FROMLIST: tracing: Add support for preempt and irq enable/disable events

Preempt and irq trace events can be used for tracing the start and
end of an atomic section which can be used by a trace viewer like
systrace to graphically view the start and end of an atomic section and
correlate them with latencies and scheduling issues.

This also serves as a prelude to using synthetic events or probes to
rewrite the preempt and irqsoff tracers, along with numerous benefits of
using trace events features for these events.

Change-Id: I718d40f7c3c48579adf9d7121b21495a669c89bd
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zilstra <peterz@infradead.org>
Cc: kernel-team@android.com
Link: https://patchwork.kernel.org/patch/9988157/
Signed-off-by: Joel Fernandes <joelaf@google.com>

Author: Joel Fernandes

include/linux/ftrace.h

@@ -702,7 +702,8 @@ static inline void __ftrace_enabled_restore(int enabled)
   static inline void time_hardirqs_off(unsigned long a0, unsigned long a1) { }
 #endif
 
-#ifdef CONFIG_PREEMPT_TRACER
+#if defined(CONFIG_PREEMPT_TRACER) || \
+	(defined(CONFIG_DEBUG_PREEMPT) && defined(CONFIG_PREEMPTIRQ_EVENTS))
   extern void trace_preempt_on(unsigned long a0, unsigned long a1);
   extern void trace_preempt_off(unsigned long a0, unsigned long a1);
 #else

include/trace/events/preemptirq.h

@@ -0,0 +1,70 @@
+#ifdef CONFIG_PREEMPTIRQ_EVENTS
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM preemptirq
+
+#if !defined(_TRACE_PREEMPTIRQ_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_PREEMPTIRQ_H
+
+#include <linux/ktime.h>
+#include <linux/tracepoint.h>
+#include <linux/string.h>
+#include <asm/sections.h>
+
+DECLARE_EVENT_CLASS(preemptirq_template,
+
+	TP_PROTO(unsigned long ip, unsigned long parent_ip),
+
+	TP_ARGS(ip, parent_ip),
+
+	TP_STRUCT__entry(
+		__field(u32, caller_offs)
+		__field(u32, parent_offs)
+	),
+
+	TP_fast_assign(
+		__entry->caller_offs = (u32)(ip - (unsigned long)_stext);
+		__entry->parent_offs = (u32)(parent_ip - (unsigned long)_stext);
+	),
+
+	TP_printk("caller=%pF parent=%pF",
+		  (void *)((unsigned long)(_stext) + __entry->caller_offs),
+		  (void *)((unsigned long)(_stext) + __entry->parent_offs))
+);
+
+#ifndef CONFIG_PROVE_LOCKING
+DEFINE_EVENT(preemptirq_template, irq_disable,
+	     TP_PROTO(unsigned long ip, unsigned long parent_ip),
+	     TP_ARGS(ip, parent_ip));
+
+DEFINE_EVENT(preemptirq_template, irq_enable,
+	     TP_PROTO(unsigned long ip, unsigned long parent_ip),
+	     TP_ARGS(ip, parent_ip));
+#endif
+
+#ifdef CONFIG_DEBUG_PREEMPT
+DEFINE_EVENT(preemptirq_template, preempt_disable,
+	     TP_PROTO(unsigned long ip, unsigned long parent_ip),
+	     TP_ARGS(ip, parent_ip));
+
+DEFINE_EVENT(preemptirq_template, preempt_enable,
+	     TP_PROTO(unsigned long ip, unsigned long parent_ip),
+	     TP_ARGS(ip, parent_ip));
+#endif
+
+#endif /* _TRACE_PREEMPTIRQ_H */
+
+#include <trace/define_trace.h>
+
+#else /* !CONFIG_PREEMPTIRQ_EVENTS */
+
+#define trace_irq_enable(...)
+#define trace_irq_disable(...)
+#define trace_preempt_enable(...)
+#define trace_preempt_disable(...)
+#define trace_irq_enable_rcuidle(...)
+#define trace_irq_disable_rcuidle(...)
+#define trace_preempt_enable_rcuidle(...)
+#define trace_preempt_disable_rcuidle(...)
+
+#endif

kernel/trace/Kconfig

@@ -165,6 +165,17 @@ config FUNCTION_GRAPH_TRACER
 	  address on the current task structure into a stack of calls.
 
