
a144ff09 | 1556 | rxs = (struct xen_netif_rx_sring *)get_zeroed_page(GFP_NOIO | __GFP_HIGH); |
0d160211 JF |
1557 | if (!rxs) { |
1558 | err = -ENOMEM; | |
1559 | xenbus_dev_fatal(dev, err, "allocating rx ring page"); | |
1ca2983a | 1560 | goto alloc_rx_ring_fail; |
0d160211 JF |
1561 | } |
1562 | SHARED_RING_INIT(rxs); | |
30c5d7f0 | 1563 | FRONT_RING_INIT(&queue->rx, rxs, XEN_PAGE_SIZE); |
0d160211 | 1564 | |
ccc9d90a | 1565 | err = xenbus_grant_ring(dev, rxs, 1, &gref); |
1ca2983a WL |
1566 | if (err < 0) |
1567 | goto grant_rx_ring_fail; | |
ccc9d90a | 1568 | queue->rx_ring_ref = gref; |
0d160211 | 1569 | |
d634bf2c | 1570 | if (feature_split_evtchn) |
2688fcb7 | 1571 | err = setup_netfront_split(queue); |
d634bf2c WL |
1572 | /* setup single event channel if |
1573 | * a) feature-split-event-channels == 0 | |
1574 | * b) feature-split-event-channels == 1 but failed to setup | |
1575 | */ | |
1576 | if (!feature_split_evtchn || (feature_split_evtchn && err)) | |
2688fcb7 | 1577 | err = setup_netfront_single(queue); |
d634bf2c | 1578 | |
0d160211 | 1579 | if (err) |
1ca2983a | 1580 | goto alloc_evtchn_fail; |
0d160211 | 1581 | |
0d160211 JF |
1582 | return 0; |
1583 | ||
1ca2983a WL |
1584 | /* If we fail to setup netfront, it is safe to just revoke access to |
1585 | * granted pages because backend is not accessing it at this point. | |
1586 | */ | |
1ca2983a | 1587 | alloc_evtchn_fail: |
2688fcb7 | 1588 | gnttab_end_foreign_access_ref(queue->rx_ring_ref, 0); |
1ca2983a WL |
1589 | grant_rx_ring_fail: |
1590 | free_page((unsigned long)rxs); | |
1591 | alloc_rx_ring_fail: | |
2688fcb7 | 1592 | gnttab_end_foreign_access_ref(queue->tx_ring_ref, 0); |
1ca2983a WL |
1593 | grant_tx_ring_fail: |
1594 | free_page((unsigned long)txs); | |
1595 | fail: | |
0d160211 JF |
1596 | return err; |
1597 | } | |
1598 | ||
2688fcb7 AB |
1599 | /* Queue-specific initialisation |
1600 | * This used to be done in xennet_create_dev() but must now | |
1601 | * be run per-queue. | |
1602 | */ | |
1603 | static int xennet_init_queue(struct netfront_queue *queue) | |
1604 | { | |
1605 | unsigned short i; | |
1606 | int err = 0; | |
1607 | ||
1608 | spin_lock_init(&queue->tx_lock); | |
1609 | spin_lock_init(&queue->rx_lock); | |
1610 | ||
493be55a VT |
1611 | setup_timer(&queue->rx_refill_timer, rx_refill_timeout, |
1612 | (unsigned long)queue); | |
2688fcb7 | 1613 | |
8b715010 WL |
1614 | snprintf(queue->name, sizeof(queue->name), "%s-q%u", |
1615 | queue->info->netdev->name, queue->id); | |
1616 | ||
2688fcb7 AB |
1617 | /* Initialise tx_skbs as a free chain containing every entry. */ |
1618 | queue->tx_skb_freelist = 0; | |
1619 | for (i = 0; i < NET_TX_RING_SIZE; i++) { | |
1620 | skb_entry_set_link(&queue->tx_skbs[i], i+1); | |
1621 | queue->grant_tx_ref[i] = GRANT_INVALID_REF; | |
1622 | queue->grant_tx_page[i] = NULL; | |
1623 | } | |
1624 | ||
1625 | /* Clear out rx_skbs */ | |
