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

9 files changed, 1344 insertions, 107 deletions

diff --git a/Makefile.objs b/Makefile.objs
index 07e46140d0..24b2f99a43 100644
--- a/Makefile.objs
+++ b/Makefile.objs
@@ -24,6 +24,7 @@ block-nested-y += parallels.o nbd.o blkdebug.o sheepdog.o blkverify.o
 block-nested-$(CONFIG_WIN32) += raw-win32.o
 block-nested-$(CONFIG_POSIX) += raw-posix.o
 block-nested-$(CONFIG_CURL) += curl.o
+block-nested-$(CONFIG_RBD) += rbd.o
 
 block-obj-y +=  $(addprefix block/, $(block-nested-y))
 
diff --git a/block.c b/block.c
index 63effd8769..e7a986c48a 100644
--- a/block.c
+++ b/block.c
@@ -215,7 +215,7 @@ int bdrv_create_file(const char* filename, QEMUOptionParameter *options)
 
     drv = bdrv_find_protocol(filename);
     if (drv == NULL) {
-        drv = bdrv_find_format("file");
+        return -ENOENT;
     }
 
     return bdrv_create(drv, filename, options);
diff --git a/block/rbd.c b/block/rbd.c
new file mode 100644
index 0000000000..249a590c98
--- /dev/null
+++ b/block/rbd.c
@@ -0,0 +1,1059 @@
+/*
+ * QEMU Block driver for RADOS (Ceph)
+ *
+ * Copyright (C) 2010 Christian Brunner <chb@muc.de>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ */
+
+#include "qemu-common.h"
+#include "qemu-error.h"
+
+#include "rbd_types.h"
+#include "block_int.h"
+
+#include <rados/librados.h>
+
+
+
+/*
+ * When specifying the image filename use:
+ *
+ * rbd:poolname/devicename
+ *
+ * poolname must be the name of an existing rados pool
+ *
+ * devicename is the basename for all objects used to
+ * emulate the raw device.
+ *
+ * Metadata information (image size, ...) is stored in an
+ * object with the name "devicename.rbd".
+ *
+ * The raw device is split into 4MB sized objects by default.
+ * The sequencenumber is encoded in a 12 byte long hex-string,
+ * and is attached to the devicename, separated by a dot.
+ * e.g. "devicename.1234567890ab"
+ *
+ */
+
+#define OBJ_MAX_SIZE (1UL << OBJ_DEFAULT_OBJ_ORDER)
+
+typedef struct RBDAIOCB {
+    BlockDriverAIOCB common;
+    QEMUBH *bh;
+    int ret;
+    QEMUIOVector *qiov;
+    char *bounce;
+    int write;
+    int64_t sector_num;
+    int aiocnt;
+    int error;
+    struct BDRVRBDState *s;
+    int cancelled;
+} RBDAIOCB;
+
+typedef struct RADOSCB {
+    int rcbid;
+    RBDAIOCB *acb;
+    struct BDRVRBDState *s;
+    int done;
+    int64_t segsize;
+    char *buf;
+    int ret;
+} RADOSCB;
+
+#define RBD_FD_READ 0
+#define RBD_FD_WRITE 1
+
+typedef struct BDRVRBDState {
+    int fds[2];
+    rados_pool_t pool;
+    rados_pool_t header_pool;
+    char name[RBD_MAX_OBJ_NAME_SIZE];
+    char block_name[RBD_MAX_BLOCK_NAME_SIZE];
+    uint64_t size;
+    uint64_t objsize;
+    int qemu_aio_count;
+    int event_reader_pos;
+    RADOSCB *event_rcb;
+} BDRVRBDState;
+
+typedef struct rbd_obj_header_ondisk RbdHeader1;
+
+static void rbd_aio_bh_cb(void *opaque);
+
+static int rbd_next_tok(char *dst, int dst_len,
+                        char *src, char delim,
+                        const char *name,
+                        char **p)
+{
+    int l;
+    char *end;
+
+    *p = NULL;
+
+    if (delim != '\0') {
+        end = strchr(src, delim);
+        if (end) {
+            *p = end + 1;
+            *end = '\0';
+        }
+    }
+    l = strlen(src);
+    if (l >= dst_len) {
+        error_report("%s too long", name);
+        return -EINVAL;
+    } else if (l == 0) {
+        error_report("%s too short", name);
+        return -EINVAL;
+    }
+
+    pstrcpy(dst, dst_len, src);
+
+    return 0;
+}
+
+static int rbd_parsename(const char *filename,
+                         char *pool, int pool_len,
+                         char *snap, int snap_len,
+                         char *name, int name_len)
+{
+    const char *start;
+    char *p, *buf;
+    int ret;
+
+    if (!strstart(filename, "rbd:", &start)) {
+        return -EINVAL;
+    }
+
+    buf = qemu_strdup(start);
+    p = buf;
+
+    ret = rbd_next_tok(pool, pool_len, p, '/', "pool name", &p);
+    if (ret < 0 || !p) {
+        ret = -EINVAL;
+        goto done;
+    }
+    ret = rbd_next_tok(name, name_len, p, '@', "object name", &p);
+    if (ret < 0) {
+        goto done;
+    }
+    if (!p) {
+        *snap = '\0';
+        goto done;
+    }
+
+    ret = rbd_next_tok(snap, snap_len, p, '\0', "snap name", &p);
+
+done:
+    qemu_free(buf);
+    return ret;
+}
+
+static int create_tmap_op(uint8_t op, const char *name, char **tmap_desc)
+{
+    uint32_t len = strlen(name);
+    uint32_t len_le = cpu_to_le32(len);
+    /* total_len = encoding op + name + empty buffer */
+    uint32_t total_len = 1 + (sizeof(uint32_t) + len) + sizeof(uint32_t);
+    uint8_t *desc = NULL;
+
+    desc = qemu_malloc(total_len);
+
+    *tmap_desc = (char *)desc;
+
+    *desc = op;
+    desc++;
+    memcpy(desc, &len_le, sizeof(len_le));
+    desc += sizeof(len_le);
+    memcpy(desc, name, len);
+    desc += len;
+    len = 0; /* no need for endian conversion for 0 */
+    memcpy(desc, &len, sizeof(len));
+    desc += sizeof(len);
+
+    return (char *)desc - *tmap_desc;
+}
+
+static void free_tmap_op(char *tmap_desc)
+{
+    qemu_free(tmap_desc);
+}
+
+static int rbd_register_image(rados_pool_t pool, const char *name)
+{
+    char *tmap_desc;
+    const char *dir = RBD_DIRECTORY;
+    int ret;
+
+    ret = create_tmap_op(CEPH_OSD_TMAP_SET, name, &tmap_desc);
+    if (ret < 0) {
+        return ret;
+    }
+
+    ret = rados_tmap_update(pool, dir, tmap_desc, ret);
+    free_tmap_op(tmap_desc);
+
+    return ret;
+}
+
+static int touch_rbd_info(rados_pool_t pool, const char *info_oid)
+{
+    int r = rados_write(pool, info_oid, 0, NULL, 0);
+    if (r < 0) {
+        return r;
+    }
+    return 0;
+}
+
+static int rbd_assign_bid(rados_pool_t pool, uint64_t *id)
+{
+    uint64_t out[1];
+    const char *info_oid = RBD_INFO;
+
+    *id = 0;
+
+    int r = touch_rbd_info(pool, info_oid);
+    if (r < 0) {
+        return r;
+    }
+
+    r = rados_exec(pool, info_oid, "rbd", "assign_bid", NULL,
+                   0, (char *)out, sizeof(out));
+    if (r < 0) {
+        return r;
+    }
+
+    le64_to_cpus(out);
+    *id = out[0];
+
+    return 0;
+}
+
+static int rbd_create(const char *filename, QEMUOptionParameter *options)
+{
+    int64_t bytes = 0;
+    int64_t objsize;
+    uint64_t size;
+    time_t mtime;
+    uint8_t obj_order = RBD_DEFAULT_OBJ_ORDER;
+    char pool[RBD_MAX_SEG_NAME_SIZE];
+    char n[RBD_MAX_SEG_NAME_SIZE];
+    char name[RBD_MAX_OBJ_NAME_SIZE];
+    char snap_buf[RBD_MAX_SEG_NAME_SIZE];
+    char *snap = NULL;
+    RbdHeader1 header;
+    rados_pool_t p;
+    uint64_t bid;
+    uint32_t hi, lo;
+    int ret;
+
+    if (rbd_parsename(filename,
+                      pool, sizeof(pool),
+                      snap_buf, sizeof(snap_buf),
+                      name, sizeof(name)) < 0) {
+        return -EINVAL;
+    }
+    if (snap_buf[0] != '\0') {
+        snap = snap_buf;
+    }
+
+    snprintf(n, sizeof(n), "%s%s", name, RBD_SUFFIX);
+
+    /* Read out options */
+    while (options && options->name) {
+        if (!strcmp(options->name, BLOCK_OPT_SIZE)) {
+            bytes = options->value.n;
+        } else if (!strcmp(options->name, BLOCK_OPT_CLUSTER_SIZE)) {
+            if (options->value.n) {
+                objsize = options->value.n;
+                if ((objsize - 1) & objsize) {    /* not a power of 2? */
+                    error_report("obj size needs to be power of 2");
+                    return -EINVAL;
+                }
+                if (objsize < 4096) {
+                    error_report("obj size too small");
+                    return -EINVAL;
+                }
