From 667000011927b4fcc359beac4a2447889db6d349 Mon Sep 17 00:00:00 2001 From: Josh Stone Date: Mon, 24 Aug 2009 14:43:11 -0700 Subject: [PATCH 1/4] 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(-) diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 2ae5d72f47ed..7238ef4c7a6b 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 fcba206529f3..4e91a2573cc4 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 ba1cab9fc1f9..07eb61b2fb3a 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 f6618e9e15ef..3ceb53c9c493 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 c5e87d891ca0..9d3dcfa79ea2 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 738bdc6b0f8b..d59cbf758f34 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 edb992ebef92..d28fad19654a 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 cee1dd2e69b2..6c86acd847a4 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 fad7d40b75f8..6f7786aea4fc 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 34dd6f15185d..a909afef44f4 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 019f380fd764..06be85a7ef8c 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 06f165a44083..be86b9a01a09 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); } From 3d27d8cb34fc156beb86de2338ca4029873a5cc6 Mon Sep 17 00:00:00 2001 From: Josh Stone Date: Mon, 24 Aug 2009 14:43:12 -0700 Subject: [PATCH 2/4] 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(-) diff --git a/include/trace/syscall.h b/include/trace/syscall.h index 9661dd406b93..5dcb7e3a544c 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 be86b9a01a09..9e0a36f0e2a9 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; From 97419875865859fd2403e66266c02ce028e2f5ab Mon Sep 17 00:00:00 2001 From: Josh Stone Date: Mon, 24 Aug 2009 14:43:13 -0700 Subject: [PATCH 3/4] 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(-) diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c index 9d3dcfa79ea2..c05b44b80c23 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 a909afef44f4..31e9b97ec4d6 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 5984ed04c03b..846a4ae501eb 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 f7a7ae1e8f90..2a969850736d 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 127400255e4c..3a0b44bdabf7 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 5dcb7e3a544c..4e1943001854 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 9e0a36f0e2a9..1a6a453b7efb 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 From 1c569f0264ea629c10bbab471dd0626ce4d3f19f Mon Sep 17 00:00:00 2001 From: Josh Stone Date: Mon, 24 Aug 2009 14:43:14 -0700 Subject: [PATCH 4/4] tracing: Create generic syscall TRACE_EVENTs This converts the syscall_enter/exit tracepoints into TRACE_EVENTs, so you can have generic ftrace events that capture all system calls with arguments and return values. These generic events are also renamed to sys_enter/exit, so they're more closely aligned to the specific sys_enter_foo events. 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: <1251150194-1713-5-git-send-email-jistone@redhat.com> Signed-off-by: Frederic Weisbecker --- arch/s390/kernel/ptrace.c | 8 ++-- arch/x86/kernel/ptrace.c | 12 +++--- include/trace/events/syscalls.h | 70 +++++++++++++++++++++++++++++++++ include/trace/syscall.h | 17 -------- kernel/trace/trace_syscalls.c | 17 ++++---- 5 files changed, 88 insertions(+), 36 deletions(-) create mode 100644 include/trace/events/syscalls.h diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c index c05b44b80c23..f3ddd7ac06c5 100644 --- a/arch/s390/kernel/ptrace.c +++ b/arch/s390/kernel/ptrace.c @@ -51,8 +51,8 @@ #include "compat_ptrace.h" #endif -DEFINE_TRACE_FN(syscall_enter, syscall_regfunc, syscall_unregfunc); -DEFINE_TRACE_FN(syscall_exit, syscall_regfunc, syscall_unregfunc); +#define CREATE_TRACE_POINTS +#include enum s390_regset { REGSET_GENERAL, @@ -665,7 +665,7 @@ asmlinkage long do_syscall_trace_enter(struct pt_regs *regs) } if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT))) - trace_syscall_enter(regs, regs->gprs[2]); + trace_sys_enter(regs, regs->gprs[2]); if (unlikely(current->audit_context)) audit_syscall_entry(is_compat_task() ? @@ -683,7 +683,7 @@ asmlinkage void do_syscall_trace_exit(struct pt_regs *regs) regs->gprs[2]); if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT))) - trace_syscall_exit(regs, regs->gprs[2]); + trace_sys_exit(regs, regs->gprs[2]); if (test_thread_flag(TIF_SYSCALL_TRACE)) tracehook_report_syscall_exit(regs, 0); diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index 31e9b97ec4d6..8d7d5c9c1be3 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -35,13 +35,11 @@ #include #include -#include - -DEFINE_TRACE_FN(syscall_enter, syscall_regfunc, syscall_unregfunc); -DEFINE_TRACE_FN(syscall_exit, syscall_regfunc, syscall_unregfunc); - #include "tls.h" +#define CREATE_TRACE_POINTS +#include + enum x86_regset { REGSET_GENERAL, REGSET_FP, @@ -1501,7 +1499,7 @@ asmregparm long syscall_trace_enter(struct pt_regs *regs) ret = -1L; if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT))) - trace_syscall_enter(regs, regs->orig_ax); + trace_sys_enter(regs, regs->orig_ax); if (unlikely(current->audit_context)) { if (IS_IA32) @@ -1527,7 +1525,7 @@ asmregparm void syscall_trace_leave(struct pt_regs *regs) audit_syscall_exit(AUDITSC_RESULT(regs->ax), regs->ax); if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT))) - trace_syscall_exit(regs, regs->ax); + trace_sys_exit(regs, regs->ax); if (test_thread_flag(TIF_SYSCALL_TRACE)) tracehook_report_syscall_exit(regs, 0); diff --git a/include/trace/events/syscalls.h b/include/trace/events/syscalls.h new file mode 100644 index 000000000000..397dff2dbd5a --- /dev/null +++ b/include/trace/events/syscalls.h @@ -0,0 +1,70 @@ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM syscalls + +#if !defined(_TRACE_EVENTS_SYSCALLS_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_EVENTS_SYSCALLS_H + +#include + +#include +#include + + +#ifdef CONFIG_HAVE_SYSCALL_TRACEPOINTS + +extern void syscall_regfunc(void); +extern void syscall_unregfunc(void); + +TRACE_EVENT_FN(sys_enter, + + TP_PROTO(struct pt_regs *regs, long id), + + TP_ARGS(regs, id), + + TP_STRUCT__entry( + __field( long, id ) + __array( unsigned long, args, 6 ) + ), + + TP_fast_assign( + __entry->id = id; + syscall_get_arguments(current, regs, 0, 6, __entry->args); + ), + + TP_printk("NR %ld (%lx, %lx, %lx, %lx, %lx, %lx)", + __entry->id, + __entry->args[0], __entry->args[1], __entry->args[2], + __entry->args[3], __entry->args[4], __entry->args[5]), + + syscall_regfunc, syscall_unregfunc +); + +TRACE_EVENT_FN(sys_exit, + + TP_PROTO(struct pt_regs *regs, long ret), + + TP_ARGS(regs, ret), + + TP_STRUCT__entry( + __field( long, id ) + __field( long, ret ) + ), + + TP_fast_assign( + __entry->id = syscall_get_nr(current, regs); + __entry->ret = ret; + ), + + TP_printk("NR %ld = %ld", + __entry->id, __entry->ret), + + syscall_regfunc, syscall_unregfunc +); + +#endif /* CONFIG_HAVE_SYSCALL_TRACEPOINTS */ + +#endif /* _TRACE_EVENTS_SYSCALLS_H */ + +/* This part must be outside protection */ +#include + diff --git a/include/trace/syscall.h b/include/trace/syscall.h index 4e1943001854..5dc283ba5ae0 100644 --- a/include/trace/syscall.h +++ b/include/trace/syscall.h @@ -8,23 +8,6 @@ #include -#ifdef CONFIG_HAVE_SYSCALL_TRACEPOINTS - -extern void syscall_regfunc(void); -extern void syscall_unregfunc(void); - -DECLARE_TRACE(syscall_enter, - TP_PROTO(struct pt_regs *regs, long id), - TP_ARGS(regs, id) -); - -DECLARE_TRACE(syscall_exit, - TP_PROTO(struct pt_regs *regs, long ret), - TP_ARGS(regs, ret) -); - -#endif - /* * A syscall entry in the ftrace syscalls array. * diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 46c1b977a2cb..2698fe401ebd 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -1,4 +1,5 @@ #include +#include #include #include #include @@ -286,7 +287,7 @@ int reg_event_syscall_enter(void *ptr) return -ENOSYS; mutex_lock(&syscall_trace_lock); if (!sys_refcount_enter) - ret = register_trace_syscall_enter(ftrace_syscall_enter); + ret = register_trace_sys_enter(ftrace_syscall_enter); if (ret) { pr_info("event trace: Could not activate" "syscall entry trace point"); @@ -311,7 +312,7 @@ void unreg_event_syscall_enter(void *ptr) sys_refcount_enter--; clear_bit(num, enabled_enter_syscalls); if (!sys_refcount_enter) - unregister_trace_syscall_enter(ftrace_syscall_enter); + unregister_trace_sys_enter(ftrace_syscall_enter); mutex_unlock(&syscall_trace_lock); } @@ -327,7 +328,7 @@ int reg_event_syscall_exit(void *ptr) return -ENOSYS; mutex_lock(&syscall_trace_lock); if (!sys_refcount_exit) - ret = register_trace_syscall_exit(ftrace_syscall_exit); + ret = register_trace_sys_exit(ftrace_syscall_exit); if (ret) { pr_info("event trace: Could not activate" "syscall exit trace point"); @@ -352,7 +353,7 @@ void unreg_event_syscall_exit(void *ptr) sys_refcount_exit--; clear_bit(num, enabled_exit_syscalls); if (!sys_refcount_exit) - unregister_trace_syscall_exit(ftrace_syscall_exit); + unregister_trace_sys_exit(ftrace_syscall_exit); mutex_unlock(&syscall_trace_lock); } @@ -418,7 +419,7 @@ int reg_prof_syscall_enter(char *name) mutex_lock(&syscall_trace_lock); if (!sys_prof_refcount_enter) - ret = register_trace_syscall_enter(prof_syscall_enter); + ret = register_trace_sys_enter(prof_syscall_enter); if (ret) { pr_info("event trace: Could not activate" "syscall entry trace point"); @@ -442,7 +443,7 @@ void unreg_prof_syscall_enter(char *name) sys_prof_refcount_enter--; clear_bit(num, enabled_prof_enter_syscalls); if (!sys_prof_refcount_enter) - unregister_trace_syscall_enter(prof_syscall_enter); + unregister_trace_sys_enter(prof_syscall_enter); mutex_unlock(&syscall_trace_lock); } @@ -479,7 +480,7 @@ int reg_prof_syscall_exit(char *name) mutex_lock(&syscall_trace_lock); if (!sys_prof_refcount_exit) - ret = register_trace_syscall_exit(prof_syscall_exit); + ret = register_trace_sys_exit(prof_syscall_exit); if (ret) { pr_info("event trace: Could not activate" "syscall entry trace point"); @@ -503,7 +504,7 @@ void unreg_prof_syscall_exit(char *name) sys_prof_refcount_exit--; clear_bit(num, enabled_prof_exit_syscalls); if (!sys_prof_refcount_exit) - unregister_trace_syscall_exit(prof_syscall_exit); + unregister_trace_sys_exit(prof_syscall_exit); mutex_unlock(&syscall_trace_lock); }