migration/next for 20140505

-----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1
 
 iQIcBAABCAAGBQJTZ/QsAAoJEPSH7xhYctcjJ4wP/31lNMpPLxdRpxpUoTPK39L0
 zEoBTUH5dzG2FG34p8/znNH0Lbmc3GALqP9UzyM0sL3QU0EDwqAP0idLJ/8zHrYN
 ZNJnNH6wexlpEypAsyQiayAFYAq6bc3hYtnEurN56dQgPVgb6jCicuOAt26JafLg
 xuGIuencgaEFZaIb8iobxloqDX9raoq0+JfnoV8r02+ES+PG+HHfZsAlbDiWpj/Z
 o06WBq9Xvf8X/zvuGHb9PCLGK36+kxPJ2G2531TJGZ6BjjewDExF2xlQH+PQ8pQB
 d2OKCP0In66JfyLu6JBz21APrks7DJxsuenvoqPhxhtlIcBSm+grZElNLbccDMLQ
 3fteWOCcri5WSwKNAkZ138D5SzIArZFBlqD9qW0GQsiaj3tCxfy162JAmzRPa94i
 R5OJCTDclwjCH6JvRsOs8NrQyYcXHcEl2rfeymfr68YOH5XBMeRPhcR91tVr03ow
 ZVgvwC/TJHntKnW+qtGIOa9Wfq4KghjddD37ayPAqKM/GQ0TV2sAPEr2htDaBZ0l
 5O/YGsAZPKax72OnMR2ObqoIMokympeanQLxG3Q8VKQSU2c+aU3rGVQrG1juiVYw
 NiC/nzarFzFWU8UYToXIu2QBDTde2vuRLX/myzL7L4OLavOcv8EXZBp60DP/bktE
 Pd+jdeLMh9kWu1tFrE8F
 =VoYL
 -----END PGP SIGNATURE-----

Merge remote-tracking branch 'remotes/juanquintela/tags/migration/20140505' into staging

migration/next for 20140505

# gpg: Signature made Mon 05 May 2014 21:27:24 BST using RSA key ID 5872D723
# gpg: Can't check signature: public key not found

* remotes/juanquintela/tags/migration/20140505: (36 commits)
  migration: expose xbzrle cache miss rate
  migration: expose the bitmap_sync_count to the end
  migration: Add counts of updating the dirty bitmap
  XBZRLE: Fix one XBZRLE corruption issues
  migration: remove duplicate code
  Coverity: Fix failure path for qemu_accept in migration
  Init the XBZRLE.lock in ram_mig_init
  Provide init function for ram migration
  Count used RAMBlock pages for migration_dirty_pages
  Make qemu_peek_buffer loop until it gets it's data
  Disallow outward migration while awaiting incoming migration
  virtio: validate config_len on load
  virtio-net: out-of-bounds buffer write on load
  openpic: avoid buffer overrun on incoming migration
  ssi-sd: fix buffer overrun on invalid state load
  savevm: Ignore minimum_version_id_old if there is no load_state_old
  usb: sanity check setup_index+setup_len in post_load
  vmstate: s/VMSTATE_INT32_LE/VMSTATE_INT32_POSITIVE_LE/
  virtio-scsi: fix buffer overrun on invalid state load
  zaurus: fix buffer overrun on invalid state load
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

arch_init.c

@@ -45,6 +45,7 @@
 #include "hw/audio/pcspk.h"
 #include "migration/page_cache.h"
 #include "qemu/config-file.h"
+#include "qemu/error-report.h"
 #include "qmp-commands.h"
 #include "trace.h"
 #include "exec/cpu-all.h"
@@ -110,6 +111,8 @@ static bool mig_throttle_on;
 static int dirty_rate_high_cnt;
 static void check_guest_throttling(void);
 
+static uint64_t bitmap_sync_count;
+
 /***********************************************************/
 /* ram save/restore */
 
@@ -167,11 +170,8 @@ static struct {
     /* Cache for XBZRLE, Protected by lock. */
     PageCache *cache;
     QemuMutex lock;
-} XBZRLE = {
+} XBZRLE;
-    .encoded_buf = NULL,
+
-    .current_buf = NULL,
-    .cache = NULL,
-};
 /* buffer used for XBZRLE decoding */
 static uint8_t *xbzrle_decoded_buf;
 
@@ -187,41 +187,44 @@ static void XBZRLE_cache_unlock(void)
         qemu_mutex_unlock(&XBZRLE.lock);
 }
 
+/*
+ * called from qmp_migrate_set_cache_size in main thread, possibly while
+ * a migration is in progress.
+ * A running migration maybe using the cache and might finish during this
+ * call, hence changes to the cache are protected by XBZRLE.lock().
+ */
 int64_t xbzrle_cache_resize(int64_t new_size)
 {
-    PageCache *new_cache, *cache_to_free;
+    PageCache *new_cache;
+    int64_t ret;
 
     if (new_size < TARGET_PAGE_SIZE) {
         return -1;
     }
 
-    /* no need to lock, the current thread holds qemu big lock */
+    XBZRLE_cache_lock();
+
     if (XBZRLE.cache != NULL) {
-        /* check XBZRLE.cache again later */
         if (pow2floor(new_size) == migrate_xbzrle_cache_size()) {
-            return pow2floor(new_size);
+            goto out_new_size;
         }
         new_cache = cache_init(new_size / TARGET_PAGE_SIZE,
                                         TARGET_PAGE_SIZE);
         if (!new_cache) {
-            DPRINTF("Error creating cache\n");
+            error_report("Error creating cache");
-            return -1;
+            ret = -1;
+            goto out;
         }
 
-        XBZRLE_cache_lock();
+        cache_fini(XBZRLE.cache);
-        /* the XBZRLE.cache may have be destroyed, check it again */
+        XBZRLE.cache = new_cache;
-        if (XBZRLE.cache != NULL) {
-            cache_to_free = XBZRLE.cache;
-            XBZRLE.cache = new_cache;
-        } else {
-            cache_to_free = new_cache;
-        }
-        XBZRLE_cache_unlock();
-
-        cache_fini(cache_to_free);
     }
 
-    return pow2floor(new_size);
+out_new_size:
+    ret = pow2floor(new_size);
+out:
+    XBZRLE_cache_unlock();
+    return ret;
 }
 
 /* accounting for migration statistics */
@@ -233,6 +236,7 @@ typedef struct AccountingInfo {
     uint64_t xbzrle_bytes;
     uint64_t xbzrle_pages;
     uint64_t xbzrle_cache_miss;
+    double xbzrle_cache_miss_rate;
     uint64_t xbzrle_overflows;
 } AccountingInfo;
 
@@ -288,6 +292,11 @@ uint64_t xbzrle_mig_pages_cache_miss(void)
     return acct_info.xbzrle_cache_miss;
 }
 
+double xbzrle_mig_cache_miss_rate(void)
+{
+    return acct_info.xbzrle_cache_miss_rate;
+}
+
 uint64_t xbzrle_mig_pages_overflow(void)
 {
     return acct_info.xbzrle_overflows;
@@ -340,7 +349,7 @@ static void xbzrle_cache_zero_page(ram_addr_t current_addr)
 
 #define ENCODING_FLAG_XBZRLE 0x1
 
-static int save_xbzrle_page(QEMUFile *f, uint8_t *current_data,
+static int save_xbzrle_page(QEMUFile *f, uint8_t **current_data,
                             ram_addr_t current_addr, RAMBlock *block,
                             ram_addr_t offset, int cont, bool last_stage)
 {
@@ -348,19 +357,23 @@ static int save_xbzrle_page(QEMUFile *f, uint8_t *current_data,
     uint8_t *prev_cached_page;
 
     if (!cache_is_cached(XBZRLE.cache, current_addr)) {
+        acct_info.xbzrle_cache_miss++;
         if (!last_stage) {
-            if (cache_insert(XBZRLE.cache, current_addr, current_data) == -1) {
+            if (cache_insert(XBZRLE.cache, current_addr, *current_data) == -1) {
                 return -1;
+            } else {
+                /* update *current_data when the page has been
+                   inserted into cache */
+                *current_data = get_cached_data(XBZRLE.cache, current_addr);
             }
         }
-        acct_info.xbzrle_cache_miss++;
         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);
+    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,
@@ -373,7 +386,10 @@ static int save_xbzrle_page(QEMUFile *f, uint8_t *current_data,
         DPRINTF("Overflow\n");
         acct_info.xbzrle_overflows++;
         /* update data in the cache */
-        memcpy(prev_cached_page, current_data, TARGET_PAGE_SIZE);
+        if (!last_stage) {
+            memcpy(prev_cached_page, *current_data, TARGET_PAGE_SIZE);
+            *current_data = prev_cached_page;
+        }
         return -1;
     }
 
