Merge branches 'tracing/ftrace' and 'tracing/syscalls'; commit 'v2.6.29-rc8' into tracing/core

arch/x86/Kconfig

@@ -35,6 +35,7 @@ config X86
 	select HAVE_FUNCTION_GRAPH_TRACER
 	select HAVE_FUNCTION_TRACE_MCOUNT_TEST
 	select HAVE_FTRACE_NMI_ENTER if DYNAMIC_FTRACE
+	select HAVE_FTRACE_SYSCALLS
 	select HAVE_KVM
 	select HAVE_ARCH_KGDB
 	select HAVE_ARCH_TRACEHOOK

arch/x86/include/asm/thread_info.h

@@ -94,6 +94,7 @@ struct thread_info {
 #define TIF_FORCED_TF		24	/* true if TF in eflags artificially */
 #define TIF_DEBUGCTLMSR		25	/* uses thread_struct.debugctlmsr */
 #define TIF_DS_AREA_MSR		26      /* uses thread_struct.ds_area_msr */
+#define TIF_SYSCALL_FTRACE	27	/* for ftrace syscall instrumentation */
 
 #define _TIF_SYSCALL_TRACE	(1 << TIF_SYSCALL_TRACE)
 #define _TIF_NOTIFY_RESUME	(1 << TIF_NOTIFY_RESUME)
@@ -115,15 +116,17 @@ struct thread_info {
 #define _TIF_FORCED_TF		(1 << TIF_FORCED_TF)
 #define _TIF_DEBUGCTLMSR	(1 << TIF_DEBUGCTLMSR)
 #define _TIF_DS_AREA_MSR	(1 << TIF_DS_AREA_MSR)
+#define _TIF_SYSCALL_FTRACE	(1 << TIF_SYSCALL_FTRACE)
 
 /* work to do in syscall_trace_enter() */
 #define _TIF_WORK_SYSCALL_ENTRY	\
-	(_TIF_SYSCALL_TRACE | _TIF_SYSCALL_EMU | \
+	(_TIF_SYSCALL_TRACE | _TIF_SYSCALL_EMU | _TIF_SYSCALL_FTRACE |	\
 	 _TIF_SYSCALL_AUDIT | _TIF_SECCOMP | _TIF_SINGLESTEP)
 
 /* work to do in syscall_trace_leave() */
 #define _TIF_WORK_SYSCALL_EXIT	\
-	(_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | _TIF_SINGLESTEP)
+	(_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | _TIF_SINGLESTEP |	\
+	 _TIF_SYSCALL_FTRACE)
 
 /* work to do on interrupt/exception return */
 #define _TIF_WORK_MASK							\
@@ -132,7 +135,7 @@ struct thread_info {
 	   _TIF_SINGLESTEP|_TIF_SECCOMP|_TIF_SYSCALL_EMU))
 
 /* work to do on any return to user space */
-#define _TIF_ALLWORK_MASK (0x0000FFFF & ~_TIF_SECCOMP)
+#define _TIF_ALLWORK_MASK ((0x0000FFFF & ~_TIF_SECCOMP) | _TIF_SYSCALL_FTRACE)
 
 /* Only used for 64 bit */
 #define _TIF_DO_NOTIFY_MASK						\

arch/x86/kernel/ptrace.c

@@ -21,6 +21,7 @@
 #include <linux/audit.h>
 #include <linux/seccomp.h>
 #include <linux/signal.h>
+#include <linux/ftrace.h>
 
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
@@ -1416,6 +1417,9 @@ asmregparm long syscall_trace_enter(struct pt_regs *regs)
 	    tracehook_report_syscall_entry(regs))
 		ret = -1L;
 
+	if (unlikely(test_thread_flag(TIF_SYSCALL_FTRACE)))
+		ftrace_syscall_enter(regs);
+
 	if (unlikely(current->audit_context)) {
 		if (IS_IA32)
 			audit_syscall_entry(AUDIT_ARCH_I386,
@@ -1439,6 +1443,9 @@ asmregparm void syscall_trace_leave(struct pt_regs *regs)
 	if (unlikely(current->audit_context))
 		audit_syscall_exit(AUDITSC_RESULT(regs->ax), regs->ax);
 
+	if (unlikely(test_thread_flag(TIF_SYSCALL_FTRACE)))
+		ftrace_syscall_exit(regs);
+
 	if (test_thread_flag(TIF_SYSCALL_TRACE))
 		tracehook_report_syscall_exit(regs, 0);
 

include/linux/ftrace.h

@@ -503,4 +503,25 @@ static inline void trace_hw_branch_oops(void) {}
 
 #endif /* CONFIG_HW_BRANCH_TRACER */
 
+/*
+ * A syscall entry in the ftrace syscalls array.
+ *
+ * @syscall_nr: syscall number
+ */
+struct syscall_trace_entry {
+	int		syscall_nr;
+};
+
+#ifdef CONFIG_FTRACE_SYSCALLS
+extern void start_ftrace_syscalls(void);
+extern void stop_ftrace_syscalls(void);
+extern void ftrace_syscall_enter(struct pt_regs *regs);
+extern void ftrace_syscall_exit(struct pt_regs *regs);
+#else
+static inline void start_ftrace_syscalls(void) { }
+static inline void stop_ftrace_syscalls(void) { }
+static inline void ftrace_syscall_enter(struct pt_regs *regs) { }
+static inline void ftrace_syscall_exit(struct pt_regs *regs) { }
+#endif
+
 #endif /* _LINUX_FTRACE_H */

kernel/trace/Kconfig

@@ -34,6 +34,9 @@ config HAVE_FTRACE_MCOUNT_RECORD
 config HAVE_HW_BRANCH_TRACER
 	bool
 
+config HAVE_FTRACE_SYSCALLS
+	bool
+
 config TRACER_MAX_TRACE
 	bool
 
@@ -175,6 +178,13 @@ config EVENT_TRACER
 	  allowing the user to pick and choose which trace point they
 	  want to trace.
 