1626 | for (i = 0; i < NET_RX_RING_SIZE; i++) { | |
1627 | queue->rx_skbs[i] = NULL; | |
1628 | queue->grant_rx_ref[i] = GRANT_INVALID_REF; | |
1629 | } | |
1630 | ||
1631 | /* A grant for every tx ring slot */ | |
1f3c2eba | 1632 | if (gnttab_alloc_grant_references(NET_TX_RING_SIZE, |
2688fcb7 AB |
1633 | &queue->gref_tx_head) < 0) { |
1634 | pr_alert("can't alloc tx grant refs\n"); | |
1635 | err = -ENOMEM; | |
1636 | goto exit; | |
1637 | } | |
1638 | ||
1639 | /* A grant for every rx ring slot */ | |
1f3c2eba | 1640 | if (gnttab_alloc_grant_references(NET_RX_RING_SIZE, |
2688fcb7 AB |
1641 | &queue->gref_rx_head) < 0) { |
1642 | pr_alert("can't alloc rx grant refs\n"); | |
1643 | err = -ENOMEM; | |
1644 | goto exit_free_tx; | |
1645 | } | |
1646 | ||
2688fcb7 AB |
1647 | return 0; |
1648 | ||
1649 | exit_free_tx: | |
1650 | gnttab_free_grant_references(queue->gref_tx_head); | |
1651 | exit: | |
1652 | return err; | |
1653 | } | |
1654 | ||
50ee6061 AB |
1655 | static int write_queue_xenstore_keys(struct netfront_queue *queue, |
1656 | struct xenbus_transaction *xbt, int write_hierarchical) | |
1657 | { | |
1658 | /* Write the queue-specific keys into XenStore in the traditional | |
1659 | * way for a single queue, or in a queue subkeys for multiple | |
1660 | * queues. | |
1661 | */ | |
1662 | struct xenbus_device *dev = queue->info->xbdev; | |
1663 | int err; | |
1664 | const char *message; | |
1665 | char *path; | |
1666 | size_t pathsize; | |
1667 | ||
1668 | /* Choose the correct place to write the keys */ | |
1669 | if (write_hierarchical) { | |
1670 | pathsize = strlen(dev->nodename) + 10; | |
1671 | path = kzalloc(pathsize, GFP_KERNEL); | |
1672 | if (!path) { | |
1673 | err = -ENOMEM; | |
1674 | message = "out of memory while writing ring references"; | |
1675 | goto error; | |
1676 | } | |
1677 | snprintf(path, pathsize, "%s/queue-%u", | |
1678 | dev->nodename, queue->id); | |
1679 | } else { | |
1680 | path = (char *)dev->nodename; | |
1681 | } | |
1682 | ||
1683 | /* Write ring references */ | |
1684 | err = xenbus_printf(*xbt, path, "tx-ring-ref", "%u", | |
1685 | queue->tx_ring_ref); | |
1686 | if (err) { | |
1687 | message = "writing tx-ring-ref"; | |
1688 | goto error; | |
1689 | } | |
1690 | ||
1691 | err = xenbus_printf(*xbt, path, "rx-ring-ref", "%u", | |
1692 | queue->rx_ring_ref); | |
1693 | if (err) { | |
1694 | message = "writing rx-ring-ref"; | |
1695 | goto error; | |
1696 | } | |
1697 | ||
1698 | /* Write event channels; taking into account both shared | |
1699 | * and split event channel scenarios. | |
1700 | */ | |
1701 | if (queue->tx_evtchn == queue->rx_evtchn) { | |
1702 | /* Shared event channel */ | |
1703 | err = xenbus_printf(*xbt, path, | |
1704 | "event-channel", "%u", queue->tx_evtchn); | |
1705 | if (err) { | |
1706 | message = "writing event-channel"; | |
1707 | goto error; | |
1708 | } | |
1709 | } else { | |
1710 | /* Split event channels */ | |
1711 | err = xenbus_printf(*xbt, path, | |
