]>
Commit | Line | Data |
---|---|---|
7f15b664 RM |
1 | Kernel driver pc87360 |
2 | ===================== | |
3 | ||
4 | Supported chips: | |
a9fc881b | 5 | |
7f15b664 | 6 | * National Semiconductor PC87360, PC87363, PC87364, PC87365 and PC87366 |
a9fc881b | 7 | |
7f15b664 | 8 | Prefixes: 'pc87360', 'pc87363', 'pc87364', 'pc87365', 'pc87366' |
a9fc881b | 9 | |
7f15b664 | 10 | Addresses scanned: none, address read from Super I/O config space |
a9fc881b | 11 | |
6aa693b8 | 12 | Datasheets: No longer available |
7f15b664 | 13 | |
7c81c60f | 14 | Authors: Jean Delvare <jdelvare@suse.de> |
7f15b664 RM |
15 | |
16 | Thanks to Sandeep Mehta, Tonko de Rooy and Daniel Ceregatti for testing. | |
a9fc881b | 17 | |
7f15b664 RM |
18 | Thanks to Rudolf Marek for helping me investigate conversion issues. |
19 | ||
20 | ||
21 | Module Parameters | |
22 | ----------------- | |
23 | ||
24 | * init int | |
a9fc881b MCC |
25 | Chip initialization level: |
26 | ||
27 | - 0: None | |
28 | - **1**: Forcibly enable internal voltage and temperature channels, | |
29 | except in9 | |
30 | - 2: Forcibly enable all voltage and temperature channels, except in9 | |
31 | - 3: Forcibly enable all voltage and temperature channels, including in9 | |
7f15b664 RM |
32 | |
33 | Note that this parameter has no effect for the PC87360, PC87363 and PC87364 | |
34 | chips. | |
35 | ||
36 | Also note that for the PC87366, initialization levels 2 and 3 don't enable | |
37 | all temperature channels, because some of them share pins with each other, | |
38 | so they can't be used at the same time. | |
39 | ||
40 | ||
41 | Description | |
42 | ----------- | |
43 | ||
44 | The National Semiconductor PC87360 Super I/O chip contains monitoring and | |
45 | PWM control circuitry for two fans. The PC87363 chip is similar, and the | |
46 | PC87364 chip has monitoring and PWM control for a third fan. | |
47 | ||
48 | The National Semiconductor PC87365 and PC87366 Super I/O chips are complete | |
49 | hardware monitoring chipsets, not only controlling and monitoring three fans, | |
50 | but also monitoring eleven voltage inputs and two (PC87365) or up to four | |
51 | (PC87366) temperatures. | |
52 | ||
a9fc881b | 53 | =========== ======= ======= ======= ======= ===== |
7f15b664 | 54 | Chip #vin #fan #pwm #temp devid |
a9fc881b | 55 | =========== ======= ======= ======= ======= ===== |
7f15b664 RM |
56 | PC87360 - 2 2 - 0xE1 |
57 | PC87363 - 2 2 - 0xE8 | |
58 | PC87364 - 3 3 - 0xE4 | |
59 | PC87365 11 3 3 2 0xE5 | |
60 | PC87366 11 3 3 3-4 0xE9 | |
a9fc881b | 61 | =========== ======= ======= ======= ======= ===== |
7f15b664 RM |
62 | |
63 | The driver assumes that no more than one chip is present, and one of the | |
64 | standard Super I/O addresses is used (0x2E/0x2F or 0x4E/0x4F) | |
65 | ||
66 | Fan Monitoring | |
67 | -------------- | |
68 | ||
69 | Fan rotation speeds are reported in RPM (revolutions per minute). An alarm | |
70 | is triggered if the rotation speed has dropped below a programmable limit. | |
71 | A different alarm is triggered if the fan speed is too low to be measured. | |
72 | ||
73 | Fan readings are affected by a programmable clock divider, giving the | |
74 | readings more range or accuracy. Usually, users have to learn how it works, | |
75 | but this driver implements dynamic clock divider selection, so you don't | |
76 | have to care no more. | |
77 | ||
78 | For reference, here are a few values about clock dividers: | |
79 | ||
a9fc881b MCC |
80 | =========== =============== =============== =========== |
81 | slowest accuracy highest | |
82 | measurable around 3000 accurate | |
7f15b664 | 83 | divider speed (RPM) RPM (RPM) speed (RPM) |
a9fc881b MCC |
84 | =========== =============== =============== =========== |
85 | 1 1882 18 6928 | |
86 | 2 941 37 4898 | |
87 | 4 470 74 3464 | |
88 | 8 235 150 2449 | |
89 | =========== =============== =============== =========== | |
7f15b664 RM |
90 | |
91 | For the curious, here is how the values above were computed: | |
a9fc881b | 92 | |
7f15b664 RM |
93 | * slowest measurable speed: clock/(255*divider) |
94 | * accuracy around 3000 RPM: 3000^2/clock | |
95 | * highest accurate speed: sqrt(clock*100) | |
a9fc881b | 96 | |
7f15b664 RM |
97 | The clock speed for the PC87360 family is 480 kHz. I arbitrarily chose 100 |
98 | RPM as the lowest acceptable accuracy. | |
99 | ||
100 | As mentioned above, you don't have to care about this no more. | |
101 | ||
102 | Note that not all RPM values can be represented, even when the best clock | |
103 | divider is selected. This is not only true for the measured speeds, but | |
104 | also for the programmable low limits, so don't be surprised if you try to | |
105 | set, say, fan1_min to 2900 and it finally reads 2909. | |
106 | ||
107 | ||
108 | Fan Control | |
109 | ----------- | |
110 | ||
