xref: /linux/drivers/net/xen-netback/common.h (revision e58e871becec2d3b04ed91c0c16fe8deac9c9dfa)
1 /*
2  * This program is free software; you can redistribute it and/or
3  * modify it under the terms of the GNU General Public License version 2
4  * as published by the Free Software Foundation; or, when distributed
5  * separately from the Linux kernel or incorporated into other
6  * software packages, subject to the following license:
7  *
8  * Permission is hereby granted, free of charge, to any person obtaining a copy
9  * of this source file (the "Software"), to deal in the Software without
10  * restriction, including without limitation the rights to use, copy, modify,
11  * merge, publish, distribute, sublicense, and/or sell copies of the Software,
12  * and to permit persons to whom the Software is furnished to do so, subject to
13  * the following conditions:
14  *
15  * The above copyright notice and this permission notice shall be included in
16  * all copies or substantial portions of the Software.
17  *
18  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
19  * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
20  * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
21  * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
22  * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
23  * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
24  * IN THE SOFTWARE.
25  */
26 
27 #ifndef __XEN_NETBACK__COMMON_H__
28 #define __XEN_NETBACK__COMMON_H__
29 
30 #define pr_fmt(fmt) KBUILD_MODNAME ":%s: " fmt, __func__
31 
32 #include <linux/module.h>
33 #include <linux/interrupt.h>
34 #include <linux/slab.h>
35 #include <linux/ip.h>
36 #include <linux/in.h>
37 #include <linux/io.h>
38 #include <linux/netdevice.h>
39 #include <linux/etherdevice.h>
40 #include <linux/wait.h>
41 #include <linux/sched.h>
42 
43 #include <xen/interface/io/netif.h>
44 #include <xen/interface/grant_table.h>
45 #include <xen/grant_table.h>
46 #include <xen/xenbus.h>
47 #include <xen/page.h>
48 #include <linux/debugfs.h>
49 
50 typedef unsigned int pending_ring_idx_t;
51 #define INVALID_PENDING_RING_IDX (~0U)
52 
53 struct pending_tx_info {
54 	struct xen_netif_tx_request req; /* tx request */
55 	unsigned int extra_count;
56 	/* Callback data for released SKBs. The callback is always
57 	 * xenvif_zerocopy_callback, desc contains the pending_idx, which is
58 	 * also an index in pending_tx_info array. It is initialized in
59 	 * xenvif_alloc and it never changes.
60 	 * skb_shinfo(skb)->destructor_arg points to the first mapped slot's
61 	 * callback_struct in this array of struct pending_tx_info's, then ctx
62 	 * to the next, or NULL if there is no more slot for this skb.
63 	 * ubuf_to_vif is a helper which finds the struct xenvif from a pointer
64 	 * to this field.
65 	 */
66 	struct ubuf_info callback_struct;
67 };
68 
69 #define XEN_NETIF_TX_RING_SIZE __CONST_RING_SIZE(xen_netif_tx, XEN_PAGE_SIZE)
70 #define XEN_NETIF_RX_RING_SIZE __CONST_RING_SIZE(xen_netif_rx, XEN_PAGE_SIZE)
71 
72 struct xenvif_rx_meta {
73 	int id;
74 	int size;
75 	int gso_type;
76 	int gso_size;
77 };
78 
79 #define GSO_BIT(type) \
80 	(1 << XEN_NETIF_GSO_TYPE_ ## type)
81 
82 /* Discriminate from any valid pending_idx value. */
83 #define INVALID_PENDING_IDX 0xFFFF
84 
85 #define MAX_BUFFER_OFFSET XEN_PAGE_SIZE
86 
87 #define MAX_PENDING_REQS XEN_NETIF_TX_RING_SIZE
88 
89 /* The maximum number of frags is derived from the size of a grant (same
90  * as a Xen page size for now).
91  */
92 #define MAX_XEN_SKB_FRAGS (65536 / XEN_PAGE_SIZE + 1)
93 
94 #define NETBACK_INVALID_HANDLE -1
95 
96 /* To avoid confusion, we define XEN_NETBK_LEGACY_SLOTS_MAX indicating
97  * the maximum slots a valid packet can use. Now this value is defined
98  * to be XEN_NETIF_NR_SLOTS_MIN, which is supposed to be supported by
99  * all backend.
