vl.c 154.3 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"
146
#include "hw/loader.h"
A
aurel32 已提交
147
#include "bt-host.h"
148 149 150 151 152 153 154 155 156
#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
157
#include "dma.h"
158 159 160 161
#include "audio/audio.h"
#include "migration.h"
#include "kvm.h"
#include "balloon.h"
K
Kevin Wolf 已提交
162
#include "qemu-option.h"
G
Gerd Hoffmann 已提交
163
#include "qemu-config.h"
164

165
#include "disas.h"
B
bellard 已提交
166

167
#include "exec-all.h"
168

169 170
#include "qemu_socket.h"

J
Jan Kiszka 已提交
171
#include "slirp/libslirp.h"
172

B
Blue Swirl 已提交
173 174
#include "qemu-queue.h"

175 176
//#define DEBUG_NET
//#define DEBUG_SLIRP
177

178
#define DEFAULT_RAM_SIZE 128
179

180 181 182
/* Maximum number of monitor devices */
#define MAX_MONITOR_DEVICES 10

P
Paul Brook 已提交
183
static const char *data_dir;
184
const char *bios_name = NULL;
T
ths 已提交
185
/* Note: drives_table[MAX_DRIVES] is a dummy block driver if none available
B
bellard 已提交
186
   to store the VM snapshots */
B
Blue Swirl 已提交
187 188
struct drivelist drives = QTAILQ_HEAD_INITIALIZER(drives);
struct driveoptlist driveopts = QTAILQ_HEAD_INITIALIZER(driveopts);
M
malc 已提交
189
enum vga_retrace_method vga_retrace_method = VGA_RETRACE_DUMB;
190
static DisplayState *display_state;
191
DisplayType display_type = DT_DEFAULT;
192
const char* keyboard_layout = NULL;
A
Anthony Liguori 已提交
193
ram_addr_t ram_size;
194
int nb_nics;
B
bellard 已提交
195
NICInfo nd_table[MAX_NICS];
196
int vm_running;
P
Paolo Bonzini 已提交
197
int autostart;
198 199
static int rtc_utc = 1;
static int rtc_date_offset = -1; /* -1 means no change */
J
Jan Kiszka 已提交
200
QEMUClock *rtc_clock;
201
int vga_interface_type = VGA_CIRRUS;
202 203 204
#ifdef TARGET_SPARC
int graphic_width = 1024;
int graphic_height = 768;
B
blueswir1 已提交
205
int graphic_depth = 8;
206
#else
207 208
int graphic_width = 800;
int graphic_height = 600;
209
int graphic_depth = 15;
B
blueswir1 已提交
210
#endif
B
blueswir1 已提交
211
static int full_screen = 0;
212
#ifdef CONFIG_SDL
B
blueswir1 已提交
213
static int no_frame = 0;
214
#endif
T
ths 已提交
215
int no_quit = 0;
216
CharDriverState *serial_hds[MAX_SERIAL_PORTS];
217
CharDriverState *parallel_hds[MAX_PARALLEL_PORTS];
218
CharDriverState *virtcon_hds[MAX_VIRTIO_CONSOLES];
219 220
#ifdef TARGET_I386
int win2k_install_hack = 0;
221
int rtc_td_hack = 0;
222
#endif
B
bellard 已提交
223
int usb_enabled = 0;
224
int singlestep = 0;
B
bellard 已提交
225
int smp_cpus = 1;
226
int max_cpus = 0;
227 228
int smp_cores = 1;
int smp_threads = 1;
229
const char *vnc_display;
B
bellard 已提交
230
int acpi_enabled = 1;
A
aliguori 已提交
231
int no_hpet = 0;
B
bellard 已提交
232
int fd_bootchk = 1;
B
bellard 已提交
233
int no_reboot = 0;
A
aurel32 已提交
234
int no_shutdown = 0;
235
int cursor_hide = 1;
236
int graphic_rotate = 0;
J
Jes Sorensen 已提交
237
uint8_t irq0override = 1;
B
blueswir1 已提交
238
#ifndef _WIN32
T
ths 已提交
239
int daemonize = 0;
B
blueswir1 已提交
240
#endif
M
Markus Armbruster 已提交
241
const char *watchdog;
242 243
const char *option_rom[MAX_OPTION_ROMS];
int nb_option_roms;
244
int semihosting_enabled = 0;
245 246 247
#ifdef TARGET_ARM
int old_param = 0;
#endif
T
ths 已提交
248
const char *qemu_name;
T
ths 已提交
249
int alt_grab = 0;
250
int ctrl_grab = 0;
251
#if defined(TARGET_SPARC) || defined(TARGET_PPC)
B
blueswir1 已提交
252 253 254
unsigned int nb_prom_envs = 0;
const char *prom_envs[MAX_PROM_ENVS];
#endif
255
int boot_menu;
256

257 258 259 260
int nb_numa_nodes;
uint64_t node_mem[MAX_NODES];
uint64_t node_cpumask[MAX_NODES];

261 262
static CPUState *cur_cpu;
static CPUState *next_cpu;
263
static int timer_alarm_pending = 1;
T
ths 已提交
264
/* Conversion factor from emulated instructions to virtual clock ticks.  */
P
pbrook 已提交
265
static int icount_time_shift;
T
ths 已提交
266
/* Arbitrarily pick 1MIPS as the minimum allowable speed.  */
P
pbrook 已提交
267 268 269
#define MAX_ICOUNT_SHIFT 10
/* Compensate for varying guest execution speed.  */
static int64_t qemu_icount_bias;
B
blueswir1 已提交
270 271
static QEMUTimer *icount_rt_timer;
static QEMUTimer *icount_vm_timer;
B
blueswir1 已提交
272
static QEMUTimer *nographic_timer;
273

274 275
uint8_t qemu_uuid[16];

J
Jan Kiszka 已提交
276 277 278
static QEMUBootSetHandler *boot_set_handler;
static void *boot_set_opaque;

279
/***********************************************************/
280 281
/* x86 ISA bus support */

A
Anthony Liguori 已提交
282
target_phys_addr_t isa_mem_base = 0;
B
bellard 已提交
283
PicState2 *isa_pic;
284 285 286 287 288

/***********************************************************/
void hw_error(const char *fmt, ...)
{
    va_list ap;
B
bellard 已提交
289
    CPUState *env;
290 291 292 293 294

    va_start(ap, fmt);
    fprintf(stderr, "qemu: hardware error: ");
    vfprintf(stderr, fmt, ap);
    fprintf(stderr, "\n");
B
bellard 已提交
295 296
    for(env = first_cpu; env != NULL; env = env->next_cpu) {
        fprintf(stderr, "CPU #%d:\n", env->cpu_index);
297
#ifdef TARGET_I386
B
bellard 已提交
298
        cpu_dump_state(env, stderr, fprintf, X86_DUMP_FPU);
299
#else
B
bellard 已提交
300
        cpu_dump_state(env, stderr, fprintf, 0);
301
#endif
B
bellard 已提交
302
    }
303 304 305
    va_end(ap);
    abort();
}
A
Andi Kleen 已提交
306 307 308

static void set_proc_name(const char *s)
{
309
#if defined(__linux__) && defined(PR_SET_NAME)
A
Andi Kleen 已提交
310 311 312 313 314 315 316 317 318 319
    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
320 321 322 323 324 325 326 327 328 329 330 331 332
 
/***************/
/* 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
333
void qemu_balloon(ram_addr_t target)
A
aliguori 已提交
334 335 336 337 338
{
    if (qemu_balloon_event)
        qemu_balloon_event(qemu_balloon_event_opaque, target);
}

A
Anthony Liguori 已提交
339
ram_addr_t qemu_balloon_status(void)
A
aliguori 已提交
340 341 342 343 344
{
    if (qemu_balloon_event)
        return qemu_balloon_event(qemu_balloon_event_opaque, 0);
    return 0;
}
345

B
bellard 已提交
346 347 348 349 350
/***********************************************************/
/* keyboard/mouse */

static QEMUPutKBDEvent *qemu_put_kbd_event;
static void *qemu_put_kbd_event_opaque;
351 352
static QEMUPutMouseEntry *qemu_put_mouse_event_head;
static QEMUPutMouseEntry *qemu_put_mouse_event_current;
B
bellard 已提交
353 354 355 356 357 358 359

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

360 361 362
QEMUPutMouseEntry *qemu_add_mouse_event_handler(QEMUPutMouseEvent *func,
                                                void *opaque, int absolute,
                                                const char *name)
B
bellard 已提交
363
{
364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419
    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
420 421 422 423 424 425 426 427 428 429 430
}

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)
{
431 432
    QEMUPutMouseEvent *mouse_event;
    void *mouse_event_opaque;
433
    int width;
434 435 436 437 438 439 440 441 442 443 444

    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) {
445 446 447 448
        if (graphic_rotate) {
            if (qemu_put_mouse_event_current->qemu_put_mouse_event_absolute)
                width = 0x7fff;
            else
449
                width = graphic_width - 1;
450 451 452 453 454
            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
455 456 457
    }
}

458 459
int kbd_mouse_is_absolute(void)
{
460 461 462 463 464 465
    if (!qemu_put_mouse_event_current)
        return 0;

    return qemu_put_mouse_event_current->qemu_put_mouse_event_absolute;
}

A
aliguori 已提交
466
void do_info_mice(Monitor *mon)
467 468 469 470 471
{
    QEMUPutMouseEntry *cursor;
    int index = 0;

    if (!qemu_put_mouse_event_head) {
A
aliguori 已提交
472
        monitor_printf(mon, "No mouse devices connected\n");
473 474 475
        return;
    }

A
aliguori 已提交
476
    monitor_printf(mon, "Mouse devices available:\n");
477 478
    cursor = qemu_put_mouse_event_head;
    while (cursor != NULL) {
A
aliguori 已提交
479 480 481
        monitor_printf(mon, "%c Mouse #%d: %s\n",
                       (cursor == qemu_put_mouse_event_current ? '*' : ' '),
                       index, cursor->qemu_put_mouse_event_name);
482 483 484 485 486
        index++;
        cursor = cursor->next;
    }
}

487
void do_mouse_set(Monitor *mon, const QDict *qdict)
488 489 490
{
    QEMUPutMouseEntry *cursor;
    int i = 0;
491
    int index = qdict_get_int(qdict, "index");
492 493

    if (!qemu_put_mouse_event_head) {
A
aliguori 已提交
494
        monitor_printf(mon, "No mouse devices connected\n");
495 496 497 498 499 500 501 502 503 504 505 506
        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
507
        monitor_printf(mon, "Mouse at given index not found\n");
508 509
}

510 511
/* compute with 96 bit intermediate result: (a*b)/c */
uint64_t muldiv64(uint64_t a, uint32_t b, uint32_t c)
512
{
513 514 515
    union {
        uint64_t ll;
        struct {
516
#ifdef HOST_WORDS_BIGENDIAN
517 518 519
            uint32_t high, low;
#else
            uint32_t low, high;
520
#endif
521 522 523
        } l;
    } u, res;
    uint64_t rl, rh;
524

525 526 527 528 529 530 531
    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;
532 533
}

534 535
/***********************************************************/
/* real time host monotonic timer */
536

J
Jan Kiszka 已提交
537 538 539 540 541 542 543 544
static int64_t get_clock_realtime(void)
{
    struct timeval tv;

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

545
#ifdef WIN32
546

547
static int64_t clock_freq;
B
bellard 已提交
548

549
static void init_get_clock(void)
B
bellard 已提交
550
{
551 552
    LARGE_INTEGER freq;
    int ret;
553 554 555 556 557 558
    ret = QueryPerformanceFrequency(&freq);
    if (ret == 0) {
        fprintf(stderr, "Could not calibrate ticks\n");
        exit(1);
    }
    clock_freq = freq.QuadPart;
B
bellard 已提交
559 560
}

561
static int64_t get_clock(void)
B
bellard 已提交
562
{
563 564
    LARGE_INTEGER ti;
    QueryPerformanceCounter(&ti);
565
    return muldiv64(ti.QuadPart, get_ticks_per_sec(), clock_freq);
B
bellard 已提交
566 567
}

568
#else
B
bellard 已提交
569

570 571 572
static int use_rt_clock;

static void init_get_clock(void)
B
bellard 已提交
573
{
574
    use_rt_clock = 0;
575 576
#if defined(__linux__) || (defined(__FreeBSD__) && __FreeBSD_version >= 500000) \
    || defined(__DragonFly__)
577 578 579 580 581 582 583
    {
        struct timespec ts;
        if (clock_gettime(CLOCK_MONOTONIC, &ts) == 0) {
            use_rt_clock = 1;
        }
    }
#endif
B
bellard 已提交
584 585
}

586
static int64_t get_clock(void)
B
bellard 已提交
587
{
588 589
#if defined(__linux__) || (defined(__FreeBSD__) && __FreeBSD_version >= 500000) \
	|| defined(__DragonFly__)
590 591 592 593
    if (use_rt_clock) {
        struct timespec ts;
        clock_gettime(CLOCK_MONOTONIC, &ts);
        return ts.tv_sec * 1000000000LL + ts.tv_nsec;
594
    } else
B
bellard 已提交
595
#endif
596 597 598
    {
        /* XXX: using gettimeofday leads to problems if the date
           changes, so it should be avoided. */
J
Jan Kiszka 已提交
599
        return get_clock_realtime();
600
    }
B
bellard 已提交
601
}
602 603
#endif

P
pbrook 已提交
604 605 606 607 608 609 610 611 612 613 614 615 616 617
/* 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);
}

618 619 620
/***********************************************************/
/* guest cycle counter */

621 622 623 624 625
typedef struct TimersState {
    int64_t cpu_ticks_prev;
    int64_t cpu_ticks_offset;
    int64_t cpu_clock_offset;
    int32_t cpu_ticks_enabled;
626
    int64_t dummy;
627 628 629
} TimersState;

TimersState timers_state;
630

631 632
/* return the host CPU cycle counter and handle stop/restart */
int64_t cpu_get_ticks(void)
633
{
P
pbrook 已提交
634 635 636
    if (use_icount) {
        return cpu_get_icount();
    }
637 638
    if (!timers_state.cpu_ticks_enabled) {
        return timers_state.cpu_ticks_offset;
639
    } else {
640 641
        int64_t ticks;
        ticks = cpu_get_real_ticks();
642
        if (timers_state.cpu_ticks_prev > ticks) {
643 644
            /* Note: non increasing ticks may happen if the host uses
               software suspend */
645
            timers_state.cpu_ticks_offset += timers_state.cpu_ticks_prev - ticks;
646
        }
647 648
        timers_state.cpu_ticks_prev = ticks;
        return ticks + timers_state.cpu_ticks_offset;
649
    }
650 651
}

652 653 654 655
/* return the host CPU monotonic timer and handle stop/restart */
static int64_t cpu_get_clock(void)
{
    int64_t ti;
656 657
    if (!timers_state.cpu_ticks_enabled) {
        return timers_state.cpu_clock_offset;
658 659
    } else {
        ti = get_clock();
660
        return ti + timers_state.cpu_clock_offset;
661 662 663
    }
}

664 665 666
/* enable cpu_get_ticks() */
void cpu_enable_ticks(void)
{
667 668 669 670
    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;
671
    }
672 673 674 675 676 677
}

/* disable cpu_get_ticks() : the clock is stopped. You must not call
   cpu_get_ticks() after that.  */
void cpu_disable_ticks(void)
{
678 679 680 681
    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;
682
    }
683 684
}

685 686
/***********************************************************/
/* timers */
687

J
Jan Kiszka 已提交
688 689
#define QEMU_CLOCK_REALTIME 0
#define QEMU_CLOCK_VIRTUAL  1
J
Jan Kiszka 已提交
690
#define QEMU_CLOCK_HOST     2
691 692 693 694 695 696 697 698 699 700 701 702 703 704

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

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

705 706
struct qemu_alarm_timer {
    char const *name;
707
    unsigned int flags;
708 709 710

    int (*start)(struct qemu_alarm_timer *t);
    void (*stop)(struct qemu_alarm_timer *t);
711
    void (*rearm)(struct qemu_alarm_timer *t);
712 713 714
    void *priv;
};

715
#define ALARM_FLAG_DYNTICKS  0x1
716
#define ALARM_FLAG_EXPIRED   0x2
717 718 719

static inline int alarm_has_dynticks(struct qemu_alarm_timer *t)
{
720
    return t && (t->flags & ALARM_FLAG_DYNTICKS);
721 722 723 724 725 726 727 728 729 730 731 732 733
}

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

734
static struct qemu_alarm_timer *alarm_timer;
735

B
bellard 已提交
736
#ifdef _WIN32
737 738 739 740

struct qemu_alarm_win32 {
    MMRESULT timerId;
    unsigned int period;
B
Blue Swirl 已提交
741
} alarm_win32_data = {0, -1};
742 743 744

static int win32_start_timer(struct qemu_alarm_timer *t);
static void win32_stop_timer(struct qemu_alarm_timer *t);
745
static void win32_rearm_timer(struct qemu_alarm_timer *t);
746

B
bellard 已提交
747
#else
748 749 750 751

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

752 753
#ifdef __linux__

754 755 756 757
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
758 759 760
static int hpet_start_timer(struct qemu_alarm_timer *t);
static void hpet_stop_timer(struct qemu_alarm_timer *t);

