From: andrew-sh.cheng <andrew-sh.cheng@mediatek.com>
To: Chanwoo Choi <cw00.choi@samsung.com>
Cc: Mark Rutland <mark.rutland@arm.com>,
devicetree@vger.kernel.org, srv_heupstream@mediatek.com,
linux-pm@vger.kernel.org, Viresh Kumar <viresh.kumar@linaro.org>,
"Rafael J. Wysocki" <rjw@rjwysocki.net>,
linux-kernel@vger.kernel.org, Rob Herring <robh+dt@kernel.org>,
fan.chen@mediatek.com, Kyungmin Park <kyungmin.park@samsung.com>,
MyungJoo Ham <myungjoo.ham@samsung.com>,
linux-mediatek@lists.infradead.org,
Matthias Brugger <matthias.bgg@gmail.com>,
linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH 3/3] devfreq: add mediatek cci devfreq
Date: Sat, 16 Feb 2019 14:16:39 +0800 [thread overview]
Message-ID: <1550297799.23797.29.camel@mtksdaap41> (raw)
In-Reply-To: <3ff7ac1c-6234-1b82-d38d-389906d295e9@samsung.com>
On Tue, 2019-02-12 at 13:06 +0900, Chanwoo Choi wrote:
> Hi Andrew-sh.Cheng,
>
> On 19. 1. 29. 오후 3:35, Andrew-sh Cheng wrote:
> > From: "Andrew-sh.Cheng" <andrew-sh.cheng@mediatek.com>
> >
> > For big/little cpu cluster architecture,
> > not only CPU frequency, but CCI frequency will also affect performance.
> >
> > Little cores and cci share the same buck in Mediatek mt8183 IC,
> > so we add a CCI devfreq which will get notification when buck voltage
> > is changed, then CCI devfreq can set cci frequency as high as possible.
> >
> > Signed-off-by: Andrew-sh.Cheng <andrew-sh.cheng@mediatek.com>
> > ---
> > drivers/devfreq/Kconfig | 9 ++
> > drivers/devfreq/Makefile | 1 +
> > drivers/devfreq/mt8183-cci-devfreq.c | 224 +++++++++++++++++++++++++++++++++++
> > 3 files changed, 234 insertions(+)
> > create mode 100644 drivers/devfreq/mt8183-cci-devfreq.c
> >
> > diff --git a/drivers/devfreq/Kconfig b/drivers/devfreq/Kconfig
> > index 6a172d3..0b14aab 100644
> > --- a/drivers/devfreq/Kconfig
> > +++ b/drivers/devfreq/Kconfig
> > @@ -113,6 +113,15 @@ config ARM_RK3399_DMC_DEVFREQ
> > It sets the frequency for the memory controller and reads the usage counts
> > from hardware.
> >
> > +config ARM_MT8183_CCI_DEVFREQ
> > + tristate "MT8183 CCI DEVFREQ Driver"
> > + depends on ARM_MEDIATEK_CPUFREQ
> > + help
> > + This adds devfreq for cci clock
> > + which is shared the same regulator with cpu cluster.
> > + It can track buck voltage and update a proper cci frequency.
> > + Use notification to get regulator status.
> > +
>
> Move ARM_MT8183_CCI_DEVFREQ under ARM_EXYNOS_BUS_DEVFREQ.
> On later, I'll change the order between ARM_TEGRA_DEVFREQ and ARM_RK3399_DMC_DEVFREQ
> alphabetically. So, you better to keep the order of name.
OK.
>
>
> > source "drivers/devfreq/event/Kconfig"
> >
> > endif # PM_DEVFREQ
> > diff --git a/drivers/devfreq/Makefile b/drivers/devfreq/Makefile
> > index 32b8d4d..25afe8c 100644
> > --- a/drivers/devfreq/Makefile
> > +++ b/drivers/devfreq/Makefile
> > @@ -11,6 +11,7 @@ obj-$(CONFIG_DEVFREQ_GOV_PASSIVE) += governor_passive.o
> > obj-$(CONFIG_ARM_EXYNOS_BUS_DEVFREQ) += exynos-bus.o
> > obj-$(CONFIG_ARM_RK3399_DMC_DEVFREQ) += rk3399_dmc.o
> > obj-$(CONFIG_ARM_TEGRA_DEVFREQ) += tegra-devfreq.o
> > +obj-$(CONFIG_ARM_MT8183_CCI_DEVFREQ) += mt8183-cci-devfreq.o
>
> Move it under obj-$(CONFIG_ARM_EXYNOS_BUS_DEVFREQ).
OK.