100  */
101 #define XEN_NETBK_LEGACY_SLOTS_MAX XEN_NETIF_NR_SLOTS_MIN
102 
103 /* Queue name is interface name with "-qNNN" appended */
104 #define QUEUE_NAME_SIZE (IFNAMSIZ + 5)
105 
106 /* IRQ name is queue name with "-tx" or "-rx" appended */
107 #define IRQ_NAME_SIZE (QUEUE_NAME_SIZE + 3)
108 
109 struct xenvif;
110 
111 struct xenvif_stats {
112 	/* Stats fields to be updated per-queue.
113 	 * A subset of struct net_device_stats that contains only the
114 	 * fields that are updated in netback.c for each queue.
115 	 */
116 	u64 rx_bytes;
117 	u64 rx_packets;
118 	u64 tx_bytes;
119 	u64 tx_packets;
120 
121 	/* Additional stats used by xenvif */
122 	unsigned long rx_gso_checksum_fixup;
123 	unsigned long tx_zerocopy_sent;
124 	unsigned long tx_zerocopy_success;
125 	unsigned long tx_zerocopy_fail;
126 	unsigned long tx_frag_overflow;
127 };
128 
129 #define COPY_BATCH_SIZE 64
130 
131 struct xenvif_copy_state {
132 	struct gnttab_copy op[COPY_BATCH_SIZE];
133 	RING_IDX idx[COPY_BATCH_SIZE];
134 	unsigned int num;
135 	struct sk_buff_head *completed;
136 };
137 
138 struct xenvif_queue { /* Per-queue data for xenvif */
139 	unsigned int id; /* Queue ID, 0-based */
140 	char name[QUEUE_NAME_SIZE]; /* DEVNAME-qN */
141 	struct xenvif *vif; /* Parent VIF */
142 
143 	/* Use NAPI for guest TX */
144 	struct napi_struct napi;
145 	/* When feature-split-event-channels = 0, tx_irq = rx_irq. */
146 	unsigned int tx_irq;
147 	/* Only used when feature-split-event-channels = 1 */
148 	char tx_irq_name[IRQ_NAME_SIZE]; /* DEVNAME-qN-tx */
149 	struct xen_netif_tx_back_ring tx;
150 	struct sk_buff_head tx_queue;
151 	struct page *mmap_pages[MAX_PENDING_REQS];
152 	pending_ring_idx_t pending_prod;
153 	pending_ring_idx_t pending_cons;
154 	u16 pending_ring[MAX_PENDING_REQS];
155 	struct pending_tx_info pending_tx_info[MAX_PENDING_REQS];
156 	grant_handle_t grant_tx_handle[MAX_PENDING_REQS];
157 
158 	struct gnttab_copy tx_copy_ops[MAX_PENDING_REQS];
159 	struct gnttab_map_grant_ref tx_map_ops[MAX_PENDING_REQS];
160 	struct gnttab_unmap_grant_ref tx_unmap_ops[MAX_PENDING_REQS];
161 	/* passed to gnttab_[un]map_refs with pages under (un)mapping */
162 	struct page *pages_to_map[MAX_PENDING_REQS];
163 	struct page *pages_to_unmap[MAX_PENDING_REQS];
164 
165 	/* This prevents zerocopy callbacks  to race over dealloc_ring */
166 	spinlock_t callback_lock;
167 	/* This prevents dealloc thread and NAPI instance to race over response
168 	 * creation and pending_ring in xenvif_idx_release. In xenvif_tx_err
169 	 * it only protect response creation
170 	 */
171 	spinlock_t response_lock;
172 	pending_ring_idx_t dealloc_prod;
173 	pending_ring_idx_t dealloc_cons;
174 	u16 dealloc_ring[MAX_PENDING_REQS];
175 	struct task_struct *dealloc_task;
176 	wait_queue_head_t dealloc_wq;
177 	atomic_t inflight_packets;
178 
179 	/* Use kthread for guest RX */
180 	struct task_struct *task;
181 	wait_queue_head_t wq;
182 	/* When feature-split-event-channels = 0, tx_irq = rx_irq. */
183 	unsigned int rx_irq;
184 	/* Only used when feature-split-event-channels = 1 */
185 	char rx_irq_name[IRQ_NAME_SIZE]; /* DEVNAME-qN-rx */
186 	struct xen_netif_rx_back_ring rx;
187 	struct sk_buff_head rx_queue;
188 
189 	unsigned int rx_queue_max;
190 	unsigned int rx_queue_len;
191 	unsigned long last_rx_time;
192 	bool stalled;
193 