761 762 763
static int rtc_start_timer(struct qemu_alarm_timer *t);
static void rtc_stop_timer(struct qemu_alarm_timer *t);

764
#endif /* __linux__ */
765

766 767
#endif /* _WIN32 */

P
pbrook 已提交
768
/* Correlation between real and virtual time is always going to be
T
ths 已提交
769
   fairly approximate, so ignore small variation.
P
pbrook 已提交
770 771
   When the guest is idle real and virtual time will be aligned in
   the IO wait loop.  */
772
#define ICOUNT_WOBBLE (get_ticks_per_sec() / 10)
P
pbrook 已提交
773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813

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,
814
                   qemu_get_clock(vm_clock) + get_ticks_per_sec() / 10);
P
pbrook 已提交
815 816 817 818 819 820 821 822 823 824 825 826 827 828 829
    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,
830
                   qemu_get_clock(vm_clock) + get_ticks_per_sec() / 10);
P
pbrook 已提交
831 832
}

833
static struct qemu_alarm_timer alarm_timers[] = {
834
#ifndef _WIN32
835
#ifdef __linux__
836 837
    {"dynticks", ALARM_FLAG_DYNTICKS, dynticks_start_timer,
     dynticks_stop_timer, dynticks_rearm_timer, NULL},
T
ths 已提交
838
    /* HPET - if available - is preferred */
839
    {"hpet", 0, hpet_start_timer, hpet_stop_timer, NULL, NULL},
T
ths 已提交
840
    /* ...otherwise try RTC */
841
    {"rtc", 0, rtc_start_timer, rtc_stop_timer, NULL, NULL},
842
#endif
843
    {"unix", 0, unix_start_timer, unix_stop_timer, NULL, NULL},
844
#else
845 846 847 848
    {"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},
849 850 851 852
#endif
    {NULL, }
};

853
static void show_available_alarms(void)
854 855 856 857 858 859 860 861 862 863 864 865
{
    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;
866
    int count = ARRAY_SIZE(alarm_timers) - 1;
867 868
    char *arg;
    char *name;
P
pbrook 已提交
869
    struct qemu_alarm_timer tmp;
870

871
    if (!strcmp(opt, "?")) {
872 873 874 875
        show_available_alarms();
        exit(0);
    }

876
    arg = qemu_strdup(opt);
877 878 879 880

    /* Reorder the array */
    name = strtok(arg, ",");
    while (name) {
881
        for (i = 0; i < count && alarm_timers[i].name; i++) {
882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904
            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, ",");
    }

905
    qemu_free(arg);
906 907

    if (cur) {
P
pbrook 已提交
908
        /* Disable remaining timers */
909 910
        for (i = cur; i < count; i++)
            alarm_timers[i].name = NULL;
911 912 913
    } else {
        show_available_alarms();
        exit(1);
914 915 916
    }
}

J
Jan Kiszka 已提交
917 918
#define QEMU_NUM_CLOCKS 3

919 920
QEMUClock *rt_clock;
QEMUClock *vm_clock;
J
Jan Kiszka 已提交
921
QEMUClock *host_clock;
922

J
Jan Kiszka 已提交
923
static QEMUTimer *active_timers[QEMU_NUM_CLOCKS];
924

925
static QEMUClock *qemu_new_clock(int type)
926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984
{
    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;
985
        if (t->expire_time > expire_time)
986 987 988 989 990 991
            break;
        pt = &t->next;
    }
    ts->expire_time = expire_time;
    ts->next = *pt;
    *pt = ts;
992 993

    /* Rearm if necessary  */
P
pbrook 已提交
994 995 996 997 998
    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.  */
999 1000
        if (use_icount)
            qemu_notify_event();
P
pbrook 已提交
1001
    }
1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013
}

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
1014
int qemu_timer_expired(QEMUTimer *timer_head, int64_t current_time)
1015 1016 1017 1018 1019 1020 1021 1022 1023
{
    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;
1024

1025 1026
    for(;;) {
        ts = *ptimer_head;
1027
        if (!ts || ts->expire_time > current_time)
1028 1029 1030 1031
            break;
        /* remove timer from the list before calling the callback */
        *ptimer_head = ts->next;
        ts->next = NULL;
1032

1033 1034 1035 1036 1037 1038 1039 1040
        /* 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
1041
    case QEMU_CLOCK_REALTIME:
1042
        return get_clock() / 1000000;
1043
    default:
J
Jan Kiszka 已提交
1044
    case QEMU_CLOCK_VIRTUAL:
P
pbrook 已提交
1045 1046 1047 1048 1049
        if (use_icount) {
            return cpu_get_icount();
        } else {
            return cpu_get_clock();
        }
J
Jan Kiszka 已提交
1050 1051
    case QEMU_CLOCK_HOST:
        return get_clock_realtime();
1052 1053 1054
    }
}

J
Jan Kiszka 已提交
1055
static void init_clocks(void)
1056 1057
{
    init_get_clock();
J
Jan Kiszka 已提交
1058 1059
    rt_clock = qemu_new_clock(QEMU_CLOCK_REALTIME);
    vm_clock = qemu_new_clock(QEMU_CLOCK_VIRTUAL);
J
Jan Kiszka 已提交
1060
    host_clock = qemu_new_clock(QEMU_CLOCK_HOST);
J
Jan Kiszka 已提交
1061 1062

    rtc_clock = host_clock;
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 1088 1089
/* 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
1090 1091 1092 1093 1094 1095 1096
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),
1097
        VMSTATE_INT64(dummy, TimersState),
J
Juan Quintela 已提交
1098 1099
        VMSTATE_INT64_V(cpu_clock_offset, TimersState, 2),
        VMSTATE_END_OF_LIST()
1100
    }
J
Juan Quintela 已提交
1101
};
1102

1103 1104
static void qemu_event_increment(void);

B
bellard 已提交
1105
#ifdef _WIN32
B
blueswir1 已提交
1106 1107 1108
static void CALLBACK host_alarm_handler(UINT uTimerID, UINT uMsg,
                                        DWORD_PTR dwUser, DWORD_PTR dw1,
                                        DWORD_PTR dw2)
B
bellard 已提交
1109
#else
1110
static void host_alarm_handler(int host_signum)
B
bellard 已提交
1111
#endif
1112
{
1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127
#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
1128
                printf("timer: min=%" PRId64 " us max=%" PRId64 " us avg=%" PRId64 " us avg_freq=%0.3f Hz\n",
1129 1130 1131 1132
                       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));
1133 1134 1135 1136 1137 1138 1139 1140 1141
                count = 0;
                delta_min = INT64_MAX;
                delta_max = 0;
                delta_cum = 0;
            }
        }
        last_clock = ti;
    }
#endif
1142
    if (alarm_has_dynticks(alarm_timer) ||
P
pbrook 已提交
1143
        (!use_icount &&
J
Jan Kiszka 已提交
1144
            qemu_timer_expired(active_timers[QEMU_CLOCK_VIRTUAL],
P
pbrook 已提交
1145
                               qemu_get_clock(vm_clock))) ||
J
Jan Kiszka 已提交
1146
        qemu_timer_expired(active_timers[QEMU_CLOCK_REALTIME],
J
Jan Kiszka 已提交
1147 1148 1149
                           qemu_get_clock(rt_clock)) ||
        qemu_timer_expired(active_timers[QEMU_CLOCK_HOST],
                           qemu_get_clock(host_clock))) {
1150
        qemu_event_increment();
1151
        if (alarm_timer) alarm_timer->flags |= ALARM_FLAG_EXPIRED;
1152

1153 1154
#ifndef CONFIG_IOTHREAD
        if (next_cpu) {
1155
            /* stop the currently executing cpu because a timer occured */
1156
            cpu_exit(next_cpu);
1157
        }
1158
#endif
1159
        timer_alarm_pending = 1;
1160
        qemu_notify_event();
1161 1162 1163
    }
}

P
pbrook 已提交
1164
static int64_t qemu_next_deadline(void)
1165
{
J
Jan Kiszka 已提交
1166 1167
    /* To avoid problems with overflow limit this to 2^32.  */
    int64_t delta = INT32_MAX;
1168

J
Jan Kiszka 已提交
1169 1170
    if (active_timers[QEMU_CLOCK_VIRTUAL]) {
        delta = active_timers[QEMU_CLOCK_VIRTUAL]->expire_time -
P
pbrook 已提交
1171
                     qemu_get_clock(vm_clock);
J
Jan Kiszka 已提交
1172 1173 1174 1175 1176 1177
    }
    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;
1178 1179
    }

P
pbrook 已提交
1180 1181
    if (delta < 0)
        delta = 0;
1182

P
pbrook 已提交
1183 1184 1185
    return delta;
}

J
Jan Kiszka 已提交
1186
#if defined(__linux__)
P
pbrook 已提交
1187 1188 1189 1190 1191 1192 1193 1194 1195 1196
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
1197 1198
    if (active_timers[QEMU_CLOCK_REALTIME]) {
        rtdelta = (active_timers[QEMU_CLOCK_REALTIME]->expire_time -
P
pbrook 已提交
1199 1200 1201 1202 1203 1204 1205 1206 1207
                 qemu_get_clock(rt_clock))*1000;
        if (rtdelta < delta)
            delta = rtdelta;
    }

    if (delta < MIN_TIMER_REARM_US)
        delta = MIN_TIMER_REARM_US;

    return delta;
1208
}
1209
#endif
1210

1211 1212
#ifndef _WIN32

1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227
/* 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
1228 1229
#if defined(__linux__)

1230 1231
#define RTC_FREQ 1024

1232
static void enable_sigio_timer(int fd)
1233 1234 1235 1236 1237 1238 1239 1240 1241
{
    struct sigaction act;

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

    sigaction(SIGIO, &act, NULL);
1242
    fcntl_setfl(fd, O_ASYNC);
1243 1244
    fcntl(fd, F_SETOWN, getpid());
}
B
BSD fix  
bellard 已提交
1245

T
ths 已提交
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 1278 1279
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);
1280
    t->priv = (void *)(long)fd;
T
ths 已提交
1281 1282 1283 1284 1285 1286 1287 1288 1289

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

static void hpet_stop_timer(struct qemu_alarm_timer *t)
{
1290
    int fd = (long)t->priv;
T
ths 已提交
1291 1292 1293 1294

    close(fd);
}

1295
static int rtc_start_timer(struct qemu_alarm_timer *t)
1296
{
1297
    int rtc_fd;
1298
    unsigned long current_rtc_freq = 0;
1299

1300
    TFR(rtc_fd = open("/dev/rtc", O_RDONLY));
1301 1302
    if (rtc_fd < 0)
        return -1;
1303 1304 1305
    ioctl(rtc_fd, RTC_IRQP_READ, &current_rtc_freq);
    if (current_rtc_freq != RTC_FREQ &&
        ioctl(rtc_fd, RTC_IRQP_SET, RTC_FREQ) < 0) {
1306 1307 1308 1309 1310 1311 1312 1313 1314 1315
        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;
    }
1316 1317 1318

    enable_sigio_timer(rtc_fd);

1319
    t->priv = (void *)(long)rtc_fd;
1320

1321 1322 1323
    return 0;
}

1324
static void rtc_stop_timer(struct qemu_alarm_timer *t)
B
BSD fix  
bellard 已提交
1325
{
1326
    int rtc_fd = (long)t->priv;
1327 1328

    close(rtc_fd);
B
BSD fix  
bellard 已提交
1329 1330
}

1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342
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);

1343 1344 1345 1346 1347
    /* 
     * Initialize ev struct to 0 to avoid valgrind complaining
     * about uninitialized data in timer_create call
     */
    memset(&ev, 0, sizeof(ev));
1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360
    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;
    }

1361
    t->priv = (void *)(long)host_timer;
1362 1363 1364 1365 1366 1367

    return 0;
}

static void dynticks_stop_timer(struct qemu_alarm_timer *t)
{
1368
    timer_t host_timer = (timer_t)(long)t->priv;
1369 1370 1371 1372 1373 1374

    timer_delete(host_timer);
}

static void dynticks_rearm_timer(struct qemu_alarm_timer *t)
{
1375
    timer_t host_timer = (timer_t)(long)t->priv;
1376 1377 1378 1379
    struct itimerspec timeout;
    int64_t nearest_delta_us = INT64_MAX;
    int64_t current_us;

J
Jan Kiszka 已提交
1380
    if (!active_timers[QEMU_CLOCK_REALTIME] &&
J
Jan Kiszka 已提交
1381 1382
        !active_timers[QEMU_CLOCK_VIRTUAL] &&
        !active_timers[QEMU_CLOCK_HOST])
1383
        return;
1384

P
pbrook 已提交
1385
    nearest_delta_us = qemu_next_deadline_dyntick();
1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407

    /* 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
1408
#endif /* defined(__linux__) */
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 1442 1443
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
1444
#endif /* !defined(_WIN32) */
1445

1446

1447 1448 1449 1450 1451 1452
#ifdef _WIN32

static int win32_start_timer(struct qemu_alarm_timer *t)
{
    TIMECAPS tc;
    struct qemu_alarm_win32 *data = t->priv;
1453
    UINT flags;
1454 1455 1456 1457 1458 1459 1460 1461 1462

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

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

    timeBeginPeriod(data->period);

1463 1464 1465 1466 1467 1468
    flags = TIME_CALLBACK_FUNCTION;
    if (alarm_has_dynticks(t))
        flags |= TIME_ONESHOT;
    else
        flags |= TIME_PERIODIC;

1469 1470 1471 1472
    data->timerId = timeSetEvent(1,         // interval (ms)
                        data->period,       // resolution
                        host_alarm_handler, // function
                        (DWORD)t,           // parameter
1473
                        flags);
1474 1475

    if (!data->timerId) {
B
Blue Swirl 已提交
1476
        fprintf(stderr, "Failed to initialize win32 alarm timer: %ld\n",
1477
                GetLastError());
1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492
        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);
}

1493 1494 1495 1496
static void win32_rearm_timer(struct qemu_alarm_timer *t)
{
    struct qemu_alarm_win32 *data = t->priv;

J
Jan Kiszka 已提交
1497
    if (!active_timers[QEMU_CLOCK_REALTIME] &&
J
Jan Kiszka 已提交
1498 1499
        !active_timers[QEMU_CLOCK_VIRTUAL] &&
        !active_timers[QEMU_CLOCK_HOST])
1500
        return;
1501 1502 1503 1504 1505 1506 1507 1508 1509 1510

    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
1511
        fprintf(stderr, "Failed to re-arm win32 alarm timer %ld\n",
1512
                GetLastError());
1513 1514 1515 1516 1517 1518

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

1519 1520
#endif /* _WIN32 */

1521
static int init_timer_alarm(void)
1522
{
1523
    struct qemu_alarm_timer *t = NULL;
1524
    int i, err = -1;
1525

1526 1527 1528 1529 1530 1531
    for (i = 0; alarm_timers[i].name; i++) {
        t = &alarm_timers[i];

        err = t->start(t);
        if (!err)
            break;
B
bellard 已提交
1532
    }
1533

1534
    if (err) {
1535 1536
        err = -ENOENT;
        goto fail;
B
bellard 已提交
1537
    }
1538 1539

    alarm_timer = t;
1540

1541
    return 0;
1542 1543 1544

fail:
    return err;
1545 1546
}

1547
static void quit_timers(void)
B
bellard 已提交
1548
{
1549 1550
    alarm_timer->stop(alarm_timer);
    alarm_timer = NULL;
B
bellard 已提交
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 1588 1589
/***********************************************************/
/* 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
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 1641 1642
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
1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653
    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
1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668
#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
1669 1670 1671 1672 1673
#ifdef _WIN32
static void socket_cleanup(void)
{
    WSACleanup();
}
B
bellard 已提交
1674

B
bellard 已提交
1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688
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;
}
1689 1690
#endif

1691 1692 1693 1694 1695
/***********************************************************/
/* Bluetooth support */
static int nb_hcis;
static int cur_hci;
static struct HCIInfo *hci_table[MAX_NICS];
1696

1697 1698 1699 1700 1701 1702 1703
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
1704
static struct bt_scatternet_s *qemu_find_bt_vlan(int id)
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 1742 1743
{
    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++];
}

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 1775 1776
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
1777
    bdaddr_t bdaddr;
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 1880 1881

    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;
}

1882 1883 1884
/***********************************************************/
/* QEMU Block devices */

1885
#define HD_ALIAS "index=%d,media=disk"
T
ths 已提交
1886 1887
#define CDROM_ALIAS "index=2,media=cdrom"
#define FD_ALIAS "index=%d,if=floppy"
1888 1889
#define PFLASH_ALIAS "if=pflash"
#define MTD_ALIAS "if=mtd"
1890
#define SD_ALIAS "index=0,if=sd"
T
ths 已提交
1891

G
Gerd Hoffmann 已提交
1892
QemuOpts *drive_add(const char *file, const char *fmt, ...)
T
ths 已提交
1893 1894
{
    va_list ap;
G
Gerd Hoffmann 已提交
1895 1896
    char optstr[1024];
    QemuOpts *opts;
T
ths 已提交
1897 1898

    va_start(ap, fmt);
G
Gerd Hoffmann 已提交
1899
    vsnprintf(optstr, sizeof(optstr), fmt, ap);
T
ths 已提交
1900 1901
    va_end(ap);

G
Gerd Hoffmann 已提交
1902
    opts = qemu_opts_parse(&qemu_drive_opts, optstr, NULL);
G
Gerd Hoffmann 已提交
1903 1904 1905 1906 1907 1908 1909 1910
    if (!opts) {
        fprintf(stderr, "%s: huh? duplicate? (%s)\n",
                __FUNCTION__, optstr);
        return NULL;
    }
    if (file)
        qemu_opt_set(opts, "file", file);
    return opts;
1911 1912
}

G
Gerd Hoffmann 已提交
1913
DriveInfo *drive_get(BlockInterfaceType type, int bus, int unit)
T
ths 已提交
1914
{
G
Gerd Hoffmann 已提交
1915
    DriveInfo *dinfo;
T
ths 已提交
1916 1917 1918

    /* seek interface, bus and unit */

B
Blue Swirl 已提交
1919
    QTAILQ_FOREACH(dinfo, &drives, next) {
G
Gerd Hoffmann 已提交
1920 1921 1922 1923 1924
        if (dinfo->type == type &&
	    dinfo->bus == bus &&
	    dinfo->unit == unit)
            return dinfo;
    }
T
ths 已提交
1925

G
Gerd Hoffmann 已提交
1926
    return NULL;
T
ths 已提交
1927 1928
}

G
Gerd Hoffmann 已提交
1929
DriveInfo *drive_get_by_id(const char *id)
G
Gerd Hoffmann 已提交
1930 1931 1932
{
    DriveInfo *dinfo;

B
Blue Swirl 已提交
1933
    QTAILQ_FOREACH(dinfo, &drives, next) {
G
Gerd Hoffmann 已提交
1934 1935 1936 1937 1938 1939 1940
        if (strcmp(id, dinfo->id))
            continue;
        return dinfo;
    }
    return NULL;
}

1941
int drive_get_max_bus(BlockInterfaceType type)
T
ths 已提交
1942 1943
{
    int max_bus;
G
Gerd Hoffmann 已提交
1944
    DriveInfo *dinfo;
T
ths 已提交
1945 1946

    max_bus = -1;
B
Blue Swirl 已提交
1947
    QTAILQ_FOREACH(dinfo, &drives, next) {
G
Gerd Hoffmann 已提交
1948 1949 1950
        if(dinfo->type == type &&
           dinfo->bus > max_bus)
            max_bus = dinfo->bus;
T
ths 已提交
1951 1952 1953 1954
    }
    return max_bus;
}

1955 1956
const char *drive_get_serial(BlockDriverState *bdrv)
{
G
Gerd Hoffmann 已提交
1957
    DriveInfo *dinfo;
1958

B
Blue Swirl 已提交
1959
    QTAILQ_FOREACH(dinfo, &drives, next) {
G
Gerd Hoffmann 已提交
1960 1961 1962
        if (dinfo->bdrv == bdrv)
            return dinfo->serial;
    }
1963 1964 1965 1966

    return "\0";
}

1967 1968
BlockInterfaceErrorAction drive_get_onerror(BlockDriverState *bdrv)
{
G
Gerd Hoffmann 已提交
1969
    DriveInfo *dinfo;
1970

B
Blue Swirl 已提交
1971
    QTAILQ_FOREACH(dinfo, &drives, next) {
G
Gerd Hoffmann 已提交
1972 1973 1974
        if (dinfo->bdrv == bdrv)
            return dinfo->onerror;
    }
1975

1976
    return BLOCK_ERR_STOP_ENOSPC;
1977 1978
}

1979 1980 1981 1982 1983
static void bdrv_format_print(void *opaque, const char *name)
{
    fprintf(stderr, " %s", name);
}

G
Gerd Hoffmann 已提交
1984
void drive_uninit(DriveInfo *dinfo)
1985
{
G
Gerd Hoffmann 已提交
1986 1987 1988 1989
    qemu_opts_del(dinfo->opts);
    bdrv_delete(dinfo->bdrv);
    QTAILQ_REMOVE(&drives, dinfo, next);
    qemu_free(dinfo);
1990 1991
}

G
Gerd Hoffmann 已提交
1992
DriveInfo *drive_init(QemuOpts *opts, void *opaque,
G
Gerd Hoffmann 已提交
1993
                      int *fatal_error)
T
ths 已提交
1994
{
G
Gerd Hoffmann 已提交
1995 1996
    const char *buf;
    const char *file = NULL;
1997
    char devname[128];
G
Gerd Hoffmann 已提交
1998
    const char *serial;
1999
    const char *mediastr = "";
2000
    BlockInterfaceType type;
T
ths 已提交
2001 2002 2003
    enum { MEDIA_DISK, MEDIA_CDROM } media;
    int bus_id, unit_id;
    int cyls, heads, secs, translation;
2004
    BlockDriver *drv = NULL;
2005
    QEMUMachine *machine = opaque;
T
ths 已提交
2006 2007
    int max_devs;
    int index;
2008
    int cache;
2009
    int aio = 0;
2010
    int bdrv_flags, onerror;
2011
    const char *devaddr;
G
Gerd Hoffmann 已提交
2012
    DriveInfo *dinfo;
G
Gerd Hoffmann 已提交
2013
    int snapshot = 0;
T
ths 已提交
2014

G
Gerd Hoffmann 已提交
2015
    *fatal_error = 1;
T
ths 已提交
2016 2017

    translation = BIOS_ATA_TRANSLATION_AUTO;
2018
    cache = 1;
T
ths 已提交
2019

2020
    if (machine && machine->use_scsi) {
2021
        type = IF_SCSI;
T
ths 已提交
2022
        max_devs = MAX_SCSI_DEVS;
B
blueswir1 已提交
2023
        pstrcpy(devname, sizeof(devname), "scsi");
T
ths 已提交
2024
    } else {
2025
        type = IF_IDE;
T
ths 已提交
2026
        max_devs = MAX_IDE_DEVS;
B
blueswir1 已提交
2027
        pstrcpy(devname, sizeof(devname), "ide");
T
ths 已提交
2028 2029 2030 2031
    }
    media = MEDIA_DISK;

    /* extract parameters */
G
Gerd Hoffmann 已提交
2032 2033 2034
    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
2035

G
Gerd Hoffmann 已提交
2036 2037 2038
    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
2039

G
Gerd Hoffmann 已提交
2040
    snapshot = qemu_opt_get_bool(opts, "snapshot", 0);
T
ths 已提交
2041

G
Gerd Hoffmann 已提交
2042 2043 2044 2045
    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
2046
        pstrcpy(devname, sizeof(devname), buf);
T
ths 已提交
2047
        if (!strcmp(buf, "ide")) {
2048
	    type = IF_IDE;
T
ths 已提交
2049 2050
            max_devs = MAX_IDE_DEVS;
        } else if (!strcmp(buf, "scsi")) {
2051
	    type = IF_SCSI;
T
ths 已提交
2052 2053
            max_devs = MAX_SCSI_DEVS;
        } else if (!strcmp(buf, "floppy")) {
2054
	    type = IF_FLOPPY;
T
ths 已提交
2055 2056
            max_devs = 0;
        } else if (!strcmp(buf, "pflash")) {
2057
	    type = IF_PFLASH;
T
ths 已提交
2058 2059
            max_devs = 0;
	} else if (!strcmp(buf, "mtd")) {
2060
	    type = IF_MTD;
T
ths 已提交
2061 2062
            max_devs = 0;
	} else if (!strcmp(buf, "sd")) {
2063
	    type = IF_SD;
T
ths 已提交
2064
            max_devs = 0;
A
aliguori 已提交
2065 2066 2067
        } else if (!strcmp(buf, "virtio")) {
            type = IF_VIRTIO;
            max_devs = 0;
2068 2069 2070
	} else if (!strcmp(buf, "xen")) {
	    type = IF_XEN;
            max_devs = 0;
G
Gerd Hoffmann 已提交
2071 2072 2073
	} else if (!strcmp(buf, "none")) {
	    type = IF_NONE;
            max_devs = 0;
2074
	} else {
G
Gerd Hoffmann 已提交
2075
            fprintf(stderr, "qemu: unsupported bus type '%s'\n", buf);
G
Gerd Hoffmann 已提交
2076
            return NULL;
T
ths 已提交
2077 2078 2079 2080 2081
	}
    }

    if (cyls || heads || secs) {
        if (cyls < 1 || cyls > 16383) {
G
Gerd Hoffmann 已提交
2082
            fprintf(stderr, "qemu: '%s' invalid physical cyls number\n", buf);
G
Gerd Hoffmann 已提交
2083
	    return NULL;
T
ths 已提交
2084 2085
	}
        if (heads < 1 || heads > 16) {
G
Gerd Hoffmann 已提交
2086
            fprintf(stderr, "qemu: '%s' invalid physical heads number\n", buf);
G
Gerd Hoffmann 已提交
2087
	    return NULL;
T
ths 已提交
2088 2089
	}
        if (secs < 1 || secs > 63) {
G
Gerd Hoffmann 已提交
2090
            fprintf(stderr, "qemu: '%s' invalid physical secs number\n", buf);
G
Gerd Hoffmann 已提交
2091
	    return NULL;
T
ths 已提交
2092 2093 2094
	}
    }

G
Gerd Hoffmann 已提交
2095
    if ((buf = qemu_opt_get(opts, "trans")) != NULL) {
T
ths 已提交
2096 2097 2098
        if (!cyls) {
            fprintf(stderr,
                    "qemu: '%s' trans must be used with cyls,heads and secs\n",
G
Gerd Hoffmann 已提交
2099
                    buf);
G
Gerd Hoffmann 已提交
2100
            return NULL;
T
ths 已提交
2101 2102 2103 2104 2105 2106 2107 2108
        }
        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
2109
            fprintf(stderr, "qemu: '%s' invalid translation type\n", buf);
G
Gerd Hoffmann 已提交
2110
	    return NULL;
T
ths 已提交
2111 2112 2113
	}
    }

G
Gerd Hoffmann 已提交
2114
    if ((buf = qemu_opt_get(opts, "media")) != NULL) {
T
ths 已提交
2115 2116 2117 2118 2119
        if (!strcmp(buf, "disk")) {
	    media = MEDIA_DISK;
	} else if (!strcmp(buf, "cdrom")) {
            if (cyls || secs || heads) {
                fprintf(stderr,
G
Gerd Hoffmann 已提交
2120
                        "qemu: '%s' invalid physical CHS format\n", buf);
G
Gerd Hoffmann 已提交
2121
	        return NULL;
T
ths 已提交
2122 2123 2124
            }
	    media = MEDIA_CDROM;
	} else {
G
Gerd Hoffmann 已提交
2125
	    fprintf(stderr, "qemu: '%s' invalid media\n", buf);
G
Gerd Hoffmann 已提交
2126
	    return NULL;
T
ths 已提交
2127 2128 2129
	}
    }

G
Gerd Hoffmann 已提交
2130
    if ((buf = qemu_opt_get(opts, "cache")) != NULL) {
2131
        if (!strcmp(buf, "off") || !strcmp(buf, "none"))
2132
            cache = 0;
2133
        else if (!strcmp(buf, "writethrough"))
2134
            cache = 1;
2135 2136
        else if (!strcmp(buf, "writeback"))
            cache = 2;
2137 2138
        else {
           fprintf(stderr, "qemu: invalid cache option\n");
G
Gerd Hoffmann 已提交
2139
           return NULL;
2140 2141 2142
        }
    }

2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155
#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
2156
    if ((buf = qemu_opt_get(opts, "format")) != NULL) {
2157 2158 2159 2160
       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
2161
	    return NULL;
2162
        }
2163 2164 2165
        drv = bdrv_find_format(buf);
        if (!drv) {
            fprintf(stderr, "qemu: '%s' invalid format\n", buf);
G
Gerd Hoffmann 已提交
2166
            return NULL;
2167 2168 2169
        }
    }

2170
    onerror = BLOCK_ERR_STOP_ENOSPC;
G
Gerd Hoffmann 已提交
2171
    if ((buf = qemu_opt_get(opts, "werror")) != NULL) {
2172
        if (type != IF_IDE && type != IF_SCSI && type != IF_VIRTIO) {
2173
            fprintf(stderr, "werror is no supported by this format\n");
G
Gerd Hoffmann 已提交
2174
            return NULL;
2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185
        }
        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
2186
            return NULL;
2187 2188 2189
        }
    }

G
Gerd Hoffmann 已提交
2190
    if ((devaddr = qemu_opt_get(opts, "addr")) != NULL) {
2191
        if (type != IF_VIRTIO) {
G
Gerd Hoffmann 已提交
2192
            fprintf(stderr, "addr is not supported\n");
G
Gerd Hoffmann 已提交
2193
            return NULL;
2194 2195 2196
        }
    }

T
ths 已提交
2197 2198 2199 2200 2201
    /* compute bus and unit according index */

    if (index != -1) {
        if (bus_id != 0 || unit_id != -1) {
            fprintf(stderr,
G
Gerd Hoffmann 已提交
2202
                    "qemu: index cannot be used with bus and unit\n");
G
Gerd Hoffmann 已提交
2203
            return NULL;
T
ths 已提交
2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220
        }
        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
2221
       while (drive_get(type, bus_id, unit_id) != NULL) {
T
ths 已提交
2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232
           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
2233 2234
        fprintf(stderr, "qemu: unit %d too big (max is %d)\n",
                unit_id, max_devs - 1);
G
Gerd Hoffmann 已提交
2235
        return NULL;
T
ths 已提交
2236 2237 2238 2239 2240 2241
    }

    /*
     * ignore multiple definitions
     */

G
Gerd Hoffmann 已提交
2242 2243 2244 2245
    if (drive_get(type, bus_id, unit_id) != NULL) {
        *fatal_error = 0;
        return NULL;
    }
T
ths 已提交
2246 2247 2248

    /* init */

G
Gerd Hoffmann 已提交
2249
    dinfo = qemu_mallocz(sizeof(*dinfo));
2250
    if ((buf = qemu_opts_id(opts)) != NULL) {
G
Gerd Hoffmann 已提交
2251 2252
        dinfo->id = qemu_strdup(buf);
    } else {
G
Gerd Hoffmann 已提交
2253
        /* no id supplied -> create one */
G
Gerd Hoffmann 已提交
2254
        dinfo->id = qemu_mallocz(32);
G
Gerd Hoffmann 已提交
2255 2256 2257
        if (type == IF_IDE || type == IF_SCSI)
            mediastr = (media == MEDIA_CDROM) ? "-cd" : "-hd";
        if (max_devs)
G
Gerd Hoffmann 已提交
2258
            snprintf(dinfo->id, 32, "%s%i%s%i",
G
Gerd Hoffmann 已提交
2259 2260
                     devname, bus_id, mediastr, unit_id);
        else
G
Gerd Hoffmann 已提交
2261
            snprintf(dinfo->id, 32, "%s%s%i",
G
Gerd Hoffmann 已提交
2262 2263 2264
                     devname, mediastr, unit_id);
    }
    dinfo->bdrv = bdrv_new(dinfo->id);
G
Gerd Hoffmann 已提交
2265 2266 2267 2268 2269
    dinfo->devaddr = devaddr;
    dinfo->type = type;
    dinfo->bus = bus_id;
    dinfo->unit = unit_id;
    dinfo->onerror = onerror;
G
Gerd Hoffmann 已提交
2270 2271 2272
    dinfo->opts = opts;
    if (serial)
        strncpy(dinfo->serial, serial, sizeof(serial));
B
Blue Swirl 已提交
2273
    QTAILQ_INSERT_TAIL(&drives, dinfo, next);
T
ths 已提交
2274

2275
    switch(type) {
T
ths 已提交
2276 2277
    case IF_IDE:
    case IF_SCSI:
2278
    case IF_XEN:
G
Gerd Hoffmann 已提交
2279
    case IF_NONE:
T
ths 已提交
2280 2281 2282
        switch(media) {
	case MEDIA_DISK:
            if (cyls != 0) {
G
Gerd Hoffmann 已提交
2283 2284
                bdrv_set_geometry_hint(dinfo->bdrv, cyls, heads, secs);
                bdrv_set_translation_hint(dinfo->bdrv, translation);
T
ths 已提交
2285 2286 2287
            }
	    break;
	case MEDIA_CDROM:
G
Gerd Hoffmann 已提交
2288
            bdrv_set_type_hint(dinfo->bdrv, BDRV_TYPE_CDROM);
T
ths 已提交
2289 2290 2291 2292 2293 2294 2295
	    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
2296
        bdrv_set_type_hint(dinfo->bdrv, BDRV_TYPE_FLOPPY);
T
ths 已提交
2297 2298 2299 2300
        break;
    case IF_PFLASH:
    case IF_MTD:
        break;
G
Gerd Hoffmann 已提交
2301 2302 2303 2304 2305 2306 2307 2308
    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
2309 2310
    case IF_COUNT:
        abort();
T
ths 已提交
2311
    }
G
Gerd Hoffmann 已提交
2312
    if (!file) {
G
Gerd Hoffmann 已提交
2313 2314 2315
        *fatal_error = 0;
        return NULL;
    }
2316
    bdrv_flags = 0;
2317
    if (snapshot) {
2318
        bdrv_flags |= BDRV_O_SNAPSHOT;
2319 2320 2321 2322 2323 2324
        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;
2325 2326 2327 2328 2329 2330 2331

