ANDROID: FIXUP: sched/tune: fix payoff calculation for boost region

The definition of the acceptance regions as well as the translation of
these regions into a payoff value was both wrong which turned out in:
a) a wrong definition of payoff for the performance boost region
b) a correct "by chance" definition of the payoff for the performance
   constraint region (i.e. two sign errors together fixing the formula)

This patch provides a better description of the cut regions as well as
a fixed version of the payoff computations, which are now reduced to a
single formula usable for both cases.

Reported-by: Leo Yan <leo.yan@linaro.org>
Reviewed-by: Leo Yan <leo.yan@linaro.org>
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Signed-off-by: Patrick Bellasi <patrick.bellasi@arm.com>
Signed-off-by: Andres Oportus <andresoportus@google.com>
This commit is contained in:
Patrick Bellasi 2016-07-28 17:38:25 +01:00 committed by Dmitry Shmidt
parent e71c425516
commit 2ed513e7b5

View file

@ -51,50 +51,51 @@ __schedtune_accept_deltas(int nrg_delta, int cap_delta,
int perf_boost_idx, int perf_constrain_idx)
{
int payoff = -INT_MAX;
int gain_idx = -1;
/* Performance Boost (B) region */
if (nrg_delta > 0 && cap_delta > 0) {
/*
* Evaluate "Performance Boost" vs "Energy Increase"
* payoff criteria:
* cap_delta / nrg_delta < cap_gain / nrg_gain
* which is:
* nrg_delta * cap_gain > cap_delta * nrg_gain
*/
payoff = nrg_delta * threshold_gains[perf_boost_idx].cap_gain;
payoff -= cap_delta * threshold_gains[perf_boost_idx].nrg_gain;
trace_sched_tune_filter(
nrg_delta, cap_delta,
threshold_gains[perf_boost_idx].nrg_gain,
threshold_gains[perf_boost_idx].cap_gain,
payoff, 8);
return payoff;
}
if (nrg_delta >= 0 && cap_delta > 0)
gain_idx = perf_boost_idx;
/* Performance Constraint (C) region */
if (nrg_delta < 0 && cap_delta < 0) {
/*
* Evaluate "Performance Boost" vs "Energy Increase"
* payoff criteria:
* cap_delta / nrg_delta > cap_gain / nrg_gain
* which is:
* cap_delta * nrg_gain > nrg_delta * cap_gain
*/
payoff = cap_delta * threshold_gains[perf_constrain_idx].nrg_gain;
payoff -= nrg_delta * threshold_gains[perf_constrain_idx].cap_gain;
trace_sched_tune_filter(
nrg_delta, cap_delta,
threshold_gains[perf_constrain_idx].nrg_gain,
threshold_gains[perf_constrain_idx].cap_gain,
payoff, 6);
return payoff;
}
else if (nrg_delta < 0 && cap_delta <= 0)
gain_idx = perf_constrain_idx;
/* Default: reject schedule candidate */
if (gain_idx == -1)
return payoff;
/*
* Evaluate "Performance Boost" vs "Energy Increase"
*
* - Performance Boost (B) region
*
* Condition: nrg_delta > 0 && cap_delta > 0
* Payoff criteria:
* cap_gain / nrg_gain < cap_delta / nrg_delta =
* cap_gain * nrg_delta < cap_delta * nrg_gain
* Note that since both nrg_gain and nrg_delta are positive, the
* inequality does not change. Thus:
*
* payoff = (cap_delta * nrg_gain) - (cap_gain * nrg_delta)
*
* - Performance Constraint (C) region
*
* Condition: nrg_delta < 0 && cap_delta < 0
* payoff criteria:
* cap_gain / nrg_gain > cap_delta / nrg_delta =
* cap_gain * nrg_delta < cap_delta * nrg_gain
* Note that since nrg_gain > 0 while nrg_delta < 0, the
* inequality change. Thus:
*
* payoff = (cap_delta * nrg_gain) - (cap_gain * nrg_delta)
*
* This means that, in case of same positive defined {cap,nrg}_gain
* for both the B and C regions, we can use the same payoff formula
* where a positive value represents the accept condition.
*/
payoff = cap_delta * threshold_gains[gain_idx].nrg_gain;
payoff -= nrg_delta * threshold_gains[gain_idx].cap_gain;
return payoff;
}