[lm-sensors] w83627ehf fan controll?

[lm-sensors] w83627ehf fan controll?

[Sebastian Band]

Hello!

Is there any possible way, to controll the fan speed with the w83627ehf chip?
I'm able to monitor the fan and the temp without any problem, but I don't have any *pwm 
entries in /sys/bus/i2c/devices/9191-0290/.
I use linux-2.6.14 and lm_sensors version 2.9.2.

Any help would be appreciated.

Sebastian

[lm-sensors] w83627ehf fan controll?

[Rudolf Marek]

Hello,

Cadu any progress on this?
Thanks

Regards
Rudolf

[lm-sensors] w83627ehf fan controll?

[Ymu]

Hello Sebastian,

> Is there any possible way, to controll the fan speed with the
w83627ehf chip?
> I'm able to monitor the fan and the temp without any problem, but I
don't have any *pwm
> entries in /sys/bus/i2c/devices/9191-0290/.
Yes, there should be :) But as you see there are no pwm entries yet. Do
you have interest in using a test version?

Hi Jean,
I have add some entries to test the chip (in my local CVS), But I have
not finished the test. Did anyone do the same work on this driver?
Actually I'm quiet confusing about the naming rule of device file, it
did not cover all the PWM/SmartFan functions.
And how about to backport w83627ehf driver to lm_sensors2 , Or there's
no need to do it....

Best Regards
Yuan Mu


==============================================The privileged confidential information contained in this email is intended for use only by the addressees as indicated by the original sender of this email. If you are not the addressee indicated in this email or are not responsible for delivery of the email to such  a person, please kindly reply to the sender indicating this fact and delete all copies of it from your computer and network server immediately. Your cooperation is highly appreciated. It is advised that any unauthorized use of confidential information of Winbond is strictly prohibited; and any information in this email irrelevant to the official business of Winbond shall be deemed as neither given nor endorsed by Winbond.
==============================================If your computer is unable to decode Chinese font, please ignore the following message.It essentially repeats the statement in English given above.本信件內所含華邦電子的財產性機密性資訊, 僅授權原發信人指定之收信人取閱\之用. 假使您並非被指定之收信人或因任何原因在未經授權的情形之下收到本信件, 請您告知原發信人並立即將信件從電腦與網路伺服器中予以消除. 對於您的合作, 我們先此致謝. 特此提醒, 任何未經授權擅自使用華邦電子的機密資訊的行為是被嚴格禁止的. 信件與華邦電子營業無關之內容,不得視為華邦電子之立場或意見.

[lm-sensors] w83627ehf fan controll?

[Jean Delvare]

Hi Yuan & all,

> I have add some entries to test the chip (in my local CVS), But I have
> not finished the test. Did anyone do the same work on this driver?

A number of persons are supposedly working on improving this driver.
Cadu said he would improve the driver back in July but we didn't hear
back since. I think Rudolf has a W83627EHF chip and may be willing to
help too. And I also seem to remember that Aurelien has such a chip and
wanted to help with the driver some times ago.

It's rare to have that many persons working on the same driver, so it's
quite sad that nothing has been merged into the driver yet :(

If you have a patch to contribute, please do! :)

> Actually I'm quiet confusing about the naming rule of device file, it
> did not cover all the PWM/SmartFan functions.

Same old story. We tried to standardize the PWM interface, but in the
end nobody seems to be happy with that standard. I'm a bit tired of all
this, so just implement any interface you like as long as it makes
sense. All I am asking for is that fan stopped is PWM=0 and fan at full
speed is PWM%5.

> And how about to backport w83627ehf driver to lm_sensors2 , Or there's
> no need to do it....

Until a user asks for it, I consider there is no need. If several users
do ask for it and someone wants to work on backporting it, fine with
me. But this is a recent chip and the 2.4 kernel is aging, so I
wouldn't be surprised if we had no request for a backport, ever.

As far as I am concerned, I started working on the w83627ehf driver
long ago because I happened to have some spare time and someone needed
support for that chip, but as I do not have a chip to test this driver,
it's really better is someone else goes on working on the driver. And I
also don't have that much spare time these days, as most of you
probably have noticed already.

Thanks,
-- 
Jean Delvare

[lm-sensors] w83627ehf fan controll?

[Aurelien Jarno]

On Sat, Dec 31, 2005 at 08:12:16PM +0100, Jean Delvare wrote:
> Hi Yuan & all,
Hi!

