aboutsummaryrefslogtreecommitdiff
path: root/hw/block/dataplane/virtio-blk.c
blob: f2d7350a503dd5059db2a61fff4f8da993cdbb52 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
/*
 * Dedicated thread for virtio-blk I/O processing
 *
 * Copyright 2012 IBM, Corp.
 * Copyright 2012 Red Hat, Inc. and/or its affiliates
 *
 * Authors:
 *   Stefan Hajnoczi <stefanha@redhat.com>
 *
 * This work is licensed under the terms of the GNU GPL, version 2 or later.
 * See the COPYING file in the top-level directory.
 *
 */

#include "trace.h"
#include "qemu/iov.h"
#include "qemu/thread.h"
#include "qemu/error-report.h"
#include "hw/virtio/dataplane/vring.h"
#include "ioq.h"
#include "block/block.h"
#include "hw/virtio/virtio-blk.h"
#include "virtio-blk.h"
#include "block/aio.h"
#include "hw/virtio/virtio-bus.h"

enum {
    SEG_MAX = 126,                  /* maximum number of I/O segments */
    VRING_MAX = SEG_MAX + 2,        /* maximum number of vring descriptors */
    REQ_MAX = VRING_MAX,            /* maximum number of requests in the vring,
                                     * is VRING_MAX / 2 with traditional and
                                     * VRING_MAX with indirect descriptors */
};

typedef struct {
    struct iocb iocb;               /* Linux AIO control block */
    QEMUIOVector *inhdr;            /* iovecs for virtio_blk_inhdr */
    unsigned int head;              /* vring descriptor index */
    struct iovec *bounce_iov;       /* used if guest buffers are unaligned */
    QEMUIOVector *read_qiov;        /* for read completion /w bounce buffer */
} VirtIOBlockRequest;

struct VirtIOBlockDataPlane {
    bool started;
    bool starting;
    bool stopping;
    QEMUBH *start_bh;
    QemuThread thread;

    VirtIOBlkConf *blk;
    int fd;                         /* image file descriptor */

    VirtIODevice *vdev;
    Vring vring;                    /* virtqueue vring */
    EventNotifier *guest_notifier;  /* irq */

    /* Note that these EventNotifiers are assigned by value.  This is
     * fine as long as you do not call event_notifier_cleanup on them
     * (because you don't own the file descriptor or handle; you just
     * use it).
     */
    AioContext *ctx;
    EventNotifier io_notifier;      /* Linux AIO completion */
    EventNotifier host_notifier;    /* doorbell */

    IOQueue ioqueue;                /* Linux AIO queue (should really be per
                                       dataplane thread) */
    VirtIOBlockRequest requests[REQ_MAX]; /* pool of requests, managed by the
                                             queue */

    unsigned int num_reqs;
};

/* Raise an interrupt to signal guest, if necessary */
static void notify_guest(VirtIOBlockDataPlane *s)
{
    if (!vring_should_notify(s->vdev, &s->vring)) {
        return;
    }

    event_notifier_set(s->guest_notifier);
}

static void complete_request(struct iocb *iocb, ssize_t ret, void *opaque)
{
    VirtIOBlockDataPlane *s = opaque;
    VirtIOBlockRequest *req = container_of(iocb, VirtIOBlockRequest, iocb);
    struct virtio_blk_inhdr hdr;
    int len;

    if (likely(ret >= 0)) {
        hdr.status = VIRTIO_BLK_S_OK;
        len = ret;
    } else {
        hdr.status = VIRTIO_BLK_S_IOERR;
        len = 0;
    }

    trace_virtio_blk_data_plane_complete_request(s, req->head, ret);

    if (req->read_qiov) {
        assert(req->bounce_iov);
        qemu_iovec_from_buf(req->read_qiov, 0, req->bounce_iov->iov_base, len);
        qemu_iovec_destroy(req->read_qiov);
        g_slice_free(QEMUIOVector, req->read_qiov);
    }

    if (req->bounce_iov) {
        qemu_vfree(req->bounce_iov->iov_base);
        g_slice_free(struct iovec, req->bounce_iov);
    }

    qemu_iovec_from_buf(req->inhdr, 0, &hdr, sizeof(hdr));
    qemu_iovec_destroy(req->inhdr);
    g_slice_free(QEMUIOVector, req->inhdr);

