arch_init.c 29.3 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
/*
 * QEMU System Emulator
 *
 * Copyright (c) 2003-2008 Fabrice Bellard
 *
 * Permission is hereby granted, free of charge, to any person obtaining a copy
 * of this software and associated documentation files (the "Software"), to deal
 * in the Software without restriction, including without limitation the rights
 * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
 * copies of the Software, and to permit persons to whom the Software is
 * furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice shall be included in
 * all copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 * THE SOFTWARE.
 */
#include <stdint.h>
#include <stdarg.h>
26
#include <stdlib.h>
27
#ifndef _WIN32
28
#include <sys/types.h>
29 30 31
#include <sys/mman.h>
#endif
#include "config.h"
32
#include "monitor/monitor.h"
33
#include "sysemu.h"
J
Juan Quintela 已提交
34 35
#include "bitops.h"
#include "bitmap.h"
36 37 38
#include "arch_init.h"
#include "audio/audio.h"
#include "hw/pc.h"
39
#include "hw/pci/pci.h"
40 41
#include "hw/audiodev.h"
#include "kvm.h"
42
#include "migration/migration.h"
43
#include "exec/gdbstub.h"
44
#include "hw/smbios.h"
45
#include "exec/address-spaces.h"
J
Jan Kiszka 已提交
46
#include "hw/pcspk.h"
47
#include "migration/page_cache.h"
48
#include "qemu-config.h"
49
#include "qmp-commands.h"
50
#include "trace.h"
51

O
Orit Wasserman 已提交
52 53 54 55 56 57 58 59
#ifdef DEBUG_ARCH_INIT
#define DPRINTF(fmt, ...) \
    do { fprintf(stdout, "arch_init: " fmt, ## __VA_ARGS__); } while (0)
#else
#define DPRINTF(fmt, ...) \
    do { } while (0)
#endif

60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80
#ifdef TARGET_SPARC
int graphic_width = 1024;
int graphic_height = 768;
int graphic_depth = 8;
#else
int graphic_width = 800;
int graphic_height = 600;
int graphic_depth = 15;
#endif


#if defined(TARGET_ALPHA)
#define QEMU_ARCH QEMU_ARCH_ALPHA
#elif defined(TARGET_ARM)
#define QEMU_ARCH QEMU_ARCH_ARM
#elif defined(TARGET_CRIS)
#define QEMU_ARCH QEMU_ARCH_CRIS
#elif defined(TARGET_I386)
#define QEMU_ARCH QEMU_ARCH_I386
#elif defined(TARGET_M68K)
#define QEMU_ARCH QEMU_ARCH_M68K
M
Michael Walle 已提交
81 82
#elif defined(TARGET_LM32)
#define QEMU_ARCH QEMU_ARCH_LM32
83 84 85 86
#elif defined(TARGET_MICROBLAZE)
#define QEMU_ARCH QEMU_ARCH_MICROBLAZE
#elif defined(TARGET_MIPS)
#define QEMU_ARCH QEMU_ARCH_MIPS
87 88
#elif defined(TARGET_OPENRISC)
#define QEMU_ARCH QEMU_ARCH_OPENRISC
89 90 91 92 93 94 95 96
#elif defined(TARGET_PPC)
#define QEMU_ARCH QEMU_ARCH_PPC
#elif defined(TARGET_S390X)
#define QEMU_ARCH QEMU_ARCH_S390X
#elif defined(TARGET_SH4)
#define QEMU_ARCH QEMU_ARCH_SH4
#elif defined(TARGET_SPARC)
#define QEMU_ARCH QEMU_ARCH_SPARC
M
Max Filippov 已提交
97 98
#elif defined(TARGET_XTENSA)
#define QEMU_ARCH QEMU_ARCH_XTENSA
99 100
#elif defined(TARGET_UNICORE32)
#define QEMU_ARCH QEMU_ARCH_UNICORE32
101 102 103 104 105 106 107
#endif

const uint32_t arch_type = QEMU_ARCH;

/***********************************************************/
/* ram save/restore */

108 109 110 111 112 113
#define RAM_SAVE_FLAG_FULL     0x01 /* Obsolete, not used anymore */
#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
#define RAM_SAVE_FLAG_CONTINUE 0x20
114
#define RAM_SAVE_FLAG_XBZRLE   0x40
115

