Dear Guenter,
Thank you for your comments,
Guenter Roeck <linux@xxxxxxxxxxxx> 於 2024年12月10日 週二 下午11:58寫道：
>
> On Tue, Dec 10, 2024 at 06:45:23PM +0800, Ming Yu wrote:
> > This driver supports Hardware monitor functionality for NCT6694 MFD
> > device based on USB interface.
> >
> > Signed-off-by: Ming Yu <tmyu0@xxxxxxxxxxx>
>
> checkpatch:
>
> WARNING: From:/Signed-off-by: email address mismatch: 'From: Ming Yu <a0282524688@xxxxxxxxx>' != 'Signed-off-by: Ming Yu <tmyu0@xxxxxxxxxxx>'
>
Fix it in v4.
...
> > +static inline long in_from_reg(u8 reg)
> > +{
> > + return reg * 16;
> > +}
> > +
> > +static inline u8 in_to_reg(long val)
> > +{
> > + if (val <= 0)
> > + return 0;
>
> This is pointless since the calling code already clamps the value to [0, 2032].
>
Drop it in v4.
> > + return val / 16;
>
> DIV_ROUND_CLOSEST() ?
>
Fix it in v4.
> > +}
> > +
> > +static inline long temp_from_reg(u8 reg)
> > +{
> > + return reg * 1000;
>
> This always returns a positive value, even though the temperature is
> supposed to be signed. More on that below.
>
I will modify the return value to (sign_extend32(reg, 7) * 1000) in v4.
> > +}
> > +
> > +static inline u8 temp_to_reg(long val)
> > +{
> > + return val / 1000;
>
> DIV_ROUND_CLOSEST() ?
>
Fix it in v4.
> > +}
...
> > +static int nct6694_in_read(struct device *dev, u32 attr, int channel,
> > + long *val)
> > +{
> > + struct nct6694_hwmon_data *data = dev_get_drvdata(dev);
> > + unsigned char vin_en;
> > + int ret;
> > +
> > + guard(mutex)(&data->lock);
> > +
> > + switch (attr) {
> > + case hwmon_in_enable:
> > + vin_en = data->hwmon_en[NCT6694_VIN_EN(channel / 8)];
> > + *val = !!(vin_en & BIT(channel % 8)) ? 1 : 0;
>
> Drop "? 1 : 0"
>
Drop it in v4.
> > +
> > + return 0;
> > + case hwmon_in_input:
> > + ret = nct6694_read_msg(data->nct6694, NCT6694_RPT_MOD,
> > + NCT6694_VIN_IDX(channel), 1,
> > + data->xmit_buf);
>
> I am curious: Since the received data length is not returned, and the
> expected minimum length is not passed to nct6694_read_msg(), the driver
> has no means to check if the received data actually includes the expected
> values at the expected index. That doesn't matter here, but some of the
> returned data is located far into the buffer. Is it indeed not necessary
> for the driver to check if the received data is actually as long as
> expected ?
>
The fourth parameter is expected length of nct6694_read_msg(), which
in this case is 1.
I will add the data length validation code for handling the
response_header in both nct6694_read_msg() and nct6694_write_msg() in
the next patch.
> > + if (ret)
> > + return ret;
...
> > +static int nct6694_temp_read(struct device *dev, u32 attr, int channel,
> > + long *val)
> > +{
> > + struct nct6694_hwmon_data *data = dev_get_drvdata(dev);
> > + unsigned char temp_en, temp_hyst;
> > + int ret, int_part, frac_part;
> > + signed char temp_max;
> > +
> > + guard(mutex)(&data->lock);
> > +
> > + switch (attr) {
> > + case hwmon_temp_enable:
> > + temp_en = data->hwmon_en[NCT6694_TIN_EN(channel / 8)];
> > + *val = !!(temp_en & BIT(channel % 8)) ? 1 : 0;
>
> Drop "? 1 : 0"
>
Drop it in v4.