1712 | "event-channel-tx", "%u", queue->tx_evtchn); | |
1713 | if (err) { | |
1714 | message = "writing event-channel-tx"; | |
1715 | goto error; | |
1716 | } | |
1717 | ||
1718 | err = xenbus_printf(*xbt, path, | |
1719 | "event-channel-rx", "%u", queue->rx_evtchn); | |
1720 | if (err) { | |
1721 | message = "writing event-channel-rx"; | |
1722 | goto error; | |
1723 | } | |
1724 | } | |
1725 | ||
1726 | if (write_hierarchical) | |
1727 | kfree(path); | |
1728 | return 0; | |
1729 | ||
1730 | error: | |
1731 | if (write_hierarchical) | |
1732 | kfree(path); | |
1733 | xenbus_dev_fatal(dev, err, "%s", message); | |
1734 | return err; | |
1735 | } | |
1736 | ||
ce58725f DV |
1737 | static void xennet_destroy_queues(struct netfront_info *info) |
1738 | { | |
1739 | unsigned int i; | |
1740 | ||
1741 | rtnl_lock(); | |
1742 | ||
1743 | for (i = 0; i < info->netdev->real_num_tx_queues; i++) { | |
1744 | struct netfront_queue *queue = &info->queues[i]; | |
1745 | ||
1746 | if (netif_running(info->netdev)) | |
1747 | napi_disable(&queue->napi); | |
ad068118 | 1748 | del_timer_sync(&queue->rx_refill_timer); |
ce58725f DV |
1749 | netif_napi_del(&queue->napi); |
1750 | } | |
1751 | ||
1752 | rtnl_unlock(); | |
1753 | ||
1754 | kfree(info->queues); | |
1755 | info->queues = NULL; | |
1756 | } | |
1757 | ||
1758 | static int xennet_create_queues(struct netfront_info *info, | |
ca88ea12 | 1759 | unsigned int *num_queues) |
ce58725f DV |
1760 | { |
1761 | unsigned int i; | |
1762 | int ret; | |
1763 | ||
ca88ea12 | 1764 | info->queues = kcalloc(*num_queues, sizeof(struct netfront_queue), |
ce58725f DV |
1765 | GFP_KERNEL); |
1766 | if (!info->queues) | |
1767 | return -ENOMEM; | |
1768 | ||
1769 | rtnl_lock(); | |
1770 | ||
ca88ea12 | 1771 | for (i = 0; i < *num_queues; i++) { |
ce58725f DV |
1772 | struct netfront_queue *queue = &info->queues[i]; |
1773 | ||
1774 | queue->id = i; | |
1775 | queue->info = info; | |
1776 | ||
1777 | ret = xennet_init_queue(queue); | |
1778 | if (ret < 0) { | |
69cb8524 DV |
1779 | dev_warn(&info->netdev->dev, |
1780 | "only created %d queues\n", i); | |
ca88ea12 | 1781 | *num_queues = i; |
ce58725f DV |
1782 | break; |
1783 | } | |
1784 | ||
1785 | netif_napi_add(queue->info->netdev, &queue->napi, | |
1786 | xennet_poll, 64); | |
1787 | if (netif_running(info->netdev)) | |
1788 | napi_enable(&queue->napi); | |
1789 | } | |
1790 | ||
ca88ea12 | 1791 | netif_set_real_num_tx_queues(info->netdev, *num_queues); |
ce58725f DV |
1792 | |
1793 | rtnl_unlock(); | |
1794 | ||
ca88ea12 | 1795 | if (*num_queues == 0) { |
ce58725f DV |
1796 | dev_err(&info->netdev->dev, "no queues\n"); |
1797 | return -EINVAL; | |
1798 | } | |
1799 | return 0; | |
1800 | } | |
1801 | ||
0d160211 | 1802 | /* Common code used when first setting up, and when resuming. */ |
f502bf2b | 1803 | static int talk_to_netback(struct xenbus_device *dev, |
0d160211 JF |
1804 | struct netfront_info *info) |
