Hi Constantine,
On 07/25/2015 02:45 PM, Constantine Shulyupin wrote:
Introduced REG_PWM, pwm[1..3]_auto_point[1..5]_temp,
pwm[1..3]_auto_point[1..5]_pwm, nct7802_auto_point_attrs,
nct7802_auto_point_group, updated nct7802_regmap_is_volatile
Signed-off-by: Constantine Shulyupin <const@xxxxxxxxxxxxx>
---
Changed in v4:
- fixed expression in nct7802_regmap_is_volatile
- added pwmX_auto_temp to driver's documentation
Changed in v3:
- removed nct7802_auto_point_is_visible
- removed usage of sysfs_update_group
- introduced REG_PWM
- removed S_IWUSR from RO attributes
- added PWM registers to nct7802_regmap_is_volatile
Changed in v2:
- removed PWM_REG, TEMP_REG
- removed auto_point[1..4]_temp, auto_point[1..4]_pwm
and auto_point_crit_temp
- introduced pwm[1..3]_auto_point[1..5]_temp
and pwm[1..3]_auto_point[1..5]_pwm.
- introduced nct7802_auto_point_is_visible
- used sysfs_update_group in store_pwm_enable
Default values on my platform:
fan1_alarm:0 pwm2_auto_point1_temp:25000
fan1_beep:0 pwm2_auto_point2_pwm:170
fan1_input:150000 pwm2_auto_point2_temp:35000
fan1_input is a bit high, isn't it ?
fan1_min:0 pwm2_auto_point3_pwm:200
fan2_alarm:0 pwm2_auto_point3_temp:45000
fan2_beep:0 pwm2_auto_point4_pwm:230
fan2_input:0 pwm2_auto_point4_temp:55000
fan2_min:0 pwm2_auto_point5_pwm:255
fan3_alarm:0 pwm2_auto_point5_temp:60000
fan3_beep:0 pwm2_enable:1
fan3_input:0 pwm2_mode:1
fan3_min:0 pwm3:127
in0_alarm:0 pwm3_auto_point1_pwm:140
in0_beep:0 pwm3_auto_point1_temp:25000
in0_input:3312 pwm3_auto_point2_pwm:170
in0_max:4092 pwm3_auto_point2_temp:35000
in0_min:0 pwm3_auto_point3_pwm:200
in1_input:902 pwm3_auto_point3_temp:45000
in3_alarm:0 pwm3_auto_point4_pwm:230
in3_beep:0 pwm3_auto_point4_temp:55000
in3_input:1808 pwm3_auto_point5_pwm:255
in3_max:2046 pwm3_auto_point5_temp:60000
in3_min:0 pwm3_enable:1
in4_alarm:0 pwm3_mode:1
in4_beep:0 temp1_beep:0
in4_input:1508 temp1_crit:100000
in4_max:2046 temp1_crit_alarm:0
in4_min:0 temp1_fault:0
name:nct7802 temp1_input:35500
pwm1:127 temp1_max:85000
pwm1_auto_point1_pwm:140 temp1_max_alarm:0
pwm1_auto_point1_temp:25000 temp1_min:0
pwm1_auto_point2_pwm:170 temp1_min_alarm:0
pwm1_auto_point2_temp:35000 temp1_type:4
pwm1_auto_point3_pwm:200 temp4_beep:0
pwm1_auto_point3_temp:45000 temp4_crit:100000
pwm1_auto_point4_pwm:230 temp4_crit_alarm:0
pwm1_auto_point4_temp:55000 temp4_input:39000
pwm1_auto_point5_pwm:255 temp4_max:85000
pwm1_auto_point5_temp:60000 temp4_max_alarm:0
pwm1_enable:1 temp4_min:0
pwm1_mode:1 temp4_min_alarm:0
pwm2:127 temp6_beep:0
pwm2_auto_point1_pwm:140 temp6_input:0
Question is if the values above are consistent for you when you reload
the driver (without updating the pwm values, which I think you did above).
I am having trouble with both the real target and with my module test code.
It fails for me with all kernel versions from 3.16 to 4.1 if the regmap
cache type is set to REGCACHE_RBTREE. I am currently trying to track this down.
---
Documentation/hwmon/nct7802 | 5 +-
drivers/hwmon/nct7802.c | 130 ++++++++++++++++++++++++++++++++++++++++++--
2 files changed, 128 insertions(+), 7 deletions(-)
diff --git a/Documentation/hwmon/nct7802 b/Documentation/hwmon/nct7802
index 2e00f5e..c5c25a5 100644
--- a/Documentation/hwmon/nct7802
+++ b/Documentation/hwmon/nct7802
@@ -5,7 +5,7 @@ Supported chips:
* Nuvoton NCT7802Y
Prefix: 'nct7802'
Addresses scanned: I2C 0x28..0x2f
- Datasheet: Available from Nuvoton web site
+ Datasheet: http://www.nuvoton.com/hq/products/cloud-computing/hardware-monitors/desktop-server-series/nct7802y/
Reason for not listing the exact datasheet link is that it keeps changing.
Please drop this change.
Authors:
Guenter Roeck <linux@xxxxxxxxxxxx>
@@ -17,8 +17,7 @@ This driver implements support for the Nuvoton NCT7802Y hardware monitoring
chip. NCT7802Y supports 6 temperature sensors, 5 voltage sensors, and 3 fan
speed sensors.
-The chip also supports intelligent fan speed control. This functionality is
-not currently supported by the driver.
+Smart Fan™ speed control is avilabel via pwmX_auto_temp attributes.
s/avilabel/available/
No need to resubmit at this time. The rest of the patch looks good, but
I want to track down the regmap problem before accepting it. If it is a
regmap problem, kind of odd that others don't see it, so I am not sure
if that is the case. If it is a driver problem, we'll need to fix it
before applying the patch. I'll let you know what I find.
Thanks,
Guenter
_______________________________________________
lm-sensors mailing list
lm-sensors@xxxxxxxxxxxxxx
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
