vl.c 153.9 KB
Newer Older
1
/*
B
bellard 已提交
2
 * QEMU System Emulator
3
 *
B
bellard 已提交
4
 * Copyright (c) 2003-2008 Fabrice Bellard
5
 *
B
bellard 已提交
6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
 * 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.
23 24 25 26 27 28
 */
#include <unistd.h>
#include <fcntl.h>
#include <signal.h>
#include <time.h>
#include <errno.h>
B
bellard 已提交
29
#include <sys/time.h>
30
#include <zlib.h>
B
bellard 已提交
31

J
Juan Quintela 已提交
32
/* Needed early for CONFIG_BSD etc. */
B
blueswir1 已提交
33 34
#include "config-host.h"

B
bellard 已提交
35
#ifndef _WIN32
P
Paul Brook 已提交
36
#include <libgen.h>
37
#include <pwd.h>
B
bellard 已提交
38
#include <sys/times.h>
B
bellard 已提交
39
#include <sys/wait.h>
B
bellard 已提交
40 41
#include <termios.h>
#include <sys/mman.h>
B
bellard 已提交
42
#include <sys/ioctl.h>
B
blueswir1 已提交
43
#include <sys/resource.h>
B
bellard 已提交
44
#include <sys/socket.h>
B
bellard 已提交
45
#include <netinet/in.h>
B
blueswir1 已提交
46 47 48 49 50 51 52 53
#include <net/if.h>
#if defined(__NetBSD__)
#include <net/if_tap.h>
#endif
#ifdef __linux__
#include <linux/if_tun.h>
#endif
#include <arpa/inet.h>
B
bellard 已提交
54
#include <dirent.h>
B
bellard 已提交
55
#include <netdb.h>
56
#include <sys/select.h>
J
Juan Quintela 已提交
57
#ifdef CONFIG_BSD
B
bellard 已提交
58
#include <sys/stat.h>
59
#if defined(__FreeBSD__) || defined(__DragonFly__)
B
bellard 已提交
60
#include <libutil.h>
B
blueswir1 已提交
61 62
#else
#include <util.h>
63
#endif
T
ths 已提交
64 65
#elif defined (__GLIBC__) && defined (__FreeBSD_kernel__)
#include <freebsd/stdlib.h>
B
bellard 已提交
66
#else
67
#ifdef __linux__
B
bellard 已提交
68 69
#include <pty.h>
#include <malloc.h>
70
#include <linux/rtc.h>
A
Andi Kleen 已提交
71
#include <sys/prctl.h>
T
ths 已提交
72 73 74 75 76 77

/* For the benefit of older linux systems which don't supply it,
   we use a local copy of hpet.h. */
/* #include <linux/hpet.h> */
#include "hpet.h"

B
bellard 已提交
78
#include <linux/ppdev.h>
79
#include <linux/parport.h>
80 81
#endif
#ifdef __sun__
82 83 84 85 86 87 88 89 90 91 92 93 94
#include <sys/stat.h>
#include <sys/ethernet.h>
#include <sys/sockio.h>
#include <netinet/arp.h>
#include <netinet/in.h>
#include <netinet/in_systm.h>
#include <netinet/ip.h>
#include <netinet/ip_icmp.h> // must come after ip.h
#include <netinet/udp.h>
#include <netinet/tcp.h>
#include <net/if.h>
#include <syslog.h>
#include <stropts.h>
95 96 97
/* See MySQL bug #7156 (http://bugs.mysql.com/bug.php?id=7156) for
   discussion about Solaris header problems */
extern int madvise(caddr_t, size_t, int);
B
bellard 已提交
98
#endif
B
bellard 已提交
99
#endif
B
bellard 已提交
100
#endif
B
bellard 已提交
101

102 103 104 105
#if defined(__OpenBSD__)
#include <util.h>
#endif

106 107 108 109
#if defined(CONFIG_VDE)
#include <libvdeplug.h>
#endif

B
bellard 已提交
110
#ifdef _WIN32
111
#include <windows.h>
T
ths 已提交
112
#include <mmsystem.h>
B
bellard 已提交
113 114
#endif

B
bellard 已提交
115
#ifdef CONFIG_SDL
S
Stefan Weil 已提交
116
#if defined(__APPLE__) || defined(main)
S
Stefan Weil 已提交
117
#include <SDL.h>
M
malc 已提交
118 119 120
int qemu_main(int argc, char **argv, char **envp);
int main(int argc, char **argv)
{
S
Stefan Weil 已提交
121
    return qemu_main(argc, argv, NULL);
M
malc 已提交
122 123 124
}
#undef main
#define main qemu_main
B
bellard 已提交
125
#endif
B
bellard 已提交
126
#endif /* CONFIG_SDL */
127

128 129 130 131 132
#ifdef CONFIG_COCOA
#undef main
#define main qemu_main
#endif /* CONFIG_COCOA */

133 134 135 136 137 138 139 140 141
#include "hw/hw.h"
#include "hw/boards.h"
#include "hw/usb.h"
#include "hw/pcmcia.h"
#include "hw/pc.h"
#include "hw/audiodev.h"
#include "hw/isa.h"
#include "hw/baum.h"
#include "hw/bt.h"
R
Richard W.M. Jones 已提交
142
#include "hw/watchdog.h"
143
#include "hw/smbios.h"
144
#include "hw/xen.h"
145
#include "hw/qdev.h"
A
aurel32 已提交
146
#include "bt-host.h"
147 148 149 150 151 152 153 154 155
#include "net.h"
#include "monitor.h"
#include "console.h"
#include "sysemu.h"
#include "gdbstub.h"
#include "qemu-timer.h"
#include "qemu-char.h"
#include "cache-utils.h"
#include "block.h"
B
blueswir1 已提交
156
#include "dma.h"
157 158 159 160
#include "audio/audio.h"
#include "migration.h"
#include "kvm.h"
#include "balloon.h"
K
Kevin Wolf 已提交
161
#include "qemu-option.h"
G
Gerd Hoffmann 已提交
162
#include "qemu-config.h"
163

164
#include "disas.h"
B
bellard 已提交
165

166
#include "exec-all.h"
167

168 169
#include "qemu_socket.h"

J
Jan Kiszka 已提交
170
#include "slirp/libslirp.h"
171

B
Blue Swirl 已提交
172 173
#include "qemu-queue.h"

174 175
//#define DEBUG_NET
//#define DEBUG_SLIRP
176

177
#define DEFAULT_RAM_SIZE 128
178

179 180 181
/* Maximum number of monitor devices */
#define MAX_MONITOR_DEVICES 10

P
Paul Brook 已提交
182
static const char *data_dir;
183
const char *bios_name = NULL;
T
ths 已提交
184
/* Note: drives_table[MAX_DRIVES] is a dummy block driver if none available
B
bellard 已提交
185
   to store the VM snapshots */
B
Blue Swirl 已提交
186 187
struct drivelist drives = QTAILQ_HEAD_INITIALIZER(drives);
struct driveoptlist driveopts = QTAILQ_HEAD_INITIALIZER(driveopts);
M
malc 已提交
188
enum vga_retrace_method vga_retrace_method = VGA_RETRACE_DUMB;
189
static DisplayState *display_state;
190
DisplayType display_type = DT_DEFAULT;
191
const char* keyboard_layout = NULL;
A
Anthony Liguori 已提交
192
ram_addr_t ram_size;
193
int nb_nics;
B
bellard 已提交
194
NICInfo nd_table[MAX_NICS];
195
int vm_running;
P
Paolo Bonzini 已提交
196
int autostart;
197 198
static int rtc_utc = 1;
static int rtc_date_offset = -1; /* -1 means no change */
J
Jan Kiszka 已提交
199
QEMUClock *rtc_clock;
200
int vga_interface_type = VGA_CIRRUS;
201 202 203
#ifdef TARGET_SPARC
int graphic_width = 1024;
int graphic_height = 768;
B
blueswir1 已提交
204
int graphic_depth = 8;
205
#else
206 207
int graphic_width = 800;
int graphic_height = 600;
208
int graphic_depth = 15;
B
blueswir1 已提交
209
#endif
B
blueswir1 已提交
210
static int full_screen = 0;
211
#ifdef CONFIG_SDL
B
blueswir1 已提交
212
static int no_frame = 0;
213
#endif
T
ths 已提交
214
int no_quit = 0;
215
CharDriverState *serial_hds[MAX_SERIAL_PORTS];
216
CharDriverState *parallel_hds[MAX_PARALLEL_PORTS];
217
CharDriverState *virtcon_hds[MAX_VIRTIO_CONSOLES];
218 219
#ifdef TARGET_I386
int win2k_install_hack = 0;
220
int rtc_td_hack = 0;
221
#endif
B
bellard 已提交
222
int usb_enabled = 0;
223
int singlestep = 0;
B
bellard 已提交
224
int smp_cpus = 1;
225
int max_cpus = 0;
226 227
int smp_cores = 1;
int smp_threads = 1;
228
const char *vnc_display;
B
bellard 已提交
229
int acpi_enabled = 1;
A
aliguori 已提交
230
int no_hpet = 0;
B
bellard 已提交
231
int fd_bootchk = 1;
B
bellard 已提交
232
int no_reboot = 0;
A
aurel32 已提交
233
int no_shutdown = 0;
234
int cursor_hide = 1;
235
int graphic_rotate = 0;
J
Jes Sorensen 已提交
236
uint8_t irq0override = 1;
B
blueswir1 已提交
237
#ifndef _WIN32
T
ths 已提交
238
int daemonize = 0;
B
blueswir1 已提交
239
#endif
M
Markus Armbruster 已提交
240
const char *watchdog;
241 242
const char *option_rom[MAX_OPTION_ROMS];
int nb_option_roms;
243
int semihosting_enabled = 0;
244 245 246
#ifdef TARGET_ARM
int old_param = 0;
#endif
T
ths 已提交
247
const char *qemu_name;
T
ths 已提交
248
int alt_grab = 0;
249
#if defined(TARGET_SPARC) || defined(TARGET_PPC)
B
blueswir1 已提交
250 251 252
unsigned int nb_prom_envs = 0;
const char *prom_envs[MAX_PROM_ENVS];
#endif
253
int boot_menu;
254

255 256 257 258
int nb_numa_nodes;
uint64_t node_mem[MAX_NODES];
uint64_t node_cpumask[MAX_NODES];

259 260
static CPUState *cur_cpu;
static CPUState *next_cpu;
261
static int timer_alarm_pending = 1;
T
ths 已提交
262
/* Conversion factor from emulated instructions to virtual clock ticks.  */
P
pbrook 已提交
263
static int icount_time_shift;
T
ths 已提交
264
/* Arbitrarily pick 1MIPS as the minimum allowable speed.  */
P
pbrook 已提交
265 266 267
#define MAX_ICOUNT_SHIFT 10
/* Compensate for varying guest execution speed.  */
static int64_t qemu_icount_bias;
B
blueswir1 已提交
268 269
static QEMUTimer *icount_rt_timer;
static QEMUTimer *icount_vm_timer;
B
blueswir1 已提交
270
static QEMUTimer *nographic_timer;
271

272 273
uint8_t qemu_uuid[16];

J
Jan Kiszka 已提交
274 275 276
static QEMUBootSetHandler *boot_set_handler;
static void *boot_set_opaque;

277
/***********************************************************/
278 279
/* x86 ISA bus support */

A
Anthony Liguori 已提交
280
target_phys_addr_t isa_mem_base = 0;
B
bellard 已提交
281
PicState2 *isa_pic;
282 283 284 285 286

/***********************************************************/
void hw_error(const char *fmt, ...)
{
    va_list ap;
B
bellard 已提交
287
    CPUState *env;
288 289 290 291 292

    va_start(ap, fmt);
    fprintf(stderr, "qemu: hardware error: ");
    vfprintf(stderr, fmt, ap);
    fprintf(stderr, "\n");
B
bellard 已提交
293 294
    for(env = first_cpu; env != NULL; env = env->next_cpu) {
        fprintf(stderr, "CPU #%d:\n", env->cpu_index);
295
#ifdef TARGET_I386
B
bellard 已提交
296
        cpu_dump_state(env, stderr, fprintf, X86_DUMP_FPU);
297
#else
B
bellard 已提交
298
        cpu_dump_state(env, stderr, fprintf, 0);
299
#endif
B
bellard 已提交
300
    }
301 302 303
    va_end(ap);
    abort();
}
A
Andi Kleen 已提交
304 305 306

static void set_proc_name(const char *s)
{
307
#if defined(__linux__) && defined(PR_SET_NAME)
A
Andi Kleen 已提交
308 309 310 311 312 313 314 315 316 317
    char name[16];
    if (!s)
        return;
    name[sizeof(name) - 1] = 0;
    strncpy(name, s, sizeof(name));
    /* Could rewrite argv[0] too, but that's a bit more complicated.
       This simple way is enough for `top'. */
    prctl(PR_SET_NAME, name);
#endif    	
}
A
aliguori 已提交
318 319 320 321 322 323 324 325 326 327 328 329 330
 
/***************/
/* ballooning */

static QEMUBalloonEvent *qemu_balloon_event;
void *qemu_balloon_event_opaque;

void qemu_add_balloon_handler(QEMUBalloonEvent *func, void *opaque)
{
    qemu_balloon_event = func;
    qemu_balloon_event_opaque = opaque;
}

A
Anthony Liguori 已提交
331
void qemu_balloon(ram_addr_t target)
A
aliguori 已提交
332 333 334 335 336
{
    if (qemu_balloon_event)
        qemu_balloon_event(qemu_balloon_event_opaque, target);
}

A
Anthony Liguori 已提交
337
ram_addr_t qemu_balloon_status(void)
A
aliguori 已提交
338 339 340 341 342
{
    if (qemu_balloon_event)
        return qemu_balloon_event(qemu_balloon_event_opaque, 0);
    return 0;
}
343

B
bellard 已提交
344 345 346 347 348
/***********************************************************/
/* keyboard/mouse */

static QEMUPutKBDEvent *qemu_put_kbd_event;
static void *qemu_put_kbd_event_opaque;
349 350
static QEMUPutMouseEntry *qemu_put_mouse_event_head;
static QEMUPutMouseEntry *qemu_put_mouse_event_current;
B
bellard 已提交
351 352 353 354 355 356 357

void qemu_add_kbd_event_handler(QEMUPutKBDEvent *func, void *opaque)
{
    qemu_put_kbd_event_opaque = opaque;
    qemu_put_kbd_event = func;
}

358 359 360
QEMUPutMouseEntry *qemu_add_mouse_event_handler(QEMUPutMouseEvent *func,
                                                void *opaque, int absolute,
                                                const char *name)
B
bellard 已提交
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
    QEMUPutMouseEntry *s, *cursor;

    s = qemu_mallocz(sizeof(QEMUPutMouseEntry));

    s->qemu_put_mouse_event = func;
    s->qemu_put_mouse_event_opaque = opaque;
    s->qemu_put_mouse_event_absolute = absolute;
    s->qemu_put_mouse_event_name = qemu_strdup(name);
    s->next = NULL;

    if (!qemu_put_mouse_event_head) {
        qemu_put_mouse_event_head = qemu_put_mouse_event_current = s;
        return s;
    }

    cursor = qemu_put_mouse_event_head;
    while (cursor->next != NULL)
        cursor = cursor->next;

    cursor->next = s;
    qemu_put_mouse_event_current = s;

    return s;
}

void qemu_remove_mouse_event_handler(QEMUPutMouseEntry *entry)
{
    QEMUPutMouseEntry *prev = NULL, *cursor;

    if (!qemu_put_mouse_event_head || entry == NULL)
        return;

    cursor = qemu_put_mouse_event_head;
    while (cursor != NULL && cursor != entry) {
        prev = cursor;
        cursor = cursor->next;
    }

    if (cursor == NULL) // does not exist or list empty
        return;
    else if (prev == NULL) { // entry is head
        qemu_put_mouse_event_head = cursor->next;
        if (qemu_put_mouse_event_current == entry)
            qemu_put_mouse_event_current = cursor->next;
        qemu_free(entry->qemu_put_mouse_event_name);
        qemu_free(entry);
        return;
    }

    prev->next = entry->next;

    if (qemu_put_mouse_event_current == entry)
        qemu_put_mouse_event_current = prev;

    qemu_free(entry->qemu_put_mouse_event_name);
    qemu_free(entry);
B
bellard 已提交
418 419 420 421 422 423 424 425 426 427 428
}

void kbd_put_keycode(int keycode)
{
    if (qemu_put_kbd_event) {
        qemu_put_kbd_event(qemu_put_kbd_event_opaque, keycode);
    }
}

void kbd_mouse_event(int dx, int dy, int dz, int buttons_state)
{
429 430
    QEMUPutMouseEvent *mouse_event;
    void *mouse_event_opaque;
431
    int width;
432 433 434 435 436 437 438 439 440 441 442

    if (!qemu_put_mouse_event_current) {
        return;
    }

    mouse_event =
        qemu_put_mouse_event_current->qemu_put_mouse_event;
    mouse_event_opaque =
        qemu_put_mouse_event_current->qemu_put_mouse_event_opaque;

    if (mouse_event) {
443 444 445 446
        if (graphic_rotate) {
            if (qemu_put_mouse_event_current->qemu_put_mouse_event_absolute)
                width = 0x7fff;
            else
447
                width = graphic_width - 1;
448 449 450 451 452
            mouse_event(mouse_event_opaque,
                                 width - dy, dx, dz, buttons_state);
        } else
            mouse_event(mouse_event_opaque,
                                 dx, dy, dz, buttons_state);
B
bellard 已提交
453 454 455
    }
}

456 457
int kbd_mouse_is_absolute(void)
{
458 459 460 461 462 463
    if (!qemu_put_mouse_event_current)
        return 0;

    return qemu_put_mouse_event_current->qemu_put_mouse_event_absolute;
}

A
aliguori 已提交
464
void do_info_mice(Monitor *mon)
465 466 467 468 469
{
    QEMUPutMouseEntry *cursor;
    int index = 0;

    if (!qemu_put_mouse_event_head) {
A
aliguori 已提交
470
        monitor_printf(mon, "No mouse devices connected\n");
471 472 473
        return;
    }

A
aliguori 已提交
474
    monitor_printf(mon, "Mouse devices available:\n");
475 476
    cursor = qemu_put_mouse_event_head;
    while (cursor != NULL) {
A
aliguori 已提交
477 478 479
        monitor_printf(mon, "%c Mouse #%d: %s\n",
                       (cursor == qemu_put_mouse_event_current ? '*' : ' '),
                       index, cursor->qemu_put_mouse_event_name);
480 481 482 483 484
        index++;
        cursor = cursor->next;
    }
}

485
void do_mouse_set(Monitor *mon, const QDict *qdict)
486 487 488
{
    QEMUPutMouseEntry *cursor;
    int i = 0;
489
    int index = qdict_get_int(qdict, "index");
490 491

    if (!qemu_put_mouse_event_head) {
A
aliguori 已提交
492
        monitor_printf(mon, "No mouse devices connected\n");
493 494 495 496 497 498 499 500 501 502 503 504
        return;
    }

    cursor = qemu_put_mouse_event_head;
    while (cursor != NULL && index != i) {
        i++;
        cursor = cursor->next;
    }

    if (cursor != NULL)
        qemu_put_mouse_event_current = cursor;
    else
A
aliguori 已提交
505
        monitor_printf(mon, "Mouse at given index not found\n");
506 507
}

508 509
/* compute with 96 bit intermediate result: (a*b)/c */
uint64_t muldiv64(uint64_t a, uint32_t b, uint32_t c)
510
{
511 512 513
    union {
        uint64_t ll;
        struct {
514
#ifdef HOST_WORDS_BIGENDIAN
515 516 517
            uint32_t high, low;
#else
            uint32_t low, high;
518
#endif
519 520 521
        } l;
    } u, res;
    uint64_t rl, rh;
522

523 524 525 526 527 528 529
    u.ll = a;
    rl = (uint64_t)u.l.low * (uint64_t)b;
    rh = (uint64_t)u.l.high * (uint64_t)b;
    rh += (rl >> 32);
    res.l.high = rh / c;
    res.l.low = (((rh % c) << 32) + (rl & 0xffffffff)) / c;
    return res.ll;
530 531
}

532 533
/***********************************************************/
/* real time host monotonic timer */
534

J
Jan Kiszka 已提交
535 536 537 538 539 540 541 542
static int64_t get_clock_realtime(void)
{
    struct timeval tv;

    gettimeofday(&tv, NULL);
    return tv.tv_sec * 1000000000LL + (tv.tv_usec * 1000);
}

543
#ifdef WIN32
544

545
static int64_t clock_freq;
B
bellard 已提交
546

547
static void init_get_clock(void)
B
bellard 已提交
548
{
549 550
    LARGE_INTEGER freq;
    int ret;
551 552 553 554 555 556
    ret = QueryPerformanceFrequency(&freq);
    if (ret == 0) {
        fprintf(stderr, "Could not calibrate ticks\n");
        exit(1);
    }
    clock_freq = freq.QuadPart;
B
bellard 已提交
557 558
}

559
static int64_t get_clock(void)
B
bellard 已提交
560
{
561 562
    LARGE_INTEGER ti;
    QueryPerformanceCounter(&ti);
563
    return muldiv64(ti.QuadPart, get_ticks_per_sec(), clock_freq);
B
bellard 已提交
564 565
}

566
#else
B
bellard 已提交
567

568 569 570
static int use_rt_clock;

static void init_get_clock(void)
B
bellard 已提交
571
{
572
    use_rt_clock = 0;
573 574
#if defined(__linux__) || (defined(__FreeBSD__) && __FreeBSD_version >= 500000) \
    || defined(__DragonFly__)
575 576 577 578 579 580 581
    {
        struct timespec ts;
        if (clock_gettime(CLOCK_MONOTONIC, &ts) == 0) {
            use_rt_clock = 1;
        }
    }
#endif
B
bellard 已提交
582 583
}

584
static int64_t get_clock(void)
B
bellard 已提交
585
{
586 587
#if defined(__linux__) || (defined(__FreeBSD__) && __FreeBSD_version >= 500000) \
	|| defined(__DragonFly__)
588 589 590 591
    if (use_rt_clock) {
        struct timespec ts;
        clock_gettime(CLOCK_MONOTONIC, &ts);
        return ts.tv_sec * 1000000000LL + ts.tv_nsec;
592
    } else
B
bellard 已提交
593
#endif
594 595 596
    {
        /* XXX: using gettimeofday leads to problems if the date
           changes, so it should be avoided. */
J
Jan Kiszka 已提交
597
        return get_clock_realtime();
598
    }
B
bellard 已提交
599
}
600 601
#endif

P
pbrook 已提交
602 603 604 605 606 607 608 609 610 611 612 613 614 615
/* Return the virtual CPU time, based on the instruction counter.  */
static int64_t cpu_get_icount(void)
{
    int64_t icount;
    CPUState *env = cpu_single_env;;
    icount = qemu_icount;
    if (env) {
        if (!can_do_io(env))
            fprintf(stderr, "Bad clock read\n");
        icount -= (env->icount_decr.u16.low + env->icount_extra);
    }
    return qemu_icount_bias + (icount << icount_time_shift);
}

616 617 618
/***********************************************************/
/* guest cycle counter */

619 620 621 622 623
typedef struct TimersState {
    int64_t cpu_ticks_prev;
    int64_t cpu_ticks_offset;
    int64_t cpu_clock_offset;
    int32_t cpu_ticks_enabled;
624
    int64_t dummy;
625 626 627
} TimersState;

TimersState timers_state;
628

629 630
/* return the host CPU cycle counter and handle stop/restart */
int64_t cpu_get_ticks(void)
631
{
P
pbrook 已提交
632 633 634
    if (use_icount) {
        return cpu_get_icount();
    }
635 636
    if (!timers_state.cpu_ticks_enabled) {
        return timers_state.cpu_ticks_offset;
637
    } else {
638 639
        int64_t ticks;
        ticks = cpu_get_real_ticks();
640
        if (timers_state.cpu_ticks_prev > ticks) {
641 642
            /* Note: non increasing ticks may happen if the host uses
               software suspend */
643
            timers_state.cpu_ticks_offset += timers_state.cpu_ticks_prev - ticks;
644
        }
645 646
        timers_state.cpu_ticks_prev = ticks;
        return ticks + timers_state.cpu_ticks_offset;
647
    }
648 649
}

650 651 652 653
/* return the host CPU monotonic timer and handle stop/restart */
static int64_t cpu_get_clock(void)
{
    int64_t ti;
654 655
    if (!timers_state.cpu_ticks_enabled) {
        return timers_state.cpu_clock_offset;
656 657
    } else {
        ti = get_clock();
658
        return ti + timers_state.cpu_clock_offset;
659 660 661
    }
}

662 663 664
/* enable cpu_get_ticks() */
void cpu_enable_ticks(void)
{
665 666 667 668
    if (!timers_state.cpu_ticks_enabled) {
        timers_state.cpu_ticks_offset -= cpu_get_real_ticks();
        timers_state.cpu_clock_offset -= get_clock();
        timers_state.cpu_ticks_enabled = 1;
669
    }
670 671 672 673 674 675
}

/* disable cpu_get_ticks() : the clock is stopped. You must not call
   cpu_get_ticks() after that.  */
void cpu_disable_ticks(void)
{
676 677 678 679
    if (timers_state.cpu_ticks_enabled) {
        timers_state.cpu_ticks_offset = cpu_get_ticks();
        timers_state.cpu_clock_offset = cpu_get_clock();
        timers_state.cpu_ticks_enabled = 0;
680
    }
681 682
}

683 684
/***********************************************************/
/* timers */
685

J
Jan Kiszka 已提交
686 687
#define QEMU_CLOCK_REALTIME 0
#define QEMU_CLOCK_VIRTUAL  1
J
Jan Kiszka 已提交
688
#define QEMU_CLOCK_HOST     2
689 690 691 692 693 694 695 696 697 698 699 700 701 702

struct QEMUClock {
    int type;
    /* XXX: add frequency */
};

struct QEMUTimer {
    QEMUClock *clock;
    int64_t expire_time;
    QEMUTimerCB *cb;
    void *opaque;
    struct QEMUTimer *next;
};

703 704
struct qemu_alarm_timer {
    char const *name;
705
    unsigned int flags;
706 707 708

    int (*start)(struct qemu_alarm_timer *t);
    void (*stop)(struct qemu_alarm_timer *t);
709
    void (*rearm)(struct qemu_alarm_timer *t);
710 711 712
    void *priv;
};

713
#define ALARM_FLAG_DYNTICKS  0x1
714
#define ALARM_FLAG_EXPIRED   0x2
715 716 717

static inline int alarm_has_dynticks(struct qemu_alarm_timer *t)
{
718
    return t && (t->flags & ALARM_FLAG_DYNTICKS);
719 720 721 722 723 724 725 726 727 728 729 730 731
}

static void qemu_rearm_alarm_timer(struct qemu_alarm_timer *t)
{
    if (!alarm_has_dynticks(t))
        return;

    t->rearm(t);
}

/* TODO: MIN_TIMER_REARM_US should be optimized */
#define MIN_TIMER_REARM_US 250

732
static struct qemu_alarm_timer *alarm_timer;
733

B
bellard 已提交
734
#ifdef _WIN32
735 736 737 738

struct qemu_alarm_win32 {
    MMRESULT timerId;
    unsigned int period;
B
Blue Swirl 已提交
739
} alarm_win32_data = {0, -1};
740 741 742

static int win32_start_timer(struct qemu_alarm_timer *t);
static void win32_stop_timer(struct qemu_alarm_timer *t);
743
static void win32_rearm_timer(struct qemu_alarm_timer *t);
744

