Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging

Block patches

# gpg: Signature made Fri 21 Feb 2014 21:42:24 GMT using RSA key ID C88F2FD6
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>"

* remotes/kevin/tags/for-upstream: (54 commits)
  iotests: Mixed quorum child device specifications
  quorum: Simplify quorum_open()
  quorum: Add unit test.
  quorum: Add quorum_open() and quorum_close().
  quorum: Implement recursive .bdrv_recurse_is_first_non_filter in quorum.
  quorum: Add quorum_co_flush().
  quorum: Add quorum_invalidate_cache().
  quorum: Add quorum_getlength().
  quorum: Add quorum mechanism.
  quorum: Add quorum_aio_readv.
  blkverify: Extract qemu_iovec_clone() and qemu_iovec_compare() from blkverify.
  quorum: Add quorum_aio_writev and its dependencies.
  quorum: Create BDRVQuorumState and BlkDriver and do init.
  quorum: Create quorum.c, add QuorumChildRequest and QuorumAIOCB.
  check-qdict: Test termination of qdict_array_split()
  check-qdict: Adjust test for qdict_array_split()
  qdict: Extract non-QDicts in qdict_array_split()
  qemu-config: Sections must consist of keys
  qemu-iotests: Check qemu-img command line parsing
  qemu-img: Allow -o help with incomplete argument list
  ...

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

21 files changed, 1195 insertions, 348 deletions

diff --git a/block/Makefile.objs b/block/Makefile.objs
index 673aa7c8c9..fd88c03ece 100644
--- a/block/Makefile.objs
+++ b/block/Makefile.objs
@@ -3,6 +3,7 @@ block-obj-y += qcow2.o qcow2-refcount.o qcow2-cluster.o qcow2-snapshot.o qcow2-c
 block-obj-y += qed.o qed-gencb.o qed-l2-cache.o qed-table.o qed-cluster.o
 block-obj-y += qed-check.o
 block-obj-$(CONFIG_VHDX) += vhdx.o vhdx-endian.o vhdx-log.o
+block-obj-$(CONFIG_QUORUM) += quorum.o
 block-obj-y += parallels.o blkdebug.o blkverify.o
 block-obj-y += snapshot.o qapi.o
 block-obj-$(CONFIG_WIN32) += raw-win32.o win32-aio.o
diff --git a/block/blkdebug.c b/block/blkdebug.c
index ee10013362..380c736101 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -410,8 +410,9 @@ static int blkdebug_open(BlockDriverState *bs, QDict *options, int flags,
     s->state = 1;
 
     /* Open the backing file */
+    assert(bs->file == NULL);
     ret = bdrv_open_image(&bs->file, qemu_opt_get(opts, "x-image"), options, "image",
-                          flags, true, false, &local_err);
+                          flags | BDRV_O_PROTOCOL, false, &local_err);
     if (ret < 0) {
         error_propagate(errp, local_err);
         goto out;
diff --git a/block/blkverify.c b/block/blkverify.c
index 1563c88324..b98b08bedf 100644
--- a/block/blkverify.c
+++ b/block/blkverify.c
@@ -135,16 +135,18 @@ static int blkverify_open(BlockDriverState *bs, QDict *options, int flags,
     }
 
     /* Open the raw file */
+    assert(bs->file == NULL);
     ret = bdrv_open_image(&bs->file, qemu_opt_get(opts, "x-raw"), options,
-                          "raw", flags, true, false, &local_err);
+                          "raw", flags | BDRV_O_PROTOCOL, false, &local_err);
     if (ret < 0) {
         error_propagate(errp, local_err);
         goto fail;
     }
 
     /* Open the test file */
+    assert(s->test_file == NULL);
     ret = bdrv_open_image(&s->test_file, qemu_opt_get(opts, "x-image"), options,
-                          "test", flags, false, false, &local_err);
+                          "test", flags, false, &local_err);
     if (ret < 0) {
         error_propagate(errp, local_err);
         s->test_file = NULL;
@@ -171,110 +173,6 @@ static int64_t blkverify_getlength(BlockDriverState *bs)
     return bdrv_getlength(s->test_file);
 }
 
-/**
- * Check that I/O vector contents are identical
- *
- * @a:          I/O vector
- * @b:          I/O vector
- * @ret:        Offset to first mismatching byte or -1 if match
- */
-static ssize_t blkverify_iovec_compare(QEMUIOVector *a, QEMUIOVector *b)
-{
-    int i;
-    ssize_t offset = 0;
-
-    assert(a->niov == b->niov);
-    for (i = 0; i < a->niov; i++) {
-        size_t len = 0;
-        uint8_t *p = (uint8_t *)a->iov[i].iov_base;
-        uint8_t *q = (uint8_t *)b->iov[i].iov_base;
-
-        assert(a->iov[i].iov_len == b->iov[i].iov_len);
-        while (len < a->iov[i].iov_len && *p++ == *q++) {
-            len++;
-        }
-
-        offset += len;
-
-        if (len != a->iov[i].iov_len) {
-            return offset;
-        }
-    }
-    return -1;
-}
-
-typedef struct {
-    int src_index;
-    struct iovec *src_iov;
-    void *dest_base;
-} IOVectorSortElem;
-
-static int sortelem_cmp_src_base(const void *a, const void *b)
-{
-    const IOVectorSortElem *elem_a = a;
-    const IOVectorSortElem *elem_b = b;
-
-    /* Don't overflow */
-    if (elem_a->src_iov->iov_base < elem_b->src_iov->iov_base) {
-        return -1;
-    } else if (elem_a->src_iov->iov_base > elem_b->src_iov->iov_base) {
-        return 1;
-    } else {
-        return 0;
-    }
-}
-
-static int sortelem_cmp_src_index(const void *a, const void *b)
-{
-    const IOVectorSortElem *elem_a = a;
-    const IOVectorSortElem *elem_b = b;
-
-    return elem_a->src_index - elem_b->src_index;
-}
-
-/**
- * Copy contents of I/O vector
- *
- * The relative relationships of overlapping iovecs are preserved.  This is
- * necessary to ensure identical semantics in the cloned I/O vector.
- */
-static void blkverify_iovec_clone(QEMUIOVector *dest, const QEMUIOVector *src,
-                                  void *buf)
-{
-    IOVectorSortElem sortelems[src->niov];
-    void *last_end;
-    int i;
-
-    /* Sort by source iovecs by base address */
-    for (i = 0; i < src->niov; i++) {
-        sortelems[i].src_index = i;
-        sortelems[i].src_iov = &src->iov[i];
-    }
-    qsort(sortelems, src->niov, sizeof(sortelems[0]), sortelem_cmp_src_base);
-
-    /* Allocate buffer space taking into account overlapping iovecs */
-    last_end = NULL;
-    for (i = 0; i < src->niov; i++) {
-        struct iovec *cur = sortelems[i].src_iov;
-        ptrdiff_t rewind = 0;
-
-        /* Detect overlap */
-        if (last_end && last_end > cur->iov_base) {
-            rewind = last_end - cur->iov_base;
-        }
-
-        sortelems[i].dest_base = buf - rewind;
-        buf += cur->iov_len - MIN(rewind, cur->iov_len);
-        last_end = MAX(cur->iov_base + cur->iov_len, last_end);
-    }
-
-    /* Sort by source iovec index and build destination iovec */
-    qsort(sortelems, src->niov, sizeof(sortelems[0]), sortelem_cmp_src_index);
-    for (i = 0; i < src->niov; i++) {
-        qemu_iovec_add(dest, sortelems[i].dest_base, src->iov[i].iov_len);
-    }
-}
-
 static BlkverifyAIOCB *blkverify_aio_get(BlockDriverState *bs, bool is_write,
                                          int64_t sector_num, QEMUIOVector *qiov,
                                          int nb_sectors,
@@ -338,7 +236,7 @@ static void blkverify_aio_cb(void *opaque, int ret)
 
 static void blkverify_verify_readv(BlkverifyAIOCB *acb)
 {
-    ssize_t offset = blkverify_iovec_compare(acb->qiov, &acb->raw_qiov);
+    ssize_t offset = qemu_iovec_compare(acb->qiov, &acb->raw_qiov);
     if (offset != -1) {
         blkverify_err(acb, "contents mismatch in sector %" PRId64,
                       acb->sector_num + (int64_t)(offset / BDRV_SECTOR_SIZE));
@@ -356,7 +254,7 @@ static BlockDriverAIOCB *blkverify_aio_readv(BlockDriverState *bs,
     acb->verify = blkverify_verify_readv;
     acb->buf = qemu_blockalign(bs->file, qiov->size);
     qemu_iovec_init(&acb->raw_qiov, acb->qiov->niov);
-    blkverify_iovec_clone(&acb->raw_qiov, qiov, acb->buf);
+    qemu_iovec_clone(&acb->raw_qiov, qiov, acb->buf);
 
     bdrv_aio_readv(s->test_file, sector_num, qiov, nb_sectors,
                    blkverify_aio_cb, acb);
diff --git a/block/bochs.c b/block/bochs.c
index 51d9a90577..4d6403f904 100644
--- a/block/bochs.c
+++ b/block/bochs.c
@@ -129,7 +129,8 @@ static int bochs_open(BlockDriverState *bs, QDict *options, int flags,
         strcmp(bochs.subtype, GROWING_TYPE) ||
 	((le32_to_cpu(bochs.version) != HEADER_VERSION) &&
 	(le32_to_cpu(bochs.version) != HEADER_V1))) {
-        return -EMEDIUMTYPE;
+        error_setg(errp, "Image not in Bochs format");
+        return -EINVAL;
     }
 
     if (le32_to_cpu(bochs.version) == HEADER_V1) {
diff --git a/block/cow.c b/block/cow.c
index 7fc0b12163..30deb88deb 100644
--- a/block/cow.c
+++ b/block/cow.c
@@ -74,7 +74,8 @@ static int cow_open(BlockDriverState *bs, QDict *options, int flags,
     }
 
     if (be32_to_cpu(cow_header.magic) != COW_MAGIC) {
-        ret = -EMEDIUMTYPE;
+        error_setg(errp, "Image not in COW format");
+        ret = -EINVAL;
         goto fail;
     }
 
@@ -82,7 +83,7 @@ static int cow_open(BlockDriverState *bs, QDict *options, int flags,
         char version[64];
         snprintf(version, sizeof(version),
                "COW version %d", cow_header.version);
-        qerror_report(QERR_UNKNOWN_BLOCK_FORMAT_FEATURE,
+        error_set(errp, QERR_UNKNOWN_BLOCK_FORMAT_FEATURE,
             bs->device_name, "cow", version);
         ret = -ENOTSUP;
         goto fail;
@@ -346,16 +347,15 @@ static int cow_create(const char *filename, QEMUOptionParameter *options,
 
     ret = bdrv_create_file(filename, options, &local_err);
     if (ret < 0) {
-        qerror_report_err(local_err);
-        error_free(local_err);
+        error_propagate(errp, local_err);
         return ret;
     }
 
