From: Amit Kucheria <amitk@kernel.org>
To: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>,
Viresh Kumar <viresh.kumar@linaro.org>,
Rob Herring <robh+dt@kernel.org>, Andy Gross <agross@kernel.org>,
Bjorn Andersson <bjorn.andersson@linaro.org>,
Linux PM list <linux-pm@vger.kernel.org>,
"open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS"
<devicetree@vger.kernel.org>, LKML <linux-kernel@vger.kernel.org>,
linux-arm-msm <linux-arm-msm@vger.kernel.org>,
Dmitry Baryshkov <dmitry.baryshkov@linaro.org>,
Taniya Das <tdas@codeaurora.org>
Subject: Re: [PATCH 4/7] cpufreq: qcom-hw: Make use of of_match data for offsets and row size
Date: Tue, 8 Sep 2020 22:36:53 +0530 [thread overview]
Message-ID: <CAHLCerN+sfk5bOg6-Exgdy+fWJO4e29qRYkaycjWbG3MoZFikg@mail.gmail.com> (raw)
In-Reply-To: <20200908150945.GB2352@mani-NUC7i5DNKE>
On Tue, Sep 8, 2020 at 8:40 PM Manivannan Sadhasivam
<manivannan.sadhasivam@linaro.org> wrote:
>
> On 0908, Amit Kucheria wrote:
> > On Tue, Sep 8, 2020 at 1:27 PM Manivannan Sadhasivam
> > <manivannan.sadhasivam@linaro.org> wrote:
> > >
> > > For preparing the driver to handle further SoC revisions, let's use the
> > > of_match data for getting the device specific offsets and row size instead
> > > of defining them globally.
> > >
> > > Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
> >
> >
> >
> > > ---
> > > drivers/cpufreq/qcom-cpufreq-hw.c | 96 +++++++++++++++++++++----------
> > > 1 file changed, 66 insertions(+), 30 deletions(-)
> > >
> > > diff --git a/drivers/cpufreq/qcom-cpufreq-hw.c b/drivers/cpufreq/qcom-cpufreq-hw.c
> > > index ccea34f61152..41853db7c9b8 100644
> > > --- a/drivers/cpufreq/qcom-cpufreq-hw.c
> > > +++ b/drivers/cpufreq/qcom-cpufreq-hw.c
> > > @@ -19,15 +19,21 @@
> > > #define LUT_L_VAL GENMASK(7, 0)
> > > #define LUT_CORE_COUNT GENMASK(18, 16)
> > > #define LUT_VOLT GENMASK(11, 0)
> > > -#define LUT_ROW_SIZE 32
> > > #define CLK_HW_DIV 2
> > > #define LUT_TURBO_IND 1
> > >
> > > -/* Register offsets */
> > > -#define REG_ENABLE 0x0
> > > -#define REG_FREQ_LUT 0x110
> > > -#define REG_VOLT_LUT 0x114
> > > -#define REG_PERF_STATE 0x920
> > > +struct qcom_cpufreq_soc_data {
> > > + u32 reg_enable;
> > > + u32 reg_freq_lut;
> > > + u32 reg_volt_lut;
> > > + u32 reg_perf_state;
> > > + u8 lut_row_size;
> > > +};
> > > +
> > > +struct qcom_cpufreq_data {
> > > + void __iomem *base;
> > > + const struct qcom_cpufreq_soc_data *soc_data;
> > > +};
> > >
> > > static unsigned long cpu_hw_rate, xo_rate;
> > > static bool icc_scaling_enabled;
> > > @@ -76,10 +82,11 @@ static int qcom_cpufreq_update_opp(struct device *cpu_dev,
> > > static int qcom_cpufreq_hw_target_index(struct cpufreq_policy *policy,
> > > unsigned int index)
> > > {
> > > - void __iomem *perf_state_reg = policy->driver_data;
> > > + struct qcom_cpufreq_data *data = policy->driver_data;
> > > + const struct qcom_cpufreq_soc_data *soc_data = data->soc_data;
> > > unsigned long freq = policy->freq_table[index].frequency;
> > >
> > > - writel_relaxed(index, perf_state_reg);
> > > + writel_relaxed(index, data->base + soc_data->reg_perf_state);
> > >
> > > if (icc_scaling_enabled)
> > > qcom_cpufreq_set_bw(policy, freq);
> > > @@ -91,7 +98,8 @@ static int qcom_cpufreq_hw_target_index(struct cpufreq_policy *policy,
> > >
> > > static unsigned int qcom_cpufreq_hw_get(unsigned int cpu)
> > > {
> > > - void __iomem *perf_state_reg;
> > > + struct qcom_cpufreq_data *data;
> > > + const struct qcom_cpufreq_soc_data *soc_data;
