Pull request

v2:
  * Fix C11 typedef redefinitions in ahci and libqos malloc [Peter]
  * Fix lx -> PRIx64 format specifiers in ahci [Peter]
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1
 
 iQEcBAABAgAGBQJU4hCxAAoJEJykq7OBq3PIh2EH/02LojwlJwtsUoFuDDOdqVu+
 iQsGbGpmtjeaBkiiiQnlN0UHFJk+I05JohagzivchB6p1v0BDnGPMGCRiY2a/URU
 ZrBt6kvIL6ro6n6ebrMq4+MFbyfgXC6tC8UuBnHBOlJMArHCue8w3BD+FWyYOJGY
 +zzvWKSNxEkjwPyXuBPJI0eqLCmANnr31LVrh+4gTvXRnpArT+fNTBooCgA6zsRe
 ChHObhstM/ljy7XZz5TmTXFf/68+WM8AYF+xIHiU3olCcBgH8VPn7lsdBSEuOtxH
 HNe9tt2M91R+3Zdr2KXu+D4BRARKMlft9GDwLGAvGrYcBr7NNJo3KbKbxoecw4M=
 =2rE4
 -----END PGP SIGNATURE-----

Merge remote-tracking branch 'remotes/stefanha/tags/block-pull-request' into staging

Pull request

v2:
 * Fix C11 typedef redefinitions in ahci and libqos malloc [Peter]
 * Fix lx -> PRIx64 format specifiers in ahci [Peter]

# gpg: Signature made Mon Feb 16 15:45:53 2015 GMT using RSA key ID 81AB73C8
# gpg: Good signature from "Stefan Hajnoczi <stefanha@redhat.com>"
# gpg:                 aka "Stefan Hajnoczi <stefanha@gmail.com>"

* remotes/stefanha/tags/block-pull-request: (65 commits)
  block: Keep bdrv_check*_request()'s return value
  block: Remove "growable" from BDS
  block: Clamp BlockBackend requests
  qemu-io: Use BlockBackend
  qemu-io: Remove "growable" option
  qemu-io: Use blk_new_open() in openfile()
  qemu-nbd: Use blk_new_open() in main()
  qemu-img: Use BlockBackend as far as possible
  qemu-img: Use blk_new_open() in img_rebase()
  qemu-img: Use blk_new_open() in img_open()
  block/xen: Use blk_new_open() in blk_connect()
  blockdev: Use blk_new_open() in blockdev_init()
  iotests: Add test for driver=qcow2, format=qcow2
  block: Add Error parameter to bdrv_find_protocol()
  block: Add blk_new_open()
  block: Lift some BDS functions to the BlockBackend
  iotests: Add test for qemu-img convert to NBD
  qemu-img: Fix qemu-img convert -n
  qemu-iotests: Add 093 for IO throttling
  qemu-iotests: Allow caller to disable underscore convertion for qmp
  ...

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

block.c

@@ -508,9 +508,8 @@ int bdrv_create_file(const char *filename, QemuOpts *opts, Error **errp)
     Error *local_err = NULL;
     int ret;
 
-    drv = bdrv_find_protocol(filename, true);
+    drv = bdrv_find_protocol(filename, true, errp);
     if (drv == NULL) {
-        error_setg(errp, "Could not find protocol for file '%s'", filename);
         return -ENOENT;
     }
 
@@ -628,7 +627,8 @@ static BlockDriver *find_hdev_driver(const char *filename)
 }
 
 BlockDriver *bdrv_find_protocol(const char *filename,
-                                bool allow_protocol_prefix)
+                                bool allow_protocol_prefix,
+                                Error **errp)
 {
     BlockDriver *drv1;
     char protocol[128];
@@ -666,6 +666,8 @@ BlockDriver *bdrv_find_protocol(const char *filename,
             return drv1;
         }
     }
+
+    error_setg(errp, "Unknown protocol '%s'", protocol);
     return NULL;
 }
 
@@ -970,7 +972,6 @@ static int bdrv_open_common(BlockDriverState *bs, BlockDriverState *file,
     bs->zero_beyond_eof = true;
     open_flags = bdrv_open_flags(bs, flags);
     bs->read_only = !(open_flags & BDRV_O_RDWR);
-    bs->growable = !!(flags & BDRV_O_PROTOCOL);
 
     if (use_bdrv_whitelist && !bdrv_is_whitelisted(drv, bs->read_only)) {
         error_setg(errp,
@@ -1136,9 +1137,8 @@ static int bdrv_fill_options(QDict **options, const char **pfilename, int flags,
     } else {
         if (!drvname && protocol) {
             if (filename) {
-                drv = bdrv_find_protocol(filename, parse_filename);
+                drv = bdrv_find_protocol(filename, parse_filename, errp);
                 if (!drv) {
-                    error_setg(errp, "Unknown protocol");
                     return -EINVAL;
                 }
 
@@ -1885,7 +1885,6 @@ void bdrv_close(BlockDriverState *bs)
         bs->encrypted = 0;
         bs->valid_key = 0;
         bs->sg = 0;
-        bs->growable = 0;
         bs->zero_beyond_eof = false;
         QDECREF(bs->options);
         bs->options = NULL;
@@ -2645,25 +2644,17 @@ exit:
 static int bdrv_check_byte_request(BlockDriverState *bs, int64_t offset,
                                    size_t size)
 {
-    int64_t len;
-
     if (size > BDRV_REQUEST_MAX_SECTORS << BDRV_SECTOR_BITS) {
         return -EIO;
     }
 
-    if (!bdrv_is_inserted(bs))
+    if (!bdrv_is_inserted(bs)) {
         return -ENOMEDIUM;
+    }
 
-    if (bs->growable)
+    if (offset < 0) {
-        return 0;
-
-    len = bdrv_getlength(bs);
-
-    if (offset < 0)
-        return -EIO;
-
-    if ((offset > len) || (len - offset < size))
         return -EIO;
+    }
 
     return 0;
 }
@@ -3042,10 +3033,10 @@ static int coroutine_fn bdrv_aligned_preadv(BlockDriverState *bs,
     }
 
     /* Forward the request to the BlockDriver */
-    if (!(bs->zero_beyond_eof && bs->growable)) {
+    if (!bs->zero_beyond_eof) {
         ret = drv->bdrv_co_readv(bs, sector_num, nb_sectors, qiov);
     } else {
-        /* Read zeros after EOF of growable BDSes */
+        /* Read zeros after EOF */
         int64_t total_sectors, max_nb_sectors;
 
         total_sectors = bdrv_nb_sectors(bs);
@@ -3107,8 +3098,10 @@ static int coroutine_fn bdrv_co_do_preadv(BlockDriverState *bs,
     if (!drv) {
         return -ENOMEDIUM;
     }
-    if (bdrv_check_byte_request(bs, offset, bytes)) {
+
-        return -EIO;
+    ret = bdrv_check_byte_request(bs, offset, bytes);
+    if (ret < 0) {
+        return ret;
     }
 
     if (bs->copy_on_read) {
@@ -3322,7 +3315,7 @@ static int coroutine_fn bdrv_aligned_pwritev(BlockDriverState *bs,
 
     block_acct_highest_sector(&bs->stats, sector_num, nb_sectors);
 
-    if (bs->growable && ret >= 0) {
+    if (ret >= 0) {
         bs->total_sectors = MAX(bs->total_sectors, sector_num + nb_sectors);
     }
 
@@ -3351,8 +3344,10 @@ static int coroutine_fn bdrv_co_do_pwritev(BlockDriverState *bs,
     if (bs->read_only) {
         return -EACCES;
     }
-    if (bdrv_check_byte_request(bs, offset, bytes)) {
+
-        return -EIO;
+    ret = bdrv_check_byte_request(bs, offset, bytes);
+    if (ret < 0) {
+        return ret;
     }
 
     /* throttling disk I/O */
@@ -4206,12 +4201,18 @@ int bdrv_write_compressed(BlockDriverState *bs, int64_t sector_num,
                           const uint8_t *buf, int nb_sectors)
 {
     BlockDriver *drv = bs->drv;
-    if (!drv)
+    int ret;
+
+    if (!drv) {
         return -ENOMEDIUM;
-    if (!drv->bdrv_write_compressed)
+    }
+    if (!drv->bdrv_write_compressed) {
         return -ENOTSUP;
-    if (bdrv_check_request(bs, sector_num, nb_sectors))
+    }
-        return -EIO;
+    ret = bdrv_check_request(bs, sector_num, nb_sectors);
+    if (ret < 0) {
+        return ret;
+    }
 
     assert(QLIST_EMPTY(&bs->dirty_bitmaps));
 
@@ -5126,12 +5127,15 @@ static void coroutine_fn bdrv_discard_co_entry(void *opaque)
 int coroutine_fn bdrv_co_discard(BlockDriverState *bs, int64_t sector_num,
                                  int nb_sectors)
 {
-    int max_discard;
+    int max_discard, ret;
 
     if (!bs->drv) {
         return -ENOMEDIUM;
-    } else if (bdrv_check_request(bs, sector_num, nb_sectors)) {
+    }
-        return -EIO;
+
+    ret = bdrv_check_request(bs, sector_num, nb_sectors);
+    if (ret < 0) {
+        return ret;
     } else if (bs->read_only) {
         return -EROFS;
     }
@@ -5623,9 +5627,8 @@ void bdrv_img_create(const char *filename, const char *fmt,
         return;
     }
 
-    proto_drv = bdrv_find_protocol(filename, true);
+    proto_drv = bdrv_find_protocol(filename, true, errp);
     if (!proto_drv) {
-        error_setg(errp, "Unknown protocol '%s'", filename);
         return;
     }
 

block/block-backend.c

@@ -31,6 +31,16 @@ struct BlockBackend {
     void *dev_opaque;
 };
 
+typedef struct BlockBackendAIOCB {
+    BlockAIOCB common;
+    QEMUBH *bh;
+    int ret;
+} BlockBackendAIOCB;
+
+static const AIOCBInfo block_backend_aiocb_info = {
+    .aiocb_size = sizeof(BlockBackendAIOCB),
+};
+
 static void drive_info_del(DriveInfo *dinfo);
 
 /* All the BlockBackends (except for hidden ones) */
@@ -91,6 +101,40 @@ BlockBackend *blk_new_with_bs(const char *name, Error **errp)
     return blk;
 }
 
+/*
+ * Calls blk_new_with_bs() and then calls bdrv_open() on the BlockDriverState.
+ *
+ * Just as with bdrv_open(), after having called this function the reference to
+ * @options belongs to the block layer (even on failure).
+ *
+ * TODO: Remove @filename and @flags; it should be possible to specify a whole
+ * BDS tree just by specifying the @options QDict (or @reference,
+ * alternatively). At the time of adding this function, this is not possible,
+ * though, so callers of this function have to be able to specify @filename and
+ * @flags.
+ */
+BlockBackend *blk_new_open(const char *name, const char *filename,
+                           const char *reference, QDict *options, int flags,
+                           Error **errp)
+{
+    BlockBackend *blk;
+    int ret;
+
+    blk = blk_new_with_bs(name, errp);
+    if (!blk) {
+        QDECREF(options);
+        return NULL;
+    }
+
+    ret = bdrv_open(&blk->bs, filename, reference, options, flags, NULL, errp);
+    if (ret < 0) {
+        blk_unref(blk);
+        return NULL;
+    }
+
+    return blk;
+}
+
 static void blk_delete(BlockBackend *blk)
 {
     assert(!blk->refcnt);
@@ -394,39 +438,137 @@ void blk_iostatus_enable(BlockBackend *blk)
     bdrv_iostatus_enable(blk->bs);
 }
 
+static int blk_check_byte_request(BlockBackend *blk, int64_t offset,
+                                  size_t size)
+{
+    int64_t len;
+
+    if (size > INT_MAX) {
+        return -EIO;
+    }
+
+    if (!blk_is_inserted(blk)) {
+        return -ENOMEDIUM;
+    }
+
+    len = blk_getlength(blk);
+    if (len < 0) {
+        return len;
+    }
+
+    if (offset < 0) {
+        return -EIO;
+    }
+
+    if (offset > len || len - offset < size) {
+        return -EIO;
+    }
+
+    return 0;
+}
+
+static int blk_check_request(BlockBackend *blk, int64_t sector_num,
+                             int nb_sectors)
+{
+    if (sector_num < 0 || sector_num > INT64_MAX / BDRV_SECTOR_SIZE) {
+        return -EIO;
+    }
+
+    if (nb_sectors < 0 || nb_sectors > INT_MAX / BDRV_SECTOR_SIZE) {
+        return -EIO;
+    }
+
+    return blk_check_byte_request(blk, sector_num * BDRV_SECTOR_SIZE,
+                                  nb_sectors * BDRV_SECTOR_SIZE);
+}
+
 int blk_read(BlockBackend *blk, int64_t sector_num, uint8_t *buf,
              int nb_sectors)
 {
+    int ret = blk_check_request(blk, sector_num, nb_sectors);
+    if (ret < 0) {
+        return ret;
+    }
+
     return bdrv_read(blk->bs, sector_num, buf, nb_sectors);
 }
 
 int blk_read_unthrottled(BlockBackend *blk, int64_t sector_num, uint8_t *buf,
                          int nb_sectors)
 {
+    int ret = blk_check_request(blk, sector_num, nb_sectors);
+    if (ret < 0) {
+        return ret;
+    }
+
     return bdrv_read_unthrottled(blk->bs, sector_num, buf, nb_sectors);
 }
 
 int blk_write(BlockBackend *blk, int64_t sector_num, const uint8_t *buf,
               int nb_sectors)
 {
+    int ret = blk_check_request(blk, sector_num, nb_sectors);
+    if (ret < 0) {
+        return ret;
+    }
+
     return bdrv_write(blk->bs, sector_num, buf, nb_sectors);
 }
 
+static void error_callback_bh(void *opaque)
+{
+    struct BlockBackendAIOCB *acb = opaque;
+    qemu_bh_delete(acb->bh);
+    acb->common.cb(acb->common.opaque, acb->ret);
+    qemu_aio_unref(acb);
+}
+
+static BlockAIOCB *abort_aio_request(BlockBackend *blk, BlockCompletionFunc *cb,
+                                     void *opaque, int ret)
+{
+    struct BlockBackendAIOCB *acb;
+    QEMUBH *bh;
+
+    acb = blk_aio_get(&block_backend_aiocb_info, blk, cb, opaque);
+    acb->ret = ret;
+
+    bh = aio_bh_new(blk_get_aio_context(blk), error_callback_bh, acb);
+    acb->bh = bh;
+    qemu_bh_schedule(bh);
+
+    return &acb->common;
+}
+
 BlockAIOCB *blk_aio_write_zeroes(BlockBackend *blk, int64_t sector_num,
                                  int nb_sectors, BdrvRequestFlags flags,
                                  BlockCompletionFunc *cb, void *opaque)
 {
+    int ret = blk_check_request(blk, sector_num, nb_sectors);
+    if (ret < 0) {
+        return abort_aio_request(blk, cb, opaque, ret);
+    }
+
     return bdrv_aio_write_zeroes(blk->bs, sector_num, nb_sectors, flags,
                                  cb, opaque);
 }
 
 int blk_pread(BlockBackend *blk, int64_t offset, void *buf, int count)
 {
+    int ret = blk_check_byte_request(blk, offset, count);
+    if (ret < 0) {
+        return ret;
+    }
+
     return bdrv_pread(blk->bs, offset, buf, count);
 }
 
 int blk_pwrite(BlockBackend *blk, int64_t offset, const void *buf, int count)
 {
+    int ret = blk_check_byte_request(blk, offset, count);
+    if (ret < 0) {
+        return ret;
+    }
+
     return bdrv_pwrite(blk->bs, offset, buf, count);
 }
 
@@ -440,10 +582,20 @@ void blk_get_geometry(BlockBackend *blk, uint64_t *nb_sectors_ptr)
     bdrv_get_geometry(blk->bs, nb_sectors_ptr);
 }
 
+int64_t blk_nb_sectors(BlockBackend *blk)
+{
+    return bdrv_nb_sectors(blk->bs);
+}
+
 BlockAIOCB *blk_aio_readv(BlockBackend *blk, int64_t sector_num,
                           QEMUIOVector *iov, int nb_sectors,
                           BlockCompletionFunc *cb, void *opaque)
 {
+    int ret = blk_check_request(blk, sector_num, nb_sectors);
+    if (ret < 0) {
+        return abort_aio_request(blk, cb, opaque, ret);
+    }
+
     return bdrv_aio_readv(blk->bs, sector_num, iov, nb_sectors, cb, opaque);
 }
 
@@ -451,6 +603,11 @@ BlockAIOCB *blk_aio_writev(BlockBackend *blk, int64_t sector_num,
                            QEMUIOVector *iov, int nb_sectors,
                            BlockCompletionFunc *cb, void *opaque)
 {
+    int ret = blk_check_request(blk, sector_num, nb_sectors);
+    if (ret < 0) {
+        return abort_aio_request(blk, cb, opaque, ret);
+    }
+
     return bdrv_aio_writev(blk->bs, sector_num, iov, nb_sectors, cb, opaque);
 }
 
@@ -464,6 +621,11 @@ BlockAIOCB *blk_aio_discard(BlockBackend *blk,
                             int64_t sector_num, int nb_sectors,
                             BlockCompletionFunc *cb, void *opaque)
 {
+    int ret = blk_check_request(blk, sector_num, nb_sectors);
+    if (ret < 0) {
+        return abort_aio_request(blk, cb, opaque, ret);
+    }
+
     return bdrv_aio_discard(blk->bs, sector_num, nb_sectors, cb, opaque);
 }
 
@@ -479,6 +641,15 @@ void blk_aio_cancel_async(BlockAIOCB *acb)
 
 int blk_aio_multiwrite(BlockBackend *blk, BlockRequest *reqs, int num_reqs)
 {
+    int i, ret;
+
+    for (i = 0; i < num_reqs; i++) {
+        ret = blk_check_request(blk, reqs[i].sector, reqs[i].nb_sectors);
+        if (ret < 0) {
+            return ret;
+        }
+    }
+
     return bdrv_aio_multiwrite(blk->bs, reqs, num_reqs);
 }
 
@@ -495,6 +666,11 @@ BlockAIOCB *blk_aio_ioctl(BlockBackend *blk, unsigned long int req, void *buf,
 
 int blk_co_discard(BlockBackend *blk, int64_t sector_num, int nb_sectors)
 {
+    int ret = blk_check_request(blk, sector_num, nb_sectors);
+    if (ret < 0) {
+        return ret;
+    }
+
     return bdrv_co_discard(blk->bs, sector_num, nb_sectors);
 }
 
@@ -668,3 +844,51 @@ void *blk_aio_get(const AIOCBInfo *aiocb_info, BlockBackend *blk,
 {
     return qemu_aio_get(aiocb_info, blk_bs(blk), cb, opaque);
 }
+
+int coroutine_fn blk_co_write_zeroes(BlockBackend *blk, int64_t sector_num,
+                                     int nb_sectors, BdrvRequestFlags flags)
+{
+    int ret = blk_check_request(blk, sector_num, nb_sectors);
+    if (ret < 0) {
+        return ret;
+    }
+
+    return bdrv_co_write_zeroes(blk->bs, sector_num, nb_sectors, flags);
+}
+
+int blk_write_compressed(BlockBackend *blk, int64_t sector_num,
+                         const uint8_t *buf, int nb_sectors)
+{
+    int ret = blk_check_request(blk, sector_num, nb_sectors);
+    if (ret < 0) {
+        return ret;
+    }
+
+    return bdrv_write_compressed(blk->bs, sector_num, buf, nb_sectors);
+}
+
+int blk_truncate(BlockBackend *blk, int64_t offset)
+{
+    return bdrv_truncate(blk->bs, offset);
+}
+
+int blk_discard(BlockBackend *blk, int64_t sector_num, int nb_sectors)
+{
+    int ret = blk_check_request(blk, sector_num, nb_sectors);
+    if (ret < 0) {
+        return ret;
+    }
+
+    return bdrv_discard(blk->bs, sector_num, nb_sectors);
+}
+
+int blk_save_vmstate(BlockBackend *blk, const uint8_t *buf,
+                     int64_t pos, int size)
+{
+    return bdrv_save_vmstate(blk->bs, buf, pos, size);
+}
+
+int blk_load_vmstate(BlockBackend *blk, uint8_t *buf, int64_t pos, int size)
+{
+    return bdrv_load_vmstate(blk->bs, buf, pos, size);
+}

block/nbd-client.c

@@ -43,20 +43,23 @@ static void nbd_recv_coroutines_enter_all(NbdClientSession *s)
     }
 }
 
-static void nbd_teardown_connection(NbdClientSession *client)
+static void nbd_teardown_connection(BlockDriverState *bs)
 {
+    NbdClientSession *client = nbd_get_client_session(bs);
+
     /* finish any pending coroutines */
     shutdown(client->sock, 2);
     nbd_recv_coroutines_enter_all(client);
 
-    nbd_client_session_detach_aio_context(client);
+    nbd_client_detach_aio_context(bs);
     closesocket(client->sock);
     client->sock = -1;
 }
 
 static void nbd_reply_ready(void *opaque)
 {
-    NbdClientSession *s = opaque;
+    BlockDriverState *bs = opaque;
+    NbdClientSession *s = nbd_get_client_session(bs);
     uint64_t i;
     int ret;
 
@@ -89,28 +92,40 @@ static void nbd_reply_ready(void *opaque)
     }
 
 fail:
-    nbd_teardown_connection(s);
+    nbd_teardown_connection(bs);
 }
 
 static void nbd_restart_write(void *opaque)
 {
-    NbdClientSession *s = opaque;
+    BlockDriverState *bs = opaque;
 
-    qemu_coroutine_enter(s->send_coroutine, NULL);
+    qemu_coroutine_enter(nbd_get_client_session(bs)->send_coroutine, NULL);
 }
 
-static int nbd_co_send_request(NbdClientSession *s,
+static int nbd_co_send_request(BlockDriverState *bs,
-    struct nbd_request *request,
+                               struct nbd_request *request,
-    QEMUIOVector *qiov, int offset)
+                               QEMUIOVector *qiov, int offset)
 {
+    NbdClientSession *s = nbd_get_client_session(bs);
     AioContext *aio_context;
-    int rc, ret;
+    int rc, ret, i;
 
     qemu_co_mutex_lock(&s->send_mutex);
+
+    for (i = 0; i < MAX_NBD_REQUESTS; i++) {
+        if (s->recv_coroutine[i] == NULL) {
+            s->recv_coroutine[i] = qemu_coroutine_self();
+            break;
+        }
+    }
+
+    assert(i < MAX_NBD_REQUESTS);
+    request->handle = INDEX_TO_HANDLE(s, i);
     s->send_coroutine = qemu_coroutine_self();
-    aio_context = bdrv_get_aio_context(s->bs);
+    aio_context = bdrv_get_aio_context(bs);
+
     aio_set_fd_handler(aio_context, s->sock,
-                       nbd_reply_ready, nbd_restart_write, s);
+                       nbd_reply_ready, nbd_restart_write, bs);
     if (qiov) {
         if (!s->is_unix) {
             socket_set_cork(s->sock, 1);
@@ -129,7 +144,7 @@ static int nbd_co_send_request(NbdClientSession *s,
     } else {
         rc = nbd_send_request(s->sock, request);
     }
-    aio_set_fd_handler(aio_context, s->sock, nbd_reply_ready, NULL, s);
+    aio_set_fd_handler(aio_context, s->sock, nbd_reply_ready, NULL, bs);
     s->send_coroutine = NULL;
     qemu_co_mutex_unlock(&s->send_mutex);
     return rc;
@@ -164,8 +179,6 @@ static void nbd_co_receive_reply(NbdClientSession *s,
 static void nbd_coroutine_start(NbdClientSession *s,
    struct nbd_request *request)
 {
-    int i;
-
     /* Poor man semaphore.  The free_sema is locked when no other request
      * can be accepted, and unlocked after receiving one reply.  */
     if (s->in_flight >= MAX_NBD_REQUESTS - 1) {
@@ -174,15 +187,7 @@ static void nbd_coroutine_start(NbdClientSession *s,
     }
     s->in_flight++;
 
-    for (i = 0; i < MAX_NBD_REQUESTS; i++) {
+    /* s->recv_coroutine[i] is set as soon as we get the send_lock.  */
-        if (s->recv_coroutine[i] == NULL) {
-            s->recv_coroutine[i] = qemu_coroutine_self();
-            break;
-        }
-    }
-
-    assert(i < MAX_NBD_REQUESTS);
-    request->handle = INDEX_TO_HANDLE(s, i);
 }
 
 static void nbd_coroutine_end(NbdClientSession *s,
@@ -195,10 +200,11 @@ static void nbd_coroutine_end(NbdClientSession *s,
     }
 }
 
-static int nbd_co_readv_1(NbdClientSession *client, int64_t sector_num,
+static int nbd_co_readv_1(BlockDriverState *bs, int64_t sector_num,
                           int nb_sectors, QEMUIOVector *qiov,
                           int offset)
 {
+    NbdClientSession *client = nbd_get_client_session(bs);
     struct nbd_request request = { .type = NBD_CMD_READ };
     struct nbd_reply reply;
     ssize_t ret;
@@ -207,7 +213,7 @@ static int nbd_co_readv_1(NbdClientSession *client, int64_t sector_num,
     request.len = nb_sectors * 512;
 
     nbd_coroutine_start(client, &request);
-    ret = nbd_co_send_request(client, &request, NULL, 0);
+    ret = nbd_co_send_request(bs, &request, NULL, 0);
     if (ret < 0) {
         reply.error = -ret;
     } else {
@@ -218,15 +224,16 @@ static int nbd_co_readv_1(NbdClientSession *client, int64_t sector_num,
 
 }
 
-static int nbd_co_writev_1(NbdClientSession *client, int64_t sector_num,
+static int nbd_co_writev_1(BlockDriverState *bs, int64_t sector_num,
                            int nb_sectors, QEMUIOVector *qiov,
                            int offset)
 {
+    NbdClientSession *client = nbd_get_client_session(bs);
     struct nbd_request request = { .type = NBD_CMD_WRITE };
     struct nbd_reply reply;
     ssize_t ret;
 
-    if (!bdrv_enable_write_cache(client->bs) &&
+    if (!bdrv_enable_write_cache(bs) &&
         (client->nbdflags & NBD_FLAG_SEND_FUA)) {
         request.type |= NBD_CMD_FLAG_FUA;
     }
@@ -235,7 +242,7 @@ static int nbd_co_writev_1(NbdClientSession *client, int64_t sector_num,
     request.len = nb_sectors * 512;
 
     nbd_coroutine_start(client, &request);
-    ret = nbd_co_send_request(client, &request, qiov, offset);
+    ret = nbd_co_send_request(bs, &request, qiov, offset);
     if (ret < 0) {
         reply.error = -ret;
     } else {
@@ -249,14 +256,13 @@ static int nbd_co_writev_1(NbdClientSession *client, int64_t sector_num,
  * remain aligned to 4K. */
 #define NBD_MAX_SECTORS 2040
 
-int nbd_client_session_co_readv(NbdClientSession *client, int64_t sector_num,
+int nbd_client_co_readv(BlockDriverState *bs, int64_t sector_num,
-    int nb_sectors, QEMUIOVector *qiov)
+                        int nb_sectors, QEMUIOVector *qiov)
 {
     int offset = 0;
     int ret;
     while (nb_sectors > NBD_MAX_SECTORS) {
-        ret = nbd_co_readv_1(client, sector_num,
+        ret = nbd_co_readv_1(bs, sector_num, NBD_MAX_SECTORS, qiov, offset);
-                             NBD_MAX_SECTORS, qiov, offset);
         if (ret < 0) {
             return ret;
         }
@@ -264,17 +270,16 @@ int nbd_client_session_co_readv(NbdClientSession *client, int64_t sector_num,
         sector_num += NBD_MAX_SECTORS;
         nb_sectors -= NBD_MAX_SECTORS;
     }
-    return nbd_co_readv_1(client, sector_num, nb_sectors, qiov, offset);
+    return nbd_co_readv_1(bs, sector_num, nb_sectors, qiov, offset);
 }
 
-int nbd_client_session_co_writev(NbdClientSession *client, int64_t sector_num,
+int nbd_client_co_writev(BlockDriverState *bs, int64_t sector_num,
-                                 int nb_sectors, QEMUIOVector *qiov)
+                         int nb_sectors, QEMUIOVector *qiov)
 {
     int offset = 0;
     int ret;
     while (nb_sectors > NBD_MAX_SECTORS) {
-        ret = nbd_co_writev_1(client, sector_num,
+        ret = nbd_co_writev_1(bs, sector_num, NBD_MAX_SECTORS, qiov, offset);
-                              NBD_MAX_SECTORS, qiov, offset);
         if (ret < 0) {
             return ret;
         }
@@ -282,11 +287,12 @@ int nbd_client_session_co_writev(NbdClientSession *client, int64_t sector_num,
         sector_num += NBD_MAX_SECTORS;
         nb_sectors -= NBD_MAX_SECTORS;
     }
-    return nbd_co_writev_1(client, sector_num, nb_sectors, qiov, offset);
+    return nbd_co_writev_1(bs, sector_num, nb_sectors, qiov, offset);
 }
 
