On 05/31/2013 09:54 AM, John Ferlan wrote:
> Commit '6afdfc8e' adjusted the exit and error paths to go through the error
> and cleanup labels, but neglected to remove the return ret prior to cleanup.
> Also noted the 'type' xml string fetch was never checked for NULL which
> could lead to some interesting results.
> ---
> src/conf/storage_conf.c | 8 ++++++--
> 1 file changed, 6 insertions(+), 2 deletions(-)
>
> diff --git a/src/conf/storage_conf.c b/src/conf/storage_conf.c
> index f0ea41d..94eb69a 100644
> --- a/src/conf/storage_conf.c
> +++ b/src/conf/storage_conf.c
> @@ -834,6 +834,12 @@ virStoragePoolDefParseXML(xmlXPathContextPtr ctxt)
> }
>
> type = virXPathString("string(./@type)", ctxt);
> + if (type == NULL) {
> + virReportError(VIR_ERR_XML_ERROR,
> + _("storage pool missing type attribute"));
> + goto error;
> + }
> +
> if ((ret->type = virStoragePoolTypeFromString(type)) < 0) {
> virReportError(VIR_ERR_XML_ERROR,
> _("unknown storage pool type %s"), type);
> @@ -956,8 +962,6 @@ virStoragePoolDefParseXML(xmlXPathContextPtr ctxt)
> goto error;
> }
>
> - return ret;
> -
> cleanup:
> VIR_FREE(uuid);
> VIR_FREE(type);
>
Peter also pointed out I was missing the "%s" in the new type check
error path. Something syntax-check would have caught (of course I had
run that on my other patch sent earlier, but forgot to on this one, sigh).
In any case, with the "%s" added, syntax-check passing, and rerun of
valgrind, I pushed.
Thanks,
John
--
libvir-list mailing list
libvir-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/libvir-list