[PATCH 2/2] cr: handle CONFIG_SYSVIPC=n

Matt Helsley matthltc at us.ibm.com
Fri Apr 9 18:42:07 PDT 2010


On Fri, Apr 09, 2010 at 04:12:58PM -0500, Serge Hallyn wrote:
> From: Serge E. Hallyn <serue at us.ibm.com>
> 
> If CONFIG_IPC_NS=n, then checkpoint a sysvipcns of 0.  If we
> end up restarted on a CONFIG_SYSVIPC=y host, we'll just end up
> using the parent sysvipcns, which is fine as we won't be trying
> to restore any sysvipc objects.
> 
> (The existing code worked fine with CONFIG_IPC_NS=n, but not
> with CONFIG_SYSVIPC=n and CONFIG_POSIX_MQUEUE=n.  But we
> trigger on CONFIG_IPC_NS since even if CONFIG_SYSVIPC=y but
> CONFIG_IPC_NS=n, checkpointing 0 is right).
> 
> Signed-off-by: Serge E. Hallyn <serue at us.ibm.com>

Applied, thanks!

Cheers,
	-Matt Helsley

> ---
>  kernel/nsproxy.c |    4 ++++
>  1 files changed, 4 insertions(+), 0 deletions(-)
> 
> diff --git a/kernel/nsproxy.c b/kernel/nsproxy.c
> index 5f9d809..0b43f4d 100644
> --- a/kernel/nsproxy.c
> +++ b/kernel/nsproxy.c
> @@ -290,9 +290,13 @@ static int do_checkpoint_ns(struct ckpt_ctx *ctx, struct nsproxy *nsproxy)
>  	if (ret <= 0)
>  		goto out;
>  	h->uts_objref = ret;
> +#ifdef CONFIG_IPC_NS
>  	ret = checkpoint_obj(ctx, nsproxy->ipc_ns, CKPT_OBJ_IPC_NS);
>  	if (ret < 0)
>  		goto out;
> +#else	/* CONFIG_IPC_NS */
> +	ret = 0;
> +#endif	/* CONFIG_IPC_NS */
>  	h->ipc_objref = ret;
>  #ifdef CONFIG_NETNS_CHECKPOINT
>  	if (!(ctx->uflags & CHECKPOINT_NONETNS))
> -- 
> 1.7.0.4
> 


More information about the Containers mailing list