1805 | { | |
1806 | const char *message; | |
1807 | struct xenbus_transaction xbt; | |
1808 | int err; | |
2688fcb7 AB |
1809 | unsigned int feature_split_evtchn; |
1810 | unsigned int i = 0; | |
50ee6061 | 1811 | unsigned int max_queues = 0; |
2688fcb7 AB |
1812 | struct netfront_queue *queue = NULL; |
1813 | unsigned int num_queues = 1; | |
0d160211 | 1814 | |
2688fcb7 AB |
1815 | info->netdev->irq = 0; |
1816 | ||
50ee6061 AB |
1817 | /* Check if backend supports multiple queues */ |
1818 | err = xenbus_scanf(XBT_NIL, info->xbdev->otherend, | |
1819 | "multi-queue-max-queues", "%u", &max_queues); | |
1820 | if (err < 0) | |
1821 | max_queues = 1; | |
1822 | num_queues = min(max_queues, xennet_max_queues); | |
1823 | ||
2688fcb7 AB |
1824 | /* Check feature-split-event-channels */ |
1825 | err = xenbus_scanf(XBT_NIL, info->xbdev->otherend, | |
1826 | "feature-split-event-channels", "%u", | |
1827 | &feature_split_evtchn); | |
1828 | if (err < 0) | |
1829 | feature_split_evtchn = 0; | |
1830 | ||
1831 | /* Read mac addr. */ | |
1832 | err = xen_net_read_mac(dev, info->netdev->dev_addr); | |
1833 | if (err) { | |
1834 | xenbus_dev_fatal(dev, err, "parsing %s/mac", dev->nodename); | |
1835 | goto out; | |
1836 | } | |
1837 | ||
ce58725f DV |
1838 | if (info->queues) |
1839 | xennet_destroy_queues(info); | |
1840 | ||
ca88ea12 | 1841 | err = xennet_create_queues(info, &num_queues); |
ce58725f DV |
1842 | if (err < 0) |
1843 | goto destroy_ring; | |
2688fcb7 AB |
1844 | |
1845 | /* Create shared ring, alloc event channel -- for each queue */ | |
1846 | for (i = 0; i < num_queues; ++i) { | |
1847 | queue = &info->queues[i]; | |
2688fcb7 AB |
1848 | err = setup_netfront(dev, queue, feature_split_evtchn); |
1849 | if (err) { | |
ce58725f DV |
1850 | /* setup_netfront() will tidy up the current |
1851 | * queue on error, but we need to clean up | |
2688fcb7 AB |
1852 | * those already allocated. |
1853 | */ | |
1854 | if (i > 0) { | |
1855 | rtnl_lock(); | |
1856 | netif_set_real_num_tx_queues(info->netdev, i); | |
1857 | rtnl_unlock(); | |
1858 | goto destroy_ring; | |
1859 | } else { | |
1860 | goto out; | |
1861 | } | |
1862 | } | |
1863 | } | |
0d160211 JF |
1864 | |
1865 | again: | |
1866 | err = xenbus_transaction_start(&xbt); | |
1867 | if (err) { | |
1868 | xenbus_dev_fatal(dev, err, "starting transaction"); | |
1869 | goto destroy_ring; | |
1870 | } | |
1871 | ||
812494d9 | 1872 | if (xenbus_exists(XBT_NIL, |
1873 | info->xbdev->otherend, "multi-queue-max-queues")) { | |
50ee6061 | 1874 | /* Write the number of queues */ |
812494d9 | 1875 | err = xenbus_printf(xbt, dev->nodename, |
1876 | "multi-queue-num-queues", "%u", num_queues); | |
d634bf2c | 1877 | if (err) { |
50ee6061 AB |
1878 | message = "writing multi-queue-num-queues"; |
1879 | goto abort_transaction_no_dev_fatal; | |
d634bf2c | 1880 | } |
812494d9 | 1881 | } |
50ee6061 | 1882 | |
812494d9 | 1883 | if (num_queues == 1) { |
