linux-toradex.git/drivers/hwmon/applesmc.c, branch v3.7.9

hwmon: Add missing inclusions of

2012-10-10T13:25:56+00:00

These drivers use IS_ERR so they should include .

Signed-off-by: Jean Delvare 
Acked-by: Guenter Roeck 
Acked-by: Luca Tettamanti 
Cc: Henrik Rydberg

hwmon: (applesmc) Bump max wait

2012-09-18T20:20:49+00:00

A heavy-load test on a MacBookPro6,1 is still showing a substantial
amount of read errors.  Increasing the maximum wait time to 128 ms
resolves the issue.

Signed-off-by: Parag Warudkar 
Signed-off-by: Henrik Rydberg 
Signed-off-by: Guenter Roeck

hwmon: (applesmc) Decode and act on read/write status codes

2012-07-27T21:00:12+00:00

The behavior of the SMC has changed several times over the years,
causing read failures in the driver. It seems the problem can be
explained by a shift in SMC speed combined with improper action on
status codes.

We should first wait for the SMC to settle, which was the most
frequent response on the old slow machines. Then, if the SMC is busy,
we need to try again later by resending the command. This was the most
likely response until 2012. Now, with a shorter wait time, we are
again most likely to poll while the SMC is settling, and as a result
we see high failure rates on many old and new models.

With the distinction between busy and failure, we can also wait longer
before retrying, without sacrificing speed.  This seems to bring
failures down to virtually zero on all models.

Tested on: MBA1,1 MBA3,1 MBA5,1 MBA5,2 MBP9,2

Tested-by: Adam Somerville 
Tested-by: Hubert Eichner 
Signed-off-by: Henrik Rydberg 
Signed-off-by: Guenter Roeck

hwmon: (applesmc) Ignore some temperature registers

2012-07-22T04:48:45+00:00

Not all sensors in the T range are useful temperatures.  This patch
creates a subset of sensors to be exported to userland, excluding the
unknown types.

Signed-off-by: Henrik Rydberg 
Signed-off-by: Guenter Roeck

hwmon: (applesmc) Allow negative temperature values

2012-07-22T04:48:45+00:00

There are many userland reports of sensors with unreasonably small and
large temperatures. There seem to be several reasons for this:

Firstly, the major sensor type (sp78) is actually a signed number.
This explains why some sensors show very small or large values - they
are in fact all small, but of different sign.

Secondly, the other sensor type (1-hex) is not properly understood; it
may be that it is not a temperature after all.

Thirdly, some sensors are differential in nature, showing changes over
time rather than absolute numbers.  This explains why those values are
small and of varying sign.

This patch interprets the sp78 type as signed short, but keeps the
original scaling. For other types, -EINVAL is returned, since the
nature of those sensors is unknown.

Signed-off-by: Henrik Rydberg 
Signed-off-by: Guenter Roeck

hwmon: (applesmc) Shorten minimum wait time

2012-07-22T04:48:45+00:00

The 2012 series of MacBooks have a faster SMC, and the current driver
timings do not work at all. Tests show that decreasing the minimum
wait time, from 64 us to 16 us, works well. Since this is still larger
than the original minimum of 10 us used before 2008, there is nothing
inherently problematic with changing it. The fail frequency on older
machines seems to increase slightly, but not enough to be noticeable.

Tested on MBA11, MBA31, MBA5,2, MBP9,2.

The patch was originally written by adamski99 (ubuntuforums.org) and
later tested by janhouse (bbs.archlinux.org).

Signed-off-by: Henrik Rydberg 
Signed-off-by: Guenter Roeck

hwmon: (applesmc) Skip sensor mapping

2012-07-22T04:48:44+00:00

The special motion sensor mapping is unnecessary; remove it.

Signed-off-by: Henrik Rydberg 
Signed-off-by: Guenter Roeck

hwmon: (applesmc) correct email address for Jesper Juhl

2012-06-25T06:36:03+00:00

I've not had a gmail address for years. This commit updates the
address to my actual working one.

Signed-off-by: Jesper Juhl 
Signed-off-by: Guenter Roeck

hwmon: (applesmc) Limit key length in warning messages

2012-06-18T15:48:00+00:00

Key lookups may call read_smc() with a fixed-length key string,
and if the lookup fails, trailing stack content may appear in the
kernel log. Fixed with this patch.

Signed-off-by: Henrik Rydberg 
Cc: stable@vger.kernel.org
Signed-off-by: Guenter Roeck

hwmon: (applesmc) Silence uninitialized warnings

2012-03-19T01:27:41+00:00

Some error paths do not set a result, leading to the (false)
assumption that the value may be used uninitialized. Set results for
those paths as well.

Signed-off-by: Henrik Rydberg 
Signed-off-by: Guenter Roeck