Loading...
1/* Copyright (C) 2009 Red Hat, Inc.
2 * Copyright (C) 2006 Rusty Russell IBM Corporation
3 *
4 * Author: Michael S. Tsirkin <mst@redhat.com>
5 *
6 * Inspiration, some code, and most witty comments come from
7 * Documentation/virtual/lguest/lguest.c, by Rusty Russell
8 *
9 * This work is licensed under the terms of the GNU GPL, version 2.
10 *
11 * Generic code for virtio server in host kernel.
12 */
13
14#include <linux/eventfd.h>
15#include <linux/vhost.h>
16#include <linux/virtio_net.h>
17#include <linux/mm.h>
18#include <linux/mmu_context.h>
19#include <linux/miscdevice.h>
20#include <linux/mutex.h>
21#include <linux/rcupdate.h>
22#include <linux/poll.h>
23#include <linux/file.h>
24#include <linux/highmem.h>
25#include <linux/slab.h>
26#include <linux/kthread.h>
27#include <linux/cgroup.h>
28
29#include <linux/net.h>
30#include <linux/if_packet.h>
31#include <linux/if_arp.h>
32
33#include "vhost.h"
34
35enum {
36 VHOST_MEMORY_MAX_NREGIONS = 64,
37 VHOST_MEMORY_F_LOG = 0x1,
38};
39
40static unsigned vhost_zcopy_mask __read_mostly;
41
42#define vhost_used_event(vq) ((u16 __user *)&vq->avail->ring[vq->num])
43#define vhost_avail_event(vq) ((u16 __user *)&vq->used->ring[vq->num])
44
45static void vhost_poll_func(struct file *file, wait_queue_head_t *wqh,
46 poll_table *pt)
47{
48 struct vhost_poll *poll;
49
50 poll = container_of(pt, struct vhost_poll, table);
51 poll->wqh = wqh;
52 add_wait_queue(wqh, &poll->wait);
53}
54
55static int vhost_poll_wakeup(wait_queue_t *wait, unsigned mode, int sync,
56 void *key)
57{
58 struct vhost_poll *poll = container_of(wait, struct vhost_poll, wait);
59
60 if (!((unsigned long)key & poll->mask))
61 return 0;
62
63 vhost_poll_queue(poll);
64 return 0;
65}
66
67static void vhost_work_init(struct vhost_work *work, vhost_work_fn_t fn)
68{
69 INIT_LIST_HEAD(&work->node);
70 work->fn = fn;
71 init_waitqueue_head(&work->done);
72 work->flushing = 0;
73 work->queue_seq = work->done_seq = 0;
74}
75
76/* Init poll structure */
77void vhost_poll_init(struct vhost_poll *poll, vhost_work_fn_t fn,
78 unsigned long mask, struct vhost_dev *dev)
79{
80 init_waitqueue_func_entry(&poll->wait, vhost_poll_wakeup);
81 init_poll_funcptr(&poll->table, vhost_poll_func);
82 poll->mask = mask;
83 poll->dev = dev;
84
85 vhost_work_init(&poll->work, fn);
86}
87
88/* Start polling a file. We add ourselves to file's wait queue. The caller must
89 * keep a reference to a file until after vhost_poll_stop is called. */
90void vhost_poll_start(struct vhost_poll *poll, struct file *file)
91{
92 unsigned long mask;
93
94 mask = file->f_op->poll(file, &poll->table);
95 if (mask)
96 vhost_poll_wakeup(&poll->wait, 0, 0, (void *)mask);
97}
98
99/* Stop polling a file. After this function returns, it becomes safe to drop the
100 * file reference. You must also flush afterwards. */
101void vhost_poll_stop(struct vhost_poll *poll)
102{
103 remove_wait_queue(poll->wqh, &poll->wait);
104}
105
106static bool vhost_work_seq_done(struct vhost_dev *dev, struct vhost_work *work,
107 unsigned seq)
108{
109 int left;
110
111 spin_lock_irq(&dev->work_lock);
112 left = seq - work->done_seq;
113 spin_unlock_irq(&dev->work_lock);
114 return left <= 0;
115}
116
117static void vhost_work_flush(struct vhost_dev *dev, struct vhost_work *work)
118{
119 unsigned seq;
120 int flushing;
121
122 spin_lock_irq(&dev->work_lock);
123 seq = work->queue_seq;
124 work->flushing++;
125 spin_unlock_irq(&dev->work_lock);
126 wait_event(work->done, vhost_work_seq_done(dev, work, seq));
127 spin_lock_irq(&dev->work_lock);
128 flushing = --work->flushing;
129 spin_unlock_irq(&dev->work_lock);
130 BUG_ON(flushing < 0);
131}
132
133/* Flush any work that has been scheduled. When calling this, don't hold any
134 * locks that are also used by the callback. */
135void vhost_poll_flush(struct vhost_poll *poll)
136{
137 vhost_work_flush(poll->dev, &poll->work);
138}
139
140static inline void vhost_work_queue(struct vhost_dev *dev,
141 struct vhost_work *work)
142{
143 unsigned long flags;
144
145 spin_lock_irqsave(&dev->work_lock, flags);
146 if (list_empty(&work->node)) {
147 list_add_tail(&work->node, &dev->work_list);
148 work->queue_seq++;
149 wake_up_process(dev->worker);
150 }
151 spin_unlock_irqrestore(&dev->work_lock, flags);
152}
153
154void vhost_poll_queue(struct vhost_poll *poll)
155{
156 vhost_work_queue(poll->dev, &poll->work);
157}
158
159static void vhost_vq_reset(struct vhost_dev *dev,
160 struct vhost_virtqueue *vq)
161{
162 vq->num = 1;
163 vq->desc = NULL;
164 vq->avail = NULL;
165 vq->used = NULL;
166 vq->last_avail_idx = 0;
167 vq->avail_idx = 0;
168 vq->last_used_idx = 0;
169 vq->signalled_used = 0;
170 vq->signalled_used_valid = false;
171 vq->used_flags = 0;
172 vq->log_used = false;
173 vq->log_addr = -1ull;
174 vq->vhost_hlen = 0;
175 vq->sock_hlen = 0;
176 vq->private_data = NULL;
177 vq->log_base = NULL;
178 vq->error_ctx = NULL;
179 vq->error = NULL;
180 vq->kick = NULL;
181 vq->call_ctx = NULL;
182 vq->call = NULL;
183 vq->log_ctx = NULL;
184 vq->upend_idx = 0;
185 vq->done_idx = 0;
186 vq->ubufs = NULL;
187}
188
189static int vhost_worker(void *data)
190{
191 struct vhost_dev *dev = data;
192 struct vhost_work *work = NULL;
193 unsigned uninitialized_var(seq);
194
195 use_mm(dev->mm);
196
197 for (;;) {
198 /* mb paired w/ kthread_stop */
199 set_current_state(TASK_INTERRUPTIBLE);
200
201 spin_lock_irq(&dev->work_lock);
202 if (work) {
203 work->done_seq = seq;
204 if (work->flushing)
205 wake_up_all(&work->done);
206 }
207
208 if (kthread_should_stop()) {
209 spin_unlock_irq(&dev->work_lock);
210 __set_current_state(TASK_RUNNING);
211 break;
212 }
213 if (!list_empty(&dev->work_list)) {
214 work = list_first_entry(&dev->work_list,
215 struct vhost_work, node);
216 list_del_init(&work->node);
217 seq = work->queue_seq;
218 } else
219 work = NULL;
220 spin_unlock_irq(&dev->work_lock);
221
222 if (work) {
223 __set_current_state(TASK_RUNNING);
224 work->fn(work);
225 } else
226 schedule();
227
228 }
229 unuse_mm(dev->mm);
230 return 0;
231}
232
233static void vhost_vq_free_iovecs(struct vhost_virtqueue *vq)
234{
235 kfree(vq->indirect);
236 vq->indirect = NULL;
237 kfree(vq->log);
238 vq->log = NULL;
239 kfree(vq->heads);
240 vq->heads = NULL;
241 kfree(vq->ubuf_info);
242 vq->ubuf_info = NULL;
243}
244
245void vhost_enable_zcopy(int vq)
246{
247 vhost_zcopy_mask |= 0x1 << vq;
248}
249
250/* Helper to allocate iovec buffers for all vqs. */
251static long vhost_dev_alloc_iovecs(struct vhost_dev *dev)
252{
253 int i;
254 bool zcopy;
255
256 for (i = 0; i < dev->nvqs; ++i) {
257 dev->vqs[i].indirect = kmalloc(sizeof *dev->vqs[i].indirect *
258 UIO_MAXIOV, GFP_KERNEL);
259 dev->vqs[i].log = kmalloc(sizeof *dev->vqs[i].log * UIO_MAXIOV,
260 GFP_KERNEL);
261 dev->vqs[i].heads = kmalloc(sizeof *dev->vqs[i].heads *
262 UIO_MAXIOV, GFP_KERNEL);
263 zcopy = vhost_zcopy_mask & (0x1 << i);
264 if (zcopy)
265 dev->vqs[i].ubuf_info =
266 kmalloc(sizeof *dev->vqs[i].ubuf_info *
267 UIO_MAXIOV, GFP_KERNEL);
268 if (!dev->vqs[i].indirect || !dev->vqs[i].log ||
269 !dev->vqs[i].heads ||
270 (zcopy && !dev->vqs[i].ubuf_info))
271 goto err_nomem;
272 }
273 return 0;
274
275err_nomem:
276 for (; i >= 0; --i)
277 vhost_vq_free_iovecs(&dev->vqs[i]);
278 return -ENOMEM;
279}
280
281static void vhost_dev_free_iovecs(struct vhost_dev *dev)
282{
283 int i;
284
285 for (i = 0; i < dev->nvqs; ++i)
286 vhost_vq_free_iovecs(&dev->vqs[i]);
287}
288
289long vhost_dev_init(struct vhost_dev *dev,
290 struct vhost_virtqueue *vqs, int nvqs)
291{
292 int i;
293
294 dev->vqs = vqs;
295 dev->nvqs = nvqs;
296 mutex_init(&dev->mutex);
297 dev->log_ctx = NULL;
298 dev->log_file = NULL;
299 dev->memory = NULL;
300 dev->mm = NULL;
301 spin_lock_init(&dev->work_lock);
302 INIT_LIST_HEAD(&dev->work_list);
303 dev->worker = NULL;
304
305 for (i = 0; i < dev->nvqs; ++i) {
306 dev->vqs[i].log = NULL;
307 dev->vqs[i].indirect = NULL;
308 dev->vqs[i].heads = NULL;
309 dev->vqs[i].ubuf_info = NULL;
310 dev->vqs[i].dev = dev;
311 mutex_init(&dev->vqs[i].mutex);
312 vhost_vq_reset(dev, dev->vqs + i);
313 if (dev->vqs[i].handle_kick)
314 vhost_poll_init(&dev->vqs[i].poll,
315 dev->vqs[i].handle_kick, POLLIN, dev);
316 }
317
318 return 0;
319}
320
321/* Caller should have device mutex */
322long vhost_dev_check_owner(struct vhost_dev *dev)
323{
324 /* Are you the owner? If not, I don't think you mean to do that */
325 return dev->mm == current->mm ? 0 : -EPERM;
326}
327
328struct vhost_attach_cgroups_struct {
329 struct vhost_work work;
330 struct task_struct *owner;
331 int ret;
332};
333
334static void vhost_attach_cgroups_work(struct vhost_work *work)
335{
336 struct vhost_attach_cgroups_struct *s;
337
338 s = container_of(work, struct vhost_attach_cgroups_struct, work);
339 s->ret = cgroup_attach_task_all(s->owner, current);
340}
341
342static int vhost_attach_cgroups(struct vhost_dev *dev)
343{
344 struct vhost_attach_cgroups_struct attach;
345
346 attach.owner = current;
347 vhost_work_init(&attach.work, vhost_attach_cgroups_work);
348 vhost_work_queue(dev, &attach.work);
349 vhost_work_flush(dev, &attach.work);
350 return attach.ret;
351}
352
353/* Caller should have device mutex */
354static long vhost_dev_set_owner(struct vhost_dev *dev)
355{
356 struct task_struct *worker;
357 int err;
358
359 /* Is there an owner already? */
360 if (dev->mm) {
361 err = -EBUSY;
362 goto err_mm;
363 }
364
365 /* No owner, become one */
366 dev->mm = get_task_mm(current);
367 worker = kthread_create(vhost_worker, dev, "vhost-%d", current->pid);
368 if (IS_ERR(worker)) {
369 err = PTR_ERR(worker);
370 goto err_worker;
371 }
372
373 dev->worker = worker;
374 wake_up_process(worker); /* avoid contributing to loadavg */
375
376 err = vhost_attach_cgroups(dev);
377 if (err)
378 goto err_cgroup;
379
380 err = vhost_dev_alloc_iovecs(dev);
381 if (err)
382 goto err_cgroup;
383
384 return 0;
385err_cgroup:
386 kthread_stop(worker);
387 dev->worker = NULL;
388err_worker:
389 if (dev->mm)
390 mmput(dev->mm);
391 dev->mm = NULL;
392err_mm:
393 return err;
394}
395
396/* Caller should have device mutex */
397long vhost_dev_reset_owner(struct vhost_dev *dev)
398{
399 struct vhost_memory *memory;
400
401 /* Restore memory to default empty mapping. */
402 memory = kmalloc(offsetof(struct vhost_memory, regions), GFP_KERNEL);
403 if (!memory)
404 return -ENOMEM;
405
406 vhost_dev_cleanup(dev);
407
408 memory->nregions = 0;
409 RCU_INIT_POINTER(dev->memory, memory);
410 return 0;
411}
412
413/* In case of DMA done not in order in lower device driver for some reason.
414 * upend_idx is used to track end of used idx, done_idx is used to track head
415 * of used idx. Once lower device DMA done contiguously, we will signal KVM
416 * guest used idx.