> > > struct cpufreq_policy *policy;
> > > unsigned int index;
> > >
> > > @@ -99,9 +107,10 @@ static unsigned int qcom_cpufreq_hw_get(unsigned int cpu)
> > > if (!policy)
> > > return 0;
> > >
> > > - perf_state_reg = policy->driver_data;
> > > + data = policy->driver_data;
> > > + soc_data = data->soc_data;
> > >
> > > - index = readl_relaxed(perf_state_reg);
> > > + index = readl_relaxed(data->base + soc_data->reg_perf_state);
> > > index = min(index, LUT_MAX_ENTRIES - 1);
> > >
> > > return policy->freq_table[index].frequency;
> > > @@ -110,12 +119,13 @@ static unsigned int qcom_cpufreq_hw_get(unsigned int cpu)
> > > static unsigned int qcom_cpufreq_hw_fast_switch(struct cpufreq_policy *policy,
> > > unsigned int target_freq)
> > > {
> > > - void __iomem *perf_state_reg = policy->driver_data;
> > > + struct qcom_cpufreq_data *data = policy->driver_data;
> > > + const struct qcom_cpufreq_soc_data *soc_data = data->soc_data;
> > > unsigned int index;
> > > unsigned long freq;
> > >
> > > index = policy->cached_resolved_idx;
> > > - writel_relaxed(index, perf_state_reg);
> > > + writel_relaxed(index, data->base + soc_data->reg_perf_state);
> > >
> > > freq = policy->freq_table[index].frequency;
> > > arch_set_freq_scale(policy->related_cpus, freq,
> > > @@ -125,8 +135,7 @@ static unsigned int qcom_cpufreq_hw_fast_switch(struct cpufreq_policy *policy,
> > > }
> > >
> > > static int qcom_cpufreq_hw_read_lut(struct device *cpu_dev,
> > > - struct cpufreq_policy *policy,
> > > - void __iomem *base)
> > > + struct cpufreq_policy *policy)
> > > {
> > > u32 data, src, lval, i, core_count, prev_freq = 0, freq;
> > > u32 volt;
> > > @@ -134,6 +143,8 @@ static int qcom_cpufreq_hw_read_lut(struct device *cpu_dev,
> > > struct dev_pm_opp *opp;
> > > unsigned long rate;
> > > int ret;
> > > + struct qcom_cpufreq_data *drv_data = policy->driver_data;
> > > + const struct qcom_cpufreq_soc_data *soc_data = drv_data->soc_data;
> > >
> > > table = kcalloc(LUT_MAX_ENTRIES + 1, sizeof(*table), GFP_KERNEL);
> > > if (!table)
> > > @@ -160,14 +171,14 @@ static int qcom_cpufreq_hw_read_lut(struct device *cpu_dev,
> > > }
> > >
> > > for (i = 0; i < LUT_MAX_ENTRIES; i++) {
> > > - data = readl_relaxed(base + REG_FREQ_LUT +
> > > - i * LUT_ROW_SIZE);
> > > + data = readl_relaxed(drv_data->base + soc_data->reg_freq_lut +
> > > + i * soc_data->lut_row_size);
> > > src = FIELD_GET(LUT_SRC, data);
> > > lval = FIELD_GET(LUT_L_VAL, data);
> > > core_count = FIELD_GET(LUT_CORE_COUNT, data);
> > >
> > > - data = readl_relaxed(base + REG_VOLT_LUT +
> > > - i * LUT_ROW_SIZE);
> > > + data = readl_relaxed(drv_data->base + soc_data->reg_volt_lut +
> > > + i * soc_data->lut_row_size);
> > > volt = FIELD_GET(LUT_VOLT, data) * 1000;
> > >
> > > if (src)
> > > @@ -237,6 +248,20 @@ static void qcom_get_related_cpus(int index, struct cpumask *m)
> > > }
> > > }
> > >
> > > +static const struct qcom_cpufreq_soc_data qcom_soc_data = {
> >
> > rename this to sdm845_soc_data?
> >
>
> Nah, this is not specific to SDM845. Atleast in mainline, there are 3 SoCs
> using this compatible.
>
> > Or even better, maybe just use the IP version number for this IP block
> > so that all SoCs using that IP version can use this struct?
> >
>
> Since the SoCs are using the same compatible it makes sense to use the same
> name for the of_data. I don't think it is a good idea to use different name
> for the of_data since the differentiation has to happen at compatible level.
