main-loop.c 12.7 KB
Newer Older
P
Paolo Bonzini 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
/*
 * QEMU System Emulator
 *
 * Copyright (c) 2003-2008 Fabrice Bellard
 *
 * Permission is hereby granted, free of charge, to any person obtaining a copy
 * of this software and associated documentation files (the "Software"), to deal
 * in the Software without restriction, including without limitation the rights
 * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
 * copies of the Software, and to permit persons to whom the Software is
 * furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice shall be included in
 * all copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 * THE SOFTWARE.
 */

S
Stefan Weil 已提交
25
#include "qemu-common.h"
P
Paolo Bonzini 已提交
26
#include "qemu-timer.h"
S
Stefan Weil 已提交
27 28
#include "slirp/slirp.h"
#include "main-loop.h"
P
Paolo Bonzini 已提交
29 30 31

#ifndef _WIN32

S
Stefan Weil 已提交
32 33
#include "compatfd.h"

P
Paolo Bonzini 已提交
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
static int io_thread_fd = -1;

void qemu_notify_event(void)
{
    /* Write 8 bytes to be compatible with eventfd.  */
    static const uint64_t val = 1;
    ssize_t ret;

    if (io_thread_fd == -1) {
        return;
    }
    do {
        ret = write(io_thread_fd, &val, sizeof(val));
    } while (ret < 0 && errno == EINTR);

    /* EAGAIN is fine, a read must be pending.  */
    if (ret < 0 && errno != EAGAIN) {
        fprintf(stderr, "qemu_notify_event: write() failed: %s\n",
                strerror(errno));
        exit(1);
    }
}

static void qemu_event_read(void *opaque)
{
    int fd = (intptr_t)opaque;
    ssize_t len;
    char buffer[512];

    /* Drain the notify pipe.  For eventfd, only 8 bytes will be read.  */
    do {
        len = read(fd, buffer, sizeof(buffer));
    } while ((len == -1 && errno == EINTR) || len == sizeof(buffer));
}

static int qemu_event_init(void)
{
    int err;
    int fds[2];

    err = qemu_eventfd(fds);
    if (err == -1) {
        return -errno;
    }
    err = fcntl_setfl(fds[0], O_NONBLOCK);
    if (err < 0) {
        goto fail;
    }
    err = fcntl_setfl(fds[1], O_NONBLOCK);
    if (err < 0) {
        goto fail;
    }
    qemu_set_fd_handler2(fds[0], NULL, qemu_event_read, NULL,
                         (void *)(intptr_t)fds[0]);

    io_thread_fd = fds[1];
    return 0;

fail:
    close(fds[0]);
    close(fds[1]);
    return err;
}

/* If we have signalfd, we mask out the signals we want to handle and then
 * use signalfd to listen for them.  We rely on whatever the current signal
 * handler is to dispatch the signals when we receive them.
 */
static void sigfd_handler(void *opaque)
{
    int fd = (intptr_t)opaque;
    struct qemu_signalfd_siginfo info;
    struct sigaction action;
    ssize_t len;

    while (1) {
        do {
            len = read(fd, &info, sizeof(info));
        } while (len == -1 && errno == EINTR);

        if (len == -1 && errno == EAGAIN) {
            break;
        }

        if (len != sizeof(info)) {
            printf("read from sigfd returned %zd: %m\n", len);
            return;
        }

        sigaction(info.ssi_signo, NULL, &action);
        if ((action.sa_flags & SA_SIGINFO) && action.sa_sigaction) {
            action.sa_sigaction(info.ssi_signo,
                                (siginfo_t *)&info, NULL);
        } else if (action.sa_handler) {
            action.sa_handler(info.ssi_signo);
        }
    }
}

static int qemu_signal_init(void)
{
    int sigfd;
    sigset_t set;

    /*
     * SIG_IPI must be blocked in the main thread and must not be caught
     * by sigwait() in the signal thread. Otherwise, the cpu thread will
     * not catch it reliably.
     */
    sigemptyset(&set);
    sigaddset(&set, SIG_IPI);
    sigaddset(&set, SIGIO);
    sigaddset(&set, SIGALRM);
    sigaddset(&set, SIGBUS);
    pthread_sigmask(SIG_BLOCK, &set, NULL);