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

G
Gerd Hoffmann 已提交
2332
    if (bdrv_open2(dinfo->bdrv, file, bdrv_flags, drv) < 0) {
2333 2334
        fprintf(stderr, "qemu: could not open disk image %s: %s\n",
                        file, strerror(errno));
G
Gerd Hoffmann 已提交
2335
        return NULL;
T
ths 已提交
2336
    }
2337

G
Gerd Hoffmann 已提交
2338
    if (bdrv_key_required(dinfo->bdrv))
2339
        autostart = 0;
G
Gerd Hoffmann 已提交
2340 2341
    *fatal_error = 0;
    return dinfo;
T
ths 已提交
2342 2343
}

G
Gerd Hoffmann 已提交
2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363
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
2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377
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
2378 2379 2380 2381 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
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
2407 2408 2409 2410 2411 2412 2413 2414 2415 2416
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);
}

2417 2418 2419 2420 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
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;
}

2473 2474 2475 2476 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
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
2523 2524 2525
/***********************************************************/
/* USB devices */

2526 2527 2528 2529 2530
static void usb_msd_password_cb(void *opaque, int err)
{
    USBDevice *dev = opaque;

    if (!err)
2531
        usb_device_attach(dev);
2532
    else
2533
        dev->info->handle_destroy(dev);
2534 2535
}

2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554
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",
    }
};

2555
static int usb_device_add(const char *devname, int is_hotplug)
B
bellard 已提交
2556 2557
{
    const char *p;
2558 2559 2560
    USBBus *bus = usb_bus_find(-1 /* any */);
    USBDevice *dev = NULL;
    int i;
B
bellard 已提交
2561

2562
    if (!usb_enabled)
B
bellard 已提交
2563 2564
        return -1;

2565 2566 2567 2568 2569 2570 2571 2572 2573
    /* 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
2574 2575
    if (strstart(devname, "host:", &p)) {
        dev = usb_host_device_open(p);
P
pbrook 已提交
2576
    } else if (strstart(devname, "disk:", &p)) {
2577 2578
        BlockDriverState *bs;

2579
        dev = usb_msd_init(p);
2580 2581
        if (!dev)
            return -1;
2582
        bs = usb_msd_get_bdrv(dev);
2583 2584
        if (bdrv_key_required(bs)) {
            autostart = 0;
2585
            if (is_hotplug) {
A
aliguori 已提交
2586 2587
                monitor_read_bdrv_key_start(cur_mon, bs, usb_msd_password_cb,
                                            dev);
2588
                return 0;
2589 2590
            }
        }
B
balrog 已提交
2591 2592
    } else if (strstart(devname, "serial:", &p)) {
        dev = usb_serial_init(p);
A
aurel32 已提交
2593 2594 2595 2596
#ifdef CONFIG_BRLAPI
    } else if (!strcmp(devname, "braille")) {
        dev = usb_baum_init();
#endif
2597
    } else if (strstart(devname, "net:", &p)) {
M
Mark McLoughlin 已提交
2598 2599
        QemuOpts *opts;
        int idx;
2600

M
Mark McLoughlin 已提交
2601 2602
        opts = qemu_opts_parse(&qemu_net_opts, p, NULL);
        if (!opts) {
2603
            return -1;
M
Mark McLoughlin 已提交
2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614
        }

        qemu_opt_set(opts, "type", "nic");
        qemu_opt_set(opts, "model", "usb");

        idx = net_client_init_from_opts(NULL, opts);
        if (idx == -1) {
            return -1;
        }

        dev = usb_net_init(&nd_table[idx]);
2615 2616 2617
    } 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
2618 2619 2620
    } else {
        return -1;
    }
P
pbrook 已提交
2621 2622 2623
    if (!dev)
        return -1;

2624
done:
B
bellard 已提交
2625 2626 2627
    return 0;
}

2628 2629 2630 2631 2632
static int usb_device_del(const char *devname)
{
    int bus_num, addr;
    const char *p;

2633 2634 2635
    if (strstart(devname, "host:", &p))
        return usb_host_device_close(p);

2636
    if (!usb_enabled)
2637 2638 2639 2640 2641 2642 2643 2644
        return -1;

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

2645
    return usb_device_delete_addr(bus_num, addr);
2646 2647
}

2648 2649 2650 2651 2652
static int usb_parse(const char *cmdline)
{
    return usb_device_add(cmdline, 0);
}

2653
void do_usb_add(Monitor *mon, const QDict *qdict)
B
bellard 已提交
2654
{
2655
    usb_device_add(qdict_get_str(qdict, "devname"), 1);
B
bellard 已提交
2656 2657
}

2658
void do_usb_del(Monitor *mon, const QDict *qdict)
B
bellard 已提交
2659
{
2660
    usb_device_del(qdict_get_str(qdict, "devname"));
B
bellard 已提交
2661 2662
}

2663 2664 2665 2666
/***********************************************************/
/* PCMCIA/Cardbus */

static struct pcmcia_socket_entry_s {
P
Paul Brook 已提交
2667
    PCMCIASocket *socket;
2668 2669 2670
    struct pcmcia_socket_entry_s *next;
} *pcmcia_sockets = 0;

P
Paul Brook 已提交
2671
void pcmcia_socket_register(PCMCIASocket *socket)
2672 2673 2674 2675 2676 2677 2678 2679 2680
{
    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
2681
void pcmcia_socket_unregister(PCMCIASocket *socket)
2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692
{
    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
2693
void pcmcia_info(Monitor *mon)
2694 2695
{
    struct pcmcia_socket_entry_s *iter;
A
aliguori 已提交
2696

2697
    if (!pcmcia_sockets)
A
aliguori 已提交
2698
        monitor_printf(mon, "No PCMCIA sockets\n");
2699 2700

    for (iter = pcmcia_sockets; iter; iter = iter->next)
A
aliguori 已提交
2701 2702 2703
        monitor_printf(mon, "%s: %s\n", iter->socket->slot_string,
                       iter->socket->attached ? iter->socket->card_string :
                       "Empty");
2704 2705
}

2706
/***********************************************************/
2707 2708
/* register display */

2709 2710 2711 2712 2713 2714
struct DisplayAllocator default_allocator = {
    defaultallocator_create_displaysurface,
    defaultallocator_resize_displaysurface,
    defaultallocator_free_displaysurface
};

2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729
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;
}

2730 2731 2732 2733 2734 2735
DisplayAllocator *register_displayallocator(DisplayState *ds, DisplayAllocator *da)
{
    if(ds->allocator ==  &default_allocator) ds->allocator = da;
    return ds->allocator;
}

2736 2737
/* dumb display */

2738
static void dumb_display_init(void)
2739
{
2740
    DisplayState *ds = qemu_mallocz(sizeof(DisplayState));
2741 2742
    ds->allocator = &default_allocator;
    ds->surface = qemu_create_displaysurface(ds, 640, 480);
2743
    register_displaystate(ds);
2744 2745
}

2746 2747
/***********************************************************/
/* I/O handling */
2748

2749 2750
typedef struct IOHandlerRecord {
    int fd;
B
bellard 已提交
2751 2752 2753
    IOCanRWHandler *fd_read_poll;
    IOHandler *fd_read;
    IOHandler *fd_write;
2754
    int deleted;
2755 2756 2757
    void *opaque;
    /* temporary data */
    struct pollfd *ufd;
2758
    struct IOHandlerRecord *next;
2759 2760
} IOHandlerRecord;

2761
static IOHandlerRecord *first_io_handler;
2762

B
bellard 已提交
2763 2764
/* XXX: fd_read_poll should be suppressed, but an API change is
   necessary in the character devices to suppress fd_can_read(). */
2765 2766 2767 2768
int qemu_set_fd_handler2(int fd,
                         IOCanRWHandler *fd_read_poll,
                         IOHandler *fd_read,
                         IOHandler *fd_write,
B
bellard 已提交
2769
                         void *opaque)
2770
{
B
bellard 已提交
2771
    IOHandlerRecord **pioh, *ioh;
2772

B
bellard 已提交
2773 2774 2775 2776 2777 2778 2779
    if (!fd_read && !fd_write) {
        pioh = &first_io_handler;
        for(;;) {
            ioh = *pioh;
            if (ioh == NULL)
                break;
            if (ioh->fd == fd) {
2780
                ioh->deleted = 1;
B
bellard 已提交
2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798
                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;
2799
        ioh->deleted = 0;
B
bellard 已提交
2800
    }
2801 2802 2803
    return 0;
}

2804 2805 2806
int qemu_set_fd_handler(int fd,
                        IOHandler *fd_read,
                        IOHandler *fd_write,
B
bellard 已提交
2807
                        void *opaque)
2808
{
B
bellard 已提交
2809
    return qemu_set_fd_handler2(fd, NULL, fd_read, fd_write, opaque);
2810 2811
}

A
aliguori 已提交
2812
#ifdef _WIN32
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 2841 2842 2843 2844 2845 2846 2847
/***********************************************************/
/* 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;
        }
    }
}

2848 2849 2850 2851 2852 2853 2854 2855 2856 2857
/***********************************************************/
/* 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};
2858

2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884
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];
2885
        }
2886 2887 2888 2889 2890 2891
    }
    if (found)
        w->num--;
}
#endif

2892 2893 2894
/***********************************************************/
/* ram save/restore */

2895
#define RAM_SAVE_FLAG_FULL	0x01 /* Obsolete, not used anymore */
2896 2897 2898 2899 2900 2901
#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)
2902
{
2903 2904 2905
    uint32_t val = ch << 24 | ch << 16 | ch << 8 | ch;
    uint32_t *array = (uint32_t *)page;
    int i;
2906

2907 2908 2909 2910 2911 2912 2913 2914 2915 2916
    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
2917 2918 2919
    static ram_addr_t current_addr = 0;
    ram_addr_t saved_addr = current_addr;
    ram_addr_t addr = 0;
2920 2921
    int found = 0;

P
pbrook 已提交
2922
    while (addr < last_ram_offset) {
2923
        if (cpu_physical_memory_get_dirty(current_addr, MIGRATION_DIRTY_FLAG)) {
P
pbrook 已提交
2924
            uint8_t *p;
2925 2926 2927 2928 2929

            cpu_physical_memory_reset_dirty(current_addr,
                                            current_addr + TARGET_PAGE_SIZE,
                                            MIGRATION_DIRTY_FLAG);

P
pbrook 已提交
2930
            p = qemu_get_ram_ptr(current_addr);
2931

P
pbrook 已提交
2932
            if (is_dup_page(p, *p)) {
2933
                qemu_put_be64(f, current_addr | RAM_SAVE_FLAG_COMPRESS);
P
pbrook 已提交
2934
                qemu_put_byte(f, *p);
2935 2936
            } else {
                qemu_put_be64(f, current_addr | RAM_SAVE_FLAG_PAGE);
P
pbrook 已提交
2937
                qemu_put_buffer(f, p, TARGET_PAGE_SIZE);
2938
            }
2939 2940 2941

            found = 1;
            break;
2942
        }
2943
        addr += TARGET_PAGE_SIZE;
P
pbrook 已提交
2944
        current_addr = (saved_addr + addr) % last_ram_offset;
2945
    }
2946 2947

    return found;
2948 2949
}

2950
static uint64_t bytes_transferred = 0;
2951

A
Anthony Liguori 已提交
2952
static ram_addr_t ram_save_remaining(void)
2953
{
A
Anthony Liguori 已提交
2954 2955
    ram_addr_t addr;
    ram_addr_t count = 0;
2956

P
pbrook 已提交
2957
    for (addr = 0; addr < last_ram_offset; addr += TARGET_PAGE_SIZE) {
2958 2959 2960 2961 2962 2963 2964
        if (cpu_physical_memory_get_dirty(addr, MIGRATION_DIRTY_FLAG))
            count++;
    }

    return count;
}

2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979
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;
}

2980 2981
static int ram_save_live(QEMUFile *f, int stage, void *opaque)
{
A
Anthony Liguori 已提交
2982
    ram_addr_t addr;
2983 2984 2985
    uint64_t bytes_transferred_last;
    double bwidth = 0;
    uint64_t expected_time = 0;
2986

2987
    if (cpu_physical_sync_dirty_bitmap(0, TARGET_PHYS_ADDR_MAX) != 0) {
2988 2989 2990 2991
        qemu_file_set_error(f);
        return 0;
    }

2992 2993
    if (stage == 1) {
        /* Make sure all dirty bits are set */
P
pbrook 已提交
2994
        for (addr = 0; addr < last_ram_offset; addr += TARGET_PAGE_SIZE) {
2995 2996 2997
            if (!cpu_physical_memory_get_dirty(addr, MIGRATION_DIRTY_FLAG))
                cpu_physical_memory_set_dirty(addr);
        }
2998

2999 3000 3001
        /* Enable dirty memory tracking */
        cpu_physical_memory_set_dirty_tracking(1);

P
pbrook 已提交
3002
        qemu_put_be64(f, last_ram_offset | RAM_SAVE_FLAG_MEM_SIZE);
3003 3004
    }

3005 3006 3007
    bytes_transferred_last = bytes_transferred;
    bwidth = get_clock();

3008 3009 3010 3011
    while (!qemu_file_rate_limit(f)) {
        int ret;

        ret = ram_save_block(f);
3012
        bytes_transferred += ret * TARGET_PAGE_SIZE;
3013 3014 3015 3016
        if (ret == 0) /* no more blocks */
            break;
    }

3017 3018 3019 3020 3021 3022 3023 3024
    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;

3025 3026 3027 3028 3029
    /* try transferring iterative blocks of memory */

    if (stage == 3) {

        /* flush all remaining blocks regardless of rate limiting */
3030 3031 3032
        while (ram_save_block(f) != 0) {
            bytes_transferred += TARGET_PAGE_SIZE;
        }
3033
        cpu_physical_memory_set_dirty_tracking(0);
3034 3035 3036 3037
    }

    qemu_put_be64(f, RAM_SAVE_FLAG_EOS);

3038 3039 3040
    expected_time = ram_save_remaining() * TARGET_PAGE_SIZE / bwidth;

    return (stage == 2) && (expected_time <= migrate_max_downtime());
3041 3042 3043 3044
}

static int ram_load(QEMUFile *f, void *opaque, int version_id)
{
A
Anthony Liguori 已提交
3045
    ram_addr_t addr;
3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057
    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
3058
            if (addr != last_ram_offset)
3059 3060 3061 3062 3063
                return -EINVAL;
        }

        if (flags & RAM_SAVE_FLAG_COMPRESS) {
            uint8_t ch = qemu_get_byte(f);
3064 3065
            memset(qemu_get_ram_ptr(addr), ch, TARGET_PAGE_SIZE);
#ifndef _WIN32
3066 3067 3068
            if (ch == 0 &&
                (!kvm_enabled() || kvm_has_sync_mmu())) {
                madvise(qemu_get_ram_ptr(addr), TARGET_PAGE_SIZE, MADV_DONTNEED);
3069
            }
3070
#endif
3071
        } else if (flags & RAM_SAVE_FLAG_PAGE)
P
pbrook 已提交
3072
            qemu_get_buffer(f, qemu_get_ram_ptr(addr), TARGET_PAGE_SIZE);
3073 3074
    } while (!(flags & RAM_SAVE_FLAG_EOS));

3075 3076 3077
    return 0;
}

A
aliguori 已提交
3078 3079
void qemu_service_io(void)
{
3080
    qemu_notify_event();
A
aliguori 已提交
3081 3082
}

B
bellard 已提交
3083 3084 3085 3086 3087 3088 3089
/***********************************************************/
/* bottom halves (can be seen as timers which expire ASAP) */

struct QEMUBH {
    QEMUBHFunc *cb;
    void *opaque;
    int scheduled;
A
aliguori 已提交
3090 3091
    int idle;
    int deleted;
B
bellard 已提交
3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102
    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
3103 3104
    bh->next = first_bh;
    first_bh = bh;
B
bellard 已提交
3105 3106 3107
    return bh;
}

B
bellard 已提交
3108
int qemu_bh_poll(void)
B
bellard 已提交
3109
{
A
aliguori 已提交
3110
    QEMUBH *bh, **bhp;
B
bellard 已提交
3111
    int ret;
B
bellard 已提交
3112

B
bellard 已提交
3113
    ret = 0;
A
aliguori 已提交
3114 3115 3116 3117 3118 3119 3120 3121
    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
3122
    }
A
aliguori 已提交
3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134

    /* 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
3135
    return ret;
B
bellard 已提交
3136 3137
}

A
aliguori 已提交
3138 3139 3140 3141 3142 3143 3144 3145
void qemu_bh_schedule_idle(QEMUBH *bh)
{
    if (bh->scheduled)
        return;
    bh->scheduled = 1;
    bh->idle = 1;
}