+		obj_order = ffs(objsize) - 1;
+            }
+        }
+        options++;
+    }
+
+    memset(&header, 0, sizeof(header));
+    pstrcpy(header.text, sizeof(header.text), RBD_HEADER_TEXT);
+    pstrcpy(header.signature, sizeof(header.signature), RBD_HEADER_SIGNATURE);
+    pstrcpy(header.version, sizeof(header.version), RBD_HEADER_VERSION);
+    header.image_size = cpu_to_le64(bytes);
+    header.options.order = obj_order;
+    header.options.crypt_type = RBD_CRYPT_NONE;
+    header.options.comp_type = RBD_COMP_NONE;
+    header.snap_seq = 0;
+    header.snap_count = 0;
+
+    if (rados_initialize(0, NULL) < 0) {
+        error_report("error initializing");
+        return -EIO;
+    }
+
+    if (rados_open_pool(pool, &p)) {
+        error_report("error opening pool %s", pool);
+        rados_deinitialize();
+        return -EIO;
+    }
+
+    /* check for existing rbd header file */
+    ret = rados_stat(p, n, &size, &mtime);
+    if (ret == 0) {
+        ret=-EEXIST;
+        goto done;
+    }
+
+    ret = rbd_assign_bid(p, &bid);
+    if (ret < 0) {
+        error_report("failed assigning block id");
+        rados_deinitialize();
+        return -EIO;
+    }
+    hi = bid >> 32;
+    lo = bid & 0xFFFFFFFF;
+    snprintf(header.block_name, sizeof(header.block_name), "rb.%x.%x", hi, lo);
+
+    /* create header file */
+    ret = rados_write(p, n, 0, (const char *)&header, sizeof(header));
+    if (ret < 0) {
+        goto done;
+    }
+
+    ret = rbd_register_image(p, name);
+done:
+    rados_close_pool(p);
+    rados_deinitialize();
+
+    return ret;
+}
+
+/*
+ * This aio completion is being called from rbd_aio_event_reader() and
+ * runs in qemu context. It schedules a bh, but just in case the aio
+ * was not cancelled before.
+ */
+static void rbd_complete_aio(RADOSCB *rcb)
+{
+    RBDAIOCB *acb = rcb->acb;
+    int64_t r;
+
+    acb->aiocnt--;
+
+    if (acb->cancelled) {
+        if (!acb->aiocnt) {
+            qemu_vfree(acb->bounce);
+            qemu_aio_release(acb);
+        }
+        goto done;
+    }
+
+    r = rcb->ret;
+
+    if (acb->write) {
+        if (r < 0) {
+            acb->ret = r;
+            acb->error = 1;
+        } else if (!acb->error) {
+            acb->ret += rcb->segsize;
+        }
+    } else {
+        if (r == -ENOENT) {
+            memset(rcb->buf, 0, rcb->segsize);
+            if (!acb->error) {
+                acb->ret += rcb->segsize;
+            }
+        } else if (r < 0) {
+	    memset(rcb->buf, 0, rcb->segsize);
+            acb->ret = r;
+            acb->error = 1;
+        } else if (r < rcb->segsize) {
+            memset(rcb->buf + r, 0, rcb->segsize - r);
+            if (!acb->error) {
+                acb->ret += rcb->segsize;
+            }
+        } else if (!acb->error) {
+            acb->ret += r;
+        }
+    }
+    /* Note that acb->bh can be NULL in case where the aio was cancelled */
+    if (!acb->aiocnt) {
+        acb->bh = qemu_bh_new(rbd_aio_bh_cb, acb);
+        qemu_bh_schedule(acb->bh);
+    }
+done:
+    qemu_free(rcb);
+}
+
+/*
+ * aio fd read handler. It runs in the qemu context and calls the
+ * completion handling of completed rados aio operations.
+ */
+static void rbd_aio_event_reader(void *opaque)
+{
+    BDRVRBDState *s = opaque;
+
+    ssize_t ret;
+
+    do {
+        char *p = (char *)&s->event_rcb;
+
+        /* now read the rcb pointer that was sent from a non qemu thread */
+        if ((ret = read(s->fds[RBD_FD_READ], p + s->event_reader_pos,
+                        sizeof(s->event_rcb) - s->event_reader_pos)) > 0) {
+            if (ret > 0) {
+                s->event_reader_pos += ret;
+                if (s->event_reader_pos == sizeof(s->event_rcb)) {
+                    s->event_reader_pos = 0;
+                    rbd_complete_aio(s->event_rcb);
+                    s->qemu_aio_count --;
+                }
+            }
+        }
+    } while (ret < 0 && errno == EINTR);
+}
+
+static int rbd_aio_flush_cb(void *opaque)
+{
+    BDRVRBDState *s = opaque;
+
+    return (s->qemu_aio_count > 0);
+}
+
+
+static int rbd_set_snapc(rados_pool_t pool, const char *snap, RbdHeader1 *header)
+{
+    uint32_t snap_count = le32_to_cpu(header->snap_count);
+    rados_snap_t *snaps = NULL;
+    rados_snap_t seq;
+    uint32_t i;
+    uint64_t snap_names_len = le64_to_cpu(header->snap_names_len);
+    int r;
+    rados_snap_t snapid = 0;
+
+    if (snap_count) {
+        const char *header_snap = (const char *)&header->snaps[snap_count];
+        const char *end = header_snap + snap_names_len;
+        snaps = qemu_malloc(sizeof(rados_snap_t) * header->snap_count);
+
+        for (i=0; i < snap_count; i++) {
+            snaps[i] = le64_to_cpu(header->snaps[i].id);
+
+            if (snap && strcmp(snap, header_snap) == 0) {
+                snapid = snaps[i];
+            }
+
+            header_snap += strlen(header_snap) + 1;
+            if (header_snap > end) {
+                error_report("bad header, snapshot list broken");
+            }
+        }
+    }
+
+    if (snap && !snapid) {
+        error_report("snapshot not found");
+        qemu_free(snaps);
+        return -ENOENT;
+    }
+    seq = le32_to_cpu(header->snap_seq);
+
+    r = rados_set_snap_context(pool, seq, snaps, snap_count);
+
+    rados_set_snap(pool, snapid);
+
+    qemu_free(snaps);
+
+    return r;
+}
+
+#define BUF_READ_START_LEN    4096
+
+static int rbd_read_header(BDRVRBDState *s, char **hbuf)
+{
+    char *buf = NULL;
+    char n[RBD_MAX_SEG_NAME_SIZE];
+    uint64_t len = BUF_READ_START_LEN;
+    int r;
+
+    snprintf(n, sizeof(n), "%s%s", s->name, RBD_SUFFIX);
+
+    buf = qemu_malloc(len);
+
+    r = rados_read(s->header_pool, n, 0, buf, len);
+    if (r < 0) {
+        goto failed;
+    }
+
+    if (r < len) {
+        goto done;
+    }
+
+    qemu_free(buf);
+    buf = qemu_malloc(len);
+
+    r = rados_stat(s->header_pool, n, &len, NULL);
+    if (r < 0) {
+        goto failed;
+    }
+
+    r = rados_read(s->header_pool, n, 0, buf, len);
+    if (r < 0) {
+        goto failed;
+    }
+
+done:
+    *hbuf = buf;
+    return 0;
+
+failed:
+    qemu_free(buf);
+    return r;
+}
+
+static int rbd_open(BlockDriverState *bs, const char *filename, int flags)
+{
+    BDRVRBDState *s = bs->opaque;
+    RbdHeader1 *header;
+    char pool[RBD_MAX_SEG_NAME_SIZE];
+    char snap_buf[RBD_MAX_SEG_NAME_SIZE];
+    char *snap = NULL;
+    char *hbuf = NULL;
+    int r;
+
+    if (rbd_parsename(filename, pool, sizeof(pool),
+                      snap_buf, sizeof(snap_buf),
+                      s->name, sizeof(s->name)) < 0) {
+        return -EINVAL;
+    }
+    if (snap_buf[0] != '\0') {
+        snap = snap_buf;
+    }
+
+    if ((r = rados_initialize(0, NULL)) < 0) {
+        error_report("error initializing");
+        return r;
+    }
+
+    if ((r = rados_open_pool(pool, &s->pool))) {
+        error_report("error opening pool %s", pool);
+        rados_deinitialize();
+        return r;
+    }
+
+    if ((r = rados_open_pool(pool, &s->header_pool))) {
+        error_report("error opening pool %s", pool);
+        rados_deinitialize();
+        return r;
+    }
+
+    if ((r = rbd_read_header(s, &hbuf)) < 0) {
+        error_report("error reading header from %s", s->name);
+        goto failed;
+    }
+
+    if (memcmp(hbuf + 64, RBD_HEADER_SIGNATURE, 4)) {
+        error_report("Invalid header signature");
+        r = -EMEDIUMTYPE;
+        goto failed;
+    }
+
+    if (memcmp(hbuf + 68, RBD_HEADER_VERSION, 8)) {
+        error_report("Unknown image version");
+        r = -EMEDIUMTYPE;
+        goto failed;
+    }
+
+    header = (RbdHeader1 *) hbuf;
