Re: [Xen-ia64-devel] [RFC] SMP issues

[Isaku Yamahata <yamahata@xxxxxxxxxxxxx>]

On Wed, Jun 14, 2006 at 10:33:10AM +0200, Tristan Gingold wrote:
> Le Mercredi 14 Juin 2006 10:30, Tristan Gingold a écrit :
> > Le Mercredi 14 Juin 2006 07:58, Isaku Yamahata a écrit :

> > > - races between global tlb purge and tlb insert
> > >   This is a race between reading/writing vcpu->arch.{d, i}tlb or VHPT
> > > entry. When a vcpu is about to insert tlb, another vcpu may purge tlb
> > >   cache globally. Inserting tlb (vcpu_itc_no_srlz()) or global tlb purge
> > >   (domain_flush_vtlb_range() and domain_flush_vtlb_all()) can't update
> > >   cpu->arch.{d, i}tlb, VHPT and mTLB. So there is a race here.
> > >   Use sequence lock to avoid this race.
> > >   After inserting tlb entry, check the sequence lock and retry to insert.
> > >   This means that when global tlb purge and tlb insert are issued
> > >   simultaneously, always tlb insert happens after global tlb purge.
> > >
> > >   There was an attempt to resolve this race by checking only
> > >   vcpu->arch.{d, i}tlb.p bit. However it was incomplete because it
> > > doesn't take care of VHPT.
> >
> > I don't agree with the last paragraph.

Sorry for that the paragraph was misleading.
I think checking p bit works for ia64_do_page_fault().

I'd like to say that 
- vcpu_itc_i(), vcpu_itc_d() should be modified.
- vcpu_itc_no_srlz() sets p bit when it is called by
  vcpu_itc_i(), vcpu_itc_d().

So checking p bit doesn't work for vcpu_itc_i(), vcpu_itc_d()
Sequence lock is needed to fix vcpu_itc_i() and vcpu_itc_d().


-- 
yamahata

_______________________________________________
Xen-ia64-devel mailing list
Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ia64-devel