111 | PWM (pulse width modulation) values range from 0 to 255, with 0 meaning | |
112 | that the fan is stopped, and 255 meaning that the fan goes at full speed. | |
113 | ||
114 | Be extremely careful when changing PWM values. Low PWM values, even | |
115 | non-zero, can stop the fan, which may cause irreversible damage to your | |
116 | hardware if temperature increases too much. When changing PWM values, go | |
117 | step by step and keep an eye on temperatures. | |
118 | ||
119 | One user reported problems with PWM. Changing PWM values would break fan | |
120 | speed readings. No explanation nor fix could be found. | |
121 | ||
122 | ||
123 | Temperature Monitoring | |
124 | ---------------------- | |
125 | ||
126 | Temperatures are reported in degrees Celsius. Each temperature measured has | |
127 | associated low, high and overtemperature limits, each of which triggers an | |
128 | alarm when crossed. | |
129 | ||
130 | The first two temperature channels are external. The third one (PC87366 | |
131 | only) is internal. | |
132 | ||
133 | The PC87366 has three additional temperature channels, based on | |
134 | thermistors (as opposed to thermal diodes for the first three temperature | |
135 | channels). For technical reasons, these channels are held by the VLM | |
136 | (voltage level monitor) logical device, not the TMS (temperature | |
137 | measurement) one. As a consequence, these temperatures are exported as | |
138 | voltages, and converted into temperatures in user-space. | |
139 | ||
140 | Note that these three additional channels share their pins with the | |
141 | external thermal diode channels, so you (physically) can't use them all at | |
142 | the same time. Although it should be possible to mix the two sensor types, | |
143 | the documents from National Semiconductor suggest that motherboard | |
144 | manufacturers should choose one type and stick to it. So you will more | |
145 | likely have either channels 1 to 3 (thermal diodes) or 3 to 6 (internal | |
146 | thermal diode, and thermistors). | |
147 | ||
148 | ||
149 | Voltage Monitoring | |
150 | ------------------ | |
151 | ||
152 | Voltages are reported relatively to a reference voltage, either internal or | |
153 | external. Some of them (in7:Vsb, in8:Vdd and in10:AVdd) are divided by two | |
154 | internally, you will have to compensate in sensors.conf. Others (in0 to in6) | |
155 | are likely to be divided externally. The meaning of each of these inputs as | |
156 | well as the values of the resistors used for division is left to the | |
157 | motherboard manufacturers, so you will have to document yourself and edit | |
158 | sensors.conf accordingly. National Semiconductor has a document with | |
159 | recommended resistor values for some voltages, but this still leaves much | |
160 | room for per motherboard specificities, unfortunately. Even worse, | |
161 | motherboard manufacturers don't seem to care about National Semiconductor's | |
162 | recommendations. | |
163 | ||
164 | Each voltage measured has associated low and high limits, each of which | |
165 | triggers an alarm when crossed. | |
166 | ||
167 | When available, VID inputs are used to provide the nominal CPU Core voltage. | |
168 | The driver will default to VRM 9.0, but this can be changed from user-space. | |
169 | The chipsets can handle two sets of VID inputs (on dual-CPU systems), but | |
170 | the driver will only export one for now. This may change later if there is | |
171 | a need. | |
172 | ||
173 | ||
174 | General Remarks | |
175 | --------------- | |
176 | ||
177 | If an alarm triggers, it will remain triggered until the hardware register | |
178 | is read at least once. This means that the cause for the alarm may already | |
179 | have disappeared! Note that all hardware registers are read whenever any | |
180 | data is read (unless it is less than 2 seconds since the last update, in | |
181 | which case cached values are returned instead). As a consequence, when | |
182 | a once-only alarm triggers, it may take 2 seconds for it to show, and 2 | |
183 | more seconds for it to disappear. | |
184 | ||
185 | Monitoring of in9 isn't enabled at lower init levels (<3) because that | |
186 | channel measures the battery voltage (Vbat). It is a known fact that | |
187 | repeatedly sampling the battery voltage reduces its lifetime. National | |
188 | Semiconductor smartly designed their chipset so that in9 is sampled only | |
189 | once every 1024 sampling cycles (that is every 34 minutes at the default | |
190 | sampling rate), so the effect is attenuated, but still present. | |
191 | ||
192 | ||
193 | Limitations | |
194 | ----------- | |
195 | ||
196 | The datasheets suggests that some values (fan mins, fan dividers) | |
197 | shouldn't be changed once the monitoring has started, but we ignore that | |
198 | recommendation. We'll reconsider if it actually causes trouble. |