-    ret = bdrv_file_open(&cow_bs, filename, NULL, NULL, BDRV_O_RDWR,
-                         &local_err);
+    cow_bs = NULL;
+    ret = bdrv_open(&cow_bs, filename, NULL, NULL,
+                    BDRV_O_RDWR | BDRV_O_PROTOCOL, NULL, &local_err);
     if (ret < 0) {
-        qerror_report_err(local_err);
-        error_free(local_err);
+        error_propagate(errp, local_err);
         return ret;
     }
 
diff --git a/block/curl.c b/block/curl.c
index bb1fc4ae28..3494c6d662 100644
--- a/block/curl.c
+++ b/block/curl.c
@@ -456,30 +456,27 @@ static int curl_open(BlockDriverState *bs, QDict *options, int flags,
     static int inited = 0;
 
     if (flags & BDRV_O_RDWR) {
-        qerror_report(ERROR_CLASS_GENERIC_ERROR,
-                      "curl block device does not support writes");
+        error_setg(errp, "curl block device does not support writes");
         return -EROFS;
     }
 
     opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
     qemu_opts_absorb_qdict(opts, options, &local_err);
     if (local_err) {
-        qerror_report_err(local_err);
-        error_free(local_err);
+        error_propagate(errp, local_err);
         goto out_noclean;
     }
 
     s->readahead_size = qemu_opt_get_size(opts, "readahead", READ_AHEAD_SIZE);
     if ((s->readahead_size & 0x1ff) != 0) {
-        fprintf(stderr, "HTTP_READAHEAD_SIZE %zd is not a multiple of 512\n",
-                s->readahead_size);
+        error_setg(errp, "HTTP_READAHEAD_SIZE %zd is not a multiple of 512",
+                   s->readahead_size);
         goto out_noclean;
     }
 
     file = qemu_opt_get(opts, "url");
     if (file == NULL) {
-        qerror_report(ERROR_CLASS_GENERIC_ERROR, "curl block driver requires "
-                      "an 'url' option");
+        error_setg(errp, "curl block driver requires an 'url' option");
         goto out_noclean;
     }
 
diff --git a/block/gluster.c b/block/gluster.c
index 58eab07829..14d390b4c7 100644
--- a/block/gluster.c
+++ b/block/gluster.c
@@ -127,7 +127,7 @@ static int qemu_gluster_parseuri(GlusterConf *gconf, const char *filename)
     }
 
     /* transport */
-    if (!strcmp(uri->scheme, "gluster")) {
+    if (!uri->scheme || !strcmp(uri->scheme, "gluster")) {
         gconf->transport = g_strdup("tcp");
     } else if (!strcmp(uri->scheme, "gluster+tcp")) {
         gconf->transport = g_strdup("tcp");
@@ -163,7 +163,7 @@ static int qemu_gluster_parseuri(GlusterConf *gconf, const char *filename)
         }
         gconf->server = g_strdup(qp->p[0].value);
     } else {
-        gconf->server = g_strdup(uri->server);
+        gconf->server = g_strdup(uri->server ? uri->server : "localhost");
         gconf->port = uri->port;
     }
 
@@ -175,7 +175,8 @@ out:
     return ret;
 }
 
-static struct glfs *qemu_gluster_init(GlusterConf *gconf, const char *filename)
+static struct glfs *qemu_gluster_init(GlusterConf *gconf, const char *filename,
+                                      Error **errp)
 {
     struct glfs *glfs = NULL;
     int ret;
@@ -183,8 +184,8 @@ static struct glfs *qemu_gluster_init(GlusterConf *gconf, const char *filename)
 
     ret = qemu_gluster_parseuri(gconf, filename);
     if (ret < 0) {
-        error_report("Usage: file=gluster[+transport]://[server[:port]]/"
-            "volname/image[?socket=...]");
+        error_setg(errp, "Usage: file=gluster[+transport]://[server[:port]]/"
+                   "volname/image[?socket=...]");
         errno = -ret;
         goto out;
     }
@@ -211,9 +212,11 @@ static struct glfs *qemu_gluster_init(GlusterConf *gconf, const char *filename)
 
     ret = glfs_init(glfs);
     if (ret) {
-        error_report("Gluster connection failed for server=%s port=%d "
-             "volume=%s image=%s transport=%s", gconf->server, gconf->port,
-             gconf->volname, gconf->image, gconf->transport);
+        error_setg_errno(errp, errno,
+                         "Gluster connection failed for server=%s port=%d "
+                         "volume=%s image=%s transport=%s", gconf->server,
+                         gconf->port, gconf->volname, gconf->image,
+                         gconf->transport);
         goto out;
     }
     return glfs;
@@ -283,15 +286,14 @@ static int qemu_gluster_open(BlockDriverState *bs,  QDict *options,
     opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
     qemu_opts_absorb_qdict(opts, options, &local_err);
     if (local_err) {
-        qerror_report_err(local_err);
-        error_free(local_err);
+        error_propagate(errp, local_err);
         ret = -EINVAL;
         goto out;
     }
 
     filename = qemu_opt_get(opts, "filename");
 
-    s->glfs = qemu_gluster_init(gconf, filename);
+    s->glfs = qemu_gluster_init(gconf, filename, errp);
     if (!s->glfs) {
         ret = -errno;
         goto out;
@@ -389,9 +391,9 @@ static int qemu_gluster_create(const char *filename,
     int64_t total_size = 0;
     GlusterConf *gconf = g_malloc0(sizeof(GlusterConf));
 
-    glfs = qemu_gluster_init(gconf, filename);
+    glfs = qemu_gluster_init(gconf, filename, errp);
     if (!glfs) {
-        ret = -errno;
+        ret = -EINVAL;
         goto out;
     }
 
diff --git a/block/iscsi.c b/block/iscsi.c
index f8e496f8ef..41ec09709d 100644
--- a/block/iscsi.c
+++ b/block/iscsi.c
@@ -856,7 +856,8 @@ retry:
 
 #endif /* SCSI_SENSE_ASCQ_CAPACITY_DATA_HAS_CHANGED */
 
-static int parse_chap(struct iscsi_context *iscsi, const char *target)
+static void parse_chap(struct iscsi_context *iscsi, const char *target,
+                       Error **errp)
 {
     QemuOptsList *list;
     QemuOpts *opts;
@@ -865,37 +866,35 @@ static int parse_chap(struct iscsi_context *iscsi, const char *target)
 
     list = qemu_find_opts("iscsi");
     if (!list) {
-        return 0;
+        return;
     }
 
     opts = qemu_opts_find(list, target);
     if (opts == NULL) {
         opts = QTAILQ_FIRST(&list->head);
         if (!opts) {
-            return 0;
+            return;
         }
     }
 
     user = qemu_opt_get(opts, "user");
     if (!user) {
-        return 0;
+        return;
     }
 
     password = qemu_opt_get(opts, "password");
     if (!password) {
-        error_report("CHAP username specified but no password was given");
-        return -1;
+        error_setg(errp, "CHAP username specified but no password was given");
+        return;
     }
 
     if (iscsi_set_initiator_username_pwd(iscsi, user, password)) {
-        error_report("Failed to set initiator username and password");
-        return -1;
+        error_setg(errp, "Failed to set initiator username and password");
     }
-
-    return 0;
 }
 
-static void parse_header_digest(struct iscsi_context *iscsi, const char *target)
+static void parse_header_digest(struct iscsi_context *iscsi, const char *target,
+                                Error **errp)
 {
     QemuOptsList *list;
     QemuOpts *opts;
@@ -928,7 +927,7 @@ static void parse_header_digest(struct iscsi_context *iscsi, const char *target)
     } else if (!strcmp(digest, "NONE-CRC32C")) {
         iscsi_set_header_digest(iscsi, ISCSI_HEADER_DIGEST_NONE_CRC32C);
     } else {
-        error_report("Invalid header-digest setting : %s", digest);
+        error_setg(errp, "Invalid header-digest setting : %s", digest);
     }
 }
 
@@ -986,12 +985,11 @@ static void iscsi_nop_timed_event(void *opaque)
 }
 #endif
 
-static int iscsi_readcapacity_sync(IscsiLun *iscsilun)
+static void iscsi_readcapacity_sync(IscsiLun *iscsilun, Error **errp)
 {
     struct scsi_task *task = NULL;
     struct scsi_readcapacity10 *rc10 = NULL;
     struct scsi_readcapacity16 *rc16 = NULL;
-    int ret = 0;
     int retries = ISCSI_CMD_RETRIES; 
 
     do {
@@ -1006,8 +1004,7 @@ static int iscsi_readcapacity_sync(IscsiLun *iscsilun)
             if (task != NULL && task->status == SCSI_STATUS_GOOD) {
                 rc16 = scsi_datain_unmarshall(task);
                 if (rc16 == NULL) {
-                    error_report("iSCSI: Failed to unmarshall readcapacity16 data.");
-                    ret = -EINVAL;
+                    error_setg(errp, "iSCSI: Failed to unmarshall readcapacity16 data.");
                 } else {
                     iscsilun->block_size = rc16->block_length;
                     iscsilun->num_blocks = rc16->returned_lba + 1;
@@ -1021,8 +1018,7 @@ static int iscsi_readcapacity_sync(IscsiLun *iscsilun)
             if (task != NULL && task->status == SCSI_STATUS_GOOD) {
                 rc10 = scsi_datain_unmarshall(task);
                 if (rc10 == NULL) {
-                    error_report("iSCSI: Failed to unmarshall readcapacity10 data.");
-                    ret = -EINVAL;
+                    error_setg(errp, "iSCSI: Failed to unmarshall readcapacity10 data.");
                 } else {
                     iscsilun->block_size = rc10->block_size;
                     if (rc10->lba == 0) {
@@ -1035,20 +1031,18 @@ static int iscsi_readcapacity_sync(IscsiLun *iscsilun)
             }
             break;
         default:
-            return 0;
+            return;
         }
     } while (task != NULL && task->status == SCSI_STATUS_CHECK_CONDITION
              && task->sense.key == SCSI_SENSE_UNIT_ATTENTION
              && retries-- > 0);
 
     if (task == NULL || task->status != SCSI_STATUS_GOOD) {
-        error_report("iSCSI: failed to send readcapacity10 command.");
-        ret = -EINVAL;
+        error_setg(errp, "iSCSI: failed to send readcapacity10 command.");
     }
     if (task) {
         scsi_free_scsi_task(task);
     }
-    return ret;
 }
 
 /* TODO Convert to fine grained options */
@@ -1065,35 +1059,36 @@ static QemuOptsList runtime_opts = {
     },
 };
 
-static struct scsi_task *iscsi_do_inquiry(struct iscsi_context *iscsi,
-                                          int lun, int evpd, int pc) {
-        int full_size;
-        struct scsi_task *task = NULL;
-        task = iscsi_inquiry_sync(iscsi, lun, evpd, pc, 64);
+static struct scsi_task *iscsi_do_inquiry(struct iscsi_context *iscsi, int lun,
+                                          int evpd, int pc, Error **errp)
+{
+    int full_size;
+    struct scsi_task *task = NULL;
+    task = iscsi_inquiry_sync(iscsi, lun, evpd, pc, 64);
+    if (task == NULL || task->status != SCSI_STATUS_GOOD) {
+        goto fail;
+    }
+    full_size = scsi_datain_getfullsize(task);
+    if (full_size > task->datain.size) {
+        scsi_free_scsi_task(task);
+
+        /* we need more data for the full list */
+        task = iscsi_inquiry_sync(iscsi, lun, evpd, pc, full_size);
         if (task == NULL || task->status != SCSI_STATUS_GOOD) {
             goto fail;
         }
-        full_size = scsi_datain_getfullsize(task);
-        if (full_size > task->datain.size) {
-            scsi_free_scsi_task(task);
-
-            /* we need more data for the full list */
-            task = iscsi_inquiry_sync(iscsi, lun, evpd, pc, full_size);
-            if (task == NULL || task->status != SCSI_STATUS_GOOD) {
-                goto fail;
-            }
-        }
+    }
 
-        return task;
+    return task;
 
 fail:
-        error_report("iSCSI: Inquiry command failed : %s",
-                     iscsi_get_error(iscsi));
-        if (task) {
-            scsi_free_scsi_task(task);
-            return NULL;
-        }
+    error_setg(errp, "iSCSI: Inquiry command failed : %s",
+               iscsi_get_error(iscsi));
+    if (task) {
+        scsi_free_scsi_task(task);
         return NULL;
+    }
+    return NULL;
 }
 
