Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <a03d6823-2084-e4c6-1c9d-853127956f69@quicinc.com>
Date:   Mon, 10 Jul 2023 18:52:00 +0530
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:102.0) Gecko/20100101
 Thunderbird/102.13.0
Subject: Re: [PATCH 2/6] thermal/drivers/tsens: Add TSENS enable and
 calibration support for V2
Content-Language: en-US
To:     Dmitry Baryshkov <dmitry.baryshkov@linaro.org>,
        <agross@kernel.org>, <andersson@kernel.org>,
        <konrad.dybcio@linaro.org>, <amitk@kernel.org>,
        <thara.gopinath@gmail.com>, <rafael@kernel.org>,
        <daniel.lezcano@linaro.org>, <rui.zhang@intel.com>,
        <robh+dt@kernel.org>, <krzysztof.kozlowski+dt@linaro.org>,
        <conor+dt@kernel.org>, <linux-arm-msm@vger.kernel.org>,
        <linux-pm@vger.kernel.org>, <devicetree@vger.kernel.org>,
        <linux-kernel@vger.kernel.org>
CC:     <quic_varada@quicinc.com>
References: <20230710103735.1375847-1-quic_ipkumar@quicinc.com>
 <20230710103735.1375847-3-quic_ipkumar@quicinc.com>
 <cf2cae88-e226-8df1-6c4f-543b683cbc8c@linaro.org>
From:   Praveenkumar I <quic_ipkumar@quicinc.com>
In-Reply-To: <cf2cae88-e226-8df1-6c4f-543b683cbc8c@linaro.org>
Content-Type: text/plain; charset="UTF-8"; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk


On 7/10/2023 4:49 PM, Dmitry Baryshkov wrote:
> On 10/07/2023 13:37, Praveenkumar I wrote:
>> SoCs without RPM have to enable sensors and calibrate from the kernel.
>> Though TSENS IP supports 16 sensors, not all are used. So added
>> sensors_to_en in tsens data help enable the relevant sensors.
>>
>> Added new calibration function for V2 as the tsens.c calib function
>> only supports V1.
>>
>> Signed-off-by: Praveenkumar I <quic_ipkumar@quicinc.com>
>> ---
>>   drivers/thermal/qcom/tsens-v2.c | 116 ++++++++++++++++++++++++++++++++
>>   drivers/thermal/qcom/tsens.c    |  37 +++++++++-
>>   drivers/thermal/qcom/tsens.h    |  56 +++++++++++++++
>>   3 files changed, 208 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/thermal/qcom/tsens-v2.c 
>> b/drivers/thermal/qcom/tsens-v2.c
>> index 29a61d2d6ca3..db48b1d95348 100644
>> --- a/drivers/thermal/qcom/tsens-v2.c
>> +++ b/drivers/thermal/qcom/tsens-v2.c
>> @@ -6,11 +6,20 @@
>>     #include <linux/bitops.h>
>>   #include <linux/regmap.h>
>> +#include <linux/nvmem-consumer.h>
>>   #include "tsens.h"
>>     /* ----- SROT ------ */
>>   #define SROT_HW_VER_OFF    0x0000
>>   #define SROT_CTRL_OFF        0x0004
>> +#define SROT_MEASURE_PERIOD    0x0008
>> +#define SROT_Sn_CONVERSION    0x0060
>> +#define V2_SHIFT_DEFAULT    0x0003
>> +#define V2_SLOPE_DEFAULT    0x0cd0
>> +#define V2_CZERO_DEFAULT    0x016a
>> +#define ONE_PT_SLOPE        0x0cd0
>> +#define TWO_PT_SHIFTED_GAIN    921600
>> +#define ONE_PT_CZERO_CONST    94
>>     /* ----- TM ------ */
>>   #define TM_INT_EN_OFF            0x0004
>> @@ -59,6 +68,16 @@ static const struct reg_field 
>> tsens_v2_regfields[MAX_REGFIELDS] = {
>>       /* CTRL_OFF */
>>       [TSENS_EN]     = REG_FIELD(SROT_CTRL_OFF,    0,  0),
>>       [TSENS_SW_RST] = REG_FIELD(SROT_CTRL_OFF,    1,  1),
>> +    [SENSOR_EN]    = REG_FIELD(SROT_CTRL_OFF,    3,  18),
>> +    [CODE_OR_TEMP] = REG_FIELD(SROT_CTRL_OFF,    21, 21),
>> +
>> +    /* MAIN_MEASURE_PERIOD */
>> +    [MAIN_MEASURE_PERIOD] = REG_FIELD(SROT_MEASURE_PERIOD, 0, 7),
>> +
>> +    /* Sn Conversion */
>> +    REG_FIELD_FOR_EACH_SENSOR16(SHIFT, SROT_Sn_CONVERSION, 23, 24),
>> +    REG_FIELD_FOR_EACH_SENSOR16(SLOPE, SROT_Sn_CONVERSION, 10, 22),
>> +    REG_FIELD_FOR_EACH_SENSOR16(CZERO, SROT_Sn_CONVERSION, 0, 9),
>>         /* ----- TM ------ */
>>       /* INTERRUPT ENABLE */
>> @@ -104,6 +123,103 @@ static const struct reg_field 
>> tsens_v2_regfields[MAX_REGFIELDS] = {
>>       [TRDY] = REG_FIELD(TM_TRDY_OFF, 0, 0),
>>   };
>>   +static int tsens_v2_calibration(struct tsens_priv *priv)
>> +{
>> +    struct device *dev = priv->dev;