B
bellard 已提交
745
#else
746 747 748 749

static int unix_start_timer(struct qemu_alarm_timer *t);
static void unix_stop_timer(struct qemu_alarm_timer *t);

750 751
#ifdef __linux__

752 753 754 755
static int dynticks_start_timer(struct qemu_alarm_timer *t);
static void dynticks_stop_timer(struct qemu_alarm_timer *t);
static void dynticks_rearm_timer(struct qemu_alarm_timer *t);

T
ths 已提交
756 757 758
static int hpet_start_timer(struct qemu_alarm_timer *t);
static void hpet_stop_timer(struct qemu_alarm_timer *t);

759 760 761
static int rtc_start_timer(struct qemu_alarm_timer *t);
static void rtc_stop_timer(struct qemu_alarm_timer *t);

762
#endif /* __linux__ */
763

764 765
#endif /* _WIN32 */

P
pbrook 已提交
766
/* Correlation between real and virtual time is always going to be
T
ths 已提交
767
   fairly approximate, so ignore small variation.
P
pbrook 已提交
768 769
   When the guest is idle real and virtual time will be aligned in
   the IO wait loop.  */
770
#define ICOUNT_WOBBLE (get_ticks_per_sec() / 10)
P
pbrook 已提交
771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811

static void icount_adjust(void)
{
    int64_t cur_time;
    int64_t cur_icount;
    int64_t delta;
    static int64_t last_delta;
    /* If the VM is not running, then do nothing.  */
    if (!vm_running)
        return;

    cur_time = cpu_get_clock();
    cur_icount = qemu_get_clock(vm_clock);
    delta = cur_icount - cur_time;
    /* FIXME: This is a very crude algorithm, somewhat prone to oscillation.  */
    if (delta > 0
        && last_delta + ICOUNT_WOBBLE < delta * 2
        && icount_time_shift > 0) {
        /* The guest is getting too far ahead.  Slow time down.  */
        icount_time_shift--;
    }
    if (delta < 0
        && last_delta - ICOUNT_WOBBLE > delta * 2
        && icount_time_shift < MAX_ICOUNT_SHIFT) {
        /* The guest is getting too far behind.  Speed time up.  */
        icount_time_shift++;
    }
    last_delta = delta;
    qemu_icount_bias = cur_icount - (qemu_icount << icount_time_shift);
}

static void icount_adjust_rt(void * opaque)
{
    qemu_mod_timer(icount_rt_timer,
                   qemu_get_clock(rt_clock) + 1000);
    icount_adjust();
}

static void icount_adjust_vm(void * opaque)
{
    qemu_mod_timer(icount_vm_timer,
812
                   qemu_get_clock(vm_clock) + get_ticks_per_sec() / 10);
P
pbrook 已提交
813 814 815 816 817 818 819 820 821 822 823 824 825 826 827
    icount_adjust();
}

static void init_icount_adjust(void)
{
    /* Have both realtime and virtual time triggers for speed adjustment.
       The realtime trigger catches emulated time passing too slowly,
       the virtual time trigger catches emulated time passing too fast.
       Realtime triggers occur even when idle, so use them less frequently
       than VM triggers.  */
    icount_rt_timer = qemu_new_timer(rt_clock, icount_adjust_rt, NULL);
    qemu_mod_timer(icount_rt_timer,
                   qemu_get_clock(rt_clock) + 1000);
    icount_vm_timer = qemu_new_timer(vm_clock, icount_adjust_vm, NULL);
    qemu_mod_timer(icount_vm_timer,
828
                   qemu_get_clock(vm_clock) + get_ticks_per_sec() / 10);
P
pbrook 已提交
829 830
}

831
static struct qemu_alarm_timer alarm_timers[] = {
832
#ifndef _WIN32
833
#ifdef __linux__
834 835
    {"dynticks", ALARM_FLAG_DYNTICKS, dynticks_start_timer,
     dynticks_stop_timer, dynticks_rearm_timer, NULL},
T
ths 已提交
836
    /* HPET - if available - is preferred */
837
    {"hpet", 0, hpet_start_timer, hpet_stop_timer, NULL, NULL},
T
ths 已提交
838
    /* ...otherwise try RTC */
839
    {"rtc", 0, rtc_start_timer, rtc_stop_timer, NULL, NULL},
840
#endif
841
    {"unix", 0, unix_start_timer, unix_stop_timer, NULL, NULL},
842
#else
843 844 845 846
    {"dynticks", ALARM_FLAG_DYNTICKS, win32_start_timer,
     win32_stop_timer, win32_rearm_timer, &alarm_win32_data},
    {"win32", 0, win32_start_timer,
     win32_stop_timer, NULL, &alarm_win32_data},
847 848 849 850
#endif
    {NULL, }
};

851
static void show_available_alarms(void)
852 853 854 855 856 857 858 859 860 861 862 863
{
    int i;

    printf("Available alarm timers, in order of precedence:\n");
    for (i = 0; alarm_timers[i].name; i++)
        printf("%s\n", alarm_timers[i].name);
}

static void configure_alarms(char const *opt)
{
    int i;
    int cur = 0;
864
    int count = ARRAY_SIZE(alarm_timers) - 1;
865 866
    char *arg;
    char *name;
P
pbrook 已提交
867
    struct qemu_alarm_timer tmp;
868

869
    if (!strcmp(opt, "?")) {
870 871 872 873
        show_available_alarms();
        exit(0);
    }

874
    arg = qemu_strdup(opt);
875 876 877 878

    /* Reorder the array */
    name = strtok(arg, ",");
    while (name) {
879
        for (i = 0; i < count && alarm_timers[i].name; i++) {
880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902
            if (!strcmp(alarm_timers[i].name, name))
                break;
        }

        if (i == count) {
            fprintf(stderr, "Unknown clock %s\n", name);
            goto next;
        }

        if (i < cur)
            /* Ignore */
            goto next;

	/* Swap */
        tmp = alarm_timers[i];
        alarm_timers[i] = alarm_timers[cur];
        alarm_timers[cur] = tmp;

        cur++;
next:
        name = strtok(NULL, ",");
    }

903
    qemu_free(arg);
904 905

    if (cur) {
P
pbrook 已提交
906
        /* Disable remaining timers */
907 908
        for (i = cur; i < count; i++)
            alarm_timers[i].name = NULL;
909 910 911
    } else {
        show_available_alarms();
        exit(1);
912 913 914
    }
}

J
Jan Kiszka 已提交
915 916
#define QEMU_NUM_CLOCKS 3

917 918
QEMUClock *rt_clock;
QEMUClock *vm_clock;
J
Jan Kiszka 已提交
919
QEMUClock *host_clock;
920

J
Jan Kiszka 已提交
921
static QEMUTimer *active_timers[QEMU_NUM_CLOCKS];
922

923
static QEMUClock *qemu_new_clock(int type)
924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982
{
    QEMUClock *clock;
    clock = qemu_mallocz(sizeof(QEMUClock));
    clock->type = type;
    return clock;
}

QEMUTimer *qemu_new_timer(QEMUClock *clock, QEMUTimerCB *cb, void *opaque)
{
    QEMUTimer *ts;

    ts = qemu_mallocz(sizeof(QEMUTimer));
    ts->clock = clock;
    ts->cb = cb;
    ts->opaque = opaque;
    return ts;
}

void qemu_free_timer(QEMUTimer *ts)
{
    qemu_free(ts);
}

/* stop a timer, but do not dealloc it */
void qemu_del_timer(QEMUTimer *ts)
{
    QEMUTimer **pt, *t;

    /* NOTE: this code must be signal safe because
       qemu_timer_expired() can be called from a signal. */
    pt = &active_timers[ts->clock->type];
    for(;;) {
        t = *pt;
        if (!t)
            break;
        if (t == ts) {
            *pt = t->next;
            break;
        }
        pt = &t->next;
    }
}

/* modify the current timer so that it will be fired when current_time
   >= expire_time. The corresponding callback will be called. */
void qemu_mod_timer(QEMUTimer *ts, int64_t expire_time)
{
    QEMUTimer **pt, *t;

    qemu_del_timer(ts);

    /* add the timer in the sorted list */
    /* NOTE: this code must be signal safe because
       qemu_timer_expired() can be called from a signal. */
    pt = &active_timers[ts->clock->type];
    for(;;) {
        t = *pt;
        if (!t)
            break;
983
        if (t->expire_time > expire_time)
984 985 986 987 988 989
            break;
        pt = &t->next;
    }
    ts->expire_time = expire_time;
    ts->next = *pt;
    *pt = ts;
990 991

    /* Rearm if necessary  */
P
pbrook 已提交
992 993 994 995 996
    if (pt == &active_timers[ts->clock->type]) {
        if ((alarm_timer->flags & ALARM_FLAG_EXPIRED) == 0) {
            qemu_rearm_alarm_timer(alarm_timer);
        }
        /* Interrupt execution to force deadline recalculation.  */
997 998
        if (use_icount)
            qemu_notify_event();
P
pbrook 已提交
999
    }
1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011
}

int qemu_timer_pending(QEMUTimer *ts)
{
    QEMUTimer *t;
    for(t = active_timers[ts->clock->type]; t != NULL; t = t->next) {
        if (t == ts)
            return 1;
    }
    return 0;
}

S
Stefano Stabellini 已提交
1012
int qemu_timer_expired(QEMUTimer *timer_head, int64_t current_time)
1013 1014 1015 1016 1017 1018 1019 1020 1021
{
    if (!timer_head)
        return 0;
    return (timer_head->expire_time <= current_time);
}

static void qemu_run_timers(QEMUTimer **ptimer_head, int64_t current_time)
{
    QEMUTimer *ts;
1022

1023 1024
    for(;;) {
        ts = *ptimer_head;
1025
        if (!ts || ts->expire_time > current_time)
1026 1027 1028 1029
            break;
        /* remove timer from the list before calling the callback */
        *ptimer_head = ts->next;
        ts->next = NULL;
1030

1031 1032 1033 1034 1035 1036 1037 1038
        /* run the callback (the timer list can be modified) */
        ts->cb(ts->opaque);
    }
}

int64_t qemu_get_clock(QEMUClock *clock)
{
    switch(clock->type) {
J
Jan Kiszka 已提交
1039
    case QEMU_CLOCK_REALTIME:
1040
        return get_clock() / 1000000;
1041
    default:
J
Jan Kiszka 已提交
1042
    case QEMU_CLOCK_VIRTUAL:
P
pbrook 已提交
1043 1044 1045 1046 1047
        if (use_icount) {
            return cpu_get_icount();
        } else {
            return cpu_get_clock();
        }
J
Jan Kiszka 已提交
1048 1049
    case QEMU_CLOCK_HOST:
        return get_clock_realtime();
1050 1051 1052
    }
}

J
Jan Kiszka 已提交
1053
static void init_clocks(void)
1054 1055
{
    init_get_clock();
J
Jan Kiszka 已提交
1056 1057
    rt_clock = qemu_new_clock(QEMU_CLOCK_REALTIME);
    vm_clock = qemu_new_clock(QEMU_CLOCK_VIRTUAL);
J
Jan Kiszka 已提交
1058
    host_clock = qemu_new_clock(QEMU_CLOCK_HOST);
J
Jan Kiszka 已提交
1059 1060

    rtc_clock = host_clock;
1061 1062
}

1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087
/* save a timer */
void qemu_put_timer(QEMUFile *f, QEMUTimer *ts)
{
    uint64_t expire_time;

    if (qemu_timer_pending(ts)) {
        expire_time = ts->expire_time;
    } else {
        expire_time = -1;
    }
    qemu_put_be64(f, expire_time);
}

void qemu_get_timer(QEMUFile *f, QEMUTimer *ts)
{
    uint64_t expire_time;

    expire_time = qemu_get_be64(f);
    if (expire_time != -1) {
        qemu_mod_timer(ts, expire_time);
    } else {
        qemu_del_timer(ts);
    }
}

J
Juan Quintela 已提交
1088 1089 1090 1091 1092 1093 1094
static const VMStateDescription vmstate_timers = {
    .name = "timer",
    .version_id = 2,
    .minimum_version_id = 1,
    .minimum_version_id_old = 1,
    .fields      = (VMStateField []) {
        VMSTATE_INT64(cpu_ticks_offset, TimersState),
1095
        VMSTATE_INT64(dummy, TimersState),
J
Juan Quintela 已提交
1096 1097
        VMSTATE_INT64_V(cpu_clock_offset, TimersState, 2),
        VMSTATE_END_OF_LIST()
1098
    }
J
Juan Quintela 已提交
1099
};
1100

1101 1102
static void qemu_event_increment(void);

B
bellard 已提交
1103
#ifdef _WIN32
B
blueswir1 已提交
1104 1105 1106
static void CALLBACK host_alarm_handler(UINT uTimerID, UINT uMsg,
                                        DWORD_PTR dwUser, DWORD_PTR dw1,
                                        DWORD_PTR dw2)
B
bellard 已提交
1107
#else
1108
static void host_alarm_handler(int host_signum)
B
bellard 已提交
1109
#endif
1110
{
1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125
#if 0
#define DISP_FREQ 1000
    {
        static int64_t delta_min = INT64_MAX;
        static int64_t delta_max, delta_cum, last_clock, delta, ti;
        static int count;
        ti = qemu_get_clock(vm_clock);
        if (last_clock != 0) {
            delta = ti - last_clock;
            if (delta < delta_min)
                delta_min = delta;
            if (delta > delta_max)
                delta_max = delta;
            delta_cum += delta;
            if (++count == DISP_FREQ) {
B
bellard 已提交
1126
                printf("timer: min=%" PRId64 " us max=%" PRId64 " us avg=%" PRId64 " us avg_freq=%0.3f Hz\n",
1127 1128 1129 1130
                       muldiv64(delta_min, 1000000, get_ticks_per_sec()),
                       muldiv64(delta_max, 1000000, get_ticks_per_sec()),
                       muldiv64(delta_cum, 1000000 / DISP_FREQ, get_ticks_per_sec()),
                       (double)get_ticks_per_sec() / ((double)delta_cum / DISP_FREQ));
1131 1132 1133 1134 1135 1136 1137 1138 1139
                count = 0;
                delta_min = INT64_MAX;
                delta_max = 0;
                delta_cum = 0;
            }
        }
        last_clock = ti;
    }
#endif
1140
    if (alarm_has_dynticks(alarm_timer) ||
P
pbrook 已提交
1141
        (!use_icount &&
J
Jan Kiszka 已提交
1142
            qemu_timer_expired(active_timers[QEMU_CLOCK_VIRTUAL],
P
pbrook 已提交
1143
                               qemu_get_clock(vm_clock))) ||
J
Jan Kiszka 已提交
1144
        qemu_timer_expired(active_timers[QEMU_CLOCK_REALTIME],
J
Jan Kiszka 已提交
1145 1146 1147
                           qemu_get_clock(rt_clock)) ||
        qemu_timer_expired(active_timers[QEMU_CLOCK_HOST],
                           qemu_get_clock(host_clock))) {
1148
        qemu_event_increment();
1149
        if (alarm_timer) alarm_timer->flags |= ALARM_FLAG_EXPIRED;
1150

1151 1152
#ifndef CONFIG_IOTHREAD
        if (next_cpu) {
1153
            /* stop the currently executing cpu because a timer occured */
1154
            cpu_exit(next_cpu);
1155
        }
1156
#endif
1157
        timer_alarm_pending = 1;
1158
        qemu_notify_event();
1159 1160 1161
    }
}

P
pbrook 已提交
1162
static int64_t qemu_next_deadline(void)
1163
{
J
Jan Kiszka 已提交
1164 1165
    /* To avoid problems with overflow limit this to 2^32.  */
    int64_t delta = INT32_MAX;
1166

J
Jan Kiszka 已提交
1167 1168
    if (active_timers[QEMU_CLOCK_VIRTUAL]) {
        delta = active_timers[QEMU_CLOCK_VIRTUAL]->expire_time -
P
pbrook 已提交
1169
                     qemu_get_clock(vm_clock);
J
Jan Kiszka 已提交
1170 1171 1172 1173 1174 1175
    }
    if (active_timers[QEMU_CLOCK_HOST]) {
        int64_t hdelta = active_timers[QEMU_CLOCK_HOST]->expire_time -
                 qemu_get_clock(host_clock);
        if (hdelta < delta)
            delta = hdelta;
1176 1177
    }

P
pbrook 已提交
1178 1179
    if (delta < 0)
        delta = 0;
1180

P
pbrook 已提交
1181 1182 1183
    return delta;
}

J
Jan Kiszka 已提交
1184
#if defined(__linux__)
P
pbrook 已提交
1185 1186 1187 1188 1189 1190 1191 1192 1193 1194
static uint64_t qemu_next_deadline_dyntick(void)
{
    int64_t delta;
    int64_t rtdelta;

    if (use_icount)
        delta = INT32_MAX;
    else
        delta = (qemu_next_deadline() + 999) / 1000;

J
Jan Kiszka 已提交
1195 1196
    if (active_timers[QEMU_CLOCK_REALTIME]) {
        rtdelta = (active_timers[QEMU_CLOCK_REALTIME]->expire_time -
P
pbrook 已提交
1197 1198 1199 1200 1201 1202 1203 1204 1205
                 qemu_get_clock(rt_clock))*1000;
        if (rtdelta < delta)
            delta = rtdelta;
    }

    if (delta < MIN_TIMER_REARM_US)
        delta = MIN_TIMER_REARM_US;

    return delta;
1206
}
1207
#endif
1208

1209 1210
#ifndef _WIN32

1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225
/* Sets a specific flag */
static int fcntl_setfl(int fd, int flag)
{
    int flags;

    flags = fcntl(fd, F_GETFL);
    if (flags == -1)
        return -errno;

    if (fcntl(fd, F_SETFL, flags | flag) == -1)
        return -errno;

    return 0;
}

B
BSD fix  
bellard 已提交
1226 1227
#if defined(__linux__)

1228 1229
#define RTC_FREQ 1024

1230
static void enable_sigio_timer(int fd)
1231 1232 1233 1234 1235 1236 1237 1238 1239
{
    struct sigaction act;

    /* timer signal */
    sigfillset(&act.sa_mask);
    act.sa_flags = 0;
    act.sa_handler = host_alarm_handler;

    sigaction(SIGIO, &act, NULL);
1240
    fcntl_setfl(fd, O_ASYNC);
1241 1242
    fcntl(fd, F_SETOWN, getpid());
}
B
BSD fix  
bellard 已提交
1243

T
ths 已提交
1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277
static int hpet_start_timer(struct qemu_alarm_timer *t)
{
    struct hpet_info info;
    int r, fd;

    fd = open("/dev/hpet", O_RDONLY);
    if (fd < 0)
        return -1;

    /* Set frequency */
    r = ioctl(fd, HPET_IRQFREQ, RTC_FREQ);
    if (r < 0) {
        fprintf(stderr, "Could not configure '/dev/hpet' to have a 1024Hz timer. This is not a fatal\n"
                "error, but for better emulation accuracy type:\n"
                "'echo 1024 > /proc/sys/dev/hpet/max-user-freq' as root.\n");
        goto fail;
    }

    /* Check capabilities */
    r = ioctl(fd, HPET_INFO, &info);
    if (r < 0)
        goto fail;

    /* Enable periodic mode */
    r = ioctl(fd, HPET_EPI, 0);
    if (info.hi_flags && (r < 0))
        goto fail;

    /* Enable interrupt */
    r = ioctl(fd, HPET_IE_ON, 0);
    if (r < 0)
        goto fail;

    enable_sigio_timer(fd);
1278
    t->priv = (void *)(long)fd;
T
ths 已提交
1279 1280 1281 1282 1283 1284 1285 1286 1287

    return 0;
fail:
    close(fd);
    return -1;
}

static void hpet_stop_timer(struct qemu_alarm_timer *t)
{
1288
    int fd = (long)t->priv;
T
ths 已提交
1289 1290 1291 1292

    close(fd);
}

1293
static int rtc_start_timer(struct qemu_alarm_timer *t)
1294
{
1295
    int rtc_fd;
1296
    unsigned long current_rtc_freq = 0;
1297

1298
    TFR(rtc_fd = open("/dev/rtc", O_RDONLY));
1299 1300
    if (rtc_fd < 0)
        return -1;
1301 1302 1303
    ioctl(rtc_fd, RTC_IRQP_READ, &current_rtc_freq);
    if (current_rtc_freq != RTC_FREQ &&
        ioctl(rtc_fd, RTC_IRQP_SET, RTC_FREQ) < 0) {
1304 1305 1306 1307 1308 1309 1310 1311 1312 1313
        fprintf(stderr, "Could not configure '/dev/rtc' to have a 1024 Hz timer. This is not a fatal\n"
                "error, but for better emulation accuracy either use a 2.6 host Linux kernel or\n"
                "type 'echo 1024 > /proc/sys/dev/rtc/max-user-freq' as root.\n");
        goto fail;
    }
    if (ioctl(rtc_fd, RTC_PIE_ON, 0) < 0) {
    fail:
        close(rtc_fd);
        return -1;
    }
1314 1315 1316

    enable_sigio_timer(rtc_fd);

1317
    t->priv = (void *)(long)rtc_fd;
1318

1319 1320 1321
    return 0;
}

1322
static void rtc_stop_timer(struct qemu_alarm_timer *t)
B
BSD fix  
bellard 已提交
1323
{
1324
    int rtc_fd = (long)t->priv;
1325 1326

    close(rtc_fd);
B
BSD fix  
bellard 已提交
1327 1328
}

1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340
static int dynticks_start_timer(struct qemu_alarm_timer *t)
{
    struct sigevent ev;
    timer_t host_timer;
    struct sigaction act;

    sigfillset(&act.sa_mask);
    act.sa_flags = 0;
    act.sa_handler = host_alarm_handler;

    sigaction(SIGALRM, &act, NULL);

1341 1342 1343 1344 1345
    /* 
     * Initialize ev struct to 0 to avoid valgrind complaining
     * about uninitialized data in timer_create call
     */
    memset(&ev, 0, sizeof(ev));
1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358
    ev.sigev_value.sival_int = 0;
    ev.sigev_notify = SIGEV_SIGNAL;
    ev.sigev_signo = SIGALRM;

    if (timer_create(CLOCK_REALTIME, &ev, &host_timer)) {
        perror("timer_create");

        /* disable dynticks */
        fprintf(stderr, "Dynamic Ticks disabled\n");

        return -1;
    }

1359
    t->priv = (void *)(long)host_timer;
1360 1361 1362 1363 1364 1365

    return 0;
}

static void dynticks_stop_timer(struct qemu_alarm_timer *t)
{
1366
    timer_t host_timer = (timer_t)(long)t->priv;
1367 1368 1369 1370 1371 1372

    timer_delete(host_timer);
}

static void dynticks_rearm_timer(struct qemu_alarm_timer *t)
{
1373
    timer_t host_timer = (timer_t)(long)t->priv;
1374 1375 1376 1377
    struct itimerspec timeout;
    int64_t nearest_delta_us = INT64_MAX;
    int64_t current_us;

J
Jan Kiszka 已提交
1378
    if (!active_timers[QEMU_CLOCK_REALTIME] &&
J
Jan Kiszka 已提交
1379 1380
        !active_timers[QEMU_CLOCK_VIRTUAL] &&
        !active_timers[QEMU_CLOCK_HOST])
1381
        return;
1382

P
pbrook 已提交
1383
    nearest_delta_us = qemu_next_deadline_dyntick();
1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405

    /* check whether a timer is already running */
    if (timer_gettime(host_timer, &timeout)) {
        perror("gettime");
        fprintf(stderr, "Internal timer error: aborting\n");
        exit(1);
    }
    current_us = timeout.it_value.tv_sec * 1000000 + timeout.it_value.tv_nsec/1000;
    if (current_us && current_us <= nearest_delta_us)
        return;

    timeout.it_interval.tv_sec = 0;
    timeout.it_interval.tv_nsec = 0; /* 0 for one-shot timer */
    timeout.it_value.tv_sec =  nearest_delta_us / 1000000;
    timeout.it_value.tv_nsec = (nearest_delta_us % 1000000) * 1000;
    if (timer_settime(host_timer, 0 /* RELATIVE */, &timeout, NULL)) {
        perror("settime");
        fprintf(stderr, "Internal timer error: aborting\n");
        exit(1);
    }
}

T
ths 已提交
1406
#endif /* defined(__linux__) */
1407

1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441
static int unix_start_timer(struct qemu_alarm_timer *t)
{
    struct sigaction act;
    struct itimerval itv;
    int err;

    /* timer signal */
    sigfillset(&act.sa_mask);
    act.sa_flags = 0;
    act.sa_handler = host_alarm_handler;

    sigaction(SIGALRM, &act, NULL);

    itv.it_interval.tv_sec = 0;
    /* for i386 kernel 2.6 to get 1 ms */
    itv.it_interval.tv_usec = 999;
    itv.it_value.tv_sec = 0;
    itv.it_value.tv_usec = 10 * 1000;

    err = setitimer(ITIMER_REAL, &itv, NULL);
    if (err)
        return -1;

    return 0;
}

static void unix_stop_timer(struct qemu_alarm_timer *t)
{
    struct itimerval itv;

    memset(&itv, 0, sizeof(itv));
    setitimer(ITIMER_REAL, &itv, NULL);
}

B
BSD fix  
bellard 已提交
1442
#endif /* !defined(_WIN32) */
1443

1444

1445 1446 1447 1448 1449 1450
#ifdef _WIN32

static int win32_start_timer(struct qemu_alarm_timer *t)
{
    TIMECAPS tc;
    struct qemu_alarm_win32 *data = t->priv;
1451
    UINT flags;
1452 1453 1454 1455 1456 1457 1458 1459 1460

    memset(&tc, 0, sizeof(tc));
    timeGetDevCaps(&tc, sizeof(tc));

    if (data->period < tc.wPeriodMin)
        data->period = tc.wPeriodMin;

    timeBeginPeriod(data->period);

1461 1462 1463 1464 1465 1466
    flags = TIME_CALLBACK_FUNCTION;
    if (alarm_has_dynticks(t))
        flags |= TIME_ONESHOT;
    else
        flags |= TIME_PERIODIC;

1467 1468 1469 1470
    data->timerId = timeSetEvent(1,         // interval (ms)
                        data->period,       // resolution
                        host_alarm_handler, // function
                        (DWORD)t,           // parameter
1471
                        flags);
1472 1473

    if (!data->timerId) {
B
Blue Swirl 已提交
1474
        fprintf(stderr, "Failed to initialize win32 alarm timer: %ld\n",
1475
                GetLastError());
1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490
        timeEndPeriod(data->period);
        return -1;
    }

    return 0;
}

static void win32_stop_timer(struct qemu_alarm_timer *t)
{
    struct qemu_alarm_win32 *data = t->priv;

    timeKillEvent(data->timerId);
    timeEndPeriod(data->period);
}

1491 1492 1493 1494
static void win32_rearm_timer(struct qemu_alarm_timer *t)
{
    struct qemu_alarm_win32 *data = t->priv;

J
Jan Kiszka 已提交
1495
    if (!active_timers[QEMU_CLOCK_REALTIME] &&
J
Jan Kiszka 已提交
1496 1497
        !active_timers[QEMU_CLOCK_VIRTUAL] &&
        !active_timers[QEMU_CLOCK_HOST])
1498
        return;
1499 1500 1501 1502 1503 1504 1505 1506 1507 1508

    timeKillEvent(data->timerId);

    data->timerId = timeSetEvent(1,
                        data->period,
                        host_alarm_handler,
                        (DWORD)t,
                        TIME_ONESHOT | TIME_PERIODIC);

    if (!data->timerId) {
B
Blue Swirl 已提交
1509
        fprintf(stderr, "Failed to re-arm win32 alarm timer %ld\n",
1510
                GetLastError());
1511 1512 1513 1514 1515 1516

        timeEndPeriod(data->period);
        exit(1);
    }
}

1517 1518
#endif /* _WIN32 */

1519
static int init_timer_alarm(void)
1520
{
1521
    struct qemu_alarm_timer *t = NULL;
1522
    int i, err = -1;
1523

