From a871bd33a6c0bc86fb47cd02ea2650dd43d3d95f Mon Sep 17 00:00:00 2001 From: Jason Baron Date: Mon, 10 Aug 2009 16:52:31 -0400 Subject: tracing: Add syscall tracepoints add two tracepoints in syscall exit and entry path, conditioned on TIF_SYSCALL_FTRACE. Supports the syscall trace event code. Signed-off-by: Jason Baron Cc: Lai Jiangshan Cc: Steven Rostedt Cc: Peter Zijlstra Cc: Mathieu Desnoyers Cc: Jiaying Zhang Cc: Martin Bligh Cc: Li Zefan Cc: Masami Hiramatsu Signed-off-by: Frederic Weisbecker --- kernel/tracepoint.c | 38 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 38 insertions(+) (limited to 'kernel/tracepoint.c') diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index 1ef5d3a601c..070a42bb892 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -24,6 +24,7 @@ #include #include #include +#include extern struct tracepoint __start___tracepoints[]; extern struct tracepoint __stop___tracepoints[]; @@ -577,3 +578,40 @@ static int init_tracepoints(void) __initcall(init_tracepoints); #endif /* CONFIG_MODULES */ + +static DEFINE_MUTEX(regfunc_mutex); +static int sys_tracepoint_refcount; + +void syscall_regfunc(void) +{ + unsigned long flags; + struct task_struct *g, *t; + + mutex_lock(®func_mutex); + if (!sys_tracepoint_refcount) { + 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); + } + sys_tracepoint_refcount++; + mutex_unlock(®func_mutex); +} + +void syscall_unregfunc(void) +{ + unsigned long flags; + struct task_struct *g, *t; + + mutex_lock(®func_mutex); + sys_tracepoint_refcount--; + if (!sys_tracepoint_refcount) { + 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); + } + mutex_unlock(®func_mutex); +} -- cgit v1.2.3-70-g09d2 From 60d970c254b95ec7a0fc4c590b510253987b64a0 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 13 Aug 2009 23:37:26 +0200 Subject: tracing: Fix syscall tracing on !HAVE_FTRACE_SYSCALLS architectures The new syscall_regfunc()/unregfunc() functions rely on the existence of TIF_SYSCALL_FTRACE - but that TIF flag is only offered by HAVE_FTRACE_SYSCALLS. Cc: Frederic Weisbecker Cc: Jason Baron Cc: Steven Rostedt Cc: Peter Zijlstra LKML-Reference: Signed-off-by: Ingo Molnar --- kernel/tracepoint.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'kernel/tracepoint.c') diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index 070a42bb892..35dd27adb82 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -579,6 +579,8 @@ __initcall(init_tracepoints); #endif /* CONFIG_MODULES */ +#ifdef CONFIG_FTRACE_SYSCALLS + static DEFINE_MUTEX(regfunc_mutex); static int sys_tracepoint_refcount; @@ -615,3 +617,4 @@ void syscall_unregfunc(void) } mutex_unlock(®func_mutex); } +#endif -- cgit v1.2.3-70-g09d2 From d88cb582325830698de5071fa8b8c9e933dbbcad Mon Sep 17 00:00:00 2001 From: Anirban Sinha Date: Tue, 25 Aug 2009 07:00:02 -0700 Subject: tracing: Eliminate code duplication in kernel/tracepoint.c Signed-off-by: Anirban Sinha Reviewed-by: Li Zefan Cc: "Oleg Nesterov" LKML-Reference: Signed-off-by: Ingo Molnar --- kernel/tracepoint.