From f036be96dd9ce442ffb9ab33e3c165f5178815c0 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 5 Feb 2009 13:45:43 +0100 Subject: printk: introduce printk_once() This pattern shows up frequently in the kernel: static int once = 1; ... if (once) { once = 0; printk(KERN_ERR "message\n"); } ... So add a printk_once() helper macro that reduces this to a single line of: printk_once(KERN_ERR "message\n"); It works analogously to WARN_ONCE() & friends. (We use a macro not an inline because vararg expansion in inlines looks awkward and the macro is simple enough.) Signed-off-by: Ingo Molnar --- include/linux/kernel.h | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) (limited to 'include/linux/kernel.h') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 343df9ef2412..3c183d9864ae 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -242,6 +242,19 @@ extern struct ratelimit_state printk_ratelimit_state; extern int printk_ratelimit(void); extern bool printk_timed_ratelimit(unsigned long *caller_jiffies, unsigned int interval_msec); + +/* + * Print a one-time message (analogous to WARN_ONCE() et al): + */ +#define printk_once(x...) ({ \ + static int __print_once = 1; \ + \ + if (__print_once) { \ + __print_once = 0; \ + printk(x); \ + } \ +}) + #else static inline int vprintk(const char *s, va_list args) __attribute__ ((format (printf, 1, 0))); @@ -253,6 +266,10 @@ static inline int printk_ratelimit(void) { return 0; } static inline bool printk_timed_ratelimit(unsigned long *caller_jiffies, \ unsigned int interval_msec) \ { return false; } + +/* No effect, but we still get type checking even in the !PRINTK case: */ +#define printk_once(x...) printk(x) + #endif extern int printk_needs_cpu(int cpu); -- cgit v1.2.3 From 526211bc58c4b3265352801c5a7f469af5c34711 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 5 Mar 2009 10:28:45 +0100 Subject: tracing: move utility functions from ftrace.h to kernel.h Make common utility functions such as trace_printk() and tracing_start()/tracing_stop() generally available to kernel code. Cc: Steven Rostedt Signed-off-by: Ingo Molnar --- include/linux/ftrace.h | 58 ++------------------------------------------------ include/linux/kernel.h | 58 ++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 60 insertions(+), 56 deletions(-) (limited to 'include/linux/kernel.h') diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h index fbb9c364e166..5b64303ec9f2 100644 --- a/include/linux/ftrace.h +++ b/include/linux/ftrace.h @@ -318,62 +318,6 @@ static inline void __ftrace_enabled_restore(int enabled) # define trace_preempt_off(a0, a1) do { } while (0) #endif -#ifdef CONFIG_TRACING -extern int ftrace_dump_on_oops; - -extern void tracing_start(void); -extern void tracing_stop(void); -extern void ftrace_off_permanent(void); - -extern void -ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3); - -/** - * trace_printk - printf formatting in the ftrace buffer - * @fmt: the printf format for printing - * - * Note: __trace_printk is an internal function for trace_printk and - * the @ip is passed in via the trace_printk macro. - * - * This function allows a kernel developer to debug fast path sections - * that printk is not appropriate for. By scattering in various - * printk like tracing in the code, a developer can quickly see - * where problems are occurring. - * - * This is intended as a debugging tool for the developer only. - * Please refrain from leaving trace_printks scattered around in - * your code. - */ -# define trace_printk(fmt...) __trace_printk(_THIS_IP_, fmt) -extern int -__trace_printk(unsigned long ip, const char *fmt, ...) - __attribute__ ((format (printf, 2, 3))); -# define ftrace_vprintk(fmt, ap) __trace_printk(_THIS_IP_, fmt, ap) -extern int -__ftrace_vprintk(unsigned long ip, const char *fmt, va_list ap); -extern void ftrace_dump(void); -#else -static inline void -ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3) { } -static inline int -trace_printk(const char *fmt, ...) __attribute__ ((format (printf, 1, 2))); - -static inline void tracing_start(void) { } -static inline void tracing_stop(void) { } -static inline void ftrace_off_permanent(void) { } -static inline int -trace_printk(const char *fmt, ...) -{ - return 0; -} -static inline int -ftrace_vprintk(const char *fmt, va_list ap) -{ - return 0; -} -static inline void ftrace_dump(void) { } -#endif - #ifdef CONFIG_FTRACE_MCOUNT_RECORD extern void ftrace_init(void); extern void ftrace_init_module(struct module *mod, @@ -542,6 +486,8 @@ static inline int test_tsk_trace_graph(struct task_struct *tsk) return tsk->trace & TSK_TRACE_FL_GRAPH; } +extern int ftrace_dump_on_oops; + #endif /* CONFIG_TRACING */ diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 7fa371898e3e..08bf5da86676 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -367,6 +367,64 @@ static inline char *pack_hex_byte(char *buf, u8 byte) ({ if (0) printk(KERN_DEBUG pr_fmt(fmt), ##__VA_ARGS__); 0; }) #endif +/* + * General tracing related utility functions - trace_printk(), + * tracing_start()/tracing_stop: + */ +#ifdef CONFIG_TRACING +extern void tracing_start(void); +extern void tracing_stop(void); +extern void ftrace_off_permanent(void); + +extern void +ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3); + +/** + * trace_printk - printf formatting in the ftrace buffer + * @fmt: the printf format for printing + * + * Note: __trace_printk is an internal function for trace_printk and + * the @ip is passed in via the trace_printk macro. + * + * This function allows a kernel developer to debug fast path sections + * that printk is not appropriate for. By scattering in various + * printk like tracing in the code, a developer can quickly see + * where problems are occurring. + * + * This is intended as a debugging tool for the developer only. + * Please refrain from leaving trace_printks scattered around in + * your code. + */ +# define trace_printk(fmt...) __trace_printk(_THIS_IP_, fmt) +extern int +__trace_printk(unsigned long ip, const char *fmt, ...) + __attribute__ ((format (printf, 2, 3))); +# define ftrace_vprintk(fmt, ap) __trace_printk(_THIS_IP_, fmt, ap) +extern int +__ftrace_vprintk(unsigned long ip, const char *fmt, va_list ap); +extern void ftrace_dump(void); +#else +static inline void +ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3) { } +static inline int +trace_printk(const char *fmt, ...) __attribute__ ((format (printf, 1, 2))); + +static inline void tracing_start(void) { } +static inline void tracing_stop(void) { } +static inline void ftrace_off_permanent(void) { } +static inline int +trace_printk(const char *fmt, ...) +{ + return 0; +} +static inline int +ftrace_vprintk(const char *fmt, va_list ap) +{ + return 0; +} +static inline void ftrace_dump(void) { } +#endif + /* * Display an IP address in readable format. */ -- cgit v1.2.3 From 2002c258faaa8f89543df284fdbaa9e4b171547f Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Thu, 5 Mar 2009 10:35:56 -0500 Subject: tracing: add tracing_on/tracing_off to kernel.h Impact: cleanup The functions tracing_start/tracing_stop have been moved to kernel.h. These are not the functions a developer most likely wants to use when