-int nbd_client_session_co_flush(NbdClientSession *client)
+int nbd_client_co_flush(BlockDriverState *bs)
 {
+    NbdClientSession *client = nbd_get_client_session(bs);
     struct nbd_request request = { .type = NBD_CMD_FLUSH };
     struct nbd_reply reply;
     ssize_t ret;
@@ -303,7 +309,7 @@ int nbd_client_session_co_flush(NbdClientSession *client)
     request.len = 0;
 
     nbd_coroutine_start(client, &request);
-    ret = nbd_co_send_request(client, &request, NULL, 0);
+    ret = nbd_co_send_request(bs, &request, NULL, 0);
     if (ret < 0) {
         reply.error = -ret;
     } else {
@@ -313,9 +319,10 @@ int nbd_client_session_co_flush(NbdClientSession *client)
     return -reply.error;
 }
 
-int nbd_client_session_co_discard(NbdClientSession *client, int64_t sector_num,
+int nbd_client_co_discard(BlockDriverState *bs, int64_t sector_num,
-    int nb_sectors)
+                          int nb_sectors)
 {
+    NbdClientSession *client = nbd_get_client_session(bs);
     struct nbd_request request = { .type = NBD_CMD_TRIM };
     struct nbd_reply reply;
     ssize_t ret;
@@ -327,7 +334,7 @@ int nbd_client_session_co_discard(NbdClientSession *client, int64_t sector_num,
     request.len = nb_sectors * 512;
 
     nbd_coroutine_start(client, &request);
-    ret = nbd_co_send_request(client, &request, NULL, 0);
+    ret = nbd_co_send_request(bs, &request, NULL, 0);
     if (ret < 0) {
         reply.error = -ret;
     } else {
@@ -338,43 +345,41 @@ int nbd_client_session_co_discard(NbdClientSession *client, int64_t sector_num,
 
 }
 
-void nbd_client_session_detach_aio_context(NbdClientSession *client)
+void nbd_client_detach_aio_context(BlockDriverState *bs)
 {
-    aio_set_fd_handler(bdrv_get_aio_context(client->bs), client->sock,
+    aio_set_fd_handler(bdrv_get_aio_context(bs),
-                       NULL, NULL, NULL);
+                       nbd_get_client_session(bs)->sock, NULL, NULL, NULL);
 }
 
-void nbd_client_session_attach_aio_context(NbdClientSession *client,
+void nbd_client_attach_aio_context(BlockDriverState *bs,
-                                           AioContext *new_context)
+                                   AioContext *new_context)
 {
-    aio_set_fd_handler(new_context, client->sock,
+    aio_set_fd_handler(new_context, nbd_get_client_session(bs)->sock,
-                       nbd_reply_ready, NULL, client);
+                       nbd_reply_ready, NULL, bs);
 }
 
-void nbd_client_session_close(NbdClientSession *client)
+void nbd_client_close(BlockDriverState *bs)
 {
+    NbdClientSession *client = nbd_get_client_session(bs);
     struct nbd_request request = {
         .type = NBD_CMD_DISC,
         .from = 0,
         .len = 0
     };
 
-    if (!client->bs) {
-        return;
-    }
     if (client->sock == -1) {
         return;
     }
 
     nbd_send_request(client->sock, &request);
 
-    nbd_teardown_connection(client);
+    nbd_teardown_connection(bs);
-    client->bs = NULL;
 }
 
-int nbd_client_session_init(NbdClientSession *client, BlockDriverState *bs,
+int nbd_client_init(BlockDriverState *bs, int sock, const char *export,
-                            int sock, const char *export, Error **errp)
+                    Error **errp)
 {
+    NbdClientSession *client = nbd_get_client_session(bs);
     int ret;
 
     /* NBD handshake */
@@ -391,13 +396,12 @@ int nbd_client_session_init(NbdClientSession *client, BlockDriverState *bs,
 
     qemu_co_mutex_init(&client->send_mutex);
     qemu_co_mutex_init(&client->free_sema);
-    client->bs = bs;
     client->sock = sock;
 
     /* Now that we're connected, set the socket to be non-blocking and
      * kick the reply mechanism.  */
     qemu_set_nonblock(sock);
-    nbd_client_session_attach_aio_context(client, bdrv_get_aio_context(bs));
+    nbd_client_attach_aio_context(bs, bdrv_get_aio_context(bs));
 
     logout("Established connection with NBD server\n");
     return 0;

block/nbd-client.h

@@ -31,24 +31,24 @@ typedef struct NbdClientSession {
     struct nbd_reply reply;
 
     bool is_unix;
-
-    BlockDriverState *bs;
 } NbdClientSession;
 
-int nbd_client_session_init(NbdClientSession *client, BlockDriverState *bs,
+NbdClientSession *nbd_get_client_session(BlockDriverState *bs);
-                            int sock, const char *export_name, Error **errp);
-void nbd_client_session_close(NbdClientSession *client);
 
-int nbd_client_session_co_discard(NbdClientSession *client, int64_t sector_num,
+int nbd_client_init(BlockDriverState *bs, int sock, const char *export_name,
-                                  int nb_sectors);
+                    Error **errp);
-int nbd_client_session_co_flush(NbdClientSession *client);
+void nbd_client_close(BlockDriverState *bs);
-int nbd_client_session_co_writev(NbdClientSession *client, int64_t sector_num,
-                                 int nb_sectors, QEMUIOVector *qiov);
-int nbd_client_session_co_readv(NbdClientSession *client, int64_t sector_num,
-                                int nb_sectors, QEMUIOVector *qiov);
 
-void nbd_client_session_detach_aio_context(NbdClientSession *client);
+int nbd_client_co_discard(BlockDriverState *bs, int64_t sector_num,
-void nbd_client_session_attach_aio_context(NbdClientSession *client,
+                          int nb_sectors);
-                                           AioContext *new_context);
+int nbd_client_co_flush(BlockDriverState *bs);
+int nbd_client_co_writev(BlockDriverState *bs, int64_t sector_num,
+                         int nb_sectors, QEMUIOVector *qiov);
+int nbd_client_co_readv(BlockDriverState *bs, int64_t sector_num,
+                        int nb_sectors, QEMUIOVector *qiov);
+
+void nbd_client_detach_aio_context(BlockDriverState *bs);
+void nbd_client_attach_aio_context(BlockDriverState *bs,
+                                   AioContext *new_context);
 
 #endif /* NBD_CLIENT_H */

block/nbd.c

@@ -224,6 +224,12 @@ static void nbd_config(BDRVNBDState *s, QDict *options, char **export,
     }
 }
 
+NbdClientSession *nbd_get_client_session(BlockDriverState *bs)
+{
+    BDRVNBDState *s = bs->opaque;
+    return &s->client;
+}
+
 static int nbd_establish_connection(BlockDriverState *bs, Error **errp)
 {
     BDRVNBDState *s = bs->opaque;
@@ -271,7 +277,7 @@ static int nbd_open(BlockDriverState *bs, QDict *options, int flags,
     }
 
     /* NBD handshake */
-    result = nbd_client_session_init(&s->client, bs, sock, export, errp);
+    result = nbd_client_init(bs, sock, export, errp);
     g_free(export);
     return result;
 }
@@ -279,26 +285,18 @@ static int nbd_open(BlockDriverState *bs, QDict *options, int flags,
 static int nbd_co_readv(BlockDriverState *bs, int64_t sector_num,
                         int nb_sectors, QEMUIOVector *qiov)
 {
-    BDRVNBDState *s = bs->opaque;
+    return nbd_client_co_readv(bs, sector_num, nb_sectors, qiov);
-
-    return nbd_client_session_co_readv(&s->client, sector_num,
-                                       nb_sectors, qiov);
 }
 
 static int nbd_co_writev(BlockDriverState *bs, int64_t sector_num,
                          int nb_sectors, QEMUIOVector *qiov)
 {
-    BDRVNBDState *s = bs->opaque;
+    return nbd_client_co_writev(bs, sector_num, nb_sectors, qiov);
-
-    return nbd_client_session_co_writev(&s->client, sector_num,
-                                        nb_sectors, qiov);
 }
 
 static int nbd_co_flush(BlockDriverState *bs)
 {
-    BDRVNBDState *s = bs->opaque;
+    return nbd_client_co_flush(bs);
-
-    return nbd_client_session_co_flush(&s->client);
 }
 
 static void nbd_refresh_limits(BlockDriverState *bs, Error **errp)
@@ -310,10 +308,7 @@ static void nbd_refresh_limits(BlockDriverState *bs, Error **errp)
 static int nbd_co_discard(BlockDriverState *bs, int64_t sector_num,
                           int nb_sectors)
 {
-    BDRVNBDState *s = bs->opaque;
+    return nbd_client_co_discard(bs, sector_num, nb_sectors);
-
-    return nbd_client_session_co_discard(&s->client, sector_num,
-                                         nb_sectors);
 }
 
 static void nbd_close(BlockDriverState *bs)
@@ -321,7 +316,7 @@ static void nbd_close(BlockDriverState *bs)
     BDRVNBDState *s = bs->opaque;
 
     qemu_opts_del(s->socket_opts);
-    nbd_client_session_close(&s->client);
+    nbd_client_close(bs);
 }
 
 static int64_t nbd_getlength(BlockDriverState *bs)
@@ -333,17 +328,13 @@ static int64_t nbd_getlength(BlockDriverState *bs)
 
 static void nbd_detach_aio_context(BlockDriverState *bs)
 {
-    BDRVNBDState *s = bs->opaque;
+    nbd_client_detach_aio_context(bs);
-
-    nbd_client_session_detach_aio_context(&s->client);
 }
 
 static void nbd_attach_aio_context(BlockDriverState *bs,
                                    AioContext *new_context)
 {
-    BDRVNBDState *s = bs->opaque;
+    nbd_client_attach_aio_context(bs, new_context);
-
-    nbd_client_session_attach_aio_context(&s->client, new_context);
 }
 
 static void nbd_refresh_filename(BlockDriverState *bs)

block/qcow2.c

@@ -2521,15 +2521,12 @@ static int qcow2_save_vmstate(BlockDriverState *bs, QEMUIOVector *qiov,
 {
     BDRVQcowState *s = bs->opaque;
     int64_t total_sectors = bs->total_sectors;
-    int growable = bs->growable;
     bool zero_beyond_eof = bs->zero_beyond_eof;
     int ret;
 
     BLKDBG_EVENT(bs->file, BLKDBG_VMSTATE_SAVE);
-    bs->growable = 1;
     bs->zero_beyond_eof = false;
     ret = bdrv_pwritev(bs, qcow2_vm_state_offset(s) + pos, qiov);
-    bs->growable = growable;
     bs->zero_beyond_eof = zero_beyond_eof;
 
     /* bdrv_co_do_writev will have increased the total_sectors value to include
@@ -2544,15 +2541,12 @@ static int qcow2_load_vmstate(BlockDriverState *bs, uint8_t *buf,
                               int64_t pos, int size)
 {
     BDRVQcowState *s = bs->opaque;
-    int growable = bs->growable;
     bool zero_beyond_eof = bs->zero_beyond_eof;
     int ret;
 
     BLKDBG_EVENT(bs->file, BLKDBG_VMSTATE_LOAD);
-    bs->growable = 1;
     bs->zero_beyond_eof = false;
     ret = bdrv_pread(bs, qcow2_vm_state_offset(s) + pos, buf, size);
-    bs->growable = growable;
     bs->zero_beyond_eof = zero_beyond_eof;
 
     return ret;

block/raw-posix.c

@@ -1047,7 +1047,7 @@ static int aio_worker(void *arg)
     switch (aiocb->aio_type & QEMU_AIO_TYPE_MASK) {
     case QEMU_AIO_READ:
         ret = handle_aiocb_rw(aiocb);
-        if (ret >= 0 && ret < aiocb->aio_nbytes && aiocb->bs->growable) {
+        if (ret >= 0 && ret < aiocb->aio_nbytes) {
             iov_memset(aiocb->aio_iov, aiocb->aio_niov, ret,
                       0, aiocb->aio_nbytes - ret);
 

block/raw-win32.c

@@ -101,7 +101,7 @@ static int aio_worker(void *arg)
     switch (aiocb->aio_type & QEMU_AIO_TYPE_MASK) {
     case QEMU_AIO_READ:
         count = handle_aiocb_rw(aiocb);
-        if (count < aiocb->aio_nbytes && aiocb->bs->growable) {
+        if (count < aiocb->aio_nbytes) {
             /* A short read means that we have reached EOF. Pad the buffer
              * with zeros for bytes after EOF. */
             iov_memset(aiocb->aio_iov, aiocb->aio_niov, count,

block/sheepdog.c

@@ -1730,7 +1730,7 @@ static int sd_create(const char *filename, QemuOpts *opts,
         BlockDriver *drv;
 
         /* Currently, only Sheepdog backing image is supported. */
-        drv = bdrv_find_protocol(backing_file, true);
+        drv = bdrv_find_protocol(backing_file, true, NULL);
         if (!drv || strcmp(drv->protocol_name, "sheepdog") != 0) {
             error_setg(errp, "backing_file must be a sheepdog image");
             ret = -EINVAL;
@@ -2117,7 +2117,7 @@ static coroutine_fn int sd_co_writev(BlockDriverState *bs, int64_t sector_num,
     int64_t offset = (sector_num + nb_sectors) * BDRV_SECTOR_SIZE;
     BDRVSheepdogState *s = bs->opaque;
 
-    if (bs->growable && offset > s->inode.vdi_size) {
+    if (offset > s->inode.vdi_size) {
         ret = sd_truncate(bs, offset);
         if (ret < 0) {
             return ret;

block/vmdk.c

@@ -843,8 +843,7 @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs,
         }
 
         extent_path = g_malloc0(PATH_MAX);
-        path_combine(extent_path, sizeof(extent_path),
+        path_combine(extent_path, PATH_MAX, desc_file_path, fname);
-                desc_file_path, fname);
         extent_file = NULL;
         ret = bdrv_open(&extent_file, extent_path, NULL, NULL,
                         bs->open_flags | BDRV_O_PROTOCOL, NULL, errp);

blockdev.c

@@ -354,13 +354,11 @@ static BlockBackend *blockdev_init(const char *file, QDict *bs_opts,
     ThrottleConfig cfg;
     int snapshot = 0;
     bool copy_on_read;
-    int ret;
     Error *error = NULL;
     QemuOpts *opts;
     const char *id;
     bool has_driver_specific_opts;
     BlockdevDetectZeroesOptions detect_zeroes;
-    BlockDriver *drv = NULL;
 
     /* Check common options by copying from bs_opts to opts, all other options
      * stay in bs_opts for processing by bdrv_open(). */
@@ -426,11 +424,11 @@ static BlockBackend *blockdev_init(const char *file, QDict *bs_opts,
             goto early_err;
         }
 
-        drv = bdrv_find_format(buf);
+        if (qdict_haskey(bs_opts, "driver")) {
-        if (!drv) {
+            error_setg(errp, "Cannot specify both 'driver' and 'format'");
-            error_setg(errp, "'%s' invalid format", buf);
             goto early_err;
         }
+        qdict_put(bs_opts, "driver", qstring_from_str(buf));
     }
 
     /* disk I/O throttling */
@@ -505,13 +503,46 @@ static BlockBackend *blockdev_init(const char *file, QDict *bs_opts,
     }
 
     /* init */
-    blk = blk_new_with_bs(qemu_opts_id(opts), errp);
+    if ((!file || !*file) && !has_driver_specific_opts) {
-    if (!blk) {
+        blk = blk_new_with_bs(qemu_opts_id(opts), errp);
-        goto early_err;
+        if (!blk) {
+            goto early_err;
+        }
+
+        bs = blk_bs(blk);
+        bs->open_flags = snapshot ? BDRV_O_SNAPSHOT : 0;
+        bs->read_only = ro;
+
+        QDECREF(bs_opts);
+    } else {
+        if (file && !*file) {
+            file = NULL;
+        }
+
+        if (snapshot) {
+            /* always use cache=unsafe with snapshot */
+            bdrv_flags &= ~BDRV_O_CACHE_MASK;
+            bdrv_flags |= (BDRV_O_SNAPSHOT|BDRV_O_CACHE_WB|BDRV_O_NO_FLUSH);
+        }
+
+        if (copy_on_read) {
+            bdrv_flags |= BDRV_O_COPY_ON_READ;
+        }
+
+        if (runstate_check(RUN_STATE_INMIGRATE)) {
+            bdrv_flags |= BDRV_O_INCOMING;
+        }
+
+        bdrv_flags |= ro ? 0 : BDRV_O_RDWR;
+
+        blk = blk_new_open(qemu_opts_id(opts), file, NULL, bs_opts, bdrv_flags,
+                           errp);
+        if (!blk) {
+            goto err_no_bs_opts;
+        }
+        bs = blk_bs(blk);
     }
-    bs = blk_bs(blk);
+
-    bs->open_flags = snapshot ? BDRV_O_SNAPSHOT : 0;
-    bs->read_only = ro;
     bs->detect_zeroes = detect_zeroes;
 
     bdrv_set_on_error(bs, on_read_error, on_write_error);
@@ -522,53 +553,14 @@ static BlockBackend *blockdev_init(const char *file, QDict *bs_opts,
         bdrv_set_io_limits(bs, &cfg);
     }
 
-    if (!file || !*file) {
-        if (has_driver_specific_opts) {
-            file = NULL;
-        } else {
-            QDECREF(bs_opts);
-            qemu_opts_del(opts);
-            return blk;
-        }
-    }
-    if (snapshot) {
-        /* always use cache=unsafe with snapshot */
-        bdrv_flags &= ~BDRV_O_CACHE_MASK;
-        bdrv_flags |= (BDRV_O_SNAPSHOT|BDRV_O_CACHE_WB|BDRV_O_NO_FLUSH);
-    }
-
-    if (copy_on_read) {
-        bdrv_flags |= BDRV_O_COPY_ON_READ;
-    }
-
-    if (runstate_check(RUN_STATE_INMIGRATE)) {
-        bdrv_flags |= BDRV_O_INCOMING;
-    }
-
-    bdrv_flags |= ro ? 0 : BDRV_O_RDWR;
-
-    QINCREF(bs_opts);
-    ret = bdrv_open(&bs, file, NULL, bs_opts, bdrv_flags, drv, &error);
-    assert(bs == blk_bs(blk));
-
-    if (ret < 0) {
-        error_setg(errp, "could not open disk image %s: %s",
-                   file ?: blk_name(blk), error_get_pretty(error));
-        error_free(error);
-        goto err;
-    }
-
     if (bdrv_key_required(bs)) {
         autostart = 0;
     }
 
-    QDECREF(bs_opts);
+err_no_bs_opts:
     qemu_opts_del(opts);
-
     return blk;
 
-err:
-    blk_unref(blk);
 early_err:
     qemu_opts_del(opts);
 err_no_opts:

cpus.c

@@ -361,15 +361,19 @@ static void icount_warp_rt(void *opaque)
 void qtest_clock_warp(int64_t dest)
 {
     int64_t clock = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL);
+    AioContext *aio_context;
     assert(qtest_enabled());
+    aio_context = qemu_get_aio_context();
     while (clock < dest) {
         int64_t deadline = qemu_clock_deadline_ns_all(QEMU_CLOCK_VIRTUAL);
         int64_t warp = qemu_soonest_timeout(dest - clock, deadline);
+
         seqlock_write_lock(&timers_state.vm_clock_seqlock);
         timers_state.qemu_icount_bias += warp;
         seqlock_write_unlock(&timers_state.vm_clock_seqlock);
 
         qemu_clock_run_timers(QEMU_CLOCK_VIRTUAL);
+        timerlist_run_timers(aio_context->tlg.tl[QEMU_CLOCK_VIRTUAL]);
         clock = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL);
     }
     qemu_clock_notify(QEMU_CLOCK_VIRTUAL);

hmp.c

@@ -16,6 +16,7 @@
 #include "hmp.h"
 #include "net/net.h"
 #include "sysemu/char.h"
+#include "sysemu/block-backend.h"
 #include "qemu/option.h"
 #include "qemu/timer.h"
 #include "qmp-commands.h"
@@ -1720,14 +1721,14 @@ void hmp_chardev_remove(Monitor *mon, const QDict *qdict)
 
 void hmp_qemu_io(Monitor *mon, const QDict *qdict)
 {
-    BlockDriverState *bs;
+    BlockBackend *blk;
     const char* device = qdict_get_str(qdict, "device");
     const char* command = qdict_get_str(qdict, "command");
     Error *err = NULL;
 
-    bs = bdrv_find(device);
+    blk = blk_by_name(device);
-    if (bs) {
+    if (blk) {
-        qemuio_command(bs, command);
+        qemuio_command(blk, command);
     } else {
         error_set(&err, QERR_DEVICE_NOT_FOUND, device);
     }

hw/block/dataplane/virtio-blk.c

@@ -16,7 +16,9 @@
 #include "qemu/iov.h"
 #include "qemu/thread.h"
 #include "qemu/error-report.h"
+#include "hw/virtio/virtio-access.h"
 #include "hw/virtio/dataplane/vring.h"
+#include "hw/virtio/dataplane/vring-accessors.h"
 #include "sysemu/block-backend.h"
 #include "hw/virtio/virtio-blk.h"
 #include "virtio-blk.h"
@@ -75,7 +77,7 @@ static void complete_request_vring(VirtIOBlockReq *req, unsigned char status)
     VirtIOBlockDataPlane *s = req->dev->dataplane;
     stb_p(&req->in->status, status);
 
-    vring_push(&req->dev->dataplane->vring, &req->elem,
+    vring_push(s->vdev, &req->dev->dataplane->vring, &req->elem,
                req->qiov.size + sizeof(*req->in));
 
     /* Suppress notification to guest by BH and its scheduled

hw/block/xen_disk.c

@@ -40,6 +40,8 @@
 #include "xen_blkif.h"
 #include "sysemu/blockdev.h"
 #include "sysemu/block-backend.h"
+#include "qapi/qmp/qdict.h"
+#include "qapi/qmp/qstring.h"
 
 /* ------------------------------------------------------------- */
 
@@ -897,30 +899,23 @@ static int blk_connect(struct XenDevice *xendev)
     blkdev->dinfo = drive_get(IF_XEN, 0, index);
     if (!blkdev->dinfo) {
         Error *local_err = NULL;
-        BlockBackend *blk;
+        QDict *options = NULL;
-        BlockDriver *drv;
+
-        BlockDriverState *bs;
+        if (strcmp(blkdev->fileproto, "<unset>")) {
+            options = qdict_new();
+            qdict_put(options, "driver", qstring_from_str(blkdev->fileproto));
+        }
 
         /* setup via xenbus -> create new block driver instance */
         xen_be_printf(&blkdev->xendev, 2, "create new bdrv (xenbus setup)\n");
-        blk = blk_new_with_bs(blkdev->dev, NULL);
+        blkdev->blk = blk_new_open(blkdev->dev, blkdev->filename, NULL, options,
-        if (!blk) {
+                                   qflags, &local_err);
-            return -1;
+        if (!blkdev->blk) {
-        }
-        blkdev->blk = blk;
-
-        bs = blk_bs(blk);
-        drv = bdrv_find_whitelisted_format(blkdev->fileproto, readonly);
-        if (bdrv_open(&bs, blkdev->filename, NULL, NULL, qflags,
-                      drv, &local_err) != 0) {
             xen_be_printf(&blkdev->xendev, 0, "error: %s\n",
                           error_get_pretty(local_err));
             error_free(local_err);
-            blk_unref(blk);
-            blkdev->blk = NULL;
             return -1;
         }
-        assert(bs == blk_bs(blk));
     } else {
         /* setup via qemu cmdline -> already setup for us */
         xen_be_printf(&blkdev->xendev, 2, "get configured bdrv (cmdline setup)\n");

hw/scsi/virtio-scsi-dataplane.c

@@ -94,7 +94,7 @@ void virtio_scsi_vring_push_notify(VirtIOSCSIReq *req)
 {
     VirtIODevice *vdev = VIRTIO_DEVICE(req->vring->parent);
 
-    vring_push(&req->vring->vring, &req->elem,
+    vring_push(vdev, &req->vring->vring, &req->elem,
                req->qsgl.size + req->resp_iov.size);
 
     if (vring_should_notify(vdev, &req->vring->vring)) {

hw/virtio/Makefile.objs

@@ -2,7 +2,7 @@ common-obj-y += virtio-rng.o
 common-obj-$(CONFIG_VIRTIO_PCI) += virtio-pci.o
 common-obj-y += virtio-bus.o
 common-obj-y += virtio-mmio.o
-common-obj-$(CONFIG_VIRTIO) += dataplane/
+obj-$(CONFIG_VIRTIO) += dataplane/
 
 obj-y += virtio.o virtio-balloon.o 
 obj-$(CONFIG_LINUX) += vhost.o vhost-backend.o vhost-user.o

hw/virtio/dataplane/Makefile.objs

@@ -1 +1 @@
-common-obj-y += vring.o
+obj-y += vring.o

hw/virtio/dataplane/vring.c

@@ -18,7 +18,9 @@
 #include "hw/hw.h"
 #include "exec/memory.h"
 #include "exec/address-spaces.h"
+#include "hw/virtio/virtio-access.h"
 #include "hw/virtio/dataplane/vring.h"
+#include "hw/virtio/dataplane/vring-accessors.h"
 #include "qemu/error-report.h"
 
 /* vring_map can be coupled with vring_unmap or (if you still have the
@@ -83,7 +85,7 @@ bool vring_setup(Vring *vring, VirtIODevice *vdev, int n)
     vring_init(&vring->vr, virtio_queue_get_num(vdev, n), vring_ptr, 4096);
 
     vring->last_avail_idx = virtio_queue_get_last_avail_idx(vdev, n);
-    vring->last_used_idx = vring->vr.used->idx;
+    vring->last_used_idx = vring_get_used_idx(vdev, vring);
     vring->signalled_used = 0;
     vring->signalled_used_valid = false;
 
@@ -104,7 +106,7 @@ void vring_teardown(Vring *vring, VirtIODevice *vdev, int n)
 void vring_disable_notification(VirtIODevice *vdev, Vring *vring)
 {
     if (!(vdev->guest_features & (1 << VIRTIO_RING_F_EVENT_IDX))) {
-        vring->vr.used->flags |= VRING_USED_F_NO_NOTIFY;
+        vring_set_used_flags(vdev, vring, VRING_USED_F_NO_NOTIFY);
     }
 }
 
@@ -117,10 +119,10 @@ bool vring_enable_notification(VirtIODevice *vdev, Vring *vring)
     if (vdev->guest_features & (1 << VIRTIO_RING_F_EVENT_IDX)) {
         vring_avail_event(&vring->vr) = vring->vr.avail->idx;
     } else {
-        vring->vr.used->flags &= ~VRING_USED_F_NO_NOTIFY;
+        vring_clear_used_flags(vdev, vring, VRING_USED_F_NO_NOTIFY);
     }
     smp_mb(); /* ensure update is seen before reading avail_idx */
-    return !vring_more_avail(vring);
+    return !vring_more_avail(vdev, vring);
 }
 
 /* This is stolen from linux/drivers/vhost/vhost.c:vhost_notify() */
@@ -134,12 +136,13 @@ bool vring_should_notify(VirtIODevice *vdev, Vring *vring)
     smp_mb();
 
     if ((vdev->guest_features & (1 << VIRTIO_F_NOTIFY_ON_EMPTY)) &&
-        unlikely(vring->vr.avail->idx == vring->last_avail_idx)) {
+        unlikely(!vring_more_avail(vdev, vring))) {
         return true;
     }
 
     if (!(vdev->guest_features & (1 << VIRTIO_RING_F_EVENT_IDX))) {
-        return !(vring->vr.avail->flags & VRING_AVAIL_F_NO_INTERRUPT);
+        return !(vring_get_avail_flags(vdev, vring) &
+                 VRING_AVAIL_F_NO_INTERRUPT);
     }
     old = vring->signalled_used;
     v = vring->signalled_used_valid;
@@ -202,9 +205,19 @@ static int get_desc(Vring *vring, VirtQueueElement *elem,
     return 0;
 }
 
+static void copy_in_vring_desc(VirtIODevice *vdev,
+                               const struct vring_desc *guest,
+                               struct vring_desc *host)
+{
+    host->addr = virtio_ldq_p(vdev, &guest->addr);
+    host->len = virtio_ldl_p(vdev, &guest->len);
+    host->flags = virtio_lduw_p(vdev, &guest->flags);
+    host->next = virtio_lduw_p(vdev, &guest->next);
+}
+
 /* This is stolen from linux/drivers/vhost/vhost.c. */
-static int get_indirect(Vring *vring, VirtQueueElement *elem,
+static int get_indirect(VirtIODevice *vdev, Vring *vring,
-                        struct vring_desc *indirect)
+                        VirtQueueElement *elem, struct vring_desc *indirect)
 {
     struct vring_desc desc;
     unsigned int i = 0, count, found = 0;
@@ -244,7 +257,7 @@ static int get_indirect(Vring *vring, VirtQueueElement *elem,
             vring->broken = true;
             return -EFAULT;
         }
-        desc = *desc_ptr;
+        copy_in_vring_desc(vdev, desc_ptr, &desc);
         memory_region_unref(mr);
 
         /* Ensure descriptor has been loaded before accessing fields */
@@ -320,7 +333,7 @@ int vring_pop(VirtIODevice *vdev, Vring *vring,
 
     /* Check it isn't doing very strange things with descriptor numbers. */
     last_avail_idx = vring->last_avail_idx;
-    avail_idx = vring->vr.avail->idx;
+    avail_idx = vring_get_avail_idx(vdev, vring);
     barrier(); /* load indices now and not again later */
 
     if (unlikely((uint16_t)(avail_idx - last_avail_idx) > num)) {
@@ -341,7 +354,7 @@ int vring_pop(VirtIODevice *vdev, Vring *vring,
 
     /* Grab the next descriptor number they're advertising, and increment
      * the index we've seen. */
-    head = vring->vr.avail->ring[last_avail_idx % num];
+    head = vring_get_avail_ring(vdev, vring, last_avail_idx % num);
 
     elem->index = head;
 
@@ -365,13 +378,13 @@ int vring_pop(VirtIODevice *vdev, Vring *vring,
             ret = -EFAULT;
             goto out;
         }
-        desc = vring->vr.desc[i];
+        copy_in_vring_desc(vdev, &vring->vr.desc[i], &desc);
 
         /* Ensure descriptor is loaded before accessing fields */
         barrier();
 
         if (desc.flags & VRING_DESC_F_INDIRECT) {
-            ret = get_indirect(vring, elem, &desc);
+            ret = get_indirect(vdev, vring, elem, &desc);
             if (ret < 0) {
                 goto out;
             }
@@ -407,9 +420,9 @@ out:
  *
  * Stolen from linux/drivers/vhost/vhost.c.
  */
-void vring_push(Vring *vring, VirtQueueElement *elem, int len)
+void vring_push(VirtIODevice *vdev, Vring *vring, VirtQueueElement *elem,
+                int len)
 {
-    struct vring_used_elem *used;
     unsigned int head = elem->index;
     uint16_t new;
 
@@ -422,14 +435,16 @@ void vring_push(Vring *vring, VirtQueueElement *elem, int len)
 
     /* The virtqueue contains a ring of used buffers.  Get a pointer to the
      * next entry in that used ring. */
-    used = &vring->vr.used->ring[vring->last_used_idx % vring->vr.num];
+    vring_set_used_ring_id(vdev, vring, vring->last_used_idx % vring->vr.num,
-    used->id = head;
+                           head);
-    used->len = len;
+    vring_set_used_ring_len(vdev, vring, vring->last_used_idx % vring->vr.num,
+                            len);
 
     /* Make sure buffer is written before we update index. */
     smp_wmb();
 
-    new = vring->vr.used->idx = ++vring->last_used_idx;
+    new = ++vring->last_used_idx;
+    vring_set_used_idx(vdev, vring, new);
     if (unlikely((int16_t)(new - vring->signalled_used) < (uint16_t)1)) {
         vring->signalled_used_valid = false;
     }

include/block/block.h

@@ -168,7 +168,8 @@ void bdrv_io_limits_disable(BlockDriverState *bs);
 void bdrv_init(void);
 void bdrv_init_with_whitelist(void);
 BlockDriver *bdrv_find_protocol(const char *filename,
-                                bool allow_protocol_prefix);
+                                bool allow_protocol_prefix,
+                                Error **errp);
 BlockDriver *bdrv_find_format(const char *format_name);
 BlockDriver *bdrv_find_whitelisted_format(const char *format_name,
                                           bool readonly);

include/block/block_int.h

@@ -369,9 +369,6 @@ struct BlockDriverState {
     /* I/O Limits */
     BlockLimits bl;
 
-    /* Whether the disk can expand beyond total_sectors */
-    int growable;
-
     /* Whether produces zeros when read beyond eof */
     bool zero_beyond_eof;
 

include/block/nbd.h

@@ -99,7 +99,6 @@ void nbd_export_close_all(void);
 
 NBDClient *nbd_client_new(NBDExport *exp, int csock,
                           void (*close)(NBDClient *));
-void nbd_client_close(NBDClient *client);
 void nbd_client_get(NBDClient *client);
 void nbd_client_put(NBDClient *client);
 

include/hw/virtio/dataplane/vring-accessors.h

@@ -0,0 +1,75 @@
+#ifndef VRING_ACCESSORS_H
+#define VRING_ACCESSORS_H
+
+#include "hw/virtio/virtio_ring.h"
+#include "hw/virtio/virtio.h"
+#include "hw/virtio/virtio-access.h"
+
+static inline uint16_t vring_get_used_idx(VirtIODevice *vdev, Vring *vring)
+{
+    return virtio_tswap16(vdev, vring->vr.used->idx);
+}
+
+static inline void vring_set_used_idx(VirtIODevice *vdev, Vring *vring,
+                                      uint16_t idx)
+{
+    vring->vr.used->idx = virtio_tswap16(vdev, idx);
+}
+
+static inline uint16_t vring_get_avail_idx(VirtIODevice *vdev, Vring *vring)
+{
+    return virtio_tswap16(vdev, vring->vr.avail->idx);
+}
+
+static inline uint16_t vring_get_avail_ring(VirtIODevice *vdev, Vring *vring,
+                                            int i)
+{
+    return virtio_tswap16(vdev, vring->vr.avail->ring[i]);
+}
+
+static inline void vring_set_used_ring_id(VirtIODevice *vdev, Vring *vring,
+                                          int i, uint32_t id)
+{
+    vring->vr.used->ring[i].id = virtio_tswap32(vdev, id);
+}
+
+static inline void vring_set_used_ring_len(VirtIODevice *vdev, Vring *vring,
+                                          int i, uint32_t len)
+{
+    vring->vr.used->ring[i].len = virtio_tswap32(vdev, len);
+}
+
+static inline uint16_t vring_get_used_flags(VirtIODevice *vdev, Vring *vring)
+{
+    return virtio_tswap16(vdev, vring->vr.used->flags);
+}
+
+static inline uint16_t vring_get_avail_flags(VirtIODevice *vdev, Vring *vring)
+{
+    return virtio_tswap16(vdev, vring->vr.avail->flags);
+}
+
+static inline void vring_set_used_flags(VirtIODevice *vdev, Vring *vring,
+                                        uint16_t flags)
+{
+    vring->vr.used->flags |= virtio_tswap16(vdev, flags);
+}
+
+static inline void vring_clear_used_flags(VirtIODevice *vdev, Vring *vring,
+                                          uint16_t flags)
+{
+    vring->vr.used->flags &= virtio_tswap16(vdev, ~flags);
+}
+
+static inline unsigned int vring_get_num(Vring *vring)
+{
+    return vring->vr.num;
+}
+
+/* Are there more descriptors available? */
+static inline bool vring_more_avail(VirtIODevice *vdev, Vring *vring)
+{
+    return vring_get_avail_idx(vdev, vring) != vring->last_avail_idx;
+}
+
+#endif

include/hw/virtio/dataplane/vring.h

@@ -31,17 +31,6 @@ typedef struct {
     bool broken;                    /* was there a fatal error? */
 } Vring;
 
-static inline unsigned int vring_get_num(Vring *vring)
-{
-    return vring->vr.num;
-}
-
-/* Are there more descriptors available? */
-static inline bool vring_more_avail(Vring *vring)
-{
-    return vring->vr.avail->idx != vring->last_avail_idx;
-}
-
 /* Fail future vring_pop() and vring_push() calls until reset */
 static inline void vring_set_broken(Vring *vring)
 {
@@ -54,6 +43,7 @@ void vring_disable_notification(VirtIODevice *vdev, Vring *vring);
 bool vring_enable_notification(VirtIODevice *vdev, Vring *vring);
 bool vring_should_notify(VirtIODevice *vdev, Vring *vring);
 int vring_pop(VirtIODevice *vdev, Vring *vring, VirtQueueElement *elem);
-void vring_push(Vring *vring, VirtQueueElement *elem, int len);
+void vring_push(VirtIODevice *vdev, Vring *vring, VirtQueueElement *elem,
+                int len);
 
 #endif /* VRING_H */

include/qemu-io.h

@@ -22,7 +22,7 @@
 
 #define CMD_FLAG_GLOBAL ((int)0x80000000) /* don't iterate "args" */
 
-typedef int (*cfunc_t)(BlockDriverState *bs, int argc, char **argv);
+typedef int (*cfunc_t)(BlockBackend *blk, int argc, char **argv);
 typedef void (*helpfunc_t)(void);
 
 typedef struct cmdinfo {
@@ -40,7 +40,7 @@ typedef struct cmdinfo {
 
 extern bool qemuio_misalign;
 
-bool qemuio_command(BlockDriverState *bs, const char *cmd);
+bool qemuio_command(BlockBackend *blk, const char *cmd);
 
 void qemuio_add_command(const cmdinfo_t *ci);
 int qemuio_command_usage(const cmdinfo_t *ci);

include/sysemu/block-backend.h

@@ -62,6 +62,9 @@ typedef struct BlockDevOps {
 
 BlockBackend *blk_new(const char *name, Error **errp);
 BlockBackend *blk_new_with_bs(const char *name, Error **errp);
+BlockBackend *blk_new_open(const char *name, const char *filename,
+                           const char *reference, QDict *options, int flags,
+                           Error **errp);
 void blk_ref(BlockBackend *blk);
 void blk_unref(BlockBackend *blk);
 const char *blk_name(BlockBackend *blk);
@@ -91,6 +94,7 @@ int blk_pread(BlockBackend *blk, int64_t offset, void *buf, int count);
 int blk_pwrite(BlockBackend *blk, int64_t offset, const void *buf, int count);
 int64_t blk_getlength(BlockBackend *blk);
 void blk_get_geometry(BlockBackend *blk, uint64_t *nb_sectors_ptr);
+int64_t blk_nb_sectors(BlockBackend *blk);
 BlockAIOCB *blk_aio_readv(BlockBackend *blk, int64_t sector_num,
                           QEMUIOVector *iov, int nb_sectors,
                           BlockCompletionFunc *cb, void *opaque);
@@ -151,5 +155,14 @@ BlockAcctStats *blk_get_stats(BlockBackend *blk);
 
 void *blk_aio_get(const AIOCBInfo *aiocb_info, BlockBackend *blk,
                   BlockCompletionFunc *cb, void *opaque);
+int coroutine_fn blk_co_write_zeroes(BlockBackend *blk, int64_t sector_num,
+                                     int nb_sectors, BdrvRequestFlags flags);
+int blk_write_compressed(BlockBackend *blk, int64_t sector_num,
+                         const uint8_t *buf, int nb_sectors);
+int blk_truncate(BlockBackend *blk, int64_t offset);
+int blk_discard(BlockBackend *blk, int64_t sector_num, int nb_sectors);
+int blk_save_vmstate(BlockBackend *blk, const uint8_t *buf,
+                     int64_t pos, int size);
+int blk_load_vmstate(BlockBackend *blk, uint8_t *buf, int64_t pos, int size);
 
 #endif

nbd.c

@@ -874,7 +874,7 @@ void nbd_client_put(NBDClient *client)
 {
     if (--client->refcount == 0) {
         /* The last reference should be dropped by client->close,
-         * which is called by nbd_client_close.
+         * which is called by client_close.
          */
         assert(client->closing);
 
@@ -889,7 +889,7 @@ void nbd_client_put(NBDClient *client)
     }
 }
 
-void nbd_client_close(NBDClient *client)
+static void client_close(NBDClient *client)
 {
     if (client->closing) {
         return;
@@ -1026,7 +1026,7 @@ void nbd_export_close(NBDExport *exp)
 
     nbd_export_get(exp);
     QTAILQ_FOREACH_SAFE(client, &exp->clients, next, next) {
-        nbd_client_close(client);
+        client_close(client);
     }
     nbd_export_set_name(exp, NULL);
     nbd_export_put(exp);
@@ -1311,7 +1311,7 @@ done:
 
 out:
     nbd_request_put(req);
-    nbd_client_close(client);
+    client_close(client);
 }
 
 static void nbd_read(void *opaque)

qemu-img.c

@@ -261,6 +261,7 @@ static int print_block_option_help(const char *filename, const char *fmt)
 {
     BlockDriver *drv, *proto_drv;
     QemuOptsList *create_opts = NULL;
+    Error *local_err = NULL;
 
     /* Find driver and parse its options */
     drv = bdrv_find_format(fmt);
@@ -271,9 +272,10 @@ static int print_block_option_help(const char *filename, const char *fmt)
 
     create_opts = qemu_opts_append(create_opts, drv->create_opts);
     if (filename) {
-        proto_drv = bdrv_find_protocol(filename, true);
+        proto_drv = bdrv_find_protocol(filename, true, &local_err);
         if (!proto_drv) {
-            error_report("Unknown protocol '%s'", filename);
+            qerror_report_err(local_err);
+            error_free(local_err);
             qemu_opts_free(create_opts);
             return 1;
         }
@@ -291,32 +293,24 @@ static BlockBackend *img_open(const char *id, const char *filename,
 {
     BlockBackend *blk;
     BlockDriverState *bs;
-    BlockDriver *drv;
     char password[256];
     Error *local_err = NULL;
-    int ret;
+    QDict *options = NULL;
-
-    blk = blk_new_with_bs(id, &error_abort);
-    bs = blk_bs(blk);
 
     if (fmt) {
-        drv = bdrv_find_format(fmt);
+        options = qdict_new();
-        if (!drv) {
+        qdict_put(options, "driver", qstring_from_str(fmt));
-            error_report("Unknown file format '%s'", fmt);
-            goto fail;
-        }
-    } else {
-        drv = NULL;
     }
 
-    ret = bdrv_open(&bs, filename, NULL, NULL, flags, drv, &local_err);
+    blk = blk_new_open(id, filename, NULL, options, flags, &local_err);
-    if (ret < 0) {
+    if (!blk) {
         error_report("Could not open '%s': %s", filename,
                      error_get_pretty(local_err));
         error_free(local_err);
         goto fail;
     }
 
+    bs = blk_bs(blk);
     if (bdrv_is_encrypted(bs) && require_io) {
         qprintf(quiet, "Disk image '%s' is encrypted.\n", filename);
         if (read_password(password, sizeof(password)) < 0) {
@@ -1016,19 +1010,19 @@ static int64_t sectors_to_process(int64_t total, int64_t from)
  * Returns 0 in case sectors are filled with 0, 1 if sectors contain non-zero
  * data and negative value on error.
  *
- * @param bs:  Driver used for accessing file
+ * @param blk:  BlockBackend for the image
  * @param sect_num: Number of first sector to check
  * @param sect_count: Number of sectors to check
  * @param filename: Name of disk file we are checking (logging purpose)
  * @param buffer: Allocated buffer for storing read data
  * @param quiet: Flag for quiet mode
  */
-static int check_empty_sectors(BlockDriverState *bs, int64_t sect_num,
+static int check_empty_sectors(BlockBackend *blk, int64_t sect_num,
                                int sect_count, const char *filename,
                                uint8_t *buffer, bool quiet)
 {
     int pnum, ret = 0;
-    ret = bdrv_read(bs, sect_num, buffer, sect_count);
+    ret = blk_read(blk, sect_num, buffer, sect_count);
     if (ret < 0) {
         error_report("Error while reading offset %" PRId64 " of %s: %s",
                      sectors_to_bytes(sect_num), filename, strerror(-ret));
@@ -1138,16 +1132,16 @@ static int img_compare(int argc, char **argv)
     }
     bs2 = blk_bs(blk2);
 
-    buf1 = qemu_blockalign(bs1, IO_BUF_SIZE);
+    buf1 = blk_blockalign(blk1, IO_BUF_SIZE);
-    buf2 = qemu_blockalign(bs2, IO_BUF_SIZE);
+    buf2 = blk_blockalign(blk2, IO_BUF_SIZE);
-    total_sectors1 = bdrv_nb_sectors(bs1);
+    total_sectors1 = blk_nb_sectors(blk1);
     if (total_sectors1 < 0) {
         error_report("Can't get size of %s: %s",
                      filename1, strerror(-total_sectors1));
         ret = 4;
         goto out;
     }
-    total_sectors2 = bdrv_nb_sectors(bs2);
+    total_sectors2 = blk_nb_sectors(blk2);
     if (total_sectors2 < 0) {
         error_report("Can't get size of %s: %s",
                      filename2, strerror(-total_sectors2));
@@ -1189,7 +1183,7 @@ static int img_compare(int argc, char **argv)
 
         if (allocated1 == allocated2) {
             if (allocated1) {
-                ret = bdrv_read(bs1, sector_num, buf1, nb_sectors);
+                ret = blk_read(blk1, sector_num, buf1, nb_sectors);
                 if (ret < 0) {
                     error_report("Error while reading offset %" PRId64 " of %s:"
                                  " %s", sectors_to_bytes(sector_num), filename1,
@@ -1197,7 +1191,7 @@ static int img_compare(int argc, char **argv)
                     ret = 4;
                     goto out;
                 }
-                ret = bdrv_read(bs2, sector_num, buf2, nb_sectors);
+                ret = blk_read(blk2, sector_num, buf2, nb_sectors);
                 if (ret < 0) {
                     error_report("Error while reading offset %" PRId64
                                  " of %s: %s", sectors_to_bytes(sector_num),
@@ -1224,10 +1218,10 @@ static int img_compare(int argc, char **argv)
             }
 
             if (allocated1) {
-                ret = check_empty_sectors(bs1, sector_num, nb_sectors,
+                ret = check_empty_sectors(blk1, sector_num, nb_sectors,
                                           filename1, buf1, quiet);
             } else {
-                ret = check_empty_sectors(bs2, sector_num, nb_sectors,
+                ret = check_empty_sectors(blk2, sector_num, nb_sectors,
                                           filename2, buf1, quiet);
             }
             if (ret) {
@@ -1244,18 +1238,18 @@ static int img_compare(int argc, char **argv)
     }
 
     if (total_sectors1 != total_sectors2) {
-        BlockDriverState *bs_over;
+        BlockBackend *blk_over;
         int64_t total_sectors_over;
         const char *filename_over;
 
         qprintf(quiet, "Warning: Image size mismatch!\n");
         if (total_sectors1 > total_sectors2) {
             total_sectors_over = total_sectors1;
-            bs_over = bs1;
+            blk_over = blk1;
             filename_over = filename1;
         } else {
             total_sectors_over = total_sectors2;
-            bs_over = bs2;
+            blk_over = blk2;
             filename_over = filename2;
         }
 
@@ -1264,7 +1258,7 @@ static int img_compare(int argc, char **argv)
             if (nb_sectors <= 0) {
                 break;
             }
-            ret = bdrv_is_allocated_above(bs_over, NULL, sector_num,
+            ret = bdrv_is_allocated_above(blk_bs(blk_over), NULL, sector_num,
                                           nb_sectors, &pnum);
             if (ret < 0) {
                 ret = 3;
@@ -1275,7 +1269,7 @@ static int img_compare(int argc, char **argv)
             }
             nb_sectors = pnum;
             if (ret) {
-                ret = check_empty_sectors(bs_over, sector_num, nb_sectors,
+                ret = check_empty_sectors(blk_over, sector_num, nb_sectors,
                                           filename_over, buf1, quiet);
                 if (ret) {
                     if (ret < 0) {
@@ -1484,7 +1478,7 @@ static int img_convert(int argc, char **argv)
             goto out;
         }
         bs[bs_i] = blk_bs(blk[bs_i]);
-        bs_sectors[bs_i] = bdrv_nb_sectors(bs[bs_i]);
+        bs_sectors[bs_i] = blk_nb_sectors(blk[bs_i]);
         if (bs_sectors[bs_i] < 0) {
             error_report("Could not get size of %s: %s",
                          argv[optind + bs_i], strerror(-bs_sectors[bs_i]));
@@ -1524,41 +1518,44 @@ static int img_convert(int argc, char **argv)
         goto out;
     }
 
-    proto_drv = bdrv_find_protocol(out_filename, true);
+    proto_drv = bdrv_find_protocol(out_filename, true, &local_err);
     if (!proto_drv) {
-        error_report("Unknown protocol '%s'", out_filename);
+        qerror_report_err(local_err);
+        error_free(local_err);
         ret = -1;
         goto out;
     }
 
-    if (!drv->create_opts) {
+    if (!skip_create) {
-        error_report("Format driver '%s' does not support image creation",
+        if (!drv->create_opts) {
-                     drv->format_name);
+            error_report("Format driver '%s' does not support image creation",
-        ret = -1;
+                         drv->format_name);
-        goto out;
+            ret = -1;
-    }
+            goto out;
+        }
 
-    if (!proto_drv->create_opts) {
+        if (!proto_drv->create_opts) {
-        error_report("Protocol driver '%s' does not support image creation",
+            error_report("Protocol driver '%s' does not support image creation",
-                     proto_drv->format_name);
+                         proto_drv->format_name);
-        ret = -1;
+            ret = -1;
-        goto out;
+            goto out;
-    }
+        }
 
-    create_opts = qemu_opts_append(create_opts, drv->create_opts);
+        create_opts = qemu_opts_append(create_opts, drv->create_opts);
-    create_opts = qemu_opts_append(create_opts, proto_drv->create_opts);
+        create_opts = qemu_opts_append(create_opts, proto_drv->create_opts);
 
-    opts = qemu_opts_create(create_opts, NULL, 0, &error_abort);
+        opts = qemu_opts_create(create_opts, NULL, 0, &error_abort);
-    if (options && qemu_opts_do_parse(opts, options, NULL)) {
+        if (options && qemu_opts_do_parse(opts, options, NULL)) {
-        error_report("Invalid options for file format '%s'", out_fmt);
+            error_report("Invalid options for file format '%s'", out_fmt);
-        ret = -1;
+            ret = -1;
-        goto out;
+            goto out;
-    }
+        }
 
-    qemu_opt_set_number(opts, BLOCK_OPT_SIZE, total_sectors * 512);
+        qemu_opt_set_number(opts, BLOCK_OPT_SIZE, total_sectors * 512);
-    ret = add_old_style_options(out_fmt, opts, out_baseimg, NULL);
+        ret = add_old_style_options(out_fmt, opts, out_baseimg, NULL);
-    if (ret < 0) {
+        if (ret < 0) {
-        goto out;
+            goto out;
+        }
     }
 
     /* Get backing file name if -o backing_file was used */
@@ -1633,10 +1630,10 @@ static int img_convert(int argc, char **argv)
                                          out_bs->bl.discard_alignment))
                     );
 
-    buf = qemu_blockalign(out_bs, bufsectors * BDRV_SECTOR_SIZE);
+    buf = blk_blockalign(out_blk, bufsectors * BDRV_SECTOR_SIZE);
 
     if (skip_create) {
-        int64_t output_sectors = bdrv_nb_sectors(out_bs);
+        int64_t output_sectors = blk_nb_sectors(out_blk);
         if (output_sectors < 0) {
             error_report("unable to get output image length: %s\n",
                          strerror(-output_sectors));
@@ -1704,7 +1701,7 @@ static int img_convert(int argc, char **argv)
                 nlow = remainder > bs_sectors[bs_i] - bs_num
                     ? bs_sectors[bs_i] - bs_num : remainder;
 
-                ret = bdrv_read(bs[bs_i], bs_num, buf2, nlow);
+                ret = blk_read(blk[bs_i], bs_num, buf2, nlow);
                 if (ret < 0) {
                     error_report("error while reading sector %" PRId64 ": %s",
                                  bs_num, strerror(-ret));
@@ -1719,7 +1716,7 @@ static int img_convert(int argc, char **argv)
             assert (remainder == 0);
 
             if (!buffer_is_zero(buf, n * BDRV_SECTOR_SIZE)) {
-                ret = bdrv_write_compressed(out_bs, sector_num, buf, n);
+                ret = blk_write_compressed(out_blk, sector_num, buf, n);
                 if (ret != 0) {
                     error_report("error while compressing sector %" PRId64
                                  ": %s", sector_num, strerror(-ret));
@@ -1730,7 +1727,7 @@ static int img_convert(int argc, char **argv)
             qemu_progress_print(100.0 * sector_num / total_sectors, 0);
         }
         /* signal EOF to align */
-        bdrv_write_compressed(out_bs, 0, NULL, 0);
+        blk_write_compressed(out_blk, 0, NULL, 0);
     } else {
         int64_t sectors_to_read, sectors_read, sector_num_next_status;
         bool count_allocated_sectors;
@@ -1831,7 +1828,7 @@ restart:
             }
 
             n1 = n;
-            ret = bdrv_read(bs[bs_i], sector_num - bs_offset, buf, n);
+            ret = blk_read(blk[bs_i], sector_num - bs_offset, buf, n);
             if (ret < 0) {
                 error_report("error while reading sector %" PRId64 ": %s",
                              sector_num - bs_offset, strerror(-ret));
@@ -1844,7 +1841,7 @@ restart:
             while (n > 0) {
                 if (!has_zero_init ||
                     is_allocated_sectors_min(buf1, n, &n1, min_sparse)) {
-                    ret = bdrv_write(out_bs, sector_num, buf1, n1);
+                    ret = blk_write(out_blk, sector_num, buf1, n1);
                     if (ret < 0) {
                         error_report("error while writing sector %" PRId64
                                      ": %s", sector_num, strerror(-ret));
@@ -2268,7 +2265,7 @@ static int img_map(int argc, char **argv)
         printf("%-16s%-16s%-16s%s\n", "Offset", "Length", "Mapped to", "File");
     }
 
-    length = bdrv_getlength(bs);
+    length = blk_getlength(blk);
     while (curr.start + curr.length < length) {
         int64_t nsectors_left;
         int64_t sector_num;
@@ -2437,8 +2434,7 @@ static int img_snapshot(int argc, char **argv)
 static int img_rebase(int argc, char **argv)
 {
     BlockBackend *blk = NULL, *blk_old_backing = NULL, *blk_new_backing = NULL;
-    BlockDriverState *bs = NULL, *bs_old_backing = NULL, *bs_new_backing = NULL;
+    BlockDriverState *bs = NULL;
-    BlockDriver *old_backing_drv, *new_backing_drv;
     char *filename;
     const char *fmt, *cache, *src_cache, *out_basefmt, *out_baseimg;
     int c, flags, src_flags, ret;
@@ -2532,22 +2528,8 @@ static int img_rebase(int argc, char **argv)
     }
     bs = blk_bs(blk);
 
-    /* Find the right drivers for the backing files */
-    old_backing_drv = NULL;
-    new_backing_drv = NULL;
-
-    if (!unsafe && bs->backing_format[0] != '\0') {
-        old_backing_drv = bdrv_find_format(bs->backing_format);
-        if (old_backing_drv == NULL) {
-            error_report("Invalid format name: '%s'", bs->backing_format);
-            ret = -1;
-            goto out;
-        }
-    }
-
     if (out_basefmt != NULL) {
-        new_backing_drv = bdrv_find_format(out_basefmt);
+        if (bdrv_find_format(out_basefmt) == NULL) {
-        if (new_backing_drv == NULL) {
             error_report("Invalid format name: '%s'", out_basefmt);
             ret = -1;
             goto out;
@@ -2557,24 +2539,34 @@ static int img_rebase(int argc, char **argv)
     /* For safe rebasing we need to compare old and new backing file */
     if (!unsafe) {
         char backing_name[PATH_MAX];
+        QDict *options = NULL;
+
+        if (bs->backing_format[0] != '\0') {
+            options = qdict_new();
+            qdict_put(options, "driver", qstring_from_str(bs->backing_format));
+        }
 
-        blk_old_backing = blk_new_with_bs("old_backing", &error_abort);
-        bs_old_backing = blk_bs(blk_old_backing);
         bdrv_get_backing_filename(bs, backing_name, sizeof(backing_name));
-        ret = bdrv_open(&bs_old_backing, backing_name, NULL, NULL, src_flags,
+        blk_old_backing = blk_new_open("old_backing", backing_name, NULL,
-                        old_backing_drv, &local_err);
+                                       options, src_flags, &local_err);
-        if (ret) {
+        if (!blk_old_backing) {
             error_report("Could not open old backing file '%s': %s",
                          backing_name, error_get_pretty(local_err));
             error_free(local_err);
             goto out;
         }
+
         if (out_baseimg[0]) {
-            blk_new_backing = blk_new_with_bs("new_backing", &error_abort);
+            if (out_basefmt) {
-            bs_new_backing = blk_bs(blk_new_backing);
+                options = qdict_new();
-            ret = bdrv_open(&bs_new_backing, out_baseimg, NULL, NULL, src_flags,
+                qdict_put(options, "driver", qstring_from_str(out_basefmt));
-                            new_backing_drv, &local_err);
+            } else {
-            if (ret) {
+                options = NULL;
+            }
+
+            blk_new_backing = blk_new_open("new_backing", out_baseimg, NULL,
+                                           options, src_flags, &local_err);
+            if (!blk_new_backing) {
                 error_report("Could not open new backing file '%s': %s",
                              out_baseimg, error_get_pretty(local_err));
                 error_free(local_err);
@@ -2602,17 +2594,17 @@ static int img_rebase(int argc, char **argv)
         uint8_t * buf_new;
         float local_progress = 0;
 
-        buf_old = qemu_blockalign(bs, IO_BUF_SIZE);
+        buf_old = blk_blockalign(blk, IO_BUF_SIZE);
-        buf_new = qemu_blockalign(bs, IO_BUF_SIZE);
+        buf_new = blk_blockalign(blk, IO_BUF_SIZE);
 
-        num_sectors = bdrv_nb_sectors(bs);
+        num_sectors = blk_nb_sectors(blk);
         if (num_sectors < 0) {
             error_report("Could not get size of '%s': %s",
                          filename, strerror(-num_sectors));
             ret = -1;
             goto out;
         }
-        old_backing_num_sectors = bdrv_nb_sectors(bs_old_backing);
+        old_backing_num_sectors = blk_nb_sectors(blk_old_backing);
         if (old_backing_num_sectors < 0) {
             char backing_name[PATH_MAX];
 
@@ -2622,8 +2614,8 @@ static int img_rebase(int argc, char **argv)
             ret = -1;
             goto out;
         }
-        if (bs_new_backing) {
+        if (blk_new_backing) {
-            new_backing_num_sectors = bdrv_nb_sectors(bs_new_backing);
+            new_backing_num_sectors = blk_nb_sectors(blk_new_backing);
             if (new_backing_num_sectors < 0) {
                 error_report("Could not get size of '%s': %s",
                              out_baseimg, strerror(-new_backing_num_sectors));
@@ -2668,21 +2660,21 @@ static int img_rebase(int argc, char **argv)
                     n = old_backing_num_sectors - sector;
                 }
 