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'kernel/tracepoint.c') diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index 35dd27adb82..06f165a4408 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -555,9 +555,6 @@ int tracepoint_module_notify(struct notifier_block *self, switch (val) { case MODULE_STATE_COMING: - tracepoint_update_probe_range(mod->tracepoints, - mod->tracepoints + mod->num_tracepoints); - break; case MODULE_STATE_GOING: tracepoint_update_probe_range(mod->tracepoints, mod->tracepoints + mod->num_tracepoints); -- cgit v1.2.3-70-g09d2 From 667000011927b4fcc359beac4a2447889db6d349 Mon Sep 17 00:00:00 2001 From: Josh Stone Date: Mon, 24 Aug 2009 14:43:11 -0700 Subject: tracing: Rename FTRACE_SYSCALLS for tracepoints s/HAVE_FTRACE_SYSCALLS/HAVE_SYSCALL_TRACEPOINTS/g s/TIF_SYSCALL_FTRACE/TIF_SYSCALL_TRACEPOINT/g The syscall enter/exit tracing is no longer specific to just ftrace, so they now have names that reflect their tie to tracepoints instead. Signed-off-by: Josh Stone Cc: Jason Baron Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Li Zefan Cc: Steven Rostedt Cc: Peter Zijlstra Cc: Mathieu Desnoyers Cc: Jiaying Zhang Cc: Martin Bligh Cc: Lai Jiangshan Cc: Paul Mundt Cc: Martin Schwidefsky Cc: Heiko Carstens LKML-Reference: <1251150194-1713-2-git-send-email-jistone@redhat.com> Signed-off-by: Frederic Weisbecker --- arch/s390/Kconfig | 2 +- arch/s390/defconfig | 2 +- arch/s390/include/asm/thread_info.h | 4 ++-- arch/s390/kernel/entry.S | 2 +- arch/s390/kernel/entry64.S | 2 +- arch/s390/kernel/ptrace.c | 4 ++-- arch/x86/Kconfig | 2 +- arch/x86/configs/i386_defconfig | 2 +- arch/x86/configs/x86_64_defconfig | 2 +- arch/x86/include/asm/thread_info.h | 13 +++++++------ arch/x86/kernel/ptrace.c | 4 ++-- kernel/trace/Kconfig | 4 ++-- kernel/tracepoint.c | 4 ++-- 13 files changed, 24 insertions(+), 23 deletions(-) (limited to 'kernel/tracepoint.c') diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 2ae5d72f47e..7238ef4c7a6 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -84,7 +84,7 @@ config S390 select HAVE_FUNCTION_TRACER select HAVE_FUNCTION_TRACE_MCOUNT_TEST select HAVE_FTRACE_MCOUNT_RECORD - select HAVE_FTRACE_SYSCALLS + select HAVE_SYSCALL_TRACEPOINTS select HAVE_DYNAMIC_FTRACE select HAVE_FUNCTION_GRAPH_TRACER select HAVE_DEFAULT_NO_SPIN_MUTEXES diff --git a/arch/s390/defconfig b/arch/s390/defconfig index fcba206529f..4e91a2573cc 100644 --- a/arch/s390/defconfig +++ b/arch/s390/defconfig @@ -900,7 +900,7 @@ CONFIG_HAVE_FUNCTION_GRAPH_TRACER=y CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST=y CONFIG_HAVE_DYNAMIC_FTRACE=y CONFIG_HAVE_FTRACE_MCOUNT_RECORD=y -CONFIG_HAVE_FTRACE_SYSCALLS=y +CONFIG_HAVE_SYSCALL_TRACEPOINTS=y CONFIG_TRACING_SUPPORT=y CONFIG_FTRACE=y # CONFIG_FUNCTION_TRACER is not set diff --git a/arch/s390/include/asm/thread_info.h b/arch/s390/include/asm/thread_info.h index ba1cab9fc1f..07eb61b2fb3 100644 --- a/arch/s390/include/asm/thread_info.h +++ b/arch/s390/include/asm/thread_info.h @@ -92,7 +92,7 @@ static inline struct thread_info *current_thread_info(void) #define TIF_SYSCALL_TRACE 8 /* syscall trace active */ #define TIF_SYSCALL_AUDIT 9 /* syscall auditing active */ #define TIF_SECCOMP 10 /* secure computing */ -#define TIF_SYSCALL_FTRACE 11 /* ftrace syscall instrumentation */ +#define TIF_SYSCALL_TRACEPOINT 11 /* syscall tracepoint instrumentation */ #define TIF_USEDFPU 16 /* FPU was used by this task this quantum (SMP) */ #define TIF_POLLING_NRFLAG 17 /* true if poll_idle() is polling TIF_NEED_RESCHED */ @@ -111,7 +111,7 @@ static inline struct thread_info *current_thread_info(void) #define _TIF_SYSCALL_TRACE (1<>8 | _TIF_SYSCALL_AUDIT>>8 | \ - _TIF_SECCOMP>>8 | _TIF_SYSCALL_FTRACE>>8) + _TIF_SECCOMP>>8 | _TIF_SYSCALL_TRACEPOINT>>8) STACK_SHIFT = PAGE_SHIFT + THREAD_ORDER STACK_SIZE = 1 << STACK_SHIFT diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S index f6618e9e15e..3ceb53c9c49 100644 --- a/arch/s390/kernel/entry64.S +++ b/arch/s390/kernel/entry64.S @@ -57,7 +57,7 @@ _TIF_WORK_SVC = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \ _TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \ _TIF_MCCK_PENDING) _TIF_SYSCALL = (_TIF_SYSCALL_TRACE>>8 | _TIF_SYSCALL_AUDIT>>8 | \ - _TIF_SECCOMP>>8 | _TIF_SYSCALL_FTRACE>>8) + _TIF_SECCOMP>>8 | _TIF_SYSCALL_TRACEPOINT>>8) #define BASED(name) name-system_call(%r13) diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c index c5e87d891ca..9d3dcfa79ea 100644 --- a/arch/s390/kernel/ptrace.c +++ b/arch/s390/kernel/ptrace.c @@ -664,7 +664,7 @@ asmlinkage long do_syscall_trace_enter(struct pt_regs *regs) ret = -1; } - if (unlikely(test_thread_flag(TIF_SYSCALL_FTRACE))) + if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT))) trace_syscall_enter(regs, regs->gprs[2]); if (unlikely(current->audit_context)) @@ -682,7 +682,7 @@ asmlinkage void do_syscall_trace_exit(struct pt_regs *regs) audit_syscall_exit(AUDITSC_RESULT(regs->gprs[2]), regs->gprs[2]); - if (unlikely(test_thread_flag(TIF_SYSCALL_FTRACE))) + if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT))) trace_syscall_exit(regs, regs->gprs[2]); if (test_thread_flag(TIF_SYSCALL_TRACE)) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 738bdc6b0f8..d59cbf758f3 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -37,7 +37,7 @@ config X86 select HAVE_FUNCTION_GRAPH_FP_TEST select HAVE_FUNCTION_TRACE_MCOUNT_TEST select HAVE_FTRACE_NMI_ENTER if DYNAMIC_FTRACE - select HAVE_FTRACE_SYSCALLS + select HAVE_SYSCALL_TRACEPOINTS select HAVE_KVM select HAVE_ARCH_KGDB select HAVE_ARCH_TRACEHOOK diff --git a/arch/x86/configs/i386_defconfig b/arch/x86/configs/i386_defconfig index edb992ebef9..d28fad19654 100644 --- a/arch/x86/configs/i386_defconfig +++ b/arch/x86/configs/i386_defconfig @@ -2355,7 +2355,7 @@ CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST=y CONFIG_HAVE_DYNAMIC_FTRACE=y CONFIG_HAVE_FTRACE_MCOUNT_RECORD=y CONFIG_HAVE_HW_BRANCH_TRACER=y -CONFIG_HAVE_FTRACE_SYSCALLS=y +CONFIG_HAVE_SYSCALL_TRACEPOINTS=y CONFIG_RING_BUFFER=y CONFIG_TRACING=y CONFIG_TRACING_SUPPORT=y diff --git a/arch/x86/configs/x86_64_defconfig b/arch/x86/configs/x86_64_defconfig index cee1dd2e69b..6c86acd847a 100644 --- a/arch/x86/configs/x86_64_defconfig +++ b/arch/x86/configs/x86_64_defconfig @@ -2329,7 +2329,7 @@ CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST=y CONFIG_HAVE_DYNAMIC_FTRACE=y CONFIG_HAVE_FTRACE_MCOUNT_RECORD=y CONFIG_HAVE_HW_BRANCH_TRACER=y -CONFIG_HAVE_FTRACE_SYSCALLS=y +CONFIG_HAVE_SYSCALL_TRACEPOINTS=y CONFIG_RING_BUFFER=y CONFIG_TRACING=y CONFIG_TRACING_SUPPORT=y diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h index fad7d40b75f..6f7786aea4f 100644 --- a/arch/x86/include/asm/thread_info.h +++ b/arch/x86/include/asm/thread_info.h @@ -95,7 +95,7 @@ struct thread_info { #define TIF_DEBUGCTLMSR 25 /* uses thread_struct.debugctlmsr */ #define TIF_DS_AREA_MSR 26 /* uses thread_struct.ds_area_msr */ #define TIF_LAZY_MMU_UPDATES 27 /* task is updating the mmu lazily */ -#define TIF_SYSCALL_FTRACE 28 /* for ftrace syscall instrumentation */ +#define TIF_SYSCALL_TRACEPOINT 28 /* syscall tracepoint instrumentation */ #define _TIF_SYSCALL_TRACE (1 << TIF_SYSCALL_TRACE) #define _TIF_NOTIFY_RESUME (1 << TIF_NOTIFY_RESUME) @@ -118,17 +118,17 @@ struct thread_info { #define _TIF_DEBUGCTLMSR (1 << TIF_DEBUGCTLMSR) #define _TIF_DS_AREA_MSR (1 << TIF_DS_AREA_MSR) #define _TIF_LAZY_MMU_UPDATES (1 << TIF_LAZY_MMU_UPDATES) -#define _TIF_SYSCALL_FTRACE (1 << TIF_SYSCALL_FTRACE) +#define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) /* work to do in syscall_trace_enter() */ #define _TIF_WORK_SYSCALL_ENTRY \ - (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_EMU | _TIF_SYSCALL_FTRACE | \ - _TIF_SYSCALL_AUDIT | _TIF_SECCOMP | _TIF_SINGLESTEP) + (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_EMU | _TIF_SYSCALL_AUDIT | \ + _TIF_SECCOMP | _TIF_SINGLESTEP | _TIF_SYSCALL_TRACEPOINT) /* work to do in syscall_trace_leave() */ #define _TIF_WORK_SYSCALL_EXIT \ (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | _TIF_SINGLESTEP | \ - _TIF_SYSCALL_FTRACE) + _TIF_SYSCALL_TRACEPOINT) /* work to do on interrupt/exception return */ #define _TIF_WORK_MASK \ @@ -137,7 +137,8 @@ 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) | _TIF_SYSCALL_FTRACE) +#define _TIF_ALLWORK_MASK \ + ((0x0000FFFF & ~_TIF_SECCOMP) | _TIF_SYSCALL_TRACEPOINT) /* Only used for 64 bit */ #define _TIF_DO_NOTIFY_MASK \ diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index 34dd6f15185..a909afef44f 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -1500,7 +1500,7 @@ asmregparm long syscall_trace_enter(struct pt_regs *regs) tracehook_report_syscall_entry(regs)) ret = -1L; - if (unlikely(test_thread_flag(TIF_SYSCALL_FTRACE))) + if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT))) trace_syscall_enter(regs, regs->orig_ax); if (unlikely(current->audit_context)) { @@ -1526,7 +1526,7 @@ 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))) + if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT))) trace_syscall_exit(regs, regs->ax); if (test_thread_flag(TIF_SYSCALL_TRACE)) diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index 019f380fd76..06be85a7ef8 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -41,7 +41,7 @@ config HAVE_FTRACE_MCOUNT_RECORD config HAVE_HW_BRANCH_TRACER bool -config HAVE_FTRACE_SYSCALLS +config HAVE_SYSCALL_TRACEPOINTS bool config TRACER_MAX_TRACE @@ -211,7 +211,7 @@ config ENABLE_DEFAULT_TRACERS config FTRACE_SYSCALLS bool "Trace syscalls" - depends on HAVE_FTRACE_SYSCALLS + depends on HAVE_SYSCALL_TRACEPOINTS select GENERIC_TRACER select KALLSYMS help diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index 06f165a4408..be86b9a01a0 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -590,7 +590,7 @@ void syscall_regfunc(void) if (!sys_tracepoint_refcount) { read_lock_irqsave(&tasklist_lock, flags); do_each_thread(g, t) { - set_tsk_thread_flag(t, TIF_SYSCALL_FTRACE); + set_tsk_thread_flag(t, TIF_SYSCALL_TRACEPOINT); } while_each_thread(g, t); read_unlock_irqrestore(&tasklist_lock, flags); } @@ -608,7 +608,7 @@ void syscall_unregfunc(void) if (!sys_tracepoint_refcount) { read_lock_irqsave(&tasklist_lock, flags); do_each_thread(g, t) { - clear_tsk_thread_flag(t, TIF_SYSCALL_FTRACE); + clear_tsk_thread_flag(t, TIF_SYSCALL_TRACEPOINT); } while_each_thread(g, t); read_unlock_irqrestore(&tasklist_lock, flags); } -- cgit v1.2.3-70-g09d2 From 3d27d8cb34fc156beb86de2338ca4029873a5cc6 Mon Sep 17 00:00:00 2001 From: Josh Stone Date: Mon, 24 Aug 