1884 | err = write_queue_xenstore_keys(&info->queues[0], &xbt, 0); /* flat */ | |
1885 | if (err) | |
1886 | goto abort_transaction_no_dev_fatal; | |
1887 | } else { | |
50ee6061 AB |
1888 | /* Write the keys for each queue */ |
1889 | for (i = 0; i < num_queues; ++i) { | |
1890 | queue = &info->queues[i]; | |
1891 | err = write_queue_xenstore_keys(queue, &xbt, 1); /* hierarchical */ | |
1892 | if (err) | |
1893 | goto abort_transaction_no_dev_fatal; | |
d634bf2c | 1894 | } |
0d160211 JF |
1895 | } |
1896 | ||
50ee6061 | 1897 | /* The remaining keys are not queue-specific */ |
0d160211 JF |
1898 | err = xenbus_printf(xbt, dev->nodename, "request-rx-copy", "%u", |
1899 | 1); | |
1900 | if (err) { | |
1901 | message = "writing request-rx-copy"; | |
1902 | goto abort_transaction; | |
1903 | } | |
1904 | ||
1905 | err = xenbus_printf(xbt, dev->nodename, "feature-rx-notify", "%d", 1); | |
1906 | if (err) { | |
1907 | message = "writing feature-rx-notify"; | |
1908 | goto abort_transaction; | |
1909 | } | |
1910 | ||
1911 | err = xenbus_printf(xbt, dev->nodename, "feature-sg", "%d", 1); | |
1912 | if (err) { | |
1913 | message = "writing feature-sg"; | |
1914 | goto abort_transaction; | |
1915 | } | |
1916 | ||
1917 | err = xenbus_printf(xbt, dev->nodename, "feature-gso-tcpv4", "%d", 1); | |
1918 | if (err) { | |
1919 | message = "writing feature-gso-tcpv4"; | |
1920 | goto abort_transaction; | |
1921 | } | |
1922 | ||
2c0057de PD |
1923 | err = xenbus_write(xbt, dev->nodename, "feature-gso-tcpv6", "1"); |
1924 | if (err) { | |
1925 | message = "writing feature-gso-tcpv6"; | |
1926 | goto abort_transaction; | |
1927 | } | |
1928 | ||
1929 | err = xenbus_write(xbt, dev->nodename, "feature-ipv6-csum-offload", | |
1930 | "1"); | |
1931 | if (err) { | |
1932 | message = "writing feature-ipv6-csum-offload"; | |
1933 | goto abort_transaction; | |
1934 | } | |
1935 | ||
0d160211 JF |
1936 | err = xenbus_transaction_end(xbt, 0); |
1937 | if (err) { | |
1938 | if (err == -EAGAIN) | |
1939 | goto again; | |
1940 | xenbus_dev_fatal(dev, err, "completing transaction"); | |
1941 | goto destroy_ring; | |
1942 | } | |
1943 | ||
1944 | return 0; | |
1945 | ||
1946 | abort_transaction: | |
0d160211 | 1947 | xenbus_dev_fatal(dev, err, "%s", message); |
50ee6061 AB |
1948 | abort_transaction_no_dev_fatal: |
1949 | xenbus_transaction_end(xbt, 1); | |
0d160211 JF |
1950 | destroy_ring: |
1951 | xennet_disconnect_backend(info); | |
2688fcb7 AB |
1952 | kfree(info->queues); |
1953 | info->queues = NULL; | |
0d160211 JF |
1954 | out: |
1955 | return err; | |
1956 | } | |
1957 | ||
0d160211 JF |
1958 | static int xennet_connect(struct net_device *dev) |
1959 | { | |
1960 | struct netfront_info *np = netdev_priv(dev); | |
2688fcb7 | 1961 | unsigned int num_queues = 0; |
a5b5dc3c | 1962 | int err; |
0d160211 | 1963 | unsigned int feature_rx_copy; |
2688fcb7 AB |
1964 | unsigned int j = 0; |