> > +
> > + return 0;
> > + case hwmon_temp_input:
> > + ret = nct6694_read_msg(data->nct6694, NCT6694_RPT_MOD,
> > + NCT6694_TIN_IDX(channel), 2,
> > + data->xmit_buf);
> > + if (ret)
> > + return ret;
> > +
> > + int_part = sign_extend32(data->xmit_buf[0], 7);
> > + frac_part = FIELD_GET(NCT6694_LSB_REG_MASK, data->xmit_buf[1]);
> > + if (int_part < 0)
> > + *val = (int_part + 1) * 1000 - (8 - frac_part) * 125;
> > + else
> > + *val = int_part * 1000 + frac_part * 125;
> > +
> > + return 0;
> > + case hwmon_temp_max:
> > + ret = nct6694_read_msg(data->nct6694, NCT6694_HWMON_MOD,
> > + NCT6694_HWMON_CMD2_OFFSET,
> > + NCT6694_HWMON_CMD2_LEN,
> > + data->xmit_buf);
> > + if (ret)
> > + return ret;
> > +
> > + *val = temp_from_reg(data->xmit_buf[NCT6694_TIN_HL(channel)]);
> > +
>
> If the value in NCT6694_TIN_HL() is signed, this will return a large positive
> value instead.
>
Understood! I will fix it in v4.
> > + return 0;
> > + case hwmon_temp_max_hyst:
> > + ret = nct6694_read_msg(data->nct6694, NCT6694_HWMON_MOD,
> > + NCT6694_HWMON_CMD2_OFFSET,
> > + NCT6694_HWMON_CMD2_LEN,
> > + data->xmit_buf);
> > + if (ret)
> > + return ret;
> > +
> > + temp_max = (signed char)data->xmit_buf[NCT6694_TIN_HL(channel)];
> > + temp_hyst = FIELD_GET(NCT6694_TIN_HYST_MASK,
> > + data->xmit_buf[NCT6694_TIN_HYST(channel)]);
> > + if (temp_max < 0)
>
> This suggests that the temperature limit is signed, suggesting in turn that
> temp_from_reg() is wrong.
>
Understood! I will fix it in v4.
> > + *val = temp_from_reg(temp_max + temp_hyst);
> > + else
> > + *val = temp_from_reg(temp_max - temp_hyst);
> > +
> > + return 0;
> > + case hwmon_temp_max_alarm:
> > + ret = nct6694_read_msg(data->nct6694, NCT6694_RPT_MOD,
> > + NCT6694_TIN_STS(channel / 8), 1,
> > + data->xmit_buf);
> > + if (ret)
> > + return ret;
> > +
> > + *val = !!(data->xmit_buf[0] & BIT(channel % 8));
> > +
> > + return 0;
> > + default:
> > + return -EOPNOTSUPP;
> > + }
> > +}
> > +
> > +static int nct6694_fan_read(struct device *dev, u32 attr, int channel,
> > + long *val)
> > +{
> > + struct nct6694_hwmon_data *data = dev_get_drvdata(dev);
> > + unsigned char fanin_en;
> > + int ret;
> > +
> > + guard(mutex)(&data->lock);
> > +
> > + switch (attr) {
> > + case hwmon_fan_enable:
> > + fanin_en = data->hwmon_en[NCT6694_FIN_EN(channel / 8)];
> > + *val = !!(fanin_en & BIT(channel % 8)) ? 1 : 0;
>
> Drop "? 1 : 0"
>
Drop it in v4.
> > +
> > + return 0;
> > + case hwmon_fan_input:
> > + ret = nct6694_read_msg(data->nct6694, NCT6694_RPT_MOD,
> > + NCT6694_FIN_IDX(channel), 2,
> > + data->xmit_buf);
> > + if (ret)
> > + return ret;
> > +
> > + *val = (data->xmit_buf[1] |
> > + (data->xmit_buf[0] << 8)) & 0xFFFF;
>
> The "& 0xffff" is pointless.
>
Fix it in v4.
> > +
> > + return 0;
> > + case hwmon_fan_min:
> > + ret = nct6694_read_msg(data->nct6694, NCT6694_HWMON_MOD,
> > + NCT6694_HWMON_CMD2_OFFSET,
> > + NCT6694_HWMON_CMD2_LEN,
> > + data->xmit_buf);
> > + if (ret)
> > + return ret;
> > +
> > + *val = (data->xmit_buf[NCT6694_FIN_LL(channel)] |
> > + data->xmit_buf[NCT6694_FIN_HL(channel)] << 8) & 0xFFFF;
>
> Same here.
>
Fix it in v4.