-                ret = bdrv_read(bs_old_backing, sector, buf_old, n);
+                ret = blk_read(blk_old_backing, sector, buf_old, n);
                 if (ret < 0) {
                     error_report("error while reading from old backing file");
                     goto out;
                 }
             }
 
-            if (sector >= new_backing_num_sectors || !bs_new_backing) {
+            if (sector >= new_backing_num_sectors || !blk_new_backing) {
                 memset(buf_new, 0, n * BDRV_SECTOR_SIZE);
             } else {
                 if (sector + n > new_backing_num_sectors) {
                     n = new_backing_num_sectors - sector;
                 }
 
-                ret = bdrv_read(bs_new_backing, sector, buf_new, n);
+                ret = blk_read(blk_new_backing, sector, buf_new, n);
                 if (ret < 0) {
                     error_report("error while reading from new backing file");
                     goto out;
@@ -2698,8 +2690,8 @@ static int img_rebase(int argc, char **argv)
                 if (compare_sectors(buf_old + written * 512,
                     buf_new + written * 512, n - written, &pnum))
                 {
-                    ret = bdrv_write(bs, sector + written,
+                    ret = blk_write(blk, sector + written,
-                        buf_old + written * 512, pnum);
+                                    buf_old + written * 512, pnum);
                     if (ret < 0) {
                         error_report("Error while writing to COW image: %s",
                             strerror(-ret));
@@ -2764,7 +2756,6 @@ static int img_resize(int argc, char **argv)
     int64_t n, total_size;
     bool quiet = false;
     BlockBackend *blk = NULL;
-    BlockDriverState *bs = NULL;
     QemuOpts *param;
     static QemuOptsList resize_options = {
         .name = "resize_options",
@@ -2846,10 +2837,9 @@ static int img_resize(int argc, char **argv)
         ret = -1;
         goto out;
     }
-    bs = blk_bs(blk);
 
     if (relative) {
-        total_size = bdrv_getlength(bs) + n * relative;
+        total_size = blk_getlength(blk) + n * relative;
     } else {
         total_size = n;
     }
@@ -2859,7 +2849,7 @@ static int img_resize(int argc, char **argv)
         goto out;
     }
 
-    ret = bdrv_truncate(bs, total_size);
+    ret = blk_truncate(blk, total_size);
     switch (ret) {
     case 0:
         qprintf(quiet, "Image resized.\n");

qemu-io-cmds.c

@@ -9,10 +9,13 @@
  */
 
 #include "qemu-io.h"
-#include "block/block_int.h"
+#include "sysemu/block-backend.h"
+#include "block/block.h"
+#include "block/block_int.h" /* for info_f() */
 #include "block/qapi.h"
 #include "qemu/main-loop.h"
 #include "qemu/timer.h"
+#include "sysemu/block-backend.h"
 
 #define CMD_NOFILE_OK   0x01
 
@@ -40,24 +43,24 @@ int qemuio_command_usage(const cmdinfo_t *ci)
     return 0;
 }
 
-static int init_check_command(BlockDriverState *bs, const cmdinfo_t *ct)
+static int init_check_command(BlockBackend *blk, const cmdinfo_t *ct)
 {
     if (ct->flags & CMD_FLAG_GLOBAL) {
         return 1;
     }
-    if (!(ct->flags & CMD_NOFILE_OK) && !bs) {
+    if (!(ct->flags & CMD_NOFILE_OK) && !blk) {
         fprintf(stderr, "no file open, try 'help open'\n");
         return 0;
     }
     return 1;
 }
 
-static int command(BlockDriverState *bs, const cmdinfo_t *ct, int argc,
+static int command(BlockBackend *blk, const cmdinfo_t *ct, int argc,
                    char **argv)
 {
     char *cmd = argv[0];
 
-    if (!init_check_command(bs, ct)) {
+    if (!init_check_command(blk, ct)) {
         return 0;
     }
 
@@ -78,7 +81,7 @@ static int command(BlockDriverState *bs, const cmdinfo_t *ct, int argc,
         return 0;
     }
     optind = 0;
-    return ct->cfunc(bs, argc, argv);
+    return ct->cfunc(blk, argc, argv);
 }
 
 static const cmdinfo_t *find_command(const char *cmd)
@@ -267,14 +270,14 @@ static int parse_pattern(const char *arg)
  */
 
 #define MISALIGN_OFFSET     16
-static void *qemu_io_alloc(BlockDriverState *bs, size_t len, int pattern)
+static void *qemu_io_alloc(BlockBackend *blk, size_t len, int pattern)
 {
     void *buf;
 
     if (qemuio_misalign) {
         len += MISALIGN_OFFSET;
     }
-    buf = qemu_blockalign(bs, len);
+    buf = blk_blockalign(blk, len);
     memset(buf, pattern, len);
     if (qemuio_misalign) {
         buf += MISALIGN_OFFSET;
@@ -340,7 +343,7 @@ static void print_report(const char *op, struct timeval *t, int64_t offset,
  * vector matching it.
  */
 static void *
-create_iovec(BlockDriverState *bs, QEMUIOVector *qiov, char **argv, int nr_iov,
+create_iovec(BlockBackend *blk, QEMUIOVector *qiov, char **argv, int nr_iov,
              int pattern)
 {
     size_t *sizes = g_new0(size_t, nr_iov);
@@ -377,7 +380,7 @@ create_iovec(BlockDriverState *bs, QEMUIOVector *qiov, char **argv, int nr_iov,
 
     qemu_iovec_init(qiov, nr_iov);
 
-    buf = p = qemu_io_alloc(bs, count, pattern);
+    buf = p = qemu_io_alloc(blk, count, pattern);
 
     for (i = 0; i < nr_iov; i++) {
         qemu_iovec_add(qiov, p, sizes[i]);
@@ -389,12 +392,12 @@ fail:
     return buf;
 }
 
-static int do_read(BlockDriverState *bs, char *buf, int64_t offset, int count,
+static int do_read(BlockBackend *blk, char *buf, int64_t offset, int count,
                    int *total)
 {
     int ret;
 
-    ret = bdrv_read(bs, offset >> 9, (uint8_t *)buf, count >> 9);
+    ret = blk_read(blk, offset >> 9, (uint8_t *)buf, count >> 9);
     if (ret < 0) {
         return ret;
     }
@@ -402,12 +405,12 @@ static int do_read(BlockDriverState *bs, char *buf, int64_t offset, int count,
     return 1;
 }
 
-static int do_write(BlockDriverState *bs, char *buf, int64_t offset, int count,
+static int do_write(BlockBackend *blk, char *buf, int64_t offset, int count,
                     int *total)
 {
     int ret;
 
-    ret = bdrv_write(bs, offset >> 9, (uint8_t *)buf, count >> 9);
+    ret = blk_write(blk, offset >> 9, (uint8_t *)buf, count >> 9);
     if (ret < 0) {
         return ret;
     }
@@ -415,20 +418,20 @@ static int do_write(BlockDriverState *bs, char *buf, int64_t offset, int count,
     return 1;
 }
 
-static int do_pread(BlockDriverState *bs, char *buf, int64_t offset, int count,
+static int do_pread(BlockBackend *blk, char *buf, int64_t offset, int count,
                     int *total)
 {
-    *total = bdrv_pread(bs, offset, (uint8_t *)buf, count);
+    *total = blk_pread(blk, offset, (uint8_t *)buf, count);
     if (*total < 0) {
         return *total;
     }
     return 1;
 }
 
-static int do_pwrite(BlockDriverState *bs, char *buf, int64_t offset, int count,
+static int do_pwrite(BlockBackend *blk, char *buf, int64_t offset, int count,
                      int *total)
 {
-    *total = bdrv_pwrite(bs, offset, (uint8_t *)buf, count);
+    *total = blk_pwrite(blk, offset, (uint8_t *)buf, count);
     if (*total < 0) {
         return *total;
     }
@@ -436,7 +439,7 @@ static int do_pwrite(BlockDriverState *bs, char *buf, int64_t offset, int count,
 }
 
 typedef struct {
-    BlockDriverState *bs;
+    BlockBackend *blk;
     int64_t offset;
     int count;
     int *total;
@@ -448,8 +451,8 @@ static void coroutine_fn co_write_zeroes_entry(void *opaque)
 {
     CoWriteZeroes *data = opaque;
 
-    data->ret = bdrv_co_write_zeroes(data->bs, data->offset / BDRV_SECTOR_SIZE,
+    data->ret = blk_co_write_zeroes(data->blk, data->offset / BDRV_SECTOR_SIZE,
-                                     data->count / BDRV_SECTOR_SIZE, 0);
+                                    data->count / BDRV_SECTOR_SIZE, 0);
     data->done = true;
     if (data->ret < 0) {
         *data->total = data->ret;
@@ -459,12 +462,12 @@ static void coroutine_fn co_write_zeroes_entry(void *opaque)
     *data->total = data->count;
 }
 
-static int do_co_write_zeroes(BlockDriverState *bs, int64_t offset, int count,
+static int do_co_write_zeroes(BlockBackend *blk, int64_t offset, int count,
                               int *total)
 {
     Coroutine *co;
     CoWriteZeroes data = {
-        .bs     = bs,
+        .blk    = blk,
         .offset = offset,
         .count  = count,
         .total  = total,
@@ -474,7 +477,7 @@ static int do_co_write_zeroes(BlockDriverState *bs, int64_t offset, int count,
     co = qemu_coroutine_create(co_write_zeroes_entry);
     qemu_coroutine_enter(co, &data);
     while (!data.done) {
-        aio_poll(bdrv_get_aio_context(bs), true);
+        aio_poll(blk_get_aio_context(blk), true);
     }
     if (data.ret < 0) {
         return data.ret;
@@ -483,12 +486,12 @@ static int do_co_write_zeroes(BlockDriverState *bs, int64_t offset, int count,
     }
 }
 
-static int do_write_compressed(BlockDriverState *bs, char *buf, int64_t offset,
+static int do_write_compressed(BlockBackend *blk, char *buf, int64_t offset,
                                int count, int *total)
 {
     int ret;
 
-    ret = bdrv_write_compressed(bs, offset >> 9, (uint8_t *)buf, count >> 9);
+    ret = blk_write_compressed(blk, offset >> 9, (uint8_t *)buf, count >> 9);
     if (ret < 0) {
         return ret;
     }
@@ -496,20 +499,20 @@ static int do_write_compressed(BlockDriverState *bs, char *buf, int64_t offset,
     return 1;
 }
 
-static int do_load_vmstate(BlockDriverState *bs, char *buf, int64_t offset,
+static int do_load_vmstate(BlockBackend *blk, char *buf, int64_t offset,
                            int count, int *total)
 {
-    *total = bdrv_load_vmstate(bs, (uint8_t *)buf, offset, count);
+    *total = blk_load_vmstate(blk, (uint8_t *)buf, offset, count);
     if (*total < 0) {
         return *total;
     }
     return 1;
 }
 
-static int do_save_vmstate(BlockDriverState *bs, char *buf, int64_t offset,
+static int do_save_vmstate(BlockBackend *blk, char *buf, int64_t offset,
                            int count, int *total)
 {
-    *total = bdrv_save_vmstate(bs, (uint8_t *)buf, offset, count);
+    *total = blk_save_vmstate(blk, (uint8_t *)buf, offset, count);
     if (*total < 0) {
         return *total;
     }
@@ -522,13 +525,13 @@ static void aio_rw_done(void *opaque, int ret)
     *(int *)opaque = ret;
 }
 
-static int do_aio_readv(BlockDriverState *bs, QEMUIOVector *qiov,
+static int do_aio_readv(BlockBackend *blk, QEMUIOVector *qiov,
                         int64_t offset, int *total)
 {
     int async_ret = NOT_DONE;
 
-    bdrv_aio_readv(bs, offset >> 9, qiov, qiov->size >> 9,
+    blk_aio_readv(blk, offset >> 9, qiov, qiov->size >> 9,
-                   aio_rw_done, &async_ret);
+                  aio_rw_done, &async_ret);
     while (async_ret == NOT_DONE) {
         main_loop_wait(false);
     }
@@ -537,13 +540,13 @@ static int do_aio_readv(BlockDriverState *bs, QEMUIOVector *qiov,
     return async_ret < 0 ? async_ret : 1;
 }
 
-static int do_aio_writev(BlockDriverState *bs, QEMUIOVector *qiov,
+static int do_aio_writev(BlockBackend *blk, QEMUIOVector *qiov,
                          int64_t offset, int *total)
 {
     int async_ret = NOT_DONE;
 
-    bdrv_aio_writev(bs, offset >> 9, qiov, qiov->size >> 9,
+    blk_aio_writev(blk, offset >> 9, qiov, qiov->size >> 9,
-                    aio_rw_done, &async_ret);
+                   aio_rw_done, &async_ret);
     while (async_ret == NOT_DONE) {
         main_loop_wait(false);
     }
@@ -567,7 +570,7 @@ static void multiwrite_cb(void *opaque, int ret)
     }
 }
 
-static int do_aio_multiwrite(BlockDriverState *bs, BlockRequest* reqs,
+static int do_aio_multiwrite(BlockBackend *blk, BlockRequest* reqs,
                              int num_reqs, int *total)
 {
     int i, ret;
@@ -583,7 +586,7 @@ static int do_aio_multiwrite(BlockDriverState *bs, BlockRequest* reqs,
         *total += reqs[i].qiov->size;
     }
 
-    ret = bdrv_aio_multiwrite(bs, reqs, num_reqs);
+    ret = blk_aio_multiwrite(blk, reqs, num_reqs);
     if (ret < 0) {
         return ret;
     }
@@ -609,7 +612,7 @@ static void read_help(void)
 " -b, -- read from the VM state rather than the virtual disk\n"
 " -C, -- report statistics in a machine parsable format\n"
 " -l, -- length for pattern verification (only with -P)\n"
-" -p, -- use bdrv_pread to read the file\n"
+" -p, -- use blk_pread to read the file\n"
 " -P, -- use a pattern to verify read data\n"
 " -q, -- quiet mode, do not show I/O statistics\n"
 " -s, -- start offset for pattern verification (only with -P)\n"
@@ -617,7 +620,7 @@ static void read_help(void)
 "\n");
 }
 
-static int read_f(BlockDriverState *bs, int argc, char **argv);
+static int read_f(BlockBackend *blk, int argc, char **argv);
 
 static const cmdinfo_t read_cmd = {
     .name       = "read",
@@ -630,7 +633,7 @@ static const cmdinfo_t read_cmd = {
     .help       = read_help,
 };
 
-static int read_f(BlockDriverState *bs, int argc, char **argv)
+static int read_f(BlockBackend *blk, int argc, char **argv)
 {
     struct timeval t1, t2;
     int Cflag = 0, pflag = 0, qflag = 0, vflag = 0;
@@ -736,15 +739,15 @@ static int read_f(BlockDriverState *bs, int argc, char **argv)
         }
     }
 
-    buf = qemu_io_alloc(bs, count, 0xab);
+    buf = qemu_io_alloc(blk, count, 0xab);
 
     gettimeofday(&t1, NULL);
     if (pflag) {
-        cnt = do_pread(bs, buf, offset, count, &total);
+        cnt = do_pread(blk, buf, offset, count, &total);
     } else if (bflag) {
-        cnt = do_load_vmstate(bs, buf, offset, count, &total);
+        cnt = do_load_vmstate(blk, buf, offset, count, &total);
     } else {
-        cnt = do_read(bs, buf, offset, count, &total);
+        cnt = do_read(blk, buf, offset, count, &total);
     }
     gettimeofday(&t2, NULL);
 
@@ -801,7 +804,7 @@ static void readv_help(void)
 "\n");
 }
 
-static int readv_f(BlockDriverState *bs, int argc, char **argv);
+static int readv_f(BlockBackend *blk, int argc, char **argv);
 
 static const cmdinfo_t readv_cmd = {
     .name       = "readv",
@@ -813,7 +816,7 @@ static const cmdinfo_t readv_cmd = {
     .help       = readv_help,
 };
 
-static int readv_f(BlockDriverState *bs, int argc, char **argv)
+static int readv_f(BlockBackend *blk, int argc, char **argv)
 {
     struct timeval t1, t2;
     int Cflag = 0, qflag = 0, vflag = 0;
@@ -869,13 +872,13 @@ static int readv_f(BlockDriverState *bs, int argc, char **argv)
     }
 
     nr_iov = argc - optind;
-    buf = create_iovec(bs, &qiov, &argv[optind], nr_iov, 0xab);
+    buf = create_iovec(blk, &qiov, &argv[optind], nr_iov, 0xab);
     if (buf == NULL) {
         return 0;
     }
 
     gettimeofday(&t1, NULL);
-    cnt = do_aio_readv(bs, &qiov, offset, &total);
+    cnt = do_aio_readv(blk, &qiov, offset, &total);
     gettimeofday(&t2, NULL);
 
     if (cnt < 0) {
@@ -923,16 +926,16 @@ static void write_help(void)
 " Writes into a segment of the currently open file, using a buffer\n"
 " filled with a set pattern (0xcdcdcdcd).\n"
 " -b, -- write to the VM state rather than the virtual disk\n"
-" -c, -- write compressed data with bdrv_write_compressed\n"
+" -c, -- write compressed data with blk_write_compressed\n"
-" -p, -- use bdrv_pwrite to write the file\n"
+" -p, -- use blk_pwrite to write the file\n"
 " -P, -- use different pattern to fill file\n"
 " -C, -- report statistics in a machine parsable format\n"
 " -q, -- quiet mode, do not show I/O statistics\n"
-" -z, -- write zeroes using bdrv_co_write_zeroes\n"
+" -z, -- write zeroes using blk_co_write_zeroes\n"
 "\n");
 }
 
-static int write_f(BlockDriverState *bs, int argc, char **argv);
+static int write_f(BlockBackend *blk, int argc, char **argv);
 
 static const cmdinfo_t write_cmd = {
     .name       = "write",
@@ -945,7 +948,7 @@ static const cmdinfo_t write_cmd = {
     .help       = write_help,
 };
 
-static int write_f(BlockDriverState *bs, int argc, char **argv)
+static int write_f(BlockBackend *blk, int argc, char **argv)
 {
     struct timeval t1, t2;
     int Cflag = 0, pflag = 0, qflag = 0, bflag = 0, Pflag = 0, zflag = 0;
@@ -1032,20 +1035,20 @@ static int write_f(BlockDriverState *bs, int argc, char **argv)
     }
 
     if (!zflag) {
-        buf = qemu_io_alloc(bs, count, pattern);
+        buf = qemu_io_alloc(blk, count, pattern);
     }
 
     gettimeofday(&t1, NULL);
     if (pflag) {
-        cnt = do_pwrite(bs, buf, offset, count, &total);
+        cnt = do_pwrite(blk, buf, offset, count, &total);
     } else if (bflag) {
-        cnt = do_save_vmstate(bs, buf, offset, count, &total);
+        cnt = do_save_vmstate(blk, buf, offset, count, &total);
     } else if (zflag) {
-        cnt = do_co_write_zeroes(bs, offset, count, &total);
+        cnt = do_co_write_zeroes(blk, offset, count, &total);
     } else if (cflag) {
-        cnt = do_write_compressed(bs, buf, offset, count, &total);
+        cnt = do_write_compressed(blk, buf, offset, count, &total);
     } else {
-        cnt = do_write(bs, buf, offset, count, &total);
+        cnt = do_write(blk, buf, offset, count, &total);
     }
     gettimeofday(&t2, NULL);
 
@@ -1088,7 +1091,7 @@ writev_help(void)
 "\n");
 }
 
-static int writev_f(BlockDriverState *bs, int argc, char **argv);
+static int writev_f(BlockBackend *blk, int argc, char **argv);
 
 static const cmdinfo_t writev_cmd = {
     .name       = "writev",
@@ -1100,7 +1103,7 @@ static const cmdinfo_t writev_cmd = {
     .help       = writev_help,
 };
 
-static int writev_f(BlockDriverState *bs, int argc, char **argv)
+static int writev_f(BlockBackend *blk, int argc, char **argv)
 {
     struct timeval t1, t2;
     int Cflag = 0, qflag = 0;
@@ -1150,13 +1153,13 @@ static int writev_f(BlockDriverState *bs, int argc, char **argv)
     }
 
     nr_iov = argc - optind;
-    buf = create_iovec(bs, &qiov, &argv[optind], nr_iov, pattern);
+    buf = create_iovec(blk, &qiov, &argv[optind], nr_iov, pattern);
     if (buf == NULL) {
         return 0;
     }
 
     gettimeofday(&t1, NULL);
-    cnt = do_aio_writev(bs, &qiov, offset, &total);
+    cnt = do_aio_writev(blk, &qiov, offset, &total);
     gettimeofday(&t2, NULL);
 
     if (cnt < 0) {
@@ -1197,7 +1200,7 @@ static void multiwrite_help(void)
 "\n");
 }
 
-static int multiwrite_f(BlockDriverState *bs, int argc, char **argv);
+static int multiwrite_f(BlockBackend *blk, int argc, char **argv);
 
 static const cmdinfo_t multiwrite_cmd = {
     .name       = "multiwrite",
@@ -1209,7 +1212,7 @@ static const cmdinfo_t multiwrite_cmd = {
     .help       = multiwrite_help,
 };
 
-static int multiwrite_f(BlockDriverState *bs, int argc, char **argv)
+static int multiwrite_f(BlockBackend *blk, int argc, char **argv)
 {
     struct timeval t1, t2;
     int Cflag = 0, qflag = 0;
@@ -1290,7 +1293,7 @@ static int multiwrite_f(BlockDriverState *bs, int argc, char **argv)
         nr_iov = j - optind;
 
         /* Build request */
-        buf[i] = create_iovec(bs, &qiovs[i], &argv[optind], nr_iov, pattern);
+        buf[i] = create_iovec(blk, &qiovs[i], &argv[optind], nr_iov, pattern);
         if (buf[i] == NULL) {
             goto out;
         }
@@ -1308,7 +1311,7 @@ static int multiwrite_f(BlockDriverState *bs, int argc, char **argv)
     nr_reqs = i;
 
     gettimeofday(&t1, NULL);
-    cnt = do_aio_multiwrite(bs, reqs, nr_reqs, &total);
+    cnt = do_aio_multiwrite(blk, reqs, nr_reqs, &total);
     gettimeofday(&t2, NULL);
 
     if (cnt < 0) {
@@ -1337,6 +1340,7 @@ out:
 }
 
 struct aio_ctx {
+    BlockBackend *blk;
     QEMUIOVector qiov;
     int64_t offset;
     char *buf;
@@ -1344,6 +1348,7 @@ struct aio_ctx {
     int vflag;
     int Cflag;
     int Pflag;
+    BlockAcctCookie acct;
     int pattern;
     struct timeval t1;
 };
@@ -1361,6 +1366,8 @@ static void aio_write_done(void *opaque, int ret)
         goto out;
     }
 
+    block_acct_done(blk_get_stats(ctx->blk), &ctx->acct);
+
     if (ctx->qflag) {
         goto out;
     }
@@ -1398,6 +1405,8 @@ static void aio_read_done(void *opaque, int ret)
         g_free(cmp_buf);
     }
 
+    block_acct_done(blk_get_stats(ctx->blk), &ctx->acct);
+
     if (ctx->qflag) {
         goto out;
     }
@@ -1436,7 +1445,7 @@ static void aio_read_help(void)
 "\n");
 }
 
-static int aio_read_f(BlockDriverState *bs, int argc, char **argv);
+static int aio_read_f(BlockBackend *blk, int argc, char **argv);
 
 static const cmdinfo_t aio_read_cmd = {
     .name       = "aio_read",
@@ -1448,11 +1457,12 @@ static const cmdinfo_t aio_read_cmd = {
     .help       = aio_read_help,
 };
 
-static int aio_read_f(BlockDriverState *bs, int argc, char **argv)
+static int aio_read_f(BlockBackend *blk, int argc, char **argv)
 {
     int nr_iov, c;
     struct aio_ctx *ctx = g_new0(struct aio_ctx, 1);
 
+    ctx->blk = blk;
     while ((c = getopt(argc, argv, "CP:qv")) != EOF) {
         switch (c) {
         case 'C':
@@ -1499,15 +1509,17 @@ static int aio_read_f(BlockDriverState *bs, int argc, char **argv)
     }
 
     nr_iov = argc - optind;
-    ctx->buf = create_iovec(bs, &ctx->qiov, &argv[optind], nr_iov, 0xab);
+    ctx->buf = create_iovec(blk, &ctx->qiov, &argv[optind], nr_iov, 0xab);
     if (ctx->buf == NULL) {
         g_free(ctx);
         return 0;
     }
 
     gettimeofday(&ctx->t1, NULL);
-    bdrv_aio_readv(bs, ctx->offset >> 9, &ctx->qiov,
+    block_acct_start(blk_get_stats(blk), &ctx->acct, ctx->qiov.size,
-                   ctx->qiov.size >> 9, aio_read_done, ctx);
+                     BLOCK_ACCT_READ);
+    blk_aio_readv(blk, ctx->offset >> 9, &ctx->qiov,
+                  ctx->qiov.size >> 9, aio_read_done, ctx);
     return 0;
 }
 
@@ -1531,7 +1543,7 @@ static void aio_write_help(void)
 "\n");
 }
 
-static int aio_write_f(BlockDriverState *bs, int argc, char **argv);
+static int aio_write_f(BlockBackend *blk, int argc, char **argv);
 
 static const cmdinfo_t aio_write_cmd = {
     .name       = "aio_write",
@@ -1543,12 +1555,13 @@ static const cmdinfo_t aio_write_cmd = {
     .help       = aio_write_help,
 };
 
-static int aio_write_f(BlockDriverState *bs, int argc, char **argv)
+static int aio_write_f(BlockBackend *blk, int argc, char **argv)
 {
     int nr_iov, c;
     int pattern = 0xcd;
     struct aio_ctx *ctx = g_new0(struct aio_ctx, 1);
 
+    ctx->blk = blk;
     while ((c = getopt(argc, argv, "CqP:")) != EOF) {
         switch (c) {
         case 'C':
@@ -1591,21 +1604,23 @@ static int aio_write_f(BlockDriverState *bs, int argc, char **argv)
     }
 
     nr_iov = argc - optind;
-    ctx->buf = create_iovec(bs, &ctx->qiov, &argv[optind], nr_iov, pattern);
+    ctx->buf = create_iovec(blk, &ctx->qiov, &argv[optind], nr_iov, pattern);
     if (ctx->buf == NULL) {
         g_free(ctx);
         return 0;
     }
 
     gettimeofday(&ctx->t1, NULL);
-    bdrv_aio_writev(bs, ctx->offset >> 9, &ctx->qiov,
+    block_acct_start(blk_get_stats(blk), &ctx->acct, ctx->qiov.size,
-                    ctx->qiov.size >> 9, aio_write_done, ctx);
+                     BLOCK_ACCT_WRITE);
+    blk_aio_writev(blk, ctx->offset >> 9, &ctx->qiov,
+                   ctx->qiov.size >> 9, aio_write_done, ctx);
     return 0;
 }
 
-static int aio_flush_f(BlockDriverState *bs, int argc, char **argv)
+static int aio_flush_f(BlockBackend *blk, int argc, char **argv)
 {
-    bdrv_drain_all();
+    blk_drain_all();
     return 0;
 }
 
@@ -1615,9 +1630,9 @@ static const cmdinfo_t aio_flush_cmd = {
     .oneline    = "completes all outstanding aio requests"
 };
 
-static int flush_f(BlockDriverState *bs, int argc, char **argv)
+static int flush_f(BlockBackend *blk, int argc, char **argv)
 {
-    bdrv_flush(bs);
+    blk_flush(blk);
     return 0;
 }
 
@@ -1628,7 +1643,7 @@ static const cmdinfo_t flush_cmd = {
     .oneline    = "flush all in-core file state to disk",
 };
 
-static int truncate_f(BlockDriverState *bs, int argc, char **argv)
+static int truncate_f(BlockBackend *blk, int argc, char **argv)
 {
     int64_t offset;
     int ret;
@@ -1639,7 +1654,7 @@ static int truncate_f(BlockDriverState *bs, int argc, char **argv)
         return 0;
     }
 
-    ret = bdrv_truncate(bs, offset);
+    ret = blk_truncate(blk, offset);
     if (ret < 0) {
         printf("truncate: %s\n", strerror(-ret));
         return 0;
@@ -1658,12 +1673,12 @@ static const cmdinfo_t truncate_cmd = {
     .oneline    = "truncates the current file at the given offset",
 };
 
-static int length_f(BlockDriverState *bs, int argc, char **argv)
+static int length_f(BlockBackend *blk, int argc, char **argv)
 {
     int64_t size;
     char s1[64];
 
-    size = bdrv_getlength(bs);
+    size = blk_getlength(blk);
     if (size < 0) {
         printf("getlength: %s\n", strerror(-size));
         return 0;
@@ -1683,8 +1698,9 @@ static const cmdinfo_t length_cmd = {
 };
 
 
-static int info_f(BlockDriverState *bs, int argc, char **argv)
+static int info_f(BlockBackend *blk, int argc, char **argv)
 {
+    BlockDriverState *bs = blk_bs(blk);
     BlockDriverInfo bdi;
     ImageInfoSpecific *spec_info;
     char s1[64], s2[64];
@@ -1742,7 +1758,7 @@ static void discard_help(void)
 "\n");
 }
 