1965 | struct netfront_queue *queue = NULL; | |
0d160211 JF |
1966 | |
1967 | err = xenbus_scanf(XBT_NIL, np->xbdev->otherend, | |
1968 | "feature-rx-copy", "%u", &feature_rx_copy); | |
1969 | if (err != 1) | |
1970 | feature_rx_copy = 0; | |
1971 | ||
1972 | if (!feature_rx_copy) { | |
1973 | dev_info(&dev->dev, | |
898eb71c | 1974 | "backend does not support copying receive path\n"); |
0d160211 JF |
1975 | return -ENODEV; |
1976 | } | |
1977 | ||
f502bf2b | 1978 | err = talk_to_netback(np->xbdev, np); |
0d160211 JF |
1979 | if (err) |
1980 | return err; | |
1981 | ||
2688fcb7 AB |
1982 | /* talk_to_netback() sets the correct number of queues */ |
1983 | num_queues = dev->real_num_tx_queues; | |
1984 | ||
1ba37c51 | 1985 | rtnl_lock(); |
fb507934 | 1986 | netdev_update_features(dev); |
1ba37c51 | 1987 | rtnl_unlock(); |
0d160211 | 1988 | |
0d160211 | 1989 | /* |
a5b5dc3c | 1990 | * All public and private state should now be sane. Get |
0d160211 JF |
1991 | * ready to start sending and receiving packets and give the driver |
1992 | * domain a kick because we've probably just requeued some | |
1993 | * packets. | |
1994 | */ | |
1995 | netif_carrier_on(np->netdev); | |
2688fcb7 AB |
1996 | for (j = 0; j < num_queues; ++j) { |
1997 | queue = &np->queues[j]; | |
f50b4076 | 1998 | |
2688fcb7 AB |
1999 | notify_remote_via_irq(queue->tx_irq); |
2000 | if (queue->tx_irq != queue->rx_irq) | |
2001 | notify_remote_via_irq(queue->rx_irq); | |
2688fcb7 | 2002 | |
f50b4076 DV |
2003 | spin_lock_irq(&queue->tx_lock); |
2004 | xennet_tx_buf_gc(queue); | |
2688fcb7 | 2005 | spin_unlock_irq(&queue->tx_lock); |
f50b4076 DV |
2006 | |
2007 | spin_lock_bh(&queue->rx_lock); | |
2008 | xennet_alloc_rx_buffers(queue); | |
2688fcb7 AB |
2009 | spin_unlock_bh(&queue->rx_lock); |
2010 | } | |
0d160211 JF |
2011 | |
2012 | return 0; | |
2013 | } | |
2014 | ||
2015 | /** | |
2016 | * Callback received when the backend's state changes. | |
2017 | */ | |
f502bf2b | 2018 | static void netback_changed(struct xenbus_device *dev, |
0d160211 JF |
2019 | enum xenbus_state backend_state) |
2020 | { | |
1b713e00 | 2021 | struct netfront_info *np = dev_get_drvdata(&dev->dev); |
0d160211 JF |
2022 | struct net_device *netdev = np->netdev; |
2023 | ||
2024 | dev_dbg(&dev->dev, "%s\n", xenbus_strstate(backend_state)); | |
2025 | ||
2026 | switch (backend_state) { | |
2027 | case XenbusStateInitialising: | |
2028 | case XenbusStateInitialised: | |
b78c9512 NI |
2029 | case XenbusStateReconfiguring: |
2030 | case XenbusStateReconfigured: | |
0d160211 | 2031 | case XenbusStateUnknown: |
0d160211 JF |
2032 | break; |
2033 | ||
2034 | case XenbusStateInitWait: | |
2035 | if (dev->state != XenbusStateInitialising) | |
2036 | break; | |
2037 | if (xennet_connect(netdev) != 0) | |
2038 | break; | |
2039 | xenbus_switch_state(dev, XenbusStateConnected); | |
08e34eb1 LE |
2040 | break; |
2041 | ||
2042 | case XenbusStateConnected: | |