@@ -479,6 +495,10 @@ static void migration_bitmap_sync(void)
     static int64_t num_dirty_pages_period;
     int64_t end_time;
     int64_t bytes_xfer_now;
+    static uint64_t xbzrle_cache_miss_prev;
+    static uint64_t iterations_prev;
+
+    bitmap_sync_count++;
 
     if (!bytes_xfer_prev) {
         bytes_xfer_prev = ram_bytes_transferred();
@@ -520,11 +540,22 @@ static void migration_bitmap_sync(void)
         } else {
              mig_throttle_on = false;
         }
+        if (migrate_use_xbzrle()) {
+            if (iterations_prev != 0) {
+                acct_info.xbzrle_cache_miss_rate =
+                   (double)(acct_info.xbzrle_cache_miss -
+                            xbzrle_cache_miss_prev) /
+                   (acct_info.iterations - iterations_prev);
+            }
+            iterations_prev = acct_info.iterations;
+            xbzrle_cache_miss_prev = acct_info.xbzrle_cache_miss;
+        }
         s->dirty_pages_rate = num_dirty_pages_period * 1000
             / (end_time - start_time);
         s->dirty_bytes_rate = s->dirty_pages_rate * TARGET_PAGE_SIZE;
         start_time = end_time;
         num_dirty_pages_period = 0;
+        s->dirty_sync_count = bitmap_sync_count;
     }
 }
 
@@ -598,15 +629,9 @@ static int ram_save_block(QEMUFile *f, bool last_stage)
                  */
                 xbzrle_cache_zero_page(current_addr);
             } else if (!ram_bulk_stage && migrate_use_xbzrle()) {
-                bytes_sent = save_xbzrle_page(f, p, current_addr, block,
+                bytes_sent = save_xbzrle_page(f, &p, current_addr, block,
                                               offset, cont, last_stage);
                 if (!last_stage) {
-                    /* We must send exactly what's in the xbzrle cache
-                     * even if the page wasn't xbzrle compressed, so that
-                     * it's right next time.
-                     */
-                    p = get_cached_data(XBZRLE.cache, current_addr);
-
                     /* Can't send this cached data async, since the cache page
                      * might get updated before it gets to the wire
                      */
@@ -726,37 +751,34 @@ static void reset_ram_globals(void)
 static int ram_save_setup(QEMUFile *f, void *opaque)
 {
     RAMBlock *block;
-    int64_t ram_pages = last_ram_offset() >> TARGET_PAGE_BITS;
+    int64_t ram_bitmap_pages; /* Size of bitmap in pages, including gaps */
 
-    migration_bitmap = bitmap_new(ram_pages);
-    bitmap_set(migration_bitmap, 0, ram_pages);
-    migration_dirty_pages = ram_pages;
     mig_throttle_on = false;
     dirty_rate_high_cnt = 0;
+    bitmap_sync_count = 0;
 
     if (migrate_use_xbzrle()) {
-        qemu_mutex_lock_iothread();
+        XBZRLE_cache_lock();
         XBZRLE.cache = cache_init(migrate_xbzrle_cache_size() /
                                   TARGET_PAGE_SIZE,
                                   TARGET_PAGE_SIZE);
         if (!XBZRLE.cache) {
-            qemu_mutex_unlock_iothread();
+            XBZRLE_cache_unlock();
-            DPRINTF("Error creating cache\n");
+            error_report("Error creating cache");
             return -1;
         }
-        qemu_mutex_init(&XBZRLE.lock);
+        XBZRLE_cache_unlock();
-        qemu_mutex_unlock_iothread();
 
         /* We prefer not to abort if there is no memory */
         XBZRLE.encoded_buf = g_try_malloc0(TARGET_PAGE_SIZE);
         if (!XBZRLE.encoded_buf) {
-            DPRINTF("Error allocating encoded_buf\n");
+            error_report("Error allocating encoded_buf");
             return -1;
         }
 
         XBZRLE.current_buf = g_try_malloc(TARGET_PAGE_SIZE);
         if (!XBZRLE.current_buf) {
-            DPRINTF("Error allocating current_buf\n");
+            error_report("Error allocating current_buf");
             g_free(XBZRLE.encoded_buf);
             XBZRLE.encoded_buf = NULL;
             return -1;
@@ -770,6 +792,22 @@ static int ram_save_setup(QEMUFile *f, void *opaque)
     bytes_transferred = 0;
     reset_ram_globals();
 
+    ram_bitmap_pages = last_ram_offset() >> TARGET_PAGE_BITS;
+    migration_bitmap = bitmap_new(ram_bitmap_pages);
+    bitmap_set(migration_bitmap, 0, ram_bitmap_pages);
+
+    /*
+     * Count the total number of pages used by ram blocks not including any
+     * gaps due to alignment or unplugs.
+     */
+    migration_dirty_pages = 0;
+    QTAILQ_FOREACH(block, &ram_list.blocks, next) {
+        uint64_t block_pages;
+
+        block_pages = block->length >> TARGET_PAGE_BITS;
+        migration_dirty_pages += block_pages;
+    }
+
     memory_global_dirty_log_start();
     migration_bitmap_sync();
     qemu_mutex_unlock_iothread();
@@ -997,7 +1035,7 @@ static int ram_load(QEMUFile *f, void *opaque, int version_id)
 
     seq_iter++;
 
-    if (version_id < 4 || version_id > 4) {
+    if (version_id != 4) {
         return -EINVAL;
     }
 
@@ -1008,44 +1046,42 @@ static int ram_load(QEMUFile *f, void *opaque, int version_id)
         addr &= TARGET_PAGE_MASK;
 
         if (flags & RAM_SAVE_FLAG_MEM_SIZE) {
-            if (version_id == 4) {
+            /* Synchronize RAM block list */
-                /* Synchronize RAM block list */
+            char id[256];
-                char id[256];
+            ram_addr_t length;
-                ram_addr_t length;
+            ram_addr_t total_ram_bytes = addr;
-                ram_addr_t total_ram_bytes = addr;
 
-                while (total_ram_bytes) {
+            while (total_ram_bytes) {
-                    RAMBlock *block;
+                RAMBlock *block;
-                    uint8_t len;
+                uint8_t len;
 
-                    len = qemu_get_byte(f);
+                len = qemu_get_byte(f);
-                    qemu_get_buffer(f, (uint8_t *)id, len);
+                qemu_get_buffer(f, (uint8_t *)id, len);
-                    id[len] = 0;
+                id[len] = 0;
-                    length = qemu_get_be64(f);
+                length = qemu_get_be64(f);
 
-                    QTAILQ_FOREACH(block, &ram_list.blocks, next) {
+                QTAILQ_FOREACH(block, &ram_list.blocks, next) {
-                        if (!strncmp(id, block->idstr, sizeof(id))) {
+                    if (!strncmp(id, block->idstr, sizeof(id))) {
-                            if (block->length != length) {
+                        if (block->length != length) {
-                                fprintf(stderr,
+                            fprintf(stderr,
-                                        "Length mismatch: %s: " RAM_ADDR_FMT
+                                    "Length mismatch: %s: " RAM_ADDR_FMT
-                                        " in != " RAM_ADDR_FMT "\n", id, length,
+                                    " in != " RAM_ADDR_FMT "\n", id, length,
-                                        block->length);
+                                    block->length);
-                                ret =  -EINVAL;
+                            ret =  -EINVAL;
-                                goto done;
+                            goto done;
-                            }
-                            break;
                         }
+                        break;
                     }
-
-                    if (!block) {
-                        fprintf(stderr, "Unknown ramblock \"%s\", cannot "
-                                "accept migration\n", id);
-                        ret = -EINVAL;
-                        goto done;
-                    }
-
-                    total_ram_bytes -= length;
                 }
+
+                if (!block) {
+                    fprintf(stderr, "Unknown ramblock \"%s\", cannot "
+                            "accept migration\n", id);
+                    ret = -EINVAL;
+                    goto done;
+                }
+
+                total_ram_bytes -= length;
             }
         }
 
