Re: [PATCH v4 21/33] drm/msm: Add _NO_SHARE flag

["Christian König" <christian.koenig@amd.com>]

On 5/5/25 16:15, Rob Clark wrote:
> On Mon, May 5, 2025 at 12:54 AM Christian König
> <christian.koenig@amd.com> wrote:
>>
>> On 5/2/25 18:56, Rob Clark wrote:
>>> From: Rob Clark <robdclark@chromium.org>
>>>
>>> Buffers that are not shared between contexts can share a single resv
>>> object.  This way drm_gpuvm will not track them as external objects, and
>>> submit-time validating overhead will be O(1) for all N non-shared BOs,
>>> instead of O(n).
>>>
>>> Signed-off-by: Rob Clark <robdclark@chromium.org>
>>> ---
>>>  drivers/gpu/drm/msm/msm_drv.h       |  1 +
>>>  drivers/gpu/drm/msm/msm_gem.c       | 23 +++++++++++++++++++++++
>>>  drivers/gpu/drm/msm/msm_gem_prime.c | 15 +++++++++++++++
>>>  include/uapi/drm/msm_drm.h          | 14 ++++++++++++++
>>>  4 files changed, 53 insertions(+)
>>>
>>> diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h
>>> index b77fd2c531c3..b0add236cbb3 100644
>>> --- a/drivers/gpu/drm/msm/msm_drv.h
>>> +++ b/drivers/gpu/drm/msm/msm_drv.h
>>> @@ -246,6 +246,7 @@ int msm_gem_prime_vmap(struct drm_gem_object *obj, struct iosys_map *map);
>>>  void msm_gem_prime_vunmap(struct drm_gem_object *obj, struct iosys_map *map);
>>>  struct drm_gem_object *msm_gem_prime_import_sg_table(struct drm_device *dev,
>>>               struct dma_buf_attachment *attach, struct sg_table *sg);
>>> +struct dma_buf *msm_gem_prime_export(struct drm_gem_object *obj, int flags);
>>>  int msm_gem_prime_pin(struct drm_gem_object *obj);
>>>  void msm_gem_prime_unpin(struct drm_gem_object *obj);
>>>
>>> diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c
>>> index 3708d4579203..d0f44c981351 100644
>>> --- a/drivers/gpu/drm/msm/msm_gem.c
>>> +++ b/drivers/gpu/drm/msm/msm_gem.c
>>> @@ -532,6 +532,9 @@ static int get_and_pin_iova_range_locked(struct drm_gem_object *obj,
>>>
>>>       msm_gem_assert_locked(obj);
>>>
>>> +     if (to_msm_bo(obj)->flags & MSM_BO_NO_SHARE)
>>> +             return -EINVAL;
>>> +
>>>       vma = get_vma_locked(obj, vm, range_start, range_end);
>>>       if (IS_ERR(vma))
>>>               return PTR_ERR(vma);
>>> @@ -1060,6 +1063,16 @@ static void msm_gem_free_object(struct drm_gem_object *obj)
>>>               put_pages(obj);
>>>       }
>>>
>>> +     if (msm_obj->flags & MSM_BO_NO_SHARE) {
>>> +             struct drm_gem_object *r_obj =
>>> +                     container_of(obj->resv, struct drm_gem_object, _resv);
>>> +
>>> +             BUG_ON(obj->resv == &obj->_resv);
>>> +
>>> +             /* Drop reference we hold to shared resv obj: */
>>> +             drm_gem_object_put(r_obj);
>>> +     }
>>> +
>>>       drm_gem_object_release(obj);
>>>
>>>       kfree(msm_obj->metadata);
>>> @@ -1092,6 +1105,15 @@ int msm_gem_new_handle(struct drm_device *dev, struct drm_file *file,
>>>       if (name)
>>>               msm_gem_object_set_name(obj, "%s", name);
>>>
>>> +     if (flags & MSM_BO_NO_SHARE) {
>>> +             struct msm_context *ctx = file->driver_priv;
>>> +             struct drm_gem_object *r_obj = drm_gpuvm_resv_obj(ctx->vm);
>>> +
>>> +             drm_gem_object_get(r_obj);
>>> +
>>> +             obj->resv = r_obj->resv;
>>> +     }
>>> +
>>>       ret = drm_gem_handle_create(file, obj, handle);
>>>
>>>       /* drop reference from allocate - handle holds it now */
>>> @@ -1124,6 +1146,7 @@ static const struct drm_gem_object_funcs msm_gem_object_funcs = {
>>>       .free = msm_gem_free_object,
>>>       .open = msm_gem_open,
>>>       .close = msm_gem_close,
>>> +     .export = msm_gem_prime_export,
>>>       .pin = msm_gem_prime_pin,
>>>       .unpin = msm_gem_prime_unpin,
>>>       .get_sg_table = msm_gem_prime_get_sg_table,
>>> diff --git a/drivers/gpu/drm/msm/msm_gem_prime.c b/drivers/gpu/drm/msm/msm_gem_prime.c
>>> index ee267490c935..1a6d8099196a 100644
>>> --- a/drivers/gpu/drm/msm/msm_gem_prime.c
>>> +++ b/drivers/gpu/drm/msm/msm_gem_prime.c
>>> @@ -16,6 +16,9 @@ struct sg_table *msm_gem_prime_get_sg_table(struct drm_gem_object *obj)
>>>       struct msm_gem_object *msm_obj = to_msm_bo(obj);
>>>       int npages = obj->size >> PAGE_SHIFT;
>>>
>>> +     if (msm_obj->flags & MSM_BO_NO_SHARE)
>>> +             return ERR_PTR(-EINVAL);
>>> +
>>>       if (WARN_ON(!msm_obj->pages))  /* should have already pinned! */
>>>               return ERR_PTR(-ENOMEM);
>>>
>>> @@ -45,6 +48,15 @@ struct drm_gem_object *msm_gem_prime_import_sg_table(struct drm_device *dev,
>>>       return msm_gem_import(dev, attach->dmabuf, sg);
>>>  }
>>>
>>> +
>>> +struct dma_buf *msm_gem_prime_export(struct drm_gem_object *obj, int flags)
>>> +{
>>> +     if (to_msm_bo(obj)->flags & MSM_BO_NO_SHARE)
>>> +             return ERR_PTR(-EPERM);
>>> +
>>> +     return drm_gem_prime_export(obj, flags);
>>> +}
>>> +
>>>  int msm_gem_prime_pin(struct drm_gem_object *obj)
>>>  {
>>>       struct page **pages;
>>> @@ -53,6 +65,9 @@ int msm_gem_prime_pin(struct drm_gem_object *obj)
>>>       if (obj->import_attach)
>>>               return 0;
>>>
>>> +     if (to_msm_bo(obj)->flags & MSM_BO_NO_SHARE)
>>> +             return -EINVAL;
>>> +
>>>       pages = msm_gem_pin_pages_locked(obj);
>>>       if (IS_ERR(pages))
>>>               ret = PTR_ERR(pages);
>>> diff --git a/include/uapi/drm/msm_drm.h b/include/uapi/drm/msm_drm.h
>>> index b974f5a24dbc..1bccc347945c 100644
>>> --- a/include/uapi/drm/msm_drm.h
>>> +++ b/include/uapi/drm/msm_drm.h
>>> @@ -140,6 +140,19 @@ struct drm_msm_param {
>>>
>>>  #define MSM_BO_SCANOUT       0x00000001     /* scanout capable */
>>>  #define MSM_BO_GPU_READONLY  0x00000002
>>> +/* Private buffers do not need to be explicitly listed in the SUBMIT
>>> + * ioctl, unless referenced by a drm_msm_gem_submit_cmd.  Private
>>> + * buffers may NOT be imported/exported or used for scanout (or any
>>> + * other situation where buffers can be indefinitely pinned, but
>>> + * cases other than scanout are all kernel owned BOs which are not
>>> + * visible to userspace).
>>
>> Why is pinning for scanout a problem with those?
>>
>> Maybe I missed something but for other drivers that doesn't seem to be a problem.
> 
> I guess _technically_ it could be ok because we track pin-count
> separately from dma_resv.  But the motivation for that statement was
> simply that _NO_SHARE buffers share a resv obj with the VM, so they
> should not be used in a different VM (in this case, the display, which
> has it's own VM).

