Merge remote-tracking branch 'kwolf/tags/for-anthony' into staging

Block patches for 2.0 (flushing block-next)

# gpg: Signature made Fri 29 Nov 2013 08:43:18 AM PST using RSA key ID C88F2FD6
# gpg: Can't check signature: public key not found

# By Peter Lieven (17) and others
# Via Kevin Wolf
* kwolf/tags/for-anthony: (41 commits)
  qemu-iotests: Add sample image and test for VMDK version 3
  vmdk: Allow read only open of VMDK version 3
  qemu-iotests: Filter out 'qemu-io> ' prompt
  qemu-iotests: Filter qemu-io output in 025
  block: Use BDRV_O_NO_BACKING where appropriate
  qemu-iotests: Test snapshot mode
  block: Enable BDRV_O_SNAPSHOT with driver-specific options
  qemu-iotests: Make test case 030, 040 and 055 deterministic
  qemu-iotest: Add pause_drive and resume_drive methods
  blkdebug: add "remove_break" command
  qemu-iotests: Drop local version of cancel_and_wait from 040
  sheepdog: support user-defined redundancy option
  sheepdog: refactor do_sd_create()
  qdict: Optimise qdict_do_flatten()
  qdict: Fix memory leak in qdict_do_flatten()
  MAINTAINERS: add sheepdog development mailing list
  COW: Extend checking allocated bits to beyond one sector
  COW: Speed up writes
  qapi: Change BlockDirtyInfo to list
  block: per caller dirty bitmap
  ...

Message-id: 1385743555-27888-1-git-send-email-kwolf@redhat.com
Signed-off-by: Anthony Liguori <aliguori@amazon.com>

13 files changed, 364 insertions, 137 deletions

diff --git a/block/backup.c b/block/backup.c
index cad14c90b2..0198514043 100644
--- a/block/backup.c
+++ b/block/backup.c
@@ -138,7 +138,8 @@ static int coroutine_fn backup_do_cow(BlockDriverState *bs,
 
         if (buffer_is_zero(iov.iov_base, iov.iov_len)) {
             ret = bdrv_co_write_zeroes(job->target,
-                                       start * BACKUP_SECTORS_PER_CLUSTER, n);
+                                       start * BACKUP_SECTORS_PER_CLUSTER,
+                                       n, BDRV_REQ_MAY_UNMAP);
         } else {
             ret = bdrv_co_writev(job->target,
                                  start * BACKUP_SECTORS_PER_CLUSTER, n,
diff --git a/block/blkdebug.c b/block/blkdebug.c
index 16d2b91ac9..37cf028545 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -605,6 +605,31 @@ static int blkdebug_debug_resume(BlockDriverState *bs, const char *tag)
     return -ENOENT;
 }
 
+static int blkdebug_debug_remove_breakpoint(BlockDriverState *bs,
+                                            const char *tag)
+{
+    BDRVBlkdebugState *s = bs->opaque;
+    BlkdebugSuspendedReq *r;
+    BlkdebugRule *rule, *next;
+    int i, ret = -ENOENT;
+
+    for (i = 0; i < BLKDBG_EVENT_MAX; i++) {
+        QLIST_FOREACH_SAFE(rule, &s->rules[i], next, next) {
+            if (rule->action == ACTION_SUSPEND &&
+                !strcmp(rule->options.suspend.tag, tag)) {
+                remove_rule(rule);
+                ret = 0;
+            }
+        }
+    }
+    QLIST_FOREACH(r, &s->suspended_reqs, next) {
+        if (!strcmp(r->tag, tag)) {
+            qemu_coroutine_enter(r->co, NULL);
+            ret = 0;
+        }
+    }
+    return ret;
+}
 
 static bool blkdebug_debug_is_suspended(BlockDriverState *bs, const char *tag)
 {
@@ -639,6 +664,8 @@ static BlockDriver bdrv_blkdebug = {
 
     .bdrv_debug_event           = blkdebug_debug_event,
     .bdrv_debug_breakpoint      = blkdebug_debug_breakpoint,
+    .bdrv_debug_remove_breakpoint
+                                = blkdebug_debug_remove_breakpoint,
     .bdrv_debug_resume          = blkdebug_debug_resume,
     .bdrv_debug_is_suspended    = blkdebug_debug_is_suspended,
 };
diff --git a/block/cow.c b/block/cow.c
index 909c3e7182..dc15e46b6c 100644
--- a/block/cow.c
+++ b/block/cow.c
@@ -103,40 +103,18 @@ static int cow_open(BlockDriverState *bs, QDict *options, int flags,
     return ret;
 }
 