@@ -1095,7 +1131,7 @@ done:
     return ret;
 }
 
-SaveVMHandlers savevm_ram_handlers = {
+static SaveVMHandlers savevm_ram_handlers = {
     .save_live_setup = ram_save_setup,
     .save_live_iterate = ram_save_iterate,
     .save_live_complete = ram_save_complete,
@@ -1104,6 +1140,12 @@ SaveVMHandlers savevm_ram_handlers = {
     .cancel = ram_migration_cancel,
 };
 
+void ram_mig_init(void)
+{
+    qemu_mutex_init(&XBZRLE.lock);
+    register_savevm_live(NULL, "ram", 0, 4, &savevm_ram_handlers, NULL);
+}
+
 struct soundhw {
     const char *name;
     const char *descr;

docs/migration.txt

@@ -139,7 +139,6 @@ static const VMStateDescription vmstate_kbd = {
     .name = "pckbd",
     .version_id = 3,
     .minimum_version_id = 3,
-    .minimum_version_id_old = 3,
     .fields      = (VMStateField []) {
         VMSTATE_UINT8(write_cmd, KBDState),
         VMSTATE_UINT8(status, KBDState),
@@ -168,12 +167,13 @@ You can see that there are several version fields:
 - minimum_version_id: the minimum version_id that VMState is able to understand
   for that device.
 - minimum_version_id_old: For devices that were not able to port to vmstate, we can
-  assign a function that knows how to read this old state.
+  assign a function that knows how to read this old state. This field is
+  ignored if there is no load_state_old handler.
 
 So, VMState is able to read versions from minimum_version_id to
-version_id.  And the function load_state_old() is able to load state
+version_id.  And the function load_state_old() (if present) is able to
-from minimum_version_id_old to minimum_version_id.  This function is
+load state from minimum_version_id_old to minimum_version_id.  This
-deprecated and will be removed when no more users are left.
+function is deprecated and will be removed when no more users are left.
 
 ===  Massaging functions ===
 
@@ -255,7 +255,6 @@ const VMStateDescription vmstate_ide_drive_pio_state = {
     .name = "ide_drive/pio_state",
     .version_id = 1,
     .minimum_version_id = 1,
-    .minimum_version_id_old = 1,
     .pre_save = ide_drive_pio_pre_save,
     .post_load = ide_drive_pio_post_load,
     .fields      = (VMStateField []) {
@@ -275,7 +274,6 @@ const VMStateDescription vmstate_ide_drive = {
     .name = "ide_drive",
     .version_id = 3,
     .minimum_version_id = 0,
-    .minimum_version_id_old = 0,
     .post_load = ide_drive_post_load,
     .fields      = (VMStateField []) {
         .... several fields ....

hmp.c

@@ -188,6 +188,8 @@ void hmp_info_migrate(Monitor *mon, const QDict *qdict)
                        info->ram->normal);
         monitor_printf(mon, "normal bytes: %" PRIu64 " kbytes\n",
                        info->ram->normal_bytes >> 10);
+        monitor_printf(mon, "dirty sync count: %" PRIu64 "\n",
+                       info->ram->dirty_sync_count);
         if (info->ram->dirty_pages_rate) {
             monitor_printf(mon, "dirty pages rate: %" PRIu64 " pages\n",
                            info->ram->dirty_pages_rate);
@@ -212,6 +214,8 @@ void hmp_info_migrate(Monitor *mon, const QDict *qdict)
                        info->xbzrle_cache->pages);
         monitor_printf(mon, "xbzrle cache miss: %" PRIu64 "\n",
                        info->xbzrle_cache->cache_miss);
+        monitor_printf(mon, "xbzrle cache miss rate: %0.2f\n",
+                       info->xbzrle_cache->cache_miss_rate);
         monitor_printf(mon, "xbzrle overflow : %" PRIu64 "\n",
                        info->xbzrle_cache->overflow);
     }

hw/arm/pxa2xx.c

@@ -732,7 +732,7 @@ static void pxa2xx_ssp_save(QEMUFile *f, void *opaque)
 static int pxa2xx_ssp_load(QEMUFile *f, void *opaque, int version_id)
 {
     PXA2xxSSPState *s = (PXA2xxSSPState *) opaque;
-    int i;
+    int i, v;
 
     s->enable = qemu_get_be32(f);
 
@@ -746,7 +746,11 @@ static int pxa2xx_ssp_load(QEMUFile *f, void *opaque, int version_id)
     qemu_get_8s(f, &s->ssrsa);
     qemu_get_8s(f, &s->ssacd);
 
-    s->rx_level = qemu_get_byte(f);
+    v = qemu_get_byte(f);
+    if (v < 0 || v > ARRAY_SIZE(s->rx_fifo)) {
+        return -EINVAL;
+    }
+    s->rx_level = v;
     s->rx_start = 0;
     for (i = 0; i < s->rx_level; i ++)
         s->rx_fifo[i] = qemu_get_byte(f);

hw/display/ssd0323.c

@@ -312,18 +312,42 @@ static int ssd0323_load(QEMUFile *f, void *opaque, int version_id)
         return -EINVAL;
 
     s->cmd_len = qemu_get_be32(f);
+    if (s->cmd_len < 0 || s->cmd_len > ARRAY_SIZE(s->cmd_data)) {
+        return -EINVAL;
+    }
     s->cmd = qemu_get_be32(f);
     for (i = 0; i < 8; i++)
         s->cmd_data[i] = qemu_get_be32(f);
     s->row = qemu_get_be32(f);
+    if (s->row < 0 || s->row >= 80) {
+        return -EINVAL;
+    }
     s->row_start = qemu_get_be32(f);
+    if (s->row_start < 0 || s->row_start >= 80) {
+        return -EINVAL;
+    }
     s->row_end = qemu_get_be32(f);
+    if (s->row_end < 0 || s->row_end >= 80) {
+        return -EINVAL;
+    }
     s->col = qemu_get_be32(f);
+    if (s->col < 0 || s->col >= 64) {
+        return -EINVAL;
+    }
     s->col_start = qemu_get_be32(f);
+    if (s->col_start < 0 || s->col_start >= 64) {
+        return -EINVAL;
+    }
     s->col_end = qemu_get_be32(f);
+    if (s->col_end < 0 || s->col_end >= 64) {
+        return -EINVAL;
+    }
     s->redraw = qemu_get_be32(f);
     s->remap = qemu_get_be32(f);
     s->mode = qemu_get_be32(f);
+    if (s->mode != SSD0323_CMD && s->mode != SSD0323_DATA) {
+        return -EINVAL;
+    }
     qemu_get_buffer(f, s->framebuffer, sizeof(s->framebuffer));
 
     ss->cs = qemu_get_be32(f);

hw/gpio/zaurus.c

@@ -203,6 +203,15 @@ static bool is_version_0 (void *opaque, int version_id)
     return version_id == 0;
 }
 