417 */
418int vhost_zerocopy_signal_used(struct vhost_virtqueue *vq)
419{
420 int i;
421 int j = 0;
422
423 for (i = vq->done_idx; i != vq->upend_idx; i = (i + 1) % UIO_MAXIOV) {
424 if ((vq->heads[i].len == VHOST_DMA_DONE_LEN)) {
425 vq->heads[i].len = VHOST_DMA_CLEAR_LEN;
426 vhost_add_used_and_signal(vq->dev, vq,
427 vq->heads[i].id, 0);
428 ++j;
429 } else
430 break;
431 }
432 if (j)
433 vq->done_idx = i;
434 return j;
435}
436
437/* Caller should have device mutex */
438void vhost_dev_cleanup(struct vhost_dev *dev)
439{
440 int i;
441
442 for (i = 0; i < dev->nvqs; ++i) {
443 if (dev->vqs[i].kick && dev->vqs[i].handle_kick) {
444 vhost_poll_stop(&dev->vqs[i].poll);
445 vhost_poll_flush(&dev->vqs[i].poll);
446 }
447 /* Wait for all lower device DMAs done. */
448 if (dev->vqs[i].ubufs)
449 vhost_ubuf_put_and_wait(dev->vqs[i].ubufs);
450
451 /* Signal guest as appropriate. */
452 vhost_zerocopy_signal_used(&dev->vqs[i]);
453
454 if (dev->vqs[i].error_ctx)
455 eventfd_ctx_put(dev->vqs[i].error_ctx);
456 if (dev->vqs[i].error)
457 fput(dev->vqs[i].error);
458 if (dev->vqs[i].kick)
459 fput(dev->vqs[i].kick);
460 if (dev->vqs[i].call_ctx)
461 eventfd_ctx_put(dev->vqs[i].call_ctx);
462 if (dev->vqs[i].call)
463 fput(dev->vqs[i].call);
464 vhost_vq_reset(dev, dev->vqs + i);
465 }
466 vhost_dev_free_iovecs(dev);
467 if (dev->log_ctx)
468 eventfd_ctx_put(dev->log_ctx);
469 dev->log_ctx = NULL;
470 if (dev->log_file)
471 fput(dev->log_file);
472 dev->log_file = NULL;
473 /* No one will access memory at this point */
474 kfree(rcu_dereference_protected(dev->memory,
475 lockdep_is_held(&dev->mutex)));
476 RCU_INIT_POINTER(dev->memory, NULL);
477 WARN_ON(!list_empty(&dev->work_list));
478 if (dev->worker) {
479 kthread_stop(dev->worker);
480 dev->worker = NULL;
481 }
482 if (dev->mm)
483 mmput(dev->mm);
484 dev->mm = NULL;
485}
486
487static int log_access_ok(void __user *log_base, u64 addr, unsigned long sz)
488{
489 u64 a = addr / VHOST_PAGE_SIZE / 8;
490
491 /* Make sure 64 bit math will not overflow. */
492 if (a > ULONG_MAX - (unsigned long)log_base ||
493 a + (unsigned long)log_base > ULONG_MAX)
494 return 0;
495
496 return access_ok(VERIFY_WRITE, log_base + a,
497 (sz + VHOST_PAGE_SIZE * 8 - 1) / VHOST_PAGE_SIZE / 8);
498}
499
500/* Caller should have vq mutex and device mutex. */
501static int vq_memory_access_ok(void __user *log_base, struct vhost_memory *mem,
502 int log_all)
503{
504 int i;
505
506 if (!mem)
507 return 0;
508
509 for (i = 0; i < mem->nregions; ++i) {
510 struct vhost_memory_region *m = mem->regions + i;
511 unsigned long a = m->userspace_addr;
512 if (m->memory_size > ULONG_MAX)
513 return 0;
514 else if (!access_ok(VERIFY_WRITE, (void __user *)a,
515 m->memory_size))
516 return 0;
517 else if (log_all && !log_access_ok(log_base,
518 m->guest_phys_addr,
519 m->memory_size))
520 return 0;
521 }
522 return 1;
523}
524
525/* Can we switch to this memory table? */
526/* Caller should have device mutex but not vq mutex */
527static int memory_access_ok(struct vhost_dev *d, struct vhost_memory *mem,
528 int log_all)
529{
530 int i;
531
532 for (i = 0; i < d->nvqs; ++i) {
533 int ok;
534 mutex_lock(&d->vqs[i].mutex);
535 /* If ring is inactive, will check when it's enabled. */
536 if (d->vqs[i].private_data)
537 ok = vq_memory_access_ok(d->vqs[i].log_base, mem,
538 log_all);
539 else
540 ok = 1;
541 mutex_unlock(&d->vqs[i].mutex);
542 if (!ok)
543 return 0;
544 }
545 return 1;
546}
547
548static int vq_access_ok(struct vhost_dev *d, unsigned int num,
549 struct vring_desc __user *desc,
550 struct vring_avail __user *avail,
551 struct vring_used __user *used)
552{
553 size_t s = vhost_has_feature(d, VIRTIO_RING_F_EVENT_IDX) ? 2 : 0;
554 return access_ok(VERIFY_READ, desc, num * sizeof *desc) &&
555 access_ok(VERIFY_READ, avail,
556 sizeof *avail + num * sizeof *avail->ring + s) &&
557 access_ok(VERIFY_WRITE, used,
558 sizeof *used + num * sizeof *used->ring + s);
559}
560
561/* Can we log writes? */
562/* Caller should have device mutex but not vq mutex */
563int vhost_log_access_ok(struct vhost_dev *dev)
564{
565 struct vhost_memory *mp;
566
567 mp = rcu_dereference_protected(dev->memory,
568 lockdep_is_held(&dev->mutex));
569 return memory_access_ok(dev, mp, 1);
570}
571
572/* Verify access for write logging. */
573/* Caller should have vq mutex and device mutex */
574static int vq_log_access_ok(struct vhost_dev *d, struct vhost_virtqueue *vq,
575 void __user *log_base)
576{
577 struct vhost_memory *mp;
578 size_t s = vhost_has_feature(d, VIRTIO_RING_F_EVENT_IDX) ? 2 : 0;
579
580 mp = rcu_dereference_protected(vq->dev->memory,
581 lockdep_is_held(&vq->mutex));
582 return vq_memory_access_ok(log_base, mp,
583 vhost_has_feature(vq->dev, VHOST_F_LOG_ALL)) &&
584 (!vq->log_used || log_access_ok(log_base, vq->log_addr,
585 sizeof *vq->used +
586 vq->num * sizeof *vq->used->ring + s));
587}
588
589/* Can we start vq? */
590/* Caller should have vq mutex and device mutex */
591int vhost_vq_access_ok(struct vhost_virtqueue *vq)
592{
593 return vq_access_ok(vq->dev, vq->num, vq->desc, vq->avail, vq->used) &&
594 vq_log_access_ok(vq->dev, vq, vq->log_base);
595}
596
597static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m)
598{
599 struct vhost_memory mem, *newmem, *oldmem;
600 unsigned long size = offsetof(struct vhost_memory, regions);
601
602 if (copy_from_user(&mem, m, size))
603 return -EFAULT;
604 if (mem.padding)
605 return -EOPNOTSUPP;
606 if (mem.nregions > VHOST_MEMORY_MAX_NREGIONS)
607 return -E2BIG;
608 newmem = kmalloc(size + mem.nregions * sizeof *m->regions, GFP_KERNEL);
609 if (!newmem)
610 return -ENOMEM;
611
612 memcpy(newmem, &mem, size);
613 if (copy_from_user(newmem->regions, m->regions,
614 mem.nregions * sizeof *m->regions)) {
615 kfree(newmem);
616 return -EFAULT;
617 }
618
619 if (!memory_access_ok(d, newmem,
620 vhost_has_feature(d, VHOST_F_LOG_ALL))) {
621 kfree(newmem);
622 return -EFAULT;
623 }
624 oldmem = rcu_dereference_protected(d->memory,
625 lockdep_is_held(&d->mutex));
626 rcu_assign_pointer(d->memory, newmem);
627 synchronize_rcu();
628 kfree(oldmem);
629 return 0;
630}
631
632static long vhost_set_vring(struct vhost_dev *d, int ioctl, void __user *argp)
633{
634 struct file *eventfp, *filep = NULL,
635 *pollstart = NULL, *pollstop = NULL;
636 struct eventfd_ctx *ctx = NULL;
637 u32 __user *idxp = argp;
638 struct vhost_virtqueue *vq;
639 struct vhost_vring_state s;
640 struct vhost_vring_file f;
641 struct vhost_vring_addr a;
642 u32 idx;
643 long r;
644
645 r = get_user(idx, idxp);
646 if (r < 0)
647 return r;
648 if (idx >= d->nvqs)
649 return -ENOBUFS;
650
651 vq = d->vqs + idx;
652
653 mutex_lock(&vq->mutex);
654
655 switch (ioctl) {
656 case VHOST_SET_VRING_NUM:
657 /* Resizing ring with an active backend?
658 * You don't want to do that. */
659 if (vq->private_data) {
660 r = -EBUSY;
661 break;
662 }
663 if (copy_from_user(&s, argp, sizeof s)) {
664 r = -EFAULT;
665 break;
666 }
667 if (!s.num || s.num > 0xffff || (s.num & (s.num - 1))) {
668 r = -EINVAL;
669 break;
670 }
671 vq->num = s.num;
672 break;
673 case VHOST_SET_VRING_BASE:
674 /* Moving base with an active backend?
675 * You don't want to do that. */
676 if (vq->private_data) {
677 r = -EBUSY;
678 break;
679 }
680 if (copy_from_user(&s, argp, sizeof s)) {
681 r = -EFAULT;
682 break;
683 }
684 if (s.num > 0xffff) {
685 r = -EINVAL;
686 break;
687 }
688 vq->last_avail_idx = s.num;
689 /* Forget the cached index value. */
690 vq->avail_idx = vq->last_avail_idx;
691 break;
692 case VHOST_GET_VRING_BASE:
693 s.index = idx;
694 s.num = vq->last_avail_idx;
695 if (copy_to_user(argp, &s, sizeof s))
696 r = -EFAULT;
697 break;
698 case VHOST_SET_VRING_ADDR:
699 if (copy_from_user(&a, argp, sizeof a)) {
700 r = -EFAULT;
701 break;
702 }
703 if (a.flags & ~(0x1 << VHOST_VRING_F_LOG)) {
704 r = -EOPNOTSUPP;
705 break;
706 }
707 /* For 32bit, verify that the top 32bits of the user
708 data are set to zero. */
709 if ((u64)(unsigned long)a.desc_user_addr != a.desc_user_addr ||
710 (u64)(unsigned long)a.used_user_addr != a.used_user_addr ||
711 (u64)(unsigned long)a.avail_user_addr != a.avail_user_addr) {
712 r = -EFAULT;
713 break;
714 }
715 if ((a.avail_user_addr & (sizeof *vq->avail->ring - 1)) ||
716 (a.used_user_addr & (sizeof *vq->used->ring - 1)) ||
717 (a.log_guest_addr & (sizeof *vq->used->ring - 1))) {
718 r = -EINVAL;
719 break;
720 }
721
722 /* We only verify access here if backend is configured.
723 * If it is not, we don't as size might not have been setup.
724 * We will verify when backend is configured. */
725 if (vq->private_data) {
726 if (!vq_access_ok(d, vq->num,
727 (void __user *)(unsigned long)a.desc_user_addr,
728 (void __user *)(unsigned long)a.avail_user_addr,
729 (void __user *)(unsigned long)a.used_user_addr)) {
730 r = -EINVAL;
731 break;
732 }
733
734 /* Also validate log access for used ring if enabled. */
735 if ((a.flags & (0x1 << VHOST_VRING_F_LOG)) &&
736 !log_access_ok(vq->log_base, a.log_guest_addr,
737 sizeof *vq->used +
738 vq->num * sizeof *vq->used->ring)) {
739 r = -EINVAL;
740 break;
741 }
742 }
743
744 vq->log_used = !!(a.flags & (0x1 << VHOST_VRING_F_LOG));
745 vq->desc = (void __user *)(unsigned long)a.desc_user_addr;
746 vq->avail = (void __user *)(unsigned long)a.avail_user_addr;
747 vq->log_addr = a.log_guest_addr;
748 vq->used = (void __user *)(unsigned long)a.used_user_addr;
749 break;
750 case VHOST_SET_VRING_KICK:
751 if (copy_from_user(&f, argp, sizeof f)) {
752 r = -EFAULT;
753 break;
754 }
755 eventfp = f.fd == -1 ? NULL : eventfd_fget(f.fd);
756 if (IS_ERR(eventfp)) {
757 r = PTR_ERR(eventfp);
758 break;
759 }
760 if (eventfp != vq->kick) {
761 pollstop = filep = vq->kick;
762 pollstart = vq->kick = eventfp;
763 } else
764 filep = eventfp;
765 break;
766 case VHOST_SET_VRING_CALL:
767 if (copy_from_user(&f, argp, sizeof f)) {
768 r = -EFAULT;
769 break;
770 }
771 eventfp = f.fd == -1 ? NULL : eventfd_fget(f.fd);
772 if (IS_ERR(eventfp)) {
773 r = PTR_ERR(eventfp);
774 break;
775 }
776 if (eventfp != vq->call) {
777 filep = vq->call;
778 ctx = vq->call_ctx;
779 vq->call = eventfp;
780 vq->call_ctx = eventfp ?
781 eventfd_ctx_fileget(eventfp) : NULL;
782 } else
783 filep = eventfp;
784 break;
785 case VHOST_SET_VRING_ERR:
786 if (copy_from_user(&f, argp, sizeof f)) {
787 r = -EFAULT;
788 break;
789 }
790 eventfp = f.fd == -1 ? NULL : eventfd_fget(f.fd);
791 if (IS_ERR(eventfp)) {
792 r = PTR_ERR(eventfp);
793 break;
794 }
795 if (eventfp != vq->error) {
796 filep = vq->error;
797 vq->error = eventfp;
798 ctx = vq->error_ctx;
799 vq->error_ctx = eventfp ?
800 eventfd_ctx_fileget(eventfp) : NULL;
801 } else
802 filep = eventfp;
803 break;
804 default:
805 r = -ENOIOCTLCMD;
806 }
807
808 if (pollstop && vq->handle_kick)
809 vhost_poll_stop(&vq->poll);
810
811 if (ctx)
812 eventfd_ctx_put(ctx);
813 if (filep)
814 fput(filep);
815
816 if (pollstart && vq->handle_kick)
817 vhost_poll_start(&vq->poll, vq->kick);
818
819 mutex_unlock(&vq->mutex);
820
821 if (pollstop && vq->handle_kick)
822 vhost_poll_flush(&vq->poll);
823 return r;
824}
825
826/* Caller must have device mutex */
827long vhost_dev_ioctl(struct vhost_dev *d, unsigned int ioctl, unsigned long arg)
828{
829 void __user *argp = (void __user *)arg;
830 struct file *eventfp, *filep = NULL;
831 struct eventfd_ctx *ctx = NULL;
832 u64 p;
833 long r;
834 int i, fd;
835
836 /* If you are not the owner, you can become one */
837 if (ioctl == VHOST_SET_OWNER) {
838 r = vhost_dev_set_owner(d);
839 goto done;
840 }
841
842 /* You must be the owner to do anything else */
843 r = vhost_dev_check_owner(d);
844 if (r)
845 goto done;
846
847 switch (ioctl) {
848 case VHOST_SET_MEM_TABLE:
849 r = vhost_set_memory(d, argp);
850 break;
851 case VHOST_SET_LOG_BASE:
852 if (copy_from_user(&p, argp, sizeof p)) {
853 r = -EFAULT;
854 break;
855 }
856 if ((u64)(unsigned long)p != p) {
857 r = -EFAULT;
858 break;
859 }
860 for (i = 0; i < d->nvqs; ++i) {
861 struct vhost_virtqueue *vq;
862 void __user *base = (void __user *)(unsigned long)p;
863 vq = d->vqs + i;
864 mutex_lock(&vq->mutex);
865 /* If ring is inactive, will check when it's enabled. */
866 if (vq->private_data && !vq_log_access_ok(d, vq, base))
867 r = -EFAULT;
868 else
869 vq->log_base = base;
870 mutex_unlock(&vq->mutex);
871 }
872 break;
873 case VHOST_SET_LOG_FD:
874 r = get_user(fd, (int __user *)argp);
875 if (r < 0)
876 break;
877 eventfp = fd == -1 ? NULL : eventfd_fget(fd);
878 if (IS_ERR(eventfp)) {
879 r = PTR_ERR(eventfp);
880 break;
881 }
882 if (eventfp != d->log_file) {
883 filep = d->log_file;
884 ctx = d->log_ctx;
885 d->log_ctx = eventfp ?
886 eventfd_ctx_fileget(eventfp) : NULL;
887 } else
888 filep = eventfp;
889 for (i = 0; i < d->nvqs; ++i) {
890 mutex_lock(&d->vqs[i].mutex);
891 d->vqs[i].log_ctx = d->log_ctx;
892 mutex_unlock(&d->vqs[i].mutex);
893 }
894 if (ctx)
895 eventfd_ctx_put(ctx);
896 if (filep)
897 fput(filep);
898 break;
899 default:
900 r = vhost_set_vring(d, ioctl, argp);
901 break;
902 }
903done:
904 return r;
905}
906
907static const struct vhost_memory_region *find_region(struct vhost_memory *mem,
908 __u64 addr, __u32 len)
909{
910 struct vhost_memory_region *reg;
911 int i;
912
913 /* linear search is not brilliant, but we really have on the order of 6
914 * regions in practice */
915 for (i = 0; i < mem->nregions; ++i) {
916 reg = mem->regions + i;
917 if (reg->guest_phys_addr <= addr &&
918 reg->guest_phys_addr + reg->memory_size - 1 >= addr)
919 return reg;
920 }
921 return NULL;
922}
923
924/* TODO: This is really inefficient. We need something like get_user()
925 * (instruction directly accesses the data, with an exception table entry
926 * returning -EFAULT). See Documentation/x86/exception-tables.txt.