-/*
- * XXX(hch): right now these functions are extremely inefficient.
- * We should just read the whole bitmap we'll need in one go instead.
- */
-static inline int cow_set_bit(BlockDriverState *bs, int64_t bitnum, bool *first)
+static inline void cow_set_bits(uint8_t *bitmap, int start, int64_t nb_sectors)
 {
-    uint64_t offset = sizeof(struct cow_header_v2) + bitnum / 8;
-    uint8_t bitmap;
-    int ret;
-
-    ret = bdrv_pread(bs->file, offset, &bitmap, sizeof(bitmap));
-    if (ret < 0) {
-       return ret;
-    }
-
-    if (bitmap & (1 << (bitnum % 8))) {
-        return 0;
-    }
-
-    if (*first) {
-        ret = bdrv_flush(bs->file);
-        if (ret < 0) {
-            return ret;
+    int64_t bitnum = start, last = start + nb_sectors;
+    while (bitnum < last) {
+        if ((bitnum & 7) == 0 && bitnum + 8 <= last) {
+            bitmap[bitnum / 8] = 0xFF;
+            bitnum += 8;
+            continue;
         }
-        *first = false;
+        bitmap[bitnum/8] |= (1 << (bitnum % 8));
+        bitnum++;
     }
-
-    bitmap |= (1 << (bitnum % 8));
-
-    ret = bdrv_pwrite(bs->file, offset, &bitmap, sizeof(bitmap));
-    if (ret < 0) {
-       return ret;
-    }
-    return 0;
 }
 
 #define BITS_PER_BITMAP_SECTOR (512 * 8)
@@ -174,18 +152,34 @@ static int coroutine_fn cow_co_is_allocated(BlockDriverState *bs,
 {
     int64_t bitnum = sector_num + sizeof(struct cow_header_v2) * 8;
     uint64_t offset = (bitnum / 8) & -BDRV_SECTOR_SIZE;
-    uint8_t bitmap[BDRV_SECTOR_SIZE];
-    int ret;
-    int changed;
+    bool first = true;
+    int changed = 0, same = 0;
 
-    ret = bdrv_pread(bs->file, offset, &bitmap, sizeof(bitmap));
-    if (ret < 0) {
-        return ret;
-    }
+    do {
+        int ret;
+        uint8_t bitmap[BDRV_SECTOR_SIZE];
+
+        bitnum &= BITS_PER_BITMAP_SECTOR - 1;
+        int sector_bits = MIN(nb_sectors, BITS_PER_BITMAP_SECTOR - bitnum);
+
+        ret = bdrv_pread(bs->file, offset, &bitmap, sizeof(bitmap));
+        if (ret < 0) {
+            return ret;
+        }
+
+        if (first) {
+            changed = cow_test_bit(bitnum, bitmap);
+            first = false;
+        }
+
+        same += cow_find_streak(bitmap, changed, bitnum, nb_sectors);
+
+        bitnum += sector_bits;
+        nb_sectors -= sector_bits;
+        offset += BDRV_SECTOR_SIZE;
+    } while (nb_sectors);
 
-    bitnum &= BITS_PER_BITMAP_SECTOR - 1;
-    changed = cow_test_bit(bitnum, bitmap);
-    *num_same = cow_find_streak(bitmap, changed, bitnum, nb_sectors);
+    *num_same = same;
     return changed;
 }
 
@@ -204,18 +198,52 @@ static int64_t coroutine_fn cow_co_get_block_status(BlockDriverState *bs,
 static int cow_update_bitmap(BlockDriverState *bs, int64_t sector_num,
         int nb_sectors)
 {
-    int error = 0;
-    int i;
+    int64_t bitnum = sector_num + sizeof(struct cow_header_v2) * 8;
+    uint64_t offset = (bitnum / 8) & -BDRV_SECTOR_SIZE;
     bool first = true;
+    int sector_bits;
+
+    for ( ; nb_sectors;
+            bitnum += sector_bits,
+            nb_sectors -= sector_bits,
+            offset += BDRV_SECTOR_SIZE) {
+        int ret, set;
+        uint8_t bitmap[BDRV_SECTOR_SIZE];
+
+        bitnum &= BITS_PER_BITMAP_SECTOR - 1;
+        sector_bits = MIN(nb_sectors, BITS_PER_BITMAP_SECTOR - bitnum);
 
-    for (i = 0; i < nb_sectors; i++) {
-        error = cow_set_bit(bs, sector_num + i, &first);
-        if (error) {
-            break;
+        ret = bdrv_pread(bs->file, offset, &bitmap, sizeof(bitmap));
+        if (ret < 0) {
+            return ret;
+        }
+
+        /* Skip over any already set bits */
+        set = cow_find_streak(bitmap, 1, bitnum, sector_bits);
+        bitnum += set;
+        sector_bits -= set;
+        nb_sectors -= set;
+        if (!sector_bits) {
+            continue;
+        }
+
+        if (first) {
+            ret = bdrv_flush(bs->file);
+            if (ret < 0) {
+                return ret;
+            }
+            first = false;
+        }
+
+        cow_set_bits(bitmap, bitnum, sector_bits);
+
+        ret = bdrv_pwrite(bs->file, offset, &bitmap, sizeof(bitmap));
+        if (ret < 0) {
+            return ret;
         }
     }
 
