2021-06-16 19:08:52

by Robert Foss

[permalink] [raw]
Subject: [RFC v1 02/11] clk: qcom: rcg2: Add support for flags

These changes are ported from the downstream driver, and are used on SM8350
for CAMCC, DISPCC, GCC, GPUCC & VIDEOCC.

Signed-off-by: Robert Foss <[email protected]>
---
drivers/clk/qcom/clk-rcg.h | 4 ++++
drivers/clk/qcom/clk-rcg2.c | 3 +++
2 files changed, 7 insertions(+)

diff --git a/drivers/clk/qcom/clk-rcg.h b/drivers/clk/qcom/clk-rcg.h
index 99efcc7f8d88..a1f05281d950 100644
--- a/drivers/clk/qcom/clk-rcg.h
+++ b/drivers/clk/qcom/clk-rcg.h
@@ -149,6 +149,10 @@ struct clk_rcg2 {
const struct freq_tbl *freq_tbl;
struct clk_regmap clkr;
u8 cfg_off;
+ u8 flags;
+#define FORCE_ENABLE_RCG BIT(0)
+#define HW_CLK_CTRL_MODE BIT(1)
+#define DFS_SUPPORT BIT(2)
};

#define to_clk_rcg2(_hw) container_of(to_clk_regmap(_hw), struct clk_rcg2, clkr)
diff --git a/drivers/clk/qcom/clk-rcg2.c b/drivers/clk/qcom/clk-rcg2.c
index 42f13a2d1cc1..ed2c9b6659cc 100644
--- a/drivers/clk/qcom/clk-rcg2.c
+++ b/drivers/clk/qcom/clk-rcg2.c
@@ -295,6 +295,9 @@ static int __clk_rcg2_configure(struct clk_rcg2 *rcg, const struct freq_tbl *f)
cfg |= rcg->parent_map[index].cfg << CFG_SRC_SEL_SHIFT;
if (rcg->mnd_width && f->n && (f->m != f->n))
cfg |= CFG_MODE_DUAL_EDGE;
+ if (rcg->flags & HW_CLK_CTRL_MODE)
+ cfg |= CFG_HW_CLK_CTRL_MASK;
+
return regmap_update_bits(rcg->clkr.regmap, RCG_CFG_OFFSET(rcg),
mask, cfg);
}
--
2.30.2


2021-06-16 19:27:03

by Konrad Dybcio

[permalink] [raw]
Subject: Re: [RFC v1 02/11] clk: qcom: rcg2: Add support for flags


On 16.06.2021 16:10, Robert Foss wrote:
> These changes are ported from the downstream driver, and are used on SM8350
> for CAMCC, DISPCC, GCC, GPUCC & VIDEOCC.
>
> Signed-off-by: Robert Foss <[email protected]>
> ---
> drivers/clk/qcom/clk-rcg.h | 4 ++++
> drivers/clk/qcom/clk-rcg2.c | 3 +++
> 2 files changed, 7 insertions(+)
>
> diff --git a/drivers/clk/qcom/clk-rcg.h b/drivers/clk/qcom/clk-rcg.h
> index 99efcc7f8d88..a1f05281d950 100644
> --- a/drivers/clk/qcom/clk-rcg.h
> +++ b/drivers/clk/qcom/clk-rcg.h
> @@ -149,6 +149,10 @@ struct clk_rcg2 {
> const struct freq_tbl *freq_tbl;
> struct clk_regmap clkr;
> u8 cfg_off;
> + u8 flags;
> +#define FORCE_ENABLE_RCG BIT(0)
> +#define HW_CLK_CTRL_MODE BIT(1)
> +#define DFS_SUPPORT BIT(2)
> };
>
> #define to_clk_rcg2(_hw) container_of(to_clk_regmap(_hw), struct clk_rcg2, clkr)
> diff --git a/drivers/clk/qcom/clk-rcg2.c b/drivers/clk/qcom/clk-rcg2.c
> index 42f13a2d1cc1..ed2c9b6659cc 100644
> --- a/drivers/clk/qcom/clk-rcg2.c
> +++ b/drivers/clk/qcom/clk-rcg2.c
> @@ -295,6 +295,9 @@ static int __clk_rcg2_configure(struct clk_rcg2 *rcg, const struct freq_tbl *f)
> cfg |= rcg->parent_map[index].cfg << CFG_SRC_SEL_SHIFT;
> if (rcg->mnd_width && f->n && (f->m != f->n))
> cfg |= CFG_MODE_DUAL_EDGE;
> + if (rcg->flags & HW_CLK_CTRL_MODE)
> + cfg |= CFG_HW_CLK_CTRL_MASK;
> +
> return regmap_update_bits(rcg->clkr.regmap, RCG_CFG_OFFSET(rcg),
> mask, cfg);
> }