+config FTRACE_SYSCALLS
+	bool "Trace syscalls"
+	depends on HAVE_FTRACE_SYSCALLS
+	select TRACING
+	help
+	  Basic tracer to catch the syscall entry and exit events.
+
 config BOOT_TRACER
 	bool "Trace boot initcalls"
 	select TRACING

kernel/trace/Makefile

@@ -43,5 +43,6 @@ obj-$(CONFIG_BLK_DEV_IO_TRACE)	+= blktrace.o
 obj-$(CONFIG_EVENT_TRACER) += trace_events.o
 obj-$(CONFIG_EVENT_TRACER) += events.o
 obj-$(CONFIG_EVENT_TRACER) += trace_export.o
+obj-$(CONFIG_FTRACE_SYSCALLS) += trace_syscalls.o
 
 libftrace-y := ftrace.o

kernel/trace/trace.h

@@ -31,6 +31,8 @@ enum trace_type {
 	TRACE_GRAPH_ENT,
 	TRACE_USER_STACK,
 	TRACE_HW_BRANCHES,
+	TRACE_SYSCALL_ENTER,
+	TRACE_SYSCALL_EXIT,
 	TRACE_KMEM_ALLOC,
 	TRACE_KMEM_FREE,
 	TRACE_POWER,

kernel/trace/trace_syscalls.c

@@ -0,0 +1,113 @@
+#include <linux/ftrace.h>
+#include <linux/kernel.h>
+
+#include <asm/syscall.h>
+
+#include "trace_output.h"
+#include "trace.h"
+
+static atomic_t refcount;
+
+void start_ftrace_syscalls(void)
+{
+	unsigned long flags;
+	struct task_struct *g, *t;
+
+	if (atomic_inc_return(&refcount) != 1)
+		goto out;
+
+	read_lock_irqsave(&tasklist_lock, flags);
+
+	do_each_thread(g, t) {
+		set_tsk_thread_flag(t, TIF_SYSCALL_FTRACE);
+	} while_each_thread(g, t);
+
+	read_unlock_irqrestore(&tasklist_lock, flags);
+out:
+	atomic_dec(&refcount);
+}
+
+void stop_ftrace_syscalls(void)
+{
+	unsigned long flags;
+	struct task_struct *g, *t;
+
+	if (atomic_dec_return(&refcount))
+		goto out;
+
+	read_lock_irqsave(&tasklist_lock, flags);
+
+	do_each_thread(g, t) {
+		clear_tsk_thread_flag(t, TIF_SYSCALL_FTRACE);
+	} while_each_thread(g, t);
+
+	read_unlock_irqrestore(&tasklist_lock, flags);
+out:
+	atomic_inc(&refcount);
+}
+
+void ftrace_syscall_enter(struct pt_regs *regs)
+{
+	int syscall_nr;
+
+	syscall_nr = syscall_get_nr(current, regs);
+
+	trace_printk("syscall %d enter\n", syscall_nr);
+}
+
+void ftrace_syscall_exit(struct pt_regs *regs)
+{
+	int syscall_nr;
+
+	syscall_nr = syscall_get_nr(current, regs);
+
+	trace_printk("syscall %d exit\n", syscall_nr);
+}
+
+static int init_syscall_tracer(struct trace_array *tr)
+{
+	start_ftrace_syscalls();
+
+	return 0;
+}
+
+static void reset_syscall_tracer(struct trace_array *tr)
+{
+	stop_ftrace_syscalls();
+}
+
+static struct trace_event syscall_enter_event = {
+	.type		= TRACE_SYSCALL_ENTER,
+};
+
+static struct trace_event syscall_exit_event = {
+	.type		= TRACE_SYSCALL_EXIT,
+};
+
+static struct tracer syscall_tracer __read_mostly = {
+	.name		= "syscall",
+	.init		= init_syscall_tracer,
+	.reset		= reset_syscall_tracer
+};
+
+__init int register_ftrace_syscalls(void)
+{
+	int ret;
+
+	ret = register_ftrace_event(&syscall_enter_event);
+	if (!ret) {
+		printk(KERN_WARNING "event %d failed to register\n",
+		       syscall_enter_event.type);
+		WARN_ON_ONCE(1);
+	}
+
+	ret = register_ftrace_event(&syscall_exit_event);
+	if (!ret) {
+		printk(KERN_WARNING "event %d failed to register\n",
+		       syscall_exit_event.type);
+		WARN_ON_ONCE(1);
+	}
+
+	return register_tracer(&syscall_tracer);
+}
+device_initcall(register_ftrace_syscalls);