>> +    u32 mode, base0, base1;
>> +    u32 slope, czero;
>> +    char name[15];
>> +    int i, j, ret;
>> +
>> +    if (priv->num_sensors > MAX_SENSORS)
>> +        return -EINVAL;
>> +
>> +    ret = nvmem_cell_read_variable_le_u32(priv->dev, "mode", &mode);
>> +    if (ret == -ENOENT)
>> +        dev_warn(priv->dev, "Calibration data not present in DT\n");
>> +    if (ret < 0)
>> +        return ret;
>> +
>> +    dev_dbg(priv->dev, "calibration mode is %d\n", mode);
>> +
>> +    ret = nvmem_cell_read_variable_le_u32(priv->dev, "base0", &base0);
>> +    if (ret < 0)
>> +        return ret;
>> +
>> +    ret = nvmem_cell_read_variable_le_u32(priv->dev, "base1", &base1);
>> +    if (ret < 0)
>> +        return ret;
>> +
>> +    /* Read offset values and allocate SHIFT, SLOPE & CZERO regmap 
>> for enabled sensors */
>> +    for (i = 0; i < priv->num_sensors; i++) {
>> +        if (!(priv->sensors_to_en & (0x1 << i)))
>> +            continue;
>> +
>> +        ret = snprintf(name, sizeof(name), "s%d_offset", 
>> priv->sensor[i].hw_id);
>> +        if (ret < 0)
>> +            return ret;
>> +
>> +        ret = nvmem_cell_read_variable_le_u32(priv->dev, name, 
>> &priv->sensor[i].offset);
>> +        if (ret)
>> +            return ret;
>> +
>> +        for (j = SHIFT_0; j <= CZERO_0; j++) {
>> +            int idx = (i * 3) + j;
>> +
>> +            priv->rf[idx] = devm_regmap_field_alloc(dev, 
>> priv->srot_map,
>> +                                priv->fields[idx]);
>> +            if (IS_ERR(priv->rf[idx]))
>> +                return PTR_ERR(priv->rf[idx]);
>
> I think, allocating data structures for 48 regfields, which are 
> written just once, to be an overkill.
Can we change it to single field for each sensor. For example, 
CONVERSION_0 instead of SHIFT_0, SLOPE_0 and CZERO_0? This way it will 
be max 16 regfields.
>
>> +        }
>> +    }
>> +
>> +    /* Based on calib mode, program SHIFT, SLOPE and CZERO for 
>> enabled sensors */
>> +    switch (mode) {
>> +    case TWO_PT_CALIB:
>> +        slope = (TWO_PT_SHIFTED_GAIN / (base1 - base0));
>> +
>> +        for (i = 0; i < priv->num_sensors; i++) {
>> +            if (!(priv->sensors_to_en & (0x1 << i)))
>> +                continue;
>> +
>> +            int idx = i * 3;
>> +
>> +            czero = (base0 + priv->sensor[i].offset - ((base1 - 
>> base0) / 3));
>> +            regmap_field_write(priv->rf[SHIFT_0 + idx], 
>> V2_SHIFT_DEFAULT);
>> +            regmap_field_write(priv->rf[SLOPE_0 + idx], slope);
>> +            regmap_field_write(priv->rf[CZERO_0 + idx], czero);
>> +        }
>> +        fallthrough;
>> +    case ONE_PT_CALIB2:
>> +        for (i = 0; i < priv->num_sensors; i++) {
>> +            if (!(priv->sensors_to_en & (0x1 << i)))
>> +                continue;
>> +
>> +            int idx = i * 3;
>> +
>> +            czero = base0 + priv->sensor[i].offset - 
>> ONE_PT_CZERO_CONST;
>> +            regmap_field_write(priv->rf[SHIFT_0 + idx], 
>> V2_SHIFT_DEFAULT);
>> +            regmap_field_write(priv->rf[SLOPE_0 + idx], ONE_PT_SLOPE);
>> +            regmap_field_write(priv->rf[CZERO_0 + idx], czero);
>> +        }
>> +        break;
>> +    default:
>> +        dev_dbg(priv->dev, "calibrationless mode\n");
>> +        for (i = 0; i < priv->num_sensors; i++) {
>> +            if (!(priv->sensors_to_en & (0x1 << i)))
>> +                continue;
>> +
>> +            int idx = i * 3;
>> +
>> +            regmap_field_write(priv->rf[SHIFT_0 + idx], 
>> V2_SHIFT_DEFAULT);
>> +            regmap_field_write(priv->rf[SLOPE_0 + idx], 
>> V2_SLOPE_DEFAULT);
>> +            regmap_field_write(priv->rf[CZERO_0 + idx], 
>> V2_CZERO_DEFAULT);
>> +        }
>> +    }
>
> This code iterates over the sensors field several times. Please 
> consider extracting a function that handles all setup for a single 
> sensor, then calling it in a loop (I should probably do the same for 
> tsens-v0/v1 too).
