diff options
Diffstat (limited to 'kernel')
-rw-r--r-- | kernel/perf_event.c | 80 | ||||
-rw-r--r-- | kernel/trace/ftrace.c | 370 | ||||
-rw-r--r-- | kernel/trace/ring_buffer.c | 15 | ||||
-rw-r--r-- | kernel/trace/trace.c | 4 | ||||
-rw-r--r-- | kernel/trace/trace.h | 48 | ||||
-rw-r--r-- | kernel/trace/trace_events.c | 47 | ||||
-rw-r--r-- | kernel/trace/trace_events_filter.c | 423 | ||||
-rw-r--r-- | kernel/trace/trace_export.c | 25 | ||||
-rw-r--r-- | kernel/trace/trace_syscalls.c | 106 |
9 files changed, 780 insertions, 338 deletions
diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 7f29643c8985..a69d4ed6a666 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -28,6 +28,7 @@ #include <linux/anon_inodes.h> #include <linux/kernel_stat.h> #include <linux/perf_event.h> +#include <linux/ftrace_event.h> #include <asm/irq_regs.h> @@ -1658,6 +1659,8 @@ static struct perf_event_context *find_get_context(pid_t pid, int cpu) return ERR_PTR(err); } +static void perf_event_free_filter(struct perf_event *event); + static void free_event_rcu(struct rcu_head *head) { struct perf_event *event; @@ -1665,6 +1668,7 @@ static void free_event_rcu(struct rcu_head *head) event = container_of(head, struct perf_event, rcu_head); if (event->ns) put_pid_ns(event->ns); + perf_event_free_filter(event); kfree(event); } @@ -1974,7 +1978,8 @@ unlock: return ret; } -int perf_event_set_output(struct perf_event *event, int output_fd); +static int perf_event_set_output(struct perf_event *event, int output_fd); +static int perf_event_set_filter(struct perf_event *event, void __user *arg); static long perf_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { @@ -2002,6 +2007,9 @@ static long perf_ioctl(struct file *file, unsigned int cmd, unsigned long arg) case PERF_EVENT_IOC_SET_OUTPUT: return perf_event_set_output(event, arg); + case PERF_EVENT_IOC_SET_FILTER: + return perf_event_set_filter(event, (void __user *)arg); + default: return -ENOTTY; } @@ -3806,9 +3814,14 @@ static int perf_swevent_is_counting(struct perf_event *event) return 1; } +static int perf_tp_event_match(struct perf_event *event, + struct perf_sample_data *data); + static int perf_swevent_match(struct perf_event *event, enum perf_type_id type, - u32 event_id, struct pt_regs *regs) + u32 event_id, + struct perf_sample_data *data, + struct pt_regs *regs) { if (!perf_swevent_is_counting(event)) return 0; @@ -3826,6 +3839,10 @@ static int perf_swevent_match(struct perf_event *event, return 0; } + if (event->attr.type == PERF_TYPE_TRACEPOINT && + !perf_tp_event_match(event, data)) + return 0; + return 1; } @@ -3842,7 +3859,7 @@ static void perf_swevent_ctx_event(struct perf_event_context *ctx, rcu_read_lock(); list_for_each_entry_rcu(event, &ctx->event_list, event_entry) { - if (perf_swevent_match(event, type, event_id, regs)) + if (perf_swevent_match(event, type, event_id, data, regs)) perf_swevent_add(event, nr, nmi, data, regs); } rcu_read_unlock(); @@ -4108,6 +4125,7 @@ static const struct pmu perf_ops_task_clock = { }; #ifdef CONFIG_EVENT_PROFILE + void perf_tp_event(int event_id, u64 addr, u64 count, void *record, int entry_size) { @@ -4131,8 +4149,15 @@ void perf_tp_event(int event_id, u64 addr, u64 count, void *record, } EXPORT_SYMBOL_GPL(perf_tp_event); -extern int ftrace_profile_enable(int); -extern void ftrace_profile_disable(int); +static int perf_tp_event_match(struct perf_event *event, + struct perf_sample_data *data) +{ + void *record = data->raw->data; + + if (likely(!event->filter) || filter_match_preds(event->filter, record)) + return 1; + return 0; +} static void tp_perf_event_destroy(struct perf_event *event) { @@ -4157,12 +4182,53 @@ static const struct pmu *tp_perf_event_init(struct perf_event *event) return &perf_ops_generic; } + +static int perf_event_set_filter(struct perf_event *event, void __user *arg) +{ + char *filter_str; + int ret; + + if (event->attr.type != PERF_TYPE_TRACEPOINT) + return -EINVAL; + + filter_str = strndup_user(arg, PAGE_SIZE); + if (IS_ERR(filter_str)) + return PTR_ERR(filter_str); + + ret = ftrace_profile_set_filter(event, event->attr.config, filter_str); + + kfree(filter_str); + return ret; +} + +static void perf_event_free_filter(struct perf_event *event) +{ + ftrace_profile_free_filter(event); +} + #else + +static int perf_tp_event_match(struct perf_event *event, + struct perf_sample_data *data) +{ + return 1; +} + static const struct pmu *tp_perf_event_init(struct perf_event *event) { return NULL; } -#endif + +static int perf_event_set_filter(struct perf_event *event, void __user *arg) +{ + return -ENOENT; +} + +static void perf_event_free_filter(struct perf_event *event) +{ +} + +#endif /* CONFIG_EVENT_PROFILE */ atomic_t perf_swevent_enabled[PERF_COUNT_SW_MAX]; @@ -4416,7 +4482,7 @@ err_size: goto out; } -int perf_event_set_output(struct perf_event *event, int output_fd) +static int perf_event_set_output(struct perf_event *event, int output_fd) { struct perf_event *output_event = NULL; struct file *output_file = NULL; diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 9c451a1930b6..1ed514fe3a30 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -60,6 +60,13 @@ static int last_ftrace_enabled; /* Quick disabling of function tracer. */ int function_trace_stop; +/* List for set_ftrace_pid's pids. */ +LIST_HEAD(ftrace_pids); +struct ftrace_pid { + struct list_head list; + struct pid *pid; +}; + /* * ftrace_disabled is set when an anomaly is discovered. * ftrace_disabled is much stronger than ftrace_enabled. @@ -78,6 +85,10 @@ ftrace_func_t ftrace_trace_function __read_mostly = ftrace_stub; ftrace_func_t __ftrace_trace_function __read_mostly = ftrace_stub; ftrace_func_t ftrace_pid_function __read_mostly = ftrace_stub; +#ifdef CONFIG_FUNCTION_GRAPH_TRACER +static int ftrace_set_func(unsigned long *array, int *idx, char *buffer); +#endif + static void ftrace_list_func(unsigned long ip, unsigned long parent_ip) { struct ftrace_ops *op = ftrace_list; @@ -155,7 +166,7 @@ static int __register_ftrace_function(struct ftrace_ops *ops) else func = ftrace_list_func; - if (ftrace_pid_trace) { + if (!list_empty(&ftrace_pids)) { set_ftrace_pid_function(func); func = ftrace_pid_func; } @@ -203,7 +214,7 @@ static int __unregister_ftrace_function(struct ftrace_ops *ops) if (ftrace_list->next == &ftrace_list_end) { ftrace_func_t func = ftrace_list->func; - if (ftrace_pid_trace) { + if (!list_empty(&ftrace_pids)) { set_ftrace_pid_function(func); func = ftrace_pid_func; } @@ -231,7 +242,7 @@ static void ftrace_update_pid_func(void) func = __ftrace_trace_function; #endif - if (ftrace_pid_trace) { + if (!list_empty(&ftrace_pids)) { set_ftrace_pid_function(func); func = ftrace_pid_func; } else { @@ -821,8 +832,6 @@ static __init void ftrace_profile_debugfs(struct dentry *d_tracer) } #endif /* CONFIG_FUNCTION_PROFILER */ -/* set when tracing only a pid */ -struct pid *ftrace_pid_trace; static struct pid * const ftrace_swapper_pid = &init_struct_pid; #ifdef CONFIG_DYNAMIC_FTRACE @@ -1261,12 +1270,34 @@ static int ftrace_update_code(struct module *mod) ftrace_new_addrs = p->newlist; p->flags = 0L; - /* convert record (i.e, patch mcount-call with NOP) */ - if (ftrace_code_disable(mod, p)) { - p->flags |= FTRACE_FL_CONVERTED; - ftrace_update_cnt++; - } else + /* + * Do the initial record convertion from mcount jump + * to the NOP instructions. + */ + if (!ftrace_code_disable(mod, p)) { ftrace_free_rec(p); + continue; + } + + p->flags |= FTRACE_FL_CONVERTED; + ftrace_update_cnt++; + + /* + * If the tracing is enabled, go ahead and enable the record. + * + * The reason not to enable the record immediatelly is the + * inherent check of ftrace_make_nop/ftrace_make_call for + * correct previous instructions. Making first the NOP + * conversion puts the module to the correct state, thus + * passing the ftrace_make_call check. + */ + if (ftrace_start_up) { + int failed = __ftrace_replace_code(p, 1); + if (failed) { + ftrace_bug(failed, p->ip); + ftrace_free_rec(p); + } + } } stop = ftrace_now(raw_smp_processor_id()); @@ -1656,60 +1687,6 @@ ftrace_regex_lseek(struct file *file, loff_t offset, int origin) return ret; } -enum { - MATCH_FULL, - MATCH_FRONT_ONLY, - MATCH_MIDDLE_ONLY, - MATCH_END_ONLY, -}; - -/* - * (static function - no need for kernel doc) - * - * Pass in a buffer containing a glob and this function will - * set search to point to the search part of the buffer and - * return the type of search it is (see enum above). - * This does modify buff. - * - * Returns enum type. - * search returns the pointer to use for comparison. - * not returns 1 if buff started with a '!' - * 0 otherwise. - */ -static int -ftrace_setup_glob(char *buff, int len, char **search, int *not) -{ - int type = MATCH_FULL; - int i; - - if (buff[0] == '!') { - *not = 1; - buff++; - len--; - } else - *not = 0; - - *search = buff; - - for (i = 0; i < len; i++) { - if (buff[i] == '*') { - if (!i) { - *search = buff + 1; - type = MATCH_END_ONLY; - } else { - if (type == MATCH_END_ONLY) - type = MATCH_MIDDLE_ONLY; - else - type = MATCH_FRONT_ONLY; - buff[i] = 0; - break; - } - } - } - - return type; -} - static int ftrace_match(char *str, char *regex, int len, int type) { int matched = 0; @@ -1758,7 +1735,7 @@ static void ftrace_match_records(char *buff, int len, int enable) int not; flag = enable ? FTRACE_FL_FILTER : FTRACE_FL_NOTRACE; - type = ftrace_setup_glob(buff, len, &search, ¬); + type = filter_parse_regex(buff, len, &search, ¬); search_len = strlen(search); @@ -1826,7 +1803,7 @@ static void ftrace_match_module_records(char *buff, char *mod, int enable) } if (strlen(buff)) { - type = ftrace_setup_glob(buff, strlen(buff), &search, ¬); + type = filter_parse_regex(buff, strlen(buff), &search, ¬); search_len = strlen(search); } @@ -1991,7 +1968,7 @@ register_ftrace_function_probe(char *glob, struct ftrace_probe_ops *ops, int count = 0; char *search; - type = ftrace_setup_glob(glob, strlen(glob), &search, ¬); + type = filter_parse_regex(glob, strlen(glob), &search, ¬); len = strlen(search); /* we do not support '!' for function probes */ @@ -2068,7 +2045,7 @@ __unregister_ftrace_function_probe(char *glob, struct ftrace_probe_ops *ops, else if (glob) { int not; - type = ftrace_setup_glob(glob, strlen(glob), &search, ¬); + type = filter_parse_regex(glob, strlen(glob), &search, ¬); len = strlen(search); /* we do not support '!' for function probes */ @@ -2297,6 +2274,7 @@ void