blob: cbac32b59c8c96a1bc86a3455a3ee4cbe3342db5 [file] [log] [blame]
R.Marek@sh.cvut.cz7f15b662005-05-26 12:42:19 +00001Kernel driver pc87360
2=====================
3
4Supported chips:
5 * National Semiconductor PC87360, PC87363, PC87364, PC87365 and PC87366
6 Prefixes: 'pc87360', 'pc87363', 'pc87364', 'pc87365', 'pc87366'
7 Addresses scanned: none, address read from Super I/O config space
Jean Delvare6aa693b2008-10-17 17:51:17 +02008 Datasheets: No longer available
R.Marek@sh.cvut.cz7f15b662005-05-26 12:42:19 +00009
10Authors: Jean Delvare <khali@linux-fr.org>
11
12Thanks to Sandeep Mehta, Tonko de Rooy and Daniel Ceregatti for testing.
13Thanks to Rudolf Marek for helping me investigate conversion issues.
14
15
16Module Parameters
17-----------------
18
19* init int
20 Chip initialization level:
21 0: None
22 *1: Forcibly enable internal voltage and temperature channels, except in9
23 2: Forcibly enable all voltage and temperature channels, except in9
24 3: Forcibly enable all voltage and temperature channels, including in9
25
26Note that this parameter has no effect for the PC87360, PC87363 and PC87364
27chips.
28
29Also note that for the PC87366, initialization levels 2 and 3 don't enable
30all temperature channels, because some of them share pins with each other,
31so they can't be used at the same time.
32
33
34Description
35-----------
36
37The National Semiconductor PC87360 Super I/O chip contains monitoring and
38PWM control circuitry for two fans. The PC87363 chip is similar, and the
39PC87364 chip has monitoring and PWM control for a third fan.
40
41The National Semiconductor PC87365 and PC87366 Super I/O chips are complete
42hardware monitoring chipsets, not only controlling and monitoring three fans,
43but also monitoring eleven voltage inputs and two (PC87365) or up to four
44(PC87366) temperatures.
45
46 Chip #vin #fan #pwm #temp devid
47
48 PC87360 - 2 2 - 0xE1
49 PC87363 - 2 2 - 0xE8
50 PC87364 - 3 3 - 0xE4
51 PC87365 11 3 3 2 0xE5
52 PC87366 11 3 3 3-4 0xE9
53
54The driver assumes that no more than one chip is present, and one of the
55standard Super I/O addresses is used (0x2E/0x2F or 0x4E/0x4F)
56
57Fan Monitoring
58--------------
59
60Fan rotation speeds are reported in RPM (revolutions per minute). An alarm
61is triggered if the rotation speed has dropped below a programmable limit.
62A different alarm is triggered if the fan speed is too low to be measured.
63
64Fan readings are affected by a programmable clock divider, giving the
65readings more range or accuracy. Usually, users have to learn how it works,
66but this driver implements dynamic clock divider selection, so you don't
67have to care no more.
68
69For reference, here are a few values about clock dividers:
70
71 slowest accuracy highest
72 measurable around 3000 accurate
73 divider speed (RPM) RPM (RPM) speed (RPM)
74 1 1882 18 6928
75 2 941 37 4898
76 4 470 74 3464
77 8 235 150 2449
78
79For the curious, here is how the values above were computed:
80 * slowest measurable speed: clock/(255*divider)
81 * accuracy around 3000 RPM: 3000^2/clock
82 * highest accurate speed: sqrt(clock*100)
83The clock speed for the PC87360 family is 480 kHz. I arbitrarily chose 100
84RPM as the lowest acceptable accuracy.
85
86As mentioned above, you don't have to care about this no more.
87
88Note that not all RPM values can be represented, even when the best clock
89divider is selected. This is not only true for the measured speeds, but
90also for the programmable low limits, so don't be surprised if you try to
91set, say, fan1_min to 2900 and it finally reads 2909.
92
93
94Fan Control
95-----------
96
97PWM (pulse width modulation) values range from 0 to 255, with 0 meaning
98that the fan is stopped, and 255 meaning that the fan goes at full speed.
99
100Be extremely careful when changing PWM values. Low PWM values, even
101non-zero, can stop the fan, which may cause irreversible damage to your
102hardware if temperature increases too much. When changing PWM values, go
103step by step and keep an eye on temperatures.
104
105One user reported problems with PWM. Changing PWM values would break fan
106speed readings. No explanation nor fix could be found.
107
108
109Temperature Monitoring
110----------------------
111
112Temperatures are reported in degrees Celsius. Each temperature measured has
113associated low, high and overtemperature limits, each of which triggers an
114alarm when crossed.
115
116The first two temperature channels are external. The third one (PC87366
117only) is internal.
118
119The PC87366 has three additional temperature channels, based on
120thermistors (as opposed to thermal diodes for the first three temperature
121channels). For technical reasons, these channels are held by the VLM
122(voltage level monitor) logical device, not the TMS (temperature
123measurement) one. As a consequence, these temperatures are exported as
124voltages, and converted into temperatures in user-space.
125
126Note that these three additional channels share their pins with the
127external thermal diode channels, so you (physically) can't use them all at
128the same time. Although it should be possible to mix the two sensor types,
129the documents from National Semiconductor suggest that motherboard
130manufacturers should choose one type and stick to it. So you will more
131likely have either channels 1 to 3 (thermal diodes) or 3 to 6 (internal
132thermal diode, and thermistors).
133
134
135Voltage Monitoring
136------------------
137
138Voltages are reported relatively to a reference voltage, either internal or
139external. Some of them (in7:Vsb, in8:Vdd and in10:AVdd) are divided by two
140internally, you will have to compensate in sensors.conf. Others (in0 to in6)
141are likely to be divided externally. The meaning of each of these inputs as
142well as the values of the resistors used for division is left to the
143motherboard manufacturers, so you will have to document yourself and edit
144sensors.conf accordingly. National Semiconductor has a document with
145recommended resistor values for some voltages, but this still leaves much
146room for per motherboard specificities, unfortunately. Even worse,
147motherboard manufacturers don't seem to care about National Semiconductor's
148recommendations.
149
150Each voltage measured has associated low and high limits, each of which
151triggers an alarm when crossed.
152
153When available, VID inputs are used to provide the nominal CPU Core voltage.
154The driver will default to VRM 9.0, but this can be changed from user-space.
155The chipsets can handle two sets of VID inputs (on dual-CPU systems), but
156the driver will only export one for now. This may change later if there is
157a need.
158
159
160General Remarks
161---------------
162
163If an alarm triggers, it will remain triggered until the hardware register
164is read at least once. This means that the cause for the alarm may already
165have disappeared! Note that all hardware registers are read whenever any
166data is read (unless it is less than 2 seconds since the last update, in
167which case cached values are returned instead). As a consequence, when
168a once-only alarm triggers, it may take 2 seconds for it to show, and 2
169more seconds for it to disappear.
170
171Monitoring of in9 isn't enabled at lower init levels (<3) because that
172channel measures the battery voltage (Vbat). It is a known fact that
173repeatedly sampling the battery voltage reduces its lifetime. National
174Semiconductor smartly designed their chipset so that in9 is sampled only
175once every 1024 sampling cycles (that is every 34 minutes at the default
176sampling rate), so the effect is attenuated, but still present.
177
178
179Limitations
180-----------
181
182The datasheets suggests that some values (fan mins, fan dividers)
183shouldn't be changed once the monitoring has started, but we ignore that
184recommendation. We'll reconsider if it actually causes trouble.