On 24.06.25 10:45, Prike Liang wrote:
> The user queue object destroy requires ensuring its
> VA has been unmapped prior to the queue being destroyed.
> Otherwise, the kernel driver should report an invalidated
> error to the user IOCLT request.

Well that is completely nonsense.

It must be just the other way around, you need to destroy the queue first and 
*then* unmap the VAs!

Regards,
Christian.

> 
> Signed-off-by: Prike Liang <[email protected]>
> ---
>  drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 15 +++++++++++++++
>  1 file changed, 15 insertions(+)
> 
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c 
> b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
> index cbbd1860a69a..79b263c18eb1 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
> @@ -477,12 +477,24 @@ amdgpu_userq_destroy(struct drm_file *filp, int 
> queue_id)
>               amdgpu_bo_unpin(queue->db_obj.obj);
>               amdgpu_bo_unreserve(queue->db_obj.obj);
>       }
> +
> +     /*
> +      * At this point the userq obj va should be unmapped,
> +      * otherwise will return error to user.
> +      */
> +     if (amdgpu_userq_buffer_vas_mapped(&fpriv->vm, queue)) {
> +             drm_warn(adev_to_drm(uq_mgr->adev), "the userq obj va shouldn't 
> be mapped during destroying userq\n");
> +             r=  -EINVAL;
> +     }
> +
>       r = amdgpu_userq_unmap_helper(uq_mgr, queue);
>       /*TODO: It requires a reset for unmap error*/
>       if (r != AMDGPU_USERQ_STATE_UNMAPPED) {
>               drm_warn(adev_to_drm(uq_mgr->adev), "trying to destroy a HW 
> mapping userq\n");
>               r = -ETIMEDOUT;
>       }
> +
> +     amdgpu_userq_buffer_vas_put(&fpriv->vm, queue);
>       amdgpu_userq_cleanup(uq_mgr, queue, queue_id);
>       mutex_unlock(&uq_mgr->userq_mutex);
>  
> @@ -603,6 +615,9 @@ amdgpu_userq_create(struct drm_file *filp, union 
> drm_amdgpu_userq *args)
>               goto unlock;
>       }
>  
> +     /* refer to the userq objects vm bo*/
> +     amdgpu_userq_buffer_vas_get(queue->vm, queue);
> +
>       qid = idr_alloc(&uq_mgr->userq_idr, queue, 1, AMDGPU_MAX_USERQ_COUNT, 
> GFP_KERNEL);
>       if (qid < 0) {
>               drm_file_err(uq_mgr->file, "Failed to allocate a queue id\n");

Reply via email to