Re: [PATCH v2 6/8] drm/msm: Attach a driver to the GMU

[Konrad Dybcio]

On 10/30/26 6:09 PM, Akhil P Oommen wrote:
> With the introduction of sync_state in the genpd framework, any consumer
> device of GCC and GPUCC which is not bound to a driver will result in
> bootup warnings like below:
> 
> [   24.362666] gcc-kaanapali 100000.clock-controller: sync_state() pending due to 3d37000.gmu
> [   24.371210] gxclkctl-kaanapali 3d64000.clock-controller: sync_state() pending due to 3d37000.gmu
> [   24.380268] gpucc-kaanapali 3d90000.clock-controller: sync_state() pending due to 3d37000.gmu
> 
> To silence these warnings and also to have a proper state in driver core,
> attach a driver to the GMU and set it up as a component device for the drm
> master device.
> 
> Since both GPU and GMU should be bound to a driver before a
> devfreq/thermal mitigation request can be handled, move the devfreq
> registration to the bind call of the master device.

Can we do that in a preceding patch to make this one smaller?

[...]

> @@ -2308,6 +2298,8 @@ int a6xx_gmu_wrapper_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node)
>  		goto err_mmio;
>  	}
>  
> +	mutex_init(&gmu->lock);

Can we do the common init (just like this mutex init) in the common
function?

Sashiko has a number of comments on this patch

Konrad