927 */
928static int set_bit_to_user(int nr, void __user *addr)
929{
930 unsigned long log = (unsigned long)addr;
931 struct page *page;
932 void *base;
933 int bit = nr + (log % PAGE_SIZE) * 8;
934 int r;
935
936 r = get_user_pages_fast(log, 1, 1, &page);
937 if (r < 0)
938 return r;
939 BUG_ON(r != 1);
940 base = kmap_atomic(page, KM_USER0);
941 set_bit(bit, base);
942 kunmap_atomic(base, KM_USER0);
943 set_page_dirty_lock(page);
944 put_page(page);
945 return 0;
946}
947
948static int log_write(void __user *log_base,
949 u64 write_address, u64 write_length)
950{
951 u64 write_page = write_address / VHOST_PAGE_SIZE;
952 int r;
953
954 if (!write_length)
955 return 0;
956 write_length += write_address % VHOST_PAGE_SIZE;
957 for (;;) {
958 u64 base = (u64)(unsigned long)log_base;
959 u64 log = base + write_page / 8;
960 int bit = write_page % 8;
961 if ((u64)(unsigned long)log != log)
962 return -EFAULT;
963 r = set_bit_to_user(bit, (void __user *)(unsigned long)log);
964 if (r < 0)
965 return r;
966 if (write_length <= VHOST_PAGE_SIZE)
967 break;
968 write_length -= VHOST_PAGE_SIZE;
969 write_page += 1;
970 }
971 return r;
972}
973
974int vhost_log_write(struct vhost_virtqueue *vq, struct vhost_log *log,
975 unsigned int log_num, u64 len)
976{
977 int i, r;
978
979 /* Make sure data written is seen before log. */
980 smp_wmb();
981 for (i = 0; i < log_num; ++i) {
982 u64 l = min(log[i].len, len);
983 r = log_write(vq->log_base, log[i].addr, l);
984 if (r < 0)
985 return r;
986 len -= l;
987 if (!len) {
988 if (vq->log_ctx)
989 eventfd_signal(vq->log_ctx, 1);
990 return 0;
991 }
992 }
993 /* Length written exceeds what we have stored. This is a bug. */
994 BUG();
995 return 0;
996}
997
998static int vhost_update_used_flags(struct vhost_virtqueue *vq)
999{
1000 void __user *used;
1001 if (__put_user(vq->used_flags, &vq->used->flags) < 0)
1002 return -EFAULT;
1003 if (unlikely(vq->log_used)) {
1004 /* Make sure the flag is seen before log. */
1005 smp_wmb();
1006 /* Log used flag write. */
1007 used = &vq->used->flags;
1008 log_write(vq->log_base, vq->log_addr +
1009 (used - (void __user *)vq->used),
1010 sizeof vq->used->flags);
1011 if (vq->log_ctx)
1012 eventfd_signal(vq->log_ctx, 1);
1013 }
1014 return 0;
1015}
1016
1017static int vhost_update_avail_event(struct vhost_virtqueue *vq, u16 avail_event)
1018{
1019 if (__put_user(vq->avail_idx, vhost_avail_event(vq)))
1020 return -EFAULT;
1021 if (unlikely(vq->log_used)) {
1022 void __user *used;
1023 /* Make sure the event is seen before log. */
1024 smp_wmb();
1025 /* Log avail event write */
1026 used = vhost_avail_event(vq);
1027 log_write(vq->log_base, vq->log_addr +
1028 (used - (void __user *)vq->used),
1029 sizeof *vhost_avail_event(vq));
1030 if (vq->log_ctx)
1031 eventfd_signal(vq->log_ctx, 1);
1032 }
1033 return 0;
1034}
1035
1036int vhost_init_used(struct vhost_virtqueue *vq)
1037{
1038 int r;
1039 if (!vq->private_data)
1040 return 0;
1041
1042 r = vhost_update_used_flags(vq);
1043 if (r)
1044 return r;
1045 vq->signalled_used_valid = false;
1046 return get_user(vq->last_used_idx, &vq->used->idx);
1047}
1048
1049static int translate_desc(struct vhost_dev *dev, u64 addr, u32 len,
1050 struct iovec iov[], int iov_size)
1051{
1052 const struct vhost_memory_region *reg;
1053 struct vhost_memory *mem;
1054 struct iovec *_iov;
1055 u64 s = 0;
1056 int ret = 0;
1057
1058 rcu_read_lock();
1059
1060 mem = rcu_dereference(dev->memory);
1061 while ((u64)len > s) {
1062 u64 size;
1063 if (unlikely(ret >= iov_size)) {
1064 ret = -ENOBUFS;
1065 break;
1066 }
1067 reg = find_region(mem, addr, len);
1068 if (unlikely(!reg)) {
1069 ret = -EFAULT;
1070 break;
1071 }
1072 _iov = iov + ret;
1073 size = reg->memory_size - addr + reg->guest_phys_addr;
1074 _iov->iov_len = min((u64)len, size);
1075 _iov->iov_base = (void __user *)(unsigned long)
1076 (reg->userspace_addr + addr - reg->guest_phys_addr);
1077 s += size;
1078 addr += size;
1079 ++ret;
1080 }
1081
1082 rcu_read_unlock();
1083 return ret;
1084}
1085
1086/* Each buffer in the virtqueues is actually a chain of descriptors. This
1087 * function returns the next descriptor in the chain,
1088 * or -1U if we're at the end. */
1089static unsigned next_desc(struct vring_desc *desc)
1090{
1091 unsigned int next;
1092
1093 /* If this descriptor says it doesn't chain, we're done. */
1094 if (!(desc->flags & VRING_DESC_F_NEXT))
1095 return -1U;
1096
1097 /* Check they're not leading us off end of descriptors. */
1098 next = desc->next;
1099 /* Make sure compiler knows to grab that: we don't want it changing! */
1100 /* We will use the result as an index in an array, so most
1101 * architectures only need a compiler barrier here. */
1102 read_barrier_depends();
1103
1104 return next;
1105}
1106
1107static int get_indirect(struct vhost_dev *dev, struct vhost_virtqueue *vq,
1108 struct iovec iov[], unsigned int iov_size,
1109 unsigned int *out_num, unsigned int *in_num,
1110 struct vhost_log *log, unsigned int *log_num,
1111 struct vring_desc *indirect)
1112{
1113 struct vring_desc desc;
1114 unsigned int i = 0, count, found = 0;
1115 int ret;
1116
1117 /* Sanity check */
1118 if (unlikely(indirect->len % sizeof desc)) {
1119 vq_err(vq, "Invalid length in indirect descriptor: "
1120 "len 0x%llx not multiple of 0x%zx\n",
1121 (unsigned long long)indirect->len,
1122 sizeof desc);
1123 return -EINVAL;
1124 }
1125
1126 ret = translate_desc(dev, indirect->addr, indirect->len, vq->indirect,
1127 UIO_MAXIOV);
1128 if (unlikely(ret < 0)) {
1129 vq_err(vq, "Translation failure %d in indirect.\n", ret);
1130 return ret;
1131 }
1132
1133 /* We will use the result as an address to read from, so most
1134 * architectures only need a compiler barrier here. */
1135 read_barrier_depends();
1136
1137 count = indirect->len / sizeof desc;
1138 /* Buffers are chained via a 16 bit next field, so
1139 * we can have at most 2^16 of these. */
1140 if (unlikely(count > USHRT_MAX + 1)) {
1141 vq_err(vq, "Indirect buffer length too big: %d\n",
1142 indirect->len);
1143 return -E2BIG;
1144 }
1145
1146 do {
1147 unsigned iov_count = *in_num + *out_num;
1148 if (unlikely(++found > count)) {
1149 vq_err(vq, "Loop detected: last one at %u "
1150 "indirect size %u\n",
1151 i, count);
1152 return -EINVAL;
1153 }
1154 if (unlikely(memcpy_fromiovec((unsigned char *)&desc,
1155 vq->indirect, sizeof desc))) {
1156 vq_err(vq, "Failed indirect descriptor: idx %d, %zx\n",
1157 i, (size_t)indirect->addr + i * sizeof desc);
1158 return -EINVAL;
1159 }
1160 if (unlikely(desc.flags & VRING_DESC_F_INDIRECT)) {
1161 vq_err(vq, "Nested indirect descriptor: idx %d, %zx\n",
1162 i, (size_t)indirect->addr + i * sizeof desc);
1163 return -EINVAL;
1164 }
1165
1166 ret = translate_desc(dev, desc.addr, desc.len, iov + iov_count,
1167 iov_size - iov_count);
1168 if (unlikely(ret < 0)) {
1169 vq_err(vq, "Translation failure %d indirect idx %d\n",
1170 ret, i);
1171 return ret;
1172 }
1173 /* If this is an input descriptor, increment that count. */
1174 if (desc.flags & VRING_DESC_F_WRITE) {
1175 *in_num += ret;
1176 if (unlikely(log)) {
1177 log[*log_num].addr = desc.addr;
1178 log[*log_num].len = desc.len;
1179 ++*log_num;
1180 }
1181 } else {
1182 /* If it's an output descriptor, they're all supposed
1183 * to come before any input descriptors. */
1184 if (unlikely(*in_num)) {
1185 vq_err(vq, "Indirect descriptor "
1186 "has out after in: idx %d\n", i);
1187 return -EINVAL;
1188 }
1189 *out_num += ret;
1190 }
1191 } while ((i = next_desc(&desc)) != -1);
1192 return 0;
1193}
1194
1195/* This looks in the virtqueue and for the first available buffer, and converts
1196 * it to an iovec for convenient access. Since descriptors consist of some
1197 * number of output then some number of input descriptors, it's actually two
1198 * iovecs, but we pack them into one and note how many of each there were.
1199 *
1200 * This function returns the descriptor number found, or vq->num (which is
1201 * never a valid descriptor number) if none was found. A negative code is
1202 * returned on error. */
1203int vhost_get_vq_desc(struct vhost_dev *dev, struct vhost_virtqueue *vq,
1204 struct iovec iov[], unsigned int iov_size,
1205 unsigned int *out_num, unsigned int *in_num,
1206 struct vhost_log *log, unsigned int *log_num)
1207{
1208 struct vring_desc desc;
1209 unsigned int i, head, found = 0;
1210 u16 last_avail_idx;
1211 int ret;
1212
1213 /* Check it isn't doing very strange things with descriptor numbers. */
1214 last_avail_idx = vq->last_avail_idx;
1215 if (unlikely(__get_user(vq->avail_idx, &vq->avail->idx))) {
1216 vq_err(vq, "Failed to access avail idx at %p\n",
1217 &vq->avail->idx);
1218 return -EFAULT;
1219 }
1220
1221 if (unlikely((u16)(vq->avail_idx - last_avail_idx) > vq->num)) {
1222 vq_err(vq, "Guest moved used index from %u to %u",
1223 last_avail_idx, vq->avail_idx);
1224 return -EFAULT;
1225 }
1226
1227 /* If there's nothing new since last we looked, return invalid. */
1228 if (vq->avail_idx == last_avail_idx)
1229 return vq->num;
1230
1231 /* Only get avail ring entries after they have been exposed by guest. */
1232 smp_rmb();
1233
1234 /* Grab the next descriptor number they're advertising, and increment
1235 * the index we've seen. */
1236 if (unlikely(__get_user(head,
1237 &vq->avail->ring[last_avail_idx % vq->num]))) {
1238 vq_err(vq, "Failed to read head: idx %d address %p\n",
1239 last_avail_idx,
1240 &vq->avail->ring[last_avail_idx % vq->num]);
1241 return -EFAULT;
1242 }
1243
1244 /* If their number is silly, that's an error. */
1245 if (unlikely(head >= vq->num)) {
1246 vq_err(vq, "Guest says index %u > %u is available",
1247 head, vq->num);
1248 return -EINVAL;
1249 }
1250
1251 /* When we start there are none of either input nor output. */
1252 *out_num = *in_num = 0;
1253 if (unlikely(log))
1254 *log_num = 0;
1255
1256 i = head;
1257 do {
1258 unsigned iov_count = *in_num + *out_num;
1259 if (unlikely(i >= vq->num)) {
1260 vq_err(vq, "Desc index is %u > %u, head = %u",
1261 i, vq->num, head);
1262 return -EINVAL;
1263 }
1264 if (unlikely(++found > vq->num)) {
1265 vq_err(vq, "Loop detected: last one at %u "
1266 "vq size %u head %u\n",
1267 i, vq->num, head);
1268 return -EINVAL;
1269 }
1270 ret = __copy_from_user(&desc, vq->desc + i, sizeof desc);
1271 if (unlikely(ret)) {
1272 vq_err(vq, "Failed to get descriptor: idx %d addr %p\n",
1273 i, vq->desc + i);
1274 return -EFAULT;
1275 }
1276 if (desc.flags & VRING_DESC_F_INDIRECT) {
1277 ret = get_indirect(dev, vq, iov, iov_size,
1278 out_num, in_num,
1279 log, log_num, &desc);
1280 if (unlikely(ret < 0)) {
1281 vq_err(vq, "Failure detected "
1282 "in indirect descriptor at idx %d\n", i);
1283 return ret;
1284 }
1285 continue;
1286 }
1287
1288 ret = translate_desc(dev, desc.addr, desc.len, iov + iov_count,
1289 iov_size - iov_count);
1290 if (unlikely(ret < 0)) {
1291 vq_err(vq, "Translation failure %d descriptor idx %d\n",
1292 ret, i);
1293 return ret;
1294 }
1295 if (desc.flags & VRING_DESC_F_WRITE) {
1296 /* If this is an input descriptor,
1297 * increment that count. */
1298 *in_num += ret;
1299 if (unlikely(log)) {
1300 log[*log_num].addr = desc.addr;
1301 log[*log_num].len = desc.len;
1302 ++*log_num;
1303 }
1304 } else {
1305 /* If it's an output descriptor, they're all supposed
1306 * to come before any input descriptors. */
1307 if (unlikely(*in_num)) {
1308 vq_err(vq, "Descriptor has out after in: "
1309 "idx %d\n", i);
1310 return -EINVAL;
1311 }
1312 *out_num += ret;
1313 }
1314 } while ((i = next_desc(&desc)) != -1);
1315
1316 /* On success, increment avail index. */
1317 vq->last_avail_idx++;
1318
1319 /* Assume notifications from guest are disabled at this point,
1320 * if they aren't we would need to update avail_event index. */
1321 BUG_ON(!(vq->used_flags & VRING_USED_F_NO_NOTIFY));
1322 return head;
1323}
1324
1325/* Reverse the effect of vhost_get_vq_desc. Useful for error handling. */
1326void vhost_discard_vq_desc(struct vhost_virtqueue *vq, int n)
1327{
1328 vq->last_avail_idx -= n;
1329}
1330
1331/* After we've used one of their buffers, we tell them about it. We'll then
1332 * want to notify the guest, using eventfd. */
1333int vhost_add_used(struct vhost_virtqueue *vq, unsigned int head, int len)
1334{
1335 struct vring_used_elem __user *used;
1336
1337 /* The virtqueue contains a ring of used buffers. Get a pointer to the
1338 * next entry in that used ring. */
1339 used = &vq->used->ring[vq->last_used_idx % vq->num];
1340 if (__put_user(head, &used->id)) {
1341 vq_err(vq, "Failed to write used id");
1342 return -EFAULT;
1343 }
1344 if (__put_user(len, &used->len)) {
1345 vq_err(vq, "Failed to write used len");
1346 return -EFAULT;
1347 }
1348 /* Make sure buffer is written before we update index. */
1349 smp_wmb();
1350 if (__put_user(vq->last_used_idx + 1, &vq->used->idx)) {
1351 vq_err(vq, "Failed to increment used idx");
1352 return -EFAULT;
1353 }
1354 if (unlikely(vq->log_used)) {
1355 /* Make sure data is seen before log. */
1356 smp_wmb();
1357 /* Log used ring entry write. */
1358 log_write(vq->log_base,
1359 vq->log_addr +
1360 ((void __user *)used - (void __user *)vq->used),
1361 sizeof *used);
1362 /* Log used index update. */
1363 log_write(vq->log_base,
1364 vq->log_addr + offsetof(struct vring_used, idx),
1365 sizeof vq->used->idx);
1366 if (vq->log_ctx)
1367 eventfd_signal(vq->log_ctx, 1);
1368 }
1369 vq->last_used_idx++;
1370 /* If the driver never bothers to signal in a very long while,
1371 * used index might wrap around. If that happens, invalidate
1372 * signalled_used index we stored. TODO: make sure driver
1373 * signals at least once in 2^16 and remove this. */
1374 if (unlikely(vq->last_used_idx == vq->signalled_used))
1375 vq->signalled_used_valid = false;
1376 return 0;
1377}
1378
1379static int __vhost_add_used_n(struct vhost_virtqueue *vq,
1380 struct vring_used_elem *heads,
1381 unsigned count)
1382{
1383 struct vring_used_elem __user *used;
1384 u16 old, new;
1385 int start;
1386
1387 start = vq->last_used_idx % vq->num;
1388 used = vq->used->ring + start;
1389 if (__copy_to_user(used, heads, count * sizeof *used)) {
1390 vq_err(vq, "Failed to write used");
1391 return -EFAULT;
1392 }
1393 if (unlikely(vq->log_used)) {
1394 /* Make sure data is seen before log. */
1395 smp_wmb();
1396 /* Log used ring entry write. */
1397 log_write(vq->log_base,
1398 vq->log_addr +
1399 ((void __user *)used - (void __user *)vq->used),
1400 count * sizeof *used);
1401 }
1402 old = vq->last_used_idx;
1403 new = (vq->last_used_idx += count);
1404 /* If the driver never bothers to signal in a very long while,
1405 * used index might wrap around. If that happens, invalidate
1406 * signalled_used index we stored. TODO: make sure driver
1407 * signals at least once in 2^16 and remove this. */
1408 if (unlikely((u16)(new - vq->signalled_used) < (u16)(new - old)))
1409 vq->signalled_used_valid = false;
1410 return 0;
1411}
1412
1413/* After we've used one of their buffers, we tell them about it. We'll then
1414 * want to notify the guest, using eventfd. */
1415int vhost_add_used_n(struct vhost_virtqueue *vq, struct vring_used_elem *heads,
1416 unsigned count)
1417{
1418 int start, n, r;
1419
1420 start = vq->last_used_idx % vq->num;
1421 n = vq->num - start;
1422 if (n < count) {
1423 r = __vhost_add_used_n(vq, heads, n);
1424 if (r < 0)
1425 return r;
1426 heads += n;
1427 count -= n;
1428 }
1429 r = __vhost_add_used_n(vq, heads, count);
1430
1431 /* Make sure buffer is written before we update index. */
1432 smp_wmb();
1433 if (put_user(vq->last_used_idx, &vq->used->idx)) {
1434 vq_err(vq, "Failed to increment used idx");
1435 return -EFAULT;
1436 }
1437 if (unlikely(vq->log_used)) {
1438 /* Log used index update. */
1439 log_write(vq->log_base,
1440 vq->log_addr + offsetof(struct vring_used, idx),
1441 sizeof vq->used->idx);
1442 if (vq->log_ctx)
1443 eventfd_signal(vq->log_ctx, 1);
1444 }
1445 return r;
1446}
1447
1448static bool vhost_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
1449{
1450 __u16 old, new, event;
1451 bool v;
1452 /* Flush out used index updates. This is paired
1453 * with the barrier that the Guest executes when enabling
1454 * interrupts. */
1455 smp_mb();
1456
1457 if (vhost_has_feature(dev, VIRTIO_F_NOTIFY_ON_EMPTY) &&
1458 unlikely(vq->avail_idx == vq->last_avail_idx))
1459 return true;
1460
1461 if (!vhost_has_feature(dev, VIRTIO_RING_F_EVENT_IDX)) {
1462 __u16 flags;
1463 if (__get_user(flags, &vq->avail->flags)) {
1464 vq_err(vq, "Failed to get flags");
1465 return true;
1466 }
1467 return !(flags & VRING_AVAIL_F_NO_INTERRUPT);
1468 }
1469 old = vq->signalled_used;
1470 v = vq->signalled_used_valid;
1471 new = vq->signalled_used = vq->last_used_idx;
1472 vq->signalled_used_valid = true;
1473
1474 if (unlikely(!v))
1475 return true;
1476
1477 if (get_user(event, vhost_used_event(vq))) {
1478 vq_err(vq, "Failed to get used event idx");
1479 return true;
1480 }
1481 return vring_need_event(event, new, old);
1482}
1483
1484/* This actually signals the guest, using eventfd. */
1485void vhost_signal(struct vhost_dev *dev, struct vhost_virtqueue *vq)
1486{
1487 /* Signal the Guest tell them we used something up. */
1488 if (vq->call_ctx && vhost_notify(dev, vq))
1489 eventfd_signal(vq->call_ctx, 1);
1490}
1491
1492/* And here's the combo meal deal. Supersize me! */
1493void vhost_add_used_and_signal(struct vhost_dev *dev,
1494 struct vhost_virtqueue *vq,
1495 unsigned int head, int len)
1496{
1497 vhost_add_used(vq, head, len);
1498 vhost_signal(dev, vq);
1499}
1500
1501/* multi-buffer version of vhost_add_used_and_signal */
1502void vhost_add_used_and_signal_n(struct vhost_dev *dev,
1503 struct vhost_virtqueue *vq,
1504 struct vring_used_elem *heads, unsigned count)
1505{
1506 vhost_add_used_n(vq, heads, count);
1507 vhost_signal(dev, vq);
1508}
1509
1510/* OK, now we need to know about added descriptors. */
1511bool vhost_enable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
1512{
1513 u16 avail_idx;
1514 int r;
1515
1516 if (!(vq->used_flags & VRING_USED_F_NO_NOTIFY))
1517 return false;
1518 vq->used_flags &= ~VRING_USED_F_NO_NOTIFY;
1519 if (!vhost_has_feature(dev, VIRTIO_RING_F_EVENT_IDX)) {
1520 r = vhost_update_used_flags(vq);
1521 if (r) {
1522 vq_err(vq, "Failed to enable notification at %p: %d\n",
1523 &vq->used->flags, r);
1524 return false;
1525 }
1526 } else {
1527 r = vhost_update_avail_event(vq, vq->avail_idx);
1528 if (r) {
1529 vq_err(vq, "Failed to update avail event index at %p: %d\n",
1530 vhost_avail_event(vq), r);
1531 return false;
1532 }
1533 }
1534 /* They could have slipped one in as we were doing that: make
1535 * sure it's written, then check again. */
1536 smp_mb();
1537 r = __get_user(avail_idx, &vq->avail->idx);
1538 if (r) {
1539 vq_err(vq, "Failed to check avail idx at %p: %d\n",
1540 &vq->avail->idx, r);
1541 return false;
1542 }
1543
1544 return avail_idx != vq->avail_idx;
1545}
1546
1547/* We don't need to be notified again. */
1548void vhost_disable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
1549{
1550 int r;
1551
1552 if (vq->used_flags & VRING_USED_F_NO_NOTIFY)
1553 return;
1554 vq->used_flags |= VRING_USED_F_NO_NOTIFY;
1555 if (!vhost_has_feature(dev, VIRTIO_RING_F_EVENT_IDX)) {
1556 r = vhost_update_used_flags(vq);
1557 if (r)
1558 vq_err(vq, "Failed to enable notification at %p: %d\n",
1559 &vq->used->flags, r);
1560 }
1561}
1562
1563static void vhost_zerocopy_done_signal(struct kref *kref)
1564{
1565 struct vhost_ubuf_ref *ubufs = container_of(kref, struct vhost_ubuf_ref,
1566 kref);
1567 wake_up(&ubufs->wait);
1568}
1569
1570struct vhost_ubuf_ref *vhost_ubuf_alloc(struct vhost_virtqueue *vq,
1571 bool zcopy)
1572{
1573 struct vhost_ubuf_ref *ubufs;
1574 /* No zero copy backend? Nothing to count. */
1575 if (!zcopy)
1576 return NULL;
1577 ubufs = kmalloc(sizeof *ubufs, GFP_KERNEL);
1578 if (!ubufs)
1579 return ERR_PTR(-ENOMEM);
1580 kref_init(&ubufs->kref);
1581 init_waitqueue_head(&ubufs->wait);
1582 ubufs->vq = vq;
1583 return ubufs;
1584}
1585
1586void vhost_ubuf_put(struct vhost_ubuf_ref *ubufs)
1587{
1588 kref_put(&ubufs->kref, vhost_zerocopy_done_signal);
1589}
1590
1591void vhost_ubuf_put_and_wait(struct vhost_ubuf_ref *ubufs)
1592{
1593 kref_put(&ubufs->kref, vhost_zerocopy_done_signal);
1594 wait_event(ubufs->wait, !atomic_read(&ubufs->kref.refcount));
1595 kfree(ubufs);
1596}
1597
1598void vhost_zerocopy_callback(void *arg)
1599{
1600 struct ubuf_info *ubuf = arg;
1601 struct vhost_ubuf_ref *ubufs = ubuf->arg;
1602 struct vhost_virtqueue *vq = ubufs->vq;
1603
1604 /* set len = 1 to mark this desc buffers done DMA */
1605 vq->heads[ubuf->desc].len = VHOST_DMA_DONE_LEN;
1606 kref_put(&ubufs->kref, vhost_zerocopy_done_signal);
1607}
1/* Copyright (C) 2009 Red Hat, Inc.
