On 11/25/25 08:49, Dave Airlie wrote:> On Tue, 25 Nov 2025 at 17:45, Christian K?nig <christian.koenig at
amd.com> wrote:
>>
>> On 11/25/25 00:44, Joel Fernandes wrote:
>>> Move the DRM buddy allocator one level up so that it can be used by
GPU
>>> drivers (example, nova-core) that have usecases other than DRM
(such as
>>> VFIO vGPU support). Modify the API, structures and Kconfigs to use
>>> "gpu_buddy" terminology. Adapt the drivers and tests to
use the new API.
>>>
>>> The commit cannot be split due to bissectability, however no
functional
>>> change is intended. Verified by running K-UNIT tests and build
tested
>>> various configurations.
>>
>> Well such a change needs to be part of the patch set which actually
starts to use it outside of drivers/gpu/drm.
>
> I'm not sure it really makes sense yet to force that for this, since
> the next patch would be just adding rust bindings to it, then a patch
> to nova-core would be needed to use it, I'd rather not merge something
> like this via the rust or nova trees as it breaks the principle of
> least surprise.
>
> Maybe this + rust bindings might be enough to justify it, but it
> should be reviewable as a standalone idea.
My question is why exactly is nova separated into nova-core and nova-drm? That
doesn't seem to be necessary in the first place.
E.g. neither amdgpu not XE need that and we are doing pretty much the same
thing.
Regards,
Christian.
>
> Dave.