Re: [Xen-devel] [RFC PATCH 2/9] iommu: Add ability to map/unmap the number of pages

[Oleksandr Tyshchenko <olekstysh@xxxxxxxxx>]

On Thu, Mar 23, 2017 at 11:07 AM, Jan Beulich <JBeulich@xxxxxxxx> wrote:
>>>> On 22.03.17 at 19:01, <olekstysh@xxxxxxxxx> wrote:
>> Hi Jan
>>
>> On Wed, Mar 22, 2017 at 5:44 PM, Jan Beulich <JBeulich@xxxxxxxx> wrote:
>>>>>> On 15.03.17 at 21:05, <olekstysh@xxxxxxxxx> wrote:
>>>> From: Oleksandr Tyshchenko <oleksandr_tyshchenko@xxxxxxxx>
>>>>
>>>> Extend the IOMMU code with new APIs and platform callbacks.
>>>> These new map_pages/unmap_pages API do almost the same thing
>>>> as existing map_page/unmap_page ones except the formers can
>>>> handle the number of pages. So do new platform callbacks.
>>>>
>>>> Currently, this patch requires to modify neither
>>>> existing IOMMU drivers nor P2M code.
>>>> But, the patch might be rewritten to replace existing
>>>> single-page stuff with the multi-page one followed by modifications
>>>> of all related parts.
>>>
>>> Yes please. However, unless you strictly need a page count, please
>>> instead use an "order" parameter. Doing that has been on my todo
>>> list for quite a while.
>>
>> The patch introduces new iommu_map_pages/iommu_unmap_pages APIs as well
>> as new map_pages/unmap_pages platform callbacks.
>>
>> So, we just need to replace single-page APIs with multi-page ones, but
>> leave both "new" (map_pages/unmap_pages) and "old"
>> (map_page/unmap_page) platform callbacks.
>> This way doesn't require to modify existing IOMMU drivers right now,
>> just P2M code. Or we need to replace platform callbacks too?
>
> That was the "yes please" part of my answer: I don't see why we
> would want two API / callback flavors, with one being a strict
> superset of the other.

Agree. I'll be back if I have questions that need to be clarified.

>
> Jan
>


-- 
Regards,

Oleksandr Tyshchenko

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