+    s->size = le64_to_cpu(header->image_size);
+    s->objsize = 1ULL << header->options.order;
+    memcpy(s->block_name, header->block_name, sizeof(header->block_name));
+
+    r = rbd_set_snapc(s->pool, snap, header);
+    if (r < 0) {
+        error_report("failed setting snap context: %s", strerror(-r));
+        goto failed;
+    }
+
+    bs->read_only = (snap != NULL);
+
+    s->event_reader_pos = 0;
+    r = qemu_pipe(s->fds);
+    if (r < 0) {
+        error_report("error opening eventfd");
+        goto failed;
+    }
+    fcntl(s->fds[0], F_SETFL, O_NONBLOCK);
+    fcntl(s->fds[1], F_SETFL, O_NONBLOCK);
+    qemu_aio_set_fd_handler(s->fds[RBD_FD_READ], rbd_aio_event_reader, NULL,
+        rbd_aio_flush_cb, NULL, s);
+
+    qemu_free(hbuf);
+
+    return 0;
+
+failed:
+    qemu_free(hbuf);
+
+    rados_close_pool(s->header_pool);
+    rados_close_pool(s->pool);
+    rados_deinitialize();
+    return r;
+}
+
+static void rbd_close(BlockDriverState *bs)
+{
+    BDRVRBDState *s = bs->opaque;
+
+    close(s->fds[0]);
+    close(s->fds[1]);
+    qemu_aio_set_fd_handler(s->fds[RBD_FD_READ], NULL , NULL, NULL, NULL,
+        NULL);
+
+    rados_close_pool(s->header_pool);
+    rados_close_pool(s->pool);
+    rados_deinitialize();
+}
+
+/*
+ * Cancel aio. Since we don't reference acb in a non qemu threads,
+ * it is safe to access it here.
+ */
+static void rbd_aio_cancel(BlockDriverAIOCB *blockacb)
+{
+    RBDAIOCB *acb = (RBDAIOCB *) blockacb;
+    acb->cancelled = 1;
+}
+
+static AIOPool rbd_aio_pool = {
+    .aiocb_size = sizeof(RBDAIOCB),
+    .cancel = rbd_aio_cancel,
+};
+
+/*
+ * This is the callback function for rados_aio_read and _write
+ *
+ * Note: this function is being called from a non qemu thread so
+ * we need to be careful about what we do here. Generally we only
+ * write to the block notification pipe, and do the rest of the
+ * io completion handling from rbd_aio_event_reader() which
+ * runs in a qemu context.
+ */
+static void rbd_finish_aiocb(rados_completion_t c, RADOSCB *rcb)
+{
+    int ret;
+    rcb->ret = rados_aio_get_return_value(c);
+    rados_aio_release(c);
+    while (1) {
+        fd_set wfd;
+        int fd = rcb->s->fds[RBD_FD_WRITE];
+
+        /* send the rcb pointer to the qemu thread that is responsible
+           for the aio completion. Must do it in a qemu thread context */
+        ret = write(fd, (void *)&rcb, sizeof(rcb));
+        if (ret >= 0) {
+            break;
+        }
+        if (errno == EINTR) {
+            continue;
+	}
+        if (errno != EAGAIN) {
+            break;
+	}
+
+        FD_ZERO(&wfd);
+        FD_SET(fd, &wfd);
+        do {
+            ret = select(fd + 1, NULL, &wfd, NULL, NULL);
+        } while (ret < 0 && errno == EINTR);
+    }
+
+    if (ret < 0) {
+        error_report("failed writing to acb->s->fds\n");
+        qemu_free(rcb);
+    }
+}
+
+/* Callback when all queued rados_aio requests are complete */
+
+static void rbd_aio_bh_cb(void *opaque)
+{
+    RBDAIOCB *acb = opaque;
+
+    if (!acb->write) {
+        qemu_iovec_from_buffer(acb->qiov, acb->bounce, acb->qiov->size);
+    }
+    qemu_vfree(acb->bounce);
+    acb->common.cb(acb->common.opaque, (acb->ret > 0 ? 0 : acb->ret));
+    qemu_bh_delete(acb->bh);
+    acb->bh = NULL;
+
+    qemu_aio_release(acb);
+}
+
+static BlockDriverAIOCB *rbd_aio_rw_vector(BlockDriverState *bs,
+                                           int64_t sector_num,
+                                           QEMUIOVector *qiov,
+                                           int nb_sectors,
+                                           BlockDriverCompletionFunc *cb,
+                                           void *opaque, int write)
+{
+    RBDAIOCB *acb;
+    RADOSCB *rcb;
+    rados_completion_t c;
+    char n[RBD_MAX_SEG_NAME_SIZE];
+    int64_t segnr, segoffs, segsize, last_segnr;
+    int64_t off, size;
+    char *buf;
+
+    BDRVRBDState *s = bs->opaque;
+
+    acb = qemu_aio_get(&rbd_aio_pool, bs, cb, opaque);
+    acb->write = write;
+    acb->qiov = qiov;
+    acb->bounce = qemu_blockalign(bs, qiov->size);
+    acb->aiocnt = 0;
+    acb->ret = 0;
+    acb->error = 0;
+    acb->s = s;
+    acb->cancelled = 0;
+    acb->bh = NULL;
+
+    if (write) {
+        qemu_iovec_to_buffer(acb->qiov, acb->bounce);
+    }
+
+    buf = acb->bounce;
+
+    off = sector_num * BDRV_SECTOR_SIZE;
+    size = nb_sectors * BDRV_SECTOR_SIZE;
+    segnr = off / s->objsize;
+    segoffs = off % s->objsize;
+    segsize = s->objsize - segoffs;
+
+    last_segnr = ((off + size - 1) / s->objsize);
+    acb->aiocnt = (last_segnr - segnr) + 1;
+
+    s->qemu_aio_count += acb->aiocnt; /* All the RADOSCB */
+
+    while (size > 0) {
+        if (size < segsize) {
+            segsize = size;
+        }
+
+        snprintf(n, sizeof(n), "%s.%012" PRIx64, s->block_name,
+                 segnr);
+
+        rcb = qemu_malloc(sizeof(RADOSCB));
+        rcb->done = 0;
+        rcb->acb = acb;
+        rcb->segsize = segsize;
+        rcb->buf = buf;
+        rcb->s = acb->s;
+
+        if (write) {
+            rados_aio_create_completion(rcb, NULL,
+                                        (rados_callback_t) rbd_finish_aiocb,
+                                        &c);
+            rados_aio_write(s->pool, n, segoffs, buf, segsize, c);
+        } else {
+            rados_aio_create_completion(rcb,
+                                        (rados_callback_t) rbd_finish_aiocb,
+                                        NULL, &c);
+            rados_aio_read(s->pool, n, segoffs, buf, segsize, c);
+        }
+
+        buf += segsize;
+        size -= segsize;
+        segoffs = 0;
+        segsize = s->objsize;
+        segnr++;
+    }
+
+    return &acb->common;
+}
+
+static BlockDriverAIOCB *rbd_aio_readv(BlockDriverState * bs,
+                                       int64_t sector_num, QEMUIOVector * qiov,
+                                       int nb_sectors,
+                                       BlockDriverCompletionFunc * cb,
+                                       void *opaque)
+{
+    return rbd_aio_rw_vector(bs, sector_num, qiov, nb_sectors, cb, opaque, 0);
+}
+
+static BlockDriverAIOCB *rbd_aio_writev(BlockDriverState * bs,
+                                        int64_t sector_num, QEMUIOVector * qiov,
+                                        int nb_sectors,
+                                        BlockDriverCompletionFunc * cb,
+                                        void *opaque)
+{
+    return rbd_aio_rw_vector(bs, sector_num, qiov, nb_sectors, cb, opaque, 1);
+}
+
+static int rbd_getinfo(BlockDriverState * bs, BlockDriverInfo * bdi)
+{
+    BDRVRBDState *s = bs->opaque;
+    bdi->cluster_size = s->objsize;
+    return 0;
+}
+
+static int64_t rbd_getlength(BlockDriverState * bs)
+{
+    BDRVRBDState *s = bs->opaque;
+
+    return s->size;
+}
+
+static int rbd_snap_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)
+{
+    BDRVRBDState *s = bs->opaque;
+    char inbuf[512], outbuf[128];
+    uint64_t snap_id;
+    int r;
+    char *p = inbuf;
+    char *end = inbuf + sizeof(inbuf);
+    char n[RBD_MAX_SEG_NAME_SIZE];
+    char *hbuf = NULL;
+    RbdHeader1 *header;
+
+    if (sn_info->name[0] == '\0') {
+        return -EINVAL; /* we need a name for rbd snapshots */
+    }
+
+    /*
+     * rbd snapshots are using the name as the user controlled unique identifier
+     * we can't use the rbd snapid for that purpose, as it can't be set
+     */
+    if (sn_info->id_str[0] != '\0' &&
+        strcmp(sn_info->id_str, sn_info->name) != 0) {
+        return -EINVAL;
+    }
+
+    if (strlen(sn_info->name) >= sizeof(sn_info->id_str)) {