-static int discard_f(BlockDriverState *bs, int argc, char **argv);
+static int discard_f(BlockBackend *blk, int argc, char **argv);
 
 static const cmdinfo_t discard_cmd = {
     .name       = "discard",
@@ -1755,7 +1771,7 @@ static const cmdinfo_t discard_cmd = {
     .help       = discard_help,
 };
 
-static int discard_f(BlockDriverState *bs, int argc, char **argv)
+static int discard_f(BlockBackend *blk, int argc, char **argv)
 {
     struct timeval t1, t2;
     int Cflag = 0, qflag = 0;
@@ -1794,8 +1810,8 @@ static int discard_f(BlockDriverState *bs, int argc, char **argv)
     }
 
     gettimeofday(&t1, NULL);
-    ret = bdrv_discard(bs, offset >> BDRV_SECTOR_BITS,
+    ret = blk_discard(blk, offset >> BDRV_SECTOR_BITS,
-                       count >> BDRV_SECTOR_BITS);
+                      count >> BDRV_SECTOR_BITS);
     gettimeofday(&t2, NULL);
 
     if (ret < 0) {
@@ -1813,8 +1829,9 @@ out:
     return 0;
 }
 
-static int alloc_f(BlockDriverState *bs, int argc, char **argv)
+static int alloc_f(BlockBackend *blk, int argc, char **argv)
 {
+    BlockDriverState *bs = blk_bs(blk);
     int64_t offset, sector_num;
     int nb_sectors, remaining;
     char s1[64];
@@ -1910,20 +1927,27 @@ static int map_is_allocated(BlockDriverState *bs, int64_t sector_num,
     return firstret;
 }
 
-static int map_f(BlockDriverState *bs, int argc, char **argv)
+static int map_f(BlockBackend *blk, int argc, char **argv)
 {
     int64_t offset;
-    int64_t nb_sectors;
+    int64_t nb_sectors, total_sectors;
     char s1[64];
     int64_t num;
     int ret;
     const char *retstr;
 
     offset = 0;
-    nb_sectors = bs->total_sectors;
+    total_sectors = blk_nb_sectors(blk);
+    if (total_sectors < 0) {
+        error_report("Failed to query image length: %s",
+                     strerror(-total_sectors));
+        return 0;
+    }
+
+    nb_sectors = total_sectors;
 
     do {
-        ret = map_is_allocated(bs, offset, nb_sectors, &num);
+        ret = map_is_allocated(blk_bs(blk), offset, nb_sectors, &num);
         if (ret < 0) {
             error_report("Failed to get allocation status: %s", strerror(-ret));
             return 0;
@@ -1940,7 +1964,7 @@ static int map_f(BlockDriverState *bs, int argc, char **argv)
 
         offset += num;
         nb_sectors -= num;
-    } while (offset < bs->total_sectors);
+    } while (offset < total_sectors);
 
     return 0;
 }
@@ -1954,11 +1978,11 @@ static const cmdinfo_t map_cmd = {
        .oneline        = "prints the allocated areas of a file",
 };
 
-static int break_f(BlockDriverState *bs, int argc, char **argv)
+static int break_f(BlockBackend *blk, int argc, char **argv)
 {
     int ret;
 
-    ret = bdrv_debug_breakpoint(bs, argv[1], argv[2]);
+    ret = bdrv_debug_breakpoint(blk_bs(blk), argv[1], argv[2]);
     if (ret < 0) {
         printf("Could not set breakpoint: %s\n", strerror(-ret));
     }
@@ -1966,11 +1990,11 @@ static int break_f(BlockDriverState *bs, int argc, char **argv)
     return 0;
 }
 
-static int remove_break_f(BlockDriverState *bs, int argc, char **argv)
+static int remove_break_f(BlockBackend *blk, int argc, char **argv)
 {
     int ret;
 
-    ret = bdrv_debug_remove_breakpoint(bs, argv[1]);
+    ret = bdrv_debug_remove_breakpoint(blk_bs(blk), argv[1]);
     if (ret < 0) {
         printf("Could not remove breakpoint %s: %s\n", argv[1], strerror(-ret));
     }
@@ -1997,11 +2021,11 @@ static const cmdinfo_t remove_break_cmd = {
        .oneline        = "remove a breakpoint by tag",
 };
 
-static int resume_f(BlockDriverState *bs, int argc, char **argv)
+static int resume_f(BlockBackend *blk, int argc, char **argv)
 {
     int ret;
 
-    ret = bdrv_debug_resume(bs, argv[1]);
+    ret = bdrv_debug_resume(blk_bs(blk), argv[1]);
     if (ret < 0) {
         printf("Could not resume request: %s\n", strerror(-ret));
     }
@@ -2018,10 +2042,10 @@ static const cmdinfo_t resume_cmd = {
        .oneline        = "resumes the request tagged as tag",
 };
 
-static int wait_break_f(BlockDriverState *bs, int argc, char **argv)
+static int wait_break_f(BlockBackend *blk, int argc, char **argv)
 {
-    while (!bdrv_debug_is_suspended(bs, argv[1])) {
+    while (!bdrv_debug_is_suspended(blk_bs(blk), argv[1])) {
-        aio_poll(bdrv_get_aio_context(bs), true);
+        aio_poll(blk_get_aio_context(blk), true);
     }
 
     return 0;
@@ -2036,7 +2060,7 @@ static const cmdinfo_t wait_break_cmd = {
        .oneline        = "waits for the suspension of a request",
 };
 
-static int abort_f(BlockDriverState *bs, int argc, char **argv)
+static int abort_f(BlockBackend *blk, int argc, char **argv)
 {
     abort();
 }
@@ -2062,7 +2086,7 @@ static void sigraise_help(void)
 "\n", SIGTERM);
 }
 
-static int sigraise_f(BlockDriverState *bs, int argc, char **argv);
+static int sigraise_f(BlockBackend *blk, int argc, char **argv);
 
 static const cmdinfo_t sigraise_cmd = {
     .name       = "sigraise",
@@ -2075,7 +2099,7 @@ static const cmdinfo_t sigraise_cmd = {
     .help       = sigraise_help,
 };
 
-static int sigraise_f(BlockDriverState *bs, int argc, char **argv)
+static int sigraise_f(BlockBackend *blk, int argc, char **argv)
 {
     int sig = cvtnum(argv[1]);
     if (sig < 0) {
@@ -2099,7 +2123,7 @@ static void sleep_cb(void *opaque)
     *expired = true;
 }
 
-static int sleep_f(BlockDriverState *bs, int argc, char **argv)
+static int sleep_f(BlockBackend *blk, int argc, char **argv)
 {
     char *endptr;
     long ms;
@@ -2168,7 +2192,7 @@ static void help_all(void)
     printf("\nUse 'help commandname' for extended help.\n");
 }
 
-static int help_f(BlockDriverState *bs, int argc, char **argv)
+static int help_f(BlockBackend *blk, int argc, char **argv)
 {
     const cmdinfo_t *ct;
 
@@ -2198,7 +2222,7 @@ static const cmdinfo_t help_cmd = {
     .oneline    = "help for one or all commands",
 };
 
-bool qemuio_command(BlockDriverState *bs, const char *cmd)
+bool qemuio_command(BlockBackend *blk, const char *cmd)
 {
     char *input;
     const cmdinfo_t *ct;
@@ -2211,7 +2235,7 @@ bool qemuio_command(BlockDriverState *bs, const char *cmd)
     if (c) {
         ct = find_command(v[0]);
         if (ct) {
-            done = command(bs, ct, c, v);
+            done = command(blk, ct, c, v);
         } else {
             fprintf(stderr, "command \"%s\" not found\n", v[0]);
         }

qemu-io.c

@@ -28,7 +28,6 @@
 static char *progname;
 
 static BlockBackend *qemuio_blk;
-static BlockDriverState *qemuio_bs;
 
 /* qemu-io commands passed using -c */
 static int ncmdline;
@@ -36,10 +35,9 @@ static char **cmdline;
 
 static ReadLineState *readline_state;
 
-static int close_f(BlockDriverState *bs, int argc, char **argv)
+static int close_f(BlockBackend *blk, int argc, char **argv)
 {
     blk_unref(qemuio_blk);
-    qemuio_bs = NULL;
     qemuio_blk = NULL;
     return 0;
 }
@@ -51,32 +49,22 @@ static const cmdinfo_t close_cmd = {
     .oneline    = "close the current open file",
 };
 
-static int openfile(char *name, BlockDriver *drv, int flags, int growable,
+static int openfile(char *name, int flags, QDict *opts)
-                    QDict *opts)
 {
     Error *local_err = NULL;
 
-    if (qemuio_bs) {
+    if (qemuio_blk) {
         fprintf(stderr, "file open already, try 'help close'\n");
         QDECREF(opts);
         return 1;
     }
 
-    qemuio_blk = blk_new_with_bs("hda", &error_abort);
+    qemuio_blk = blk_new_open("hda", name, NULL, opts, flags, &local_err);
-    qemuio_bs = blk_bs(qemuio_blk);
+    if (!qemuio_blk) {
-
-    if (growable) {
-        flags |= BDRV_O_PROTOCOL;
-    }
-
-    if (bdrv_open(&qemuio_bs, name, NULL, opts, flags, drv, &local_err) < 0) {
         fprintf(stderr, "%s: can't open%s%s: %s\n", progname,
                 name ? " device " : "", name ?: "",
                 error_get_pretty(local_err));
         error_free(local_err);
-        blk_unref(qemuio_blk);
-        qemuio_bs = NULL;
-        qemuio_blk = NULL;
         return 1;
     }
 
@@ -96,12 +84,11 @@ static void open_help(void)
 " -r, -- open file read-only\n"
 " -s, -- use snapshot file\n"
 " -n, -- disable host cache\n"
-" -g, -- allow file to grow (only applies to protocols)\n"
 " -o, -- options to be given to the block driver"
 "\n");
 }
 
-static int open_f(BlockDriverState *bs, int argc, char **argv);
+static int open_f(BlockBackend *blk, int argc, char **argv);
 
 static const cmdinfo_t open_cmd = {
     .name       = "open",
@@ -125,11 +112,10 @@ static QemuOptsList empty_opts = {
     },
 };
 
-static int open_f(BlockDriverState *bs, int argc, char **argv)
+static int open_f(BlockBackend *blk, int argc, char **argv)
 {
     int flags = 0;
     int readonly = 0;
-    int growable = 0;
     int c;
     QemuOpts *qopts;
     QDict *opts;
@@ -145,9 +131,6 @@ static int open_f(BlockDriverState *bs, int argc, char **argv)
         case 'r':
             readonly = 1;
             break;
-        case 'g':
-            growable = 1;
-            break;
         case 'o':
             if (!qemu_opts_parse(&empty_opts, optarg, 0)) {
                 printf("could not parse option list -- %s\n", optarg);
@@ -170,16 +153,16 @@ static int open_f(BlockDriverState *bs, int argc, char **argv)
     qemu_opts_reset(&empty_opts);
 
     if (optind == argc - 1) {
-        return openfile(argv[optind], NULL, flags, growable, opts);
+        return openfile(argv[optind], flags, opts);
     } else if (optind == argc) {
-        return openfile(NULL, NULL, flags, growable, opts);
+        return openfile(NULL, flags, opts);
     } else {
         QDECREF(opts);
         return qemuio_command_usage(&open_cmd);
     }
 }
 
-static int quit_f(BlockDriverState *bs, int argc, char **argv)
+static int quit_f(BlockBackend *blk, int argc, char **argv)
 {
     return 1;
 }
@@ -206,7 +189,6 @@ static void usage(const char *name)
 "  -r, --read-only      export read-only\n"
 "  -s, --snapshot       use snapshot file\n"
 "  -n, --nocache        disable host cache\n"
-"  -g, --growable       allow file to grow (only applies to protocols)\n"
 "  -m, --misalign       misalign allocations for O_DIRECT\n"
 "  -k, --native-aio     use kernel AIO implementation (on Linux only)\n"
 "  -t, --cache=MODE     use the given cache mode for the image\n"
@@ -317,7 +299,7 @@ static void command_loop(void)
     char *input;
 
     for (i = 0; !done && i < ncmdline; i++) {
-        done = qemuio_command(qemuio_bs, cmdline[i]);
+        done = qemuio_command(qemuio_blk, cmdline[i]);
     }
     if (cmdline) {
         g_free(cmdline);
@@ -342,7 +324,7 @@ static void command_loop(void)
         if (input == NULL) {
             break;
         }
-        done = qemuio_command(qemuio_bs, input);
+        done = qemuio_command(qemuio_blk, input);
         g_free(input);
 
         prompted = 0;
@@ -365,7 +347,6 @@ static void reenable_tty_echo(void)
 int main(int argc, char **argv)
 {
     int readonly = 0;
-    int growable = 0;
     const char *sopt = "hVc:d:f:rsnmgkt:T:";
     const struct option lopt[] = {
         { "help", 0, NULL, 'h' },
@@ -377,7 +358,6 @@ int main(int argc, char **argv)
         { "snapshot", 0, NULL, 's' },
         { "nocache", 0, NULL, 'n' },
         { "misalign", 0, NULL, 'm' },
-        { "growable", 0, NULL, 'g' },
         { "native-aio", 0, NULL, 'k' },
         { "discard", 1, NULL, 'd' },
         { "cache", 1, NULL, 't' },
@@ -387,8 +367,8 @@ int main(int argc, char **argv)
     int c;
     int opt_index = 0;
     int flags = BDRV_O_UNMAP;
-    BlockDriver *drv = NULL;
     Error *local_error = NULL;
+    QDict *opts = NULL;
 
 #ifdef CONFIG_POSIX
     signal(SIGPIPE, SIG_IGN);
@@ -414,11 +394,10 @@ int main(int argc, char **argv)
             }
             break;
         case 'f':
-            drv = bdrv_find_format(optarg);
+            if (!opts) {
-            if (!drv) {
+                opts = qdict_new();
-                error_report("Invalid format '%s'", optarg);
-                exit(EXIT_FAILURE);
             }
+            qdict_put(opts, "driver", qstring_from_str(optarg));
             break;
         case 'c':
             add_user_command(optarg);
@@ -429,9 +408,6 @@ int main(int argc, char **argv)
         case 'm':
             qemuio_misalign = true;
             break;
-        case 'g':
-            growable = 1;
-            break;
         case 'k':
             flags |= BDRV_O_NATIVE_AIO;
             break;
@@ -489,7 +465,7 @@ int main(int argc, char **argv)
     }
 
     if ((argc - optind) == 1) {
-        openfile(argv[optind], drv, flags, growable, NULL);
+        openfile(argv[optind], flags, opts);
     }
     command_loop();
 

qemu-nbd.c

@@ -391,7 +391,6 @@ int main(int argc, char **argv)
 {
     BlockBackend *blk;
     BlockDriverState *bs;
-    BlockDriver *drv;
     off_t dev_offset = 0;
     uint32_t nbdflags = 0;
     bool disconnect = false;
@@ -434,7 +433,7 @@ int main(int argc, char **argv)
     char *end;
     int flags = BDRV_O_RDWR;
     int partition = -1;
-    int ret;
+    int ret = 0;
     int fd;
     bool seen_cache = false;
     bool seen_discard = false;
@@ -445,6 +444,7 @@ int main(int argc, char **argv)
     const char *fmt = NULL;
     Error *local_err = NULL;
     BlockdevDetectZeroesOptions detect_zeroes = BLOCKDEV_DETECT_ZEROES_OPTIONS_OFF;
+    QDict *options = NULL;
 
     /* The client thread uses SIGTERM to interrupt the server.  A signal
      * handler ensures that "qemu-nbd -v -c" exits with a nice status code.
@@ -689,24 +689,17 @@ int main(int argc, char **argv)
     atexit(bdrv_close_all);
 
     if (fmt) {
-        drv = bdrv_find_format(fmt);
+        options = qdict_new();
-        if (!drv) {
+        qdict_put(options, "driver", qstring_from_str(fmt));
-            errx(EXIT_FAILURE, "Unknown file format '%s'", fmt);
-        }
-    } else {
-        drv = NULL;
     }
 
-    blk = blk_new_with_bs("hda", &error_abort);
-    bs = blk_bs(blk);
-
     srcpath = argv[optind];
-    ret = bdrv_open(&bs, srcpath, NULL, NULL, flags, drv, &local_err);
+    blk = blk_new_open("hda", srcpath, NULL, options, flags, &local_err);
-    if (ret < 0) {
+    if (!blk) {
-        errno = -ret;
+        errx(EXIT_FAILURE, "Failed to blk_new_open '%s': %s", argv[optind],
-        err(EXIT_FAILURE, "Failed to bdrv_open '%s': %s", argv[optind],
+             error_get_pretty(local_err));
-            error_get_pretty(local_err));
     }
+    bs = blk_bs(blk);
 
     if (sn_opts) {
         ret = bdrv_snapshot_load_tmp(bs,

savevm.c

@@ -821,7 +821,7 @@ void qemu_savevm_state_cancel(void)
     }
 }
 
-static int qemu_savevm_state(QEMUFile *f)
+static int qemu_savevm_state(QEMUFile *f, Error **errp)
 {
     int ret;
     MigrationParams params = {
@@ -829,7 +829,7 @@ static int qemu_savevm_state(QEMUFile *f)
         .shared = 0
     };
 
-    if (qemu_savevm_state_blocked(NULL)) {
+    if (qemu_savevm_state_blocked(errp)) {
         return -EINVAL;
     }
 
@@ -850,6 +850,7 @@ static int qemu_savevm_state(QEMUFile *f)
     }
     if (ret != 0) {
         qemu_savevm_state_cancel();
+        error_setg_errno(errp, -ret, "Error while writing VM state");
     }
     return ret;
 }
@@ -1102,6 +1103,7 @@ void do_savevm(Monitor *mon, const QDict *qdict)
     qemu_timeval tv;
     struct tm tm;
     const char *name = qdict_get_try_str(qdict, "name");
+    Error *local_err = NULL;
 
     /* Verify if there is a device that doesn't support snapshots and is writable */
     bs = NULL;
@@ -1160,11 +1162,12 @@ void do_savevm(Monitor *mon, const QDict *qdict)
         monitor_printf(mon, "Could not open VM state file\n");
         goto the_end;
     }
-    ret = qemu_savevm_state(f);
+    ret = qemu_savevm_state(f, &local_err);
     vm_state_size = qemu_ftell(f);
     qemu_fclose(f);
     if (ret < 0) {
-        monitor_printf(mon, "Error %d while writing VM\n", ret);
+        monitor_printf(mon, "%s\n", error_get_pretty(local_err));
+        error_free(local_err);
         goto the_end;
     }
 

scripts/qtest.py

@@ -0,0 +1,71 @@
+# QEMU qtest library
+#
+# Copyright (C) 2015 Red Hat Inc.
+#
+# Authors:
+#  Fam Zheng <famz@redhat.com>
+#
+# This work is licensed under the terms of the GNU GPL, version 2.  See
+# the COPYING file in the top-level directory.
+#
+# Based on qmp.py.
+#
+
+import errno
+import socket
+
+class QEMUQtestProtocol(object):
+    def __init__(self, address, server=False):
+        """
+        Create a QEMUQtestProtocol object.
+
+        @param address: QEMU address, can be either a unix socket path (string)
+                        or a tuple in the form ( address, port ) for a TCP
+                        connection
+        @param server: server mode, listens on the socket (bool)
+        @raise socket.error on socket connection errors
+        @note No connection is established, this is done by the connect() or
+              accept() methods
+        """
+        self._address = address
+        self._sock = self._get_sock()
+        if server:
+            self._sock.bind(self._address)
+            self._sock.listen(1)
+
+    def _get_sock(self):
+        if isinstance(self._address, tuple):
+            family = socket.AF_INET
+        else:
+            family = socket.AF_UNIX
+        return socket.socket(family, socket.SOCK_STREAM)
+
+    def connect(self):
+        """
+        Connect to the qtest socket.
+
+        @raise socket.error on socket connection errors
+        """
+        self._sock.connect(self._address)
+
+    def accept(self):
+        """
+        Await connection from QEMU.
+
+        @raise socket.error on socket connection errors
+        """
+        self._sock, _ = self._sock.accept()
+
+    def cmd(self, qtest_cmd):
+        """
+        Send a qtest command on the wire.
+
+        @param qtest_cmd: qtest command text to be sent
+        """
+        self._sock.sendall(qtest_cmd + "\n")
+
+    def close(self):
+        self._sock.close()
+
+    def settimeout(self, timeout):
+        self._sock.settimeout(timeout)

tests/Makefile

@@ -307,9 +307,10 @@ tests/test-mul64$(EXESUF): tests/test-mul64.o libqemuutil.a
 tests/test-bitops$(EXESUF): tests/test-bitops.o libqemuutil.a
 
 libqos-obj-y = tests/libqos/pci.o tests/libqos/fw_cfg.o tests/libqos/malloc.o
-libqos-obj-y += tests/libqos/i2c.o
+libqos-obj-y += tests/libqos/i2c.o tests/libqos/libqos.o
 libqos-pc-obj-y = $(libqos-obj-y) tests/libqos/pci-pc.o
-libqos-pc-obj-y += tests/libqos/malloc-pc.o
+libqos-pc-obj-y += tests/libqos/malloc-pc.o tests/libqos/libqos-pc.o
+libqos-pc-obj-y += tests/libqos/ahci.o
 libqos-omap-obj-y = $(libqos-obj-y) tests/libqos/i2c-omap.o
 libqos-virtio-obj-y = $(libqos-obj-y) $(libqos-pc-obj-y) tests/libqos/virtio.o tests/libqos/virtio-pci.o
 libqos-usb-obj-y = $(libqos-pc-obj-y) tests/libqos/usb.o