    /* According to the virtio specification len should be the number of bytes
     * written to, but for virtio-blk it seems to be the number of bytes
     * transferred plus the status bytes.
     */
    vring_push(&s->vring, req->head, len + sizeof(hdr));

    s->num_reqs--;
}

static void complete_request_early(VirtIOBlockDataPlane *s, unsigned int head,
                                   QEMUIOVector *inhdr, unsigned char status)
{
    struct virtio_blk_inhdr hdr = {
        .status = status,
    };

    qemu_iovec_from_buf(inhdr, 0, &hdr, sizeof(hdr));
    qemu_iovec_destroy(inhdr);
    g_slice_free(QEMUIOVector, inhdr);

    vring_push(&s->vring, head, sizeof(hdr));
    notify_guest(s);
}

/* Get disk serial number */
static void do_get_id_cmd(VirtIOBlockDataPlane *s,
                          struct iovec *iov, unsigned int iov_cnt,
                          unsigned int head, QEMUIOVector *inhdr)
{
    char id[VIRTIO_BLK_ID_BYTES];

    /* Serial number not NUL-terminated when shorter than buffer */
    strncpy(id, s->blk->serial ? s->blk->serial : "", sizeof(id));
    iov_from_buf(iov, iov_cnt, 0, id, sizeof(id));
    complete_request_early(s, head, inhdr, VIRTIO_BLK_S_OK);
}

static int do_rdwr_cmd(VirtIOBlockDataPlane *s, bool read,
                       struct iovec *iov, unsigned int iov_cnt,
                       long long offset, unsigned int head,
                       QEMUIOVector *inhdr)
{
    struct iocb *iocb;
    QEMUIOVector qiov;
    struct iovec *bounce_iov = NULL;
    QEMUIOVector *read_qiov = NULL;

    qemu_iovec_init_external(&qiov, iov, iov_cnt);
    if (!bdrv_qiov_is_aligned(s->blk->conf.bs, &qiov)) {
        void *bounce_buffer = qemu_blockalign(s->blk->conf.bs, qiov.size);

        if (read) {
            /* Need to copy back from bounce buffer on completion */
            read_qiov = g_slice_new(QEMUIOVector);
            qemu_iovec_init(read_qiov, iov_cnt);
            qemu_iovec_concat_iov(read_qiov, iov, iov_cnt, 0, qiov.size);
        } else {
            qemu_iovec_to_buf(&qiov, 0, bounce_buffer, qiov.size);
        }

        /* Redirect I/O to aligned bounce buffer */
        bounce_iov = g_slice_new(struct iovec);
        bounce_iov->iov_base = bounce_buffer;
        bounce_iov->iov_len = qiov.size;
        iov = bounce_iov;
        iov_cnt = 1;
    }

    iocb = ioq_rdwr(&s->ioqueue, read, iov, iov_cnt, offset);

    /* Fill in virtio block metadata needed for completion */
    VirtIOBlockRequest *req = container_of(iocb, VirtIOBlockRequest, iocb);
    req->head = head;
    req->inhdr = inhdr;
    req->bounce_iov = bounce_iov;
    req->read_qiov = read_qiov;
    return 0;
}

static int process_request(IOQueue *ioq, struct iovec iov[],
                           unsigned int out_num, unsigned int in_num,
                           unsigned int head)
{
    VirtIOBlockDataPlane *s = container_of(ioq, VirtIOBlockDataPlane, ioqueue);
    struct iovec *in_iov = &iov[out_num];
    struct virtio_blk_outhdr outhdr;
    QEMUIOVector *inhdr;
    size_t in_size;

    /* Copy in outhdr */
    if (unlikely(iov_to_buf(iov, out_num, 0, &outhdr,
                            sizeof(outhdr)) != sizeof(outhdr))) {
        error_report("virtio-blk request outhdr too short");
        return -EFAULT;
    }
    iov_discard_front(&iov, &out_num, sizeof(outhdr));

    /* Grab inhdr for later */
    in_size = iov_size(in_iov, in_num);
    if (in_size < sizeof(struct virtio_blk_inhdr)) {
        error_report("virtio_blk request inhdr too short");
        return -EFAULT;
    }
    inhdr = g_slice_new(QEMUIOVector);
    qemu_iovec_init(inhdr, 1);
    qemu_iovec_concat_iov(inhdr, in_iov, in_num,
            in_size - sizeof(struct virtio_blk_inhdr),
            sizeof(struct virtio_blk_inhdr));
    iov_discard_back(in_iov, &in_num, sizeof(struct virtio_blk_inhdr));

