Message ID | 20160901025328.376-3-d-gerlach@ti.com (mailing list archive) |
---|---|
State | Changes Requested, archived |
Headers | show |
On 31-08-16, 21:53, Dave Gerlach wrote: > Some TI SoCs, like those in the AM335x, AM437x, DRA7x, and AM57x families, > have different OPPs available for the MPU depending on which specific > variant of the SoC is in use. This can be determined through use of the > revision and an eFuse register present in the silicon. Introduce a > ti-cpufreq driver that can read the aformentioned values and provide > them as version matching data to the opp framework. Through this the > opp-supported-hw dt binding that is part of the operating-points-v2 > table can be used to indicate availability of OPPs for each device. > > This driver also creates the "cpufreq-dt" platform_device after passing > the version matching data to the OPP framework so that the cpufreq-dt > handles the actual cpufreq implementation. Even without the necessary > data to pass the version matching data the driver will still create this > device to maintain backwards compatibility with operating-points v1 > tables. > > Signed-off-by: Dave Gerlach <d-gerlach@ti.com> > --- > v1->v2: > - Convert to module_platform_driver to match against new compatibles > in patch 1 > - Cleaned up some bit shifts > - of_property_read_u32_array used rather than reading values individually > > drivers/cpufreq/Kconfig.arm | 11 ++ > drivers/cpufreq/Makefile | 1 + > drivers/cpufreq/ti-cpufreq.c | 308 +++++++++++++++++++++++++++++++++++++++++++ > 3 files changed, 320 insertions(+) > create mode 100644 drivers/cpufreq/ti-cpufreq.c I am wondering if we should start writing OPP drivers instead. As this patch doesn't have anything to do with cpufreq really :) But its fine for now.. > +static const struct of_device_id ti_cpufreq_of_match[] = { > + { .compatible = "operating-points-v2-ti-am3352-cpu", > + .data = &am3x_soc_data, }, > + { .compatible = "operating-points-v2-ti-am4372-cpu", > + .data = &am4x_soc_data, }, > + { .compatible = "operating-points-v2-ti-dra7-cpu", > + .data = &dra7_soc_data }, You should be using your SoC compatible strings here. OPP compatible property isn't supposed to be (mis)used for this purpose.
On 09/07/2016 12:20 AM, Viresh Kumar wrote: > On 31-08-16, 21:53, Dave Gerlach wrote: >> Some TI SoCs, like those in the AM335x, AM437x, DRA7x, and AM57x families, >> have different OPPs available for the MPU depending on which specific >> variant of the SoC is in use. This can be determined through use of the >> revision and an eFuse register present in the silicon. Introduce a >> ti-cpufreq driver that can read the aformentioned values and provide >> them as version matching data to the opp framework. Through this the >> opp-supported-hw dt binding that is part of the operating-points-v2 >> table can be used to indicate availability of OPPs for each device. >> >> This driver also creates the "cpufreq-dt" platform_device after passing >> the version matching data to the OPP framework so that the cpufreq-dt >> handles the actual cpufreq implementation. Even without the necessary >> data to pass the version matching data the driver will still create this >> device to maintain backwards compatibility with operating-points v1 >> tables. >> >> Signed-off-by: Dave Gerlach <d-gerlach@ti.com> >> --- >> v1->v2: >> - Convert to module_platform_driver to match against new compatibles >> in patch 1 >> - Cleaned up some bit shifts >> - of_property_read_u32_array used rather than reading values individually >> >> drivers/cpufreq/Kconfig.arm | 11 ++ >> drivers/cpufreq/Makefile | 1 + >> drivers/cpufreq/ti-cpufreq.c | 308 +++++++++++++++++++++++++++++++++++++++++++ >> 3 files changed, 320 insertions(+) >> create mode 100644 drivers/cpufreq/ti-cpufreq.c > > I am wondering if we should start writing OPP drivers instead. As this > patch doesn't have anything to do with cpufreq really :) > > But its fine for now.. > >> +static const struct of_device_id ti_cpufreq_of_match[] = { >> + { .compatible = "operating-points-v2-ti-am3352-cpu", >> + .data = &am3x_soc_data, }, >> + { .compatible = "operating-points-v2-ti-am4372-cpu", >> + .data = &am4x_soc_data, }, >> + { .compatible = "operating-points-v2-ti-dra7-cpu", >> + .data = &dra7_soc_data }, > > You should be using your SoC compatible strings here. OPP compatible > property isn't supposed to be (mis)used for this purpose. > Referring to my comments in patch 1, what if we end up changing the bindings based on DT maintainer comments? We will have these compatible strings, and at that point is it acceptable to match against them? Or is it still better to match to SoC compatibles? I think it makes sense to just probe against these. Regards, Dave -- 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
On 07-09-16, 10:04, Dave Gerlach wrote: > >>+static const struct of_device_id ti_cpufreq_of_match[] = { > >>+ { .compatible = "operating-points-v2-ti-am3352-cpu", > >>+ .data = &am3x_soc_data, }, > >>+ { .compatible = "operating-points-v2-ti-am4372-cpu", > >>+ .data = &am4x_soc_data, }, > >>+ { .compatible = "operating-points-v2-ti-dra7-cpu", > >>+ .data = &dra7_soc_data }, > > > >You should be using your SoC compatible strings here. OPP compatible > >property isn't supposed to be (mis)used for this purpose. > > > > Referring to my comments in patch 1, what if we end up changing the bindings > based on DT maintainer comments? We will have these compatible strings, and > at that point is it acceptable to match against them? Or is it still better > to match to SoC compatibles? I think it makes sense to just probe against > these. But even then I think these are not correct. You should have added a single compatible string: operating-points-v2-ti-cpu. As the properties will stay the same across machines. And then you need to use SoC strings here.
Viresh, On 09/07/2016 10:39 PM, Viresh Kumar wrote: > On 07-09-16, 10:04, Dave Gerlach wrote: >>>> +static const struct of_device_id ti_cpufreq_of_match[] = { >>>> + { .compatible = "operating-points-v2-ti-am3352-cpu", >>>> + .data = &am3x_soc_data, }, >>>> + { .compatible = "operating-points-v2-ti-am4372-cpu", >>>> + .data = &am4x_soc_data, }, >>>> + { .compatible = "operating-points-v2-ti-dra7-cpu", >>>> + .data = &dra7_soc_data }, >>> >>> You should be using your SoC compatible strings here. OPP compatible >>> property isn't supposed to be (mis)used for this purpose. >>> >> >> Referring to my comments in patch 1, what if we end up changing the bindings >> based on DT maintainer comments? We will have these compatible strings, and >> at that point is it acceptable to match against them? Or is it still better >> to match to SoC compatibles? I think it makes sense to just probe against >> these. > > But even then I think these are not correct. You should have added a > single compatible string: operating-points-v2-ti-cpu. > > As the properties will stay the same across machines. And then you > need to use SoC strings here. > Are you opposed to moving _of_get_opp_desc_node from drivers/base/power/opp/opp.h to include/linux/pm_opp.h and renaming it appropriately? If I move the ti properties out of the cpu node, as discussed in patch 1 of this series, and into the operating-points-v2 table, I need a way to get the operating-points-v2 device node and I think it makes sense to reuse this as it is what the opp framework uses internally to parse the phandle to the opp table. Regards, Dave -- 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
