[lm-sensors] Call for 3.3.0

[lm-sensors] Call for 3.3.0

[Jean Delvare]

Hi all,

Last release of lm-sensors (3.2.0) was in October 2010, 5 months ago.
The list of changes since then is getting large, so it's time for a new
release. The changes are important enough that I would go for 3.3.0
rather than 3.2.1 as originally planned.

This decision is motivated by the long list of new attributes and even
new sensor types handled by libsensors and sensors. Other changes worth
mentioning are Guenter's generic code to display limit values in
"sensors", the removal of some arbitrary limits in libsensors attribute
enumeration code, and the usual addition of many new devices to
sensors-detect.

The planned schedule is as follows:
* Saturday, March 19th, 2011: SVN freeze
* March 20th to 25th: Testing and bug fixing
* Saturday, March 26th, 2011: Release

Thanks,
-- 
Jean Delvare

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

Re: [lm-sensors] Call for 3.3.0

[Hans de Goede]

Hi,

On 03/16/2011 10:41 PM, Jean Delvare wrote:
> Hi all,
>
> Last release of lm-sensors (3.2.0) was in October 2010, 5 months ago.
> The list of changes since then is getting large, so it's time for a new
> release. The changes are important enough that I would go for 3.3.0
> rather than 3.2.1 as originally planned.
>
> This decision is motivated by the long list of new attributes and even
> new sensor types handled by libsensors and sensors. Other changes worth
> mentioning are Guenter's generic code to display limit values in
> "sensors", the removal of some arbitrary limits in libsensors attribute
> enumeration code, and the usual addition of many new devices to
> sensors-detect.
>
> The planned schedule is as follows:
> * Saturday, March 19th, 2011: SVN freeze

I have not been able to make time yet to add detection of the new
models handled by f71882fg, in as far as they are not already
present (some probably are haven't checked). I think it would be
good to have this in place before 3.3.0. Note I cannot test
these changes, so they might as well be done by someone else
(hint hint).

Here are the new models supported by the f71882fg driver:

device-id  device-name     Fintek model
0x0814     f71869          F71869F + F71689E (register compatible)
0x0901     f71808e         F71808E
0x0909     f71889ed        F71889ED

And here is a bunch of new models I've datasheets for, but
no requests to support yet, and no support yet:

device-id  device-name(*)  Fintek model
0x1001     f71808a         F71808A
0x1007     f71869a         F71869A
0x1005     f71889a         F71889A

And some more, special models with 6 uarts, but also with
hwmon function:

device-id  device-name(*)  Fintek model
0x0704     f81865f         F81865F
0x1010     f81866d         F81866D

Wrt device-name versus Fintek model, this of course is the same
everywhere, except for the f71869 special case which is why I've
differentiated between the 2 in the above table. The LDN value for
the hwmon part for all of these is 4.

(*) In these cases the device name is not set in stone yet,
as there is no driver support yet.

Regards,

Hans

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

Re: [lm-sensors] Call for 3.3.0

[Jean Delvare]

On Thu, 17 Mar 2011 09:26:57 +0100, Hans de Goede wrote:
> Hi,
> 
> On 03/16/2011 10:41 PM, Jean Delvare wrote:
> > Hi all,
> >
> > Last release of lm-sensors (3.2.0) was in October 2010, 5 months ago.
> > The list of changes since then is getting large, so it's time for a new
> > release. The changes are important enough that I would go for 3.3.0
> > rather than 3.2.1 as originally planned.
> >
> > This decision is motivated by the long list of new attributes and even
> > new sensor types handled by libsensors and sensors. Other changes worth
> > mentioning are Guenter's generic code to display limit values in
> > "sensors", the removal of some arbitrary limits in libsensors attribute
> > enumeration code, and the usual addition of many new devices to
> > sensors-detect.
> >
> > The planned schedule is as follows:
> > * Saturday, March 19th, 2011: SVN freeze
> 
> I have not been able to make time yet to add detection of the new
> models handled by f71882fg, in as far as they are not already
> present (some probably are haven't checked). I think it would be
> good to have this in place before 3.3.0. Note I cannot test
> these changes, so they might as well be done by someone else
> (hint hint).