    /* TODO Linux sets the barrier bit even when not advertised! */
    outhdr.type &= ~VIRTIO_BLK_T_BARRIER;

    switch (outhdr.type) {
    case VIRTIO_BLK_T_IN:
        do_rdwr_cmd(s, true, in_iov, in_num, outhdr.sector * 512, head, inhdr);
        return 0;

    case VIRTIO_BLK_T_OUT:
        do_rdwr_cmd(s, false, iov, out_num, outhdr.sector * 512, head, inhdr);
        return 0;

    case VIRTIO_BLK_T_SCSI_CMD:
        /* TODO support SCSI commands */
        complete_request_early(s, head, inhdr, VIRTIO_BLK_S_UNSUPP);
        return 0;

    case VIRTIO_BLK_T_FLUSH:
        /* TODO fdsync not supported by Linux AIO, do it synchronously here! */
        if (qemu_fdatasync(s->fd) < 0) {
            complete_request_early(s, head, inhdr, VIRTIO_BLK_S_IOERR);
        } else {
            complete_request_early(s, head, inhdr, VIRTIO_BLK_S_OK);
        }
        return 0;

    case VIRTIO_BLK_T_GET_ID:
        do_get_id_cmd(s, in_iov, in_num, head, inhdr);
        return 0;

    default:
        error_report("virtio-blk unsupported request type %#x", outhdr.type);
        qemu_iovec_destroy(inhdr);
        g_slice_free(QEMUIOVector, inhdr);
        return -EFAULT;
    }
}

static void handle_notify(EventNotifier *e)
{
    VirtIOBlockDataPlane *s = container_of(e, VirtIOBlockDataPlane,
                                           host_notifier);

    /* There is one array of iovecs into which all new requests are extracted
     * from the vring.  Requests are read from the vring and the translated
     * descriptors are written to the iovecs array.  The iovecs do not have to
     * persist across handle_notify() calls because the kernel copies the
     * iovecs on io_submit().
     *
     * Handling io_submit() EAGAIN may require storing the requests across
     * handle_notify() calls until the kernel has sufficient resources to
     * accept more I/O.  This is not implemented yet.
     */
    struct iovec iovec[VRING_MAX];
    struct iovec *end = &iovec[VRING_MAX];
    struct iovec *iov = iovec;

    /* When a request is read from the vring, the index of the first descriptor
     * (aka head) is returned so that the completed request can be pushed onto
     * the vring later.
     *
     * The number of hypervisor read-only iovecs is out_num.  The number of
     * hypervisor write-only iovecs is in_num.
     */
    int head;
    unsigned int out_num = 0, in_num = 0;
    unsigned int num_queued;

    event_notifier_test_and_clear(&s->host_notifier);
    for (;;) {
        /* Disable guest->host notifies to avoid unnecessary vmexits */
        vring_disable_notification(s->vdev, &s->vring);

        for (;;) {
            head = vring_pop(s->vdev, &s->vring, iov, end, &out_num, &in_num);
            if (head < 0) {
                break; /* no more requests */
            }

            trace_virtio_blk_data_plane_process_request(s, out_num, in_num,
                                                        head);

            if (process_request(&s->ioqueue, iov, out_num, in_num, head) < 0) {
                vring_set_broken(&s->vring);
                break;
            }
            iov += out_num + in_num;
        }

        if (likely(head == -EAGAIN)) { /* vring emptied */
            /* Re-enable guest->host notifies and stop processing the vring.
             * But if the guest has snuck in more descriptors, keep processing.
             */
            if (vring_enable_notification(s->vdev, &s->vring)) {
                break;
            }
        } else { /* head == -ENOBUFS or fatal error, iovecs[] is depleted */
            /* Since there are no iovecs[] left, stop processing for now.  Do
             * not re-enable guest->host notifies since the I/O completion
             * handler knows to check for more vring descriptors anyway.
             */
            break;
        }
    }

    num_queued = ioq_num_queued(&s->ioqueue);
    if (num_queued > 0) {
        s->num_reqs += num_queued;

        int rc = ioq_submit(&s->ioqueue);
        if (unlikely(rc < 0)) {
            fprintf(stderr, "ioq_submit failed %d\n", rc);
            exit(1);
        }
    }
}

static void handle_io(EventNotifier *e)
{
    VirtIOBlockDataPlane *s = container_of(e, VirtIOBlockDataPlane,
                                           io_notifier);

    event_notifier_test_and_clear(&s->io_notifier);
    if (ioq_run_completion(&s->ioqueue, complete_request, s) > 0) {
        notify_guest(s);
    }

    /* If there were more requests than iovecs, the vring will not be empty yet
     * so check again.  There should now be enough resources to process more
     * requests.
     */
    if (unlikely(vring_more_avail(&s->vring))) {
        handle_notify(&s->host_notifier);
    }
}

static void *data_plane_thread(void *opaque)
{
    VirtIOBlockDataPlane *s = opaque;

    while (!s->stopping || s->num_reqs > 0) {
        aio_poll(s->ctx, true);
    }
    return NULL;
}

static void start_data_plane_bh(void *opaque)
{
    VirtIOBlockDataPlane *s = opaque;

    qemu_bh_delete(s->start_bh);
    s->start_bh = NULL;
    qemu_thread_create(&s->thread, data_plane_thread,
                       s, QEMU_THREAD_JOINABLE);
}

bool virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *blk,
                                  VirtIOBlockDataPlane **dataplane)
{
    VirtIOBlockDataPlane *s;
    int fd;

