Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 15d5f839 authored by Dmitriy Zavin's avatar Dmitriy Zavin Committed by Andi Kleen
Browse files

[PATCH] x86: Refactor thermal throttle processing 



Refactor the event processing (syslog messaging and rate limiting)
into separate file therm_throt.c. This allows consistent reporting
of CPU thermal throttle events.

After ACK'ing the interrupt, if the event is current, the user
(p4.c/mce_intel.c) calls therm_throt_process to log (and rate limit)
the event. If that function returns 1, the user has the option to log
things further (such as to mce_log in x86_64).

AK: minor cleanup

Signed-off-by: default avatarDmitriy Zavin <dmitriyz@google.com>
Signed-off-by: default avatarAndi Kleen <ak@suse.de>
parent 3b171672

arch/i386/kernel/cpu/mcheck/Makefile

+1 −1
Original line number Original line Diff line number Diff line 
obj-y	=	mce.o k7.o p4.o p5.o p6.o winchip.o

obj-y	=	mce.o k7.o p4.o p5.o p6.o winchip.o therm_throt.o

obj-$(CONFIG_X86_MCE_NONFATAL)	+=	non-fatal.o
 
obj-$(CONFIG_X86_MCE_NONFATAL)	+=	non-fatal.o

arch/i386/kernel/cpu/mcheck/p4.c

+6 −17
Original line number Original line Diff line number Diff line
@@ -13,6 +13,8 @@ 
#include <asm/msr.h>

#include <asm/msr.h>

#include <asm/apic.h>

#include <asm/apic.h>





#include <asm/therm_throt.h>



#include "mce.h"

#include "mce.h"





/* as supported by the P4/Xeon family */

/* as supported by the P4/Xeon family */
@@ -44,25 +46,12 @@ static void unexpected_thermal_interrupt(struct pt_regs *regs) 
/* P4/Xeon Thermal transition interrupt handler */

/* P4/Xeon Thermal transition interrupt handler */

static void intel_thermal_interrupt(struct pt_regs *regs)

static void intel_thermal_interrupt(struct pt_regs *regs)

{

{

	u32 l, h;

	__u64 msr_val;

	unsigned int cpu = smp_processor_id();

	static unsigned long next[NR_CPUS];





	ack_APIC_irq();

	ack_APIC_irq();





	if (time_after(next[cpu], jiffies))

	rdmsrl(MSR_IA32_THERM_STATUS, msr_val);

		return;

	therm_throt_process(msr_val & 0x1);



	next[cpu] = jiffies + HZ*5;

	rdmsr(MSR_IA32_THERM_STATUS, l, h);

	if (l & 0x1) {

		printk(KERN_EMERG "CPU%d: Temperature above threshold\n", cpu);

		printk(KERN_EMERG "CPU%d: Running in modulated clock mode\n",

				cpu);

		add_taint(TAINT_MACHINE_CHECK);

	} else {

		printk(KERN_INFO "CPU%d: Temperature/speed normal\n", cpu);

	}

}

}





/* Thermal interrupt handler for this CPU setup */

/* Thermal interrupt handler for this CPU setup */

arch/i386/kernel/cpu/mcheck/therm_throt.c

0 → 100644
+58 −0
Original line number Original line Diff line number Diff line 
/*

 * linux/arch/i386/kerne/cpu/mcheck/therm_throt.c

 *

 * Thermal throttle event support code.

 *

 * Author: Dmitriy Zavin (dmitriyz@google.com)

 *

 * Credits: Adapted from Zwane Mwaikambo's original code in mce_intel.c.

 *

 */



#include <linux/percpu.h>

#include <linux/cpu.h>

#include <asm/cpu.h>

#include <linux/notifier.h>

#include <asm/therm_throt.h>



/* How long to wait between reporting thermal events */

#define CHECK_INTERVAL              (300 * HZ)



static DEFINE_PER_CPU(unsigned long, next_check);



/***

 * therm_throt_process - Process thermal throttling event

 * @curr: Whether the condition is current or not (boolean), since the

 *        thermal interrupt normally gets called both when the thermal

 *        event begins and once the event has ended.

 *

 * This function is normally called by the thermal interrupt after the

 * IRQ has been acknowledged.

 *

 * It will take care of rate limiting and printing messages to the syslog.

 *

 * Returns: 0 : Event should NOT be further logged, i.e. still in

 *              "timeout" from previous log message.

 *          1 : Event should be logged further, and a message has been

 *              printed to the syslog.