 /*
@@ -1119,27 +1114,25 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
     int ret;
 
     if ((BDRV_SECTOR_SIZE % 512) != 0) {
-        error_report("iSCSI: Invalid BDRV_SECTOR_SIZE. "
-                     "BDRV_SECTOR_SIZE(%lld) is not a multiple "
-                     "of 512", BDRV_SECTOR_SIZE);
+        error_setg(errp, "iSCSI: Invalid BDRV_SECTOR_SIZE. "
+                   "BDRV_SECTOR_SIZE(%lld) is not a multiple "
+                   "of 512", BDRV_SECTOR_SIZE);
         return -EINVAL;
     }
 
     opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
     qemu_opts_absorb_qdict(opts, options, &local_err);
     if (local_err) {
-        qerror_report_err(local_err);
-        error_free(local_err);
+        error_propagate(errp, local_err);
         ret = -EINVAL;
         goto out;
     }
 
     filename = qemu_opt_get(opts, "filename");
 
-
     iscsi_url = iscsi_parse_full_url(iscsi, filename);
     if (iscsi_url == NULL) {
-        error_report("Failed to parse URL : %s", filename);
+        error_setg(errp, "Failed to parse URL : %s", filename);
         ret = -EINVAL;
         goto out;
     }
@@ -1150,13 +1143,13 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
 
     iscsi = iscsi_create_context(initiator_name);
     if (iscsi == NULL) {
-        error_report("iSCSI: Failed to create iSCSI context.");
+        error_setg(errp, "iSCSI: Failed to create iSCSI context.");
         ret = -ENOMEM;
         goto out;
     }
 
     if (iscsi_set_targetname(iscsi, iscsi_url->target)) {
-        error_report("iSCSI: Failed to set target name.");
+        error_setg(errp, "iSCSI: Failed to set target name.");
         ret = -EINVAL;
         goto out;
     }
@@ -1165,21 +1158,22 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
         ret = iscsi_set_initiator_username_pwd(iscsi, iscsi_url->user,
                                               iscsi_url->passwd);
         if (ret != 0) {
-            error_report("Failed to set initiator username and password");
+            error_setg(errp, "Failed to set initiator username and password");
             ret = -EINVAL;
             goto out;
         }
     }
 
     /* check if we got CHAP username/password via the options */
-    if (parse_chap(iscsi, iscsi_url->target) != 0) {
-        error_report("iSCSI: Failed to set CHAP user/password");
+    parse_chap(iscsi, iscsi_url->target, &local_err);
+    if (local_err != NULL) {
+        error_propagate(errp, local_err);
         ret = -EINVAL;
         goto out;
     }
 
     if (iscsi_set_session_type(iscsi, ISCSI_SESSION_NORMAL) != 0) {
-        error_report("iSCSI: Failed to set session type to normal.");
+        error_setg(errp, "iSCSI: Failed to set session type to normal.");
         ret = -EINVAL;
         goto out;
     }
@@ -1187,10 +1181,15 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
     iscsi_set_header_digest(iscsi, ISCSI_HEADER_DIGEST_NONE_CRC32C);
 
     /* check if we got HEADER_DIGEST via the options */
-    parse_header_digest(iscsi, iscsi_url->target);
+    parse_header_digest(iscsi, iscsi_url->target, &local_err);
+    if (local_err != NULL) {
+        error_propagate(errp, local_err);
+        ret = -EINVAL;
+        goto out;
+    }
 
     if (iscsi_full_connect_sync(iscsi, iscsi_url->portal, iscsi_url->lun) != 0) {
-        error_report("iSCSI: Failed to connect to LUN : %s",
+        error_setg(errp, "iSCSI: Failed to connect to LUN : %s",
             iscsi_get_error(iscsi));
         ret = -EINVAL;
         goto out;
@@ -1202,14 +1201,14 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
     task = iscsi_inquiry_sync(iscsi, iscsilun->lun, 0, 0, 36);
 
     if (task == NULL || task->status != SCSI_STATUS_GOOD) {
-        error_report("iSCSI: failed to send inquiry command.");
+        error_setg(errp, "iSCSI: failed to send inquiry command.");
         ret = -EINVAL;
         goto out;
     }
 
     inq = scsi_datain_unmarshall(task);
     if (inq == NULL) {
-        error_report("iSCSI: Failed to unmarshall inquiry data.");
+        error_setg(errp, "iSCSI: Failed to unmarshall inquiry data.");
         ret = -EINVAL;
         goto out;
     }
@@ -1217,7 +1216,9 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
     iscsilun->type = inq->periperal_device_type;
     iscsilun->has_write_same = true;
 
-    if ((ret = iscsi_readcapacity_sync(iscsilun)) != 0) {
+    iscsi_readcapacity_sync(iscsilun, &local_err);
+    if (local_err != NULL) {
+        error_propagate(errp, local_err);
         goto out;
     }
     bs->total_sectors = sector_lun2qemu(iscsilun->num_blocks, iscsilun);
@@ -1235,14 +1236,15 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
     if (iscsilun->lbpme) {
         struct scsi_inquiry_logical_block_provisioning *inq_lbp;
         task = iscsi_do_inquiry(iscsilun->iscsi, iscsilun->lun, 1,
-                                SCSI_INQUIRY_PAGECODE_LOGICAL_BLOCK_PROVISIONING);
+                                SCSI_INQUIRY_PAGECODE_LOGICAL_BLOCK_PROVISIONING,
+                                errp);
         if (task == NULL) {
             ret = -EINVAL;
             goto out;
         }
         inq_lbp = scsi_datain_unmarshall(task);
         if (inq_lbp == NULL) {
-            error_report("iSCSI: failed to unmarshall inquiry datain blob");
+            error_setg(errp, "iSCSI: failed to unmarshall inquiry datain blob");
             ret = -EINVAL;
             goto out;
         }
@@ -1255,14 +1257,14 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
     if (iscsilun->lbp.lbpu || iscsilun->lbp.lbpws) {
         struct scsi_inquiry_block_limits *inq_bl;
         task = iscsi_do_inquiry(iscsilun->iscsi, iscsilun->lun, 1,
-                                SCSI_INQUIRY_PAGECODE_BLOCK_LIMITS);
+                                SCSI_INQUIRY_PAGECODE_BLOCK_LIMITS, errp);
         if (task == NULL) {
             ret = -EINVAL;
             goto out;
         }
         inq_bl = scsi_datain_unmarshall(task);
         if (inq_bl == NULL) {
-            error_report("iSCSI: failed to unmarshall inquiry datain blob");
+            error_setg(errp, "iSCSI: failed to unmarshall inquiry datain blob");
             ret = -EINVAL;
             goto out;
         }
@@ -1353,14 +1355,16 @@ static int iscsi_reopen_prepare(BDRVReopenState *state,
 static int iscsi_truncate(BlockDriverState *bs, int64_t offset)
 {
     IscsiLun *iscsilun = bs->opaque;
-    int ret = 0;
+    Error *local_err = NULL;
 
     if (iscsilun->type != TYPE_DISK) {
         return -ENOTSUP;
     }
 
-    if ((ret = iscsi_readcapacity_sync(iscsilun)) != 0) {
-        return ret;
+    iscsi_readcapacity_sync(iscsilun, &local_err);
+    if (local_err != NULL) {
+        error_free(local_err);
+        return -EIO;
     }
 
     if (offset > iscsi_getlength(bs)) {
diff --git a/block/nbd.c b/block/nbd.c
index abae506f04..55124239df 100644
--- a/block/nbd.c
+++ b/block/nbd.c
@@ -188,31 +188,28 @@ out:
     g_free(file);
 }
 
-static int nbd_config(BDRVNBDState *s, QDict *options, char **export)
+static void nbd_config(BDRVNBDState *s, QDict *options, char **export,
+                       Error **errp)
 {
     Error *local_err = NULL;
 
-    if (qdict_haskey(options, "path")) {
-        if (qdict_haskey(options, "host")) {
-            qerror_report(ERROR_CLASS_GENERIC_ERROR, "path and host may not "
-                          "be used at the same time.");
-            return -EINVAL;
+    if (qdict_haskey(options, "path") == qdict_haskey(options, "host")) {
+        if (qdict_haskey(options, "path")) {
+            error_setg(errp, "path and host may not be used at the same time.");
+        } else {
+            error_setg(errp, "one of path and host must be specified.");
         }
-        s->client.is_unix = true;
-    } else if (qdict_haskey(options, "host")) {
-        s->client.is_unix = false;
-    } else {
-        return -EINVAL;
+        return;
     }
 
+    s->client.is_unix = qdict_haskey(options, "path");
     s->socket_opts = qemu_opts_create(&socket_optslist, NULL, 0,
                                       &error_abort);
 
     qemu_opts_absorb_qdict(s->socket_opts, options, &local_err);
     if (local_err) {
-        qerror_report_err(local_err);
-        error_free(local_err);
-        return -EINVAL;
+        error_propagate(errp, local_err);
+        return;
     }
 
     if (!qemu_opt_get(s->socket_opts, "port")) {
@@ -223,19 +220,17 @@ static int nbd_config(BDRVNBDState *s, QDict *options, char **export)
     if (*export) {
         qdict_del(options, "export");
     }
-
-    return 0;
 }
 
-static int nbd_establish_connection(BlockDriverState *bs)
+static int nbd_establish_connection(BlockDriverState *bs, Error **errp)
 {
     BDRVNBDState *s = bs->opaque;
     int sock;
 
     if (s->client.is_unix) {
-        sock = unix_socket_outgoing(qemu_opt_get(s->socket_opts, "path"));
+        sock = unix_connect_opts(s->socket_opts, errp, NULL, NULL);
     } else {
-        sock = tcp_socket_outgoing_opts(s->socket_opts);
+        sock = inet_connect_opts(s->socket_opts, errp, NULL, NULL);
         if (sock >= 0) {
             socket_set_nodelay(sock);
         }
@@ -256,17 +251,19 @@ static int nbd_open(BlockDriverState *bs, QDict *options, int flags,
     BDRVNBDState *s = bs->opaque;
     char *export = NULL;
     int result, sock;
+    Error *local_err = NULL;
 
     /* Pop the config into our state object. Exit if invalid. */
-    result = nbd_config(s, options, &export);
-    if (result != 0) {
-        return result;
+    nbd_config(s, options, &export, &local_err);
+    if (local_err) {
+        error_propagate(errp, local_err);
+        return -EINVAL;
     }
 