>
> >
> > # DEVFREQ Event Drivers
> > obj-$(CONFIG_PM_DEVFREQ_EVENT) += event/
> > diff --git a/drivers/devfreq/mt8183-cci-devfreq.c b/drivers/devfreq/mt8183-cci-devfreq.c
> > new file mode 100644
> > index 0000000..4837892
> > --- /dev/null
> > +++ b/drivers/devfreq/mt8183-cci-devfreq.c
> > @@ -0,0 +1,224 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +//
> > +// Copyright (c) 2018 MediaTek Inc.
>
> Add blank line and change year 2018 -> 2019.
OK.
>
> > +#include <linux/cpu.h>
> > +#include <linux/cpufreq.h>
>
> This driver doesn't use any helper function of cpufreq.h/cpu.h.
> It depends on devfreq/regulator framework. Please remove them.
I will remove it in next patch.
>
> > +#include <linux/devfreq.h>
> > +#include <linux/module.h>
> > +#include <linux/of.h>
> > +#include <linux/platform_device.h>
> > +#include <linux/regulator/consumer.h>
> > +
> > +#include "governor.h"
> > +
> > +struct cci_devfreq_data {
> > + struct devfreq *devfreq;
> > + struct regulator *proc_reg;
> > + struct clk *cci_clk;
> > + unsigned long buck_volt;
> > + int volt_increasing;
> > + struct notifier_block nb;
> > +};
> > +
> > +static int mtk_cci_governor_get_target(struct devfreq *devfreq,
> > + unsigned long *freq)
> > +{
> > + struct cci_devfreq_data *cci_devdata;
> > + int i, opp_count;
> > + struct dev_pm_opp *opp;
> > + unsigned long opp_rate, opp_voltage;
>
> Above three definitions use the tab for indentation. Please use space.
OK.
>
> > +
> > + cci_devdata = dev_get_drvdata(devfreq->dev.parent);
> > +
> > + /* find available frequency */
> > + opp_count = dev_pm_opp_get_opp_count(devfreq->dev.parent);
> > + for (i = 0, opp_rate = ULONG_MAX; i < opp_count; i++, opp_rate--) {
> > + opp = dev_pm_opp_find_freq_floor(devfreq->dev.parent,
> > + &opp_rate);
> > + opp_voltage = dev_pm_opp_get_voltage(opp);
> > + dev_pm_opp_put(opp);
> > +
> > + if (opp_voltage <= cci_devdata->buck_volt)
> > + break;
> > + }
> > + *freq = opp_rate;
> > +
> > + return 0;
> > +}
> > +
> > +static int mtk_cci_governor_event_handler(struct devfreq *devfreq,
> > + unsigned int event, void *data)
> > +{
> > + return 0;
> > +}
>
> The devfreq core enables the governor with using DEVFREQ_GOV_START.
> But, this driver just start after proving the device driver. It break
> the previous sequence of governor.
>
> So, if you register/unregister the regulator as following,
> it can keep the existing sequence for governor starting.
>
> case DEVFREQ_GOV_START:
> case DEVFREQ_GOV_RESUME:
> regulator_register_notifier(...);
>
> case DEVFREQ_GOV_STOP:
> case DEVFREQ_GOV_SUSPEND:
> regulator_unregister_notifier(...);
> break;
I will move regulator register/unregister notifier to
governor event handler as above in next patch.
>
> > +
> > +static struct devfreq_governor mtk_cci_devfreq_governor = {
> > + .name = "voltage_monitor",
> > + .get_target_freq = mtk_cci_governor_get_target,
> > + .event_handler = mtk_cci_governor_event_handler,
> > +};
> > +
> > +static int mtk_cci_devfreq_target(struct device *dev, unsigned long *freq,
> > + u32 flags)
> > +{
> > + struct cci_devfreq_data *cci_devdata;
> > +
> > + cci_devdata = dev_get_drvdata(dev);
>
> You better to modify it as following:
>
> struct cci_devfreq_data *cci_devdata = dev_get_drvdata(dev);
>
> if (!cci_devdata)
> return -EINVAL;
I will modify it in next patch
>
> > +
> > + clk_set_rate(cci_devdata->cci_clk, *freq);
> > +
> > + return 0;
> > +}
> > +
> > +static struct devfreq_dev_profile cci_devfreq_profile = {
> > + .polling_ms = 0,
>
> Dont' need to initialize it as zero. just remove this line.
OK.
>
> > + .target = mtk_cci_devfreq_target,
> > +};
> > +
> > +static int cci_devfreq_regulator_notifier(struct notifier_block *nb,
> > + unsigned long val, void *data)
> > +{
> > + struct cci_devfreq_data *cci_devdata =
> > + container_of(nb, struct cci_devfreq_data, nb);
> > +
> > + /* deal with reduce frequency */
> > + if (val & REGULATOR_EVENT_PRE_VOLTAGE_CHANGE) {
> > + struct pre_voltage_change_data *pvc_data = data;
> > +
> > + if (pvc_data->old_uV > pvc_data->min_uV) {
> > + cci_devdata->volt_increasing = 0;
>
> IMO, 'volt_increasing' is not clear. Instead, cci_devdata
> could have the 'previous_voltage' variable instead of volt_increasing.
