On Sun, Jan 07, 2018 at 01:23:35PM +0100, Richard Palethorpe wrote:
> Add QAPI wrapper functions for the existing snapshot functionality. These
> functions behave the same way as the HMP savevm, loadvm and delvm
> commands. This will allow applications, such as OpenQA, to programmatically
> revert the VM to a previous state with no dependence on HMP or qemu-img.
>
> I used the term snapshot instead of VM because I think it is less ambiguous
> and matches the internal function names.
>
> Signed-off-by: Richard Palethorpe <richi...@f-m.fm>
> ---
> migration/savevm.c | 27 ++++++++++++++++++++++
> qapi-schema.json | 66
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++
> 2 files changed, 93 insertions(+)
>
> diff --git a/migration/savevm.c b/migration/savevm.c
> index b7908f62be..d7bc0f0d41 100644
> --- a/migration/savevm.c
> +++ b/migration/savevm.c
> @@ -2242,6 +2242,11 @@ int save_snapshot(const char *name, Error **errp)
> return ret;
> }
>
> +void qmp_save_snapshot(const char *name, Error **errp)
> +{
> + save_snapshot(name, errp);
> +}
> +
> void qmp_xen_save_devices_state(const char *filename, bool has_live, bool
> live,
> Error **errp)
> {
> @@ -2404,6 +2409,28 @@ err_drain:
> return ret;
> }
>
> +void qmp_load_snapshot(const char *name, Error **errp)
> +{
> + int saved_vm_running = runstate_is_running();
> +
> + vm_stop(RUN_STATE_RESTORE_VM);
> +
> + if (!load_snapshot(name, errp) && saved_vm_running) {
> + vm_start();
> + }
> +}
Hi, Richard,
I think it's good to have, but from interface POV for sure I'll leave
it for maintainers.
For the code, I would prefer at least unify the code instead of
duplicating it. I think calling qmp_*() functions in hmp_*() would be
good since that's what we do now mostly AFAICT.
Also, comparing to exposing snapshot operations, I am curious about
what would be the difference if we just migrate (which can use QMP) to
a file and then load that file using incoming migration. Thanks,
--
Peter Xu