L
Lai Jiangshan 已提交
150
    sigdelset(&set, SIG_IPI);
P
Paolo Bonzini 已提交
151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166
    sigfd = qemu_signalfd(&set);
    if (sigfd == -1) {
        fprintf(stderr, "failed to create signalfd\n");
        return -errno;
    }

    fcntl_setfl(sigfd, O_NONBLOCK);

    qemu_set_fd_handler2(sigfd, NULL, sigfd_handler, NULL,
                         (void *)(intptr_t)sigfd);

    return 0;
}

#else /* _WIN32 */

167
static HANDLE qemu_event_handle = NULL;
P
Paolo Bonzini 已提交
168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185

static void dummy_event_handler(void *opaque)
{
}

static int qemu_event_init(void)
{
    qemu_event_handle = CreateEvent(NULL, FALSE, FALSE, NULL);
    if (!qemu_event_handle) {
        fprintf(stderr, "Failed CreateEvent: %ld\n", GetLastError());
        return -1;
    }
    qemu_add_wait_object(qemu_event_handle, dummy_event_handler, NULL);
    return 0;
}

void qemu_notify_event(void)
{
186 187 188
    if (!qemu_event_handle) {
        return;
    }
P
Paolo Bonzini 已提交
189 190 191 192 193 194 195 196 197 198 199 200 201
    if (!SetEvent(qemu_event_handle)) {
        fprintf(stderr, "qemu_notify_event: SetEvent failed: %ld\n",
                GetLastError());
        exit(1);
    }
}

static int qemu_signal_init(void)
{
    return 0;
}
#endif

202
int main_loop_init(void)
P
Paolo Bonzini 已提交
203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220
{
    int ret;

    qemu_mutex_lock_iothread();
    ret = qemu_signal_init();
    if (ret) {
        return ret;
    }

    /* Note eventfd must be drained before signalfd handlers run */
    ret = qemu_event_init();
    if (ret) {
        return ret;
    }

    return 0;
}

221 222
static fd_set rfds, wfds, xfds;
static int nfds;
223
static GPollFD poll_fds[1024 * 2]; /* this is probably overkill */
P
Paolo Bonzini 已提交
224 225 226
static int n_poll_fds;
static int max_priority;

227
#ifndef _WIN32
P
Paolo Bonzini 已提交
228
static void glib_select_fill(int *max_fd, fd_set *rfds, fd_set *wfds,
229
                             fd_set *xfds, uint32_t *cur_timeout)
P
Paolo Bonzini 已提交
230 231 232
{
    GMainContext *context = g_main_context_default();
    int i;
233
    int timeout = 0;
P
Paolo Bonzini 已提交
234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257

    g_main_context_prepare(context, &max_priority);

    n_poll_fds = g_main_context_query(context, max_priority, &timeout,
                                      poll_fds, ARRAY_SIZE(poll_fds));
    g_assert(n_poll_fds <= ARRAY_SIZE(poll_fds));

    for (i = 0; i < n_poll_fds; i++) {
        GPollFD *p = &poll_fds[i];

        if ((p->events & G_IO_IN)) {
            FD_SET(p->fd, rfds);
            *max_fd = MAX(*max_fd, p->fd);
        }
        if ((p->events & G_IO_OUT)) {
            FD_SET(p->fd, wfds);
            *max_fd = MAX(*max_fd, p->fd);
        }
        if ((p->events & G_IO_ERR)) {
            FD_SET(p->fd, xfds);
            *max_fd = MAX(*max_fd, p->fd);
        }
    }

258 259
    if (timeout >= 0 && timeout < *cur_timeout) {
        *cur_timeout = timeout;
P
Paolo Bonzini 已提交
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
    }
}

static void glib_select_poll(fd_set *rfds, fd_set *wfds, fd_set *xfds,
                             bool err)
{
    GMainContext *context = g_main_context_default();

    if (!err) {
        int i;

        for (i = 0; i < n_poll_fds; i++) {
            GPollFD *p = &poll_fds[i];

            if ((p->events & G_IO_IN) && FD_ISSET(p->fd, rfds)) {
                p->revents |= G_IO_IN;
            }
            if ((p->events & G_IO_OUT) && FD_ISSET(p->fd, wfds)) {
                p->revents |= G_IO_OUT;
            }
            if ((p->events & G_IO_ERR) && FD_ISSET(p->fd, xfds)) {
                p->revents |= G_IO_ERR;
            }
        }
    }

    if (g_main_context_check(context, max_priority, poll_fds, n_poll_fds)) {
        g_main_context_dispatch(context);
    }
}

