Merge remote-tracking branch 'stefanha/block' into staging

* stefanha/block:
  rbd: add an asynchronous flush
  iotests: Add 'check -ssh' option to test Secure Shell block device.
  block: ssh: Use libssh2_sftp_fsync (if supported by libssh2) to flush to disk.
  block: Add support for Secure Shell (ssh) block device.
  ide: refuse WIN_READ_NATIVE_MAX on empty device
  qemu-iotests: filter QEMU_PROG in 051.out
  qemu-iotests: Add test for -drive options
  qemu-iotests: A few more bdrv_pread/pwrite tests
  block: Introduce bdrv_pwritev() for qcow2_save_vmstate
  savevm: Implement block_writev_buffer()
  block: Introduce bdrv_writev_vmstate

Conflicts:
	savevm.c

aliguori: add f->pos parameter to writev_buffer().

Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>

block.c

@@ -2131,27 +2131,21 @@ static void coroutine_fn bdrv_rw_co_entry(void *opaque)
 }
 
 /*
- * Process a synchronous request using coroutines
+ * Process a vectored synchronous request using coroutines
  */
-static int bdrv_rw_co(BlockDriverState *bs, int64_t sector_num, uint8_t *buf,
+static int bdrv_rwv_co(BlockDriverState *bs, int64_t sector_num,
-                      int nb_sectors, bool is_write)
+                       QEMUIOVector *qiov, bool is_write)
 {
-    QEMUIOVector qiov;
-    struct iovec iov = {
-        .iov_base = (void *)buf,
-        .iov_len = nb_sectors * BDRV_SECTOR_SIZE,
-    };
     Coroutine *co;
     RwCo rwco = {
         .bs = bs,
         .sector_num = sector_num,
-        .nb_sectors = nb_sectors,
+        .nb_sectors = qiov->size >> BDRV_SECTOR_BITS,
-        .qiov = &qiov,
+        .qiov = qiov,
         .is_write = is_write,
         .ret = NOT_DONE,
     };
-
+    assert((qiov->size & (BDRV_SECTOR_SIZE - 1)) == 0);
-    qemu_iovec_init_external(&qiov, &iov, 1);
 
     /**
      * In sync call context, when the vcpu is blocked, this throttling timer
@@ -2177,6 +2171,22 @@ static int bdrv_rw_co(BlockDriverState *bs, int64_t sector_num, uint8_t *buf,
     return rwco.ret;
 }
 
+/*
+ * Process a synchronous request using coroutines
+ */
+static int bdrv_rw_co(BlockDriverState *bs, int64_t sector_num, uint8_t *buf,
+                      int nb_sectors, bool is_write)
+{
+    QEMUIOVector qiov;
+    struct iovec iov = {
+        .iov_base = (void *)buf,
+        .iov_len = nb_sectors * BDRV_SECTOR_SIZE,
+    };
+
+    qemu_iovec_init_external(&qiov, &iov, 1);
+    return bdrv_rwv_co(bs, sector_num, &qiov, is_write);
+}
+
 /* return < 0 if error. See bdrv_write() for the return codes */
 int bdrv_read(BlockDriverState *bs, int64_t sector_num,
               uint8_t *buf, int nb_sectors)
@@ -2210,6 +2220,11 @@ int bdrv_write(BlockDriverState *bs, int64_t sector_num,
     return bdrv_rw_co(bs, sector_num, (uint8_t *)buf, nb_sectors, true);
 }
 
+int bdrv_writev(BlockDriverState *bs, int64_t sector_num, QEMUIOVector *qiov)
+{
+    return bdrv_rwv_co(bs, sector_num, qiov, true);
+}
+
 int bdrv_pread(BlockDriverState *bs, int64_t offset,
                void *buf, int count1)
 {
@@ -2255,15 +2270,15 @@ int bdrv_pread(BlockDriverState *bs, int64_t offset,
     return count1;
 }
 
-int bdrv_pwrite(BlockDriverState *bs, int64_t offset,
+int bdrv_pwritev(BlockDriverState *bs, int64_t offset, QEMUIOVector *qiov)
-                const void *buf, int count1)
 {
     uint8_t tmp_buf[BDRV_SECTOR_SIZE];
     int len, nb_sectors, count;
     int64_t sector_num;
     int ret;
 
-    count = count1;
+    count = qiov->size;
+
     /* first write to align to sector start */
     len = (BDRV_SECTOR_SIZE - offset) & (BDRV_SECTOR_SIZE - 1);
     if (len > count)
@@ -2272,24 +2287,32 @@ int bdrv_pwrite(BlockDriverState *bs, int64_t offset,
     if (len > 0) {
         if ((ret = bdrv_read(bs, sector_num, tmp_buf, 1)) < 0)
             return ret;
-        memcpy(tmp_buf + (offset & (BDRV_SECTOR_SIZE - 1)), buf, len);
+        qemu_iovec_to_buf(qiov, 0, tmp_buf + (offset & (BDRV_SECTOR_SIZE - 1)),
+                          len);
         if ((ret = bdrv_write(bs, sector_num, tmp_buf, 1)) < 0)
             return ret;
         count -= len;
         if (count == 0)
-            return count1;
+            return qiov->size;
         sector_num++;
-        buf += len;
     }
 