tests/libqos/ahci.c

@@ -0,0 +1,838 @@
+/*
+ * libqos AHCI functions
+ *
+ * Copyright (c) 2014 John Snow <jsnow@redhat.com>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+#include <glib.h>
+
+#include "libqtest.h"
+#include "libqos/ahci.h"
+#include "libqos/pci-pc.h"
+
+#include "qemu-common.h"
+#include "qemu/host-utils.h"
+
+#include "hw/pci/pci_ids.h"
+#include "hw/pci/pci_regs.h"
+
+typedef struct AHCICommandProp {
+    uint8_t  cmd;        /* Command Code */
+    bool     data;       /* Data transfer command? */
+    bool     pio;
+    bool     dma;
+    bool     lba28;
+    bool     lba48;
+    bool     read;
+    bool     write;
+    bool     atapi;
+    bool     ncq;
+    uint64_t size;       /* Static transfer size, for commands like IDENTIFY. */
+    uint32_t interrupts; /* Expected interrupts for this command. */
+} AHCICommandProp;
+
+AHCICommandProp ahci_command_properties[] = {
+    { .cmd = CMD_READ_PIO,      .data = true,  .pio = true,
+                                .lba28 = true, .read = true },
+    { .cmd = CMD_WRITE_PIO,     .data = true,  .pio = true,
+                                .lba28 = true, .write = true },
+    { .cmd = CMD_READ_PIO_EXT,  .data = true,  .pio = true,
+                                .lba48 = true, .read = true },
+    { .cmd = CMD_WRITE_PIO_EXT, .data = true,  .pio = true,
+                                .lba48 = true, .write = true },
+    { .cmd = CMD_READ_DMA,      .data = true,  .dma = true,
+                                .lba28 = true, .read = true },
+    { .cmd = CMD_WRITE_DMA,     .data = true,  .dma = true,
+                                .lba28 = true, .write = true },
+    { .cmd = CMD_READ_DMA_EXT,  .data = true,  .dma = true,
+                                .lba48 = true, .read = true },
+    { .cmd = CMD_WRITE_DMA_EXT, .data = true,  .dma = true,
+                                .lba48 = true, .write = true },
+    { .cmd = CMD_IDENTIFY,      .data = true,  .pio = true,
+                                .size = 512,   .read = true },
+    { .cmd = CMD_READ_MAX,      .lba28 = true },
+    { .cmd = CMD_READ_MAX_EXT,  .lba48 = true },
+    { .cmd = CMD_FLUSH_CACHE,   .data = false }
+};
+
+/**
+ * Allocate space in the guest using information in the AHCIQState object.
+ */
+uint64_t ahci_alloc(AHCIQState *ahci, size_t bytes)
+{
+    g_assert(ahci);
+    g_assert(ahci->parent);
+    return qmalloc(ahci->parent, bytes);
+}
+
+void ahci_free(AHCIQState *ahci, uint64_t addr)
+{
+    g_assert(ahci);
+    g_assert(ahci->parent);
+    qfree(ahci->parent, addr);
+}
+
+/**
+ * Locate, verify, and return a handle to the AHCI device.
+ */
+QPCIDevice *get_ahci_device(uint32_t *fingerprint)
+{
+    QPCIDevice *ahci;
+    uint32_t ahci_fingerprint;
+    QPCIBus *pcibus;
+
+    pcibus = qpci_init_pc();
+
+    /* Find the AHCI PCI device and verify it's the right one. */
+    ahci = qpci_device_find(pcibus, QPCI_DEVFN(0x1F, 0x02));
+    g_assert(ahci != NULL);
+
+    ahci_fingerprint = qpci_config_readl(ahci, PCI_VENDOR_ID);
+
+    switch (ahci_fingerprint) {
+    case AHCI_INTEL_ICH9:
+        break;
+    default:
+        /* Unknown device. */
+        g_assert_not_reached();
+    }
+
+    if (fingerprint) {
+        *fingerprint = ahci_fingerprint;
+    }
+    return ahci;
+}
+
+void free_ahci_device(QPCIDevice *dev)
+{
+    QPCIBus *pcibus = dev ? dev->bus : NULL;
+
+    /* libqos doesn't have a function for this, so free it manually */
+    g_free(dev);
+    qpci_free_pc(pcibus);
+}
+
+/* Free all memory in-use by the AHCI device. */
+void ahci_clean_mem(AHCIQState *ahci)
+{
+    uint8_t port, slot;
+
+    for (port = 0; port < 32; ++port) {
+        if (ahci->port[port].fb) {
+            ahci_free(ahci, ahci->port[port].fb);
+        }
+        if (ahci->port[port].clb) {
+            for (slot = 0; slot < 32; slot++) {
+                ahci_destroy_command(ahci, port, slot);
+            }
+            ahci_free(ahci, ahci->port[port].clb);
+        }
+    }
+}
+
+/*** Logical Device Initialization ***/
+
+/**
+ * Start the PCI device and sanity-check default operation.
+ */
+void ahci_pci_enable(AHCIQState *ahci)
+{
+    uint8_t reg;
+
+    start_ahci_device(ahci);
+
+    switch (ahci->fingerprint) {
+    case AHCI_INTEL_ICH9:
+        /* ICH9 has a register at PCI 0x92 that
+         * acts as a master port enabler mask. */
+        reg = qpci_config_readb(ahci->dev, 0x92);
+        reg |= 0x3F;
+        qpci_config_writeb(ahci->dev, 0x92, reg);
+        /* 0...0111111b -- bit significant, ports 0-5 enabled. */
+        ASSERT_BIT_SET(qpci_config_readb(ahci->dev, 0x92), 0x3F);
+        break;
+    }
+
+}
+
+/**
+ * Map BAR5/ABAR, and engage the PCI device.
+ */
+void start_ahci_device(AHCIQState *ahci)
+{
+    /* Map AHCI's ABAR (BAR5) */
+    ahci->hba_base = qpci_iomap(ahci->dev, 5, &ahci->barsize);
+    g_assert(ahci->hba_base);
+
+    /* turns on pci.cmd.iose, pci.cmd.mse and pci.cmd.bme */
+    qpci_device_enable(ahci->dev);
+}
+
+/**
+ * Test and initialize the AHCI's HBA memory areas.
+ * Initialize and start any ports with devices attached.
+ * Bring the HBA into the idle state.
+ */
+void ahci_hba_enable(AHCIQState *ahci)
+{
+    /* Bits of interest in this section:
+     * GHC.AE     Global Host Control / AHCI Enable
+     * PxCMD.ST   Port Command: Start
+     * PxCMD.SUD  "Spin Up Device"
+     * PxCMD.POD  "Power On Device"
+     * PxCMD.FRE  "FIS Receive Enable"
+     * PxCMD.FR   "FIS Receive Running"
+     * PxCMD.CR   "Command List Running"
+     */
+    uint32_t reg, ports_impl;
+    uint16_t i;
+    uint8_t num_cmd_slots;
+
+    g_assert(ahci != NULL);
+
+    /* Set GHC.AE to 1 */
+    ahci_set(ahci, AHCI_GHC, AHCI_GHC_AE);
+    reg = ahci_rreg(ahci, AHCI_GHC);
+    ASSERT_BIT_SET(reg, AHCI_GHC_AE);
+
+    /* Cache CAP and CAP2. */
+    ahci->cap = ahci_rreg(ahci, AHCI_CAP);
+    ahci->cap2 = ahci_rreg(ahci, AHCI_CAP2);
+
+    /* Read CAP.NCS, how many command slots do we have? */
+    num_cmd_slots = ((ahci->cap & AHCI_CAP_NCS) >> ctzl(AHCI_CAP_NCS)) + 1;
+    g_test_message("Number of Command Slots: %u", num_cmd_slots);
+
+    /* Determine which ports are implemented. */
+    ports_impl = ahci_rreg(ahci, AHCI_PI);
+
+    for (i = 0; ports_impl; ports_impl >>= 1, ++i) {
+        if (!(ports_impl & 0x01)) {
+            continue;
+        }
+
+        g_test_message("Initializing port %u", i);
+
+        reg = ahci_px_rreg(ahci, i, AHCI_PX_CMD);
+        if (BITCLR(reg, AHCI_PX_CMD_ST | AHCI_PX_CMD_CR |
+                   AHCI_PX_CMD_FRE | AHCI_PX_CMD_FR)) {
+            g_test_message("port is idle");
+        } else {
+            g_test_message("port needs to be idled");
+            ahci_px_clr(ahci, i, AHCI_PX_CMD,
+                        (AHCI_PX_CMD_ST | AHCI_PX_CMD_FRE));
+            /* The port has 500ms to disengage. */
+            usleep(500000);
+            reg = ahci_px_rreg(ahci, i, AHCI_PX_CMD);
+            ASSERT_BIT_CLEAR(reg, AHCI_PX_CMD_CR);
+            ASSERT_BIT_CLEAR(reg, AHCI_PX_CMD_FR);
+            g_test_message("port is now idle");
+            /* The spec does allow for possibly needing a PORT RESET
+             * or HBA reset if we fail to idle the port. */
+        }
+
+        /* Allocate Memory for the Command List Buffer & FIS Buffer */
+        /* PxCLB space ... 0x20 per command, as in 4.2.2 p 36 */
+        ahci->port[i].clb = ahci_alloc(ahci, num_cmd_slots * 0x20);
+        qmemset(ahci->port[i].clb, 0x00, 0x100);
+        g_test_message("CLB: 0x%08" PRIx64, ahci->port[i].clb);
+        ahci_px_wreg(ahci, i, AHCI_PX_CLB, ahci->port[i].clb);
+        g_assert_cmphex(ahci->port[i].clb, ==,
+                        ahci_px_rreg(ahci, i, AHCI_PX_CLB));
+
+        /* PxFB space ... 0x100, as in 4.2.1 p 35 */
+        ahci->port[i].fb = ahci_alloc(ahci, 0x100);
+        qmemset(ahci->port[i].fb, 0x00, 0x100);
+        g_test_message("FB: 0x%08" PRIx64, ahci->port[i].fb);
+        ahci_px_wreg(ahci, i, AHCI_PX_FB, ahci->port[i].fb);
+        g_assert_cmphex(ahci->port[i].fb, ==,
+                        ahci_px_rreg(ahci, i, AHCI_PX_FB));
+
+        /* Clear PxSERR, PxIS, then IS.IPS[x] by writing '1's. */
+        ahci_px_wreg(ahci, i, AHCI_PX_SERR, 0xFFFFFFFF);
+        ahci_px_wreg(ahci, i, AHCI_PX_IS, 0xFFFFFFFF);
+        ahci_wreg(ahci, AHCI_IS, (1 << i));
+
+        /* Verify Interrupts Cleared */
+        reg = ahci_px_rreg(ahci, i, AHCI_PX_SERR);
+        g_assert_cmphex(reg, ==, 0);
+
+        reg = ahci_px_rreg(ahci, i, AHCI_PX_IS);
+        g_assert_cmphex(reg, ==, 0);
+
+        reg = ahci_rreg(ahci, AHCI_IS);
+        ASSERT_BIT_CLEAR(reg, (1 << i));
+
+        /* Enable All Interrupts: */
+        ahci_px_wreg(ahci, i, AHCI_PX_IE, 0xFFFFFFFF);
+        reg = ahci_px_rreg(ahci, i, AHCI_PX_IE);
+        g_assert_cmphex(reg, ==, ~((uint32_t)AHCI_PX_IE_RESERVED));
+
+        /* Enable the FIS Receive Engine. */
+        ahci_px_set(ahci, i, AHCI_PX_CMD, AHCI_PX_CMD_FRE);
+        reg = ahci_px_rreg(ahci, i, AHCI_PX_CMD);
+        ASSERT_BIT_SET(reg, AHCI_PX_CMD_FR);
+
+        /* AHCI 1.3 spec: if !STS.BSY, !STS.DRQ and PxSSTS.DET indicates
+         * physical presence, a device is present and may be started. However,
+         * PxSERR.DIAG.X /may/ need to be cleared a priori. */
+        reg = ahci_px_rreg(ahci, i, AHCI_PX_SERR);
+        if (BITSET(reg, AHCI_PX_SERR_DIAG_X)) {
+            ahci_px_set(ahci, i, AHCI_PX_SERR, AHCI_PX_SERR_DIAG_X);
+        }
+
+        reg = ahci_px_rreg(ahci, i, AHCI_PX_TFD);
+        if (BITCLR(reg, AHCI_PX_TFD_STS_BSY | AHCI_PX_TFD_STS_DRQ)) {
+            reg = ahci_px_rreg(ahci, i, AHCI_PX_SSTS);
+            if ((reg & AHCI_PX_SSTS_DET) == SSTS_DET_ESTABLISHED) {
+                /* Device Found: set PxCMD.ST := 1 */
+                ahci_px_set(ahci, i, AHCI_PX_CMD, AHCI_PX_CMD_ST);
+                ASSERT_BIT_SET(ahci_px_rreg(ahci, i, AHCI_PX_CMD),
+                               AHCI_PX_CMD_CR);
+                g_test_message("Started Device %u", i);
+            } else if ((reg & AHCI_PX_SSTS_DET)) {
+                /* Device present, but in some unknown state. */
+                g_assert_not_reached();
+            }
+        }
+    }
+
+    /* Enable GHC.IE */
+    ahci_set(ahci, AHCI_GHC, AHCI_GHC_IE);
+    reg = ahci_rreg(ahci, AHCI_GHC);
+    ASSERT_BIT_SET(reg, AHCI_GHC_IE);
+
+    /* TODO: The device should now be idling and waiting for commands.
+     * In the future, a small test-case to inspect the Register D2H FIS
+     * and clear the initial interrupts might be good. */
+}
+
+/**
+ * Pick the first implemented and running port
+ */
+unsigned ahci_port_select(AHCIQState *ahci)
+{
+    uint32_t ports, reg;
+    unsigned i;
+
+    ports = ahci_rreg(ahci, AHCI_PI);
+    for (i = 0; i < 32; ports >>= 1, ++i) {
+        if (ports == 0) {
+            i = 32;
+        }
+
+        if (!(ports & 0x01)) {
+            continue;
+        }
+
+        reg = ahci_px_rreg(ahci, i, AHCI_PX_CMD);
+        if (BITSET(reg, AHCI_PX_CMD_ST)) {
+            break;
+        }
+    }
+    g_assert(i < 32);
+    return i;
+}
+
+/**
+ * Clear a port's interrupts and status information prior to a test.
+ */
+void ahci_port_clear(AHCIQState *ahci, uint8_t port)
+{
+    uint32_t reg;
+
+    /* Clear out this port's interrupts (ignore the init register d2h fis) */
+    reg = ahci_px_rreg(ahci, port, AHCI_PX_IS);
+    ahci_px_wreg(ahci, port, AHCI_PX_IS, reg);
+    g_assert_cmphex(ahci_px_rreg(ahci, port, AHCI_PX_IS), ==, 0);
+
+    /* Wipe the FIS-Recieve Buffer */
+    qmemset(ahci->port[port].fb, 0x00, 0x100);
+}
+
+/**
+ * Check a port for errors.
+ */
+void ahci_port_check_error(AHCIQState *ahci, uint8_t port)
+{
+    uint32_t reg;
+
+    /* The upper 9 bits of the IS register all indicate errors. */
+    reg = ahci_px_rreg(ahci, port, AHCI_PX_IS);
+    reg >>= 23;
+    g_assert_cmphex(reg, ==, 0);
+
+    /* The Sata Error Register should be empty. */
+    reg = ahci_px_rreg(ahci, port, AHCI_PX_SERR);
+    g_assert_cmphex(reg, ==, 0);
+
+    /* The TFD also has two error sections. */
+    reg = ahci_px_rreg(ahci, port, AHCI_PX_TFD);
+    ASSERT_BIT_CLEAR(reg, AHCI_PX_TFD_STS_ERR);
+    ASSERT_BIT_CLEAR(reg, AHCI_PX_TFD_ERR);
+}
+
+void ahci_port_check_interrupts(AHCIQState *ahci, uint8_t port,
+                                uint32_t intr_mask)
+{
+    uint32_t reg;
+
+    /* Check for expected interrupts */
+    reg = ahci_px_rreg(ahci, port, AHCI_PX_IS);
+    ASSERT_BIT_SET(reg, intr_mask);
+
+    /* Clear expected interrupts and assert all interrupts now cleared. */
+    ahci_px_wreg(ahci, port, AHCI_PX_IS, intr_mask);
+    g_assert_cmphex(ahci_px_rreg(ahci, port, AHCI_PX_IS), ==, 0);
+}
+
+void ahci_port_check_nonbusy(AHCIQState *ahci, uint8_t port, uint8_t slot)
+{
+    uint32_t reg;
+
+    /* Assert that the command slot is no longer busy (NCQ) */
+    reg = ahci_px_rreg(ahci, port, AHCI_PX_SACT);
+    ASSERT_BIT_CLEAR(reg, (1 << slot));
+
+    /* Non-NCQ */
+    reg = ahci_px_rreg(ahci, port, AHCI_PX_CI);
+    ASSERT_BIT_CLEAR(reg, (1 << slot));
+
+    /* And assert that we are generally not busy. */
+    reg = ahci_px_rreg(ahci, port, AHCI_PX_TFD);
+    ASSERT_BIT_CLEAR(reg, AHCI_PX_TFD_STS_BSY);
+    ASSERT_BIT_CLEAR(reg, AHCI_PX_TFD_STS_DRQ);
+}
+
+void ahci_port_check_d2h_sanity(AHCIQState *ahci, uint8_t port, uint8_t slot)
+{
+    RegD2HFIS *d2h = g_malloc0(0x20);
+    uint32_t reg;
+
+    memread(ahci->port[port].fb + 0x40, d2h, 0x20);
+    g_assert_cmphex(d2h->fis_type, ==, 0x34);
+
+    reg = ahci_px_rreg(ahci, port, AHCI_PX_TFD);
+    g_assert_cmphex((reg & AHCI_PX_TFD_ERR) >> 8, ==, d2h->error);
+    g_assert_cmphex((reg & AHCI_PX_TFD_STS), ==, d2h->status);
+
+    g_free(d2h);
+}
+
+void ahci_port_check_pio_sanity(AHCIQState *ahci, uint8_t port,
+                                uint8_t slot, size_t buffsize)
+{
+    PIOSetupFIS *pio = g_malloc0(0x20);
+
+    /* We cannot check the Status or E_Status registers, becuase
+     * the status may have again changed between the PIO Setup FIS
+     * and the conclusion of the command with the D2H Register FIS. */
+    memread(ahci->port[port].fb + 0x20, pio, 0x20);
+    g_assert_cmphex(pio->fis_type, ==, 0x5f);
+
+    /* BUG: PIO Setup FIS as utilized by QEMU tries to fit the entire
+     * transfer size in a uint16_t field. The maximum transfer size can
+     * eclipse this; the field is meant to convey the size of data per
+     * each Data FIS, not the entire operation as a whole. For now,
+     * we will sanity check the broken case where applicable. */
+    if (buffsize <= UINT16_MAX) {
+        g_assert_cmphex(le16_to_cpu(pio->tx_count), ==, buffsize);
+    }
+
+    g_free(pio);
+}
+
+void ahci_port_check_cmd_sanity(AHCIQState *ahci, uint8_t port,
+                                uint8_t slot, size_t buffsize)
+{
+    AHCICommandHeader cmd;
+
+    ahci_get_command_header(ahci, port, slot, &cmd);
+    g_assert_cmphex(buffsize, ==, cmd.prdbc);
+}
+
+/* Get the command in #slot of port #port. */
+void ahci_get_command_header(AHCIQState *ahci, uint8_t port,
+                             uint8_t slot, AHCICommandHeader *cmd)
+{
+    uint64_t ba = ahci->port[port].clb;
+    ba += slot * sizeof(AHCICommandHeader);
+    memread(ba, cmd, sizeof(AHCICommandHeader));
+
+    cmd->flags = le16_to_cpu(cmd->flags);
+    cmd->prdtl = le16_to_cpu(cmd->prdtl);
+    cmd->prdbc = le32_to_cpu(cmd->prdbc);
+    cmd->ctba = le64_to_cpu(cmd->ctba);
+}
+
+/* Set the command in #slot of port #port. */
+void ahci_set_command_header(AHCIQState *ahci, uint8_t port,
+                             uint8_t slot, AHCICommandHeader *cmd)
+{
+    AHCICommandHeader tmp;
+    uint64_t ba = ahci->port[port].clb;
+    ba += slot * sizeof(AHCICommandHeader);
+
+    tmp.flags = cpu_to_le16(cmd->flags);
+    tmp.prdtl = cpu_to_le16(cmd->prdtl);
+    tmp.prdbc = cpu_to_le32(cmd->prdbc);
+    tmp.ctba = cpu_to_le64(cmd->ctba);
+
+    memwrite(ba, &tmp, sizeof(AHCICommandHeader));
+}
+
+void ahci_destroy_command(AHCIQState *ahci, uint8_t port, uint8_t slot)
+{
+    AHCICommandHeader cmd;
+
+    /* Obtain the Nth Command Header */
+    ahci_get_command_header(ahci, port, slot, &cmd);
+    if (cmd.ctba == 0) {
+        /* No address in it, so just return -- it's empty. */
+        goto tidy;
+    }
+
+    /* Free the Table */
+    ahci_free(ahci, cmd.ctba);
+
+ tidy:
+    /* NULL the header. */
+    memset(&cmd, 0x00, sizeof(cmd));
+    ahci_set_command_header(ahci, port, slot, &cmd);
+    ahci->port[port].ctba[slot] = 0;
+    ahci->port[port].prdtl[slot] = 0;
+}
+
+void ahci_write_fis(AHCIQState *ahci, RegH2DFIS *fis, uint64_t addr)
+{
+    RegH2DFIS tmp = *fis;
+
+    /* The auxiliary FIS fields are defined per-command and are not
+     * currently implemented in libqos/ahci.o, but may or may not need
+     * to be flipped. */
+
+    /* All other FIS fields are 8 bit and do not need to be flipped. */
+    tmp.count = cpu_to_le16(tmp.count);
+
+    memwrite(addr, &tmp, sizeof(tmp));
+}
+
+unsigned ahci_pick_cmd(AHCIQState *ahci, uint8_t port)
+{
+    unsigned i;
+    unsigned j;
+    uint32_t reg;
+
+    reg = ahci_px_rreg(ahci, port, AHCI_PX_CI);
+
+    /* Pick the least recently used command slot that's available */
+    for (i = 0; i < 32; ++i) {
+        j = ((ahci->port[port].next + i) % 32);
+        if (reg & (1 << j)) {
+            continue;
+        }
+        ahci_destroy_command(ahci, port, i);
+        ahci->port[port].next = (j + 1) % 32;
+        return j;
+    }
+
+    g_test_message("All command slots were busy.");
+    g_assert_not_reached();
+}
+
+inline unsigned size_to_prdtl(unsigned bytes, unsigned bytes_per_prd)
+{
+    /* Each PRD can describe up to 4MiB */
+    g_assert_cmphex(bytes_per_prd, <=, 4096 * 1024);
+    g_assert_cmphex(bytes_per_prd & 0x01, ==, 0x00);
+    return (bytes + bytes_per_prd - 1) / bytes_per_prd;
+}
+
+/* Given a guest buffer address, perform an IO operation */
+void ahci_guest_io(AHCIQState *ahci, uint8_t port, uint8_t ide_cmd,
+                   uint64_t buffer, size_t bufsize)
+{
+    AHCICommand *cmd;
+
+    cmd = ahci_command_create(ide_cmd);
+    ahci_command_set_buffer(cmd, buffer);
+    ahci_command_set_size(cmd, bufsize);
+    ahci_command_commit(ahci, cmd, port);
+    ahci_command_issue(ahci, cmd);
+    ahci_command_verify(ahci, cmd);
+    ahci_command_free(cmd);
+}
+
+struct AHCICommand {
+    /* Test Management Data */
+    uint8_t name;
+    uint8_t port;
+    uint8_t slot;
+    uint32_t interrupts;
+    uint64_t xbytes;
+    uint32_t prd_size;
+    uint64_t buffer;
+    AHCICommandProp *props;
+    /* Data to be transferred to the guest */
+    AHCICommandHeader header;
+    RegH2DFIS fis;
+    void *atapi_cmd;
+};
+
+static AHCICommandProp *ahci_command_find(uint8_t command_name)
+{
+    int i;
+
+    for (i = 0; i < ARRAY_SIZE(ahci_command_properties); i++) {
+        if (ahci_command_properties[i].cmd == command_name) {
+            return &ahci_command_properties[i];
+        }
+    }
+
+    return NULL;
+}
+
+/* Given a HOST buffer, create a buffer address and perform an IO operation. */
+void ahci_io(AHCIQState *ahci, uint8_t port, uint8_t ide_cmd,
+             void *buffer, size_t bufsize)
+{
+    uint64_t ptr;
+    AHCICommandProp *props;
+
+    props = ahci_command_find(ide_cmd);
+    g_assert(props);
+    ptr = ahci_alloc(ahci, bufsize);
+    g_assert(ptr);
+
+    if (props->write) {
+        memwrite(ptr, buffer, bufsize);
+    }
+
+    ahci_guest_io(ahci, port, ide_cmd, ptr, bufsize);
+
+    if (props->read) {
+        memread(ptr, buffer, bufsize);
+    }
+
+    ahci_free(ahci, ptr);
+}
+
+/**
+ * Initializes a basic command header in memory.
+ * We assume that this is for an ATA command using RegH2DFIS.
+ */
+static void command_header_init(AHCICommand *cmd)
+{
+    AHCICommandHeader *hdr = &cmd->header;
+    AHCICommandProp *props = cmd->props;
+
+    hdr->flags = 5;             /* RegH2DFIS is 5 DW long. Must be < 32 */
+    hdr->flags |= CMDH_CLR_BSY; /* Clear the BSY bit when done */
+    if (props->write) {
+        hdr->flags |= CMDH_WRITE;
+    }
+    if (props->atapi) {
+        hdr->flags |= CMDH_ATAPI;
+    }
+    /* Other flags: PREFETCH, RESET, and BIST */
+    hdr->prdtl = size_to_prdtl(cmd->xbytes, cmd->prd_size);
+    hdr->prdbc = 0;
+    hdr->ctba = 0;
+}
+
+static void command_table_init(AHCICommand *cmd)
+{
+    RegH2DFIS *fis = &(cmd->fis);
+
+    fis->fis_type = REG_H2D_FIS;
+    fis->flags = REG_H2D_FIS_CMD; /* "Command" bit */
+    fis->command = cmd->name;
+    cmd->fis.feature_low = 0x00;
+    cmd->fis.feature_high = 0x00;
+    if (cmd->props->lba28 || cmd->props->lba48) {
+        cmd->fis.device = ATA_DEVICE_LBA;
+    }
+    cmd->fis.count = (cmd->xbytes / AHCI_SECTOR_SIZE);
+    cmd->fis.icc = 0x00;
+    cmd->fis.control = 0x00;
+    memset(cmd->fis.aux, 0x00, ARRAY_SIZE(cmd->fis.aux));
+}
+
+AHCICommand *ahci_command_create(uint8_t command_name)
+{
+    AHCICommandProp *props = ahci_command_find(command_name);
+    AHCICommand *cmd;
+
+    g_assert(props);
+    cmd = g_malloc0(sizeof(AHCICommand));
+    g_assert(!(props->dma && props->pio));
+    g_assert(!(props->lba28 && props->lba48));
+    g_assert(!(props->read && props->write));
+    g_assert(!props->size || props->data);
+
+    /* Defaults and book-keeping */
+    cmd->props = props;
+    cmd->name = command_name;
+    cmd->xbytes = props->size;
+    cmd->prd_size = 4096;
+    cmd->buffer = 0xabad1dea;
+
+    cmd->interrupts = AHCI_PX_IS_DHRS;
+    /* BUG: We expect the DPS interrupt for data commands */
+    /* cmd->interrupts |= props->data ? AHCI_PX_IS_DPS : 0; */
+    /* BUG: We expect the DMA Setup interrupt for DMA commands */
+    /* cmd->interrupts |= props->dma ? AHCI_PX_IS_DSS : 0; */
+    cmd->interrupts |= props->pio ? AHCI_PX_IS_PSS : 0;
+
+    command_header_init(cmd);
+    command_table_init(cmd);
+
+    return cmd;
+}
+
+void ahci_command_free(AHCICommand *cmd)
+{
+    g_free(cmd);
+}
+
+void ahci_command_set_buffer(AHCICommand *cmd, uint64_t buffer)
+{
+    cmd->buffer = buffer;
+}
+
+void ahci_command_set_sizes(AHCICommand *cmd, uint64_t xbytes,
+                            unsigned prd_size)
+{
+    /* Each PRD can describe up to 4MiB, and must not be odd. */
+    g_assert_cmphex(prd_size, <=, 4096 * 1024);
+    g_assert_cmphex(prd_size & 0x01, ==, 0x00);
+    cmd->prd_size = prd_size;
+    cmd->xbytes = xbytes;
+    cmd->fis.count = (cmd->xbytes / AHCI_SECTOR_SIZE);
+    cmd->header.prdtl = size_to_prdtl(cmd->xbytes, cmd->prd_size);
+}
+
+void ahci_command_set_size(AHCICommand *cmd, uint64_t xbytes)
+{
+    ahci_command_set_sizes(cmd, xbytes, cmd->prd_size);
+}
+
+void ahci_command_set_prd_size(AHCICommand *cmd, unsigned prd_size)
+{
+    ahci_command_set_sizes(cmd, cmd->xbytes, prd_size);
+}
+
+void ahci_command_commit(AHCIQState *ahci, AHCICommand *cmd, uint8_t port)
+{
+    uint16_t i, prdtl;
+    uint64_t table_size, table_ptr, remaining;
+    PRD prd;
+
+    /* This command is now tied to this port/command slot */
+    cmd->port = port;
+    cmd->slot = ahci_pick_cmd(ahci, port);
+
+    /* Create a buffer for the command table */
+    prdtl = size_to_prdtl(cmd->xbytes, cmd->prd_size);
+    table_size = CMD_TBL_SIZ(prdtl);
+    table_ptr = ahci_alloc(ahci, table_size);
+    g_assert(table_ptr);
+    /* AHCI 1.3: Must be aligned to 0x80 */
+    g_assert((table_ptr & 0x7F) == 0x00);
+    cmd->header.ctba = table_ptr;
+
+    /* Commit the command header and command FIS */
+    ahci_set_command_header(ahci, port, cmd->slot, &(cmd->header));
+    ahci_write_fis(ahci, &(cmd->fis), table_ptr);
+
+    /* Construct and write the PRDs to the command table */
+    g_assert_cmphex(prdtl, ==, cmd->header.prdtl);
+    remaining = cmd->xbytes;
+    for (i = 0; i < prdtl; ++i) {
+        prd.dba = cpu_to_le64(cmd->buffer + (cmd->prd_size * i));
+        prd.res = 0;
+        if (remaining > cmd->prd_size) {
+            /* Note that byte count is 0-based. */
+            prd.dbc = cpu_to_le32(cmd->prd_size - 1);
+            remaining -= cmd->prd_size;
+        } else {
+            /* Again, dbc is 0-based. */
+            prd.dbc = cpu_to_le32(remaining - 1);
+            remaining = 0;
+        }
+        prd.dbc |= cpu_to_le32(0x80000000); /* Request DPS Interrupt */
+
+        /* Commit the PRD entry to the Command Table */
+        memwrite(table_ptr + 0x80 + (i * sizeof(PRD)),
+                 &prd, sizeof(PRD));
+    }
+
+    /* Bookmark the PRDTL and CTBA values */
+    ahci->port[port].ctba[cmd->slot] = table_ptr;
+    ahci->port[port].prdtl[cmd->slot] = prdtl;
+}
+
+void ahci_command_issue_async(AHCIQState *ahci, AHCICommand *cmd)
+{
+    if (cmd->props->ncq) {
+        ahci_px_wreg(ahci, cmd->port, AHCI_PX_SACT, (1 << cmd->slot));
+    }
+
+    ahci_px_wreg(ahci, cmd->port, AHCI_PX_CI, (1 << cmd->slot));
+}
+
+void ahci_command_wait(AHCIQState *ahci, AHCICommand *cmd)
+{
+    /* We can't rely on STS_BSY until the command has started processing.
+     * Therefore, we also use the Command Issue bit as indication of
+     * a command in-flight. */
+    while (BITSET(ahci_px_rreg(ahci, cmd->port, AHCI_PX_TFD),
+                  AHCI_PX_TFD_STS_BSY) ||
+           BITSET(ahci_px_rreg(ahci, cmd->port, AHCI_PX_CI), (1 << cmd->slot))) {
+        usleep(50);
+    }
+}
+
+void ahci_command_issue(AHCIQState *ahci, AHCICommand *cmd)
+{
+    ahci_command_issue_async(ahci, cmd);
+    ahci_command_wait(ahci, cmd);
+}
+
+void ahci_command_verify(AHCIQState *ahci, AHCICommand *cmd)
+{
+    uint8_t slot = cmd->slot;
+    uint8_t port = cmd->port;
+
+    ahci_port_check_error(ahci, port);
+    ahci_port_check_interrupts(ahci, port, cmd->interrupts);
+    ahci_port_check_nonbusy(ahci, port, slot);
+    ahci_port_check_cmd_sanity(ahci, port, slot, cmd->xbytes);
+    ahci_port_check_d2h_sanity(ahci, port, slot);
+    if (cmd->props->pio) {
+        ahci_port_check_pio_sanity(ahci, port, slot, cmd->xbytes);
+    }
+}
+
+uint8_t ahci_command_slot(AHCICommand *cmd)
+{
+    return cmd->slot;
+}