+static bool vmstate_scoop_validate(void *opaque, int version_id)
+{
+    ScoopInfo *s = opaque;
+
+    return !(s->prev_level & 0xffff0000) &&
+        !(s->gpio_level & 0xffff0000) &&
+        !(s->gpio_dir & 0xffff0000);
+}
+
 static const VMStateDescription vmstate_scoop_regs = {
     .name = "scoop",
     .version_id = 1,
@@ -215,6 +224,7 @@ static const VMStateDescription vmstate_scoop_regs = {
         VMSTATE_UINT32(gpio_level, ScoopInfo),
         VMSTATE_UINT32(gpio_dir, ScoopInfo),
         VMSTATE_UINT32(prev_level, ScoopInfo),
+        VMSTATE_VALIDATE("irq levels are 16 bit", vmstate_scoop_validate),
         VMSTATE_UINT16(mcr, ScoopInfo),
         VMSTATE_UINT16(cdr, ScoopInfo),
         VMSTATE_UINT16(ccr, ScoopInfo),

hw/ide/ahci.c

@@ -1293,7 +1293,7 @@ const VMStateDescription vmstate_ahci = {
         VMSTATE_UINT32(control_regs.impl, AHCIState),
         VMSTATE_UINT32(control_regs.version, AHCIState),
         VMSTATE_UINT32(idp_index, AHCIState),
-        VMSTATE_INT32(ports, AHCIState),
+        VMSTATE_INT32_EQUAL(ports, AHCIState),
         VMSTATE_END_OF_LIST()
     },
 };

hw/input/tsc210x.c

@@ -1070,9 +1070,21 @@ static int tsc210x_load(QEMUFile *f, void *opaque, int version_id)
     s->enabled = qemu_get_byte(f);
     s->host_mode = qemu_get_byte(f);
     s->function = qemu_get_byte(f);
+    if (s->function < 0 || s->function >= ARRAY_SIZE(mode_regs)) {
+        return -EINVAL;
+    }
     s->nextfunction = qemu_get_byte(f);
+    if (s->nextfunction < 0 || s->nextfunction >= ARRAY_SIZE(mode_regs)) {
+        return -EINVAL;
+    }
     s->precision = qemu_get_byte(f);
+    if (s->precision < 0 || s->precision >= ARRAY_SIZE(resolution)) {
+        return -EINVAL;
+    }
     s->nextprecision = qemu_get_byte(f);
+    if (s->nextprecision < 0 || s->nextprecision >= ARRAY_SIZE(resolution)) {
+        return -EINVAL;
+    }
     s->filter = qemu_get_byte(f);
     s->pin_func = qemu_get_byte(f);
     s->ref = qemu_get_byte(f);

hw/intc/openpic.c

@@ -41,6 +41,7 @@
 #include "hw/sysbus.h"
 #include "hw/pci/msi.h"
 #include "qemu/bitops.h"
+#include "qapi/qmp/qerror.h"
 
 //#define DEBUG_OPENPIC
 
@@ -1416,7 +1417,7 @@ static void openpic_load_IRQ_queue(QEMUFile* f, IRQQueue *q)
 static int openpic_load(QEMUFile* f, void *opaque, int version_id)
 {
     OpenPICState *opp = (OpenPICState *)opaque;
-    unsigned int i;
+    unsigned int i, nb_cpus;
 
     if (version_id != 1) {
         return -EINVAL;
@@ -1428,7 +1429,11 @@ static int openpic_load(QEMUFile* f, void *opaque, int version_id)
     qemu_get_be32s(f, &opp->spve);
     qemu_get_be32s(f, &opp->tfrr);
 
-    qemu_get_be32s(f, &opp->nb_cpus);
+    qemu_get_be32s(f, &nb_cpus);
+    if (opp->nb_cpus != nb_cpus) {
+        return -EINVAL;
+    }
+    assert(nb_cpus > 0 && nb_cpus <= MAX_CPU);
 
     for (i = 0; i < opp->nb_cpus; i++) {
         qemu_get_sbe32s(f, &opp->dst[i].ctpr);
@@ -1567,6 +1572,13 @@ static void openpic_realize(DeviceState *dev, Error **errp)
         {NULL}
     };
 
+    if (opp->nb_cpus > MAX_CPU) {
+        error_set(errp, QERR_PROPERTY_VALUE_OUT_OF_RANGE,
+                  TYPE_OPENPIC, "nb_cpus", (uint64_t)opp->nb_cpus,
+                  (uint64_t)0, (uint64_t)MAX_CPU);
+        return;
+    }
+
     switch (opp->model) {
     case OPENPIC_MODEL_FSL_MPIC_20:
     default:

hw/net/virtio-net.c

@@ -1362,10 +1362,17 @@ static int virtio_net_load(QEMUFile *f, void *opaque, int version_id)
         if (n->mac_table.in_use <= MAC_TABLE_ENTRIES) {
             qemu_get_buffer(f, n->mac_table.macs,
                             n->mac_table.in_use * ETH_ALEN);
-        } else if (n->mac_table.in_use) {
+        } else {
-            uint8_t *buf = g_malloc0(n->mac_table.in_use);
+            int64_t i;
-            qemu_get_buffer(f, buf, n->mac_table.in_use * ETH_ALEN);
+
-            g_free(buf);
+            /* Overflow detected - can happen if source has a larger MAC table.
+             * We simply set overflow flag so there's no need to maintain the
+             * table of addresses, discard them all.
+             * Note: 64 bit math to avoid integer overflow.
+             */
+            for (i = 0; i < (int64_t)n->mac_table.in_use * ETH_ALEN; ++i) {
+                qemu_get_byte(f);
+            }
             n->mac_table.multi_overflow = n->mac_table.uni_overflow = 1;
             n->mac_table.in_use = 0;
         }
@@ -1407,6 +1414,11 @@ static int virtio_net_load(QEMUFile *f, void *opaque, int version_id)
         }
 
         n->curr_queues = qemu_get_be16(f);
+        if (n->curr_queues > n->max_queues) {
+            error_report("virtio-net: curr_queues %x > max_queues %x",
+                         n->curr_queues, n->max_queues);
+            return -1;
+        }
         for (i = 1; i < n->curr_queues; i++) {
             n->vqs[i].tx_waiting = qemu_get_be32(f);
         }

hw/pci/pci.c

@@ -475,7 +475,7 @@ const VMStateDescription vmstate_pci_device = {
     .minimum_version_id = 1,
     .minimum_version_id_old = 1,
     .fields      = (VMStateField []) {
-        VMSTATE_INT32_LE(version_id, PCIDevice),
+        VMSTATE_INT32_POSITIVE_LE(version_id, PCIDevice),
         VMSTATE_BUFFER_UNSAFE_INFO(config, PCIDevice, 0,
                                    vmstate_info_pci_config,
                                    PCI_CONFIG_SPACE_SIZE),
@@ -492,7 +492,7 @@ const VMStateDescription vmstate_pcie_device = {
     .minimum_version_id = 1,
     .minimum_version_id_old = 1,
     .fields      = (VMStateField []) {
-        VMSTATE_INT32_LE(version_id, PCIDevice),
+        VMSTATE_INT32_POSITIVE_LE(version_id, PCIDevice),
         VMSTATE_BUFFER_UNSAFE_INFO(config, PCIDevice, 0,
                                    vmstate_info_pci_config,
                                    PCIE_CONFIG_SPACE_SIZE),

hw/pci/pcie_aer.c

@@ -795,6 +795,13 @@ static const VMStateDescription vmstate_pcie_aer_err = {
     }
 };
 