What about code for handling other flags? If it's not a part of the series,

I don't think it makes sense to define them. Or perhaps you sent the

wrong revision?


Konrad

2021-06-16 22:50:32

by Dmitry Baryshkov

[permalink] [raw]
Subject: Re: [RFC v1 02/11] clk: qcom: rcg2: Add support for flags

On 16/06/2021 17:10, Robert Foss wrote:
> These changes are ported from the downstream driver, and are used on SM8350
> for CAMCC, DISPCC, GCC, GPUCC & VIDEOCC.
>
> Signed-off-by: Robert Foss <[email protected]>
> ---
> drivers/clk/qcom/clk-rcg.h | 4 ++++
> drivers/clk/qcom/clk-rcg2.c | 3 +++
> 2 files changed, 7 insertions(+)
>
> diff --git a/drivers/clk/qcom/clk-rcg.h b/drivers/clk/qcom/clk-rcg.h
> index 99efcc7f8d88..a1f05281d950 100644
> --- a/drivers/clk/qcom/clk-rcg.h
> +++ b/drivers/clk/qcom/clk-rcg.h
> @@ -149,6 +149,10 @@ struct clk_rcg2 {
> const struct freq_tbl *freq_tbl;
> struct clk_regmap clkr;
> u8 cfg_off;
> + u8 flags;
> +#define FORCE_ENABLE_RCG BIT(0)
> +#define HW_CLK_CTRL_MODE BIT(1)

Downstream also has these flags for 8250, but the upstream driver ended
up not using them for the dispcc clocks. Could you please check that you
realy need HW_CLK_CTRL for dispcc clocks?

> +#define DFS_SUPPORT BIT(2)
> };
>
> #define to_clk_rcg2(_hw) container_of(to_clk_regmap(_hw), struct clk_rcg2, clkr)
> diff --git a/drivers/clk/qcom/clk-rcg2.c b/drivers/clk/qcom/clk-rcg2.c
> index 42f13a2d1cc1..ed2c9b6659cc 100644
> --- a/drivers/clk/qcom/clk-rcg2.c
> +++ b/drivers/clk/qcom/clk-rcg2.c
> @@ -295,6 +295,9 @@ static int __clk_rcg2_configure(struct clk_rcg2 *rcg, const struct freq_tbl *f)
> cfg |= rcg->parent_map[index].cfg << CFG_SRC_SEL_SHIFT;
> if (rcg->mnd_width && f->n && (f->m != f->n))
> cfg |= CFG_MODE_DUAL_EDGE;
> + if (rcg->flags & HW_CLK_CTRL_MODE)
> + cfg |= CFG_HW_CLK_CTRL_MASK;
> +
> return regmap_update_bits(rcg->clkr.regmap, RCG_CFG_OFFSET(rcg),
> mask, cfg);
> }
>


--
With best wishes
Dmitry

2021-06-17 08:02:19

by Robert Foss

[permalink] [raw]
Subject: Re: [RFC v1 02/11] clk: qcom: rcg2: Add support for flags