tests/libqos/ahci.h

@@ -0,0 +1,549 @@
+#ifndef __libqos_ahci_h
+#define __libqos_ahci_h
+
+/*
+ * AHCI qtest library functions and definitions
+ *
+ * Copyright (c) 2014 John Snow <jsnow@redhat.com>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+#include <stdint.h>
+#include <stdlib.h>
+#include <stdbool.h>
+#include "libqos/libqos.h"
+#include "libqos/pci.h"
+#include "libqos/malloc-pc.h"
+
+/*** Supplementary PCI Config Space IDs & Masks ***/
+#define PCI_DEVICE_ID_INTEL_Q35_AHCI   (0x2922)
+#define PCI_MSI_FLAGS_RESERVED         (0xFF00)
+#define PCI_PM_CTRL_RESERVED             (0xFC)
+#define PCI_BCC(REG32)          ((REG32) >> 24)
+#define PCI_PI(REG32)   (((REG32) >> 8) & 0xFF)
+#define PCI_SCC(REG32) (((REG32) >> 16) & 0xFF)
+
+/*** Recognized AHCI Device Types ***/
+#define AHCI_INTEL_ICH9 (PCI_DEVICE_ID_INTEL_Q35_AHCI << 16 | \
+                         PCI_VENDOR_ID_INTEL)
+
+/*** AHCI/HBA Register Offsets and Bitmasks ***/
+#define AHCI_CAP                          (0)
+#define AHCI_CAP_NP                    (0x1F)
+#define AHCI_CAP_SXS                   (0x20)
+#define AHCI_CAP_EMS                   (0x40)
+#define AHCI_CAP_CCCS                  (0x80)
+#define AHCI_CAP_NCS                 (0x1F00)
+#define AHCI_CAP_PSC                 (0x2000)
+#define AHCI_CAP_SSC                 (0x4000)
+#define AHCI_CAP_PMD                 (0x8000)
+#define AHCI_CAP_FBSS               (0x10000)
+#define AHCI_CAP_SPM                (0x20000)
+#define AHCI_CAP_SAM                (0x40000)
+#define AHCI_CAP_RESERVED           (0x80000)
+#define AHCI_CAP_ISS               (0xF00000)
+#define AHCI_CAP_SCLO             (0x1000000)
+#define AHCI_CAP_SAL              (0x2000000)
+#define AHCI_CAP_SALP             (0x4000000)
+#define AHCI_CAP_SSS              (0x8000000)
+#define AHCI_CAP_SMPS            (0x10000000)
+#define AHCI_CAP_SSNTF           (0x20000000)
+#define AHCI_CAP_SNCQ            (0x40000000)
+#define AHCI_CAP_S64A            (0x80000000)
+
+#define AHCI_GHC                          (1)
+#define AHCI_GHC_HR                    (0x01)
+#define AHCI_GHC_IE                    (0x02)
+#define AHCI_GHC_MRSM                  (0x04)
+#define AHCI_GHC_RESERVED        (0x7FFFFFF8)
+#define AHCI_GHC_AE              (0x80000000)
+
+#define AHCI_IS                           (2)
+#define AHCI_PI                           (3)
+#define AHCI_VS                           (4)
+
+#define AHCI_CCCCTL                       (5)
+#define AHCI_CCCCTL_EN                 (0x01)
+#define AHCI_CCCCTL_RESERVED           (0x06)
+#define AHCI_CCCCTL_CC               (0xFF00)
+#define AHCI_CCCCTL_TV           (0xFFFF0000)
+
+#define AHCI_CCCPORTS                     (6)
+#define AHCI_EMLOC                        (7)
+
+#define AHCI_EMCTL                        (8)
+#define AHCI_EMCTL_STSMR               (0x01)
+#define AHCI_EMCTL_CTLTM              (0x100)
+#define AHCI_EMCTL_CTLRST             (0x200)
+#define AHCI_EMCTL_RESERVED      (0xF0F0FCFE)
+
+#define AHCI_CAP2                         (9)
+#define AHCI_CAP2_BOH                  (0x01)
+#define AHCI_CAP2_NVMP                 (0x02)
+#define AHCI_CAP2_APST                 (0x04)
+#define AHCI_CAP2_RESERVED       (0xFFFFFFF8)
+
+#define AHCI_BOHC                        (10)
+#define AHCI_RESERVED                    (11)
+#define AHCI_NVMHCI                      (24)
+#define AHCI_VENDOR                      (40)
+#define AHCI_PORTS                       (64)
+
+/*** Port Memory Offsets & Bitmasks ***/
+#define AHCI_PX_CLB                       (0)
+#define AHCI_PX_CLB_RESERVED          (0x1FF)
+
+#define AHCI_PX_CLBU                      (1)
+
+#define AHCI_PX_FB                        (2)
+#define AHCI_PX_FB_RESERVED            (0xFF)
+
+#define AHCI_PX_FBU                       (3)
+
+#define AHCI_PX_IS                        (4)
+#define AHCI_PX_IS_DHRS                 (0x1)
+#define AHCI_PX_IS_PSS                  (0x2)
+#define AHCI_PX_IS_DSS                  (0x4)
+#define AHCI_PX_IS_SDBS                 (0x8)
+#define AHCI_PX_IS_UFS                 (0x10)
+#define AHCI_PX_IS_DPS                 (0x20)
+#define AHCI_PX_IS_PCS                 (0x40)
+#define AHCI_PX_IS_DMPS                (0x80)
+#define AHCI_PX_IS_RESERVED       (0x23FFF00)
+#define AHCI_PX_IS_PRCS            (0x400000)
+#define AHCI_PX_IS_IPMS            (0x800000)
+#define AHCI_PX_IS_OFS            (0x1000000)
+#define AHCI_PX_IS_INFS           (0x4000000)
+#define AHCI_PX_IS_IFS            (0x8000000)
+#define AHCI_PX_IS_HBDS          (0x10000000)
+#define AHCI_PX_IS_HBFS          (0x20000000)
+#define AHCI_PX_IS_TFES          (0x40000000)
+#define AHCI_PX_IS_CPDS          (0x80000000)
+
+#define AHCI_PX_IE                        (5)
+#define AHCI_PX_IE_DHRE                 (0x1)
+#define AHCI_PX_IE_PSE                  (0x2)
+#define AHCI_PX_IE_DSE                  (0x4)
+#define AHCI_PX_IE_SDBE                 (0x8)
+#define AHCI_PX_IE_UFE                 (0x10)
+#define AHCI_PX_IE_DPE                 (0x20)
+#define AHCI_PX_IE_PCE                 (0x40)
+#define AHCI_PX_IE_DMPE                (0x80)
+#define AHCI_PX_IE_RESERVED       (0x23FFF00)
+#define AHCI_PX_IE_PRCE            (0x400000)
+#define AHCI_PX_IE_IPME            (0x800000)
+#define AHCI_PX_IE_OFE            (0x1000000)
+#define AHCI_PX_IE_INFE           (0x4000000)
+#define AHCI_PX_IE_IFE            (0x8000000)
+#define AHCI_PX_IE_HBDE          (0x10000000)
+#define AHCI_PX_IE_HBFE          (0x20000000)
+#define AHCI_PX_IE_TFEE          (0x40000000)
+#define AHCI_PX_IE_CPDE          (0x80000000)
+
+#define AHCI_PX_CMD                       (6)
+#define AHCI_PX_CMD_ST                  (0x1)
+#define AHCI_PX_CMD_SUD                 (0x2)
+#define AHCI_PX_CMD_POD                 (0x4)
+#define AHCI_PX_CMD_CLO                 (0x8)
+#define AHCI_PX_CMD_FRE                (0x10)
+#define AHCI_PX_CMD_RESERVED           (0xE0)
+#define AHCI_PX_CMD_CCS              (0x1F00)
+#define AHCI_PX_CMD_MPSS             (0x2000)
+#define AHCI_PX_CMD_FR               (0x4000)
+#define AHCI_PX_CMD_CR               (0x8000)
+#define AHCI_PX_CMD_CPS             (0x10000)
+#define AHCI_PX_CMD_PMA             (0x20000)
+#define AHCI_PX_CMD_HPCP            (0x40000)
+#define AHCI_PX_CMD_MPSP            (0x80000)
+#define AHCI_PX_CMD_CPD            (0x100000)
+#define AHCI_PX_CMD_ESP            (0x200000)
+#define AHCI_PX_CMD_FBSCP          (0x400000)
+#define AHCI_PX_CMD_APSTE          (0x800000)
+#define AHCI_PX_CMD_ATAPI         (0x1000000)
+#define AHCI_PX_CMD_DLAE          (0x2000000)
+#define AHCI_PX_CMD_ALPE          (0x4000000)
+#define AHCI_PX_CMD_ASP           (0x8000000)
+#define AHCI_PX_CMD_ICC          (0xF0000000)
+
+#define AHCI_PX_RES1                      (7)
+
+#define AHCI_PX_TFD                       (8)
+#define AHCI_PX_TFD_STS                (0xFF)
+#define AHCI_PX_TFD_STS_ERR            (0x01)
+#define AHCI_PX_TFD_STS_CS1            (0x06)
+#define AHCI_PX_TFD_STS_DRQ            (0x08)
+#define AHCI_PX_TFD_STS_CS2            (0x70)
+#define AHCI_PX_TFD_STS_BSY            (0x80)
+#define AHCI_PX_TFD_ERR              (0xFF00)
+#define AHCI_PX_TFD_RESERVED     (0xFFFF0000)
+
+#define AHCI_PX_SIG                       (9)
+#define AHCI_PX_SIG_SECTOR_COUNT       (0xFF)
+#define AHCI_PX_SIG_LBA_LOW          (0xFF00)
+#define AHCI_PX_SIG_LBA_MID        (0xFF0000)
+#define AHCI_PX_SIG_LBA_HIGH     (0xFF000000)
+
+#define AHCI_PX_SSTS                     (10)
+#define AHCI_PX_SSTS_DET               (0x0F)
+#define AHCI_PX_SSTS_SPD               (0xF0)
+#define AHCI_PX_SSTS_IPM              (0xF00)
+#define AHCI_PX_SSTS_RESERVED    (0xFFFFF000)
+#define SSTS_DET_NO_DEVICE             (0x00)
+#define SSTS_DET_PRESENT               (0x01)
+#define SSTS_DET_ESTABLISHED           (0x03)
+#define SSTS_DET_OFFLINE               (0x04)
+
+#define AHCI_PX_SCTL                     (11)
+
+#define AHCI_PX_SERR                     (12)
+#define AHCI_PX_SERR_ERR             (0xFFFF)
+#define AHCI_PX_SERR_DIAG        (0xFFFF0000)
+#define AHCI_PX_SERR_DIAG_X      (0x04000000)
+
+#define AHCI_PX_SACT                     (13)
+#define AHCI_PX_CI                       (14)
+#define AHCI_PX_SNTF                     (15)
+
+#define AHCI_PX_FBS                      (16)
+#define AHCI_PX_FBS_EN                  (0x1)
+#define AHCI_PX_FBS_DEC                 (0x2)
+#define AHCI_PX_FBS_SDE                 (0x4)
+#define AHCI_PX_FBS_DEV               (0xF00)
+#define AHCI_PX_FBS_ADO              (0xF000)
+#define AHCI_PX_FBS_DWE             (0xF0000)
+#define AHCI_PX_FBS_RESERVED     (0xFFF000F8)
+
+#define AHCI_PX_RES2                     (17)
+#define AHCI_PX_VS                       (28)
+
+#define HBA_DATA_REGION_SIZE            (256)
+#define HBA_PORT_DATA_SIZE              (128)
+#define HBA_PORT_NUM_REG (HBA_PORT_DATA_SIZE/4)
+
+#define AHCI_VERSION_0_95        (0x00000905)
+#define AHCI_VERSION_1_0         (0x00010000)
+#define AHCI_VERSION_1_1         (0x00010100)
+#define AHCI_VERSION_1_2         (0x00010200)
+#define AHCI_VERSION_1_3         (0x00010300)
+
+#define AHCI_SECTOR_SIZE                (512)
+
+/* FIS types */
+enum {
+    REG_H2D_FIS = 0x27,
+    REG_D2H_FIS = 0x34,
+    DMA_ACTIVATE_FIS = 0x39,
+    DMA_SETUP_FIS = 0x41,
+    DATA_FIS = 0x46,
+    BIST_ACTIVATE_FIS = 0x58,
+    PIO_SETUP_FIS = 0x5F,
+    SDB_FIS = 0xA1
+};
+
+/* FIS flags */
+#define REG_H2D_FIS_CMD  0x80
+
+/* ATA Commands */
+enum {
+    /* DMA */
+    CMD_READ_DMA      = 0xC8,
+    CMD_READ_DMA_EXT  = 0x25,
+    CMD_WRITE_DMA     = 0xCA,
+    CMD_WRITE_DMA_EXT = 0x35,
+    /* PIO */
+    CMD_READ_PIO      = 0x20,
+    CMD_READ_PIO_EXT  = 0x24,
+    CMD_WRITE_PIO     = 0x30,
+    CMD_WRITE_PIO_EXT = 0x34,
+    /* Misc */
+    CMD_READ_MAX      = 0xF8,
+    CMD_READ_MAX_EXT  = 0x27,
+    CMD_FLUSH_CACHE   = 0xE7,
+    CMD_IDENTIFY      = 0xEC
+};
+
+/* AHCI Command Header Flags & Masks*/
+#define CMDH_CFL        (0x1F)
+#define CMDH_ATAPI      (0x20)
+#define CMDH_WRITE      (0x40)
+#define CMDH_PREFETCH   (0x80)
+#define CMDH_RESET     (0x100)
+#define CMDH_BIST      (0x200)
+#define CMDH_CLR_BSY   (0x400)
+#define CMDH_RES       (0x800)
+#define CMDH_PMP      (0xF000)
+
+/* ATA device register masks */
+#define ATA_DEVICE_MAGIC 0xA0
+#define ATA_DEVICE_LBA   0x40
+#define ATA_DEVICE_DRIVE 0x10
+#define ATA_DEVICE_HEAD  0x0F
+
+/*** Structures ***/
+
+typedef struct AHCIPortQState {
+    uint64_t fb;
+    uint64_t clb;
+    uint64_t ctba[32];
+    uint16_t prdtl[32];
+    uint8_t next; /** Next Command Slot to Use **/
+} AHCIPortQState;
+
+typedef struct AHCIQState {
+    QOSState *parent;
+    QPCIDevice *dev;
+    void *hba_base;
+    uint64_t barsize;
+    uint32_t fingerprint;
+    uint32_t cap;
+    uint32_t cap2;
+    AHCIPortQState port[32];
+} AHCIQState;
+
+/**
+ * Generic FIS structure.
+ */
+typedef struct FIS {
+    uint8_t fis_type;
+    uint8_t flags;
+    char data[0];
+} __attribute__((__packed__)) FIS;
+
+/**
+ * Register device-to-host FIS structure.
+ */
+typedef struct RegD2HFIS {
+    /* DW0 */
+    uint8_t fis_type;
+    uint8_t flags;
+    uint8_t status;
+    uint8_t error;
+    /* DW1 */
+    uint8_t lba_lo[3];
+    uint8_t device;
+    /* DW2 */
+    uint8_t lba_hi[3];
+    uint8_t res0;
+    /* DW3 */
+    uint16_t count;
+    uint16_t res1;
+    /* DW4 */
+    uint32_t res2;
+} __attribute__((__packed__)) RegD2HFIS;
+
+/**
+ * Register device-to-host FIS structure;
+ * PIO Setup variety.
+ */
+typedef struct PIOSetupFIS {
+    /* DW0 */
+    uint8_t fis_type;
+    uint8_t flags;
+    uint8_t status;
+    uint8_t error;
+    /* DW1 */
+    uint8_t lba_lo[3];
+    uint8_t device;
+    /* DW2 */
+    uint8_t lba_hi[3];
+    uint8_t res0;
+    /* DW3 */
+    uint16_t count;
+    uint8_t res1;
+    uint8_t e_status;
+    /* DW4 */
+    uint16_t tx_count;
+    uint16_t res2;
+} __attribute__((__packed__)) PIOSetupFIS;
+
+/**
+ * Register host-to-device FIS structure.
+ */
+typedef struct RegH2DFIS {
+    /* DW0 */
+    uint8_t fis_type;
+    uint8_t flags;
+    uint8_t command;
+    uint8_t feature_low;
+    /* DW1 */
+    uint8_t lba_lo[3];
+    uint8_t device;
+    /* DW2 */
+    uint8_t lba_hi[3];
+    uint8_t feature_high;
+    /* DW3 */
+    uint16_t count;
+    uint8_t icc;
+    uint8_t control;
+    /* DW4 */
+    uint8_t aux[4];
+} __attribute__((__packed__)) RegH2DFIS;
+
+/**
+ * Command List entry structure.
+ * The command list contains between 1-32 of these structures.
+ */
+typedef struct AHCICommandHeader {
+    uint16_t flags; /* Cmd-Fis-Len, PMP#, and flags. */
+    uint16_t prdtl; /* Phys Region Desc. Table Length */
+    uint32_t prdbc; /* Phys Region Desc. Byte Count */
+    uint64_t ctba;  /* Command Table Descriptor Base Address */
+    uint32_t res[4];
+} __attribute__((__packed__)) AHCICommandHeader;
+
+/**
+ * Physical Region Descriptor; pointed to by the Command List Header,
+ * struct ahci_command.
+ */
+typedef struct PRD {
+    uint64_t dba;  /* Data Base Address */
+    uint32_t res;  /* Reserved */
+    uint32_t dbc;  /* Data Byte Count (0-indexed) & Interrupt Flag (bit 2^31) */
+} __attribute__((__packed__)) PRD;
+
+/* Opaque, defined within ahci.c */
+typedef struct AHCICommand AHCICommand;
+
+/*** Macro Utilities ***/
+#define BITANY(data, mask) (((data) & (mask)) != 0)
+#define BITSET(data, mask) (((data) & (mask)) == (mask))
+#define BITCLR(data, mask) (((data) & (mask)) == 0)
+#define ASSERT_BIT_SET(data, mask) g_assert_cmphex((data) & (mask), ==, (mask))
+#define ASSERT_BIT_CLEAR(data, mask) g_assert_cmphex((data) & (mask), ==, 0)
+
+/* For calculating how big the PRD table needs to be: */
+#define CMD_TBL_SIZ(n) ((0x80 + ((n) * sizeof(PRD)) + 0x7F) & ~0x7F)
+
+/* Helpers for reading/writing AHCI HBA register values */
+
+static inline uint32_t ahci_mread(AHCIQState *ahci, size_t offset)
+{
+    return qpci_io_readl(ahci->dev, ahci->hba_base + offset);
+}
+
+static inline void ahci_mwrite(AHCIQState *ahci, size_t offset, uint32_t value)
+{
+    qpci_io_writel(ahci->dev, ahci->hba_base + offset, value);
+}
+
+static inline uint32_t ahci_rreg(AHCIQState *ahci, uint32_t reg_num)
+{
+    return ahci_mread(ahci, 4 * reg_num);
+}
+
+static inline void ahci_wreg(AHCIQState *ahci, uint32_t reg_num, uint32_t value)
+{
+    ahci_mwrite(ahci, 4 * reg_num, value);
+}
+
+static inline void ahci_set(AHCIQState *ahci, uint32_t reg_num, uint32_t mask)
+{
+    ahci_wreg(ahci, reg_num, ahci_rreg(ahci, reg_num) | mask);
+}
+
+static inline void ahci_clr(AHCIQState *ahci, uint32_t reg_num, uint32_t mask)
+{
+    ahci_wreg(ahci, reg_num, ahci_rreg(ahci, reg_num) & ~mask);
+}
+
+static inline size_t ahci_px_offset(uint8_t port, uint32_t reg_num)
+{
+    return AHCI_PORTS + (HBA_PORT_NUM_REG * port) + reg_num;
+}
+
+static inline uint32_t ahci_px_rreg(AHCIQState *ahci, uint8_t port,
+                                    uint32_t reg_num)
+{
+    return ahci_rreg(ahci, ahci_px_offset(port, reg_num));
+}
+
+static inline void ahci_px_wreg(AHCIQState *ahci, uint8_t port,
+                                uint32_t reg_num, uint32_t value)
+{
+    ahci_wreg(ahci, ahci_px_offset(port, reg_num), value);
+}
+
+static inline void ahci_px_set(AHCIQState *ahci, uint8_t port,
+                               uint32_t reg_num, uint32_t mask)
+{
+    ahci_px_wreg(ahci, port, reg_num,
+                 ahci_px_rreg(ahci, port, reg_num) | mask);
+}
+
+static inline void ahci_px_clr(AHCIQState *ahci, uint8_t port,
+                               uint32_t reg_num, uint32_t mask)
+{
+    ahci_px_wreg(ahci, port, reg_num,
+                 ahci_px_rreg(ahci, port, reg_num) & ~mask);
+}
+
+/*** Prototypes ***/
+uint64_t ahci_alloc(AHCIQState *ahci, size_t bytes);
+void ahci_free(AHCIQState *ahci, uint64_t addr);
+QPCIDevice *get_ahci_device(uint32_t *fingerprint);
+void free_ahci_device(QPCIDevice *dev);
+void ahci_clean_mem(AHCIQState *ahci);
+void ahci_pci_enable(AHCIQState *ahci);
+void start_ahci_device(AHCIQState *ahci);
+void ahci_hba_enable(AHCIQState *ahci);
+unsigned ahci_port_select(AHCIQState *ahci);
+void ahci_port_clear(AHCIQState *ahci, uint8_t port);
+void ahci_port_check_error(AHCIQState *ahci, uint8_t port);
+void ahci_port_check_interrupts(AHCIQState *ahci, uint8_t port,
+                                uint32_t intr_mask);
+void ahci_port_check_nonbusy(AHCIQState *ahci, uint8_t port, uint8_t slot);
+void ahci_port_check_d2h_sanity(AHCIQState *ahci, uint8_t port, uint8_t slot);
+void ahci_port_check_pio_sanity(AHCIQState *ahci, uint8_t port,
+                                uint8_t slot, size_t buffsize);
+void ahci_port_check_cmd_sanity(AHCIQState *ahci, uint8_t port,
+                                uint8_t slot, size_t buffsize);
+void ahci_get_command_header(AHCIQState *ahci, uint8_t port,
+                             uint8_t slot, AHCICommandHeader *cmd);
+void ahci_set_command_header(AHCIQState *ahci, uint8_t port,
+                             uint8_t slot, AHCICommandHeader *cmd);
+void ahci_destroy_command(AHCIQState *ahci, uint8_t port, uint8_t slot);
+void ahci_write_fis(AHCIQState *ahci, RegH2DFIS *fis, uint64_t addr);
+unsigned ahci_pick_cmd(AHCIQState *ahci, uint8_t port);
+unsigned size_to_prdtl(unsigned bytes, unsigned bytes_per_prd);
+void ahci_guest_io(AHCIQState *ahci, uint8_t port, uint8_t ide_cmd,
+                   uint64_t gbuffer, size_t size);
+void ahci_io(AHCIQState *ahci, uint8_t port, uint8_t ide_cmd,
+             void *buffer, size_t bufsize);
+
+/* Command Lifecycle */
+AHCICommand *ahci_command_create(uint8_t command_name);
+void ahci_command_commit(AHCIQState *ahci, AHCICommand *cmd, uint8_t port);
+void ahci_command_issue(AHCIQState *ahci, AHCICommand *cmd);
+void ahci_command_issue_async(AHCIQState *ahci, AHCICommand *cmd);
+void ahci_command_wait(AHCIQState *ahci, AHCICommand *cmd);
+void ahci_command_verify(AHCIQState *ahci, AHCICommand *cmd);
+void ahci_command_free(AHCICommand *cmd);
+
+/* Command adjustments */
+void ahci_command_set_buffer(AHCICommand *cmd, uint64_t buffer);
+void ahci_command_set_size(AHCICommand *cmd, uint64_t xbytes);
+void ahci_command_set_prd_size(AHCICommand *cmd, unsigned prd_size);
+void ahci_command_set_sizes(AHCICommand *cmd, uint64_t xbytes,
+                            unsigned prd_size);
+
+/* Command Misc */
+uint8_t ahci_command_slot(AHCICommand *cmd);
+
+#endif

tests/libqos/libqos-pc.c

@@ -0,0 +1,24 @@
+#include "libqos/libqos-pc.h"
+#include "libqos/malloc-pc.h"
+
+static QOSOps qos_ops = {
+    .init_allocator = pc_alloc_init_flags,
+    .uninit_allocator = pc_alloc_uninit
+};
+
+QOSState *qtest_pc_boot(const char *cmdline_fmt, ...)
+{
+    QOSState *qs;
+    va_list ap;
+
+    va_start(ap, cmdline_fmt);
+    qs = qtest_vboot(&qos_ops, cmdline_fmt, ap);
+    va_end(ap);
+
+    return qs;
+}
+
+void qtest_pc_shutdown(QOSState *qs)
+{
+    return qtest_shutdown(qs);
+}

tests/libqos/libqos-pc.h

@@ -0,0 +1,9 @@
+#ifndef __libqos_pc_h
+#define __libqos_pc_h
+
+#include "libqos/libqos.h"
+
+QOSState *qtest_pc_boot(const char *cmdline_fmt, ...);
+void qtest_pc_shutdown(QOSState *qs);
+
+#endif

tests/libqos/libqos.c

@@ -0,0 +1,63 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <glib.h>
+#include <unistd.h>
+#include <fcntl.h>
+#include <sys/wait.h>
+
+#include "libqtest.h"
+#include "libqos/libqos.h"
+#include "libqos/pci.h"
+
+/*** Test Setup & Teardown ***/
+
+/**
+ * Launch QEMU with the given command line,
+ * and then set up interrupts and our guest malloc interface.
+ */
+QOSState *qtest_vboot(QOSOps *ops, const char *cmdline_fmt, va_list ap)
+{
+    char *cmdline;
+
+    struct QOSState *qs = g_malloc(sizeof(QOSState));
+
+    cmdline = g_strdup_vprintf(cmdline_fmt, ap);
+    qs->qts = qtest_start(cmdline);
+    qs->ops = ops;
+    qtest_irq_intercept_in(global_qtest, "ioapic");
+    if (ops && ops->init_allocator) {
+        qs->alloc = ops->init_allocator(ALLOC_NO_FLAGS);
+    }
+
+    g_free(cmdline);
+    return qs;
+}
+
+/**
+ * Launch QEMU with the given command line,
+ * and then set up interrupts and our guest malloc interface.
+ */
+QOSState *qtest_boot(QOSOps *ops, const char *cmdline_fmt, ...)
+{
+    QOSState *qs;
+    va_list ap;
+
+    va_start(ap, cmdline_fmt);
+    qs = qtest_vboot(ops, cmdline_fmt, ap);
+    va_end(ap);
+
+    return qs;
+}
+
+/**
+ * Tear down the QEMU instance.
+ */
+void qtest_shutdown(QOSState *qs)
+{
+    if (qs->alloc && qs->ops && qs->ops->uninit_allocator) {
+        qs->ops->uninit_allocator(qs->alloc);
+        qs->alloc = NULL;
+    }
+    qtest_quit(qs->qts);
+    g_free(qs);
+}

tests/libqos/libqos.h

@@ -0,0 +1,33 @@
+#ifndef __libqos_h
+#define __libqos_h
+
+#include "libqtest.h"
+#include "libqos/pci.h"
+#include "libqos/malloc-pc.h"
+
+typedef struct QOSOps {
+    QGuestAllocator *(*init_allocator)(QAllocOpts);
+    void (*uninit_allocator)(QGuestAllocator *);
+} QOSOps;
+
+typedef struct QOSState {
+    QTestState *qts;
+    QGuestAllocator *alloc;
+    QOSOps *ops;
+} QOSState;
+
+QOSState *qtest_vboot(QOSOps *ops, const char *cmdline_fmt, va_list ap);
+QOSState *qtest_boot(QOSOps *ops, const char *cmdline_fmt, ...);
+void qtest_shutdown(QOSState *qs);
+
+static inline uint64_t qmalloc(QOSState *q, size_t bytes)
+{
+    return guest_alloc(q->alloc, bytes);
+}
+
+static inline void qfree(QOSState *q, uint64_t addr)
+{
+    guest_free(q->alloc, addr);
+}
+
+#endif

tests/libqos/malloc-pc.c

@@ -32,31 +32,17 @@ void pc_alloc_uninit(QGuestAllocator *allocator)
 
 QGuestAllocator *pc_alloc_init_flags(QAllocOpts flags)
 {
-    QGuestAllocator *s = g_malloc0(sizeof(*s));
+    QGuestAllocator *s;
     uint64_t ram_size;
     QFWCFG *fw_cfg = pc_fw_cfg_init();
-    MemBlock *node;
-
-    s->opts = flags;
-    s->page_size = PAGE_SIZE;
 
     ram_size = qfw_cfg_get_u64(fw_cfg, FW_CFG_RAM_SIZE);
-
+    s = alloc_init_flags(flags, 1 << 20, MIN(ram_size, 0xE0000000));
-    /* Start at 1MB */
+    alloc_set_page_size(s, PAGE_SIZE);
-    s->start = 1 << 20;
-
-    /* Respect PCI hole */
-    s->end = MIN(ram_size, 0xE0000000);
 
     /* clean-up */
     g_free(fw_cfg);
 
-    QTAILQ_INIT(&s->used);
-    QTAILQ_INIT(&s->free);
-
-    node = mlist_new(s->start, s->end - s->start);
-    QTAILQ_INSERT_HEAD(&s->free, node, MLIST_ENTNAME);
-
     return s;
 }
 

tests/libqos/malloc.c

@@ -16,6 +16,26 @@
 #include <inttypes.h>
 #include <glib.h>
 
+typedef QTAILQ_HEAD(MemList, MemBlock) MemList;
+
+typedef struct MemBlock {
+    QTAILQ_ENTRY(MemBlock) MLIST_ENTNAME;
+    uint64_t size;
+    uint64_t addr;
+} MemBlock;
+
+struct QGuestAllocator {
+    QAllocOpts opts;
+    uint64_t start;
+    uint64_t end;
+    uint32_t page_size;
+
+    MemList used;
+    MemList free;
+};
+
+#define DEFAULT_PAGE_SIZE 4096
+
 static void mlist_delete(MemList *list, MemBlock *node)
 {
     g_assert(list && node);
@@ -103,6 +123,21 @@ static void mlist_coalesce(MemList *head, MemBlock *node)
     } while (merge);
 }
 
+static MemBlock *mlist_new(uint64_t addr, uint64_t size)
+{
+    MemBlock *block;
+
+    if (!size) {
+        return NULL;
+    }
+    block = g_malloc0(sizeof(MemBlock));
+
+    block->addr = addr;
+    block->size = size;
+
+    return block;
+}
+
 static uint64_t mlist_fulfill(QGuestAllocator *s, MemBlock *freenode,
                                                                 uint64_t size)
 {
@@ -187,21 +222,6 @@ static void mlist_free(QGuestAllocator *s, uint64_t addr)
     mlist_coalesce(&s->free, node);
 }
 
-MemBlock *mlist_new(uint64_t addr, uint64_t size)
-{
-    MemBlock *block;
-
-    if (!size) {
-        return NULL;
-    }
-    block = g_malloc0(sizeof(MemBlock));
-
-    block->addr = addr;
-    block->size = size;
-
-    return block;
-}
-
 /*
  * Mostly for valgrind happiness, but it does offer
  * a chokepoint for debugging guest memory leaks, too.
@@ -268,3 +288,44 @@ void guest_free(QGuestAllocator *allocator, uint64_t addr)
         mlist_check(allocator);
     }
 }
+
+QGuestAllocator *alloc_init(uint64_t start, uint64_t end)
+{
+    QGuestAllocator *s = g_malloc0(sizeof(*s));
+    MemBlock *node;
+
+    s->start = start;
+    s->end = end;
+
+    QTAILQ_INIT(&s->used);
+    QTAILQ_INIT(&s->free);
+
+    node = mlist_new(s->start, s->end - s->start);
+    QTAILQ_INSERT_HEAD(&s->free, node, MLIST_ENTNAME);
+
+    s->page_size = DEFAULT_PAGE_SIZE;
+
+    return s;
+}
+
+QGuestAllocator *alloc_init_flags(QAllocOpts opts,
+                                  uint64_t start, uint64_t end)
+{
+    QGuestAllocator *s = alloc_init(start, end);
+    s->opts = opts;
+    return s;
+}
+
+void alloc_set_page_size(QGuestAllocator *allocator, size_t page_size)
+{
+    /* Can't alter the page_size for an allocator in-use */
+    g_assert(QTAILQ_EMPTY(&allocator->used));
+
+    g_assert(is_power_of_2(page_size));
+    allocator->page_size = page_size;
+}
+
+void alloc_set_flags(QGuestAllocator *allocator, QAllocOpts opts)
+{
+    allocator->opts |= opts;
+}

tests/libqos/malloc.h

@@ -17,8 +17,6 @@
 #include <sys/types.h>
 #include "qemu/queue.h"
 
-#define MLIST_ENTNAME entries
-
 typedef enum {
     ALLOC_NO_FLAGS    = 0x00,
     ALLOC_LEAK_WARN   = 0x01,
@@ -26,28 +24,18 @@ typedef enum {
     ALLOC_PARANOID    = 0x04
 } QAllocOpts;
 
-typedef QTAILQ_HEAD(MemList, MemBlock) MemList;
+typedef struct QGuestAllocator QGuestAllocator;
-typedef struct MemBlock {
-    QTAILQ_ENTRY(MemBlock) MLIST_ENTNAME;
-    uint64_t size;
-    uint64_t addr;
-} MemBlock;
 