ee89bab1 | 2043 | netdev_notify_peers(netdev); |
0d160211 JF |
2044 | break; |
2045 | ||
bce3ea81 DV |
2046 | case XenbusStateClosed: |
2047 | if (dev->state == XenbusStateClosed) | |
2048 | break; | |
2049 | /* Missed the backend's CLOSING state -- fallthrough */ | |
0d160211 JF |
2050 | case XenbusStateClosing: |
2051 | xenbus_frontend_closed(dev); | |
2052 | break; | |
2053 | } | |
2054 | } | |
2055 | ||
e0ce4af9 IC |
2056 | static const struct xennet_stat { |
2057 | char name[ETH_GSTRING_LEN]; | |
2058 | u16 offset; | |
2059 | } xennet_stats[] = { | |
2060 | { | |
2061 | "rx_gso_checksum_fixup", | |
2062 | offsetof(struct netfront_info, rx_gso_checksum_fixup) | |
2063 | }, | |
2064 | }; | |
2065 | ||
2066 | static int xennet_get_sset_count(struct net_device *dev, int string_set) | |
2067 | { | |
2068 | switch (string_set) { | |
2069 | case ETH_SS_STATS: | |
2070 | return ARRAY_SIZE(xennet_stats); | |
2071 | default: | |
2072 | return -EINVAL; | |
2073 | } | |
2074 | } | |
2075 | ||
2076 | static void xennet_get_ethtool_stats(struct net_device *dev, | |
2077 | struct ethtool_stats *stats, u64 * data) | |
2078 | { | |
2079 | void *np = netdev_priv(dev); | |
2080 | int i; | |
2081 | ||
2082 | for (i = 0; i < ARRAY_SIZE(xennet_stats); i++) | |
2688fcb7 | 2083 | data[i] = atomic_read((atomic_t *)(np + xennet_stats[i].offset)); |
e0ce4af9 IC |
2084 | } |
2085 | ||
2086 | static void xennet_get_strings(struct net_device *dev, u32 stringset, u8 * data) | |
2087 | { | |
2088 | int i; | |
2089 | ||
2090 | switch (stringset) { | |
2091 | case ETH_SS_STATS: | |
2092 | for (i = 0; i < ARRAY_SIZE(xennet_stats); i++) | |
2093 | memcpy(data + i * ETH_GSTRING_LEN, | |
2094 | xennet_stats[i].name, ETH_GSTRING_LEN); | |
2095 | break; | |
2096 | } | |
2097 | } | |
2098 | ||
0fc0b732 | 2099 | static const struct ethtool_ops xennet_ethtool_ops = |
0d160211 | 2100 | { |
0d160211 | 2101 | .get_link = ethtool_op_get_link, |
e0ce4af9 IC |
2102 | |
2103 | .get_sset_count = xennet_get_sset_count, | |
2104 | .get_ethtool_stats = xennet_get_ethtool_stats, | |
2105 | .get_strings = xennet_get_strings, | |
0d160211 JF |
2106 | }; |
2107 | ||
2108 | #ifdef CONFIG_SYSFS | |
1f3c2eba DV |
2109 | static ssize_t show_rxbuf(struct device *dev, |
2110 | struct device_attribute *attr, char *buf) | |
0d160211 | 2111 | { |
1f3c2eba | 2112 | return sprintf(buf, "%lu\n", NET_RX_RING_SIZE); |
0d160211 JF |
2113 | } |
2114 | ||
1f3c2eba DV |
2115 | static ssize_t store_rxbuf(struct device *dev, |
2116 | struct device_attribute *attr, | |
2117 | const char *buf, size_t len) | |
0d160211 | 2118 | { |
0d160211 JF |
2119 | char *endp; |
2120 | unsigned long target; | |
2121 | ||
2122 | if (!capable(CAP_NET_ADMIN)) | |
2123 | return -EPERM; | |
2124 | ||
2125 | target = simple_strtoul(buf, &endp, 0); | |
2126 | if (endp == buf) | |
2127 | return -EBADMSG; | |
2128 | ||
1f3c2eba | 2129 | /* rxbuf_min and rxbuf_max are no longer configurable. */ |
0d160211 | 2130 | |
