Re: [PATCH V2 6/6] scsi: ufs: host : Introduce phy_power_on/off wrapper function

[Bart Van Assche]

On 3/18/25 7:49 AM, Nitin Rawat wrote:

Just like the other patches in this series, the subject of this patch
should have the prefix "scsi: ufs: qcom:" instead of "scsi: ufs: host:"

> diff --git a/drivers/ufs/host/ufs-qcom.h b/drivers/ufs/host/ufs-qcom.h
> index d0e6ec9128e7..3db29fbcd40b 100644
> --- a/drivers/ufs/host/ufs-qcom.h
> +++ b/drivers/ufs/host/ufs-qcom.h
> @@ -252,6 +252,10 @@ struct ufs_qcom_host {
>   	u32 phy_gear;
>
>   	bool esi_enabled;
> +	/* flag to check if phy is powered on */
> +	bool is_phy_pwr_on;
> +	/* Protect the usage of is_phy_pwr_on against racing */
> +	struct mutex phy_mutex;
>   };

Please reorder the above two structure members. Synchronization objects
should occur before the data members protected by these synchronization
objects in structure definitions.

It seems to me that phy_mutex not only serializes is_phy_pwr_on accesses
but that it also serializes phy_power_on() / phy_power_off() calls. If
this is the case, please mention this.

Thanks,

Bart.