Re: [Cbe-oss-dev] powerpc/cell/cpufreq: add spu aware cpufreq governor

[Eric Blossom <eb@comsec.com>]

On Mon, Jul 07, 2008 at 05:02:30PM +0200, Arnd Bergmann wrote:
> From: Christian Krafft <krafft@de.ibm.com>
> 
> This patch adds a cpufreq governor that takes the number of running spus
> into account. It's very similar to the ondemand governor, but not as complex.
> Instead of hacking spu load into the ondemand governor it might be easier to
> have cpufreq accepting multiple governors per cpu in future.
> Don't know if this is the right way, but it would keep the governors simple.

I've got a basic question about this idea:

Does it throttle only idle spus or is there a single control for
the entire PPE + SPE complex?

I can think of many situations in our code where at certain times we
may only be able to keep N out of M spes occupied (because of how
we've expressed our task-specific parallelism), but we're counting on
the ones we're using running at full speed so that we can maintain our
real-time throughput.  (Our application does real-time signal
processing of continuously streaming data)

Is there a way to disable the "throttling SPEs" feature short of
configuring it out of the kernel?

Eric