|
|
|
|
|
|
|
|
|
|
xen-devel
Re: [Xen-devel] question on iommu _map_page
On May 28, 2010, at 3:08 AM, Keir Fraser wrote:
> On 27/05/2010 21:59, "Kaushik Kumar Ram" <kaushik@xxxxxxxx> wrote:
>
>> On May 27, 2010, at 3:46 PM, Keir Fraser wrote:
>>
>>> It looks like the person who implemented that never needed the read-only
>>> case. We need a function to map pages into the iommu read-only and to call
>>> that from the grant code.
>>>
>>> A patch to just call iommu_map_page() for any kind of grant mapping would be
>>> acceptable for now, if you want to submit a patch.
>>
>> OK. Attached.
>
> Actually I've implemented the better fix as xen-unstable:21476. Hope that
> works okay for you.
Checking if act->pin is zero before calling iommu_unmap_page() (in
grant_table.c) is not sufficient since there can be multiple
active grants all referring to the same mfn. In fact I came across iommu page
faults because pages were
getting unmapped from the IOMMU when active grants referring to these pages
were still around.
Ideally, there needs to be a per-page count of how many IOMMU mappings exists
for a page.
I can't think of an obvious fix for this problem. For my purposes, I hacked
page_info to add another counter which I guess is not an acceptable solution!
-Kaushik
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
<Prev in Thread] |
Current Thread |
[Next in Thread>
|
- Re: [Xen-devel] question on iommu _map_page,
Kaushik Kumar Ram <=
|
|
|
|
|