> > I have add some entries to test the chip (in my local CVS), But I have
> > not finished the test. Did anyone do the same work on this driver?
> 
> A number of persons are supposedly working on improving this driver.
> Cadu said he would improve the driver back in July but we didn't hear
> back since. I think Rudolf has a W83627EHF chip and may be willing to
> help too. And I also seem to remember that Aurelien has such a chip and
> wanted to help with the driver some times ago.
> 
> It's rare to have that many persons working on the same driver, so it's
> quite sad that nothing has been merged into the driver yet :(
>
> If you have a patch to contribute, please do! :)
>

Yes I have started to work on it, but I have then stopped due to lack of
time. Please find attached a patch between the current version in the
kernel and my local version. It adds support for voltage inputs and it
factorizes some code.

> > Actually I'm quiet confusing about the naming rule of device file, it
> > did not cover all the PWM/SmartFan functions.
> 
> Same old story. We tried to standardize the PWM interface, but in the
> end nobody seems to be happy with that standard. I'm a bit tired of all
> this, so just implement any interface you like as long as it makes
> sense. All I am asking for is that fan stopped is PWM=0 and fan at full
> speed is PWM%5.
> 
I have also started to work on pwm, until I understood the Asus guys
have wired the fans directly to +12V. That's why I finally used a
hardware solution instead of a software one :(

Bye,
Aurelien

-- 
  .''`.  Aurelien Jarno	            | GPG: 1024D/F1BCDB73
 : :' :  Debian developer           | Electrical Engineer
 `. `'   aurel32 at debian.org         | aurelien at aurel32.net
   `-    people.debian.org/~aurel32 | www.aurel32.net
-------------- next part --------------
--- /usr/src/linux-2.6.14/drivers/hwmon/w83627ehf.c	2005-10-28 02:02:08.000000000 +0200
+++ w83627ehf.c	2006-01-09 11:14:20.000000000 +0100
@@ -30,7 +30,7 @@
     Supports the following chips:
 
     Chip        #vin    #fan    #pwm    #temp   chip_id man_id
-    w83627ehf   -       5       -       3       0x88    0x5ca3
+    w83627ehf   10      5       -       3       0x88    0x5ca3
 
     This is a preliminary version of the driver, only supporting the
     fan and temperature inputs. The chip does much more than that.
@@ -114,6 +114,10 @@
 #define W83627EHF_REG_CHIP_ID		0x49
 #define W83627EHF_REG_MAN_ID		0x4F
 
+static const u16 W83627EHF_REG_IN[] = { 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x550, 0x551, 0x552 };
+static const u16 W83627EHF_REG_IN_MIN[] = { 0x2B, 0x2D, 0x2F, 0x31, 0x33, 0x35, 0x37, 0x554, 0x556, 0x558 };
+static const u16 W83627EHF_REG_IN_MAX[] = { 0x2C, 0x2E, 0x30, 0x32, 0x34, 0x36, 0x38, 0x555, 0x557, 0x559 };
+
 static const u16 W83627EHF_REG_FAN[] = { 0x28, 0x29, 0x2a, 0x3f, 0x553 };
 static const u16 W83627EHF_REG_FAN_MIN[] = { 0x3b, 0x3c, 0x3d, 0x3e, 0x55c };
 
@@ -137,6 +141,22 @@
  */
 
 static inline unsigned int
+in_from_reg(u8 reg)
+{
+	return reg * 8;
+}
+
+static inline u8
+reg_from_in(int val)
+{
+	if (val < 0)
+		return 0;
+	if (val > 2040)
+		return 255;
+	return (val + 4) / 8;
+}
+
+static inline unsigned int
 fan_from_reg(u8 reg, unsigned int div)
 {
 	if (reg = 0 || reg = 255)
@@ -182,6 +202,9 @@
 	unsigned long last_updated;	/* In jiffies */
 
 	/* Register values */
+	u8 in[10];		/* Register value */
+	u8 in_max[10];		/* Register value */
+	u8 in_min[10];		/* Register value */
 	u8 fan[5];
 	u8 fan_min[5];
 	u8 fan_div[5];
@@ -324,6 +347,16 @@
 