     /* write the sectors "in place" */
     nb_sectors = count >> BDRV_SECTOR_BITS;
     if (nb_sectors > 0) {
-        if ((ret = bdrv_write(bs, sector_num, buf, nb_sectors)) < 0)
+        QEMUIOVector qiov_inplace;
+
+        qemu_iovec_init(&qiov_inplace, qiov->niov);
+        qemu_iovec_concat(&qiov_inplace, qiov, len,
+                          nb_sectors << BDRV_SECTOR_BITS);
+        ret = bdrv_writev(bs, sector_num, &qiov_inplace);
+        qemu_iovec_destroy(&qiov_inplace);
+        if (ret < 0) {
             return ret;
+        }
+
         sector_num += nb_sectors;
         len = nb_sectors << BDRV_SECTOR_BITS;
-        buf += len;
         count -= len;
     }
 
@@ -2297,11 +2320,24 @@ int bdrv_pwrite(BlockDriverState *bs, int64_t offset,
     if (count > 0) {
         if ((ret = bdrv_read(bs, sector_num, tmp_buf, 1)) < 0)
             return ret;
-        memcpy(tmp_buf, buf, count);
+        qemu_iovec_to_buf(qiov, qiov->size - count, tmp_buf, count);
         if ((ret = bdrv_write(bs, sector_num, tmp_buf, 1)) < 0)
             return ret;
     }
-    return count1;
+    return qiov->size;
+}
+
+int bdrv_pwrite(BlockDriverState *bs, int64_t offset,
+                const void *buf, int count1)
+{
+    QEMUIOVector qiov;
+    struct iovec iov = {
+        .iov_base   = (void *) buf,
+        .iov_len    = count1,
+    };
+
+    qemu_iovec_init_external(&qiov, &iov, 1);
+    return bdrv_pwritev(bs, offset, &qiov);
 }
 
 /*
@@ -3183,14 +3219,29 @@ int bdrv_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
 
 int bdrv_save_vmstate(BlockDriverState *bs, const uint8_t *buf,
                       int64_t pos, int size)
+{
+    QEMUIOVector qiov;
+    struct iovec iov = {
+        .iov_base   = (void *) buf,
+        .iov_len    = size,
+    };
+
+    qemu_iovec_init_external(&qiov, &iov, 1);
+    return bdrv_writev_vmstate(bs, &qiov, pos);
+}
+
+int bdrv_writev_vmstate(BlockDriverState *bs, QEMUIOVector *qiov, int64_t pos)
 {
     BlockDriver *drv = bs->drv;
-    if (!drv)
+
+    if (!drv) {
         return -ENOMEDIUM;
-    if (drv->bdrv_save_vmstate)
+    } else if (drv->bdrv_save_vmstate) {
-        return drv->bdrv_save_vmstate(bs, buf, pos, size);
+        return drv->bdrv_save_vmstate(bs, qiov, pos);
-    if (bs->file)
+    } else if (bs->file) {
-        return bdrv_save_vmstate(bs->file, buf, pos, size);
+        return bdrv_writev_vmstate(bs->file, qiov, pos);
+    }
+
     return -ENOTSUP;
 }
 

block/Makefile.objs

@@ -13,6 +13,7 @@ block-obj-$(CONFIG_LIBISCSI) += iscsi.o
 block-obj-$(CONFIG_CURL) += curl.o
 block-obj-$(CONFIG_RBD) += rbd.o
 block-obj-$(CONFIG_GLUSTERFS) += gluster.o
+block-obj-$(CONFIG_LIBSSH2) += ssh.o
 endif
 
 common-obj-y += stream.o

block/qcow2.c

@@ -1652,8 +1652,8 @@ static void dump_refcounts(BlockDriverState *bs)
 }
 #endif
 
-static int qcow2_save_vmstate(BlockDriverState *bs, const uint8_t *buf,
+static int qcow2_save_vmstate(BlockDriverState *bs, QEMUIOVector *qiov,
-                              int64_t pos, int size)
+                              int64_t pos)
 {
     BDRVQcowState *s = bs->opaque;
     int growable = bs->growable;
@@ -1661,7 +1661,7 @@ static int qcow2_save_vmstate(BlockDriverState *bs, const uint8_t *buf,
 
     BLKDBG_EVENT(bs->file, BLKDBG_VMSTATE_SAVE);
     bs->growable = 1;
-    ret = bdrv_pwrite(bs, qcow2_vm_state_offset(s) + pos, buf, size);
+    ret = bdrv_pwritev(bs, qcow2_vm_state_offset(s) + pos, qiov);
     bs->growable = growable;
 
     return ret;

block/rbd.c

@@ -63,7 +63,8 @@
 typedef enum {
     RBD_AIO_READ,
     RBD_AIO_WRITE,
-    RBD_AIO_DISCARD
+    RBD_AIO_DISCARD,
+    RBD_AIO_FLUSH
 } RBDAIOCmd;
 
 typedef struct RBDAIOCB {
@@ -379,8 +380,7 @@ static void qemu_rbd_complete_aio(RADOSCB *rcb)
 
     r = rcb->ret;
 
-    if (acb->cmd == RBD_AIO_WRITE ||
+    if (acb->cmd != RBD_AIO_READ) {
-        acb->cmd == RBD_AIO_DISCARD) {
         if (r < 0) {
             acb->ret = r;
             acb->error = 1;
@@ -659,6 +659,16 @@ static int rbd_aio_discard_wrapper(rbd_image_t image,
 #endif
 }
 
