15  * struct em_perf_state - Performance state of a performance domain
16  * @performance:	CPU performance (capacity) at a given frequency
25 	unsigned long performance;  member
35  * EM_PERF_STATE_INEFFICIENT: The performance state is inefficient. There is
36  * in this em_perf_domain, another performance state with a higher frequency
43  * struct em_perf_table - Performance states table
46  * @state:	List of performance states, in ascending order
55  * struct em_perf_domain - Performance domain
57  * @nr_perf_states:	Number of performance states
58  * @min_perf_state:	Minimum allowed Performance State index
59  * @max_perf_state:	Maximum allowed Performance State index
62  *			for performance reasons to avoid potential cache
66  * In case of CPU device, a "performance domain" represents a group of CPUs
67  * whose performance is scaled together. All CPUs of a performance domain
68  * must have the same micro-architecture. Performance domains often have
122 	 * active_power() - Provide power at the next performance state of
125 	 * @power	: Active power at the performance state
127 	 * @freq	: Frequency at the performance state in kHz
130 	 * active_power() must find the lowest performance state of 'dev' above
144 	 * get_cost() - Provide the cost at the given performance state of
147 	 * @freq	: Frequency at the performance state in kHz
148 	 * @cost	: The cost value for the performance state
185  * em_pd_get_efficient_state() - Get an efficient performance state from the EM
186  * @table:		List of performance states, in ascending order
187  * @pd:			performance domain for which this must be done
193  * Return: An efficient performance state id, high enough to meet @max_util
208 		if (ps->performance >= max_util) {  in em_pd_get_efficient_state()
221  *		performance domain
222  * @pd		: performance domain for which energy has to be estimated
251 	 * In order to predict the performance state, map the utilization of  in em_cpu_energy()
252 	 * the most utilized CPU of the performance domain to a requested  in em_cpu_energy()
253 	 * performance, like schedutil. Take also into account that the real  in em_cpu_energy()
254 	 * performance might be set lower (due to thermal capping). Thus, clamp  in em_cpu_energy()
256 	 * effective performance.  in em_cpu_energy()
261 	 * Find the lowest performance state of the Energy Model above the  in em_cpu_energy()
262 	 * requested performance.  in em_cpu_energy()
269 	 * The performance (capacity) of a CPU in the domain at the performance  in em_cpu_energy()
273 	 *   ps->performance = --------------------                  (1)  in em_cpu_energy()
277 	 * the EM), the energy consumed by this CPU at that performance state  in em_cpu_energy()
282 	 *               ps->performance  in em_cpu_energy()
284 	 * since 'cpu_util / ps->performance' represents its percentage of busy  in em_cpu_energy()
313  * em_pd_nr_perf_states() - Get the number of performance states of a perf.
315  * @pd		: performance domain for which this must be done
317  * Return: the number of performance states in the performance domain table
325  * em_perf_state_from_pd() - Get the performance states table of perf.
327  * @pd		: performance domain for which this must be done
330  * of the performance states table is finished, the rcu_read_unlock() should
333  * Return: the pointer to performance states table of the performance domain