On Wed, 16 Jun 2021 at 17:33, Konrad Dybcio
<[email protected]> wrote:
>
>
> On 16.06.2021 16:10, Robert Foss wrote:
> > These changes are ported from the downstream driver, and are used on SM8350
> > for CAMCC, DISPCC, GCC, GPUCC & VIDEOCC.
> >
> > Signed-off-by: Robert Foss <[email protected]>
> > ---
> > drivers/clk/qcom/clk-rcg.h | 4 ++++
> > drivers/clk/qcom/clk-rcg2.c | 3 +++
> > 2 files changed, 7 insertions(+)
> >
> > diff --git a/drivers/clk/qcom/clk-rcg.h b/drivers/clk/qcom/clk-rcg.h
> > index 99efcc7f8d88..a1f05281d950 100644
> > --- a/drivers/clk/qcom/clk-rcg.h
> > +++ b/drivers/clk/qcom/clk-rcg.h
> > @@ -149,6 +149,10 @@ struct clk_rcg2 {
> > const struct freq_tbl *freq_tbl;
> > struct clk_regmap clkr;
> > u8 cfg_off;
> > + u8 flags;
> > +#define FORCE_ENABLE_RCG BIT(0)
> > +#define HW_CLK_CTRL_MODE BIT(1)
> > +#define DFS_SUPPORT BIT(2)
> > };
> >
> > #define to_clk_rcg2(_hw) container_of(to_clk_regmap(_hw), struct clk_rcg2, clkr)
> > diff --git a/drivers/clk/qcom/clk-rcg2.c b/drivers/clk/qcom/clk-rcg2.c
> > index 42f13a2d1cc1..ed2c9b6659cc 100644
> > --- a/drivers/clk/qcom/clk-rcg2.c
> > +++ b/drivers/clk/qcom/clk-rcg2.c
> > @@ -295,6 +295,9 @@ static int __clk_rcg2_configure(struct clk_rcg2 *rcg, const struct freq_tbl *f)
> > cfg |= rcg->parent_map[index].cfg << CFG_SRC_SEL_SHIFT;
> > if (rcg->mnd_width && f->n && (f->m != f->n))
> > cfg |= CFG_MODE_DUAL_EDGE;
> > + if (rcg->flags & HW_CLK_CTRL_MODE)
> > + cfg |= CFG_HW_CLK_CTRL_MASK;
> > +
> > return regmap_update_bits(rcg->clkr.regmap, RCG_CFG_OFFSET(rcg),
> > mask, cfg);
> > }
>
> What about code for handling other flags? If it's not a part of the series,
>
> I don't think it makes sense to define them. Or perhaps you sent the
>
> wrong revision?

I opted to add all of the flags just to document them existing, but
only introducing the ones that will immediately be used is the better
way to go.

2021-06-17 18:00:18

by Robert Foss

[permalink] [raw]
Subject: Re: [RFC v1 02/11] clk: qcom: rcg2: Add support for flags

On Wed, 16 Jun 2021 at 18:07, Dmitry Baryshkov
<[email protected]> wrote:
> > diff --git a/drivers/clk/qcom/clk-rcg.h b/drivers/clk/qcom/clk-rcg.h
> > index 99efcc7f8d88..a1f05281d950 100644
> > --- a/drivers/clk/qcom/clk-rcg.h
> > +++ b/drivers/clk/qcom/clk-rcg.h
> > @@ -149,6 +149,10 @@ struct clk_rcg2 {
> > const struct freq_tbl *freq_tbl;
> > struct clk_regmap clkr;
> > u8 cfg_off;
> > + u8 flags;
> > +#define FORCE_ENABLE_RCG BIT(0)
> > +#define HW_CLK_CTRL_MODE BIT(1)
>
> Downstream also has these flags for 8250, but the upstream driver ended
> up not using them for the dispcc clocks. Could you please check that you
> realy need HW_CLK_CTRL for dispcc clocks?

HW_CLK_CTRL being flagged in dispcc causes the CFG_HW_CLK_CTRL flag to
be set in the RCG_CFG registers of dispcc.

This flag simply marks the clock as having hardware control enabled or disabled.

As for the question if it is really needed, I can't answer that since
no documentation or downstream comments explain the exact behaviour.
As far as I know the only way to figure out if it is required is
disabling the flag and checking for bugs. I did find this[1] patch,
which enabled HW_CLK_CTRL_MODE.

Should I err on the side of the downstream implementation, or try to
create a minimum functional driver based on the downstream driver?

[1] https://patchwork.kernel.org/project/linux-arm-msm/patch/[email protected]/