-    return error;
+    return 0;
 }
 
 static int coroutine_fn cow_read(BlockDriverState *bs, int64_t sector_num,
diff --git a/block/iscsi.c b/block/iscsi.c
index a2d578c0a7..b7b52381d6 100644
--- a/block/iscsi.c
+++ b/block/iscsi.c
@@ -56,6 +56,7 @@ typedef struct IscsiLun {
     uint8_t lbprz;
     struct scsi_inquiry_logical_block_provisioning lbp;
     struct scsi_inquiry_block_limits bl;
+    unsigned char *zeroblock;
 } IscsiLun;
 
 typedef struct IscsiTask {
@@ -87,7 +88,6 @@ typedef struct IscsiAIOCB {
 #define NOP_INTERVAL 5000
 #define MAX_NOP_FAILURES 3
 #define ISCSI_CMD_RETRIES 5
-#define ISCSI_MAX_UNMAP 131072
 
 static void
 iscsi_bh_cb(void *p)
@@ -912,8 +912,6 @@ coroutine_fn iscsi_co_discard(BlockDriverState *bs, int64_t sector_num,
     IscsiLun *iscsilun = bs->opaque;
     struct IscsiTask iTask;
     struct unmap_list list;
-    uint32_t nb_blocks;
-    uint32_t max_unmap;
 
     if (!is_request_lun_aligned(sector_num, nb_sectors, iscsilun)) {
         return -EINVAL;
@@ -925,57 +923,102 @@ coroutine_fn iscsi_co_discard(BlockDriverState *bs, int64_t sector_num,
     }
 
     list.lba = sector_qemu2lun(sector_num, iscsilun);
-    nb_blocks = sector_qemu2lun(nb_sectors, iscsilun);
+    list.num = sector_qemu2lun(nb_sectors, iscsilun);
 
-    max_unmap = iscsilun->bl.max_unmap;
-    if (max_unmap == 0xffffffff) {
-        max_unmap = ISCSI_MAX_UNMAP;
+    iscsi_co_init_iscsitask(iscsilun, &iTask);
+retry:
+    if (iscsi_unmap_task(iscsilun->iscsi, iscsilun->lun, 0, 0, &list, 1,
+                     iscsi_co_generic_cb, &iTask) == NULL) {
+        return -EIO;
     }
 
-    while (nb_blocks > 0) {
-        iscsi_co_init_iscsitask(iscsilun, &iTask);
-        list.num = nb_blocks;
-        if (list.num > max_unmap) {
-            list.num = max_unmap;
-        }
-retry:
-        if (iscsi_unmap_task(iscsilun->iscsi, iscsilun->lun, 0, 0, &list, 1,
-                         iscsi_co_generic_cb, &iTask) == NULL) {
-            return -EIO;
-        }
+    while (!iTask.complete) {
+        iscsi_set_events(iscsilun);
+        qemu_coroutine_yield();
+    }
 
-        while (!iTask.complete) {
-            iscsi_set_events(iscsilun);
-            qemu_coroutine_yield();
-        }
+    if (iTask.task != NULL) {
+        scsi_free_scsi_task(iTask.task);
+        iTask.task = NULL;
+    }
 
-        if (iTask.task != NULL) {
-            scsi_free_scsi_task(iTask.task);
-            iTask.task = NULL;
-        }
+    if (iTask.do_retry) {
+        goto retry;
+    }
 
-        if (iTask.do_retry) {
-            goto retry;
-        }
+    if (iTask.status == SCSI_STATUS_CHECK_CONDITION) {
+        /* the target might fail with a check condition if it
+           is not happy with the alignment of the UNMAP request
+           we silently fail in this case */
+        return 0;
+    }
 
-        if (iTask.status == SCSI_STATUS_CHECK_CONDITION) {
-            /* the target might fail with a check condition if it
-               is not happy with the alignment of the UNMAP request
-               we silently fail in this case */
-            return 0;
-        }
+    if (iTask.status != SCSI_STATUS_GOOD) {
+        return -EIO;
+    }
 
-        if (iTask.status != SCSI_STATUS_GOOD) {
-            return -EIO;
-        }
+    return 0;
+}
+
+#if defined(SCSI_SENSE_ASCQ_CAPACITY_DATA_HAS_CHANGED)
+
+static int
+coroutine_fn iscsi_co_write_zeroes(BlockDriverState *bs, int64_t sector_num,
+                                   int nb_sectors, BdrvRequestFlags flags)
+{
+    IscsiLun *iscsilun = bs->opaque;
+    struct IscsiTask iTask;
+    uint64_t lba;
+    uint32_t nb_blocks;
 
-        list.lba += list.num;
-        nb_blocks -= list.num;
+    if (!is_request_lun_aligned(sector_num, nb_sectors, iscsilun)) {
+        return -EINVAL;
+    }
+
+    if (!iscsilun->lbp.lbpws) {
+        /* WRITE SAME is not supported by the target */
+        return -ENOTSUP;
+    }
+
+    lba = sector_qemu2lun(sector_num, iscsilun);
+    nb_blocks = sector_qemu2lun(nb_sectors, iscsilun);
+
+    if (iscsilun->zeroblock == NULL) {
+        iscsilun->zeroblock = g_malloc0(iscsilun->block_size);
+    }
+
+    iscsi_co_init_iscsitask(iscsilun, &iTask);
+retry:
+    if (iscsi_writesame16_task(iscsilun->iscsi, iscsilun->lun, lba,
+                               iscsilun->zeroblock, iscsilun->block_size,
+                               nb_blocks, 0, !!(flags & BDRV_REQ_MAY_UNMAP),
+                               0, 0, iscsi_co_generic_cb, &iTask) == NULL) {
+        return -EIO;
+    }
+
+    while (!iTask.complete) {
+        iscsi_set_events(iscsilun);
+        qemu_coroutine_yield();
+    }
+
+    if (iTask.task != NULL) {
+        scsi_free_scsi_task(iTask.task);
+        iTask.task = NULL;
+    }
+
+    if (iTask.do_retry) {
+        goto retry;
+    }
+
+    if (iTask.status != SCSI_STATUS_GOOD) {
+        return -EIO;
     }
 
