From: dawei chien <dawei.chien@mediatek.com>
To: Eduardo Valentin <edubezval@gmail.com>
Cc: Viresh Kumar <viresh.kumar@linaro.org>,
"Rafael J. Wysocki" <rjw@rjwysocki.net>,
Rob Herring <robh+dt@kernel.org>, Pawel Moll <pawel.moll@arm.com>,
Mark Rutland <mark.rutland@arm.com>,
Ian Campbell <ijc+devicetree@hellion.org.uk>,
Kumar Gala <galak@codeaurora.org>,
Matthias Brugger <matthias.bgg@gmail.com>,
Daniel Kurtz <djkurtz@chromium.org>,
Sascha Hauer <s.hauer@pengutronix.de>,
"Daniel Lezcano" <daniel.lezcano@linaro.org>,
<devicetree@vger.kernel.org>,
<linux-arm-kernel@lists.infradead.org>,
<linux-kernel@vger.kernel.org>, <linux-pm@vger.kernel.org>,
<linux-mediatek@lists.infradead.org>,
<srv_heupstream@mediatek.com>,
Sascha Hauer <kernel@pengutronix.de>
Subject: Re: [PATCH v3 1/2] thermal: mediatek: Add cpu power cooling model.
Date: Thu, 5 Nov 2015 19:10:37 +0800 [thread overview]
Message-ID: <1446721837.4016.57.camel@mtksdaap41> (raw)
In-Reply-To: <20151104193958.GC14127@localhost.localdomain>
On Wed, 2015-11-04 at 11:40 -0800, Eduardo Valentin wrote:
> On Thu, Oct 22, 2015 at 08:02:38PM +0800, Dawei Chien wrote:
> > This power model is base on Intelligent Power Allocation (IPA)
> > technical, requires that the operating-points of the CPUs are
> > registered using the kernel's opp library and the
> > `cpufreq_frequency_table` is assigned to the `struct device`
> > of the cpu MT8173.
> >
> > Signed-off-by: Dawei.Chien <dawei.chien@mediatek.com>
> > ---
> > This patch is base on
> > https://patchwork.kernel.org/patch/7034601/
> > ---
> > drivers/cpufreq/mt8173-cpufreq.c | 152 ++++++++++++++++++++++++++++++++++++--
>
> Given that you are proposing this on top a DT binding, why reading a
> table of static power model applicable only to mt8173 cpufreq driver?
>
> I have not seen anything specific (formula etc) that prevents this code
> to be generalized to other CPUs.
>
> Can you please help me to understand?
>
> BR,
This is because our platform currently only support mt8173_cpufreq.c, so
that I only add static power model for our owner IC.
Please understanding that I wouldn't give a DT binding document since I
will remove static power table on next version, but I can try to explain
it.
As far as I know, static power is somewhat leakage of CPU clusters, so
that we hardly to find a formula, which can suitable all kinds of CPUs
since leakage is different. In ARM IPA framework, static power only need
to be defined by who register cpufreq_power_cooling_register. The
voltage/power table is just one way to present leakage power of CPUs.
Actually, static power is optional since dynamic power is much more than
static power.