+        return -ERANGE;
+    }
+
+    r = rados_selfmanaged_snap_create(s->header_pool, &snap_id);
+    if (r < 0) {
+        error_report("failed to create snap id: %s", strerror(-r));
+        return r;
+    }
+
+    *(uint32_t *)p = strlen(sn_info->name);
+    cpu_to_le32s((uint32_t *)p);
+    p += sizeof(uint32_t);
+    strncpy(p, sn_info->name, end - p);
+    p += strlen(p);
+    if (p + sizeof(snap_id) > end) {
+        error_report("invalid input parameter");
+        return -EINVAL;
+    }
+
+    *(uint64_t *)p = snap_id;
+    cpu_to_le64s((uint64_t *)p);
+
+    snprintf(n, sizeof(n), "%s%s", s->name, RBD_SUFFIX);
+
+    r = rados_exec(s->header_pool, n, "rbd", "snap_add", inbuf,
+                   sizeof(inbuf), outbuf, sizeof(outbuf));
+    if (r < 0) {
+        error_report("rbd.snap_add execution failed failed: %s", strerror(-r));
+        return r;
+    }
+
+    sprintf(sn_info->id_str, "%s", sn_info->name);
+
+    r = rbd_read_header(s, &hbuf);
+    if (r < 0) {
+        error_report("failed reading header: %s", strerror(-r));
+        return r;
+    }
+
+    header = (RbdHeader1 *) hbuf;
+    r = rbd_set_snapc(s->pool, sn_info->name, header);
+    if (r < 0) {
+        error_report("failed setting snap context: %s", strerror(-r));
+        goto failed;
+    }
+
+    return 0;
+
+failed:
+    qemu_free(header);
+    return r;
+}
+
+static int decode32(char **p, const char *end, uint32_t *v)
+{
+    if (*p + 4 > end) {
+	return -ERANGE;
+    }
+
+    *v = *(uint32_t *)(*p);
+    le32_to_cpus(v);
+    *p += 4;
+    return 0;
+}
+
+static int decode64(char **p, const char *end, uint64_t *v)
+{
+    if (*p + 8 > end) {
+        return -ERANGE;
+    }
+
+    *v = *(uint64_t *)(*p);
+    le64_to_cpus(v);
+    *p += 8;
+    return 0;
+}
+
+static int decode_str(char **p, const char *end, char **s)
+{
+    uint32_t len;
+    int r;
+
+    if ((r = decode32(p, end, &len)) < 0) {
+        return r;
+    }
+
+    *s = qemu_malloc(len + 1);
+    memcpy(*s, *p, len);
+    *p += len;
+    (*s)[len] = '\0';
+
+    return len;
+}
+
+static int rbd_snap_list(BlockDriverState *bs, QEMUSnapshotInfo **psn_tab)
+{
+    BDRVRBDState *s = bs->opaque;
+    char n[RBD_MAX_SEG_NAME_SIZE];
+    QEMUSnapshotInfo *sn_info, *sn_tab = NULL;
+    RbdHeader1 *header;
+    char *hbuf = NULL;
+    char *outbuf = NULL, *end, *buf;
+    uint64_t len;
+    uint64_t snap_seq;
+    uint32_t snap_count;
+    int r, i;
+
+    /* read header to estimate how much space we need to read the snap
+     * list */
+    if ((r = rbd_read_header(s, &hbuf)) < 0) {
+        goto done_err;
+    }
+    header = (RbdHeader1 *)hbuf;
+    len = le64_to_cpu(header->snap_names_len);
+    len += 1024; /* should have already been enough, but new snapshots might
+                    already been created since we read the header. just allocate
+                    a bit more, so that in most cases it'll suffice anyway */
+    qemu_free(hbuf);
+
+    snprintf(n, sizeof(n), "%s%s", s->name, RBD_SUFFIX);
+    while (1) {
+        qemu_free(outbuf);
+        outbuf = qemu_malloc(len);
+
+        r = rados_exec(s->header_pool, n, "rbd", "snap_list", NULL, 0,
+                       outbuf, len);
+        if (r < 0) {
+            error_report("rbd.snap_list execution failed failed: %s", strerror(-r));
+            goto done_err;
+        }
+        if (r != len) {
+            break;
+	}
+
+        /* if we're here, we probably raced with some snaps creation */
+        len *= 2;
+    }
+    buf = outbuf;
+    end = buf + len;
+
+    if ((r = decode64(&buf, end, &snap_seq)) < 0) {
+        goto done_err;
+    }
+    if ((r = decode32(&buf, end, &snap_count)) < 0) {
+        goto done_err;
+    }
+
+    sn_tab = qemu_mallocz(snap_count * sizeof(QEMUSnapshotInfo));
+    for (i = 0; i < snap_count; i++) {
+        uint64_t id, image_size;
+        char *snap_name;
+
+        if ((r = decode64(&buf, end, &id)) < 0) {
+            goto done_err;
+        }
+        if ((r = decode64(&buf, end, &image_size)) < 0) {
+            goto done_err;
+        }
+        if ((r = decode_str(&buf, end, &snap_name)) < 0) {
+            goto done_err;
+        }
+
+        sn_info = sn_tab + i;
+        pstrcpy(sn_info->id_str, sizeof(sn_info->id_str), snap_name);
+        pstrcpy(sn_info->name, sizeof(sn_info->name), snap_name);
+        qemu_free(snap_name);
+
+        sn_info->vm_state_size = image_size;
+        sn_info->date_sec = 0;
+        sn_info->date_nsec = 0;
+        sn_info->vm_clock_nsec = 0;
+    }
+    *psn_tab = sn_tab;
+    qemu_free(outbuf);
+    return snap_count;
+done_err:
+    qemu_free(sn_tab);
+    qemu_free(outbuf);
+    return r;
+}
+
+static QEMUOptionParameter rbd_create_options[] = {
+    {
+     .name = BLOCK_OPT_SIZE,
+     .type = OPT_SIZE,
+     .help = "Virtual disk size"
+    },
+    {
+     .name = BLOCK_OPT_CLUSTER_SIZE,
+     .type = OPT_SIZE,
+     .help = "RBD object size"
+    },
+    {NULL}
+};
+
+static BlockDriver bdrv_rbd = {
+    .format_name        = "rbd",
+    .instance_size      = sizeof(BDRVRBDState),
+    .bdrv_file_open     = rbd_open,
+    .bdrv_close         = rbd_close,
+    .bdrv_create        = rbd_create,
+    .bdrv_get_info      = rbd_getinfo,
+    .create_options     = rbd_create_options,
+    .bdrv_getlength     = rbd_getlength,
+    .protocol_name      = "rbd",
+
+    .bdrv_aio_readv     = rbd_aio_readv,
+    .bdrv_aio_writev    = rbd_aio_writev,
+
+    .bdrv_snapshot_create = rbd_snap_create,
+    .bdrv_snapshot_list = rbd_snap_list,
+};
+
+static void bdrv_rbd_init(void)
+{
+    bdrv_register(&bdrv_rbd);
+}
+
+block_init(bdrv_rbd_init);
diff --git a/block/rbd_types.h b/block/rbd_types.h
new file mode 100644
index 0000000000..f4cca9970c
--- /dev/null
+++ b/block/rbd_types.h
@@ -0,0 +1,71 @@
+/*
+ * Ceph - scalable distributed file system
+ *
+ * Copyright (C) 2004-2010 Sage Weil <sage@newdream.net>
+ *
+ * This is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License version 2.1, as published by the Free Software
+ * Foundation.  See file COPYING.LIB.
+ *
+ */
+
+#ifndef CEPH_RBD_TYPES_H
+#define CEPH_RBD_TYPES_H
+
+
+/*
+ * rbd image 'foo' consists of objects
+ *   foo.rbd      - image metadata
+ *   foo.00000000
+ *   foo.00000001
+ *   ...          - data
+ */
+
+#define RBD_SUFFIX              ".rbd"
+#define RBD_DIRECTORY           "rbd_directory"
+#define RBD_INFO                "rbd_info"
+
+#define RBD_DEFAULT_OBJ_ORDER   22   /* 4MB */
+
+#define RBD_MAX_OBJ_NAME_SIZE   96
+#define RBD_MAX_BLOCK_NAME_SIZE 24
+#define RBD_MAX_SEG_NAME_SIZE   128
+
+#define RBD_COMP_NONE           0
+#define RBD_CRYPT_NONE          0
+
+#define RBD_HEADER_TEXT         "<<< Rados Block Device Image >>>\n"
+#define RBD_HEADER_SIGNATURE    "RBD"
+#define RBD_HEADER_VERSION      "001.005"
+
+struct rbd_info {
+    uint64_t max_id;
+} __attribute__ ((packed));
+
+struct rbd_obj_snap_ondisk {
+    uint64_t id;
+    uint64_t image_size;
+} __attribute__((packed));
+
+struct rbd_obj_header_ondisk {
+    char text[40];
+    char block_name[RBD_MAX_BLOCK_NAME_SIZE];
+    char signature[4];
+    char version[8];
+    struct {
+        uint8_t order;
+        uint8_t crypt_type;
+        uint8_t comp_type;
+        uint8_t unused;
+    } __attribute__((packed)) options;
+    uint64_t image_size;
+    uint64_t snap_seq;
+    uint32_t snap_count;
+    uint32_t reserved;
+    uint64_t snap_names_len;
+    struct rbd_obj_snap_ondisk snaps[0];
+} __attribute__((packed));
+
+
+#endif
diff --git a/block_int.h b/block_int.h
index 0a0e47d56c..6b3b09808b 100644
--- a/block_int.h
+++ b/block_int.h
@@ -29,7 +29,6 @@
 #include "qemu-queue.h"
 
