On 2/9/22 07:45, Agathe Porte wrote:
Hi Guenter,
Le 09/02/2022 à 15:46, Guenter Roeck a écrit :
On 2/9/22 05:36, Agathe Porte wrote:
Add support for Texas Instruments TMP464 temperature sensor IC.
TI's TMP464 is an I2C temperature sensor chip. This chip is
similar to TI's TMP421 chip, but with 16bit-wide registers (instead
of 8bit-wide registers). The chip have one local sensor and four
remote sensors.
Signed-off-by: Agathe Porte <agathe.porte@xxxxxxxxx>
Acked-by: Krzysztof Adamski <krzysztof.adamski@xxxxxxxxx>
---
Documentation/hwmon/tmp464.rst | 43 +++
MAINTAINERS | 7 +
drivers/hwmon/Kconfig | 10 +
drivers/hwmon/Makefile | 1 +
drivers/hwmon/tmp464.c | 472 +++++++++++++++++++++++++++++++++
5 files changed, 533 insertions(+)
create mode 100644 Documentation/hwmon/tmp464.rst
create mode 100644 drivers/hwmon/tmp464.c
diff --git a/Documentation/hwmon/tmp464.rst b/Documentation/hwmon/tmp464.rst
Needs to be added to Documentation/hwmon/index.rst.
ACK.
new file mode 100644
index 000000000000..8990554194de
--- /dev/null
+++ b/Documentation/hwmon/tmp464.rst
@@ -0,0 +1,43 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+Kernel driver tmp421
+====================
+
+Supported chips:
+
+ * Texas Instruments TMP464
+
+ Prefix: 'tmp464'
+
+ Addresses scanned: I2C 0x48, 0x49, 0x4a and 0x4b
+
+ Datasheet: http://focus.ti.com/docs/prod/folders/print/tmp464.html
+
+Authors:
+
+ Agathe Porte <agathe.porte@xxxxxxxxx>
+
+Description
+-----------
+
+This driver implements support for Texas Instruments TMP464
+temperature sensor chip. This chip implement one local four remote
+sensors. Temperature is measured in degrees Celsius. The chips are
+wired over I2C/SMBus and specified over a temperature range of -40 to
++125 degrees Celsius. Resolution for both the local and remote
+channels is 0.0625 degree C.
+
+The chips support only temperature measurement. The driver exports
+the temperature values via the following sysfs files:
+
+**temp[1-5]_input**
+
+Each sensor can be individually disabled via Devicetree or from sysfs
+via:
+
+**temp[1-5]_enable**
+
+If labels were specified in Devicetree, additional sysfs files will
+be present:
+
+**temp[1-5]_label**
diff --git a/MAINTAINERS b/MAINTAINERS
index 3e461db9cd91..fa0b27a8fe28 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -19457,6 +19457,13 @@ S: Maintained
F: Documentation/hwmon/tmp401.rst
F: drivers/hwmon/tmp401.c
+TMP464 HARDWARE MONITOR DRIVER
+M: Agathe Porte <agathe.porte@xxxxxxxxx>
+L: linux-hwmon@xxxxxxxxxxxxxxx
+S: Maintained
+F: Documentation/hwmon/tmp464.rst
+F: drivers/hwmon/tmp464.c
+
TMP513 HARDWARE MONITOR DRIVER
M: Eric Tremblay <etremblay@xxxxxxxxxxxxxxxxxxxx>
L: linux-hwmon@xxxxxxxxxxxxxxx
diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig
index 8df25f1079ba..52b4f5688b45 100644
--- a/drivers/hwmon/Kconfig
+++ b/drivers/hwmon/Kconfig
@@ -1979,6 +1979,16 @@ config SENSORS_TMP421
This driver can also be built as a module. If so, the module
will be called tmp421.
+config SENSORS_TMP464
+ tristate "Texas Instruments TMP464 and compatible"
+ depends on I2C
+ help
+ If you say yes here you get support for Texas Instruments TMP464
+ temperature sensor chip.
+
+ This driver can also be built as a module. If so, the module
+ will be called tmp464.
