Re: [Xen-devel] [PATCH RESEND v5 5/6] xen/arm: Implement hypercall for dirty page tracing

[Ian Campbell <Ian.Campbell@xxxxxxxxxx>]

On Fri, 2013-11-15 at 13:15 +0900, Jaeyong Yoo wrote:
> I accidently missed this comment in the previous email.
> 
> > > diff --git a/xen/arch/arm/traps.c b/xen/arch/arm/traps.c index
> > > 287dd7b..1a7ed11 100644
> > > --- a/xen/arch/arm/traps.c
> > > +++ b/xen/arch/arm/traps.c
> > > @@ -1321,6 +1321,8 @@ static void do_trap_data_abort_guest(struct
> > cpu_user_regs *regs,
> > >      const char *msg;
> > >      int rc, level = -1;
> > >      mmio_info_t info;
> > > +    int page_fault = ( (dabt.dfsc & FSC_MASK) ==
> > > +                          (FSC_FLT_PERM | FSC_3D_LEVEL) && dabt.write
> > > + );
> > 
> > I think you can use FSC_TYPE_MASK and FSC_LL_MASK here, can't you?
> > 
> 
> Something like
> 
> dabt.dfsc & FSC_TYPE_MASK == FSC_TYPE_FAULT &&
> dabt.dfsc & FSC_LL_MASK == 0x3 /* third level */,
> 
> right?

Something like that, yes. You'd need the bit about it being a perm fault
in there somewhere too.

> 
> > I think this would be better off refactored into a
> > dabt_is_page_fault(dabt), used in the test below. That would allow you to
> > more easily comment on why these particular conditions are the ones we
> > care about etc.
> > 
> 
> Sure.
> 
> Jaeyong
> 



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