Call nbd_co_send_extents() with correct length parameter
(extent.length may be smaller than original length).

Also, switch length parameter type to uint32_t, to correspond with
request->len and similar nbd_co_send_bitmap().

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
---
 nbd/server.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/nbd/server.c b/nbd/server.c
index e52b76bd1a..ea5fe0eb33 100644
--- a/nbd/server.c
+++ b/nbd/server.c
@@ -1910,7 +1910,7 @@ static int nbd_co_send_extents(NBDClient *client, uint64_t handle,
 /* Get block status from the exported device and send it to the client */
 static int nbd_co_send_block_status(NBDClient *client, uint64_t handle,
                                     BlockDriverState *bs, uint64_t offset,
-                                    uint64_t length, bool last,
+                                    uint32_t length, bool last,
                                     uint32_t context_id, Error **errp)
 {
     int ret;
@@ -1922,7 +1922,8 @@ static int nbd_co_send_block_status(NBDClient *client, uint64_t handle,
                 client, handle, -ret, "can't get block status", errp);
     }
 
-    return nbd_co_send_extents(client, handle, &extent, 1, length, last,
+    return nbd_co_send_extents(client, handle, &extent, 1,
+                               be32_to_cpu(extent.length), last,
                                context_id, errp);
 }
 
-- 
2.11.1

On 07/04/2018 06:23 AM, Vladimir Sementsov-Ogievskiy wrote:
> Call nbd_co_send_extents() with correct length parameter
> (extent.length may be smaller than original length).

Only matters for tracing purposes, but indeed worth an accurate trace.

> 
> Also, switch length parameter type to uint32_t, to correspond with
> request->len and similar nbd_co_send_bitmap().

No semantic impact.

> 
> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
> ---
>   nbd/server.c | 5 +++--
>   1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/nbd/server.c b/nbd/server.c
> index e52b76bd1a..ea5fe0eb33 100644
> --- a/nbd/server.c
> +++ b/nbd/server.c
> @@ -1910,7 +1910,7 @@ static int nbd_co_send_extents(NBDClient *client, uint64_t handle,
>   /* Get block status from the exported device and send it to the client */
>   static int nbd_co_send_block_status(NBDClient *client, uint64_t handle,
>                                       BlockDriverState *bs, uint64_t offset,
> -                                    uint64_t length, bool last,
> +                                    uint32_t length, bool last,
>                                       uint32_t context_id, Error **errp)
>   {
>       int ret;
> @@ -1922,7 +1922,8 @@ static int nbd_co_send_block_status(NBDClient *client, uint64_t handle,
>                   client, handle, -ret, "can't get block status", errp);
>       }
>   
> -    return nbd_co_send_extents(client, handle, &extent, 1, length, last,
> +    return nbd_co_send_extents(client, handle, &extent, 1,
> +                               be32_to_cpu(extent.length), last,

be32_to_cpu() is necessary to undo the fact that extent is already 
encoded for transmission over the wire.  Alternatively, 
blockstatus_to_extent_be() could be modified to return negative on 
error, or the total length encoded on success, or set the final length 
via a pointer parameter.  But the next patch will probably deal with 
that more directly.

This one is fine for 3.0,
Reviewed-by: Eric Blake <eblake@redhat.com>

-- 
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3266
Virtualization:  qemu.org | libvirt.org