Isaku Yamahata wrote:
On Thu, May 24, 2007 at 11:38:19AM +0200, Jes Sorensen wrote:
P=M will break the current grant table api. It means all of virtual
io device (balloon, vbd, vnif, ...) will be broken.
What do you think about it?
Then we have to fix it - the Linux kernel requires this, it's not
optional. Sure it works for over simplified systems such as DIG, but
anything moderately advanced will break.
In fact, Virtual Physical address model was introduced to
resolve the grant table issue. And the ia64 default dma api and
hp zx1 iommu are already paravirtulized.
There's much more to it than that. We need it for things like
alloc_pages_node() in the Linux kernel if we want any level of
performance.
In addition we want this
feature when we want to support multiple dom0's in the future.
Do you mean driver domain by "multiple dom0"?
I.e. domain which can serves as backend.
If so, it is already supported by pci backend/frontend driver.
I mean if we want to let individual dom's have direct access to a PCI
device. They will need to know where in the system it's physically
located.
Regards,
Jes
12 messages in com.xensource.lists.xen-ia64-develRe: [Xen-ia64-devel] [patch] calculat...
.diff