2009 14:43:12 -0700 Subject: tracing: Make syscall tracepoints conditional The syscall enter/exit tracepoints are only supported on archs that HAVE_SYSCALL_TRACEPOINTS, so the declarations should be #ifdef'ed. Also, the definition of syscall_regfunc and syscall_unregfunc should depend on this same config, rather than the ftrace-specific one. Signed-off-by: Josh Stone Cc: Jason Baron Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Li Zefan Cc: Steven Rostedt Cc: Peter Zijlstra Cc: Mathieu Desnoyers Cc: Jiaying Zhang Cc: Martin Bligh Cc: Lai Jiangshan LKML-Reference: <1251150194-1713-3-git-send-email-jistone@redhat.com> Signed-off-by: Frederic Weisbecker --- include/trace/syscall.h | 4 ++++ kernel/tracepoint.c | 2 +- 2 files changed, 5 insertions(+), 1 deletion(-) (limited to 'kernel/tracepoint.c') diff --git a/include/trace/syscall.h b/include/trace/syscall.h index 9661dd406b9..5dcb7e3a544 100644 --- a/include/trace/syscall.h +++ b/include/trace/syscall.h @@ -8,6 +8,8 @@ #include +#ifdef CONFIG_HAVE_SYSCALL_TRACEPOINTS + extern void syscall_regfunc(void); extern void syscall_unregfunc(void); @@ -25,6 +27,8 @@ DECLARE_TRACE_WITH_CALLBACK(syscall_exit, syscall_unregfunc ); +#endif + /* * A syscall entry in the ftrace syscalls array. * diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index be86b9a01a0..9e0a36f0e2a 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -576,7 +576,7 @@ __initcall(init_tracepoints); #endif /* CONFIG_MODULES */ -#ifdef CONFIG_FTRACE_SYSCALLS +#ifdef CONFIG_HAVE_SYSCALL_TRACEPOINTS static DEFINE_MUTEX(regfunc_mutex); static int sys_tracepoint_refcount; -- cgit v1.2.3-70-g09d2 From 97419875865859fd2403e66266c02ce028e2f5ab Mon Sep 17 00:00:00 2001 From: Josh Stone Date: Mon, 24 Aug 2009 14:43:13 -0700 Subject: tracing: Move tracepoint callbacks from declaration to definition It's not strictly correct for the tracepoint reg/unreg callbacks to occur when a client is hooking up, because the actual tracepoint may not be present yet. This happens to be fine for syscall, since that's in the core kernel, but it would cause problems for tracepoints defined in a module that hasn't been loaded yet. It also means the reg/unreg has to be EXPORTed for any modules to use the tracepoint (as in SystemTap). This patch removes DECLARE_TRACE_WITH_CALLBACK, and instead introduces DEFINE_TRACE_FN which stores the callbacks in struct tracepoint. The callbacks are used now when the active state of the tracepoint changes in set_tracepoint & disable_tracepoint. This also introduces TRACE_EVENT_FN, so ftrace events can also provide registration callbacks if needed. Signed-off-by: Josh Stone Cc: Jason Baron Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Li Zefan Cc: Steven Rostedt Cc: Peter Zijlstra Cc: Mathieu Desnoyers Cc: Jiaying Zhang Cc: Martin Bligh Cc: Lai Jiangshan Cc: Paul Mundt Cc: Martin Schwidefsky Cc: Heiko Carstens LKML-Reference: <1251150194-1713-4-git-send-email-jistone@redhat.com> Signed-off-by: Frederic Weisbecker --- arch/s390/kernel/ptrace.c | 4 ++-- arch/x86/kernel/ptrace.c | 4 ++-- include/linux/tracepoint.h | 46 +++++++++++++++++--------------------------- include/trace/define_trace.h | 5 +++++ include/trace/ftrace.h | 9 +++++++++ include/trace/syscall.h | 12 ++++-------- kernel/tracepoint.c | 14 +++++++++----- 7 files changed, 49 insertions(+), 45 deletions(-) (limited to 