mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2025-02-05 03:45:42 +07:00
7fe2f6399a
CPU power consumption vs performance tuning is no longer limited to CPU frequency switching anymore: deep sleep states, traditional dynamic frequency scaling and hidden turbo/boost frequencies are tied close together and depend on each other. The first two exist on different architectures like PPC, Itanium and ARM, the latter (so far) only on X86. On X86 the APU (CPU+GPU) will only run most efficiently if CPU and GPU has proper power management in place. Users and Developers want to have *one* tool to get an overview what their system supports and to monitor and debug CPU power management in detail. The tool should compile and work on as many architectures as possible. Once this tool stabilizes a bit, it is intended to replace the Intel-specific tools in tools/power/x86 Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
114 lines
3.0 KiB
C
114 lines
3.0 KiB
C
/*
|
|
* test module to check whether the TSC-based delay routine continues
|
|
* to work properly after cpufreq transitions. Needs ACPI to work
|
|
* properly.
|
|
*
|
|
* Based partly on the Power Management Timer (PMTMR) code to be found
|
|
* in arch/i386/kernel/timers/timer_pm.c on recent 2.6. kernels, especially
|
|
* code written by John Stultz. The read_pmtmr function was copied verbatim
|
|
* from that file.
|
|
*
|
|
* (C) 2004 Dominik Brodowski
|
|
*
|
|
* To use:
|
|
* 1.) pass clock=tsc to the kernel on your bootloader
|
|
* 2.) modprobe this module (it'll fail)
|
|
* 3.) change CPU frequency
|
|
* 4.) modprobe this module again
|
|
* 5.) if the third value, "diff_pmtmr", changes between 2. and 4., the
|
|
* TSC-based delay routine on the Linux kernel does not correctly
|
|
* handle the cpufreq transition. Please report this to
|
|
* cpufreq@vger.kernel.org
|
|
*/
|
|
|
|
#include <linux/kernel.h>
|
|
#include <linux/module.h>
|
|
#include <linux/init.h>
|
|
#include <linux/delay.h>
|
|
|
|
#include <asm/io.h>
|
|
|
|
#include <acpi/acpi_bus.h>
|
|
#include <acpi/acpi_drivers.h>
|
|
|
|
static int pm_tmr_ioport = 0;
|
|
|
|
/*helper function to safely read acpi pm timesource*/
|
|
static u32 read_pmtmr(void)
|
|
{
|
|
u32 v1=0,v2=0,v3=0;
|
|
/* It has been reported that because of various broken
|
|
* chipsets (ICH4, PIIX4 and PIIX4E) where the ACPI PM time
|
|
* source is not latched, so you must read it multiple
|
|
* times to insure a safe value is read.
|
|
*/
|
|
do {
|
|
v1 = inl(pm_tmr_ioport);
|
|
v2 = inl(pm_tmr_ioport);
|
|
v3 = inl(pm_tmr_ioport);
|
|
} while ((v1 > v2 && v1 < v3) || (v2 > v3 && v2 < v1)
|
|
|| (v3 > v1 && v3 < v2));
|
|
|
|
/* mask the output to 24 bits */
|
|
return (v2 & 0xFFFFFF);
|
|
}
|
|
|
|
static int __init cpufreq_test_tsc(void)
|
|
{
|
|
u32 now, then, diff;
|
|
u64 now_tsc, then_tsc, diff_tsc;
|
|
int i;
|
|
|
|
/* the following code snipped is copied from arch/x86/kernel/acpi/boot.c
|
|
of Linux v2.6.25. */
|
|
|
|
/* detect the location of the ACPI PM Timer */
|
|
if (acpi_gbl_FADT.header.revision >= FADT2_REVISION_ID) {
|
|
/* FADT rev. 2 */
|
|
if (acpi_gbl_FADT.xpm_timer_block.space_id !=
|
|
ACPI_ADR_SPACE_SYSTEM_IO)
|
|
return 0;
|
|
|
|
pm_tmr_ioport = acpi_gbl_FADT.xpm_timer_block.address;
|
|
/*
|
|
* "X" fields are optional extensions to the original V1.0
|
|
* fields, so we must selectively expand V1.0 fields if the
|
|
* corresponding X field is zero.
|
|
*/
|
|
if (!pm_tmr_ioport)
|
|
pm_tmr_ioport = acpi_gbl_FADT.pm_timer_block;
|
|
} else {
|
|
/* FADT rev. 1 */
|
|
pm_tmr_ioport = acpi_gbl_FADT.pm_timer_block;
|
|
}
|
|
|
|
printk(KERN_DEBUG "start--> \n");
|
|
then = read_pmtmr();
|
|
rdtscll(then_tsc);
|
|
for (i=0;i<20;i++) {
|
|
mdelay(100);
|
|
now = read_pmtmr();
|
|
rdtscll(now_tsc);
|
|
diff = (now - then) & 0xFFFFFF;
|
|
diff_tsc = now_tsc - then_tsc;
|
|
printk(KERN_DEBUG "t1: %08u t2: %08u diff_pmtmr: %08u diff_tsc: %016llu\n", then, now, diff, diff_tsc);
|
|
then = now;
|
|
then_tsc = now_tsc;
|
|
}
|
|
printk(KERN_DEBUG "<-- end \n");
|
|
return -ENODEV;
|
|
}
|
|
|
|
static void __exit cpufreq_none(void)
|
|
{
|
|
return;
|
|
}
|
|
|
|
module_init(cpufreq_test_tsc)
|
|
module_exit(cpufreq_none)
|
|
|
|
|
|
MODULE_AUTHOR("Dominik Brodowski");
|
|
MODULE_DESCRIPTION("Verify the TSC cpufreq notifier working correctly -- needs ACPI-enabled system");
|
|
MODULE_LICENSE ("GPL");
|