     /* establish TCP connection, return error if it fails
      * TODO: Configurable retry-until-timeout behaviour.
      */
-    sock = nbd_establish_connection(bs);
+    sock = nbd_establish_connection(bs, errp);
     if (sock < 0) {
         return sock;
     }
diff --git a/block/parallels.c b/block/parallels.c
index 2121e43204..3f588f58dc 100644
--- a/block/parallels.c
+++ b/block/parallels.c
@@ -85,7 +85,8 @@ static int parallels_open(BlockDriverState *bs, QDict *options, int flags,
 
     if (memcmp(ph.magic, HEADER_MAGIC, 16) ||
         (le32_to_cpu(ph.version) != HEADER_VERSION)) {
-        ret = -EMEDIUMTYPE;
+        error_setg(errp, "Image not in Parallels format");
+        ret = -EINVAL;
         goto fail;
     }
 
diff --git a/block/qcow.c b/block/qcow.c
index 948b0c5601..1e128becf0 100644
--- a/block/qcow.c
+++ b/block/qcow.c
@@ -113,23 +113,26 @@ static int qcow_open(BlockDriverState *bs, QDict *options, int flags,
     be64_to_cpus(&header.l1_table_offset);
 
     if (header.magic != QCOW_MAGIC) {
-        ret = -EMEDIUMTYPE;
+        error_setg(errp, "Image not in qcow format");
+        ret = -EINVAL;
         goto fail;
     }
     if (header.version != QCOW_VERSION) {
         char version[64];
         snprintf(version, sizeof(version), "QCOW version %d", header.version);
-        qerror_report(QERR_UNKNOWN_BLOCK_FORMAT_FEATURE,
-            bs->device_name, "qcow", version);
+        error_set(errp, QERR_UNKNOWN_BLOCK_FORMAT_FEATURE,
+                  bs->device_name, "qcow", version);
         ret = -ENOTSUP;
         goto fail;
     }
 
     if (header.size <= 1 || header.cluster_bits < 9) {
+        error_setg(errp, "invalid value in qcow header");
         ret = -EINVAL;
         goto fail;
     }
     if (header.crypt_method > QCOW_CRYPT_AES) {
+        error_setg(errp, "invalid encryption method in qcow header");
         ret = -EINVAL;
         goto fail;
     }
@@ -686,16 +689,15 @@ static int qcow_create(const char *filename, QEMUOptionParameter *options,
 
     ret = bdrv_create_file(filename, options, &local_err);
     if (ret < 0) {
-        qerror_report_err(local_err);
-        error_free(local_err);
+        error_propagate(errp, local_err);
         return ret;
     }
 
-    ret = bdrv_file_open(&qcow_bs, filename, NULL, NULL, BDRV_O_RDWR,
-                         &local_err);
+    qcow_bs = NULL;
+    ret = bdrv_open(&qcow_bs, filename, NULL, NULL,
+                    BDRV_O_RDWR | BDRV_O_PROTOCOL, NULL, &local_err);
     if (ret < 0) {
-        qerror_report_err(local_err);
-        error_free(local_err);
+        error_propagate(errp, local_err);
         return ret;
     }
 
diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
index c57f39dd2b..36c1bed350 100644
--- a/block/qcow2-cluster.c
+++ b/block/qcow2-cluster.c
@@ -1367,13 +1367,31 @@ static int discard_single_l2(BlockDriverState *bs, uint64_t offset,
         uint64_t old_offset;
 
         old_offset = be64_to_cpu(l2_table[l2_index + i]);
-        if ((old_offset & L2E_OFFSET_MASK) == 0) {
+
+        /*
+         * Make sure that a discarded area reads back as zeroes for v3 images
+         * (we cannot do it for v2 without actually writing a zero-filled
+         * buffer). We can skip the operation if the cluster is already marked
+         * as zero, or if it's unallocated and we don't have a backing file.
+         *
+         * TODO We might want to use bdrv_get_block_status(bs) here, but we're
+         * holding s->lock, so that doesn't work today.
+         */
+        if (old_offset & QCOW_OFLAG_ZERO) {
+            continue;
+        }
+
+        if ((old_offset & L2E_OFFSET_MASK) == 0 && !bs->backing_hd) {
             continue;
         }
 
         /* First remove L2 entries */
         qcow2_cache_entry_mark_dirty(s->l2_table_cache, l2_table);
-        l2_table[l2_index + i] = cpu_to_be64(0);
+        if (s->qcow_version >= 3) {
+            l2_table[l2_index + i] = cpu_to_be64(QCOW_OFLAG_ZERO);
+        } else {
+            l2_table[l2_index + i] = cpu_to_be64(0);
+        }
 
         /* Then decrease the refcount */
         qcow2_free_any_clusters(bs, old_offset, 1, type);
diff --git a/block/qcow2.c b/block/qcow2.c
index b1dbdb120e..cfe80befa0 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -449,7 +449,7 @@ static int qcow2_open(BlockDriverState *bs, QDict *options, int flags,
 
     if (header.magic != QCOW_MAGIC) {
         error_setg(errp, "Image is not in qcow2 format");
-        ret = -EMEDIUMTYPE;
+        ret = -EINVAL;
         goto fail;
     }
     if (header.version < 2 || header.version > 3) {
@@ -1493,7 +1493,9 @@ static int qcow2_create2(const char *filename, int64_t total_size,
         return ret;
     }
 
-    ret = bdrv_file_open(&bs, filename, NULL, NULL, BDRV_O_RDWR, &local_err);
+    bs = NULL;
+    ret = bdrv_open(&bs, filename, NULL, NULL, BDRV_O_RDWR | BDRV_O_PROTOCOL,
+                    NULL, &local_err);
     if (ret < 0) {
         error_propagate(errp, local_err);
         return ret;
@@ -1543,7 +1545,8 @@ static int qcow2_create2(const char *filename, int64_t total_size,
         goto out;
     }
 
-    bdrv_close(bs);
+    bdrv_unref(bs);
+    bs = NULL;
 
     /*
      * And now open the image and make it consistent first (i.e. increase the
@@ -1552,7 +1555,7 @@ static int qcow2_create2(const char *filename, int64_t total_size,
      */
     BlockDriver* drv = bdrv_find_format("qcow2");
     assert(drv != NULL);
-    ret = bdrv_open(bs, filename, NULL,
+    ret = bdrv_open(&bs, filename, NULL, NULL,
         BDRV_O_RDWR | BDRV_O_CACHE_WB | BDRV_O_NO_FLUSH, drv, &local_err);
     if (ret < 0) {
         error_propagate(errp, local_err);
@@ -1599,10 +1602,11 @@ static int qcow2_create2(const char *filename, int64_t total_size,
         }
     }
 
-    bdrv_close(bs);
+    bdrv_unref(bs);
+    bs = NULL;
 
     /* Reopen the image without BDRV_O_NO_FLUSH to flush it before returning */
-    ret = bdrv_open(bs, filename, NULL,
+    ret = bdrv_open(&bs, filename, NULL, NULL,
                     BDRV_O_RDWR | BDRV_O_CACHE_WB | BDRV_O_NO_BACKING,
                     drv, &local_err);
     if (local_err) {
@@ -1612,7 +1616,9 @@ static int qcow2_create2(const char *filename, int64_t total_size,
 
     ret = 0;
 out:
-    bdrv_unref(bs);
+    if (bs) {
+        bdrv_unref(bs);
+    }
     return ret;
 }
 
diff --git a/block/qed.c b/block/qed.c
index b9ca7ac0da..8802ad3845 100644
--- a/block/qed.c
+++ b/block/qed.c
@@ -391,14 +391,15 @@ static int bdrv_qed_open(BlockDriverState *bs, QDict *options, int flags,
     qed_header_le_to_cpu(&le_header, &s->header);
 
     if (s->header.magic != QED_MAGIC) {
-        return -EMEDIUMTYPE;
+        error_setg(errp, "Image not in QED format");
+        return -EINVAL;
     }
     if (s->header.features & ~QED_FEATURE_MASK) {
         /* image uses unsupported feature bits */
         char buf[64];
         snprintf(buf, sizeof(buf), "%" PRIx64,
             s->header.features & ~QED_FEATURE_MASK);
-        qerror_report(QERR_UNKNOWN_BLOCK_FORMAT_FEATURE,
+        error_set(errp, QERR_UNKNOWN_BLOCK_FORMAT_FEATURE,
             bs->device_name, "QED", buf);
         return -ENOTSUP;
     }
@@ -545,7 +546,8 @@ static void bdrv_qed_close(BlockDriverState *bs)
 
 static int qed_create(const char *filename, uint32_t cluster_size,
                       uint64_t image_size, uint32_t table_size,
-                      const char *backing_file, const char *backing_fmt)
+                      const char *backing_file, const char *backing_fmt,
+                      Error **errp)
 {
     QEDHeader header = {
         .magic = QED_MAGIC,
@@ -562,20 +564,20 @@ static int qed_create(const char *filename, uint32_t cluster_size,
     size_t l1_size = header.cluster_size * header.table_size;
     Error *local_err = NULL;
     int ret = 0;
-    BlockDriverState *bs = NULL;
+    BlockDriverState *bs;
 
     ret = bdrv_create_file(filename, NULL, &local_err);
     if (ret < 0) {
-        qerror_report_err(local_err);
-        error_free(local_err);
+        error_propagate(errp, local_err);
         return ret;
     }
 
-    ret = bdrv_file_open(&bs, filename, NULL, NULL,
-                         BDRV_O_RDWR | BDRV_O_CACHE_WB, &local_err);
+    bs = NULL;
+    ret = bdrv_open(&bs, filename, NULL, NULL,
+                    BDRV_O_RDWR | BDRV_O_CACHE_WB | BDRV_O_PROTOCOL, NULL,
+                    &local_err);
     if (ret < 0) {
-        qerror_report_err(local_err);
-        error_free(local_err);
+        error_propagate(errp, local_err);
         return ret;
     }
 
@@ -665,7 +667,7 @@ static int bdrv_qed_create(const char *filename, QEMUOptionParameter *options,
     }
 
     return qed_create(filename, cluster_size, image_size, table_size,
-                      backing_file, backing_fmt);
+                      backing_file, backing_fmt, errp);
 }
 