+static bool pcie_aer_state_log_num_valid(void *opaque, int version_id)
+{
+    PCIEAERLog *s = opaque;
+
+    return s->log_num <= s->log_max;
+}
+
 const VMStateDescription vmstate_pcie_aer_log = {
     .name = "PCIE_AER_ERROR_LOG",
     .version_id = 1,
@@ -802,7 +809,8 @@ const VMStateDescription vmstate_pcie_aer_log = {
     .minimum_version_id_old = 1,
     .fields     = (VMStateField[]) {
         VMSTATE_UINT16(log_num, PCIEAERLog),
-        VMSTATE_UINT16(log_max, PCIEAERLog),
+        VMSTATE_UINT16_EQUAL(log_max, PCIEAERLog),
+        VMSTATE_VALIDATE("log_num <= log_max", pcie_aer_state_log_num_valid),
         VMSTATE_STRUCT_VARRAY_POINTER_UINT16(log, PCIEAERLog, log_num,
                               vmstate_pcie_aer_err, PCIEAERErr),
         VMSTATE_END_OF_LIST()

hw/scsi/virtio-scsi.c

@@ -147,6 +147,15 @@ static void *virtio_scsi_load_request(QEMUFile *f, SCSIRequest *sreq)
     qemu_get_be32s(f, &n);
     assert(n < vs->conf.num_queues);
     qemu_get_buffer(f, (unsigned char *)&req->elem, sizeof(req->elem));
+    /* TODO: add a way for SCSIBusInfo's load_request to fail,
+     * and fail migration instead of asserting here.
+     * When we do, we might be able to re-enable NDEBUG below.
+     */
+#ifdef NDEBUG
+#error building with NDEBUG is not supported
+#endif
+    assert(req->elem.in_num <= ARRAY_SIZE(req->elem.in_sg));
+    assert(req->elem.out_num <= ARRAY_SIZE(req->elem.out_sg));
     virtio_scsi_parse_req(s, vs->cmd_vqs[n], req);
 
     scsi_req_ref(sreq);

hw/sd/ssi-sd.c

@@ -230,8 +230,17 @@ static int ssi_sd_load(QEMUFile *f, void *opaque, int version_id)
     for (i = 0; i < 5; i++)
         s->response[i] = qemu_get_be32(f);
     s->arglen = qemu_get_be32(f);
+    if (s->mode == SSI_SD_CMDARG &&
+        (s->arglen < 0 || s->arglen >= ARRAY_SIZE(s->cmdarg))) {
+        return -EINVAL;
+    }
     s->response_pos = qemu_get_be32(f);
     s->stopping = qemu_get_be32(f);
+    if (s->mode == SSI_SD_RESPONSE &&
+        (s->response_pos < 0 || s->response_pos >= ARRAY_SIZE(s->response) ||
+        (!s->stopping && s->arglen > ARRAY_SIZE(s->response)))) {
+        return -EINVAL;
+    }
 
     ss->cs = qemu_get_be32(f);
 

hw/ssi/pl022.c

@@ -240,11 +240,25 @@ static const MemoryRegionOps pl022_ops = {
     .endianness = DEVICE_NATIVE_ENDIAN,
 };
 
+static int pl022_post_load(void *opaque, int version_id)
+{
+    PL022State *s = opaque;
+
+    if (s->tx_fifo_head < 0 ||
+        s->tx_fifo_head >= ARRAY_SIZE(s->tx_fifo) ||
+        s->rx_fifo_head < 0 ||
+        s->rx_fifo_head >= ARRAY_SIZE(s->rx_fifo)) {
+        return -1;
+    }
+    return 0;
+}
+
 static const VMStateDescription vmstate_pl022 = {
     .name = "pl022_ssp",
     .version_id = 1,
     .minimum_version_id = 1,
     .minimum_version_id_old = 1,
+    .post_load = pl022_post_load,
     .fields      = (VMStateField[]) {
         VMSTATE_UINT32(cr0, PL022State),
         VMSTATE_UINT32(cr1, PL022State),

hw/timer/hpet.c

@@ -239,6 +239,18 @@ static int hpet_pre_load(void *opaque)
     return 0;
 }
 
+static bool hpet_validate_num_timers(void *opaque, int version_id)
+{
+    HPETState *s = opaque;
+
+    if (s->num_timers < HPET_MIN_TIMERS) {
+        return false;
+    } else if (s->num_timers > HPET_MAX_TIMERS) {
+        return false;
+    }
+    return true;
+}
+
 static int hpet_post_load(void *opaque, int version_id)
 {
     HPETState *s = opaque;
@@ -307,6 +319,7 @@ static const VMStateDescription vmstate_hpet = {
         VMSTATE_UINT64(isr, HPETState),
         VMSTATE_UINT64(hpet_counter, HPETState),
         VMSTATE_UINT8_V(num_timers, HPETState, 2),
+        VMSTATE_VALIDATE("num_timers in range", hpet_validate_num_timers),
         VMSTATE_STRUCT_VARRAY_UINT8(timer, HPETState, num_timers, 0,
                                     vmstate_hpet_timer, HPETTimer),
         VMSTATE_END_OF_LIST()

hw/usb/bus.c

@@ -49,7 +49,9 @@ static int usb_device_post_load(void *opaque, int version_id)
     } else {
         dev->attached = 1;
     }
-    if (dev->setup_index >= sizeof(dev->data_buf) ||
+    if (dev->setup_index < 0 ||
+        dev->setup_len < 0 ||
+        dev->setup_index >= sizeof(dev->data_buf) ||
         dev->setup_len >= sizeof(dev->data_buf)) {
         return -EINVAL;
     }

hw/virtio/virtio.c

@@ -430,6 +430,12 @@ void virtqueue_map_sg(struct iovec *sg, hwaddr *addr,
     unsigned int i;
     hwaddr len;
 
+    if (num_sg >= VIRTQUEUE_MAX_SIZE) {
+        error_report("virtio: map attempt out of bounds: %zd > %d",
+                     num_sg, VIRTQUEUE_MAX_SIZE);
+        exit(1);
+    }
+
     for (i = 0; i < num_sg; i++) {
         len = sg[i].iov_len;
         sg[i].iov_base = cpu_physical_memory_map(addr[i], &len, is_write);
@@ -891,7 +897,9 @@ int virtio_set_features(VirtIODevice *vdev, uint32_t val)
 
 int virtio_load(VirtIODevice *vdev, QEMUFile *f)
 {
-    int num, i, ret;
+    int i, ret;
+    int32_t config_len;
+    uint32_t num;
     uint32_t features;
     uint32_t supported_features;
     BusState *qbus = qdev_get_parent_bus(DEVICE(vdev));
@@ -906,6 +914,9 @@ int virtio_load(VirtIODevice *vdev, QEMUFile *f)
     qemu_get_8s(f, &vdev->status);
     qemu_get_8s(f, &vdev->isr);
     qemu_get_be16s(f, &vdev->queue_sel);
+    if (vdev->queue_sel >= VIRTIO_PCI_QUEUE_MAX) {
+        return -1;
+    }
     qemu_get_be32s(f, &features);
 
     if (virtio_set_features(vdev, features) < 0) {
@@ -914,11 +925,21 @@ int virtio_load(VirtIODevice *vdev, QEMUFile *f)
                      features, supported_features);
         return -1;
     }
-    vdev->config_len = qemu_get_be32(f);
+    config_len = qemu_get_be32(f);
+    if (config_len != vdev->config_len) {
+        error_report("Unexpected config length 0x%x. Expected 0x%zx",
+                     config_len, vdev->config_len);
+        return -1;
+    }
     qemu_get_buffer(f, vdev->config, vdev->config_len);
 
     num = qemu_get_be32(f);
 
+    if (num > VIRTIO_PCI_QUEUE_MAX) {
+        error_report("Invalid number of PCI queues: 0x%x", num);
+        return -1;
+    }
+
     for (i = 0; i < num; i++) {
         vdev->vq[i].vring.num = qemu_get_be32(f);
         if (k->has_variable_vring_alignment) {

include/hw/virtio/virtio-net.h

@@ -176,8 +176,8 @@ typedef struct VirtIONet {
     uint8_t nobcast;
     uint8_t vhost_started;
     struct {
-        int in_use;
+        uint32_t in_use;
-        int first_multi;
+        uint32_t first_multi;
         uint8_t multi_overflow;
         uint8_t uni_overflow;
         uint8_t *macs;

include/migration/migration.h

@@ -61,6 +61,7 @@ struct MigrationState
     bool enabled_capabilities[MIGRATION_CAPABILITY_MAX];
     int64_t xbzrle_cache_size;
     int64_t setup_time;
+    int64_t dirty_sync_count;
 };
 
 void process_incoming_migration(QEMUFile *f);
@@ -113,8 +114,6 @@ void free_xbzrle_decoded_buf(void);
 
 void acct_update_position(QEMUFile *f, size_t size, bool zero);
 
