On 01/06/16 08:52, Paulina Szubarczyk wrote: > On Tue, 2016-05-31 at 10:37 +0100, David Vrabel wrote: >> On 31/05/2016 05:44, Paulina Szubarczyk wrote: >>> Grant copy operation is divided into two phases different for >>> 'read' and 'write' operation. >>> >>> For a 'read' operation the flow is as follow: >>> 1. allocate local buffers for all the segments contained in >>> a request. >> >> Allocating buffers page-by-page looks sub-optimal to me. Why not >> allocate one large buffer for the whole request? > I thought about caching the pages and reuse them if there are more request. > I did the change in the next patch 4/4.
Ok. I hadn't looked at 4/4 yet. David _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