On 21-09-16, 14:34, Dave Gerlach wrote: > Viresh, > On 09/07/2016 10:39 PM, Viresh Kumar wrote: > >On 07-09-16, 10:04, Dave Gerlach wrote: > >>>>+static const struct of_device_id ti_cpufreq_of_match[] = { > >>>>+ { .compatible = "operating-points-v2-ti-am3352-cpu", > >>>>+ .data = &am3x_soc_data, }, > >>>>+ { .compatible = "operating-points-v2-ti-am4372-cpu", > >>>>+ .data = &am4x_soc_data, }, > >>>>+ { .compatible = "operating-points-v2-ti-dra7-cpu", > >>>>+ .data = &dra7_soc_data }, > >>> > >>>You should be using your SoC compatible strings here. OPP compatible > >>>property isn't supposed to be (mis)used for this purpose. > >>> > >> > >>Referring to my comments in patch 1, what if we end up changing the bindings > >>based on DT maintainer comments? We will have these compatible strings, and > >>at that point is it acceptable to match against them? Or is it still better > >>to match to SoC compatibles? I think it makes sense to just probe against > >>these. > > > >But even then I think these are not correct. You should have added a > >single compatible string: operating-points-v2-ti-cpu. > > > >As the properties will stay the same across machines. And then you > >need to use SoC strings here. > > > > Are you opposed to moving _of_get_opp_desc_node from > drivers/base/power/opp/opp.h to include/linux/pm_opp.h and renaming it > appropriately? I am not opposed to that, but ... > If I move the ti properties out of the cpu node, as discussed in patch 1 of > this series, and into the operating-points-v2 table, I need a way to get the > operating-points-v2 device node and I think it makes sense to reuse this as > it is what the opp framework uses internally to parse the phandle to the opp > table. I am not sure if those registers belong to the OPP bindings. What are those registers really? What all can be read from them? Why shouldn't they be present as a separate node in DT on the respective bus? Look at how it is done for sti-cpufreq driver.
On 09/23/2016 12:19 AM, Viresh Kumar wrote: > On 21-09-16, 14:34, Dave Gerlach wrote: >> Viresh, >> On 09/07/2016 10:39 PM, Viresh Kumar wrote: >>> On 07-09-16, 10:04, Dave Gerlach wrote: >>>>>> +static const struct of_device_id ti_cpufreq_of_match[] = { >>>>>> + { .compatible = "operating-points-v2-ti-am3352-cpu", >>>>>> + .data = &am3x_soc_data, }, >>>>>> + { .compatible = "operating-points-v2-ti-am4372-cpu", >>>>>> + .data = &am4x_soc_data, }, >>>>>> + { .compatible = "operating-points-v2-ti-dra7-cpu", >>>>>> + .data = &dra7_soc_data }, >>>>> >>>>> You should be using your SoC compatible strings here. OPP compatible >>>>> property isn't supposed to be (mis)used for this purpose. >>>>> >>>> >>>> Referring to my comments in patch 1, what if we end up changing the bindings >>>> based on DT maintainer comments? We will have these compatible strings, and >>>> at that point is it acceptable to match against them? Or is it still better >>>> to match to SoC compatibles? I think it makes sense to just probe against >>>> these. >>> >>> But even then I think these are not correct. You should have added a >>> single compatible string: operating-points-v2-ti-cpu. >>> >>> As the properties will stay the same across machines. And then you >>> need to use SoC strings here. >>> >> >> Are you opposed to moving _of_get_opp_desc_node from >> drivers/base/power/opp/opp.h to include/linux/pm_opp.h and renaming it >> appropriately? > > I