On Fri, Feb 21, 2025 at 11:42:06AM +, Jean-Philippe Brucker wrote:
> > There is no requirement that an iommu driver implement identity, this
> > is all an optimization. If it doesn't work in alot of cases can we
> > just remove it? It would simplfy a lot..
>
> No objection. Identity domains a
On Wed, Feb 12, 2025 at 08:50:07AM -0400, Jason Gunthorpe wrote:
> On Tue, Feb 11, 2025 at 04:41:51PM -0800, Jacob Pan wrote:
>
> > I see the same on arm64 with v9.0, assigned an ixgbe nic via VFIO.
>
> Huh, I figured it worked on ARM..
Crosvm would be a more useful test because it doesn't imple
On 2025-02-12 6:50 pm, Jacob Pan wrote:
Hi Jason,
On Wed, 12 Feb 2025 08:50:07 -0400
Jason Gunthorpe wrote:
On Tue, Feb 11, 2025 at 04:41:51PM -0800, Jacob Pan wrote:
I see the same on arm64 with v9.0, assigned an ixgbe nic via VFIO.
Huh, I figured it worked on ARM..
There is no requirem
On Wed, Feb 12, 2025 at 09:05:35AM -0800, Jacob Pan wrote:
> > Does reproducing this require
> > modifying QEMU or Linux to disable support for F_BYPASS_CONFIG, or
> > just passing QEMU/kernel parameters?
> I only modified guest code to pretend F_BYPASS_CONFIG is not supported,
> a couple of plac
Hi Jason,
On Wed, 12 Feb 2025 08:50:07 -0400
Jason Gunthorpe wrote:
> On Tue, Feb 11, 2025 at 04:41:51PM -0800, Jacob Pan wrote:
>
> > I see the same on arm64 with v9.0, assigned an ixgbe nic via VFIO.
>
> Huh, I figured it worked on ARM..
>
> There is no requirement that an iommu driver im
Hi Jean-Philippe,
On Wed, 12 Feb 2025 11:58:45 +
Jean-Philippe Brucker wrote:
> Hi Jacob,
>
> On Tue, Feb 11, 2025 at 04:41:51PM -0800, Jacob Pan wrote:
> > > However, there seem to be some unrelated qemu bugs. 7.0.0 does not
> > > have working VIRTIO_IOMMU_F_BYPASS_CONFIG and neither qemu
On Tue, Feb 11, 2025 at 04:41:51PM -0800, Jacob Pan wrote:
> I see the same on arm64 with v9.0, assigned an ixgbe nic via VFIO.
Huh, I figured it worked on ARM..
There is no requirement that an iommu driver implement identity, this
is all an optimization. If it doesn't work in alot of cases can
Hi Jacob,
On Tue, Feb 11, 2025 at 04:41:51PM -0800, Jacob Pan wrote:
> > However, there seem to be some unrelated qemu bugs. 7.0.0 does not
> > have working VIRTIO_IOMMU_F_BYPASS_CONFIG and neither qemu works with
> > an IDENTITY domain using !VIRTIO_IOMMU_F_BYPASS_CONFIG. It prints:
> >
> > qem
Hi Jason,
On Fri, 7 Feb 2025 10:46:00 -0400
Jason Gunthorpe wrote:
> virtio-iommu and fsl_pamu are the only two drivers left that still
> rely on ops.domain_alloc()
>
> Update virtio-iommu to have a global static identity domain, implement
> domain_alloc_paging(), and finalize its domain durin
virtio-iommu and fsl_pamu are the only two drivers left that still
rely on ops.domain_alloc()
Update virtio-iommu to have a global static identity domain, implement
domain_alloc_paging(), and finalize its domain during allocation instead
of on first attach.
As virtio-iommu was the last real iommu
10 matches
Mail list logo