1524 1525 1526 1527 1528 1529
    for (i = 0; alarm_timers[i].name; i++) {
        t = &alarm_timers[i];

        err = t->start(t);
        if (!err)
            break;
B
bellard 已提交
1530
    }
1531

1532
    if (err) {
1533 1534
        err = -ENOENT;
        goto fail;
B
bellard 已提交
1535
    }
1536 1537

    alarm_timer = t;
1538

1539
    return 0;
1540 1541 1542

fail:
    return err;
1543 1544
}

1545
static void quit_timers(void)
B
bellard 已提交
1546
{
1547 1548
    alarm_timer->stop(alarm_timer);
    alarm_timer = NULL;
B
bellard 已提交
1549 1550
}

1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587
/***********************************************************/
/* host time/date access */
void qemu_get_timedate(struct tm *tm, int offset)
{
    time_t ti;
    struct tm *ret;

    time(&ti);
    ti += offset;
    if (rtc_date_offset == -1) {
        if (rtc_utc)
            ret = gmtime(&ti);
        else
            ret = localtime(&ti);
    } else {
        ti -= rtc_date_offset;
        ret = gmtime(&ti);
    }

    memcpy(tm, ret, sizeof(struct tm));
}

int qemu_timedate_diff(struct tm *tm)
{
    time_t seconds;

    if (rtc_date_offset == -1)
        if (rtc_utc)
            seconds = mktimegm(tm);
        else
            seconds = mktime(tm);
    else
        seconds = mktimegm(tm) + rtc_date_offset;

    return seconds - time(NULL);
}

J
Jan Kiszka 已提交
1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640
static void configure_rtc_date_offset(const char *startdate, int legacy)
{
    time_t rtc_start_date;
    struct tm tm;

    if (!strcmp(startdate, "now") && legacy) {
        rtc_date_offset = -1;
    } else {
        if (sscanf(startdate, "%d-%d-%dT%d:%d:%d",
                   &tm.tm_year,
                   &tm.tm_mon,
                   &tm.tm_mday,
                   &tm.tm_hour,
                   &tm.tm_min,
                   &tm.tm_sec) == 6) {
            /* OK */
        } else if (sscanf(startdate, "%d-%d-%d",
                          &tm.tm_year,
                          &tm.tm_mon,
                          &tm.tm_mday) == 3) {
            tm.tm_hour = 0;
            tm.tm_min = 0;
            tm.tm_sec = 0;
        } else {
            goto date_fail;
        }
        tm.tm_year -= 1900;
        tm.tm_mon--;
        rtc_start_date = mktimegm(&tm);
        if (rtc_start_date == -1) {
        date_fail:
            fprintf(stderr, "Invalid date format. Valid formats are:\n"
                            "'2006-06-17T16:01:21' or '2006-06-17'\n");
            exit(1);
        }
        rtc_date_offset = time(NULL) - rtc_start_date;
    }
}

static void configure_rtc(QemuOpts *opts)
{
    const char *value;

    value = qemu_opt_get(opts, "base");
    if (value) {
        if (!strcmp(value, "utc")) {
            rtc_utc = 1;
        } else if (!strcmp(value, "localtime")) {
            rtc_utc = 0;
        } else {
            configure_rtc_date_offset(value, 0);
        }
    }
J
Jan Kiszka 已提交
1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651
    value = qemu_opt_get(opts, "clock");
    if (value) {
        if (!strcmp(value, "host")) {
            rtc_clock = host_clock;
        } else if (!strcmp(value, "vm")) {
            rtc_clock = vm_clock;
        } else {
            fprintf(stderr, "qemu: invalid option value '%s'\n", value);
            exit(1);
        }
    }
J
Jan Kiszka 已提交
1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666
#ifdef CONFIG_TARGET_I386
    value = qemu_opt_get(opts, "driftfix");
    if (value) {
        if (!strcmp(buf, "slew")) {
            rtc_td_hack = 1;
        } else if (!strcmp(buf, "none")) {
            rtc_td_hack = 0;
        } else {
            fprintf(stderr, "qemu: invalid option value '%s'\n", value);
            exit(1);
        }
    }
#endif
}

B
bellard 已提交
1667 1668 1669 1670 1671
#ifdef _WIN32
static void socket_cleanup(void)
{
    WSACleanup();
}
B
bellard 已提交
1672

B
bellard 已提交
1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686
static int socket_init(void)
{
    WSADATA Data;
    int ret, err;

    ret = WSAStartup(MAKEWORD(2,2), &Data);
    if (ret != 0) {
        err = WSAGetLastError();
        fprintf(stderr, "WSAStartup: %d\n", err);
        return -1;
    }
    atexit(socket_cleanup);
    return 0;
}
1687 1688
#endif

1689 1690 1691 1692 1693
/***********************************************************/
/* Bluetooth support */
static int nb_hcis;
static int cur_hci;
static struct HCIInfo *hci_table[MAX_NICS];
1694

1695 1696 1697 1698 1699 1700 1701
static struct bt_vlan_s {
    struct bt_scatternet_s net;
    int id;
    struct bt_vlan_s *next;
} *first_bt_vlan;

/* find or alloc a new bluetooth "VLAN" */
B
blueswir1 已提交
1702
static struct bt_scatternet_s *qemu_find_bt_vlan(int id)
1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741
{
    struct bt_vlan_s **pvlan, *vlan;
    for (vlan = first_bt_vlan; vlan != NULL; vlan = vlan->next) {
        if (vlan->id == id)
            return &vlan->net;
    }
    vlan = qemu_mallocz(sizeof(struct bt_vlan_s));
    vlan->id = id;
    pvlan = &first_bt_vlan;
    while (*pvlan != NULL)
        pvlan = &(*pvlan)->next;
    *pvlan = vlan;
    return &vlan->net;
}

static void null_hci_send(struct HCIInfo *hci, const uint8_t *data, int len)
{
}

static int null_hci_addr_set(struct HCIInfo *hci, const uint8_t *bd_addr)
{
    return -ENOTSUP;
}

static struct HCIInfo null_hci = {
    .cmd_send = null_hci_send,
    .sco_send = null_hci_send,
    .acl_send = null_hci_send,
    .bdaddr_set = null_hci_addr_set,
};

struct HCIInfo *qemu_next_hci(void)
{
    if (cur_hci == nb_hcis)
        return &null_hci;

    return hci_table[cur_hci++];
}

1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774
static struct HCIInfo *hci_init(const char *str)
{
    char *endp;
    struct bt_scatternet_s *vlan = 0;

    if (!strcmp(str, "null"))
        /* null */
        return &null_hci;
    else if (!strncmp(str, "host", 4) && (str[4] == '\0' || str[4] == ':'))
        /* host[:hciN] */
        return bt_host_hci(str[4] ? str + 5 : "hci0");
    else if (!strncmp(str, "hci", 3)) {
        /* hci[,vlan=n] */
        if (str[3]) {
            if (!strncmp(str + 3, ",vlan=", 6)) {
                vlan = qemu_find_bt_vlan(strtol(str + 9, &endp, 0));
                if (*endp)
                    vlan = 0;
            }
        } else
            vlan = qemu_find_bt_vlan(0);
        if (vlan)
           return bt_new_hci(vlan);
    }

    fprintf(stderr, "qemu: Unknown bluetooth HCI `%s'.\n", str);

    return 0;
}

static int bt_hci_parse(const char *str)
{
    struct HCIInfo *hci;
A
Anthony Liguori 已提交
1775
    bdaddr_t bdaddr;
1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879

    if (nb_hcis >= MAX_NICS) {
        fprintf(stderr, "qemu: Too many bluetooth HCIs (max %i).\n", MAX_NICS);
        return -1;
    }

    hci = hci_init(str);
    if (!hci)
        return -1;

    bdaddr.b[0] = 0x52;
    bdaddr.b[1] = 0x54;
    bdaddr.b[2] = 0x00;
    bdaddr.b[3] = 0x12;
    bdaddr.b[4] = 0x34;
    bdaddr.b[5] = 0x56 + nb_hcis;
    hci->bdaddr_set(hci, bdaddr.b);

    hci_table[nb_hcis++] = hci;

    return 0;
}

static void bt_vhci_add(int vlan_id)
{
    struct bt_scatternet_s *vlan = qemu_find_bt_vlan(vlan_id);

    if (!vlan->slave)
        fprintf(stderr, "qemu: warning: adding a VHCI to "
                        "an empty scatternet %i\n", vlan_id);

    bt_vhci_init(bt_new_hci(vlan));
}

static struct bt_device_s *bt_device_add(const char *opt)
{
    struct bt_scatternet_s *vlan;
    int vlan_id = 0;
    char *endp = strstr(opt, ",vlan=");
    int len = (endp ? endp - opt : strlen(opt)) + 1;
    char devname[10];

    pstrcpy(devname, MIN(sizeof(devname), len), opt);

    if (endp) {
        vlan_id = strtol(endp + 6, &endp, 0);
        if (*endp) {
            fprintf(stderr, "qemu: unrecognised bluetooth vlan Id\n");
            return 0;
        }
    }

    vlan = qemu_find_bt_vlan(vlan_id);

    if (!vlan->slave)
        fprintf(stderr, "qemu: warning: adding a slave device to "
                        "an empty scatternet %i\n", vlan_id);

    if (!strcmp(devname, "keyboard"))
        return bt_keyboard_init(vlan);

    fprintf(stderr, "qemu: unsupported bluetooth device `%s'\n", devname);
    return 0;
}

static int bt_parse(const char *opt)
{
    const char *endp, *p;
    int vlan;

    if (strstart(opt, "hci", &endp)) {
        if (!*endp || *endp == ',') {
            if (*endp)
                if (!strstart(endp, ",vlan=", 0))
                    opt = endp + 1;

            return bt_hci_parse(opt);
       }
    } else if (strstart(opt, "vhci", &endp)) {
        if (!*endp || *endp == ',') {
            if (*endp) {
                if (strstart(endp, ",vlan=", &p)) {
                    vlan = strtol(p, (char **) &endp, 0);
                    if (*endp) {
                        fprintf(stderr, "qemu: bad scatternet '%s'\n", p);
                        return 1;
                    }
                } else {
                    fprintf(stderr, "qemu: bad parameter '%s'\n", endp + 1);
                    return 1;
                }
            } else
                vlan = 0;

            bt_vhci_add(vlan);
            return 0;
        }
    } else if (strstart(opt, "device:", &endp))
        return !bt_device_add(endp);

    fprintf(stderr, "qemu: bad bluetooth parameter '%s'\n", opt);
    return 1;
}

1880 1881 1882
/***********************************************************/
/* QEMU Block devices */

1883
#define HD_ALIAS "index=%d,media=disk"
T
ths 已提交
1884 1885
#define CDROM_ALIAS "index=2,media=cdrom"
#define FD_ALIAS "index=%d,if=floppy"
1886 1887
#define PFLASH_ALIAS "if=pflash"
#define MTD_ALIAS "if=mtd"
1888
#define SD_ALIAS "index=0,if=sd"
T
ths 已提交
1889

G
Gerd Hoffmann 已提交
1890
QemuOpts *drive_add(const char *file, const char *fmt, ...)
T
ths 已提交
1891 1892
{
    va_list ap;
G
Gerd Hoffmann 已提交
1893 1894
    char optstr[1024];
    QemuOpts *opts;
T
ths 已提交
1895 1896

    va_start(ap, fmt);
G
Gerd Hoffmann 已提交
1897
    vsnprintf(optstr, sizeof(optstr), fmt, ap);
T
ths 已提交
1898 1899
    va_end(ap);

G
Gerd Hoffmann 已提交
1900
    opts = qemu_opts_parse(&qemu_drive_opts, optstr, NULL);
G
Gerd Hoffmann 已提交
1901 1902 1903 1904 1905 1906 1907 1908
    if (!opts) {
        fprintf(stderr, "%s: huh? duplicate? (%s)\n",
                __FUNCTION__, optstr);
        return NULL;
    }
    if (file)
        qemu_opt_set(opts, "file", file);
    return opts;
1909 1910
}

G
Gerd Hoffmann 已提交
1911
DriveInfo *drive_get(BlockInterfaceType type, int bus, int unit)
T
ths 已提交
1912
{
G
Gerd Hoffmann 已提交
1913
    DriveInfo *dinfo;
T
ths 已提交
1914 1915 1916

    /* seek interface, bus and unit */

B
Blue Swirl 已提交
1917
    QTAILQ_FOREACH(dinfo, &drives, next) {
G
Gerd Hoffmann 已提交
1918 1919 1920 1921 1922
        if (dinfo->type == type &&
	    dinfo->bus == bus &&
	    dinfo->unit == unit)
            return dinfo;
    }
T
ths 已提交
1923

G
Gerd Hoffmann 已提交
1924
    return NULL;
T
ths 已提交
1925 1926
}

G
Gerd Hoffmann 已提交
1927
DriveInfo *drive_get_by_id(const char *id)
G
Gerd Hoffmann 已提交
1928 1929 1930
{
    DriveInfo *dinfo;

B
Blue Swirl 已提交
1931
    QTAILQ_FOREACH(dinfo, &drives, next) {
G
Gerd Hoffmann 已提交
1932 1933 1934 1935 1936 1937 1938
        if (strcmp(id, dinfo->id))
            continue;
        return dinfo;
    }
    return NULL;
}

1939
int drive_get_max_bus(BlockInterfaceType type)
T
ths 已提交
1940 1941
{
    int max_bus;
G
Gerd Hoffmann 已提交
1942
    DriveInfo *dinfo;
T
ths 已提交
1943 1944

    max_bus = -1;
B
Blue Swirl 已提交
1945
    QTAILQ_FOREACH(dinfo, &drives, next) {
G
Gerd Hoffmann 已提交
1946 1947 1948
        if(dinfo->type == type &&
           dinfo->bus > max_bus)
            max_bus = dinfo->bus;
T
ths 已提交
1949 1950 1951 1952
    }
    return max_bus;
}

1953 1954
const char *drive_get_serial(BlockDriverState *bdrv)
{
G
Gerd Hoffmann 已提交
1955
    DriveInfo *dinfo;
1956

B
Blue Swirl 已提交
1957
    QTAILQ_FOREACH(dinfo, &drives, next) {
G
Gerd Hoffmann 已提交
1958 1959 1960
        if (dinfo->bdrv == bdrv)
            return dinfo->serial;
    }
1961 1962 1963 1964

    return "\0";
}

1965 1966
BlockInterfaceErrorAction drive_get_onerror(BlockDriverState *bdrv)
{
G
Gerd Hoffmann 已提交
1967
    DriveInfo *dinfo;
1968

B
Blue Swirl 已提交
1969
    QTAILQ_FOREACH(dinfo, &drives, next) {
G
Gerd Hoffmann 已提交
1970 1971 1972
        if (dinfo->bdrv == bdrv)
            return dinfo->onerror;
    }
1973

1974
    return BLOCK_ERR_STOP_ENOSPC;
1975 1976
}

1977 1978 1979 1980 1981
static void bdrv_format_print(void *opaque, const char *name)
{
    fprintf(stderr, " %s", name);
}

1982 1983
void drive_uninit(BlockDriverState *bdrv)
{
G
Gerd Hoffmann 已提交
1984
    DriveInfo *dinfo;
1985

B
Blue Swirl 已提交
1986
    QTAILQ_FOREACH(dinfo, &drives, next) {
G
Gerd Hoffmann 已提交
1987 1988
        if (dinfo->bdrv != bdrv)
            continue;
G
Gerd Hoffmann 已提交
1989
        qemu_opts_del(dinfo->opts);
B
Blue Swirl 已提交
1990
        QTAILQ_REMOVE(&drives, dinfo, next);
G
Gerd Hoffmann 已提交
1991 1992 1993
        qemu_free(dinfo);
        break;
    }
1994 1995
}

G
Gerd Hoffmann 已提交
1996
DriveInfo *drive_init(QemuOpts *opts, void *opaque,
G
Gerd Hoffmann 已提交
1997
                      int *fatal_error)
T
ths 已提交
1998
{
G
Gerd Hoffmann 已提交
1999 2000
    const char *buf;
    const char *file = NULL;
2001
    char devname[128];
G
Gerd Hoffmann 已提交
2002
    const char *serial;
2003
    const char *mediastr = "";
2004
    BlockInterfaceType type;
T
ths 已提交
2005 2006 2007
    enum { MEDIA_DISK, MEDIA_CDROM } media;
    int bus_id, unit_id;
    int cyls, heads, secs, translation;
2008
    BlockDriver *drv = NULL;
2009
    QEMUMachine *machine = opaque;
T
ths 已提交
2010 2011
    int max_devs;
    int index;
2012
    int cache;
2013
    int aio = 0;
2014
    int bdrv_flags, onerror;
2015
    const char *devaddr;
G
Gerd Hoffmann 已提交
2016
    DriveInfo *dinfo;
G
Gerd Hoffmann 已提交
2017
    int snapshot = 0;
T
ths 已提交
2018

G
Gerd Hoffmann 已提交
2019
    *fatal_error = 1;
T
ths 已提交
2020 2021

    translation = BIOS_ATA_TRANSLATION_AUTO;
2022
    cache = 1;
T
ths 已提交
2023

2024
    if (machine && machine->use_scsi) {
2025
        type = IF_SCSI;
T
ths 已提交
2026
        max_devs = MAX_SCSI_DEVS;
B
blueswir1 已提交
2027
        pstrcpy(devname, sizeof(devname), "scsi");
T
ths 已提交
2028
    } else {
2029
        type = IF_IDE;
T
ths 已提交
2030
        max_devs = MAX_IDE_DEVS;
B
blueswir1 已提交
2031
        pstrcpy(devname, sizeof(devname), "ide");
T
ths 已提交
2032 2033 2034 2035
    }
    media = MEDIA_DISK;

    /* extract parameters */
G
Gerd Hoffmann 已提交
2036 2037 2038
    bus_id  = qemu_opt_get_number(opts, "bus", 0);
    unit_id = qemu_opt_get_number(opts, "unit", -1);
    index   = qemu_opt_get_number(opts, "index", -1);
T
ths 已提交
2039

G
Gerd Hoffmann 已提交
2040 2041 2042
    cyls  = qemu_opt_get_number(opts, "cyls", 0);
    heads = qemu_opt_get_number(opts, "heads", 0);
    secs  = qemu_opt_get_number(opts, "secs", 0);
T
ths 已提交
2043

G
Gerd Hoffmann 已提交
2044
    snapshot = qemu_opt_get_bool(opts, "snapshot", 0);
T
ths 已提交
2045

G
Gerd Hoffmann 已提交
2046 2047 2048 2049
    file = qemu_opt_get(opts, "file");
    serial = qemu_opt_get(opts, "serial");

    if ((buf = qemu_opt_get(opts, "if")) != NULL) {
B
bellard 已提交
2050
        pstrcpy(devname, sizeof(devname), buf);
T
ths 已提交
2051
        if (!strcmp(buf, "ide")) {
2052
	    type = IF_IDE;
T
ths 已提交
2053 2054
            max_devs = MAX_IDE_DEVS;
        } else if (!strcmp(buf, "scsi")) {
2055
	    type = IF_SCSI;
T
ths 已提交
2056 2057
            max_devs = MAX_SCSI_DEVS;
        } else if (!strcmp(buf, "floppy")) {
2058
	    type = IF_FLOPPY;
T
ths 已提交
2059 2060
            max_devs = 0;
        } else if (!strcmp(buf, "pflash")) {
2061
	    type = IF_PFLASH;
T
ths 已提交
2062 2063
            max_devs = 0;
	} else if (!strcmp(buf, "mtd")) {
2064
	    type = IF_MTD;
T
ths 已提交
2065 2066
            max_devs = 0;
	} else if (!strcmp(buf, "sd")) {
2067
	    type = IF_SD;
T
ths 已提交
2068
            max_devs = 0;
A
aliguori 已提交
2069 2070 2071
        } else if (!strcmp(buf, "virtio")) {
            type = IF_VIRTIO;
            max_devs = 0;
2072 2073 2074
	} else if (!strcmp(buf, "xen")) {
	    type = IF_XEN;
            max_devs = 0;
G
Gerd Hoffmann 已提交
2075 2076 2077
	} else if (!strcmp(buf, "none")) {
	    type = IF_NONE;
            max_devs = 0;
2078
	} else {
G
Gerd Hoffmann 已提交
2079
            fprintf(stderr, "qemu: unsupported bus type '%s'\n", buf);
G
Gerd Hoffmann 已提交
2080
            return NULL;
T
ths 已提交
2081 2082 2083 2084 2085
	}
    }

    if (cyls || heads || secs) {
        if (cyls < 1 || cyls > 16383) {
G
Gerd Hoffmann 已提交
2086
            fprintf(stderr, "qemu: '%s' invalid physical cyls number\n", buf);
G
Gerd Hoffmann 已提交
2087
	    return NULL;
T
ths 已提交
2088 2089
	}
        if (heads < 1 || heads > 16) {
G
Gerd Hoffmann 已提交
2090
            fprintf(stderr, "qemu: '%s' invalid physical heads number\n", buf);
G
Gerd Hoffmann 已提交
2091
	    return NULL;
T
ths 已提交
2092 2093
	}
        if (secs < 1 || secs > 63) {
G
Gerd Hoffmann 已提交
2094
            fprintf(stderr, "qemu: '%s' invalid physical secs number\n", buf);
G
Gerd Hoffmann 已提交
2095
	    return NULL;
T
ths 已提交
2096 2097 2098
	}
    }

G
Gerd Hoffmann 已提交
2099
    if ((buf = qemu_opt_get(opts, "trans")) != NULL) {
T
ths 已提交
2100 2101 2102
        if (!cyls) {
            fprintf(stderr,
                    "qemu: '%s' trans must be used with cyls,heads and secs\n",
G
Gerd Hoffmann 已提交
2103
                    buf);
G
Gerd Hoffmann 已提交
2104
            return NULL;
T
ths 已提交
2105 2106 2107 2108 2109 2110 2111 2112
        }
        if (!strcmp(buf, "none"))
            translation = BIOS_ATA_TRANSLATION_NONE;
        else if (!strcmp(buf, "lba"))
            translation = BIOS_ATA_TRANSLATION_LBA;
        else if (!strcmp(buf, "auto"))
            translation = BIOS_ATA_TRANSLATION_AUTO;
	else {
G
Gerd Hoffmann 已提交
2113
            fprintf(stderr, "qemu: '%s' invalid translation type\n", buf);
G
Gerd Hoffmann 已提交
2114
	    return NULL;
T
ths 已提交
2115 2116 2117
	}
    }

G
Gerd Hoffmann 已提交
2118
    if ((buf = qemu_opt_get(opts, "media")) != NULL) {
T
ths 已提交
2119 2120 2121 2122 2123
        if (!strcmp(buf, "disk")) {
	    media = MEDIA_DISK;
	} else if (!strcmp(buf, "cdrom")) {
            if (cyls || secs || heads) {
                fprintf(stderr,
G
Gerd Hoffmann 已提交
2124
                        "qemu: '%s' invalid physical CHS format\n", buf);
G
Gerd Hoffmann 已提交
2125
	        return NULL;
T
ths 已提交
2126 2127 2128
            }
	    media = MEDIA_CDROM;
	} else {
G
Gerd Hoffmann 已提交
2129
	    fprintf(stderr, "qemu: '%s' invalid media\n", buf);
G
Gerd Hoffmann 已提交
2130
	    return NULL;
T
ths 已提交
2131 2132 2133
	}
    }

G
Gerd Hoffmann 已提交
2134
    if ((buf = qemu_opt_get(opts, "cache")) != NULL) {
2135
        if (!strcmp(buf, "off") || !strcmp(buf, "none"))
2136
            cache = 0;
2137
        else if (!strcmp(buf, "writethrough"))
2138
            cache = 1;
2139 2140
        else if (!strcmp(buf, "writeback"))
            cache = 2;
2141 2142
        else {
           fprintf(stderr, "qemu: invalid cache option\n");
G
Gerd Hoffmann 已提交
2143
           return NULL;
2144 2145 2146
        }
    }

2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159
#ifdef CONFIG_LINUX_AIO
    if ((buf = qemu_opt_get(opts, "aio")) != NULL) {
        if (!strcmp(buf, "threads"))
            aio = 0;
        else if (!strcmp(buf, "native"))
            aio = 1;
        else {
           fprintf(stderr, "qemu: invalid aio option\n");
           return NULL;
        }
    }
#endif

G
Gerd Hoffmann 已提交
2160
    if ((buf = qemu_opt_get(opts, "format")) != NULL) {
2161 2162 2163 2164
       if (strcmp(buf, "?") == 0) {
            fprintf(stderr, "qemu: Supported formats:");
            bdrv_iterate_format(bdrv_format_print, NULL);
            fprintf(stderr, "\n");
G
Gerd Hoffmann 已提交
2165
	    return NULL;
2166
        }
2167 2168 2169
        drv = bdrv_find_format(buf);
        if (!drv) {
            fprintf(stderr, "qemu: '%s' invalid format\n", buf);
G
Gerd Hoffmann 已提交
2170
            return NULL;
2171 2172 2173
        }
    }

2174
    onerror = BLOCK_ERR_STOP_ENOSPC;
G
Gerd Hoffmann 已提交
2175
    if ((buf = qemu_opt_get(opts, "werror")) != NULL) {
2176
        if (type != IF_IDE && type != IF_SCSI && type != IF_VIRTIO) {
2177
            fprintf(stderr, "werror is no supported by this format\n");
G
Gerd Hoffmann 已提交
2178
            return NULL;
2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189
        }
        if (!strcmp(buf, "ignore"))
            onerror = BLOCK_ERR_IGNORE;
        else if (!strcmp(buf, "enospc"))
            onerror = BLOCK_ERR_STOP_ENOSPC;
        else if (!strcmp(buf, "stop"))
            onerror = BLOCK_ERR_STOP_ANY;
        else if (!strcmp(buf, "report"))
            onerror = BLOCK_ERR_REPORT;
        else {
            fprintf(stderr, "qemu: '%s' invalid write error action\n", buf);
G
Gerd Hoffmann 已提交
2190
            return NULL;
2191 2192 2193
        }
    }

G
Gerd Hoffmann 已提交
2194
    if ((devaddr = qemu_opt_get(opts, "addr")) != NULL) {
2195
        if (type != IF_VIRTIO) {
G
Gerd Hoffmann 已提交
2196
            fprintf(stderr, "addr is not supported\n");
G
Gerd Hoffmann 已提交
2197
            return NULL;
2198 2199 2200
        }
    }

T
ths 已提交
2201 2202 2203 2204 2205
    /* compute bus and unit according index */

    if (index != -1) {
        if (bus_id != 0 || unit_id != -1) {
            fprintf(stderr,
G
Gerd Hoffmann 已提交
2206
                    "qemu: index cannot be used with bus and unit\n");
G
Gerd Hoffmann 已提交
2207
            return NULL;
T
ths 已提交
2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224
        }
        if (max_devs == 0)
        {
            unit_id = index;
            bus_id = 0;
        } else {
            unit_id = index % max_devs;
            bus_id = index / max_devs;
        }
    }

