Re: [PATCH v1 7/8]: PVH privcmd changes

[Mukesh Rathor <mukesh.rathor@oracle.com>]

On Thu, 4 Oct 2012 09:50:42 +0100
Ian Campbell <Ian.Campbell@citrix.com> wrote:

> On Wed, 2012-10-03 at 23:31 +0100, Mukesh Rathor wrote:
> > On Wed, 3 Oct 2012 14:21:35 +0100
> > Ian Campbell <Ian.Campbell@citrix.com> wrote:
> > 
> > > On Fri, 2012-09-21 at 20:21 +0100, Mukesh Rathor wrote:
> > > > +static int pvh_privcmd_resv_pfns(struct vm_area_struct *vma,
> > > > int numpgs)
> > > ...
> > > > +       pvhp->pi_num_pgs = numpgs;
> > > > +       BUG_ON(vma->vm_private_data != (void *)1);
> > > > +       vma->vm_private_data = pvhp; 
> > > 
> > > How does this interact with:
> > > 
> > > static int privcmd_enforce_singleshot_mapping(struct
> > > vm_area_struct *vma) {
> > > 	return (xchg(&vma->vm_private_data, (void *)1) == NULL);
> > > }
> > > 
> > > If someone tries to map a second time then won't this correct the
> > > pvhp in vm_private_data by resetting it to 1? Then when the
> > > original mapping is torn down things all fall apart?
> > > 
> > > Perhaps we need a cmpxchg here? Or to rework the callers a little
> > > bit perhaps.
> > 
> > Right, that's why I had it originally checking for auto xlated and
> > doing something different. I think that is better than to change
> > this and change again. I'll change it back to just putting the ptr
> > here.
> 
> Won't that break because on the second call you will pass in the
> freshly allocated pointer and overwrite the exiting (useful) one with
> it?

No, for xlate, I just check for NULL. I didn't think it was big 
deal to special case xlate in this case. We got so many if xlate 
cases already thru the code. It leaves the semantics easy to 
understand: NULL == avail. 1 == locked PV. PTR == Locked PVH. I'll add
a comment this time :).

thanks,
Mukesh