they want to insert a place to stop tracing and restart it from user space. tracing_start/tracing_stop was created to work with things like suspend to ram, where even calling smp_processor_id() can crash the system. The tracing_start/tracing_stop was used to stop the tracer from doing anything. These are still light weight functions, but add a bit more overhead to be able to stop the tracers. They also have no interface back to userland. That is, if the kernel calls tracing_stop, userland can not start tracing. What a developer most likely wants to use is tracing_on/tracing_off. These are very light weight functions (simply sets or clears a bit). These functions just stop recording into the ring buffer. The tracers don't even know that this happens except that they would receive NULL from the ring_buffer_lock_reserve function. Also, there's a way for the user land to enable or disable this bit. In debugfs/tracing/tracing_on, a user may echo "0" (same as tracing_off()) or echo "1" (same as tracing_on()) into this file. This becomes handy when a kernel developer is debugging and wants tracing to turn off when it hits an anomaly. Then the developer can examine the trace, and restart tracing if they want to try again (echo 1 > tracing_on). This patch moves the prototypes for tracing_on/tracing_off to kernel.h and comments their use, so that a kernel developer will know how to use them. Signed-off-by: Steven Rostedt --- include/linux/kernel.h | 29 ++++++++++++++++++++++++++++- include/linux/ring_buffer.h | 15 --------------- 2 files changed, 28 insertions(+), 16 deletions(-) (limited to 'include/linux/kernel.h') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 08bf5da86676..d4614a8a034b 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -369,8 +369,35 @@ static inline char *pack_hex_byte(char *buf, u8 byte) /* * General tracing related utility functions - trace_printk(), - * tracing_start()/tracing_stop: + * tracing_on/tracing_off and tracing_start()/tracing_stop + * + * Use tracing_on/tracing_off when you want to quickly turn on or off + * tracing. It simply enables or disables the recording of the trace events. + * This also corresponds to the user space debugfs/tracing/tracing_on + * file, which gives a means for the kernel and userspace to interact. + * Place a tracing_off() in the kernel where you want tracing to end. + * From user space, examine the trace, and then echo 1 > tracing_on + * to continue tracing. + * + * tracing_stop/tracing_start has slightly more overhead. It is used + * by things like suspend to ram where disabling the recording of the + * trace is not enough, but tracing must actually stop because things + * like calling smp_processor_id() may crash the system. + * + * Most likely, you want to use tracing_on/tracing_off. */ +#ifdef CONFIG_RING_BUFFER +void tracing_on(void); +void tracing_off(void); +/* trace_off_permanent stops recording with no way to bring it back */ +void tracing_off_permanent(void); +int tracing_is_on(void); +#else +static inline void tracing_on(void) { } +static inline void tracing_off(void) { } +static inline void tracing_off_permanent(void) { } +static inline int tracing_is_on(void) { return 0; } +#endif #ifdef CONFIG_TRACING extern void tracing_start(void); extern void tracing_stop(void); diff --git a/include/linux/ring_buffer.h b/include/linux/ring_buffer.h index 79fcbc4b09d6..b1a0068a5557 100644 --- a/include/linux/ring_buffer.h +++ b/include/linux/ring_buffer.h @@ -124,21 +124,6 @@ void ring_buffer_normalize_time_stamp(int cpu, u64 *ts); size_t ring_buffer_page_len(void *page); -/* - * The below functions are fine to use outside the tracing facility. - */ -#ifdef CONFIG_RING_BUFFER -void tracing_on(void); -void tracing_off(void); -void tracing_off_permanent(void); -int tracing_is_on(void); -#else -static inline void tracing_on(void) { } -static inline void tracing_off(void) { } -static inline void tracing_off_permanent(void) { } -static inline int tracing_is_on(void) { return 0; } -#endif - void *ring_buffer_alloc_read_page(struct ring_buffer *buffer); void ring_buffer_free_read_page(struct ring_buffer *buffer, void *data); int ring_buffer_read_page(struct ring_buffer *buffer, void **data_page, -- cgit v1.2.3 From 769b0441f438c4bb4872cb8560eb6fe51bcc09ee Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Fri, 6 Mar 2009 17:21:49 +0100 Subject: tracing/core: drop the old trace_printk() implementation in favour of trace_bprintk() Impact: faster and lighter tracing Now that we have trace_bprintk() which is faster and consume lesser memory than trace_printk() and has the same purpose, we can now drop the old implementation in favour of the binary one from trace_bprintk(), which means we move all the implementation of trace_bprintk() to trace_printk(), so the Api doesn't change except that we must now use trace_seq_bprintk() to print the TRACE_PRINT entries. Some changes result of this: - Previously, trace_bprintk depended of a single tracer and couldn't work without. This tracer has been dropped and the whole implementation of trace_printk() (like the module formats management) is now integrated in the tracing core (comes with CONFIG_TRACING), though we keep the file trace_printk (previously trace_bprintk.c) where we can find the module management. Thus we don't overflow trace.c - changes some parts to use trace_seq_bprintk() to print TRACE_PRINT entries. - change a bit trace_printk/trace_vprintk macros to support non-builtin formats constants, and fix 'const' qualifiers warnings. But this is all transparent for developers. - etc... V2: - Rebase against last changes - Fix mispell on the changelog V3: - Rebase against last changes (moving trace_printk() to kernel.h) Signed-off-by: Frederic Weisbecker Acked-by: Steven Rostedt LKML-Reference: <1236356510-8381-5-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar --- include/linux/ftrace.h | 25 ----- include/linux/kernel.h | 34 +++++- include/linux/module.h | 2 +- kernel/trace/Kconfig | 7 +- kernel/trace/Makefile | 2 +- kernel/trace/trace.c | 212 ++++++++++------------------------- kernel/trace/trace.h | 14 +-- kernel/trace/trace_bprintk.c | 154 ------------------------- kernel/trace/trace_functions_graph.c | 6 +- kernel/trace/trace_mmiotrace.c | 9 +- kernel/trace/trace_output.c | 70 ++---------- kernel/trace/trace_output.h | 2 + kernel/trace/trace_printk.c | 138 +++++++++++++++++++++++ 13 files changed, 262 insertions(+), 413 deletions(-) delete mode 100644 kernel/trace/trace_bprintk.c create mode 100644 kernel/trace/trace_printk.c (limited to 'include/linux/kernel.h') diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h index 1cc8ca453a9b..e1583f2639b0 100644 --- a/include/linux/ftrace.h +++ b/include/linux/ftrace.h @@ -223,31 +223,6 @@ extern int ftrace_make_nop(struct module *mod, */ extern int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr); -#ifdef CONFIG_TRACE_BPRINTK -extern int trace_vbprintk(unsigned long ip, const char *fmt, va_list args); -extern int __trace_bprintk(unsigned long ip, const char *fmt, ...) - __attribute__ ((format (printf, 2, 3))); - -static inline void ____trace_bprintk_check_format(const char *fmt, ...) - __attribute__ ((format (printf, 1, 2))); -static inline void ____trace_bprintk_check_format(const char *fmt, ...) {} -#define __trace_bprintk_check_format(fmt, args...) \ -do { \ - if (0) \ - ____trace_bprintk_check_format(fmt, ##args); \ -} while (0) - -#define trace_bprintk(fmt, args...) \ -do { \ - static char *__attribute__((section("__trace_bprintk_fmt"))) \ - trace_bprintk_fmt = fmt; \ - __trace_bprintk_check_format(fmt, ##args); \ - __trace_bprintk(_THIS_IP_, trace_bprintk_fmt, ##args); \ -} while (0) -#else -#define trace_bprintk trace_printk -#endif - /* May be defined in arch */ extern int ftrace_arch_read_dyn_info(char *buf, int size); diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 7aef15c4645e..4e726b9a71ec 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -423,6 +423,16 @@ extern void ftrace_off_permanent(void); extern void ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3); +static inline void __attribute__ ((format (printf, 1, 2))) +____trace_printk_check_format(const char *fmt, ...) +{ +} +#define __trace_printk_check_format(fmt, args...) \ +do { \ + if (0) \ + ____trace_printk_check_format(fmt, ##args); \ +} while (0) + /** * trace_printk - printf formatting in the ftrace buffer * @fmt: the printf format for printing @@ -439,13 +449,31 @@ ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3); * Please refrain from leaving trace_printks scattered around in * your code. */ -# define trace_printk(fmt...) __trace_printk(_THIS_IP_, fmt) + +#define trace_printk(fmt, args...) \ +do { \ + static const char *trace_printk_fmt \ + __attribute__((section("__trace_printk_fmt"))); \ + trace_printk_fmt = fmt; \ + __trace_printk_check_format(fmt, ##args); \ + __trace_printk(_THIS_IP_, trace_printk_fmt, ##args); \ +} while (0) + extern int __trace_printk(unsigned long ip, const char *fmt, ...) __attribute__ ((format (printf, 2, 3))); -# define ftrace_vprintk(fmt, ap) __trace_printk(_THIS_IP_, fmt, ap) + +#define ftrace_vprintk(fmt, vargs) \ +do { \ + static const char *trace_printk_fmt \ + __attribute__((section("__trace_printk_fmt"))); \ + trace_printk_fmt = fmt; \ + __ftrace_vprintk(_THIS_IP_, trace_printk_fmt, vargs); \ +} while (0) + extern int __ftrace_vprintk(unsigned long ip, const char *fmt, va_list ap); + extern void ftrace_dump(void); #else static inline void @@ -467,7 +495,7 @@ ftrace_vprintk(const char *fmt, va_list ap) return 0; } static inline void ftrace_dump(void) { } -#endif +#endif /* CONFIG_TRACING */ /* * Display an IP address in readable format. diff --git a/include/linux/module.h b/include/linux/module.h index 8cbec972d8e7..22d9878e868c 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -329,7 +329,7 @@ struct module unsigned int num_tracepoints; #endif -#ifdef CONFIG_TRACE_BPRINTK +#ifdef CONFIG_TRACING const char **trace_bprintk_fmt_start; unsigned int num_trace_bprintk_fmt; #endif diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index ad8d3617d0a6..8e4a2a61cd75 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -52,6 +52,7 @@ config TRACING select STACKTRACE if STACKTRACE_SUPPORT select TRACEPOINTS select NOP_TRACER + select BINARY_PRINTF # # Minimum requirements an architecture has to meet for us to @@ -97,12 +98,6 @@ config FUNCTION_GRAPH_TRACER This is done by setting the current return address on the current task structure into a stack of calls. -config TRACE_BPRINTK - bool "Binary printk for tracing" - default y - depends on TRACING - select BINARY_PRINTF - config IRQSOFF_TRACER bool "Interrupts-off Latency Tracer" default n diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile index 46557ef4c379..c7a2943796eb 100644 --- a/kernel/trace/Makefile +++ b/kernel/trace/Makefile @@ -22,7 +22,7 @@ obj-$(CONFIG_TRACING) += trace.o obj-$(CONFIG_TRACING) += trace_clock.o obj-$(CONFIG_TRACING) += trace_output.o obj-$(CONFIG_TRACING) += trace_stat.o -obj-$(CONFIG_TRACE_BPRINTK) += trace_bprintk.o +obj-$(CONFIG_TRACING) += trace_printk.o obj-$(CONFIG_CONTEXT_SWITCH_TRACER) += trace_sched_switch.o obj-$(CONFIG_SYSPROF_TRACER) += trace_sysprof.o obj-$(CONFIG_FUNCTION_TRACER) += trace_functions.o diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 46b3cd7a5752..cc94f8642485 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1169,6 +1169,67 @@ void trace_graph_return(struct ftrace_graph_ret *trace) } #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ + +/** + * trace_vprintk - write binary msg to tracing buffer + * + */ +int trace_vprintk(unsigned long ip, int depth, const char *fmt, va_list args) +{ + static DEFINE_SPINLOCK(trace_buf_lock); + static u32 trace_buf[TRACE_BUF_SIZE]; + + struct ring_buffer_event *event; + struct trace_array *tr = &global_trace; + struct trace_array_cpu *data; + struct print_entry *entry; + unsigned long flags; + int resched; + int cpu, len = 0, size, pc; + + if (unlikely(tracing_selftest_running || tracing_disabled)) + return 0; + + /* Don't pollute graph traces with trace_vprintk internals */ + pause_graph_tracing(); + + pc = preempt_count(); + resched = ftrace_preempt_disable(); + cpu = raw_smp_processor_id(); + data = tr->data[cpu]; + + if (unlikely(atomic_read(&data->disabled))) + goto out; + + spin_lock_irqsave(&trace_buf_lock, flags); + len = vbin_printf(trace_buf, TRACE_BUF_SIZE, fmt, args); + + if (len > TRACE_BUF_SIZE || len < 0) + goto out_unlock; + + size = sizeof(*entry) + sizeof(u32) * len; + event = trace_buffer_lock_reserve(tr, TRACE_PRINT, size, flags, pc); + if (!event) + goto out_unlock; + entry = ring_buffer_event_data(event); + entry->ip = ip; + entry->depth = depth; + entry->fmt = fmt; + + memcpy(entry->buf, trace_buf, sizeof(u32) * len); + ring_buffer_unlock_commit(tr->buffer, event); + +out_unlock: + spin_unlock_irqrestore(&trace_buf_lock, flags); + +out: + ftrace_preempt_enable(resched); + unpause_graph_tracing(); + + return len; +} +EXPORT_SYMBOL_GPL(trace_vprintk); + enum trace_file_type { TRACE_FILE_LAT_FMT = 1, TRACE_FILE_ANNOTATE = 2, @@ -1564,7 +1625,7 @@ static enum print_line_t print_printk_msg_only(struct trace_iterator *iter) trace_assign_type(field, entry); - ret = trace_seq_printf(s, "%s", field->buf); + ret = trace_seq_bprintf(s, field->fmt, field->buf); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -3714,155 +3775,6 @@ static __init int tracer_init_debugfs(void) return 0; } -int trace_vprintk(unsigned long ip, int depth, const char *fmt, va_list args) -{ - static raw_spinlock_t trace_buf_lock = __RAW_SPIN_LOCK_UNLOCKED; - static char trace_buf[TRACE_BUF_SIZE]; - - struct ring_buffer_event *event; - struct trace_array *tr = &global_trace; - struct trace_array_cpu *data; - int cpu, len = 0, size, pc; - struct print_entry *entry; - unsigned long irq_flags; - - if (tracing_disabled || tracing_selftest_running) - return 0; - - pc = preempt_count(); - preempt_disable_notrace(); - cpu = raw_smp_processor_id(); - data = tr->data[cpu]; - - if (unlikely(atomic_read(&data->disabled))) - goto out; - - pause_graph_tracing(); - raw_local_irq_save(irq_flags); - __raw_spin_lock(&trace_buf_lock); - len = vsnprintf(trace_buf, TRACE_BUF_SIZE, fmt, args); - - len = min(len, TRACE_BUF_SIZE-1); - trace_buf[len] = 0; - - size = sizeof(*entry) + len + 1; - event = trace_buffer_lock_reserve(tr, TRACE_PRINT, size, irq_flags, pc); - if (!event) - goto out_unlock; - entry = ring_buffer_event_data(event); - entry->ip = ip; - entry->depth = depth; - - memcpy(&entry->buf, trace_buf, len); - entry->buf[len] = 0; - ring_buffer_unlock_commit(tr->buffer, event); - - out_unlock: - __raw_spin_unlock(&trace_buf_lock); - raw_local_irq_restore(irq_flags); - unpause_graph_tracing(); - out: - preempt_enable_notrace(); - - return len; -} -EXPORT_SYMBOL_GPL(trace_vprintk); - -int __trace_printk(unsigned long ip, const char *fmt, ...) -{ - int ret; - va_list ap; - - if (!