On Mon, Jan 11, 2021 at 03:00:09PM +0200, Mikko Perttunen wrote:
> Add a callback field to the job structure, to be called just before
> the job is to be freed. This allows the job's submitter to clean
> up any of its own state, like decrement runtime PM refcounts.
> 
> Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
> ---
>  drivers/gpu/host1x/job.c | 3 +++
>  include/linux/host1x.h   | 4 ++++
>  2 files changed, 7 insertions(+)
> 
> diff --git a/drivers/gpu/host1x/job.c b/drivers/gpu/host1x/job.c
> index 8f59b34672c2..09097e19c0d0 100644
> --- a/drivers/gpu/host1x/job.c
> +++ b/drivers/gpu/host1x/job.c
> @@ -79,6 +79,9 @@ static void job_free(struct kref *ref)
>  {
>  	struct host1x_job *job = container_of(ref, struct host1x_job, ref);
>  
> +	if (job->release)
> +		job->release(job);
> +
>  	if (job->waiter)
>  		host1x_intr_put_ref(job->syncpt->host, job->syncpt->id,
>  				    job->waiter, false);
> diff --git a/include/linux/host1x.h b/include/linux/host1x.h
> index 81ca70066c76..d48cab563d5c 100644
> --- a/include/linux/host1x.h
> +++ b/include/linux/host1x.h
> @@ -265,6 +265,10 @@ struct host1x_job {
>  
>  	/* Fast-forward syncpoint increments on job timeout */
>  	bool syncpt_recovery;
> +
> +	/* Callback called when job is freed */
> +	void (*release)(struct host1x_job *job);
> +	void *user_data;

It's not clean to me what the user_data is used for. It's not used in
this patch at all, but perhaps it'll become relevant later on. I guess
I'll see.

Thierry