P
Paolo Bonzini 已提交
116 117 118 119 120
#ifdef __ALTIVEC__
#include <altivec.h>
#define VECTYPE        vector unsigned char
#define SPLAT(p)       vec_splat(vec_ld(0, p), 0)
#define ALL_EQ(v1, v2) vec_all_eq(v1, v2)
121 122 123 124
/* altivec.h may redefine the bool macro as vector type.
 * Reset it to POSIX semantics. */
#undef bool
#define bool _Bool
P
Paolo Bonzini 已提交
125 126 127 128 129 130 131 132 133 134 135
#elif defined __SSE2__
#include <emmintrin.h>
#define VECTYPE        __m128i
#define SPLAT(p)       _mm_set1_epi8(*(p))
#define ALL_EQ(v1, v2) (_mm_movemask_epi8(_mm_cmpeq_epi8(v1, v2)) == 0xFFFF)
#else
#define VECTYPE        unsigned long
#define SPLAT(p)       (*(p) * (~0UL / 255))
#define ALL_EQ(v1, v2) ((v1) == (v2))
#endif

136

137 138
static struct defconfig_file {
    const char *filename;
139 140
    /* Indicates it is an user config file (disabled by -no-user-config) */
    bool userconfig;
141
} default_config_files[] = {
142 143
    { CONFIG_QEMU_CONFDIR "/qemu.conf",                   true },
    { CONFIG_QEMU_CONFDIR "/target-" TARGET_ARCH ".conf", true },
144 145 146 147
    { NULL }, /* end of list */
};


148
int qemu_read_default_config_files(bool userconfig)
149 150
{
    int ret;
151
    struct defconfig_file *f;
152

153
    for (f = default_config_files; f->filename; f++) {
154 155 156
        if (!userconfig && f->userconfig) {
            continue;
        }
157 158 159 160
        ret = qemu_read_config_file(f->filename);
        if (ret < 0 && ret != -ENOENT) {
            return ret;
        }
161
    }
162
    
163 164 165
    return 0;
}

P
Paolo Bonzini 已提交
166
static int is_dup_page(uint8_t *page)
167
{
P
Paolo Bonzini 已提交
168 169
    VECTYPE *p = (VECTYPE *)page;
    VECTYPE val = SPLAT(page);
170 171
    int i;

P
Paolo Bonzini 已提交
172 173
    for (i = 0; i < TARGET_PAGE_SIZE / sizeof(VECTYPE); i++) {
        if (!ALL_EQ(val, p[i])) {
174 175 176 177 178 179 180
            return 0;
        }
    }

    return 1;
}

181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198
/* struct contains XBZRLE cache and a static page
   used by the compression */
static struct {
    /* buffer used for XBZRLE encoding */
    uint8_t *encoded_buf;
    /* buffer for storing page content */
    uint8_t *current_buf;
    /* buffer used for XBZRLE decoding */
    uint8_t *decoded_buf;
    /* Cache for XBZRLE */
    PageCache *cache;
} XBZRLE = {
    .encoded_buf = NULL,
    .current_buf = NULL,
    .decoded_buf = NULL,
    .cache = NULL,
};

199 200 201 202 203 204 205 206 207 208

int64_t xbzrle_cache_resize(int64_t new_size)
{
    if (XBZRLE.cache != NULL) {
        return cache_resize(XBZRLE.cache, new_size / TARGET_PAGE_SIZE) *
            TARGET_PAGE_SIZE;
    }
    return pow2floor(new_size);
}

209 210 211 212 213
/* accounting for migration statistics */
typedef struct AccountingInfo {
    uint64_t dup_pages;
    uint64_t norm_pages;
    uint64_t iterations;
O
Orit Wasserman 已提交
214 215 216 217
    uint64_t xbzrle_bytes;
    uint64_t xbzrle_pages;
    uint64_t xbzrle_cache_miss;
    uint64_t xbzrle_overflows;
218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246
} AccountingInfo;

static AccountingInfo acct_info;

static void acct_clear(void)
{
    memset(&acct_info, 0, sizeof(acct_info));
}

uint64_t dup_mig_bytes_transferred(void)
{
    return acct_info.dup_pages * TARGET_PAGE_SIZE;
}

uint64_t dup_mig_pages_transferred(void)
{
    return acct_info.dup_pages;
}

uint64_t norm_mig_bytes_transferred(void)
{
    return acct_info.norm_pages * TARGET_PAGE_SIZE;
}

uint64_t norm_mig_pages_transferred(void)
{
    return acct_info.norm_pages;
}

O
Orit Wasserman 已提交
247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266
uint64_t xbzrle_mig_bytes_transferred(void)
{
    return acct_info.xbzrle_bytes;
}

uint64_t xbzrle_mig_pages_transferred(void)
{
    return acct_info.xbzrle_pages;
}

uint64_t xbzrle_mig_pages_cache_miss(void)
{
    return acct_info.xbzrle_cache_miss;
}

uint64_t xbzrle_mig_pages_overflow(void)
{
    return acct_info.xbzrle_overflows;
}

O
Orit Wasserman 已提交
267 268 269 270 271 272 273 274 275 276 277 278
static void save_block_hdr(QEMUFile *f, RAMBlock *block, ram_addr_t offset,
        int cont, int flag)
{
        qemu_put_be64(f, offset | cont | flag);
        if (!cont) {
                qemu_put_byte(f, strlen(block->idstr));
                qemu_put_buffer(f, (uint8_t *)block->idstr,
                                strlen(block->idstr));
        }