(trace_flags & TRACE_ITER_PRINTK)) - return 0; - - va_start(ap, fmt); - ret = trace_vprintk(ip, task_curr_ret_stack(current), fmt, ap); - va_end(ap); - return ret; -} -EXPORT_SYMBOL_GPL(__trace_printk); - -int __ftrace_vprintk(unsigned long ip, const char *fmt, va_list ap) -{ - if (!(trace_flags & TRACE_ITER_PRINTK)) - return 0; - - return trace_vprintk(ip, task_curr_ret_stack(current), fmt, ap); -} -EXPORT_SYMBOL_GPL(__ftrace_vprintk); - -/** - * trace_vbprintk - write binary msg to tracing buffer - * - * Caller must insure @fmt are valid when msg is in tracing buffer. - */ -int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) -{ - static DEFINE_SPINLOCK(trace_buf_lock); - static u32 trace_buf[TRACE_BUF_SIZE]; - - struct ring_buffer_event *event; - struct trace_array *tr = &global_trace; - struct trace_array_cpu *data; - struct bprintk_entry *entry; - unsigned long flags; - int resched; - int cpu, len = 0, size, pc; - - if (tracing_disabled || !trace_bprintk_enable) - return 0; - - pc = preempt_count(); - resched = ftrace_preempt_disable(); - cpu = raw_smp_processor_id(); - data = tr->data[cpu]; - - if (unlikely(atomic_read(&data->disabled))) - goto out; - - spin_lock_irqsave(&trace_buf_lock, flags); - len = vbin_printf(trace_buf, TRACE_BUF_SIZE, fmt, args); - - if (len > TRACE_BUF_SIZE || len < 0) - goto out_unlock; - - size = sizeof(*entry) + sizeof(u32) * len; - event = trace_buffer_lock_reserve(tr, TRACE_BPRINTK, size, flags, pc); - if (!event) - goto out_unlock; - entry = ring_buffer_event_data(event); - entry->ip = ip; - entry->fmt = fmt; - - memcpy(entry->buf, trace_buf, sizeof(u32) * len); - ring_buffer_unlock_commit(tr->buffer, event); - -out_unlock: - spin_unlock_irqrestore(&trace_buf_lock, flags); - -out: - ftrace_preempt_enable(resched); - - return len; -} -EXPORT_SYMBOL_GPL(trace_vbprintk); - -int __trace_bprintk(unsigned long ip, const char *fmt, ...) -{ - int ret; - va_list ap; - - if (!fmt) - return 0; - - va_start(ap, fmt); - ret = trace_vbprintk(ip, fmt, ap); - va_end(ap); - return ret; -} -EXPORT_SYMBOL_GPL(__trace_bprintk); - static int trace_panic_handler(struct notifier_block *this, unsigned long event, void *unused) { diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 0f5077f8f957..6140922392c8 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -20,7 +20,6 @@ enum trace_type { TRACE_WAKE, TRACE_STACK, TRACE_PRINT, - TRACE_BPRINTK, TRACE_SPECIAL, TRACE_MMIO_RW, TRACE_MMIO_MAP, @@ -120,16 +119,10 @@ struct userstack_entry { */ struct print_entry { struct trace_entry ent; - unsigned long ip; + unsigned long ip; int depth; - char buf[]; -}; - -struct bprintk_entry { - struct trace_entry ent; - unsigned long ip; - const char *fmt; - u32 buf[]; + const char *fmt; + u32 buf[]; }; #ifdef CONFIG_TRACE_BPRINTK extern int trace_bprintk_enable; @@ -296,7 +289,6 @@ extern void __ftrace_bad_type(void); IF_ASSIGN(var, ent, struct stack_entry, TRACE_STACK); \ IF_ASSIGN(var, ent, struct userstack_entry, TRACE_USER_STACK);\ IF_ASSIGN(var, ent, struct print_entry, TRACE_PRINT); \ - IF_ASSIGN(var, ent, struct bprintk_entry, TRACE_BPRINTK);\ IF_ASSIGN(var, ent, struct special_entry, 0); \ IF_ASSIGN(var, ent, struct trace_mmiotrace_rw, \ TRACE_MMIO_RW); \ diff --git a/kernel/trace/trace_bprintk.c b/kernel/trace/trace_bprintk.c deleted file mode 100644 index f4c245a5cd33..000000000000 --- a/kernel/trace/trace_bprintk.c +++ /dev/null @@ -1,154 +0,0 @@ -/* - * trace binary printk - * - * Copyright (C) 2008 Lai Jiangshan - * - */ -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include "trace.h" - -#ifdef CONFIG_MODULES - -/* binary printk basic */ -static DEFINE_MUTEX(btrace_mutex); -/* - * modules trace_bprintk()'s formats are autosaved in struct trace_bprintk_fmt - * which are queued on trace_bprintk_fmt_list. - */ -static LIST_HEAD(trace_bprintk_fmt_list); - -struct trace_bprintk_fmt { - struct list_head list; - char fmt[0]; -}; - - -static inline void lock_btrace(void) -{ - mutex_lock(&btrace_mutex); -} - -static inline void unlock_btrace(void) -{ - mutex_unlock(&btrace_mutex); -} - - -static inline struct trace_bprintk_fmt *lookup_format(const char *fmt) -{ - struct trace_bprintk_fmt *pos; - list_for_each_entry(pos, &trace_bprintk_fmt_list, list) { - if (!strcmp(pos->fmt, fmt)) - return pos; - } - return NULL; -} - -static -void hold_module_trace_bprintk_format(const char **start, const char **end) -{ - const char **iter; - lock_btrace(); - for (iter = start; iter < end; iter++) { - struct trace_bprintk_fmt *tb_fmt = lookup_format(*iter); - if (tb_fmt) { - *iter = tb_fmt->fmt; - continue; - } - - tb_fmt = kmalloc(offsetof(struct trace_bprintk_fmt, fmt) - + strlen(*iter) + 1, GFP_KERNEL); - if (tb_fmt) { - list_add_tail(&tb_fmt->list, &trace_bprintk_fmt_list); - strcpy(tb_fmt->fmt, *iter); - *iter = tb_fmt->fmt; - } else - *iter = NULL; - } - unlock_btrace(); -} - -static int module_trace_bprintk_format_notify(struct notifier_block *self, - unsigned long val, void *data) -{ - struct module *mod = data; - if (mod->num_trace_bprintk_fmt) { - const char **start = mod->trace_bprintk_fmt_start; - const char **end = start + mod->num_trace_bprintk_fmt; - - if (val == MODULE_STATE_COMING) - hold_module_trace_bprintk_format(start, end); - } - return 0; -} - -#else /* !CONFIG_MODULES */ -__init static int -module_trace_bprintk_format_notify(struct notifier_block *self, - unsigned long val, void *data) -{ - return 0; -} -#endif /* CONFIG_MODULES */ - - -__initdata_or_module static -struct notifier_block module_trace_bprintk_format_nb = { - .notifier_call = module_trace_bprintk_format_notify, -}; - -/* events tracer */ -int trace_bprintk_enable; - -static void start_bprintk_trace(struct trace_array *tr) -{ - tracing_reset_online_cpus(tr); - trace_bprintk_enable = 1; -} - -static void stop_bprintk_trace(struct trace_array *tr) -{ - trace_bprintk_enable = 0; - tracing_reset_online_cpus(tr); -} - -static int init_bprintk_trace(struct trace_array *tr) -{ - start_bprintk_trace(tr); - return 0; -} - -static struct tracer bprintk_trace __read_mostly = -{ - .name = "events", - .init = init_bprintk_trace, - .reset = stop_bprintk_trace, - .start = start_bprintk_trace, - .stop = stop_bprintk_trace, -}; - -static __init int init_bprintk(void) -{ - int ret = register_module_notifier(&module_trace_bprintk_format_nb); - if (ret) - return ret; - - ret = register_tracer(&bprintk_trace); - if (ret) - unregister_module_notifier(&module_trace_bprintk_format_nb); - return ret; -} - -device_initcall(init_bprintk); diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index e527f2f66c73..453ebd3b636e 