Re: [PATCH] x86/PoD: skip eager reclaim when possible

[George Dunlap <george.dunlap@citrix.com>]

On 12/05/16 16:08, Jan Beulich wrote:
>>>> On 12.05.16 at 16:40, <george.dunlap@citrix.com> wrote:
>> On 10/05/16 14:44, Jan Beulich wrote:
>>> Reclaiming pages is pointless when the cache can already satisfy all
>>> outstanding PoD entries, and doing reclaims in that case can be very
>>> harmful to performance when that memory gets used by the guest, but
>>> only to store zeroes there.
>>>
>>> Signed-off-by: Jan Beulich <jbeulich@suse.com>
>>
>> Sorry for the delay.  Just one question...
>>
>>>
>>> --- a/xen/arch/x86/mm/p2m-pod.c
>>> +++ b/xen/arch/x86/mm/p2m-pod.c
>>> @@ -1027,7 +1027,6 @@ static void pod_eager_record(struct p2m_
>>>  {
>>>      struct pod_mrp_list *mrp = &p2m->pod.mrp;
>>>  
>>> -    ASSERT(mrp->list[mrp->idx] == INVALID_GFN);
>>>      ASSERT(gfn != INVALID_GFN);
>>
>> What's this for?
> 
> The deletion of the ASSERT()? Since we no longer always insert a
> GFN through pod_eager_reclaim(), we also can no longer assume
> all (used) entries hold a valid GFN.

Oh, right.  Should have taken a closer look, sorry for the noise:

Reviewed-by: George Dunlap <george.dunlap@citrix.com>

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