summaryrefslogtreecommitdiffstats
path: root/qemu/aio-win32.c
blob: 50a68674589aa3c6418604bfb9320e915f624796 (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
/*
 * QEMU aio implementation
 *
 * Copyright IBM Corp., 2008
 * Copyright Red Hat Inc., 2012
 *
 * Authors:
 *  Anthony Liguori   <aliguori@us.ibm.com>
 *  Paolo Bonzini     <pbonzini@redhat.com>
 *
 * This work is licensed under the terms of the GNU GPL, version 2.  See
 * the COPYING file in the top-level directory.
 *
 * Contributions after 2012-01-13 are licensed under the terms of the
 * GNU GPL, version 2 or (at your option) any later version.
 */

#include "qemu-common.h"
#include "block/block.h"
#include "qemu/queue.h"
#include "qemu/sockets.h"

struct AioHandler {
    EventNotifier *e;
    IOHandler *io_read;
    IOHandler *io_write;
    EventNotifierHandler *io_notify;
    GPollFD pfd;
    int deleted;
    void *opaque;
    QLIST_ENTRY(AioHandler) node;
};

void aio_set_fd_handler(AioContext *ctx,
                        int fd,
                        IOHandler *io_read,
                        IOHandler *io_write,
                        void *opaque)
{
    /* fd is a SOCKET in our case */
    AioHandler *node;

    QLIST_FOREACH(node, &ctx->aio_handlers, node) {
        if (node->pfd.fd == fd && !node->deleted) {
            break;
        }
    }

    /* Are we deleting the fd handler? */
    if (!io_read && !io_write) {
        if (node) {
            /* If the lock is held, just mark the node as deleted */
            if (ctx->walking_handlers) {
                node->deleted = 1;
                node->pfd.revents = 0;
            } else {
                /* Otherwise, delete it for real.  We can't just mark it as
                 * deleted because deleted nodes are only cleaned up after
                 * releasing the walking_handlers lock.
                 */
                QLIST_REMOVE(node, node);
                g_free(node);
            }
        }
    } else {
        HANDLE event;

        if (node == NULL) {
            /* Alloc and insert if it's not already there */
            node = g_new0(AioHandler, 1);
            node->pfd.fd = fd;
            QLIST_INSERT_HEAD(&ctx->aio_handlers, node, node);
        }

        node->pfd.events = 0;
        if (node->io_read) {
            node->pfd.events |= G_IO_IN;
        }
        if (node->io_write) {
            node->pfd.events |= G_IO_OUT;
        }

        node->e = &ctx->notifier;

        /* Update handler with latest information */
        node->opaque = opaque;
        node->io_read = io_read;
        node->io_write = io_write;

        event = event_notifier_get_handle(&ctx->notifier);
        WSAEventSelect(node->pfd.fd, event,
                       FD_READ | FD_ACCEPT | FD_CLOSE |
                       FD_CONNECT | FD_WRITE | FD_OOB);
    }

    aio_notify(ctx);
}

void aio_set_event_notifier(AioContext *ctx,
                            EventNotifier *e,
                            EventNotifierHandler *io_notify)
{
    AioHandler *node;

    QLIST_FOREACH(node, &ctx->aio_handlers, node) {
        if (node->e == e && !node->deleted) {
            break;
        }
    }

    /* Are we deleting the fd handler? */
    if (!io_notify) {
        if (node) {
            g_source_remove_poll(&ctx->source, &node->pfd);

            /* If the lock is held, just mark the node as deleted */
            if (ctx->walking_handlers) {
                node->deleted = 1;
                node->pfd.revents = 0;
            } else {
                /* Otherwise, delete it for real.  We can't just mark it as
                 * deleted because deleted nodes are only cleaned up after
                 * releasing the walking_handlers lock.
                 */
                QLIST_REMOVE(node, node);
                g_free(node);
            }
        }
    } else {
        if (node == NULL) {
            /* Alloc and insert if it's not already there */
            node = g_new0(AioHandler, 1);
            node->e = e;
            node->pfd.fd = (uintptr_t)event_notifier_get_handle(e);
            node->pfd.events = G_IO_IN;
            QLIST_INSERT_HEAD(&ctx->aio_handlers, node, node);

            g_source_add_poll(&ctx->source, &node->pfd);
        }
        /* Update handler with latest information */
        node->io_notify = io_notify;
    }

    aio_notify(ctx);
}

bool aio_prepare(AioContext *ctx)
{
    static struct timeval tv0;
    AioHandler *node;
    bool have_select_revents = false;
    fd_set rfds, wfds;

    /* fill fd sets */
    FD_ZERO(&rfds);
    FD_ZERO(&wfds);
    QLIST_FOREACH(node, &ctx->aio_handlers, node) {
        if (node->io_read) {
            FD_SET ((SOCKET)node->pfd.fd, &rfds);
        }
        if (node->io_write) {
            FD_SET ((SOCKET)node->pfd.fd, &wfds);
        }
    }

    if (select(0, &rfds, &wfds, NULL, &tv0) > 0) {
        QLIST_FOREACH(node, &ctx->aio_handlers, node) {
            node->pfd.revents = 0;
            if (FD_ISSET(node->pfd.fd, &rfds)) {
                node->pfd.revents |= G_IO_IN;
                have_select_revents = true;
            }

            if (FD_ISSET(node->pfd.fd, &wfds)) {
                node->pfd.revents |= G_IO_OUT;
                have_select_revents = true;
            }
        }
    }

    return have_select_revents;
}

bool aio_pending(AioContext *ctx)
{
    AioHandler *node;

