Re: [Xen-devel] [PATCH v2 1/4] xen: make logdirty and iommu mutually exclusive

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 23.04.14 at 10:16, <roger.pau@xxxxxxxxxx> wrote:
> --- a/xen/arch/x86/mm/hap/hap.c
> +++ b/xen/arch/x86/mm/hap/hap.c
> @@ -175,6 +175,15 @@ out:
>   */
>  static int hap_enable_log_dirty(struct domain *d, bool_t log_global)
>  {
> +
> +    if ( iommu_enabled && need_iommu(d) && log_global )

need_iommu() implies iommu_enabled, and hence checking only the
former ought to be sufficient.

Also, why do you do this here and in shadow code rather than in
paging_log_dirty_enable()?

> --- a/xen/drivers/passthrough/iommu.c
> +++ b/xen/drivers/passthrough/iommu.c
> @@ -292,7 +292,8 @@ static int assign_device(struct domain *d, u16 seg, u8 
> bus, u8 devfn)
>       * enabled for this domain */
>      if ( unlikely(!need_iommu(d) &&
>              (d->arch.hvm_domain.mem_sharing_enabled ||
> -             d->mem_event->paging.ring_page)) )
> +             d->mem_event->paging.ring_page ||
> +             paging_mode_log_dirty(d))) )

I'm afraid that would return true also for non-global log dirty mode.
This patch series
http://lists.xenproject.org/archives/html/xen-devel/2014-04/msg02750.html
(patch 1) introduces a way to know whether global mode is enabled
- maybe you should base your series on top of that one?

Jan


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