2 * Copyright (C) 2006 Rusty Russell IBM Corporation
3 *
4 * Author: Michael S. Tsirkin <mst@redhat.com>
5 *
6 * Inspiration, some code, and most witty comments come from
7 * Documentation/virtual/lguest/lguest.c, by Rusty Russell
8 *
9 * This work is licensed under the terms of the GNU GPL, version 2.
10 *
11 * Generic code for virtio server in host kernel.
12 */
13
14#include <linux/eventfd.h>
15#include <linux/vhost.h>
16#include <linux/uio.h>
17#include <linux/mm.h>
18#include <linux/mmu_context.h>
19#include <linux/miscdevice.h>
20#include <linux/mutex.h>
21#include <linux/poll.h>
22#include <linux/file.h>
23#include <linux/highmem.h>
24#include <linux/slab.h>
25#include <linux/vmalloc.h>
26#include <linux/kthread.h>
27#include <linux/cgroup.h>
28#include <linux/module.h>
29#include <linux/sort.h>
30#include <linux/sched/mm.h>
31#include <linux/sched/signal.h>
32#include <linux/interval_tree_generic.h>
33
34#include "vhost.h"
35
36static ushort max_mem_regions = 64;
37module_param(max_mem_regions, ushort, 0444);
38MODULE_PARM_DESC(max_mem_regions,
39 "Maximum number of memory regions in memory map. (default: 64)");
40static int max_iotlb_entries = 2048;
41module_param(max_iotlb_entries, int, 0444);
42MODULE_PARM_DESC(max_iotlb_entries,
43 "Maximum number of iotlb entries. (default: 2048)");
44
45enum {
46 VHOST_MEMORY_F_LOG = 0x1,
47};
48
49#define vhost_used_event(vq) ((__virtio16 __user *)&vq->avail->ring[vq->num])
50#define vhost_avail_event(vq) ((__virtio16 __user *)&vq->used->ring[vq->num])
51
52INTERVAL_TREE_DEFINE(struct vhost_umem_node,
53 rb, __u64, __subtree_last,
54 START, LAST, static inline, vhost_umem_interval_tree);
55
56#ifdef CONFIG_VHOST_CROSS_ENDIAN_LEGACY
57static void vhost_disable_cross_endian(struct vhost_virtqueue *vq)
58{
59 vq->user_be = !virtio_legacy_is_little_endian();
60}
61
62static void vhost_enable_cross_endian_big(struct vhost_virtqueue *vq)
63{
64 vq->user_be = true;
65}
66
67static void vhost_enable_cross_endian_little(struct vhost_virtqueue *vq)
68{
69 vq->user_be = false;
70}
71
72static long vhost_set_vring_endian(struct vhost_virtqueue *vq, int __user *argp)
73{
74 struct vhost_vring_state s;
75
76 if (vq->private_data)
77 return -EBUSY;
78
79 if (copy_from_user(&s, argp, sizeof(s)))
80 return -EFAULT;
81
82 if (s.num != VHOST_VRING_LITTLE_ENDIAN &&
83 s.num != VHOST_VRING_BIG_ENDIAN)
84 return -EINVAL;
85
86 if (s.num == VHOST_VRING_BIG_ENDIAN)
87 vhost_enable_cross_endian_big(vq);
88 else
89 vhost_enable_cross_endian_little(vq);
90
91 return 0;
92}
93
94static long vhost_get_vring_endian(struct vhost_virtqueue *vq, u32 idx,
95 int __user *argp)
96{
97 struct vhost_vring_state s = {
98 .index = idx,
99 .num = vq->user_be
100 };
101
102 if (copy_to_user(argp, &s, sizeof(s)))
103 return -EFAULT;
104
105 return 0;
106}
107
108static void vhost_init_is_le(struct vhost_virtqueue *vq)
109{
110 /* Note for legacy virtio: user_be is initialized at reset time
111 * according to the host endianness. If userspace does not set an
112 * explicit endianness, the default behavior is native endian, as
113 * expected by legacy virtio.
114 */
115 vq->is_le = vhost_has_feature(vq, VIRTIO_F_VERSION_1) || !vq->user_be;
116}
117#else
118static void vhost_disable_cross_endian(struct vhost_virtqueue *vq)
119{
120}
121
122static long vhost_set_vring_endian(struct vhost_virtqueue *vq, int __user *argp)
123{
124 return -ENOIOCTLCMD;
125}
126
127static long vhost_get_vring_endian(struct vhost_virtqueue *vq, u32 idx,
128 int __user *argp)
129{
130 return -ENOIOCTLCMD;
131}
132
133static void vhost_init_is_le(struct vhost_virtqueue *vq)
134{
135 vq->is_le = vhost_has_feature(vq, VIRTIO_F_VERSION_1)
136 || virtio_legacy_is_little_endian();
137}
138#endif /* CONFIG_VHOST_CROSS_ENDIAN_LEGACY */
139
140static void vhost_reset_is_le(struct vhost_virtqueue *vq)
141{
142 vhost_init_is_le(vq);
143}
144
145struct vhost_flush_struct {
146 struct vhost_work work;
147 struct completion wait_event;
148};
149
150static void vhost_flush_work(struct vhost_work *work)
151{
152 struct vhost_flush_struct *s;
153
154 s = container_of(work, struct vhost_flush_struct, work);
155 complete(&s->wait_event);
156}
157
158static void vhost_poll_func(struct file *file, wait_queue_head_t *wqh,
159 poll_table *pt)
160{
161 struct vhost_poll *poll;
162
163 poll = container_of(pt, struct vhost_poll, table);
164 poll->wqh = wqh;
165 add_wait_queue(wqh, &poll->wait);
166}
167
168static int vhost_poll_wakeup(wait_queue_entry_t *wait, unsigned mode, int sync,
169 void *key)
170{
171 struct vhost_poll *poll = container_of(wait, struct vhost_poll, wait);
172
173 if (!(key_to_poll(key) & poll->mask))
174 return 0;
175
176 vhost_poll_queue(poll);
177 return 0;
178}
179
180void vhost_work_init(struct vhost_work *work, vhost_work_fn_t fn)
181{
182 clear_bit(VHOST_WORK_QUEUED, &work->flags);
183 work->fn = fn;
184}
185EXPORT_SYMBOL_GPL(vhost_work_init);
186
187/* Init poll structure */
188void vhost_poll_init(struct vhost_poll *poll, vhost_work_fn_t fn,
189 __poll_t mask, struct vhost_dev *dev)
190{
191 init_waitqueue_func_entry(&poll->wait, vhost_poll_wakeup);
192 init_poll_funcptr(&poll->table, vhost_poll_func);
193 poll->mask = mask;
194 poll->dev = dev;
195 poll->wqh = NULL;
196
197 vhost_work_init(&poll->work, fn);
198}
199EXPORT_SYMBOL_GPL(vhost_poll_init);
200
201/* Start polling a file. We add ourselves to file's wait queue. The caller must
202 * keep a reference to a file until after vhost_poll_stop is called. */
203int vhost_poll_start(struct vhost_poll *poll, struct file *file)
204{
205 __poll_t mask;
206 int ret = 0;
207
208 if (poll->wqh)
209 return 0;
210
211 mask = file->f_op->poll(file, &poll->table);
212 if (mask)
213 vhost_poll_wakeup(&poll->wait, 0, 0, poll_to_key(mask));
214 if (mask & EPOLLERR) {
215 vhost_poll_stop(poll);
216 ret = -EINVAL;
217 }
218
219 return ret;
220}
221EXPORT_SYMBOL_GPL(vhost_poll_start);
222
223/* Stop polling a file. After this function returns, it becomes safe to drop the
224 * file reference. You must also flush afterwards. */
225void vhost_poll_stop(struct vhost_poll *poll)
226{
227 if (poll->wqh) {
228 remove_wait_queue(poll->wqh, &poll->wait);
229 poll->wqh = NULL;
230 }
231}
232EXPORT_SYMBOL_GPL(vhost_poll_stop);
233
234void vhost_work_flush(struct vhost_dev *dev, struct vhost_work *work)
235{
236 struct vhost_flush_struct flush;
237
238 if (dev->worker) {
239 init_completion(&flush.wait_event);
240 vhost_work_init(&flush.work, vhost_flush_work);
241
242 vhost_work_queue(dev, &flush.work);
243 wait_for_completion(&flush.wait_event);
244 }
245}
246EXPORT_SYMBOL_GPL(vhost_work_flush);
247
248/* Flush any work that has been scheduled. When calling this, don't hold any
249 * locks that are also used by the callback. */
250void vhost_poll_flush(struct vhost_poll *poll)
251{
252 vhost_work_flush(poll->dev, &poll->work);
253}
254EXPORT_SYMBOL_GPL(vhost_poll_flush);
255
256void vhost_work_queue(struct vhost_dev *dev, struct vhost_work *work)
257{
258 if (!dev->worker)
259 return;
260
261 if (!test_and_set_bit(VHOST_WORK_QUEUED, &work->flags)) {
262 /* We can only add the work to the list after we're
263 * sure it was not in the list.
264 * test_and_set_bit() implies a memory barrier.
265 */
266 llist_add(&work->node, &dev->work_list);
267 wake_up_process(dev->worker);
268 }
269}
270EXPORT_SYMBOL_GPL(vhost_work_queue);
271
272/* A lockless hint for busy polling code to exit the loop */
273bool vhost_has_work(struct vhost_dev *dev)
274{
275 return !llist_empty(&dev->work_list);
276}
277EXPORT_SYMBOL_GPL(vhost_has_work);
278
279void vhost_poll_queue(struct vhost_poll *poll)
280{
281 vhost_work_queue(poll->dev, &poll->work);
282}
283EXPORT_SYMBOL_GPL(vhost_poll_queue);
284
285static void __vhost_vq_meta_reset(struct vhost_virtqueue *vq)
286{
287 int j;
288
289 for (j = 0; j < VHOST_NUM_ADDRS; j++)
290 vq->meta_iotlb[j] = NULL;
291}
292
293static void vhost_vq_meta_reset(struct vhost_dev *d)
294{
295 int i;
296
297 for (i = 0; i < d->nvqs; ++i)
298 __vhost_vq_meta_reset(d->vqs[i]);
299}
300
301static void vhost_vq_reset(struct vhost_dev *dev,
302 struct vhost_virtqueue *vq)
303{
304 vq->num = 1;
305 vq->desc = NULL;
306 vq->avail = NULL;
307 vq->used = NULL;
308 vq->last_avail_idx = 0;
309 vq->avail_idx = 0;
310 vq->last_used_idx = 0;
311 vq->signalled_used = 0;
312 vq->signalled_used_valid = false;
313 vq->used_flags = 0;
314 vq->log_used = false;
315 vq->log_addr = -1ull;
316 vq->private_data = NULL;
317 vq->acked_features = 0;
318 vq->log_base = NULL;
319 vq->error_ctx = NULL;
320 vq->kick = NULL;
321 vq->call_ctx = NULL;
322 vq->log_ctx = NULL;
323 vhost_reset_is_le(vq);
324 vhost_disable_cross_endian(vq);
325 vq->busyloop_timeout = 0;
326 vq->umem = NULL;
327 vq->iotlb = NULL;
328 __vhost_vq_meta_reset(vq);
329}
330
331static int vhost_worker(void *data)
332{
333 struct vhost_dev *dev = data;
334 struct vhost_work *work, *work_next;
335 struct llist_node *node;
336 mm_segment_t oldfs = get_fs();
337
338 set_fs(USER_DS);
339 use_mm(dev->mm);
340
341 for (;;) {
342 /* mb paired w/ kthread_stop */
343 set_current_state(TASK_INTERRUPTIBLE);
344
345 if (kthread_should_stop()) {
346 __set_current_state(TASK_RUNNING);
347 break;
348 }
349
350 node = llist_del_all(&dev->work_list);
351 if (!node)
352 schedule();
353
354 node = llist_reverse_order(node);
355 /* make sure flag is seen after deletion */
356 smp_wmb();
357 llist_for_each_entry_safe(work, work_next, node, node) {
358 clear_bit(VHOST_WORK_QUEUED, &work->flags);
359 __set_current_state(TASK_RUNNING);
360 work->fn(work);
361 if (need_resched())
362 schedule();
363 }
364 }
365 unuse_mm(dev->mm);
366 set_fs(oldfs);
367 return 0;
368}
369
370static void vhost_vq_free_iovecs(struct vhost_virtqueue *vq)
371{
372 kfree(vq->indirect);
373 vq->indirect = NULL;
374 kfree(vq->log);
375 vq->log = NULL;
376 kfree(vq->heads);
377 vq->heads = NULL;
378}
379
380/* Helper to allocate iovec buffers for all vqs. */
381static long vhost_dev_alloc_iovecs(struct vhost_dev *dev)
382{
383 struct vhost_virtqueue *vq;
384 int i;
385
386 for (i = 0; i < dev->nvqs; ++i) {
387 vq = dev->vqs[i];
388 vq->indirect = kmalloc(sizeof *vq->indirect * UIO_MAXIOV,
389 GFP_KERNEL);
390 vq->log = kmalloc(sizeof *vq->log * UIO_MAXIOV, GFP_KERNEL);
391 vq->heads = kmalloc(sizeof *vq->heads * UIO_MAXIOV, GFP_KERNEL);
392 if (!vq->indirect || !vq->log || !vq->heads)
393 goto err_nomem;
394 }
395 return 0;
396
397err_nomem:
398 for (; i >= 0; --i)
399 vhost_vq_free_iovecs(dev->vqs[i]);
400 return -ENOMEM;
401}
402
403static void vhost_dev_free_iovecs(struct vhost_dev *dev)
404{
405 int i;
406
407 for (i = 0; i < dev->nvqs; ++i)
408 vhost_vq_free_iovecs(dev->vqs[i]);
409}
410
411void vhost_dev_init(struct vhost_dev *dev,
412 struct vhost_virtqueue **vqs, int nvqs)
413{
414 struct vhost_virtqueue *vq;
415 int i;
416
417 dev->vqs = vqs;
418 dev->nvqs = nvqs;
419 mutex_init(&dev->mutex);
420 dev->log_ctx = NULL;
421 dev->umem = NULL;
422 dev->iotlb = NULL;
423 dev->mm = NULL;
424 dev->worker = NULL;
425 init_llist_head(&dev->work_list);
426 init_waitqueue_head(&dev->wait);
427 INIT_LIST_HEAD(&dev->read_list);
428 INIT_LIST_HEAD(&dev->pending_list);
429 spin_lock_init(&dev->iotlb_lock);
430
431
432 for (i = 0; i < dev->nvqs; ++i) {
433 vq = dev->vqs[i];
434 vq->log = NULL;
435 vq->indirect = NULL;
436 vq->heads = NULL;
437 vq->dev = dev;
438 mutex_init(&vq->mutex);
439 vhost_vq_reset(dev, vq);
440 if (vq->handle_kick)
441 vhost_poll_init(&vq->poll, vq->handle_kick,
442 EPOLLIN, dev);
443 }
444}
445EXPORT_SYMBOL_GPL(vhost_dev_init);
446
447/* Caller should have device mutex */
448long vhost_dev_check_owner(struct vhost_dev *dev)
449{
450 /* Are you the owner? If not, I don't think you mean to do that */
451 return dev->mm == current->mm ? 0 : -EPERM;
452}
453EXPORT_SYMBOL_GPL(vhost_dev_check_owner);
454
455struct vhost_attach_cgroups_struct {
456 struct vhost_work work;
457 struct task_struct *owner;
458 int ret;
459};
460
461static void vhost_attach_cgroups_work(struct vhost_work *work)
462{
463 struct vhost_attach_cgroups_struct *s;
464
465 s = container_of(work, struct vhost_attach_cgroups_struct, work);
466 s->ret = cgroup_attach_task_all(s->owner, current);
467}
468
469static int vhost_attach_cgroups(struct vhost_dev *dev)
470{
471 struct vhost_attach_cgroups_struct attach;
472
473 attach.owner = current;
474 vhost_work_init(&attach.work, vhost_attach_cgroups_work);
475 vhost_work_queue(dev, &attach.work);
476 vhost_work_flush(dev, &attach.work);
477 return attach.ret;
478}
479
480/* Caller should have device mutex */
481bool vhost_dev_has_owner(struct vhost_dev *dev)
482{
483 return dev->mm;
484}
485EXPORT_SYMBOL_GPL(vhost_dev_has_owner);
486
487/* Caller should have device mutex */
488long vhost_dev_set_owner(struct vhost_dev *dev)
489{
490 struct task_struct *worker;
491 int err;
492
493 /* Is there an owner already? */
494 if (vhost_dev_has_owner(dev)) {
495 err = -EBUSY;
496 goto err_mm;
497 }
498
499 /* No owner, become one */
500 dev->mm = get_task_mm(current);
501 worker = kthread_create(vhost_worker, dev, "vhost-%d", current->pid);
502 if (IS_ERR(worker)) {
503 err = PTR_ERR(worker);
504 goto err_worker;
505 }
506
507 dev->worker = worker;
508 wake_up_process(worker); /* avoid contributing to loadavg */
509
510 err = vhost_attach_cgroups(dev);
511 if (err)
512 goto err_cgroup;
513
514 err = vhost_dev_alloc_iovecs(dev);
515 if (err)
516 goto err_cgroup;
517
518 return 0;
519err_cgroup:
520 kthread_stop(worker);
521 dev->worker = NULL;
522err_worker:
523 if (dev->mm)
524 mmput(dev->mm);
525 dev->mm = NULL;
526err_mm:
527 return err;
528}
529EXPORT_SYMBOL_GPL(vhost_dev_set_owner);
530
531struct vhost_umem *vhost_dev_reset_owner_prepare(void)
532{
533 return kvzalloc(sizeof(struct vhost_umem), GFP_KERNEL);
534}
535EXPORT_SYMBOL_GPL(vhost_dev_reset_owner_prepare);
536
537/* Caller should have device mutex */
538void vhost_dev_reset_owner(struct vhost_dev *dev, struct vhost_umem *umem)
539{
540 int i;
541
542 vhost_dev_cleanup(dev);
543
544 /* Restore memory to default empty mapping. */
545 INIT_LIST_HEAD(&umem->umem_list);
546 dev->umem = umem;
547 /* We don't need VQ locks below since vhost_dev_cleanup makes sure
548 * VQs aren't running.