-typedef struct QGuestAllocator {
-    QAllocOpts opts;
-    uint64_t start;
-    uint64_t end;
-    uint32_t page_size;
-
-    MemList used;
-    MemList free;
-} QGuestAllocator;
-
-MemBlock *mlist_new(uint64_t addr, uint64_t size);
 void alloc_uninit(QGuestAllocator *allocator);
 
 /* Always returns page aligned values */
 uint64_t guest_alloc(QGuestAllocator *allocator, size_t size);
 void guest_free(QGuestAllocator *allocator, uint64_t addr);
 
+QGuestAllocator *alloc_init(uint64_t start, uint64_t end);
+QGuestAllocator *alloc_init_flags(QAllocOpts flags,
+                                  uint64_t start, uint64_t end);
+void alloc_set_page_size(QGuestAllocator *allocator, size_t page_size);
+void alloc_set_flags(QGuestAllocator *allocator, QAllocOpts opts);
+
 #endif

tests/qemu-iotests/016.out

@@ -1,23 +0,0 @@
-QA output created by 016
-Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
-
-== reading at EOF ==
-read 512/512 bytes at offset 134217728
-512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
-
-== reading far past EOF ==
-read 512/512 bytes at offset 268435456
-512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
-
-== writing at EOF ==
-wrote 512/512 bytes at offset 134217728
-512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
-read 512/512 bytes at offset 134217728
-512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
-
-== writing far past EOF ==
-wrote 512/512 bytes at offset 268435456
-512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
-read 512/512 bytes at offset 268435456
-512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
-*** done

tests/qemu-iotests/051

@@ -93,6 +93,7 @@ echo
 run_qemu -drive file="$TEST_IMG",format=foo
 run_qemu -drive file="$TEST_IMG",driver=foo
 run_qemu -drive file="$TEST_IMG",driver=raw,format=qcow2
+run_qemu -drive file="$TEST_IMG",driver=qcow2,format=qcow2
 
 echo
 echo === Overriding backing file ===

tests/qemu-iotests/051.out

@@ -5,43 +5,46 @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728 backing_file='TEST_DIR
 === Unknown option ===
 
 Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,unknown_opt=
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,unknown_opt=: could not open disk image TEST_DIR/t.qcow2: Block format 'qcow2' used by device 'ide0-hd0' doesn't support the option 'unknown_opt'
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,unknown_opt=: Block format 'qcow2' used by device 'ide0-hd0' doesn't support the option 'unknown_opt'
 
 Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,unknown_opt=on
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,unknown_opt=on: could not open disk image TEST_DIR/t.qcow2: Block format 'qcow2' used by device 'ide0-hd0' doesn't support the option 'unknown_opt'
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,unknown_opt=on: Block format 'qcow2' used by device 'ide0-hd0' doesn't support the option 'unknown_opt'
 
 Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,unknown_opt=1234
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,unknown_opt=1234: could not open disk image TEST_DIR/t.qcow2: Block format 'qcow2' used by device 'ide0-hd0' doesn't support the option 'unknown_opt'
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,unknown_opt=1234: Block format 'qcow2' used by device 'ide0-hd0' doesn't support the option 'unknown_opt'
 
 Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,unknown_opt=foo
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,unknown_opt=foo: could not open disk image TEST_DIR/t.qcow2: Block format 'qcow2' used by device 'ide0-hd0' doesn't support the option 'unknown_opt'
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,unknown_opt=foo: Block format 'qcow2' used by device 'ide0-hd0' doesn't support the option 'unknown_opt'
 
 
 === Unknown protocol option ===
 
 Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,file.unknown_opt=
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,file.unknown_opt=: could not open disk image TEST_DIR/t.qcow2: Block protocol 'file' doesn't support the option 'unknown_opt'
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,file.unknown_opt=: Block protocol 'file' doesn't support the option 'unknown_opt'
 
 Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,file.unknown_opt=on
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,file.unknown_opt=on: could not open disk image TEST_DIR/t.qcow2: Block protocol 'file' doesn't support the option 'unknown_opt'
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,file.unknown_opt=on: Block protocol 'file' doesn't support the option 'unknown_opt'
 
 Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,file.unknown_opt=1234
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,file.unknown_opt=1234: could not open disk image TEST_DIR/t.qcow2: Block protocol 'file' doesn't support the option 'unknown_opt'
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,file.unknown_opt=1234: Block protocol 'file' doesn't support the option 'unknown_opt'
 
 Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,file.unknown_opt=foo
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,file.unknown_opt=foo: could not open disk image TEST_DIR/t.qcow2: Block protocol 'file' doesn't support the option 'unknown_opt'
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,file.unknown_opt=foo: Block protocol 'file' doesn't support the option 'unknown_opt'
 
 
 === Invalid format ===
 
 Testing: -drive file=TEST_DIR/t.qcow2,format=foo
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=foo: 'foo' invalid format
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=foo: Unknown driver 'foo'
 
 Testing: -drive file=TEST_DIR/t.qcow2,driver=foo
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,driver=foo: could not open disk image TEST_DIR/t.qcow2: Unknown driver 'foo'
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,driver=foo: Unknown driver 'foo'
 
 Testing: -drive file=TEST_DIR/t.qcow2,driver=raw,format=qcow2
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,driver=raw,format=qcow2: could not open disk image TEST_DIR/t.qcow2: Driver specified twice
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,driver=raw,format=qcow2: Cannot specify both 'driver' and 'format'
+
+Testing: -drive file=TEST_DIR/t.qcow2,driver=qcow2,format=qcow2
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,driver=qcow2,format=qcow2: Cannot specify both 'driver' and 'format'
 
 
 === Overriding backing file ===
@@ -55,13 +58,13 @@ ide0-hd0: TEST_DIR/t.qcow2 (qcow2)
 (qemu) qququiquit
 
 Testing: -drive file=TEST_DIR/t.qcow2,driver=raw,backing.file.filename=TEST_DIR/t.qcow2.orig
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,driver=raw,backing.file.filename=TEST_DIR/t.qcow2.orig: could not open disk image TEST_DIR/t.qcow2: Driver doesn't support backing files
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,driver=raw,backing.file.filename=TEST_DIR/t.qcow2.orig: Driver doesn't support backing files
 
 Testing: -drive file=TEST_DIR/t.qcow2,file.backing.driver=file,file.backing.filename=TEST_DIR/t.qcow2.orig
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,file.backing.driver=file,file.backing.filename=TEST_DIR/t.qcow2.orig: could not open disk image TEST_DIR/t.qcow2: Driver doesn't support backing files
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,file.backing.driver=file,file.backing.filename=TEST_DIR/t.qcow2.orig: Driver doesn't support backing files
 
 Testing: -drive file=TEST_DIR/t.qcow2,file.backing.driver=qcow2,file.backing.file.filename=TEST_DIR/t.qcow2.orig
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,file.backing.driver=qcow2,file.backing.file.filename=TEST_DIR/t.qcow2.orig: could not open disk image TEST_DIR/t.qcow2: Driver doesn't support backing files
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,file.backing.driver=qcow2,file.backing.file.filename=TEST_DIR/t.qcow2.orig: Driver doesn't support backing files
 
 
 === Enable and disable lazy refcounting on the command line, plus some invalid values ===
@@ -75,20 +78,20 @@ QEMU X.Y.Z monitor - type 'help' for more information
 (qemu) qququiquit
 
 Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy-refcounts=
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy-refcounts=: could not open disk image TEST_DIR/t.qcow2: Parameter 'lazy-refcounts' expects 'on' or 'off'
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy-refcounts=: Parameter 'lazy-refcounts' expects 'on' or 'off'
 
 Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy-refcounts=42
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy-refcounts=42: could not open disk image TEST_DIR/t.qcow2: Parameter 'lazy-refcounts' expects 'on' or 'off'
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy-refcounts=42: Parameter 'lazy-refcounts' expects 'on' or 'off'
 
 Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy-refcounts=foo
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy-refcounts=foo: could not open disk image TEST_DIR/t.qcow2: Parameter 'lazy-refcounts' expects 'on' or 'off'
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy-refcounts=foo: Parameter 'lazy-refcounts' expects 'on' or 'off'
 
 
 === With version 2 images enabling lazy refcounts must fail ===
 
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
 Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy-refcounts=on
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy-refcounts=on: could not open disk image TEST_DIR/t.qcow2: Lazy refcounts require a qcow2 image with at least qemu 1.1 compatibility level
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy-refcounts=on: Lazy refcounts require a qcow2 image with at least qemu 1.1 compatibility level
 
 Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy-refcounts=off
 QEMU X.Y.Z monitor - type 'help' for more information
@@ -248,31 +251,31 @@ QEMU X.Y.Z monitor - type 'help' for more information
 (qemu) qququiquit
 
 Testing: -drive file=TEST_DIR/t.qcow2,file.driver=qcow2
-QEMU_PROG: -drive file=TEST_DIR/t.qcow2,file.driver=qcow2: could not open disk image TEST_DIR/t.qcow2: Block format 'qcow2' used by device '' doesn't support the option 'filename'
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,file.driver=qcow2: Block format 'qcow2' used by device '' doesn't support the option 'filename'
 
 
 === Leaving out required options ===
 
 Testing: -drive driver=file
-QEMU_PROG: -drive driver=file: could not open disk image ide0-hd0: The 'file' block driver requires a file name
+QEMU_PROG: -drive driver=file: The 'file' block driver requires a file name
 
 Testing: -drive driver=nbd
-QEMU_PROG: -drive driver=nbd: could not open disk image ide0-hd0: one of path and host must be specified.
+QEMU_PROG: -drive driver=nbd: one of path and host must be specified.
 
 Testing: -drive driver=raw
-QEMU_PROG: -drive driver=raw: could not open disk image ide0-hd0: Can't use 'raw' as a block driver for the protocol level
+QEMU_PROG: -drive driver=raw: Can't use 'raw' as a block driver for the protocol level
 
 Testing: -drive file.driver=file
-QEMU_PROG: -drive file.driver=file: could not open disk image ide0-hd0: The 'file' block driver requires a file name
+QEMU_PROG: -drive file.driver=file: The 'file' block driver requires a file name
 
 Testing: -drive file.driver=nbd
-QEMU_PROG: -drive file.driver=nbd: could not open disk image ide0-hd0: one of path and host must be specified.
+QEMU_PROG: -drive file.driver=nbd: one of path and host must be specified.
 
 Testing: -drive file.driver=raw
-QEMU_PROG: -drive file.driver=raw: could not open disk image ide0-hd0: Can't use 'raw' as a block driver for the protocol level
+QEMU_PROG: -drive file.driver=raw: Can't use 'raw' as a block driver for the protocol level
 
 Testing: -drive foo=bar
-QEMU_PROG: -drive foo=bar: could not open disk image ide0-hd0: Must specify either driver or file
+QEMU_PROG: -drive foo=bar: Must specify either driver or file
 
 
 === Specifying both an option and its legacy alias ===
@@ -323,13 +326,13 @@ QEMU_PROG: -drive file=TEST_DIR/t.qcow2,readonly=on,read-only=off: 'read-only' a
 === Parsing protocol from file name ===
 
 Testing: -hda foo:bar
-QEMU_PROG: -hda foo:bar: could not open disk image foo:bar: Unknown protocol
+QEMU_PROG: -hda foo:bar: Unknown protocol 'foo'
 
 Testing: -drive file=foo:bar
-QEMU_PROG: -drive file=foo:bar: could not open disk image foo:bar: Unknown protocol
+QEMU_PROG: -drive file=foo:bar: Unknown protocol 'foo'
 
 Testing: -drive file.filename=foo:bar
-QEMU_PROG: -drive file.filename=foo:bar: could not open disk image ide0-hd0: Could not open 'foo:bar': No such file or directory
+QEMU_PROG: -drive file.filename=foo:bar: Could not open 'foo:bar': No such file or directory
 
 Testing: -hda file:TEST_DIR/t.qcow2
 QEMU X.Y.Z monitor - type 'help' for more information
@@ -340,7 +343,7 @@ QEMU X.Y.Z monitor - type 'help' for more information
 (qemu) qququiquit
 
 Testing: -drive file.filename=file:TEST_DIR/t.qcow2
-QEMU_PROG: -drive file.filename=file:TEST_DIR/t.qcow2: could not open disk image ide0-hd0: Could not open 'file:TEST_DIR/t.qcow2': No such file or directory
+QEMU_PROG: -drive file.filename=file:TEST_DIR/t.qcow2: Could not open 'file:TEST_DIR/t.qcow2': No such file or directory
 
 
 === Snapshot mode ===

tests/qemu-iotests/087.out

@@ -21,9 +21,9 @@ QMP_VERSION
 {"return": {}}
 {"error": {"class": "GenericError", "desc": "Device with id 'disk' already exists"}}
 {"error": {"class": "GenericError", "desc": "Device name 'test-node' conflicts with an existing node name"}}
-{"error": {"class": "GenericError", "desc": "could not open disk image disk2: node-name=disk is conflicting with a device id"}}
+{"error": {"class": "GenericError", "desc": "node-name=disk is conflicting with a device id"}}
-{"error": {"class": "GenericError", "desc": "could not open disk image disk2: Duplicate node name"}}
+{"error": {"class": "GenericError", "desc": "Duplicate node name"}}
-{"error": {"class": "GenericError", "desc": "could not open disk image disk3: node-name=disk3 is conflicting with a device id"}}
+{"error": {"class": "GenericError", "desc": "node-name=disk3 is conflicting with a device id"}}
 {"return": {}}
 {"timestamp": {"seconds":  TIMESTAMP, "microseconds":  TIMESTAMP}, "event": "SHUTDOWN"}
 {"timestamp": {"seconds":  TIMESTAMP, "microseconds":  TIMESTAMP}, "event": "DEVICE_TRAY_MOVED", "data": {"device": "ide1-cd0", "tray-open": true}}
@@ -57,7 +57,7 @@ QMP_VERSION
 Testing:
 QMP_VERSION
 {"return": {}}
-{"error": {"class": "GenericError", "desc": "could not open disk image disk: Guest must be stopped for opening of encrypted image"}}
+{"error": {"class": "GenericError", "desc": "Guest must be stopped for opening of encrypted image"}}
 {"return": {}}
 {"timestamp": {"seconds":  TIMESTAMP, "microseconds":  TIMESTAMP}, "event": "SHUTDOWN"}
 {"timestamp": {"seconds":  TIMESTAMP, "microseconds":  TIMESTAMP}, "event": "DEVICE_TRAY_MOVED", "data": {"device": "ide1-cd0", "tray-open": true}}

tests/qemu-iotests/093

@@ -0,0 +1,114 @@
+#!/usr/bin/env python
+#
+# Tests for IO throttling
+#
+# Copyright (C) 2015 Red Hat, Inc.
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+#
+
+import iotests
+
+class ThrottleTestCase(iotests.QMPTestCase):
+    test_img = "null-aio://"
+
+    def blockstats(self, device):
+        result = self.vm.qmp("query-blockstats")
+        for r in result['return']:
+            if r['device'] == device:
+                stat = r['stats']
+                return stat['rd_bytes'], stat['rd_operations'], stat['wr_bytes'], stat['wr_operations']
+        raise Exception("Device not found for blockstats: %s" % device)
+
+    def setUp(self):
+        self.vm = iotests.VM().add_drive(self.test_img)
+        self.vm.launch()
+
+    def tearDown(self):
+        self.vm.shutdown()
+
+    def do_test_throttle(self, seconds, params):
+        def check_limit(limit, num):
+            # IO throttling algorithm is discrete, allow 10% error so the test
+            # is more robust
+            return limit == 0 or \
+                   (num < seconds * limit * 1.1
+                   and num > seconds * limit * 0.9)
+
+        nsec_per_sec = 1000000000
+
+        params['device'] = 'drive0'
+
+        result = self.vm.qmp("block_set_io_throttle", conv_keys=False, **params)
+        self.assert_qmp(result, 'return', {})
+
+        # Set vm clock to a known value
+        ns = seconds * nsec_per_sec
+        self.vm.qtest("clock_step %d" % ns)
+
+        # Submit enough requests. They will drain bps_max and iops_max, but the
+        # rest requests won't get executed until we advance the virtual clock
+        # with qtest interface
+        rq_size = 512
+        rd_nr = max(params['bps'] / rq_size / 2,
+                    params['bps_rd'] / rq_size,
+                    params['iops'] / 2,
+                    params['iops_rd'])
+        rd_nr *= seconds * 2
+        wr_nr = max(params['bps'] / rq_size / 2,
+                    params['bps_wr'] / rq_size,
+                    params['iops'] / 2,
+                    params['iops_wr'])
+        wr_nr *= seconds * 2
+        for i in range(rd_nr):
+            self.vm.hmp_qemu_io("drive0", "aio_read %d %d" % (i * rq_size, rq_size))
+        for i in range(wr_nr):
+            self.vm.hmp_qemu_io("drive0", "aio_write %d %d" % (i * rq_size, rq_size))
+
+        start_rd_bytes, start_rd_iops, start_wr_bytes, start_wr_iops = self.blockstats('drive0')
+
+        self.vm.qtest("clock_step %d" % ns)
+        end_rd_bytes, end_rd_iops, end_wr_bytes, end_wr_iops = self.blockstats('drive0')
+
+        rd_bytes = end_rd_bytes - start_rd_bytes
+        rd_iops = end_rd_iops - start_rd_iops
+        wr_bytes = end_wr_bytes - start_wr_bytes
+        wr_iops = end_wr_iops - start_wr_iops
+
+        self.assertTrue(check_limit(params['bps'], rd_bytes + wr_bytes))
+        self.assertTrue(check_limit(params['bps_rd'], rd_bytes))
+        self.assertTrue(check_limit(params['bps_wr'], wr_bytes))
+        self.assertTrue(check_limit(params['iops'], rd_iops + wr_iops))
+        self.assertTrue(check_limit(params['iops_rd'], rd_iops))
+        self.assertTrue(check_limit(params['iops_wr'], wr_iops))
+
+    def test_all(self):
+        params = {"bps": 4096,
+                  "bps_rd": 4096,
+                  "bps_wr": 4096,
+                  "iops": 10,
+                  "iops_rd": 10,
+                  "iops_wr": 10,
+                 }
+        # Pick each out of all possible params and test
+        for tk in params:
+            limits = dict([(k, 0) for k in params])
+            limits[tk] = params[tk]
+            self.do_test_throttle(5, limits)
+
+class ThrottleTestCoroutine(ThrottleTestCase):
+    test_img = "null-co://"
+
+if __name__ == '__main__':
+    iotests.main(supported_fmts=["raw"])

tests/qemu-iotests/093.out

@@ -0,0 +1,5 @@
+..
+----------------------------------------------------------------------
+Ran 2 tests
+
+OK

tests/qemu-iotests/094

@@ -0,0 +1,81 @@
+#!/bin/bash
+#
+# Test case for drive-mirror to NBD (especially bdrv_swap() on NBD BDS)
+#
+# Copyright (C) 2015 Red Hat, Inc.
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+#
+
+# creator
+owner=mreitz@redhat.com
+
+seq="$(basename $0)"
+echo "QA output created by $seq"
+
+here="$PWD"
+tmp=/tmp/$$
+status=1	# failure is the default!
+
+trap "exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ./common.rc
+. ./common.filter
+. ./common.qemu
+
+_supported_fmt generic
+_supported_proto nbd
+_supported_os Linux
+_unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat"
+
+_make_test_img 64M
+$QEMU_IMG create -f $IMGFMT "$TEST_DIR/source.$IMGFMT" 64M | _filter_img_create
+
+_launch_qemu -drive if=none,id=src,file="$TEST_DIR/source.$IMGFMT",format=raw \
+             -nodefaults
+
+_send_qemu_cmd $QEMU_HANDLE \
+    "{'execute': 'qmp_capabilities'}" \
+    'return'
+
+# 'format': 'nbd' is not actually "correct", but this is probably the only way
+# to test bdrv_swap() on an NBD BDS
+_send_qemu_cmd $QEMU_HANDLE  \
+    "{'execute': 'drive-mirror',
+      'arguments': {'device': 'src',
+                    'target': '$TEST_IMG',
+                    'format': 'nbd',
+                    'sync':'full',
+                    'mode':'existing'}}" \
+    'BLOCK_JOB_READY'
+
+_send_qemu_cmd $QEMU_HANDLE  \
+    "{'execute': 'block-job-complete',
+      'arguments': {'device': 'src'}}" \
+    'BLOCK_JOB_COMPLETE'
+
+_send_qemu_cmd $QEMU_HANDLE \
+    "{'execute': 'quit'}" \
+    'return'
+
+wait=1 _cleanup_qemu
+
+_cleanup_test_img
+rm -f "$TEST_DIR/source.$IMGFMT"
+
+# success, all done
+echo '*** done'
+rm -f $seq.full
+status=0

tests/qemu-iotests/094.out

@@ -0,0 +1,11 @@
+QA output created by 094
+Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864
+Formatting 'TEST_DIR/source.IMGFMT', fmt=IMGFMT size=67108864
+{"return": {}}
+{"return": {}}
+{"timestamp": {"seconds":  TIMESTAMP, "microseconds":  TIMESTAMP}, "event": "BLOCK_JOB_READY", "data": {"device": "src", "len": 67108864, "offset": 67108864, "speed": 0, "type": "mirror"}}
+{"return": {}}
+{"timestamp": {"seconds":  TIMESTAMP, "microseconds":  TIMESTAMP}, "event": "BLOCK_JOB_COMPLETED", "data": {"device": "src", "len": 67108864, "offset": 67108864, "speed": 0, "type": "mirror"}}
+{"return": {}}
+{"timestamp": {"seconds":  TIMESTAMP, "microseconds":  TIMESTAMP}, "event": "SHUTDOWN"}
+*** done

tests/qemu-iotests/123

@@ -1,8 +1,8 @@
 #!/bin/bash
 #
-# Test I/O after EOF for growable images.
+# Test case for qemu-img convert to NBD
 #
-# Copyright (C) 2009 Red Hat, Inc.
+# Copyright (C) 2015 Red Hat, Inc.
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -19,18 +19,19 @@
 #
 
 # creator
-owner=hch@lst.de
+owner=mreitz@redhat.com
 
-seq=`basename $0`
+seq="$(basename $0)"
 echo "QA output created by $seq"
 
-here=`pwd`
+here="$PWD"
 tmp=/tmp/$$
 status=1	# failure is the default!
 
 _cleanup()
 {
-	_cleanup_test_img
+    _cleanup_test_img
+    rm -f "$SRC_IMG"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
@@ -39,35 +40,23 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 . ./common.filter
 
 _supported_fmt raw
-_supported_proto file sheepdog nfs
+_supported_proto nbd
 _supported_os Linux
 
+SRC_IMG="$TEST_DIR/source.$IMGFMT"
 
-# No -f, use probing for the protocol driver
+_make_test_img 1M
-QEMU_IO_PROTO="$QEMU_IO_PROG -g --cache $CACHEMODE"
+$QEMU_IMG create -f $IMGFMT "$SRC_IMG" 1M | _filter_img_create
 
-size=128M
+$QEMU_IO -c 'write -P 42 0 1M' "$SRC_IMG" | _filter_qemu_io
-_make_test_img $size
 
-echo
+$QEMU_IMG convert -n -f $IMGFMT -O raw "$SRC_IMG" "$TEST_IMG"
-echo "== reading at EOF =="
-$QEMU_IO_PROTO -c "read -P 0 $size 512" "$TEST_IMG" | _filter_qemu_io
 
-echo
+$QEMU_IO -c 'read -P 42 0 1M' "$TEST_IMG" | _filter_qemu_io
-echo "== reading far past EOF =="
-$QEMU_IO_PROTO -c "read -P 0 256M 512" "$TEST_IMG" | _filter_qemu_io
 
-echo
-echo "== writing at EOF =="
-$QEMU_IO_PROTO -c "write -P 66 $size 512" "$TEST_IMG" | _filter_qemu_io
-$QEMU_IO -c "read -P 66 $size 512" "$TEST_IMG" | _filter_qemu_io
-
-echo
-echo "== writing far past EOF =="
-$QEMU_IO_PROTO -c "write -P 66 256M 512" "$TEST_IMG" | _filter_qemu_io
-$QEMU_IO -c "read -P 66 256M 512" "$TEST_IMG" | _filter_qemu_io
 
 # success, all done
-echo "*** done"
+echo
+echo '*** done'
 rm -f $seq.full
 status=0

tests/qemu-iotests/123.out

@@ -0,0 +1,9 @@
+QA output created by 123
+Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=1048576
+Formatting 'TEST_DIR/source.IMGFMT', fmt=IMGFMT size=1048576
+wrote 1048576/1048576 bytes at offset 0
+1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 1048576/1048576 bytes at offset 0
+1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+*** done

tests/qemu-iotests/common.qemu

@@ -187,13 +187,23 @@ function _launch_qemu()
 
 
 # Silenty kills the QEMU process
+#
+# If $wait is set to anything other than the empty string, the process will not
+# be killed but only waited for, and any output will be forwarded to stdout. If
+# $wait is empty, the process will be killed and all output will be suppressed.
 function _cleanup_qemu()
 {
     # QEMU_PID[], QEMU_IN[], QEMU_OUT[] all use same indices
     for i in "${!QEMU_OUT[@]}"
     do
-        kill -KILL ${QEMU_PID[$i]} 2>/dev/null
+        if [ -z "${wait}" ]; then
+            kill -KILL ${QEMU_PID[$i]} 2>/dev/null
+        fi
         wait ${QEMU_PID[$i]} 2>/dev/null # silent kill
+        if [ -n "${wait}" ]; then
+            cat <&${QEMU_OUT[$i]} | _filter_testdir | _filter_qemu \
+                                  | _filter_qemu_io | _filter_qmp
+        fi
         rm -f "${QEMU_FIFO_IN}_${i}" "${QEMU_FIFO_OUT}_${i}"
         eval "exec ${QEMU_IN[$i]}<&-"   # close file descriptors
         eval "exec ${QEMU_OUT[$i]}<&-"

tests/qemu-iotests/group

@@ -22,7 +22,7 @@
 013 rw auto
 014 rw auto
 015 rw snapshot auto
-016 rw auto quick
+# 016 was removed, do not reuse
 017 rw backing auto quick
 018 rw backing auto quick
 019 rw backing auto quick
@@ -99,6 +99,8 @@
 090 rw auto quick
 091 rw auto
 092 rw auto quick
+093 auto
+094 rw auto quick
 095 rw auto quick
 097 rw auto backing
 098 rw auto backing quick
@@ -117,3 +119,4 @@
 113 rw auto quick
 114 rw auto quick
 116 rw auto quick
+123 rw auto quick

tests/qemu-iotests/iotests.py

@@ -21,8 +21,11 @@ import re
 import subprocess
 import string
 import unittest
-import sys; sys.path.append(os.path.join(os.path.dirname(__file__), '..', '..', 'scripts', 'qmp'))
+import sys
+sys.path.append(os.path.join(os.path.dirname(__file__), '..', '..', 'scripts'))
+sys.path.append(os.path.join(os.path.dirname(__file__), '..', '..', 'scripts', 'qmp'))
 import qmp
+import qtest
 import struct
 
 __all__ = ['imgfmt', 'imgproto', 'test_dir' 'qemu_img', 'qemu_io',
@@ -81,10 +84,12 @@ class VM(object):
     def __init__(self):
         self._monitor_path = os.path.join(test_dir, 'qemu-mon.%d' % os.getpid())
         self._qemu_log_path = os.path.join(test_dir, 'qemu-log.%d' % os.getpid())
+        self._qtest_path = os.path.join(test_dir, 'qemu-qtest.%d' % os.getpid())
         self._args = qemu_args + ['-chardev',
                      'socket,id=mon,path=' + self._monitor_path,
                      '-mon', 'chardev=mon,mode=control',
-                     '-qtest', 'stdio', '-machine', 'accel=qtest',
+                     '-qtest', 'unix:path=' + self._qtest_path,
+                     '-machine', 'accel=qtest',
                      '-display', 'none', '-vga', 'none']
         self._num_drives = 0
 
@@ -160,9 +165,11 @@ class VM(object):
         qemulog = open(self._qemu_log_path, 'wb')
         try:
             self._qmp = qmp.QEMUMonitorProtocol(self._monitor_path, server=True)
+            self._qtest = qtest.QEMUQtestProtocol(self._qtest_path, server=True)
             self._popen = subprocess.Popen(self._args, stdin=devnull, stdout=qemulog,
                                            stderr=subprocess.STDOUT)
             self._qmp.accept()
+            self._qtest.accept()
         except:
             os.remove(self._monitor_path)
             raise
@@ -173,18 +180,26 @@ class VM(object):
             self._qmp.cmd('quit')
             self._popen.wait()
             os.remove(self._monitor_path)
+            os.remove(self._qtest_path)
             os.remove(self._qemu_log_path)
             self._popen = None
 
     underscore_to_dash = string.maketrans('_', '-')
-    def qmp(self, cmd, **args):
+    def qmp(self, cmd, conv_keys=True, **args):
         '''Invoke a QMP command and return the result dict'''
         qmp_args = dict()
         for k in args.keys():
-            qmp_args[k.translate(self.underscore_to_dash)] = args[k]
+            if conv_keys:
+                qmp_args[k.translate(self.underscore_to_dash)] = args[k]
+            else:
+                qmp_args[k] = args[k]
 
         return self._qmp.cmd(cmd, args=qmp_args)
 
+    def qtest(self, cmd):
+        '''Send a qtest command to guest'''
+        return self._qtest.cmd(cmd)
+
     def get_qmp_event(self, wait=False):
         '''Poll for one queued QMP events and return it'''
         return self._qmp.pull_event(wait=wait)