WARNING - OLD ARCHIVES

This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-ia64-devel

Re: [Xen-ia64-devel] [PATCH] discontig memory support

On Sun, 2006-04-16 at 21:33 -0600, Alex Williamson wrote:
> On Mon, 2006-04-17 at 11:21 +0900, Kouya SHIMURA wrote:
> > BTW, I think that it is unnecessary to support large holes on
> > dom0 in p2m/vp model. Or dom0 still requires to be configured with 
> > virtual memory map because of MMIO. Is it true?
> 
>    You may be right.  The page at the end of memory in dom0 goes away
> with the p2m/vp patches and that's what causes me to need virtual
> memmap/discontig/sparsemem in dom0.  MMIO at high addresses shouldn't
> require any special handling of memory in dom0.  Thanks,

Hi Kouya,

   I couple data points.  When I merge these patches in w/ the P2M/VP
patches, as expected dom0 gets a memory map that no longer needs virtual
memmap on my system.  Even better, I can enable virtual memmap on a
P2M/VP kernel and I don't see the Oops from the loop driver.  I think
maybe those of us on systems that need virtual memmap can use the
max_addr= boot option to limit the available machine memory until the
P2M/VP patches are merged in-tree.  Also, my performance tests show no
degradation in either dom0 or domU performance (with the kernel compile
benchmark).  Thanks,

        Alex

-- 
Alex Williamson                             HP Linux & Open Source Lab


_______________________________________________
Xen-ia64-devel mailing list
Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ia64-devel