k315rc7dirk2_r1: CPU Frequency Verses load experiment: Powersave mode;
Note: This version includes Dirk's patches for Kernel 3.16RC1
Note that for this experiment "perf record" data was acquired also.
Overview:
Detail A - Start of Experiment:
Detail B - End of Experiment:
. The issues evident on the graph are the subject of another web page, however there is additional concern for systems with HZ = 250 as the timer call tends to run at 0.02 seconds instead of 0.01 seconds, leading to a different response curve and less buffer on the 3X delay for "bla bla bla"
Load / sleep frequency sweep test - 85% average load with load / sleep frequency sweeping from 2 hertz to 250 hertz:
Other data:
The time difference between timer calls. (warning very large file, slow server)
The time difference between timer calls. (warning large file, slow server)
The time difference histogram file. (I haven't actually made the histogram graph yet.)
References:
Dirks's (3.15RC7) intel_pstate.c 2. (proposed 3.16RC1)
Script used to ramp cpu load.
Script used to ramp cpu sleep / load frequency.
Program used to create load ramp script.
Program used to create sleep / load frequency script.
Consume Program.
CPU frequency monitoring program (CPU 7).