Open darcagn opened 2 years ago
Summary for this board.
The UPEC
method fills in 9 sensors, numbered 19 to 27 (0x13 to 0x1b):
Method (UPEC, 0, Serialized)
{
If ((Acquire (_GL, 0x03E8) == Zero))
{
WECB (Zero)
Local0 = REC1 (0x90)
Local1 = REC1 (0x91)
Local0 = ((Local0 << 0x08) | Local1)
Local0 *= 0x0C35
NU19 = Local0
Local0 = REC1 (0x3A)
NU20 = Local0
Local0 = REC1 (0x3D)
NU21 = Local0
Local0 = REC1 (0x3E)
NU22 = Local0
Local0 = REC1 (0xF4)
NU23 = Local0
Local0 = REC1 (0xBC)
Local1 = REC1 (0xBD)
Local0 = ((Local0 << 0x08) | Local1)
NU24 = Local0
Local0 = REC2 (One, 0x0D)
NU25 = Local0
Local0 = REC2 (One, 0x0B)
NU26 = Local0
Local0 = REC2 (Zero, 0xB4)
Local1 = REC2 (Zero, 0xB5)
Local0 = ((Local0 << 0x08) | Local1)
NU27 = Local0
Release (_GL)
Return (Zero)
}
Return (Ones)
}
Corresponding records from the INFO
package:
Package (0x06)
{
"SB 1.05V Voltage",
Zero,
0x03,
0x02,
0x03,
0x13
},
Package (0x06)
{
"Chipset Temperature",
One,
0x04,
0x02,
Zero,
0x14
},
Package (0x06)
{
"Tsensor 1 Temperature",
One,
0x05,
0x02,
Zero,
0x15
},
Package (0x06)
{
"CPU VRM Temperature",
One,
0x06,
0x02,
Zero,
0x16
},
Package (0x06)
{
"CPU VRM Output Current",
0x03,
0x06,
0x02,
One,
0x17
},
Package (0x06)
{
"CPU OPT",
0x02,
0x03,
0x02,
One,
0x18
},
Package (0x06)
{
"Water In",
One,
0x07,
0x02,
Zero,
0x19
},
Package (0x06)
{
"Water Out",
One,
0x07,
0x02,
Zero,
0x1A
},
Package (0x06)
{
"Water Flow",
0x04,
0x07,
0x02,
One,
0x1B
}
Label | source | size | address | notes |
---|---|---|---|---|
SB 1.05V Voltage | REC1 | 2 | 0x90 | *= 0x0C35 |
Chipset Temperature | REC1 | 1 | 0x3a | |
Tsensor 1 Temperature | REC1 | 1 | 0x3d | |
CPU VRM Temperature | REC1 | 1 | 0x3e | |
CPU VRM Output Current | REC1 | 1 | 0xf4 | |
CPU OPT | REC1 | 2 | 0xbc | |
Water In | REC2 | 1 | 0x01, 0x0d | |
Water Out | REC2 | 1 | 0x01, 0x0b | |
Water Flow | REC2 | 2 | 0xb4 |
Now, what does REC2()
do?
Recompiling and testing with these new values:
Now, what does REC2() do?
Here is its code:
Method (REC2, 2, Serialized)
{
IBFY ()
EC6C = 0x91
IBFY ()
EC68 = 0xFF
IBFY ()
EC68 = Arg0
IBFY ()
EC6C = 0x90
IBFY ()
EC68 = Arg1
OBFY ()
Local0 = EC68 /* \AMW0.EC68 */
Return (Local0)
}
Looks like it reads from a EC controller via ports 0x68 and 0x6c, but what are constants 0x91 and 0x90 for?
- SB 1.05V Voltage shows up, but it's only showing about 338mV when asus-wmi-sensors shows 1.06V. I don't know what the *=0x0C35 means, though.
I guess it is a conversion factor to volts. 0x0C35 (=3125) * 0.338 = 1.056, which is pretty close to 1.06.
- Chipset Temp, Tsensor 1 Temp, CPU VRM Temp, CPU VRM Current, and CPU OPT values are working properly and agree with asus-wmi-sensors.
Great!
- The three sensors showing REC2 don't work, I guess something else has to be changed.
They need to be read via other EC ports (see REC2() above).
I guess it is a conversion factor to volts. 0x0C35 (=3125) * 0.338 = 1.056, which is pretty close to 1.06.
Sorry, to µV, of course: 3125 338 1e-6 = 1.05625
board_name:
ROG CROSSHAIR VI HERO (WI-FI AC)
mutex name:\AMW0.ASMX
(same as others)This board is being used in an Ubuntu LTS server with an older 5.4 kernel -- it was unable to compile at first because linux/dev_printk.h and linux/units.h weren't found. I removed the includes for those two header files and it seems to compile and work just fine, but I am unsure if functionality has been affected.
These sensors work just fine and match the same output from asus-wmi-sensors:
SENSOR_SET_TEMP_CHIPSET_CPU_MB | SENSOR_TEMP_T_SENSOR | SENSOR_CURR_CPU
SENSOR_FAN_CPU_OPT_X470
is the same situation as the X470 board, CPU_OPT is at 0x00bc (in place of Water_Flow for 500-series).I plugged in fans/probes to the other sensor connectors and could not get a reading from any other sensor, including CPU voltage. I didn't get to poke around the registers with this board though, so there may be other undiscovered sensors exposed in the EC.
I have attached the DSDT file for this board if it helps. ROG CROSSHAIR VI HERO (WI-FI AC).zip
Example output with fan plugged into CPU_OPT and temp probe attached to T_SENSOR: