status of the calgary iommu driver

status of the calgary iommu driver

[Christoph Hellwig]

Hi Muli and Jon,

do you know if there are user of systems with the Calgary iommu
around still? It seems like the last non-drive by changes to it
are from 2010 and I'm not sure how common these systems were.

Re: status of the calgary iommu driver

[Jon Mason]

On Tue, Apr 9, 2019 at 3:03 PM Christoph Hellwig <hch@lst.de> wrote:
>
> Hi Muli and Jon,
>
> do you know if there are user of systems with the Calgary iommu
> around still? It seems like the last non-drive by changes to it
> are from 2010 and I'm not sure how common these systems were.


These systems were plentiful for 2-4 years after the original series
made it in.  After that, the Intel and AMD IOMMUs should were shipped
and were superior to this chip.  So, even in systems where these might
be present, the AMD/Intel ones should be used (unknown if they were
shipped on the same ones, as both Muli and I have left IBM).

You thinking about removing the code?

Thanks,
Jon

Re: status of the calgary iommu driver

[Christoph Hellwig]

On Wed, May 08, 2019 at 06:42:39PM +0100, Jon Mason wrote:
> These systems were plentiful for 2-4 years after the original series
> made it in.  After that, the Intel and AMD IOMMUs should were shipped
> and were superior to this chip.  So, even in systems where these might
> be present, the AMD/Intel ones should be used (unknown if they were
> shipped on the same ones, as both Muli and I have left IBM).
> 
> You thinking about removing the code?

I'm wondering if we could remove it.  I've been done lots of
maintainance on various dma mapping and iommu drivers, and the calgary
one seems like it didn't get a whole lot of love, and I've not seen
any recent users, so I mostly wonder if I should bother at all.

Re: status of the calgary iommu driver

[Jon Mason]

On Wed, May 8, 2019 at 6:52 PM Christoph Hellwig <hch@lst.de> wrote:
>
> On Wed, May 08, 2019 at 06:42:39PM +0100, Jon Mason wrote:
> > These systems were plentiful for 2-4 years after the original series
> > made it in.  After that, the Intel and AMD IOMMUs should were shipped
> > and were superior to this chip.  So, even in systems where these might
> > be present, the AMD/Intel ones should be used (unknown if they were
> > shipped on the same ones, as both Muli and I have left IBM).
> >
> > You thinking about removing the code?
>
> I'm wondering if we could remove it.  I've been done lots of
> maintainance on various dma mapping and iommu drivers, and the calgary
> one seems like it didn't get a whole lot of love, and I've not seen
> any recent users, so I mostly wonder if I should bother at all.


I do have a system.  So, it could be tested.  However given the age of
the HW, I would say it is not worth the effort to update and it would
be best to remove it from the kernel.

I can send a patch to do this, unless you would prefer to do it (or
already have something handy).

Thanks,
Jon

Re: status of the calgary iommu driver

[Christoph Hellwig]

On Wed, May 08, 2019 at 06:56:46PM +0100, Jon Mason wrote:
> I do have a system.  So, it could be tested.  However given the age of
> the HW, I would say it is not worth the effort to update and it would
> be best to remove it from the kernel.
> 
> I can send a patch to do this, unless you would prefer to do it (or
> already have something handy).

I don't have anything, and at least to me it is not urgent.  So feel
free to send it.

Re: status of the calgary iommu driver

[Christoph Hellwig]

On Wed, May 08, 2019 at 06:56:46PM +0100, Jon Mason wrote:
> I do have a system.  So, it could be tested.  However given the age of
> the HW, I would say it is not worth the effort to update and it would
> be best to remove it from the kernel.
> 
> I can send a patch to do this, unless you would prefer to do it (or
> already have something handy).

Do you still plan to send that patch?