+static int rbd_aio_flush_wrapper(rbd_image_t image,
+                                 rbd_completion_t comp)
+{
+#ifdef LIBRBD_SUPPORTS_AIO_FLUSH
+    return rbd_aio_flush(image, comp);
+#else
+    return -ENOTSUP;
+#endif
+}
+
 static BlockDriverAIOCB *rbd_start_aio(BlockDriverState *bs,
                                        int64_t sector_num,
                                        QEMUIOVector *qiov,
@@ -679,7 +689,7 @@ static BlockDriverAIOCB *rbd_start_aio(BlockDriverState *bs,
     acb = qemu_aio_get(&rbd_aiocb_info, bs, cb, opaque);
     acb->cmd = cmd;
     acb->qiov = qiov;
-    if (cmd == RBD_AIO_DISCARD) {
+    if (cmd == RBD_AIO_DISCARD || cmd == RBD_AIO_FLUSH) {
         acb->bounce = NULL;
     } else {
         acb->bounce = qemu_blockalign(bs, qiov->size);
@@ -723,6 +733,9 @@ static BlockDriverAIOCB *rbd_start_aio(BlockDriverState *bs,
     case RBD_AIO_DISCARD:
         r = rbd_aio_discard_wrapper(s->image, off, size, c);
         break;
+    case RBD_AIO_FLUSH:
+        r = rbd_aio_flush_wrapper(s->image, c);
+        break;
     default:
         r = -EINVAL;
     }
@@ -762,6 +775,16 @@ static BlockDriverAIOCB *qemu_rbd_aio_writev(BlockDriverState *bs,
                          RBD_AIO_WRITE);
 }
 
+#ifdef LIBRBD_SUPPORTS_AIO_FLUSH
+static BlockDriverAIOCB *qemu_rbd_aio_flush(BlockDriverState *bs,
+                                            BlockDriverCompletionFunc *cb,
+                                            void *opaque)
+{
+    return rbd_start_aio(bs, 0, NULL, 0, cb, opaque, RBD_AIO_FLUSH);
+}
+
+#else
+
 static int qemu_rbd_co_flush(BlockDriverState *bs)
 {
 #if LIBRBD_VERSION_CODE >= LIBRBD_VERSION(0, 1, 1)
@@ -772,6 +795,7 @@ static int qemu_rbd_co_flush(BlockDriverState *bs)
     return 0;
 #endif
 }
+#endif
 
 static int qemu_rbd_getinfo(BlockDriverState *bs, BlockDriverInfo *bdi)
 {
@@ -949,7 +973,12 @@ static BlockDriver bdrv_rbd = {
 
     .bdrv_aio_readv         = qemu_rbd_aio_readv,
     .bdrv_aio_writev        = qemu_rbd_aio_writev,
+
+#ifdef LIBRBD_SUPPORTS_AIO_FLUSH
+    .bdrv_aio_flush         = qemu_rbd_aio_flush,
+#else
     .bdrv_co_flush_to_disk  = qemu_rbd_co_flush,
+#endif
 
 #ifdef LIBRBD_SUPPORTS_DISCARD
     .bdrv_aio_discard       = qemu_rbd_aio_discard,

block/sheepdog.c

@@ -2054,12 +2054,19 @@ cleanup:
     return ret;
 }
 
-static int sd_save_vmstate(BlockDriverState *bs, const uint8_t *data,
+static int sd_save_vmstate(BlockDriverState *bs, QEMUIOVector *qiov,
-                           int64_t pos, int size)
+                           int64_t pos)
 {
     BDRVSheepdogState *s = bs->opaque;
+    void *buf;
+    int ret;
 
-    return do_load_save_vmstate(s, (uint8_t *)data, pos, size, 0);
+    buf = qemu_blockalign(bs, qiov->size);
+    qemu_iovec_to_buf(qiov, 0, buf, qiov->size);
+    ret = do_load_save_vmstate(s, (uint8_t *) buf, pos, qiov->size, 0);
+    qemu_vfree(buf);
+
+    return ret;
 }
 
 static int sd_load_vmstate(BlockDriverState *bs, uint8_t *data,

configure

@@ -239,6 +239,7 @@ virtio_blk_data_plane=""
 gtk=""
 gtkabi="2.0"
 tpm="no"
+libssh2=""
 
 # parse CC options first
 for opt do
@@ -922,6 +923,10 @@ for opt do
   ;;
   --enable-tpm) tpm="yes"
   ;;
+  --disable-libssh2) libssh2="no"
+  ;;
+  --enable-libssh2) libssh2="yes"
+  ;;
   *) echo "ERROR: unknown option $opt"; show_help="yes"
   ;;
   esac
@@ -1182,6 +1187,8 @@ echo "  --disable-glusterfs      disable GlusterFS backend"
 echo "  --enable-gcov            enable test coverage analysis with gcov"
 echo "  --gcov=GCOV              use specified gcov [$gcov_tool]"
 echo "  --enable-tpm             enable TPM support"
+echo "  --disable-libssh2        disable ssh block device support"
+echo "  --enable-libssh2         enable ssh block device support"
 echo ""
 echo "NOTE: The object files are built at the place where configure is launched"
 exit 1
@@ -2313,6 +2320,67 @@ EOF
   fi
 fi
 