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -742,7 +742,11 @@ print_graph_comment(struct print_entry *trace, struct trace_seq *s, } /* The comment */ - ret = trace_seq_printf(s, "/* %s", trace->buf); + ret = trace_seq_printf(s, "/* "); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + ret = trace_seq_bprintf(s, trace->fmt, trace->buf); if (!ret) return TRACE_TYPE_PARTIAL_LINE; diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c index c401b908e805..23e346a734ca 100644 --- a/kernel/trace/trace_mmiotrace.c +++ b/kernel/trace/trace_mmiotrace.c @@ -254,15 +254,18 @@ static enum print_line_t mmio_print_mark(struct trace_iterator *iter) { struct trace_entry *entry = iter->ent; struct print_entry *print = (struct print_entry *)entry; - const char *msg = print->buf; struct trace_seq *s = &iter->seq; unsigned long long t = ns2usecs(iter->ts); - unsigned long usec_rem = do_div(t, 1000000ULL); + unsigned long usec_rem = do_div(t, USEC_PER_SEC); unsigned secs = (unsigned long)t; int ret; /* The trailing newline must be in the message. */ - ret = trace_seq_printf(s, "MARK %u.%06lu %s", secs, usec_rem, msg); + ret = trace_seq_printf(s, "MARK %u.%06lu ", secs, usec_rem); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + ret = trace_seq_bprintf(s, print->fmt, print->buf); if (!ret) return TRACE_TYPE_PARTIAL_LINE; diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index 4ab71201862e..ef8fd661b217 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -53,8 +53,7 @@ trace_seq_printf(struct trace_seq *s, const char *fmt, ...) return len; } -static int -trace_seq_bprintf(struct trace_seq *s, const char *fmt, const u32 *binary) +int trace_seq_bprintf(struct trace_seq *s, const char *fmt, const u32 *binary) { int len = (PAGE_SIZE - 1) - s->len; int ret; @@ -834,54 +833,12 @@ static struct trace_event trace_user_stack_event = { }; /* TRACE_PRINT */ -static enum print_line_t trace_print_print(struct trace_iterator *iter, - int flags) -{ - struct print_entry *field; - struct trace_seq *s = &iter->seq; - - trace_assign_type(field, iter->ent); - - if (!seq_print_ip_sym(s, field->ip, flags)) - goto partial; - - if (!trace_seq_printf(s, ": %s", field->buf)) - goto partial; - - return TRACE_TYPE_HANDLED; - - partial: - return TRACE_TYPE_PARTIAL_LINE; -} - -static enum print_line_t trace_print_raw(struct trace_iterator *iter, int flags) -{ - struct print_entry *field; - - trace_assign_type(field, iter->ent); - - if (!trace_seq_printf(&iter->seq, "# %lx %s", field->ip, field->buf)) - goto partial; - - return TRACE_TYPE_HANDLED; - - partial: - return TRACE_TYPE_PARTIAL_LINE; -} - -static struct trace_event trace_print_event = { - .type = TRACE_PRINT, - .trace = trace_print_print, - .raw = trace_print_raw, -}; - -/* TRACE_BPRINTK */ static enum print_line_t -trace_bprintk_print(struct trace_iterator *iter, int flags) +trace_print_print(struct trace_iterator *iter, int flags) { struct trace_entry *entry = iter->ent; struct trace_seq *s = &iter->seq; - struct bprintk_entry *field; + struct print_entry *field; trace_assign_type(field, entry); @@ -900,14 +857,13 @@ trace_bprintk_print(struct trace_iterator *iter, int flags) return TRACE_TYPE_PARTIAL_LINE; } -static enum print_line_t -trace_bprintk_raw(struct trace_iterator *iter, int flags) + +static enum print_line_t trace_print_raw(struct trace_iterator *iter, int flags) { - struct trace_entry *entry = iter->ent; + struct print_entry *field; struct trace_seq *s = &iter->seq; - struct bprintk_entry *field; - trace_assign_type(field, entry); + trace_assign_type(field, iter->ent); if (!trace_seq_printf(s, ": %lx : ", field->ip)) goto partial; @@ -921,12 +877,11 @@ trace_bprintk_raw(struct trace_iterator *iter, int flags) return TRACE_TYPE_PARTIAL_LINE; } -static struct trace_event trace_bprintk_event = { - .type = TRACE_BPRINTK, - .trace = trace_bprintk_print, - .raw = trace_bprintk_raw, - .hex = trace_nop_print, - .binary = trace_nop_print, + +static struct trace_event trace_print_event = { + .type = TRACE_PRINT, + .trace = trace_print_print, + .raw = trace_print_raw, }; static struct trace_event *events[] __initdata = { @@ -937,7 +892,6 @@ static struct trace_event *events[] __initdata = { &trace_stack_event, &trace_user_stack_event, &trace_print_event, - &trace_bprintk_event, NULL }; diff --git a/kernel/trace/trace_output.h b/kernel/trace/trace_output.h index 8a34d688ed63..3b90e6ade1aa 100644 --- a/kernel/trace/trace_output.h +++ b/kernel/trace/trace_output.h @@ -18,6 +18,8 @@ struct trace_event { extern int trace_seq_printf(struct trace_seq *s, const char *fmt, ...) __attribute__ ((format (printf, 2, 3))); extern int +trace_seq_bprintf(struct trace_seq *s, const char *fmt, const u32 *binary); +extern int seq_print_ip_sym(struct trace_seq *s, unsigned long ip, unsigned long sym_flags); extern ssize_t trace_seq_to_user(struct trace_seq *s, char __user *ubuf, diff --git a/kernel/trace/trace_printk.c b/kernel/trace/trace_printk.c new file mode 100644 index 000000000000..a50aea22e929 --- /dev/null +++ b/kernel/trace/trace_printk.c @@ -0,0 +1,138 @@ +/* + * trace binary printk + * + * Copyright (C) 2008 Lai Jiangshan + * + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "trace.h" + +#ifdef CONFIG_MODULES + +/* + * modules trace_printk()'s formats are autosaved in struct trace_bprintk_fmt + * which are queued on trace_bprintk_fmt_list. + */ +static LIST_HEAD(trace_bprintk_fmt_list); + +/* serialize accesses to trace_bprintk_fmt_list */ +static DEFINE_MUTEX(btrace_mutex); + +struct trace_bprintk_fmt { + struct list_head list; + char fmt[0]; +}; + +static inline struct trace_bprintk_fmt *lookup_format(const char *fmt) +{ + struct trace_bprintk_fmt *pos; + list_for_each_entry(pos, &trace_bprintk_fmt_list, list) { + if (!strcmp(pos->fmt, fmt)) + return pos; + } + return NULL; +} + +static +void hold_module_trace_bprintk_format(const char **start, const char **end) +{ + const char **iter; + + mutex_lock(&btrace_mutex); + for (iter = start; iter < end; iter++) { + struct trace_bprintk_fmt *tb_fmt = lookup_format(*iter); + if (tb_fmt) { + *iter = tb_fmt->fmt; + continue; + } + + tb_fmt = kmalloc(offsetof(struct trace_bprintk_fmt, fmt) + + strlen(*iter) + 1, GFP_KERNEL); + if (tb_fmt) { + list_add_tail(&tb_fmt->list, &trace_bprintk_fmt_list); + strcpy(tb_fmt->fmt, *iter); + *iter = tb_fmt->fmt; + } else + *iter = NULL; + } + mutex_unlock(&btrace_mutex); +} + +static int module_trace_bprintk_format_notify(struct notifier_block *self, + unsigned long val, void *data) +{ + struct module *mod = data; + if (mod->num_trace_bprintk_fmt) { + const char **start = mod->trace_bprintk_fmt_start; + const char **end = start + mod->num_trace_bprintk_fmt; + + if (val == MODULE_STATE_COMING) + hold_module_trace_bprintk_format(start, end); + } + return 0; +} + +#else /* !CONFIG_MODULES */ +__init static int +module_trace_bprintk_format_notify(struct notifier_block *self, + unsigned long val, void *data) +{ + return 0; +} +#endif /* CONFIG_MODULES */ + + +__initdata_or_module static +struct notifier_block module_trace_bprintk_format_nb = { + .notifier_call = module_trace_bprintk_format_notify, +}; + +int __trace_printk(unsigned long ip, const char *fmt, ...) + { + int ret; + va_list ap; + + if (unlikely(!fmt)) + return 0; + + if (!