-extern SaveVMHandlers savevm_ram_handlers;
-
 uint64_t dup_mig_bytes_transferred(void);
 uint64_t dup_mig_pages_transferred(void);
 uint64_t skipped_mig_bytes_transferred(void);
@@ -125,6 +124,7 @@ uint64_t xbzrle_mig_bytes_transferred(void);
 uint64_t xbzrle_mig_pages_transferred(void);
 uint64_t xbzrle_mig_pages_overflow(void);
 uint64_t xbzrle_mig_pages_cache_miss(void);
+double xbzrle_mig_cache_miss_rate(void);
 
 void ram_handle_compressed(void *host, uint8_t ch, uint64_t size);
 

include/migration/qemu-file.h

@@ -123,6 +123,11 @@ void qemu_put_be32(QEMUFile *f, unsigned int v);
 void qemu_put_be64(QEMUFile *f, uint64_t v);
 int qemu_peek_buffer(QEMUFile *f, uint8_t *buf, int size, size_t offset);
 int qemu_get_buffer(QEMUFile *f, uint8_t *buf, int size);
+/*
+ * Note that you can only peek continuous bytes from where the current pointer
+ * is; you aren't guaranteed to be able to peak to +n bytes unless you've
+ * previously peeked +n-1.
+ */
 int qemu_peek_byte(QEMUFile *f, int offset);
 int qemu_get_byte(QEMUFile *f);
 void qemu_file_skip(QEMUFile *f, int size);

include/migration/vmstate.h

@@ -100,6 +100,7 @@ enum VMStateFlags {
     VMS_MULTIPLY         = 0x200,  /* multiply "size" field by field_size */
     VMS_VARRAY_UINT8     = 0x400,  /* Array with size in uint8_t field*/
     VMS_VARRAY_UINT32    = 0x800,  /* Array with size in uint32_t field*/
+    VMS_MUST_EXIST       = 0x1000, /* Field must exist in input */
 };
 
 typedef struct {
@@ -203,6 +204,14 @@ extern const VMStateInfo vmstate_info_bitmap;
     .offset       = vmstate_offset_value(_state, _field, _type),     \
 }
 
+/* Validate state using a boolean predicate. */
+#define VMSTATE_VALIDATE(_name, _test) { \
+    .name         = (_name),                                         \
+    .field_exists = (_test),                                         \
+    .flags        = VMS_ARRAY | VMS_MUST_EXIST,                      \
+    .num          = 0, /* 0 elements: no data, only run _test */     \
+}
+
 #define VMSTATE_POINTER(_field, _state, _version, _info, _type) {    \
     .name       = (stringify(_field)),                               \
     .version_id = (_version),                                        \
@@ -592,7 +601,7 @@ extern const VMStateInfo vmstate_info_bitmap;
 #define VMSTATE_UINT64_EQUAL(_f, _s)                                  \
     VMSTATE_UINT64_EQUAL_V(_f, _s, 0)
 
-#define VMSTATE_INT32_LE(_f, _s)                                   \
+#define VMSTATE_INT32_POSITIVE_LE(_f, _s)                             \
     VMSTATE_SINGLE(_f, _s, 0, vmstate_info_int32_le, int32_t)
 
 #define VMSTATE_UINT8_TEST(_f, _s, _t)                               \

include/sysemu/arch_init.h

@@ -29,6 +29,7 @@ extern const uint32_t arch_type;
 void select_soundhw(const char *optarg);
 void do_acpitable_option(const QemuOpts *opts);
 void do_smbios_option(QemuOpts *opts);
+void ram_mig_init(void);
 void cpudef_init(void);
 void audio_init(void);
 int tcg_available(void);

migration-tcp.c

@@ -13,7 +13,10 @@
  * GNU GPL, version 2 or (at your option) any later version.
  */
 
+#include <string.h>
+
 #include "qemu-common.h"
+#include "qemu/error-report.h"
 #include "qemu/sockets.h"
 #include "migration/migration.h"
 #include "migration/qemu-file.h"
@@ -56,24 +59,26 @@ static void tcp_accept_incoming_migration(void *opaque)
     socklen_t addrlen = sizeof(addr);
     int s = (intptr_t)opaque;
     QEMUFile *f;
-    int c;
+    int c, err;
 
     do {
         c = qemu_accept(s, (struct sockaddr *)&addr, &addrlen);
-    } while (c == -1 && socket_error() == EINTR);
+        err = socket_error();
+    } while (c < 0 && err == EINTR);
     qemu_set_fd_handler2(s, NULL, NULL, NULL, NULL);
     closesocket(s);
 
     DPRINTF("accepted migration\n");
 
-    if (c == -1) {
+    if (c < 0) {
-        fprintf(stderr, "could not accept migration connection\n");
+        error_report("could not accept migration connection (%s)",
-        goto out;
+                     strerror(err));
+        return;
     }
 
     f = qemu_fopen_socket(c, "rb");
     if (f == NULL) {
-        fprintf(stderr, "could not qemu_fopen socket\n");
+        error_report("could not qemu_fopen socket");
         goto out;
     }
 

migration-unix.c

@@ -13,7 +13,10 @@
  * GNU GPL, version 2 or (at your option) any later version.
  */
 
+#include <string.h>
+
 #include "qemu-common.h"
+#include "qemu/error-report.h"
 #include "qemu/sockets.h"
 #include "qemu/main-loop.h"
 #include "migration/migration.h"
@@ -56,24 +59,26 @@ static void unix_accept_incoming_migration(void *opaque)
     socklen_t addrlen = sizeof(addr);
     int s = (intptr_t)opaque;
     QEMUFile *f;
-    int c;
+    int c, err;
 
     do {
         c = qemu_accept(s, (struct sockaddr *)&addr, &addrlen);
-    } while (c == -1 && errno == EINTR);
+        err = errno;
+    } while (c < 0 && err == EINTR);
     qemu_set_fd_handler2(s, NULL, NULL, NULL, NULL);
     close(s);
 
     DPRINTF("accepted migration\n");
 
-    if (c == -1) {
+    if (c < 0) {
-        fprintf(stderr, "could not accept migration connection\n");
+        error_report("could not accept migration connection (%s)",
-        goto out;
+                     strerror(err));
+        return;
     }
 
     f = qemu_fopen_socket(c, "rb");
     if (f == NULL) {
-        fprintf(stderr, "could not qemu_fopen socket\n");
+        error_report("could not qemu_fopen socket");
         goto out;
     }
 

migration.c

@@ -174,6 +174,7 @@ static void get_xbzrle_cache_stats(MigrationInfo *info)
         info->xbzrle_cache->bytes = xbzrle_mig_bytes_transferred();
         info->xbzrle_cache->pages = xbzrle_mig_pages_transferred();
         info->xbzrle_cache->cache_miss = xbzrle_mig_pages_cache_miss();
+        info->xbzrle_cache->cache_miss_rate = xbzrle_mig_cache_miss_rate();
         info->xbzrle_cache->overflow = xbzrle_mig_pages_overflow();
     }
 }
@@ -215,6 +216,7 @@ MigrationInfo *qmp_query_migrate(Error **errp)
         info->ram->normal_bytes = norm_mig_bytes_transferred();
         info->ram->dirty_pages_rate = s->dirty_pages_rate;
         info->ram->mbps = s->mbps;
+        info->ram->dirty_sync_count = s->dirty_sync_count;
 
         if (blk_mig_active()) {
             info->has_disk = true;
@@ -248,6 +250,7 @@ MigrationInfo *qmp_query_migrate(Error **errp)
         info->ram->normal = norm_mig_pages_transferred();
         info->ram->normal_bytes = norm_mig_bytes_transferred();
         info->ram->mbps = s->mbps;
+        info->ram->dirty_sync_count = s->dirty_sync_count;
         break;
     case MIG_STATE_ERROR:
         info->has_status = true;
@@ -419,6 +422,11 @@ void qmp_migrate(const char *uri, bool has_blk, bool blk,
         return;
     }
 