+##########################################
+# libssh2 probe
+if test "$libssh2" != "no" ; then
+  cat > $TMPC <<EOF
+#include <stdio.h>
+#include <libssh2.h>
+#include <libssh2_sftp.h>
+int main(void) {
+    LIBSSH2_SESSION *session;
+    session = libssh2_session_init ();
+    (void) libssh2_sftp_init (session);
+    return 0;
+}
+EOF
+
+  if $pkg_config libssh2 --modversion >/dev/null 2>&1; then
+    libssh2_cflags=`$pkg_config libssh2 --cflags`
+    libssh2_libs=`$pkg_config libssh2 --libs`
+  else
+    libssh2_cflags=
+    libssh2_libs="-lssh2"
+  fi
+
+  if compile_prog "$libssh2_cflags" "$libssh2_libs" ; then
+    libssh2=yes
+    libs_tools="$libssh2_libs $libs_tools"
+    libs_softmmu="$libssh2_libs $libs_softmmu"
+    QEMU_CFLAGS="$QEMU_CFLAGS $libssh2_cflags"
+  else
+    if test "$libssh2" = "yes" ; then
+      feature_not_found "libssh2"
+    fi
+    libssh2=no
+  fi
+fi
+
+##########################################
+# libssh2_sftp_fsync probe
+
+if test "$libssh2" = "yes"; then
+  cat > $TMPC <<EOF
+#include <stdio.h>
+#include <libssh2.h>
+#include <libssh2_sftp.h>
+int main(void) {
+    LIBSSH2_SESSION *session;
+    LIBSSH2_SFTP *sftp;
+    LIBSSH2_SFTP_HANDLE *sftp_handle;
+    session = libssh2_session_init ();
+    sftp = libssh2_sftp_init (session);
+    sftp_handle = libssh2_sftp_open (sftp, "/", 0, 0);
+    libssh2_sftp_fsync (sftp_handle);
+    return 0;
+}
+EOF
+  # libssh2_cflags/libssh2_libs defined in previous test.
+  if compile_prog "$libssh2_cflags" "$libssh2_libs" ; then
+    QEMU_CFLAGS="-DHAS_LIBSSH2_SFTP_FSYNC $QEMU_CFLAGS"
+  fi
+fi
+
 ##########################################
 # linux-aio probe
 
@@ -3435,6 +3503,7 @@ echo "virtio-blk-data-plane $virtio_blk_data_plane"
 echo "gcov              $gcov_tool"
 echo "gcov enabled      $gcov"
 echo "TPM support       $tpm"
+echo "libssh2 support   $libssh2"
 
 if test "$sdl_too_old" = "yes"; then
 echo "-> Your SDL version is too old - please upgrade to have SDL support"
@@ -3793,6 +3862,10 @@ if test "$glusterfs" = "yes" ; then
   echo "CONFIG_GLUSTERFS=y" >> $config_host_mak
 fi
 
+if test "$libssh2" = "yes" ; then
+  echo "CONFIG_LIBSSH2=y" >> $config_host_mak
+fi
+
 if test "$virtio_blk_data_plane" = "yes" ; then
   echo 'CONFIG_VIRTIO_BLK_DATA_PLANE=$(CONFIG_VIRTIO)' >> $config_host_mak
 fi

hw/ide/core.c

@@ -1262,6 +1262,10 @@ void ide_exec_cmd(IDEBus *bus, uint32_t val)
         lba48 = 1;
         /* fall through */
     case WIN_READ_NATIVE_MAX:
+        /* Refuse if no sectors are addressable (e.g. medium not inserted) */
+        if (s->nb_sectors == 0) {
+            goto abort_cmd;
+        }
 	ide_cmd_lba48_transform(s, lba48);
         ide_set_sector(s, s->nb_sectors - 1);
         s->status = READY_STAT | SEEK_STAT;

include/block/block.h

@@ -166,10 +166,12 @@ int bdrv_read_unthrottled(BlockDriverState *bs, int64_t sector_num,
                           uint8_t *buf, int nb_sectors);
 int bdrv_write(BlockDriverState *bs, int64_t sector_num,
                const uint8_t *buf, int nb_sectors);
+int bdrv_writev(BlockDriverState *bs, int64_t sector_num, QEMUIOVector *qiov);
 int bdrv_pread(BlockDriverState *bs, int64_t offset,
                void *buf, int count);
 int bdrv_pwrite(BlockDriverState *bs, int64_t offset,
                 const void *buf, int count);
+int bdrv_pwritev(BlockDriverState *bs, int64_t offset, QEMUIOVector *qiov);
 int bdrv_pwrite_sync(BlockDriverState *bs, int64_t offset,
     const void *buf, int count);
 int coroutine_fn bdrv_co_readv(BlockDriverState *bs, int64_t sector_num,
@@ -348,6 +350,7 @@ void path_combine(char *dest, int dest_size,
                   const char *base_path,
                   const char *filename);
 
+int bdrv_writev_vmstate(BlockDriverState *bs, QEMUIOVector *qiov, int64_t pos);
 int bdrv_save_vmstate(BlockDriverState *bs, const uint8_t *buf,
                       int64_t pos, int size);
 

include/block/block_int.h

@@ -164,8 +164,8 @@ struct BlockDriver {
                                   const char *snapshot_name);
     int (*bdrv_get_info)(BlockDriverState *bs, BlockDriverInfo *bdi);
 
-    int (*bdrv_save_vmstate)(BlockDriverState *bs, const uint8_t *buf,
+    int (*bdrv_save_vmstate)(BlockDriverState *bs, QEMUIOVector *qiov,
-                             int64_t pos, int size);
+                             int64_t pos);
     int (*bdrv_load_vmstate)(BlockDriverState *bs, uint8_t *buf,
                              int64_t pos, int size);
 

include/migration/qemu-file.h

@@ -55,7 +55,7 @@ typedef int (QEMUFileGetFD)(void *opaque);
  * This function writes an iovec to file.
  */
 typedef ssize_t (QEMUFileWritevBufferFunc)(void *opaque, struct iovec *iov,
-                                           int iovcnt);
+                                           int iovcnt, int64_t pos);
 
 typedef struct QEMUFileOps {
     QEMUFilePutBufferFunc *put_buffer;

qemu-doc.texi

@@ -423,6 +423,7 @@ snapshots.
 * disk_images_sheepdog::      Sheepdog disk images
 * disk_images_iscsi::         iSCSI LUNs
 * disk_images_gluster::       GlusterFS disk images
+* disk_images_ssh::           Secure Shell (ssh) disk images
 @end menu
 