 	if (time_after(jiffies, data->last_updated + HZ)
 	 || !data->valid) {
+		/* Measured voltages and limits */
+		for (i = 0; i < 10; i++) {
+			data->in[i] = w83627ehf_read_value(client,
+				      W83627EHF_REG_IN[i]);
+			data->in_min[i] = w83627ehf_read_value(client,
+				          W83627EHF_REG_IN_MIN[i]);
+			data->in_max[i] = w83627ehf_read_value(client,
+				          W83627EHF_REG_IN_MAX[i]);
+		}
+
 		/* Fan clock dividers */
 		i = w83627ehf_read_value(client, W83627EHF_REG_FANDIV1);
 		data->fan_div[0] = (i >> 4) & 0x03;
@@ -402,6 +435,81 @@
 /*
  * Sysfs callback functions
  */
+#define show_in_reg(reg) \
+static ssize_t show_##reg (struct device *dev, char *buf, int nr) \
+{ \
+	struct w83627ehf_data *data = w83627ehf_update_device(dev); \
+	return sprintf(buf,"%d\n", in_from_reg(data->reg[nr])); \
+}
+show_in_reg(in)
+show_in_reg(in_min)
+show_in_reg(in_max)
+
+#define store_in_reg(REG, reg) \
+static ssize_t \
+store_in_##reg (struct device *dev, const char *buf, size_t count, int nr) \
+{ \
+	struct i2c_client *client = to_i2c_client(dev); \
+	struct w83627ehf_data *data = i2c_get_clientdata(client); \
+	u32 val; \
+	 \
+	val = simple_strtoul(buf, NULL, 10); \
+	 \
+	down(&data->update_lock); \
+	data->in_##reg[nr] = reg_from_in(val); \
+	w83627ehf_write_value(client, W83627EHF_REG_IN_##REG[nr], \
+			    data->in_##reg[nr]); \
+	 \
+	up(&data->update_lock); \
+	return count; \
+}
+store_in_reg(MIN, min)
+store_in_reg(MAX, max)
+	
+#define sysfs_in_offset(offset) \
+static ssize_t \
+show_regs_in_##offset (struct device *dev, struct device_attribute *attr, char *buf) \
+{ \
+	return show_in(dev, buf, offset); \
+} \
+static DEVICE_ATTR(in##offset##_input, S_IRUGO, show_regs_in_##offset, NULL);
+
+#define sysfs_in_reg_offset(reg, offset) \
+static ssize_t show_regs_in_##reg##offset (struct device *dev, struct device_attribute *attr, char *buf) \
+{ \
+	return show_in_##reg (dev, buf, offset); \
+} \
+static ssize_t \
+store_regs_in_##reg##offset (struct device *dev, struct device_attribute *attr, \
+			    const char *buf, size_t count) \
+{ \
+	return store_in_##reg (dev, buf, count, offset); \
+} \
+static DEVICE_ATTR(in##offset##_##reg, S_IRUGO | S_IWUSR, \
+		  show_regs_in_##reg##offset, store_regs_in_##reg##offset);
+
+#define sysfs_in_offsets(offset) \
+sysfs_in_offset(offset) \
+sysfs_in_reg_offset(min, offset) \
+sysfs_in_reg_offset(max, offset)
+	
+sysfs_in_offsets(0);
+sysfs_in_offsets(1);
+sysfs_in_offsets(2);
+sysfs_in_offsets(3);
+sysfs_in_offsets(4);
+sysfs_in_offsets(5);
+sysfs_in_offsets(6);
+sysfs_in_offsets(7);
+sysfs_in_offsets(8);
+sysfs_in_offsets(9);
+
+#define device_create_file_in(client, offset) \
+do { \
+	device_create_file(&client->dev, &dev_attr_in##offset##_input); \
+	device_create_file(&client->dev, &dev_attr_in##offset##_max); \
+	device_create_file(&client->dev, &dev_attr_in##offset##_min); \
+} while (0)
 
 #define show_fan_reg(reg) \
 static ssize_t \
@@ -522,21 +630,23 @@
 static DEVICE_ATTR(fan##offset##_div, S_IRUGO, \
 		   show_reg_fan##offset##_div, NULL);
 
-sysfs_fan_offset(1);
-sysfs_fan_min_offset(1);
-sysfs_fan_div_offset(1);
-sysfs_fan_offset(2);
-sysfs_fan_min_offset(2);
-sysfs_fan_div_offset(2);
-sysfs_fan_offset(3);
-sysfs_fan_min_offset(3);
-sysfs_fan_div_offset(3);
-sysfs_fan_offset(4);
-sysfs_fan_min_offset(4);
-sysfs_fan_div_offset(4);
-sysfs_fan_offset(5);
-sysfs_fan_min_offset(5);
-sysfs_fan_div_offset(5);
+#define sysfs_fan_offsets(offset) \
+sysfs_fan_offset(offset) \
+sysfs_fan_min_offset(offset) \
+sysfs_fan_div_offset(offset)
+
+sysfs_fan_offsets(1);
+sysfs_fan_offsets(2);
+sysfs_fan_offsets(3);
+sysfs_fan_offsets(4);
+sysfs_fan_offsets(5);
+
+#define device_create_file_fan(client, offset) \
+do { \
+	device_create_file(&client->dev, &dev_attr_fan##offset##_input); \
+	device_create_file(&client->dev, &dev_attr_fan##offset##_min); \
+	device_create_file(&client->dev, &dev_attr_fan##offset##_div); \
+} while (0)
 