> > +
> > + return 0;
> > + case hwmon_fan_min_alarm:
> > + ret = nct6694_read_msg(data->nct6694, NCT6694_RPT_MOD,
> > + NCT6694_FIN_STS(channel / 8),
> > + 1, data->xmit_buf);
> > + if (ret)
> > + return ret;
> > +
> > + *val = !!(data->xmit_buf[0] & BIT(channel % 8));
> > +
> > + return 0;
> > + default:
> > + return -EOPNOTSUPP;
> > + }
> > +}
> > +
> > +static int nct6694_pwm_read(struct device *dev, u32 attr, int channel,
> > + long *val)
> > +{
> > + struct nct6694_hwmon_data *data = dev_get_drvdata(dev);
> > + unsigned char pwm_en;
> > + int ret;
> > +
> > + guard(mutex)(&data->lock);
> > +
> > + switch (attr) {
> > + case hwmon_pwm_enable:
> > + pwm_en = data->hwmon_en[NCT6694_PWM_EN(channel / 8)];
> > + *val = !!(pwm_en & BIT(channel % 8)) ? 1 : 0;
>
> Drop "? 1 : 0".
>
Drop it in v4.
> > +
...
> > + case hwmon_temp_max_hyst:
> > + ret = nct6694_read_msg(data->nct6694, NCT6694_HWMON_MOD,
> > + NCT6694_HWMON_CMD2_OFFSET,
> > + NCT6694_HWMON_CMD2_LEN,
> > + data->xmit_buf);
> > +
> > + val = clamp_val(val, -127000, 127000);
> > + temp_max = (signed char)data->xmit_buf[NCT6694_TIN_HL(channel)];
> > + temp_hyst = (temp_max < 0) ? (temp_max + val / 1000) :
> > + (temp_max - val / 1000);
>
> DIV_ROUND_CLOSEST() ?
>
Fix it in v4.
> > + temp_hyst = clamp_val(temp_hyst, 0, 7);
> > + data->xmit_buf[NCT6694_TIN_HYST(channel)] =
> > + (data->xmit_buf[NCT6694_TIN_HYST(channel)] & ~NCT6694_TIN_HYST_MASK) |
> > + FIELD_PREP(NCT6694_TIN_HYST_MASK, temp_hyst);
> > +
...
> > +
> > +static int nct6694_hwmon_init(struct nct6694_hwmon_data *data)
> > +{
> > + int ret;
> > +
> > + guard(mutex)(&data->lock);
> > +
>
> Not significant, but this is unnecessary because the function is only
> called once from probe.
>
Understood! I will drop it in the next patch.
> > + /*
...
> > +static int nct6694_hwmon_probe(struct platform_device *pdev)
> > +{
> > + struct nct6694_hwmon_data *data;
> > + struct nct6694 *nct6694 = dev_get_drvdata(pdev->dev.parent);
> > + struct device *hwmon_dev;
> > + int ret;
> > +
> > + data = devm_kzalloc(&pdev->dev, sizeof(*data), GFP_KERNEL);
> > + if (!data)
> > + return -ENOMEM;
> > +
> > + data->xmit_buf = devm_kcalloc(&pdev->dev, NCT6694_MAX_PACKET_SZ,
> > + sizeof(unsigned char), GFP_KERNEL);
>
> Wondering ... why not just devm_kzalloc(..,. NCT6694_MAX_PACKET_SZ, ...) ?
> sizeof(unsigned char) is always 1, after all.
>
Okay, I'll fix it.
> > + if (!data->xmit_buf)
> > + return -ENOMEM;
> > +
> > + data->nct6694 = nct6694;
> > + mutex_init(&data->lock);
> > + platform_set_drvdata(pdev, data);
>
> This is unnecessary unless I am missing something.
>
Drop it in v4.
> > +
> > + ret = nct6694_hwmon_init(data);
> > + if (ret)
> > + return ret;
> > +
> > + /* Register hwmon device to HWMON framework */
> > + hwmon_dev = devm_hwmon_device_register_with_info(&pdev->dev,
> > + "nct6694", data,
> > + &nct6694_chip_info,
> > + NULL);
> > + return PTR_ERR_OR_ZERO(hwmon_dev);
> > +}
> > +
> > +static struct platform_driver nct6694_hwmon_driver = {
> > + .driver = {
> > + .name = "nct6694-hwmon",
> > + },
> > + .probe = nct6694_hwmon_probe,
> > +};
> > +
> > +module_platform_driver(nct6694_hwmon_driver);
> > +
> > +MODULE_DESCRIPTION("USB-HWMON driver for NCT6694");
> > +MODULE_AUTHOR("Ming Yu <tmyu0@xxxxxxxxxxx>");
> > +MODULE_LICENSE("GPL");
> > +MODULE_ALIAS("platform:nct6694-hwmon");
> > --
> > 2.34.1
> >
> >
Best regards,
Ming