B
bellard 已提交
3146 3147 3148 3149 3150
void qemu_bh_schedule(QEMUBH *bh)
{
    if (bh->scheduled)
        return;
    bh->scheduled = 1;
A
aliguori 已提交
3151
    bh->idle = 0;
B
bellard 已提交
3152
    /* stop the currently executing CPU to execute the BH ASAP */
3153
    qemu_notify_event();
B
bellard 已提交
3154 3155 3156 3157
}

void qemu_bh_cancel(QEMUBH *bh)
{
A
aliguori 已提交
3158
    bh->scheduled = 0;
B
bellard 已提交
3159 3160 3161 3162
}

void qemu_bh_delete(QEMUBH *bh)
{
A
aliguori 已提交
3163 3164
    bh->scheduled = 0;
    bh->deleted = 1;
B
bellard 已提交
3165 3166
}

A
aliguori 已提交
3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186
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;
            }
        }
    }
}

3187 3188 3189
/***********************************************************/
/* machine registration */

B
blueswir1 已提交
3190
static QEMUMachine *first_machine = NULL;
3191
QEMUMachine *current_machine = NULL;
3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203

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

3204
static QEMUMachine *find_machine(const char *name)
3205 3206 3207 3208 3209 3210
{
    QEMUMachine *m;

    for(m = first_machine; m != NULL; m = m->next) {
        if (!strcmp(m->name, name))
            return m;
M
Mark McLoughlin 已提交
3211 3212
        if (m->alias && !strcmp(m->alias, name))
            return m;
3213 3214 3215 3216
    }
    return NULL;
}

3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228
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;
}

3229 3230 3231
/***********************************************************/
/* main execution loop */

3232
static void gui_update(void *opaque)
3233
{
3234
    uint64_t interval = GUI_REFRESH_INTERVAL;
3235
    DisplayState *ds = opaque;
3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246
    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));
3247 3248
}

B
blueswir1 已提交
3249 3250 3251 3252 3253 3254 3255
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
3256 3257 3258
struct vm_change_state_entry {
    VMChangeStateHandler *cb;
    void *opaque;
B
Blue Swirl 已提交
3259
    QLIST_ENTRY (vm_change_state_entry) entries;
B
bellard 已提交
3260 3261
};

B
Blue Swirl 已提交
3262
static QLIST_HEAD(vm_change_state_head, vm_change_state_entry) vm_change_state_head;
B
bellard 已提交
3263 3264 3265 3266 3267 3268 3269 3270 3271 3272

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
3273
    QLIST_INSERT_HEAD(&vm_change_state_head, e, entries);
B
bellard 已提交
3274 3275 3276 3277 3278
    return e;
}

void qemu_del_vm_change_state_handler(VMChangeStateEntry *e)
{
B
Blue Swirl 已提交
3279
    QLIST_REMOVE (e, entries);
B
bellard 已提交
3280 3281 3282
    qemu_free (e);
}

3283
static void vm_state_notify(int running, int reason)
B
bellard 已提交
3284 3285 3286 3287
{
    VMChangeStateEntry *e;

    for (e = vm_change_state_head.lh_first; e; e = e->entries.le_next) {
3288
        e->cb(e->opaque, running, reason);
B
bellard 已提交
3289 3290 3291
    }
}

3292 3293 3294
static void resume_all_vcpus(void);
static void pause_all_vcpus(void);

3295 3296 3297 3298 3299
void vm_start(void)
{
    if (!vm_running) {
        cpu_enable_ticks();
        vm_running = 1;
3300
        vm_state_notify(1, 0);
3301
        qemu_rearm_alarm_timer(alarm_timer);
3302
        resume_all_vcpus();
3303 3304 3305
    }
}

3306 3307 3308
/* reset/shutdown handler */

typedef struct QEMUResetEntry {
B
Blue Swirl 已提交
3309
    QTAILQ_ENTRY(QEMUResetEntry) entry;
3310 3311 3312 3313
    QEMUResetHandler *func;
    void *opaque;
} QEMUResetEntry;

B
Blue Swirl 已提交
3314 3315
static QTAILQ_HEAD(reset_handlers, QEMUResetEntry) reset_handlers =
    QTAILQ_HEAD_INITIALIZER(reset_handlers);
3316 3317
static int reset_requested;
static int shutdown_requested;
B
bellard 已提交
3318
static int powerdown_requested;
3319
static int debug_requested;
3320
static int vmstop_requested;
3321

A
aurel32 已提交
3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342
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;
}

3343 3344 3345 3346 3347 3348 3349
static int qemu_debug_requested(void)
{
    int r = debug_requested;
    debug_requested = 0;
    return r;
}

3350 3351 3352 3353 3354 3355 3356 3357 3358 3359 3360 3361
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;
3362
        pause_all_vcpus();
3363 3364 3365 3366
        vm_state_notify(0, reason);
    }
}

3367
void qemu_register_reset(QEMUResetHandler *func, void *opaque)
3368
{
J
Jan Kiszka 已提交
3369
    QEMUResetEntry *re = qemu_mallocz(sizeof(QEMUResetEntry));
3370 3371 3372

    re->func = func;
    re->opaque = opaque;
B
Blue Swirl 已提交
3373
    QTAILQ_INSERT_TAIL(&reset_handlers, re, entry);
3374 3375
}

J
Jan Kiszka 已提交
3376
void qemu_unregister_reset(QEMUResetHandler *func, void *opaque)
3377 3378 3379
{
    QEMUResetEntry *re;

B
Blue Swirl 已提交
3380
    QTAILQ_FOREACH(re, &reset_handlers, entry) {
J
Jan Kiszka 已提交
3381
        if (re->func == func && re->opaque == opaque) {
B
Blue Swirl 已提交
3382
            QTAILQ_REMOVE(&reset_handlers, re, entry);
J
Jan Kiszka 已提交
3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393
            qemu_free(re);
            return;
        }
    }
}

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

    /* reset all devices */
B
Blue Swirl 已提交
3394
    QTAILQ_FOREACH_SAFE(re, &reset_handlers, entry, nre) {
3395 3396 3397 3398 3399 3400
        re->func(re->opaque);
    }
}

void qemu_system_reset_request(void)
{
B
bellard 已提交
3401 3402 3403 3404 3405
    if (no_reboot) {
        shutdown_requested = 1;
    } else {
        reset_requested = 1;
    }
3406
    qemu_notify_event();
3407 3408 3409 3410 3411
}

void qemu_system_shutdown_request(void)
{
    shutdown_requested = 1;
3412
    qemu_notify_event();
3413 3414
}

B
bellard 已提交
3415 3416 3417
void qemu_system_powerdown_request(void)
{
    powerdown_requested = 1;
3418 3419 3420
    qemu_notify_event();
}

3421 3422
#ifdef CONFIG_IOTHREAD
static void qemu_system_vmstop_request(int reason)
3423
{
3424 3425
    vmstop_requested = reason;
    qemu_notify_event();
3426
}
3427
#endif
3428

3429 3430 3431 3432
#ifndef _WIN32
static int io_thread_fd = -1;

static void qemu_event_increment(void)
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 3468 3469 3470 3471 3472 3473 3474
    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
3475 3476
    return 0;

3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492
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
3493
        fprintf(stderr, "Failed CreateEvent: %ld\n", GetLastError());
3494 3495 3496
        return -1;
    }
    qemu_add_wait_object(qemu_event_handle, dummy_event_handler, NULL);
3497 3498 3499
    return 0;
}

3500 3501
static void qemu_event_increment(void)
{
M
malc 已提交
3502
    if (!SetEvent(qemu_event_handle)) {
B
Blue Swirl 已提交
3503
        fprintf(stderr, "qemu_event_increment: SetEvent failed: %ld\n",
M
malc 已提交
3504 3505 3506
                GetLastError());
        exit (1);
    }
3507 3508 3509
}
#endif

3510 3511 3512 3513 3514 3515 3516 3517 3518 3519
static int cpu_can_run(CPUState *env)
{
    if (env->stop)
        return 0;
    if (env->stopped)
        return 0;
    return 1;
}

#ifndef CONFIG_IOTHREAD
3520 3521 3522 3523 3524
static int qemu_init_main_loop(void)
{
    return qemu_event_init();
}

3525 3526 3527 3528 3529 3530
void qemu_init_vcpu(void *_env)
{
    CPUState *env = _env;

    if (kvm_enabled())
        kvm_init_vcpu(env);
3531 3532
    env->nr_cores = smp_cores;
    env->nr_threads = smp_threads;
3533 3534 3535
    return;
}

3536 3537 3538 3539 3540
int qemu_cpu_self(void *env)
{
    return 1;
}

3541 3542 3543 3544 3545 3546 3547 3548
static void resume_all_vcpus(void)
{
}

static void pause_all_vcpus(void)
{
}

3549 3550 3551 3552 3553
void qemu_cpu_kick(void *env)
{
    return;
}

3554 3555 3556 3557 3558 3559
void qemu_notify_event(void)
{
    CPUState *env = cpu_single_env;

    if (env) {
        cpu_exit(env);
3560
    }
3561 3562
}

3563 3564 3565
#define qemu_mutex_lock_iothread() do { } while (0)
#define qemu_mutex_unlock_iothread() do { } while (0)

3566 3567 3568 3569 3570
void vm_stop(int reason)
{
    do_vm_stop(reason);
}

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 3637 3638 3639 3640 3641 3642 3643
#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);
3644 3645
    if (kvm_enabled())
        kvm_init_vcpu(env);
3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656 3657 3658

    /* 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);
3659
        qemu_wait_io_event(env);
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 3693 3694 3695 3696 3697 3698 3699
    }

    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
3700
int qemu_cpu_self(void *_env)
3701
{
G
Glauber Costa 已提交
3702 3703 3704 3705 3706 3707
    CPUState *env = _env;
    QemuThread this;
 
    qemu_thread_self(&this);
 
    return qemu_thread_equal(&this, env->thread);
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 3855 3856 3857 3858 3859 3860 3861
}

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);
3862 3863
    env->nr_cores = smp_cores;
    env->nr_threads = smp_threads;
3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893
}

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
3894
#ifdef _WIN32
3895
static void host_main_loop_wait(int *timeout)
A
aliguori 已提交
3896 3897
{
    int ret, ret2, i;
3898 3899
    PollingEntry *pe;

3900

3901 3902 3903 3904 3905
    /* 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);
    }
3906
    if (ret == 0) {
3907 3908
        int err;
        WaitObjects *w = &wait_objects;
3909

A
aliguori 已提交
3910
        ret = WaitForMultipleObjects(w->num, w->events, FALSE, *timeout);
3911 3912 3913
        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]);
3914

3915
            /* Check for additional signaled events */
3916
            for(i = (ret - WAIT_OBJECT_0 + 1); i < w->num; i++) {
3917

3918 3919 3920 3921 3922 3923 3924 3925 3926
                /* 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);
3927 3928
                }
            }
3929 3930 3931
        } else if (ret == WAIT_TIMEOUT) {
        } else {
            err = GetLastError();
3932
            fprintf(stderr, "WaitForMultipleObjects error %d %d\n", ret, err);
3933
        }
3934
    }
A
aliguori 已提交
3935 3936 3937 3938

    *timeout = 0;
}
#else
3939
static void host_main_loop_wait(int *timeout)
A
aliguori 已提交
3940 3941
{
}
B
bellard 已提交
3942
#endif
A
aliguori 已提交
3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954

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
3955 3956
    /* poll any events */
    /* XXX: separate device handlers from system ones */
3957
    nfds = -1;
B
bellard 已提交
3958 3959
    FD_ZERO(&rfds);
    FD_ZERO(&wfds);
3960
    FD_ZERO(&xfds);
B
bellard 已提交
3961
    for(ioh = first_io_handler; ioh != NULL; ioh = ioh->next) {
3962 3963
        if (ioh->deleted)
            continue;
B
bellard 已提交
3964 3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976
        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;
        }
    }
3977

A
aliguori 已提交
3978 3979 3980
    tv.tv_sec = timeout / 1000;
    tv.tv_usec = (timeout % 1000) * 1000;

J
Jan Kiszka 已提交
3981 3982
    slirp_select_fill(&nfds, &rfds, &wfds, &xfds);

3983
    qemu_mutex_unlock_iothread();
3984
    ret = select(nfds + 1, &rfds, &wfds, &xfds, &tv);
3985
    qemu_mutex_lock_iothread();
B
bellard 已提交
3986
    if (ret > 0) {
3987 3988 3989
        IOHandlerRecord **pioh;

        for(ioh = first_io_handler; ioh != NULL; ioh = ioh->next) {
3990
            if (!ioh->deleted && ioh->fd_read && FD_ISSET(ioh->fd, &rfds)) {
B
bellard 已提交
3991
                ioh->fd_read(ioh->opaque);
B
bellard 已提交
3992
            }
3993
            if (!ioh->deleted && ioh->fd_write && FD_ISSET(ioh->fd, &wfds)) {
B
bellard 已提交
3994
                ioh->fd_write(ioh->opaque);
3995
            }
B
bellard 已提交
3996
        }
3997 3998 3999 4000 4001 4002 4003 4004

	/* remove deleted IO handlers */
	pioh = &first_io_handler;
	while (*pioh) {
            ioh = *pioh;
            if (ioh->deleted) {
                *pioh = ioh->next;
                qemu_free(ioh);
4005
            } else
4006 4007
                pioh = &ioh->next;
        }
B
bellard 已提交
4008
    }
J
Jan Kiszka 已提交
4009 4010

    slirp_select_poll(&rfds, &wfds, &xfds, (ret < 0));
B
bellard 已提交
4011

4012 4013 4014 4015 4016 4017
    /* rearm timer, if not periodic */
    if (alarm_timer->flags & ALARM_FLAG_EXPIRED) {
        alarm_timer->flags &= ~ALARM_FLAG_EXPIRED;
        qemu_rearm_alarm_timer(alarm_timer);
    }

4018
    /* vm time timers */
4019 4020
    if (vm_running) {
        if (!cur_cpu || likely(!(cur_cpu->singlestep_enabled & SSTEP_NOTIMER)))
J
Jan Kiszka 已提交
4021 4022
            qemu_run_timers(&active_timers[QEMU_CLOCK_VIRTUAL],
                            qemu_get_clock(vm_clock));
4023
    }
4024 4025

    /* real time timers */
J
Jan Kiszka 已提交
4026
    qemu_run_timers(&active_timers[QEMU_CLOCK_REALTIME],
4027 4028
                    qemu_get_clock(rt_clock));

J
Jan Kiszka 已提交
4029 4030 4031
    qemu_run_timers(&active_timers[QEMU_CLOCK_HOST],
                    qemu_get_clock(host_clock));

P
pbrook 已提交
4032 4033 4034
    /* Check bottom-halves last in case any of the earlier events triggered
       them.  */
    qemu_bh_poll();
4035

B
bellard 已提交
4036 4037
}

4038
static int qemu_cpu_exec(CPUState *env)
B
bellard 已提交
4039
{
4040
    int ret;
4041 4042 4043
#ifdef CONFIG_PROFILER
    int64_t ti;
#endif
B
bellard 已提交
4044

4045
#ifdef CONFIG_PROFILER
4046
    ti = profile_getclock();
4047
#endif
4048 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063
    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);
4064
#ifdef CONFIG_PROFILER
4065
    qemu_time += profile_getclock() - ti;
4066
#endif
4067 4068 4069 4070 4071 4072 4073 4074 4075 4076 4077
    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;
}

4078 4079
static void tcg_cpu_exec(void)
{
4080
    int ret = 0;
4081 4082 4083 4084 4085 4086 4087 4088 4089 4090 4091 4092

    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;
        }
4093 4094
        if (cpu_can_run(env))
            ret = qemu_cpu_exec(env);
4095 4096 4097 4098 4099 4100 4101 4102
        if (ret == EXCP_DEBUG) {
            gdb_set_stop_cpu(env);
            debug_requested = 1;
            break;
        }
    }
}

4103 4104
static int cpu_has_work(CPUState *env)
{
4105 4106 4107 4108
    if (env->stop)
        return 1;
    if (env->stopped)
        return 0;
4109 4110 4111 4112 4113 4114 4115 4116 4117 4118 4119 4120 4121 4122 4123 4124 4125 4126 4127
    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)
{
4128
#ifndef CONFIG_IOTHREAD
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 4167 4168 4169 4170 4171 4172 4173
    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;
4174 4175 4176
#else /* CONFIG_IOTHREAD */
    return 1000;
#endif
4177 4178 4179 4180 4181 4182 4183 4184 4185 4186
}

static int vm_can_run(void)
{
    if (powerdown_requested)
        return 0;
    if (reset_requested)
        return 0;
    if (shutdown_requested)
        return 0;
4187 4188
    if (debug_requested)
        return 0;
4189 4190 4191
    return 1;
}

B
Blue Swirl 已提交
4192 4193
qemu_irq qemu_system_powerdown;

4194 4195
static void main_loop(void)
{
4196
    int r;
4197

4198 4199 4200 4201 4202
#ifdef CONFIG_IOTHREAD
    qemu_system_ready = 1;
    qemu_cond_broadcast(&qemu_system_cond);
#endif

