>>> On 15.06.17 at 19:21, wrote:
> On 14/06/17 10:12, Jan Beulich wrote:
> On 13.06.17 at 17:25, wrote:
>>> @@ -0,0 +1,116 @@
>>> +Atomic operations in Xen
>>> +
>>> +
>>> +Data structures in Xen memory which can be accessed by multiple CPUs
>>> +at the same time need
On Thu, 15 Jun 2017, Jan Beulich wrote:
> >>> Stefano Stabellini 06/15/17 2:27 AM >>>
> >On Wed, 14 Jun 2017, Jan Beulich wrote:
> >> >>> Stefano Stabellini 06/14/17 8:45 PM >>>
> >> >On Wed, 14 Jun 2017, Jan Beulich wrote:
> >> >> > +What ACCESS_ONCE does *not* guarantee though is this access is
Hi Jan,
thanks for spending your time on this mind boggling exercise!
On 14/06/17 10:12, Jan Beulich wrote:
On 13.06.17 at 17:25, wrote:
>> as mentioned in my previous mail, I consider this more of a discussion
>> base that an actual patch. I am by no means an expert in this area, so
>> par
>>> Stefano Stabellini 06/15/17 2:27 AM >>>
>On Wed, 14 Jun 2017, Jan Beulich wrote:
>> >>> Stefano Stabellini 06/14/17 8:45 PM >>>
>> >On Wed, 14 Jun 2017, Jan Beulich wrote:
>> >> > +What ACCESS_ONCE does *not* guarantee though is this access is done in
>> >> > a
>> >> > +single instruction, s
On Wed, 14 Jun 2017, Jan Beulich wrote:
> >>> Stefano Stabellini 06/14/17 8:45 PM >>>
> >On Wed, 14 Jun 2017, Jan Beulich wrote:
> >> > +What ACCESS_ONCE does *not* guarantee though is this access is done in a
> >> > +single instruction, so complex or non-native or unaligned data types are
> >> >
>>> Stefano Stabellini 06/14/17 8:45 PM >>>
>On Wed, 14 Jun 2017, Jan Beulich wrote:
>> > +What ACCESS_ONCE does *not* guarantee though is this access is done in a
>> > +single instruction, so complex or non-native or unaligned data types are
>> > +not guaranteed to be atomic. If for instance coun
On Wed, 14 Jun 2017, Jan Beulich wrote:
> > +What ACCESS_ONCE does *not* guarantee though is this access is done in a
> > +single instruction, so complex or non-native or unaligned data types are
> > +not guaranteed to be atomic. If for instance counter would be a 64-bit
> > value
> > +on a 32-bit
>>> On 13.06.17 at 17:25, wrote:
> as mentioned in my previous mail, I consider this more of a discussion
> base that an actual patch. I am by no means an expert in this area, so
> part of this exercise here is to write down my understanding and see it
> corrected by more knowledgable people ;-)
On Tue, 13 Jun 2017, Andre Przywara wrote:
> Recently there were some discussions about the nature and guarantees of
> the atomic primitives that Xen provides.
> This README.atomic file tries to document our expectations in those
> functions and macros.
>
> Signed-off-by: Andre Przywara
> ---
> H
