On 06/16/2014 04:05 AM, Hu Tao wrote:
> Add qmp command query-memdev to query for information
> of memory devices
>
> Signed-off-by: Hu Tao <hu...@cn.fujitsu.com>
> ---
> numa.c | 84
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> qapi-schema.json | 40 +++++++++++++++++++++++++++
> qmp-commands.hx | 38 +++++++++++++++++++++++++
> 3 files changed, 162 insertions(+)
>
> diff --git a/numa.c b/numa.c
> index 8ba5a38..bcbafd2 100644
> --- a/numa.c
> +++ b/numa.c
> @@ -34,6 +34,7 @@
> #include "qapi/qmp/qerror.h"
> #include "hw/boards.h"
> #include "sysemu/hostmem.h"
> +#include "qmp-commands.h"
>
> QemuOptsList qemu_numa_opts = {
> .name = "numa",
> @@ -280,3 +281,86 @@ void memory_region_allocate_system_memory(MemoryRegion
> *mr, Object *owner,
> addr += size;
> }
> }
> +
> +static int query_memdev(Object *obj, void *opaque)
> +{
> + MemdevList **list = opaque;
> + Error *err = NULL;
This variable is usually named local_err.
> +
> + if (object_dynamic_cast(obj, TYPE_MEMORY_BACKEND)) {
> + MemdevList *m = g_malloc0(sizeof(*m));
> +
> + m->value = g_malloc0(sizeof(*m->value));
> +
> + m->value->size = object_property_get_int(obj, "size",
> + &err);
> + if (err) {
> + goto error;
> + }
Is this error ever likely, or would it represent a programming error, in
which case passing &error_abort would avoid the need to check for
failure afterwords. Similarly for all the other properties.
> +query-memdev
> +------------
> +
> +Show memory devices information.
Makes sense, but took me a couple reads. Might be better as:
Show information on memory devices.
> +
> +
> +Example (1):
What is the "(1)" for? I'd just drop it (probably copy-and-paste from
another command that actually had more than one example)
--
Eric Blake eblake redhat com +1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