+
config SENSORS_TMP513
tristate "Texas Instruments TMP513 and compatibles"
depends on I2C
diff --git a/drivers/hwmon/Makefile b/drivers/hwmon/Makefile
index 185f946d698b..a1f2d6686227 100644
--- a/drivers/hwmon/Makefile
+++ b/drivers/hwmon/Makefile
@@ -194,6 +194,7 @@ obj-$(CONFIG_SENSORS_TMP103) += tmp103.o
obj-$(CONFIG_SENSORS_TMP108) += tmp108.o
obj-$(CONFIG_SENSORS_TMP401) += tmp401.o
obj-$(CONFIG_SENSORS_TMP421) += tmp421.o
+obj-$(CONFIG_SENSORS_TMP464) += tmp464.o
obj-$(CONFIG_SENSORS_TMP513) += tmp513.o
obj-$(CONFIG_SENSORS_VEXPRESS) += vexpress-hwmon.o
obj-$(CONFIG_SENSORS_VIA_CPUTEMP)+= via-cputemp.o
diff --git a/drivers/hwmon/tmp464.c b/drivers/hwmon/tmp464.c
new file mode 100644
index 000000000000..f294e12b1e39
--- /dev/null
+++ b/drivers/hwmon/tmp464.c
@@ -0,0 +1,472 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+/* Driver for the Texas Instruments TMP464 SMBus temperature sensor IC.
+ * Supported models: TMP464
+
+ * Copyright (C) 2022 Agathe Porte <agathe.porte@xxxxxxxxx>
+ * Preliminary support by:
+ * Lionel Pouliquen <lionel.lp.pouliquen@xxxxxxxxx>
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/jiffies.h>
+#include <linux/i2c.h>
+#include <linux/hwmon.h>
+#include <linux/hwmon-sysfs.h>
Unnecessary include.
ACK.
+#include <linux/err.h>
+#include <linux/mutex.h>
+#include <linux/of_device.h>
+#include <linux/sysfs.h>
Is this include necessary ?
Removed.
Include files in alphabetic order, please.
ACK.
+
+/* Addresses to scan */
+static const unsigned short normal_i2c[] = { 0x48, 0x49, 0x4a, 0x4b,
+ I2C_CLIENT_END };
Unnecessary continuation line.
ACK.
+
+enum chips { tmp464 };
+
+#define MAX_CHANNELS 5 /* chan 0 is internal, 1-4 are remote */
+
+/* TMP464 registers */
+#define TMP464_REG_THERM_STATUS 0x21
+#define TMP464_REG_THERM2_STATUS 0x22
+
+static const u8 TMP464_TEMP[MAX_CHANNELS] = { 0x00, 0x01, 0x02, 0x03, 0x04 };
+static const u8 TMP464_THERM_LIMIT[MAX_CHANNELS] = { 0x39, 0x42, 0x4A, 0x52, 0x5A };
+static const u8 TMP464_THERM2_LIMIT[MAX_CHANNELS] = { 0x3A, 0x43, 0x4B, 0x53, 0x5B };
+static const u8 TMP464_OFFSET_REMOTE[MAX_CHANNELS] = { 0x40, 0x48, 0x50, 0x58 };
Most of those are unused (which would result in static analyzer warnings).
More on that below.
+#define TMP464_N_FACTOR_REG_1 0x41
+#define TMP464_CONFIG_REG 0x30
+
+/* alarm offset in THERM_STATUS/THERM2_STATUS register */
+#define TMP464_ALARM_BASE_NUMBER 7
+
+#define TMP464_REG_THERM_HYSTERESIS 0x38
+
+#define TMP464_MANUFACTURER_ID_REG 0xFE
+#define TMP464_DEVICE_ID_REG 0xFF
+
+/* Flags */
+#define TMP464_CONFIG_SHUTDOWN BIT(5)
+#define TMP464_CONFIG_RANGE 0x04
+#define TMP464_CONFIG_REG_REN(x) (BIT(7 + (x)))
+#define TMP464_CONFIG_REG_REN_MASK GENMASK(11, 7)
+#define TMP464_CONFIG_CONVERSION_RATE_B0 2
+#define TMP464_CONFIG_CONVERSION_RATE_MASK GENMASK(TMP464_CONFIG_CONVERSION_RATE_B0, \
+ TMP464_CONFIG_CONVERSION_RATE_B0 + 2)
+
+/* Manufacturer / Device ID's */
+#define TMP464_MANUFACTURER_ID 0x5449
+#define TMP464_DEVICE_ID 0x1468
+
+
No double empty lines, please.