0d160211 JF |
2131 | return len; |
2132 | } | |
2133 | ||
27b917e5 TI |
2134 | static DEVICE_ATTR(rxbuf_min, S_IRUGO|S_IWUSR, show_rxbuf, store_rxbuf); |
2135 | static DEVICE_ATTR(rxbuf_max, S_IRUGO|S_IWUSR, show_rxbuf, store_rxbuf); | |
2136 | static DEVICE_ATTR(rxbuf_cur, S_IRUGO, show_rxbuf, NULL); | |
0d160211 | 2137 | |
27b917e5 TI |
2138 | static struct attribute *xennet_dev_attrs[] = { |
2139 | &dev_attr_rxbuf_min.attr, | |
2140 | &dev_attr_rxbuf_max.attr, | |
2141 | &dev_attr_rxbuf_cur.attr, | |
2142 | NULL | |
2143 | }; | |
0d160211 | 2144 | |
27b917e5 TI |
2145 | static const struct attribute_group xennet_dev_group = { |
2146 | .attrs = xennet_dev_attrs | |
2147 | }; | |
0d160211 JF |
2148 | #endif /* CONFIG_SYSFS */ |
2149 | ||
8e0e46bb | 2150 | static int xennet_remove(struct xenbus_device *dev) |
0d160211 | 2151 | { |
1b713e00 | 2152 | struct netfront_info *info = dev_get_drvdata(&dev->dev); |
0d160211 JF |
2153 | |
2154 | dev_dbg(&dev->dev, "%s\n", dev->nodename); | |
2155 | ||
0d160211 JF |
2156 | xennet_disconnect_backend(info); |
2157 | ||
6bc96d04 IC |
2158 | unregister_netdev(info->netdev); |
2159 | ||
9a873c71 CW |
2160 | if (info->queues) |
2161 | xennet_destroy_queues(info); | |
900e1833 | 2162 | xennet_free_netdev(info->netdev); |
0d160211 JF |
2163 | |
2164 | return 0; | |
2165 | } | |
2166 | ||
95afae48 DV |
2167 | static const struct xenbus_device_id netfront_ids[] = { |
2168 | { "vif" }, | |
2169 | { "" } | |
2170 | }; | |
2171 | ||
2172 | static struct xenbus_driver netfront_driver = { | |
2173 | .ids = netfront_ids, | |
0d160211 | 2174 | .probe = netfront_probe, |
8e0e46bb | 2175 | .remove = xennet_remove, |
0d160211 | 2176 | .resume = netfront_resume, |
f502bf2b | 2177 | .otherend_changed = netback_changed, |
95afae48 | 2178 | }; |
0d160211 JF |
2179 | |
2180 | static int __init netif_init(void) | |
2181 | { | |
6e833587 | 2182 | if (!xen_domain()) |
0d160211 JF |
2183 | return -ENODEV; |
2184 | ||
51c71a3b | 2185 | if (!xen_has_pv_nic_devices()) |
b9136d20 IM |
2186 | return -ENODEV; |
2187 | ||
383eda32 | 2188 | pr_info("Initialising Xen virtual ethernet driver\n"); |
0d160211 | 2189 | |
32a84405 WL |
2190 | /* Allow as many queues as there are CPUs if user has not |
2191 | * specified a value. | |
2192 | */ | |
2193 | if (xennet_max_queues == 0) | |
2194 | xennet_max_queues = num_online_cpus(); | |
50ee6061 | 2195 | |
ffb78a26 | 2196 | return xenbus_register_frontend(&netfront_driver); |
0d160211 JF |
2197 | } |
2198 | module_init(netif_init); | |
2199 | ||
2200 | ||
2201 | static void __exit netif_exit(void) | |
2202 | { | |
ffb78a26 | 2203 | xenbus_unregister_driver(&netfront_driver); |
0d160211 JF |
2204 | } |
2205 | module_exit(netif_exit); | |
2206 | ||
2207 | MODULE_DESCRIPTION("Xen virtual network device frontend"); | |
2208 | MODULE_LICENSE("GPL"); | |
d2f0c52b | 2209 | MODULE_ALIAS("xen:vif"); |
4f93f09b | 2210 | MODULE_ALIAS("xennet"); |