 typedef struct {
diff --git a/block/quorum.c b/block/quorum.c
new file mode 100644
index 0000000000..6c28239718
--- /dev/null
+++ b/block/quorum.c
@@ -0,0 +1,870 @@
+/*
+ * Quorum Block filter
+ *
+ * Copyright (C) 2012-2014 Nodalink, EURL.
+ *
+ * Author:
+ *   Benoît Canet <benoit.canet@irqsave.net>
+ *
+ * Based on the design and code of blkverify.c (Copyright (C) 2010 IBM, Corp)
+ * and blkmirror.c (Copyright (C) 2011 Red Hat, Inc).
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+
+#include <gnutls/gnutls.h>
+#include <gnutls/crypto.h>
+#include "block/block_int.h"
+#include "qapi/qmp/qjson.h"
+
+#define HASH_LENGTH 32
+
+#define QUORUM_OPT_VOTE_THRESHOLD "vote-threshold"
+#define QUORUM_OPT_BLKVERIFY      "blkverify"
+
+/* This union holds a vote hash value */
+typedef union QuorumVoteValue {
+    char h[HASH_LENGTH];       /* SHA-256 hash */
+    int64_t l;                 /* simpler 64 bits hash */
+} QuorumVoteValue;
+
+/* A vote item */
+typedef struct QuorumVoteItem {
+    int index;
+    QLIST_ENTRY(QuorumVoteItem) next;
+} QuorumVoteItem;
+
+/* this structure is a vote version. A version is the set of votes sharing the
+ * same vote value.
+ * The set of votes will be tracked with the items field and its cardinality is
+ * vote_count.
+ */
+typedef struct QuorumVoteVersion {
+    QuorumVoteValue value;
+    int index;
+    int vote_count;
+    QLIST_HEAD(, QuorumVoteItem) items;
+    QLIST_ENTRY(QuorumVoteVersion) next;
+} QuorumVoteVersion;
+
+/* this structure holds a group of vote versions together */
+typedef struct QuorumVotes {
+    QLIST_HEAD(, QuorumVoteVersion) vote_list;
+    bool (*compare)(QuorumVoteValue *a, QuorumVoteValue *b);
+} QuorumVotes;
+
+/* the following structure holds the state of one quorum instance */
+typedef struct BDRVQuorumState {
+    BlockDriverState **bs; /* children BlockDriverStates */
+    int num_children;      /* children count */
+    int threshold;         /* if less than threshold children reads gave the
+                            * same result a quorum error occurs.
+                            */
+    bool is_blkverify;     /* true if the driver is in blkverify mode
+                            * Writes are mirrored on two children devices.
+                            * On reads the two children devices' contents are
+                            * compared and if a difference is spotted its
+                            * location is printed and the code aborts.
+                            * It is useful to debug other block drivers by
+                            * comparing them with a reference one.
+                            */
+} BDRVQuorumState;
+
+typedef struct QuorumAIOCB QuorumAIOCB;
+
+/* Quorum will create one instance of the following structure per operation it
+ * performs on its children.
+ * So for each read/write operation coming from the upper layer there will be
+ * $children_count QuorumChildRequest.
+ */
+typedef struct QuorumChildRequest {
+    BlockDriverAIOCB *aiocb;
+    QEMUIOVector qiov;
+    uint8_t *buf;
+    int ret;
+    QuorumAIOCB *parent;
+} QuorumChildRequest;
+
+/* Quorum will use the following structure to track progress of each read/write
+ * operation received by the upper layer.
+ * This structure hold pointers to the QuorumChildRequest structures instances
+ * used to do operations on each children and track overall progress.
+ */
+struct QuorumAIOCB {
+    BlockDriverAIOCB common;
+
+    /* Request metadata */
+    uint64_t sector_num;
+    int nb_sectors;
+
+    QEMUIOVector *qiov;         /* calling IOV */
+
+    QuorumChildRequest *qcrs;   /* individual child requests */
+    int count;                  /* number of completed AIOCB */
+    int success_count;          /* number of successfully completed AIOCB */
+
+    QuorumVotes votes;
+
+    bool is_read;
+    int vote_ret;
+};
+
+static void quorum_vote(QuorumAIOCB *acb);
+
+static void quorum_aio_cancel(BlockDriverAIOCB *blockacb)
+{
+    QuorumAIOCB *acb = container_of(blockacb, QuorumAIOCB, common);
+    BDRVQuorumState *s = acb->common.bs->opaque;
+    int i;
+
+    /* cancel all callbacks */
+    for (i = 0; i < s->num_children; i++) {
+        bdrv_aio_cancel(acb->qcrs[i].aiocb);
+    }
+
+    g_free(acb->qcrs);
+    qemu_aio_release(acb);
+}
+
+static AIOCBInfo quorum_aiocb_info = {
+    .aiocb_size         = sizeof(QuorumAIOCB),
+    .cancel             = quorum_aio_cancel,
+};
+
+static void quorum_aio_finalize(QuorumAIOCB *acb)
+{
+    BDRVQuorumState *s = acb->common.bs->opaque;
+    int i, ret = 0;
+
+    if (acb->vote_ret) {
+        ret = acb->vote_ret;
+    }
+
+    acb->common.cb(acb->common.opaque, ret);
+
+    if (acb->is_read) {
+        for (i = 0; i < s->num_children; i++) {
+            qemu_vfree(acb->qcrs[i].buf);
+            qemu_iovec_destroy(&acb->qcrs[i].qiov);
+        }
+    }
+
+    g_free(acb->qcrs);
+    qemu_aio_release(acb);
+}
+
+static bool quorum_sha256_compare(QuorumVoteValue *a, QuorumVoteValue *b)
+{
+    return !memcmp(a->h, b->h, HASH_LENGTH);
+}
+
+static bool quorum_64bits_compare(QuorumVoteValue *a, QuorumVoteValue *b)
+{
+    return a->l == b->l;
+}
+
+static QuorumAIOCB *quorum_aio_get(BDRVQuorumState *s,
+                                   BlockDriverState *bs,
+                                   QEMUIOVector *qiov,
+                                   uint64_t sector_num,
+                                   int nb_sectors,
+                                   BlockDriverCompletionFunc *cb,
+                                   void *opaque)
+{
+    QuorumAIOCB *acb = qemu_aio_get(&quorum_aiocb_info, bs, cb, opaque);
+    int i;
+
+    acb->common.bs->opaque = s;
+    acb->sector_num = sector_num;
+    acb->nb_sectors = nb_sectors;
+    acb->qiov = qiov;
+    acb->qcrs = g_new0(QuorumChildRequest, s->num_children);
+    acb->count = 0;
+    acb->success_count = 0;
+    acb->votes.compare = quorum_sha256_compare;
+    QLIST_INIT(&acb->votes.vote_list);
+    acb->is_read = false;
+    acb->vote_ret = 0;
+
+    for (i = 0; i < s->num_children; i++) {
+        acb->qcrs[i].buf = NULL;
+        acb->qcrs[i].ret = 0;
+        acb->qcrs[i].parent = acb;
+    }
+
+    return acb;
+}
+
+static void quorum_report_bad(QuorumAIOCB *acb, char *node_name, int ret)
+{
+    QObject *data;
+    assert(node_name);
+    data = qobject_from_jsonf("{ 'ret': %d"
+                              ", 'node-name': %s"
+                              ", 'sector-num': %" PRId64
+                              ", 'sectors-count': %d }",
+                              ret, node_name, acb->sector_num, acb->nb_sectors);
+    monitor_protocol_event(QEVENT_QUORUM_REPORT_BAD, data);
+    qobject_decref(data);
+}
+
+static void quorum_report_failure(QuorumAIOCB *acb)
+{
+    QObject *data;
+    const char *reference = acb->common.bs->device_name[0] ?
+                            acb->common.bs->device_name :
+                            acb->common.bs->node_name;
+    data = qobject_from_jsonf("{ 'reference': %s"
+                              ", 'sector-num': %" PRId64
+                              ", 'sectors-count': %d }",
+                              reference, acb->sector_num, acb->nb_sectors);
+    monitor_protocol_event(QEVENT_QUORUM_FAILURE, data);
+    qobject_decref(data);
+}
+
+static int quorum_vote_error(QuorumAIOCB *acb);
+
+static bool quorum_has_too_much_io_failed(QuorumAIOCB *acb)
+{
+    BDRVQuorumState *s = acb->common.bs->opaque;
+
+    if (acb->success_count < s->threshold) {
+        acb->vote_ret = quorum_vote_error(acb);
+        quorum_report_failure(acb);
+        return true;
+    }
+
+    return false;
+}
+
+static void quorum_aio_cb(void *opaque, int ret)
+{
+    QuorumChildRequest *sacb = opaque;
+    QuorumAIOCB *acb = sacb->parent;
+    BDRVQuorumState *s = acb->common.bs->opaque;
+
+    sacb->ret = ret;
+    acb->count++;
+    if (ret == 0) {
+        acb->success_count++;
+    } else {
+        quorum_report_bad(acb, sacb->aiocb->bs->node_name, ret);
+    }
+    assert(acb->count <= s->num_children);
+    assert(acb->success_count <= s->num_children);
+    if (acb->count < s->num_children) {
+        return;
+    }
+
+    /* Do the vote on read */
+    if (acb->is_read) {
+        quorum_vote(acb);
+    } else {
+        quorum_has_too_much_io_failed(acb);
+    }
+
+    quorum_aio_finalize(acb);
+}
+
+static void quorum_report_bad_versions(BDRVQuorumState *s,
+                                       QuorumAIOCB *acb,
+                                       QuorumVoteValue *value)
+{
+    QuorumVoteVersion *version;
+    QuorumVoteItem *item;
+
+    QLIST_FOREACH(version, &acb->votes.vote_list, next) {
+        if (acb->votes.compare(&version->value, value)) {
+            continue;
+        }
+        QLIST_FOREACH(item, &version->items, next) {
+            quorum_report_bad(acb, s->bs[item->index]->node_name, 0);
+        }
+    }
+}
+
+static void quorum_copy_qiov(QEMUIOVector *dest, QEMUIOVector *source)
+{
+    int i;
+    assert(dest->niov == source->niov);
+    assert(dest->size == source->size);
+    for (i = 0; i < source->niov; i++) {
+        assert(dest->iov[i].iov_len == source->iov[i].iov_len);
+        memcpy(dest->iov[i].iov_base,
+               source->iov[i].iov_base,
+               source->iov[i].iov_len);
+    }
+}
+
+static void quorum_count_vote(QuorumVotes *votes,
+                              QuorumVoteValue *value,
+                              int index)
+{
+    QuorumVoteVersion *v = NULL, *version = NULL;
+    QuorumVoteItem *item;
+
+    /* look if we have something with this hash */
+    QLIST_FOREACH(v, &votes->vote_list, next) {
+        if (votes->compare(&v->value, value)) {
+            version = v;
+            break;
+        }
+    }
+
+    /* It's a version not yet in the list add it */
+    if (!version) {
+        version = g_new0(QuorumVoteVersion, 1);
+        QLIST_INIT(&version->items);
+        memcpy(&version->value, value, sizeof(version->value));
+        version->index = index;
+        version->vote_count = 0;
+        QLIST_INSERT_HEAD(&votes->vote_list, version, next);
+    }
+
+    version->vote_count++;
+
+    item = g_new0(QuorumVoteItem, 1);
+    item->index = index;
+    QLIST_INSERT_HEAD(&version->items, item, next);
+}
+
+static void quorum_free_vote_list(QuorumVotes *votes)
+{
+    QuorumVoteVersion *version, *next_version;
+    QuorumVoteItem *item, *next_item;
+
+    QLIST_FOREACH_SAFE(version, &votes->vote_list, next, next_version) {
+        QLIST_REMOVE(version, next);
+        QLIST_FOREACH_SAFE(item, &version->items, next, next_item) {
+            QLIST_REMOVE(item, next);
+            g_free(item);
+        }
+        g_free(version);
+    }
+}
+
+static int quorum_compute_hash(QuorumAIOCB *acb, int i, QuorumVoteValue *hash)
+{
+    int j, ret;
+    gnutls_hash_hd_t dig;
+    QEMUIOVector *qiov = &acb->qcrs[i].qiov;
+
+    ret = gnutls_hash_init(&dig, GNUTLS_DIG_SHA256);
+
+    if (ret < 0) {
+        return ret;
+    }
+
+    for (j = 0; j < qiov->niov; j++) {
+        ret = gnutls_hash(dig, qiov->iov[j].iov_base, qiov->iov[j].iov_len);
+        if (ret < 0) {
+            break;
+        }
+    }
+
+    gnutls_hash_deinit(dig, (void *) hash);
+    return ret;
+}
+
+static QuorumVoteVersion *quorum_get_vote_winner(QuorumVotes *votes)
+{
+    int max = 0;
+    QuorumVoteVersion *candidate, *winner = NULL;
+
+    QLIST_FOREACH(candidate, &votes->vote_list, next) {
+        if (candidate->vote_count > max) {
+            max = candidate->vote_count;
+            winner = candidate;
+        }
+    }
+
+    return winner;
+}
+
+/* qemu_iovec_compare is handy for blkverify mode because it returns the first
+ * differing byte location. Yet it is handcoded to compare vectors one byte
+ * after another so it does not benefit from the libc SIMD optimizations.
+ * quorum_iovec_compare is written for speed and should be used in the non
+ * blkverify mode of quorum.
+ */
+static bool quorum_iovec_compare(QEMUIOVector *a, QEMUIOVector *b)
+{
+    int i;
+    int result;
+
+    assert(a->niov == b->niov);
+    for (i = 0; i < a->niov; i++) {
+        assert(a->iov[i].iov_len == b->iov[i].iov_len);
+        result = memcmp(a->iov[i].iov_base,
+                        b->iov[i].iov_base,
+                        a->iov[i].iov_len);
+        if (result) {
+            return false;
+        }
+    }
+
+    return true;
+}
+
+static void GCC_FMT_ATTR(2, 3) quorum_err(QuorumAIOCB *acb,
+                                          const char *fmt, ...)
+{
+    va_list ap;
+
+    va_start(ap, fmt);
+    fprintf(stderr, "quorum: sector_num=%" PRId64 " nb_sectors=%d ",
+            acb->sector_num, acb->nb_sectors);
+    vfprintf(stderr, fmt, ap);
+    fprintf(stderr, "\n");
+    va_end(ap);
+    exit(1);
+}
+
+static bool quorum_compare(QuorumAIOCB *acb,
+                           QEMUIOVector *a,