    *dataplane = NULL;

    if (!blk->data_plane) {
        return true;
    }

    if (blk->scsi) {
        error_report("device is incompatible with x-data-plane, use scsi=off");
        return false;
    }

    if (blk->config_wce) {
        error_report("device is incompatible with x-data-plane, "
                     "use config-wce=off");
        return false;
    }

    /* If dataplane is (re-)enabled while the guest is running there could be
     * block jobs that can conflict.
     */
    if (bdrv_in_use(blk->conf.bs)) {
        error_report("cannot start dataplane thread while device is in use");
        return false;
    }

    fd = raw_get_aio_fd(blk->conf.bs);
    if (fd < 0) {
        error_report("drive is incompatible with x-data-plane, "
                     "use format=raw,cache=none,aio=native");
        return false;
    }

    s = g_new0(VirtIOBlockDataPlane, 1);
    s->vdev = vdev;
    s->fd = fd;
    s->blk = blk;

    /* Prevent block operations that conflict with data plane thread */
    bdrv_set_in_use(blk->conf.bs, 1);

    *dataplane = s;
    return true;
}

void virtio_blk_data_plane_destroy(VirtIOBlockDataPlane *s)
{
    if (!s) {
        return;
    }

    virtio_blk_data_plane_stop(s);
    bdrv_set_in_use(s->blk->conf.bs, 0);
    g_free(s);
}

void virtio_blk_data_plane_start(VirtIOBlockDataPlane *s)
{
    BusState *qbus = BUS(qdev_get_parent_bus(DEVICE(s->vdev)));
    VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(qbus);
    VirtQueue *vq;
    int i;

    if (s->started) {
        return;
    }

    if (s->starting) {
        return;
    }

    s->starting = true;

    vq = virtio_get_queue(s->vdev, 0);
    if (!vring_setup(&s->vring, s->vdev, 0)) {
        s->starting = false;
        return;
    }

    s->ctx = aio_context_new();

    /* Set up guest notifier (irq) */
    if (k->set_guest_notifiers(qbus->parent, 1, true) != 0) {
        fprintf(stderr, "virtio-blk failed to set guest notifier, "
                "ensure -enable-kvm is set\n");
        exit(1);
    }
    s->guest_notifier = virtio_queue_get_guest_notifier(vq);

    /* Set up virtqueue notify */
    if (k->set_host_notifier(qbus->parent, 0, true) != 0) {
        fprintf(stderr, "virtio-blk failed to set host notifier\n");
        exit(1);
    }
    s->host_notifier = *virtio_queue_get_host_notifier(vq);
    aio_set_event_notifier(s->ctx, &s->host_notifier, handle_notify);

    /* Set up ioqueue */
    ioq_init(&s->ioqueue, s->fd, REQ_MAX);
    for (i = 0; i < ARRAY_SIZE(s->requests); i++) {
        ioq_put_iocb(&s->ioqueue, &s->requests[i].iocb);
    }
    s->io_notifier = *ioq_get_notifier(&s->ioqueue);
    aio_set_event_notifier(s->ctx, &s->io_notifier, handle_io);

    s->starting = false;
    s->started = true;
    trace_virtio_blk_data_plane_start(s);

    /* Kick right away to begin processing requests already in vring */
    event_notifier_set(virtio_queue_get_host_notifier(vq));

    /* Spawn thread in BH so it inherits iothread cpusets */
    s->start_bh = qemu_bh_new(start_data_plane_bh, s);
    qemu_bh_schedule(s->start_bh);
}

void virtio_blk_data_plane_stop(VirtIOBlockDataPlane *s)
{
    BusState *qbus = BUS(qdev_get_parent_bus(DEVICE(s->vdev)));
    VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(qbus);
    if (!s->started || s->stopping) {
        return;
    }
    s->stopping = true;
    trace_virtio_blk_data_plane_stop(s);

    /* Stop thread or cancel pending thread creation BH */
    if (s->start_bh) {
        qemu_bh_delete(s->start_bh);
        s->start_bh = NULL;
    } else {
        aio_notify(s->ctx);
        qemu_thread_join(&s->thread);
    }

    aio_set_event_notifier(s->ctx, &s->io_notifier, NULL);
    ioq_cleanup(&s->ioqueue);

    aio_set_event_notifier(s->ctx, &s->host_notifier, NULL);
    k->set_host_notifier(qbus->parent, 0, false);

    aio_context_unref(s->ctx);

    /* Clean up guest notifier (irq) */
    k->set_guest_notifiers(qbus->parent, 1, false);

    vring_teardown(&s->vring, s->vdev, 0);
    s->started = false;
    s->stopping = false;
}