    /* if user doesn't specify a unit_id,
     * try to find the first free
     */

    if (unit_id == -1) {
       unit_id = 0;
G
Gerd Hoffmann 已提交
2225
       while (drive_get(type, bus_id, unit_id) != NULL) {
T
ths 已提交
2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236
           unit_id++;
           if (max_devs && unit_id >= max_devs) {
               unit_id -= max_devs;
               bus_id++;
           }
       }
    }

    /* check unit id */

    if (max_devs && unit_id >= max_devs) {
G
Gerd Hoffmann 已提交
2237 2238
        fprintf(stderr, "qemu: unit %d too big (max is %d)\n",
                unit_id, max_devs - 1);
G
Gerd Hoffmann 已提交
2239
        return NULL;
T
ths 已提交
2240 2241 2242 2243 2244 2245
    }

    /*
     * ignore multiple definitions
     */

G
Gerd Hoffmann 已提交
2246 2247 2248 2249
    if (drive_get(type, bus_id, unit_id) != NULL) {
        *fatal_error = 0;
        return NULL;
    }
T
ths 已提交
2250 2251 2252

    /* init */

G
Gerd Hoffmann 已提交
2253
    dinfo = qemu_mallocz(sizeof(*dinfo));
2254
    if ((buf = qemu_opts_id(opts)) != NULL) {
G
Gerd Hoffmann 已提交
2255 2256
        dinfo->id = qemu_strdup(buf);
    } else {
G
Gerd Hoffmann 已提交
2257
        /* no id supplied -> create one */
G
Gerd Hoffmann 已提交
2258
        dinfo->id = qemu_mallocz(32);
G
Gerd Hoffmann 已提交
2259 2260 2261
        if (type == IF_IDE || type == IF_SCSI)
            mediastr = (media == MEDIA_CDROM) ? "-cd" : "-hd";
        if (max_devs)
G
Gerd Hoffmann 已提交
2262
            snprintf(dinfo->id, 32, "%s%i%s%i",
G
Gerd Hoffmann 已提交
2263 2264
                     devname, bus_id, mediastr, unit_id);
        else
G
Gerd Hoffmann 已提交
2265
            snprintf(dinfo->id, 32, "%s%s%i",
G
Gerd Hoffmann 已提交
2266 2267 2268
                     devname, mediastr, unit_id);
    }
    dinfo->bdrv = bdrv_new(dinfo->id);
G
Gerd Hoffmann 已提交
2269 2270 2271 2272 2273
    dinfo->devaddr = devaddr;
    dinfo->type = type;
    dinfo->bus = bus_id;
    dinfo->unit = unit_id;
    dinfo->onerror = onerror;
G
Gerd Hoffmann 已提交
2274 2275 2276
    dinfo->opts = opts;
    if (serial)
        strncpy(dinfo->serial, serial, sizeof(serial));
B
Blue Swirl 已提交
2277
    QTAILQ_INSERT_TAIL(&drives, dinfo, next);
T
ths 已提交
2278

2279
    switch(type) {
T
ths 已提交
2280 2281
    case IF_IDE:
    case IF_SCSI:
2282
    case IF_XEN:
G
Gerd Hoffmann 已提交
2283
    case IF_NONE:
T
ths 已提交
2284 2285 2286
        switch(media) {
	case MEDIA_DISK:
            if (cyls != 0) {
G
Gerd Hoffmann 已提交
2287 2288
                bdrv_set_geometry_hint(dinfo->bdrv, cyls, heads, secs);
                bdrv_set_translation_hint(dinfo->bdrv, translation);
T
ths 已提交
2289 2290 2291
            }
	    break;
	case MEDIA_CDROM:
G
Gerd Hoffmann 已提交
2292
            bdrv_set_type_hint(dinfo->bdrv, BDRV_TYPE_CDROM);
T
ths 已提交
2293 2294 2295 2296 2297 2298 2299
	    break;
	}
        break;
    case IF_SD:
        /* FIXME: This isn't really a floppy, but it's a reasonable
           approximation.  */
    case IF_FLOPPY:
G
Gerd Hoffmann 已提交
2300
        bdrv_set_type_hint(dinfo->bdrv, BDRV_TYPE_FLOPPY);
T
ths 已提交
2301 2302 2303 2304
        break;
    case IF_PFLASH:
    case IF_MTD:
        break;
G
Gerd Hoffmann 已提交
2305 2306 2307 2308 2309 2310 2311 2312
    case IF_VIRTIO:
        /* add virtio block device */
        opts = qemu_opts_create(&qemu_device_opts, NULL, 0);
        qemu_opt_set(opts, "driver", "virtio-blk-pci");
        qemu_opt_set(opts, "drive", dinfo->id);
        if (devaddr)
            qemu_opt_set(opts, "addr", devaddr);
        break;
P
Paul Brook 已提交
2313 2314
    case IF_COUNT:
        abort();
T
ths 已提交
2315
    }
G
Gerd Hoffmann 已提交
2316
    if (!file) {
G
Gerd Hoffmann 已提交
2317 2318 2319
        *fatal_error = 0;
        return NULL;
    }
2320
    bdrv_flags = 0;
2321
    if (snapshot) {
2322
        bdrv_flags |= BDRV_O_SNAPSHOT;
2323 2324 2325 2326 2327 2328
        cache = 2; /* always use write-back with snapshot */
    }
    if (cache == 0) /* no caching */
        bdrv_flags |= BDRV_O_NOCACHE;
    else if (cache == 2) /* write-back */
        bdrv_flags |= BDRV_O_CACHE_WB;
2329 2330 2331 2332 2333 2334 2335

    if (aio == 1) {
        bdrv_flags |= BDRV_O_NATIVE_AIO;
    } else {
        bdrv_flags &= ~BDRV_O_NATIVE_AIO;
    }

G
Gerd Hoffmann 已提交
2336
    if (bdrv_open2(dinfo->bdrv, file, bdrv_flags, drv) < 0) {
T
ths 已提交
2337 2338
        fprintf(stderr, "qemu: could not open disk image %s\n",
                        file);
G
Gerd Hoffmann 已提交
2339
        return NULL;
T
ths 已提交
2340
    }
2341

G
Gerd Hoffmann 已提交
2342
    if (bdrv_key_required(dinfo->bdrv))
2343
        autostart = 0;
G
Gerd Hoffmann 已提交
2344 2345
    *fatal_error = 0;
    return dinfo;
T
ths 已提交
2346 2347
}

G
Gerd Hoffmann 已提交
2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367
static int drive_init_func(QemuOpts *opts, void *opaque)
{
    QEMUMachine *machine = opaque;
    int fatal_error = 0;

    if (drive_init(opts, machine, &fatal_error) == NULL) {
        if (fatal_error)
            return 1;
    }
    return 0;
}

static int drive_enable_snapshot(QemuOpts *opts, void *opaque)
{
    if (NULL == qemu_opt_get(opts, "snapshot")) {
        qemu_opt_set(opts, "snapshot", "on");
    }
    return 0;
}

J
Jan Kiszka 已提交
2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381
void qemu_register_boot_set(QEMUBootSetHandler *func, void *opaque)
{
    boot_set_handler = func;
    boot_set_opaque = opaque;
}

int qemu_boot_set(const char *boot_devices)
{
    if (!boot_set_handler) {
        return -EINVAL;
    }
    return boot_set_handler(boot_set_opaque, boot_devices);
}

J
Jan Kiszka 已提交
2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410
static int parse_bootdevices(char *devices)
{
    /* We just do some generic consistency checks */
    const char *p;
    int bitmap = 0;

    for (p = devices; *p != '\0'; p++) {
        /* Allowed boot devices are:
         * a-b: floppy disk drives
         * c-f: IDE disk drives
         * g-m: machine implementation dependant drives
         * n-p: network devices
         * It's up to each machine implementation to check if the given boot
         * devices match the actual hardware implementation and firmware
         * features.
         */
        if (*p < 'a' || *p > 'p') {
            fprintf(stderr, "Invalid boot device '%c'\n", *p);
            exit(1);
        }
        if (bitmap & (1 << (*p - 'a'))) {
            fprintf(stderr, "Boot device '%c' was given twice\n", *p);
            exit(1);
        }
        bitmap |= 1 << (*p - 'a');
    }
    return bitmap;
}

J
Jan Kiszka 已提交
2411 2412 2413 2414 2415 2416 2417 2418 2419 2420
static void restore_boot_devices(void *opaque)
{
    char *standard_boot_devices = opaque;

    qemu_boot_set(standard_boot_devices);

    qemu_unregister_reset(restore_boot_devices, standard_boot_devices);
    qemu_free(standard_boot_devices);
}

2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476
static void numa_add(const char *optarg)
{
    char option[128];
    char *endptr;
    unsigned long long value, endvalue;
    int nodenr;

    optarg = get_opt_name(option, 128, optarg, ',') + 1;
    if (!strcmp(option, "node")) {
        if (get_param_value(option, 128, "nodeid", optarg) == 0) {
            nodenr = nb_numa_nodes;
        } else {
            nodenr = strtoull(option, NULL, 10);
        }

        if (get_param_value(option, 128, "mem", optarg) == 0) {
            node_mem[nodenr] = 0;
        } else {
            value = strtoull(option, &endptr, 0);
            switch (*endptr) {
            case 0: case 'M': case 'm':
                value <<= 20;
                break;
            case 'G': case 'g':
                value <<= 30;
                break;
            }
            node_mem[nodenr] = value;
        }
        if (get_param_value(option, 128, "cpus", optarg) == 0) {
            node_cpumask[nodenr] = 0;
        } else {
            value = strtoull(option, &endptr, 10);
            if (value >= 64) {
                value = 63;
                fprintf(stderr, "only 64 CPUs in NUMA mode supported.\n");
            } else {
                if (*endptr == '-') {
                    endvalue = strtoull(endptr+1, &endptr, 10);
                    if (endvalue >= 63) {
                        endvalue = 62;
                        fprintf(stderr,
                            "only 63 CPUs in NUMA mode supported.\n");
                    }
                    value = (1 << (endvalue + 1)) - (1 << value);
                } else {
                    value = 1 << value;
                }
            }
            node_cpumask[nodenr] = value;
        }
        nb_numa_nodes++;
    }
    return;
}

2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526
static void smp_parse(const char *optarg)
{
    int smp, sockets = 0, threads = 0, cores = 0;
    char *endptr;
    char option[128];

    smp = strtoul(optarg, &endptr, 10);
    if (endptr != optarg) {
        if (*endptr == ',') {
            endptr++;
        }
    }
    if (get_param_value(option, 128, "sockets", endptr) != 0)
        sockets = strtoull(option, NULL, 10);
    if (get_param_value(option, 128, "cores", endptr) != 0)
        cores = strtoull(option, NULL, 10);
    if (get_param_value(option, 128, "threads", endptr) != 0)
        threads = strtoull(option, NULL, 10);
    if (get_param_value(option, 128, "maxcpus", endptr) != 0)
        max_cpus = strtoull(option, NULL, 10);

    /* compute missing values, prefer sockets over cores over threads */
    if (smp == 0 || sockets == 0) {
        sockets = sockets > 0 ? sockets : 1;
        cores = cores > 0 ? cores : 1;
        threads = threads > 0 ? threads : 1;
        if (smp == 0) {
            smp = cores * threads * sockets;
        } else {
            sockets = smp / (cores * threads);
        }
    } else {
        if (cores == 0) {
            threads = threads > 0 ? threads : 1;
            cores = smp / (sockets * threads);
        } else {
            if (sockets == 0) {
                sockets = smp / (cores * threads);
            } else {
                threads = smp / (cores * sockets);
            }
        }
    }
    smp_cpus = smp;
    smp_cores = cores > 0 ? cores : 1;
    smp_threads = threads > 0 ? threads : 1;
    if (max_cpus == 0)
        max_cpus = smp_cpus;
}

B
bellard 已提交
2527 2528 2529
/***********************************************************/
/* USB devices */

2530 2531 2532 2533 2534
static void usb_msd_password_cb(void *opaque, int err)
{
    USBDevice *dev = opaque;

    if (!err)
2535
        usb_device_attach(dev);
2536
    else
2537
        dev->info->handle_destroy(dev);
2538 2539
}

2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558
static struct {
    const char *name;
    const char *qdev;
} usbdevs[] = {
    {
        .name = "mouse",
        .qdev = "QEMU USB Mouse",
    },{
        .name = "tablet",
        .qdev = "QEMU USB Tablet",
    },{
        .name = "keyboard",
        .qdev = "QEMU USB Keyboard",
    },{
        .name = "wacom-tablet",
        .qdev = "QEMU PenPartner Tablet",
    }
};

2559
static int usb_device_add(const char *devname, int is_hotplug)
B
bellard 已提交
2560 2561
{
    const char *p;
2562 2563 2564
    USBBus *bus = usb_bus_find(-1 /* any */);
    USBDevice *dev = NULL;
    int i;
B
bellard 已提交
2565

2566
    if (!usb_enabled)
B
bellard 已提交
2567 2568
        return -1;

2569 2570 2571 2572 2573 2574 2575 2576 2577
    /* simple devices which don't need extra care */
    for (i = 0; i < ARRAY_SIZE(usbdevs); i++) {
        if (strcmp(devname, usbdevs[i].name) != 0)
            continue;
        dev = usb_create_simple(bus, usbdevs[i].qdev);
        goto done;
    }

    /* the other ones */
B
bellard 已提交
2578 2579
    if (strstart(devname, "host:", &p)) {
        dev = usb_host_device_open(p);
P
pbrook 已提交
2580
    } else if (strstart(devname, "disk:", &p)) {
2581 2582
        BlockDriverState *bs;

2583
        dev = usb_msd_init(p);
2584 2585
        if (!dev)
            return -1;
2586
        bs = usb_msd_get_bdrv(dev);
2587 2588
        if (bdrv_key_required(bs)) {
            autostart = 0;
2589
            if (is_hotplug) {
A
aliguori 已提交
2590 2591
                monitor_read_bdrv_key_start(cur_mon, bs, usb_msd_password_cb,
                                            dev);
2592
                return 0;
2593 2594
            }
        }
B
balrog 已提交
2595 2596
    } else if (strstart(devname, "serial:", &p)) {
        dev = usb_serial_init(p);
A
aurel32 已提交
2597 2598 2599 2600
#ifdef CONFIG_BRLAPI
    } else if (!strcmp(devname, "braille")) {
        dev = usb_baum_init();
#endif
2601
    } else if (strstart(devname, "net:", &p)) {
2602
        int nic = nb_nics;
2603

2604
        if (net_client_init(NULL, "nic", p) < 0)
2605
            return -1;
2606 2607
        nd_table[nic].model = "usb";
        dev = usb_net_init(&nd_table[nic]);
2608 2609 2610
    } else if (!strcmp(devname, "bt") || strstart(devname, "bt:", &p)) {
        dev = usb_bt_init(devname[2] ? hci_init(p) :
                        bt_new_hci(qemu_find_bt_vlan(0)));
B
bellard 已提交
2611 2612 2613
    } else {
        return -1;
    }
P
pbrook 已提交
2614 2615 2616
    if (!dev)
        return -1;

2617
done:
B
bellard 已提交
2618 2619 2620
    return 0;
}

2621 2622 2623 2624 2625
static int usb_device_del(const char *devname)
{
    int bus_num, addr;
    const char *p;

2626 2627 2628
    if (strstart(devname, "host:", &p))
        return usb_host_device_close(p);

2629
    if (!usb_enabled)
2630 2631 2632 2633 2634 2635 2636 2637
        return -1;

    p = strchr(devname, '.');
    if (!p)
        return -1;
    bus_num = strtoul(devname, NULL, 0);
    addr = strtoul(p + 1, NULL, 0);

2638
    return usb_device_delete_addr(bus_num, addr);
2639 2640
}

2641 2642 2643 2644 2645
static int usb_parse(const char *cmdline)
{
    return usb_device_add(cmdline, 0);
}

2646
void do_usb_add(Monitor *mon, const QDict *qdict)
B
bellard 已提交
2647
{
2648
    usb_device_add(qdict_get_str(qdict, "devname"), 1);
B
bellard 已提交
2649 2650
}

2651
void do_usb_del(Monitor *mon, const QDict *qdict)
B
bellard 已提交
2652
{
2653
    usb_device_del(qdict_get_str(qdict, "devname"));
B
bellard 已提交
2654 2655
}

2656 2657 2658 2659
/***********************************************************/
/* PCMCIA/Cardbus */

static struct pcmcia_socket_entry_s {
P
Paul Brook 已提交
2660
    PCMCIASocket *socket;
2661 2662 2663
    struct pcmcia_socket_entry_s *next;
} *pcmcia_sockets = 0;

P
Paul Brook 已提交
2664
void pcmcia_socket_register(PCMCIASocket *socket)
2665 2666 2667 2668 2669 2670 2671 2672 2673
{
    struct pcmcia_socket_entry_s *entry;

    entry = qemu_malloc(sizeof(struct pcmcia_socket_entry_s));
    entry->socket = socket;
    entry->next = pcmcia_sockets;
    pcmcia_sockets = entry;
}

P
Paul Brook 已提交
2674
void pcmcia_socket_unregister(PCMCIASocket *socket)
2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685
{
    struct pcmcia_socket_entry_s *entry, **ptr;

    ptr = &pcmcia_sockets;
    for (entry = *ptr; entry; ptr = &entry->next, entry = *ptr)
        if (entry->socket == socket) {
            *ptr = entry->next;
            qemu_free(entry);
        }
}

A
aliguori 已提交
2686
void pcmcia_info(Monitor *mon)
2687 2688
{
    struct pcmcia_socket_entry_s *iter;
A
aliguori 已提交
2689

2690
    if (!pcmcia_sockets)
A
aliguori 已提交
2691
        monitor_printf(mon, "No PCMCIA sockets\n");
2692 2693

    for (iter = pcmcia_sockets; iter; iter = iter->next)
A
aliguori 已提交
2694 2695 2696
        monitor_printf(mon, "%s: %s\n", iter->socket->slot_string,
                       iter->socket->attached ? iter->socket->card_string :
                       "Empty");
2697 2698
}

2699
/***********************************************************/
2700 2701
/* register display */

2702 2703 2704 2705 2706 2707
struct DisplayAllocator default_allocator = {
    defaultallocator_create_displaysurface,
    defaultallocator_resize_displaysurface,
    defaultallocator_free_displaysurface
};

2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722
void register_displaystate(DisplayState *ds)
{
    DisplayState **s;
    s = &display_state;
    while (*s != NULL)
        s = &(*s)->next;
    ds->next = NULL;
    *s = ds;
}

DisplayState *get_displaystate(void)
{
    return display_state;
}

2723 2724 2725 2726 2727 2728
DisplayAllocator *register_displayallocator(DisplayState *ds, DisplayAllocator *da)
{
    if(ds->allocator ==  &default_allocator) ds->allocator = da;
    return ds->allocator;
}

2729 2730
/* dumb display */

2731
static void dumb_display_init(void)
2732
{
2733
    DisplayState *ds = qemu_mallocz(sizeof(DisplayState));
2734 2735
    ds->allocator = &default_allocator;
    ds->surface = qemu_create_displaysurface(ds, 640, 480);
2736
    register_displaystate(ds);
2737 2738
}

2739 2740
/***********************************************************/
/* I/O handling */
2741

2742 2743
typedef struct IOHandlerRecord {
    int fd;
B
bellard 已提交
2744 2745 2746
    IOCanRWHandler *fd_read_poll;
    IOHandler *fd_read;
    IOHandler *fd_write;
2747
    int deleted;
2748 2749 2750
    void *opaque;
    /* temporary data */
    struct pollfd *ufd;
2751
    struct IOHandlerRecord *next;
2752 2753
} IOHandlerRecord;

2754
static IOHandlerRecord *first_io_handler;
2755

B
bellard 已提交
2756 2757
/* XXX: fd_read_poll should be suppressed, but an API change is
   necessary in the character devices to suppress fd_can_read(). */
2758 2759 2760 2761
int qemu_set_fd_handler2(int fd,
                         IOCanRWHandler *fd_read_poll,
                         IOHandler *fd_read,
                         IOHandler *fd_write,
B
bellard 已提交
2762
                         void *opaque)
2763
{
B
bellard 已提交
2764
    IOHandlerRecord **pioh, *ioh;
2765

B
bellard 已提交
2766 2767 2768 2769 2770 2771 2772
    if (!fd_read && !fd_write) {
        pioh = &first_io_handler;
        for(;;) {
            ioh = *pioh;
            if (ioh == NULL)
                break;
            if (ioh->fd == fd) {
2773
                ioh->deleted = 1;
B
bellard 已提交
2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791
                break;
            }
            pioh = &ioh->next;
        }
    } else {
        for(ioh = first_io_handler; ioh != NULL; ioh = ioh->next) {
            if (ioh->fd == fd)
                goto found;
        }
        ioh = qemu_mallocz(sizeof(IOHandlerRecord));
        ioh->next = first_io_handler;
        first_io_handler = ioh;
    found:
        ioh->fd = fd;
        ioh->fd_read_poll = fd_read_poll;
        ioh->fd_read = fd_read;
        ioh->fd_write = fd_write;
        ioh->opaque = opaque;
2792
        ioh->deleted = 0;
B
bellard 已提交
2793
    }
2794 2795 2796
    return 0;
}

2797 2798 2799
int qemu_set_fd_handler(int fd,
                        IOHandler *fd_read,
                        IOHandler *fd_write,
B
bellard 已提交
2800
                        void *opaque)
2801
{
B
bellard 已提交
2802
    return qemu_set_fd_handler2(fd, NULL, fd_read, fd_write, opaque);
2803 2804
}

A
aliguori 已提交
2805
#ifdef _WIN32
2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840
/***********************************************************/
/* 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 = qemu_mallocz(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;
            qemu_free(pe);
            break;
        }
    }
}

2841 2842 2843 2844 2845 2846 2847 2848 2849 2850
/***********************************************************/
/* Wait objects support */
typedef struct WaitObjects {
    int num;
    HANDLE events[MAXIMUM_WAIT_OBJECTS + 1];
    WaitObjectFunc *func[MAXIMUM_WAIT_OBJECTS + 1];
    void *opaque[MAXIMUM_WAIT_OBJECTS + 1];
} WaitObjects;

static WaitObjects wait_objects = {0};
2851

2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877
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;
    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];
2878
        }
2879 2880 2881 2882 2883 2884
    }
    if (found)
        w->num--;
}
#endif

2885 2886 2887
/***********************************************************/
/* ram save/restore */

2888
#define RAM_SAVE_FLAG_FULL	0x01 /* Obsolete, not used anymore */
2889 2890 2891 2892 2893 2894
#define RAM_SAVE_FLAG_COMPRESS	0x02
#define RAM_SAVE_FLAG_MEM_SIZE	0x04
#define RAM_SAVE_FLAG_PAGE	0x08
#define RAM_SAVE_FLAG_EOS	0x10

static int is_dup_page(uint8_t *page, uint8_t ch)
2895
{
2896 2897 2898
    uint32_t val = ch << 24 | ch << 16 | ch << 8 | ch;
    uint32_t *array = (uint32_t *)page;
    int i;
2899

2900 2901 2902 2903 2904 2905 2906 2907 2908 2909
    for (i = 0; i < (TARGET_PAGE_SIZE / 4); i++) {
        if (array[i] != val)
            return 0;
    }

    return 1;
}

static int ram_save_block(QEMUFile *f)
{
A
Anthony Liguori 已提交
2910 2911 2912
    static ram_addr_t current_addr = 0;
    ram_addr_t saved_addr = current_addr;
    ram_addr_t addr = 0;
2913 2914
    int found = 0;

P
pbrook 已提交
2915
    while (addr < last_ram_offset) {
2916
        if (cpu_physical_memory_get_dirty(current_addr, MIGRATION_DIRTY_FLAG)) {
P
pbrook 已提交
2917
            uint8_t *p;
2918 2919 2920 2921 2922

            cpu_physical_memory_reset_dirty(current_addr,
                                            current_addr + TARGET_PAGE_SIZE,
                                            MIGRATION_DIRTY_FLAG);

P
pbrook 已提交
2923
            p = qemu_get_ram_ptr(current_addr);
2924

P
pbrook 已提交
2925
            if (is_dup_page(p, *p)) {
2926
                qemu_put_be64(f, current_addr | RAM_SAVE_FLAG_COMPRESS);
P
pbrook 已提交
2927
                qemu_put_byte(f, *p);
2928 2929
            } else {
                qemu_put_be64(f, current_addr | RAM_SAVE_FLAG_PAGE);
P
pbrook 已提交
2930
                qemu_put_buffer(f, p, TARGET_PAGE_SIZE);
2931
            }
2932 2933 2934

            found = 1;
            break;
2935
        }
2936
        addr += TARGET_PAGE_SIZE;
P
pbrook 已提交
2937
        current_addr = (saved_addr + addr) % last_ram_offset;
2938
    }
2939 2940

    return found;
2941 2942
}

2943
static uint64_t bytes_transferred = 0;
2944

A
Anthony Liguori 已提交
2945
static ram_addr_t ram_save_remaining(void)
2946
{
A
Anthony Liguori 已提交
2947 2948
    ram_addr_t addr;
    ram_addr_t count = 0;
2949

P
pbrook 已提交
2950
    for (addr = 0; addr < last_ram_offset; addr += TARGET_PAGE_SIZE) {
2951 2952 2953 2954 2955 2956 2957
        if (cpu_physical_memory_get_dirty(addr, MIGRATION_DIRTY_FLAG))
            count++;
    }

    return count;
}

2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972
uint64_t ram_bytes_remaining(void)
{
    return ram_save_remaining() * TARGET_PAGE_SIZE;
}

uint64_t ram_bytes_transferred(void)
{
    return bytes_transferred;
}

uint64_t ram_bytes_total(void)
{
    return last_ram_offset;
}

2973 2974
static int ram_save_live(QEMUFile *f, int stage, void *opaque)
{
A
Anthony Liguori 已提交
2975
    ram_addr_t addr;
2976 2977 2978
    uint64_t bytes_transferred_last;
    double bwidth = 0;
    uint64_t expected_time = 0;
2979

2980
    if (cpu_physical_sync_dirty_bitmap(0, TARGET_PHYS_ADDR_MAX) != 0) {
2981 2982 2983 2984
        qemu_file_set_error(f);
        return 0;
    }

2985 2986
    if (stage == 1) {
        /* Make sure all dirty bits are set */
P
pbrook 已提交
2987
        for (addr = 0; addr < last_ram_offset; addr += TARGET_PAGE_SIZE) {
2988 2989 2990
            if (!cpu_physical_memory_get_dirty(addr, MIGRATION_DIRTY_FLAG))
                cpu_physical_memory_set_dirty(addr);
        }
2991

2992 2993 2994
        /* Enable dirty memory tracking */
        cpu_physical_memory_set_dirty_tracking(1);

P
pbrook 已提交
2995
        qemu_put_be64(f, last_ram_offset | RAM_SAVE_FLAG_MEM_SIZE);
2996 2997
    }

2998 2999 3000
    bytes_transferred_last = bytes_transferred;
    bwidth = get_clock();

3001 3002 3003 3004
    while (!qemu_file_rate_limit(f)) {
        int ret;

        ret = ram_save_block(f);
3005
        bytes_transferred += ret * TARGET_PAGE_SIZE;
3006 3007 3008 3009
        if (ret == 0) /* no more blocks */
            break;
    }

3010 3011 3012 3013 3014 3015 3016 3017
    bwidth = get_clock() - bwidth;
    bwidth = (bytes_transferred - bytes_transferred_last) / bwidth;

    /* if we haven't transferred anything this round, force expected_time to a
     * a very high value, but without crashing */
    if (bwidth == 0)
        bwidth = 0.000001;

