Re: [PATCH v7 2/3] ufs: host: Add ICE clock scaling during UFS clock changes

[Harshal Dev <harshal.dev@oss.qualcomm.com>]

Hi Abhinaba,

On 3/2/2026 4:19 PM, Abhinaba Rakshit wrote:
> Implement ICE (Inline Crypto Engine) clock scaling in sync with
> UFS controller clock scaling. This ensures that the ICE operates at
> an appropriate frequency when the UFS clocks are scaled up or down,
> improving performance and maintaining stability for crypto operations.
> 
> For scale_up operation ensure to pass ~round_ceil (round_floor)
> and vice-versa for scale_down operations.
> 
> Incase of OPP scaling is not supported by ICE, ensure to not prevent
> devfreq for UFS, as ICE OPP-table is optional.
> 
> Acked-by: Manivannan Sadhasivam <mani@kernel.org>
> Signed-off-by: Abhinaba Rakshit <abhinaba.rakshit@oss.qualcomm.com>
> ---
>  drivers/ufs/host/ufs-qcom.c | 19 ++++++++++++++++++-
>  1 file changed, 18 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/ufs/host/ufs-qcom.c b/drivers/ufs/host/ufs-qcom.c
> index 8d119b3223cbdaa3297d2beabced0962a1a847d5..776444f46fe5f00f947e4b0b4dfe6d64e2ad2150 100644
> --- a/drivers/ufs/host/ufs-qcom.c
> +++ b/drivers/ufs/host/ufs-qcom.c
> @@ -305,6 +305,15 @@ static int ufs_qcom_ice_prepare_key(struct blk_crypto_profile *profile,
>  	return qcom_ice_prepare_key(host->ice, lt_key, lt_key_size, eph_key);
>  }
>  
> +static int ufs_qcom_ice_scale_clk(struct ufs_qcom_host *host, unsigned long target_freq,
> +				  bool round_ceil)
> +{
> +	if (host->hba->caps & UFSHCD_CAP_CRYPTO)
> +		return qcom_ice_scale_clk(host->ice, target_freq, round_ceil);
> +
> +	return 0;
> +}
> +
>  static const struct blk_crypto_ll_ops ufs_qcom_crypto_ops = {
>  	.keyslot_program	= ufs_qcom_ice_keyslot_program,
>  	.keyslot_evict		= ufs_qcom_ice_keyslot_evict,
> @@ -339,6 +348,12 @@ static void ufs_qcom_config_ice_allocator(struct ufs_qcom_host *host)
>  {
>  }
>  
> +static int ufs_qcom_ice_scale_clk(struct ufs_qcom_host *host, unsigned long target_freq,
> +				  bool round_ceil)
> +{
> +	return 0;
> +}
> +
>  #endif
>  
>  static void ufs_qcom_disable_lane_clks(struct ufs_qcom_host *host)
> @@ -1646,8 +1661,10 @@ static int ufs_qcom_clk_scale_notify(struct ufs_hba *hba, bool scale_up,
>  		else
>  			err = ufs_qcom_clk_scale_down_post_change(hba, target_freq);
>  
> +		if (!err)
> +			err = ufs_qcom_ice_scale_clk(host, target_freq, !scale_up);
>  
> -		if (err) {
> +		if (err && err != -EOPNOTSUPP) {

Using -EOPNOTSUPP here works fine for now. But anyone touching any of the lower APIs called by
ufs_qcom_clk_scale_up/down_post_change() needs to ensure they don't return -EOPNOTSUPP, otherwise
hibernate exit will be skipped. So this carries a minor risk of breaking.

Since regardless of whether ufs_qcom_clk_scale_up/down_post_change() fails or ufs_qcom_ice_scale_clk()
fails, we exit from hibernate and return from this function, I suggest you handle the error for ice_scale
separately.

>  			ufshcd_uic_hibern8_exit(hba);
>  			return err;
>  		}
> 

Add the call to ufs_qcom_ice_scale_clk() along with error handle here, and let the above error handle
remain untouched.

		err = ufs_qcom_ice_scale_clk(host, target_freq, !scale_up);
		if (err && err != -EOPNOTSUPP) {
			ufshcd_uic_hibern8_exit(hba);
  			return err;
  		}

Regards,
Harshal