On Wed, Jun 30, 2010 at 03:34:14PM -0400, Ted Unangst wrote: > On Wed, Jun 30, 2010 at 3:23 PM, Luis Henriques <[email protected]> > wrote: > > Probably, a silly question, but here it goes: > > > > With this patch, I will not be able to set the perflevel to, say, 50% and > > keep the system using that performance level forever. Is this correct? > > I guess that with current apmd we are able to do this. > > > > If both of these two statements are true (maybe they are not!), we should > > also have a mechanism to disable this code (either at runtime or compile > > time). > > You are correct, but I wonder why you would ever want a machine only > running at 50% all the time. When it is busy, it's still slow, and > when it's not, it's still using power.
In my X31 it scales between 600 and 1600 MHz, or something like that. And when I keep it around setperf 40 it avoids turning the fan on. Which is handy when for example you're compiling something and leave it by bed ;) I didn't say it was a very good reason ;) > We are thinking that states > other than 0 and 100 are not very useful. But this is not the final > diff. It will probably have some means of control eventually, until > then, it'd be nice if people evaluated if this meets their needs. -- viq [demime 1.01d removed an attachment of type application/pgp-signature]