ACK.
+static const struct i2c_device_id tmp464_id[] = {
+ { "tmp464", 0 },
+ { }
+};
+MODULE_DEVICE_TABLE(i2c, tmp464_id);
+
+static const struct of_device_id __maybe_unused tmp464_of_match[] = {
+ {
+ .compatible = "ti,tmp464",
+ .data = (void *)5
+ },
+ {},
+};
+MODULE_DEVICE_TABLE(of, tmp464_of_match);
+
+
No double empty lines, please. checkpatch --strict should tell you that. Please
run and fix what it reports.
ACK. I only ran checkpatch.pl without the --strict version before. I will fix all --strict warnings in v2 patchset
+struct tmp464_channel {
+ const char *label;
+ bool enabled;
+ s16 temp;
+};
+
+struct tmp464_data {
+ struct i2c_client *client;
+ struct mutex update_lock;
+ u32 temp_config[MAX_CHANNELS + 1];
+ struct hwmon_channel_info temp_info;
+ const struct hwmon_channel_info *info[2];
+ struct hwmon_chip_info chip;
+ bool valid;
+ unsigned long last_updated;
+ unsigned long channels;
+ u16 config;
+ struct tmp464_channel channel[MAX_CHANNELS];
+};
+
+
+/* funcs */
+
Useless comment. Please drop.
ACK.
+static int temp_from_raw(u16 reg)
+{
+ int temp = 0;
+
+ temp = (reg >> 8) * 2000;
+ temp += ((reg & 0xFF) >> 3) * 625 / 10;
+
This doesn't correctly handle negative temperatures. reg needs to be s16
to be able to handle those. Also, why not just
return (reg >> 3) * 625 / 10;
or
return DIV_ROUND_CLOSEST((reg >> 3) * 625, 10);
?
ACK. I did not think about simplifying the math. Next patchset will use DIV_ROUND_CLOSEST and s16.
+ return temp;
+}
+
+static int tmp464_update_device(struct tmp464_data *data)
+{
+ struct i2c_client *client = data->client;
+ int ret = 0;
+ int i;
+
+ mutex_lock(&data->update_lock);
+
+ if (time_after(jiffies, data->last_updated + (HZ / 2)) ||
+ !data->valid) {
+ ret = i2c_smbus_read_word_swapped(client, TMP464_CONFIG_REG);
+ if (ret < 0)
+ goto exit;
+ data->config = ret;
+
+ for (i = 0; i < data->channels; i++) {
+ ret = i2c_smbus_read_word_swapped(client, TMP464_TEMP[i]);
+ if (ret < 0)
+ goto exit;
+ data->channel[i].temp = ret;
+ }
+ data->last_updated = jiffies;
+ data->valid = true;
+ }
+
+exit:
+ mutex_unlock(&data->update_lock);
+
+ if (ret < 0) {
+ data->valid = false;
+ return ret;
+ }
+
+ return 0;
+}
+
We are moving away from local caching. Please consider using regmap instead.
Seems like I was wrong to base my work on the tmp421.c driver. Do you have a driver in mind in hwmon that is state-of-the-art and uses regmap so that I can inspire myself? Will propose v2 patchset without regmap first.
Any driver using regmap would be fine. A couple of driver conversions
to use regmap are queued in linux-next, and there are a few earlier
conversions as well.