 #define show_temp1_reg(reg) \
 static ssize_t \
@@ -639,6 +749,13 @@
 sysfs_temp_reg_offset(max, 3);
 sysfs_temp_reg_offset(max_hyst, 3);
 
+#define device_create_file_temp(client, offset) \
+do { \
+	device_create_file(&client->dev, &dev_attr_temp##offset##_input); \
+	device_create_file(&client->dev, &dev_attr_temp##offset##_max); \
+	device_create_file(&client->dev, &dev_attr_temp##offset##_max_hyst); \
+} while (0)
+
 /*
  * Driver and client management
  */
@@ -665,6 +782,13 @@
 					      W83627EHF_REG_TEMP_CONFIG[i],
 					      tmp & 0xfe);
 	}
+	w83627ehf_write_value(client, 0x00, 0x04);
+	w83627ehf_write_value(client, 0x02, 0x04);
+	w83627ehf_write_value(client, 0x04, 0x00);
+	w83627ehf_write_value(client, 0x10, 0x04);
+	w83627ehf_write_value(client, 0x12, 0x00);
+	w83627ehf_write_value(client, 0x60, 0x04);
+	w83627ehf_write_value(client, 0x62, 0x00);
 }
 
 static int w83627ehf_detect(struct i2c_adapter *adapter)
@@ -724,36 +848,31 @@
 		goto exit_detach;
 	}
 
-	device_create_file(&client->dev, &dev_attr_fan1_input);
-	device_create_file(&client->dev, &dev_attr_fan1_min);
-	device_create_file(&client->dev, &dev_attr_fan1_div);
-	device_create_file(&client->dev, &dev_attr_fan2_input);
-	device_create_file(&client->dev, &dev_attr_fan2_min);
-	device_create_file(&client->dev, &dev_attr_fan2_div);
-	device_create_file(&client->dev, &dev_attr_fan3_input);
-	device_create_file(&client->dev, &dev_attr_fan3_min);
-	device_create_file(&client->dev, &dev_attr_fan3_div);
+	device_create_file_in(client, 0);
+	device_create_file_in(client, 1);
+	device_create_file_in(client, 2);
+	device_create_file_in(client, 3);
+	device_create_file_in(client, 4);
+	device_create_file_in(client, 5);
+	device_create_file_in(client, 6);
+	device_create_file_in(client, 7);
+	device_create_file_in(client, 8);
+	device_create_file_in(client, 9);
+
+	device_create_file_fan(client, 1);
+	device_create_file_fan(client, 2);
+	device_create_file_fan(client, 3);
 
 	if (data->has_fan & (1 << 3)) {
-		device_create_file(&client->dev, &dev_attr_fan4_input);
-		device_create_file(&client->dev, &dev_attr_fan4_min);
-		device_create_file(&client->dev, &dev_attr_fan4_div);
+		device_create_file_fan(client, 4);
 	}
 	if (data->has_fan & (1 << 4)) {
-		device_create_file(&client->dev, &dev_attr_fan5_input);
-		device_create_file(&client->dev, &dev_attr_fan5_min);
-		device_create_file(&client->dev, &dev_attr_fan5_div);
+		device_create_file_fan(client, 5);
 	}
-
-	device_create_file(&client->dev, &dev_attr_temp1_input);
-	device_create_file(&client->dev, &dev_attr_temp1_max);
-	device_create_file(&client->dev, &dev_attr_temp1_max_hyst);
-	device_create_file(&client->dev, &dev_attr_temp2_input);
-	device_create_file(&client->dev, &dev_attr_temp2_max);
-	device_create_file(&client->dev, &dev_attr_temp2_max_hyst);
-	device_create_file(&client->dev, &dev_attr_temp3_input);
-	device_create_file(&client->dev, &dev_attr_temp3_max);
-	device_create_file(&client->dev, &dev_attr_temp3_max_hyst);
+	
+	device_create_file_temp(client, 1);
+	device_create_file_temp(client, 2);
+	device_create_file_temp(client, 3);
 
 	return 0;
 

[lm-sensors] w83627ehf fan controll?

[Rudolf Marek]

Hi all,

Sorry for late response. I did some quick review of this code. Maybe Yuan can find it useful.

1) max and min regs are switched
2) code does not use sensor_dev_attr and uses old macro system
3) it contains some tweaks to force PWM (seems from Aurelian attempts to get his PWM working :)

I have a question now:

1) does anyone else has something to show?
2) does anyone has userspace voltage support?

I cant work much on this next moth (feb) busy with final exams. Well maybe I will - as brain relax subroutine :)

Yuan have you this on your own schedule?