 
+config PREEMPTIRQ_EVENTS
+	bool "Enable trace events for preempt and irq disable/enable"
+	select TRACE_IRQFLAGS
+	depends on DEBUG_PREEMPT || !PROVE_LOCKING
+	default n
+	help
+	  Enable tracing of disable and enable events for preemption and irqs.
+	  For tracing preempt disable/enable events, DEBUG_PREEMPT must be
+	  enabled. For tracing irq disable/enable events, PROVE_LOCKING must
+	  be disabled.
+
 config IRQSOFF_TRACER
 	bool "Interrupts-off Latency Tracer"
 	default n

kernel/trace/Makefile

@@ -37,6 +37,7 @@ obj-$(CONFIG_TRACING) += trace_stat.o
 obj-$(CONFIG_TRACING) += trace_printk.o
 obj-$(CONFIG_CONTEXT_SWITCH_TRACER) += trace_sched_switch.o
 obj-$(CONFIG_FUNCTION_TRACER) += trace_functions.o
+obj-$(CONFIG_PREEMPTIRQ_EVENTS) += trace_irqsoff.o
 obj-$(CONFIG_IRQSOFF_TRACER) += trace_irqsoff.o
 obj-$(CONFIG_PREEMPT_TRACER) += trace_irqsoff.o
 obj-$(CONFIG_SCHED_TRACER) += trace_sched_wakeup.o

kernel/trace/trace_irqsoff.c

@@ -16,6 +16,9 @@
 
 #include "trace.h"
 
+#define CREATE_TRACE_POINTS
+#include <trace/events/preemptirq.h>
+
 #if defined(CONFIG_IRQSOFF_TRACER) || defined(CONFIG_PREEMPT_TRACER)
 static struct trace_array		*irqsoff_trace __read_mostly;
 static int				tracer_enabled __read_mostly;
@@ -765,27 +768,54 @@ static inline void tracer_preempt_on(unsigned long a0, unsigned long a1) { }
 static inline void tracer_preempt_off(unsigned long a0, unsigned long a1) { }
 #endif
 
+/* Per-cpu variable to prevent redundant calls when IRQs already off */
+static DEFINE_PER_CPU(int, tracing_irq_cpu);
+
 #if defined(CONFIG_TRACE_IRQFLAGS) && !defined(CONFIG_PROVE_LOCKING)
 void trace_hardirqs_on(void)
 {
+	if (!this_cpu_read(tracing_irq_cpu))
+		return;
+
+	trace_irq_enable_rcuidle(CALLER_ADDR0, CALLER_ADDR1);
 	tracer_hardirqs_on();
+
+	this_cpu_write(tracing_irq_cpu, 0);
 }
 EXPORT_SYMBOL(trace_hardirqs_on);
 
 void trace_hardirqs_off(void)
 {
+	if (this_cpu_read(tracing_irq_cpu))
+		return;
+
+	this_cpu_write(tracing_irq_cpu, 1);
+
+	trace_irq_disable_rcuidle(CALLER_ADDR0, CALLER_ADDR1);
 	tracer_hardirqs_off();
 }
 EXPORT_SYMBOL(trace_hardirqs_off);
 
 __visible void trace_hardirqs_on_caller(unsigned long caller_addr)
 {
+	if (!this_cpu_read(tracing_irq_cpu))
+		return;
+
+	trace_irq_enable_rcuidle(CALLER_ADDR0, caller_addr);
 	tracer_hardirqs_on_caller(caller_addr);
+
+	this_cpu_write(tracing_irq_cpu, 0);
 }
 EXPORT_SYMBOL(trace_hardirqs_on_caller);
 
 __visible void trace_hardirqs_off_caller(unsigned long caller_addr)
 {
+	if (this_cpu_read(tracing_irq_cpu))
+		return;
+
+	this_cpu_write(tracing_irq_cpu, 1);
+
+	trace_irq_disable_rcuidle(CALLER_ADDR0, caller_addr);
 	tracer_hardirqs_off_caller(caller_addr);
 }
 EXPORT_SYMBOL(trace_hardirqs_off_caller);
@@ -807,14 +837,17 @@ inline void print_irqtrace_events(struct task_struct *curr)
 }
 #endif
 
-#ifdef CONFIG_PREEMPT_TRACER
+#if defined(CONFIG_PREEMPT_TRACER) || \
+	(defined(CONFIG_DEBUG_PREEMPT) && defined(CONFIG_PREEMPTIRQ_EVENTS))
 void trace_preempt_on(unsigned long a0, unsigned long a1)
 {
+	trace_preempt_enable_rcuidle(a0, a1);
 	tracer_preempt_on(a0, a1);
 }
 
 void trace_preempt_off(unsigned long a0, unsigned long a1)
 {
+	trace_preempt_disable_rcuidle(a0, a1);
 	tracer_preempt_off(a0, a1);
 }
 #endif