 @node disk_images_quickstart
@@ -1038,6 +1039,59 @@ qemu-system-x86_64 -drive file=gluster+unix:///testvol/dir/a.img?socket=/tmp/glu
 qemu-system-x86_64 -drive file=gluster+rdma://1.2.3.4:24007/testvol/a.img
 @end example
 
+@node disk_images_ssh
+@subsection Secure Shell (ssh) disk images
+
+You can access disk images located on a remote ssh server
+by using the ssh protocol:
+
+@example
+qemu-system-x86_64 -drive file=ssh://[@var{user}@@]@var{server}[:@var{port}]/@var{path}[?host_key_check=@var{host_key_check}]
+@end example
+
+Alternative syntax using properties:
+
+@example
+qemu-system-x86_64 -drive file.driver=ssh[,file.user=@var{user}],file.host=@var{server}[,file.port=@var{port}],file.path=@var{path}[,file.host_key_check=@var{host_key_check}]
+@end example
+
+@var{ssh} is the protocol.
+
+@var{user} is the remote user.  If not specified, then the local
+username is tried.
+
+@var{server} specifies the remote ssh server.  Any ssh server can be
+used, but it must implement the sftp-server protocol.  Most Unix/Linux
+systems should work without requiring any extra configuration.
+
+@var{port} is the port number on which sshd is listening.  By default
+the standard ssh port (22) is used.
+
+@var{path} is the path to the disk image.
+
+The optional @var{host_key_check} parameter controls how the remote
+host's key is checked.  The default is @code{yes} which means to use
+the local @file{.ssh/known_hosts} file.  Setting this to @code{no}
+turns off known-hosts checking.  Or you can check that the host key
+matches a specific fingerprint:
+@code{host_key_check=md5:78:45:8e:14:57:4f:d5:45:83:0a:0e:f3:49:82:c9:c8}
+(@code{sha1:} can also be used as a prefix, but note that OpenSSH
+tools only use MD5 to print fingerprints).
+
+Currently authentication must be done using ssh-agent.  Other
+authentication methods may be supported in future.
+
+Note: Many ssh servers do not support an @code{fsync}-style operation.
+The ssh driver cannot guarantee that disk flush requests are
+obeyed, and this causes a risk of disk corruption if the remote
+server or network goes down during writes.  The driver will
+print a warning when @code{fsync} is not supported:
+
+warning: ssh server @code{ssh.example.com:22} does not support fsync
+
+With sufficiently new versions of libssh2 and OpenSSH, @code{fsync} is
+supported.
+
 @node pcsys_network
 @section Network emulation
 

qemu-options.hx

@@ -2107,6 +2107,18 @@ Example for Unix Domain Sockets
 qemu-system-i386 --drive file=nbd:unix:/tmp/nbd-socket
 @end example
 
+@item SSH
+QEMU supports SSH (Secure Shell) access to remote disks.
+
+Examples:
+@example
+qemu-system-i386 -drive file=ssh://user@@host/path/to/disk.img
+qemu-system-i386 -drive file.driver=ssh,file.user=user,file.host=host,file.port=22,file.path=/path/to/disk.img
+@end example
+
+Currently authentication must be done using ssh-agent.  Other
+authentication methods may be supported in future.
+
 @item Sheepdog
 Sheepdog is a distributed storage system for QEMU.
 QEMU supports using either local sheepdog devices or remote networked

savevm.c

@@ -175,7 +175,8 @@ static void coroutine_fn yield_until_fd_readable(int fd)
     qemu_coroutine_yield();
 }
 
-static ssize_t socket_writev_buffer(void *opaque, struct iovec *iov, int iovcnt)
+static ssize_t socket_writev_buffer(void *opaque, struct iovec *iov, int iovcnt,
+                                    int64_t pos)
 {
     QEMUFileSocket *s = opaque;
     ssize_t len;
@@ -457,6 +458,21 @@ fail:
     return NULL;
 }
 
+static ssize_t block_writev_buffer(void *opaque, struct iovec *iov, int iovcnt,
+                                   int64_t pos)
+{
+    int ret;
+    QEMUIOVector qiov;
+
+    qemu_iovec_init_external(&qiov, iov, iovcnt);
+    ret = bdrv_writev_vmstate(opaque, &qiov, pos);
+    if (ret < 0) {
+        return ret;
+    }
+
+    return qiov.size;
+}
+
 static int block_put_buffer(void *opaque, const uint8_t *buf,
                            int64_t pos, int size)
 {
@@ -481,6 +497,7 @@ static const QEMUFileOps bdrv_read_ops = {
 
 static const QEMUFileOps bdrv_write_ops = {
     .put_buffer     = block_put_buffer,
+    .writev_buffer  = block_writev_buffer,
     .close          = bdrv_fclose
 };
 