Hint received. Here's the patch I have come up with:

---
 prog/detect/sensors-detect |   40 +++++++++++++++++++++++++++++++++++-----
 1 file changed, 35 insertions(+), 5 deletions(-)

--- lm-sensors.orig/prog/detect/sensors-detect	2011-03-13 14:25:54.000000000 +0100
+++ lm-sensors/prog/detect/sensors-detect	2011-03-17 10:56:43.000000000 +0100
@@ -1933,23 +1933,35 @@ use constant FEAT_SMBUS	=> (1 << 7);
 		features => FEAT_IN | FEAT_FAN | FEAT_TEMP,
 	}, {
 		name => "Fintek F71808E Super IO Sensors",
-		driver => "to-be-written",	# f71882fg
+		driver => "f71882fg",
 		devid => 0x0901,
 		logdev => 0x04,
 		features => FEAT_IN | FEAT_FAN | FEAT_TEMP,
 	}, {
+		name => "Fintek F71808A Super IO Sensors",
+		driver => "to-be-written",
+		devid => 0x1001,
+		logdev => 0x04,
+		features => FEAT_IN | FEAT_FAN | FEAT_TEMP,
+	}, {
 		name => "Fintek F71862FG Super IO Sensors",
 		driver => "f71882fg",
 		devid => 0x0601,
 		logdev => 0x04,
 		features => FEAT_IN | FEAT_FAN | FEAT_TEMP,
 	}, {
-		name => "Fintek F71869FG Super IO Sensors",
-		driver => "to-be-written",
+		name => "Fintek F71869F/E Super IO Sensors",
+		driver => "f71882fg",
 		devid => 0x0814,
 		logdev => 0x04,
 		features => FEAT_IN | FEAT_FAN | FEAT_TEMP,
 	}, {
+		name => "Fintek F71869A Super IO Sensors",
+		driver => "to-be-written",
+		devid => 0x1007,
+		logdev => 0x04,
+		features => FEAT_IN | FEAT_FAN | FEAT_TEMP,
+	}, {
 		name => "Fintek F71806FG/F71872FG Super IO Sensors",
 		driver => "f71805f",
 		devid => 0x0341,
@@ -1974,12 +1986,18 @@ use constant FEAT_SMBUS	=> (1 << 7);
 		logdev => 0x04,
 		features => FEAT_IN | FEAT_FAN | FEAT_TEMP,
 	}, {
-		name => "Fintek F71889E Super IO Sensors",
-		driver => "to-be-written",
+		name => "Fintek F71889ED Super IO Sensors",
+		driver => "f71882fg",
 		devid => 0x0909,
 		logdev => 0x04,
 		features => FEAT_IN | FEAT_FAN | FEAT_TEMP,
 	}, {
+		name => "Fintek F71889A Super IO Sensors",
+		driver => "to-be-written",
+		devid => 0x1005,
+		logdev => 0x04,
+		features => FEAT_IN | FEAT_FAN | FEAT_TEMP,
+	}, {
 		name => "Fintek F81216D Super IO",
 		driver => "not-a-sensor",
 		devid => 0x0208,
@@ -1988,6 +2006,18 @@ use constant FEAT_SMBUS	=> (1 << 7);
 		driver => "not-a-sensor",
 		devid => 0x0206,
 	}, {
+		name => "Fintek F81865F Super IO Sensors",
+		driver => "to-be-written",
+		devid => 0x0704,
+		logdev => 0x04,
+		features => FEAT_IN | FEAT_FAN | FEAT_TEMP,
+	}, {
+		name => "Fintek F81866D Super IO Sensors",
+		driver => "to-be-written",
+		devid => 0x1010,
+		logdev => 0x04,
+		features => FEAT_IN | FEAT_FAN | FEAT_TEMP,
+	}, {
 		name => "Asus F8000 Super IO",
 		driver => "f71882fg",
 		devid => 0x0581,

I've also made a test version of sensors-detect available at:
  http://khali.linux-fr.org/devel/misc/sensors-detect

Anyone with any of these chips is invited to test.