495da5954e15 hwmon: (adt7x10) Convert to use regmap
a166d8e6f5b7 hwmon: (lm83) Use regmap
50152fb6c1a1 hwmon: (tmp401) Use regmap
ef67959c4253 hwmon: (adt7470) Convert to use regmap
df885d912f67 hwmon: (adm9240) Convert to regmap
+static int tmp464_enable_channels(struct tmp464_data *data)
+{
+ int err;
+ struct i2c_client *client = data->client;
+ struct device *dev = &client->dev;
+ int old = i2c_smbus_read_word_swapped(client, TMP464_CONFIG_REG);
+ int new, i;
+
+ if (old < 0) {
+ dev_err(dev, "error reading register, can't disable channels\n");
+ return old;
+ }
+
+ new = old & ~TMP464_CONFIG_REG_REN_MASK;
+ for (i = 0; i < data->channels; i++)
+ if (data->channel[i].enabled)
+ new |= TMP464_CONFIG_REG_REN(i);
+
+ if (new == old)
+ return 0;
+
+ err = i2c_smbus_write_word_data(client, TMP464_CONFIG_REG, new);
+ if (err < 0)
+ dev_err(dev, "error writing register, can't disable channels\n");
+
+ return err;
+}
+
+static int tmp464_read(struct device *dev, enum hwmon_sensor_types type,
+ u32 attr, int channel, long *val)
+{
+ struct tmp464_data *tmp464 = dev_get_drvdata(dev);
+ int ret = 0;
+
+ ret = tmp464_update_device(tmp464);
+ if (ret)
+ return ret;
+
+ switch (attr) {
+ case hwmon_temp_input:
+ if (!tmp464->channel[channel].enabled)
+ return -ENODATA;
+ *val = temp_from_raw(tmp464->channel[channel].temp);
+ return 0;
+ case hwmon_temp_enable:
+ *val = tmp464->channel[channel].enabled;
+ return 0;
+ default:
+ return -EOPNOTSUPP;
+ }
+
+}
+
+static int tmp464_read_string(struct device *dev, enum hwmon_sensor_types type,
+ u32 attr, int channel, const char **str)
+{
+ struct tmp464_data *data = dev_get_drvdata(dev);
+
+ *str = data->channel[channel].label;
+
+ return 0;
+}
+
+static int tmp464_write(struct device *dev, enum hwmon_sensor_types type,
+ u32 attr, int channel, long val)
+{
+ struct tmp464_data *data = dev_get_drvdata(dev);
+ int ret;
+
+ switch (attr) {
+ case hwmon_temp_enable:
+ data->channel[channel].enabled = val;
+ ret = tmp464_enable_channels(data);
+ break;
+ default:
+ ret = -EOPNOTSUPP;
+ }
+
+ return ret;
+}
+
+static umode_t tmp464_is_visible(const void *data, enum hwmon_sensor_types type,
+ u32 attr, int channel)
+{
+ switch (attr) {
+ case hwmon_temp_input:
+ return 0444;
+ case hwmon_temp_label:
+ return 0444;
+ case hwmon_temp_enable:
+ return 0644;
+ default:
+ return 0;
+ }
+}
+
+static int tmp464_init_client(struct tmp464_data *data)
+{
+ int config, config_orig;
+ struct i2c_client *client = data->client;
+
+ /* Set the conversion rate to 2 Hz */
+ config = i2c_smbus_read_word_swapped(client, TMP464_CONFIG_REG);
+ if (config < 0) {
+ dev_err(&client->dev,
+ "Could not read configuration register (%d)\n", config);
+ return config;
+ }
+
+ config_orig = config;
+ config &= ~TMP464_CONFIG_CONVERSION_RATE_MASK;
+ config |= (0x05) << TMP464_CONFIG_CONVERSION_RATE_B0;
+
+ if (config != config_orig) {
+ dev_info(&client->dev, "Set conversion rate to 2 Hz\n");
+ i2c_smbus_write_word_data(client, TMP464_CONFIG_REG, config);
+ }
+
+ /* Start conversions (disable shutdown if necessary) */
+ config = i2c_smbus_read_word_swapped(client, TMP464_CONFIG_REG);
Why read this register twice ?
+ if (config < 0) {
+ dev_err(&client->dev,
+ "Could not read configuration register (%d)\n", config);
+ return config;
+ }
+
+ config_orig = config;
+ config &= ~TMP464_CONFIG_SHUTDOWN;
+
+ if (config != config_orig) {
+ dev_info(&client->dev, "Enable monitoring chip\n");
Please only write the configuration register once, and drop the messages.