 #define BLOCK_FLAG_ENCRYPT	1
-#define BLOCK_FLAG_COMPRESS	2
 #define BLOCK_FLAG_COMPAT6	4
 
 #define BLOCK_OPT_SIZE          "size"
diff --git a/configure b/configure
index 2917874f96..62defc41b8 100755
--- a/configure
+++ b/configure
@@ -332,6 +332,7 @@ zero_malloc=""
 trace_backend="nop"
 trace_file="trace"
 spice=""
+rbd=""
 
 # OS specific
 if check_define __linux__ ; then
@@ -741,6 +742,10 @@ for opt do
   ;;
   --*dir)
   ;;
+  --disable-rbd) rbd="no"
+  ;;
+  --enable-rbd) rbd="yes"
+  ;;
   *) echo "ERROR: unknown option $opt"; show_help="yes"
   ;;
   esac
@@ -934,6 +939,7 @@ echo "  --trace-file=NAME        Full PATH,NAME of file to store traces"
 echo "                           Default:trace-<pid>"
 echo "  --disable-spice          disable spice"
 echo "  --enable-spice           enable spice"
+echo "  --enable-rbd             enable building the rados block device (rbd)"
 echo ""
 echo "NOTE: The object files are built at the place where configure is launched"
 exit 1
@@ -1746,6 +1752,48 @@ if test "$mingw32" != yes -a "$pthread" = no; then
 fi
 