     return 0;
 }
 
+#endif /* SCSI_SENSE_ASCQ_CAPACITY_DATA_HAS_CHANGED */
+
 static int parse_chap(struct iscsi_context *iscsi, const char *target)
 {
     QemuOptsList *list;
@@ -1384,6 +1427,20 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
                sizeof(struct scsi_inquiry_block_limits));
         scsi_free_scsi_task(task);
         task = NULL;
+
+        if (iscsilun->bl.max_unmap < 0xffffffff) {
+            bs->bl.max_discard = sector_lun2qemu(iscsilun->bl.max_unmap,
+                                                 iscsilun);
+        }
+        bs->bl.discard_alignment = sector_lun2qemu(iscsilun->bl.opt_unmap_gran,
+                                                   iscsilun);
+
+        if (iscsilun->bl.max_ws_len < 0xffffffff) {
+            bs->bl.max_write_zeroes = sector_lun2qemu(iscsilun->bl.max_ws_len,
+                                                      iscsilun);
+        }
+        bs->bl.write_zeroes_alignment = sector_lun2qemu(iscsilun->bl.opt_unmap_gran,
+                                                        iscsilun);
     }
 
 #if defined(LIBISCSI_FEATURE_NOP_COUNTER)
@@ -1424,6 +1481,7 @@ static void iscsi_close(BlockDriverState *bs)
     }
     qemu_aio_set_fd_handler(iscsi_get_fd(iscsi), NULL, NULL, NULL);
     iscsi_destroy_context(iscsi);
+    g_free(iscsilun->zeroblock);
     memset(iscsilun, 0, sizeof(IscsiLun));
 }
 
@@ -1506,6 +1564,14 @@ out:
     return ret;
 }
 
+static int iscsi_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
+{
+    IscsiLun *iscsilun = bs->opaque;
+    bdi->unallocated_blocks_are_zero = !!iscsilun->lbprz;
+    bdi->can_write_zeroes_with_unmap = iscsilun->lbprz && iscsilun->lbp.lbpws;
+    return 0;
+}
+
 static QEMUOptionParameter iscsi_create_options[] = {
     {
         .name = BLOCK_OPT_SIZE,
@@ -1527,12 +1593,16 @@ static BlockDriver bdrv_iscsi = {
     .create_options  = iscsi_create_options,
 
     .bdrv_getlength  = iscsi_getlength,
+    .bdrv_get_info   = iscsi_get_info,
     .bdrv_truncate   = iscsi_truncate,
 
 #if defined(LIBISCSI_FEATURE_IOVECTOR)
     .bdrv_co_get_block_status = iscsi_co_get_block_status,
 #endif
     .bdrv_co_discard      = iscsi_co_discard,
+#if defined(SCSI_SENSE_ASCQ_CAPACITY_DATA_HAS_CHANGED)
+    .bdrv_co_write_zeroes = iscsi_co_write_zeroes,
+#endif
 
     .bdrv_aio_readv  = iscsi_aio_readv,
     .bdrv_aio_writev = iscsi_aio_writev,
diff --git a/block/mirror.c b/block/mirror.c
index 7b95acf88c..6dc27ad35d 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -39,6 +39,7 @@ typedef struct MirrorBlockJob {
     int64_t granularity;
     size_t buf_size;
     unsigned long *cow_bitmap;
+    BdrvDirtyBitmap *dirty_bitmap;
     HBitmapIter hbi;
     uint8_t *buf;
     QSIMPLEQ_HEAD(, MirrorBuffer) buf_free;
@@ -145,9 +146,10 @@ static void coroutine_fn mirror_iteration(MirrorBlockJob *s)
 
     s->sector_num = hbitmap_iter_next(&s->hbi);
     if (s->sector_num < 0) {
-        bdrv_dirty_iter_init(source, &s->hbi);
+        bdrv_dirty_iter_init(source, s->dirty_bitmap, &s->hbi);
         s->sector_num = hbitmap_iter_next(&s->hbi);
-        trace_mirror_restart_iter(s, bdrv_get_dirty_count(source));
+        trace_mirror_restart_iter(s,
+                                  bdrv_get_dirty_count(source, s->dirty_bitmap));
         assert(s->sector_num >= 0);
     }
 
@@ -183,7 +185,7 @@ static void coroutine_fn mirror_iteration(MirrorBlockJob *s)
     do {
         int added_sectors, added_chunks;
 
-        if (!bdrv_get_dirty(source, next_sector) ||
+        if (!bdrv_get_dirty(source, s->dirty_bitmap, next_sector) ||
             test_bit(next_chunk, s->in_flight_bitmap)) {
             assert(nb_sectors > 0);
             break;
@@ -249,7 +251,8 @@ static void coroutine_fn mirror_iteration(MirrorBlockJob *s)
         /* Advance the HBitmapIter in parallel, so that we do not examine
          * the same sector twice.
          */
-        if (next_sector > hbitmap_next_sector && bdrv_get_dirty(source, next_sector)) {
+        if (next_sector > hbitmap_next_sector
+            && bdrv_get_dirty(source, s->dirty_bitmap, next_sector)) {
             hbitmap_next_sector = hbitmap_iter_next(&s->hbi);
         }
 
