Re: [Xen-devel] [RFC XEN PATCH v3 01/39] x86_64/mm: fix the PDX group check in mem_hotadd_check()

[Haozhong Zhang <haozhong.zhang@xxxxxxxxx>]

On 10/27/17 14:49 +0800, Chao Peng wrote:
> On Mon, 2017-09-11 at 12:37 +0800, Haozhong Zhang wrote:
> > The current check refuses the hot-plugged memory that falls in one
> > unused PDX group, which should be allowed.
> 
> Looks reasonable to me. The only thing I can think of is you can double
> check if the following find_next_zero_bit/find_next_bit will still
> work. 

The first check in mem_hotadd_check() ensures spfn < epfn, so sidx <=
eidx here. Compared with the previous code, the only added case is
sidx == eidx, which is what this patch intends to allow and tested.

Haozhong

> 
> Chao
> > 
> > Signed-off-by: Haozhong Zhang <haozhong.zhang@xxxxxxxxx>
> > ---
> > Cc: Jan Beulich <jbeulich@xxxxxxxx>
> > Cc: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
> > ---
> >  xen/arch/x86/x86_64/mm.c | 6 +-----
> >  1 file changed, 1 insertion(+), 5 deletions(-)
> > 
> > diff --git a/xen/arch/x86/x86_64/mm.c b/xen/arch/x86/x86_64/mm.c
> > index 11746730b4..6c5221f90c 100644
> > --- a/xen/arch/x86/x86_64/mm.c
> > +++ b/xen/arch/x86/x86_64/mm.c
> > @@ -1296,12 +1296,8 @@ static int mem_hotadd_check(unsigned long spfn,
> > unsigned long epfn)
> >          return 0;
> >  
> >      /* Make sure the new range is not present now */
> > -    sidx = ((pfn_to_pdx(spfn) + PDX_GROUP_COUNT - 1)  &
> > ~(PDX_GROUP_COUNT - 1))
> > -            / PDX_GROUP_COUNT;
> > +    sidx = (pfn_to_pdx(spfn) & ~(PDX_GROUP_COUNT - 1)) /
> > PDX_GROUP_COUNT;
> >      eidx = (pfn_to_pdx(epfn - 1) & ~(PDX_GROUP_COUNT - 1)) /
> > PDX_GROUP_COUNT;
> > -    if (sidx >= eidx)
> > -        return 0;
> > -
> >      s = find_next_zero_bit(pdx_group_valid, eidx, sidx);
> >      if ( s > eidx )
> >          return 0;

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel