mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2024-11-30 17:26:38 +07:00
97ae60bb38
The LM99 differs from the LM86, LM89 and LM90 in that it reports remote temperatures (temp2) 16 degrees lower than they really are. So far we have been cheating and handled this in userspace but it really should be handled by the driver directly. Signed-off-by: Jean Delvare <khali@linux-fr.org> Cc: Matthew Garrett <mjg59@srcf.ucam.org>
190 lines
7.4 KiB
Plaintext
190 lines
7.4 KiB
Plaintext
Kernel driver lm90
|
|
==================
|
|
|
|
Supported chips:
|
|
* National Semiconductor LM90
|
|
Prefix: 'lm90'
|
|
Addresses scanned: I2C 0x4c
|
|
Datasheet: Publicly available at the National Semiconductor website
|
|
http://www.national.com/pf/LM/LM90.html
|
|
* National Semiconductor LM89
|
|
Prefix: 'lm89' (no auto-detection)
|
|
Addresses scanned: I2C 0x4c and 0x4d
|
|
Datasheet: Publicly available at the National Semiconductor website
|
|
http://www.national.com/mpf/LM/LM89.html
|
|
* National Semiconductor LM99
|
|
Prefix: 'lm99'
|
|
Addresses scanned: I2C 0x4c and 0x4d
|
|
Datasheet: Publicly available at the National Semiconductor website
|
|
http://www.national.com/pf/LM/LM99.html
|
|
* National Semiconductor LM86
|
|
Prefix: 'lm86'
|
|
Addresses scanned: I2C 0x4c
|
|
Datasheet: Publicly available at the National Semiconductor website
|
|
http://www.national.com/mpf/LM/LM86.html
|
|
* Analog Devices ADM1032
|
|
Prefix: 'adm1032'
|
|
Addresses scanned: I2C 0x4c and 0x4d
|
|
Datasheet: Publicly available at the ON Semiconductor website
|
|
http://www.onsemi.com/PowerSolutions/product.do?id=ADM1032
|
|
* Analog Devices ADT7461
|
|
Prefix: 'adt7461'
|
|
Addresses scanned: I2C 0x4c and 0x4d
|
|
Datasheet: Publicly available at the ON Semiconductor website
|
|
http://www.onsemi.com/PowerSolutions/product.do?id=ADT7461
|
|
* Maxim MAX6646
|
|
Prefix: 'max6646'
|
|
Addresses scanned: I2C 0x4d
|
|
Datasheet: Publicly available at the Maxim website
|
|
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3497
|
|
* Maxim MAX6647
|
|
Prefix: 'max6646'
|
|
Addresses scanned: I2C 0x4e
|
|
Datasheet: Publicly available at the Maxim website
|
|
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3497
|
|
* Maxim MAX6649
|
|
Prefix: 'max6646'
|
|
Addresses scanned: I2C 0x4c
|
|
Datasheet: Publicly available at the Maxim website
|
|
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3497
|
|
* Maxim MAX6657
|
|
Prefix: 'max6657'
|
|
Addresses scanned: I2C 0x4c
|
|
Datasheet: Publicly available at the Maxim website
|
|
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/2578
|
|
* Maxim MAX6658
|
|
Prefix: 'max6657'
|
|
Addresses scanned: I2C 0x4c
|
|
Datasheet: Publicly available at the Maxim website
|
|
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/2578
|
|
* Maxim MAX6659
|
|
Prefix: 'max6657'
|
|
Addresses scanned: I2C 0x4c, 0x4d (unsupported 0x4e)
|
|
Datasheet: Publicly available at the Maxim website
|
|
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/2578
|
|
* Maxim MAX6680
|
|
Prefix: 'max6680'
|
|
Addresses scanned: I2C 0x18, 0x19, 0x1a, 0x29, 0x2a, 0x2b,
|
|
0x4c, 0x4d and 0x4e
|
|
Datasheet: Publicly available at the Maxim website
|
|
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3370
|
|
* Maxim MAX6681
|
|
Prefix: 'max6680'
|
|
Addresses scanned: I2C 0x18, 0x19, 0x1a, 0x29, 0x2a, 0x2b,
|
|
0x4c, 0x4d and 0x4e
|
|
Datasheet: Publicly available at the Maxim website
|
|
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3370
|
|
|
|
|
|
Author: Jean Delvare <khali@linux-fr.org>
|
|
|
|
|
|
Description
|
|
-----------
|
|
|
|
The LM90 is a digital temperature sensor. It senses its own temperature as
|
|
well as the temperature of up to one external diode. It is compatible
|
|
with many other devices, many of which are supported by this driver.