> And you could use 'previous_voltage' for comparision.
OK, I will remove volt_increasing in next patch.
>
> > + cci_devdata->buck_volt =
> > + (unsigned long)(pvc_data->min_uV);
> > + mutex_lock(&cci_devdata->devfreq->lock);
> > + update_devfreq(cci_devdata->devfreq);
> > + mutex_unlock(&cci_devdata->devfreq->lock);
> > + } else {
> > + cci_devdata->volt_increasing = 1;
> > + }
> > + }
> > + /* deal with abort reduce frequency */
> > + if ((val & REGULATOR_EVENT_ABORT_VOLTAGE_CHANGE) ||
>
> if -> else if
>
> At the same time, receives only one notification.
> It is not necessary to check the two if statement always, .
OK, I will separate it in next patch.
>
>
>
> > + /* deal with increase frequency */
> > + ((val & REGULATOR_EVENT_VOLTAGE_CHANGE) &&
> > + cci_devdata->volt_increasing == 1)) {
> > + cci_devdata->buck_volt = (unsigned long)data;
> > + mutex_lock(&cci_devdata->devfreq->lock);
> > + update_devfreq(cci_devdata->devfreq);
> > + mutex_unlock(&cci_devdata->devfreq->lock);
> > + }
> > +
> > + return 0;
> > +}
> > +
> > +static int mtk_cci_devfreq_probe(struct platform_device *pdev)
> > +{
> > + struct device *cci_dev = &pdev->dev;
> > + struct cci_devfreq_data *cci_devdata;
> > + struct notifier_block *nb;
> > + int ret;
> > +
> > + dev_pm_opp_of_add_table(&pdev->dev);
>
> Move it after getting clock/regulator. After checking that all resourcess are available,
> and then you better to get the opp table. And check the return value of dev_pm_opp...()
>
OK. I will move it after checking clock/regulator are available, and add
return value checking.
> > +
> > + cci_devdata = devm_kzalloc(cci_dev, sizeof(*cci_devdata), GFP_KERNEL);
> > + if (!cci_devdata)
> > + return -ENOMEM;
> > + nb = &cci_devdata->nb;
> > + cci_devdata->cci_clk = ERR_PTR(-ENODEV);
>
> It should be handled when error happen. It is not good for readability.
I will remove it in next patch.
>
> > + cci_devdata->proc_reg = ERR_PTR(-ENODEV);
>
> ditto.
I will remove it in next patch.
>
> > +
> > + cci_devdata->cci_clk = clk_get(cci_dev, "cci_clock");
> > + ret = PTR_ERR_OR_ZERO(cci_devdata->cci_clk);
> > + if (ret) {
> > + if (ret == -EPROBE_DEFER)
> > + pr_err("cci clock not ready, retry\n");
>
> You should use 'dev_err' instead of pr_err on this driver.
>
> I think that don't show the error when EPROBE_DEFER. you better to just return.
I will change pr_err to dev_err, and remove the log with EPROBE_DEFER in
next patch.
>
> > + else
> > + pr_err("no clock for cci: %d\n", ret);
>
> How about changing the log like below?
> "failed to get clock for CCI"
OK.
>
> pr_err -> dev_err
>
> > +
> > + goto out_free_resources;
> > + }
> > +
> > + cci_devdata->proc_reg = regulator_get_optional(cci_dev, "proc");
> > + ret = PTR_ERR_OR_ZERO(cci_devdata->proc_reg);
> > + if (ret) {
> > + if (ret == -EPROBE_DEFER)
> > + pr_err("cci regulator not ready, retry\n");
>
> ditto.
>
> > + else
> > + pr_err("no regulator for cci: %d\n", ret);
>
> How about changing the log like below?
> "failed to get regulator for CCI"
OK.
>
> > +
> > + goto out_free_resources;
> > + }
> > +
> > + platform_set_drvdata(pdev, cci_devdata);
> > +
> > + /* create cci_devfreq and add to cci device.
> > + * governor is voltage_monitor.
> > + * governor will get platform_device data to make decision.
> > + */
>
> I think that you don't need to explain the kind of governor and how to
> decide the next frequency by governor. It is enough to call the devm_devfreq_add_device.
I will remove these explanation in next patch.
>
> > + cci_devdata->devfreq = devm_devfreq_add_device(cci_dev,
> > + &cci_devfreq_profile,
> > + "voltage_monitor",
> > + NULL);
> > +
> > + nb->notifier_call = cci_devfreq_regulator_notifier;
> > + regulator_register_notifier(cci_devdata->proc_reg,
> > + nb);
>
> As I commented, regulator_register_notifier() should be handled in event_handler function.
