On Thu, Sep 20, 2012 at 6:01 AM, Marc Zyngier <marc.zyngier@xxxxxxx> wrote: > On 18/09/12 22:53, Christoffer Dall wrote: >> On Tue, Sep 18, 2012 at 5:04 PM, Russell King - ARM Linux >> <linux@xxxxxxxxxxxxxxxx> wrote: >>> On Sat, Sep 15, 2012 at 11:34:36AM -0400, Christoffer Dall wrote: >>>> From: Marc Zyngier <marc.zyngier@xxxxxxx> >>>> >>>> The KVM hypervisor mmu code requires access to the mem_type prot_pte >>>> field when setting up page tables pointing to a device. Unfortunately, >>>> the mem_type structure is opaque. >>>> >>>> Add an accessor (get_mem_type_prot_pte()) to retrieve the prot_pte >>>> value. >>>> >>>> Signed-off-by: Marc Zyngier <marc.zyngier@xxxxxxx> >>>> Signed-off-by: Christoffer Dall <c.dall@xxxxxxxxxxxxxxxxxxxxxx> >>> >>> Is there a reason why we need this to be exposed, along with all the >>> page table manipulation in patch 7? >>> >>> Is there a reason why we can't have new MT_ types for PAGE_HYP and >>> the HYP MT_DEVICE type (which is the same as MT_DEVICE but with >>> PTE_USER set) and have the standard ARM/generic kernel code build >>> those mappings? >> >> For hyp mode we can do this, but we cannot do this for the cpu >> interfaces that need to be mapped into each VM as they have each their >> own pgd. > > Isn't that the same problem? The HYP mode has its own pgd too. I think > this is the main issue with the generic code. If we can come up with an > interface that allows the generic code to work on alternative pgds, we > could pretty much do what Russell suggests here. > Hyp mode has its own pgd, but there will only be one of them which can be allocated at boot and setup at boot in mmu.c This will not be the case for guest pages. On the other hand, this locks us to only one user of such mappings and more users could potentially (I know it's a stretch) clutter mmu.c later on, which is why I suggest the PAGE_KVM_DEVICE approach for now, which should then be renamed, PAGE_S2_DEVICE probably. -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