+                           QEMUIOVector *b)
+{
+    BDRVQuorumState *s = acb->common.bs->opaque;
+    ssize_t offset;
+
+    /* This driver will replace blkverify in this particular case */
+    if (s->is_blkverify) {
+        offset = qemu_iovec_compare(a, b);
+        if (offset != -1) {
+            quorum_err(acb, "contents mismatch in sector %" PRId64,
+                       acb->sector_num +
+                       (uint64_t)(offset / BDRV_SECTOR_SIZE));
+        }
+        return true;
+    }
+
+    return quorum_iovec_compare(a, b);
+}
+
+/* Do a vote to get the error code */
+static int quorum_vote_error(QuorumAIOCB *acb)
+{
+    BDRVQuorumState *s = acb->common.bs->opaque;
+    QuorumVoteVersion *winner = NULL;
+    QuorumVotes error_votes;
+    QuorumVoteValue result_value;
+    int i, ret = 0;
+    bool error = false;
+
+    QLIST_INIT(&error_votes.vote_list);
+    error_votes.compare = quorum_64bits_compare;
+
+    for (i = 0; i < s->num_children; i++) {
+        ret = acb->qcrs[i].ret;
+        if (ret) {
+            error = true;
+            result_value.l = ret;
+            quorum_count_vote(&error_votes, &result_value, i);
+        }
+    }
+
+    if (error) {
+        winner = quorum_get_vote_winner(&error_votes);
+        ret = winner->value.l;
+    }
+
+    quorum_free_vote_list(&error_votes);
+
+    return ret;
+}
+
+static void quorum_vote(QuorumAIOCB *acb)
+{
+    bool quorum = true;
+    int i, j, ret;
+    QuorumVoteValue hash;
+    BDRVQuorumState *s = acb->common.bs->opaque;
+    QuorumVoteVersion *winner;
+
+    if (quorum_has_too_much_io_failed(acb)) {
+        return;
+    }
+
+    /* get the index of the first successful read */
+    for (i = 0; i < s->num_children; i++) {
+        if (!acb->qcrs[i].ret) {
+            break;
+        }
+    }
+
+    assert(i < s->num_children);
+
+    /* compare this read with all other successful reads stopping at quorum
+     * failure
+     */
+    for (j = i + 1; j < s->num_children; j++) {
+        if (acb->qcrs[j].ret) {
+            continue;
+        }
+        quorum = quorum_compare(acb, &acb->qcrs[i].qiov, &acb->qcrs[j].qiov);
+        if (!quorum) {
+            break;
+       }
+    }
+
+    /* Every successful read agrees */
+    if (quorum) {
+        quorum_copy_qiov(acb->qiov, &acb->qcrs[i].qiov);
+        return;
+    }
+
+    /* compute hashes for each successful read, also store indexes */
+    for (i = 0; i < s->num_children; i++) {
+        if (acb->qcrs[i].ret) {
+            continue;
+        }
+        ret = quorum_compute_hash(acb, i, &hash);
+        /* if ever the hash computation failed */
+        if (ret < 0) {
+            acb->vote_ret = ret;
+            goto free_exit;
+        }
+        quorum_count_vote(&acb->votes, &hash, i);
+    }
+
+    /* vote to select the most represented version */
+    winner = quorum_get_vote_winner(&acb->votes);
+
+    /* if the winner count is smaller than threshold the read fails */
+    if (winner->vote_count < s->threshold) {
+        quorum_report_failure(acb);
+        acb->vote_ret = -EIO;
+        goto free_exit;
+    }
+
+    /* we have a winner: copy it */
+    quorum_copy_qiov(acb->qiov, &acb->qcrs[winner->index].qiov);
+
+    /* some versions are bad print them */
+    quorum_report_bad_versions(s, acb, &winner->value);
+
+free_exit:
+    /* free lists */
+    quorum_free_vote_list(&acb->votes);
+}
+
+static BlockDriverAIOCB *quorum_aio_readv(BlockDriverState *bs,
+                                         int64_t sector_num,
+                                         QEMUIOVector *qiov,
+                                         int nb_sectors,
+                                         BlockDriverCompletionFunc *cb,
+                                         void *opaque)
+{
+    BDRVQuorumState *s = bs->opaque;
+    QuorumAIOCB *acb = quorum_aio_get(s, bs, qiov, sector_num,
+                                      nb_sectors, cb, opaque);
+    int i;
+
+    acb->is_read = true;
+
+    for (i = 0; i < s->num_children; i++) {
+        acb->qcrs[i].buf = qemu_blockalign(s->bs[i], qiov->size);
+        qemu_iovec_init(&acb->qcrs[i].qiov, qiov->niov);
+        qemu_iovec_clone(&acb->qcrs[i].qiov, qiov, acb->qcrs[i].buf);
+    }
+
+    for (i = 0; i < s->num_children; i++) {
+        bdrv_aio_readv(s->bs[i], sector_num, &acb->qcrs[i].qiov, nb_sectors,
+                       quorum_aio_cb, &acb->qcrs[i]);
+    }
+
+    return &acb->common;
+}
+
+static BlockDriverAIOCB *quorum_aio_writev(BlockDriverState *bs,
+                                          int64_t sector_num,
+                                          QEMUIOVector *qiov,
+                                          int nb_sectors,
+                                          BlockDriverCompletionFunc *cb,
+                                          void *opaque)
+{
+    BDRVQuorumState *s = bs->opaque;
+    QuorumAIOCB *acb = quorum_aio_get(s, bs, qiov, sector_num, nb_sectors,
+                                      cb, opaque);
+    int i;
+
+    for (i = 0; i < s->num_children; i++) {
+        acb->qcrs[i].aiocb = bdrv_aio_writev(s->bs[i], sector_num, qiov,
+                                             nb_sectors, &quorum_aio_cb,
+                                             &acb->qcrs[i]);
+    }
+
+    return &acb->common;
+}
+
+static int64_t quorum_getlength(BlockDriverState *bs)
+{
+    BDRVQuorumState *s = bs->opaque;
+    int64_t result;
+    int i;
+
+    /* check that all file have the same length */
+    result = bdrv_getlength(s->bs[0]);
+    if (result < 0) {
+        return result;
+    }
+    for (i = 1; i < s->num_children; i++) {
+        int64_t value = bdrv_getlength(s->bs[i]);
+        if (value < 0) {
+            return value;
+        }
+        if (value != result) {
+            return -EIO;
+        }
+    }
+
+    return result;
+}
+
+static void quorum_invalidate_cache(BlockDriverState *bs)
+{
+    BDRVQuorumState *s = bs->opaque;
+    int i;
+
+    for (i = 0; i < s->num_children; i++) {
+        bdrv_invalidate_cache(s->bs[i]);
+    }
+}
+
+static coroutine_fn int quorum_co_flush(BlockDriverState *bs)
+{
+    BDRVQuorumState *s = bs->opaque;
+    QuorumVoteVersion *winner = NULL;
+    QuorumVotes error_votes;
+    QuorumVoteValue result_value;
+    int i;
+    int result = 0;
+
+    QLIST_INIT(&error_votes.vote_list);
+    error_votes.compare = quorum_64bits_compare;
+
+    for (i = 0; i < s->num_children; i++) {
+        result = bdrv_co_flush(s->bs[i]);
+        result_value.l = result;
+        quorum_count_vote(&error_votes, &result_value, i);
+    }
+
+    winner = quorum_get_vote_winner(&error_votes);
+    result = winner->value.l;
+
+    quorum_free_vote_list(&error_votes);
+
+    return result;
+}
+
+static bool quorum_recurse_is_first_non_filter(BlockDriverState *bs,
+                                               BlockDriverState *candidate)
+{
+    BDRVQuorumState *s = bs->opaque;
+    int i;
+
+    for (i = 0; i < s->num_children; i++) {
+        bool perm = bdrv_recurse_is_first_non_filter(s->bs[i],
+                                                     candidate);
+        if (perm) {
+            return true;
+        }
+    }
+
+    return false;
+}
+
+static int quorum_valid_threshold(int threshold, int num_children, Error **errp)
+{
+
+    if (threshold < 1) {
+        error_set(errp, QERR_INVALID_PARAMETER_VALUE,
+                  "vote-threshold", "value >= 1");
+        return -ERANGE;
+    }
+
+    if (threshold > num_children) {
+        error_setg(errp, "threshold may not exceed children count");
+        return -ERANGE;
+    }
+
+    return 0;
+}
+
+static QemuOptsList quorum_runtime_opts = {
+    .name = "quorum",
+    .head = QTAILQ_HEAD_INITIALIZER(quorum_runtime_opts.head),
+    .desc = {
+        {
+            .name = QUORUM_OPT_VOTE_THRESHOLD,
+            .type = QEMU_OPT_NUMBER,
+            .help = "The number of vote needed for reaching quorum",
+        },
+        {
+            .name = QUORUM_OPT_BLKVERIFY,
+            .type = QEMU_OPT_BOOL,
+            .help = "Trigger block verify mode if set",
+        },
+        { /* end of list */ }
+    },
+};
+
+static int quorum_open(BlockDriverState *bs, QDict *options, int flags,
+                       Error **errp)
+{
+    BDRVQuorumState *s = bs->opaque;
+    Error *local_err = NULL;
+    QemuOpts *opts;
+    bool *opened;
+    QDict *sub = NULL;
+    QList *list = NULL;
+    const QListEntry *lentry;
+    int i;
+    int ret = 0;
+
+    qdict_flatten(options);
+    qdict_extract_subqdict(options, &sub, "children.");
+    qdict_array_split(sub, &list);
+
+    if (qdict_size(sub)) {
+        error_setg(&local_err, "Invalid option children.%s",
+                   qdict_first(sub)->key);
+        ret = -EINVAL;
+        goto exit;
+    }
+
+    /* count how many different children are present */
+    s->num_children = qlist_size(list);
+    if (s->num_children < 2) {
+        error_setg(&local_err,
+                   "Number of provided children must be greater than 1");
+        ret = -EINVAL;
+        goto exit;
+    }
+
+    opts = qemu_opts_create(&quorum_runtime_opts, NULL, 0, &error_abort);
+    qemu_opts_absorb_qdict(opts, options, &local_err);
+    if (error_is_set(&local_err)) {
+        ret = -EINVAL;
+        goto exit;
+    }
+
+    s->threshold = qemu_opt_get_number(opts, QUORUM_OPT_VOTE_THRESHOLD, 0);
+
+    /* and validate it against s->num_children */
+    ret = quorum_valid_threshold(s->threshold, s->num_children, &local_err);
+    if (ret < 0) {
+        goto exit;
+    }
+
+    /* is the driver in blkverify mode */
+    if (qemu_opt_get_bool(opts, QUORUM_OPT_BLKVERIFY, false) &&
+        s->num_children == 2 && s->threshold == 2) {
+        s->is_blkverify = true;
+    } else if (qemu_opt_get_bool(opts, QUORUM_OPT_BLKVERIFY, false)) {
+        fprintf(stderr, "blkverify mode is set by setting blkverify=on "
+                "and using two files with vote_threshold=2\n");
+    }
+
+    /* allocate the children BlockDriverState array */
+    s->bs = g_new0(BlockDriverState *, s->num_children);
+    opened = g_new0(bool, s->num_children);
+
+    for (i = 0, lentry = qlist_first(list); lentry;
+         lentry = qlist_next(lentry), i++) {
+        QDict *d;
+        QString *string;
+
+        switch (qobject_type(lentry->value))
+        {
+            /* List of options */
+            case QTYPE_QDICT:
+                d = qobject_to_qdict(lentry->value);
+                QINCREF(d);
+                ret = bdrv_open(&s->bs[i], NULL, NULL, d, flags, NULL,
+                                &local_err);
+                break;
+
+            /* QMP reference */
+            case QTYPE_QSTRING:
+                string = qobject_to_qstring(lentry->value);
+                ret = bdrv_open(&s->bs[i], NULL, qstring_get_str(string), NULL,
+                                flags, NULL, &local_err);
+                break;
+
+            default:
+                error_setg(&local_err, "Specification of child block device %i "
+                           "is invalid", i);
+                ret = -EINVAL;
+        }
+
+        if (ret < 0) {
+            goto close_exit;
+        }
+        opened[i] = true;
+    }
+
+    g_free(opened);
+    goto exit;
+
+close_exit:
+    /* cleanup on error */
+    for (i = 0; i < s->num_children; i++) {
+        if (!opened[i]) {
+            continue;
+        }
+        bdrv_unref(s->bs[i]);
+    }
+    g_free(s->bs);
+    g_free(opened);
+exit:
+    /* propagate error */
+    if (error_is_set(&local_err)) {
+        error_propagate(errp, local_err);
+    }
+    QDECREF(list);
+    QDECREF(sub);
+    return ret;
+}
+
+static void quorum_close(BlockDriverState *bs)
+{
+    BDRVQuorumState *s = bs->opaque;
+    int i;
+
+    for (i = 0; i < s->num_children; i++) {
+        bdrv_unref(s->bs[i]);
+    }
+
+    g_free(s->bs);
+}
+
+static BlockDriver bdrv_quorum = {
+    .format_name        = "quorum",
+    .protocol_name      = "quorum",
+
+    .instance_size      = sizeof(BDRVQuorumState),
+
+    .bdrv_file_open     = quorum_open,
+    .bdrv_close         = quorum_close,
+
+    .authorizations     = { true, true },
+
+    .bdrv_co_flush_to_disk = quorum_co_flush,
+
+    .bdrv_getlength     = quorum_getlength,
+
+    .bdrv_aio_readv     = quorum_aio_readv,
+    .bdrv_aio_writev    = quorum_aio_writev,
+    .bdrv_invalidate_cache = quorum_invalidate_cache,
+
+    .bdrv_recurse_is_first_non_filter = quorum_recurse_is_first_non_filter,
+};
+
+static void bdrv_quorum_init(void)
+{
+    bdrv_register(&bdrv_quorum);
+}
+
+block_init(bdrv_quorum_init);
diff --git a/block/sheepdog.c b/block/sheepdog.c
index e6c0376566..f7bd0242e5 100644
--- a/block/sheepdog.c
+++ b/block/sheepdog.c
@@ -1534,7 +1534,8 @@ static int sd_prealloc(const char *filename)
     Error *local_err = NULL;
     int ret;
 