3018 3019 3020 3021 3022
    /* try transferring iterative blocks of memory */

    if (stage == 3) {

        /* flush all remaining blocks regardless of rate limiting */
3023 3024 3025
        while (ram_save_block(f) != 0) {
            bytes_transferred += TARGET_PAGE_SIZE;
        }
3026
        cpu_physical_memory_set_dirty_tracking(0);
3027 3028 3029 3030
    }

    qemu_put_be64(f, RAM_SAVE_FLAG_EOS);

3031 3032 3033
    expected_time = ram_save_remaining() * TARGET_PAGE_SIZE / bwidth;

    return (stage == 2) && (expected_time <= migrate_max_downtime());
3034 3035 3036 3037
}

static int ram_load(QEMUFile *f, void *opaque, int version_id)
{
A
Anthony Liguori 已提交
3038
    ram_addr_t addr;
3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050
    int flags;

    if (version_id != 3)
        return -EINVAL;

    do {
        addr = qemu_get_be64(f);

        flags = addr & ~TARGET_PAGE_MASK;
        addr &= TARGET_PAGE_MASK;

        if (flags & RAM_SAVE_FLAG_MEM_SIZE) {
P
pbrook 已提交
3051
            if (addr != last_ram_offset)
3052 3053 3054 3055 3056
                return -EINVAL;
        }

        if (flags & RAM_SAVE_FLAG_COMPRESS) {
            uint8_t ch = qemu_get_byte(f);
3057 3058
            memset(qemu_get_ram_ptr(addr), ch, TARGET_PAGE_SIZE);
#ifndef _WIN32
3059 3060 3061
            if (ch == 0 &&
                (!kvm_enabled() || kvm_has_sync_mmu())) {
                madvise(qemu_get_ram_ptr(addr), TARGET_PAGE_SIZE, MADV_DONTNEED);
3062
            }
3063
#endif
3064
        } else if (flags & RAM_SAVE_FLAG_PAGE)
P
pbrook 已提交
3065
            qemu_get_buffer(f, qemu_get_ram_ptr(addr), TARGET_PAGE_SIZE);
3066 3067
    } while (!(flags & RAM_SAVE_FLAG_EOS));

3068 3069 3070
    return 0;
}

A
aliguori 已提交
3071 3072
void qemu_service_io(void)
{
3073
    qemu_notify_event();
A
aliguori 已提交
3074 3075
}

B
bellard 已提交
3076 3077 3078 3079 3080 3081 3082
/***********************************************************/
/* bottom halves (can be seen as timers which expire ASAP) */

struct QEMUBH {
    QEMUBHFunc *cb;
    void *opaque;
    int scheduled;
A
aliguori 已提交
3083 3084
    int idle;
    int deleted;
B
bellard 已提交
3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095
    QEMUBH *next;
};

static QEMUBH *first_bh = NULL;

QEMUBH *qemu_bh_new(QEMUBHFunc *cb, void *opaque)
{
    QEMUBH *bh;
    bh = qemu_mallocz(sizeof(QEMUBH));
    bh->cb = cb;
    bh->opaque = opaque;
A
aliguori 已提交
3096 3097
    bh->next = first_bh;
    first_bh = bh;
B
bellard 已提交
3098 3099 3100
    return bh;
}

B
bellard 已提交
3101
int qemu_bh_poll(void)
B
bellard 已提交
3102
{
A
aliguori 已提交
3103
    QEMUBH *bh, **bhp;
B
bellard 已提交
3104
    int ret;
B
bellard 已提交
3105

B
bellard 已提交
3106
    ret = 0;
A
aliguori 已提交
3107 3108 3109 3110 3111 3112 3113 3114
    for (bh = first_bh; bh; bh = bh->next) {
        if (!bh->deleted && bh->scheduled) {
            bh->scheduled = 0;
            if (!bh->idle)
                ret = 1;
            bh->idle = 0;
            bh->cb(bh->opaque);
        }
B
bellard 已提交
3115
    }
A
aliguori 已提交
3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127

    /* remove deleted bhs */
    bhp = &first_bh;
    while (*bhp) {
        bh = *bhp;
        if (bh->deleted) {
            *bhp = bh->next;
            qemu_free(bh);
        } else
            bhp = &bh->next;
    }

B
bellard 已提交
3128
    return ret;
B
bellard 已提交
3129 3130
}

A
aliguori 已提交
3131 3132 3133 3134 3135 3136 3137 3138
void qemu_bh_schedule_idle(QEMUBH *bh)
{
    if (bh->scheduled)
        return;
    bh->scheduled = 1;
    bh->idle = 1;
}

B
bellard 已提交
3139 3140 3141 3142 3143
void qemu_bh_schedule(QEMUBH *bh)
{
    if (bh->scheduled)
        return;
    bh->scheduled = 1;
A
aliguori 已提交
3144
    bh->idle = 0;
B
bellard 已提交
3145
    /* stop the currently executing CPU to execute the BH ASAP */
3146
    qemu_notify_event();
B
bellard 已提交
3147 3148 3149 3150
}

void qemu_bh_cancel(QEMUBH *bh)
{
A
aliguori 已提交
3151
    bh->scheduled = 0;
B
bellard 已提交
3152 3153 3154 3155
}

void qemu_bh_delete(QEMUBH *bh)
{
A
aliguori 已提交
3156 3157
    bh->scheduled = 0;
    bh->deleted = 1;
B
bellard 已提交
3158 3159
}

A
aliguori 已提交
3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179
static void qemu_bh_update_timeout(int *timeout)
{
    QEMUBH *bh;

    for (bh = first_bh; bh; bh = bh->next) {
        if (!bh->deleted && bh->scheduled) {
            if (bh->idle) {
                /* idle bottom halves will be polled at least
                 * every 10ms */
                *timeout = MIN(10, *timeout);
            } else {
                /* non-idle bottom halves will be executed
                 * immediately */
                *timeout = 0;
                break;
            }
        }
    }
}

3180 3181 3182
/***********************************************************/
/* machine registration */

B
blueswir1 已提交
3183
static QEMUMachine *first_machine = NULL;
3184
QEMUMachine *current_machine = NULL;
3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196

int qemu_register_machine(QEMUMachine *m)
{
    QEMUMachine **pm;
    pm = &first_machine;
    while (*pm != NULL)
        pm = &(*pm)->next;
    m->next = NULL;
    *pm = m;
    return 0;
}

3197
static QEMUMachine *find_machine(const char *name)
3198 3199 3200 3201 3202 3203
{
    QEMUMachine *m;

    for(m = first_machine; m != NULL; m = m->next) {
        if (!strcmp(m->name, name))
            return m;
M
Mark McLoughlin 已提交
3204 3205
        if (m->alias && !strcmp(m->alias, name))
            return m;
3206 3207 3208 3209
    }
    return NULL;
}

3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221
static QEMUMachine *find_default_machine(void)
{
    QEMUMachine *m;

    for(m = first_machine; m != NULL; m = m->next) {
        if (m->is_default) {
            return m;
        }
    }
    return NULL;
}

3222 3223 3224
/***********************************************************/
/* main execution loop */

3225
static void gui_update(void *opaque)
3226
{
3227
    uint64_t interval = GUI_REFRESH_INTERVAL;
3228
    DisplayState *ds = opaque;
3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239
    DisplayChangeListener *dcl = ds->listeners;

    dpy_refresh(ds);

    while (dcl != NULL) {
        if (dcl->gui_timer_interval &&
            dcl->gui_timer_interval < interval)
            interval = dcl->gui_timer_interval;
        dcl = dcl->next;
    }
    qemu_mod_timer(ds->gui_timer, interval + qemu_get_clock(rt_clock));
3240 3241
}

B
blueswir1 已提交
3242 3243 3244 3245 3246 3247 3248
static void nographic_update(void *opaque)
{
    uint64_t interval = GUI_REFRESH_INTERVAL;

    qemu_mod_timer(nographic_timer, interval + qemu_get_clock(rt_clock));
}

B
bellard 已提交
3249 3250 3251
struct vm_change_state_entry {
    VMChangeStateHandler *cb;
    void *opaque;
B
Blue Swirl 已提交
3252
    QLIST_ENTRY (vm_change_state_entry) entries;
B
bellard 已提交
3253 3254
};

B
Blue Swirl 已提交
3255
static QLIST_HEAD(vm_change_state_head, vm_change_state_entry) vm_change_state_head;
B
bellard 已提交
3256 3257 3258 3259 3260 3261 3262 3263 3264 3265

VMChangeStateEntry *qemu_add_vm_change_state_handler(VMChangeStateHandler *cb,
                                                     void *opaque)
{
    VMChangeStateEntry *e;

    e = qemu_mallocz(sizeof (*e));

    e->cb = cb;
    e->opaque = opaque;
B
Blue Swirl 已提交
3266
    QLIST_INSERT_HEAD(&vm_change_state_head, e, entries);
B
bellard 已提交
3267 3268 3269 3270 3271
    return e;
}

void qemu_del_vm_change_state_handler(VMChangeStateEntry *e)
{
B
Blue Swirl 已提交
3272
    QLIST_REMOVE (e, entries);
B
bellard 已提交
3273 3274 3275
    qemu_free (e);
}

3276
static void vm_state_notify(int running, int reason)
B
bellard 已提交
3277 3278 3279 3280
{
    VMChangeStateEntry *e;

    for (e = vm_change_state_head.lh_first; e; e = e->entries.le_next) {
3281
        e->cb(e->opaque, running, reason);
B
bellard 已提交
3282 3283 3284
    }
}

3285 3286 3287
static void resume_all_vcpus(void);
static void pause_all_vcpus(void);

3288 3289 3290 3291 3292
void vm_start(void)
{
    if (!vm_running) {
        cpu_enable_ticks();
        vm_running = 1;
3293
        vm_state_notify(1, 0);
3294
        qemu_rearm_alarm_timer(alarm_timer);
3295
        resume_all_vcpus();
3296 3297 3298
    }
}

3299 3300 3301
/* reset/shutdown handler */

typedef struct QEMUResetEntry {
B
Blue Swirl 已提交
3302
    QTAILQ_ENTRY(QEMUResetEntry) entry;
3303 3304 3305 3306
    QEMUResetHandler *func;
    void *opaque;
} QEMUResetEntry;

B
Blue Swirl 已提交
3307 3308
static QTAILQ_HEAD(reset_handlers, QEMUResetEntry) reset_handlers =
    QTAILQ_HEAD_INITIALIZER(reset_handlers);
3309 3310
static int reset_requested;
static int shutdown_requested;
B
bellard 已提交
3311
static int powerdown_requested;
3312
static int debug_requested;
3313
static int vmstop_requested;
3314

A
aurel32 已提交
3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335
int qemu_shutdown_requested(void)
{
    int r = shutdown_requested;
    shutdown_requested = 0;
    return r;
}

int qemu_reset_requested(void)
{
    int r = reset_requested;
    reset_requested = 0;
    return r;
}

int qemu_powerdown_requested(void)
{
    int r = powerdown_requested;
    powerdown_requested = 0;
    return r;
}

3336 3337 3338 3339 3340 3341 3342
static int qemu_debug_requested(void)
{
    int r = debug_requested;
    debug_requested = 0;
    return r;
}

3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354
static int qemu_vmstop_requested(void)
{
    int r = vmstop_requested;
    vmstop_requested = 0;
    return r;
}

static void do_vm_stop(int reason)
{
    if (vm_running) {
        cpu_disable_ticks();
        vm_running = 0;
3355
        pause_all_vcpus();
3356 3357 3358 3359
        vm_state_notify(0, reason);
    }
}

3360
void qemu_register_reset(QEMUResetHandler *func, void *opaque)
3361
{
J
Jan Kiszka 已提交
3362
    QEMUResetEntry *re = qemu_mallocz(sizeof(QEMUResetEntry));
3363 3364 3365

    re->func = func;
    re->opaque = opaque;
B
Blue Swirl 已提交
3366
    QTAILQ_INSERT_TAIL(&reset_handlers, re, entry);
3367 3368
}

J
Jan Kiszka 已提交
3369
void qemu_unregister_reset(QEMUResetHandler *func, void *opaque)
3370 3371 3372
{
    QEMUResetEntry *re;

B
Blue Swirl 已提交
3373
    QTAILQ_FOREACH(re, &reset_handlers, entry) {
J
Jan Kiszka 已提交
3374
        if (re->func == func && re->opaque == opaque) {
B
Blue Swirl 已提交
3375
            QTAILQ_REMOVE(&reset_handlers, re, entry);
J
Jan Kiszka 已提交
3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386
            qemu_free(re);
            return;
        }
    }
}

void qemu_system_reset(void)
{
    QEMUResetEntry *re, *nre;

    /* reset all devices */
B
Blue Swirl 已提交
3387
    QTAILQ_FOREACH_SAFE(re, &reset_handlers, entry, nre) {
3388 3389 3390 3391 3392 3393
        re->func(re->opaque);
    }
}

void qemu_system_reset_request(void)
{
B
bellard 已提交
3394 3395 3396 3397 3398
    if (no_reboot) {
        shutdown_requested = 1;
    } else {
        reset_requested = 1;
    }
3399
    qemu_notify_event();
3400 3401 3402 3403 3404
}

void qemu_system_shutdown_request(void)
{
    shutdown_requested = 1;
3405
    qemu_notify_event();
3406 3407
}

B
bellard 已提交
3408 3409 3410
void qemu_system_powerdown_request(void)
{
    powerdown_requested = 1;
3411 3412 3413
    qemu_notify_event();
}

3414 3415
#ifdef CONFIG_IOTHREAD
static void qemu_system_vmstop_request(int reason)
3416
{
3417 3418
    vmstop_requested = reason;
    qemu_notify_event();
3419
}
3420
#endif
3421

3422 3423 3424 3425
#ifndef _WIN32
static int io_thread_fd = -1;

static void qemu_event_increment(void)
3426
{
3427 3428 3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467
    static const char byte = 0;

    if (io_thread_fd == -1)
        return;

    write(io_thread_fd, &byte, sizeof(byte));
}

static void qemu_event_read(void *opaque)
{
    int fd = (unsigned long)opaque;
    ssize_t len;

    /* Drain the notify pipe */
    do {
        char buffer[512];
        len = read(fd, buffer, sizeof(buffer));
    } while ((len == -1 && errno == EINTR) || len > 0);
}

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

    err = pipe(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 *)(unsigned long)fds[0]);

    io_thread_fd = fds[1];
J
Jan Kiszka 已提交
3468 3469
    return 0;

3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485
fail:
    close(fds[0]);
    close(fds[1]);
    return err;
}
#else
HANDLE qemu_event_handle;

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) {
B
Blue Swirl 已提交
3486
        fprintf(stderr, "Failed CreateEvent: %ld\n", GetLastError());
3487 3488 3489
        return -1;
    }
    qemu_add_wait_object(qemu_event_handle, dummy_event_handler, NULL);
3490 3491 3492
    return 0;
}

3493 3494
static void qemu_event_increment(void)
{
M
malc 已提交
3495
    if (!SetEvent(qemu_event_handle)) {
B
Blue Swirl 已提交
3496
        fprintf(stderr, "qemu_event_increment: SetEvent failed: %ld\n",
M
malc 已提交
3497 3498 3499
                GetLastError());
        exit (1);
    }
3500 3501 3502
}
#endif

3503 3504 3505 3506 3507 3508 3509 3510 3511 3512
static int cpu_can_run(CPUState *env)
{
    if (env->stop)
        return 0;
    if (env->stopped)
        return 0;
    return 1;
}

#ifndef CONFIG_IOTHREAD
3513 3514 3515 3516 3517
static int qemu_init_main_loop(void)
{
    return qemu_event_init();
}

3518 3519 3520 3521 3522 3523
void qemu_init_vcpu(void *_env)
{
    CPUState *env = _env;

    if (kvm_enabled())
        kvm_init_vcpu(env);
3524 3525
    env->nr_cores = smp_cores;
    env->nr_threads = smp_threads;
3526 3527 3528
    return;
}

3529 3530 3531 3532 3533
int qemu_cpu_self(void *env)
{
    return 1;
}

3534 3535 3536 3537 3538 3539 3540 3541
static void resume_all_vcpus(void)
{
}

static void pause_all_vcpus(void)
{
}

3542 3543 3544 3545 3546
void qemu_cpu_kick(void *env)
{
    return;
}

3547 3548 3549 3550 3551 3552
void qemu_notify_event(void)
{
    CPUState *env = cpu_single_env;

    if (env) {
        cpu_exit(env);
3553
    }
3554 3555
}

3556 3557 3558
#define qemu_mutex_lock_iothread() do { } while (0)
#define qemu_mutex_unlock_iothread() do { } while (0)

3559 3560 3561 3562 3563
void vm_stop(int reason)
{
    do_vm_stop(reason);
}

3564 3565 3566 3567 3568 3569 3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 3604 3605 3606 3607 3608 3609 3610 3611 3612 3613 3614 3615 3616 3617 3618 3619 3620 3621 3622 3623 3624 3625 3626 3627 3628 3629 3630 3631 3632 3633 3634 3635 3636
#else /* CONFIG_IOTHREAD */

#include "qemu-thread.h"

QemuMutex qemu_global_mutex;
static QemuMutex qemu_fair_mutex;

static QemuThread io_thread;

static QemuThread *tcg_cpu_thread;
static QemuCond *tcg_halt_cond;

static int qemu_system_ready;
/* cpu creation */
static QemuCond qemu_cpu_cond;
/* system init */
static QemuCond qemu_system_cond;
static QemuCond qemu_pause_cond;

static void block_io_signals(void);
static void unblock_io_signals(void);
static int tcg_has_work(void);

static int qemu_init_main_loop(void)
{
    int ret;

    ret = qemu_event_init();
    if (ret)
        return ret;

    qemu_cond_init(&qemu_pause_cond);
    qemu_mutex_init(&qemu_fair_mutex);
    qemu_mutex_init(&qemu_global_mutex);
    qemu_mutex_lock(&qemu_global_mutex);

    unblock_io_signals();
    qemu_thread_self(&io_thread);

    return 0;
}

static void qemu_wait_io_event(CPUState *env)
{
    while (!tcg_has_work())
        qemu_cond_timedwait(env->halt_cond, &qemu_global_mutex, 1000);

    qemu_mutex_unlock(&qemu_global_mutex);

    /*
     * Users of qemu_global_mutex can be starved, having no chance
     * to acquire it since this path will get to it first.
     * So use another lock to provide fairness.
     */
    qemu_mutex_lock(&qemu_fair_mutex);
    qemu_mutex_unlock(&qemu_fair_mutex);

    qemu_mutex_lock(&qemu_global_mutex);
    if (env->stop) {
        env->stop = 0;
        env->stopped = 1;
        qemu_cond_signal(&qemu_pause_cond);
    }
}

static int qemu_cpu_exec(CPUState *env);

static void *kvm_cpu_thread_fn(void *arg)
{
    CPUState *env = arg;

    block_io_signals();
    qemu_thread_self(env->thread);
3637 3638
    if (kvm_enabled())
        kvm_init_vcpu(env);
3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650 3651

    /* signal CPU creation */
    qemu_mutex_lock(&qemu_global_mutex);
    env->created = 1;
    qemu_cond_signal(&qemu_cpu_cond);

    /* and wait for machine initialization */
    while (!qemu_system_ready)
        qemu_cond_timedwait(&qemu_system_cond, &qemu_global_mutex, 100);

    while (1) {
        if (cpu_can_run(env))
            qemu_cpu_exec(env);
3652
        qemu_wait_io_event(env);
3653 3654 3655 3656 3657 3658 3659 3660 3661 3662 3663 3664 3665 3666 3667 3668 3669 3670 3671 3672 3673 3674 3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692
    }

    return NULL;
}

static void tcg_cpu_exec(void);

static void *tcg_cpu_thread_fn(void *arg)
{
    CPUState *env = arg;

    block_io_signals();
    qemu_thread_self(env->thread);

    /* signal CPU creation */
    qemu_mutex_lock(&qemu_global_mutex);
    for (env = first_cpu; env != NULL; env = env->next_cpu)
        env->created = 1;
    qemu_cond_signal(&qemu_cpu_cond);

    /* and wait for machine initialization */
    while (!qemu_system_ready)
        qemu_cond_timedwait(&qemu_system_cond, &qemu_global_mutex, 100);

    while (1) {
        tcg_cpu_exec();
        qemu_wait_io_event(cur_cpu);
    }

    return NULL;
}

void qemu_cpu_kick(void *_env)
{
    CPUState *env = _env;
    qemu_cond_broadcast(env->halt_cond);
    if (kvm_enabled())
        qemu_thread_signal(env->thread, SIGUSR1);
}

G
Glauber Costa 已提交
3693
int qemu_cpu_self(void *_env)
3694
{
G
Glauber Costa 已提交
3695 3696 3697 3698 3699 3700
    CPUState *env = _env;
    QemuThread this;
 
    qemu_thread_self(&this);
 
    return qemu_thread_equal(&this, env->thread);
3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 3720 3721 3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734 3735 3736 3737 3738 3739 3740 3741 3742 3743 3744 3745 3746 3747 3748 3749 3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854
}

static void cpu_signal(int sig)
{
    if (cpu_single_env)
        cpu_exit(cpu_single_env);
}

static void block_io_signals(void)
{
    sigset_t set;
    struct sigaction sigact;

    sigemptyset(&set);
    sigaddset(&set, SIGUSR2);
    sigaddset(&set, SIGIO);
    sigaddset(&set, SIGALRM);
    pthread_sigmask(SIG_BLOCK, &set, NULL);

    sigemptyset(&set);
    sigaddset(&set, SIGUSR1);
    pthread_sigmask(SIG_UNBLOCK, &set, NULL);

    memset(&sigact, 0, sizeof(sigact));
    sigact.sa_handler = cpu_signal;
    sigaction(SIGUSR1, &sigact, NULL);
}

static void unblock_io_signals(void)
{
    sigset_t set;

    sigemptyset(&set);
    sigaddset(&set, SIGUSR2);
    sigaddset(&set, SIGIO);
    sigaddset(&set, SIGALRM);
    pthread_sigmask(SIG_UNBLOCK, &set, NULL);

    sigemptyset(&set);
    sigaddset(&set, SIGUSR1);
    pthread_sigmask(SIG_BLOCK, &set, NULL);
}

static void qemu_signal_lock(unsigned int msecs)
{
    qemu_mutex_lock(&qemu_fair_mutex);

    while (qemu_mutex_trylock(&qemu_global_mutex)) {
        qemu_thread_signal(tcg_cpu_thread, SIGUSR1);
        if (!qemu_mutex_timedlock(&qemu_global_mutex, msecs))
            break;
    }
    qemu_mutex_unlock(&qemu_fair_mutex);
}

static void qemu_mutex_lock_iothread(void)
{
    if (kvm_enabled()) {
        qemu_mutex_lock(&qemu_fair_mutex);
        qemu_mutex_lock(&qemu_global_mutex);
        qemu_mutex_unlock(&qemu_fair_mutex);
    } else
        qemu_signal_lock(100);
}

static void qemu_mutex_unlock_iothread(void)
{
    qemu_mutex_unlock(&qemu_global_mutex);
}

static int all_vcpus_paused(void)
{
    CPUState *penv = first_cpu;

    while (penv) {
        if (!penv->stopped)
            return 0;
        penv = (CPUState *)penv->next_cpu;
    }

    return 1;
}

static void pause_all_vcpus(void)
{
    CPUState *penv = first_cpu;

    while (penv) {
        penv->stop = 1;
        qemu_thread_signal(penv->thread, SIGUSR1);
        qemu_cpu_kick(penv);
        penv = (CPUState *)penv->next_cpu;
    }

    while (!all_vcpus_paused()) {
        qemu_cond_timedwait(&qemu_pause_cond, &qemu_global_mutex, 100);
        penv = first_cpu;
        while (penv) {
            qemu_thread_signal(penv->thread, SIGUSR1);
            penv = (CPUState *)penv->next_cpu;
        }
    }
}

static void resume_all_vcpus(void)
{
    CPUState *penv = first_cpu;

    while (penv) {
        penv->stop = 0;
        penv->stopped = 0;
        qemu_thread_signal(penv->thread, SIGUSR1);
        qemu_cpu_kick(penv);
        penv = (CPUState *)penv->next_cpu;
    }
}

static void tcg_init_vcpu(void *_env)
{
    CPUState *env = _env;
    /* share a single thread for all cpus with TCG */
    if (!tcg_cpu_thread) {
        env->thread = qemu_mallocz(sizeof(QemuThread));
        env->halt_cond = qemu_mallocz(sizeof(QemuCond));
        qemu_cond_init(env->halt_cond);
        qemu_thread_create(env->thread, tcg_cpu_thread_fn, env);
        while (env->created == 0)
            qemu_cond_timedwait(&qemu_cpu_cond, &qemu_global_mutex, 100);
        tcg_cpu_thread = env->thread;
        tcg_halt_cond = env->halt_cond;
    } else {
        env->thread = tcg_cpu_thread;
        env->halt_cond = tcg_halt_cond;
    }
}

static void kvm_start_vcpu(CPUState *env)
{
    env->thread = qemu_mallocz(sizeof(QemuThread));
    env->halt_cond = qemu_mallocz(sizeof(QemuCond));
    qemu_cond_init(env->halt_cond);
    qemu_thread_create(env->thread, kvm_cpu_thread_fn, env);
    while (env->created == 0)
        qemu_cond_timedwait(&qemu_cpu_cond, &qemu_global_mutex, 100);
}

void qemu_init_vcpu(void *_env)
{
    CPUState *env = _env;

    if (kvm_enabled())
        kvm_start_vcpu(env);
    else
        tcg_init_vcpu(env);
3855 3856
    env->nr_cores = smp_cores;
    env->nr_threads = smp_threads;
3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886
}

void qemu_notify_event(void)
{
    qemu_event_increment();
}

void vm_stop(int reason)
{
    QemuThread me;
    qemu_thread_self(&me);

    if (!qemu_thread_equal(&me, &io_thread)) {
        qemu_system_vmstop_request(reason);
        /*
         * FIXME: should not return to device code in case
         * vm_stop() has been requested.
         */
        if (cpu_single_env) {
            cpu_exit(cpu_single_env);
            cpu_single_env->stop = 1;
        }
        return;
    }
    do_vm_stop(reason);
}

#endif


T
ths 已提交
3887
#ifdef _WIN32
3888
static void host_main_loop_wait(int *timeout)
A
aliguori 已提交
3889 3890
{
    int ret, ret2, i;
3891 3892
    PollingEntry *pe;

3893

3894 3895 3896 3897 3898
    /* 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);
    }
3899
    if (ret == 0) {
3900 3901
        int err;
        WaitObjects *w = &wait_objects;
3902

A
aliguori 已提交
3903
        ret = WaitForMultipleObjects(w->num, w->events, FALSE, *timeout);
3904 3905 3906
        if (WAIT_OBJECT_0 + 0 <= ret && ret <= WAIT_OBJECT_0 + w->num - 1) {
            if (w->func[ret - WAIT_OBJECT_0])
                w->func[ret - WAIT_OBJECT_0](w->opaque[ret - WAIT_OBJECT_0]);
3907

3908
            /* Check for additional signaled events */
3909
            for(i = (ret - WAIT_OBJECT_0 + 1); i < w->num; i++) {
3910