@@ -535,7 +552,7 @@ static void qemu_fflush(QEMUFile *f)
 
     if (f->ops->writev_buffer) {
         if (f->iovcnt > 0) {
-            ret = f->ops->writev_buffer(f->opaque, f->iov, f->iovcnt);
+            ret = f->ops->writev_buffer(f->opaque, f->iov, f->iovcnt, f->pos);
         }
     } else {
         if (f->buf_index > 0) {

tests/qemu-iotests/002

@@ -61,10 +61,23 @@ $QEMU_IO -c "read -pP 0xa 0 $size" $TEST_IMG | _filter_qemu_io
 echo
 echo "unaligned pwrite"
 $QEMU_IO -c 'write -pP 0xab 66 42' $TEST_IMG | _filter_qemu_io
+$QEMU_IO -c 'write -pP 0xac 512 288' $TEST_IMG | _filter_qemu_io
+$QEMU_IO -c 'write -pP 0xad 800 224' $TEST_IMG | _filter_qemu_io
+$QEMU_IO -c 'write -pP 0xae 66000 128k' $TEST_IMG | _filter_qemu_io
+$QEMU_IO -c 'write -pP 0xaf 256k 42' $TEST_IMG | _filter_qemu_io
 
 echo
 echo "verify pattern"
+$QEMU_IO -c 'read -pP 0xa 0 66' $TEST_IMG | _filter_qemu_io
 $QEMU_IO -c 'read -pP 0xab 66 42' $TEST_IMG | _filter_qemu_io
+$QEMU_IO -c 'read -pP 0xa 108 404' $TEST_IMG | _filter_qemu_io
+$QEMU_IO -c 'read -pP 0xac 512 288' $TEST_IMG | _filter_qemu_io
+$QEMU_IO -c 'read -pP 0xad 800 224' $TEST_IMG | _filter_qemu_io
+$QEMU_IO -c 'read -pP 0xa 1k 64976' $TEST_IMG | _filter_qemu_io
+$QEMU_IO -c 'read -pP 0xae 66000 128k' $TEST_IMG | _filter_qemu_io
+$QEMU_IO -c 'read -pP 0xa 197072 65072' $TEST_IMG | _filter_qemu_io
+$QEMU_IO -c 'read -pP 0xaf 256k 42' $TEST_IMG | _filter_qemu_io
+$QEMU_IO -c 'read -pP 0xa 262186 470' $TEST_IMG | _filter_qemu_io
 
 # success, all done
 echo "*** done"

tests/qemu-iotests/002.out

@@ -16,8 +16,34 @@ read 134217728/134217728 bytes at offset 0
 unaligned pwrite
 wrote 42/42 bytes at offset 66
 42 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 288/288 bytes at offset 512
+288 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 224/224 bytes at offset 800
+224 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 131072/131072 bytes at offset 66000
+128 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 42/42 bytes at offset 262144
+42 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 
 verify pattern
+read 66/66 bytes at offset 0
+66 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 read 42/42 bytes at offset 66
 42 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 404/404 bytes at offset 108
+404 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 288/288 bytes at offset 512
+288 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 224/224 bytes at offset 800
+224 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 64976/64976 bytes at offset 1024
+63.453 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 131072/131072 bytes at offset 66000
+128 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 65072/65072 bytes at offset 197072
+63.547 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 42/42 bytes at offset 262144
+42 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 470/470 bytes at offset 262186
+470 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 *** done