> > 1 file changed, 144 insertions(+), 8 deletions(-)
> >
> > diff --git a/drivers/cpufreq/mt8173-cpufreq.c b/drivers/cpufreq/mt8173-cpufreq.c
> > index 49caed2..23c19c5 100644
> > --- a/drivers/cpufreq/mt8173-cpufreq.c
> > +++ b/drivers/cpufreq/mt8173-cpufreq.c
> > @@ -29,6 +29,16 @@
> > #define MAX_VOLT_LIMIT (1150000)
> > #define VOLT_TOL (10000)
> >
> > +struct mtk_cpu_static_power {
> > + unsigned long voltage;
> > + unsigned int power;
> > +};
> > +
> > +static struct mtk_cpu_static_power *mtk_ca53_static_power_table;
> > +static struct mtk_cpu_static_power *mtk_ca57_static_power_table;
> > +static int mtk_ca53_static_table_length;
> > +static int mtk_ca57_static_table_length;
> > +
> > /*
> > * The struct mtk_cpu_dvfs_info holds necessary information for doing CPU DVFS
> > * on each CPU power/clock domain of Mediatek SoCs. Each CPU cluster in
> > @@ -51,6 +61,110 @@ struct mtk_cpu_dvfs_info {
> > bool need_voltage_tracking;
> > };
> >
> > +unsigned int mtk_cpufreq_lookup_power(const struct mtk_cpu_static_power *table,
> > + unsigned int count, unsigned long voltage)
> > +{
> > + int i;
> > +
> > + for (i = 0; i < count; i++) {
> > + if (voltage <= table[i].voltage)
> > + return table[i].power;
> > + }
> > +
> > + return table[count - 1].power;
> > +}
> > +
> > +int mtk_cpufreq_get_static(cpumask_t *cpumask, int interval,
> > + unsigned long voltage, u32 *power)
> > +{
> > + int nr_cpus = cpumask_weight(cpumask);
> > +
> > + *power = 0;
> > +
> > + if (nr_cpus) {
> > + if (cpumask_test_cpu(0, cpumask))
> > + *power += mtk_cpufreq_lookup_power(
> > + mtk_ca53_static_power_table,
> > + mtk_ca53_static_table_length,
> > + voltage);
> > +
> > + if (cpumask_test_cpu(2, cpumask))
> > + *power += mtk_cpufreq_lookup_power(
> > + mtk_ca57_static_power_table,
> > + mtk_ca57_static_table_length,
> > + voltage);
> > + }
> > +
> > + return 0;
> > +}
> > +
> > +unsigned int mtk_get_power_table_info(struct cpufreq_policy *policy,
> > + struct device_node *np, const char *node_name)
> > +{
> > + int mtk_static_table_length;
> > + const struct property *prop;
> > + struct mtk_cpu_dvfs_info *info = policy->driver_data;
> > + struct device *cpu_dev = info->cpu_dev;
> > + const __be32 *val;
> > + int nr, i;
> > +
> > + prop = of_find_property(np, node_name, NULL);
> > +
> > + if (!prop) {
> > + pr_err("failed to get static-power-points\n");
> > + return -ENODEV;
> > + }
> > +
> > + if (!prop->value) {
> > + pr_err("failed to get static power array data\n");
> > + return -EINVAL;
> > + }
> > +
> > + nr = prop->length / sizeof(u32);
> > +
> > + if (nr % 2) {
> > + pr_err("Invalid OPP list\n");
> > + return -EINVAL;
> > + }
> > +
> > + mtk_static_table_length = nr / 2;
> > +
> > + if (cpumask_test_cpu(0, policy->related_cpus)) {
> > + mtk_ca53_static_table_length = mtk_static_table_length;
> > + mtk_ca53_static_power_table = devm_kcalloc(cpu_dev,
> > + mtk_static_table_length,
> > + sizeof(*mtk_ca53_static_power_table),
> > + GFP_KERNEL);
> > +
> > + val = prop->value;
> > + for (i = 0; i < mtk_static_table_length; ++i) {
> > + unsigned long voltage = be32_to_cpup(val++);
> > + unsigned int power = be32_to_cpup(val++);
> > +
> > + mtk_ca53_static_power_table[i].voltage = voltage;
> > + mtk_ca53_static_power_table[i].power = power;
> > + pr_info("volt:%ld uv, power:%d mW\n", voltage, power);
> > + }
> > + } else {
> > + mtk_ca57_static_table_length = mtk_static_table_length;
> > + mtk_ca57_static_power_table = devm_kcalloc(cpu_dev,
> > + mtk_static_table_length,
> > + sizeof(*mtk_ca57_static_power_table),
> > + GFP_KERNEL);
> > + val = prop->value;
> > + for (i = 0; i < mtk_static_table_length; ++i) {
> > + unsigned long voltage = be32_to_cpup(val++);
> > + unsigned int power = be32_to_cpup(val++);
> > +
> > + mtk_ca57_static_power_table[i].voltage = voltage;