am not opposed to that, but ... > >> If I move the ti properties out of the cpu node, as discussed in patch 1 of >> this series, and into the operating-points-v2 table, I need a way to get the >> operating-points-v2 device node and I think it makes sense to reuse this as >> it is what the opp framework uses internally to parse the phandle to the opp >> table. > > I am not sure if those registers belong to the OPP bindings. What are those > registers really? What all can be read from them? Why shouldn't they be present > as a separate node in DT on the respective bus? Look at how it is done for > sti-cpufreq driver. > The sti-cpufreq driver in v4.8-rc7 appears to do what I am already doing in this revision of the patch, reading from a syscon phandle that is part of the cpu node in the DT which is what I was told not to do. The register I am referencing in the syscon is a bit-field describing which OPPs are valid for the system, so it is very relevant to the OPP binding. They really are already present in a separate node, I'm just indexing into a syscon, same as the sti-cpufreq driver appears to be doing. Regards, Dave -- 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
On 23-09-16, 11:17, Dave Gerlach wrote: > On 09/23/2016 12:19 AM, Viresh Kumar wrote: > >On 21-09-16, 14:34, Dave Gerlach wrote: > >>Viresh, > >>On 09/07/2016 10:39 PM, Viresh Kumar wrote: > >>>On 07-09-16, 10:04, Dave Gerlach wrote: > >>>>>>+static const struct of_device_id ti_cpufreq_of_match[] = { > >>>>>>+ { .compatible = "operating-points-v2-ti-am3352-cpu", > >>>>>>+ .data = &am3x_soc_data, }, > >>>>>>+ { .compatible = "operating-points-v2-ti-am4372-cpu", > >>>>>>+ .data = &am4x_soc_data, }, > >>>>>>+ { .compatible = "operating-points-v2-ti-dra7-cpu", > >>>>>>+ .data = &dra7_soc_data }, > >>>>> > >>>>>You should be using your SoC compatible strings here. OPP compatible > >>>>>property isn't supposed to be (mis)used for this purpose. > >>>>> > >>>> > >>>>Referring to my comments in patch 1, what if we end up changing the bindings > >>>>based on DT maintainer comments? We will have these compatible strings, and > >>>>at that point is it acceptable to match against them? Or is it still better > >>>>to match to SoC compatibles? I think it makes sense to just probe against > >>>>these. > >>> > >>>But even then I think these are not correct. You should have added a > >>>single compatible string: operating-points-v2-ti-cpu. > >>> > >>>As the properties will stay the same across machines. And then you > >>>need to use SoC strings here. > >>> > >> > >>Are you opposed to moving _of_get_opp_desc_node from > >>drivers/base/power/opp/opp.h to include/linux/pm_opp.h and renaming it > >>appropriately? > > > >I am not opposed to that, but ... > > > >>If I move the ti properties out of the cpu node, as discussed in patch 1 of > >>this series, and into the operating-points-v2 table, I need a way to get the > >>operating-points-v2 device node and I think it makes sense to reuse this as > >>it is what the opp framework uses internally to parse the phandle to the opp > >>table. > > > >I am not sure if those registers belong to the OPP bindings. What are those > >registers really? What all can be read from them? Why shouldn't they be present > >as a separate node in DT on the respective bus? Look at how it is done for > >sti-cpufreq driver. > > > > The sti-cpufreq driver in v4.8-rc7 appears to do what I am already doing in > this revision of the patch, reading from a syscon phandle that is part of > the cpu node in the DT which is what I was told not to do. > > The register I am referencing in the syscon is a bit-field describing which > OPPs are valid for the system, so it is very relevant to the OPP binding. > They really are already present in a separate node, I'm just indexing into a > syscon, same as the sti-cpufreq driver appears to be doing. Okay, you can move that function out.