(trace_flags & TRACE_ITER_PRINTK)) + return 0; + + va_start(ap, fmt); + ret = trace_vprintk(ip, task_curr_ret_stack(current), fmt, ap); + va_end(ap); + return ret; +} +EXPORT_SYMBOL_GPL(__trace_printk); + +int __ftrace_vprintk(unsigned long ip, const char *fmt, va_list ap) + { + if (unlikely(!fmt)) + return 0; + + if (!(trace_flags & TRACE_ITER_PRINTK)) + return 0; + + return trace_vprintk(ip, task_curr_ret_stack(current), fmt, ap); +} +EXPORT_SYMBOL_GPL(__ftrace_vprintk); + + +static __init int init_trace_printk(void) +{ + return register_module_notifier(&module_trace_bprintk_format_nb); +} + +early_initcall(init_trace_printk); -- cgit v1.2.3 From 7bffc23e56e92c14b787bf4d95249a32085bfed5 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Mon, 9 Mar 2009 10:11:36 +0100 Subject: tracing: optimize trace_printk() Impact: micro-optimization trace_printk() does this unconditionally: trace_printk_fmt = fmt; Where trace_printk_fmt is an entry into a global array. This is very SMP-unfriendly. So only write it once per bootup. Cc: Frederic Weisbecker Cc: Steven Rostedt Cc: Peter Zijlstra LKML-Reference: <1236356510-8381-5-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar --- include/linux/kernel.h | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'include/linux/kernel.h') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 4e726b9a71ec..7742798c9208 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -454,7 +454,10 @@ do { \ do { \ static const char *trace_printk_fmt \ __attribute__((section("__trace_printk_fmt"))); \ - trace_printk_fmt = fmt; \ + \ + if (!trace_printk_fmt) \ + trace_printk_fmt = fmt; \ + \ __trace_printk_check_format(fmt, ##args); \ __trace_printk(_THIS_IP_, trace_printk_fmt, ##args); \ } while (0) @@ -467,7 +470,10 @@ __trace_printk(unsigned long ip, const char *fmt, ...) do { \ static const char *trace_printk_fmt \ __attribute__((section("__trace_printk_fmt"))); \ - trace_printk_fmt = fmt; \ + \ + if (!trace_printk_fmt) \ + trace_printk_fmt = fmt; \ + \ __ftrace_vprintk(_THIS_IP_, trace_printk_fmt, vargs); \ } while (0) -- cgit v1.2.3 From 48ead02030f849d011259244bb4ea9b985479006 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Thu, 12 Mar 2009 18:24:49 +0100 Subject: tracing/core: bring back raw trace_printk for dynamic formats strings Impact: fix callsites with dynamic format strings Since its new binary implementation, trace_printk() internally uses static containers for the format strings on each callsites. But the value is assigned once at build time, which means that it can't take dynamic formats. So this patch unearthes the raw trace_printk implementation for the callers that will need trace_printk to be able to carry these dynamic format strings. The trace_printk() macro will use the appropriate implementation for each callsite. Most of the time however, the binary implementation will still be used. The other impact of this patch is that mmiotrace_printk() will use the old implementation because it calls the low level trace_vprintk and we can't guess here whether the format passed in it is dynamic or not. Some parts of this patch have been written by Steven Rostedt (most notably the part that chooses the appropriate implementation for each callsites). Signed-off-by: Frederic Weisbecker Signed-off-by: Steven Rostedt --- include/linux/kernel.h | 40 +++++++++++------ kernel/trace/trace.c | 85 +++++++++++++++++++++++++++++++++--- kernel/trace/trace.h | 13 +++++- kernel/trace/trace_event_types.h | 11 ++++- kernel/trace/trace_functions_graph.c | 6 +-- kernel/trace/trace_mmiotrace.c | 7 +-- kernel/trace/trace_output.c | 57 +++++++++++++++++++++--- kernel/trace/trace_printk.c | 33 +++++++++++--- 8 files changed, 213 insertions(+), 39 deletions(-) (limited to 'include/linux/kernel.h') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 7742798c9208..1daca3b062bb 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -452,31 +452,45 @@ do { \ #define trace_printk(fmt, args...) \ do { \ - static const char *trace_printk_fmt \ - __attribute__((section("__trace_printk_fmt"))); \ - \ - if (!trace_printk_fmt) \ - trace_printk_fmt = fmt; \ - \ __trace_printk_check_format(fmt, ##args); \ - __trace_printk(_THIS_IP_, trace_printk_fmt, ##args); \ + if (__builtin_constant_p(fmt)) { \ + static const char *trace_printk_fmt \ + __attribute__((section("__trace_printk_fmt"))) = \ + __builtin_constant_p(fmt) ? fmt : NULL; \ + \ + __trace_bprintk(_THIS_IP_, trace_printk_fmt, ##args); \ + } else \ + __trace_printk(_THIS_IP_, fmt, ##args); \ } while (0) +extern int +__trace_bprintk(unsigned long ip, const char *fmt, ...) + __attribute__ ((format (printf, 2, 3))); + extern int __trace_printk(unsigned long ip, const char *fmt, ...) __attribute__ ((format (printf, 2, 3))); +/* + * The double __builtin_constant_p is because gcc will give us an error + * if we try to allocate the static variable to fmt if it is not a + * constant. Even with the outer if statement. + */ #define ftrace_vprintk(fmt, vargs) \ do { \ - static const char *trace_printk_fmt \ - __attribute__((section("__trace_printk_fmt"))); \ - \ - if (!trace_printk_fmt) \ - trace_printk_fmt = fmt; \ + if (__builtin_constant_p(fmt)) { \ + static const char *trace_printk_fmt \ + __attribute__((section("__trace_printk_fmt"))) = \ + __builtin_constant_p(fmt) ? fmt : NULL; \ \ - __ftrace_vprintk(_THIS_IP_, trace_printk_fmt, vargs); \ + __ftrace_vbprintk(_THIS_IP_, trace_printk_fmt, vargs); \ + } else \ + __ftrace_vprintk(_THIS_IP_, fmt, vargs); \ } while (0) +extern int +__ftrace_vbprintk(unsigned long ip, const char *fmt, va_list ap); + extern int __ftrace_vprintk(unsigned long ip, const char *fmt, va_list ap); diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 62a63b2b33dd..dbb077d8a172 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1179,10 +1179,10 @@ void trace_graph_return(struct ftrace_graph_ret *trace) /** - * trace_vprintk - write binary msg to tracing buffer + * trace_vbprintk - write binary msg to tracing buffer * */ -int trace_vprintk(unsigned long ip, int depth, const char *fmt, va_list args) +int trace_vbprintk(unsigned long ip, int depth, const char *fmt, va_list args) { static raw_spinlock_t trace_buf_lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; @@ -1191,7 +1191,7 @@ int trace_vprintk(unsigned long ip, int depth, const char *fmt, va_list args) struct ring_buffer_event *event; struct trace_array *tr = &global_trace; struct trace_array_cpu *data; - struct print_entry *entry; + struct bprint_entry *entry; unsigned long flags; int resched; int cpu, len = 0, size, pc; @@ -1219,7 +1219,7 @@ int trace_vprintk(unsigned long ip, int depth, const char *fmt, va_list args) goto out_unlock; size = sizeof(*entry) + sizeof(u32) * len; - event = trace_buffer_lock_reserve(tr, TRACE_PRINT, size, flags, pc); + event = trace_buffer_lock_reserve(tr, TRACE_BPRINT, size, flags, pc); if (!event) goto out_unlock; entry = ring_buffer_event_data(event); @@ -1240,6 +1240,60 @@ out: return len; } +EXPORT_SYMBOL_GPL(trace_vbprintk); + +int trace_vprintk(unsigned long ip, int depth, const char *fmt, va_list args) +{ + static