On 2016-03-01 22:16, Peter Feiner wrote:
> On Tue, Mar 1, 2016 at 1:12 PM, Paolo Bonzini <pbonzini@xxxxxxxxxx> wrote:
>>
>>
>> On 01/03/2016 20:30, Peter Feiner wrote:
>>> Signed-off-by: Peter Feiner <pfeiner@xxxxxxxxxx>
>>> ---
>>> x86/vmx.c | 6 +++---
>>> x86/vmx.h | 1 +
>>> 2 files changed, 4 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/x86/vmx.c b/x86/vmx.c
>>> index 3fa1a73..9811a28 100644
>>> --- a/x86/vmx.c
>>> +++ b/x86/vmx.c
>>> @@ -246,7 +246,7 @@ void install_ept_entry(unsigned long *pml4,
>>> | EPT_RA | EPT_WA | EPT_EA;
>>> } else
>>> pt[offset] &= ~EPT_LARGE_PAGE;
>>> - pt = phys_to_virt(pt[offset] & 0xffffffffff000ull);
>>> + pt = phys_to_virt(pt[offset] & EPT_ADDR_MASK);
>>> }
>>> offset = ((unsigned long)guest_addr >> ((level-1) *
>>> EPT_PGDIR_WIDTH + 12)) & EPT_PGDIR_MASK;
>>> @@ -334,7 +334,7 @@ unsigned long get_ept_pte(unsigned long *pml4,
>>> break;
>>> if (l < 4 && (pte & EPT_LARGE_PAGE))
>>> return pte;
>>> - pt = (unsigned long *)(pte & 0xffffffffff000ull);
>>> + pt = (unsigned long *)(pte & EPT_ADDR_MASK);
>>> }
>>> offset = (guest_addr >> (((l-1) * EPT_PGDIR_WIDTH) + 12))
>>> & EPT_PGDIR_MASK;
>>> @@ -378,7 +378,7 @@ int set_ept_pte(unsigned long *pml4, unsigned long guest_addr,
>>> break;
>>> if (!(pt[offset] & (EPT_PRESENT)))
>>> return -1;
>>> - pt = (unsigned long *)(pt[offset] & 0xffffffffff000ull);
>>> + pt = (unsigned long *)(pt[offset] & EPT_ADDR_MASK);
>>> }
>>> offset = (guest_addr >> (((l-1) * EPT_PGDIR_WIDTH) + 12))
>>> & EPT_PGDIR_MASK;
>>> diff --git a/x86/vmx.h b/x86/vmx.h
>>> index cae9274..c2a2121 100644
>>> --- a/x86/vmx.h
>>> +++ b/x86/vmx.h
>>> @@ -466,6 +466,7 @@ enum Ctrl1 {
>>> #define EPT_PAGE_LEVEL 4
>>> #define EPT_PGDIR_WIDTH 9
>>> #define EPT_PGDIR_MASK 511
>>> +#define EPT_ADDR_MASK 0xffffffffff000ul
>>
>> This is just PAGE_MASK, defined one line below. :)
>>
>> Paolo
>>
>>> #define PAGE_MASK (~(PAGE_SIZE-1))
>>> #define PAGE_MASK_2M (~(PAGE_SIZE_2M-1))
>>>
>>>
>
> It's not precisely the same thing since PAGE_MASK sets the upper
> reserved bits :-)
>
> I'll send v2.
Consider defining the constant as GENMASK(51, 12) (from linux/bitops.h).
A while back, I re-invented this macro for Jailhouse, and I just
realized that the kernel has the very same thing. It makes such bitmasks
much more readable (and verifiable when looking at specs).
Jan
--
Siemens AG, Corporate Technology, CT RDA ITP SES-DE
Corporate Competence Center Embedded Linux
--
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