549 */
550 for (i = 0; i < dev->nvqs; ++i)
551 dev->vqs[i]->umem = umem;
552}
553EXPORT_SYMBOL_GPL(vhost_dev_reset_owner);
554
555void vhost_dev_stop(struct vhost_dev *dev)
556{
557 int i;
558
559 for (i = 0; i < dev->nvqs; ++i) {
560 if (dev->vqs[i]->kick && dev->vqs[i]->handle_kick) {
561 vhost_poll_stop(&dev->vqs[i]->poll);
562 vhost_poll_flush(&dev->vqs[i]->poll);
563 }
564 }
565}
566EXPORT_SYMBOL_GPL(vhost_dev_stop);
567
568static void vhost_umem_free(struct vhost_umem *umem,
569 struct vhost_umem_node *node)
570{
571 vhost_umem_interval_tree_remove(node, &umem->umem_tree);
572 list_del(&node->link);
573 kfree(node);
574 umem->numem--;
575}
576
577static void vhost_umem_clean(struct vhost_umem *umem)
578{
579 struct vhost_umem_node *node, *tmp;
580
581 if (!umem)
582 return;
583
584 list_for_each_entry_safe(node, tmp, &umem->umem_list, link)
585 vhost_umem_free(umem, node);
586
587 kvfree(umem);
588}
589
590static void vhost_clear_msg(struct vhost_dev *dev)
591{
592 struct vhost_msg_node *node, *n;
593
594 spin_lock(&dev->iotlb_lock);
595
596 list_for_each_entry_safe(node, n, &dev->read_list, node) {
597 list_del(&node->node);
598 kfree(node);
599 }
600
601 list_for_each_entry_safe(node, n, &dev->pending_list, node) {
602 list_del(&node->node);
603 kfree(node);
604 }
605
606 spin_unlock(&dev->iotlb_lock);
607}
608
609void vhost_dev_cleanup(struct vhost_dev *dev)
610{
611 int i;
612
613 for (i = 0; i < dev->nvqs; ++i) {
614 if (dev->vqs[i]->error_ctx)
615 eventfd_ctx_put(dev->vqs[i]->error_ctx);
616 if (dev->vqs[i]->kick)
617 fput(dev->vqs[i]->kick);
618 if (dev->vqs[i]->call_ctx)
619 eventfd_ctx_put(dev->vqs[i]->call_ctx);
620 vhost_vq_reset(dev, dev->vqs[i]);
621 }
622 vhost_dev_free_iovecs(dev);
623 if (dev->log_ctx)
624 eventfd_ctx_put(dev->log_ctx);
625 dev->log_ctx = NULL;
626 /* No one will access memory at this point */
627 vhost_umem_clean(dev->umem);
628 dev->umem = NULL;
629 vhost_umem_clean(dev->iotlb);
630 dev->iotlb = NULL;
631 vhost_clear_msg(dev);
632 wake_up_interruptible_poll(&dev->wait, EPOLLIN | EPOLLRDNORM);
633 WARN_ON(!llist_empty(&dev->work_list));
634 if (dev->worker) {
635 kthread_stop(dev->worker);
636 dev->worker = NULL;
637 }
638 if (dev->mm)
639 mmput(dev->mm);
640 dev->mm = NULL;
641}
642EXPORT_SYMBOL_GPL(vhost_dev_cleanup);
643
644static bool log_access_ok(void __user *log_base, u64 addr, unsigned long sz)
645{
646 u64 a = addr / VHOST_PAGE_SIZE / 8;
647
648 /* Make sure 64 bit math will not overflow. */
649 if (a > ULONG_MAX - (unsigned long)log_base ||
650 a + (unsigned long)log_base > ULONG_MAX)
651 return false;
652
653 return access_ok(VERIFY_WRITE, log_base + a,
654 (sz + VHOST_PAGE_SIZE * 8 - 1) / VHOST_PAGE_SIZE / 8);
655}
656
657static bool vhost_overflow(u64 uaddr, u64 size)
658{
659 /* Make sure 64 bit math will not overflow. */
660 return uaddr > ULONG_MAX || size > ULONG_MAX || uaddr > ULONG_MAX - size;
661}
662
663/* Caller should have vq mutex and device mutex. */
664static bool vq_memory_access_ok(void __user *log_base, struct vhost_umem *umem,
665 int log_all)
666{
667 struct vhost_umem_node *node;
668
669 if (!umem)
670 return false;
671
672 list_for_each_entry(node, &umem->umem_list, link) {
673 unsigned long a = node->userspace_addr;
674
675 if (vhost_overflow(node->userspace_addr, node->size))
676 return false;
677
678
679 if (!access_ok(VERIFY_WRITE, (void __user *)a,
680 node->size))
681 return false;
682 else if (log_all && !log_access_ok(log_base,
683 node->start,
684 node->size))
685 return false;
686 }
687 return true;
688}
689
690static inline void __user *vhost_vq_meta_fetch(struct vhost_virtqueue *vq,
691 u64 addr, unsigned int size,
692 int type)
693{
694 const struct vhost_umem_node *node = vq->meta_iotlb[type];
695
696 if (!node)
697 return NULL;
698
699 return (void *)(uintptr_t)(node->userspace_addr + addr - node->start);
700}
701
702/* Can we switch to this memory table? */
703/* Caller should have device mutex but not vq mutex */
704static bool memory_access_ok(struct vhost_dev *d, struct vhost_umem *umem,
705 int log_all)
706{
707 int i;
708
709 for (i = 0; i < d->nvqs; ++i) {
710 bool ok;
711 bool log;
712
713 mutex_lock(&d->vqs[i]->mutex);
714 log = log_all || vhost_has_feature(d->vqs[i], VHOST_F_LOG_ALL);
715 /* If ring is inactive, will check when it's enabled. */
716 if (d->vqs[i]->private_data)
717 ok = vq_memory_access_ok(d->vqs[i]->log_base,
718 umem, log);
719 else
720 ok = true;
721 mutex_unlock(&d->vqs[i]->mutex);
722 if (!ok)
723 return false;
724 }
725 return true;
726}
727
728static int translate_desc(struct vhost_virtqueue *vq, u64 addr, u32 len,
729 struct iovec iov[], int iov_size, int access);
730
731static int vhost_copy_to_user(struct vhost_virtqueue *vq, void __user *to,
732 const void *from, unsigned size)
733{
734 int ret;
735
736 if (!vq->iotlb)
737 return __copy_to_user(to, from, size);
738 else {
739 /* This function should be called after iotlb
740 * prefetch, which means we're sure that all vq
741 * could be access through iotlb. So -EAGAIN should
742 * not happen in this case.
743 */
744 struct iov_iter t;
745 void __user *uaddr = vhost_vq_meta_fetch(vq,
746 (u64)(uintptr_t)to, size,
747 VHOST_ADDR_USED);
748
749 if (uaddr)
750 return __copy_to_user(uaddr, from, size);
751
752 ret = translate_desc(vq, (u64)(uintptr_t)to, size, vq->iotlb_iov,
753 ARRAY_SIZE(vq->iotlb_iov),
754 VHOST_ACCESS_WO);
755 if (ret < 0)
756 goto out;
757 iov_iter_init(&t, WRITE, vq->iotlb_iov, ret, size);
758 ret = copy_to_iter(from, size, &t);
759 if (ret == size)
760 ret = 0;
761 }
762out:
763 return ret;
764}
765
766static int vhost_copy_from_user(struct vhost_virtqueue *vq, void *to,
767 void __user *from, unsigned size)
768{
769 int ret;
770
771 if (!vq->iotlb)
772 return __copy_from_user(to, from, size);
773 else {
774 /* This function should be called after iotlb
775 * prefetch, which means we're sure that vq
776 * could be access through iotlb. So -EAGAIN should
777 * not happen in this case.
778 */
779 void __user *uaddr = vhost_vq_meta_fetch(vq,
780 (u64)(uintptr_t)from, size,
781 VHOST_ADDR_DESC);
782 struct iov_iter f;
783
784 if (uaddr)
785 return __copy_from_user(to, uaddr, size);
786
787 ret = translate_desc(vq, (u64)(uintptr_t)from, size, vq->iotlb_iov,
788 ARRAY_SIZE(vq->iotlb_iov),
789 VHOST_ACCESS_RO);
790 if (ret < 0) {
791 vq_err(vq, "IOTLB translation failure: uaddr "
792 "%p size 0x%llx\n", from,
793 (unsigned long long) size);
794 goto out;
795 }
796 iov_iter_init(&f, READ, vq->iotlb_iov, ret, size);
797 ret = copy_from_iter(to, size, &f);
798 if (ret == size)
799 ret = 0;
800 }
801
802out:
803 return ret;
804}
805
806static void __user *__vhost_get_user_slow(struct vhost_virtqueue *vq,
807 void __user *addr, unsigned int size,
808 int type)
809{
810 int ret;
811
812 ret = translate_desc(vq, (u64)(uintptr_t)addr, size, vq->iotlb_iov,
813 ARRAY_SIZE(vq->iotlb_iov),
814 VHOST_ACCESS_RO);
815 if (ret < 0) {
816 vq_err(vq, "IOTLB translation failure: uaddr "
817 "%p size 0x%llx\n", addr,
818 (unsigned long long) size);
819 return NULL;
820 }
821
822 if (ret != 1 || vq->iotlb_iov[0].iov_len != size) {
823 vq_err(vq, "Non atomic userspace memory access: uaddr "
824 "%p size 0x%llx\n", addr,
825 (unsigned long long) size);
826 return NULL;
827 }
828
829 return vq->iotlb_iov[0].iov_base;
830}
831
832/* This function should be called after iotlb
833 * prefetch, which means we're sure that vq
834 * could be access through iotlb. So -EAGAIN should
835 * not happen in this case.