}

279 280 281 282
#define ENCODING_FLAG_XBZRLE 0x1

static int save_xbzrle_page(QEMUFile *f, uint8_t *current_data,
                            ram_addr_t current_addr, RAMBlock *block,
283
                            ram_addr_t offset, int cont, bool last_stage)
284 285 286 287 288
{
    int encoded_len = 0, bytes_sent = -1;
    uint8_t *prev_cached_page;

    if (!cache_is_cached(XBZRLE.cache, current_addr)) {
289 290 291 292
        if (!last_stage) {
            cache_insert(XBZRLE.cache, current_addr,
                         g_memdup(current_data, TARGET_PAGE_SIZE));
        }
O
Orit Wasserman 已提交
293
        acct_info.xbzrle_cache_miss++;
294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310
        return -1;
    }

    prev_cached_page = get_cached_data(XBZRLE.cache, current_addr);

    /* save current buffer into memory */
    memcpy(XBZRLE.current_buf, current_data, TARGET_PAGE_SIZE);

    /* XBZRLE encoding (if there is no overflow) */
    encoded_len = xbzrle_encode_buffer(prev_cached_page, XBZRLE.current_buf,
                                       TARGET_PAGE_SIZE, XBZRLE.encoded_buf,
                                       TARGET_PAGE_SIZE);
    if (encoded_len == 0) {
        DPRINTF("Skipping unmodified page\n");
        return 0;
    } else if (encoded_len == -1) {
        DPRINTF("Overflow\n");
O
Orit Wasserman 已提交
311
        acct_info.xbzrle_overflows++;
312 313 314 315 316 317
        /* update data in the cache */
        memcpy(prev_cached_page, current_data, TARGET_PAGE_SIZE);
        return -1;
    }

    /* we need to update the data in the cache, in order to get the same data */
318 319 320
    if (!last_stage) {
        memcpy(prev_cached_page, XBZRLE.current_buf, TARGET_PAGE_SIZE);
    }
321 322 323 324 325 326 327

    /* Send XBZRLE based compressed page */
    save_block_hdr(f, block, offset, cont, RAM_SAVE_FLAG_XBZRLE);
    qemu_put_byte(f, ENCODING_FLAG_XBZRLE);
    qemu_put_be16(f, encoded_len);
    qemu_put_buffer(f, XBZRLE.encoded_buf, encoded_len);
    bytes_sent = encoded_len + 1 + 2;
O
Orit Wasserman 已提交
328 329
    acct_info.xbzrle_pages++;
    acct_info.xbzrle_bytes += bytes_sent;
330 331 332 333

    return bytes_sent;
}

334 335
static RAMBlock *last_block;
static ram_addr_t last_offset;
J
Juan Quintela 已提交
336 337
static unsigned long *migration_bitmap;
static uint64_t migration_dirty_pages;
338

339 340 341
static inline bool migration_bitmap_test_and_reset_dirty(MemoryRegion *mr,
                                                         ram_addr_t offset)
{
J
Juan Quintela 已提交
342 343 344 345
    bool ret;
    int nr = (mr->ram_addr + offset) >> TARGET_PAGE_BITS;

    ret = test_and_clear_bit(nr, migration_bitmap);
346 347

    if (ret) {
J
Juan Quintela 已提交
348
        migration_dirty_pages--;
349 350 351 352
    }
    return ret;
}

J
Juan Quintela 已提交
353 354
static inline bool migration_bitmap_set_dirty(MemoryRegion *mr,
                                              ram_addr_t offset)
355
{
J
Juan Quintela 已提交
356 357
    bool ret;
    int nr = (mr->ram_addr + offset) >> TARGET_PAGE_BITS;
358

J
Juan Quintela 已提交
359 360 361 362
    ret = test_and_set_bit(nr, migration_bitmap);

    if (!ret) {
        migration_dirty_pages++;
363
    }
J
Juan Quintela 已提交
364
    return ret;
365 366
}

367 368
static void migration_bitmap_sync(void)
{
J
Juan Quintela 已提交
369 370 371
    RAMBlock *block;
    ram_addr_t addr;
    uint64_t num_dirty_pages_init = migration_dirty_pages;
372 373 374 375 376 377 378 379
    MigrationState *s = migrate_get_current();
    static int64_t start_time;
    static int64_t num_dirty_pages_period;
    int64_t end_time;

    if (!start_time) {
        start_time = qemu_get_clock_ms(rt_clock);
    }
380 381

    trace_migration_bitmap_sync_start();
382
    memory_global_sync_dirty_bitmap(get_system_memory());
J
Juan Quintela 已提交
383 384 385 386 387 388 389 390 391 392 393 394