+    if (runstate_check(RUN_STATE_INMIGRATE)) {
+        error_setg(errp, "Guest is waiting for an incoming migration");
+        return;
+    }
+
     if (qemu_savevm_state_blocked(errp)) {
         return;
     }

qapi-schema.json

@@ -651,13 +651,15 @@
 #
 # @mbps: throughput in megabits/sec. (since 1.6)
 #
+# @dirty-sync-count: number of times that dirty ram was synchronized (since 2.1)
+#
 # Since: 0.14.0
 ##
 { 'type': 'MigrationStats',
   'data': {'transferred': 'int', 'remaining': 'int', 'total': 'int' ,
            'duplicate': 'int', 'skipped': 'int', 'normal': 'int',
            'normal-bytes': 'int', 'dirty-pages-rate' : 'int',
-           'mbps' : 'number' } }
+           'mbps' : 'number', 'dirty-sync-count' : 'int' } }
 
 ##
 # @XBZRLECacheStats
@@ -672,13 +674,16 @@
 #
 # @cache-miss: number of cache miss
 #
+# @cache-miss-rate: rate of cache miss (since 2.1)
+#
 # @overflow: number of overflows
 #
 # Since: 1.2
 ##
 { 'type': 'XBZRLECacheStats',
   'data': {'cache-size': 'int', 'bytes': 'int', 'pages': 'int',
-           'cache-miss': 'int', 'overflow': 'int' } }
+           'cache-miss': 'int', 'cache-miss-rate': 'number',
+           'overflow': 'int' } }
 
 ##
 # @MigrationInfo

qemu-file.c

@@ -530,7 +530,15 @@ size_t ram_control_save_page(QEMUFile *f, ram_addr_t block_offset,
     return RAM_SAVE_CONTROL_NOT_SUPP;
 }
 
-static void qemu_fill_buffer(QEMUFile *f)
+/*
+ * Attempt to fill the buffer from the underlying file
+ * Returns the number of bytes read, or negative value for an error.
+ *
+ * Note that it can return a partially full buffer even in a not error/not EOF
+ * case if the underlying file descriptor gives a short read, and that can
+ * happen even on a blocking fd.
+ */
+static ssize_t qemu_fill_buffer(QEMUFile *f)
 {
     int len;
     int pending;
@@ -554,6 +562,8 @@ static void qemu_fill_buffer(QEMUFile *f)
     } else if (len != -EAGAIN) {
         qemu_file_set_error(f, len);
     }
+
+    return len;
 }
 
 int qemu_get_fd(QEMUFile *f)
@@ -685,17 +695,39 @@ void qemu_file_skip(QEMUFile *f, int size)
     }
 }
 
+/*
+ * Read 'size' bytes from file (at 'offset') into buf without moving the
+ * pointer.
+ *
+ * It will return size bytes unless there was an error, in which case it will
+ * return as many as it managed to read (assuming blocking fd's which
+ * all current QEMUFile are)
+ */
 int qemu_peek_buffer(QEMUFile *f, uint8_t *buf, int size, size_t offset)
 {
     int pending;
     int index;
 
     assert(!qemu_file_is_writable(f));
+    assert(offset < IO_BUF_SIZE);
+    assert(size <= IO_BUF_SIZE - offset);
 
+    /* The 1st byte to read from */
     index = f->buf_index + offset;
+    /* The number of available bytes starting at index */
     pending = f->buf_size - index;
-    if (pending < size) {
+
-        qemu_fill_buffer(f);
+    /*
+     * qemu_fill_buffer might return just a few bytes, even when there isn't
+     * an error, so loop collecting them until we get enough.
+     */
+    while (pending < size) {
+        int received = qemu_fill_buffer(f);
+
+        if (received <= 0) {
+            break;
+        }
+
         index = f->buf_index + offset;
         pending = f->buf_size - index;
     }
@@ -711,6 +743,14 @@ int qemu_peek_buffer(QEMUFile *f, uint8_t *buf, int size, size_t offset)
     return size;
 }
 
+/*
+ * Read 'size' bytes of data from the file into buf.
+ * 'size' can be larger than the internal buffer.
+ *
+ * It will return size bytes unless there was an error, in which case it will
+ * return as many as it managed to read (assuming blocking fd's which
+ * all current QEMUFile are)
+ */
 int qemu_get_buffer(QEMUFile *f, uint8_t *buf, int size)
 {
     int pending = size;
@@ -719,7 +759,7 @@ int qemu_get_buffer(QEMUFile *f, uint8_t *buf, int size)
     while (pending > 0) {
         int res;
 
-        res = qemu_peek_buffer(f, buf, pending, 0);
+        res = qemu_peek_buffer(f, buf, MIN(pending, IO_BUF_SIZE), 0);
         if (res == 0) {
             return done;
         }
@@ -731,11 +771,16 @@ int qemu_get_buffer(QEMUFile *f, uint8_t *buf, int size)
     return done;
 }
 
+/*
+ * Peeks a single byte from the buffer; this isn't guaranteed to work if
+ * offset leaves a gap after the previous read/peeked data.
+ */
 int qemu_peek_byte(QEMUFile *f, int offset)
 {
     int index = f->buf_index + offset;
 
     assert(!qemu_file_is_writable(f));
+    assert(offset < IO_BUF_SIZE);
 
     if (index >= f->buf_size) {
         qemu_fill_buffer(f);

qmp-commands.hx

@@ -2967,6 +2967,7 @@ The main json-object contains the following:
             pages. This is just normal pages times size of one page,
             but this way upper levels don't need to care about page
             size (json-int)
+         - "dirty-sync-count": times that dirty ram was synchronized (json-int)
 - "disk": only present if "status" is "active" and it is a block migration,
   it is a json-object with the following disk information:
          - "transferred": amount transferred in bytes (json-int)
@@ -2978,6 +2979,7 @@ The main json-object contains the following:
          - "bytes": number of bytes transferred for XBZRLE compressed pages
          - "pages": number of XBZRLE compressed pages
          - "cache-miss": number of XBRZRLE page cache misses
+         - "cache-miss-rate": rate of XBRZRLE page cache misses
          - "overflow": number of times XBZRLE overflows.  This means
            that the XBZRLE encoding was bigger than just sent the
            whole page, and then we sent the whole page instead (as as
@@ -3004,7 +3006,8 @@ Examples:
           "downtime":12345,
           "duplicate":123,
           "normal":123,
-          "normal-bytes":123456
+          "normal-bytes":123456,
+          "dirty-sync-count":15
         }
      }
    }
@@ -3029,7 +3032,8 @@ Examples:
             "expected-downtime":12345,
             "duplicate":123,
             "normal":123,
-            "normal-bytes":123456
+            "normal-bytes":123456,
+            "dirty-sync-count":15
          }
       }
    }
@@ -3049,7 +3053,8 @@ Examples:
             "expected-downtime":12345,
             "duplicate":123,
             "normal":123,
-            "normal-bytes":123456
+            "normal-bytes":123456,
+            "dirty-sync-count":15
          },
          "disk":{
             "total":20971520,
@@ -3075,13 +3080,15 @@ Examples:
             "expected-downtime":12345,
             "duplicate":10,
             "normal":3333,
-            "normal-bytes":3412992
+            "normal-bytes":3412992,
+            "dirty-sync-count":15
          },
          "xbzrle-cache":{
             "cache-size":67108864,
             "bytes":20971520,
             "pages":2444343,
             "cache-miss":2244,
+            "cache-miss-rate":0.123,
             "overflow":34434
          }
       }

target-arm/machine.c

@@ -248,7 +248,7 @@ const VMStateDescription vmstate_arm_cpu = {
         /* The length-check must come before the arrays to avoid
          * incoming data possibly overflowing the array.
          */
-        VMSTATE_INT32_LE(cpreg_vmstate_array_len, ARMCPU),
+        VMSTATE_INT32_POSITIVE_LE(cpreg_vmstate_array_len, ARMCPU),
         VMSTATE_VARRAY_INT32(cpreg_vmstate_indexes, ARMCPU,
                              cpreg_vmstate_array_len,
                              0, vmstate_info_uint64, uint64_t),

vl.c

@@ -4322,6 +4322,7 @@ int main(int argc, char **argv, char **envp)
     cpu_exec_init_all();
 
     blk_mig_init();
+    ram_mig_init();
 