Thanks,
Regards
Rudolf


> --- /usr/src/linux-2.6.14/drivers/hwmon/w83627ehf.c	2005-10-28 02:02:08.000000000 +0200
> +++ w83627ehf.c	2006-01-09 11:14:20.000000000 +0100
> @@ -30,7 +30,7 @@
>      Supports the following chips:
>  
>      Chip        #vin    #fan    #pwm    #temp   chip_id man_id
> -    w83627ehf   -       5       -       3       0x88    0x5ca3
> +    w83627ehf   10      5       -       3       0x88    0x5ca3
>  
>      This is a preliminary version of the driver, only supporting the
>      fan and temperature inputs. The chip does much more than that.
> @@ -114,6 +114,10 @@
>  #define W83627EHF_REG_CHIP_ID		0x49
>  #define W83627EHF_REG_MAN_ID		0x4F
>  
> +static const u16 W83627EHF_REG_IN[] = { 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x550, 0x551, 0x552 };
> +static const u16 W83627EHF_REG_IN_MIN[] = { 0x2B, 0x2D, 0x2F, 0x31, 0x33, 0x35, 0x37, 0x554, 0x556, 0x558 };
> +static const u16 W83627EHF_REG_IN_MAX[] = { 0x2C, 0x2E, 0x30, 0x32, 0x34, 0x36, 0x38, 0x555, 0x557, 0x559 };
> +

Min and max are switched.

>  static const u16 W83627EHF_REG_FAN[] = { 0x28, 0x29, 0x2a, 0x3f, 0x553 };
>  static const u16 W83627EHF_REG_FAN_MIN[] = { 0x3b, 0x3c, 0x3d, 0x3e, 0x55c };
>  
> @@ -137,6 +141,22 @@
>   */
>  
>  static inline unsigned int
> +in_from_reg(u8 reg)
> +{
> +	return reg * 8;
> +}
> +
> +static inline u8
> +reg_from_in(int val)
> +{
> +	if (val < 0)
> +		return 0;
> +	if (val > 2040)
> +		return 255;
> +	return (val + 4) / 8;
> +}
> +
> +static inline unsigned int
>  fan_from_reg(u8 reg, unsigned int div)
>  {
>  	if (reg = 0 || reg = 255)
> @@ -182,6 +202,9 @@
>  	unsigned long last_updated;	/* In jiffies */
>  
>  	/* Register values */
> +	u8 in[10];		/* Register value */
> +	u8 in_max[10];		/* Register value */
> +	u8 in_min[10];		/* Register value */
>  	u8 fan[5];
>  	u8 fan_min[5];
>  	u8 fan_div[5];
> @@ -324,6 +347,16 @@
>  
>  	if (time_after(jiffies, data->last_updated + HZ)
>  	 || !data->valid) {
> +		/* Measured voltages and limits */
> +		for (i = 0; i < 10; i++) {
> +			data->in[i] = w83627ehf_read_value(client,
> +				      W83627EHF_REG_IN[i]);
> +			data->in_min[i] = w83627ehf_read_value(client,
> +				          W83627EHF_REG_IN_MIN[i]);
> +			data->in_max[i] = w83627ehf_read_value(client,
> +				          W83627EHF_REG_IN_MAX[i]);
Maybe one tab missing?
> +		}
> +
>  		/* Fan clock dividers */
>  		i = w83627ehf_read_value(client, W83627EHF_REG_FANDIV1);
>  		data->fan_div[0] = (i >> 4) & 0x03;
> @@ -402,6 +435,81 @@
>  /*
>   * Sysfs callback functions
>   */
> +#define show_in_reg(reg) \
> +static ssize_t show_##reg (struct device *dev, char *buf, int nr) \
> +{ \
> +	struct w83627ehf_data *data = w83627ehf_update_device(dev); \
> +	return sprintf(buf,"%d\n", in_from_reg(data->reg[nr])); \
> +}
> +show_in_reg(in)
> +show_in_reg(in_min)
> +show_in_reg(in_max)
> +
> +#define store_in_reg(REG, reg) \
> +static ssize_t \
> +store_in_##reg (struct device *dev, const char *buf, size_t count, int nr) \
> +{ \
> +	struct i2c_client *client = to_i2c_client(dev); \
> +	struct w83627ehf_data *data = i2c_get_clientdata(client); \
> +	u32 val; \
> +	 \
> +	val = simple_strtoul(buf, NULL, 10); \
> +	 \
> +	down(&data->update_lock); \
> +	data->in_##reg[nr] = reg_from_in(val); \
> +	w83627ehf_write_value(client, W83627EHF_REG_IN_##REG[nr], \
> +			    data->in_##reg[nr]); \
> +	 \
> +	up(&data->update_lock); \

I think I have seen some patch that is changing this semaphore to mutexes.