    QLIST_FOREACH(block, &ram_list.blocks, next) {
        for (addr = 0; addr < block->length; addr += TARGET_PAGE_SIZE) {
            if (memory_region_get_dirty(block->mr, addr, TARGET_PAGE_SIZE,
                                        DIRTY_MEMORY_MIGRATION)) {
                migration_bitmap_set_dirty(block->mr, addr);
            }
        }
        memory_region_reset_dirty(block->mr, 0, block->length,
                                  DIRTY_MEMORY_MIGRATION);
    }
    trace_migration_bitmap_sync_end(migration_dirty_pages
395
                                    - num_dirty_pages_init);
396 397 398 399 400 401 402 403 404 405
    num_dirty_pages_period += migration_dirty_pages - num_dirty_pages_init;
    end_time = qemu_get_clock_ms(rt_clock);

    /* more than 1 second = 1000 millisecons */
    if (end_time > start_time + 1000) {
        s->dirty_pages_rate = num_dirty_pages_period * 1000
            / (end_time - start_time);
        start_time = end_time;
        num_dirty_pages_period = 0;
    }
406 407 408
}


409 410 411 412 413 414 415 416
/*
 * ram_save_block: Writes a page of memory to the stream f
 *
 * Returns:  0: if the page hasn't changed
 *          -1: if there are no more dirty pages
 *           n: the amount of bytes written in other case
 */

417
static int ram_save_block(QEMUFile *f, bool last_stage)
418
{
419 420
    RAMBlock *block = last_block;
    ram_addr_t offset = last_offset;
421
    int bytes_sent = -1;
A
Avi Kivity 已提交
422
    MemoryRegion *mr;
423
    ram_addr_t current_addr;
424

425 426 427 428
    if (!block)
        block = QLIST_FIRST(&ram_list.blocks);

    do {
A
Avi Kivity 已提交
429
        mr = block->mr;
430
        if (migration_bitmap_test_and_reset_dirty(mr, offset)) {
431
            uint8_t *p;
432
            int cont = (block == last_block) ? RAM_SAVE_FLAG_CONTINUE : 0;
433

A
Avi Kivity 已提交
434
            p = memory_region_get_ram_ptr(mr) + offset;
435

P
Paolo Bonzini 已提交
436
            if (is_dup_page(p)) {
437
                acct_info.dup_pages++;
O
Orit Wasserman 已提交
438
                save_block_hdr(f, block, offset, cont, RAM_SAVE_FLAG_COMPRESS);
439
                qemu_put_byte(f, *p);
440
                bytes_sent = 1;
441 442 443
            } else if (migrate_use_xbzrle()) {
                current_addr = block->offset + offset;
                bytes_sent = save_xbzrle_page(f, p, current_addr, block,
444 445 446 447
                                              offset, cont, last_stage);
                if (!last_stage) {
                    p = get_cached_data(XBZRLE.cache, current_addr);
                }
448 449 450 451
            }

            /* either we didn't send yet (we may have had XBZRLE overflow) */
            if (bytes_sent == -1) {
O
Orit Wasserman 已提交
452
                save_block_hdr(f, block, offset, cont, RAM_SAVE_FLAG_PAGE);
453
                qemu_put_buffer(f, p, TARGET_PAGE_SIZE);
454
                bytes_sent = TARGET_PAGE_SIZE;
455
                acct_info.norm_pages++;
456 457
            }

458 459 460 461
            /* if page is unmodified, continue to the next */
            if (bytes_sent != 0) {
                break;
            }
462
        }
463 464 465 466 467 468 469 470

        offset += TARGET_PAGE_SIZE;
        if (offset >= block->length) {
            offset = 0;
            block = QLIST_NEXT(block, next);
            if (!block)
                block = QLIST_FIRST(&ram_list.blocks);
        }
A
Avi Kivity 已提交
471
    } while (block != last_block || offset != last_offset);
472 473 474

    last_block = block;
    last_offset = offset;
475

476
    return bytes_sent;
477 478 479 480 481 482
}

static uint64_t bytes_transferred;

static ram_addr_t ram_save_remaining(void)
{
J
Juan Quintela 已提交
483
    return migration_dirty_pages;
484 485 486 487 488 489 490 491 492 493 494 495 496 497
}

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)
{
498 499 500 501 502 503 504
    RAMBlock *block;
    uint64_t total = 0;