diff --git a/drivers/cpufreq/Kconfig.arm b/drivers/cpufreq/Kconfig.arm index d89b8afe23b6..0dea6849ac3e 100644 --- a/drivers/cpufreq/Kconfig.arm +++ b/drivers/cpufreq/Kconfig.arm @@ -234,6 +234,17 @@ config ARM_TEGRA124_CPUFREQ help This adds the CPUFreq driver support for Tegra124 SOCs. +config ARM_TI_CPUFREQ + tristate "Texas Instruments CPUFreq support" + depends on ARCH_OMAP2PLUS + help + This driver enables valid OPPs on the running platform based on + values contained within the SoC in use. Enable this in order to + use the cpufreq-dt driver on all Texas Instruments platforms that + provide dt based operating-points-v2 tables with opp-supported-hw + data provided. Required for cpufreq support on AM335x, AM437x, + DRA7x, and AM57x platforms. + config ARM_PXA2xx_CPUFREQ tristate "Intel PXA2xx CPUfreq driver" depends on PXA27x || PXA25x diff --git a/drivers/cpufreq/Makefile b/drivers/cpufreq/Makefile index 0a9b6a093646..5b1b6ec0a9f0 100644 --- a/drivers/cpufreq/Makefile +++ b/drivers/cpufreq/Makefile @@ -77,6 +77,7 @@ obj-$(CONFIG_ARM_SPEAR_CPUFREQ) += spear-cpufreq.o obj-$(CONFIG_ARM_STI_CPUFREQ) += sti-cpufreq.o obj-$(CONFIG_ARM_TEGRA20_CPUFREQ) += tegra20-cpufreq.o obj-$(CONFIG_ARM_TEGRA124_CPUFREQ) += tegra124-cpufreq.o +obj-$(CONFIG_ARM_TI_CPUFREQ) += ti-cpufreq.o obj-$(CONFIG_ARM_VEXPRESS_SPC_CPUFREQ) += vexpress-spc-cpufreq.o obj-$(CONFIG_ACPI_CPPC_CPUFREQ) += cppc_cpufreq.o obj-$(CONFIG_MACH_MVEBU_V7) += mvebu-cpufreq.o diff --git a/drivers/cpufreq/ti-cpufreq.c b/drivers/cpufreq/ti-cpufreq.c new file mode 100644 index 000000000000..819aff3ba449 --- /dev/null +++ b/drivers/cpufreq/ti-cpufreq.c @@ -0,0 +1,308 @@ +/* + * TI CPUFreq/OPP hw-supported driver + * + * Copyright (C) 2016 Texas Instruments, Inc. + * Dave Gerlach <d-gerlach@ti.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * version 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <linux/cpu.h> +#include <linux/io.h> +#include <linux/mfd/syscon.h> +#include <linux/module.h> +#include <linux/of.h> +#include <linux/of_platform.h> +#include <linux/pm_opp.h> +#include <linux/regmap.h> + +#define REVISION_MASK 0xF +#define REVISION_SHIFT 28 + +#define AM33XX_800M_ARM_MPU_MAX_FREQ 0x1E2F +#define AM43XX_600M_ARM_MPU_MAX_FREQ 0xFFA + +#define DRA7_EFUSE_HAS_OD_MPU_OPP 11 +#define DRA7_EFUSE_HAS_HIGH_MPU_OPP 15 +#define DRA7_EFUSE_HAS_ALL_MPU_OPP 23 + +#define DRA7_EFUSE_NOM_MPU_OPP BIT(0) +#define DRA7_EFUSE_OD_MPU_OPP BIT(1) +#define DRA7_EFUSE_HIGH_MPU_OPP BIT(2) + +#define VERSION_COUNT 2 + +struct ti_cpufreq_data; + +struct ti_cpufreq_soc_data { + unsigned long (*efuse_xlate)(struct ti_cpufreq_data *opp_data, + unsigned long efuse); + unsigned long efuse_fallback; +}; + +struct ti_cpufreq_data { + struct device *cpu_dev; + struct device *opp_dev; + struct regmap *opp_efuse; + struct regmap *revision; + const struct ti_cpufreq_soc_data *soc_data; +}; + +static unsigned long amx3_efuse_xlate(struct ti_cpufreq_data *opp_data, + unsigned long efuse) +{ + if (!efuse) + efuse = opp_data->soc_data->efuse_fallback; + /* AM335x and AM437x use "OPP disable" bits, so invert */ + return ~efuse; +} + +static unsigned long dra7_efuse_xlate(struct ti_cpufreq_data *opp_data, + unsigned long efuse) +{ + unsigned long calculated_efuse = DRA7_EFUSE_NOM_MPU_OPP; + + /* + * The efuse on dra7 and am57 parts contains a specific + * value indicating the highest