> +	return count; \
> +}
> +store_in_reg(MIN, min)
> +store_in_reg(MAX, max)
> +	
> +#define sysfs_in_offset(offset) \
> +static ssize_t \
> +show_regs_in_##offset (struct device *dev, struct device_attribute *attr, char *buf) \
> +{ \
> +	return show_in(dev, buf, offset); \
> +} \
> +static DEVICE_ATTR(in##offset##_input, S_IRUGO, show_regs_in_##offset, NULL);

Why this way?


What about this:

static struct sensor_device_attribute sda_in_input[] = {
	SENSOR_ATTR(in0_input, S_IRUGO, show_in, NULL, 0),
	SENSOR_ATTR(in1_input, S_IRUGO, show_in, NULL, 1),

and
static struct sensor_device_attribute sda_in_min[] = {
       SENSOR_ATTR(in0_min, S_IWUSR | S_IRUGO, show_in_min, store_in_min, 0),
       SENSOR_ATTR(in1_min, S_IWUSR | S_IRUGO, show_in_min, store_in_min, 1),
       SENSOR_ATTR(in2_min, S_IWUSR | S_IRUGO, show_in_min, store_in_min, 2),

/* following are the sysfs callback functions */
static ssize_t show_in(struct device *dev, struct device_attribute *attr,
                        char *buf)
{
        struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr);
        int nr = sensor_attr->index;
        struct w83792d_data *data = w83792d_update_device(dev);
        return sprintf(buf,"%ld\n", IN_FROM_REG(nr,(in_count_from_reg(nr, data))));
}
I think you get it :)

> +
> +#define sysfs_in_reg_offset(reg, offset) \
> +static ssize_t show_regs_in_##reg##offset (struct device *dev, struct device_attribute *attr, char *buf) \
> +{ \
> +	return show_in_##reg (dev, buf, offset); \
> +} \
> +static ssize_t \
> +store_regs_in_##reg##offset (struct device *dev, struct device_attribute *attr, \
> +			    const char *buf, size_t count) \
> +{ \
> +	return store_in_##reg (dev, buf, count, offset); \
> +} \
> +static DEVICE_ATTR(in##offset##_##reg, S_IRUGO | S_IWUSR, \
> +		  show_regs_in_##reg##offset, store_regs_in_##reg##offset);
> +
> +#define sysfs_in_offsets(offset) \
> +sysfs_in_offset(offset) \
> +sysfs_in_reg_offset(min, offset) \
> +sysfs_in_reg_offset(max, offset)
> +	
> +sysfs_in_offsets(0);
> +sysfs_in_offsets(1);
> +sysfs_in_offsets(2);
> +sysfs_in_offsets(3);
> +sysfs_in_offsets(4);
> +sysfs_in_offsets(5);
> +sysfs_in_offsets(6);
> +sysfs_in_offsets(7);
> +sysfs_in_offsets(8);
> +sysfs_in_offsets(9);
> +
> +#define device_create_file_in(client, offset) \
> +do { \
> +	device_create_file(&client->dev, &dev_attr_in##offset##_input); \
> +	device_create_file(&client->dev, &dev_attr_in##offset##_max); \
> +	device_create_file(&client->dev, &dev_attr_in##offset##_min); \
> +} while (0)
>  
>  #define show_fan_reg(reg) \
>  static ssize_t \
> @@ -522,21 +630,23 @@
>  static DEVICE_ATTR(fan##offset##_div, S_IRUGO, \
>  		   show_reg_fan##offset##_div, NULL);
>  
> -sysfs_fan_offset(1);
> -sysfs_fan_min_offset(1);
> -sysfs_fan_div_offset(1);
> -sysfs_fan_offset(2);
> -sysfs_fan_min_offset(2);
> -sysfs_fan_div_offset(2);
> -sysfs_fan_offset(3);
> -sysfs_fan_min_offset(3);
> -sysfs_fan_div_offset(3);
> -sysfs_fan_offset(4);
> -sysfs_fan_min_offset(4);
> -sysfs_fan_div_offset(4);
> -sysfs_fan_offset(5);
> -sysfs_fan_min_offset(5);
> -sysfs_fan_div_offset(5);
> +#define sysfs_fan_offsets(offset) \
> +sysfs_fan_offset(offset) \
> +sysfs_fan_min_offset(offset) \
> +sysfs_fan_div_offset(offset)
> +
> +sysfs_fan_offsets(1);
> +sysfs_fan_offsets(2);
> +sysfs_fan_offsets(3);
> +sysfs_fan_offsets(4);
> +sysfs_fan_offsets(5);
> +
> +#define device_create_file_fan(client, offset) \
> +do { \
> +	device_create_file(&client->dev, &dev_attr_fan##offset##_input); \
> +	device_create_file(&client->dev, &dev_attr_fan##offset##_min); \
> +	device_create_file(&client->dev, &dev_attr_fan##offset##_div); \
> +} while (0)
>  
>  #define show_temp1_reg(reg) \
>  static ssize_t \
> @@ -639,6 +749,13 @@
>  sysfs_temp_reg_offset(max, 3);
>  sysfs_temp_reg_offset(max_hyst, 3);
>  
> +#define device_create_file_temp(client, offset) \
> +do { \
> +	device_create_file(&client->dev, &dev_attr_temp##offset##_input); \
> +	device_create_file(&client->dev, &dev_attr_temp##offset##_max); \
> +	device_create_file(&client->dev, &dev_attr_temp##offset##_max_hyst); \
> +} while (0)
> +
>  /*
>   * Driver and client management
>   */
> @@ -665,6 +782,13 @@
>  					      W83627EHF_REG_TEMP_CONFIG[i],
>  					      tmp & 0xfe);
>  	}
> +	w83627ehf_write_value(client, 0x00, 0x04);
> +	w83627ehf_write_value(client, 0x02, 0x04);
> +	w83627ehf_write_value(client, 0x04, 0x00);
> +	w83627ehf_write_value(client, 0x10, 0x04);
> +	w83627ehf_write_value(client, 0x12, 0x00);
> +	w83627ehf_write_value(client, 0x60, 0x04);
> +	w83627ehf_write_value(client, 0x62, 0x00);