4203
    for (;;) {
4204
        do {
4205 4206 4207
#ifdef CONFIG_PROFILER
            int64_t ti;
#endif
4208
#ifndef CONFIG_IOTHREAD
4209
            tcg_cpu_exec();
4210
#endif
4211
#ifdef CONFIG_PROFILER
4212
            ti = profile_getclock();
4213
#endif
4214
            main_loop_wait(qemu_calculate_timeout());
4215
#ifdef CONFIG_PROFILER
4216
            dev_time += profile_getclock() - ti;
4217
#endif
4218
        } while (vm_can_run());
4219

4220 4221
        if (qemu_debug_requested())
            vm_stop(EXCP_DEBUG);
4222 4223 4224 4225 4226 4227 4228
        if (qemu_shutdown_requested()) {
            if (no_shutdown) {
                vm_stop(0);
                no_shutdown = 0;
            } else
                break;
        }
4229 4230
        if (qemu_reset_requested()) {
            pause_all_vcpus();
4231
            qemu_system_reset();
4232 4233
            resume_all_vcpus();
        }
B
Blue Swirl 已提交
4234 4235 4236
        if (qemu_powerdown_requested()) {
            qemu_irq_raise(qemu_system_powerdown);
        }
4237 4238
        if ((r = qemu_vmstop_requested()))
            vm_stop(r);
B
bellard 已提交
4239
    }
4240
    pause_all_vcpus();
B
bellard 已提交
4241 4242
}

P
pbrook 已提交
4243 4244
static void version(void)
{
P
pbrook 已提交
4245
    printf("QEMU PC emulator version " QEMU_VERSION QEMU_PKGVERSION ", Copyright (c) 2003-2008 Fabrice Bellard\n");
P
pbrook 已提交
4246 4247
}

4248
static void help(int exitcode)
4249
{
P
pbrook 已提交
4250 4251
    version();
    printf("usage: %s [options] [disk_image]\n"
4252
           "\n"
4253
           "'disk_image' is a raw hard image image for IDE hard disk 0\n"
B
bellard 已提交
4254
           "\n"
4255 4256 4257 4258 4259 4260 4261
#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
4262
           "\n"
B
bellard 已提交
4263
           "During emulation, the following keys are useful:\n"
B
bellard 已提交
4264 4265 4266
           "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
4267 4268 4269
           "\n"
           "When using -nographic, press 'ctrl-a h' to get some help.\n"
           ,
4270
           "qemu",
B
bellard 已提交
4271
           DEFAULT_RAM_SIZE,
B
bellard 已提交
4272
#ifndef _WIN32
B
bellard 已提交
4273
           DEFAULT_NETWORK_SCRIPT,
T
ths 已提交
4274
           DEFAULT_NETWORK_DOWN_SCRIPT,
B
bellard 已提交
4275
#endif
4276
           DEFAULT_GDBSTUB_PORT,
B
bellard 已提交
4277
           "/tmp/qemu.log");
4278
    exit(exitcode);
4279 4280
}

4281 4282 4283
#define HAS_ARG 0x0001

enum {
4284 4285 4286 4287 4288 4289 4290
#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
4291 4292 4293 4294 4295 4296 4297 4298
};

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

B
blueswir1 已提交
4299
static const QEMUOption qemu_options[] = {
4300
    { "h", 0, QEMU_OPTION_h },
4301 4302 4303 4304 4305 4306 4307
#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
4308
    { NULL },
B
bellard 已提交
4309 4310
};

4311
#ifdef HAS_AUDIO
4312
struct soundhw soundhw[] = {
4313
#ifdef HAS_AUDIO_CHOICE
A
aurel32 已提交
4314
#if defined(TARGET_I386) || defined(TARGET_MIPS)
B
bellard 已提交
4315 4316 4317 4318 4319 4320 4321 4322
    {
        "pcspk",
        "PC speaker",
        0,
        1,
        { .init_isa = pcspk_audio_init }
    },
#endif
M
malc 已提交
4323 4324

#ifdef CONFIG_SB16
4325 4326 4327 4328 4329 4330 4331
    {
        "sb16",
        "Creative Sound Blaster 16",
        0,
        1,
        { .init_isa = SB16_init }
    },
M
malc 已提交
4332
#endif
4333

M
malc 已提交
4334 4335 4336 4337 4338 4339 4340 4341 4342 4343
#ifdef CONFIG_CS4231A
    {
        "cs4231a",
        "CS4231A",
        0,
        1,
        { .init_isa = cs4231a_init }
    },
#endif

4344
#ifdef CONFIG_ADLIB
4345 4346
    {
        "adlib",
4347
#ifdef HAS_YMF262
4348
        "Yamaha YMF262 (OPL3)",
4349
#else
4350
        "Yamaha YM3812 (OPL2)",
4351
#endif
4352 4353 4354 4355
        0,
        1,
        { .init_isa = Adlib_init }
    },
4356
#endif
4357

4358
#ifdef CONFIG_GUS
4359 4360 4361 4362 4363 4364 4365
    {
        "gus",
        "Gravis Ultrasound GF1",
        0,
        1,
        { .init_isa = GUS_init }
    },
4366
#endif
4367

M
malc 已提交
4368
#ifdef CONFIG_AC97
B
balrog 已提交
4369 4370 4371 4372 4373 4374 4375
    {
        "ac97",
        "Intel 82801AA AC97 Audio",
        0,
        0,
        { .init_pci = ac97_init }
    },
M
malc 已提交
4376
#endif
B
balrog 已提交
4377

M
malc 已提交
4378
#ifdef CONFIG_ES1370
4379 4380 4381 4382 4383 4384 4385
    {
        "es1370",
        "ENSONIQ AudioPCI ES1370",
        0,
        0,
        { .init_pci = es1370_init }
    },
4386
#endif
4387

M
malc 已提交
4388 4389
#endif /* HAS_AUDIO_CHOICE */

4390 4391 4392 4393 4394 4395 4396 4397 4398 4399 4400 4401 4402 4403 4404
    { 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");
4405 4406 4407
        exit (*optarg != '?');
    }
    else {
4408
        size_t l;
4409 4410 4411 4412
        const char *p;
        char *e;
        int bad_card = 0;

4413 4414 4415 4416 4417 4418
        if (!strcmp (optarg, "all")) {
            for (c = soundhw; c->name; ++c) {
                c->enabled = 1;
            }
            return;
        }
4419

4420
        p = optarg;
4421 4422 4423
        while (*p) {
            e = strchr (p, ',');
            l = !e ? strlen (p) : (size_t) (e - p);
4424 4425

            for (c = soundhw; c->name; ++c) {
4426
                if (!strncmp (c->name, p, l) && !c->name[l]) {
4427
                    c->enabled = 1;
4428 4429 4430
                    break;
                }
            }
4431 4432

            if (!c->name) {
4433 4434 4435 4436 4437 4438 4439 4440 4441 4442 4443 4444 4445 4446 4447 4448 4449 4450 4451
                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

4452 4453 4454 4455
static void select_vgahw (const char *p)
{
    const char *opts;

4456
    vga_interface_type = VGA_NONE;
4457
    if (strstart(p, "std", &opts)) {
4458
        vga_interface_type = VGA_STD;
4459
    } else if (strstart(p, "cirrus", &opts)) {
4460
        vga_interface_type = VGA_CIRRUS;
4461
    } else if (strstart(p, "vmware", &opts)) {
4462
        vga_interface_type = VGA_VMWARE;
4463
    } else if (strstart(p, "xenfb", &opts)) {
4464
        vga_interface_type = VGA_XENFB;
4465
    } else if (!strstart(p, "none", &opts)) {
4466 4467 4468 4469
    invalid_vga:
        fprintf(stderr, "Unknown vga type: %s\n", p);
        exit(1);
    }
M
malc 已提交
4470 4471 4472 4473 4474 4475 4476 4477 4478 4479 4480 4481 4482
    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;
    }
4483 4484
}

4485 4486 4487
#ifdef TARGET_I386
static int balloon_parse(const char *arg)
{
4488
    QemuOpts *opts;
4489

4490 4491 4492 4493 4494 4495 4496 4497 4498 4499 4500 4501 4502
    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);
4503
        }
4504 4505
        qemu_opt_set(opts, "driver", "virtio-balloon-pci");
        return 0;
4506
    }
4507 4508

    return -1;
4509 4510 4511
}
#endif

4512 4513 4514 4515 4516 4517 4518 4519
#ifdef _WIN32
static BOOL WINAPI qemu_ctrl_handler(DWORD type)
{
    exit(STATUS_CONTROL_C_EXIT);
    return TRUE;
}
#endif

4520
int qemu_uuid_parse(const char *str, uint8_t *uuid)
4521 4522 4523 4524 4525 4526 4527 4528 4529 4530 4531 4532 4533
{
    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;

4534 4535 4536 4537
#ifdef TARGET_I386
    smbios_add_field(1, offsetof(struct smbios_type_1, uuid), 16, uuid);
#endif

4538 4539 4540
    return 0;
}

B
bellard 已提交
4541
#define MAX_NET_CLIENTS 32
B
bellard 已提交
4542

4543 4544 4545 4546 4547 4548 4549
#ifndef _WIN32

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

4550 4551 4552 4553 4554 4555
static void sigchld_handler(int signal)
{
    waitpid(-1, NULL, WNOHANG);
}

static void sighandler_setup(void)
4556 4557 4558 4559 4560 4561 4562 4563
{
    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);
4564 4565 4566 4567

    act.sa_handler = sigchld_handler;
    act.sa_flags = SA_NOCLDSTOP;
    sigaction(SIGCHLD, &act, NULL);
4568 4569 4570 4571
}

#endif

P
Paul Brook 已提交
4572 4573 4574 4575 4576 4577 4578 4579 4580 4581
#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
4582
        return NULL;
P
Paul Brook 已提交
4583 4584 4585 4586 4587 4588 4589 4590 4591 4592 4593 4594 4595 4596 4597 4598 4599 4600 4601 4602 4603 4604 4605 4606 4607
    }

    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];
4608
    size_t max_len;
P
Paul Brook 已提交
4609 4610 4611 4612 4613 4614 4615 4616 4617 4618 4619 4620 4621 4622 4623 4624 4625 4626 4627 4628 4629 4630 4631

#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) {
4632
        p = realpath(argv0, buf);
P
Paul Brook 已提交
4633 4634 4635 4636 4637 4638 4639
        if (!p) {
            return NULL;
        }
    }
    dir = dirname(p);
    dir = dirname(dir);

4640 4641 4642 4643
    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
4644
    if (access(res, R_OK)) {
4645
        snprintf(res, max_len, "%s%s", dir, BUILD_SUFFIX);
P
Paul Brook 已提交
4646 4647 4648 4649 4650
        if (access(res, R_OK)) {
            qemu_free(res);
            res = NULL;
        }
    }
4651

P
Paul Brook 已提交
4652 4653 4654 4655 4656 4657 4658 4659 4660 4661 4662 4663 4664 4665 4666
    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) {
4667
        return qemu_strdup(name);
P
Paul Brook 已提交
4668 4669 4670 4671 4672 4673 4674 4675 4676 4677 4678 4679 4680
    }
    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);
4681
    snprintf(buf, len, "%s/%s%s", data_dir, subdir, name);
P
Paul Brook 已提交
4682 4683 4684 4685 4686 4687 4688
    if (access(buf, R_OK)) {
        qemu_free(buf);
        return NULL;
    }
    return buf;
}

G
Gerd Hoffmann 已提交
4689 4690 4691 4692 4693 4694 4695 4696 4697 4698
static int device_init_func(QemuOpts *opts, void *opaque)
{
    DeviceState *dev;

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

4699 4700 4701 4702 4703 4704
struct device_config {
    enum {
        DEV_USB,       /* -usbdevice   */
        DEV_BT,        /* -bt          */
    } type;
    const char *cmdline;
B
Blue Swirl 已提交
4705
    QTAILQ_ENTRY(device_config) next;
4706
};
B
Blue Swirl 已提交
4707
QTAILQ_HEAD(, device_config) device_configs = QTAILQ_HEAD_INITIALIZER(device_configs);
4708 4709 4710 4711 4712 4713 4714 4715

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
4716
    QTAILQ_INSERT_TAIL(&device_configs, conf, next);
4717 4718 4719 4720 4721 4722 4723
}

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

B
Blue Swirl 已提交
4724
    QTAILQ_FOREACH(conf, &device_configs, next) {
4725 4726 4727 4728 4729 4730 4731 4732 4733
        if (conf->type != type)
            continue;
        rc = func(conf->cmdline);
        if (0 != rc)
            return rc;
    }
    return 0;
}

M
malc 已提交
4734
int main(int argc, char **argv, char **envp)
4735
{
4736
    const char *gdbstub_dev = NULL;
4737
    uint32_t boot_devices_bitmap = 0;
T
ths 已提交
4738
    int i;
4739
    int snapshot, linux_boot, net_boot;
B
bellard 已提交
4740
    const char *initrd_filename;
4741
    const char *kernel_filename, *kernel_cmdline;
J
Jan Kiszka 已提交
4742
    char boot_devices[33] = "cad"; /* default to HD->floppy->CD-ROM */
4743
    DisplayState *ds;
4744
    DisplayChangeListener *dcl;
B
bellard 已提交
4745
    int cyls, heads, secs, translation;
4746
    const char *net_clients[MAX_NET_CLIENTS];
B
bellard 已提交
4747
    int nb_net_clients;
G
Gerd Hoffmann 已提交
4748
    QemuOpts *hda_opts = NULL, *opts;
4749 4750
    int optind;
    const char *r, *optarg;
4751 4752 4753
    CharDriverState *monitor_hds[MAX_MONITOR_DEVICES];
    const char *monitor_devices[MAX_MONITOR_DEVICES];
    int monitor_device_index;
4754
    const char *serial_devices[MAX_SERIAL_PORTS];
4755
    int serial_device_index;
4756
    const char *parallel_devices[MAX_PARALLEL_PORTS];
4757
    int parallel_device_index;
4758 4759
    const char *virtio_consoles[MAX_VIRTIO_CONSOLES];
    int virtio_console_index;
B
bellard 已提交
4760
    const char *loadvm = NULL;
4761
    QEMUMachine *machine;
4762
    const char *cpu_model;
B
blueswir1 已提交
4763
#ifndef _WIN32
T
ths 已提交
4764
    int fds[2];
B
blueswir1 已提交
4765
#endif
4766
    int tb_size;
4767
    const char *pid_file = NULL;
A
aliguori 已提交
4768
    const char *incoming = NULL;
B
blueswir1 已提交
4769
#ifndef _WIN32
4770 4771
    int fd = 0;
    struct passwd *pwd = NULL;
4772 4773
    const char *chroot_dir = NULL;
    const char *run_as = NULL;
B
blueswir1 已提交
4774
#endif
4775
    CPUState *env;
4776
    int show_vnc_port = 0;
B
bellard 已提交
4777

J
Jan Kiszka 已提交
4778 4779
    init_clocks();

G
Gerd Hoffmann 已提交
4780
    qemu_errors_to_file(stderr);
M
malc 已提交
4781 4782
    qemu_cache_utils_init(envp);

B
Blue Swirl 已提交
4783
    QLIST_INIT (&vm_change_state_head);
4784 4785 4786 4787 4788 4789 4790 4791
#ifndef _WIN32
    {
        struct sigaction act;
        sigfillset(&act.sa_mask);
        act.sa_flags = 0;
        act.sa_handler = SIG_IGN;
        sigaction(SIGPIPE, &act, NULL);
    }
4792 4793
#else
    SetConsoleCtrlHandler(qemu_ctrl_handler, TRUE);
4794 4795 4796 4797 4798 4799 4800 4801 4802 4803 4804 4805 4806 4807 4808 4809 4810 4811
    /* 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
4812
#endif
4813

4814
    module_call_init(MODULE_INIT_MACHINE);
4815
    machine = find_default_machine();
4816
    cpu_model = NULL;
B
bellard 已提交
4817
    initrd_filename = NULL;
4818
    ram_size = 0;
4819
    snapshot = 0;
4820 4821
    kernel_filename = NULL;
    kernel_cmdline = "";
4822
    cyls = heads = secs = 0;
B
bellard 已提交
4823
    translation = BIOS_ATA_TRANSLATION_AUTO;
4824

A
aurel32 已提交
4825
    serial_devices[0] = "vc:80Cx24C";
4826
    for(i = 1; i < MAX_SERIAL_PORTS; i++)
4827
        serial_devices[i] = NULL;
4828
    serial_device_index = 0;
4829

4830
    parallel_devices[0] = "vc:80Cx24C";
4831
    for(i = 1; i < MAX_PARALLEL_PORTS; i++)
4832
        parallel_devices[i] = NULL;
4833
    parallel_device_index = 0;
4834

4835
    for(i = 0; i < MAX_VIRTIO_CONSOLES; i++)
4836 4837 4838
        virtio_consoles[i] = NULL;
    virtio_console_index = 0;

4839 4840 4841 4842 4843 4844
    monitor_devices[0] = "vc:80Cx24C";
    for (i = 1; i < MAX_MONITOR_DEVICES; i++) {
        monitor_devices[i] = NULL;
    }
    monitor_device_index = 0;

4845 4846 4847 4848 4849
    for (i = 0; i < MAX_NODES; i++) {
        node_mem[i] = 0;
        node_cpumask[i] = 0;
    }

