Hi, On Fri, Jul 11, 2014 at 9:30 PM, Nuno Magalhães <nunomagalhaes@xxxxxxxxx> wrote: [...] > As far as fans go i have: > * a Nox Urano 500 PSU, but it's not being monitored (i think) > * the CPU cooler (Zalman CNPS7000C-AlCu) connected to CPU_FAN (4pin) > * a 120mm fan (Noctua NF-S12A ULN) connected to CHA_FAN1 (3pin) > * a 120mm fan (Noctua NF-S12A FLX) on PWR_FAN (3pin) > * nothing connected on CHA_FAN2 (3pin) at the moment > > The BIOS has these settings: > * Q-Fan controller: enabled (otherwise all fans run at full speed) > > * Vcore voltage:1.20V > * 3.3V voltage: 3.26V > * 5V voltage: 4.98V > * 12V voltage: 11.85V > > * CPU temp: 39 > * M/B temp: 43 > * CPU fan speed: 1670 (warn at 400) > * Chassis Fan1 speed: 560 > * Power fan speed: 1196 > > There is no Chassis Fan12 in the BIOS, yet if i connect one, it'll > work, seemlingly controlled by Q-Fan. [...] > > atk0110-acpi-0 > Adapter: ACPI interface > Vcore Voltage: +1.11 V (min = +0.85 V, max = +1.60 V) > +3.3 Voltage: +3.17 V (min = +3.00 V, max = +3.60 V) > +5.0 Voltage: +5.01 V (min = +4.50 V, max = +5.50 V) > +12.0 Voltage: +11.86 V (min = +11.20 V, max = +13.20 V) > CPU FAN Speed: 1618 RPM (min = 0 RPM, max = 1800 RPM) > CHASSIS FAN Speed: 565 RPM (min = 0 RPM, max = 1800 RPM) > CPU Temperature: +38.0°C (high = +90.0°C, crit = +125.0°C) > MB Temperature: +44.0°C (high = +45.0°C, crit = +90.0°C) > > radeon-pci-0100 > Adapter: PCI adapter > temp1: +73.5°C > > atk0110 also doesn't detect the second fan. I've read somewhere > (forgot where) that ATK0110 should detect everything automagically and > loading extra modules should be avoided because it's the "old > guesswork method", is this true? Sort of. atk0100 uses the same interface that the BIOS uses to show you the reading, and the assumption is that the manufacturer configured all the sensors correctly; the interface is enumerable and items are self describing so no guess work is needed (with native drivers you basically have to poke "known" addresses and see if the thing that replies looks like something that you can handle). The downside is that you cannot tweak the driver: it just reports whatever was configured in the firmware and that's it. > [ 10.683889] ACPI: resource nForce2_smbus [io 0x4c00-0x4c3f] > conflicts with ACPI region SM00 [io 0x4c00-0x4c05] This means that there's a smbus controller that is used by ACPI so it's not safe to use it directly (it87). You can load the native driver driver if you want to get access to all the readings, but: - ACPI EC might still touch the chip - likewise SMM might fiddle with it without the OS even knowing Access to hwmon chips is usually banked, which means that you first select a "bank" and the read/write from a register in that bank; overriding acpi_enforce_resources mean that you are allowing 2 different drivers to touch the same hardware without any synchronization: - driver1 selects bank B1 - driver2 selects bank B2 - driver1 reads/writes register R1 ---> something bad might happen because the access went to the wrong bank Luca _______________________________________________ lm-sensors mailing list lm-sensors@xxxxxxxxxxxxxx http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
