On Wed, 2015-07-01 at 10:43 +0800, Yang Hongyang wrote: > > On 06/30/2015 06:19 PM, Ian Campbell wrote: > > On Thu, 2015-06-25 at 14:25 +0800, Yang Hongyang wrote: > >> Under COLO, we are doing checkpoint on demand, if this > >> callback returns 1, we will take another checkpoint. > >> 0 indicates unexpected error. > > > > Is this checkpoint therefore expected to be blocking until another > > checkpoint is desired (or an error occurs)? > > Right.
IMHO this should therefore be documented. Some care will need to be taken to avoid blocking any non-async part of the main libxl process. The means by which this is achieved (which I suppose is that the call back is in the context of the helper process) should be mentioned in the commit message. Ian. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