Sure. After reading the mode0, base0 and base1 from QFPROM, we can call 
a function in a loop to setup the calibration for each sensor.
>
>> +
>> +    return 0;
>> +}
>> +
>>   static const struct tsens_ops ops_generic_v2 = {
>>       .init        = init_common,
>>       .get_temp    = get_temp_tsens_valid,
>> diff --git a/drivers/thermal/qcom/tsens.c b/drivers/thermal/qcom/tsens.c
>> index 98c356acfe98..169690355dad 100644
>> --- a/drivers/thermal/qcom/tsens.c
>> +++ b/drivers/thermal/qcom/tsens.c
>> @@ -974,7 +974,7 @@ int __init init_common(struct tsens_priv *priv)
>>       ret = regmap_field_read(priv->rf[TSENS_EN], &enabled);
>>       if (ret)
>>           goto err_put_device;
>> -    if (!enabled) {
>> +    if (!enabled && !priv->sensors_to_en) {
>>           dev_err(dev, "%s: device not enabled\n", __func__);
>>           ret = -ENODEV;
>>           goto err_put_device;
>> @@ -1006,6 +1006,40 @@ int __init init_common(struct tsens_priv *priv)
>>           goto err_put_device;
>>       }
>>   +    /* Do TSENS initialization if required */
>> +    if (priv->sensors_to_en) {
>
> Maybe it would be better to explicitly add VER_2_X_NO_RPM and check it 
> here?
Sure, will add a separate version macro.
>
>> +        priv->rf[CODE_OR_TEMP] = devm_regmap_field_alloc(dev, 
>> priv->srot_map,
>> + priv->fields[CODE_OR_TEMP]);
>> +        if (IS_ERR(priv->rf[CODE_OR_TEMP])) {
>> +            ret = PTR_ERR(priv->rf[CODE_OR_TEMP]);
>> +            goto err_put_device;
>> +        }
>> +
>> +        priv->rf[MAIN_MEASURE_PERIOD] =
>> +            devm_regmap_field_alloc(dev, priv->srot_map,
>> +                        priv->fields[MAIN_MEASURE_PERIOD]);
>> +        if (IS_ERR(priv->rf[MAIN_MEASURE_PERIOD])) {
>> +            ret = PTR_ERR(priv->rf[MAIN_MEASURE_PERIOD]);
>> +            goto err_put_device;
>> +        }
>> +
>> +        regmap_field_write(priv->rf[TSENS_SW_RST], 0x1);
>> +
>> +        /* Update measure period to 2ms */
>> +        regmap_field_write(priv->rf[MAIN_MEASURE_PERIOD], 0x1);
>> +
>> +        /* Enable available sensors */
>> +        regmap_field_write(priv->rf[SENSOR_EN], priv->sensors_to_en);
>> +
>> +        /* Real temperature format */
>> +        regmap_field_write(priv->rf[CODE_OR_TEMP], 0x1);
>> +
>> +        regmap_field_write(priv->rf[TSENS_SW_RST], 0x0);
>> +
>> +        /* Enable TSENS */
>> +        regmap_field_write(priv->rf[TSENS_EN], 0x1);
>> +    }
>> +
>>       /* This loop might need changes if enum regfield_ids is 
>> reordered */
>>       for (j = LAST_TEMP_0; j <= UP_THRESH_15; j += 16) {
>>           for (i = 0; i < priv->feat->max_sensors; i++) {
>> @@ -1282,6 +1316,7 @@ static int tsens_probe(struct platform_device 
>> *pdev)
>>         priv->dev = dev;
>>       priv->num_sensors = num_sensors;
>> +    priv->sensors_to_en = data->sensors_to_en;
>>       priv->ops = data->ops;
>>       for (i = 0;  i < priv->num_sensors; i++) {
>>           if (data->hw_ids)