836 */
837static inline void __user *__vhost_get_user(struct vhost_virtqueue *vq,
838 void *addr, unsigned int size,
839 int type)
840{
841 void __user *uaddr = vhost_vq_meta_fetch(vq,
842 (u64)(uintptr_t)addr, size, type);
843 if (uaddr)
844 return uaddr;
845
846 return __vhost_get_user_slow(vq, addr, size, type);
847}
848
849#define vhost_put_user(vq, x, ptr) \
850({ \
851 int ret = -EFAULT; \
852 if (!vq->iotlb) { \
853 ret = __put_user(x, ptr); \
854 } else { \
855 __typeof__(ptr) to = \
856 (__typeof__(ptr)) __vhost_get_user(vq, ptr, \
857 sizeof(*ptr), VHOST_ADDR_USED); \
858 if (to != NULL) \
859 ret = __put_user(x, to); \
860 else \
861 ret = -EFAULT; \
862 } \
863 ret; \
864})
865
866#define vhost_get_user(vq, x, ptr, type) \
867({ \
868 int ret; \
869 if (!vq->iotlb) { \
870 ret = __get_user(x, ptr); \
871 } else { \
872 __typeof__(ptr) from = \
873 (__typeof__(ptr)) __vhost_get_user(vq, ptr, \
874 sizeof(*ptr), \
875 type); \
876 if (from != NULL) \
877 ret = __get_user(x, from); \
878 else \
879 ret = -EFAULT; \
880 } \
881 ret; \
882})
883
884#define vhost_get_avail(vq, x, ptr) \
885 vhost_get_user(vq, x, ptr, VHOST_ADDR_AVAIL)
886
887#define vhost_get_used(vq, x, ptr) \
888 vhost_get_user(vq, x, ptr, VHOST_ADDR_USED)
889
890static void vhost_dev_lock_vqs(struct vhost_dev *d)
891{
892 int i = 0;
893 for (i = 0; i < d->nvqs; ++i)
894 mutex_lock_nested(&d->vqs[i]->mutex, i);
895}
896
897static void vhost_dev_unlock_vqs(struct vhost_dev *d)
898{
899 int i = 0;
900 for (i = 0; i < d->nvqs; ++i)
901 mutex_unlock(&d->vqs[i]->mutex);
902}
903
904static int vhost_new_umem_range(struct vhost_umem *umem,
905 u64 start, u64 size, u64 end,
906 u64 userspace_addr, int perm)
907{
908 struct vhost_umem_node *tmp, *node = kmalloc(sizeof(*node), GFP_ATOMIC);
909
910 if (!node)
911 return -ENOMEM;
912
913 if (umem->numem == max_iotlb_entries) {
914 tmp = list_first_entry(&umem->umem_list, typeof(*tmp), link);
915 vhost_umem_free(umem, tmp);
916 }
917
918 node->start = start;
919 node->size = size;
920 node->last = end;
921 node->userspace_addr = userspace_addr;
922 node->perm = perm;
923 INIT_LIST_HEAD(&node->link);
924 list_add_tail(&node->link, &umem->umem_list);
925 vhost_umem_interval_tree_insert(node, &umem->umem_tree);
926 umem->numem++;
927
928 return 0;
929}
930
931static void vhost_del_umem_range(struct vhost_umem *umem,
932 u64 start, u64 end)
933{
934 struct vhost_umem_node *node;
935
936 while ((node = vhost_umem_interval_tree_iter_first(&umem->umem_tree,
937 start, end)))
938 vhost_umem_free(umem, node);
939}
940
941static void vhost_iotlb_notify_vq(struct vhost_dev *d,
942 struct vhost_iotlb_msg *msg)
943{
944 struct vhost_msg_node *node, *n;
945
946 spin_lock(&d->iotlb_lock);
947
948 list_for_each_entry_safe(node, n, &d->pending_list, node) {
949 struct vhost_iotlb_msg *vq_msg = &node->msg.iotlb;
950 if (msg->iova <= vq_msg->iova &&
951 msg->iova + msg->size - 1 > vq_msg->iova &&
952 vq_msg->type == VHOST_IOTLB_MISS) {
953 vhost_poll_queue(&node->vq->poll);
954 list_del(&node->node);
955 kfree(node);
956 }
957 }
958
959 spin_unlock(&d->iotlb_lock);
960}
961
962static bool umem_access_ok(u64 uaddr, u64 size, int access)
963{
964 unsigned long a = uaddr;
965
966 /* Make sure 64 bit math will not overflow. */
967 if (vhost_overflow(uaddr, size))
968 return false;
969
970 if ((access & VHOST_ACCESS_RO) &&
971 !access_ok(VERIFY_READ, (void __user *)a, size))
972 return false;
973 if ((access & VHOST_ACCESS_WO) &&
974 !access_ok(VERIFY_WRITE, (void __user *)a, size))
975 return false;
976 return true;
977}
978
979static int vhost_process_iotlb_msg(struct vhost_dev *dev,
980 struct vhost_iotlb_msg *msg)
981{
982 int ret = 0;
983
984 mutex_lock(&dev->mutex);
985 vhost_dev_lock_vqs(dev);
986 switch (msg->type) {
987 case VHOST_IOTLB_UPDATE:
988 if (!dev->iotlb) {
989 ret = -EFAULT;
990 break;
991 }
992 if (!umem_access_ok(msg->uaddr, msg->size, msg->perm)) {
993 ret = -EFAULT;
994 break;
995 }
996 vhost_vq_meta_reset(dev);
997 if (vhost_new_umem_range(dev->iotlb, msg->iova, msg->size,
998 msg->iova + msg->size - 1,
999 msg->uaddr, msg->perm)) {
1000 ret = -ENOMEM;
1001 break;
1002 }
1003 vhost_iotlb_notify_vq(dev, msg);
1004 break;
1005 case VHOST_IOTLB_INVALIDATE:
1006 if (!dev->iotlb) {
1007 ret = -EFAULT;
1008 break;
1009 }
1010 vhost_vq_meta_reset(dev);
1011 vhost_del_umem_range(dev->iotlb, msg->iova,
1012 msg->iova + msg->size - 1);
1013 break;
1014 default:
1015 ret = -EINVAL;
1016 break;
1017 }
1018
1019 vhost_dev_unlock_vqs(dev);
1020 mutex_unlock(&dev->mutex);
1021
1022 return ret;
1023}
1024ssize_t vhost_chr_write_iter(struct vhost_dev *dev,
1025 struct iov_iter *from)
1026{
1027 struct vhost_msg_node node;
1028 unsigned size = sizeof(struct vhost_msg);
1029 size_t ret;
1030 int err;
1031
1032 if (iov_iter_count(from) < size)
1033 return 0;
1034 ret = copy_from_iter(&node.msg, size, from);
1035 if (ret != size)
1036 goto done;
1037
1038 switch (node.msg.type) {
1039 case VHOST_IOTLB_MSG:
1040 err = vhost_process_iotlb_msg(dev, &node.msg.iotlb);
1041 if (err)
1042 ret = err;
1043 break;
1044 default:
1045 ret = -EINVAL;
1046 break;
1047 }
1048
1049done:
1050 return ret;
1051}
1052EXPORT_SYMBOL(vhost_chr_write_iter);
1053
1054__poll_t vhost_chr_poll(struct file *file, struct vhost_dev *dev,
1055 poll_table *wait)
1056{
1057 __poll_t mask = 0;
1058
1059 poll_wait(file, &dev->wait, wait);
1060
1061 if (!list_empty(&dev->read_list))
1062 mask |= EPOLLIN | EPOLLRDNORM;
1063
1064 return mask;
1065}
1066EXPORT_SYMBOL(vhost_chr_poll);
1067
1068ssize_t vhost_chr_read_iter(struct vhost_dev *dev, struct iov_iter *to,
1069 int noblock)
1070{
1071 DEFINE_WAIT(wait);
1072 struct vhost_msg_node *node;
1073 ssize_t ret = 0;
1074 unsigned size = sizeof(struct vhost_msg);
1075
1076 if (iov_iter_count(to) < size)
1077 return 0;
1078
1079 while (1) {
1080 if (!noblock)
1081 prepare_to_wait(&dev->wait, &wait,
1082 TASK_INTERRUPTIBLE);
1083
1084 node = vhost_dequeue_msg(dev, &dev->read_list);
1085 if (node)
1086 break;
1087 if (noblock) {
1088 ret = -EAGAIN;
1089 break;
1090 }
1091 if (signal_pending(current)) {
1092 ret = -ERESTARTSYS;
1093 break;
1094 }
1095 if (!dev->iotlb) {
1096 ret = -EBADFD;
1097 break;
1098 }
1099
1100 schedule();
1101 }
1102
1103 if (!noblock)
1104 finish_wait(&dev->wait, &wait);
1105
1106 if (node) {
1107 ret = copy_to_iter(&node->msg, size, to);
1108
1109 if (ret != size || node->msg.type != VHOST_IOTLB_MISS) {
1110 kfree(node);
1111 return ret;
1112 }
1113
1114 vhost_enqueue_msg(dev, &dev->pending_list, node);
1115 }
1116
1117 return ret;
1118}
1119EXPORT_SYMBOL_GPL(vhost_chr_read_iter);
1120
1121static int vhost_iotlb_miss(struct vhost_virtqueue *vq, u64 iova, int access)
1122{
1123 struct vhost_dev *dev = vq->dev;
1124 struct vhost_msg_node *node;
1125 struct vhost_iotlb_msg *msg;
1126
1127 node = vhost_new_msg(vq, VHOST_IOTLB_MISS);
1128 if (!node)
1129 return -ENOMEM;
1130
1131 msg = &node->msg.iotlb;
1132 msg->type = VHOST_IOTLB_MISS;
1133 msg->iova = iova;
1134 msg->perm = access;
1135
1136 vhost_enqueue_msg(dev, &dev->read_list, node);
1137
1138 return 0;
1139}
1140
1141static bool vq_access_ok(struct vhost_virtqueue *vq, unsigned int num,
1142 struct vring_desc __user *desc,
1143 struct vring_avail __user *avail,
1144 struct vring_used __user *used)
1145
1146{
1147 size_t s = vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX) ? 2 : 0;
1148
1149 return access_ok(VERIFY_READ, desc, num * sizeof *desc) &&
1150 access_ok(VERIFY_READ, avail,
1151 sizeof *avail + num * sizeof *avail->ring + s) &&
1152 access_ok(VERIFY_WRITE, used,
1153 sizeof *used + num * sizeof *used->ring + s);
1154}
1155
1156static void vhost_vq_meta_update(struct vhost_virtqueue *vq,
1157 const struct vhost_umem_node *node,
1158 int type)
1159{
1160 int access = (type == VHOST_ADDR_USED) ?
1161 VHOST_ACCESS_WO : VHOST_ACCESS_RO;
1162
1163 if (likely(node->perm & access))
1164 vq->meta_iotlb[type] = node;
1165}
1166
1167static bool iotlb_access_ok(struct vhost_virtqueue *vq,
1168 int access, u64 addr, u64 len, int type)
1169{
1170 const struct vhost_umem_node *node;
1171 struct vhost_umem *umem = vq->iotlb;
1172 u64 s = 0, size, orig_addr = addr, last = addr + len - 1;
1173
1174 if (vhost_vq_meta_fetch(vq, addr, len, type))
1175 return true;
1176
1177 while (len > s) {
1178 node = vhost_umem_interval_tree_iter_first(&umem->umem_tree,
1179 addr,
1180 last);
1181 if (node == NULL || node->start > addr) {
1182 vhost_iotlb_miss(vq, addr, access);
1183 return false;
1184 } else if (!(node->perm & access)) {
1185 /* Report the possible access violation by
1186 * request another translation from userspace.
1187 */
1188 return false;
1189 }
1190
1191 size = node->size - addr + node->start;
1192
1193 if (orig_addr == addr && size >= len)
1194 vhost_vq_meta_update(vq, node, type);
1195
1196 s += size;
1197 addr += size;
1198 }
1199
1200 return true;
1201}
1202
1203int vq_iotlb_prefetch(struct vhost_virtqueue *vq)
1204{
1205 size_t s = vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX) ? 2 : 0;
1206 unsigned int num = vq->num;
1207
1208 if (!vq->iotlb)
1209 return 1;
1210
1211 return iotlb_access_ok(vq, VHOST_ACCESS_RO, (u64)(uintptr_t)vq->desc,
1212 num * sizeof(*vq->desc), VHOST_ADDR_DESC) &&
1213 iotlb_access_ok(vq, VHOST_ACCESS_RO, (u64)(uintptr_t)vq->avail,
1214 sizeof *vq->avail +
1215 num * sizeof(*vq->avail->ring) + s,
1216 VHOST_ADDR_AVAIL) &&
1217 iotlb_access_ok(vq, VHOST_ACCESS_WO, (u64)(uintptr_t)vq->used,
1218 sizeof *vq->used +
1219 num * sizeof(*vq->used->ring) + s,
1220 VHOST_ADDR_USED);
1221}
1222EXPORT_SYMBOL_GPL(vq_iotlb_prefetch);
1223
1224/* Can we log writes? */
1225/* Caller should have device mutex but not vq mutex */
1226bool vhost_log_access_ok(struct vhost_dev *dev)
1227{
1228 return memory_access_ok(dev, dev->umem, 1);
1229}
1230EXPORT_SYMBOL_GPL(vhost_log_access_ok);
1231
1232/* Verify access for write logging. */
1233/* Caller should have vq mutex and device mutex */
1234static bool vq_log_access_ok(struct vhost_virtqueue *vq,
1235 void __user *log_base)
1236{
1237 size_t s = vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX) ? 2 : 0;
1238
1239 return vq_memory_access_ok(log_base, vq->umem,
1240 vhost_has_feature(vq, VHOST_F_LOG_ALL)) &&
1241 (!vq->log_used || log_access_ok(log_base, vq->log_addr,
1242 sizeof *vq->used +
1243 vq->num * sizeof *vq->used->ring + s));
1244}
1245
1246/* Can we start vq? */
1247/* Caller should have vq mutex and device mutex */
1248bool vhost_vq_access_ok(struct vhost_virtqueue *vq)
1249{
1250 if (!vq_log_access_ok(vq, vq->log_base))
1251 return false;
1252
1253 /* Access validation occurs at prefetch time with IOTLB */
1254 if (vq->iotlb)
1255 return true;
1256
1257 return vq_access_ok(vq, vq->num, vq->desc, vq->avail, vq->used);
1258}
1259EXPORT_SYMBOL_GPL(vhost_vq_access_ok);
1260
1261static struct vhost_umem *vhost_umem_alloc(void)
1262{
1263 struct vhost_umem *umem = kvzalloc(sizeof(*umem), GFP_KERNEL);
1264
1265 if (!umem)
1266 return NULL;
1267
1268 umem->umem_tree = RB_ROOT_CACHED;
1269 umem->numem = 0;
1270 INIT_LIST_HEAD(&umem->umem_list);
1271
1272 return umem;
1273}
1274
1275static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m)
1276{
1277 struct vhost_memory mem, *newmem;
1278 struct vhost_memory_region *region;
1279 struct vhost_umem *newumem, *oldumem;
1280 unsigned long size = offsetof(struct vhost_memory, regions);
1281 int i;
1282
1283 if (copy_from_user(&mem, m, size))
1284 return -EFAULT;
1285 if (mem.padding)
1286 return -EOPNOTSUPP;
1287 if (mem.nregions > max_mem_regions)
1288 return -E2BIG;
1289 newmem = kvzalloc(size + mem.nregions * sizeof(*m->regions), GFP_KERNEL);
1290 if (!newmem)
1291 return -ENOMEM;
1292
1293 memcpy(newmem, &mem, size);
1294 if (copy_from_user(newmem->regions, m->regions,
1295 mem.nregions * sizeof *m->regions)) {
1296 kvfree(newmem);
1297 return -EFAULT;
1298 }
1299
1300 newumem = vhost_umem_alloc();
1301 if (!newumem) {
1302 kvfree(newmem);
1303 return -ENOMEM;
1304 }
1305
1306 for (region = newmem->regions;
1307 region < newmem->regions + mem.nregions;
1308 region++) {
1309 if (vhost_new_umem_range(newumem,
1310 region->guest_phys_addr,
1311 region->memory_size,
1312 region->guest_phys_addr +
1313 region->memory_size - 1,
1314 region->userspace_addr,
1315 VHOST_ACCESS_RW))
1316 goto err;
1317 }
1318
1319 if (!memory_access_ok(d, newumem, 0))
1320 goto err;
1321
1322 oldumem = d->umem;
1323 d->umem = newumem;
1324
1325 /* All memory accesses are done under some VQ mutex. */
1326 for (i = 0; i < d->nvqs; ++i) {
1327 mutex_lock(&d->vqs[i]->mutex);
1328 d->vqs[i]->umem = newumem;
1329 mutex_unlock(&d->vqs[i]->mutex);
1330 }
1331
1332 kvfree(newmem);
1333 vhost_umem_clean(oldumem);
1334 return 0;
1335
1336err:
1337 vhost_umem_clean(newumem);
1338 kvfree(newmem);
1339 return -EFAULT;
1340}
1341
1342long vhost_vring_ioctl(struct vhost_dev *d, unsigned int ioctl, void __user *argp)
1343{
1344 struct file *eventfp, *filep = NULL;
1345 bool pollstart = false, pollstop = false;
1346 struct eventfd_ctx *ctx = NULL;
1347 u32 __user *idxp = argp;
1348 struct vhost_virtqueue *vq;
1349 struct vhost_vring_state s;
1350 struct vhost_vring_file f;
1351 struct vhost_vring_addr a;
1352 u32 idx;
1353 long r;
1354
1355 r = get_user(idx, idxp);
1356 if (r < 0)
1357 return r;
1358 if (idx >= d->nvqs)
1359 return -ENOBUFS;
1360
1361 vq = d->vqs[idx];
1362
1363 mutex_lock(&vq->mutex);
1364
1365 switch (ioctl) {
1366 case VHOST_SET_VRING_NUM:
1367 /* Resizing ring with an active backend?
1368 * You don't want to do that. */
1369 if (vq->private_data) {
1370 r = -EBUSY;
1371 break;
1372 }
1373 if (copy_from_user(&s, argp, sizeof s)) {
1374 r = -EFAULT;
1375 break;
1376 }
1377 if (!s.num || s.num > 0xffff || (s.num & (s.num - 1))) {
1378 r = -EINVAL;
1379 break;
1380 }
1381 vq->num = s.num;
1382 break;
1383 case VHOST_SET_VRING_BASE:
1384 /* Moving base with an active backend?
1385 * You don't want to do that. */
1386 if (vq->private_data) {
1387 r = -EBUSY;
1388 break;
1389 }
1390 if (copy_from_user(&s, argp, sizeof s)) {
1391 r = -EFAULT;
1392 break;
1393 }
1394 if (s.num > 0xffff) {
1395 r = -EINVAL;
1396 break;
1397 }
1398 vq->last_avail_idx = s.num;
1399 /* Forget the cached index value. */
1400 vq->avail_idx = vq->last_avail_idx;
1401 break;
1402 case VHOST_GET_VRING_BASE:
1403 s.index = idx;
1404 s.num = vq->last_avail_idx;
1405 if (copy_to_user(argp, &s, sizeof s))
1406 r = -EFAULT;
1407 break;
1408 case VHOST_SET_VRING_ADDR:
1409 if (copy_from_user(&a, argp, sizeof a)) {
1410 r = -EFAULT;
1411 break;
1412 }
1413 if (a.flags & ~(0x1 << VHOST_VRING_F_LOG)) {
1414 r = -EOPNOTSUPP;
1415 break;
1416 }
1417 /* For 32bit, verify that the top 32bits of the user
1418 data are set to zero. */
1419 if ((u64)(unsigned long)a.desc_user_addr != a.desc_user_addr ||
1420 (u64)(unsigned long)a.used_user_addr != a.used_user_addr ||
1421 (u64)(unsigned long)a.avail_user_addr != a.avail_user_addr) {
1422 r = -EFAULT;
1423 break;
1424 }
1425
1426 /* Make sure it's safe to cast pointers to vring types. */
1427 BUILD_BUG_ON(__alignof__ *vq->avail > VRING_AVAIL_ALIGN_SIZE);
1428 BUILD_BUG_ON(__alignof__ *vq->used > VRING_USED_ALIGN_SIZE);
1429 if ((a.avail_user_addr & (VRING_AVAIL_ALIGN_SIZE - 1)) ||
1430 (a.used_user_addr & (VRING_USED_ALIGN_SIZE - 1)) ||
1431 (a.log_guest_addr & (VRING_USED_ALIGN_SIZE - 1))) {
1432 r = -EINVAL;
1433 break;
1434 }
1435
1436 /* We only verify access here if backend is configured.
1437 * If it is not, we don't as size might not have been setup.