    QLIST_FOREACH(block, &ram_list.blocks, next)
        total += block->length;

    return total;
505 506
}

507 508 509 510
static int block_compar(const void *a, const void *b)
{
    RAMBlock * const *ablock = a;
    RAMBlock * const *bblock = b;
511 512

    return strcmp((*ablock)->idstr, (*bblock)->idstr);
513 514 515 516 517 518 519 520 521 522
}

static void sort_ram_list(void)
{
    RAMBlock *block, *nblock, **blocks;
    int n;
    n = 0;
    QLIST_FOREACH(block, &ram_list.blocks, next) {
        ++n;
    }
523
    blocks = g_malloc(n * sizeof *blocks);
524 525 526 527 528 529 530 531 532
    n = 0;
    QLIST_FOREACH_SAFE(block, &ram_list.blocks, next, nblock) {
        blocks[n++] = block;
        QLIST_REMOVE(block, next);
    }
    qsort(blocks, n, sizeof *blocks, block_compar);
    while (--n >= 0) {
        QLIST_INSERT_HEAD(&ram_list.blocks, blocks[n], next);
    }
533
    g_free(blocks);
534 535
}

O
Orit Wasserman 已提交
536 537 538
static void migration_end(void)
{
    memory_global_dirty_log_stop();
539 540 541 542 543 544 545 546 547

    if (migrate_use_xbzrle()) {
        cache_fini(XBZRLE.cache);
        g_free(XBZRLE.cache);
        g_free(XBZRLE.encoded_buf);
        g_free(XBZRLE.current_buf);
        g_free(XBZRLE.decoded_buf);
        XBZRLE.cache = NULL;
    }
O
Orit Wasserman 已提交
548 549
}

550 551 552 553 554
static void ram_migration_cancel(void *opaque)
{
    migration_end();
}

555 556 557 558 559 560 561 562

static void reset_ram_globals(void)
{
    last_block = NULL;
    last_offset = 0;
    sort_ram_list();
}

563 564
#define MAX_WAIT 50 /* ms, half buffered_file limit */

565
static int ram_save_setup(QEMUFile *f, void *opaque)
566
{
567
    RAMBlock *block;
J
Juan Quintela 已提交
568 569 570
    int64_t ram_pages = last_ram_offset() >> TARGET_PAGE_BITS;

    migration_bitmap = bitmap_new(ram_pages);
571
    bitmap_set(migration_bitmap, 0, ram_pages);
J
Juan Quintela 已提交
572
    migration_dirty_pages = ram_pages;
573

574
    bytes_transferred = 0;
575
    reset_ram_globals();
576

577 578 579 580 581 582 583 584 585 586
    if (migrate_use_xbzrle()) {
        XBZRLE.cache = cache_init(migrate_xbzrle_cache_size() /
                                  TARGET_PAGE_SIZE,
                                  TARGET_PAGE_SIZE);
        if (!XBZRLE.cache) {
            DPRINTF("Error creating cache\n");
            return -1;
        }
        XBZRLE.encoded_buf = g_malloc0(TARGET_PAGE_SIZE);
        XBZRLE.current_buf = g_malloc(TARGET_PAGE_SIZE);
587
        acct_clear();
588 589
    }

590
    memory_global_dirty_log_start();
J
Juan Quintela 已提交
591
    migration_bitmap_sync();
592

593
    qemu_put_be64(f, ram_bytes_total() | RAM_SAVE_FLAG_MEM_SIZE);
594

595 596 597 598
    QLIST_FOREACH(block, &ram_list.blocks, next) {
        qemu_put_byte(f, strlen(block->idstr));
        qemu_put_buffer(f, (uint8_t *)block->idstr, strlen(block->idstr));
        qemu_put_be64(f, block->length);
599 600
    }

601 602 603 604 605
    qemu_put_be64(f, RAM_SAVE_FLAG_EOS);

    return 0;
}

606
static int ram_save_iterate(QEMUFile *f, void *opaque)
607 608 609 610 611
{
    uint64_t bytes_transferred_last;
    double bwidth = 0;
    int ret;
    int i;
612
    uint64_t expected_downtime;
613
    MigrationState *s = migrate_get_current();
614

615 616 617
    bytes_transferred_last = bytes_transferred;
    bwidth = qemu_get_clock_ns(rt_clock);

618
    i = 0;
619
    while ((ret = qemu_file_rate_limit(f)) == 0) {
620
        int bytes_sent;
621

622
        bytes_sent = ram_save_block(f, false);
623 624
        /* no more blocks to sent */
        if (bytes_sent < 0) {
625 626
            break;
        }
627
        bytes_transferred += bytes_sent;
628
        acct_info.iterations++;
629 630 631 632 633 634 635 636
        /* we want to check in the 1st loop, just in case it was the 1st time
           and we had to sync the dirty bitmap.
           qemu_get_clock_ns() is a bit expensive, so we only check each some
           iterations
        */
        if ((i & 63) == 0) {
            uint64_t t1 = (qemu_get_clock_ns(rt_clock) - bwidth) / 1000000;
            if (t1 > MAX_WAIT) {
637
                DPRINTF("big wait: %" PRIu64 " milliseconds, %d iterations\n",
638 639 640 641 642
                        t1, i);
                break;
            }
        }
        i++;
643 644
    }

