Re: [V2 PATCH 3/3] P2M error code propogation

["Jan Beulich" <JBeulich@suse.com>]

>>> On 10.04.14 at 02:19, <mukesh.rathor@oracle.com> wrote:
>              ret = iomem_deny_access(d, mfn, mfn + nr_mfns - 1);
> -            if ( !ret && add )
> -                ret = -EIO;
> +            if ( !ret && tmp_rc )
> +                ret = tmp_rc;

This could be just "if ( !ret )"; no need to re-submit just because of this
though.

>  static int
>  p2m_pt_set_entry(struct p2m_domain *p2m, unsigned long gfn, mfn_t mfn,
>                   unsigned int page_order, p2m_type_t p2mt, p2m_access_t p2ma)
>  {
> -    // XXX -- this might be able to be faster iff current->domain == d
> +    /* XXX -- this might be able to be faster iff current->domain == d */
>      mfn_t table_mfn = pagetable_get_mfn(p2m_get_pagetable(p2m));
> -    void *table =map_domain_page(mfn_x(table_mfn));
> +    void *table = map_domain_page(mfn_x(table_mfn));

While I'm generally fine with you doing cleanup like this as you go, I
think this should have been done in the mechanical (renaming) patch
rather than in this one, which while not intended to change
functionality needs pretty close looking at to verify it really doesn't,
and hence would benefit from having just the changes that are
needed (that said, adjusting style when you have to touch a line
anyway is still fine).

As the comments are only of stylistic nature, nevertheless
Reviewed-by: Jan Beulich <jbeulich@suse.com>

Jan