    QLIST_FOREACH(node, &ctx->aio_handlers, node) {
        if (node->pfd.revents && node->io_notify) {
            return true;
        }

        if ((node->pfd.revents & G_IO_IN) && node->io_read) {
            return true;
        }
        if ((node->pfd.revents & G_IO_OUT) && node->io_write) {
            return true;
        }
    }

    return false;
}

static bool aio_dispatch_handlers(AioContext *ctx, HANDLE event)
{
    AioHandler *node;
    bool progress = false;

    /*
     * We have to walk very carefully in case aio_set_fd_handler is
     * called while we're walking.
     */
    node = QLIST_FIRST(&ctx->aio_handlers);
    while (node) {
        AioHandler *tmp;
        int revents = node->pfd.revents;

        ctx->walking_handlers++;

        if (!node->deleted &&
            (revents || event_notifier_get_handle(node->e) == event) &&
            node->io_notify) {
            node->pfd.revents = 0;
            node->io_notify(node->e);

            /* aio_notify() does not count as progress */
            if (node->e != &ctx->notifier) {
                progress = true;
            }
        }

        if (!node->deleted &&
            (node->io_read || node->io_write)) {
            node->pfd.revents = 0;
            if ((revents & G_IO_IN) && node->io_read) {
                node->io_read(node->opaque);
                progress = true;
            }
            if ((revents & G_IO_OUT) && node->io_write) {
                node->io_write(node->opaque);
                progress = true;
            }

            /* if the next select() will return an event, we have progressed */
            if (event == event_notifier_get_handle(&ctx->notifier)) {
                WSANETWORKEVENTS ev;
                WSAEnumNetworkEvents(node->pfd.fd, event, &ev);
                if (ev.lNetworkEvents) {
                    progress = true;
                }
            }
        }

        tmp = node;
        node = QLIST_NEXT(node, node);

        ctx->walking_handlers--;

        if (!ctx->walking_handlers && tmp->deleted) {
            QLIST_REMOVE(tmp, node);
            g_free(tmp);
        }
    }

    return progress;
}

bool aio_dispatch(AioContext *ctx)
{
    bool progress;

    progress = aio_bh_poll(ctx);
    progress |= aio_dispatch_handlers(ctx, INVALID_HANDLE_VALUE);
    progress |= timerlistgroup_run_timers(&ctx->tlg);
    return progress;
}

bool aio_poll(AioContext *ctx, bool blocking)
{
    AioHandler *node;
    HANDLE events[MAXIMUM_WAIT_OBJECTS + 1];
    bool progress, have_select_revents, first;
    int count;
    int timeout;

    aio_context_acquire(ctx);
    progress = false;

    /* aio_notify can avoid the expensive event_notifier_set if
     * everything (file descriptors, bottom halves, timers) will
     * be re-evaluated before the next blocking poll().  This is
     * already true when aio_poll is called with blocking == false;
     * if blocking == true, it is only true after poll() returns,
     * so disable the optimization now.
     */
    if (blocking) {
        atomic_add(&ctx->notify_me, 2);
    }

    have_select_revents = aio_prepare(ctx);

    ctx->walking_handlers++;

    /* fill fd sets */
    count = 0;
    QLIST_FOREACH(node, &ctx->aio_handlers, node) {
        if (!node->deleted && node->io_notify) {
            events[count++] = event_notifier_get_handle(node->e);
        }
    }

    ctx->walking_handlers--;
    first = true;

    /* ctx->notifier is always registered.  */
    assert(count > 0);

    /* Multiple iterations, all of them non-blocking except the first,
     * may be necessary to process all pending events.  After the first
     * WaitForMultipleObjects call ctx->notify_me will be decremented.
     */
    do {
        HANDLE event;
        int ret;

        timeout = blocking && !have_select_revents
            ? qemu_timeout_ns_to_ms(aio_compute_timeout(ctx)) : 0;
        if (timeout) {
            aio_context_release(ctx);
        }
        ret = WaitForMultipleObjects(count, events, FALSE, timeout);
        if (blocking) {
            assert(first);
            atomic_sub(&ctx->notify_me, 2);
        }
        if (timeout) {
            aio_context_acquire(ctx);
        }

        if (first) {
            aio_notify_accept(ctx);
            progress |= aio_bh_poll(ctx);
            first = false;
        }

        /* if we have any signaled events, dispatch event */
        event = NULL;
        if ((DWORD) (ret - WAIT_OBJECT_0) < count) {
            event = events[ret - WAIT_OBJECT_0];
            events[ret - WAIT_OBJECT_0] = events[--count];
        } else if (!have_select_revents) {
            break;
        }

        have_select_revents = false;
        blocking = false;

        progress |= aio_dispatch_handlers(ctx, event);
    } while (count > 0);

    progress |= timerlistgroup_run_timers(&ctx->tlg);

    aio_context_release(ctx);
    return progress;
}