645 646 647 648
    if (ret < 0) {
        return ret;
    }

649 650 651
    bwidth = qemu_get_clock_ns(rt_clock) - bwidth;
    bwidth = (bytes_transferred - bytes_transferred_last) / bwidth;

652 653 654
    /* if we haven't transferred anything this round, force
     * expected_downtime to a very high value, but without
     * crashing */
655 656 657 658
    if (bwidth == 0) {
        bwidth = 0.000001;
    }

659 660
    qemu_put_be64(f, RAM_SAVE_FLAG_EOS);

661 662 663
    expected_downtime = ram_save_remaining() * TARGET_PAGE_SIZE / bwidth;
    DPRINTF("ram_save_live: expected(%" PRIu64 ") <= max(" PRIu64 ")?\n",
            expected_downtime, migrate_max_downtime());
664

665
    if (expected_downtime <= migrate_max_downtime()) {
666
        migration_bitmap_sync();
667
        expected_downtime = ram_save_remaining() * TARGET_PAGE_SIZE / bwidth;
668
        s->expected_downtime = expected_downtime / 1000000; /* ns -> ms */
J
Juan Quintela 已提交
669

670
        return expected_downtime <= migrate_max_downtime();
J
Juan Quintela 已提交
671 672
    }
    return 0;
673 674 675 676
}

static int ram_save_complete(QEMUFile *f, void *opaque)
{
677
    migration_bitmap_sync();
678

679
    /* try transferring iterative blocks of memory */
O
Orit Wasserman 已提交
680

681
    /* flush all remaining blocks regardless of rate limiting */
682
    while (true) {
683 684
        int bytes_sent;

685
        bytes_sent = ram_save_block(f, true);
686 687 688
        /* no more blocks to sent */
        if (bytes_sent < 0) {
            break;
689
        }
690
        bytes_transferred += bytes_sent;
691
    }
692
    memory_global_dirty_log_stop();
693 694 695

    qemu_put_be64(f, RAM_SAVE_FLAG_EOS);

J
Juan Quintela 已提交
696 697 698
    g_free(migration_bitmap);
    migration_bitmap = NULL;

699
    return 0;
700 701
}

702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742
static int load_xbzrle(QEMUFile *f, ram_addr_t addr, void *host)
{
    int ret, rc = 0;
    unsigned int xh_len;
    int xh_flags;

    if (!XBZRLE.decoded_buf) {
        XBZRLE.decoded_buf = g_malloc(TARGET_PAGE_SIZE);
    }

    /* extract RLE header */
    xh_flags = qemu_get_byte(f);
    xh_len = qemu_get_be16(f);

    if (xh_flags != ENCODING_FLAG_XBZRLE) {
        fprintf(stderr, "Failed to load XBZRLE page - wrong compression!\n");
        return -1;
    }

    if (xh_len > TARGET_PAGE_SIZE) {
        fprintf(stderr, "Failed to load XBZRLE page - len overflow!\n");
        return -1;
    }
    /* load data and decode */
    qemu_get_buffer(f, XBZRLE.decoded_buf, xh_len);

    /* decode RLE */
    ret = xbzrle_decode_buffer(XBZRLE.decoded_buf, xh_len, host,
                               TARGET_PAGE_SIZE);
    if (ret == -1) {
        fprintf(stderr, "Failed to load XBZRLE page - decode error!\n");
        rc = -1;
    } else  if (ret > TARGET_PAGE_SIZE) {
        fprintf(stderr, "Failed to load XBZRLE page - size %d exceeds %d!\n",
                ret, TARGET_PAGE_SIZE);
        abort();
    }

    return rc;
}

743 744 745 746 747 748 749 750 751 752 753 754 755 756
static inline void *host_from_stream_offset(QEMUFile *f,
                                            ram_addr_t offset,
                                            int flags)
{
    static RAMBlock *block = NULL;
    char id[256];
    uint8_t len;

    if (flags & RAM_SAVE_FLAG_CONTINUE) {
        if (!block) {
            fprintf(stderr, "Ack, bad migration stream!\n");
            return NULL;
        }

757
        return memory_region_get_ram_ptr(block->mr) + offset;
758 759 760 761 762 763 764 765
    }

    len = qemu_get_byte(f);
    qemu_get_buffer(f, (uint8_t *)id, len);
    id[len] = 0;

    QLIST_FOREACH(block, &ram_list.blocks, next) {
        if (!strncmp(id, block->idstr, sizeof(id)))
766
            return memory_region_get_ram_ptr(block->mr) + offset;
767 768 769 770 771 772
    }

    fprintf(stderr, "Can't find block %s!\n", id);
    return NULL;
}

773
static int ram_load(QEMUFile *f, void *opaque, int version_id)
774 775
{
    ram_addr_t addr;
O
Orit Wasserman 已提交
776
    int flags, ret = 0;
777
    int error;
O
Orit Wasserman 已提交
778 779 780
    static uint64_t seq_iter;

    seq_iter++;
781

782
    if (version_id < 4 || version_id > 4) {
783 784 785 786 787 788 789 790 791 792
        return -EINVAL;
    }

    do {
        addr = qemu_get_be64(f);