|
|
|
|
Note that there is no easy way to differentiate between the MAX6657,
|
|
MAX6658 and MAX6659 variants. The extra address and features of the
|
|
MAX6659 are not supported by this driver. The MAX6680 and MAX6681 only
|
|
differ in their pinout, therefore they obviously can't (and don't need to)
|
|
be distinguished.
|
|
|
|
The specificity of this family of chipsets over the ADM1021/LM84
|
|
family is that it features critical limits with hysteresis, and an
|
|
increased resolution of the remote temperature measurement.
|
|
|
|
The different chipsets of the family are not strictly identical, although
|
|
very similar. For reference, here comes a non-exhaustive list of specific
|
|
features:
|
|
|
|
LM90:
|
|
* Filter and alert configuration register at 0xBF.
|
|
* ALERT is triggered by temperatures over critical limits.
|
|
|
|
LM86 and LM89:
|
|
* Same as LM90
|
|
* Better external channel accuracy
|
|
|
|
LM99:
|
|
* Same as LM89
|
|
* External temperature shifted by 16 degrees down
|
|
|
|
ADM1032:
|
|
* Consecutive alert register at 0x22.
|
|
* Conversion averaging.
|
|
* Up to 64 conversions/s.
|
|
* ALERT is triggered by open remote sensor.
|
|
* SMBus PEC support for Write Byte and Receive Byte transactions.
|
|
|
|
ADT7461:
|
|
* Extended temperature range (breaks compatibility)
|
|
* Lower resolution for remote temperature
|
|
|
|
MAX6657 and MAX6658:
|
|
* Better local resolution
|
|
* Remote sensor type selection
|
|
|
|
MAX6659:
|
|
* Better local resolution
|
|
* Selectable address
|
|
* Second critical temperature limit
|
|
* Remote sensor type selection
|
|
|
|
MAX6680 and MAX6681:
|
|
* Selectable address
|
|
* Remote sensor type selection
|
|
|
|
All temperature values are given in degrees Celsius. Resolution
|
|
is 1.0 degree for the local temperature, 0.125 degree for the remote
|
|
temperature, except for the MAX6657, MAX6658 and MAX6659 which have a
|
|
resolution of 0.125 degree for both temperatures.
|
|
|
|
Each sensor has its own high and low limits, plus a critical limit.
|
|
Additionally, there is a relative hysteresis value common to both critical
|
|
values. To make life easier to user-space applications, two absolute values
|
|
are exported, one for each channel, but these values are of course linked.
|
|
Only the local hysteresis can be set from user-space, and the same delta
|
|
applies to the remote hysteresis.
|
|
|
|
The lm90 driver will not update its values more frequently than every
|
|
other second; reading them more often will do no harm, but will return
|
|
'old' values.
|
|
|
|
PEC Support
|
|
-----------
|
|
|
|
The ADM1032 is the only chip of the family which supports PEC. It does
|
|
not support PEC on all transactions though, so some care must be taken.
|
|
|
|
When reading a register value, the PEC byte is computed and sent by the
|
|
ADM1032 chip. However, in the case of a combined transaction (SMBus Read
|
|
Byte), the ADM1032 computes the CRC value over only the second half of
|
|
the message rather than its entirety, because it thinks the first half
|
|
of the message belongs to a different transaction. As a result, the CRC
|
|
value differs from what the SMBus master expects, and all reads fail.
|
|
|
|
For this reason, the lm90 driver will enable PEC for the ADM1032 only if
|
|
the bus supports the SMBus Send Byte and Receive Byte transaction types.
|
|
These transactions will be used to read register values, instead of
|
|
SMBus Read Byte, and PEC will work properly.
|
|
|
|
Additionally, the ADM1032 doesn't support SMBus Send Byte with PEC.
|
|
Instead, it will try to write the PEC value to the register (because the
|
|
SMBus Send Byte transaction with PEC is similar to a Write Byte transaction
|
|
without PEC), which is not what we want. Thus, PEC is explicitly disabled
|
|
on SMBus Send Byte transactions in the lm90 driver.
|
|
|
|
PEC on byte data transactions represents a significant increase in bandwidth
|
|
usage (+33% for writes, +25% for reads) in normal conditions. With the need
|
|
to use two SMBus transaction for reads, this overhead jumps to +50%. Worse,
|
|
two transactions will typically mean twice as much delay waiting for
|
|
transaction completion, effectively doubling the register cache refresh time.
|
|
I guess reliability comes at a price, but it's quite expensive this time.
|
|
|
|
So, as not everyone might enjoy the slowdown, PEC can be disabled through
|
|
sysfs. Just write 0 to the "pec" file and PEC will be disabled. Write 1
|
|
to that file to enable PEC again.
|