Carl-Daniel Hailfinger wrote:
On 03.05.2007 19:17, Stefan Reinauer wrote:
- Carl-Daniel Hailfinger c-d.hailfinger.devel.2006@gmx.net [070503 19:16]:
The concept of code not under the control of the operating system is frightening some low-latency audio/video folks. If the operating system uses some operations heavily which trigger SMM execution, extended periods of switched off interrupts may result.
Yes, anyone doing real time systems has a problem with this.
OTOH, systems which need SMM should probably not be used anyway if latency is critical.
They can not. This is the reason a lot of people are using LinuxBIOS rather than legacy bios.
How do we deal with fan control, then? Most recent mainboards use SMM for fan control. Once you forbid SMM, you have to trust the OS to perform fan control correctly, which will fail if the machine crashes or locks up.
Regards, Carl-Daniel
In the legacy/APM world power management and fan control were mostly done in SMM.
In the ACPI world that has been moved to the OS. Proper drivers and applications and SMM all play a part but SMM's role is greatly reduced and latency much less of an issue. You might also want to look at the OpenIPMI project to see how some of this is being done now. http://openipmi.sourceforge.net/IPMI.pdf
Marc