Re: [RFC 1/2] mmc: sdhci-msm: Add support for bus bandwidth voting

[Ulf Hansson]

On Wed, 25 Sep 2019 at 13:27, <ppvk@xxxxxxxxxxxxxx> wrote:
>
> On 2019-09-12 22:15, Bjorn Andersson wrote:
> > On Fri 06 Sep 05:47 PDT 2019, Pradeep P V K wrote:
> >
> >> Vote for the MSM bus bandwidth required by SDHC driver
> >> based on the clock frequency and bus width of the card.
> >> Otherwise,the system clocks may run at minimum clock speed
> >> and thus affecting the performance.
> >>
> >> This change is based on Georgi Djakov [RFC]
> >> (https://lkml.org/lkml/2018/10/11/499)
> >>
> >> Signed-off-by: Sahitya Tummala <stummala@xxxxxxxxxxxxxx>
> >> Signed-off-by: Subhash Jadavani <subhashj@xxxxxxxxxxxxxx>
> >> Signed-off-by: Veerabhadrarao Badiganti <vbadigan@xxxxxxxxxxxxxx>
> >> Signed-off-by: Pradeep P V K <ppvk@xxxxxxxxxxxxxx>
> >
> > This says that Sahitya wrote the patch, then Subhash handled it, then
> > Veerabhadrarao handled it and finally you handled it; but you're at the
> > same time listed as author (by From:).
> >
> > Please see section 12 on Co-Developed-by in submitting-patches.rst
> >
> Thanks Bjorn, i will update this on my next patch set.
> >> ---
> >>  drivers/mmc/host/sdhci-msm.c | 393
> >> ++++++++++++++++++++++++++++++++++++++++++-
> >
> > This patch implements support for requesting bandwidth to the register
> > space and for the controllers access to DDR. To me this seems like
> > common requirements for any mmc controller, can this functionality be
> > provided by the mmc/sdhci common code?
> >
> > Regards,
> > Bjorn
> >
> Yes, this can be provided in common code but the bandwidth calculations
> (arbitrated value or average bandwidth and instantaneous value or peak
> bandwidth) for bus vote will
> consider various parameters like voltage corners, clock domains, clock
> plans etc. which may differ from
> vendor to vendor and target to target. So, these values should be
> updated properly and correctly (considering all the parameters)
> if it brings to common area.
>
> Hence the reason for implementing this in sdhci-msm.c file.
> It would be really helpful if you could suggest your insights on where
> and how exactly this needs to be
> implemented in common code area.
>
> Hi Ulf and Adrian,
>
> Can you please also suggest your recommendations on this ?

I am not sure where to put it at this point. Perhaps we can just start
with making it specific for sdhci-msm, then when new users come into
play, we can consider moving it to a common place.

[...]

Kind regards
Uffe