194 	struct xenvif_copy_state rx_copy;
195 
196 	/* Transmit shaping: allow 'credit_bytes' every 'credit_usec'. */
197 	unsigned long   credit_bytes;
198 	unsigned long   credit_usec;
199 	unsigned long   remaining_credit;
200 	struct timer_list credit_timeout;
201 	u64 credit_window_start;
202 
203 	/* Statistics */
204 	struct xenvif_stats stats;
205 };
206 
207 enum state_bit_shift {
208 	/* This bit marks that the vif is connected */
209 	VIF_STATUS_CONNECTED,
210 };
211 
212 struct xenvif_mcast_addr {
213 	struct list_head entry;
214 	struct rcu_head rcu;
215 	u8 addr[6];
216 };
217 
218 #define XEN_NETBK_MCAST_MAX 64
219 
220 #define XEN_NETBK_MAX_HASH_KEY_SIZE 40
221 #define XEN_NETBK_MAX_HASH_MAPPING_SIZE 128
222 #define XEN_NETBK_HASH_TAG_SIZE 40
223 
224 struct xenvif_hash_cache_entry {
225 	struct list_head link;
226 	struct rcu_head rcu;
227 	u8 tag[XEN_NETBK_HASH_TAG_SIZE];
228 	unsigned int len;
229 	u32 val;
230 	int seq;
231 };
232 
233 struct xenvif_hash_cache {
234 	spinlock_t lock;
235 	struct list_head list;
236 	unsigned int count;
237 	atomic_t seq;
238 };
239 
240 struct xenvif_hash {
241 	unsigned int alg;
242 	u32 flags;
243 	u8 key[XEN_NETBK_MAX_HASH_KEY_SIZE];
244 	u32 mapping[XEN_NETBK_MAX_HASH_MAPPING_SIZE];
245 	unsigned int size;
246 	struct xenvif_hash_cache cache;
247 };
248 
249 struct xenvif {
250 	/* Unique identifier for this interface. */
251 	domid_t          domid;
252 	unsigned int     handle;
253 
254 	u8               fe_dev_addr[6];
255 	struct list_head fe_mcast_addr;
256 	unsigned int     fe_mcast_count;
257 
258 	/* Frontend feature information. */
259 	int gso_mask;
260 
261 	u8 can_sg:1;
262 	u8 ip_csum:1;
263 	u8 ipv6_csum:1;
264 	u8 multicast_control:1;
265 
266 	/* Is this interface disabled? True when backend discovers
267 	 * frontend is rogue.
268 	 */
269 	bool disabled;
270 	unsigned long status;
271 	unsigned long drain_timeout;
272 	unsigned long stall_timeout;
273 
274 	/* Queues */
275 	struct xenvif_queue *queues;
276 	unsigned int num_queues; /* active queues, resource allocated */
277 	unsigned int stalled_queues;
278 
279 	struct xenvif_hash hash;
280 
281 	struct xenbus_watch credit_watch;
282 	struct xenbus_watch mcast_ctrl_watch;
283 
284 	spinlock_t lock;
285 
286 #ifdef CONFIG_DEBUG_FS
287 	struct dentry *xenvif_dbg_root;
288 #endif
289 
290 	struct xen_netif_ctrl_back_ring ctrl;
291 	unsigned int ctrl_irq;
292 
293 	/* Miscellaneous private stuff. */
294 	struct net_device *dev;
295 };
296 
297 struct xenvif_rx_cb {
298 	unsigned long expires;
299 	int meta_slots_used;
300 };
301 
302 #define XENVIF_RX_CB(skb) ((struct xenvif_rx_cb *)(skb)->cb)
303 
304 static inline struct xenbus_device *xenvif_to_xenbus_device(struct xenvif *vif)
305 {
306 	return to_xenbus_device(vif->dev->dev.parent);
307 }
308 
309 void xenvif_tx_credit_callback(unsigned long data);
310 
311 struct xenvif *xenvif_alloc(struct device *parent,
312 			    domid_t domid,
313 			    unsigned int handle);
314 
315 int xenvif_init_queue(struct xenvif_queue *queue);
316 void xenvif_deinit_queue(struct xenvif_queue *queue);
317 
318 int xenvif_connect_data(struct xenvif_queue *queue,
319 			unsigned long tx_ring_ref,
320 			unsigned long rx_ring_ref,
321 			unsigned int tx_evtchn,
322 			unsigned int rx_evtchn);
323 void xenvif_disconnect_data(struct xenvif *vif);