ACK. This was inspired from the tmp421.c driver, but there should be no reason to read twice.
+ i2c_smbus_write_word_data(client, TMP464_CONFIG_REG, config);
+ }
+
+ return tmp464_enable_channels(data);
+}
+
+static int tmp464_detect(struct i2c_client *client,
+ struct i2c_board_info *info)
+{
+ enum chips kind;
+ struct i2c_adapter *adapter = client->adapter;
+ static const char * const names[] = {
+ "TMP464"
+ };
+ u8 reg;
Should be int to enable error checks.
ACK.
+
+ if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_WORD_DATA))
+ return -ENODEV;
+
+ reg = i2c_smbus_read_word_swapped(client, TMP464_MANUFACTURER_ID_REG);
+ if (reg != TMP464_MANUFACTURER_ID)
+ return -ENODEV;
+
+ /* Check for "always return zero" bits */
+ reg = i2c_smbus_read_word_swapped(client, TMP464_REG_THERM_STATUS);
+ if (reg & 0x1f)
should also check for < 0.
ACK.
+ return -ENODEV;
+ reg = i2c_smbus_read_word_swapped(client, TMP464_REG_THERM2_STATUS);
+ if (reg & 0x1f)
+ return -ENODEV;
Should also check for < 0.
ACK.
+
+ reg = i2c_smbus_read_word_swapped(client, TMP464_DEVICE_ID_REG);
+ if (reg < 0)
+ return reg;
That won't happen with an u8 variable.
ACK.
+ switch (reg) {
+ case TMP464_DEVICE_ID:
+ kind = tmp464;
+ break;
Are there more devices with the same register set ? If not it is pointless to
have/prepare support for it.
The tmp468 is the same device with 1+8 channels (tmp464 is 1+4 channels). Adding TMP468 support would be fairly easy with the current structure, but we do not have the hardware to validate it unfortunately.
Evaluation board for TI468 is $75 over at TI. Unfortunately it looks
like TMP464 is sold out everywhere, or I'd order a few.
+ default:
+ return -ENODEV;
+ }
+
+ strscpy(info->type, tmp464_id[kind].name, I2C_NAME_SIZE);
+ dev_info(&adapter->dev, "Detected TI %s chip at 0x%02x\n",
+ names[kind], client->addr);
+
+ return 0;
+}
+
+static int tmp464_probe_child_from_dt(struct i2c_client *client,
+ struct device_node *child,
+ struct tmp464_data *data)
+
+{
+ struct device *dev = &client->dev;
+ u32 i;
+ s32 val;
+ int err;
+
+ err = of_property_read_u32(child, "reg", &i);
+ if (err) {
+ dev_err(dev, "missing reg property of %pOFn\n", child);
+ return err;
+ }
+
+ if (i >= data->channels) {
+ dev_err(dev, "invalid reg %d of %pOFn\n", i, child);
+ return -EINVAL;
+ }
+
+ of_property_read_string(child, "label", &data->channel[i].label);
+ if (data->channel[i].label)
+ data->temp_config[i] |= HWMON_T_LABEL;
+
+ data->channel[i].enabled = of_device_is_available(child);
+
+ err = of_property_read_s32(child, "ti,n-factor", &val);
+ if (!err) {
+ if (i == 0) {
+ dev_err(dev, "n-factor can't be set for internal channel\n");
+ return -EINVAL;
+ }
+
+ if (val > 127 || val < -128) {
+ dev_err(dev, "n-factor for channel %d invalid (%d)\n",
+ i, val);
+ return -EINVAL;
+ }
+ i2c_smbus_write_word_data(client, TMP464_N_FACTOR_REG_1 + i - 1,
+ val);
needs error check
ACK.