available OPP. + */ + + switch (efuse) { + case DRA7_EFUSE_HAS_ALL_MPU_OPP: + case DRA7_EFUSE_HAS_HIGH_MPU_OPP: + calculated_efuse |= DRA7_EFUSE_HIGH_MPU_OPP; + case DRA7_EFUSE_HAS_OD_MPU_OPP: + calculated_efuse |= DRA7_EFUSE_OD_MPU_OPP; + } + + return calculated_efuse; +} + +static struct ti_cpufreq_soc_data am3x_soc_data = { + .efuse_xlate = amx3_efuse_xlate, + .efuse_fallback = AM33XX_800M_ARM_MPU_MAX_FREQ, +}; + +static struct ti_cpufreq_soc_data am4x_soc_data = { + .efuse_xlate = amx3_efuse_xlate, + .efuse_fallback = AM43XX_600M_ARM_MPU_MAX_FREQ, +}; + +static struct ti_cpufreq_soc_data dra7_soc_data = { + .efuse_xlate = dra7_efuse_xlate, +}; + +/** + * ti_cpufreq_get_efuse() - Parse and return efuse value present on SoC + * @opp_data: pointer to ti_cpufreq_data context + * @efuse_value: Set to the value parsed from efuse + * + * Returns error code if efuse not read properly. + */ +static int ti_cpufreq_get_efuse(struct ti_cpufreq_data *opp_data, + u32 *efuse_value) +{ + struct device *dev = opp_data->cpu_dev; + struct device_node *np = dev->of_node; + unsigned int efuse_offset; + u32 efuse, efuse_mask, efuse_shift, vals[4]; + int ret; + + ret = of_property_read_u32_array(np, "ti,syscon-efuse", vals, 4); + if (ret) { + dev_err(dev, "ti,syscon-efuse cannot be read %s: %d\n", + np->full_name, ret); + return ret; + } + + efuse_offset = vals[1]; + efuse_mask = vals[2]; + efuse_shift = vals[3]; + + ret = regmap_read(opp_data->opp_efuse, efuse_offset, &efuse); + if (ret) { + dev_err(dev, + "Failed to read the efuse value from syscon: %d\n", + ret); + return ret; + } + + efuse = (efuse & efuse_mask) >> efuse_shift; + + *efuse_value = opp_data->soc_data->efuse_xlate(opp_data, efuse); + + return 0; +} + +/** + * ti_cpufreq_get_rev() - Parse and return rev value present on SoC + * @opp_data: pointer to ti_cpufreq_data context + * @revision_value: Set to the value parsed from revision register + * + * Returns error code if revision not read properly. + */ +static int ti_cpufreq_get_rev(struct ti_cpufreq_data *opp_data, + u32 *revision_value) +{ + struct device *dev = opp_data->cpu_dev; + struct device_node *np = dev->of_node; + unsigned int revision_offset; + u32 revision; + int ret; + + ret = of_property_read_u32_index(np, "ti,syscon-rev", + 1, &revision_offset); + if (ret) { + dev_err(dev, + "No revision offset provided %s [%d]\n", + np->full_name, ret); + return ret; + } + + ret = regmap_read(opp_data->revision, revision_offset, &revision); + if (ret) { + dev_err(dev, + "Failed to read the revision number from syscon: %d\n", + ret); + return ret; + } + + *revision_value = BIT((revision >> REVISION_SHIFT) & REVISION_MASK); + + return 0; +} + +static int ti_cpufreq_setup_syscon_registers(struct ti_cpufreq_data *opp_data) +{ + struct device *dev = opp_data->cpu_dev; + struct device_node *np = dev->of_node; + + opp_data->opp_efuse = syscon_regmap_lookup_by_phandle(np, + "ti,syscon-efuse"); + if (IS_ERR(opp_data->opp_efuse)) { + dev_err(dev, + "\"ti,syscon-efuse\" is missing, cannot use OPPv2 table.\n"); + return PTR_ERR(opp_data->opp_efuse); + } + + opp_data->revision = syscon_regmap_lookup_by_phandle(np, + "ti,syscon-rev"); + if (IS_ERR(opp_data->revision)) { + dev_err(dev, + "\"ti,syscon-rev\" is missing, cannot use OPPv2 table.