1438 * We will verify when backend is configured. */
1439 if (vq->private_data) {
1440 if (!vq_access_ok(vq, vq->num,
1441 (void __user *)(unsigned long)a.desc_user_addr,
1442 (void __user *)(unsigned long)a.avail_user_addr,
1443 (void __user *)(unsigned long)a.used_user_addr)) {
1444 r = -EINVAL;
1445 break;
1446 }
1447
1448 /* Also validate log access for used ring if enabled. */
1449 if ((a.flags & (0x1 << VHOST_VRING_F_LOG)) &&
1450 !log_access_ok(vq->log_base, a.log_guest_addr,
1451 sizeof *vq->used +
1452 vq->num * sizeof *vq->used->ring)) {
1453 r = -EINVAL;
1454 break;
1455 }
1456 }
1457
1458 vq->log_used = !!(a.flags & (0x1 << VHOST_VRING_F_LOG));
1459 vq->desc = (void __user *)(unsigned long)a.desc_user_addr;
1460 vq->avail = (void __user *)(unsigned long)a.avail_user_addr;
1461 vq->log_addr = a.log_guest_addr;
1462 vq->used = (void __user *)(unsigned long)a.used_user_addr;
1463 break;
1464 case VHOST_SET_VRING_KICK:
1465 if (copy_from_user(&f, argp, sizeof f)) {
1466 r = -EFAULT;
1467 break;
1468 }
1469 eventfp = f.fd == -1 ? NULL : eventfd_fget(f.fd);
1470 if (IS_ERR(eventfp)) {
1471 r = PTR_ERR(eventfp);
1472 break;
1473 }
1474 if (eventfp != vq->kick) {
1475 pollstop = (filep = vq->kick) != NULL;
1476 pollstart = (vq->kick = eventfp) != NULL;
1477 } else
1478 filep = eventfp;
1479 break;
1480 case VHOST_SET_VRING_CALL:
1481 if (copy_from_user(&f, argp, sizeof f)) {
1482 r = -EFAULT;
1483 break;
1484 }
1485 ctx = f.fd == -1 ? NULL : eventfd_ctx_fdget(f.fd);
1486 if (IS_ERR(ctx)) {
1487 r = PTR_ERR(ctx);
1488 break;
1489 }
1490 swap(ctx, vq->call_ctx);
1491 break;
1492 case VHOST_SET_VRING_ERR:
1493 if (copy_from_user(&f, argp, sizeof f)) {
1494 r = -EFAULT;
1495 break;
1496 }
1497 ctx = f.fd == -1 ? NULL : eventfd_ctx_fdget(f.fd);
1498 if (IS_ERR(ctx)) {
1499 r = PTR_ERR(ctx);
1500 break;
1501 }
1502 swap(ctx, vq->error_ctx);
1503 break;
1504 case VHOST_SET_VRING_ENDIAN:
1505 r = vhost_set_vring_endian(vq, argp);
1506 break;
1507 case VHOST_GET_VRING_ENDIAN:
1508 r = vhost_get_vring_endian(vq, idx, argp);
1509 break;
1510 case VHOST_SET_VRING_BUSYLOOP_TIMEOUT:
1511 if (copy_from_user(&s, argp, sizeof(s))) {
1512 r = -EFAULT;
1513 break;
1514 }
1515 vq->busyloop_timeout = s.num;
1516 break;
1517 case VHOST_GET_VRING_BUSYLOOP_TIMEOUT:
1518 s.index = idx;
1519 s.num = vq->busyloop_timeout;
1520 if (copy_to_user(argp, &s, sizeof(s)))
1521 r = -EFAULT;
1522 break;
1523 default:
1524 r = -ENOIOCTLCMD;
1525 }
1526
1527 if (pollstop && vq->handle_kick)
1528 vhost_poll_stop(&vq->poll);
1529
1530 if (!IS_ERR_OR_NULL(ctx))
1531 eventfd_ctx_put(ctx);
1532 if (filep)
1533 fput(filep);
1534
1535 if (pollstart && vq->handle_kick)
1536 r = vhost_poll_start(&vq->poll, vq->kick);
1537
1538 mutex_unlock(&vq->mutex);
1539
1540 if (pollstop && vq->handle_kick)
1541 vhost_poll_flush(&vq->poll);
1542 return r;
1543}
1544EXPORT_SYMBOL_GPL(vhost_vring_ioctl);
1545
1546int vhost_init_device_iotlb(struct vhost_dev *d, bool enabled)
1547{
1548 struct vhost_umem *niotlb, *oiotlb;
1549 int i;
1550
1551 niotlb = vhost_umem_alloc();
1552 if (!niotlb)
1553 return -ENOMEM;
1554
1555 oiotlb = d->iotlb;
1556 d->iotlb = niotlb;
1557
1558 for (i = 0; i < d->nvqs; ++i) {
1559 mutex_lock(&d->vqs[i]->mutex);
1560 d->vqs[i]->iotlb = niotlb;
1561 mutex_unlock(&d->vqs[i]->mutex);
1562 }
1563
1564 vhost_umem_clean(oiotlb);
1565
1566 return 0;
1567}
1568EXPORT_SYMBOL_GPL(vhost_init_device_iotlb);
1569
1570/* Caller must have device mutex */
1571long vhost_dev_ioctl(struct vhost_dev *d, unsigned int ioctl, void __user *argp)
1572{
1573 struct eventfd_ctx *ctx;
1574 u64 p;
1575 long r;
1576 int i, fd;
1577
1578 /* If you are not the owner, you can become one */
1579 if (ioctl == VHOST_SET_OWNER) {
1580 r = vhost_dev_set_owner(d);
1581 goto done;
1582 }
1583
1584 /* You must be the owner to do anything else */
1585 r = vhost_dev_check_owner(d);
1586 if (r)
1587 goto done;
1588
1589 switch (ioctl) {
1590 case VHOST_SET_MEM_TABLE:
1591 r = vhost_set_memory(d, argp);
1592 break;
1593 case VHOST_SET_LOG_BASE:
1594 if (copy_from_user(&p, argp, sizeof p)) {
1595 r = -EFAULT;
1596 break;
1597 }
1598 if ((u64)(unsigned long)p != p) {
1599 r = -EFAULT;
1600 break;
1601 }
1602 for (i = 0; i < d->nvqs; ++i) {
1603 struct vhost_virtqueue *vq;
1604 void __user *base = (void __user *)(unsigned long)p;
1605 vq = d->vqs[i];
1606 mutex_lock(&vq->mutex);
1607 /* If ring is inactive, will check when it's enabled. */
1608 if (vq->private_data && !vq_log_access_ok(vq, base))
1609 r = -EFAULT;
1610 else
1611 vq->log_base = base;
1612 mutex_unlock(&vq->mutex);
1613 }
1614 break;
1615 case VHOST_SET_LOG_FD:
1616 r = get_user(fd, (int __user *)argp);
1617 if (r < 0)
1618 break;
1619 ctx = fd == -1 ? NULL : eventfd_ctx_fdget(fd);
1620 if (IS_ERR(ctx)) {
1621 r = PTR_ERR(ctx);
1622 break;
1623 }
1624 swap(ctx, d->log_ctx);
1625 for (i = 0; i < d->nvqs; ++i) {
1626 mutex_lock(&d->vqs[i]->mutex);
1627 d->vqs[i]->log_ctx = d->log_ctx;
1628 mutex_unlock(&d->vqs[i]->mutex);
1629 }
1630 if (ctx)
1631 eventfd_ctx_put(ctx);
1632 break;
1633 default:
1634 r = -ENOIOCTLCMD;
1635 break;
1636 }
1637done:
1638 return r;
1639}
1640EXPORT_SYMBOL_GPL(vhost_dev_ioctl);
1641
1642/* TODO: This is really inefficient. We need something like get_user()
1643 * (instruction directly accesses the data, with an exception table entry
1644 * returning -EFAULT). See Documentation/x86/exception-tables.txt.
1645 */
1646static int set_bit_to_user(int nr, void __user *addr)
1647{
1648 unsigned long log = (unsigned long)addr;
1649 struct page *page;
1650 void *base;
1651 int bit = nr + (log % PAGE_SIZE) * 8;
1652 int r;
1653
1654 r = get_user_pages_fast(log, 1, 1, &page);
1655 if (r < 0)
1656 return r;
1657 BUG_ON(r != 1);
1658 base = kmap_atomic(page);
1659 set_bit(bit, base);
1660 kunmap_atomic(base);
1661 set_page_dirty_lock(page);
1662 put_page(page);
1663 return 0;
1664}
1665
1666static int log_write(void __user *log_base,
1667 u64 write_address, u64 write_length)
1668{
1669 u64 write_page = write_address / VHOST_PAGE_SIZE;
1670 int r;
1671
1672 if (!write_length)
1673 return 0;
1674 write_length += write_address % VHOST_PAGE_SIZE;
1675 for (;;) {
1676 u64 base = (u64)(unsigned long)log_base;
1677 u64 log = base + write_page / 8;
1678 int bit = write_page % 8;
1679 if ((u64)(unsigned long)log != log)
1680 return -EFAULT;
1681 r = set_bit_to_user(bit, (void __user *)(unsigned long)log);
1682 if (r < 0)
1683 return r;
1684 if (write_length <= VHOST_PAGE_SIZE)
1685 break;
1686 write_length -= VHOST_PAGE_SIZE;
1687 write_page += 1;
1688 }
1689 return r;
1690}
1691
1692int vhost_log_write(struct vhost_virtqueue *vq, struct vhost_log *log,
1693 unsigned int log_num, u64 len)
1694{
1695 int i, r;
1696
1697 /* Make sure data written is seen before log. */
1698 smp_wmb();
1699 for (i = 0; i < log_num; ++i) {
1700 u64 l = min(log[i].len, len);
1701 r = log_write(vq->log_base, log[i].addr, l);
1702 if (r < 0)
1703 return r;
1704 len -= l;
1705 if (!len) {
1706 if (vq->log_ctx)
1707 eventfd_signal(vq->log_ctx, 1);
1708 return 0;
1709 }
1710 }
1711 /* Length written exceeds what we have stored. This is a bug. */
1712 BUG();
1713 return 0;
1714}
1715EXPORT_SYMBOL_GPL(vhost_log_write);
1716
1717static int vhost_update_used_flags(struct vhost_virtqueue *vq)
1718{
1719 void __user *used;
1720 if (vhost_put_user(vq, cpu_to_vhost16(vq, vq->used_flags),
1721 &vq->used->flags) < 0)
1722 return -EFAULT;
1723 if (unlikely(vq->log_used)) {
1724 /* Make sure the flag is seen before log. */
1725 smp_wmb();
1726 /* Log used flag write. */
1727 used = &vq->used->flags;
1728 log_write(vq->log_base, vq->log_addr +
1729 (used - (void __user *)vq->used),
1730 sizeof vq->used->flags);
1731 if (vq->log_ctx)
1732 eventfd_signal(vq->log_ctx, 1);
1733 }
1734 return 0;
1735}
1736
1737static int vhost_update_avail_event(struct vhost_virtqueue *vq, u16 avail_event)
1738{
1739 if (vhost_put_user(vq, cpu_to_vhost16(vq, vq->avail_idx),
1740 vhost_avail_event(vq)))
1741 return -EFAULT;
1742 if (unlikely(vq->log_used)) {
1743 void __user *used;
1744 /* Make sure the event is seen before log. */
1745 smp_wmb();
1746 /* Log avail event write */
1747 used = vhost_avail_event(vq);
1748 log_write(vq->log_base, vq->log_addr +
1749 (used - (void __user *)vq->used),
1750 sizeof *vhost_avail_event(vq));
1751 if (vq->log_ctx)
1752 eventfd_signal(vq->log_ctx, 1);
1753 }
1754 return 0;
1755}
1756
1757int vhost_vq_init_access(struct vhost_virtqueue *vq)
1758{
1759 __virtio16 last_used_idx;
1760 int r;
1761 bool is_le = vq->is_le;
1762
1763 if (!vq->private_data)
1764 return 0;
1765
1766 vhost_init_is_le(vq);
1767
1768 r = vhost_update_used_flags(vq);
1769 if (r)
1770 goto err;
1771 vq->signalled_used_valid = false;
1772 if (!vq->iotlb &&
1773 !access_ok(VERIFY_READ, &vq->used->idx, sizeof vq->used->idx)) {
1774 r = -EFAULT;
1775 goto err;
1776 }
1777 r = vhost_get_used(vq, last_used_idx, &vq->used->idx);
1778 if (r) {
1779 vq_err(vq, "Can't access used idx at %p\n",
1780 &vq->used->idx);
1781 goto err;
1782 }
1783 vq->last_used_idx = vhost16_to_cpu(vq, last_used_idx);
1784 return 0;
1785
1786err:
1787 vq->is_le = is_le;
1788 return r;
1789}
1790EXPORT_SYMBOL_GPL(vhost_vq_init_access);
1791
1792static int translate_desc(struct vhost_virtqueue *vq, u64 addr, u32 len,
1793 struct iovec iov[], int iov_size, int access)
1794{
1795 const struct vhost_umem_node *node;
1796 struct vhost_dev *dev = vq->dev;
1797 struct vhost_umem *umem = dev->iotlb ? dev->iotlb : dev->umem;
1798 struct iovec *_iov;
1799 u64 s = 0;
1800 int ret = 0;
1801
1802 while ((u64)len > s) {
1803 u64 size;
1804 if (unlikely(ret >= iov_size)) {
1805 ret = -ENOBUFS;
1806 break;
1807 }
1808
1809 node = vhost_umem_interval_tree_iter_first(&umem->umem_tree,
1810 addr, addr + len - 1);
1811 if (node == NULL || node->start > addr) {
1812 if (umem != dev->iotlb) {
1813 ret = -EFAULT;
1814 break;
1815 }
1816 ret = -EAGAIN;
1817 break;
1818 } else if (!(node->perm & access)) {
1819 ret = -EPERM;
1820 break;
1821 }
1822
1823 _iov = iov + ret;
1824 size = node->size - addr + node->start;
1825 _iov->iov_len = min((u64)len - s, size);
1826 _iov->iov_base = (void __user *)(unsigned long)
1827 (node->userspace_addr + addr - node->start);
1828 s += size;
1829 addr += size;
1830 ++ret;
1831 }
1832
1833 if (ret == -EAGAIN)
1834 vhost_iotlb_miss(vq, addr, access);
1835 return ret;
1836}
1837
1838/* Each buffer in the virtqueues is actually a chain of descriptors. This
1839 * function returns the next descriptor in the chain,
1840 * or -1U if we're at the end. */
1841static unsigned next_desc(struct vhost_virtqueue *vq, struct vring_desc *desc)
1842{
1843 unsigned int next;
1844
1845 /* If this descriptor says it doesn't chain, we're done. */
1846 if (!(desc->flags & cpu_to_vhost16(vq, VRING_DESC_F_NEXT)))
1847 return -1U;
1848
1849 /* Check they're not leading us off end of descriptors. */
1850 next = vhost16_to_cpu(vq, READ_ONCE(desc->next));
1851 return next;
1852}
1853
1854static int get_indirect(struct vhost_virtqueue *vq,
1855 struct iovec iov[], unsigned int iov_size,
1856 unsigned int *out_num, unsigned int *in_num,
1857 struct vhost_log *log, unsigned int *log_num,
1858 struct vring_desc *indirect)
1859{
1860 struct vring_desc desc;
1861 unsigned int i = 0, count, found = 0;
1862 u32 len = vhost32_to_cpu(vq, indirect->len);
1863 struct iov_iter from;
1864 int ret, access;
1865
1866 /* Sanity check */
1867 if (unlikely(len % sizeof desc)) {
1868 vq_err(vq, "Invalid length in indirect descriptor: "
1869 "len 0x%llx not multiple of 0x%zx\n",
1870 (unsigned long long)len,
1871 sizeof desc);
1872 return -EINVAL;
1873 }
1874
1875 ret = translate_desc(vq, vhost64_to_cpu(vq, indirect->addr), len, vq->indirect,
1876 UIO_MAXIOV, VHOST_ACCESS_RO);
1877 if (unlikely(ret < 0)) {
1878 if (ret != -EAGAIN)
1879 vq_err(vq, "Translation failure %d in indirect.\n", ret);
1880 return ret;
1881 }
1882 iov_iter_init(&from, READ, vq->indirect, ret, len);
1883
1884 /* We will use the result as an address to read from, so most
1885 * architectures only need a compiler barrier here. */
1886 read_barrier_depends();
1887
1888 count = len / sizeof desc;
1889 /* Buffers are chained via a 16 bit next field, so
1890 * we can have at most 2^16 of these. */
1891 if (unlikely(count > USHRT_MAX + 1)) {
1892 vq_err(vq, "Indirect buffer length too big: %d\n",
1893 indirect->len);
1894 return -E2BIG;
1895 }
1896
1897 do {
1898 unsigned iov_count = *in_num + *out_num;
1899 if (unlikely(++found > count)) {
1900 vq_err(vq, "Loop detected: last one at %u "
1901 "indirect size %u\n",
1902 i, count);
1903 return -EINVAL;
1904 }
1905 if (unlikely(!copy_from_iter_full(&desc, sizeof(desc), &from))) {
1906 vq_err(vq, "Failed indirect descriptor: idx %d, %zx\n",
1907 i, (size_t)vhost64_to_cpu(vq, indirect->addr) + i * sizeof desc);
1908 return -EINVAL;
1909 }
1910 if (unlikely(desc.flags & cpu_to_vhost16(vq, VRING_DESC_F_INDIRECT))) {
1911 vq_err(vq, "Nested indirect descriptor: idx %d, %zx\n",
1912 i, (size_t)vhost64_to_cpu(vq, indirect->addr) + i * sizeof desc);
1913 return -EINVAL;
1914 }
1915
1916 if (desc.flags & cpu_to_vhost16(vq, VRING_DESC_F_WRITE))
1917 access = VHOST_ACCESS_WO;
1918 else
1919 access = VHOST_ACCESS_RO;
1920
1921 ret = translate_desc(vq, vhost64_to_cpu(vq, desc.addr),
1922 vhost32_to_cpu(vq, desc.len), iov + iov_count,
1923 iov_size - iov_count, access);
1924 if (unlikely(ret < 0)) {
1925 if (ret != -EAGAIN)
1926 vq_err(vq, "Translation failure %d indirect idx %d\n",
1927 ret, i);
1928 return ret;
1929 }
1930 /* If this is an input descriptor, increment that count. */
1931 if (access == VHOST_ACCESS_WO) {
1932 *in_num += ret;
1933 if (unlikely(log)) {
1934 log[*log_num].addr = vhost64_to_cpu(vq, desc.addr);
1935 log[*log_num].len = vhost32_to_cpu(vq, desc.len);
1936 ++*log_num;
1937 }
1938 } else {
1939 /* If it's an output descriptor, they're all supposed
1940 * to come before any input descriptors. */
1941 if (unlikely(*in_num)) {
1942 vq_err(vq, "Indirect descriptor "
1943 "has out after in: idx %d\n", i);
1944 return -EINVAL;
1945 }
1946 *out_num += ret;
1947 }
1948 } while ((i = next_desc(vq, &desc)) != -1);
1949 return 0;
1950}
1951
1952/* This looks in the virtqueue and for the first available buffer, and converts
1953 * it to an iovec for convenient access. Since descriptors consist of some
1954 * number of output then some number of input descriptors, it's actually two
1955 * iovecs, but we pack them into one and note how many of each there were.