291
static int os_host_main_loop_wait(uint32_t timeout)
292
{
293
    struct timeval tv, *tvarg = NULL;
294 295 296 297
    int ret;

    glib_select_fill(&nfds, &rfds, &wfds, &xfds, &timeout);

298 299 300 301 302 303
    if (timeout < UINT32_MAX) {
        tvarg = &tv;
        tv.tv_sec = timeout / 1000;
        tv.tv_usec = (timeout % 1000) * 1000;
    }

304 305 306 307
    if (timeout > 0) {
        qemu_mutex_unlock_iothread();
    }

308
    ret = select(nfds + 1, &rfds, &wfds, &xfds, tvarg);
309 310 311 312 313 314 315 316 317

    if (timeout > 0) {
        qemu_mutex_lock_iothread();
    }

    glib_select_poll(&rfds, &wfds, &xfds, (ret < 0));
    return ret;
}
#else
P
Paolo Bonzini 已提交
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
/***********************************************************/
/* Polling handling */

typedef struct PollingEntry {
    PollingFunc *func;
    void *opaque;
    struct PollingEntry *next;
} PollingEntry;

static PollingEntry *first_polling_entry;

int qemu_add_polling_cb(PollingFunc *func, void *opaque)
{
    PollingEntry **ppe, *pe;
    pe = g_malloc0(sizeof(PollingEntry));
    pe->func = func;
    pe->opaque = opaque;
    for(ppe = &first_polling_entry; *ppe != NULL; ppe = &(*ppe)->next);
    *ppe = pe;
    return 0;
}

void qemu_del_polling_cb(PollingFunc *func, void *opaque)
{
    PollingEntry **ppe, *pe;
    for(ppe = &first_polling_entry; *ppe != NULL; ppe = &(*ppe)->next) {
        pe = *ppe;
        if (pe->func == func && pe->opaque == opaque) {
            *ppe = pe->next;
            g_free(pe);
            break;
        }
    }
}

/***********************************************************/
/* Wait objects support */
typedef struct WaitObjects {
    int num;
357
    int revents[MAXIMUM_WAIT_OBJECTS + 1];
P
Paolo Bonzini 已提交
358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373
    HANDLE events[MAXIMUM_WAIT_OBJECTS + 1];
    WaitObjectFunc *func[MAXIMUM_WAIT_OBJECTS + 1];
    void *opaque[MAXIMUM_WAIT_OBJECTS + 1];
} WaitObjects;

static WaitObjects wait_objects = {0};

int qemu_add_wait_object(HANDLE handle, WaitObjectFunc *func, void *opaque)
{
    WaitObjects *w = &wait_objects;
    if (w->num >= MAXIMUM_WAIT_OBJECTS) {
        return -1;
    }
    w->events[w->num] = handle;
    w->func[w->num] = func;
    w->opaque[w->num] = opaque;
374
    w->revents[w->num] = 0;
P
Paolo Bonzini 已提交
375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392
    w->num++;
    return 0;
}

void qemu_del_wait_object(HANDLE handle, WaitObjectFunc *func, void *opaque)
{
    int i, found;
    WaitObjects *w = &wait_objects;

    found = 0;
    for (i = 0; i < w->num; i++) {
        if (w->events[i] == handle) {
            found = 1;
        }
        if (found) {
            w->events[i] = w->events[i + 1];
            w->func[i] = w->func[i + 1];
            w->opaque[i] = w->opaque[i + 1];
393
            w->revents[i] = w->revents[i + 1];
P
Paolo Bonzini 已提交
394 395 396 397 398 399 400
        }
    }
    if (found) {
        w->num--;
    }
}