tests/qemu-iotests/051

@@ -0,0 +1,148 @@
+#!/bin/bash
+#
+# Test command line configuration of block devices and driver-specific options
+#
+# Copyright (C) 2013 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=kwolf@redhat.com
+
+seq=`basename $0`
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1	# failure is the default!
+
+_cleanup()
+{
+	_cleanup_test_img
+}
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ./common.rc
+. ./common.filter
+
+_supported_fmt qcow2
+_supported_proto file
+_supported_os Linux
+
+function do_run_qemu()
+{
+    echo Testing: "$@"
+    echo quit | $QEMU -nographic -monitor stdio -serial none "$@"
+    echo
+}
+
+function run_qemu()
+{
+    do_run_qemu "$@" 2>&1 | _filter_testdir | _filter_qemu
+}
+
+size=128M
+
+_make_test_img $size
+
+echo
+echo === Unknown option ===
+echo
+
+run_qemu -drive file=$TEST_IMG,format=qcow2,unknown_opt=
+run_qemu -drive file=$TEST_IMG,format=qcow2,unknown_opt=on
+run_qemu -drive file=$TEST_IMG,format=qcow2,unknown_opt=1234
+run_qemu -drive file=$TEST_IMG,format=qcow2,unknown_opt=foo
+
+
+echo
+echo === Enable and disable lazy refcounting on the command line, plus some invalid values ===
+echo
+
+run_qemu -drive file=$TEST_IMG,format=qcow2,lazy_refcounts=on
+run_qemu -drive file=$TEST_IMG,format=qcow2,lazy_refcounts=off
+run_qemu -drive file=$TEST_IMG,format=qcow2,lazy_refcounts=
+run_qemu -drive file=$TEST_IMG,format=qcow2,lazy_refcounts=42
+run_qemu -drive file=$TEST_IMG,format=qcow2,lazy_refcounts=foo
+
+
+echo
+echo === With version 2 images enabling lazy refcounts must fail ===
+echo
+
+_make_test_img -ocompat=0.10 $size
+
+run_qemu -drive file=$TEST_IMG,format=qcow2,lazy_refcounts=on
+run_qemu -drive file=$TEST_IMG,format=qcow2,lazy_refcounts=off
+
+echo
+echo === No medium ===
+echo
+
+run_qemu -drive if=floppy
+run_qemu -drive if=ide,media=cdrom
+run_qemu -drive if=scsi,media=cdrom
+
+run_qemu -drive if=ide
+run_qemu -drive if=virtio
+run_qemu -drive if=scsi
+
+run_qemu -drive if=none,id=disk -device ide-cd,drive=disk
+run_qemu -drive if=none,id=disk -device lsi53c895a -device scsi-cd,drive=disk
+
+run_qemu -drive if=none,id=disk -device ide-drive,drive=disk
+run_qemu -drive if=none,id=disk -device ide-hd,drive=disk
+run_qemu -drive if=none,id=disk -device lsi53c895a -device scsi-disk,drive=disk
+run_qemu -drive if=none,id=disk -device lsi53c895a -device scsi-hd,drive=disk
+
+echo
+echo === Read-only ===
+echo
+
+run_qemu -drive file=$TEST_IMG,if=floppy,readonly=on
+run_qemu -drive file=$TEST_IMG,if=ide,media=cdrom,readonly=on
+run_qemu -drive file=$TEST_IMG,if=scsi,media=cdrom,readonly=on
+
+run_qemu -drive file=$TEST_IMG,if=ide,readonly=on
+run_qemu -drive file=$TEST_IMG,if=virtio,readonly=on
+run_qemu -drive file=$TEST_IMG,if=scsi,readonly=on
+
+run_qemu -drive file=$TEST_IMG,if=none,id=disk,readonly=on -device ide-cd,drive=disk
+run_qemu -drive file=$TEST_IMG,if=none,id=disk,readonly=on -device lsi53c895a -device scsi-cd,drive=disk
+
+run_qemu -drive file=$TEST_IMG,if=none,id=disk,readonly=on -device ide-drive,drive=disk
+run_qemu -drive file=$TEST_IMG,if=none,id=disk,readonly=on -device ide-hd,drive=disk
+run_qemu -drive file=$TEST_IMG,if=none,id=disk,readonly=on -device lsi53c895a -device scsi-disk,drive=disk
+run_qemu -drive file=$TEST_IMG,if=none,id=disk,readonly=on -device lsi53c895a -device scsi-hd,drive=disk
+
+echo
+echo === Cache modes ===
+echo
+
+# Cannot use the test image because cache=none might not work on the host FS
+# Use cdrom so that we won't get errors about missing media
+
+run_qemu -drive media=cdrom,cache=none
+run_qemu -drive media=cdrom,cache=directsync
+run_qemu -drive media=cdrom,cache=writeback
+run_qemu -drive media=cdrom,cache=writethrough
+run_qemu -drive media=cdrom,cache=unsafe
+run_qemu -drive media=cdrom,cache=invalid_value
+
+# success, all done
+echo "*** done"
+rm -f $seq.full
+status=0

