Lines Matching defs:performance

95  * struct sample -	Store performance sample
97  *			performance during last sample period
101  * @aperf:		Difference of actual performance frequency clock count
103  * @mperf:		Difference of maximum performance frequency clock count
109  * This structure is used in the cpudata structure to store performance sample
130  * @scaling:		Scaling factor between performance and frequency
210  * @epp_powersave:	Last saved HWP energy performance preference
211  *			(EPP) or energy performance bias (EPB),
212  *			when policy switched to performance
214  * @epp_default:	Power on default HWP energy performance
216  * @epp_cached:		Cached HWP energy-performance preference value
222  * @hwp_boost_min:	Last HWP boosted min performance
560  * intel_pstate_hybrid_hwp_adjust - Calibrate HWP performance levels.
563  * On hybrid processors, HWP may expose more performance levels than there are
565  * scaling factor between HWP performance levels and CPU frequency will be less
602 	 * the effective range of HWP performance levels.
652  *	1		performance
668 	[EPP_INDEX_PERFORMANCE] = "performance",
707 		 *	0x04-0x07	:	Balance performance
764 		 * from 0 (performance) if the current policy is "performance",
942 	 * performance level and otherwise leave the value set by the caller.
1094 		 * During initialization, CPU performance at full capacity needs
1118 		pr_info("Found no CPUs with nonzero maximum performance\n");
1224 	 * The maximum performance of the CPU may have changed, but assume
1225 	 * that the performance of the other CPUs has not changed.
1332 		 * In case the EPP has been set to "performance" by the
1333 		 * active mode "performance" scaling algorithm, replace that
1339 		 * However, make sure that EPP will be set to "performance" when
1340 		 * the CPU is brought back online again and the "performance"
1988 	/* Enable HWP notification interrupt for performance change */
2013 	 * - But less performance oriented than performance EPP
2556 	 * possible performance oscillations and offset possible performance
2840 	 * abstract values to represent performance rather than pure ratios.
2982 	 * If the CPU is an SMT thread and it goes offline with the performance
2984 	 * from getting to lower performance levels, so force the minimum
2985 	 * performance on CPU offline to prevent that from happening.
3070 	 * the default cpufreq governor is neither powersave nor performance.
3110  * performance field, and lablelled as such on the graphs; also known as
3114  * (or performance or core_avg_perf) have a fixed y-axis from 0 to 100%,
3702 #define HWP_SET_EPP_VALUES(powersave, balance_power, balance_perf, performance) \
3706 	 FIELD_PREP_CONST(PERFORMANCE_MASK, performance))
3878 			pr_debug("Updated EPPs powersave:%x balanced power:%x balanced perf:%x performance:%x\n",