@@ -355,7 +358,7 @@ static void coroutine_fn mirror_run(void *opaque)
         }
     }
 
-    bdrv_dirty_iter_init(bs, &s->hbi);
+    bdrv_dirty_iter_init(bs, s->dirty_bitmap, &s->hbi);
     last_pause_ns = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
     for (;;) {
         uint64_t delay_ns;
@@ -367,7 +370,7 @@ static void coroutine_fn mirror_run(void *opaque)
             goto immediate_exit;
         }
 
-        cnt = bdrv_get_dirty_count(bs);
+        cnt = bdrv_get_dirty_count(bs, s->dirty_bitmap);
 
         /* Note that even when no rate limit is applied we need to yield
          * periodically with no pending I/O so that qemu_aio_flush() returns.
@@ -409,7 +412,7 @@ static void coroutine_fn mirror_run(void *opaque)
 
                 should_complete = s->should_complete ||
                     block_job_is_cancelled(&s->common);
-                cnt = bdrv_get_dirty_count(bs);
+                cnt = bdrv_get_dirty_count(bs, s->dirty_bitmap);
             }
         }
 
@@ -424,7 +427,7 @@ static void coroutine_fn mirror_run(void *opaque)
              */
             trace_mirror_before_drain(s, cnt);
             bdrv_drain_all();
-            cnt = bdrv_get_dirty_count(bs);
+            cnt = bdrv_get_dirty_count(bs, s->dirty_bitmap);
         }
 
         ret = 0;
@@ -471,7 +474,7 @@ immediate_exit:
     qemu_vfree(s->buf);
     g_free(s->cow_bitmap);
     g_free(s->in_flight_bitmap);
-    bdrv_set_dirty_tracking(bs, 0);
+    bdrv_release_dirty_bitmap(bs, s->dirty_bitmap);
     bdrv_iostatus_disable(s->target);
     if (s->should_complete && ret == 0) {
         if (bdrv_get_flags(s->target) != bdrv_get_flags(s->common.bs)) {
@@ -575,7 +578,7 @@ void mirror_start(BlockDriverState *bs, BlockDriverState *target,
     s->granularity = granularity;
     s->buf_size = MAX(buf_size, granularity);
 
-    bdrv_set_dirty_tracking(bs, granularity);
+    s->dirty_bitmap = bdrv_create_dirty_bitmap(bs, granularity);
     bdrv_set_enable_write_cache(s->target, true);
     bdrv_set_on_error(s->target, on_target_error, on_target_error);
     bdrv_iostatus_enable(s->target);
diff --git a/block/qapi.c b/block/qapi.c
index 5880b3e42b..a32cb79db8 100644
--- a/block/qapi.c
+++ b/block/qapi.c
@@ -204,12 +204,9 @@ void bdrv_query_info(BlockDriverState *bs,
         info->io_status = bs->iostatus;
     }
 
-    if (bs->dirty_bitmap) {
-        info->has_dirty = true;
-        info->dirty = g_malloc0(sizeof(*info->dirty));
-        info->dirty->count = bdrv_get_dirty_count(bs) * BDRV_SECTOR_SIZE;
-        info->dirty->granularity =
-         ((int64_t) BDRV_SECTOR_SIZE << hbitmap_granularity(bs->dirty_bitmap));
+    if (!QLIST_EMPTY(&bs->dirty_bitmaps)) {
+        info->has_dirty_bitmaps = true;
+        info->dirty_bitmaps = bdrv_query_dirty_bitmaps(bs);
     }
 
     if (bs->drv) {
diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
index 791083a0ef..11f9c50aa7 100644
--- a/block/qcow2-cluster.c
+++ b/block/qcow2-cluster.c
@@ -1613,7 +1613,7 @@ static int expand_zero_clusters_in_l1(BlockDriverState *bs, uint64_t *l1_table,
             }
 
             ret = bdrv_write_zeroes(bs->file, offset / BDRV_SECTOR_SIZE,
-                                    s->cluster_sectors);
+                                    s->cluster_sectors, 0);
             if (ret < 0) {
                 if (!preallocated) {
                     qcow2_free_clusters(bs, offset, s->cluster_size,
diff --git a/block/qcow2.c b/block/qcow2.c
index 6e5d98dc48..8e2b6c7548 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -1588,7 +1588,8 @@ static int qcow2_create2(const char *filename, int64_t total_size,
 
     /* Reopen the image without BDRV_O_NO_FLUSH to flush it before returning */
     ret = bdrv_open(bs, filename, NULL,
-                    BDRV_O_RDWR | BDRV_O_CACHE_WB, drv, &local_err);
+                    BDRV_O_RDWR | BDRV_O_CACHE_WB | BDRV_O_NO_BACKING,
+                    drv, &local_err);
     if (error_is_set(&local_err)) {
         error_propagate(errp, local_err);
         goto out;
@@ -1696,7 +1697,7 @@ static int qcow2_make_empty(BlockDriverState *bs)
 }
 