\n"); + return PTR_ERR(opp_data->revision); + } + + return 0; +} + +static const struct of_device_id ti_cpufreq_of_match[] = { + { .compatible = "operating-points-v2-ti-am3352-cpu", + .data = &am3x_soc_data, }, + { .compatible = "operating-points-v2-ti-am4372-cpu", + .data = &am4x_soc_data, }, + { .compatible = "operating-points-v2-ti-dra7-cpu", + .data = &dra7_soc_data }, + {}, +}; +MODULE_DEVICE_TABLE(of, ti_cpufreq_of_match); + +static int ti_cpufreq_probe(struct platform_device *pdev) +{ + int ret; + u32 version[VERSION_COUNT]; + struct device *dev = &pdev->dev; + struct device_node *np = dev->of_node; + const struct of_device_id *match; + struct ti_cpufreq_data *opp_data; + + if (!np) + return -ENODEV; + + match = of_match_node(ti_cpufreq_of_match, np); + if (!match) + return -ENODEV; + + opp_data = devm_kzalloc(dev, sizeof(*opp_data), GFP_KERNEL); + if (!opp_data) + return -ENOMEM; + + opp_data->soc_data = match->data; + opp_data->opp_dev = dev; + + opp_data->cpu_dev = get_cpu_device(0); + if (!opp_data->cpu_dev) { + pr_err("%s: Failed to get device for CPU0\n", __func__); + return -ENODEV; + } + + if (!of_get_property(opp_data->cpu_dev->of_node, "operating-points-v2", + NULL)) { + dev_info(opp_data->cpu_dev, + "OPP-v2 not supported, cpufreq-dt will attempt to use legacy tables.\n"); + goto register_cpufreq_dt; + } + + ret = ti_cpufreq_setup_syscon_registers(opp_data); + if (ret) + return ret; + + /* + * OPPs determine whether or not they are supported based on + * two metrics: + * 0 - SoC Revision + * 1 - eFuse value + */ + ret = ti_cpufreq_get_rev(opp_data, &version[0]); + if (ret) + return ret; + + ret = ti_cpufreq_get_efuse(opp_data, &version[1]); + if (ret) + return ret; + + ret = dev_pm_opp_set_supported_hw(opp_data->cpu_dev, version, + VERSION_COUNT); + if (ret) { + dev_err(opp_data->cpu_dev, + "Failed to set supported hardware\n"); + return ret; + } + +register_cpufreq_dt: + dev_set_drvdata(dev, opp_data); + platform_device_register_simple("cpufreq-dt", -1, NULL, 0); + + return 0; +} + +static int ti_cpufreq_remove(struct platform_device *pdev) +{ + struct ti_cpufreq_data *opp_data = platform_get_drvdata(pdev); + + dev_pm_opp_put_supported_hw(opp_data->cpu_dev); + + return 0; +} + +static struct platform_driver ti_cpufreq_driver = { + .probe = ti_cpufreq_probe, + .remove = ti_cpufreq_remove, + .driver = { + .name = "ti_cpufreq", + .of_match_table = ti_cpufreq_of_match, + }, +}; + +module_platform_driver(ti_cpufreq_driver); + +MODULE_DESCRIPTION("TI CPUFreq/OPP hw-supported driver"); +MODULE_AUTHOR("Dave Gerlach <d-gerlach@ti.com>"); +MODULE_LICENSE("GPL v2");
Some TI SoCs, like those in the AM335x, AM437x, DRA7x, and AM57x families, have different OPPs available for the MPU depending on which specific variant of the SoC is in use. This can be determined through use of the revision and an eFuse register present in the silicon. Introduce a ti-cpufreq driver that can read the aformentioned values and provide them as version matching data to the opp framework. Through this the opp-supported-hw dt binding that is part of the operating-points-v2 table can be used to indicate availability of OPPs for each device. This driver also creates the "cpufreq-dt" platform_device after passing the version matching data to the OPP framework so that the cpufreq-dt handles the actual cpufreq implementation. Even without the necessary data to pass the version matching data the driver will still create this device to maintain backwards compatibility with operating-points v1 tables. Signed-off-by: Dave Gerlach <d-gerlach@ti.com> --- v1->v2: - Convert to module_platform_driver to match against new compatibles in patch 1 - Cleaned up some bit shifts - of_property_read_u32_array used rather than reading values individually drivers/cpufreq/Kconfig.arm | 11 ++ drivers/cpufreq/Makefile | 1 + drivers/cpufreq/ti-cpufreq.c | 308 +++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 320 insertions(+) create mode 100644 drivers/cpufreq/ti-cpufreq.c