1956 *
1957 * This function returns the descriptor number found, or vq->num (which is
1958 * never a valid descriptor number) if none was found. A negative code is
1959 * returned on error. */
1960int vhost_get_vq_desc(struct vhost_virtqueue *vq,
1961 struct iovec iov[], unsigned int iov_size,
1962 unsigned int *out_num, unsigned int *in_num,
1963 struct vhost_log *log, unsigned int *log_num)
1964{
1965 struct vring_desc desc;
1966 unsigned int i, head, found = 0;
1967 u16 last_avail_idx;
1968 __virtio16 avail_idx;
1969 __virtio16 ring_head;
1970 int ret, access;
1971
1972 /* Check it isn't doing very strange things with descriptor numbers. */
1973 last_avail_idx = vq->last_avail_idx;
1974
1975 if (vq->avail_idx == vq->last_avail_idx) {
1976 if (unlikely(vhost_get_avail(vq, avail_idx, &vq->avail->idx))) {
1977 vq_err(vq, "Failed to access avail idx at %p\n",
1978 &vq->avail->idx);
1979 return -EFAULT;
1980 }
1981 vq->avail_idx = vhost16_to_cpu(vq, avail_idx);
1982
1983 if (unlikely((u16)(vq->avail_idx - last_avail_idx) > vq->num)) {
1984 vq_err(vq, "Guest moved used index from %u to %u",
1985 last_avail_idx, vq->avail_idx);
1986 return -EFAULT;
1987 }
1988
1989 /* If there's nothing new since last we looked, return
1990 * invalid.
1991 */
1992 if (vq->avail_idx == last_avail_idx)
1993 return vq->num;
1994
1995 /* Only get avail ring entries after they have been
1996 * exposed by guest.
1997 */
1998 smp_rmb();
1999 }
2000
2001 /* Grab the next descriptor number they're advertising, and increment
2002 * the index we've seen. */
2003 if (unlikely(vhost_get_avail(vq, ring_head,
2004 &vq->avail->ring[last_avail_idx & (vq->num - 1)]))) {
2005 vq_err(vq, "Failed to read head: idx %d address %p\n",
2006 last_avail_idx,
2007 &vq->avail->ring[last_avail_idx % vq->num]);
2008 return -EFAULT;
2009 }
2010
2011 head = vhost16_to_cpu(vq, ring_head);
2012
2013 /* If their number is silly, that's an error. */
2014 if (unlikely(head >= vq->num)) {
2015 vq_err(vq, "Guest says index %u > %u is available",
2016 head, vq->num);
2017 return -EINVAL;
2018 }
2019
2020 /* When we start there are none of either input nor output. */
2021 *out_num = *in_num = 0;
2022 if (unlikely(log))
2023 *log_num = 0;
2024
2025 i = head;
2026 do {
2027 unsigned iov_count = *in_num + *out_num;
2028 if (unlikely(i >= vq->num)) {
2029 vq_err(vq, "Desc index is %u > %u, head = %u",
2030 i, vq->num, head);
2031 return -EINVAL;
2032 }
2033 if (unlikely(++found > vq->num)) {
2034 vq_err(vq, "Loop detected: last one at %u "
2035 "vq size %u head %u\n",
2036 i, vq->num, head);
2037 return -EINVAL;
2038 }
2039 ret = vhost_copy_from_user(vq, &desc, vq->desc + i,
2040 sizeof desc);
2041 if (unlikely(ret)) {
2042 vq_err(vq, "Failed to get descriptor: idx %d addr %p\n",
2043 i, vq->desc + i);
2044 return -EFAULT;
2045 }
2046 if (desc.flags & cpu_to_vhost16(vq, VRING_DESC_F_INDIRECT)) {
2047 ret = get_indirect(vq, iov, iov_size,
2048 out_num, in_num,
2049 log, log_num, &desc);
2050 if (unlikely(ret < 0)) {
2051 if (ret != -EAGAIN)
2052 vq_err(vq, "Failure detected "
2053 "in indirect descriptor at idx %d\n", i);
2054 return ret;
2055 }
2056 continue;
2057 }
2058
2059 if (desc.flags & cpu_to_vhost16(vq, VRING_DESC_F_WRITE))
2060 access = VHOST_ACCESS_WO;
2061 else
2062 access = VHOST_ACCESS_RO;
2063 ret = translate_desc(vq, vhost64_to_cpu(vq, desc.addr),
2064 vhost32_to_cpu(vq, desc.len), iov + iov_count,
2065 iov_size - iov_count, access);
2066 if (unlikely(ret < 0)) {
2067 if (ret != -EAGAIN)
2068 vq_err(vq, "Translation failure %d descriptor idx %d\n",
2069 ret, i);
2070 return ret;
2071 }
2072 if (access == VHOST_ACCESS_WO) {
2073 /* If this is an input descriptor,
2074 * increment that count. */
2075 *in_num += ret;
2076 if (unlikely(log)) {
2077 log[*log_num].addr = vhost64_to_cpu(vq, desc.addr);
2078 log[*log_num].len = vhost32_to_cpu(vq, desc.len);
2079 ++*log_num;
2080 }
2081 } else {
2082 /* If it's an output descriptor, they're all supposed
2083 * to come before any input descriptors. */
2084 if (unlikely(*in_num)) {
2085 vq_err(vq, "Descriptor has out after in: "
2086 "idx %d\n", i);
2087 return -EINVAL;
2088 }
2089 *out_num += ret;
2090 }
2091 } while ((i = next_desc(vq, &desc)) != -1);
2092
2093 /* On success, increment avail index. */
2094 vq->last_avail_idx++;
2095
2096 /* Assume notifications from guest are disabled at this point,
2097 * if they aren't we would need to update avail_event index. */
2098 BUG_ON(!(vq->used_flags & VRING_USED_F_NO_NOTIFY));
2099 return head;
2100}
2101EXPORT_SYMBOL_GPL(vhost_get_vq_desc);
2102
2103/* Reverse the effect of vhost_get_vq_desc. Useful for error handling. */
2104void vhost_discard_vq_desc(struct vhost_virtqueue *vq, int n)
2105{
2106 vq->last_avail_idx -= n;
2107}
2108EXPORT_SYMBOL_GPL(vhost_discard_vq_desc);
2109
2110/* After we've used one of their buffers, we tell them about it. We'll then
2111 * want to notify the guest, using eventfd. */
2112int vhost_add_used(struct vhost_virtqueue *vq, unsigned int head, int len)
2113{
2114 struct vring_used_elem heads = {
2115 cpu_to_vhost32(vq, head),
2116 cpu_to_vhost32(vq, len)
2117 };
2118
2119 return vhost_add_used_n(vq, &heads, 1);
2120}
2121EXPORT_SYMBOL_GPL(vhost_add_used);
2122
2123static int __vhost_add_used_n(struct vhost_virtqueue *vq,
2124 struct vring_used_elem *heads,
2125 unsigned count)
2126{
2127 struct vring_used_elem __user *used;
2128 u16 old, new;
2129 int start;
2130
2131 start = vq->last_used_idx & (vq->num - 1);
2132 used = vq->used->ring + start;
2133 if (count == 1) {
2134 if (vhost_put_user(vq, heads[0].id, &used->id)) {
2135 vq_err(vq, "Failed to write used id");
2136 return -EFAULT;
2137 }
2138 if (vhost_put_user(vq, heads[0].len, &used->len)) {
2139 vq_err(vq, "Failed to write used len");
2140 return -EFAULT;
2141 }
2142 } else if (vhost_copy_to_user(vq, used, heads, count * sizeof *used)) {
2143 vq_err(vq, "Failed to write used");
2144 return -EFAULT;
2145 }
2146 if (unlikely(vq->log_used)) {
2147 /* Make sure data is seen before log. */
2148 smp_wmb();
2149 /* Log used ring entry write. */
2150 log_write(vq->log_base,
2151 vq->log_addr +
2152 ((void __user *)used - (void __user *)vq->used),
2153 count * sizeof *used);
2154 }
2155 old = vq->last_used_idx;
2156 new = (vq->last_used_idx += count);
2157 /* If the driver never bothers to signal in a very long while,
2158 * used index might wrap around. If that happens, invalidate
2159 * signalled_used index we stored. TODO: make sure driver
2160 * signals at least once in 2^16 and remove this. */
2161 if (unlikely((u16)(new - vq->signalled_used) < (u16)(new - old)))
2162 vq->signalled_used_valid = false;
2163 return 0;
2164}
2165
2166/* After we've used one of their buffers, we tell them about it. We'll then
2167 * want to notify the guest, using eventfd. */
2168int vhost_add_used_n(struct vhost_virtqueue *vq, struct vring_used_elem *heads,
2169 unsigned count)
2170{
2171 int start, n, r;
2172
2173 start = vq->last_used_idx & (vq->num - 1);
2174 n = vq->num - start;
2175 if (n < count) {
2176 r = __vhost_add_used_n(vq, heads, n);
2177 if (r < 0)
2178 return r;
2179 heads += n;
2180 count -= n;
2181 }
2182 r = __vhost_add_used_n(vq, heads, count);
2183
2184 /* Make sure buffer is written before we update index. */
2185 smp_wmb();
2186 if (vhost_put_user(vq, cpu_to_vhost16(vq, vq->last_used_idx),
2187 &vq->used->idx)) {
2188 vq_err(vq, "Failed to increment used idx");
2189 return -EFAULT;
2190 }
2191 if (unlikely(vq->log_used)) {
2192 /* Log used index update. */
2193 log_write(vq->log_base,
2194 vq->log_addr + offsetof(struct vring_used, idx),
2195 sizeof vq->used->idx);
2196 if (vq->log_ctx)
2197 eventfd_signal(vq->log_ctx, 1);
2198 }
2199 return r;
2200}
2201EXPORT_SYMBOL_GPL(vhost_add_used_n);
2202
2203static bool vhost_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
2204{
2205 __u16 old, new;
2206 __virtio16 event;
2207 bool v;
2208 /* Flush out used index updates. This is paired
2209 * with the barrier that the Guest executes when enabling
2210 * interrupts. */
2211 smp_mb();
2212
2213 if (vhost_has_feature(vq, VIRTIO_F_NOTIFY_ON_EMPTY) &&
2214 unlikely(vq->avail_idx == vq->last_avail_idx))
2215 return true;
2216
2217 if (!vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX)) {
2218 __virtio16 flags;
2219 if (vhost_get_avail(vq, flags, &vq->avail->flags)) {
2220 vq_err(vq, "Failed to get flags");
2221 return true;
2222 }
2223 return !(flags & cpu_to_vhost16(vq, VRING_AVAIL_F_NO_INTERRUPT));
2224 }
2225 old = vq->signalled_used;
2226 v = vq->signalled_used_valid;
2227 new = vq->signalled_used = vq->last_used_idx;
2228 vq->signalled_used_valid = true;
2229
2230 if (unlikely(!v))
2231 return true;
2232
2233 if (vhost_get_avail(vq, event, vhost_used_event(vq))) {
2234 vq_err(vq, "Failed to get used event idx");
2235 return true;
2236 }
2237 return vring_need_event(vhost16_to_cpu(vq, event), new, old);
2238}
2239
2240/* This actually signals the guest, using eventfd. */
2241void vhost_signal(struct vhost_dev *dev, struct vhost_virtqueue *vq)
2242{
2243 /* Signal the Guest tell them we used something up. */
2244 if (vq->call_ctx && vhost_notify(dev, vq))
2245 eventfd_signal(vq->call_ctx, 1);
2246}
2247EXPORT_SYMBOL_GPL(vhost_signal);
2248
2249/* And here's the combo meal deal. Supersize me! */
2250void vhost_add_used_and_signal(struct vhost_dev *dev,
2251 struct vhost_virtqueue *vq,
2252 unsigned int head, int len)
2253{
2254 vhost_add_used(vq, head, len);
2255 vhost_signal(dev, vq);
2256}
2257EXPORT_SYMBOL_GPL(vhost_add_used_and_signal);
2258
2259/* multi-buffer version of vhost_add_used_and_signal */
2260void vhost_add_used_and_signal_n(struct vhost_dev *dev,
2261 struct vhost_virtqueue *vq,
2262 struct vring_used_elem *heads, unsigned count)
2263{
2264 vhost_add_used_n(vq, heads, count);
2265 vhost_signal(dev, vq);
2266}
2267EXPORT_SYMBOL_GPL(vhost_add_used_and_signal_n);
2268
2269/* return true if we're sure that avaiable ring is empty */
2270bool vhost_vq_avail_empty(struct vhost_dev *dev, struct vhost_virtqueue *vq)
2271{
2272 __virtio16 avail_idx;
2273 int r;
2274
2275 if (vq->avail_idx != vq->last_avail_idx)
2276 return false;
2277
2278 r = vhost_get_avail(vq, avail_idx, &vq->avail->idx);
2279 if (unlikely(r))
2280 return false;
2281 vq->avail_idx = vhost16_to_cpu(vq, avail_idx);
2282
2283 return vq->avail_idx == vq->last_avail_idx;
2284}
2285EXPORT_SYMBOL_GPL(vhost_vq_avail_empty);
2286
2287/* OK, now we need to know about added descriptors. */
2288bool vhost_enable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
2289{
2290 __virtio16 avail_idx;
2291 int r;
2292
2293 if (!(vq->used_flags & VRING_USED_F_NO_NOTIFY))
2294 return false;
2295 vq->used_flags &= ~VRING_USED_F_NO_NOTIFY;
2296 if (!vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX)) {
2297 r = vhost_update_used_flags(vq);
2298 if (r) {
2299 vq_err(vq, "Failed to enable notification at %p: %d\n",
2300 &vq->used->flags, r);
2301 return false;
2302 }
2303 } else {
2304 r = vhost_update_avail_event(vq, vq->avail_idx);
2305 if (r) {
2306 vq_err(vq, "Failed to update avail event index at %p: %d\n",
2307 vhost_avail_event(vq), r);
2308 return false;
2309 }
2310 }
2311 /* They could have slipped one in as we were doing that: make
2312 * sure it's written, then check again. */
2313 smp_mb();
2314 r = vhost_get_avail(vq, avail_idx, &vq->avail->idx);
2315 if (r) {
2316 vq_err(vq, "Failed to check avail idx at %p: %d\n",
2317 &vq->avail->idx, r);
2318 return false;
2319 }
2320
2321 return vhost16_to_cpu(vq, avail_idx) != vq->avail_idx;
2322}
2323EXPORT_SYMBOL_GPL(vhost_enable_notify);
2324
2325/* We don't need to be notified again. */
2326void vhost_disable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
2327{
2328 int r;
2329
2330 if (vq->used_flags & VRING_USED_F_NO_NOTIFY)
2331 return;
2332 vq->used_flags |= VRING_USED_F_NO_NOTIFY;
2333 if (!vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX)) {
2334 r = vhost_update_used_flags(vq);
2335 if (r)
2336 vq_err(vq, "Failed to enable notification at %p: %d\n",
2337 &vq->used->flags, r);
2338 }
2339}
2340EXPORT_SYMBOL_GPL(vhost_disable_notify);
2341
2342/* Create a new message. */
2343struct vhost_msg_node *vhost_new_msg(struct vhost_virtqueue *vq, int type)
2344{
2345 struct vhost_msg_node *node = kmalloc(sizeof *node, GFP_KERNEL);
2346 if (!node)
2347 return NULL;
2348 node->vq = vq;
2349 node->msg.type = type;
2350 return node;
2351}
2352EXPORT_SYMBOL_GPL(vhost_new_msg);
2353
2354void vhost_enqueue_msg(struct vhost_dev *dev, struct list_head *head,
2355 struct vhost_msg_node *node)
2356{
2357 spin_lock(&dev->iotlb_lock);
2358 list_add_tail(&node->node, head);
2359 spin_unlock(&dev->iotlb_lock);
2360
2361 wake_up_interruptible_poll(&dev->wait, EPOLLIN | EPOLLRDNORM);
2362}
2363EXPORT_SYMBOL_GPL(vhost_enqueue_msg);
2364
2365struct vhost_msg_node *vhost_dequeue_msg(struct vhost_dev *dev,
2366 struct list_head *head)
2367{
2368 struct vhost_msg_node *node = NULL;
2369
2370 spin_lock(&dev->iotlb_lock);
2371 if (!list_empty(head)) {
2372 node = list_first_entry(head, struct vhost_msg_node,
2373 node);
2374 list_del(&node->node);
2375 }
2376 spin_unlock(&dev->iotlb_lock);
2377
2378 return node;
2379}
2380EXPORT_SYMBOL_GPL(vhost_dequeue_msg);
2381
2382
2383static int __init vhost_init(void)
2384{
2385 return 0;
2386}
2387
2388static void __exit vhost_exit(void)
2389{
2390}
2391
2392module_init(vhost_init);
2393module_exit(vhost_exit);
2394
2395MODULE_VERSION("0.0.1");
2396MODULE_LICENSE("GPL v2");
2397MODULE_AUTHOR("Michael S. Tsirkin");
2398MODULE_DESCRIPTION("Host kernel accelerator for virtio");