 ##########################################
+# rbd probe
+if test "$rbd" != "no" ; then
+  cat > $TMPC <<EOF
+#include <stdio.h>
+#include <rados/librados.h>
+int main(void) { rados_initialize(0, NULL); return 0; }
+EOF
+  rbd_libs="-lrados -lcrypto"
+  if compile_prog "" "$rbd_libs" ; then
+    librados_too_old=no
+    cat > $TMPC <<EOF
+#include <stdio.h>
+#include <rados/librados.h>
+#ifndef CEPH_OSD_TMAP_SET
+#error missing CEPH_OSD_TMAP_SET
+#endif
+int main(void) {
+    int (*func)(const rados_pool_t pool, uint64_t *snapid) = rados_selfmanaged_snap_create;
+    rados_initialize(0, NULL);
+    return 0;
+}
+EOF
+    if compile_prog "" "$rbd_libs" ; then
+      rbd=yes
+      libs_tools="$rbd_libs $libs_tools"
+      libs_softmmu="$rbd_libs $libs_softmmu"
+    else
+      rbd=no
+      librados_too_old=yes
+    fi
+  else
+    if test "$rbd" = "yes" ; then
+      feature_not_found "rados block device"
+    fi
+    rbd=no
+  fi
+  if test "$librados_too_old" = "yes" ; then
+    echo "-> Your librados version is too old - upgrade needed to have rbd support"
+  fi
+fi
+
+##########################################
 # linux-aio probe
 
 if test "$linux_aio" != "no" ; then
@@ -2354,6 +2402,7 @@ echo "vhost-net support $vhost_net"
 echo "Trace backend     $trace_backend"
 echo "Trace output file $trace_file-<pid>"
 echo "spice support     $spice"
+echo "rbd support       $rbd"
 
 if test $sdl_too_old = "yes"; then
 echo "-> Your SDL version is too old - please upgrade to have SDL support"
@@ -2627,6 +2676,9 @@ echo "CONFIG_UNAME_RELEASE=\"$uname_release\"" >> $config_host_mak
 if test "$zero_malloc" = "yes" ; then
   echo "CONFIG_ZERO_MALLOC=y" >> $config_host_mak
 fi
+if test "$rbd" = "yes" ; then
+  echo "CONFIG_RBD=y" >> $config_host_mak
+fi
 
 # USB host support
 case "$usb" in
diff --git a/qemu-img.c b/qemu-img.c
index fa77ac0dbe..52282e3add 100644
--- a/qemu-img.c
+++ b/qemu-img.c
@@ -188,6 +188,33 @@ static int read_password(char *buf, int buf_size)
 }
 #endif
 
+static int print_block_option_help(const char *filename, const char *fmt)
+{
+    BlockDriver *drv, *proto_drv;
+    QEMUOptionParameter *create_options = NULL;
+
+    /* Find driver and parse its options */
+    drv = bdrv_find_format(fmt);
+    if (!drv) {
+        error("Unknown file format '%s'", fmt);
+        return 1;
+    }
+
+    proto_drv = bdrv_find_protocol(filename);
+    if (!proto_drv) {
+        error("Unknown protocol '%s'", filename);
+        return 1;
+    }
+
+    create_options = append_option_parameters(create_options,
+                                              drv->create_options);
+    create_options = append_option_parameters(create_options,
+                                              proto_drv->create_options);
+    print_option_help(create_options);
+    free_option_parameters(create_options);
+    return 0;
+}
+
 static BlockDriverState *bdrv_new_open(const char *filename,
                                        const char *fmt,
                                        int flags)
@@ -234,21 +261,9 @@ fail:
 }
 
 static int add_old_style_options(const char *fmt, QEMUOptionParameter *list,
-    int flags, const char *base_filename, const char *base_fmt)
+                                 const char *base_filename,
+                                 const char *base_fmt)
 {
-    if (flags & BLOCK_FLAG_ENCRYPT) {
-        if (set_option_parameter(list, BLOCK_OPT_ENCRYPT, "on")) {
-            error("Encryption not supported for file format '%s'", fmt);
-            return -1;
-        }
-    }
-    if (flags & BLOCK_FLAG_COMPAT6) {
-        if (set_option_parameter(list, BLOCK_OPT_COMPAT6, "on")) {
-            error("VMDK version 6 not supported for file format '%s'", fmt);
-            return -1;
-        }
-    }
-
     if (base_filename) {
         if (set_option_parameter(list, BLOCK_OPT_BACKING_FILE, base_filename)) {
             error("Backing file not supported for file format '%s'", fmt);
@@ -266,21 +281,23 @@ static int add_old_style_options(const char *fmt, QEMUOptionParameter *list,
 
 static int img_create(int argc, char **argv)
 {
-    int c, ret = 0, flags;
+    int c, ret = 0;
     const char *fmt = "raw";
     const char *base_fmt = NULL;
     const char *filename;
     const char *base_filename = NULL;
     BlockDriver *drv, *proto_drv;
     QEMUOptionParameter *param = NULL, *create_options = NULL;
+    QEMUOptionParameter *backing_fmt = NULL;
     char *options = NULL;
 
-    flags = 0;
     for(;;) {
         c = getopt(argc, argv, "F:b:f:he6o:");
-        if (c == -1)
+        if (c == -1) {
             break;
+        }
         switch(c) {
+        case '?':
         case 'h':
             help();
             break;
@@ -294,11 +311,13 @@ static int img_create(int argc, char **argv)
             fmt = optarg;
             break;
         case 'e':
-            flags |= BLOCK_FLAG_ENCRYPT;
-            break;
+            error("qemu-img: option -e is deprecated, please use \'-o "
+                  "encryption\' instead!");
+            return 1;
         case '6':
-            flags |= BLOCK_FLAG_COMPAT6;
-            break;
+            error("qemu-img: option -6 is deprecated, please use \'-o "
+                  "compat6\' instead!");
+            return 1;
         case 'o':
             options = optarg;
             break;
@@ -306,21 +325,29 @@ static int img_create(int argc, char **argv)
     }
 
     /* Get the filename */
-    if (optind >= argc)
+    if (optind >= argc) {
         help();
+    }
     filename = argv[optind++];
 
+    if (options && !strcmp(options, "?")) {
+        ret = print_block_option_help(filename, fmt);
+        goto out;
+    }
+
     /* Find driver and parse its options */
     drv = bdrv_find_format(fmt);
     if (!drv) {
         error("Unknown file format '%s'", fmt);
-        return 1;
+        ret = -1;
+        goto out;
     }
 
     proto_drv = bdrv_find_protocol(filename);
     if (!proto_drv) {
         error("Unknown protocol '%s'", filename);
-        return 1;
+        ret = -1;
+        goto out;
     }
 
     create_options = append_option_parameters(create_options,
@@ -328,11 +355,6 @@ static int img_create(int argc, char **argv)
     create_options = append_option_parameters(create_options,
                                               proto_drv->create_options);
 
-    if (options && !strcmp(options, "?")) {
-        print_option_help(create_options);
-        goto out;
-    }
-
     /* Create parameter list with default values */
     param = parse_option_parameters("", create_options, param);
     set_option_parameter_int(param, BLOCK_OPT_SIZE, -1);
@@ -353,19 +375,27 @@ static int img_create(int argc, char **argv)
     }
 
     /* Add old-style options to parameters */
-    ret = add_old_style_options(fmt, param, flags, base_filename, base_fmt);
+    ret = add_old_style_options(fmt, param, base_filename, base_fmt);
     if (ret < 0) {
         goto out;
     }
 
+    backing_fmt = get_option_parameter(param, BLOCK_OPT_BACKING_FMT);
+    if (backing_fmt && backing_fmt->value.s) {
+        if (!bdrv_find_format(backing_fmt->value.s)) {
+            error("Unknown backing file format '%s'",
+                  backing_fmt->value.s);
+            ret = -1;
+            goto out;
+        }
+    }
+
     // The size for the image must always be specified, with one exception:
     // If we are using a backing file, we can obtain the size from there
     if (get_option_parameter(param, BLOCK_OPT_SIZE)->value.n == -1) {
 