raw_spinlock_t trace_buf_lock = __RAW_SPIN_LOCK_UNLOCKED; + static char trace_buf[TRACE_BUF_SIZE]; + + struct ring_buffer_event *event; + struct trace_array *tr = &global_trace; + struct trace_array_cpu *data; + int cpu, len = 0, size, pc; + struct print_entry *entry; + unsigned long irq_flags; + + if (tracing_disabled || tracing_selftest_running) + return 0; + + pc = preempt_count(); + preempt_disable_notrace(); + cpu = raw_smp_processor_id(); + data = tr->data[cpu]; + + if (unlikely(atomic_read(&data->disabled))) + goto out; + + pause_graph_tracing(); + raw_local_irq_save(irq_flags); + __raw_spin_lock(&trace_buf_lock); + len = vsnprintf(trace_buf, TRACE_BUF_SIZE, fmt, args); + + len = min(len, TRACE_BUF_SIZE-1); + trace_buf[len] = 0; + + size = sizeof(*entry) + len + 1; + event = trace_buffer_lock_reserve(tr, TRACE_PRINT, size, irq_flags, pc); + if (!event) + goto out_unlock; + entry = ring_buffer_event_data(event); + entry->ip = ip; + entry->depth = depth; + + memcpy(&entry->buf, trace_buf, len); + entry->buf[len] = 0; + ring_buffer_unlock_commit(tr->buffer, event); + + out_unlock: + __raw_spin_unlock(&trace_buf_lock); + raw_local_irq_restore(irq_flags); + unpause_graph_tracing(); + out: + preempt_enable_notrace(); + + return len; +} EXPORT_SYMBOL_GPL(trace_vprintk); enum trace_file_type { @@ -1628,6 +1682,22 @@ static enum print_line_t print_hex_fmt(struct trace_iterator *iter) return TRACE_TYPE_HANDLED; } +static enum print_line_t print_bprintk_msg_only(struct trace_iterator *iter) +{ + struct trace_seq *s = &iter->seq; + struct trace_entry *entry = iter->ent; + struct bprint_entry *field; + int ret; + + trace_assign_type(field, entry); + + ret = trace_seq_bprintf(s, field->fmt, field->buf); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + return TRACE_TYPE_HANDLED; +} + static enum print_line_t print_printk_msg_only(struct trace_iterator *iter) { struct trace_seq *s = &iter->seq; @@ -1637,7 +1707,7 @@ static enum print_line_t print_printk_msg_only(struct trace_iterator *iter) trace_assign_type(field, entry); - ret = trace_seq_bprintf(s, field->fmt, field->buf); + ret = trace_seq_printf(s, "%s", field->buf); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -1702,6 +1772,11 @@ static enum print_line_t print_trace_line(struct trace_iterator *iter) return ret; } + if (iter->ent->type == TRACE_BPRINT && + trace_flags & TRACE_ITER_PRINTK && + trace_flags & TRACE_ITER_PRINTK_MSGONLY) + return print_bprintk_msg_only(iter); + if (iter->ent->type == TRACE_PRINT && trace_flags & TRACE_ITER_PRINTK && trace_flags & TRACE_ITER_PRINTK_MSGONLY) diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 336324d717f8..cede1ab49d07 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -20,6 +20,7 @@ enum trace_type { TRACE_WAKE, TRACE_STACK, TRACE_PRINT, + TRACE_BPRINT, TRACE_SPECIAL, TRACE_MMIO_RW, TRACE_MMIO_MAP, @@ -117,7 +118,7 @@ struct userstack_entry { /* * trace_printk entry: */ -struct print_entry { +struct bprint_entry { struct trace_entry ent; unsigned long ip; int depth; @@ -125,6 +126,13 @@ struct print_entry { u32 buf[]; }; +struct print_entry { + struct trace_entry ent; + unsigned long ip; + int depth; + char buf[]; +}; + #define TRACE_OLD_SIZE 88 struct trace_field_cont { @@ -286,6 +294,7 @@ extern void __ftrace_bad_type(void); IF_ASSIGN(var, ent, struct stack_entry, TRACE_STACK); \ IF_ASSIGN(var, ent, struct userstack_entry, TRACE_USER_STACK);\ IF_ASSIGN(var, ent, struct print_entry, TRACE_PRINT); \ + IF_ASSIGN(var, ent, struct bprint_entry, TRACE_BPRINT); \ IF_ASSIGN(var, ent, struct special_entry, 0); \ IF_ASSIGN(var, ent, struct trace_mmiotrace_rw, \ TRACE_MMIO_RW); \ @@ -570,6 +579,8 @@ extern int trace_selftest_startup_branch(struct tracer *trace, extern void *head_page(struct trace_array_cpu *data); extern long ns2usecs(cycle_t nsec); extern int +trace_vbprintk(unsigned long ip, int depth, const char *fmt, va_list args); +extern int trace_vprintk(unsigned long ip, int depth, const char *fmt, va_list args); extern unsigned long trace_flags; diff --git a/kernel/trace/trace_event_types.h b/kernel/trace/trace_event_types.h index 5cca4c978bde..d0907d746425 100644 --- a/kernel/trace/trace_event_types.h +++ b/kernel/trace/trace_event_types.h @@ -102,7 +102,7 @@ TRACE_EVENT_FORMAT(user_stack, TRACE_USER_STACK, userstack_entry, ignore, "\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n") ); -TRACE_EVENT_FORMAT(print, TRACE_PRINT, print_entry, ignore, +TRACE_EVENT_FORMAT(bprint, TRACE_PRINT, bprint_entry, ignore, TRACE_STRUCT( TRACE_FIELD(unsigned long, ip, ip) TRACE_FIELD(unsigned int, depth, depth) @@ -112,6 +112,15 @@ TRACE_EVENT_FORMAT(print, TRACE_PRINT, print_entry, ignore, TP_RAW_FMT("%08lx (%d) fmt:%p %s") ); +TRACE_EVENT_FORMAT(print, TRACE_PRINT, print_entry, ignore, + TRACE_STRUCT( + TRACE_FIELD(unsigned long, ip, ip) + TRACE_FIELD(unsigned int, depth, depth) + TRACE_FIELD_ZERO_CHAR(buf) + ), + TP_RAW_FMT("%08lx (%d) fmt:%p %s") +); + TRACE_EVENT_FORMAT(branch, TRACE_BRANCH, trace_branch, ignore, TRACE_STRUCT( TRACE_FIELD(unsigned int, line, line) diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index 8566c14b3e9a..4c388607ed67 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -684,7 +684,7 @@ print_graph_return(struct ftrace_graph_ret *trace, struct trace_seq *s, } static enum print_line_t -print_graph_comment(struct print_entry *trace, struct trace_seq *s, +print_graph_comment(struct bprint_entry *trace, struct trace_seq *s, struct trace_entry *ent, struct trace_iterator *iter) { int i; @@ -781,8 +781,8 @@ print_graph_function(struct trace_iterator *iter) trace_assign_type(field, entry); return print_graph_return(&field->ret, s, entry, iter); } - case TRACE_PRINT: { - struct print_entry *field; + case TRACE_BPRINT: { + struct bprint_entry *field; trace_assign_type(field, entry); return print_graph_comment(field, s, entry, iter); } diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c index 23e346a734ca..f095916e477f 100644 --- a/kernel/trace/trace_mmiotrace.c +++ b/kernel/trace/trace_mmiotrace.c @@ -254,6 +254,7 @@ static enum print_line_t mmio_print_mark(struct trace_iterator *iter) { struct trace_entry *entry = iter->ent; struct print_entry *print = (struct print_entry *)entry; + const char *msg = print->buf; struct trace_seq *s = &iter->seq; unsigned long long t = ns2usecs(iter->ts); unsigned long usec_rem = do_div(t, USEC_PER_SEC); @@ -261,11 +262,7 @@ static enum print_line_t mmio_print_mark(struct trace_iterator *iter) int ret; /* The trailing newline must be in the message. */ - ret = trace_seq_printf(s, "MARK %u.%06lu ", secs, usec_rem); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - - ret = trace_seq_bprintf(s, print->fmt, print->buf); + ret = trace_seq_printf(s, "MARK %u.