B
bellard 已提交
4850
    nb_net_clients = 0;
4851
    nb_numa_nodes = 0;
B
bellard 已提交
4852
    nb_nics = 0;
4853

4854
    tb_size = 0;
4855 4856
    autostart= 1;

4857
    optind = 1;
4858
    for(;;) {
4859
        if (optind >= argc)
4860
            break;
4861 4862
        r = argv[optind];
        if (r[0] != '-') {
G
Gerd Hoffmann 已提交
4863
	    hda_opts = drive_add(argv[optind++], HD_ALIAS, 0);
4864 4865 4866 4867
        } else {
            const QEMUOption *popt;

            optind++;
P
pbrook 已提交
4868 4869 4870
            /* Treat --foo the same as -foo.  */
            if (r[1] == '-')
                r++;
4871 4872 4873
            popt = qemu_options;
            for(;;) {
                if (!popt->name) {
4874
                    fprintf(stderr, "%s: invalid option -- '%s'\n",
4875 4876 4877 4878 4879 4880 4881 4882 4883 4884 4885 4886 4887 4888 4889 4890 4891 4892 4893
                            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) {
4894 4895 4896 4897 4898 4899
            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
4900 4901 4902
                        if (m->alias)
                            printf("%-10s %s (alias of %s)\n",
                                   m->alias, m->desc, m->name);
4903
                        printf("%-10s %s%s\n",
4904
                               m->name, m->desc,
4905
                               m->is_default ? " (default)" : "");
4906
                    }
4907
                    exit(*optarg != '?');
4908 4909
                }
                break;
4910 4911
            case QEMU_OPTION_cpu:
                /* hw initialization will check this */
4912
                if (*optarg == '?') {
J
j_mayer 已提交
4913 4914 4915
/* XXX: implement xxx_cpu_list for targets that still miss it */
#if defined(cpu_list)
                    cpu_list(stdout, &fprintf);
4916
#endif
4917
                    exit(0);
4918 4919 4920 4921
                } else {
                    cpu_model = optarg;
                }
                break;
4922
            case QEMU_OPTION_initrd:
B
bellard 已提交
4923 4924
                initrd_filename = optarg;
                break;
4925
            case QEMU_OPTION_hda:
T
ths 已提交
4926
                if (cyls == 0)
G
Gerd Hoffmann 已提交
4927
                    hda_opts = drive_add(optarg, HD_ALIAS, 0);
T
ths 已提交
4928
                else
G
Gerd Hoffmann 已提交
4929
                    hda_opts = drive_add(optarg, HD_ALIAS
T
ths 已提交
4930
			     ",cyls=%d,heads=%d,secs=%d%s",
4931
                             0, cyls, heads, secs,
T
ths 已提交
4932 4933 4934 4935 4936
                             translation == BIOS_ATA_TRANSLATION_LBA ?
                                 ",trans=lba" :
                             translation == BIOS_ATA_TRANSLATION_NONE ?
                                 ",trans=none" : "");
                 break;
4937
            case QEMU_OPTION_hdb:
4938 4939
            case QEMU_OPTION_hdc:
            case QEMU_OPTION_hdd:
4940
                drive_add(optarg, HD_ALIAS, popt->index - QEMU_OPTION_hda);
B
bellard 已提交
4941
                break;
T
ths 已提交
4942
            case QEMU_OPTION_drive:
4943
                drive_add(NULL, "%s", optarg);
T
ths 已提交
4944
	        break;
G
Gerd Hoffmann 已提交
4945 4946 4947 4948
            case QEMU_OPTION_set:
                if (qemu_set_option(optarg) != 0)
                    exit(1);
	        break;
4949
            case QEMU_OPTION_mtdblock:
4950
                drive_add(optarg, MTD_ALIAS);
4951
                break;
4952
            case QEMU_OPTION_sd:
4953
                drive_add(optarg, SD_ALIAS);
4954
                break;
4955
            case QEMU_OPTION_pflash:
4956
                drive_add(optarg, PFLASH_ALIAS);
4957
                break;
4958
            case QEMU_OPTION_snapshot:
4959 4960
                snapshot = 1;
                break;
4961
            case QEMU_OPTION_hdachs:
4962 4963 4964 4965
                {
                    const char *p;
                    p = optarg;
                    cyls = strtol(p, (char **)&p, 0);
B
bellard 已提交
4966 4967
                    if (cyls < 1 || cyls > 16383)
                        goto chs_fail;
4968 4969 4970 4971
                    if (*p != ',')
                        goto chs_fail;
                    p++;
                    heads = strtol(p, (char **)&p, 0);
B
bellard 已提交
4972 4973
                    if (heads < 1 || heads > 16)
                        goto chs_fail;
4974 4975 4976 4977
                    if (*p != ',')
                        goto chs_fail;
                    p++;
                    secs = strtol(p, (char **)&p, 0);
B
bellard 已提交
4978 4979 4980 4981 4982 4983 4984 4985 4986 4987 4988 4989 4990
                    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') {
4991
                    chs_fail:
B
bellard 已提交
4992 4993
                        fprintf(stderr, "qemu: invalid physical CHS format\n");
                        exit(1);
4994
                    }
G
Gerd Hoffmann 已提交
4995 4996 4997 4998 4999 5000 5001 5002 5003 5004 5005 5006 5007
		    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");
                    }
5008 5009
                }
                break;
5010 5011 5012 5013 5014 5015 5016
            case QEMU_OPTION_numa:
                if (nb_numa_nodes >= MAX_NODES) {
                    fprintf(stderr, "qemu: too many NUMA nodes\n");
                    exit(1);
                }
                numa_add(optarg);
                break;
5017
            case QEMU_OPTION_nographic:
5018
                display_type = DT_NOGRAPHIC;
5019
                break;
B
balrog 已提交
5020 5021
#ifdef CONFIG_CURSES
            case QEMU_OPTION_curses:
5022
                display_type = DT_CURSES;
B
balrog 已提交
5023 5024
                break;
#endif
5025 5026 5027
            case QEMU_OPTION_portrait:
                graphic_rotate = 1;
                break;
5028
            case QEMU_OPTION_kernel:
5029 5030
                kernel_filename = optarg;
                break;
5031
            case QEMU_OPTION_append:
5032
                kernel_cmdline = optarg;
5033
                break;
5034
            case QEMU_OPTION_cdrom:
5035
                drive_add(optarg, CDROM_ALIAS);
5036
                break;
5037
            case QEMU_OPTION_boot:
5038
                {
J
Jan Kiszka 已提交
5039
                    static const char * const params[] = {
5040
                        "order", "once", "menu", NULL
J
Jan Kiszka 已提交
5041 5042
                    };
                    char buf[sizeof(boot_devices)];
J
Jan Kiszka 已提交
5043
                    char *standard_boot_devices;
J
Jan Kiszka 已提交
5044 5045 5046 5047 5048 5049 5050 5051 5052 5053 5054 5055 5056 5057 5058 5059
                    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);
5060
                    }
J
Jan Kiszka 已提交
5061 5062 5063 5064 5065 5066 5067 5068 5069
                    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);
                        }
5070 5071 5072 5073 5074 5075 5076 5077 5078 5079 5080 5081 5082
                        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
5083
                    }
5084 5085
                }
                break;
5086 5087
            case QEMU_OPTION_fda:
            case QEMU_OPTION_fdb:
5088
                drive_add(optarg, FD_ALIAS, popt->index - QEMU_OPTION_fda);
5089
                break;
B
bellard 已提交
5090 5091 5092 5093 5094
#ifdef TARGET_I386
            case QEMU_OPTION_no_fd_bootchk:
                fd_bootchk = 0;
                break;
#endif
B
bellard 已提交
5095 5096 5097
            case QEMU_OPTION_net:
                if (nb_net_clients >= MAX_NET_CLIENTS) {
                    fprintf(stderr, "qemu: too many network clients\n");
5098 5099
                    exit(1);
                }
5100
                net_clients[nb_net_clients] = optarg;
B
bellard 已提交
5101
                nb_net_clients++;
B
bellard 已提交
5102
                break;
B
bellard 已提交
5103 5104
#ifdef CONFIG_SLIRP
            case QEMU_OPTION_tftp:
5105
                legacy_tftp_prefix = optarg;
B
bellard 已提交
5106
                break;
5107
            case QEMU_OPTION_bootp:
5108
                legacy_bootp_filename = optarg;
5109
                break;
B
bellard 已提交
5110
#ifndef _WIN32
B
bellard 已提交
5111
            case QEMU_OPTION_smb:
M
Markus Armbruster 已提交
5112 5113
                if (net_slirp_smb(optarg) < 0)
                    exit(1);
B
bellard 已提交
5114
                break;
B
bellard 已提交
5115
#endif
B
bellard 已提交
5116
            case QEMU_OPTION_redir:
M
Markus Armbruster 已提交
5117 5118
                if (net_slirp_redir(optarg) < 0)
                    exit(1);
B
bellard 已提交
5119
                break;
B
bellard 已提交
5120
#endif
5121
            case QEMU_OPTION_bt:
5122
                add_device_config(DEV_BT, optarg);
5123
                break;
5124 5125 5126 5127 5128 5129 5130 5131 5132
#ifdef HAS_AUDIO
            case QEMU_OPTION_audio_help:
                AUD_help ();
                exit (0);
                break;
            case QEMU_OPTION_soundhw:
                select_soundhw (optarg);
                break;
#endif
5133
            case QEMU_OPTION_h:
5134
                help(0);
5135
                break;
P
pbrook 已提交
5136 5137 5138 5139
            case QEMU_OPTION_version:
                version();
                exit(0);
                break;
5140 5141 5142 5143 5144 5145 5146 5147 5148 5149 5150 5151 5152 5153
            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);
5154 5155
                    exit(1);
                }
5156 5157

                /* On 32-bit hosts, QEMU is limited by virtual address space */
5158
                if (value > (2047 << 20) && HOST_LONG_BITS == 32) {
5159 5160 5161
                    fprintf(stderr, "qemu: at most 2047 MB RAM can be simulated\n");
                    exit(1);
                }
A
Anthony Liguori 已提交
5162
                if (value != (uint64_t)(ram_addr_t)value) {
5163 5164 5165 5166
                    fprintf(stderr, "qemu: ram size too large\n");
                    exit(1);
                }
                ram_size = value;
5167
                break;
5168
            }
5169 5170 5171
            case QEMU_OPTION_d:
                {
                    int mask;
B
blueswir1 已提交
5172
                    const CPULogItem *item;
5173

5174 5175 5176
                    mask = cpu_str_to_log_mask(optarg);
                    if (!mask) {
                        printf("Log items (comma separated):\n");
5177 5178 5179 5180
                    for(item = cpu_log_items; item->mask != 0; item++) {
                        printf("%-10s %s\n", item->name, item->help);
                    }
                    exit(1);
5181 5182
                    }
                    cpu_set_log(mask);
5183
                }
5184 5185
                break;
            case QEMU_OPTION_s:
5186
                gdbstub_dev = "tcp::" DEFAULT_GDBSTUB_PORT;
5187
                break;
5188 5189
            case QEMU_OPTION_gdb:
                gdbstub_dev = optarg;
5190 5191
                break;
            case QEMU_OPTION_L:
P
Paul Brook 已提交
5192
                data_dir = optarg;
5193
                break;
5194 5195 5196
            case QEMU_OPTION_bios:
                bios_name = optarg;
                break;
5197 5198 5199
            case QEMU_OPTION_singlestep:
                singlestep = 1;
                break;
5200
            case QEMU_OPTION_S:
5201
                autostart = 0;
5202
                break;
5203
#ifndef _WIN32
5204 5205 5206
	    case QEMU_OPTION_k:
		keyboard_layout = optarg;
		break;
5207
#endif
B
bellard 已提交
5208 5209 5210
            case QEMU_OPTION_localtime:
                rtc_utc = 0;
                break;
5211 5212
            case QEMU_OPTION_vga:
                select_vgahw (optarg);
5213
                break;
5214
#if defined(TARGET_PPC) || defined(TARGET_SPARC)
5215 5216 5217 5218 5219 5220 5221 5222 5223 5224 5225 5226 5227 5228 5229 5230 5231 5232 5233 5234
            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);
5235
                        if (depth != 8 && depth != 15 && depth != 16 &&
5236 5237 5238 5239 5240 5241 5242
                            depth != 24 && depth != 32)
                            goto graphic_error;
                    } else if (*p == '\0') {
                        depth = graphic_depth;
                    } else {
                        goto graphic_error;
                    }
5243

5244 5245 5246 5247 5248
                    graphic_width = w;
                    graphic_height = h;
                    graphic_depth = depth;
                }
                break;
5249
#endif
T
ths 已提交
5250 5251 5252 5253 5254 5255 5256 5257
            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
5258
            case QEMU_OPTION_monitor:
5259 5260 5261 5262 5263 5264
                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
5265
                break;
5266 5267 5268 5269 5270 5271
            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
5272
                if (qemu_chr_open_opts(opts, NULL) == NULL) {
5273 5274 5275
                    exit(1);
                }
                break;
B
bellard 已提交
5276
            case QEMU_OPTION_serial:
5277 5278 5279 5280
                if (serial_device_index >= MAX_SERIAL_PORTS) {
                    fprintf(stderr, "qemu: too many serial ports\n");
                    exit(1);
                }
5281
                serial_devices[serial_device_index] = optarg;
5282
                serial_device_index++;
B
bellard 已提交
5283
                break;
R
Richard W.M. Jones 已提交
5284
            case QEMU_OPTION_watchdog:
M
Markus Armbruster 已提交
5285 5286 5287 5288 5289 5290
                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
5291 5292 5293 5294 5295 5296 5297
                break;
            case QEMU_OPTION_watchdog_action:
                if (select_watchdog_action(optarg) == -1) {
                    fprintf(stderr, "Unknown -watchdog-action parameter\n");
                    exit(1);
                }
                break;
5298 5299 5300 5301 5302 5303 5304 5305
            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;
5306 5307 5308 5309 5310
            case QEMU_OPTION_parallel:
                if (parallel_device_index >= MAX_PARALLEL_PORTS) {
                    fprintf(stderr, "qemu: too many parallel ports\n");
                    exit(1);
                }
5311
                parallel_devices[parallel_device_index] = optarg;
5312 5313
                parallel_device_index++;
                break;
B
bellard 已提交
5314 5315 5316 5317 5318 5319
	    case QEMU_OPTION_loadvm:
		loadvm = optarg;
		break;
            case QEMU_OPTION_full_screen:
                full_screen = 1;
                break;
T
ths 已提交
5320
#ifdef CONFIG_SDL
5321 5322 5323
            case QEMU_OPTION_no_frame:
                no_frame = 1;
                break;
T
ths 已提交
5324 5325 5326
            case QEMU_OPTION_alt_grab:
                alt_grab = 1;
                break;
5327 5328 5329
            case QEMU_OPTION_ctrl_grab:
                ctrl_grab = 1;
                break;
T
ths 已提交
5330 5331 5332
            case QEMU_OPTION_no_quit:
                no_quit = 1;
                break;
5333
            case QEMU_OPTION_sdl:
5334
                display_type = DT_SDL;
5335
                break;
T
ths 已提交
5336
#endif
B
bellard 已提交
5337
            case QEMU_OPTION_pidfile:
5338
                pid_file = optarg;
B
bellard 已提交
5339
                break;
5340 5341 5342 5343
#ifdef TARGET_I386
            case QEMU_OPTION_win2k_hack:
                win2k_install_hack = 1;
                break;
5344 5345 5346
            case QEMU_OPTION_rtc_td_hack:
                rtc_td_hack = 1;
                break;
5347 5348 5349 5350 5351 5352
            case QEMU_OPTION_acpitable:
                if(acpi_table_add(optarg) < 0) {
                    fprintf(stderr, "Wrong acpi table provided\n");
                    exit(1);
                }
                break;
5353 5354 5355 5356 5357 5358
            case QEMU_OPTION_smbios:
                if(smbios_entry_add(optarg) < 0) {
                    fprintf(stderr, "Wrong smbios provided\n");
                    exit(1);
                }
                break;
5359
#endif
A
aliguori 已提交
5360 5361 5362 5363
#ifdef CONFIG_KVM
            case QEMU_OPTION_enable_kvm:
                kvm_allowed = 1;
                break;
B
bellard 已提交
5364
#endif
B
bellard 已提交
5365 5366 5367
            case QEMU_OPTION_usb:
                usb_enabled = 1;
                break;
B
bellard 已提交
5368 5369
            case QEMU_OPTION_usbdevice:
                usb_enabled = 1;
5370 5371 5372
                add_device_config(DEV_USB, optarg);
                break;
            case QEMU_OPTION_device:
5373
                if (!qemu_opts_parse(&qemu_device_opts, optarg, "driver")) {
G
Gerd Hoffmann 已提交
5374 5375
                    exit(1);
                }
B
bellard 已提交
5376
                break;
B
bellard 已提交
5377
            case QEMU_OPTION_smp:
5378
                smp_parse(optarg);
5379
                if (smp_cpus < 1) {
B
bellard 已提交
5380 5381 5382
                    fprintf(stderr, "Invalid number of CPUs\n");
                    exit(1);
                }
5383 5384 5385 5386 5387 5388 5389 5390 5391
                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
5392
                break;
B
bellard 已提交
5393
	    case QEMU_OPTION_vnc:
5394
                display_type = DT_VNC;
5395
		vnc_display = optarg;
B
bellard 已提交
5396
		break;
5397
#ifdef TARGET_I386
B
bellard 已提交
5398 5399 5400
            case QEMU_OPTION_no_acpi:
                acpi_enabled = 0;
                break;
A
aliguori 已提交
5401 5402 5403
            case QEMU_OPTION_no_hpet:
                no_hpet = 1;
                break;
5404 5405 5406 5407 5408
            case QEMU_OPTION_balloon:
                if (balloon_parse(optarg) < 0) {
                    fprintf(stderr, "Unknown -balloon argument %s\n", optarg);
                    exit(1);
                }
5409
                break;
5410
#endif
B
bellard 已提交
5411 5412 5413
            case QEMU_OPTION_no_reboot:
                no_reboot = 1;
                break;
A
aurel32 已提交
5414 5415 5416
            case QEMU_OPTION_no_shutdown:
                no_shutdown = 1;
                break;
5417 5418 5419
            case QEMU_OPTION_show_cursor:
                cursor_hide = 0;
                break;
5420 5421 5422 5423 5424 5425 5426
            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;
5427
#ifndef _WIN32
T
ths 已提交
5428 5429 5430
	    case QEMU_OPTION_daemonize:
		daemonize = 1;
		break;
5431
#endif
5432 5433 5434 5435 5436 5437 5438 5439
	    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;
5440
#if defined(TARGET_ARM) || defined(TARGET_M68K)
5441 5442 5443
            case QEMU_OPTION_semihosting:
                semihosting_enabled = 1;
                break;
5444
#endif
T
ths 已提交
5445
            case QEMU_OPTION_name:
A
Andi Kleen 已提交
5446 5447 5448 5449 5450 5451 5452 5453 5454 5455 5456 5457 5458
                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
5459
                break;
5460
#if defined(TARGET_SPARC) || defined(TARGET_PPC)
B
blueswir1 已提交
5461 5462 5463 5464 5465 5466 5467 5468
            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;
5469 5470 5471 5472
#endif
#ifdef TARGET_ARM
            case QEMU_OPTION_old_param:
                old_param = 1;
5473
                break;
B
blueswir1 已提交
5474
#endif
5475 5476 5477
            case QEMU_OPTION_clock:
                configure_alarms(optarg);
                break;
B
bellard 已提交
5478
            case QEMU_OPTION_startdate:
J
Jan Kiszka 已提交
5479 5480 5481 5482 5483 5484 5485
                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
5486
                }
J
Jan Kiszka 已提交
5487
                configure_rtc(opts);
B
bellard 已提交
5488
                break;
5489 5490 5491 5492 5493
            case QEMU_OPTION_tb_size:
                tb_size = strtol(optarg, NULL, 0);
                if (tb_size < 0)
                    tb_size = 0;
                break;
P
pbrook 已提交
5494 5495 5496 5497 5498 5499 5500 5501
            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
5502 5503 5504
            case QEMU_OPTION_incoming:
                incoming = optarg;
                break;
5505
#ifndef _WIN32
5506 5507 5508 5509 5510 5511
            case QEMU_OPTION_chroot:
                chroot_dir = optarg;
                break;
            case QEMU_OPTION_runas:
                run_as = optarg;
                break;
5512 5513 5514 5515 5516 5517 5518 5519 5520 5521 5522
#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;
5523
#endif
5524
            }
5525 5526
        }
    }
5527

