On 11/07/2012 08:49 PM, Lv Zheng wrote: > ACPICA defines the following initialization steps that must be called to > make ACPICA functional: > 1. acpi_initialize_tables > 2. acpi_initialize_subsystem > 3. acpi_load_tables > 4. acpi_enable_subsystem > 5. acpi_initialize_objects > 6. OSPM specific bus scanning > 7. acpi_update_all_gpes > This patch allows developers to check the execution time of such bootup > steps through /sys/kernel/debug/acpi/acpica_init_time. > > The ACPICA 20121018 release measurement result is as follows: > ======================================== > 0.000000 acpi_initialize_tables > 0.000034 acpi_initialize_subsystem > 0.004822 acpi_load_tables > 0.003351 acpi_enable_subsystem > 0.003204 acpi_initialize_objects > 0.045082 acpi_bus_scan > 0.000023 acpi_bus_scan_fixed > 0.000017 acpi_update_all_gpes > ======================================== > The left column is the execution time in timeval (sec.usec) format. > The right column indicates the name of the initialization step. As you've identified that 80% of the ACPI startup time goes to acpi_bus_scan(), have you identified any ways to make acpi_bus_scan() faster? Do these numbers vary greatly when run on different machines? thanks, -Len Brown, Intel Open Source Technology Center -- To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html