 static coroutine_fn int qcow2_co_write_zeroes(BlockDriverState *bs,
-    int64_t sector_num, int nb_sectors)
+    int64_t sector_num, int nb_sectors, BdrvRequestFlags flags)
 {
     int ret;
     BDRVQcowState *s = bs->opaque;
diff --git a/block/qed.c b/block/qed.c
index 6c0cba04f3..adc2736dd7 100644
--- a/block/qed.c
+++ b/block/qed.c
@@ -1397,7 +1397,8 @@ static void coroutine_fn qed_co_write_zeroes_cb(void *opaque, int ret)
 
 static int coroutine_fn bdrv_qed_co_write_zeroes(BlockDriverState *bs,
                                                  int64_t sector_num,
-                                                 int nb_sectors)
+                                                 int nb_sectors,
+                                                 BdrvRequestFlags flags)
 {
     BlockDriverAIOCB *blockacb;
     BDRVQEDState *s = bs->opaque;
diff --git a/block/raw_bsd.c b/block/raw_bsd.c
index 2265dcc03f..978ae7a102 100644
--- a/block/raw_bsd.c
+++ b/block/raw_bsd.c
@@ -68,9 +68,10 @@ static int64_t coroutine_fn raw_co_get_block_status(BlockDriverState *bs,
 }
 
 static int coroutine_fn raw_co_write_zeroes(BlockDriverState *bs,
-                                            int64_t sector_num, int nb_sectors)
+                                            int64_t sector_num, int nb_sectors,
+                                            BdrvRequestFlags flags)
 {
-    return bdrv_co_write_zeroes(bs->file, sector_num, nb_sectors);
+    return bdrv_co_write_zeroes(bs->file, sector_num, nb_sectors, flags);
 }
 
 static int coroutine_fn raw_co_discard(BlockDriverState *bs,
@@ -149,6 +150,7 @@ static int raw_open(BlockDriverState *bs, QDict *options, int flags,
                     Error **errp)
 {
     bs->sg = bs->file->sg;
+    bs->bl = bs->file->bl;
     return 0;
 }
 
diff --git a/block/sheepdog.c b/block/sheepdog.c
index ef387de71f..b4ae50f44d 100644
--- a/block/sheepdog.c
+++ b/block/sheepdog.c
@@ -91,6 +91,14 @@
 #define SD_NR_VDIS   (1U << 24)
 #define SD_DATA_OBJ_SIZE (UINT64_C(1) << 22)
 #define SD_MAX_VDI_SIZE (SD_DATA_OBJ_SIZE * MAX_DATA_OBJS)
+/*
+ * For erasure coding, we use at most SD_EC_MAX_STRIP for data strips and
+ * (SD_EC_MAX_STRIP - 1) for parity strips
+ *
+ * SD_MAX_COPIES is sum of number of data strips and parity strips.
+ */
+#define SD_EC_MAX_STRIP 16
+#define SD_MAX_COPIES (SD_EC_MAX_STRIP * 2 - 1)
 
 #define SD_INODE_SIZE (sizeof(SheepdogInode))
 #define CURRENT_VDI_ID 0
@@ -1464,9 +1472,7 @@ out:
     return ret;
 }
 
-static int do_sd_create(BDRVSheepdogState *s, char *filename, int64_t vdi_size,
-                        uint32_t base_vid, uint32_t *vdi_id, int snapshot,
-                        uint8_t copy_policy)
+static int do_sd_create(BDRVSheepdogState *s, uint32_t *vdi_id, int snapshot)
 {
     SheepdogVdiReq hdr;
     SheepdogVdiRsp *rsp = (SheepdogVdiRsp *)&hdr;
@@ -1483,11 +1489,11 @@ static int do_sd_create(BDRVSheepdogState *s, char *filename, int64_t vdi_size,
      * does not fit in buf?  For now, just truncate and avoid buffer overrun.
      */
     memset(buf, 0, sizeof(buf));
-    pstrcpy(buf, sizeof(buf), filename);
+    pstrcpy(buf, sizeof(buf), s->name);
 
     memset(&hdr, 0, sizeof(hdr));
     hdr.opcode = SD_OP_NEW_VDI;
-    hdr.vdi_id = base_vid;
+    hdr.vdi_id = s->inode.vdi_id;
 
     wlen = SD_MAX_VDI_LEN;
 
@@ -1495,8 +1501,9 @@ static int do_sd_create(BDRVSheepdogState *s, char *filename, int64_t vdi_size,
     hdr.snapid = snapshot;
 
     hdr.data_length = wlen;
-    hdr.vdi_size = vdi_size;
-    hdr.copy_policy = copy_policy;
+    hdr.vdi_size = s->inode.vdi_size;
+    hdr.copy_policy = s->inode.copy_policy;
+    hdr.copies = s->inode.nr_copies;
 
     ret = do_req(fd, (SheepdogReq *)&hdr, buf, &wlen, &rlen);
 
@@ -1507,7 +1514,7 @@ static int do_sd_create(BDRVSheepdogState *s, char *filename, int64_t vdi_size,
     }
 
     if (rsp->result != SD_RES_SUCCESS) {
-        error_report("%s, %s", sd_strerror(rsp->result), filename);
+        error_report("%s, %s", sd_strerror(rsp->result), s->inode.name);
         return -EIO;
     }
 
