On 10/12/2017 02:54 PM, Andrew Cooper wrote:
> There are currently three functions which write L4 pagetables for Xen, but
> they all behave subtly differently. sh_install_xen_entries_in_l4() in
> particular is catering for two different usecases, which makes the safety of
> the linear mappings har
>>> On 12.10.17 at 17:24, wrote:
> On 12/10/17 16:08, Jan Beulich wrote:
> On 12.10.17 at 15:54, wrote:
>>> --- a/xen/arch/x86/mm/hap/hap.c
>>> +++ b/xen/arch/x86/mm/hap/hap.c
>>> @@ -391,41 +391,24 @@ int hap_set_allocation(struct domain *d, unsigned int
> pages, bool *preempted)
>>> r
On 12/10/17 16:08, Jan Beulich wrote:
On 12.10.17 at 15:54, wrote:
>> --- a/xen/arch/x86/mm/hap/hap.c
>> +++ b/xen/arch/x86/mm/hap/hap.c
>> @@ -391,41 +391,24 @@ int hap_set_allocation(struct domain *d, unsigned int
>> pages, bool *preempted)
>> return 0;
>> }
>>
>> -static void hap_
On Thu, Oct 12, 2017 at 02:54:22PM +0100, Andrew Cooper wrote:
> There are currently three functions which write L4 pagetables for Xen, but
> they all behave subtly differently. sh_install_xen_entries_in_l4() in
> particular is catering for two different usecases, which makes the safety of
> the l
>>> On 12.10.17 at 15:54, wrote:
> --- a/xen/arch/x86/mm/hap/hap.c
> +++ b/xen/arch/x86/mm/hap/hap.c
> @@ -391,41 +391,24 @@ int hap_set_allocation(struct domain *d, unsigned int
> pages, bool *preempted)
> return 0;
> }
>
> -static void hap_install_xen_entries_in_l4(struct vcpu *v, mfn_t
There are currently three functions which write L4 pagetables for Xen, but
they all behave subtly differently. sh_install_xen_entries_in_l4() in
particular is catering for two different usecases, which makes the safety of
the linear mappings hard to follow.
By consolidating the L4 pagetable writi