%06lu %s", secs, usec_rem, msg); if (!ret) return TRACE_TYPE_PARTIAL_LINE; diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index 491832af9ba1..ea9d3b410c7a 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -832,13 +832,13 @@ static struct trace_event trace_user_stack_event = { .binary = trace_special_bin, }; -/* TRACE_PRINT */ +/* TRACE_BPRINT */ static enum print_line_t -trace_print_print(struct trace_iterator *iter, int flags) +trace_bprint_print(struct trace_iterator *iter, int flags) { struct trace_entry *entry = iter->ent; struct trace_seq *s = &iter->seq; - struct print_entry *field; + struct bprint_entry *field; trace_assign_type(field, entry); @@ -858,9 +858,10 @@ trace_print_print(struct trace_iterator *iter, int flags) } -static enum print_line_t trace_print_raw(struct trace_iterator *iter, int flags) +static enum print_line_t +trace_bprint_raw(struct trace_iterator *iter, int flags) { - struct print_entry *field; + struct bprint_entry *field; struct trace_seq *s = &iter->seq; trace_assign_type(field, iter->ent); @@ -878,12 +879,55 @@ static enum print_line_t trace_print_raw(struct trace_iterator *iter, int flags) } +static struct trace_event trace_bprint_event = { + .type = TRACE_BPRINT, + .trace = trace_bprint_print, + .raw = trace_bprint_raw, +}; + +/* TRACE_PRINT */ +static enum print_line_t trace_print_print(struct trace_iterator *iter, + int flags) +{ + struct print_entry *field; + struct trace_seq *s = &iter->seq; + + trace_assign_type(field, iter->ent); + + if (!seq_print_ip_sym(s, field->ip, flags)) + goto partial; + + if (!trace_seq_printf(s, ": %s", field->buf)) + goto partial; + + return TRACE_TYPE_HANDLED; + + partial: + return TRACE_TYPE_PARTIAL_LINE; +} + +static enum print_line_t trace_print_raw(struct trace_iterator *iter, int flags) +{ + struct print_entry *field; + + trace_assign_type(field, iter->ent); + + if (!trace_seq_printf(&iter->seq, "# %lx %s", field->ip, field->buf)) + goto partial; + + return TRACE_TYPE_HANDLED; + + partial: + return TRACE_TYPE_PARTIAL_LINE; +} + static struct trace_event trace_print_event = { - .type = TRACE_PRINT, + .type = TRACE_PRINT, .trace = trace_print_print, .raw = trace_print_raw, }; + static struct trace_event *events[] __initdata = { &trace_fn_event, &trace_ctx_event, @@ -891,6 +935,7 @@ static struct trace_event *events[] __initdata = { &trace_special_event, &trace_stack_event, &trace_user_stack_event, + &trace_bprint_event, &trace_print_event, NULL }; diff --git a/kernel/trace/trace_printk.c b/kernel/trace/trace_printk.c index a50aea22e929..f307a11e2332 100644 --- a/kernel/trace/trace_printk.c +++ b/kernel/trace/trace_printk.c @@ -99,7 +99,7 @@ struct notifier_block module_trace_bprintk_format_nb = { .notifier_call = module_trace_bprintk_format_notify, }; -int __trace_printk(unsigned long ip, const char *fmt, ...) +int __trace_bprintk(unsigned long ip, const char *fmt, ...) { int ret; va_list ap; @@ -111,13 +111,13 @@ int __trace_printk(unsigned long ip, const char *fmt, ...) return 0; va_start(ap, fmt); - ret = trace_vprintk(ip, task_curr_ret_stack(current), fmt, ap); + ret = trace_vbprintk(ip, task_curr_ret_stack(current), fmt, ap); va_end(ap); return ret; } -EXPORT_SYMBOL_GPL(__trace_printk); +EXPORT_SYMBOL_GPL(__trace_bprintk); -int __ftrace_vprintk(unsigned long ip, const char *fmt, va_list ap) +int __ftrace_vbprintk(unsigned long ip, const char *fmt, va_list ap) { if (unlikely(!fmt)) return 0; @@ -125,11 +125,34 @@ int __ftrace_vprintk(unsigned long ip, const char *fmt, va_list ap) if (!(trace_flags & TRACE_ITER_PRINTK)) return 0; + return trace_vbprintk(ip, task_curr_ret_stack(current), fmt, ap); +} +EXPORT_SYMBOL_GPL(__ftrace_vbprintk); + +int __trace_printk(unsigned long ip, const char *fmt, ...) +{ + int ret; + va_list ap; + + if (!(trace_flags & TRACE_ITER_PRINTK)) + return 0; + + va_start(ap, fmt); + ret = trace_vprintk(ip, task_curr_ret_stack(current), fmt, ap); + va_end(ap); + return ret; +} +EXPORT_SYMBOL_GPL(__trace_printk); + +int __ftrace_vprintk(unsigned long ip, const char *fmt, va_list ap) +{ + if (!(trace_flags & TRACE_ITER_PRINTK)) + return 0; + return trace_vprintk(ip, task_curr_ret_stack(current), fmt, ap); } EXPORT_SYMBOL_GPL(__ftrace_vprintk); - static __init int init_trace_printk(void) { return register_module_notifier(&module_trace_bprintk_format_nb); -- cgit v1.2.3 From 04d491ab2a53008a1aa98ac09561768c7f3adda3 Mon Sep 17 00:00:00 2001 From: Neil Horman Date: Thu, 2 Apr 2009 16:58:57 -0700 Subject: kexec: add dmesg log symbols to /proc/vmcoreinfo lists It would be nice to be able to extract the dmesg log from a vmcore file without needing to keep the debug symbols for the running kernel handy all the time. We have a facility to do this in /proc/vmcore. This patch adds the log_buf and log_end symbols to the vmcoreinfo area so that tools (like makedumpfile) can easily extract the dmesg logs from a vmcore image. [akpm@linux-foundation.org: several fixes and cleanups] [akpm@linux-foundation.org: fix unused log_buf_kexec_setup()] [akpm@linux-foundation.org: build fix] Signed-off-by: Neil Horman Cc: Simon Horman Acked-by: Vivek Goyal Cc: Neil Horman Cc: Simon Horman Cc: Vivek Goyal Cc: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kernel.h | 4 ++++ kernel/kexec.c | 1 + kernel/printk.c | 19 +++++++++++++++++++ 3 files changed, 24 insertions(+) (limited to 'include/linux/kernel.h') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index e720b0da7751..556d781e69fe 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -242,6 +242,7 @@ extern struct ratelimit_state printk_ratelimit_state; extern int printk_ratelimit(void); extern bool printk_timed_ratelimit(unsigned long *caller_jiffies, unsigned int interval_msec); +void log_buf_kexec_setup(void); #else static inline int vprintk(const char *s, va_list args) __attribute__ ((format (printf, 1, 0))); @@ -253,6 +254,9 @@ static inline int printk_ratelimit(void) { return 0; } static inline bool printk_timed_ratelimit(unsigned long *caller_jiffies, \ unsigned int interval_msec) \ { return false; } +static inline void log_buf_kexec_setup(void) +{ +} #endif extern int printk_needs_cpu(int cpu); diff --git a/kernel/kexec.c b/kernel/kexec.c index 93eed85fe017..589832aac41f 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -1409,6 +1409,7 @@ static int __init crash_save_vmcoreinfo_init(void) VMCOREINFO_OFFSET(list_head, prev); VMCOREINFO_OFFSET(vm_struct, addr); VMCOREINFO_LENGTH(zone.free_area, MAX_ORDER); + log_buf_kexec_setup(); VMCOREINFO_LENGTH(free_area.free_list, MIGRATE_TYPES); VMCOREINFO_NUMBER(NR_FREE_PAGES); VMCOREINFO_NUMBER(PG_lru); diff --git a/kernel/printk.c b/kernel/printk.c index e3602d0755b0..a5f61a9acedb 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -32,6 +32,7 @@ #include #include #include +#include #include @@ -135,6 +136,24 @@ static char *log_buf = __log_buf; static int log_buf_len = __LOG_BUF_LEN; static unsigned logged_chars; /* Number of chars produced since last read+clear operation */ +#ifdef CONFIG_KEXEC +/* + * This appends the listed symbols to /proc/vmcoreinfo + * + * /proc/vmcoreinfo is used by various utiilties, like crash and makedumpfile to + * obtain access to symbols that are otherwise very difficult to locate. These + * symbols are specifically used so that utilities can access and extract the + * dmesg log from a vmcore file after a crash. + */ +void log_buf_kexec_setup(void) +{ + VMCOREINFO_SYMBOL(log_buf); + VMCOREINFO_SYMBOL(log_end); + VMCOREINFO_SYMBOL(log_buf_len); + VMCOREINFO_SYMBOL(logged_chars); +} +#endif + static int __init log_buf_len_setup(char *str) { unsigned size = memparse(str, &str); -- cgit v1.2.3