Skip to contentSkip to conversation

w83627ehf: Fix the detection of fan5

PATCH New Greg KH 4 participants 5 messages Oct 18, 2006 – Dec 9, 2006 Hardware monitoring

lore
Share & tools
lore

Greg KH

author
#0 lore raw

PATCH 6/8 w83627ehf: Fix the detection of fan5

From: Rudolf Marek <r.marek at assembler.cz>

Fix the detection of fan5 and preserve the bit between the register writes, because the bit is write only.

Signed-off-by Rudolf Marek <r.marek at assembler.cz>
Signed-off-by Jean Delvare <khali at linux-fr.org>
Signed-off-by Greg Kroah-Hartman <gregkh at suse.de>
---
 drivers/hwmon/w83627ehf.c |   11 +++++++++--
 1 files changed, 9 insertions(+), 2 deletions(-)
drivers/hwmon/w83627ehf.c +9 -2
index 833faa2..2257806 100644
@@ -354,6 +354,8 @@ static void w83627ehf_write_fan_div(stru
 	case 0:
 		reg = (w83627ehf_read_value(client, W83627EHF_REG_FANDIV1) & 0xcf)
 		    | ((data->fan_div[0] & 0x03) << 4);
+		/* fan5 input control bit is write only, compute the value */
+		reg |= (data->has_fan & (1 << 4)) ? 1 : 0;
 		w83627ehf_write_value(client, W83627EHF_REG_FANDIV1, reg);
 		reg = (w83627ehf_read_value(client, W83627EHF_REG_VBAT) & 0xdf)
 		    | ((data->fan_div[0] & 0x04) << 3);
@@ -362,6 +364,8 @@ static void w83627ehf_write_fan_div(stru
 	case 1:
 		reg = (w83627ehf_read_value(client, W83627EHF_REG_FANDIV1) & 0x3f)
 		    | ((data->fan_div[1] & 0x03) << 6);
+		/* fan5 input control bit is write only, compute the value */
+		reg |= (data->has_fan & (1 << 4)) ? 1 : 0;
 		w83627ehf_write_value(client, W83627EHF_REG_FANDIV1, reg);
 		reg = (w83627ehf_read_value(client, W83627EHF_REG_VBAT) & 0xbf)
 		    | ((data->fan_div[1] & 0x04) << 4);
@@ -1216,13 +1220,16 @@ static int w83627ehf_detect(struct i2c_a
 	superio_exit();
 
 	/* It looks like fan4 and fan5 pins can be alternatively used
-	   as fan on/off switches */
+	   as fan on/off switches, but fan5 control is write only :/
+	   We assume that if the serial interface is disabled, designers
+	   connected fan5 as input unless they are emitting log 1, which
+	   is not the default. */
 
 	data->has_fan = 0x07; /* fan1, fan2 and fan3 */
 	i = w83627ehf_read_value(client, W83627EHF_REG_FANDIV1);
 	if ((i & (1 << 2)) && (!fan4pin))
 		data->has_fan |= (1 << 3);
-	if ((i & (1 << 0)) && (!fan5pin))
+	if (!(i & (1 << 1)) && (!fan5pin))
 		data->has_fan |= (1 << 4);
 
 	/* Register sysfs hooks */
Signature · 1 line
1.4.2.4
#1 lore raw

PATCH 6/8 w83627ehf: Fix the detection of fan5

Hi Jean, Rudolf,

I'm doing bits and pieces of the w83627dhg driver, and I want to make sure I test it on the latest patched w83627ehf driver. I believe this is Rudolf's last patch, the last update to the driver. However, the machine I develop on at home has a very slow internet connection, so I have tried linux-2.6.18-mm2 and linux-2.6.19-rc3. Neither seems to have that patch.

Has Rudolf's patch been pushed upstream? I'm replying on the thread where Greg K received the patch. Sorry, this should be an easy one, but I'm stuck.

Thanks,
David

#2 lore raw

PATCH 6/8 w83627ehf: Fix the detection of fan5

In reply to Greg KH

Hi David,

Quoted 10 lines · David Hubbard

I'm doing bits and pieces of the w83627dhg driver, and I want to make sure I test it on the latest patched w83627ehf driver. I believe this is Rudolf's last patch, the last update to the driver. However, the machine I develop on at home has a very slow internet connection, so I have tried linux-2.6.18-mm2 and linux-2.6.19-rc3. Neither seems to have that patch.

Has Rudolf's patch been pushed upstream? I'm replying on the thread where Greg K received the patch. Sorry, this should be an easy one, but I'm stuck.

The fan5 detection patch has been pushed upstream, it was merged in 2.6.19-rc3. There are no pending patches for the w83627ehf driver in my tree so the version in 2.6.19-rc3 is the latest one.

Signature · 1 line
Jean Delvare
#3 lore raw

PATCH 6/8 w83627ehf: Fix the detection of fan5

In reply to Greg KH

Hi David,

I always could read fans5 on my P5B wifi deluxe board.

I am still on 2.6.18.2 kernel with your version of w83627ehf driver (and Rudolfs coretemp patch).
anything you want me to try out?

Christian

Am Donnerstag, 7. Dezember 2006 22:01 schrieb David Hubbard:

Quoted 13 lines

Hi Jean, Rudolf,

I'm doing bits and pieces of the w83627dhg driver, and I want to make sure I test it on the latest patched w83627ehf driver. I believe this is Rudolf's last patch, the last update to the driver. However, the machine I develop on at home has a very slow internet connection, so I have tried linux-2.6.18-mm2 and linux-2.6.19-rc3. Neither seems to have that patch.

Has Rudolf's patch been pushed upstream? I'm replying on the thread where Greg K received the patch. Sorry, this should be an easy one, but I'm stuck.

Quoted 7 lines

Thanks,
David

_______________________________________________
lm-sensors mailing list
lm-sensors at lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

___________________________________________________________
Der fr?he Vogel f?ngt den Wurm. Hier gelangen Sie zum neuen Yahoo! Mail: http://mail.yahoo.de

#4 lore raw

PATCH 6/8 w83627ehf: Fix the detection of fan5

In reply to Greg KH

Hi Christian,

You have a w83627dhg, if I remember correctly. The current driver you are probably using is the testing driver from the lm-sensors mailing list in September. Although it functions correctly for the w83627dhg, it does not correctly detect some of the w83627ehf chip revisions. I've taken way too long to get the driver fixed so it can be added to the kernel. Mostly that's because my work has become very intense for the last few months -- I haven't been sleeping much. However, w83627dhg support should be in a testing release of the kernel soon. (Possibly 2.6.20)

David

Quoted 9 lines · Christian Mahr

Hi David,

I always could read fans5 on my P5B wifi deluxe board.

I am still on 2.6.18.2 kernel with your version of w83627ehf driver (and Rudolfs coretemp patch).
anything you want me to try out?

Christian