[PATCH v3 02/32] arm64: KVM: define HYP and Stage-2 translation page flags

Marc Zyngier marc.zyngier at arm.com
Fri Apr 26 13:11:07 EDT 2013


On 26/04/13 18:01, Catalin Marinas wrote:
> On Mon, Apr 08, 2013 at 05:17:04PM +0100, Marc Zyngier wrote:
>> diff --git a/arch/arm64/include/asm/pgtable-hwdef.h b/arch/arm64/include/asm/pgtable-hwdef.h
>> index 75fd13d..acb4ee5 100644
>> --- a/arch/arm64/include/asm/pgtable-hwdef.h
>> +++ b/arch/arm64/include/asm/pgtable-hwdef.h
>> @@ -68,6 +69,18 @@
>>  #define PTE_ATTRINDX_MASK	(_AT(pteval_t, 7) << 2)
>>  
>>  /*
>> + * 2nd stage PTE definitions
>> + */
>> +#define PTE_S2_RDONLY		 (_AT(pteval_t, 1) << 6)   /* HAP[1]   */
>> +#define PTE_S2_RDWR		 (_AT(pteval_t, 2) << 6)   /* HAP[2:1] */
> 
> RDWR should be 3 here (already the case in arch/arm). BTW, I would use

Yes, Will spotted this one already, and it is now fixed in my tree.

> HAP[2:1] comment in both cases since that's the attribute field.

Indeed.

>> diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h
>> index e333a24..7c84ab4 100644
>> --- a/arch/arm64/include/asm/pgtable.h
>> +++ b/arch/arm64/include/asm/pgtable.h
>> @@ -76,6 +76,12 @@ extern pgprot_t pgprot_default;
>>  #define PAGE_KERNEL		_MOD_PROT(pgprot_default, PTE_PXN | PTE_UXN | PTE_DIRTY)
>>  #define PAGE_KERNEL_EXEC	_MOD_PROT(pgprot_default, PTE_UXN | PTE_DIRTY)
>>  
>> +#define PAGE_HYP		_MOD_PROT(pgprot_default, PTE_HYP)
>> +#define PAGE_HYP_DEVICE		_MOD_PROT(__pgprot(PROT_DEVICE_nGnRE), PTE_HYP)
>> +
>> +#define PAGE_S2			_MOD_PROT(pgprot_default, PTE_USER | PTE_S2_RDONLY)
> 
> Why is this one read-only by default?

Because the guest pages start their life mapped RO. Only on the first
write access they become writeable.

>> +#define PAGE_S2_DEVICE		_MOD_PROT(__pgprot(PROT_DEVICE_nGnRE), PTE_USER | PTE_S2_RDWR)
> 
> You could write it directly as __pgprot(PROT_DEVICE_nGnRE | PTE_USER | PTE_S2_RDWR)

Good point! This code as changed a bit anyway, as it contains some other
odd things... ;-)

Thanks for reviewing,

	M.
-- 
Jazz is not dead. It just smells funny...




More information about the linux-arm-kernel mailing list