3911 3912 3913 3914 3915 3916 3917 3918 3919
                /* Check if event is signaled */
                ret2 = WaitForSingleObject(w->events[i], 0);
                if(ret2 == WAIT_OBJECT_0) {
                    if (w->func[i])
                        w->func[i](w->opaque[i]);
                } else if (ret2 == WAIT_TIMEOUT) {
                } else {
                    err = GetLastError();
                    fprintf(stderr, "WaitForSingleObject error %d %d\n", i, err);
3920 3921
                }
            }
3922 3923 3924
        } else if (ret == WAIT_TIMEOUT) {
        } else {
            err = GetLastError();
3925
            fprintf(stderr, "WaitForMultipleObjects error %d %d\n", ret, err);
3926
        }
3927
    }
A
aliguori 已提交
3928 3929 3930 3931

    *timeout = 0;
}
#else
3932
static void host_main_loop_wait(int *timeout)
A
aliguori 已提交
3933 3934
{
}
B
bellard 已提交
3935
#endif
A
aliguori 已提交
3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947

void main_loop_wait(int timeout)
{
    IOHandlerRecord *ioh;
    fd_set rfds, wfds, xfds;
    int ret, nfds;
    struct timeval tv;

    qemu_bh_update_timeout(&timeout);

    host_main_loop_wait(&timeout);

B
bellard 已提交
3948 3949
    /* poll any events */
    /* XXX: separate device handlers from system ones */
3950
    nfds = -1;
B
bellard 已提交
3951 3952
    FD_ZERO(&rfds);
    FD_ZERO(&wfds);
3953
    FD_ZERO(&xfds);
B
bellard 已提交
3954
    for(ioh = first_io_handler; ioh != NULL; ioh = ioh->next) {
3955 3956
        if (ioh->deleted)
            continue;
B
bellard 已提交
3957 3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969
        if (ioh->fd_read &&
            (!ioh->fd_read_poll ||
             ioh->fd_read_poll(ioh->opaque) != 0)) {
            FD_SET(ioh->fd, &rfds);
            if (ioh->fd > nfds)
                nfds = ioh->fd;
        }
        if (ioh->fd_write) {
            FD_SET(ioh->fd, &wfds);
            if (ioh->fd > nfds)
                nfds = ioh->fd;
        }
    }
3970

A
aliguori 已提交
3971 3972 3973
    tv.tv_sec = timeout / 1000;
    tv.tv_usec = (timeout % 1000) * 1000;

J
Jan Kiszka 已提交
3974 3975
    slirp_select_fill(&nfds, &rfds, &wfds, &xfds);

3976
    qemu_mutex_unlock_iothread();
3977
    ret = select(nfds + 1, &rfds, &wfds, &xfds, &tv);
3978
    qemu_mutex_lock_iothread();
B
bellard 已提交
3979
    if (ret > 0) {
3980 3981 3982
        IOHandlerRecord **pioh;

        for(ioh = first_io_handler; ioh != NULL; ioh = ioh->next) {
3983
            if (!ioh->deleted && ioh->fd_read && FD_ISSET(ioh->fd, &rfds)) {
B
bellard 已提交
3984
                ioh->fd_read(ioh->opaque);
B
bellard 已提交
3985
            }
3986
            if (!ioh->deleted && ioh->fd_write && FD_ISSET(ioh->fd, &wfds)) {
B
bellard 已提交
3987
                ioh->fd_write(ioh->opaque);
3988
            }
B
bellard 已提交
3989
        }
3990 3991 3992 3993 3994 3995 3996 3997

	/* remove deleted IO handlers */
	pioh = &first_io_handler;
	while (*pioh) {
            ioh = *pioh;
            if (ioh->deleted) {
                *pioh = ioh->next;
                qemu_free(ioh);
3998
            } else
3999 4000
                pioh = &ioh->next;
        }
B
bellard 已提交
4001
    }
J
Jan Kiszka 已提交
4002 4003

    slirp_select_poll(&rfds, &wfds, &xfds, (ret < 0));
B
bellard 已提交
4004

4005 4006 4007 4008 4009 4010
    /* rearm timer, if not periodic */
    if (alarm_timer->flags & ALARM_FLAG_EXPIRED) {
        alarm_timer->flags &= ~ALARM_FLAG_EXPIRED;
        qemu_rearm_alarm_timer(alarm_timer);
    }

4011
    /* vm time timers */
4012 4013
    if (vm_running) {
        if (!cur_cpu || likely(!(cur_cpu->singlestep_enabled & SSTEP_NOTIMER)))
J
Jan Kiszka 已提交
4014 4015
            qemu_run_timers(&active_timers[QEMU_CLOCK_VIRTUAL],
                            qemu_get_clock(vm_clock));
4016
    }
4017 4018

    /* real time timers */
J
Jan Kiszka 已提交
4019
    qemu_run_timers(&active_timers[QEMU_CLOCK_REALTIME],
4020 4021
                    qemu_get_clock(rt_clock));

J
Jan Kiszka 已提交
4022 4023 4024
    qemu_run_timers(&active_timers[QEMU_CLOCK_HOST],
                    qemu_get_clock(host_clock));

P
pbrook 已提交
4025 4026 4027
    /* Check bottom-halves last in case any of the earlier events triggered
       them.  */
    qemu_bh_poll();
4028

B
bellard 已提交
4029 4030
}

4031
static int qemu_cpu_exec(CPUState *env)
B
bellard 已提交
4032
{
4033
    int ret;
4034 4035 4036
#ifdef CONFIG_PROFILER
    int64_t ti;
#endif
B
bellard 已提交
4037

4038
#ifdef CONFIG_PROFILER
4039
    ti = profile_getclock();
4040
#endif
4041 4042 4043 4044 4045 4046 4047 4048 4049 4050 4051 4052 4053 4054 4055 4056
    if (use_icount) {
        int64_t count;
        int decr;
        qemu_icount -= (env->icount_decr.u16.low + env->icount_extra);
        env->icount_decr.u16.low = 0;
        env->icount_extra = 0;
        count = qemu_next_deadline();
        count = (count + (1 << icount_time_shift) - 1)
                >> icount_time_shift;
        qemu_icount += count;
        decr = (count > 0xffff) ? 0xffff : count;
        count -= decr;
        env->icount_decr.u16.low = decr;
        env->icount_extra = count;
    }
    ret = cpu_exec(env);
4057
#ifdef CONFIG_PROFILER
4058
    qemu_time += profile_getclock() - ti;
4059
#endif
4060 4061 4062 4063 4064 4065 4066 4067 4068 4069 4070
    if (use_icount) {
        /* Fold pending instructions back into the
           instruction counter, and clear the interrupt flag.  */
        qemu_icount -= (env->icount_decr.u16.low
                        + env->icount_extra);
        env->icount_decr.u32 = 0;
        env->icount_extra = 0;
    }
    return ret;
}

4071 4072
static void tcg_cpu_exec(void)
{
4073
    int ret = 0;
4074 4075 4076 4077 4078 4079 4080 4081 4082 4083 4084 4085

    if (next_cpu == NULL)
        next_cpu = first_cpu;
    for (; next_cpu != NULL; next_cpu = next_cpu->next_cpu) {
        CPUState *env = cur_cpu = next_cpu;

        if (!vm_running)
            break;
        if (timer_alarm_pending) {
            timer_alarm_pending = 0;
            break;
        }
4086 4087
        if (cpu_can_run(env))
            ret = qemu_cpu_exec(env);
4088 4089 4090 4091 4092 4093 4094 4095
        if (ret == EXCP_DEBUG) {
            gdb_set_stop_cpu(env);
            debug_requested = 1;
            break;
        }
    }
}

4096 4097
static int cpu_has_work(CPUState *env)
{
4098 4099 4100 4101
    if (env->stop)
        return 1;
    if (env->stopped)
        return 0;
4102 4103 4104 4105 4106 4107 4108 4109 4110 4111 4112 4113 4114 4115 4116 4117 4118 4119 4120
    if (!env->halted)
        return 1;
    if (qemu_cpu_has_work(env))
        return 1;
    return 0;
}

static int tcg_has_work(void)
{
    CPUState *env;

    for (env = first_cpu; env != NULL; env = env->next_cpu)
        if (cpu_has_work(env))
            return 1;
    return 0;
}

static int qemu_calculate_timeout(void)
{
4121
#ifndef CONFIG_IOTHREAD
4122 4123 4124 4125 4126 4127 4128 4129 4130 4131 4132 4133 4134 4135 4136 4137 4138 4139 4140 4141 4142 4143 4144 4145 4146 4147 4148 4149 4150 4151 4152 4153 4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166
    int timeout;

    if (!vm_running)
        timeout = 5000;
    else if (tcg_has_work())
        timeout = 0;
    else if (!use_icount)
        timeout = 5000;
    else {
     /* XXX: use timeout computed from timers */
        int64_t add;
        int64_t delta;
        /* Advance virtual time to the next event.  */
        if (use_icount == 1) {
            /* When not using an adaptive execution frequency
               we tend to get badly out of sync with real time,
               so just delay for a reasonable amount of time.  */
            delta = 0;
        } else {
            delta = cpu_get_icount() - cpu_get_clock();
        }
        if (delta > 0) {
            /* If virtual time is ahead of real time then just
               wait for IO.  */
            timeout = (delta / 1000000) + 1;
        } else {
            /* Wait for either IO to occur or the next
               timer event.  */
            add = qemu_next_deadline();
            /* We advance the timer before checking for IO.
               Limit the amount we advance so that early IO
               activity won't get the guest too far ahead.  */
            if (add > 10000000)
                add = 10000000;
            delta += add;
            add = (add + (1 << icount_time_shift) - 1)
                  >> icount_time_shift;
            qemu_icount += add;
            timeout = delta / 1000000;
            if (timeout < 0)
                timeout = 0;
        }
    }

    return timeout;
4167 4168 4169
#else /* CONFIG_IOTHREAD */
    return 1000;
#endif
4170 4171 4172 4173 4174 4175 4176 4177 4178 4179
}

static int vm_can_run(void)
{
    if (powerdown_requested)
        return 0;
    if (reset_requested)
        return 0;
    if (shutdown_requested)
        return 0;
4180 4181
    if (debug_requested)
        return 0;
4182 4183 4184
    return 1;
}

B
Blue Swirl 已提交
4185 4186
qemu_irq qemu_system_powerdown;

4187 4188
static void main_loop(void)
{
4189
    int r;
4190

4191 4192 4193 4194 4195
#ifdef CONFIG_IOTHREAD
    qemu_system_ready = 1;
    qemu_cond_broadcast(&qemu_system_cond);
#endif

4196
    for (;;) {
4197
        do {
4198 4199 4200
#ifdef CONFIG_PROFILER
            int64_t ti;
#endif
4201
#ifndef CONFIG_IOTHREAD
4202
            tcg_cpu_exec();
4203
#endif
4204
#ifdef CONFIG_PROFILER
4205
            ti = profile_getclock();
4206
#endif
4207
            main_loop_wait(qemu_calculate_timeout());
4208
#ifdef CONFIG_PROFILER
4209
            dev_time += profile_getclock() - ti;
4210
#endif
4211
        } while (vm_can_run());
4212

4213 4214
        if (qemu_debug_requested())
            vm_stop(EXCP_DEBUG);
4215 4216 4217 4218 4219 4220 4221
        if (qemu_shutdown_requested()) {
            if (no_shutdown) {
                vm_stop(0);
                no_shutdown = 0;
            } else
                break;
        }
4222 4223
        if (qemu_reset_requested()) {
            pause_all_vcpus();
4224
            qemu_system_reset();
4225 4226
            resume_all_vcpus();
        }
B
Blue Swirl 已提交
4227 4228 4229
        if (qemu_powerdown_requested()) {
            qemu_irq_raise(qemu_system_powerdown);
        }
4230 4231
        if ((r = qemu_vmstop_requested()))
            vm_stop(r);
B
bellard 已提交
4232
    }
4233
    pause_all_vcpus();
B
bellard 已提交
4234 4235
}

P
pbrook 已提交
4236 4237
static void version(void)
{
P
pbrook 已提交
4238
    printf("QEMU PC emulator version " QEMU_VERSION QEMU_PKGVERSION ", Copyright (c) 2003-2008 Fabrice Bellard\n");
P
pbrook 已提交
4239 4240
}

4241
static void help(int exitcode)
4242
{
P
pbrook 已提交
4243 4244
    version();
    printf("usage: %s [options] [disk_image]\n"
4245
           "\n"
4246
           "'disk_image' is a raw hard image image for IDE hard disk 0\n"
B
bellard 已提交
4247
           "\n"
4248 4249 4250 4251 4252 4253 4254
#define DEF(option, opt_arg, opt_enum, opt_help)        \
           opt_help
#define DEFHEADING(text) stringify(text) "\n"
#include "qemu-options.h"
#undef DEF
#undef DEFHEADING
#undef GEN_DOCS
4255
           "\n"
B
bellard 已提交
4256
           "During emulation, the following keys are useful:\n"
B
bellard 已提交
4257 4258 4259
           "ctrl-alt-f      toggle full screen\n"
           "ctrl-alt-n      switch to virtual console 'n'\n"
           "ctrl-alt        toggle mouse and keyboard grab\n"
B
bellard 已提交
4260 4261 4262
           "\n"
           "When using -nographic, press 'ctrl-a h' to get some help.\n"
           ,
4263
           "qemu",
B
bellard 已提交
4264
           DEFAULT_RAM_SIZE,
B
bellard 已提交
4265
#ifndef _WIN32
B
bellard 已提交
4266
           DEFAULT_NETWORK_SCRIPT,
T
ths 已提交
4267
           DEFAULT_NETWORK_DOWN_SCRIPT,
B
bellard 已提交
4268
#endif
4269
           DEFAULT_GDBSTUB_PORT,
B
bellard 已提交
4270
           "/tmp/qemu.log");
4271
    exit(exitcode);
4272 4273
}

4274 4275 4276
#define HAS_ARG 0x0001

enum {
4277 4278 4279 4280 4281 4282 4283
#define DEF(option, opt_arg, opt_enum, opt_help)        \
    opt_enum,
#define DEFHEADING(text)
#include "qemu-options.h"
#undef DEF
#undef DEFHEADING
#undef GEN_DOCS
4284 4285 4286 4287 4288 4289 4290 4291
};

typedef struct QEMUOption {
    const char *name;
    int flags;
    int index;
} QEMUOption;

B
blueswir1 已提交
4292
static const QEMUOption qemu_options[] = {
4293
    { "h", 0, QEMU_OPTION_h },
4294 4295 4296 4297 4298 4299 4300
#define DEF(option, opt_arg, opt_enum, opt_help)        \
    { option, opt_arg, opt_enum },
#define DEFHEADING(text)
#include "qemu-options.h"
#undef DEF
#undef DEFHEADING
#undef GEN_DOCS
4301
    { NULL },
B
bellard 已提交
4302 4303
};

4304
#ifdef HAS_AUDIO
4305
struct soundhw soundhw[] = {
4306
#ifdef HAS_AUDIO_CHOICE
A
aurel32 已提交
4307
#if defined(TARGET_I386) || defined(TARGET_MIPS)
B
bellard 已提交
4308 4309 4310 4311 4312 4313 4314 4315
    {
        "pcspk",
        "PC speaker",
        0,
        1,
        { .init_isa = pcspk_audio_init }
    },
#endif
M
malc 已提交
4316 4317

#ifdef CONFIG_SB16
4318 4319 4320 4321 4322 4323 4324
    {
        "sb16",
        "Creative Sound Blaster 16",
        0,
        1,
        { .init_isa = SB16_init }
    },
M
malc 已提交
4325
#endif
4326

M
malc 已提交
4327 4328 4329 4330 4331 4332 4333 4334 4335 4336
#ifdef CONFIG_CS4231A
    {
        "cs4231a",
        "CS4231A",
        0,
        1,
        { .init_isa = cs4231a_init }
    },
#endif

4337
#ifdef CONFIG_ADLIB
4338 4339
    {
        "adlib",
4340
#ifdef HAS_YMF262
4341
        "Yamaha YMF262 (OPL3)",
4342
#else
4343
        "Yamaha YM3812 (OPL2)",
4344
#endif
4345 4346 4347 4348
        0,
        1,
        { .init_isa = Adlib_init }
    },
4349
#endif
4350

4351
#ifdef CONFIG_GUS
4352 4353 4354 4355 4356 4357 4358
    {
        "gus",
        "Gravis Ultrasound GF1",
        0,
        1,
        { .init_isa = GUS_init }
    },
4359
#endif
4360

M
malc 已提交
4361
#ifdef CONFIG_AC97
B
balrog 已提交
4362 4363 4364 4365 4366 4367 4368
    {
        "ac97",
        "Intel 82801AA AC97 Audio",
        0,
        0,
        { .init_pci = ac97_init }
    },
M
malc 已提交
4369
#endif
B
balrog 已提交
4370

M
malc 已提交
4371
#ifdef CONFIG_ES1370
4372 4373 4374 4375 4376 4377 4378
    {
        "es1370",
        "ENSONIQ AudioPCI ES1370",
        0,
        0,
        { .init_pci = es1370_init }
    },
4379
#endif
4380

M
malc 已提交
4381 4382
#endif /* HAS_AUDIO_CHOICE */

4383 4384 4385 4386 4387 4388 4389 4390 4391 4392 4393 4394 4395 4396 4397
    { NULL, NULL, 0, 0, { NULL } }
};

static void select_soundhw (const char *optarg)
{
    struct soundhw *c;

    if (*optarg == '?') {
    show_valid_cards:

        printf ("Valid sound card names (comma separated):\n");
        for (c = soundhw; c->name; ++c) {
            printf ("%-11s %s\n", c->name, c->descr);
        }
        printf ("\n-soundhw all will enable all of the above\n");
4398 4399 4400
        exit (*optarg != '?');
    }
    else {
4401
        size_t l;
4402 4403 4404 4405
        const char *p;
        char *e;
        int bad_card = 0;

4406 4407 4408 4409 4410 4411
        if (!strcmp (optarg, "all")) {
            for (c = soundhw; c->name; ++c) {
                c->enabled = 1;
            }
            return;
        }
4412

4413
        p = optarg;
4414 4415 4416
        while (*p) {
            e = strchr (p, ',');
            l = !e ? strlen (p) : (size_t) (e - p);
4417 4418

            for (c = soundhw; c->name; ++c) {
4419
                if (!strncmp (c->name, p, l) && !c->name[l]) {
4420
                    c->enabled = 1;
4421 4422 4423
                    break;
                }
            }
4424 4425

            if (!c->name) {
4426 4427 4428 4429 4430 4431 4432 4433 4434 4435 4436 4437 4438 4439 4440 4441 4442 4443 4444
                if (l > 80) {
                    fprintf (stderr,
                             "Unknown sound card name (too big to show)\n");
                }
                else {
                    fprintf (stderr, "Unknown sound card name `%.*s'\n",
                             (int) l, p);
                }
                bad_card = 1;
            }
            p += l + (e != NULL);
        }

        if (bad_card)
            goto show_valid_cards;
    }
}
#endif

4445 4446 4447 4448
static void select_vgahw (const char *p)
{
    const char *opts;

4449
    vga_interface_type = VGA_NONE;
4450
    if (strstart(p, "std", &opts)) {
4451
        vga_interface_type = VGA_STD;
4452
    } else if (strstart(p, "cirrus", &opts)) {
4453
        vga_interface_type = VGA_CIRRUS;
4454
    } else if (strstart(p, "vmware", &opts)) {
4455
        vga_interface_type = VGA_VMWARE;
4456
    } else if (strstart(p, "xenfb", &opts)) {
4457
        vga_interface_type = VGA_XENFB;
4458
    } else if (!strstart(p, "none", &opts)) {
4459 4460 4461 4462
    invalid_vga:
        fprintf(stderr, "Unknown vga type: %s\n", p);
        exit(1);
    }
M
malc 已提交
4463 4464 4465 4466 4467 4468 4469 4470 4471 4472 4473 4474 4475
    while (*opts) {
        const char *nextopt;

        if (strstart(opts, ",retrace=", &nextopt)) {
            opts = nextopt;
            if (strstart(opts, "dumb", &nextopt))
                vga_retrace_method = VGA_RETRACE_DUMB;
            else if (strstart(opts, "precise", &nextopt))
                vga_retrace_method = VGA_RETRACE_PRECISE;
            else goto invalid_vga;
        } else goto invalid_vga;
        opts = nextopt;
    }
4476 4477
}

4478 4479 4480
#ifdef TARGET_I386
static int balloon_parse(const char *arg)
{
4481
    QemuOpts *opts;
4482

4483 4484 4485 4486 4487 4488 4489 4490 4491 4492 4493 4494 4495
    if (strcmp(arg, "none") == 0) {
        return 0;
    }

    if (!strncmp(arg, "virtio", 6)) {
        if (arg[6] == ',') {
            /* have params -> parse them */
            opts = qemu_opts_parse(&qemu_device_opts, arg+7, NULL);
            if (!opts)
                return  -1;
        } else {
            /* create empty opts */
            opts = qemu_opts_create(&qemu_device_opts, NULL, 0);
4496
        }
4497 4498
        qemu_opt_set(opts, "driver", "virtio-balloon-pci");
        return 0;
4499
    }
4500 4501

    return -1;
4502 4503 4504
}
#endif

4505 4506 4507 4508 4509 4510 4511 4512
#ifdef _WIN32
static BOOL WINAPI qemu_ctrl_handler(DWORD type)
{
    exit(STATUS_CONTROL_C_EXIT);
    return TRUE;
}
#endif

4513
int qemu_uuid_parse(const char *str, uint8_t *uuid)
4514 4515 4516 4517 4518 4519 4520 4521 4522 4523 4524 4525 4526
{
    int ret;

    if(strlen(str) != 36)
        return -1;

    ret = sscanf(str, UUID_FMT, &uuid[0], &uuid[1], &uuid[2], &uuid[3],
            &uuid[4], &uuid[5], &uuid[6], &uuid[7], &uuid[8], &uuid[9],
            &uuid[10], &uuid[11], &uuid[12], &uuid[13], &uuid[14], &uuid[15]);

    if(ret != 16)
        return -1;

4527 4528 4529 4530
#ifdef TARGET_I386
    smbios_add_field(1, offsetof(struct smbios_type_1, uuid), 16, uuid);
#endif

4531 4532 4533
    return 0;
}

B
bellard 已提交
4534
#define MAX_NET_CLIENTS 32
B
bellard 已提交
4535

4536 4537 4538 4539 4540 4541 4542
#ifndef _WIN32

static void termsig_handler(int signal)
{
    qemu_system_shutdown_request();
}

4543 4544 4545 4546 4547 4548
static void sigchld_handler(int signal)
{
    waitpid(-1, NULL, WNOHANG);
}

static void sighandler_setup(void)
4549 4550 4551 4552 4553 4554 4555 4556
{
    struct sigaction act;

    memset(&act, 0, sizeof(act));
    act.sa_handler = termsig_handler;
    sigaction(SIGINT,  &act, NULL);
    sigaction(SIGHUP,  &act, NULL);
    sigaction(SIGTERM, &act, NULL);
4557 4558 4559 4560

    act.sa_handler = sigchld_handler;
    act.sa_flags = SA_NOCLDSTOP;
    sigaction(SIGCHLD, &act, NULL);
4561 4562 4563 4564
}

#endif

P
Paul Brook 已提交
4565 4566 4567 4568 4569 4570 4571 4572 4573 4574
#ifdef _WIN32
/* Look for support files in the same directory as the executable.  */
static char *find_datadir(const char *argv0)
{
    char *p;
    char buf[MAX_PATH];
    DWORD len;

    len = GetModuleFileName(NULL, buf, sizeof(buf) - 1);
    if (len == 0) {
B
Blue Swirl 已提交
4575
        return NULL;
P
Paul Brook 已提交
4576 4577 4578 4579 4580 4581 4582 4583 4584 4585 4586 4587 4588 4589 4590 4591 4592 4593 4594 4595 4596 4597 4598 4599 4600
    }

    buf[len] = 0;
    p = buf + len - 1;
    while (p != buf && *p != '\\')
        p--;
    *p = 0;
    if (access(buf, R_OK) == 0) {
        return qemu_strdup(buf);
    }
    return NULL;
}
#else /* !_WIN32 */

/* Find a likely location for support files using the location of the binary.
   For installed binaries this will be "$bindir/../share/qemu".  When
   running from the build tree this will be "$bindir/../pc-bios".  */
#define SHARE_SUFFIX "/share/qemu"
#define BUILD_SUFFIX "/pc-bios"
static char *find_datadir(const char *argv0)
{
    char *dir;
    char *p = NULL;
    char *res;
    char buf[PATH_MAX];
4601
    size_t max_len;
P
Paul Brook 已提交
4602 4603 4604 4605 4606 4607 4608 4609 4610 4611 4612 4613 4614 4615 4616 4617 4618 4619 4620 4621 4622 4623 4624

#if defined(__linux__)
    {
        int len;
        len = readlink("/proc/self/exe", buf, sizeof(buf) - 1);
        if (len > 0) {
            buf[len] = 0;
            p = buf;
        }
    }
#elif defined(__FreeBSD__)
    {
        int len;
        len = readlink("/proc/curproc/file", buf, sizeof(buf) - 1);
        if (len > 0) {
            buf[len] = 0;
            p = buf;
        }
    }
#endif
    /* If we don't have any way of figuring out the actual executable
       location then try argv[0].  */
    if (!p) {
4625
        p = realpath(argv0, buf);
P
Paul Brook 已提交
4626 4627 4628 4629 4630 4631 4632
        if (!p) {
            return NULL;
        }
    }
    dir = dirname(p);
    dir = dirname(dir);

4633 4634 4635 4636
    max_len = strlen(dir) +
        MAX(strlen(SHARE_SUFFIX), strlen(BUILD_SUFFIX)) + 1;
    res = qemu_mallocz(max_len);
    snprintf(res, max_len, "%s%s", dir, SHARE_SUFFIX);
P
Paul Brook 已提交
4637
    if (access(res, R_OK)) {
4638
        snprintf(res, max_len, "%s%s", dir, BUILD_SUFFIX);
P
Paul Brook 已提交
4639 4640 4641 4642 4643
        if (access(res, R_OK)) {
            qemu_free(res);
            res = NULL;
        }
    }
4644

P
Paul Brook 已提交
4645 4646 4647 4648 4649 4650 4651 4652 4653 4654 4655 4656 4657 4658 4659
    return res;
}
#undef SHARE_SUFFIX
#undef BUILD_SUFFIX
#endif

char *qemu_find_file(int type, const char *name)
{
    int len;
    const char *subdir;
    char *buf;

    /* If name contains path separators then try it as a straight path.  */
    if ((strchr(name, '/') || strchr(name, '\\'))
        && access(name, R_OK) == 0) {
4660
        return qemu_strdup(name);
P
Paul Brook 已提交
4661 4662 4663 4664 4665 4666 4667 4668 4669 4670 4671 4672 4673
    }
    switch (type) {
    case QEMU_FILE_TYPE_BIOS:
        subdir = "";
        break;
    case QEMU_FILE_TYPE_KEYMAP:
        subdir = "keymaps/";
        break;
    default:
        abort();
    }
    len = strlen(data_dir) + strlen(name) + strlen(subdir) + 2;
    buf = qemu_mallocz(len);
4674
    snprintf(buf, len, "%s/%s%s", data_dir, subdir, name);
P
Paul Brook 已提交
4675 4676 4677 4678 4679 4680 4681
    if (access(buf, R_OK)) {
        qemu_free(buf);
        return NULL;
    }
    return buf;
}

