Re: [PATCH v2 16/17] scsi: ufs: qcom: Use ufshcd_rmwl() where applicable

[Manivannan Sadhasivam]

On Fri, Dec 08, 2023 at 12:02:03PM -0600, Andrew Halaney wrote:
> On Fri, Dec 08, 2023 at 12:29:01PM +0530, Manivannan Sadhasivam wrote:
> > Instead of using both ufshcd_readl() and ufshcd_writel() to read/modify/
> > write a register, let's make use of the existing helper.
> > 
> > Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@xxxxxxxxxx>
> > ---
> >  drivers/ufs/host/ufs-qcom.c | 12 ++++--------
> >  drivers/ufs/host/ufs-qcom.h |  3 +++
> >  2 files changed, 7 insertions(+), 8 deletions(-)
> > 
> > diff --git a/drivers/ufs/host/ufs-qcom.c b/drivers/ufs/host/ufs-qcom.c
> > index 26aa8904c823..549a08645391 100644
> > --- a/drivers/ufs/host/ufs-qcom.c
> > +++ b/drivers/ufs/host/ufs-qcom.c
> > @@ -387,9 +387,8 @@ static int ufs_qcom_power_up_sequence(struct ufs_hba *hba)
> >   */
> >  static void ufs_qcom_enable_hw_clk_gating(struct ufs_hba *hba)
> >  {
> > -	ufshcd_writel(hba,
> > -		ufshcd_readl(hba, REG_UFS_CFG2) | REG_UFS_CFG2_CGC_EN_ALL,
> > -		REG_UFS_CFG2);
> > +	ufshcd_rmwl(hba, REG_UFS_CFG2_CGC_EN_ALL, REG_UFS_CFG2_CGC_EN_ALL,
> > +		    REG_UFS_CFG2);
> >  
> >  	/* Ensure that HW clock gating is enabled before next operations */
> >  	mb();
> > @@ -1689,11 +1688,8 @@ static int ufs_qcom_config_esi(struct ufs_hba *hba)
> >  		platform_msi_domain_free_irqs(hba->dev);
> >  	} else {
> >  		if (host->hw_ver.major == 6 && host->hw_ver.minor == 0 &&
> > -		    host->hw_ver.step == 0) {
> > -			ufshcd_writel(hba,
> > -				      ufshcd_readl(hba, REG_UFS_CFG3) | 0x1F000,
> > -				      REG_UFS_CFG3);
> > -		}
> > +		    host->hw_ver.step == 0)
> > +			ufshcd_rmwl(hba, ESI_VEC_MASK, 0x1f00, REG_UFS_CFG3);
> 
> Is this right? I feel like you accidentally just shifted what was written
> prior >> 4 bits.
> 
> Before: 0x1f000
> After:  0x01f00
> 

Doh! you are right, I accidentally missed one 0. Since the series is already
merged, I will send a follow up patch.

> >  		ufshcd_mcq_enable_esi(hba);
> >  	}
> >  
> > diff --git a/drivers/ufs/host/ufs-qcom.h b/drivers/ufs/host/ufs-qcom.h
> > index 385480499e71..2ce63a1c7f2f 100644
> > --- a/drivers/ufs/host/ufs-qcom.h
> > +++ b/drivers/ufs/host/ufs-qcom.h
> > @@ -102,6 +102,9 @@ enum {
> >  #define TMRLUT_HW_CGC_EN	BIT(6)
> >  #define OCSC_HW_CGC_EN		BIT(7)
> >  
> > +/* bit definitions for REG_UFS_CFG3 register */
> > +#define ESI_VEC_MASK		GENMASK(22, 12)
> > +
> 
> I'll leave it to someone with the docs to review this field. It at least
> contains the bits set above, fwiw. It would be neat to see that
> converted to using the field + a FIELD_PREP to set the bits IMO, but I
> honestly don't have a lot of experience using those APIs so feel free to
> reject that.
> 

Fair point. The reason for hardcoding the value in this patch was I didn't get
information on how the value works. But now I got that info, so will address
this in the follow up patch I mentioned above.

- Mani

> >  /* bit definitions for REG_UFS_PARAM0 */
> >  #define MAX_HS_GEAR_MASK	GENMASK(6, 4)
> >  #define UFS_QCOM_MAX_GEAR(x)	FIELD_GET(MAX_HS_GEAR_MASK, (x))
> > -- 
> > 2.25.1
> > 
> 

-- 
மணிவண்ணன் சதாசிவம்