Ah, yes that makes perfect sense.

You should indeed avoid importing the BO into a different VM when it shares the reservation object with it. That will only cause trouble.

But at least amdgpu/radeon and I think i915 as well don't need to do that. Scanout is just separate from all VMs.

Regards,
Christian.


> 
> BR,
> -R
> 
>> Regards,
>> Christian.
>>
>>
>>> + *
>>> + * In exchange for those constraints, all private BOs associated with
>>> + * a single context (drm_file) share a single dma_resv, and if there
>>> + * has been no eviction since the last submit, there are no per-BO
>>> + * bookeeping to do, significantly cutting the SUBMIT overhead.
>>> + */
>>> +#define MSM_BO_NO_SHARE      0x00000004
>>>  #define MSM_BO_CACHE_MASK    0x000f0000
>>>  /* cache modes */
>>>  #define MSM_BO_CACHED        0x00010000
>>> @@ -149,6 +162,7 @@ struct drm_msm_param {
>>>
>>>  #define MSM_BO_FLAGS         (MSM_BO_SCANOUT | \
>>>                                MSM_BO_GPU_READONLY | \
>>> +                              MSM_BO_NO_SHARE | \
>>>                                MSM_BO_CACHE_MASK)
>>>
>>>  struct drm_msm_gem_new {
>>