intel_pstate: Clarify average performance computation

[cascardo/linux.git] / drivers / cpufreq / intel_pstate.c

diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c
index 36953b5..19712e2 100644 (file)
--- a/drivers/cpufreq/intel_pstate.c
+++ b/drivers/cpufreq/intel_pstate.c
@@ -49,6 +49,9 @@
 #define int_tofp(X) ((int64_t)(X) << FRAC_BITS)
 #define fp_toint(X) ((X) >> FRAC_BITS)
 
+#define EXT_BITS 6
+#define EXT_FRAC_BITS (EXT_BITS + FRAC_BITS)
+
 static inline int32_t mul_fp(int32_t x, int32_t y)
 {
        return ((int64_t)x * (int64_t)y) >> FRAC_BITS;
@@ -70,12 +73,22 @@ static inline int ceiling_fp(int32_t x)
        return ret;
 }
 
+static inline u64 mul_ext_fp(u64 x, u64 y)
+{
+       return (x * y) >> EXT_FRAC_BITS;
+}
+
+static inline u64 div_ext_fp(u64 x, u64 y)
+{
+       return div64_u64(x << EXT_FRAC_BITS, y);
+}
+
 /**
  * struct sample -     Store performance sample
- * @core_pct_busy:     Ratio of APERF/MPERF in percent, which is actual
+ * @core_avg_perf:     Ratio of APERF/MPERF which is the actual average
  *                     performance during last sample period
  * @busy_scaled:       Scaled busy value which is used to calculate next
- *                     P state. This can be different than core_pct_busy
+ *                     P state. This can be different than core_avg_perf
  *                     to account for cpu idle period
  * @aperf:             Difference of actual performance frequency clock count
  *                     read from APERF MSR between last and current sample
@@ -90,7 +103,7 @@ static inline int ceiling_fp(int32_t x)
  * data for choosing next P State.
  */
 struct sample {
-       int32_t core_pct_busy;
+       int32_t core_avg_perf;
        int32_t busy_scaled;
        u64 aperf;
        u64 mperf;
@@ -1152,15 +1165,11 @@ static void intel_pstate_get_cpu_pstates(struct cpudata *cpu)
        intel_pstate_set_min_pstate(cpu);
 }
 
-static inline void intel_pstate_calc_busy(struct cpudata *cpu)
+static inline void intel_pstate_calc_avg_perf(struct cpudata *cpu)
 {
        struct sample *sample = &cpu->sample;
-       int64_t core_pct;
-
-       core_pct = sample->aperf * int_tofp(100);
-       core_pct = div64_u64(core_pct, sample->mperf);
 
-       sample->core_pct_busy = (int32_t)core_pct;
+       sample->core_avg_perf = div_ext_fp(sample->aperf, sample->mperf);
 }
 
 static inline bool intel_pstate_sample(struct cpudata *cpu, u64 time)
@@ -1203,9 +1212,8 @@ static inline bool intel_pstate_sample(struct cpudata *cpu, u64 time)
 
 static inline int32_t get_avg_frequency(struct cpudata *cpu)
 {
-       return fp_toint(mul_fp(cpu->sample.core_pct_busy,
-                              int_tofp(cpu->pstate.max_pstate_physical *
-                                               cpu->pstate.scaling / 100)));
+       return mul_ext_fp(cpu->sample.core_avg_perf,
+                         cpu->pstate.max_pstate_physical * cpu->pstate.scaling);
 }
 
 static inline int32_t get_avg_pstate(struct cpudata *cpu)
@@ -1265,10 +1273,10 @@ static inline int32_t get_target_pstate_use_performance(struct cpudata *cpu)
         * period. The result will be a percentage of busy at a
         * specified pstate.
         */
-       core_busy = cpu->sample.core_pct_busy;
        max_pstate = cpu->pstate.max_pstate_physical;
        current_pstate = cpu->pstate.current_pstate;
-       core_busy = mul_fp(core_busy, div_fp(max_pstate, current_pstate));
+       core_busy = mul_ext_fp(cpu->sample.core_avg_perf,
+                              div_fp(100 * max_pstate, current_pstate));
 
        /*
         * Since our utilization update callback will not run unless we are
@@ -1317,7 +1325,7 @@ static inline void intel_pstate_adjust_busy_pstate(struct cpudata *cpu)
        intel_pstate_update_pstate(cpu, target_pstate);
 
        sample = &cpu->sample;
-       trace_pstate_sample(fp_toint(sample->core_pct_busy),
+       trace_pstate_sample(mul_ext_fp(100, sample->core_avg_perf),
                fp_toint(sample->busy_scaled),
                from,
                cpu->pstate.current_pstate,
@@ -1337,7 +1345,7 @@ static void intel_pstate_update_util(struct update_util_data *data, u64 time,
                bool sample_taken = intel_pstate_sample(cpu, time);
 
                if (sample_taken) {
-                       intel_pstate_calc_busy(cpu);
+                       intel_pstate_calc_avg_perf(cpu);
                        if (!hwp_active)
                                intel_pstate_adjust_busy_pstate(cpu);
                }