> > + mtk_ca57_static_power_table[i].power = power;
> > + pr_info("volt:%ld uv, power:%d mW\n", voltage, power);
> > + }
> > + }
> > +
> > + return 0;
> > +}
> > +
> > static int mtk_cpufreq_voltage_tracking(struct mtk_cpu_dvfs_info *info,
> > int new_vproc)
> > {
> > @@ -267,20 +381,40 @@ static void mtk_cpufreq_ready(struct cpufreq_policy *policy)
> > {
> > struct mtk_cpu_dvfs_info *info = policy->driver_data;
> > struct device_node *np = of_node_get(info->cpu_dev->of_node);
> > + u32 capacitance;
> > + int ret;
> >
> > if (WARN_ON(!np))
> > return;
> >
> > if (of_find_property(np, "#cooling-cells", NULL)) {
> > - info->cdev = of_cpufreq_cooling_register(np,
> > - policy->related_cpus);
> >
> > - if (IS_ERR(info->cdev)) {
> > - dev_err(info->cpu_dev,
> > - "running cpufreq without cooling device: %ld\n",
> > - PTR_ERR(info->cdev));
> > + if (!info->cdev) {
> > +
> > + of_property_read_u32(np,
> > + "dynamic-power-coefficient",
> > + &capacitance);
> > +
> > + ret = mtk_get_power_table_info(policy, np,
> > + "static-power-points");
> > + if (ret) {
> > + dev_err(info->cpu_dev,
> > + "cpufreq without static-points: %d\n",
> > + ret);
> > + }
> > +
> > + info->cdev = of_cpufreq_power_cooling_register(np,
> > + policy->related_cpus,
> > + capacitance,
> > + mtk_cpufreq_get_static);
> > +
> > + if (IS_ERR(info->cdev)) {
> > + dev_err(info->cpu_dev,
> > + "cpufreq without cdev: %ld\n",
> > + PTR_ERR(info->cdev));
> > + info->cdev = NULL;
> > + }
> >
> > - info->cdev = NULL;
> > }
> > }
> >
> > @@ -460,7 +594,9 @@ static int mtk_cpufreq_exit(struct cpufreq_policy *policy)
> > {
> > struct mtk_cpu_dvfs_info *info = policy->driver_data;
> >
> > - cpufreq_cooling_unregister(info->cdev);
> > + if (info->cdev)
> > + cpufreq_cooling_unregister(info->cdev);
> > +
> > dev_pm_opp_free_cpufreq_table(info->cpu_dev, &policy->freq_table);
> > mtk_cpu_dvfs_info_release(info);
> > kfree(info);
> > --
> > 1.7.9.5
> >
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-pm" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2015-11-05 11:10 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-10-22 12:02 [PATCH v3 0/2] thermal: mediatek: Add cpu power cooling model Dawei Chien
2015-10-22 12:02 ` [PATCH v3 1/2] " Dawei Chien
2015-11-04 19:40 ` Eduardo Valentin
2015-11-05 11:10 ` dawei chien [this message]
2015-11-06 3:24 ` Viresh Kumar
2015-11-10 11:20 ` Javi Merino
2015-11-10 18:41 ` Eduardo Valentin
2015-11-11 9:36 ` Javi Merino
2015-11-13 5:02 ` Viresh Kumar
2015-10-22 12:02 ` [PATCH v3 2/2] arm64: dts: mt8173: Add thermal zone node for mt8173 Dawei Chien
2015-10-28 15:39 ` Viresh Kumar
2015-11-02 10:51 ` dawei chien
2015-11-04 19:41 ` Eduardo Valentin
2015-11-05 11:10 ` dawei chien
2015-10-28 15:44 ` [PATCH v3 0/2] thermal: mediatek: Add cpu power cooling model Viresh Kumar
2015-11-02 10:46 ` dawei chien
2015-11-02 12:10 ` Viresh Kumar
2015-11-05 11:09 ` dawei chien
2015-11-06 3:20 ` Viresh Kumar
2015-11-11 12:00 ` dawei chien
2015-11-02 15:53 ` Punit Agrawal
2015-11-02 16:10 ` Viresh Kumar
2015-11-04 19:36 ` Eduardo Valentin
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=1446721837.4016.57.camel@mtksdaap41 \
--to=dawei.chien@mediatek.com \
--cc=daniel.lezcano@linaro.org \
--cc=devicetree@vger.kernel.org \
--cc=djkurtz@chromium.org \
--cc=edubezval@gmail.com \
--cc=galak@codeaurora.org \
--cc=ijc+devicetree@hellion.org.uk \
--cc=kernel@pengutronix.de \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mediatek@lists.infradead.org \
--cc=linux-pm@vger.kernel.org \
--cc=mark.rutland@arm.com \
--cc=matthias.bgg@gmail.com \
--cc=pawel.moll@arm.com \
--cc=rjw@rjwysocki.net \
--cc=robh+dt@kernel.org \
--cc=s.hauer@pengutronix.de \
--cc=srv_heupstream@mediatek.com \
--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).