You are using the name sm8250_soc_data in a subsequent patch, though ;-)
So I think it would make sense for compatible "qcom,cpufreq-hw" to use
data "osm_soc_data" and compatible "qcom,sm8250-epss" to use data
"epss_soc_data" as suggested by Bjorn.
Regards,
Amit
>
> > > + .reg_enable = 0x0,
> > > + .reg_freq_lut = 0x110,
> > > + .reg_volt_lut = 0x114,
> > > + .reg_perf_state = 0x920,
> > > + .lut_row_size = 32,
> > > +};
> > > +
> > > +static const struct of_device_id qcom_cpufreq_hw_match[] = {
> > > + { .compatible = "qcom,cpufreq-hw", .data = &qcom_soc_data },
> > > + {}
> > > +};
next prev parent reply other threads:[~2020-09-08 17:09 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-08 7:57 [PATCH 0/7] Add CPUFreq support for SM8250 SoC Manivannan Sadhasivam
2020-09-08 7:57 ` [PATCH 1/7] dt-bindings: cpufreq: cpufreq-qcom-hw: Document SM8250 compatible Manivannan Sadhasivam
2020-09-08 11:56 ` Amit Kucheria
2020-09-08 14:58 ` Bjorn Andersson
2020-09-08 15:10 ` Manivannan Sadhasivam
2020-09-15 16:38 ` Rob Herring
2020-09-08 7:57 ` [PATCH 2/7] arm64: dts: qcom: sm8250: Add cpufreq hw node Manivannan Sadhasivam
2020-09-08 10:39 ` Viresh Kumar
2020-09-08 11:08 ` Manivannan Sadhasivam
2020-09-08 11:12 ` Viresh Kumar
2020-09-08 11:56 ` Amit Kucheria
2020-09-08 7:57 ` [PATCH 3/7] cpufreq: qcom-hw: Make use of cpufreq driver_data for passing pdev Manivannan Sadhasivam
2020-09-08 10:31 ` Viresh Kumar
2020-09-08 7:57 ` [PATCH 4/7] cpufreq: qcom-hw: Make use of of_match data for offsets and row size Manivannan Sadhasivam
2020-09-08 12:18 ` Amit Kucheria
2020-09-08 15:09 ` Manivannan Sadhasivam
2020-09-08 17:06 ` Amit Kucheria [this message]
2020-09-08 15:31 ` Bjorn Andersson
2020-09-08 7:57 ` [PATCH 5/7] cpufreq: qcom-hw: Use regmap for accessing hardware registers Manivannan Sadhasivam
2020-09-08 10:34 ` Viresh Kumar
2020-09-08 11:11 ` Manivannan Sadhasivam
2020-09-08 11:18 ` Viresh Kumar
2020-09-08 11:28 ` Manivannan Sadhasivam
2020-09-08 11:48 ` Amit Kucheria
2020-09-08 12:08 ` Amit Kucheria
2020-09-08 15:03 ` Manivannan Sadhasivam
2020-09-09 4:35 ` Viresh Kumar
2020-09-08 14:08 ` Sudeep Holla
2020-09-08 15:39 ` Bjorn Andersson
2020-09-08 7:57 ` [PATCH 6/7] cpufreq: qcom-hw: Add cpufreq support for SM8250 SoC Manivannan Sadhasivam
2020-09-08 12:10 ` Amit Kucheria
2020-09-08 15:22 ` Bjorn Andersson
2020-09-08 15:29 ` Manivannan Sadhasivam
2020-09-08 7:57 ` [PATCH 7/7] cpufreq: qcom-hw: Use devm_platform_ioremap_resource() to simplify code Manivannan Sadhasivam
2020-09-08 10:36 ` Viresh Kumar
2020-09-08 11:12 ` Manivannan Sadhasivam
2020-09-08 12:00 ` Amit Kucheria
2020-09-08 15:35 ` Bjorn Andersson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAHLCerN+sfk5bOg6-Exgdy+fWJO4e29qRYkaycjWbG3MoZFikg@mail.gmail.com \
--to=amitk@kernel.org \
--cc=agross@kernel.org \
--cc=bjorn.andersson@linaro.org \
--cc=devicetree@vger.kernel.org \
--cc=dmitry.baryshkov@linaro.org \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=manivannan.sadhasivam@linaro.org \
--cc=rjw@rjwysocki.net \
--cc=robh+dt@kernel.org \
--cc=tdas@codeaurora.org \
--cc=viresh.kumar@linaro.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).