On Mon, Jan 10, 2011 at 10:36:34AM +0000, stefano.stabellini@xxxxxxxxxxxxx
wrote:
> From: Jeremy Fitzhardinge <jeremy.fitzhardinge@xxxxxxxxxx>
>
> Add a simple hashtable based mechanism to override some portions of the
> m2p, so that we can find out the pfn corresponding to an mfn of a
> granted page. In fact entries corresponding to granted pages in the m2p
> hold the original pfn value of the page in the source domain that
> granted it.
>
> Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@xxxxxxxxxx>
> Signed-off-by: Stefano Stabellini <stefano.stabellini@xxxxxxxxxxxxx>
> ---
> arch/x86/include/asm/xen/page.h | 16 ++++++-
> arch/x86/xen/p2m.c | 80
> +++++++++++++++++++++++++++++++++++++++
> 2 files changed, 93 insertions(+), 3 deletions(-)
>
> diff --git a/arch/x86/include/asm/xen/page.h b/arch/x86/include/asm/xen/page.h
> index 8760cc6..50f0a0f 100644
> --- a/arch/x86/include/asm/xen/page.h
> +++ b/arch/x86/include/asm/xen/page.h
> @@ -42,6 +42,11 @@ extern unsigned int machine_to_phys_order;
> extern unsigned long get_phys_to_machine(unsigned long pfn);
> extern bool set_phys_to_machine(unsigned long pfn, unsigned long mfn);
>
> +extern void m2p_add_override(unsigned long mfn, struct page *page);
> +extern void m2p_remove_override(struct page *page);
> +extern struct page *m2p_find_override(unsigned long mfn);
> +extern unsigned long m2p_find_override_pfn(unsigned long mfn, unsigned long
> pfn);
> +
> static inline unsigned long pfn_to_mfn(unsigned long pfn)
> {
> unsigned long mfn;
> @@ -72,9 +77,6 @@ static inline unsigned long mfn_to_pfn(unsigned long mfn)
> if (xen_feature(XENFEAT_auto_translated_physmap))
> return mfn;
>
> - if (unlikely((mfn >> machine_to_phys_order) != 0))
> - return ~0;
> -
> pfn = 0;
> /*
> * The array access can fail (e.g., device space beyond end of RAM).
> @@ -83,6 +85,14 @@ static inline unsigned long mfn_to_pfn(unsigned long mfn)
> */
> __get_user(pfn, &machine_to_phys_mapping[mfn]);
>
> + /*
> + * If this appears to be a foreign mfn (because the pfn
> + * doesn't map back to the mfn), then check the local override
> + * table to see if there's a better pfn to use.
> + */
> + if (get_phys_to_machine(pfn) != mfn)
Any reason to not use 'pfn_to_mfn' call instead? If we do want
to use the get_hhys_to_machine wouldn't be easier to just
check for 'FOREIGN_FRAME_BIT' set? Or is just mostly a copy-n-paste
of mfn_to_local_pfn?
Asking this b/c with some of my changes this will be triggered on
identity mappings. The reason being that the IDENTITY_FRAME_BIT
is still set when using get_phys_to_machine. However on the 'pfn_to_mfn'
call, that is cleared (along with the FOREIGN_FRAME_BIT)...
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
Home