On Tue, Apr 23, 2019 at 12:36:05PM -0500, Eric Blake wrote:
On 3/28/19 11:18 AM, Richard W.M. Jones wrote:
> Allows you to safely use nbdkit-offset-filter on top of a plugin
> supporting extents.
> ---
> filters/offset/offset.c | 36 ++++++++++++++++++++++++++++++++++++
> 1 file changed, 36 insertions(+)
>
> +
> + extents2 = nbdkit_extents_new (offs + offset, real_size - offset);
> + if (extents2 == NULL) {
> + *err = errno;
> + return -1;
> + }
Here, we are careful to set *err.
> + if (next_ops->extents (nxdata, count, offs + offset,
> + flags, extents2, err) == -1)
> + goto error;
And here.
> +
> + for (i = 0; i < nbdkit_extents_count (extents2); ++i) {
> + e = nbdkit_get_extent (extents2, i);
> + e.offset -= offset;
> + if (nbdkit_add_extent (extents, e.offset, e.length, e.type) == -1)
> + goto error;
But here, *err remains unchanged. Is this a problem? Should
nbdkit_add_extent() guarantee that errno is sane on failure (right now,
it does not)?
Affects several filters.
Yes I'm pretty certain this is a bug. I guess because this jumps to
'error:' I copied the previous pattern and thus forgot to set *err ...
Rich.
--
Richard Jones, Virtualization Group, Red Hat
http://people.redhat.com/~rjones
Read my programming and virtualization blog:
http://rwmj.wordpress.com
virt-builder quickly builds VMs from scratch
http://libguestfs.org/virt-builder.1.html