'kernel/tracepoint.c') diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c index 9d3dcfa79ea..c05b44b80c2 100644 --- a/arch/s390/kernel/ptrace.c +++ b/arch/s390/kernel/ptrace.c @@ -51,8 +51,8 @@ #include "compat_ptrace.h" #endif -DEFINE_TRACE(syscall_enter); -DEFINE_TRACE(syscall_exit); +DEFINE_TRACE_FN(syscall_enter, syscall_regfunc, syscall_unregfunc); +DEFINE_TRACE_FN(syscall_exit, syscall_regfunc, syscall_unregfunc); enum s390_regset { REGSET_GENERAL, diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index a909afef44f..31e9b97ec4d 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -37,8 +37,8 @@ #include -DEFINE_TRACE(syscall_enter); -DEFINE_TRACE(syscall_exit); +DEFINE_TRACE_FN(syscall_enter, syscall_regfunc, syscall_unregfunc); +DEFINE_TRACE_FN(syscall_exit, syscall_regfunc, syscall_unregfunc); #include "tls.h" diff --git a/include/linux/tracepoint.h b/include/linux/tracepoint.h index 5984ed04c03..846a4ae501e 100644 --- a/include/linux/tracepoint.h +++ b/include/linux/tracepoint.h @@ -23,6 +23,8 @@ struct tracepoint; struct tracepoint { const char *name; /* Tracepoint name */ int state; /* State. */ + void (*regfunc)(void); + void (*unregfunc)(void); void **funcs; } __attribute__((aligned(32))); /* * Aligned on 32 bytes because it is @@ -60,10 +62,8 @@ struct tracepoint { * Make sure the alignment of the structure in the __tracepoints section will * not add unwanted padding between the beginning of the section and the * structure. Force alignment to the same alignment as the section start. - * An optional set of (un)registration functions can be passed to perform any - * additional (un)registration work. */ -#define DECLARE_TRACE_WITH_CALLBACK(name, proto, args, reg, unreg) \ +#define DECLARE_TRACE(name, proto, args) \ extern struct tracepoint __tracepoint_##name; \ static inline void trace_##name(proto) \ { \ @@ -73,36 +73,23 @@ struct tracepoint { } \ static inline int register_trace_##name(void (*probe)(proto)) \ { \ - int ret; \ - void (*func)(void) = reg; \ - \ - ret = tracepoint_probe_register(#name, (void *)probe); \ - if (func && !ret) \ - func(); \ - return ret; \ + return tracepoint_probe_register(#name, (void *)probe); \ } \ static inline int unregister_trace_##name(void (*probe)(proto)) \ { \ - int ret; \ - void (*func)(void) = unreg; \ - \ - ret = tracepoint_probe_unregister(#name, (void *)probe);\ - if (func && !ret) \ - func(); \ - return ret; \ + return tracepoint_probe_unregister(#name, (void *)probe);\ } -#define DECLARE_TRACE(name, proto, args) \ - DECLARE_TRACE_WITH_CALLBACK(name, TP_PROTO(proto), TP_ARGS(args),\ - NULL, NULL); - -#define DEFINE_TRACE(name) \ +#define DEFINE_TRACE_FN(name, reg, unreg) \ static const char __tpstrtab_##name[] \ __attribute__((section("__tracepoints_strings"))) = #name; \ struct tracepoint __tracepoint_##name \ __attribute__((section("__tracepoints"), aligned(32))) = \ - { __tpstrtab_##name, 0, NULL } + { __tpstrtab_##name, 0, reg, unreg, NULL } + +#define DEFINE_TRACE(name) \ + DEFINE_TRACE_FN(name, NULL, NULL); #define EXPORT_TRACEPOINT_SYMBOL_GPL(name) \ EXPORT_SYMBOL_GPL(__tracepoint_##name) @@ -113,7 +100,7 @@ extern void tracepoint_update_probe_range(struct tracepoint *begin, struct tracepoint *end); #else /* !CONFIG_TRACEPOINTS */ -#define DECLARE_TRACE_WITH_CALLBACK(name, proto, args, reg, unreg) \ +#define DECLARE_TRACE(name, proto, args) \ static inline void _do_trace_##name(struct