[PATCH v5 01/40] drm/gpuvm: Don't require obj lock in destructor path

[Danilo Krummrich]

On Mon, May 19, 2025 at 10:51:24AM -0700, Rob Clark wrote:
> From: Rob Clark <robdclark at chromium.org>
> 
> See commit a414fe3a2129 ("drm/msm/gem: Drop obj lock in
> msm_gem_free_object()") for justification.

I asked for a proper commit message in v4.

Only referring to a driver commit and let the people figure out how the driver
works and what it does in order to motivate a change in the generic
infrastructure is simply unreasonable.

> Cc: Danilo Krummrich <dakr at kernel.org>
> Signed-off-by: Rob Clark <robdclark at chromium.org>
> ---
>  drivers/gpu/drm/drm_gpuvm.c | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/gpu/drm/drm_gpuvm.c b/drivers/gpu/drm/drm_gpuvm.c
> index f9eb56f24bef..1e89a98caad4 100644
> --- a/drivers/gpu/drm/drm_gpuvm.c
> +++ b/drivers/gpu/drm/drm_gpuvm.c
> @@ -1511,7 +1511,9 @@ drm_gpuvm_bo_destroy(struct kref *kref)
>  	drm_gpuvm_bo_list_del(vm_bo, extobj, lock);
>  	drm_gpuvm_bo_list_del(vm_bo, evict, lock);
>  
> -	drm_gem_gpuva_assert_lock_held(obj);
> +	if (kref_read(&obj->refcount) > 0)
> +		drm_gem_gpuva_assert_lock_held(obj);

Again, this is broken. What if the reference count drops to zero right after
the kref_read() check, but before drm_gem_gpuva_assert_lock_held() is called?

Putting conditionals on a refcount is always suspicious.

If you still really want this, please guard it with

	if (unlikely(gpuvm->flags & DRM_GPUVM_MSM_LEGACY_QUIRK))

and get an explicit waiver from Dave / Sima.