     /* open the virtual block devices */
     if (snapshot)
@@ -4336,8 +4337,6 @@ int main(int argc, char **argv, char **envp)
     default_drive(default_floppy, snapshot, IF_FLOPPY, 0, FD_OPTS);
     default_drive(default_sdcard, snapshot, IF_SD, 0, SD_OPTS);
 
-    register_savevm_live(NULL, "ram", 0, 4, &savevm_ram_handlers, NULL);
-
     if (nb_numa_nodes > 0) {
         int i;
 

vmstate.c

@@ -10,6 +10,50 @@ static void vmstate_subsection_save(QEMUFile *f, const VMStateDescription *vmsd,
 static int vmstate_subsection_load(QEMUFile *f, const VMStateDescription *vmsd,
                                    void *opaque);
 
+static int vmstate_n_elems(void *opaque, VMStateField *field)
+{
+    int n_elems = 1;
+
+    if (field->flags & VMS_ARRAY) {
+        n_elems = field->num;
+    } else if (field->flags & VMS_VARRAY_INT32) {
+        n_elems = *(int32_t *)(opaque+field->num_offset);
+    } else if (field->flags & VMS_VARRAY_UINT32) {
+        n_elems = *(uint32_t *)(opaque+field->num_offset);
+    } else if (field->flags & VMS_VARRAY_UINT16) {
+        n_elems = *(uint16_t *)(opaque+field->num_offset);
+    } else if (field->flags & VMS_VARRAY_UINT8) {
+        n_elems = *(uint8_t *)(opaque+field->num_offset);
+    }
+
+    return n_elems;
+}
+
+static int vmstate_size(void *opaque, VMStateField *field)
+{
+    int size = field->size;
+
+    if (field->flags & VMS_VBUFFER) {
+        size = *(int32_t *)(opaque+field->size_offset);
+        if (field->flags & VMS_MULTIPLY) {
+            size *= field->size;
+        }
+    }
+
+    return size;
+}
+
+static void *vmstate_base_addr(void *opaque, VMStateField *field)
+{
+    void *base_addr = opaque + field->offset;
+
+    if (field->flags & VMS_POINTER) {
+        base_addr = *(void **)base_addr + field->start;
+    }
+
+    return base_addr;
+}
+
 int vmstate_load_state(QEMUFile *f, const VMStateDescription *vmsd,
                        void *opaque, int version_id)
 {
@@ -19,11 +63,12 @@ int vmstate_load_state(QEMUFile *f, const VMStateDescription *vmsd,
     if (version_id > vmsd->version_id) {
         return -EINVAL;
     }
-    if (version_id < vmsd->minimum_version_id_old) {
-        return -EINVAL;
-    }
     if  (version_id < vmsd->minimum_version_id) {
-        return vmsd->load_state_old(f, opaque, version_id);
+        if (vmsd->load_state_old &&
+            version_id >= vmsd->minimum_version_id_old) {
+            return vmsd->load_state_old(f, opaque, version_id);
+        }
+        return -EINVAL;
     }
     if (vmsd->pre_load) {
         int ret = vmsd->pre_load(opaque);
@@ -36,30 +81,10 @@ int vmstate_load_state(QEMUFile *f, const VMStateDescription *vmsd,
              field->field_exists(opaque, version_id)) ||
             (!field->field_exists &&
              field->version_id <= version_id)) {
-            void *base_addr = opaque + field->offset;
+            void *base_addr = vmstate_base_addr(opaque, field);
-            int i, n_elems = 1;
+            int i, n_elems = vmstate_n_elems(opaque, field);
-            int size = field->size;
+            int size = vmstate_size(opaque, field);
 
-            if (field->flags & VMS_VBUFFER) {
-                size = *(int32_t *)(opaque+field->size_offset);
-                if (field->flags & VMS_MULTIPLY) {
-                    size *= field->size;
-                }
-            }
-            if (field->flags & VMS_ARRAY) {
-                n_elems = field->num;
-            } else if (field->flags & VMS_VARRAY_INT32) {
-                n_elems = *(int32_t *)(opaque+field->num_offset);
-            } else if (field->flags & VMS_VARRAY_UINT32) {
-                n_elems = *(uint32_t *)(opaque+field->num_offset);
-            } else if (field->flags & VMS_VARRAY_UINT16) {
-                n_elems = *(uint16_t *)(opaque+field->num_offset);
-            } else if (field->flags & VMS_VARRAY_UINT8) {
-                n_elems = *(uint8_t *)(opaque+field->num_offset);
-            }
-            if (field->flags & VMS_POINTER) {
-                base_addr = *(void **)base_addr + field->start;
-            }
             for (i = 0; i < n_elems; i++) {
                 void *addr = base_addr + size * i;
 
@@ -78,6 +103,10 @@ int vmstate_load_state(QEMUFile *f, const VMStateDescription *vmsd,
                     return ret;
                 }
             }
+        } else if (field->flags & VMS_MUST_EXIST) {
+            fprintf(stderr, "Input validation failed: %s/%s\n",
+                    vmsd->name, field->name);
+            return -1;
         }
         field++;
     }
@@ -102,30 +131,10 @@ void vmstate_save_state(QEMUFile *f, const VMStateDescription *vmsd,
     while (field->name) {
         if (!field->field_exists ||
             field->field_exists(opaque, vmsd->version_id)) {
-            void *base_addr = opaque + field->offset;
+            void *base_addr = vmstate_base_addr(opaque, field);
-            int i, n_elems = 1;
+            int i, n_elems = vmstate_n_elems(opaque, field);
-            int size = field->size;
+            int size = vmstate_size(opaque, field);
 
-            if (field->flags & VMS_VBUFFER) {
-                size = *(int32_t *)(opaque+field->size_offset);
-                if (field->flags & VMS_MULTIPLY) {
-                    size *= field->size;
-                }
-            }
-            if (field->flags & VMS_ARRAY) {
-                n_elems = field->num;
-            } else if (field->flags & VMS_VARRAY_INT32) {
-                n_elems = *(int32_t *)(opaque+field->num_offset);
-            } else if (field->flags & VMS_VARRAY_UINT32) {
-                n_elems = *(uint32_t *)(opaque+field->num_offset);
-            } else if (field->flags & VMS_VARRAY_UINT16) {
-                n_elems = *(uint16_t *)(opaque+field->num_offset);
-            } else if (field->flags & VMS_VARRAY_UINT8) {
-                n_elems = *(uint8_t *)(opaque+field->num_offset);
-            }
-            if (field->flags & VMS_POINTER) {
-                base_addr = *(void **)base_addr + field->start;
-            }
             for (i = 0; i < n_elems; i++) {
                 void *addr = base_addr + size * i;
 
@@ -138,6 +147,12 @@ void vmstate_save_state(QEMUFile *f, const VMStateDescription *vmsd,
                     field->info->put(f, addr, size);
                 }
             }
+        } else {
+            if (field->flags & VMS_MUST_EXIST) {
+                fprintf(stderr, "Output state validation failed: %s/%s\n",
+                        vmsd->name, field->name);
+                assert(!(field->flags & VMS_MUST_EXIST));
+            }
         }
         field++;
     }
@@ -323,8 +338,9 @@ const VMStateInfo vmstate_info_int32_equal = {
     .put  = put_int32,
 };
 
-/* 32 bit int. Check that the received value is less than or equal to
+/* 32 bit int. Check that the received value is non-negative
-   the one in the field */
+ * and less than or equal to the one in the field.
+ */
 
 static int get_int32_le(QEMUFile *f, void *pv, size_t size)
 {
@@ -332,7 +348,7 @@ static int get_int32_le(QEMUFile *f, void *pv, size_t size)
     int32_t loaded;
     qemu_get_sbe32s(f, &loaded);
 
-    if (loaded <= *cur) {
+    if (loaded >= 0 && loaded <= *cur) {
         *cur = loaded;
         return 0;
     }