@@ -1564,27 +1571,79 @@ out:
     return ret;
 }
 
+/*
+ * Sheepdog support two kinds of redundancy, full replication and erasure
+ * coding.
+ *
+ * # create a fully replicated vdi with x copies
+ * -o redundancy=x (1 <= x <= SD_MAX_COPIES)
+ *
+ * # create a erasure coded vdi with x data strips and y parity strips
+ * -o redundancy=x:y (x must be one of {2,4,8,16} and 1 <= y < SD_EC_MAX_STRIP)
+ */
+static int parse_redundancy(BDRVSheepdogState *s, const char *opt)
+{
+    struct SheepdogInode *inode = &s->inode;
+    const char *n1, *n2;
+    long copy, parity;
+    char p[10];
+
+    pstrcpy(p, sizeof(p), opt);
+    n1 = strtok(p, ":");
+    n2 = strtok(NULL, ":");
+
+    if (!n1) {
+        return -EINVAL;
+    }
+
+    copy = strtol(n1, NULL, 10);
+    if (copy > SD_MAX_COPIES || copy < 1) {
+        return -EINVAL;
+    }
+    if (!n2) {
+        inode->copy_policy = 0;
+        inode->nr_copies = copy;
+        return 0;
+    }
+
+    if (copy != 2 && copy != 4 && copy != 8 && copy != 16) {
+        return -EINVAL;
+    }
+
+    parity = strtol(n2, NULL, 10);
+    if (parity >= SD_EC_MAX_STRIP || parity < 1) {
+        return -EINVAL;
+    }
+
+    /*
+     * 4 bits for parity and 4 bits for data.
+     * We have to compress upper data bits because it can't represent 16
+     */
+    inode->copy_policy = ((copy / 2) << 4) + parity;
+    inode->nr_copies = copy + parity;
+
+    return 0;
+}
+
 static int sd_create(const char *filename, QEMUOptionParameter *options,
                      Error **errp)
 {
     int ret = 0;
-    uint32_t vid = 0, base_vid = 0;
-    int64_t vdi_size = 0;
+    uint32_t vid = 0;
     char *backing_file = NULL;
     BDRVSheepdogState *s;
-    char vdi[SD_MAX_VDI_LEN], tag[SD_MAX_VDI_TAG_LEN];
+    char tag[SD_MAX_VDI_TAG_LEN];
     uint32_t snapid;
     bool prealloc = false;
     Error *local_err = NULL;
 
     s = g_malloc0(sizeof(BDRVSheepdogState));
 
-    memset(vdi, 0, sizeof(vdi));
     memset(tag, 0, sizeof(tag));
     if (strstr(filename, "://")) {
-        ret = sd_parse_uri(s, filename, vdi, &snapid, tag);
+        ret = sd_parse_uri(s, filename, s->name, &snapid, tag);
     } else {
-        ret = parse_vdiname(s, filename, vdi, &snapid, tag);
+        ret = parse_vdiname(s, filename, s->name, &snapid, tag);
     }
     if (ret < 0) {
         goto out;
@@ -1592,7 +1651,7 @@ static int sd_create(const char *filename, QEMUOptionParameter *options,
 
     while (options && options->name) {
         if (!strcmp(options->name, BLOCK_OPT_SIZE)) {
-            vdi_size = options->value.n;
+            s->inode.vdi_size = options->value.n;
         } else if (!strcmp(options->name, BLOCK_OPT_BACKING_FILE)) {
             backing_file = options->value.s;
         } else if (!strcmp(options->name, BLOCK_OPT_PREALLOC)) {
@@ -1606,11 +1665,16 @@ static int sd_create(const char *filename, QEMUOptionParameter *options,
                 ret = -EINVAL;
                 goto out;
             }
+        } else if (!strcmp(options->name, BLOCK_OPT_REDUNDANCY)) {
+            ret = parse_redundancy(s, options->value.s);
+            if (ret < 0) {
+                goto out;
+            }
         }
         options++;
     }
 
-    if (vdi_size > SD_MAX_VDI_SIZE) {
+    if (s->inode.vdi_size > SD_MAX_VDI_SIZE) {
         error_report("too big image size");
         ret = -EINVAL;
         goto out;
@@ -1645,12 +1709,10 @@ static int sd_create(const char *filename, QEMUOptionParameter *options,
             goto out;
         }
 
-        base_vid = s->inode.vdi_id;
         bdrv_unref(bs);
     }
 
-    /* TODO: allow users to specify copy number */
-    ret = do_sd_create(s, vdi, vdi_size, base_vid, &vid, 0, 0);
+    ret = do_sd_create(s, &vid, 0);
     if (!prealloc || ret) {
         goto out;
     }
@@ -1833,8 +1895,7 @@ static int sd_create_branch(BDRVSheepdogState *s)
      * false bail out.
      */
     deleted = sd_delete(s);
-    ret = do_sd_create(s, s->name, s->inode.vdi_size, s->inode.vdi_id, &vid,
-                       !deleted, s->inode.copy_policy);
+    ret = do_sd_create(s, &vid, !deleted);
     if (ret) {
         goto out;
     }
@@ -2097,8 +2158,7 @@ static int sd_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)
         goto cleanup;
     }
 
-    ret = do_sd_create(s, s->name, s->inode.vdi_size, s->inode.vdi_id, &new_vid,
-                       1, s->inode.copy_policy);
+    ret = do_sd_create(s, &new_vid, 1);
     if (ret < 0) {
         error_report("failed to create inode for snapshot. %s",
                      strerror(errno));
@@ -2407,6 +2467,22 @@ sd_co_get_block_status(BlockDriverState *bs, int64_t sector_num, int nb_sectors,
     return ret;
 }
 