G
Gerd Hoffmann 已提交
4682 4683 4684 4685 4686 4687 4688 4689 4690 4691
static int device_init_func(QemuOpts *opts, void *opaque)
{
    DeviceState *dev;

    dev = qdev_device_add(opts);
    if (!dev)
        return -1;
    return 0;
}

4692 4693 4694 4695 4696 4697
struct device_config {
    enum {
        DEV_USB,       /* -usbdevice   */
        DEV_BT,        /* -bt          */
    } type;
    const char *cmdline;
B
Blue Swirl 已提交
4698
    QTAILQ_ENTRY(device_config) next;
4699
};
B
Blue Swirl 已提交
4700
QTAILQ_HEAD(, device_config) device_configs = QTAILQ_HEAD_INITIALIZER(device_configs);
4701 4702 4703 4704 4705 4706 4707 4708

static void add_device_config(int type, const char *cmdline)
{
    struct device_config *conf;

    conf = qemu_mallocz(sizeof(*conf));
    conf->type = type;
    conf->cmdline = cmdline;
B
Blue Swirl 已提交
4709
    QTAILQ_INSERT_TAIL(&device_configs, conf, next);
4710 4711 4712 4713 4714 4715 4716
}

static int foreach_device_config(int type, int (*func)(const char *cmdline))
{
    struct device_config *conf;
    int rc;

B
Blue Swirl 已提交
4717
    QTAILQ_FOREACH(conf, &device_configs, next) {
4718 4719 4720 4721 4722 4723 4724 4725 4726
        if (conf->type != type)
            continue;
        rc = func(conf->cmdline);
        if (0 != rc)
            return rc;
    }
    return 0;
}

M
malc 已提交
4727
int main(int argc, char **argv, char **envp)
4728
{
4729
    const char *gdbstub_dev = NULL;
4730
    uint32_t boot_devices_bitmap = 0;
T
ths 已提交
4731
    int i;
4732
    int snapshot, linux_boot, net_boot;
B
bellard 已提交
4733
    const char *initrd_filename;
4734
    const char *kernel_filename, *kernel_cmdline;
J
Jan Kiszka 已提交
4735
    char boot_devices[33] = "cad"; /* default to HD->floppy->CD-ROM */
4736
    DisplayState *ds;
4737
    DisplayChangeListener *dcl;
B
bellard 已提交
4738
    int cyls, heads, secs, translation;
4739
    const char *net_clients[MAX_NET_CLIENTS];
B
bellard 已提交
4740
    int nb_net_clients;
G
Gerd Hoffmann 已提交
4741
    QemuOpts *hda_opts = NULL, *opts;
4742 4743
    int optind;
    const char *r, *optarg;
4744 4745 4746
    CharDriverState *monitor_hds[MAX_MONITOR_DEVICES];
    const char *monitor_devices[MAX_MONITOR_DEVICES];
    int monitor_device_index;
4747
    const char *serial_devices[MAX_SERIAL_PORTS];
4748
    int serial_device_index;
4749
    const char *parallel_devices[MAX_PARALLEL_PORTS];
4750
    int parallel_device_index;
4751 4752
    const char *virtio_consoles[MAX_VIRTIO_CONSOLES];
    int virtio_console_index;
B
bellard 已提交
4753
    const char *loadvm = NULL;
4754
    QEMUMachine *machine;
4755
    const char *cpu_model;
B
blueswir1 已提交
4756
#ifndef _WIN32
T
ths 已提交
4757
    int fds[2];
B
blueswir1 已提交
4758
#endif
4759
    int tb_size;
4760
    const char *pid_file = NULL;
A
aliguori 已提交
4761
    const char *incoming = NULL;
B
blueswir1 已提交
4762
#ifndef _WIN32
4763 4764
    int fd = 0;
    struct passwd *pwd = NULL;
4765 4766
    const char *chroot_dir = NULL;
    const char *run_as = NULL;
B
blueswir1 已提交
4767
#endif
4768
    CPUState *env;
4769
    int show_vnc_port = 0;
B
bellard 已提交
4770

J
Jan Kiszka 已提交
4771 4772
    init_clocks();

G
Gerd Hoffmann 已提交
4773
    qemu_errors_to_file(stderr);
M
malc 已提交
4774 4775
    qemu_cache_utils_init(envp);

B
Blue Swirl 已提交
4776
    QLIST_INIT (&vm_change_state_head);
4777 4778 4779 4780 4781 4782 4783 4784
#ifndef _WIN32
    {
        struct sigaction act;
        sigfillset(&act.sa_mask);
        act.sa_flags = 0;
        act.sa_handler = SIG_IGN;
        sigaction(SIGPIPE, &act, NULL);
    }
4785 4786
#else
    SetConsoleCtrlHandler(qemu_ctrl_handler, TRUE);
4787 4788 4789 4790 4791 4792 4793 4794 4795 4796 4797 4798 4799 4800 4801 4802 4803 4804
    /* Note: cpu_interrupt() is currently not SMP safe, so we force
       QEMU to run on a single CPU */
    {
        HANDLE h;
        DWORD mask, smask;
        int i;
        h = GetCurrentProcess();
        if (GetProcessAffinityMask(h, &mask, &smask)) {
            for(i = 0; i < 32; i++) {
                if (mask & (1 << i))
                    break;
            }
            if (i != 32) {
                mask = 1 << i;
                SetProcessAffinityMask(h, mask);
            }
        }
    }
B
bellard 已提交
4805
#endif
4806

4807
    module_call_init(MODULE_INIT_MACHINE);
4808
    machine = find_default_machine();
4809
    cpu_model = NULL;
B
bellard 已提交
4810
    initrd_filename = NULL;
4811
    ram_size = 0;
4812
    snapshot = 0;
4813 4814
    kernel_filename = NULL;
    kernel_cmdline = "";
4815
    cyls = heads = secs = 0;
B
bellard 已提交
4816
    translation = BIOS_ATA_TRANSLATION_AUTO;
4817

A
aurel32 已提交
4818
    serial_devices[0] = "vc:80Cx24C";
4819
    for(i = 1; i < MAX_SERIAL_PORTS; i++)
4820
        serial_devices[i] = NULL;
4821
    serial_device_index = 0;
4822

4823
    parallel_devices[0] = "vc:80Cx24C";
4824
    for(i = 1; i < MAX_PARALLEL_PORTS; i++)
4825
        parallel_devices[i] = NULL;
4826
    parallel_device_index = 0;
4827

4828
    for(i = 0; i < MAX_VIRTIO_CONSOLES; i++)
4829 4830 4831
        virtio_consoles[i] = NULL;
    virtio_console_index = 0;

4832 4833 4834 4835 4836 4837
    monitor_devices[0] = "vc:80Cx24C";
    for (i = 1; i < MAX_MONITOR_DEVICES; i++) {
        monitor_devices[i] = NULL;
    }
    monitor_device_index = 0;

4838 4839 4840 4841 4842
    for (i = 0; i < MAX_NODES; i++) {
        node_mem[i] = 0;
        node_cpumask[i] = 0;
    }

B
bellard 已提交
4843
    nb_net_clients = 0;
4844
    nb_numa_nodes = 0;
B
bellard 已提交
4845
    nb_nics = 0;
4846

4847
    tb_size = 0;
4848 4849
    autostart= 1;

4850
    optind = 1;
4851
    for(;;) {
4852
        if (optind >= argc)
4853
            break;
4854 4855
        r = argv[optind];
        if (r[0] != '-') {
G
Gerd Hoffmann 已提交
4856
	    hda_opts = drive_add(argv[optind++], HD_ALIAS, 0);
4857 4858 4859 4860
        } else {
            const QEMUOption *popt;

            optind++;
P
pbrook 已提交
4861 4862 4863
            /* Treat --foo the same as -foo.  */
            if (r[1] == '-')
                r++;
4864 4865 4866
            popt = qemu_options;
            for(;;) {
                if (!popt->name) {
4867
                    fprintf(stderr, "%s: invalid option -- '%s'\n",
4868 4869 4870 4871 4872 4873 4874 4875 4876 4877 4878 4879 4880 4881 4882 4883 4884 4885 4886
                            argv[0], r);
                    exit(1);
                }
                if (!strcmp(popt->name, r + 1))
                    break;
                popt++;
            }
            if (popt->flags & HAS_ARG) {
                if (optind >= argc) {
                    fprintf(stderr, "%s: option '%s' requires an argument\n",
                            argv[0], r);
                    exit(1);
                }
                optarg = argv[optind++];
            } else {
                optarg = NULL;
            }

            switch(popt->index) {
4887 4888 4889 4890 4891 4892
            case QEMU_OPTION_M:
                machine = find_machine(optarg);
                if (!machine) {
                    QEMUMachine *m;
                    printf("Supported machines are:\n");
                    for(m = first_machine; m != NULL; m = m->next) {
M
Mark McLoughlin 已提交
4893 4894 4895
                        if (m->alias)
                            printf("%-10s %s (alias of %s)\n",
                                   m->alias, m->desc, m->name);
4896
                        printf("%-10s %s%s\n",
4897
                               m->name, m->desc,
4898
                               m->is_default ? " (default)" : "");
4899
                    }
4900
                    exit(*optarg != '?');
4901 4902
                }
                break;
4903 4904
            case QEMU_OPTION_cpu:
                /* hw initialization will check this */
4905
                if (*optarg == '?') {
J
j_mayer 已提交
4906 4907 4908
/* XXX: implement xxx_cpu_list for targets that still miss it */
#if defined(cpu_list)
                    cpu_list(stdout, &fprintf);
4909
#endif
4910
                    exit(0);
4911 4912 4913 4914
                } else {
                    cpu_model = optarg;
                }
                break;
4915
            case QEMU_OPTION_initrd:
B
bellard 已提交
4916 4917
                initrd_filename = optarg;
                break;
4918
            case QEMU_OPTION_hda:
T
ths 已提交
4919
                if (cyls == 0)
G
Gerd Hoffmann 已提交
4920
                    hda_opts = drive_add(optarg, HD_ALIAS, 0);
T
ths 已提交
4921
                else
G
Gerd Hoffmann 已提交
4922
                    hda_opts = drive_add(optarg, HD_ALIAS
T
ths 已提交
4923
			     ",cyls=%d,heads=%d,secs=%d%s",
4924
                             0, cyls, heads, secs,
T
ths 已提交
4925 4926 4927 4928 4929
                             translation == BIOS_ATA_TRANSLATION_LBA ?
                                 ",trans=lba" :
                             translation == BIOS_ATA_TRANSLATION_NONE ?
                                 ",trans=none" : "");
                 break;
4930
            case QEMU_OPTION_hdb:
4931 4932
            case QEMU_OPTION_hdc:
            case QEMU_OPTION_hdd:
4933
                drive_add(optarg, HD_ALIAS, popt->index - QEMU_OPTION_hda);
B
bellard 已提交
4934
                break;
T
ths 已提交
4935
            case QEMU_OPTION_drive:
4936
                drive_add(NULL, "%s", optarg);
T
ths 已提交
4937
	        break;
G
Gerd Hoffmann 已提交
4938 4939 4940 4941
            case QEMU_OPTION_set:
                if (qemu_set_option(optarg) != 0)
                    exit(1);
	        break;
4942
            case QEMU_OPTION_mtdblock:
4943
                drive_add(optarg, MTD_ALIAS);
4944
                break;
4945
            case QEMU_OPTION_sd:
4946
                drive_add(optarg, SD_ALIAS);
4947
                break;
4948
            case QEMU_OPTION_pflash:
4949
                drive_add(optarg, PFLASH_ALIAS);
4950
                break;
4951
            case QEMU_OPTION_snapshot:
4952 4953
                snapshot = 1;
                break;
4954
            case QEMU_OPTION_hdachs:
4955 4956 4957 4958
                {
                    const char *p;
                    p = optarg;
                    cyls = strtol(p, (char **)&p, 0);
B
bellard 已提交
4959 4960
                    if (cyls < 1 || cyls > 16383)
                        goto chs_fail;
4961 4962 4963 4964
                    if (*p != ',')
                        goto chs_fail;
                    p++;
                    heads = strtol(p, (char **)&p, 0);
B
bellard 已提交
4965 4966
                    if (heads < 1 || heads > 16)
                        goto chs_fail;
4967 4968 4969 4970
                    if (*p != ',')
                        goto chs_fail;
                    p++;
                    secs = strtol(p, (char **)&p, 0);
B
bellard 已提交
4971 4972 4973 4974 4975 4976 4977 4978 4979 4980 4981 4982 4983
                    if (secs < 1 || secs > 63)
                        goto chs_fail;
                    if (*p == ',') {
                        p++;
                        if (!strcmp(p, "none"))
                            translation = BIOS_ATA_TRANSLATION_NONE;
                        else if (!strcmp(p, "lba"))
                            translation = BIOS_ATA_TRANSLATION_LBA;
                        else if (!strcmp(p, "auto"))
                            translation = BIOS_ATA_TRANSLATION_AUTO;
                        else
                            goto chs_fail;
                    } else if (*p != '\0') {
4984
                    chs_fail:
B
bellard 已提交
4985 4986
                        fprintf(stderr, "qemu: invalid physical CHS format\n");
                        exit(1);
4987
                    }
G
Gerd Hoffmann 已提交
4988 4989 4990 4991 4992 4993 4994 4995 4996 4997 4998 4999 5000
		    if (hda_opts != NULL) {
                        char num[16];
                        snprintf(num, sizeof(num), "%d", cyls);
                        qemu_opt_set(hda_opts, "cyls", num);
                        snprintf(num, sizeof(num), "%d", heads);
                        qemu_opt_set(hda_opts, "heads", num);
                        snprintf(num, sizeof(num), "%d", secs);
                        qemu_opt_set(hda_opts, "secs", num);
                        if (translation == BIOS_ATA_TRANSLATION_LBA)
                            qemu_opt_set(hda_opts, "trans", "lba");
                        if (translation == BIOS_ATA_TRANSLATION_NONE)
                            qemu_opt_set(hda_opts, "trans", "none");
                    }
5001 5002
                }
                break;
5003 5004 5005 5006 5007 5008 5009
            case QEMU_OPTION_numa:
                if (nb_numa_nodes >= MAX_NODES) {
                    fprintf(stderr, "qemu: too many NUMA nodes\n");
                    exit(1);
                }
                numa_add(optarg);
                break;
5010
            case QEMU_OPTION_nographic:
5011
                display_type = DT_NOGRAPHIC;
5012
                break;
B
balrog 已提交
5013 5014
#ifdef CONFIG_CURSES
            case QEMU_OPTION_curses:
5015
                display_type = DT_CURSES;
B
balrog 已提交
5016 5017
                break;
#endif
5018 5019 5020
            case QEMU_OPTION_portrait:
                graphic_rotate = 1;
                break;
5021
            case QEMU_OPTION_kernel:
5022 5023
                kernel_filename = optarg;
                break;
5024
            case QEMU_OPTION_append:
5025
                kernel_cmdline = optarg;
5026
                break;
5027
            case QEMU_OPTION_cdrom:
5028
                drive_add(optarg, CDROM_ALIAS);
5029
                break;
5030
            case QEMU_OPTION_boot:
5031
                {
J
Jan Kiszka 已提交
5032
                    static const char * const params[] = {
5033
                        "order", "once", "menu", NULL
J
Jan Kiszka 已提交
5034 5035
                    };
                    char buf[sizeof(boot_devices)];
J
Jan Kiszka 已提交
5036
                    char *standard_boot_devices;
J
Jan Kiszka 已提交
5037 5038 5039 5040 5041 5042 5043 5044 5045 5046 5047 5048 5049 5050 5051 5052
                    int legacy = 0;

                    if (!strchr(optarg, '=')) {
                        legacy = 1;
                        pstrcpy(buf, sizeof(buf), optarg);
                    } else if (check_params(buf, sizeof(buf), params, optarg) < 0) {
                        fprintf(stderr,
                                "qemu: unknown boot parameter '%s' in '%s'\n",
                                buf, optarg);
                        exit(1);
                    }

                    if (legacy ||
                        get_param_value(buf, sizeof(buf), "order", optarg)) {
                        boot_devices_bitmap = parse_bootdevices(buf);
                        pstrcpy(boot_devices, sizeof(boot_devices), buf);
5053
                    }
J
Jan Kiszka 已提交
5054 5055 5056 5057 5058 5059 5060 5061 5062
                    if (!legacy) {
                        if (get_param_value(buf, sizeof(buf),
                                            "once", optarg)) {
                            boot_devices_bitmap |= parse_bootdevices(buf);
                            standard_boot_devices = qemu_strdup(boot_devices);
                            pstrcpy(boot_devices, sizeof(boot_devices), buf);
                            qemu_register_reset(restore_boot_devices,
                                                standard_boot_devices);
                        }
5063 5064 5065 5066 5067 5068 5069 5070 5071 5072 5073 5074 5075
                        if (get_param_value(buf, sizeof(buf),
                                            "menu", optarg)) {
                            if (!strcmp(buf, "on")) {
                                boot_menu = 1;
                            } else if (!strcmp(buf, "off")) {
                                boot_menu = 0;
                            } else {
                                fprintf(stderr,
                                        "qemu: invalid option value '%s'\n",
                                        buf);
                                exit(1);
                            }
                        }
J
Jan Kiszka 已提交
5076
                    }
5077 5078
                }
                break;
5079 5080
            case QEMU_OPTION_fda:
            case QEMU_OPTION_fdb:
5081
                drive_add(optarg, FD_ALIAS, popt->index - QEMU_OPTION_fda);
5082
                break;
B
bellard 已提交
5083 5084 5085 5086 5087
#ifdef TARGET_I386
            case QEMU_OPTION_no_fd_bootchk:
                fd_bootchk = 0;
                break;
#endif
B
bellard 已提交
5088 5089 5090
            case QEMU_OPTION_net:
                if (nb_net_clients >= MAX_NET_CLIENTS) {
                    fprintf(stderr, "qemu: too many network clients\n");
5091 5092
                    exit(1);
                }
5093
                net_clients[nb_net_clients] = optarg;
B
bellard 已提交
5094
                nb_net_clients++;
B
bellard 已提交
5095
                break;
B
bellard 已提交
5096 5097
#ifdef CONFIG_SLIRP
            case QEMU_OPTION_tftp:
5098
                legacy_tftp_prefix = optarg;
B
bellard 已提交
5099
                break;
5100
            case QEMU_OPTION_bootp:
5101
                legacy_bootp_filename = optarg;
5102
                break;
B
bellard 已提交
5103
#ifndef _WIN32
B
bellard 已提交
5104
            case QEMU_OPTION_smb:
5105
                net_slirp_smb(optarg);
B
bellard 已提交
5106
                break;
B
bellard 已提交
5107
#endif
B
bellard 已提交
5108
            case QEMU_OPTION_redir:
5109
                net_slirp_redir(optarg);
B
bellard 已提交
5110
                break;
B
bellard 已提交
5111
#endif
5112
            case QEMU_OPTION_bt:
5113
                add_device_config(DEV_BT, optarg);
5114
                break;
5115 5116 5117 5118 5119 5120 5121 5122 5123
#ifdef HAS_AUDIO
            case QEMU_OPTION_audio_help:
                AUD_help ();
                exit (0);
                break;
            case QEMU_OPTION_soundhw:
                select_soundhw (optarg);
                break;
#endif
5124
            case QEMU_OPTION_h:
5125
                help(0);
5126
                break;
P
pbrook 已提交
5127 5128 5129 5130
            case QEMU_OPTION_version:
                version();
                exit(0);
                break;
5131 5132 5133 5134 5135 5136 5137 5138 5139 5140 5141 5142 5143 5144
            case QEMU_OPTION_m: {
                uint64_t value;
                char *ptr;

                value = strtoul(optarg, &ptr, 10);
                switch (*ptr) {
                case 0: case 'M': case 'm':
                    value <<= 20;
                    break;
                case 'G': case 'g':
                    value <<= 30;
                    break;
                default:
                    fprintf(stderr, "qemu: invalid ram size: %s\n", optarg);
5145 5146
                    exit(1);
                }
5147 5148

                /* On 32-bit hosts, QEMU is limited by virtual address space */
5149
                if (value > (2047 << 20) && HOST_LONG_BITS == 32) {
5150 5151 5152
                    fprintf(stderr, "qemu: at most 2047 MB RAM can be simulated\n");
                    exit(1);
                }
A
Anthony Liguori 已提交
5153
                if (value != (uint64_t)(ram_addr_t)value) {
5154 5155 5156 5157
                    fprintf(stderr, "qemu: ram size too large\n");
                    exit(1);
                }
                ram_size = value;
5158
                break;
5159
            }
5160 5161 5162
            case QEMU_OPTION_d:
                {
                    int mask;
B
blueswir1 已提交
5163
                    const CPULogItem *item;
5164

5165 5166 5167
                    mask = cpu_str_to_log_mask(optarg);
                    if (!mask) {
                        printf("Log items (comma separated):\n");
5168 5169 5170 5171
                    for(item = cpu_log_items; item->mask != 0; item++) {
                        printf("%-10s %s\n", item->name, item->help);
                    }
                    exit(1);
5172 5173
                    }
                    cpu_set_log(mask);
5174
                }
5175 5176
                break;
            case QEMU_OPTION_s:
5177
                gdbstub_dev = "tcp::" DEFAULT_GDBSTUB_PORT;
5178
                break;
5179 5180
            case QEMU_OPTION_gdb:
                gdbstub_dev = optarg;
5181 5182
                break;
            case QEMU_OPTION_L:
P
Paul Brook 已提交
5183
                data_dir = optarg;
5184
                break;
5185 5186 5187
            case QEMU_OPTION_bios:
                bios_name = optarg;
                break;
5188 5189 5190
            case QEMU_OPTION_singlestep:
                singlestep = 1;
                break;
5191
            case QEMU_OPTION_S:
5192
                autostart = 0;
5193
                break;
5194
#ifndef _WIN32
5195 5196 5197
	    case QEMU_OPTION_k:
		keyboard_layout = optarg;
		break;
5198
#endif
B
bellard 已提交
5199 5200 5201
            case QEMU_OPTION_localtime:
                rtc_utc = 0;
                break;
5202 5203
            case QEMU_OPTION_vga:
                select_vgahw (optarg);
5204
                break;
5205
#if defined(TARGET_PPC) || defined(TARGET_SPARC)
5206 5207 5208 5209 5210 5211 5212 5213 5214 5215 5216 5217 5218 5219 5220 5221 5222 5223 5224 5225
            case QEMU_OPTION_g:
                {
                    const char *p;
                    int w, h, depth;
                    p = optarg;
                    w = strtol(p, (char **)&p, 10);
                    if (w <= 0) {
                    graphic_error:
                        fprintf(stderr, "qemu: invalid resolution or depth\n");
                        exit(1);
                    }
                    if (*p != 'x')
                        goto graphic_error;
                    p++;
                    h = strtol(p, (char **)&p, 10);
                    if (h <= 0)
                        goto graphic_error;
                    if (*p == 'x') {
                        p++;
                        depth = strtol(p, (char **)&p, 10);
5226
                        if (depth != 8 && depth != 15 && depth != 16 &&
5227 5228 5229 5230 5231 5232 5233
                            depth != 24 && depth != 32)
                            goto graphic_error;
                    } else if (*p == '\0') {
                        depth = graphic_depth;
                    } else {
                        goto graphic_error;
                    }
5234