>> diff --git a/drivers/thermal/qcom/tsens.h b/drivers/thermal/qcom/tsens.h
>> index 2805de1c6827..f8897bc8944e 100644
>> --- a/drivers/thermal/qcom/tsens.h
>> +++ b/drivers/thermal/qcom/tsens.h
>> @@ -168,6 +168,58 @@ enum regfield_ids {
>>       TSENS_SW_RST,
>>       SENSOR_EN,
>>       CODE_OR_TEMP,
>> +    /* MEASURE_PERIOD */
>> +    MAIN_MEASURE_PERIOD,
>> +
>> +    /* Sn_CONVERSION */
>> +    SHIFT_0,
>> +    SLOPE_0,
>> +    CZERO_0,
>> +    SHIFT_1,
>> +    SLOPE_1,
>> +    CZERO_1,
>> +    SHIFT_2,
>> +    SLOPE_2,
>> +    CZERO_2,
>> +    SHIFT_3,
>> +    SLOPE_3,
>> +    CZERO_3,
>> +    SHIFT_4,
>> +    SLOPE_4,
>> +    CZERO_4,
>> +    SHIFT_5,
>> +    SLOPE_5,
>> +    CZERO_5,
>> +    SHIFT_6,
>> +    SLOPE_6,
>> +    CZERO_6,
>> +    SHIFT_7,
>> +    SLOPE_7,
>> +    CZERO_7,
>> +    SHIFT_8,
>> +    SLOPE_8,
>> +    CZERO_8,
>> +    SHIFT_9,
>> +    SLOPE_9,
>> +    CZERO_9,
>> +    SHIFT_10,
>> +    SLOPE_10,
>> +    CZERO_10,
>> +    SHIFT_11,
>> +    SLOPE_11,
>> +    CZERO_11,
>> +    SHIFT_12,
>> +    SLOPE_12,
>> +    CZERO_12,
>> +    SHIFT_13,
>> +    SLOPE_13,
>> +    CZERO_13,
>> +    SHIFT_14,
>> +    SLOPE_14,
>> +    CZERO_14,
>> +    SHIFT_15,
>> +    SLOPE_15,
>> +    CZERO_15,
>>         /* ----- TM ------ */
>>       /* TRDY */
>> @@ -524,6 +576,7 @@ struct tsens_features {
>>   /**
>>    * struct tsens_plat_data - tsens compile-time platform data
>>    * @num_sensors: Number of sensors supported by platform
>> + * @sensors_to_en: Sensors to be enabled. Each bit represent a sensor
>>    * @ops: operations the tsens instance supports
>>    * @hw_ids: Subset of sensors ids supported by platform, if not the 
>> first n
>>    * @feat: features of the IP
>> @@ -531,6 +584,7 @@ struct tsens_features {
>>    */
>>   struct tsens_plat_data {
>>       const u32        num_sensors;
>> +    const u16        sensors_to_en;
>
> There is already a similar field, hw_ids. Can it be used instead?
Yes, it can be used. I missed to check this hw_ids. Will change the 
num_sensors to 5 and use the hw_ids.
>
>>       const struct tsens_ops    *ops;
>>       unsigned int        *hw_ids;
>>       struct tsens_features    *feat;
>> @@ -551,6 +605,7 @@ struct tsens_context {
>>    * struct tsens_priv - private data for each instance of the tsens IP
>>    * @dev: pointer to struct device
>>    * @num_sensors: number of sensors enabled on this device
>> + * @sensors_to_en: sensors to be enabled. Each bit represents a sensor
>>    * @tm_map: pointer to TM register address space
>>    * @srot_map: pointer to SROT register address space
>>    * @tm_offset: deal with old device trees that don't address TM and 
>> SROT
>> @@ -569,6 +624,7 @@ struct tsens_context {
>>   struct tsens_priv {
>>       struct device            *dev;
>>       u32                num_sensors;
>> +    u16                sensors_to_en;
>>       struct regmap            *tm_map;
>>       struct regmap            *srot_map;
>>       u32                tm_offset;
>
--
Thanks,
Praveenkumar