Re: [Qemu-devel] [PATCH v10 1/3] Add new block driver interface to add/delete a BDS's child

[Max Reitz <mreitz@redhat.com>]

[-- Attachment #1.1: Type: text/plain, Size: 5728 bytes --]

Sorry that I wasn't so pedantic last time; or maybe I should rather be
sorry that I'm so pedantic this time.

On 16.02.2016 10:37, Changlong Xie wrote:
> From: Wen Congyang <wency@cn.fujitsu.com>
> 
> In some cases, we want to take a quorum child offline, and take
> another child online.
> 
> Signed-off-by: Wen Congyang <wency@cn.fujitsu.com>
> Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
> Signed-off-by: Gonglei <arei.gonglei@huawei.com>
> Signed-off-by: Changlong Xie <xiecl.fnst@cn.fujitsu.com>
> Reviewed-by: Eric Blake <eblake@redhat.com>
> Reviewed-by: Alberto Garcia <berto@igalia.com>
> ---
>  block.c                   | 50 +++++++++++++++++++++++++++++++++++++++++++++++
>  include/block/block.h     |  5 +++++
>  include/block/block_int.h |  5 +++++
>  3 files changed, 60 insertions(+)
> 
> diff --git a/block.c b/block.c
> index efc3c43..08aa979 100644
> --- a/block.c
> +++ b/block.c
> @@ -4399,3 +4399,53 @@ void bdrv_refresh_filename(BlockDriverState *bs)
>          QDECREF(json);
>      }
>  }
> +
> +/*
> + * Hot add/remove a BDS's child. So the user can take a child offline when
> + * it is broken and take a new child online
> + */
> +void bdrv_add_child(BlockDriverState *parent_bs, BlockDriverState *child_bs,
> +                    Error **errp)
> +{
> +
> +    if (!parent_bs->drv || !parent_bs->drv->bdrv_add_child) {
> +        error_setg(errp, "The node %s doesn't support adding a child",

As I said in my reply to v9's patch 3 (and I see you've followed it), I
don't quite like contractions in error messages, so I'd rather like this
to be "does not" instead of "doesn't".

If you don't decide to change this patch, then feel free to leave this
as it is, because that way you can keep Eric's and Berto's R-bs.

> +                   bdrv_get_device_or_node_name(parent_bs));
> +        return;
> +    }
> +
> +    if (!QLIST_EMPTY(&child_bs->parents)) {
> +        error_setg(errp, "The node %s already has a parent",
> +                   child_bs->node_name);
> +        return;
> +    }
> +
> +    parent_bs->drv->bdrv_add_child(parent_bs, child_bs, errp);
> +}
> +
> +void bdrv_del_child(BlockDriverState *parent_bs, BlockDriverState *child_bs,
> +                    Error **errp)

I wondered before and now I'm wondering why I didn't say anything. Why
is this function taking a BDS pointer as child_bs? Why not just
"BdrvChild *child"? Its only caller will be qmp_x_blockdev_change()
which gets the child BDS from bdrv_find_child(), which could just as
well return a BdrvChild instead of a BDS pointer.

I see two advantages to this: First, it's a bit clearer since this
operation actually does not delete the child *BDS* but only the *edge*
between parent and child, and that edge is what a BdrvChild is.

And second, some block drivers may prefer the BdrvChild over the BDS
itself. They can always trivially get the BDS from the BdrvChild
structure, but the other way around is difficult.

The only disadvantage I can see is that it then becomes asymmetric to
bdrv_add_child(). But that's fine, it's a different function after all.
bdrv_add_child() creates a BdrvChild object (an edge), bdrv_del_child()
deletes it.

(I don't know whether you had this discussion with someone else before,
though. Sorry if you did, I missed it, then.)

> +{
> +    BdrvChild *child;
> +
> +    if (!parent_bs->drv || !parent_bs->drv->bdrv_del_child) {
> +        error_setg(errp, "The node %s doesn't support removing a child",
> +                   bdrv_get_device_or_node_name(parent_bs));

Again, optional s/doesn't/does not/.

> +        return;
> +    }
> +
> +    QLIST_FOREACH(child, &parent_bs->children, next) {
> +        if (child->bs == child_bs) {
> +            break;
> +        }
> +    }
> +
> +    if (!child) {
> +        error_setg(errp, "The node %s is not a child of %s",
> +                   bdrv_get_device_or_node_name(child_bs),
> +                   bdrv_get_device_or_node_name(parent_bs));

Is there a special reason why you are using
bdrv_get_device_or_node_name() for the child here, while in
bdrv_add_child() you directly use the node name?

Neither seems wrong to me. A child is unlikely to have a BB of its own,
but if it does, printing its name instead of the node name may be
appropriate. I'm just wondering about the difference between
bdrv_add_child() and bdrv_del_child().

Max

> +        return;
> +    }
> +
> +    parent_bs->drv->bdrv_del_child(parent_bs, child_bs, errp);
> +}
> diff --git a/include/block/block.h b/include/block/block.h
> index 1c4f4d8..ecde190 100644
> --- a/include/block/block.h
> +++ b/include/block/block.h
> @@ -582,4 +582,9 @@ void bdrv_drained_begin(BlockDriverState *bs);
>   */
>  void bdrv_drained_end(BlockDriverState *bs);
>  
> +void bdrv_add_child(BlockDriverState *parent, BlockDriverState *child,
> +                    Error **errp);
> +void bdrv_del_child(BlockDriverState *parent, BlockDriverState *child,
> +                    Error **errp);
> +
>  #endif
> diff --git a/include/block/block_int.h b/include/block/block_int.h
> index 9ef823a..89ec138 100644
> --- a/include/block/block_int.h
> +++ b/include/block/block_int.h
> @@ -305,6 +305,11 @@ struct BlockDriver {
>       */
>      void (*bdrv_drain)(BlockDriverState *bs);
>  
> +    void (*bdrv_add_child)(BlockDriverState *parent, BlockDriverState *child,
> +                           Error **errp);
> +    void (*bdrv_del_child)(BlockDriverState *parent, BlockDriverState *child,
> +                           Error **errp);
> +
>      QLIST_ENTRY(BlockDriver) list;
>  };
>  
> 



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 473 bytes --]