tracepoint *tp, proto) \ { } \ static inline void trace_##name(proto) \ @@ -127,10 +114,7 @@ extern void tracepoint_update_probe_range(struct tracepoint *begin, return -ENOSYS; \ } -#define DECLARE_TRACE(name, proto, args) \ - DECLARE_TRACE_WITH_CALLBACK(name, TP_PROTO(proto), TP_ARGS(args),\ - NULL, NULL); - +#define DEFINE_TRACE_FN(name, reg, unreg) #define DEFINE_TRACE(name) #define EXPORT_TRACEPOINT_SYMBOL_GPL(name) #define EXPORT_TRACEPOINT_SYMBOL(name) @@ -282,10 +266,16 @@ static inline void tracepoint_synchronize_unregister(void) * can also by used by generic instrumentation like SystemTap), and * it is also used to expose a structured trace record in * /sys/kernel/debug/tracing/events/. + * + * A set of (un)registration functions can be passed to the variant + * TRACE_EVENT_FN to perform any (un)registration work. */ #define TRACE_EVENT(name, proto, args, struct, assign, print) \ DECLARE_TRACE(name, PARAMS(proto), PARAMS(args)) +#define TRACE_EVENT_FN(name, proto, args, struct, \ + assign, print, reg, unreg) \ + DECLARE_TRACE(name, PARAMS(proto), PARAMS(args)) #endif #endif diff --git a/include/trace/define_trace.h b/include/trace/define_trace.h index f7a7ae1e8f9..2a969850736 100644 --- a/include/trace/define_trace.h +++ b/include/trace/define_trace.h @@ -26,6 +26,11 @@ #define TRACE_EVENT(name, proto, args, tstruct, assign, print) \ DEFINE_TRACE(name) +#undef TRACE_EVENT_FN +#define TRACE_EVENT_FN(name, proto, args, tstruct, \ + assign, print, reg, unreg) \ + DEFINE_TRACE_FN(name, reg, unreg) + #undef DECLARE_TRACE #define DECLARE_TRACE(name, proto, args) \ DEFINE_TRACE(name) diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h index 127400255e4..3a0b44bdabf 100644 --- a/include/trace/ftrace.h +++ b/include/trace/ftrace.h @@ -42,6 +42,15 @@ }; \ static struct ftrace_event_call event_##name +/* Callbacks are meaningless to ftrace. */ +#undef TRACE_EVENT_FN +#define TRACE_EVENT_FN(name, proto, args, tstruct, \ + assign, print, reg, unreg) \ + TRACE_EVENT(name, TP_PROTO(proto), TP_ARGS(args), \ + TP_STRUCT__entry(tstruct), \ + TP_fast_assign(assign), \ + TP_printk(print)) + #include TRACE_INCLUDE(TRACE_INCLUDE_FILE) diff --git a/include/trace/syscall.h b/include/trace/syscall.h index 5dcb7e3a544..4e194300185 100644 --- a/include/trace/syscall.h +++ b/include/trace/syscall.h @@ -13,18 +13,14 @@ extern void syscall_regfunc(void); extern void syscall_unregfunc(void); -DECLARE_TRACE_WITH_CALLBACK(syscall_enter, +DECLARE_TRACE(syscall_enter, TP_PROTO(struct pt_regs *regs, long id), - TP_ARGS(regs, id), - syscall_regfunc, - syscall_unregfunc + TP_ARGS(regs, id) ); -DECLARE_TRACE_WITH_CALLBACK(syscall_exit, +DECLARE_TRACE(syscall_exit, TP_PROTO(struct pt_regs *regs, long ret), - TP_ARGS(regs, ret), - syscall_regfunc, - syscall_unregfunc + TP_ARGS(regs, ret) ); #endif diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index 9e0a36f0e2a..1a6a453b7ef 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -243,6 +243,11 @@ static void set_tracepoint(struct tracepoint_entry **entry, { WARN_ON(strcmp((*entry)->name, elem->name) != 0); + if (elem->regfunc && !elem->state && active) + elem->regfunc(); + else if (elem->unregfunc && elem->state && !active) + elem->unregfunc(); + /* * rcu_assign_pointer has a smp_wmb() which makes sure that the new * probe callbacks array is consistent before setting a pointer to it. @@ -262,6 +267,9 @@ static void set_tracepoint(struct tracepoint_entry **entry, */ static void disable_tracepoint(struct tracepoint *elem) { + if (elem->unregfunc && elem->state) + elem->unregfunc(); + elem->state = 0; rcu_assign_pointer(elem->funcs, NULL); } @@ -578,7 +586,7 @@ __initcall(init_tracepoints); #ifdef CONFIG_HAVE_SYSCALL_TRACEPOINTS -static DEFINE_MUTEX(regfunc_mutex); +/* NB: reg/unreg are called while guarded with the tracepoints_mutex */ static int sys_tracepoint_refcount; void syscall_regfunc(void) @@ -586,7 +594,6 @@ void syscall_regfunc(void) unsigned long flags; struct task_struct *g, *t; - mutex_lock(®func_mutex); if (!sys_tracepoint_refcount) { read_lock_irqsave(&tasklist_lock, flags); do_each_thread(g, t) { @@ -595,7 +602,6 @@ void syscall_regfunc(void) read_unlock_irqrestore(&tasklist_lock, flags); } sys_tracepoint_refcount++; - mutex_unlock(®func_mutex); } void syscall_unregfunc(void) @@ -603,7 +609,6 @@ void syscall_unregfunc(void) unsigned long flags; struct task_struct *g, *t; - mutex_lock(®func_mutex); sys_tracepoint_refcount--; if (!sys_tracepoint_refcount) { read_lock_irqsave(&tasklist_lock, flags); @@ -612,6 +617,5 @@ void syscall_unregfunc(void) } while_each_thread(g, t); read_unlock_irqrestore(&tasklist_lock, flags); } - mutex_unlock(®func_mutex); } #endif -- cgit v1.2.3-70-g09d2 From cc3b13c11c567c69a6356be98d0c03ff11541d5c Mon Sep 17 00:00:00 2001 From: Hendrik Brueckner Date: Tue, 25 Aug 2009 18:02:37 +0200 Subject: tracing: Don't trace kernel thread syscalls Kernel threads don't call syscalls using the sysenter/sysexit path. Instead they directly call the sys_* or do_* functions that implement the syscalls inside the kernel. The current syscall tracepoints only bind the sysenter/sysexit path, then it has no effect to trace the kernel thread calls to syscalls in that path. Setting the TIF_SYSCALL_TRACEPOINT flag is then useless for these. Actually there is only one case when a kernel thread can reach the usual syscall exit tracing path: when we create a kernel thread, the child comes to ret_from_fork and is the fork() return is then traced. But this information alone is useless, then we don't want to set the TIF flags for these threads. Kernel threads have task_struct->mm set to NULL. (Thanks to Heiko for that hint ;-) The idea is then to check the mm field in syscall_regfunc() and set the flag accordingly. Signed-off-by: Hendrik Brueckner Cc: Jason Baron Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Lai Jiangshan Cc: Steven Rostedt Cc: Peter Zijlstra Cc: Mathieu Desnoyers Cc: Jiaying Zhang Cc: Martin Bligh Cc: Li Zefan Cc: Martin Schwidefsky Cc: Paul Mundt Cc: Heiko Carstens Cc: Hendrik Brueckner LKML-Reference: <20090825160237.GG4639@cetus.boeblingen.de.ibm.com> Signed-off-by: Frederic Weisbecker --- kernel/tracepoint.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'kernel/tracepoint.c') diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index 1a6a453b7ef..9489a0a9b1b 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -597,7 +597,9 @@ void syscall_regfunc(void) if (!sys_tracepoint_refcount) { read_lock_irqsave(&tasklist_lock, flags); do_each_thread(g, t) { - set_tsk_thread_flag(t, TIF_SYSCALL_TRACEPOINT); + /* Skip kernel threads. */ + if (t->mm) + set_tsk_thread_flag(t, TIF_SYSCALL_TRACEPOINT); } while_each_thread(g, t); read_unlock_irqrestore(&tasklist_lock, flags); } -- cgit v1.2.3-70-g09d2