5235 5236 5237 5238 5239
                    graphic_width = w;
                    graphic_height = h;
                    graphic_depth = depth;
                }
                break;
5240
#endif
T
ths 已提交
5241 5242 5243 5244 5245 5246 5247 5248
            case QEMU_OPTION_echr:
                {
                    char *r;
                    term_escape_char = strtol(optarg, &r, 0);
                    if (r == optarg)
                        printf("Bad argument to echr\n");
                    break;
                }
B
bellard 已提交
5249
            case QEMU_OPTION_monitor:
5250 5251 5252 5253 5254 5255
                if (monitor_device_index >= MAX_MONITOR_DEVICES) {
                    fprintf(stderr, "qemu: too many monitor devices\n");
                    exit(1);
                }
                monitor_devices[monitor_device_index] = optarg;
                monitor_device_index++;
B
bellard 已提交
5256
                break;
5257 5258 5259 5260 5261 5262
            case QEMU_OPTION_chardev:
                opts = qemu_opts_parse(&qemu_chardev_opts, optarg, "backend");
                if (!opts) {
                    fprintf(stderr, "parse error: %s\n", optarg);
                    exit(1);
                }
M
Mark McLoughlin 已提交
5263
                if (qemu_chr_open_opts(opts, NULL) == NULL) {
5264 5265 5266
                    exit(1);
                }
                break;
B
bellard 已提交
5267
            case QEMU_OPTION_serial:
5268 5269 5270 5271
                if (serial_device_index >= MAX_SERIAL_PORTS) {
                    fprintf(stderr, "qemu: too many serial ports\n");
                    exit(1);
                }
5272
                serial_devices[serial_device_index] = optarg;
5273
                serial_device_index++;
B
bellard 已提交
5274
                break;
R
Richard W.M. Jones 已提交
5275
            case QEMU_OPTION_watchdog:
M
Markus Armbruster 已提交
5276 5277 5278 5279 5280 5281
                if (watchdog) {
                    fprintf(stderr,
                            "qemu: only one watchdog option may be given\n");
                    return 1;
                }
                watchdog = optarg;
R
Richard W.M. Jones 已提交
5282 5283 5284 5285 5286 5287 5288
                break;
            case QEMU_OPTION_watchdog_action:
                if (select_watchdog_action(optarg) == -1) {
                    fprintf(stderr, "Unknown -watchdog-action parameter\n");
                    exit(1);
                }
                break;
5289 5290 5291 5292 5293 5294 5295 5296
            case QEMU_OPTION_virtiocon:
                if (virtio_console_index >= MAX_VIRTIO_CONSOLES) {
                    fprintf(stderr, "qemu: too many virtio consoles\n");
                    exit(1);
                }
                virtio_consoles[virtio_console_index] = optarg;
                virtio_console_index++;
                break;
5297 5298 5299 5300 5301
            case QEMU_OPTION_parallel:
                if (parallel_device_index >= MAX_PARALLEL_PORTS) {
                    fprintf(stderr, "qemu: too many parallel ports\n");
                    exit(1);
                }
5302
                parallel_devices[parallel_device_index] = optarg;
5303 5304
                parallel_device_index++;
                break;
B
bellard 已提交
5305 5306 5307 5308 5309 5310
	    case QEMU_OPTION_loadvm:
		loadvm = optarg;
		break;
            case QEMU_OPTION_full_screen:
                full_screen = 1;
                break;
T
ths 已提交
5311
#ifdef CONFIG_SDL
5312 5313 5314
            case QEMU_OPTION_no_frame:
                no_frame = 1;
                break;
T
ths 已提交
5315 5316 5317
            case QEMU_OPTION_alt_grab:
                alt_grab = 1;
                break;
T
ths 已提交
5318 5319 5320
            case QEMU_OPTION_no_quit:
                no_quit = 1;
                break;
5321
            case QEMU_OPTION_sdl:
5322
                display_type = DT_SDL;
5323
                break;
T
ths 已提交
5324
#endif
B
bellard 已提交
5325
            case QEMU_OPTION_pidfile:
5326
                pid_file = optarg;
B
bellard 已提交
5327
                break;
5328 5329 5330 5331
#ifdef TARGET_I386
            case QEMU_OPTION_win2k_hack:
                win2k_install_hack = 1;
                break;
5332 5333 5334
            case QEMU_OPTION_rtc_td_hack:
                rtc_td_hack = 1;
                break;
5335 5336 5337 5338 5339 5340
            case QEMU_OPTION_acpitable:
                if(acpi_table_add(optarg) < 0) {
                    fprintf(stderr, "Wrong acpi table provided\n");
                    exit(1);
                }
                break;
5341 5342 5343 5344 5345 5346
            case QEMU_OPTION_smbios:
                if(smbios_entry_add(optarg) < 0) {
                    fprintf(stderr, "Wrong smbios provided\n");
                    exit(1);
                }
                break;
5347
#endif
A
aliguori 已提交
5348 5349 5350 5351
#ifdef CONFIG_KVM
            case QEMU_OPTION_enable_kvm:
                kvm_allowed = 1;
                break;
B
bellard 已提交
5352
#endif
B
bellard 已提交
5353 5354 5355
            case QEMU_OPTION_usb:
                usb_enabled = 1;
                break;
B
bellard 已提交
5356 5357
            case QEMU_OPTION_usbdevice:
                usb_enabled = 1;
5358 5359 5360
                add_device_config(DEV_USB, optarg);
                break;
            case QEMU_OPTION_device:
G
Gerd Hoffmann 已提交
5361 5362 5363 5364 5365
                opts = qemu_opts_parse(&qemu_device_opts, optarg, "driver");
                if (!opts) {
                    fprintf(stderr, "parse error: %s\n", optarg);
                    exit(1);
                }
B
bellard 已提交
5366
                break;
B
bellard 已提交
5367
            case QEMU_OPTION_smp:
5368
                smp_parse(optarg);
5369
                if (smp_cpus < 1) {
B
bellard 已提交
5370 5371 5372
                    fprintf(stderr, "Invalid number of CPUs\n");
                    exit(1);
                }
5373 5374 5375 5376 5377 5378 5379 5380 5381
                if (max_cpus < smp_cpus) {
                    fprintf(stderr, "maxcpus must be equal to or greater than "
                            "smp\n");
                    exit(1);
                }
                if (max_cpus > 255) {
                    fprintf(stderr, "Unsupported number of maxcpus\n");
                    exit(1);
                }
B
bellard 已提交
5382
                break;
B
bellard 已提交
5383
	    case QEMU_OPTION_vnc:
5384
                display_type = DT_VNC;
5385
		vnc_display = optarg;
B
bellard 已提交
5386
		break;
5387
#ifdef TARGET_I386
B
bellard 已提交
5388 5389 5390
            case QEMU_OPTION_no_acpi:
                acpi_enabled = 0;
                break;
A
aliguori 已提交
5391 5392 5393
            case QEMU_OPTION_no_hpet:
                no_hpet = 1;
                break;
5394 5395 5396 5397 5398
            case QEMU_OPTION_balloon:
                if (balloon_parse(optarg) < 0) {
                    fprintf(stderr, "Unknown -balloon argument %s\n", optarg);
                    exit(1);
                }
5399
                break;
5400
#endif
B
bellard 已提交
5401 5402 5403
            case QEMU_OPTION_no_reboot:
                no_reboot = 1;
                break;
A
aurel32 已提交
5404 5405 5406
            case QEMU_OPTION_no_shutdown:
                no_shutdown = 1;
                break;
5407 5408 5409
            case QEMU_OPTION_show_cursor:
                cursor_hide = 0;
                break;
5410 5411 5412 5413 5414 5415 5416
            case QEMU_OPTION_uuid:
                if(qemu_uuid_parse(optarg, qemu_uuid) < 0) {
                    fprintf(stderr, "Fail to parse UUID string."
                            " Wrong format.\n");
                    exit(1);
                }
                break;
5417
#ifndef _WIN32
T
ths 已提交
5418 5419 5420
	    case QEMU_OPTION_daemonize:
		daemonize = 1;
		break;
5421
#endif
5422 5423 5424 5425 5426 5427 5428 5429
	    case QEMU_OPTION_option_rom:
		if (nb_option_roms >= MAX_OPTION_ROMS) {
		    fprintf(stderr, "Too many option ROMs\n");
		    exit(1);
		}
		option_rom[nb_option_roms] = optarg;
		nb_option_roms++;
		break;
5430
#if defined(TARGET_ARM) || defined(TARGET_M68K)
5431 5432 5433
            case QEMU_OPTION_semihosting:
                semihosting_enabled = 1;
                break;
5434
#endif
T
ths 已提交
5435
            case QEMU_OPTION_name:
A
Andi Kleen 已提交
5436 5437 5438 5439 5440 5441 5442 5443 5444 5445 5446 5447 5448
                qemu_name = qemu_strdup(optarg);
		 {
		     char *p = strchr(qemu_name, ',');
		     if (p != NULL) {
		        *p++ = 0;
			if (strncmp(p, "process=", 8)) {
			    fprintf(stderr, "Unknown subargument %s to -name", p);
			    exit(1);
			}
			p += 8;
			set_proc_name(p);
		     }	
		 }	
T
ths 已提交
5449
                break;
5450
#if defined(TARGET_SPARC) || defined(TARGET_PPC)
B
blueswir1 已提交
5451 5452 5453 5454 5455 5456 5457 5458
            case QEMU_OPTION_prom_env:
                if (nb_prom_envs >= MAX_PROM_ENVS) {
                    fprintf(stderr, "Too many prom variables\n");
                    exit(1);
                }
                prom_envs[nb_prom_envs] = optarg;
                nb_prom_envs++;
                break;
5459 5460 5461 5462
#endif
#ifdef TARGET_ARM
            case QEMU_OPTION_old_param:
                old_param = 1;
5463
                break;
B
blueswir1 已提交
5464
#endif
5465 5466 5467
            case QEMU_OPTION_clock:
                configure_alarms(optarg);
                break;
B
bellard 已提交
5468
            case QEMU_OPTION_startdate:
J
Jan Kiszka 已提交
5469 5470 5471 5472 5473 5474 5475
                configure_rtc_date_offset(optarg, 1);
                break;
            case QEMU_OPTION_rtc:
                opts = qemu_opts_parse(&qemu_rtc_opts, optarg, NULL);
                if (!opts) {
                    fprintf(stderr, "parse error: %s\n", optarg);
                    exit(1);
B
bellard 已提交
5476
                }
J
Jan Kiszka 已提交
5477
                configure_rtc(opts);
B
bellard 已提交
5478
                break;
5479 5480 5481 5482 5483
            case QEMU_OPTION_tb_size:
                tb_size = strtol(optarg, NULL, 0);
                if (tb_size < 0)
                    tb_size = 0;
                break;
P
pbrook 已提交
5484 5485 5486 5487 5488 5489 5490 5491
            case QEMU_OPTION_icount:
                use_icount = 1;
                if (strcmp(optarg, "auto") == 0) {
                    icount_time_shift = -1;
                } else {
                    icount_time_shift = strtol(optarg, NULL, 0);
                }
                break;
A
aliguori 已提交
5492 5493 5494
            case QEMU_OPTION_incoming:
                incoming = optarg;
                break;
5495
#ifndef _WIN32
5496 5497 5498 5499 5500 5501
            case QEMU_OPTION_chroot:
                chroot_dir = optarg;
                break;
            case QEMU_OPTION_runas:
                run_as = optarg;
                break;
5502 5503 5504 5505 5506 5507 5508 5509 5510 5511 5512
#endif
#ifdef CONFIG_XEN
            case QEMU_OPTION_xen_domid:
                xen_domid = atoi(optarg);
                break;
            case QEMU_OPTION_xen_create:
                xen_mode = XEN_CREATE;
                break;
            case QEMU_OPTION_xen_attach:
                xen_mode = XEN_ATTACH;
                break;
5513
#endif
5514
            }
5515 5516
        }
    }
5517

P
Paul Brook 已提交
5518 5519 5520 5521 5522 5523 5524 5525 5526 5527
    /* If no data_dir is specified then try to find it relative to the
       executable path.  */
    if (!data_dir) {
        data_dir = find_datadir(argv[0]);
    }
    /* If all else fails use the install patch specified when building.  */
    if (!data_dir) {
        data_dir = CONFIG_QEMU_SHAREDIR;
    }

5528 5529 5530 5531 5532 5533 5534
    /*
     * Default to max_cpus = smp_cpus, in case the user doesn't
     * specify a max_cpus value.
     */
    if (!max_cpus)
        max_cpus = smp_cpus;

B
balrog 已提交
5535
    machine->max_cpus = machine->max_cpus ?: 1; /* Default to UP */
5536 5537 5538 5539 5540 5541 5542
    if (smp_cpus > machine->max_cpus) {
        fprintf(stderr, "Number of SMP cpus requested (%d), exceeds max cpus "
                "supported by machine `%s' (%d)\n", smp_cpus,  machine->name,
                machine->max_cpus);
        exit(1);
    }

5543
    if (display_type == DT_NOGRAPHIC) {
5544 5545 5546 5547
       if (serial_device_index == 0)
           serial_devices[0] = "stdio";
       if (parallel_device_index == 0)
           parallel_devices[0] = "null";
5548 5549 5550
       if (strncmp(monitor_devices[0], "vc", 2) == 0) {
           monitor_devices[0] = "stdio";
       }
5551 5552
    }

T
ths 已提交
5553 5554 5555 5556 5557 5558 5559 5560 5561 5562 5563 5564 5565 5566 5567
#ifndef _WIN32
    if (daemonize) {
	pid_t pid;

	if (pipe(fds) == -1)
	    exit(1);

	pid = fork();
	if (pid > 0) {
	    uint8_t status;
	    ssize_t len;

	    close(fds[1]);

	again:
5568 5569 5570 5571 5572 5573 5574 5575 5576 5577 5578
            len = read(fds[0], &status, 1);
            if (len == -1 && (errno == EINTR))
                goto again;

            if (len != 1)
                exit(1);
            else if (status == 1) {
                fprintf(stderr, "Could not acquire pidfile\n");
                exit(1);
            } else
                exit(0);
T
ths 已提交
5579
	} else if (pid < 0)
5580
            exit(1);
T
ths 已提交
5581 5582 5583 5584 5585 5586 5587 5588 5589 5590 5591 5592 5593 5594 5595 5596

	setsid();

	pid = fork();
	if (pid > 0)
	    exit(0);
	else if (pid < 0)
	    exit(1);

	umask(027);

        signal(SIGTSTP, SIG_IGN);
        signal(SIGTTOU, SIG_IGN);
        signal(SIGTTIN, SIG_IGN);
    }

5597
    if (pid_file && qemu_create_pidfile(pid_file) != 0) {
5598 5599 5600 5601 5602 5603 5604
        if (daemonize) {
            uint8_t status = 1;
            write(fds[1], &status, 1);
        } else
            fprintf(stderr, "Could not acquire pid file\n");
        exit(1);
    }
B
blueswir1 已提交
5605
#endif
5606

M
Marcelo Tosatti 已提交
5607 5608 5609 5610 5611 5612 5613 5614 5615 5616
    if (kvm_enabled()) {
        int ret;

        ret = kvm_init(smp_cpus);
        if (ret < 0) {
            fprintf(stderr, "failed to initialize KVM\n");
            exit(1);
        }
    }

5617 5618 5619 5620
    if (qemu_init_main_loop()) {
        fprintf(stderr, "qemu_init_main_loop failed\n");
        exit(1);
    }
5621
    linux_boot = (kernel_filename != NULL);
B
balrog 已提交
5622

5623 5624 5625 5626 5627 5628 5629 5630 5631 5632
    if (!linux_boot && *kernel_cmdline != '\0') {
        fprintf(stderr, "-append only allowed with -kernel option\n");
        exit(1);
    }

    if (!linux_boot && initrd_filename != NULL) {
        fprintf(stderr, "-initrd only allowed with -kernel option\n");
        exit(1);
    }

5633 5634
#ifndef _WIN32
    /* Win32 doesn't support line-buffering and requires size >= 2 */
B
bellard 已提交
5635
    setvbuf(stdout, NULL, _IOLBF, 0);
5636
#endif
5637

5638 5639 5640 5641
    if (init_timer_alarm() < 0) {
        fprintf(stderr, "could not initialize alarm timer\n");
        exit(1);
    }
P
pbrook 已提交
5642 5643 5644 5645 5646 5647 5648
    if (use_icount && icount_time_shift < 0) {
        use_icount = 2;
        /* 125MIPS seems a reasonable initial guess at the guest speed.
           It will be corrected fairly quickly anyway.  */
        icount_time_shift = 3;
        init_icount_adjust();
    }
5649

B
bellard 已提交
5650 5651 5652 5653
#ifdef _WIN32
    socket_init();
#endif

B
bellard 已提交
5654 5655 5656
    /* init network clients */
    if (nb_net_clients == 0) {
        /* if no clients, we use a default config */
5657 5658 5659 5660
        net_clients[nb_net_clients++] = "nic";
#ifdef CONFIG_SLIRP
        net_clients[nb_net_clients++] = "user";
#endif
B
bellard 已提交
5661 5662
    }

B
bellard 已提交
5663
    for(i = 0;i < nb_net_clients; i++) {
5664
        if (net_client_parse(net_clients[i]) < 0)
B
bellard 已提交
5665
            exit(1);
B
bellard 已提交
5666
    }
B
bellard 已提交
5667

5668 5669 5670 5671
    net_boot = (boot_devices_bitmap >> ('n' - 'a')) & 0xF;
    net_set_boot_mask(net_boot);

    net_client_check();
5672

5673
    /* init the bluetooth world */
5674 5675
    if (foreach_device_config(DEV_BT, bt_parse))
        exit(1);
5676

5677
    /* init the memory */
P
pbrook 已提交
5678 5679
    if (ram_size == 0)
        ram_size = DEFAULT_RAM_SIZE * 1024 * 1024;
5680

5681 5682 5683
    /* init the dynamic translator */
    cpu_exec_init_all(tb_size * 1024 * 1024);

B
bellard 已提交
5684
    bdrv_init();
5685

T
ths 已提交
5686
    /* we always create the cdrom drive, even if no disk is there */
G
Gerd Hoffmann 已提交
5687
    drive_add(NULL, CDROM_ALIAS);
5688

5689
    /* we always create at least one floppy */
G
Gerd Hoffmann 已提交
5690
    drive_add(NULL, FD_ALIAS, 0);
5691

5692
    /* we always create one sd slot, even if no card is in it */
G
Gerd Hoffmann 已提交
5693
    drive_add(NULL, SD_ALIAS);
5694

T
ths 已提交
5695
    /* open the virtual block devices */
G
Gerd Hoffmann 已提交
5696
    if (snapshot)
G
Gerd Hoffmann 已提交
5697 5698
        qemu_opts_foreach(&qemu_drive_opts, drive_enable_snapshot, NULL, 0);
    if (qemu_opts_foreach(&qemu_drive_opts, drive_init_func, machine, 1) != 0)
G
Gerd Hoffmann 已提交
5699
        exit(1);
5700

J
Juan Quintela 已提交
5701
    vmstate_register(0, &vmstate_timers ,&timers_state);
5702
    register_savevm_live("ram", 0, 3, ram_save_live, NULL, ram_load, NULL);
5703

5704
    /* Maintain compatibility with multiple stdio monitors */
5705
    if (!strcmp(monitor_devices[0],"stdio")) {
5706 5707 5708
        for (i = 0; i < MAX_SERIAL_PORTS; i++) {
            const char *devname = serial_devices[i];
            if (devname && !strcmp(devname,"mon:stdio")) {
5709
                monitor_devices[0] = NULL;
5710 5711
                break;
            } else if (devname && !strcmp(devname,"stdio")) {
5712
                monitor_devices[0] = NULL;
5713 5714 5715 5716 5717 5718
                serial_devices[i] = "mon:stdio";
                break;
            }
        }
    }

5719 5720 5721 5722 5723 5724 5725 5726 5727 5728 5729 5730 5731 5732 5733 5734 5735 5736 5737 5738 5739 5740 5741 5742 5743 5744 5745 5746 5747 5748 5749 5750 5751 5752 5753 5754 5755 5756 5757 5758 5759 5760
    if (nb_numa_nodes > 0) {
        int i;

        if (nb_numa_nodes > smp_cpus) {
            nb_numa_nodes = smp_cpus;
        }

        /* If no memory size if given for any node, assume the default case
         * and distribute the available memory equally across all nodes
         */
        for (i = 0; i < nb_numa_nodes; i++) {
            if (node_mem[i] != 0)
                break;
        }
        if (i == nb_numa_nodes) {
            uint64_t usedmem = 0;

            /* On Linux, the each node's border has to be 8MB aligned,
             * the final node gets the rest.
             */
            for (i = 0; i < nb_numa_nodes - 1; i++) {
                node_mem[i] = (ram_size / nb_numa_nodes) & ~((1 << 23UL) - 1);
                usedmem += node_mem[i];
            }
            node_mem[i] = ram_size - usedmem;
        }

        for (i = 0; i < nb_numa_nodes; i++) {
            if (node_cpumask[i] != 0)
                break;
        }
        /* assigning the VCPUs round-robin is easier to implement, guest OSes
         * must cope with this anyway, because there are BIOSes out there in
         * real machines which also use this scheme.
         */
        if (i == nb_numa_nodes) {
            for (i = 0; i < smp_cpus; i++) {
                node_cpumask[i % nb_numa_nodes] |= 1 << i;
            }
        }
    }

5761 5762 5763 5764 5765 5766 5767 5768 5769 5770 5771 5772 5773 5774 5775
    for (i = 0; i < MAX_MONITOR_DEVICES; i++) {
        const char *devname = monitor_devices[i];
        if (devname && strcmp(devname, "none")) {
            char label[32];
            if (i == 0) {
                snprintf(label, sizeof(label), "monitor");
            } else {
                snprintf(label, sizeof(label), "monitor%d", i);
            }
            monitor_hds[i] = qemu_chr_open(label, devname, NULL);
            if (!monitor_hds[i]) {
                fprintf(stderr, "qemu: could not open monitor device '%s'\n",
                        devname);
                exit(1);
            }
A
aliguori 已提交
5776 5777 5778
        }
    }

5779 5780 5781
    for(i = 0; i < MAX_SERIAL_PORTS; i++) {
        const char *devname = serial_devices[i];
        if (devname && strcmp(devname, "none")) {
5782 5783
            char label[32];
            snprintf(label, sizeof(label), "serial%d", i);
5784
            serial_hds[i] = qemu_chr_open(label, devname, NULL);
5785 5786 5787 5788 5789
            if (!serial_hds[i]) {
                fprintf(stderr, "qemu: could not open serial device '%s'\n",
                        devname);
                exit(1);
            }
5790 5791 5792 5793 5794 5795 5796 5797
        }
    }

    for(i = 0; i < MAX_PARALLEL_PORTS; i++) {
        const char *devname = parallel_devices[i];
        if (devname && strcmp(devname, "none")) {
            char label[32];
            snprintf(label, sizeof(label), "parallel%d", i);
5798
            parallel_hds[i] = qemu_chr_open(label, devname, NULL);
5799 5800 5801 5802 5803 5804 5805 5806 5807 5808 5809 5810 5811
            if (!parallel_hds[i]) {
                fprintf(stderr, "qemu: could not open parallel device '%s'\n",
                        devname);
                exit(1);
            }
        }
    }

    for(i = 0; i < MAX_VIRTIO_CONSOLES; i++) {
        const char *devname = virtio_consoles[i];
        if (devname && strcmp(devname, "none")) {
            char label[32];
            snprintf(label, sizeof(label), "virtcon%d", i);
5812
            virtcon_hds[i] = qemu_chr_open(label, devname, NULL);
5813 5814 5815 5816 5817 5818 5819 5820
            if (!virtcon_hds[i]) {
                fprintf(stderr, "qemu: could not open virtio console '%s'\n",
                        devname);
                exit(1);
            }
        }
    }

P
Paul Brook 已提交
5821 5822
    module_call_init(MODULE_INIT_DEVICE);

M
Markus Armbruster 已提交
5823 5824 5825 5826 5827 5828
    if (watchdog) {
        i = select_watchdog(watchdog);
        if (i > 0)
            exit (i == 1 ? 1 : 0);
    }

5829 5830 5831
    if (machine->compat_props) {
        qdev_prop_register_compat(machine->compat_props);
    }
P
Paul Brook 已提交
5832
    machine->init(ram_size, boot_devices,
5833 5834
                  kernel_filename, kernel_cmdline, initrd_filename, cpu_model);

5835

J
Juan Quintela 已提交
5836 5837 5838 5839 5840
#ifndef _WIN32
    /* must be after terminal init, SDL library changes signal handlers */
    sighandler_setup();
#endif

5841 5842 5843 5844 5845 5846 5847 5848
    for (env = first_cpu; env != NULL; env = env->next_cpu) {
        for (i = 0; i < nb_numa_nodes; i++) {
            if (node_cpumask[i] & (1 << env->cpu_index)) {
                env->numa_node = i;
            }
        }
    }

5849 5850
    current_machine = machine;

5851 5852
    /* init USB devices */
    if (usb_enabled) {
5853
        foreach_device_config(DEV_USB, usb_parse);
5854 5855
    }

5856
    /* init generic devices */
G
Gerd Hoffmann 已提交
5857
    if (qemu_opts_foreach(&qemu_device_opts, device_init_func, NULL, 1) != 0)
5858 5859
        exit(1);

5860 5861
    if (!display_state)
        dumb_display_init();
5862 5863
    /* just use the first displaystate for the moment */
    ds = display_state;
5864 5865 5866 5867 5868 5869 5870 5871 5872 5873 5874 5875 5876 5877 5878

    if (display_type == DT_DEFAULT) {
#if defined(CONFIG_SDL) || defined(CONFIG_COCOA)
        display_type = DT_SDL;
#else
        display_type = DT_VNC;
        vnc_display = "localhost:0,to=99";
        show_vnc_port = 1;
#endif
    }
        

    switch (display_type) {
    case DT_NOGRAPHIC:
        break;
B
balrog 已提交
5879
#if defined(CONFIG_CURSES)
5880 5881 5882
    case DT_CURSES:
        curses_display_init(ds, full_screen);
        break;
B
balrog 已提交
5883
#endif
5884
#if defined(CONFIG_SDL)
5885 5886 5887
    case DT_SDL:
        sdl_display_init(ds, full_screen, no_frame);
        break;
5888
#elif defined(CONFIG_COCOA)
5889 5890 5891
    case DT_SDL:
        cocoa_display_init(ds, full_screen);
        break;
5892
#endif
5893 5894 5895 5896
    case DT_VNC:
        vnc_display_init(ds);
        if (vnc_display_open(ds, vnc_display) < 0)
            exit(1);
5897

5898 5899
        if (show_vnc_port) {
            printf("VNC server running on `%s'\n", vnc_display_local_addr(ds));
5900
        }
5901 5902 5903
        break;
    default:
        break;
5904
    }
5905
    dpy_resize(ds);
5906

5907 5908 5909 5910 5911
    dcl = ds->listeners;
    while (dcl != NULL) {
        if (dcl->dpy_refresh != NULL) {
            ds->gui_timer = qemu_new_timer(rt_clock, gui_update, ds);
            qemu_mod_timer(ds->gui_timer, qemu_get_clock(rt_clock));
T
ths 已提交
5912
        }
5913
        dcl = dcl->next;
T
ths 已提交
5914
    }
5915

5916
    if (display_type == DT_NOGRAPHIC || display_type == DT_VNC) {
B
blueswir1 已提交
5917 5918 5919 5920
        nographic_timer = qemu_new_timer(rt_clock, nographic_update, NULL);
        qemu_mod_timer(nographic_timer, qemu_get_clock(rt_clock));
    }

5921
    text_consoles_set_display(display_state);
A
aliguori 已提交
5922
    qemu_chr_initial_reset();
5923

5924 5925 5926 5927 5928 5929 5930
    for (i = 0; i < MAX_MONITOR_DEVICES; i++) {
        if (monitor_devices[i] && monitor_hds[i]) {
            monitor_init(monitor_hds[i],
                         MONITOR_USE_READLINE |
                         ((i == 0) ? MONITOR_IS_DEFAULT : 0));
        }
    }
B
bellard 已提交
5931

5932
    for(i = 0; i < MAX_SERIAL_PORTS; i++) {
5933
        const char *devname = serial_devices[i];
5934
        if (devname && strcmp(devname, "none")) {
5935
            if (strstart(devname, "vc", 0))
B
bellard 已提交
5936
                qemu_chr_printf(serial_hds[i], "serial%d console\r\n", i);
5937
        }
B
bellard 已提交
5938 5939
    }

5940
    for(i = 0; i < MAX_PARALLEL_PORTS; i++) {
5941
        const char *devname = parallel_devices[i];
5942
        if (devname && strcmp(devname, "none")) {
5943
            if (strstart(devname, "vc", 0))
B
bellard 已提交
5944
                qemu_chr_printf(parallel_hds[i], "parallel%d console\r\n", i);
5945 5946 5947
        }
    }

5948 5949
    for(i = 0; i < MAX_VIRTIO_CONSOLES; i++) {
        const char *devname = virtio_consoles[i];
5950
        if (virtcon_hds[i] && devname) {
5951 5952 5953 5954 5955
            if (strstart(devname, "vc", 0))
                qemu_chr_printf(virtcon_hds[i], "virtio console%d\r\n", i);
        }
    }

5956 5957 5958 5959
    if (gdbstub_dev && gdbserver_start(gdbstub_dev) < 0) {
        fprintf(stderr, "qemu: could not open gdbserver on device '%s'\n",
                gdbstub_dev);
        exit(1);
5960 5961
    }

5962 5963
    qdev_machine_creation_done();

5964 5965 5966 5967 5968
    if (loadvm) {
        if (load_vmstate(cur_mon, loadvm) < 0) {
            autostart = 0;
        }
    }
B
bellard 已提交
5969

G
Glauber Costa 已提交
5970
    if (incoming) {
A
aliguori 已提交
5971
        qemu_start_incoming_migration(incoming);
5972
    } else if (autostart) {
5973
        vm_start();
5974
    }
5975

B
blueswir1 已提交
5976
#ifndef _WIN32
T
ths 已提交
5977 5978 5979 5980 5981 5982 5983 5984 5985 5986 5987 5988
    if (daemonize) {
	uint8_t status = 0;
	ssize_t len;

    again1:
	len = write(fds[1], &status, 1);
	if (len == -1 && (errno == EINTR))
	    goto again1;

	if (len != 1)
	    exit(1);

A
aliguori 已提交
5989
	chdir("/");
5990
	TFR(fd = open("/dev/null", O_RDWR));
T
ths 已提交
5991 5992
	if (fd == -1)
	    exit(1);
5993
    }
T
ths 已提交
5994

5995 5996 5997 5998 5999 6000 6001 6002 6003 6004 6005 6006 6007 6008 6009 6010 6011 6012 6013 6014 6015 6016 6017 6018 6019 6020 6021 6022 6023 6024 6025 6026 6027 6028 6029
    if (run_as) {
        pwd = getpwnam(run_as);
        if (!pwd) {
            fprintf(stderr, "User \"%s\" doesn't exist\n", run_as);
            exit(1);
        }
    }

    if (chroot_dir) {
        if (chroot(chroot_dir) < 0) {
            fprintf(stderr, "chroot failed\n");
            exit(1);
        }
        chdir("/");
    }

    if (run_as) {
        if (setgid(pwd->pw_gid) < 0) {
            fprintf(stderr, "Failed to setgid(%d)\n", pwd->pw_gid);
            exit(1);
        }
        if (setuid(pwd->pw_uid) < 0) {
            fprintf(stderr, "Failed to setuid(%d)\n", pwd->pw_uid);
            exit(1);
        }
        if (setuid(0) != -1) {
            fprintf(stderr, "Dropping privileges failed\n");
            exit(1);
        }
    }

    if (daemonize) {
        dup2(fd, 0);
        dup2(fd, 1);
        dup2(fd, 2);
T
ths 已提交
6030

6031
        close(fd);
T
ths 已提交
6032
    }
B
blueswir1 已提交
6033
#endif
T
ths 已提交
6034

6035
    main_loop();
B
bellard 已提交
6036
    quit_timers();
6037
    net_cleanup();
T
ths 已提交
6038

6039 6040
    return 0;
}