On Tue, Sep 9, 2014 at 11:20 AM, Ian Campbell wrote: > On Mon, 2014-09-08 at 13:41 -0700, Julien Grall wrote: > > > > On 03/09/14 14:56, Tamas K Lengyel wrote: > > > I'm not entirely sure I could easily extend > > > apply_one_level/apply_p2m_changes without significant changes and a > > > potential to affect something else along the way.. I can give it a try > > > though but I'm a bit hesitant.. Do you see an immediate performance > > > reason to try to hook it into those functions instead of keeping it > > > separate? The current function is pretty straight forward in what it is > > > doing as it is. > > > > I would like to avoid standalone function that manipulate the p2m. It > > will be easier to fix bug, hence the code if everything is using the > > same code. > > Agreed, having two functions which do essentially the same thing (walk > the p2m and manipulate it) is not good from a maintenance point of view. > There would need to be a more compelling reason for this than a worry > about breaking the existing code. > > Ian. > Fair enough. I'll get it in for the next iteration. Tamas