         QEMUOptionParameter *backing_file =
             get_option_parameter(param, BLOCK_OPT_BACKING_FILE);
-        QEMUOptionParameter *backing_fmt =
-            get_option_parameter(param, BLOCK_OPT_BACKING_FMT);
 
         if (backing_file && backing_file->value.s) {
             BlockDriverState *bs;
@@ -374,14 +404,7 @@ static int img_create(int argc, char **argv)
             char buf[32];
 
             if (backing_fmt && backing_fmt->value.s) {
-                 if (bdrv_find_format(backing_fmt->value.s)) {
-                     fmt = backing_fmt->value.s;
-                } else {
-                     error("Unknown backing file format '%s'",
-                        backing_fmt->value.s);
-                     ret = -1;
-                     goto out;
-                }
+                fmt = backing_fmt->value.s;
             }
 
             bs = bdrv_new_open(backing_file->value.s, fmt, BDRV_O_FLAGS);
@@ -407,8 +430,6 @@ static int img_create(int argc, char **argv)
     puts("");
 
     ret = bdrv_create(drv, filename, param);
-    free_option_parameters(create_options);
-    free_option_parameters(param);
 
     if (ret < 0) {
         if (ret == -ENOTSUP) {
@@ -420,6 +441,8 @@ static int img_create(int argc, char **argv)
         }
     }
 out:
+    free_option_parameters(create_options);
+    free_option_parameters(param);
     if (ret) {
         return 1;
     }
@@ -444,9 +467,11 @@ static int img_check(int argc, char **argv)
     fmt = NULL;
     for(;;) {
         c = getopt(argc, argv, "f:h");
-        if (c == -1)
+        if (c == -1) {
             break;
+        }
         switch(c) {
+        case '?':
         case 'h':
             help();
             break;
@@ -455,8 +480,9 @@ static int img_check(int argc, char **argv)
             break;
         }
     }
-    if (optind >= argc)
+    if (optind >= argc) {
         help();
+    }
     filename = argv[optind++];
 
     bs = bdrv_new_open(filename, fmt, BDRV_O_FLAGS);
@@ -520,9 +546,11 @@ static int img_commit(int argc, char **argv)
     fmt = NULL;
     for(;;) {
         c = getopt(argc, argv, "f:h");
-        if (c == -1)
+        if (c == -1) {
             break;
+        }
         switch(c) {
+        case '?':
         case 'h':
             help();
             break;
@@ -531,8 +559,9 @@ static int img_commit(int argc, char **argv)
             break;
         }
     }
-    if (optind >= argc)
+    if (optind >= argc) {
         help();
+    }
     filename = argv[optind++];
 
     bs = bdrv_new_open(filename, fmt, BDRV_O_FLAGS | BDRV_O_RDWR);
@@ -636,7 +665,7 @@ static int compare_sectors(const uint8_t *buf1, const uint8_t *buf2, int n,
 
 static int img_convert(int argc, char **argv)
 {
-    int c, ret = 0, n, n1, bs_n, bs_i, flags, cluster_size, cluster_sectors;
+    int c, ret = 0, n, n1, bs_n, bs_i, compress, cluster_size, cluster_sectors;
     const char *fmt, *out_fmt, *out_baseimg, *out_filename;
     BlockDriver *drv, *proto_drv;
     BlockDriverState **bs = NULL, *out_bs = NULL;
@@ -653,12 +682,14 @@ static int img_convert(int argc, char **argv)
     fmt = NULL;
     out_fmt = "raw";
     out_baseimg = NULL;
-    flags = 0;
+    compress = 0;
     for(;;) {
         c = getopt(argc, argv, "f:O:B:s:hce6o:");
-        if (c == -1)
+        if (c == -1) {
             break;
+        }
         switch(c) {
+        case '?':
         case 'h':
             help();
             break;
@@ -672,14 +703,16 @@ static int img_convert(int argc, char **argv)
             out_baseimg = optarg;
             break;
         case 'c':
-            flags |= BLOCK_FLAG_COMPRESS;
+            compress = 1;
             break;
         case 'e':
-            flags |= BLOCK_FLAG_ENCRYPT;
-            break;
+            error("qemu-img: option -e is deprecated, please use \'-o "
+                  "encryption\' instead!");
+            return 1;
         case '6':
-            flags |= BLOCK_FLAG_COMPAT6;
-            break;
+            error("qemu-img: option -6 is deprecated, please use \'-o "
+                  "compat6\' instead!");
+            return 1;
         case 'o':
             options = optarg;
             break;
@@ -690,20 +723,24 @@ static int img_convert(int argc, char **argv)
     }
 
     bs_n = argc - optind - 1;
-    if (bs_n < 1) help();
+    if (bs_n < 1) {
+        help();
+    }
 
     out_filename = argv[argc - 1];
 
+    if (options && !strcmp(options, "?")) {
+        ret = print_block_option_help(out_filename, out_fmt);
+        goto out;
+    }
+
     if (bs_n > 1 && out_baseimg) {
         error("-B makes no sense when concatenating multiple input images");
-        return 1;
+        ret = -1;
+        goto out;
     }
         
-    bs = calloc(bs_n, sizeof(BlockDriverState *));
-    if (!bs) {
-        error("Out of memory");
-        return 1;
-    }
+    bs = qemu_mallocz(bs_n * sizeof(BlockDriverState *));
 
     total_sectors = 0;
     for (bs_i = 0; bs_i < bs_n; bs_i++) {
@@ -749,10 +786,6 @@ static int img_convert(int argc, char **argv)
                                               drv->create_options);
     create_options = append_option_parameters(create_options,
                                               proto_drv->create_options);
-    if (options && !strcmp(options, "?")) {
-        print_option_help(create_options);
-        goto out;
-    }
 
     if (options) {
         param = parse_option_parameters(options, create_options, param);
@@ -766,7 +799,7 @@ static int img_convert(int argc, char **argv)
     }
 
     set_option_parameter_int(param, BLOCK_OPT_SIZE, total_sectors * 512);
-    ret = add_old_style_options(out_fmt, param, flags, out_baseimg, NULL);
+    ret = add_old_style_options(out_fmt, param, out_baseimg, NULL);
     if (ret < 0) {
         goto out;
     }
@@ -778,7 +811,7 @@ static int img_convert(int argc, char **argv)
     }
 
     /* Check if compression is supported */
-    if (flags & BLOCK_FLAG_COMPRESS) {
+    if (compress) {
         QEMUOptionParameter *encryption =
             get_option_parameter(param, BLOCK_OPT_ENCRYPT);
 
@@ -820,7 +853,7 @@ static int img_convert(int argc, char **argv)
     bdrv_get_geometry(bs[0], &bs_sectors);
     buf = qemu_malloc(IO_BUF_SIZE);
 
-    if (flags & BLOCK_FLAG_COMPRESS) {
+    if (compress) {
         ret = bdrv_get_info(out_bs, &bdi);
         if (ret < 0) {
             error("could not get block driver info");
@@ -880,8 +913,9 @@ static int img_convert(int argc, char **argv)
             }
             assert (remainder == 0);
 
-            if (n < cluster_sectors)
+            if (n < cluster_sectors) {
                 memset(buf + n * 512, 0, cluster_size - n * 512);
+            }
             if (is_not_zero(buf, cluster_size)) {
                 ret = bdrv_write_compressed(out_bs, sector_num, buf,
                                             cluster_sectors);
@@ -901,12 +935,14 @@ static int img_convert(int argc, char **argv)
         sector_num = 0; // total number of sectors converted so far
         for(;;) {
             nb_sectors = total_sectors - sector_num;
-            if (nb_sectors <= 0)
+            if (nb_sectors <= 0) {
                 break;
-            if (nb_sectors >= (IO_BUF_SIZE / 512))
+            }
+            if (nb_sectors >= (IO_BUF_SIZE / 512)) {
                 n = (IO_BUF_SIZE / 512);
-            else
+            } else {
                 n = nb_sectors;
+            }
 
             while (sector_num - bs_offset >= bs_sectors) {
                 bs_i ++;
@@ -918,8 +954,9 @@ static int img_convert(int argc, char **argv)
                    sector_num, bs_i, bs_offset, bs_sectors); */
             }
 