> 
> Here are the new models supported by the f71882fg driver:
> 
> device-id  device-name     Fintek model
> 0x0814     f71869          F71869F + F71689E (register compatible)
> 0x0901     f71808e         F71808E
> 0x0909     f71889ed        F71889ED
> 
> And here is a bunch of new models I've datasheets for, but
> no requests to support yet, and no support yet:
> 
> device-id  device-name(*)  Fintek model
> 0x1001     f71808a         F71808A
> 0x1007     f71869a         F71869A
> 0x1005     f71889a         F71889A
> 
> And some more, special models with 6 uarts, but also with
> hwmon function:
> 
> device-id  device-name(*)  Fintek model
> 0x0704     f81865f         F81865F
> 0x1010     f81866d         F81866D
> 
> Wrt device-name versus Fintek model, this of course is the same
> everywhere, except for the f71869 special case which is why I've
> differentiated between the 2 in the above table. The LDN value for
> the hwmon part for all of these is 4.
> 
> (*) In these cases the device name is not set in stone yet,
> as there is no driver support yet.

It doesn't matter, sensors-detect doesn't care about device name
strings.

-- 
Jean Delvare

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

Re: [lm-sensors] Call for 3.3.0

[Hans de Goede]

Hi,

On 03/17/2011 10:59 AM, Jean Delvare wrote:
> On Thu, 17 Mar 2011 09:26:57 +0100, Hans de Goede wrote:
>> Hi,
>>
>> On 03/16/2011 10:41 PM, Jean Delvare wrote:
>>> Hi all,
>>>
>>> Last release of lm-sensors (3.2.0) was in October 2010, 5 months ago.
>>> The list of changes since then is getting large, so it's time for a new
>>> release. The changes are important enough that I would go for 3.3.0
>>> rather than 3.2.1 as originally planned.
>>>
>>> This decision is motivated by the long list of new attributes and even
>>> new sensor types handled by libsensors and sensors. Other changes worth
>>> mentioning are Guenter's generic code to display limit values in
>>> "sensors", the removal of some arbitrary limits in libsensors attribute
>>> enumeration code, and the usual addition of many new devices to
>>> sensors-detect.
>>>
>>> The planned schedule is as follows:
>>> * Saturday, March 19th, 2011: SVN freeze
>>
>> I have not been able to make time yet to add detection of the new
>> models handled by f71882fg, in as far as they are not already
>> present (some probably are haven't checked). I think it would be
>> good to have this in place before 3.3.0. Note I cannot test
>> these changes, so they might as well be done by someone else
>> (hint hint).
>
> Hint received. Here's the patch I have come up with:
>

Looks good, thanks.

Regards,

Hans

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

Re: [lm-sensors] Call for 3.3.0

[Jean Delvare]

Hi Hans,

On Thu, 17 Mar 2011 12:28:51 +0100, Hans de Goede wrote:
> On 03/17/2011 10:59 AM, Jean Delvare wrote:
> > On Thu, 17 Mar 2011 09:26:57 +0100, Hans de Goede wrote:
> >> I have not been able to make time yet to add detection of the new
> >> models handled by f71882fg, in as far as they are not already
> >> present (some probably are haven't checked). I think it would be
> >> good to have this in place before 3.3.0. Note I cannot test
> >> these changes, so they might as well be done by someone else
> >> (hint hint).
> >
> > Hint received. Here's the patch I have come up with:
> 
> Looks good, thanks.

Care to update wiki/Devices?

Thanks,
-- 
Jean Delvare

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

Re: [lm-sensors] Call for 3.3.0

[Hans de Goede]

Hi,

On 03/20/2011 05:19 PM, Jean Delvare wrote:
> Hi Hans,
>
> On Thu, 17 Mar 2011 12:28:51 +0100, Hans de Goede wrote:
>> On 03/17/2011 10:59 AM, Jean Delvare wrote:
>>> On Thu, 17 Mar 2011 09:26:57 +0100, Hans de Goede wrote:
>>>> I have not been able to make time yet to add detection of the new
>>>> models handled by f71882fg, in as far as they are not already
>>>> present (some probably are haven't checked). I think it would be
>>>> good to have this in place before 3.3.0. Note I cannot test
>>>> these changes, so they might as well be done by someone else
>>>> (hint hint).
>>>
>>> Hint received. Here's the patch I have come up with:
>>
>> Looks good, thanks.
>
> Care to update wiki/Devices?

Done, I've updated all Fintek models which are newly supported,
added the F718xxA + F81865 / F81866 models where are newly detected,
but not yet supported and updated the entry for the SMSC SCH5627 too.

Regards,

Hans

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