324 int xenvif_connect_ctrl(struct xenvif *vif, grant_ref_t ring_ref,
325 			unsigned int evtchn);
326 void xenvif_disconnect_ctrl(struct xenvif *vif);
327 void xenvif_free(struct xenvif *vif);
328 
329 int xenvif_xenbus_init(void);
330 void xenvif_xenbus_fini(void);
331 
332 int xenvif_schedulable(struct xenvif *vif);
333 
334 int xenvif_queue_stopped(struct xenvif_queue *queue);
335 void xenvif_wake_queue(struct xenvif_queue *queue);
336 
337 /* (Un)Map communication rings. */
338 void xenvif_unmap_frontend_data_rings(struct xenvif_queue *queue);
339 int xenvif_map_frontend_data_rings(struct xenvif_queue *queue,
340 				   grant_ref_t tx_ring_ref,
341 				   grant_ref_t rx_ring_ref);
342 
343 /* Check for SKBs from frontend and schedule backend processing */
344 void xenvif_napi_schedule_or_enable_events(struct xenvif_queue *queue);
345 
346 /* Prevent the device from generating any further traffic. */
347 void xenvif_carrier_off(struct xenvif *vif);
348 
349 int xenvif_tx_action(struct xenvif_queue *queue, int budget);
350 
351 int xenvif_kthread_guest_rx(void *data);
352 void xenvif_kick_thread(struct xenvif_queue *queue);
353 
354 int xenvif_dealloc_kthread(void *data);
355 
356 irqreturn_t xenvif_ctrl_irq_fn(int irq, void *data);
357 
358 void xenvif_rx_action(struct xenvif_queue *queue);
359 void xenvif_rx_queue_tail(struct xenvif_queue *queue, struct sk_buff *skb);
360 
361 void xenvif_carrier_on(struct xenvif *vif);
362 
363 /* Callback from stack when TX packet can be released */
364 void xenvif_zerocopy_callback(struct ubuf_info *ubuf, bool zerocopy_success);
365 
366 /* Unmap a pending page and release it back to the guest */
367 void xenvif_idx_unmap(struct xenvif_queue *queue, u16 pending_idx);
368 
369 static inline pending_ring_idx_t nr_pending_reqs(struct xenvif_queue *queue)
370 {
371 	return MAX_PENDING_REQS -
372 		queue->pending_prod + queue->pending_cons;
373 }
374 
375 irqreturn_t xenvif_interrupt(int irq, void *dev_id);
376 
377 extern bool separate_tx_rx_irq;
378 
379 extern unsigned int rx_drain_timeout_msecs;
380 extern unsigned int rx_stall_timeout_msecs;
381 extern unsigned int xenvif_max_queues;
382 extern unsigned int xenvif_hash_cache_size;
383 
384 #ifdef CONFIG_DEBUG_FS
385 extern struct dentry *xen_netback_dbg_root;
386 #endif
387 
388 void xenvif_skb_zerocopy_prepare(struct xenvif_queue *queue,
389 				 struct sk_buff *skb);
390 void xenvif_skb_zerocopy_complete(struct xenvif_queue *queue);
391 
392 /* Multicast control */
393 bool xenvif_mcast_match(struct xenvif *vif, const u8 *addr);
394 void xenvif_mcast_addr_list_free(struct xenvif *vif);
395 
396 /* Hash */
397 void xenvif_init_hash(struct xenvif *vif);
398 void xenvif_deinit_hash(struct xenvif *vif);
399 
400 u32 xenvif_set_hash_alg(struct xenvif *vif, u32 alg);
401 u32 xenvif_get_hash_flags(struct xenvif *vif, u32 *flags);
402 u32 xenvif_set_hash_flags(struct xenvif *vif, u32 flags);
403 u32 xenvif_set_hash_key(struct xenvif *vif, u32 gref, u32 len);
404 u32 xenvif_set_hash_mapping_size(struct xenvif *vif, u32 size);
405 u32 xenvif_set_hash_mapping(struct xenvif *vif, u32 gref, u32 len,
406 			    u32 off);
407 
408 void xenvif_set_skb_hash(struct xenvif *vif, struct sk_buff *skb);
409 
410 #ifdef CONFIG_DEBUG_FS
411 void xenvif_dump_hash_info(struct xenvif *vif, struct seq_file *m);
412 #endif
413 
414 #endif /* __XEN_NETBACK__COMMON_H__ */
415