Re: [PATCH V2 2/2] scsi: ufs: introduce quirk to extend PA_HIBERN8TIME for UFS devices

[Manivannan Sadhasivam]

On Tue, Apr 08, 2025 at 10:58:10PM +0530, MANISH PANDEY wrote:
> 
> 
> On 4/8/2025 10:01 PM, Manivannan Sadhasivam wrote:
> > On Tue, Apr 08, 2025 at 01:14:50PM +0530, MANISH PANDEY wrote:
> > > 
> > > 
> > > On 4/8/2025 12:53 PM, Manivannan Sadhasivam wrote:
> > > > On Tue, Apr 08, 2025 at 11:07:58AM +0530, MANISH PANDEY wrote:
> > > > > 
> > > > > 
> > > > > On 4/7/2025 12:05 AM, Manivannan Sadhasivam wrote:
> > > > > > On Fri, Apr 04, 2025 at 11:15:39PM +0530, Manish Pandey wrote:
> > > > > > > Some UFS devices need additional time in hibern8 mode before exiting,
> > > > > > > beyond the negotiated handshaking phase between the host and device.
> > > > > > > Introduce a quirk to increase the PA_HIBERN8TIME parameter by 100 µs
> > > > > > > to ensure proper hibernation process.
> > > > > > > 
> > > > > > 
> > > > > > This commit message didn't mention the UFS device for which this quirk is being
> > > > > > applied.
> > > > > > 
> > > > > Since it's a quirk and may be applicable to other vendors also in future, so
> > > > > i thought to keep it general.
> > > > > 
> > > > 
> > > > You cannot make commit message generic. It should precisely describe the change.
> > > > 
> > > > > Will update in next patch set if required.
> > > > >    >> Signed-off-by: Manish Pandey <quic_mapa@xxxxxxxxxxx>
> > > > > > > ---
> > > > > > >     drivers/ufs/core/ufshcd.c | 31 +++++++++++++++++++++++++++++++
> > > > > > >     include/ufs/ufs_quirks.h  |  6 ++++++
> > > > > > >     2 files changed, 37 insertions(+)
> > > > > > > 
> > > > > > > diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c
> > > > > > > index 464f13da259a..2b8203fe7b8c 100644
> > > > > > > --- a/drivers/ufs/core/ufshcd.c
> > > > > > > +++ b/drivers/ufs/core/ufshcd.c
> > > > > > > @@ -278,6 +278,7 @@ static const struct ufs_dev_quirk ufs_fixups[] = {
> > > > > > >     	  .model = UFS_ANY_MODEL,
> > > > > > >     	  .quirk = UFS_DEVICE_QUIRK_DELAY_BEFORE_LPM |
> > > > > > >     		   UFS_DEVICE_QUIRK_HOST_PA_TACTIVATE |
> > > > > > > +		   UFS_DEVICE_QUIRK_PA_HIBER8TIME |
> > > > > > >     		   UFS_DEVICE_QUIRK_RECOVERY_FROM_DL_NAC_ERRORS },
> > > > > > >     	{ .wmanufacturerid = UFS_VENDOR_SKHYNIX,
> > > > > > >     	  .model = UFS_ANY_MODEL,
> > > > > > > @@ -8384,6 +8385,33 @@ static int ufshcd_quirk_tune_host_pa_tactivate(struct ufs_hba *hba)
> > > > > > >     	return ret;
> > > > > > >     }
> > > > > > > +/**
> > > > > > > + * ufshcd_quirk_override_pa_h8time - Ensures proper adjustment of PA_HIBERN8TIME.
> > > > > > > + * @hba: per-adapter instance
> > > > > > > + *
> > > > > > > + * Some UFS devices require specific adjustments to the PA_HIBERN8TIME parameter
> > > > > > > + * to ensure proper hibernation timing. This function retrieves the current
> > > > > > > + * PA_HIBERN8TIME value and increments it by 100us.
> > > > > > > + */
> > > > > > > +static void ufshcd_quirk_override_pa_h8time(struct ufs_hba *hba)
> > > > > > > +{
> > > > > > > +	u32 pa_h8time = 0;
> > > > > > 
> > > > > > Why do you need to initialize it?
> > > > > > 
> > > > > Agree.. Not needed, will update.>> +	int ret;
> > > > > > > +
> > > > > > > +	ret = ufshcd_dme_get(hba, UIC_ARG_MIB(PA_HIBERN8TIME),
> > > > > > > +			&pa_h8time);
> > > > > > > +	if (ret) {
> > > > > > > +		dev_err(hba->dev, "Failed to get PA_HIBERN8TIME: %d\n", ret);
> > > > > > > +		return;
> > > > > > > +	}
> > > > > > > +
> > > > > > > +	/* Increment by 1 to increase hibernation time by 100 µs */
> > > > > > 
> > > > > >    From where the value of 100us adjustment is coming from?
> > > > > > 
> > > > > > - Mani
> > > > > > 
> > > > > These values are derived from experiments on Qualcomm SoCs.
> > > > > However this is also matching with ufs-exynos.c
> > > > > 
> > > > 
> > > > Okay. In that case, you should mention that the 100us value is derived from
> > > > experiments on Qcom and Samsung SoCs. Otherwise, it gives an assumption that
> > > > this value is universal.
> > > > 
> > > > - Mani
> > > > 
> > >   << Otherwise, it gives an assumption that this value is universal. >>
> > > So with this, should i add this quirk for Qcom only, or should add in
> > > ufshcd.c and make it common for all SoC vendors?
> > > 
> > 
> > You can add the quirk for both Qcom and Samsung. My comment was about clarifying
> > it in the kernel doc comments.
> > 
> > - Mani
> > 
> Just for conclusion, why i moved this quirk from ufs-qcom to ufshcd.c is as
> per Bart's suggestion in patchset
> https://lore.kernel.org/lkml/c0691392-1523-4863-a722-d4f4640e4e28@xxxxxxx/
> 
> << Which of these quirks are required for all host controllers and which of
> these quirks are only required for Qualcomm host controllers?
> 
> > Should we consider moving the QUIRK_PA_HIBER8TIME quirk to the ufshcd
> > driver? Please advise.
> 
> That would be appreciated. >>
> 
> Just to brief, QUIRK_PA_HIBER8TIME is required for Samsung UFS devices and
> may be applicable to all SoC vendors with Samsung ufs device.
> 
> If you suggest to move it to ufs-qcom.c, i don't have any concern.
> BTW Samsung UFS driver already has this implemented in their driver,
> So i need not have to add this quirk to Samsung driver (ufs-exynos.c).
> 

No. I think there was a miscommunication. You can add the quirk in ufshcd.c, but
I just want you to mention that the quirk is currently applicable to Samsung UFS
devices and the value of 100us is derived from experiments.

- Mani

-- 
மணிவண்ணன் சதாசிவம்