401 402 403 404 405 406
void qemu_fd_register(int fd)
{
    WSAEventSelect(fd, qemu_event_handle, FD_READ | FD_ACCEPT | FD_CLOSE |
                   FD_CONNECT | FD_WRITE | FD_OOB);
}

407
static int os_host_main_loop_wait(uint32_t timeout)
P
Paolo Bonzini 已提交
408
{
409
    GMainContext *context = g_main_context_default();
410
    int ret, i;
P
Paolo Bonzini 已提交
411
    PollingEntry *pe;
412
    WaitObjects *w = &wait_objects;
413
    gint poll_timeout;
414
    static struct timeval tv0;
P
Paolo Bonzini 已提交
415 416 417 418 419 420

    /* XXX: need to suppress polling by better using win32 events */
    ret = 0;
    for (pe = first_polling_entry; pe != NULL; pe = pe->next) {
        ret |= pe->func(pe->opaque);
    }
421 422 423
    if (ret != 0) {
        return ret;
    }
P
Paolo Bonzini 已提交
424

425 426 427
    if (nfds >= 0) {
        ret = select(nfds + 1, &rfds, &wfds, &xfds, &tv0);
        if (ret != 0) {
428
            /* TODO. */
429 430
        }
    }
P
Paolo Bonzini 已提交
431

432
    g_main_context_prepare(context, &max_priority);
433
    n_poll_fds = g_main_context_query(context, max_priority, &poll_timeout,
434 435 436
                                      poll_fds, ARRAY_SIZE(poll_fds));
    g_assert(n_poll_fds <= ARRAY_SIZE(poll_fds));

437
    for (i = 0; i < w->num; i++) {
438
        poll_fds[n_poll_fds + i].fd = (DWORD_PTR)w->events[i];
439
        poll_fds[n_poll_fds + i].events = G_IO_IN;
440 441
    }

442
    qemu_mutex_unlock_iothread();
443
    ret = g_poll(poll_fds, n_poll_fds + w->num, poll_timeout);
444
    qemu_mutex_lock_iothread();
445 446
    if (ret > 0) {
        for (i = 0; i < w->num; i++) {
447
            w->revents[i] = poll_fds[n_poll_fds + i].revents;
448
        }
449 450 451
        for (i = 0; i < w->num; i++) {
            if (w->revents[i] && w->func[i]) {
                w->func[i](w->opaque[i]);
P
Paolo Bonzini 已提交
452 453 454 455
            }
        }
    }

456 457 458 459
    if (g_main_context_check(context, max_priority, poll_fds, n_poll_fds)) {
        g_main_context_dispatch(context);
    }

460 461 462 463 464
    /* If an edge-triggered socket event occurred, select will return a
     * positive result on the next iteration.  We do not need to do anything
     * here.
     */

465
    return ret;
P
Paolo Bonzini 已提交
466 467 468 469 470
}
#endif

int main_loop_wait(int nonblocking)
{
471 472
    int ret;
    uint32_t timeout = UINT32_MAX;
P
Paolo Bonzini 已提交
473 474 475 476 477 478 479 480 481 482 483 484 485 486 487

    if (nonblocking) {
        timeout = 0;
    } else {
        qemu_bh_update_timeout(&timeout);
    }

    /* poll any events */
    /* XXX: separate device handlers from system ones */
    nfds = -1;
    FD_ZERO(&rfds);
    FD_ZERO(&wfds);
    FD_ZERO(&xfds);

#ifdef CONFIG_SLIRP
488
    slirp_update_timeout(&timeout);
P
Paolo Bonzini 已提交
489 490 491
    slirp_select_fill(&nfds, &rfds, &wfds, &xfds);
#endif
    qemu_iohandler_fill(&nfds, &rfds, &wfds, &xfds);
492
    ret = os_host_main_loop_wait(timeout);
P
Paolo Bonzini 已提交
493 494 495 496 497 498 499 500 501 502 503 504 505
    qemu_iohandler_poll(&rfds, &wfds, &xfds, ret);
#ifdef CONFIG_SLIRP
    slirp_select_poll(&rfds, &wfds, &xfds, (ret < 0));
#endif

    qemu_run_all_timers();

    /* Check bottom-halves last in case any of the earlier events triggered
       them.  */
    qemu_bh_poll();

    return ret;
}