-    ret = bdrv_file_open(&bs, filename, NULL, NULL, BDRV_O_RDWR, &local_err);
+    ret = bdrv_open(&bs, filename, NULL, NULL, BDRV_O_RDWR | BDRV_O_PROTOCOL,
+                    NULL, &local_err);
     if (ret < 0) {
         qerror_report_err(local_err);
         error_free(local_err);
@@ -1695,7 +1696,9 @@ static int sd_create(const char *filename, QEMUOptionParameter *options,
             goto out;
         }
 
-        ret = bdrv_file_open(&bs, backing_file, NULL, NULL, 0, &local_err);
+        bs = NULL;
+        ret = bdrv_open(&bs, backing_file, NULL, NULL, BDRV_O_PROTOCOL, NULL,
+                        &local_err);
         if (ret < 0) {
             qerror_report_err(local_err);
             error_free(local_err);
diff --git a/block/vdi.c b/block/vdi.c
index 2d7490f173..ae49cd83ca 100644
--- a/block/vdi.c
+++ b/block/vdi.c
@@ -395,43 +395,50 @@ static int vdi_open(BlockDriverState *bs, QDict *options, int flags,
     }
 
     if (header.signature != VDI_SIGNATURE) {
-        logout("bad vdi signature %08x\n", header.signature);
-        ret = -EMEDIUMTYPE;
+        error_setg(errp, "Image not in VDI format (bad signature %08x)", header.signature);
+        ret = -EINVAL;
         goto fail;
     } else if (header.version != VDI_VERSION_1_1) {
-        logout("unsupported version %u.%u\n",
-               header.version >> 16, header.version & 0xffff);
+        error_setg(errp, "unsupported VDI image (version %u.%u)",
+                   header.version >> 16, header.version & 0xffff);
         ret = -ENOTSUP;
         goto fail;
     } else if (header.offset_bmap % SECTOR_SIZE != 0) {
         /* We only support block maps which start on a sector boundary. */
-        logout("unsupported block map offset 0x%x B\n", header.offset_bmap);
+        error_setg(errp, "unsupported VDI image (unaligned block map offset "
+                   "0x%x)", header.offset_bmap);
         ret = -ENOTSUP;
         goto fail;
     } else if (header.offset_data % SECTOR_SIZE != 0) {
         /* We only support data blocks which start on a sector boundary. */
-        logout("unsupported data offset 0x%x B\n", header.offset_data);
+        error_setg(errp, "unsupported VDI image (unaligned data offset 0x%x)",
+                   header.offset_data);
         ret = -ENOTSUP;
         goto fail;
     } else if (header.sector_size != SECTOR_SIZE) {
-        logout("unsupported sector size %u B\n", header.sector_size);
+        error_setg(errp, "unsupported VDI image (sector size %u is not %u)",
+                   header.sector_size, SECTOR_SIZE);
         ret = -ENOTSUP;
         goto fail;
     } else if (header.block_size != 1 * MiB) {
-        logout("unsupported block size %u B\n", header.block_size);
+        error_setg(errp, "unsupported VDI image (sector size %u is not %u)",
+                   header.block_size, 1 * MiB);
         ret = -ENOTSUP;
         goto fail;
     } else if (header.disk_size >
                (uint64_t)header.blocks_in_image * header.block_size) {
-        logout("unsupported disk size %" PRIu64 " B\n", header.disk_size);
+        error_setg(errp, "unsupported VDI image (disk size %" PRIu64 ", "
+                   "image bitmap has room for %" PRIu64 ")",
+                   header.disk_size,
+                   (uint64_t)header.blocks_in_image * header.block_size);
         ret = -ENOTSUP;
         goto fail;
     } else if (!uuid_is_null(header.uuid_link)) {
-        logout("link uuid != 0, unsupported\n");
+        error_setg(errp, "unsupported VDI image (non-NULL link UUID)");
         ret = -ENOTSUP;
         goto fail;
     } else if (!uuid_is_null(header.uuid_parent)) {
-        logout("parent uuid != 0, unsupported\n");
+        error_setg(errp, "unsupported VDI image (non-NULL parent UUID)");
         ret = -ENOTSUP;
         goto fail;
     }
diff --git a/block/vhdx.c b/block/vhdx.c
index 55689cf641..5390ba6d0f 100644
--- a/block/vhdx.c
+++ b/block/vhdx.c
@@ -402,9 +402,10 @@ int vhdx_update_headers(BlockDriverState *bs, BDRVVHDXState *s,
 }
 
 /* opens the specified header block from the VHDX file header section */
-static int vhdx_parse_header(BlockDriverState *bs, BDRVVHDXState *s)
+static void vhdx_parse_header(BlockDriverState *bs, BDRVVHDXState *s,
+                              Error **errp)
 {
-    int ret = 0;
+    int ret;
     VHDXHeader *header1;
     VHDXHeader *header2;
     bool h1_valid = false;
@@ -462,7 +463,6 @@ static int vhdx_parse_header(BlockDriverState *bs, BDRVVHDXState *s)
     } else if (!h1_valid && h2_valid) {
         s->curr_header = 1;
     } else if (!h1_valid && !h2_valid) {
-        ret = -EINVAL;
         goto fail;
     } else {
         /* If both headers are valid, then we choose the active one by the
@@ -473,27 +473,22 @@ static int vhdx_parse_header(BlockDriverState *bs, BDRVVHDXState *s)
         } else if (h2_seq > h1_seq) {
             s->curr_header = 1;
         } else {
-            ret = -EINVAL;
             goto fail;
         }
     }
 
     vhdx_region_register(s, s->headers[s->curr_header]->log_offset,
                             s->headers[s->curr_header]->log_length);
-
-    ret = 0;
-
     goto exit;
 
 fail:
-    qerror_report(ERROR_CLASS_GENERIC_ERROR, "No valid VHDX header found");
+    error_setg_errno(errp, -ret, "No valid VHDX header found");
     qemu_vfree(header1);
     qemu_vfree(header2);
     s->headers[0] = NULL;
     s->headers[1] = NULL;
 exit:
     qemu_vfree(buffer);
-    return ret;
 }
 
 
@@ -878,7 +873,7 @@ static int vhdx_open(BlockDriverState *bs, QDict *options, int flags,
     int ret = 0;
     uint32_t i;
     uint64_t signature;
-
+    Error *local_err = NULL;
 
     s->bat = NULL;
     s->first_visible_write = true;
@@ -901,8 +896,10 @@ static int vhdx_open(BlockDriverState *bs, QDict *options, int flags,
      * header update */
     vhdx_guid_generate(&s->session_guid);
 
-    ret = vhdx_parse_header(bs, s);
-    if (ret < 0) {
+    vhdx_parse_header(bs, s, &local_err);
+    if (local_err != NULL) {
+        error_propagate(errp, local_err);
+        ret = -EINVAL;
         goto fail;
     }
 
@@ -1797,7 +1794,9 @@ static int vhdx_create(const char *filename, QEMUOptionParameter *options,
         goto exit;
     }
 
-    ret = bdrv_file_open(&bs, filename, NULL, NULL, BDRV_O_RDWR, &local_err);
+    bs = NULL;
+    ret = bdrv_open(&bs, filename, NULL, NULL, BDRV_O_RDWR | BDRV_O_PROTOCOL,
+                    NULL, &local_err);
     if (ret < 0) {
         error_propagate(errp, local_err);
         goto exit;
diff --git a/block/vmdk.c b/block/vmdk.c
index ff6f5ee911..83839f9b7a 100644
--- a/block/vmdk.c
+++ b/block/vmdk.c
@@ -526,8 +526,34 @@ static int vmdk_open_vmfs_sparse(BlockDriverState *bs,
     return ret;
 }
 
-static int vmdk_open_desc_file(BlockDriverState *bs, int flags,
-                               uint64_t desc_offset, Error **errp);
+static int vmdk_open_desc_file(BlockDriverState *bs, int flags, char *buf,
+                               Error **errp);
+
+static char *vmdk_read_desc(BlockDriverState *file, uint64_t desc_offset,
+                            Error **errp)
+{
+    int64_t size;
+    char *buf;
+    int ret;
+
+    size = bdrv_getlength(file);
+    if (size < 0) {
+        error_setg_errno(errp, -size, "Could not access file");
+        return NULL;
+    }
+
+    size = MIN(size, 1 << 20);  /* avoid unbounded allocation */
+    buf = g_malloc0(size + 1);
+
+    ret = bdrv_pread(file, desc_offset, buf, size);
+    if (ret < 0) {
+        error_setg_errno(errp, -ret, "Could not read from file");
+        g_free(buf);
+        return NULL;
+    }
+
+    return buf;
+}
 
 static int vmdk_open_vmdk4(BlockDriverState *bs,
                            BlockDriverState *file,
@@ -546,11 +572,18 @@ static int vmdk_open_vmdk4(BlockDriverState *bs,
         error_setg_errno(errp, -ret,
                          "Could not read header from file '%s'",
                          file->filename);
+        return -EINVAL;
     }
     if (header.capacity == 0) {
         uint64_t desc_offset = le64_to_cpu(header.desc_offset);
         if (desc_offset) {
-            return vmdk_open_desc_file(bs, flags, desc_offset << 9, errp);
+            char *buf = vmdk_read_desc(file, desc_offset << 9, errp);
+            if (!buf) {
+                return -EINVAL;
+            }
+            ret = vmdk_open_desc_file(bs, flags, buf, errp);
+            g_free(buf);
+            return ret;
         }
     }
 
