>>> On 17.12.15 at 17:31, <julien.gr...@citrix.com> wrote: > I've looked at reworking XENMEM_exchange to avoid mfn_to_gmfn. My idea would > be to allocate a temporary array to store the GFN between the two loops. > However, the array would be quite large (the max default is 18 on ARM), > so I don't know if it's acceptable.
This would quite clearly not be acceptable, yet I don't have an alternative suggestion other then implementing M2P on ARM, which you have already named as not an option. Otoh I'm not sure what two loops you talk about - I find exactly one use of mfn_to_gmfn() in memory_exchange(), and latching the GFN earlier would be wrong (as it might change in parallel to your processing). Jan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