It seems this are your own tweaks. This needs to be removed.

>  }
>  
>  static int w83627ehf_detect(struct i2c_adapter *adapter)
> @@ -724,36 +848,31 @@
>  		goto exit_detach;
>  	}
>  
> -	device_create_file(&client->dev, &dev_attr_fan1_input);
> -	device_create_file(&client->dev, &dev_attr_fan1_min);
> -	device_create_file(&client->dev, &dev_attr_fan1_div);
> -	device_create_file(&client->dev, &dev_attr_fan2_input);
> -	device_create_file(&client->dev, &dev_attr_fan2_min);
> -	device_create_file(&client->dev, &dev_attr_fan2_div);
> -	device_create_file(&client->dev, &dev_attr_fan3_input);
> -	device_create_file(&client->dev, &dev_attr_fan3_min);
> -	device_create_file(&client->dev, &dev_attr_fan3_div);
> +	device_create_file_in(client, 0);
> +	device_create_file_in(client, 1);
> +	device_create_file_in(client, 2);
> +	device_create_file_in(client, 3);
> +	device_create_file_in(client, 4);
> +	device_create_file_in(client, 5);
> +	device_create_file_in(client, 6);
> +	device_create_file_in(client, 7);
> +	device_create_file_in(client, 8);
> +	device_create_file_in(client, 9);
> +
> +	device_create_file_fan(client, 1);
> +	device_create_file_fan(client, 2);
> +	device_create_file_fan(client, 3);
>  
>  	if (data->has_fan & (1 << 3)) {
> -		device_create_file(&client->dev, &dev_attr_fan4_input);
> -		device_create_file(&client->dev, &dev_attr_fan4_min);
> -		device_create_file(&client->dev, &dev_attr_fan4_div);
> +		device_create_file_fan(client, 4);
>  	}
>  	if (data->has_fan & (1 << 4)) {
> -		device_create_file(&client->dev, &dev_attr_fan5_input);
> -		device_create_file(&client->dev, &dev_attr_fan5_min);
> -		device_create_file(&client->dev, &dev_attr_fan5_div);
> +		device_create_file_fan(client, 5);
>  	}
> -
> -	device_create_file(&client->dev, &dev_attr_temp1_input);
> -	device_create_file(&client->dev, &dev_attr_temp1_max);
> -	device_create_file(&client->dev, &dev_attr_temp1_max_hyst);
> -	device_create_file(&client->dev, &dev_attr_temp2_input);
> -	device_create_file(&client->dev, &dev_attr_temp2_max);
> -	device_create_file(&client->dev, &dev_attr_temp2_max_hyst);
> -	device_create_file(&client->dev, &dev_attr_temp3_input);
> -	device_create_file(&client->dev, &dev_attr_temp3_max);
> -	device_create_file(&client->dev, &dev_attr_temp3_max_hyst);
> +	
> +	device_create_file_temp(client, 1);
> +	device_create_file_temp(client, 2);
> +	device_create_file_temp(client, 3);
>  
>  	return 0;
>  
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> lm-sensors mailing list
> lm-sensors at lm-sensors.org
> http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

[lm-sensors] w83627ehf fan controll?