-            if (n > bs_offset + bs_sectors - sector_num)
+            if (n > bs_offset + bs_sectors - sector_num) {
                 n = bs_offset + bs_sectors - sector_num;
+            }
 
             if (has_zero_init) {
                 /* If the output image is being created as a copy on write image,
@@ -978,12 +1015,14 @@ out:
     if (out_bs) {
         bdrv_delete(out_bs);
     }
-    for (bs_i = 0; bs_i < bs_n; bs_i++) {
-        if (bs[bs_i]) {
-            bdrv_delete(bs[bs_i]);
+    if (bs) {
+        for (bs_i = 0; bs_i < bs_n; bs_i++) {
+            if (bs[bs_i]) {
+                bdrv_delete(bs[bs_i]);
+            }
         }
+        qemu_free(bs);
     }
-    free(bs);
     if (ret) {
         return 1;
     }
@@ -1053,9 +1092,11 @@ static int img_info(int argc, char **argv)
     fmt = NULL;
     for(;;) {
         c = getopt(argc, argv, "f:h");
-        if (c == -1)
+        if (c == -1) {
             break;
+        }
         switch(c) {
+        case '?':
         case 'h':
             help();
             break;
@@ -1064,8 +1105,9 @@ static int img_info(int argc, char **argv)
             break;
         }
     }
-    if (optind >= argc)
+    if (optind >= argc) {
         help();
+    }
     filename = argv[optind++];
 
     bs = bdrv_new_open(filename, fmt, BDRV_O_FLAGS | BDRV_O_NO_BACKING);
@@ -1076,11 +1118,12 @@ static int img_info(int argc, char **argv)
     bdrv_get_geometry(bs, &total_sectors);
     get_human_readable_size(size_buf, sizeof(size_buf), total_sectors * 512);
     allocated_size = get_allocated_file_size(filename);
-    if (allocated_size < 0)
+    if (allocated_size < 0) {
         snprintf(dsize_buf, sizeof(dsize_buf), "unavailable");
-    else
+    } else {
         get_human_readable_size(dsize_buf, sizeof(dsize_buf),
                                 allocated_size);
+    }
     printf("image: %s\n"
            "file format: %s\n"
            "virtual size: %s (%" PRId64 " bytes)\n"
@@ -1088,11 +1131,13 @@ static int img_info(int argc, char **argv)
            filename, fmt_name, size_buf,
            (total_sectors * 512),
            dsize_buf);
-    if (bdrv_is_encrypted(bs))
+    if (bdrv_is_encrypted(bs)) {
         printf("encrypted: yes\n");
+    }
     if (bdrv_get_info(bs, &bdi) >= 0) {
-        if (bdi.cluster_size != 0)
+        if (bdi.cluster_size != 0) {
             printf("cluster_size: %d\n", bdi.cluster_size);
+        }
     }
     bdrv_get_backing_filename(bs, backing_filename, sizeof(backing_filename));
     if (backing_filename[0] != '\0') {
@@ -1125,9 +1170,11 @@ static int img_snapshot(int argc, char **argv)
     /* Parse commandline parameters */
     for(;;) {
         c = getopt(argc, argv, "la:c:d:h");
-        if (c == -1)
+        if (c == -1) {
             break;
+        }
         switch(c) {
+        case '?':
         case 'h':
             help();
             return 0;
@@ -1166,8 +1213,9 @@ static int img_snapshot(int argc, char **argv)
         }
     }
 
-    if (optind >= argc)
+    if (optind >= argc) {
         help();
+    }
     filename = argv[optind++];
 
     /* Open the image */
@@ -1191,23 +1239,26 @@ static int img_snapshot(int argc, char **argv)
         sn.date_nsec = tv.tv_usec * 1000;
 
         ret = bdrv_snapshot_create(bs, &sn);
-        if (ret)
+        if (ret) {
             error("Could not create snapshot '%s': %d (%s)",
                 snapshot_name, ret, strerror(-ret));
+        }
         break;
 
     case SNAPSHOT_APPLY:
         ret = bdrv_snapshot_goto(bs, snapshot_name);
-        if (ret)
+        if (ret) {
             error("Could not apply snapshot '%s': %d (%s)",
                 snapshot_name, ret, strerror(-ret));
+        }
         break;
 
     case SNAPSHOT_DELETE:
         ret = bdrv_snapshot_delete(bs, snapshot_name);
-        if (ret)
+        if (ret) {
             error("Could not delete snapshot '%s': %d (%s)",
                 snapshot_name, ret, strerror(-ret));
+        }
         break;
     }
 
@@ -1235,9 +1286,11 @@ static int img_rebase(int argc, char **argv)
 
     for(;;) {
         c = getopt(argc, argv, "uhf:F:b:");
-        if (c == -1)
+        if (c == -1) {
             break;
+        }
         switch(c) {
+        case '?':
         case 'h':
             help();
             return 0;
@@ -1256,8 +1309,9 @@ static int img_rebase(int argc, char **argv)
         }
     }
 
-    if ((optind >= argc) || !out_baseimg)
+    if ((optind >= argc) || !out_baseimg) {
         help();
+    }
     filename = argv[optind++];
 
     /*
@@ -1433,7 +1487,7 @@ static int img_resize(int argc, char **argv)
     int c, ret, relative;
     const char *filename, *fmt, *size;
     int64_t n, total_size;
-    BlockDriverState *bs;
+    BlockDriverState *bs = NULL;
     QEMUOptionParameter *param;
     QEMUOptionParameter resize_options[] = {
         {
@@ -1451,6 +1505,7 @@ static int img_resize(int argc, char **argv)
             break;
         }
         switch(c) {
+        case '?':
         case 'h':
             help();
             break;
@@ -1484,14 +1539,16 @@ static int img_resize(int argc, char **argv)
     param = parse_option_parameters("", resize_options, NULL);
     if (set_option_parameter(param, BLOCK_OPT_SIZE, size)) {
         /* Error message already printed when size parsing fails */
-        exit(1);
+        ret = -1;
+        goto out;
     }
     n = get_option_parameter(param, BLOCK_OPT_SIZE)->value.n;
     free_option_parameters(param);
 
     bs = bdrv_new_open(filename, fmt, BDRV_O_FLAGS | BDRV_O_RDWR);
     if (!bs) {
-        return 1;
+        ret = -1;
+        goto out;
     }
 
     if (relative) {
@@ -1521,7 +1578,9 @@ static int img_resize(int argc, char **argv)
         break;
     }
 out:
-    bdrv_delete(bs);
+    if (bs) {
+        bdrv_delete(bs);
+    }
     if (ret) {
         return 1;
     }
diff --git a/qemu-malloc.c b/qemu-malloc.c
index 28fb05a481..b9b38514ac 100644
--- a/qemu-malloc.c
+++ b/qemu-malloc.c
@@ -64,10 +64,13 @@ void *qemu_realloc(void *ptr, size_t size)
 
 void *qemu_mallocz(size_t size)
 {
+    void *ptr;
     if (!size && !allow_zero_malloc()) {
         abort();
     }
-    return qemu_oom_check(calloc(1, size ? size : 1));
+    ptr = qemu_oom_check(calloc(1, size ? size : 1));
+    trace_qemu_malloc(size, ptr);
+    return ptr;
 }
 
 char *qemu_strdup(const char *str)
diff --git a/qemu-option.c b/qemu-option.c
index 1f8f41ae7d..65db54292b 100644
--- a/qemu-option.c
+++ b/qemu-option.c
@@ -394,8 +394,8 @@ QEMUOptionParameter *append_option_parameters(QEMUOptionParameter *dest,
 /*
  * Parses a parameter string (param) into an option list (dest).
  *
- * list is the templace is. If dest is NULL, a new copy of list is created for
- * it. If list is NULL, this function fails.
+ * list is the template option list. If dest is NULL, a new copy of list is
+ * created. If list is NULL, this function fails.
  *
  * A parameter string consists of one or more parameters, separated by commas.
  * Each parameter consists of its name and possibly of a value. In the latter
@@ -416,20 +416,13 @@ QEMUOptionParameter *parse_option_parameters(const char *param,
     char value[256];
     char *param_delim, *value_delim;
     char next_delim;
-    size_t num_options;
 
     if (list == NULL) {
         return NULL;
     }
 
     if (dest == NULL) {
-        // Count valid options
-        num_options = count_option_parameters(list);
-
-        // Create a copy of the option list to fill in values
-        dest = qemu_mallocz((num_options + 1) * sizeof(QEMUOptionParameter));
-        allocated = dest;
-        memcpy(dest, list, (num_options + 1) * sizeof(QEMUOptionParameter));
+        dest = allocated = append_option_parameters(NULL, list);
     }
 
     while (*param) {