OK. I will move it into governor event handler
>
> > +
> > + return 0;
> > +
> > +out_free_resources:
> > + if (!IS_ERR(cci_devdata->cci_clk))
> > + clk_put(cci_devdata->cci_clk);
> > + if (!IS_ERR(cci_devdata->proc_reg))
> > + regulator_put(cci_devdata->proc_reg);
> > +
> > + return ret;
> > +}
> > +
> > +static const struct of_device_id mediatek_cci_devfreq_of_match[] = {
> > + { .compatible = "mediatek,mt8183-cci" },
> > + { },
> > +};
> > +MODULE_DEVICE_TABLE(of, mediatek_cci_devfreq_of_match);
> > +
> > +static struct platform_driver cci_devfreq_driver = {
> > + .probe = mtk_cci_devfreq_probe,
> > + .driver = {
> > + .name = "mediatek-cci-devfreq",
> > + .of_match_table = mediatek_cci_devfreq_of_match,
> > + },
> > +};
> > +
> > +static int __init mtk_cci_devfreq_init(void)
> > +{
> > + int ret = 0;
> > +
> > + ret = devfreq_add_governor(&mtk_cci_devfreq_governor);
> > + if (ret) {
> > + pr_err("%s: failed to add governor: %d\n", __func__, ret);
> > + return ret;
> > + }
> > +
> > + ret = platform_driver_register(&cci_devfreq_driver);
> > + if (ret)
> > + devfreq_remove_governor(&mtk_cci_devfreq_governor);
> > +
> > + return ret;
> > +}
> > +module_init(mtk_cci_devfreq_init)
> > +
> > +static void __exit mtk_cci_devfreq_exit(void)
> > +{
> > + int ret = 0;
> > +
> > + platform_driver_unregister(&cci_devfreq_driver);
> > +
> > + ret = devfreq_remove_governor(&mtk_cci_devfreq_governor);
> > + if (ret)
> > + pr_err("%s: failed to remove governor: %d\n", __func__, ret);
> > +}
> > +module_exit(mtk_cci_devfreq_exit)
> > +
> > +MODULE_LICENSE("GPL v2");
>
> nitpick. better to move it under MODULE_AUTHOR("")
I will move it after MODULE_AUTHOR :)
>
> > +MODULE_DESCRIPTION("Mediatek CCI devfreq driver");
> > +MODULE_AUTHOR("andrew-sh.cheng");
>
> Add full name and email as following:
> - MODULE_AUTHOR("Andrew-sh.Cheng <andrew-sh.cheng@mediatek.com>"
I will change to full name and email in next patch.
>
>
> > +
> >
>
> Remove unneeded blank line.
OK.
>
>
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
next prev parent reply other threads:[~2019-02-16 6:17 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-29 6:35 [PATCH 0/3] Add cpufreq and cci devfreq for mt8183 Andrew-sh Cheng
2019-01-29 6:35 ` [PATCH 1/3] cpufreq: mediatek: add mt8183 cpufreq support Andrew-sh Cheng
2019-01-29 10:13 ` Viresh Kumar
2019-02-11 13:19 ` andrew-sh.cheng
2019-01-29 6:35 ` [PATCH 2/3] dt-bindings: devfreq: add compatible for mt8183 cci devfreq Andrew-sh Cheng
2019-02-12 1:00 ` Chanwoo Choi
2019-02-16 0:09 ` andrew-sh.cheng
2019-01-29 6:35 ` [PATCH 3/3] devfreq: add mediatek " Andrew-sh Cheng
2019-02-01 3:43 ` Nicolas Boichat
2019-02-12 9:05 ` andrew-sh.cheng
2019-02-01 20:13 ` Matthias Kaehlcke
2019-02-16 3:07 ` andrew-sh.cheng
2019-02-12 4:06 ` Chanwoo Choi
2019-02-16 6:16 ` andrew-sh.cheng [this message]
[not found] ` <CGME20190129063600epcas2p255b062c64f22555692ff895634ea4eb0@epcms1p4>
2019-01-29 8:17 ` MyungJoo Ham
2019-02-11 13:21 ` andrew-sh.cheng
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=1550297799.23797.29.camel@mtksdaap41 \
--to=andrew-sh.cheng@mediatek.com \
--cc=cw00.choi@samsung.com \
--cc=devicetree@vger.kernel.org \
--cc=fan.chen@mediatek.com \
--cc=kyungmin.park@samsung.com \
--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=myungjoo.ham@samsung.com \
--cc=rjw@rjwysocki.net \
--cc=robh+dt@kernel.org \
--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).