@@ -609,8 +642,8 @@ static int vmdk_open_vmdk4(BlockDriverState *bs,
         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);
+        error_set(errp, 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
@@ -622,7 +655,7 @@ static int vmdk_open_vmdk4(BlockDriverState *bs,
     }
 
     if (le32_to_cpu(header.num_gtes_per_gt) > 512) {
-        error_report("L2 table size too big");
+        error_setg(errp, "L2 table size too big");
         return -EINVAL;
     }
 
@@ -638,8 +671,8 @@ static int vmdk_open_vmdk4(BlockDriverState *bs,
     }
     if (bdrv_getlength(file) <
             le64_to_cpu(header.grain_offset) * BDRV_SECTOR_SIZE) {
-        error_report("File truncated, expecting at least %lld bytes",
-                le64_to_cpu(header.grain_offset) * BDRV_SECTOR_SIZE);
+        error_setg(errp, "File truncated, expecting at least %lld bytes",
+                   le64_to_cpu(header.grain_offset) * BDRV_SECTOR_SIZE);
         return -EINVAL;
     }
 
@@ -701,16 +734,12 @@ static int vmdk_parse_description(const char *desc, const char *opt_name,
 
 /* Open an extent file and append to bs array */
 static int vmdk_open_sparse(BlockDriverState *bs,
-                            BlockDriverState *file,
-                            int flags, Error **errp)
+                            BlockDriverState *file, int flags,
+                            char *buf, Error **errp)
 {
     uint32_t magic;
 
-    if (bdrv_pread(file, 0, &magic, sizeof(magic)) != sizeof(magic)) {
-        return -EIO;
-    }
-
-    magic = be32_to_cpu(magic);
+    magic = ldl_be_p(buf);
     switch (magic) {
         case VMDK3_MAGIC:
             return vmdk_open_vmfs_sparse(bs, file, flags, errp);
@@ -719,7 +748,8 @@ static int vmdk_open_sparse(BlockDriverState *bs,
             return vmdk_open_vmdk4(bs, file, flags, errp);
             break;
         default:
-            return -EMEDIUMTYPE;
+            error_setg(errp, "Image not in VMDK format");
+            return -EINVAL;
             break;
     }
 }
@@ -776,8 +806,9 @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs,
 
         path_combine(extent_path, sizeof(extent_path),
                 desc_file_path, fname);
-        ret = bdrv_file_open(&extent_file, extent_path, NULL, NULL,
-                             bs->open_flags, errp);
+        extent_file = NULL;
+        ret = bdrv_open(&extent_file, extent_path, NULL, NULL,
+                        bs->open_flags | BDRV_O_PROTOCOL, NULL, errp);
         if (ret) {
             return ret;
         }
@@ -794,8 +825,14 @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs,
             extent->flat_start_offset = flat_offset << 9;
         } else if (!strcmp(type, "SPARSE") || !strcmp(type, "VMFSSPARSE")) {
             /* SPARSE extent and VMFSSPARSE extent are both "COWD" sparse file*/
-            ret = vmdk_open_sparse(bs, extent_file, bs->open_flags, errp);
+            char *buf = vmdk_read_desc(extent_file, 0, errp);
+            if (!buf) {
+                ret = -EINVAL;
+            } else {
+                ret = vmdk_open_sparse(bs, extent_file, bs->open_flags, buf, errp);
+            }
             if (ret) {
+                g_free(buf);
                 bdrv_unref(extent_file);
                 return ret;
             }
@@ -818,29 +855,16 @@ next_line:
     return 0;
 }
 
-static int vmdk_open_desc_file(BlockDriverState *bs, int flags,
-                               uint64_t desc_offset, Error **errp)
+static int vmdk_open_desc_file(BlockDriverState *bs, int flags, char *buf,
+                               Error **errp)
 {
     int ret;
-    char *buf = NULL;
     char ct[128];
     BDRVVmdkState *s = bs->opaque;
-    int64_t size;
-
-    size = bdrv_getlength(bs->file);
-    if (size < 0) {
-        return -EINVAL;
-    }
-
-    size = MIN(size, 1 << 20);  /* avoid unbounded allocation */
-    buf = g_malloc0(size + 1);
 
-    ret = bdrv_pread(bs->file, desc_offset, buf, size);
-    if (ret < 0) {
-        goto exit;
-    }
     if (vmdk_parse_description(buf, "createType", ct, sizeof(ct))) {
-        ret = -EMEDIUMTYPE;
+        error_setg(errp, "invalid VMDK image descriptor");
+        ret = -EINVAL;
         goto exit;
     }
     if (strcmp(ct, "monolithicFlat") &&
@@ -856,24 +880,37 @@ static int vmdk_open_desc_file(BlockDriverState *bs, int flags,
     s->desc_offset = 0;
     ret = vmdk_parse_extents(buf, bs, bs->file->filename, errp);
 exit:
-    g_free(buf);
     return ret;
 }
 
 static int vmdk_open(BlockDriverState *bs, QDict *options, int flags,
                      Error **errp)
 {
+    char *buf = NULL;
     int ret;
     BDRVVmdkState *s = bs->opaque;
+    uint32_t magic;
 
-    if (vmdk_open_sparse(bs, bs->file, flags, errp) == 0) {
-        s->desc_offset = 0x200;
-    } else {
-        ret = vmdk_open_desc_file(bs, flags, 0, errp);
-        if (ret) {
-            goto fail;
-        }
+    buf = vmdk_read_desc(bs->file, 0, errp);
+    if (!buf) {
+        return -EINVAL;
     }
+
+    magic = ldl_be_p(buf);
+    switch (magic) {
+        case VMDK3_MAGIC:
+        case VMDK4_MAGIC:
+            ret = vmdk_open_sparse(bs, bs->file, flags, buf, errp);
+            s->desc_offset = 0x200;
+            break;
+        default:
+            ret = vmdk_open_desc_file(bs, flags, buf, errp);
+            break;
+    }
+    if (ret) {
+        goto fail;
+    }
+
     /* try to open parent images, if exist */
     ret = vmdk_parent_open(bs);
     if (ret) {
@@ -888,10 +925,11 @@ static int vmdk_open(BlockDriverState *bs, QDict *options, int flags,
               QERR_BLOCK_FORMAT_FEATURE_NOT_SUPPORTED,
               "vmdk", bs->device_name, "live migration");
     migrate_add_blocker(s->migration_blocker);
-
+    g_free(buf);
     return 0;
 
 fail:
+    g_free(buf);
     g_free(s->create_type);
     s->create_type = NULL;
     vmdk_free_extents(bs);
@@ -1493,7 +1531,9 @@ static int vmdk_create_extent(const char *filename, int64_t filesize,
         goto exit;
     }
 
-    ret = bdrv_file_open(&bs, filename, NULL, NULL, BDRV_O_RDWR, &local_err);
+    assert(bs == NULL);
+    ret = bdrv_open(&bs, filename, NULL, NULL, BDRV_O_RDWR | BDRV_O_PROTOCOL,
+                    NULL, &local_err);
     if (ret < 0) {
         error_propagate(errp, local_err);
         goto exit;
@@ -1755,10 +1795,10 @@ static int vmdk_create(const char *filename, QEMUOptionParameter *options,
         goto exit;
     }
     if (backing_file) {
-        BlockDriverState *bs = bdrv_new("");
-        ret = bdrv_open(bs, backing_file, NULL, BDRV_O_NO_BACKING, NULL, errp);
+        BlockDriverState *bs = NULL;
+        ret = bdrv_open(&bs, backing_file, NULL, NULL, BDRV_O_NO_BACKING, NULL,
+                        errp);
         if (ret != 0) {
-            bdrv_unref(bs);
             goto exit;
         }
         if (strcmp(bs->drv->format_name, "vmdk")) {
@@ -1831,7 +1871,9 @@ static int vmdk_create(const char *filename, QEMUOptionParameter *options,
             goto exit;
         }
     }
-    ret = bdrv_file_open(&new_bs, filename, NULL, NULL, BDRV_O_RDWR, &local_err);
+    assert(new_bs == NULL);
+    ret = bdrv_open(&new_bs, filename, NULL, NULL,
+                    BDRV_O_RDWR | BDRV_O_PROTOCOL, NULL, &local_err);
     if (ret < 0) {
         error_setg_errno(errp, -ret, "Could not write description");
         goto exit;
diff --git a/block/vpc.c b/block/vpc.c
index 1d326cbf44..82bf2485a5 100644
--- a/block/vpc.c
+++ b/block/vpc.c
@@ -190,7 +190,8 @@ static int vpc_open(BlockDriverState *bs, QDict *options, int flags,
             goto fail;
         }
         if (strncmp(footer->creator, "conectix", 8)) {
-            ret = -EMEDIUMTYPE;
+            error_setg(errp, "invalid VPC image");
+            ret = -EINVAL;
             goto fail;
         }
         disk_type = VHD_FIXED;
diff --git a/block/vvfat.c b/block/vvfat.c
index a19e4ca227..f966ea5da8 100644
--- a/block/vvfat.c
+++ b/block/vvfat.c
@@ -1086,16 +1086,14 @@ DLOG(if (stderr == NULL) {
     opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
     qemu_opts_absorb_qdict(opts, options, &local_err);
     if (local_err) {
-        qerror_report_err(local_err);
-        error_free(local_err);
+        error_propagate(errp, local_err);
         ret = -EINVAL;
         goto fail;
     }
 
     dirname = qemu_opt_get(opts, "dir");
     if (!dirname) {
-        qerror_report(ERROR_CLASS_GENERIC_ERROR, "vvfat block driver requires "
-                      "a 'dir' option");
+        error_setg(errp, "vvfat block driver requires a 'dir' option");
         ret = -EINVAL;
         goto fail;
     }
@@ -1135,8 +1133,7 @@ DLOG(if (stderr == NULL) {
     case 12:
         break;
     default:
-        qerror_report(ERROR_CLASS_GENERIC_ERROR, "Valid FAT types are only "
-                      "12, 16 and 32");
+        error_setg(errp, "Valid FAT types are only 12, 16 and 32");
         ret = -EINVAL;
         goto fail;
     }
@@ -2936,15 +2933,13 @@ static int enable_write_target(BDRVVVFATState *s)
         goto err;
     }
 
-    s->qcow = bdrv_new("");
-
-    ret = bdrv_open(s->qcow, s->qcow_filename, NULL,
+    s->qcow = NULL;
+    ret = bdrv_open(&s->qcow, s->qcow_filename, NULL, NULL,
             BDRV_O_RDWR | BDRV_O_CACHE_WB | BDRV_O_NO_FLUSH, bdrv_qcow,
             &local_err);
     if (ret < 0) {
         qerror_report_err(local_err);
         error_free(local_err);
-        bdrv_unref(s->qcow);
         goto err;
     }