Hi Jerry,
On 11/19/21 3:48 AM, Jerry Snitselaar wrote:
On Fri, 2021-11-12 at 10:59 +0800, Lu Baolu wrote:
Hi Alex,
On 11/11/21 8:32 AM, Alex Williamson wrote:
When supporting only the .map and .unmap callbacks of iommu_ops,
the IOMMU driver can make assumptions about the size and alignment
On Fri, 2021-11-12 at 10:59 +0800, Lu Baolu wrote:
> Hi Alex,
>
> On 11/11/21 8:32 AM, Alex Williamson wrote:
> > When supporting only the .map and .unmap callbacks of iommu_ops,
> > the IOMMU driver can make assumptions about the size and alignment
> > used for mappings based on the driver
Hi Alex,
On 11/11/21 8:32 AM, Alex Williamson wrote:
When supporting only the .map and .unmap callbacks of iommu_ops,
the IOMMU driver can make assumptions about the size and alignment
used for mappings based on the driver provided pgsize_bitmap. VT-d
previously used essentially PAGE_MASK for
I was able to hit the issue that this patch is fixing using the memory
allocator in QATlib (https://github.com/intel/qatlib/) which uses vfio
to map and unmap memory.
This patch fixes the problem.
On Wed, Nov 10, 2021 at 05:32:50PM -0700, Alex Williamson wrote:
> When supporting only the .map
Thanks Alex, Baolu.
The patch fixes things at my end. No kernel-flooding is seen now
(tested starting/stopping vm > 10 times).
Thanks and Regards,
Ajay
On Thu, Nov 11, 2021 at 6:03 AM Alex Williamson
wrote:
>
> When supporting only the .map and .unmap callbacks of iommu_ops,
> the IOMMU driver
When supporting only the .map and .unmap callbacks of iommu_ops,
the IOMMU driver can make assumptions about the size and alignment
used for mappings based on the driver provided pgsize_bitmap. VT-d
previously used essentially PAGE_MASK for this bitmap as any power
of two mapping was acceptably