CPUFreq addon in HAL

[Dave Jones]

On Tue, Jul 11, 2006 at 03:29:13PM +0200, Holger Macht wrote:

 > I thought about this possibility, too, but came to the conclusion that it
 > doesn't make sense. There main reasons are:
 > 
 >   1. There are several dependencies between the different CPUs. The
 >      cpufreq interfaces below /sys/devices/system/cpu/ lists all available
 >      CPUs, each in one directory. However, you cannot assume that you can
 >      simply control each CPU through it's single interface. Theres a file
 >      affected_cpus in each directory listing the CPUs which are controlled
 >      through this interface. So maybe it could look like:
 > 
 > 	cpu0/cpufreq/
 > 		--> controls cpu0, cpu1
 > 	cpu1/cpufreq/
 > 		--> controls cpu0, cpu1
 > 		...
 > 
 >      or:
 > 	cpu0/cpufreq/
 > 		--> controls cpu0, cpu1, cpu2, cpu3
 > 		...
 >      
 >      or:
 > 	cpu0/cpufreq/
 > 		--> only controls cpu0
 > 	cpu1/cpufreq/
 > 		--> only controls cpu1
 > 		...
 > 		
 >      I really don't want to export these dependencies. This would make
 >      things unnecessarily compicated for higher level applications.

The notion of exposing CPUs to the app is perhaps misguided.
Maybe it would be better to expose 'sockets' to the app.
(There may still be valid use cases for exposing individual cores
 too, so maybe this should be in addition)

 >   2. I can't imagine a use case were one might want to have different
 >      settings for different CPUs.

I've been wondering the same thing recently, wondering why we bothered
making the governors per-cpu.  For eg, it's nonsense to have ondemand on
one core, and 'powersave' as the governor for the second core.

So far I've not heard any valid reasons for supporting this oddball usecase.

 > > Comments on the patch
 > > 
 > > > +int get_cpu_count(void)
 > > 
 > > Maybe use sysconf(3) here?
 > 
 > sysconf doesn't provide this information.

x86info has been using nrCPUs = sysconf(_SC_NPROCESSORS_CONF);
for several years, and I don't recall getting a single bug report
about that getting it wrong.

		Dave

-- 
http://www.codemonkey.org.uk