+static int64_t sd_get_allocated_file_size(BlockDriverState *bs)
+{
+    BDRVSheepdogState *s = bs->opaque;
+    SheepdogInode *inode = &s->inode;
+    unsigned long i, last = DIV_ROUND_UP(inode->vdi_size, SD_DATA_OBJ_SIZE);
+    uint64_t size = 0;
+
+    for (i = 0; i < last; i++) {
+        if (inode->data_vdi_id[i] == 0) {
+            continue;
+        }
+        size += SD_DATA_OBJ_SIZE;
+    }
+    return size;
+}
+
 static QEMUOptionParameter sd_create_options[] = {
     {
         .name = BLOCK_OPT_SIZE,
@@ -2423,6 +2499,11 @@ static QEMUOptionParameter sd_create_options[] = {
         .type = OPT_STRING,
         .help = "Preallocation mode (allowed values: off, full)"
     },
+    {
+        .name = BLOCK_OPT_REDUNDANCY,
+        .type = OPT_STRING,
+        .help = "Redundancy of the image"
+    },
     { NULL }
 };
 
@@ -2436,6 +2517,7 @@ static BlockDriver bdrv_sheepdog = {
     .bdrv_create    = sd_create,
     .bdrv_has_zero_init = bdrv_has_zero_init_1,
     .bdrv_getlength = sd_getlength,
+    .bdrv_get_allocated_file_size = sd_get_allocated_file_size,
     .bdrv_truncate  = sd_truncate,
 
     .bdrv_co_readv  = sd_co_readv,
@@ -2465,6 +2547,7 @@ static BlockDriver bdrv_sheepdog_tcp = {
     .bdrv_create    = sd_create,
     .bdrv_has_zero_init = bdrv_has_zero_init_1,
     .bdrv_getlength = sd_getlength,
+    .bdrv_get_allocated_file_size = sd_get_allocated_file_size,
     .bdrv_truncate  = sd_truncate,
 
     .bdrv_co_readv  = sd_co_readv,
@@ -2494,6 +2577,7 @@ static BlockDriver bdrv_sheepdog_unix = {
     .bdrv_create    = sd_create,
     .bdrv_has_zero_init = bdrv_has_zero_init_1,
     .bdrv_getlength = sd_getlength,
+    .bdrv_get_allocated_file_size = sd_get_allocated_file_size,
     .bdrv_truncate  = sd_truncate,
 
     .bdrv_co_readv  = sd_co_readv,
diff --git a/block/stream.c b/block/stream.c
index 694fd42e41..46bec7d379 100644
--- a/block/stream.c
+++ b/block/stream.c
@@ -88,6 +88,11 @@ static void coroutine_fn stream_run(void *opaque)
     int n = 0;
     void *buf;
 
+    if (!bs->backing_hd) {
+        block_job_completed(&s->common, 0);
+        return;
+    }
+
     s->common.len = bdrv_getlength(bs);
     if (s->common.len < 0) {
         block_job_completed(&s->common, s->common.len);
diff --git a/block/vmdk.c b/block/vmdk.c
index a7ebd0f125..88d09e3e16 100644
--- a/block/vmdk.c
+++ b/block/vmdk.c
@@ -605,13 +605,20 @@ static int vmdk_open_vmdk4(BlockDriverState *bs,
         header = footer.header;
     }
 
-    if (le32_to_cpu(header.version) >= 3) {
+    if (le32_to_cpu(header.version) > 3) {
         char buf[64];
         snprintf(buf, sizeof(buf), "VMDK version %d",
                  le32_to_cpu(header.version));
         qerror_report(QERR_UNKNOWN_BLOCK_FORMAT_FEATURE,
                 bs->device_name, "vmdk", buf);
         return -ENOTSUP;
+    } else if (le32_to_cpu(header.version) == 3 && (flags & BDRV_O_RDWR)) {
+        /* VMware KB 2064959 explains that version 3 added support for
+         * persistent changed block tracking (CBT), and backup software can
+         * read it as version=1 if it doesn't care about the changed area
+         * information. So we are safe to enable read only. */
+        error_setg(errp, "VMDK version 3 must be read only");
+        return -EINVAL;
     }
 
     if (le32_to_cpu(header.num_gtes_per_gt) > 512) {
@@ -1419,7 +1426,8 @@ static coroutine_fn int vmdk_co_write(BlockDriverState *bs, int64_t sector_num,
 
 static int coroutine_fn vmdk_co_write_zeroes(BlockDriverState *bs,
                                              int64_t sector_num,
-                                             int nb_sectors)
+                                             int nb_sectors,
+                                             BdrvRequestFlags flags)
 {
     int ret;
     BDRVVmdkState *s = bs->opaque;
@@ -1689,7 +1697,7 @@ static int vmdk_create(const char *filename, QEMUOptionParameter *options,
     }
     if (backing_file) {
         BlockDriverState *bs = bdrv_new("");
-        ret = bdrv_open(bs, backing_file, NULL, 0, NULL, errp);
+        ret = bdrv_open(bs, backing_file, NULL, BDRV_O_NO_BACKING, NULL, errp);
         if (ret != 0) {
             bdrv_unref(bs);
             return ret;