On Tue, 1 May 2012 16:36:22 +0200, Luca Tettamanti wrote: > On Tue, May 1, 2012 at 4:07 PM, Ian Pilcher <arequipeno@xxxxxxxxx> wrote: > > On 04/30/2012 09:55 AM, Jean Delvare wrote: > >> Most likely they would "work", because they simply do not check the > >> ACPI conflict in the first place. So they run but not reliably, exactly > >> the same as running Linux with acpi_enforce_resources=lax. > > > > On a slightly related note, I've always been a bit bemused by the this > > message: > > > > If an ACPI driver is available for this device, you should use it > > instead of the native driver > > > > What is an "ACPI driver"? > > It means a driver that uses the interface exposed by ACPI. > > > Is there any hardware for which ACPI actually provides the same level of > > information that "legacy" access does? Every system I've dealt with > > over the last few years provides a single temperature reading via ACPI > > (generally CPU surface temperature) -- no other temperatures, no fans, > > no voltages. > > A single TZ is common, yes. Asus motherboards provide the so called > ATK0110 interface for reading temps, fans and voltages. Provided, actually, it seems they are moving away from it. There are other examples, including acpi_power_meter and many drivers under drivers/platform/x86 which do expose some hardware monitoring features based on proprietary BIOS interfaces, sometimes using ACPI, sometimes not. That being said, I can't disagree with Ian: in most cases, when you see the warning message, there is no APCI driver you can use, and this is a real problem looking for a solution. The ball is in the vendor's court though. -- Jean Delvare _______________________________________________ lm-sensors mailing list lm-sensors@xxxxxxxxxxxxxx http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