[Ymu]

Hi Rudolf, Aurelien,
> Sorry for late response. I did some quick review of this code. Maybe
Yuan can find it useful.
> 
> 1) max and min regs are switched
> 2) code does not use sensor_dev_attr and uses old macro system
Yes, that's true.
> 3) it contains some tweaks to force PWM (seems from Aurelian attempts
to get his PWM working :)
> 
I use a voltmeter to measure the output of PWM pins of the chip, I don't
expect they can wire all the pwm pins....
Maybe Aurelien can use the same way?


> I have a question now:
> 
> 1) does anyone else has something to show?
> 2) does anyone has userspace voltage support?
> 
Hmm, As what you see,  I have do some modify in my local CVS, and the
code is send to Sebastian with a user tool to give convenience in
testing, but still no reply yet.
> I cant work much on this next moth (feb) busy with final exams. Well
maybe I will - as brain relax subroutine :)
> 
> Yuan have you this on your own schedule?
I will take a holiday from 1/26/2006 to 2/5/2005, and I'm doing
something else now, maybe I can send my patch out before I leave,then
Aurelien and someone else can test and modify it?


Best Regards
Yuan Mu
 



=============================================The privileged confidential information contained in this email is intended for use only by the addressees as indicated by the original sender of this email. If you are not the addressee indicated in this email or are not responsible for delivery of the email to such  a person, please kindly reply to the sender indicating this fact and delete all copies of it from your computer and network server immediately. Your cooperation is highly appreciated. It is advised that any unauthorized use of confidential information of Winbond is strictly prohibited; and any information in this email irrelevant to the official business of Winbond shall be deemed as neither given nor endorsed by Winbond.

[lm-sensors] w83627ehf fan controll?

[Ymu]

Hi all,
The attachments is the diff of linux-2.6.16-rc1 and my local CVS, These
two patches use new sensor attr and add more functions, any comments and
modify are welcome. Hope someone can help fix the bugs in these patches
and send it out.Thanks.

It depends on "[PATCH 15/17] hwmon: Semaphore to mutex conversions"
hwmon-convert-semaphores-to-mutexes.patch
The first patch is w83627ehf_use_new_sensor_device_attr.patch.txt
w83627ehf_add_more_functions.patch.txt is the second.

Best Regards
Yuan Mu



=============================================The privileged confidential information contained in this email is intended for use only by the addressees as indicated by the original sender of this email. If you are not the addressee indicated in this email or are not responsible for delivery of the email to such  a person, please kindly reply to the sender indicating this fact and delete all copies of it from your computer and network server immediately. Your cooperation is highly appreciated. It is advised that any unauthorized use of confidential information of Winbond is strictly prohibited; and any information in this email irrelevant to the official business of Winbond shall be deemed as neither given nor endorsed by Winbond.
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: w83627ehf_use_new_sensor_device_attr.patch.txt
Url: http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060124/d5f60576/w83627ehf_use_new_sensor_device_attr.patch-0001.txt
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: w83627ehf_add_more_functions.patch.txt
Url: http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060124/d5f60576/w83627ehf_add_more_functions.patch-0001.txt

[lm-sensors] w83627ehf fan controll?

[Jean Delvare]

Hi Yuan,

> The attachments is the diff of linux-2.6.16-rc1 and my local CVS, These
> two patches use new sensor attr and add more functions, any comments and
> modify are welcome. Hope someone can help fix the bugs in these patches
> and send it out.Thanks.
> 
> It depends on "[PATCH 15/17] hwmon: Semaphore to mutex conversions"
> hwmon-convert-semaphores-to-mutexes.patch
> The first patch is w83627ehf_use_new_sensor_device_attr.patch.txt
> w83627ehf_add_more_functions.patch.txt is the second.

I've applied the first one to start with (use new sensor device attr),
it was really good. Let alone a few minor alignment issues, the only
change I'd like to do is:

 	for (i = 0; i < 5; i++) {
-		if (!(data->has_fan & (1 << i)))
-			continue;
-		device_create_file_fan(dev, i);
+		if (data->has_fan & (1 << i))
+			device_create_file_fan(dev, i);
 	}

It's clearer and shorter that way. OK?

Also, next time, please provide -p1-style patches so that I can apply
them easily using quilt. And please also include your Signed-off-by
line.

I'll try to find some time tomorrow to review the second patch, unless
Rudolf Marek (or anyone else for that matter) beats me to it.

Thanks,
-- 
Jean Delvare