[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: Dmitriy Zavin <dmitriyz@google.com>
Signed-off-by: Andi Kleen <ak@suse.de>

3 files changed, 35 insertions, 23 deletions

diff --git a/arch/x86_64/kernel/Makefile b/arch/x86_64/kernel/Makefile
index 2466fbd035ee..3c7cbff04d3d 100644
--- a/arch/x86_64/kernel/Makefile
+++ b/arch/x86_64/kernel/Makefile
@@ -11,7 +11,7 @@ obj-y	:= process.o signal.o entry.o traps.o irq.o \
 		pci-dma.o pci-nommu.o alternative.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-$(CONFIG_X86_MCE_AMD)	+= mce_amd.o
 obj-$(CONFIG_MTRR)		+= ../../i386/kernel/cpu/mtrr/
@@ -46,6 +46,7 @@ obj-y				+= intel_cacheinfo.o
 
 CFLAGS_vsyscall.o		:= $(PROFILING) -g0
 
+therm_throt-y                   += ../../i386/kernel/cpu/mcheck/therm_throt.o
 bootflag-y			+= ../../i386/kernel/bootflag.o
 cpuid-$(subst m,y,$(CONFIG_X86_CPUID))  += ../../i386/kernel/cpuid.o
 topology-y                     += ../../i386/kernel/topology.o
@@ -55,4 +56,3 @@ quirks-y			+= ../../i386/kernel/quirks.o
 i8237-y				+= ../../i386/kernel/i8237.o
 msr-$(subst m,y,$(CONFIG_X86_MSR))  += ../../i386/kernel/msr.o
 alternative-y			+= ../../i386/kernel/alternative.o
-
diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c
index 1a93c3738404..bbea88801d88 100644
--- a/arch/x86_64/kernel/mce.c
+++ b/arch/x86_64/kernel/mce.c
@@ -274,6 +274,33 @@ void do_machine_check(struct pt_regs * regs, long error_code)
 	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.
  */
diff --git a/arch/x86_64/kernel/mce_intel.c b/arch/x86_64/kernel/mce_intel.c
index 8f533d2c40cb..dec11219e278 100644
--- a/arch/x86_64/kernel/mce_intel.c
+++ b/arch/x86_64/kernel/mce_intel.c
@@ -11,36 +11,21 @@
 #include <asm/mce.h>
 #include <asm/hw_irq.h>
 #include <asm/idle.h>
-
-static DEFINE_PER_CPU(unsigned long, next_check);
+#include <asm/therm_throt.h>
 
 asmlinkage void smp_thermal_interrupt(void)
 {
-	struct mce m;
+	__u64 msr_val;
 
 	ack_APIC_irq();
 
 	exit_idle();
 	irq_enter();
-	if (time_before(jiffies, __get_cpu_var(next_check)))
-		goto done;
-
-	__get_cpu_var(next_check) = jiffies + HZ*300;
-	memset(&m, 0, sizeof(m));
-	m.cpu = smp_processor_id();
-	m.bank = MCE_THERMAL_BANK;
-	rdtscll(m.tsc);
-	rdmsrl(MSR_IA32_THERM_STATUS, m.status);
-	if (m.status & 0x1) {
-		printk(KERN_EMERG
-			"CPU%d: Temperature above threshold, cpu clock throttled\n", m.cpu);
-		add_taint(TAINT_MACHINE_CHECK);
-	} else {
-		printk(KERN_EMERG "CPU%d: Temperature/speed normal\n", m.cpu);
-	}
 
-	mce_log(&m);
-done:
+	rdmsrl(MSR_IA32_THERM_STATUS, msr_val);
+	if (therm_throt_process(msr_val & 1))
+		mce_log_therm_throt_event(smp_processor_id(), msr_val);
+
 	irq_exit();
 }