+ }
+
+ return 0;
+}
+
+static int tmp464_probe_from_dt(struct i2c_client *client, struct tmp464_data *data)
+{
+ struct device *dev = &client->dev;
+ const struct device_node *np = dev->of_node;
+ struct device_node *child;
+ int err;
+
+ for_each_child_of_node(np, child) {
+ if (strcmp(child->name, "channel"))
+ continue;
+
+ err = tmp464_probe_child_from_dt(client, child, data);
+ if (err) {
+ of_node_put(child);
+ return err;
+ }
+ }
+
+ return 0;
+}
+
+static const struct hwmon_ops tmp464_ops = {
+ .is_visible = tmp464_is_visible,
+ .read = tmp464_read,
+ .read_string = tmp464_read_string,
+ .write = tmp464_write,
+};
+
+static int tmp464_probe(struct i2c_client *client)
+{
+ struct device *dev = &client->dev;
+ struct device *hwmon_dev;
+ struct tmp464_data *data;
+ int i, err;
+
+ if (!i2c_check_functionality(client->adapter,
+ I2C_FUNC_SMBUS_WORD_DATA)) {
+ dev_err(&client->dev, "i2c functionality check failed\n");
+ return -ENODEV;
+ }
+ data = devm_kzalloc(dev, sizeof(struct tmp464_data), GFP_KERNEL);
+ if (!data)
+ return -ENOMEM;
+
+ mutex_init(&data->update_lock);
+ if (client->dev.of_node)
+ data->channels = (unsigned long)
+ of_device_get_match_data(&client->dev);
Unnecessary continuation line.
ACK.
+ else
+ data->channels = i2c_match_id(tmp464_id, client)->driver_data;
+ data->client = client;
+
+ for (i = 0; i < data->channels; i++) {
+ data->temp_config[i] = HWMON_T_INPUT | HWMON_T_ENABLE;
Why not support limits, temperature offsets, alarms, hysteresis, and conversion
rate configuration ? It is kind of off to support n-factor correction but
nothing else.
Overall it looks like some of the support was dropped since defines for
limit and offset registers are there, and the driver seems to be prepared
to support more chips. Why ?
The history of this driver is that it was using older hwmon framework practise. During the port to the new framework for upstreaming, we found out that we did not have the use of {limits,offsets,alarms} sysfs entries, so these features were not ported. They can be added later, though. I will drop the #defines for the unused registers meanwhile.
Hmm, that sounds less than perfect to me, to say it mildly.
Even if you don't use those attributes, they should still be
supported. It is not as if providing that support is
super-complicated to add (or keep), and testing it is quite
straightforward.
Can you possibly send me a couple of samples of tmp464 ? I'd rather
deal with that myself now instead of having to review the code later
again. I already ordered an evaluation board for TMP468, so that
part is covered, but I'd want to test the code with a real TMP464 as well.
Thanks,
Guenter
+ data->channel[i].enabled = true;
+ }
+
+ err = tmp464_probe_from_dt(client, data);
+ if (err)
+ return err;
+
+ err = tmp464_init_client(data);
+ if (err)
+ return err;
+
+ data->chip.ops = &tmp464_ops;
+ data->chip.info = data->info;
+
+ data->info[0] = &data->temp_info;
+
+ data->temp_info.type = hwmon_temp;
+ data->temp_info.config = data->temp_config;
+
+ hwmon_dev = devm_hwmon_device_register_with_info(dev, client->name,
+ data,
+ &data->chip,
+ NULL);
Please drop some of those continuation lines. We can go up to 100 columns and
should do so in situations like this.
ACK.
+
+ return PTR_ERR_OR_ZERO(hwmon_dev);
+}
+
+static struct i2c_driver tmp464_driver = {
+ .class = I2C_CLASS_HWMON,
+ .driver = {
+ .name = "tmp464",
+ .of_match_table = of_match_ptr(tmp464_of_match),
+ },
+ .probe_new = tmp464_probe,
+ .id_table = tmp464_id,
+ .detect = tmp464_detect,
+ .address_list = normal_i2c,
+};
+
+module_i2c_driver(tmp464_driver);
+
+MODULE_AUTHOR("Agathe Porte <agathe.porte@xxxxxxxxx>");
+MODULE_DESCRIPTION("Texas Instruments TMP464 temperature sensor driver");
+MODULE_LICENSE("GPL");
