Skip to content

GitLab

Switch branch/tag
  1. 15 Jul, 2017 1 commit
    • Guenter Roeck's avatar
      hwmon: (applesmc) Avoid buffer overruns · 1009ccdc
      Guenter Roeck authored 
      

gcc 7.1 complains that the driver uses sprintf() and thus does not validate
the length of output buffers.

drivers/hwmon/applesmc.c: In function 'applesmc_show_fan_position':
drivers/hwmon/applesmc.c:82:21: warning:
	'%d' directive writing between 1 and 5 bytes into a region of size 4

Fix the problem by using scnprintf() instead of sprintf() throughout the
driver. Also explicitly limit the number of supported fans to avoid actual
buffer overruns and thus invalid keys.
Signed-off-by: default avatarGuenter Roeck <linux@roeck-us.net>
      1009ccdc
  3. 16 Nov, 2015 1 commit
    • Shuah Khan's avatar
      hwmon : (applesmc) Fix uninitialized variables warnings · 5e0a0ee4
      Shuah Khan authored 
      

Fix the following "maybe used uninitialized" warnings by
initializing the variables to keep the compiler quiet.
There is no "used uninitialized" in this case.

  CC [M]  drivers/hwmon/applesmc.o
drivers/hwmon/applesmc.c: In function ‘applesmc_init_smcreg’:
drivers/hwmon/applesmc.c:595:43: warning: ‘right_light_sensor’
may be used uninitialized in this function [-Wmaybe-uninitialized]
  s->num_light_sensors = left_light_sensor + right_light_sensor;
                                           ^
drivers/hwmon/applesmc.c:540:26: note: ‘right_light_sensor’ was
declared here
  bool left_light_sensor, right_light_sensor;
                          ^
drivers/hwmon/applesmc.c:595:43: warning: ‘left_light_sensor’ may
be used uninitialized in this function [-Wmaybe-uninitialized]
  s->num_light_sensors = left_light_sensor + right_light_sensor;
                                           ^
drivers/hwmon/applesmc.c:540:7: note: ‘left_light_sensor’ was
declared here
  bool left_light_sensor, right_light_sensor;
       ^
Signed-off-by: default avatarShuah Khan <shuahkh@osg.samsung.com>
Signed-off-by: default avatarGuenter Roeck <linux@roeck-us.net>
      5e0a0ee4
  5. 29 Sep, 2015 1 commit
  7. 20 Oct, 2014 1 commit
  9. 09 Oct, 2013 1 commit
    • Henrik Rydberg's avatar
      hwmon: (applesmc) Always read until end of data · 25f2bd7f
      Henrik Rydberg authored 
      

The crash reported and investigated in commit 5f4513 turned out to be
caused by a change to the read interface on newer (2012) SMCs.

Tests by Chris show that simply reading the data valid line is enough
for the problem to go away. Additional tests show that the newer SMCs
no longer wait for the number of requested bytes, but start sending
data right away.  Apparently the number of bytes to read is no longer
specified as before, but instead found out by reading until end of
data. Failure to read until end of data confuses the state machine,
which eventually causes the crash.

As a remedy, assuming bit0 is the read valid line, make sure there is
nothing more to read before leaving the read function.

Tested to resolve the original problem, and runtested on MBA3,1,
MBP4,1, MBP8,2, MBP10,1, MBP10,2. The patch seems to have no effect on
machines before 2012.
Tested-by: default avatarChris Murphy <chris@cmurf.com>
Cc: stable@vger.kernel.org
Signed-off-by: default avatarHenrik Rydberg <rydberg@euromail.se>
Signed-off-by: default avatarGuenter Roeck <linux@roeck-us.net>
      25f2bd7f
  11. 26 Sep, 2013 1 commit
  13. 08 Apr, 2013 1 commit
  15. 10 Oct, 2012 1 commit
  17. 18 Sep, 2012 1 commit
  19. 27 Jul, 2012 1 commit
    • Henrik Rydberg's avatar
      hwmon: (applesmc) Decode and act on read/write status codes · 829917cd
      Henrik Rydberg authored 
      

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: default avatarAdam Somerville <adamsomerville@gmail.com>
Tested-by: default avatarHubert Eichner <hubert.georg.eichner@gmail.com>
Signed-off-by: default avatarHenrik Rydberg <rydberg@euromail.se>
Signed-off-by: default avatarGuenter Roeck <linux@roeck-us.net>
      829917cd
  21. 22 Jul, 2012 4 commits
  23. 25 Jun, 2012 1 commit
  25. 18 Jun, 2012 1 commit
  27. 19 Mar, 2012 2 commits
  29. 05 Jan, 2012 1 commit
  31. 23 Jan, 2011 1 commit
  33. 08 Jan, 2011 12 commits
  35. 27 May, 2010 4 commits
  37. 11 May, 2010 1 commit
  39. 14 Apr, 2010 1 commit
  41. 15 Dec, 2009 1 commit
  43. 22 Sep, 2009 1 commit

Imprint & Privacy Policy