 */

int therm_throt_process(int curr)

{

	unsigned int cpu = smp_processor_id();



	if (time_before(jiffies, __get_cpu_var(next_check)))

		return 0;



	__get_cpu_var(next_check) = jiffies + CHECK_INTERVAL;



	/* if we just entered the thermal event */

	if (curr) {

		printk(KERN_CRIT "CPU%d: Temperature above threshold, "

		       "cpu clock throttled\n", cpu);

		add_taint(TAINT_MACHINE_CHECK);

	} else {

		printk(KERN_CRIT "CPU%d: Temperature/speed normal\n", cpu);

	}



	return 1;

}

arch/x86_64/kernel/Makefile

+2 −2
Original line number Original line Diff line number Diff line
@@ -11,7 +11,7 @@ obj-y := process.o signal.o entry.o traps.o irq.o \ 
		pci-dma.o pci-nommu.o alternative.o

		pci-dma.o pci-nommu.o alternative.o





obj-$(CONFIG_STACKTRACE)	+= stacktrace.o

obj-$(CONFIG_STACKTRACE)	+= stacktrace.o

obj-$(CONFIG_X86_MCE)         += mce.o

obj-$(CONFIG_X86_MCE)		+= mce.o therm_throt.o

obj-$(CONFIG_X86_MCE_INTEL)	+= mce_intel.o

obj-$(CONFIG_X86_MCE_INTEL)	+= mce_intel.o

obj-$(CONFIG_X86_MCE_AMD)	+= mce_amd.o

obj-$(CONFIG_X86_MCE_AMD)	+= mce_amd.o

obj-$(CONFIG_MTRR)		+= ../../i386/kernel/cpu/mtrr/

obj-$(CONFIG_MTRR)		+= ../../i386/kernel/cpu/mtrr/
@@ -46,6 +46,7 @@ obj-y += intel_cacheinfo.o 




CFLAGS_vsyscall.o		:= $(PROFILING) -g0

CFLAGS_vsyscall.o		:= $(PROFILING) -g0





therm_throt-y                   += ../../i386/kernel/cpu/mcheck/therm_throt.o

bootflag-y			+= ../../i386/kernel/bootflag.o

bootflag-y			+= ../../i386/kernel/bootflag.o

cpuid-$(subst m,y,$(CONFIG_X86_CPUID))  += ../../i386/kernel/cpuid.o

cpuid-$(subst m,y,$(CONFIG_X86_CPUID))  += ../../i386/kernel/cpuid.o

topology-y                     += ../../i386/kernel/topology.o

topology-y                     += ../../i386/kernel/topology.o
@@ -55,4 +56,3 @@ quirks-y += ../../i386/kernel/quirks.o 
i8237-y				+= ../../i386/kernel/i8237.o

i8237-y				+= ../../i386/kernel/i8237.o

msr-$(subst m,y,$(CONFIG_X86_MSR))  += ../../i386/kernel/msr.o

msr-$(subst m,y,$(CONFIG_X86_MSR))  += ../../i386/kernel/msr.o

alternative-y			+= ../../i386/kernel/alternative.o
 
alternative-y			+= ../../i386/kernel/alternative.o

arch/x86_64/kernel/mce.c

+27 −0
Original line number Original line Diff line number Diff line
@@ -274,6 +274,33 @@ void do_machine_check(struct pt_regs * regs, long error_code) 
	atomic_dec(&mce_entry);

	atomic_dec(&mce_entry);

}

}





#ifdef CONFIG_X86_MCE_INTEL

/***

 * mce_log_therm_throt_event - Logs the thermal throttling event to mcelog

 * @cpu: The CPU on which the event occured.

 * @status: Event status information

 *

 * This function should be called by the thermal interrupt after the

 * event has been processed and the decision was made to log the event

 * further.

 *

 * The status parameter will be saved to the 'status' field of 'struct mce'

 * and historically has been the register value of the

 * MSR_IA32_THERMAL_STATUS (Intel) msr.

 */

void mce_log_therm_throt_event(unsigned int cpu, __u64 status)

{

	struct mce m;



	memset(&m, 0, sizeof(m));

	m.cpu = cpu;

	m.bank = MCE_THERMAL_BANK;

	m.status = status;

	rdtscll(m.tsc);

	mce_log(&m);

}

#endif /* CONFIG_X86_MCE_INTEL */



/*

/*

 * Periodic polling timer for "silent" machine check errors.

 * Periodic polling timer for "silent" machine check errors.

 */

 */