>>> On 30.01.15 at 11:20, wrote:
> At 08:36 + on 30 Jan (1422603387), Jan Beulich wrote:
>> >>> On 29.01.15 at 18:30, wrote:
>> > --- a/xen/arch/x86/mm/shadow/private.h
>> > +++ b/xen/arch/x86/mm/shadow/private.h
>> > @@ -318,6 +318,33 @@ static inline int mfn_oos_may_write(mfn_t gmfn)
>> >
At 08:36 + on 30 Jan (1422603387), Jan Beulich wrote:
> >>> On 29.01.15 at 18:30, wrote:
> > OK, here's what I have. It keeps the mechanism the same, threading on
> > the main page list entry, but tidies it up to use the page_list operations
> > rather than open-coding. I've tested it (light
>>> On 29.01.15 at 18:30, wrote:
> OK, here's what I have. It keeps the mechanism the same, threading on
> the main page list entry, but tidies it up to use the page_list operations
> rather than open-coding. I've tested it (lightly - basically jsut
> boot testing) with 32bit, 32pae and 64bit Wi
Hi,
At 11:50 +0100 on 27 Jan (1422355826), Tim Deegan wrote:
> Let me have a think about it on Thursday (assuming I don't have too
> many patch series to review by then!)
OK, here's what I have. It keeps the mechanism the same, threading on
the main page list entry, but tidies it up to use the
At 16:35 + on 26 Jan (1422286504), Jan Beulich wrote:
> Tim,
>
> in the course of adding >16Tb support to the hypervisor, I ran into
> issues with you having added several open-coded page list accesses
> while breaking up non-order-0 allocations there. I looked at that
> code for quite a while
