Re: [PATCH 3/3 V2] pvclock: Get rid of __pvclock_read_cycles in function pvclock_read_flags

[Minfei Huang]

On 06/09/16 at 01:28P, Borislav Petkov wrote:
> On Thu, Jun 09, 2016 at 01:21:18PM +0200, Paolo Bonzini wrote:
> > This is basically implementing a seqcount.  It needs two barriers and,
> 
> Why does it need the two barriers? More details please.

Hi, Borislav.

It's a seqcount-like. We should confirm that reading flags occurs
between two reading version. To make complier do not reorder, two
barriers are necessary.

Thanks
Minfei