summaryrefslogtreecommitdiffstats
path: root/drivers/powercap
diff options
context:
space:
mode:
authorLukasz Luba <lukasz.luba@arm.com>2022-07-07 08:15:52 +0100
committerRafael J. Wysocki <rafael.j.wysocki@intel.com>2022-07-15 19:17:30 +0200
commitae6ccaa650380d243cf43d31c864c5ced2fd4612 (patch)
tree31a1316e5a8e779db5d4697fa05bb43853fed24a /drivers/powercap
parent32346491ddf24599decca06190ebca03ff9de7f8 (diff)
downloadlinux-stable-ae6ccaa650380d243cf43d31c864c5ced2fd4612.tar.gz
linux-stable-ae6ccaa650380d243cf43d31c864c5ced2fd4612.tar.bz2
linux-stable-ae6ccaa650380d243cf43d31c864c5ced2fd4612.zip
PM: EM: convert power field to micro-Watts precision and align drivers
The milli-Watts precision causes rounding errors while calculating efficiency cost for each OPP. This is especially visible in the 'simple' Energy Model (EM), where the power for each OPP is provided from OPP framework. This can cause some OPPs to be marked inefficient, while using micro-Watts precision that might not happen. Update all EM users which access 'power' field and assume the value is in milli-Watts. Solve also an issue with potential overflow in calculation of energy estimation on 32bit machine. It's needed now since the power value (thus the 'cost' as well) are higher. Example calculation which shows the rounding error and impact: power = 'dyn-power-coeff' * volt_mV * volt_mV * freq_MHz power_a_uW = (100 * 600mW * 600mW * 500MHz) / 10^6 = 18000 power_a_mW = (100 * 600mW * 600mW * 500MHz) / 10^9 = 18 power_b_uW = (100 * 605mW * 605mW * 600MHz) / 10^6 = 21961 power_b_mW = (100 * 605mW * 605mW * 600MHz) / 10^9 = 21 max_freq = 2000MHz cost_a_mW = 18 * 2000MHz/500MHz = 72 cost_a_uW = 18000 * 2000MHz/500MHz = 72000 cost_b_mW = 21 * 2000MHz/600MHz = 70 // <- artificially better cost_b_uW = 21961 * 2000MHz/600MHz = 73203 The 'cost_b_mW' (which is based on old milli-Watts) is misleadingly better that the 'cost_b_uW' (this patch uses micro-Watts) and such would have impact on the 'inefficient OPPs' information in the Cpufreq framework. This patch set removes the rounding issue. Signed-off-by: Lukasz Luba <lukasz.luba@arm.com> Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Diffstat (limited to 'drivers/powercap')
-rw-r--r--drivers/powercap/dtpm_cpu.c5
1 files changed, 2 insertions, 3 deletions
diff --git a/drivers/powercap/dtpm_cpu.c b/drivers/powercap/dtpm_cpu.c
index f5eced0842b3..61c5ff80bd30 100644
--- a/drivers/powercap/dtpm_cpu.c
+++ b/drivers/powercap/dtpm_cpu.c
@@ -53,7 +53,7 @@ static u64 set_pd_power_limit(struct dtpm *dtpm, u64 power_limit)
for (i = 0; i < pd->nr_perf_states; i++) {
- power = pd->table[i].power * MICROWATT_PER_MILLIWATT * nr_cpus;
+ power = pd->table[i].power * nr_cpus;
if (power > power_limit)
break;
@@ -63,8 +63,7 @@ static u64 set_pd_power_limit(struct dtpm *dtpm, u64 power_limit)
freq_qos_update_request(&dtpm_cpu->qos_req, freq);
- power_limit = pd->table[i - 1].power *
- MICROWATT_PER_MILLIWATT * nr_cpus;
+ power_limit = pd->table[i - 1].power * nr_cpus;
return power_limit;
}