tests/qemu-iotests/051.out

@@ -0,0 +1,162 @@
+QA output created by 051
+Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728 
+
+=== 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=: 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=: could not open disk image TEST_DIR/t.qcow2: Invalid argument
+
+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: 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: could not open disk image TEST_DIR/t.qcow2: Invalid argument
+
+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: 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: could not open disk image TEST_DIR/t.qcow2: Invalid argument
+
+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: 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: could not open disk image TEST_DIR/t.qcow2: Invalid argument
+
+
+=== Enable and disable lazy refcounting on the command line, plus some invalid values ===
+
+Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy_refcounts=on
+qququiquit
+
+Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy_refcounts=off
+qququiquit
+
+Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy_refcounts=
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy_refcounts=: Parameter 'lazy_refcounts' expects 'on' or 'off'
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy_refcounts=: could not open disk image TEST_DIR/t.qcow2: Invalid argument
+
+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: Parameter 'lazy_refcounts' expects 'on' or 'off'
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy_refcounts=42: could not open disk image TEST_DIR/t.qcow2: Invalid argument
+
+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: Parameter 'lazy_refcounts' expects 'on' or 'off'
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy_refcounts=foo: could not open disk image TEST_DIR/t.qcow2: Invalid argument
+
+
+=== 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: 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: could not open disk image TEST_DIR/t.qcow2: Invalid argument
+
+Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy_refcounts=off
+qququiquit
+
+
+=== No medium ===
+
+Testing: -drive if=floppy
+qququiquit
+
+Testing: -drive if=ide,media=cdrom
+qququiquit
+
+Testing: -drive if=scsi,media=cdrom
+qququiquit
+
+Testing: -drive if=ide
+QEMU_PROG: Device needs media, but drive is empty
+QEMU_PROG: Initialization of device ide-hd failed
+
+Testing: -drive if=virtio
+QEMU_PROG: -drive if=virtio: Device needs media, but drive is empty
+QEMU_PROG: -drive if=virtio: Device 'virtio-blk-pci' could not be initialized
+
+Testing: -drive if=scsi
+QEMU_PROG: -drive if=scsi: Device needs media, but drive is empty
+QEMU_PROG: Initialization of device lsi53c895a failed
+
+Testing: -drive if=none,id=disk -device ide-cd,drive=disk
+qququiquit
+
+Testing: -drive if=none,id=disk -device lsi53c895a -device scsi-cd,drive=disk
+qququiquit
+
+Testing: -drive if=none,id=disk -device ide-drive,drive=disk
+QEMU_PROG: -device ide-drive,drive=disk: Device needs media, but drive is empty
+QEMU_PROG: -device ide-drive,drive=disk: Device 'ide-drive' could not be initialized
+
+Testing: -drive if=none,id=disk -device ide-hd,drive=disk
+QEMU_PROG: -device ide-hd,drive=disk: Device needs media, but drive is empty
+QEMU_PROG: -device ide-hd,drive=disk: Device 'ide-hd' could not be initialized
+
+Testing: -drive if=none,id=disk -device lsi53c895a -device scsi-disk,drive=disk
+QEMU_PROG: -device scsi-disk,drive=disk: Device needs media, but drive is empty
+QEMU_PROG: -device scsi-disk,drive=disk: Device 'scsi-disk' could not be initialized
+
+Testing: -drive if=none,id=disk -device lsi53c895a -device scsi-hd,drive=disk
+QEMU_PROG: -device scsi-hd,drive=disk: Device needs media, but drive is empty
+QEMU_PROG: -device scsi-hd,drive=disk: Device 'scsi-hd' could not be initialized
+
+
+=== Read-only ===
+
+Testing: -drive file=TEST_DIR/t.qcow2,if=floppy,readonly=on
+qququiquit
+
+Testing: -drive file=TEST_DIR/t.qcow2,if=ide,media=cdrom,readonly=on
+qququiquit
+
+Testing: -drive file=TEST_DIR/t.qcow2,if=scsi,media=cdrom,readonly=on
+qququiquit
+
+Testing: -drive file=TEST_DIR/t.qcow2,if=ide,readonly=on
+QEMU_PROG: -drive file=TEST_DIR/t.qcow2,if=ide,readonly=on: readonly not supported by this bus type
+
+Testing: -drive file=TEST_DIR/t.qcow2,if=virtio,readonly=on
+qququiquit
+
+Testing: -drive file=TEST_DIR/t.qcow2,if=scsi,readonly=on
+qququiquit
+
+Testing: -drive file=TEST_DIR/t.qcow2,if=none,id=disk,readonly=on -device ide-cd,drive=disk
+qququiquit
+
+Testing: -drive file=TEST_DIR/t.qcow2,if=none,id=disk,readonly=on -device lsi53c895a -device scsi-cd,drive=disk
+qququiquit
+
+Testing: -drive file=TEST_DIR/t.qcow2,if=none,id=disk,readonly=on -device ide-drive,drive=disk
+QEMU_PROG: -device ide-drive,drive=disk: Can't use a read-only drive
+QEMU_PROG: -device ide-drive,drive=disk: Device 'ide-drive' could not be initialized
+
+Testing: -drive file=TEST_DIR/t.qcow2,if=none,id=disk,readonly=on -device ide-hd,drive=disk
+QEMU_PROG: -device ide-hd,drive=disk: Can't use a read-only drive
+QEMU_PROG: -device ide-hd,drive=disk: Device 'ide-hd' could not be initialized
+
+Testing: -drive file=TEST_DIR/t.qcow2,if=none,id=disk,readonly=on -device lsi53c895a -device scsi-disk,drive=disk
+qququiquit
+
+Testing: -drive file=TEST_DIR/t.qcow2,if=none,id=disk,readonly=on -device lsi53c895a -device scsi-hd,drive=disk
+qququiquit
+
+
+=== Cache modes ===
+
+Testing: -drive media=cdrom,cache=none
+qququiquit
+
+Testing: -drive media=cdrom,cache=directsync
+qququiquit
+
+Testing: -drive media=cdrom,cache=writeback
+qququiquit
+
+Testing: -drive media=cdrom,cache=writethrough
+qququiquit
+
+Testing: -drive media=cdrom,cache=unsafe
+qququiquit
+
+Testing: -drive media=cdrom,cache=invalid_value
+QEMU_PROG: -drive media=cdrom,cache=invalid_value: invalid cache option
+
+*** done

tests/qemu-iotests/common

@@ -137,6 +137,7 @@ check options
     -rbd                test rbd
     -sheepdog           test sheepdog
     -nbd                test nbd
+    -ssh                test ssh
     -xdiff		graphical mode diff
     -nocache		use O_DIRECT on backing file
     -misalign		misalign memory allocations
@@ -206,6 +207,10 @@ testlist options
 	    IMGPROTO=nbd
 	    xpand=false
 	    ;;
+        -ssh)
+            IMGPROTO=ssh
+            xpand=false
+            ;;
 	-nocache)
 	    QEMU_IO_OPTIONS="$QEMU_IO_OPTIONS --nocache"
 	    xpand=false

tests/qemu-iotests/common.filter

@@ -152,5 +152,11 @@ _filter_qemu_io()
     _filter_win32 | sed -e "s/[0-9]* ops\; [0-9/:. sec]* ([0-9/.inf]* [EPTGMKiBbytes]*\/sec and [0-9/.inf]* ops\/sec)/X ops\; XX:XX:XX.X (XXX YYY\/sec and XXX ops\/sec)/"
 }
 
+# replace occurrences of QEMU_PROG with "qemu"
+_filter_qemu()
+{
+    sed -e "s#$(basename $QEMU_PROG)#QEMU_PROG#g"
+}
+
 # make sure this script returns success
 /bin/true

tests/qemu-iotests/common.rc

@@ -52,6 +52,9 @@ if [ "$IMGPROTO" = "file" ]; then
 elif [ "$IMGPROTO" = "nbd" ]; then
     TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
     TEST_IMG="nbd:127.0.0.1:10810"
+elif [ "$IMGPROTO" = "ssh" ]; then
+    TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
+    TEST_IMG="ssh://127.0.0.1$TEST_IMG_FILE"
 else
     TEST_IMG=$IMGPROTO:$TEST_DIR/t.$IMGFMT
 fi

tests/qemu-iotests/group

@@ -57,4 +57,5 @@
 048 img auto quick
 049 rw auto
 050 rw auto backing quick
+051 rw auto
 052 rw auto backing