P
Paul Brook 已提交
5528 5529 5530 5531 5532 5533 5534 5535 5536 5537
    /* 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;
    }

5538 5539 5540 5541 5542 5543 5544
    /*
     * 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
5545
    machine->max_cpus = machine->max_cpus ?: 1; /* Default to UP */
5546 5547 5548 5549 5550 5551 5552
    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);
    }

5553
    if (display_type == DT_NOGRAPHIC) {
5554 5555 5556 5557
       if (serial_device_index == 0)
           serial_devices[0] = "stdio";
       if (parallel_device_index == 0)
           parallel_devices[0] = "null";
5558 5559 5560
       if (strncmp(monitor_devices[0], "vc", 2) == 0) {
           monitor_devices[0] = "stdio";
       }
5561 5562
    }

T
ths 已提交
5563 5564 5565 5566 5567 5568 5569 5570 5571 5572 5573 5574 5575 5576 5577
#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:
5578 5579 5580 5581 5582 5583 5584
            len = read(fds[0], &status, 1);
            if (len == -1 && (errno == EINTR))
                goto again;

            if (len != 1)
                exit(1);
            else if (status == 1) {
5585
                fprintf(stderr, "Could not acquire pidfile: %s\n", strerror(errno));
5586 5587 5588
                exit(1);
            } else
                exit(0);
T
ths 已提交
5589
	} else if (pid < 0)
5590
            exit(1);
T
ths 已提交
5591 5592 5593 5594 5595 5596 5597 5598 5599 5600 5601 5602 5603 5604 5605 5606

	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);
    }

5607
    if (pid_file && qemu_create_pidfile(pid_file) != 0) {
5608 5609 5610 5611
        if (daemonize) {
            uint8_t status = 1;
            write(fds[1], &status, 1);
        } else
5612
            fprintf(stderr, "Could not acquire pid file: %s\n", strerror(errno));
5613 5614
        exit(1);
    }
B
blueswir1 已提交
5615
#endif
5616

M
Marcelo Tosatti 已提交
5617 5618 5619 5620 5621 5622 5623 5624 5625 5626
    if (kvm_enabled()) {
        int ret;

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

5627 5628 5629 5630
    if (qemu_init_main_loop()) {
        fprintf(stderr, "qemu_init_main_loop failed\n");
        exit(1);
    }
5631
    linux_boot = (kernel_filename != NULL);
B
balrog 已提交
5632

5633 5634 5635 5636 5637 5638 5639 5640 5641 5642
    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);
    }

5643 5644
#ifndef _WIN32
    /* Win32 doesn't support line-buffering and requires size >= 2 */
B
bellard 已提交
5645
    setvbuf(stdout, NULL, _IOLBF, 0);
5646
#endif
5647

5648 5649 5650 5651
    if (init_timer_alarm() < 0) {
        fprintf(stderr, "could not initialize alarm timer\n");
        exit(1);
    }
P
pbrook 已提交
5652 5653 5654 5655 5656 5657 5658
    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();
    }
5659

B
bellard 已提交
5660 5661 5662 5663
#ifdef _WIN32
    socket_init();
#endif

B
bellard 已提交
5664 5665 5666
    /* init network clients */
    if (nb_net_clients == 0) {
        /* if no clients, we use a default config */
5667 5668 5669 5670
        net_clients[nb_net_clients++] = "nic";
#ifdef CONFIG_SLIRP
        net_clients[nb_net_clients++] = "user";
#endif
B
bellard 已提交
5671 5672
    }

B
bellard 已提交
5673
    for(i = 0;i < nb_net_clients; i++) {
5674
        if (net_client_parse(net_clients[i]) < 0)
B
bellard 已提交
5675
            exit(1);
B
bellard 已提交
5676
    }
B
bellard 已提交
5677

5678 5679 5680 5681
    net_boot = (boot_devices_bitmap >> ('n' - 'a')) & 0xF;
    net_set_boot_mask(net_boot);

    net_client_check();
5682

5683
    /* init the bluetooth world */
5684 5685
    if (foreach_device_config(DEV_BT, bt_parse))
        exit(1);
5686

5687
    /* init the memory */
P
pbrook 已提交
5688 5689
    if (ram_size == 0)
        ram_size = DEFAULT_RAM_SIZE * 1024 * 1024;
5690

5691 5692 5693
    /* init the dynamic translator */
    cpu_exec_init_all(tb_size * 1024 * 1024);

B
bellard 已提交
5694
    bdrv_init();
5695

T
ths 已提交
5696
    /* we always create the cdrom drive, even if no disk is there */
G
Gerd Hoffmann 已提交
5697
    drive_add(NULL, CDROM_ALIAS);
5698

5699
    /* we always create at least one floppy */
G
Gerd Hoffmann 已提交
5700
    drive_add(NULL, FD_ALIAS, 0);
5701

5702
    /* we always create one sd slot, even if no card is in it */
G
Gerd Hoffmann 已提交
5703
    drive_add(NULL, SD_ALIAS);
5704

T
ths 已提交
5705
    /* open the virtual block devices */
G
Gerd Hoffmann 已提交
5706
    if (snapshot)
G
Gerd Hoffmann 已提交
5707 5708
        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
5709
        exit(1);
5710

J
Juan Quintela 已提交
5711
    vmstate_register(0, &vmstate_timers ,&timers_state);
5712
    register_savevm_live("ram", 0, 3, ram_save_live, NULL, ram_load, NULL);
5713

5714
    /* Maintain compatibility with multiple stdio monitors */
5715
    if (!strcmp(monitor_devices[0],"stdio")) {
5716 5717 5718
        for (i = 0; i < MAX_SERIAL_PORTS; i++) {
            const char *devname = serial_devices[i];
            if (devname && !strcmp(devname,"mon:stdio")) {
5719
                monitor_devices[0] = NULL;
5720 5721
                break;
            } else if (devname && !strcmp(devname,"stdio")) {
5722
                monitor_devices[0] = NULL;
5723 5724 5725 5726 5727 5728
                serial_devices[i] = "mon:stdio";
                break;
            }
        }
    }

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 5761 5762 5763 5764 5765 5766 5767 5768 5769 5770
    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;
            }
        }
    }

5771 5772 5773 5774 5775 5776 5777 5778 5779 5780 5781 5782 5783 5784 5785
    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
5786 5787 5788
        }
    }

5789 5790 5791
    for(i = 0; i < MAX_SERIAL_PORTS; i++) {
        const char *devname = serial_devices[i];
        if (devname && strcmp(devname, "none")) {
5792 5793
            char label[32];
            snprintf(label, sizeof(label), "serial%d", i);
5794
            serial_hds[i] = qemu_chr_open(label, devname, NULL);
5795
            if (!serial_hds[i]) {
5796 5797
                fprintf(stderr, "qemu: could not open serial device '%s': %s\n",
                        devname, strerror(errno));
5798 5799
                exit(1);
            }
5800 5801 5802 5803 5804 5805 5806 5807
        }
    }

    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);
5808
            parallel_hds[i] = qemu_chr_open(label, devname, NULL);
5809
            if (!parallel_hds[i]) {
5810 5811
                fprintf(stderr, "qemu: could not open parallel device '%s': %s\n",
                        devname, strerror(errno));
5812 5813 5814 5815 5816 5817 5818 5819 5820 5821
                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);
5822
            virtcon_hds[i] = qemu_chr_open(label, devname, NULL);
5823
            if (!virtcon_hds[i]) {
5824 5825
                fprintf(stderr, "qemu: could not open virtio console '%s': %s\n",
                        devname, strerror(errno));
5826 5827 5828 5829 5830
                exit(1);
            }
        }
    }

P
Paul Brook 已提交
5831 5832
    module_call_init(MODULE_INIT_DEVICE);

M
Markus Armbruster 已提交
5833 5834 5835 5836 5837 5838
    if (watchdog) {
        i = select_watchdog(watchdog);
        if (i > 0)
            exit (i == 1 ? 1 : 0);
    }

5839 5840 5841
    if (machine->compat_props) {
        qdev_prop_register_compat(machine->compat_props);
    }
P
Paul Brook 已提交
5842
    machine->init(ram_size, boot_devices,
5843 5844
                  kernel_filename, kernel_cmdline, initrd_filename, cpu_model);

5845

J
Juan Quintela 已提交
5846 5847 5848 5849 5850
#ifndef _WIN32
    /* must be after terminal init, SDL library changes signal handlers */
    sighandler_setup();
#endif

5851 5852 5853 5854 5855 5856 5857 5858
    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;
            }
        }
    }

5859 5860
    current_machine = machine;

5861 5862
    /* init USB devices */
    if (usb_enabled) {
M
Markus Armbruster 已提交
5863 5864
        if (foreach_device_config(DEV_USB, usb_parse) < 0)
            exit(1);
5865 5866
    }

5867
    /* init generic devices */
G
Gerd Hoffmann 已提交
5868
    if (qemu_opts_foreach(&qemu_device_opts, device_init_func, NULL, 1) != 0)
5869 5870
        exit(1);

5871 5872
    if (!display_state)
        dumb_display_init();
5873 5874
    /* just use the first displaystate for the moment */
    ds = display_state;
5875 5876 5877 5878 5879 5880 5881 5882 5883 5884 5885 5886 5887 5888 5889

    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
5890
#if defined(CONFIG_CURSES)
5891 5892 5893
    case DT_CURSES:
        curses_display_init(ds, full_screen);
        break;
B
balrog 已提交
5894
#endif
5895
#if defined(CONFIG_SDL)
5896 5897 5898
    case DT_SDL:
        sdl_display_init(ds, full_screen, no_frame);
        break;
5899
#elif defined(CONFIG_COCOA)
5900 5901 5902
    case DT_SDL:
        cocoa_display_init(ds, full_screen);
        break;
5903
#endif
5904 5905 5906 5907
    case DT_VNC:
        vnc_display_init(ds);
        if (vnc_display_open(ds, vnc_display) < 0)
            exit(1);
5908

5909 5910
        if (show_vnc_port) {
            printf("VNC server running on `%s'\n", vnc_display_local_addr(ds));
5911
        }
5912 5913 5914
        break;
    default:
        break;
5915
    }
5916
    dpy_resize(ds);
5917

5918 5919 5920 5921 5922
    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
5923
        }
5924
        dcl = dcl->next;
T
ths 已提交
5925
    }
5926

5927
    if (display_type == DT_NOGRAPHIC || display_type == DT_VNC) {
B
blueswir1 已提交
5928 5929 5930 5931
        nographic_timer = qemu_new_timer(rt_clock, nographic_update, NULL);
        qemu_mod_timer(nographic_timer, qemu_get_clock(rt_clock));
    }

5932
    text_consoles_set_display(display_state);
A
aliguori 已提交
5933
    qemu_chr_initial_reset();
5934

5935 5936 5937 5938 5939 5940 5941
    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
5942

5943
    for(i = 0; i < MAX_SERIAL_PORTS; i++) {
5944
        const char *devname = serial_devices[i];
5945
        if (devname && strcmp(devname, "none")) {
5946
            if (strstart(devname, "vc", 0))
B
bellard 已提交
5947
                qemu_chr_printf(serial_hds[i], "serial%d console\r\n", i);
5948
        }
B
bellard 已提交
5949 5950
    }

5951
    for(i = 0; i < MAX_PARALLEL_PORTS; i++) {
5952
        const char *devname = parallel_devices[i];
5953
        if (devname && strcmp(devname, "none")) {
5954
            if (strstart(devname, "vc", 0))
B
bellard 已提交
5955
                qemu_chr_printf(parallel_hds[i], "parallel%d console\r\n", i);
5956 5957 5958
        }
    }

5959 5960
    for(i = 0; i < MAX_VIRTIO_CONSOLES; i++) {
        const char *devname = virtio_consoles[i];
5961
        if (virtcon_hds[i] && devname) {
5962 5963 5964 5965 5966
            if (strstart(devname, "vc", 0))
                qemu_chr_printf(virtcon_hds[i], "virtio console%d\r\n", i);
        }
    }

5967 5968 5969 5970
    if (gdbstub_dev && gdbserver_start(gdbstub_dev) < 0) {
        fprintf(stderr, "qemu: could not open gdbserver on device '%s'\n",
                gdbstub_dev);
        exit(1);
5971 5972
    }

5973 5974
    qdev_machine_creation_done();

5975 5976
    rom_load_all();

5977 5978 5979 5980 5981
    if (loadvm) {
        if (load_vmstate(cur_mon, loadvm) < 0) {
            autostart = 0;
        }
    }
B
bellard 已提交
5982

G
Glauber Costa 已提交
5983
    if (incoming) {
A
aliguori 已提交
5984
        qemu_start_incoming_migration(incoming);
5985
    } else if (autostart) {
5986
        vm_start();
5987
    }
5988

B
blueswir1 已提交
5989
#ifndef _WIN32
T
ths 已提交
5990 5991 5992 5993 5994 5995 5996 5997 5998 5999 6000 6001
    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
6002
	chdir("/");
6003
	TFR(fd = open("/dev/null", O_RDWR));
T
ths 已提交
6004 6005
	if (fd == -1)
	    exit(1);
6006
    }
T
ths 已提交
6007

6008 6009 6010 6011 6012 6013 6014 6015 6016 6017 6018 6019 6020 6021 6022 6023 6024 6025 6026 6027 6028 6029 6030 6031 6032 6033 6034 6035 6036 6037 6038 6039 6040 6041 6042
    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
6043

6044
        close(fd);
T
ths 已提交
6045
    }
B
blueswir1 已提交
6046
#endif
T
ths 已提交
6047

6048
    main_loop();
B
bellard 已提交
6049
    quit_timers();
6050
    net_cleanup();
T
ths 已提交
6051

6052 6053
    return 0;
}