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

        if (flags & RAM_SAVE_FLAG_MEM_SIZE) {
793
            if (version_id == 4) {
794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809
                /* Synchronize RAM block list */
                char id[256];
                ram_addr_t length;
                ram_addr_t total_ram_bytes = addr;

                while (total_ram_bytes) {
                    RAMBlock *block;
                    uint8_t len;

                    len = qemu_get_byte(f);
                    qemu_get_buffer(f, (uint8_t *)id, len);
                    id[len] = 0;
                    length = qemu_get_be64(f);

                    QLIST_FOREACH(block, &ram_list.blocks, next) {
                        if (!strncmp(id, block->idstr, sizeof(id))) {
O
Orit Wasserman 已提交
810 811 812 813
                            if (block->length != length) {
                                ret =  -EINVAL;
                                goto done;
                            }
814 815 816 817 818
                            break;
                        }
                    }

                    if (!block) {
819 820
                        fprintf(stderr, "Unknown ramblock \"%s\", cannot "
                                "accept migration\n", id);
O
Orit Wasserman 已提交
821 822
                        ret = -EINVAL;
                        goto done;
823 824 825 826
                    }

                    total_ram_bytes -= length;
                }
827 828 829 830
            }
        }

        if (flags & RAM_SAVE_FLAG_COMPRESS) {
831 832 833
            void *host;
            uint8_t ch;

834
            host = host_from_stream_offset(f, addr, flags);
835 836 837
            if (!host) {
                return -EINVAL;
            }
838 839 840

            ch = qemu_get_byte(f);
            memset(host, ch, TARGET_PAGE_SIZE);
841 842
#ifndef _WIN32
            if (ch == 0 &&
843 844
                (!kvm_enabled() || kvm_has_sync_mmu()) &&
                getpagesize() <= TARGET_PAGE_SIZE) {
A
Andreas Färber 已提交
845
                qemu_madvise(host, TARGET_PAGE_SIZE, QEMU_MADV_DONTNEED);
846 847 848
            }
#endif
        } else if (flags & RAM_SAVE_FLAG_PAGE) {
849 850
            void *host;

851
            host = host_from_stream_offset(f, addr, flags);
852 853 854
            if (!host) {
                return -EINVAL;
            }
855 856

            qemu_get_buffer(f, host, TARGET_PAGE_SIZE);
857 858 859 860 861 862 863 864 865 866 867 868 869
        } else if (flags & RAM_SAVE_FLAG_XBZRLE) {
            if (!migrate_use_xbzrle()) {
                return -EINVAL;
            }
            void *host = host_from_stream_offset(f, addr, flags);
            if (!host) {
                return -EINVAL;
            }

            if (load_xbzrle(f, addr, host) < 0) {
                ret = -EINVAL;
                goto done;
            }
870
        }
871 872
        error = qemu_file_get_error(f);
        if (error) {
O
Orit Wasserman 已提交
873 874
            ret = error;
            goto done;
875 876 877
        }
    } while (!(flags & RAM_SAVE_FLAG_EOS));

O
Orit Wasserman 已提交
878
done:
879 880
    DPRINTF("Completed load of VM with exit code %d seq iteration "
            "%" PRIu64 "\n", ret, seq_iter);
O
Orit Wasserman 已提交
881
    return ret;
882 883
}

884
SaveVMHandlers savevm_ram_handlers = {
885
    .save_live_setup = ram_save_setup,
886 887
    .save_live_iterate = ram_save_iterate,
    .save_live_complete = ram_save_complete,
888
    .load_state = ram_load,
889
    .cancel = ram_migration_cancel,
890 891
};

892
#ifdef HAS_AUDIO
I
Isaku Yamahata 已提交
893 894 895 896 897 898
struct soundhw {
    const char *name;
    const char *descr;
    int enabled;
    int isa;
    union {
899
        int (*init_isa) (ISABus *bus);
I
Isaku Yamahata 已提交
900 901 902 903 904
        int (*init_pci) (PCIBus *bus);
    } init;
};

static struct soundhw soundhw[] = {
905
#ifdef HAS_AUDIO_CHOICE
906
#ifdef CONFIG_PCSPK
907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979
    {
        "pcspk",
        "PC speaker",
        0,
        1,
        { .init_isa = pcspk_audio_init }
    },
#endif

#ifdef CONFIG_SB16
    {
        "sb16",
        "Creative Sound Blaster 16",
        0,
        1,
        { .init_isa = SB16_init }
    },
#endif

#ifdef CONFIG_CS4231A
    {
        "cs4231a",
        "CS4231A",
        0,
        1,
        { .init_isa = cs4231a_init }
    },
#endif

