El Fri, 1 Jun 2012 19:44:26 +0200 Juan Francisco Cantero Hurtado <[email protected]> escribió:
> On Thu, May 31, 2012 at 11:21:20PM +0000, Stuart Henderson wrote: > > On 2012-05-31, Ted Unangst <[email protected]> wrote: > > > On Thu, May 31, 2012 at 08:28, Tomas Bodzar wrote: > > >> On Thu, May 31, 2012 at 3:45 AM, Juan Francisco Cantero Hurtado > > >> <[email protected]> wrote: > > > > > >>> This is important because when I open a web page with a lot of > > >>> javascript, the browser is very slow. Also when I compile > > >>> something with "make -j1", apmd doesn't raise the speed of my > > >>> CPU, I need use "make -j4" for raising the cpu speed to 2700Mhz. > > >> > > >> What shows top, vmstat, systat about %sys, %usr, %idle during > > >> that time? Because you can have 800MHz of CPU, but %usr and/or > > >> %sys can be eg. only 20% so there's no reason to switch to > > >> higher frequency. > > > > > > apmd scaling works like crap with more than one cpu. This was > > > fixed, then unfixed, but you can get the code from cvs if you > > > like. > > > > > > > > > > conflicts. if anyone wants it they could grab it from > > junkpile.org/apmd.diff > > Thanks for the patch! > > I've been using the version 1.51 of apmd.c since yesterday. I can feel > the difference with respect to the last version of apmd. > > Is it possible to add a new option to apmd for to select between the > new and the old code?. I mean, eg: -C for use the "new" code with > better SMP support and -O for "compatibility mode" with the old code > for people with problematic machines. > > Cheers. > Thanks for this patch. My thinkpad had the same problem as the reported one by Juan Francisco , but now it runs smoothly. I'd like to ask for more features like configurable thresholds for increase and reduce the cpu. Maybe with a couple of sysctls or with options for apmd. Is it easy? Thanks. Jes