#ifdef CONFIG_ADLIB
    {
        "adlib",
#ifdef HAS_YMF262
        "Yamaha YMF262 (OPL3)",
#else
        "Yamaha YM3812 (OPL2)",
#endif
        0,
        1,
        { .init_isa = Adlib_init }
    },
#endif

#ifdef CONFIG_GUS
    {
        "gus",
        "Gravis Ultrasound GF1",
        0,
        1,
        { .init_isa = GUS_init }
    },
#endif

#ifdef CONFIG_AC97
    {
        "ac97",
        "Intel 82801AA AC97 Audio",
        0,
        0,
        { .init_pci = ac97_init }
    },
#endif

#ifdef CONFIG_ES1370
    {
        "es1370",
        "ENSONIQ AudioPCI ES1370",
        0,
        0,
        { .init_pci = es1370_init }
    },
#endif

980 981 982 983 984 985 986 987 988 989
#ifdef CONFIG_HDA
    {
        "hda",
        "Intel HD Audio",
        0,
        0,
        { .init_pci = intel_hda_and_codec_init }
    },
#endif

990 991 992 993 994 995 996 997 998
#endif /* HAS_AUDIO_CHOICE */

    { NULL, NULL, 0, 0, { NULL } }
};

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

999
    if (is_help_option(optarg)) {
1000 1001
    show_valid_cards:

1002
#ifdef HAS_AUDIO_CHOICE
1003 1004 1005 1006 1007
        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");
1008 1009 1010 1011
#else
        printf("Machine has no user-selectable audio hardware "
               "(it may or may not have always-present audio hardware).\n");
#endif
1012
        exit(!is_help_option(optarg));
1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057
    }
    else {
        size_t l;
        const char *p;
        char *e;
        int bad_card = 0;

        if (!strcmp(optarg, "all")) {
            for (c = soundhw; c->name; ++c) {
                c->enabled = 1;
            }
            return;
        }

        p = optarg;
        while (*p) {
            e = strchr(p, ',');
            l = !e ? strlen(p) : (size_t) (e - p);

            for (c = soundhw; c->name; ++c) {
                if (!strncmp(c->name, p, l) && !c->name[l]) {
                    c->enabled = 1;
                    break;
                }
            }

            if (!c->name) {
                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;
        }
    }
}
I
Isaku Yamahata 已提交
1058

1059
void audio_init(ISABus *isa_bus, PCIBus *pci_bus)
I
Isaku Yamahata 已提交
1060 1061 1062 1063 1064 1065
{
    struct soundhw *c;

    for (c = soundhw; c->name; ++c) {
        if (c->enabled) {
            if (c->isa) {
1066 1067
                if (isa_bus) {
                    c->init.init_isa(isa_bus);
I
Isaku Yamahata 已提交
1068 1069 1070 1071 1072 1073 1074 1075 1076
                }
            } else {
                if (pci_bus) {
                    c->init.init_pci(pci_bus);
                }
            }
        }
    }
}
1077 1078 1079 1080
#else
void select_soundhw(const char *optarg)
{
}
1081
void audio_init(ISABus *isa_bus, PCIBus *pci_bus)
I
Isaku Yamahata 已提交
1082 1083
{
}
1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143
#endif

int qemu_uuid_parse(const char *str, uint8_t *uuid)
{
    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;
    }
#ifdef TARGET_I386
    smbios_add_field(1, offsetof(struct smbios_type_1, uuid), 16, uuid);
#endif
    return 0;
}

void do_acpitable_option(const char *optarg)
{
#ifdef TARGET_I386
    if (acpi_table_add(optarg) < 0) {
        fprintf(stderr, "Wrong acpi table provided\n");
        exit(1);
    }
#endif
}

void do_smbios_option(const char *optarg)
{
#ifdef TARGET_I386
    if (smbios_entry_add(optarg) < 0) {
        fprintf(stderr, "Wrong smbios provided\n");
        exit(1);
    }
#endif
}

void cpudef_init(void)
{
#if defined(cpudef_setup)
    cpudef_setup(); /* parse cpu definitions in target config file */
#endif
}

int audio_available(void)
{
#ifdef HAS_AUDIO
    return 1;
#else
    return 0;
#endif
}

1144 1145 1146 1147 1148
int tcg_available(void)
{
    return 1;
}

1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165
int kvm_available(void)
{
#ifdef CONFIG_KVM
    return 1;
#else
    return 0;
#endif
}

int xen_available(void)
{
#ifdef CONFIG_XEN
    return 1;
#else
    return 0;
#endif
}
1166 1167 1168 1169 1170 1171 1172 1173 1174 1175


TargetInfo *qmp_query_target(Error **errp)
{
    TargetInfo *info = g_malloc0(sizeof(*info));

    info->arch = TARGET_TYPE;

    return info;
}