ftrace_set_notrace(unsigned char *buf, int len, int reset) #define FTRACE_FILTER_SIZE COMMAND_LINE_SIZE static char ftrace_notrace_buf[FTRACE_FILTER_SIZE] __initdata; static char ftrace_filter_buf[FTRACE_FILTER_SIZE] __initdata; +static char ftrace_graph_buf[FTRACE_FILTER_SIZE] __initdata; static int __init set_ftrace_notrace(char *str) { @@ -2312,6 +2290,31 @@ static int __init set_ftrace_filter(char *str) } __setup("ftrace_filter=", set_ftrace_filter); +#ifdef CONFIG_FUNCTION_GRAPH_TRACER +static int __init set_graph_function(char *str) +{ + strlcpy(ftrace_graph_buf, str, FTRACE_FILTER_SIZE); + return 1; +} +__setup("ftrace_graph_filter=", set_graph_function); + +static void __init set_ftrace_early_graph(char *buf) +{ + int ret; + char *func; + + while (buf) { + func = strsep(&buf, ","); + /* we allow only one expression at a time */ + ret = ftrace_set_func(ftrace_graph_funcs, &ftrace_graph_count, + func); + if (ret) + printk(KERN_DEBUG "ftrace: function %s not " + "traceable\n", func); + } +} +#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ + static void __init set_ftrace_early_filter(char *buf, int enable) { char *func; @@ -2328,6 +2331,10 @@ static void __init set_ftrace_early_filters(void) set_ftrace_early_filter(ftrace_filter_buf, 1); if (ftrace_notrace_buf[0]) set_ftrace_early_filter(ftrace_notrace_buf, 0); +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + if (ftrace_graph_buf[0]) + set_ftrace_early_graph(ftrace_graph_buf); +#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ } static int @@ -2513,7 +2520,7 @@ ftrace_set_func(unsigned long *array, int *idx, char *buffer) return -ENODEV; /* decode regex */ - type = ftrace_setup_glob(buffer, strlen(buffer), &search, ¬); + type = filter_parse_regex(buffer, strlen(buffer), &search, ¬); if (not) return -EINVAL; @@ -2624,7 +2631,7 @@ static __init int ftrace_init_dyn_debugfs(struct dentry *d_tracer) return 0; } -static int ftrace_convert_nops(struct module *mod, +static int ftrace_process_locs(struct module *mod, unsigned long *start, unsigned long *end) { @@ -2684,7 +2691,7 @@ static void ftrace_init_module(struct module *mod, { if (ftrace_disabled || start == end) return; - ftrace_convert_nops(mod, start, end); + ftrace_process_locs(mod, start, end); } static int ftrace_module_notify(struct notifier_block *self, @@ -2745,7 +2752,7 @@ void __init ftrace_init(void) last_ftrace_enabled = ftrace_enabled = 1; - ret = ftrace_convert_nops(NULL, + ret = ftrace_process_locs(NULL, __start_mcount_loc, __stop_mcount_loc); @@ -2778,23 +2785,6 @@ static inline void ftrace_startup_enable(int command) { } # define ftrace_shutdown_sysctl() do { } while (0) #endif /* CONFIG_DYNAMIC_FTRACE */ -static ssize_t -ftrace_pid_read(struct file *file, char __user *ubuf, - size_t cnt, loff_t *ppos) -{ - char buf[64]; - int r; - - if (ftrace_pid_trace == ftrace_swapper_pid) - r = sprintf(buf, "swapper tasks\n"); - else if (ftrace_pid_trace) - r = sprintf(buf, "%u\n", pid_vnr(ftrace_pid_trace)); - else - r = sprintf(buf, "no pid\n"); - - return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); -} - static void clear_ftrace_swapper(void) { struct task_struct *p; @@ -2845,14 +2835,12 @@ static void set_ftrace_pid(struct pid *pid) rcu_read_unlock(); } -static void clear_ftrace_pid_task(struct pid **pid) +static void clear_ftrace_pid_task(struct pid *pid) { - if (*pid == ftrace_swapper_pid) + if (pid == ftrace_swapper_pid) clear_ftrace_swapper(); else - clear_ftrace_pid(*pid); - - *pid = NULL; + clear_ftrace_pid(pid); } static void set_ftrace_pid_task(struct pid *pid) @@ -2863,11 +2851,140 @@ static void set_ftrace_pid_task(struct pid *pid) set_ftrace_pid(pid); } +static int ftrace_pid_add(int p) +{ + struct pid *pid; + struct ftrace_pid *fpid; + int ret = -EINVAL; + + mutex_lock(&ftrace_lock); + + if (!p) + pid = ftrace_swapper_pid; + else + pid = find_get_pid(p); + + if (!pid) + goto out; + + ret = 0; + + list_for_each_entry(fpid, &ftrace_pids, list) + if (fpid->pid == pid) + goto out_put; + + ret = -ENOMEM; + + fpid = kmalloc(sizeof(*fpid), GFP_KERNEL); + if (!fpid) + goto out_put; + + list_add(&fpid->list, &ftrace_pids); + fpid->pid = pid; + + set_ftrace_pid_task(pid); + + ftrace_update_pid_func(); + ftrace_startup_enable(0); + + mutex_unlock(&ftrace_lock); + return 0; + +out_put: + if (pid != ftrace_swapper_pid) + put_pid(pid); + +out: + mutex_unlock(&ftrace_lock); + return ret; +} + +static void ftrace_pid_reset(void) +{ + struct ftrace_pid *fpid, *safe; + + mutex_lock(&ftrace_lock); + list_for_each_entry_safe(fpid, safe, &ftrace_pids, list) { + struct pid *pid = fpid->pid; + + clear_ftrace_pid_task(pid); + + list_del(&fpid->list); + kfree(fpid); + } + + ftrace_update_pid_func(); + ftrace_startup_enable(0); + + mutex_unlock(&ftrace_lock); +} + +static void *fpid_start(struct seq_file *m, loff_t *pos) +{ + mutex_lock(&ftrace_lock); + + if (list_empty(&ftrace_pids) && (!*pos)) + return (void *) 1; + + return seq_list_start(&ftrace_pids, *pos); +} + +static void *fpid_next(struct seq_file *m, void *v, loff_t *pos) +{ + if (v == (void *)1) + return NULL; + + return seq_list_next(v, &ftrace_pids, pos); +} + +static void fpid_stop(struct seq_file *m, void *p) +{ + mutex_unlock(&ftrace_lock); +} + +static int fpid_show(struct seq_file *m, void *v) +{ + const struct ftrace_pid *fpid = list_entry(v, struct ftrace_pid, list); + + if (v == (void *)1) { + seq_printf(m, "no pid\n"); + return 0; + } + + if (fpid->pid == ftrace_swapper_pid) + seq_printf(m, "swapper tasks\n"); + else + seq_printf(m, "%u\n", pid_vnr(fpid->pid)); + + return 0; +} + +static const struct seq_operations ftrace_pid_sops = { + .start = fpid_start, + .next = fpid_next, + .stop = fpid_stop, + .show = fpid_show, +}; + +static int +ftrace_pid_open(struct inode *inode, struct file *file) +{ + int ret = 0; + + if ((file->f_mode & FMODE_WRITE) && + (file->f_flags & O_TRUNC)) + ftrace_pid_reset(); + + if (file->f_mode & FMODE_READ) + ret = seq_open(file, &ftrace_pid_sops); + + return ret; +} + static ssize_t ftrace_pid_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { - struct pid *pid; char buf[64]; long val; int ret; @@ -2880,57 +2997,38 @@ ftrace_pid_write(struct file *filp, const char __user *ubuf, buf[cnt] = 0; + /* + * Allow "echo > set_ftrace_pid" or "echo -n '' > set_ftrace_pid" + * to clean the filter quietly. + */ + strstrip(buf); + if (strlen(buf) == 0) + return 1; + ret = strict_strtol(buf, 10, &val); if (ret < 0) return ret; - mutex_lock(&ftrace_lock); - if (val < 0) { - /* disable pid tracing */ - if (!ftrace_pid_trace) - goto out; - - clear_ftrace_pid_task(&ftrace_pid_trace); - - } else { - /* swapper task is special */ - if (!val) { - pid = ftrace_swapper_pid; - if (pid == ftrace_pid_trace) - goto out; - } else { - pid = find_get_pid(val); + ret = ftrace_pid_add(val); - if (pid == ftrace_pid_trace) { - put_pid(pid); - goto out; - } - } - - if (ftrace_pid_trace) - clear_ftrace_pid_task(&ftrace_pid_trace); - - if (!pid) - goto out; - - ftrace_pid_trace = pid; - - set_ftrace_pid_task(ftrace_pid_trace); - } - - /* update the function call */ - ftrace_update_pid_func(); - ftrace_startup_enable(0); + return ret ? ret : cnt; +} - out: - mutex_unlock(&ftrace_lock); +static int +ftrace_pid_release(struct inode *inode, struct file *file) +{ + if (file->f_mode & FMODE_READ) + seq_release(inode, file); - return cnt; + return 0; } static const struct file_operations ftrace_pid_fops = { - .read = ftrace_pid_read, - .write = ftrace_pid_write, + .open = ftrace_pid_open, + .write = ftrace_pid_write, + .read = seq_read, + .llseek = seq_lseek, + .release = ftrace_pid_release, }; static __init int ftrace_init_debugfs(void) diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 3ffa502fb243..63446f12e470 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -397,18 +397,21 @@ int ring_buffer_print_page_header(struct trace_seq *s) int ret; ret = trace_seq_printf(s, "\tfield: u64 timestamp;\t" - "offset:0;\tsize:%u;\n", - (unsigned int)sizeof(field.time_stamp)); + "offset:0;\tsize:%u;\tsigned:%u;\n", + (unsigned int)sizeof(field.time_stamp), + (unsigned int)is_signed_type(u64)); ret = trace_seq_printf(s, "\tfield: local_t commit;\t" - "offset:%u;\tsize:%u;\n", + "offset:%u;\tsize:%u;\tsigned:%u;\n", (unsigned int)offsetof(typeof(field), commit), - (unsigned int)sizeof(field.commit)); + (unsigned int)sizeof(field.commit), + (unsigned int)is_signed_type(long)); ret = trace_seq_printf(s, "\tfield: char data;\t" - "offset:%u;\tsize:%u;\n", + "offset:%u;\tsize:%u;\tsigned:%u;\n", (unsigned int)offsetof(typeof(field), data), - (unsigned int)BUF_PAGE_SIZE); + (unsigned int)BUF_PAGE_SIZE, + (unsigned int)is_signed_type(char)); return ret; } diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index b20d3ec75de9..9d3067a62d43 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -129,7 +129,7 @@ static int tracing_set_tracer(const char *buf); static char bootup_tracer_buf[MAX_TRACER_SIZE] __initdata; static char *default_bootup_tracer; -static int __init set_ftrace(char *str) +static int __init set_cmdline_ftrace(char *str) { strncpy(bootup_tracer_buf, str, MAX_TRACER_SIZE); default_bootup_tracer = bootup_tracer_buf; @@ -137,7 +137,7 @@ static int __init set_ftrace(char *str) ring_buffer_expanded = 1; return 1; } -__setup("ftrace=", set_ftrace); +__setup("ftrace=", set_cmdline_ftrace); static int __init set_ftrace_dump_on_oops(char *str) { diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 405cb850b75d..4959ada9e0bb 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -483,10 +483,6 @@ static inline int ftrace_graph_addr(unsigned long addr) return 0; } #else -static inline int ftrace_trace_addr(unsigned long addr) -{ - return 1; -} static inline int ftrace_graph_addr(unsigned long addr) { return 1; @@ -500,12 +496,12 @@ print_graph_function(struct trace_iterator *iter) } #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ -extern struct pid *ftrace_pid_trace; +extern struct list_head ftrace_pids; #ifdef CONFIG_FUNCTION_TRACER static inline int ftrace_trace_task(struct task_struct *task) { - if (!ftrace_pid_trace) + if (list_empty(&ftrace_pids)) return 1; return test_tsk_trace_trace(task); @@ -687,7 +683,6 @@ struct event_filter { int n_preds; struct filter_pred **preds; char *filter_string; - bool no_reset; }; struct event_subsystem { @@ -699,22 +694,40 @@ struct event_subsystem { }; struct filter_pred; +struct regex; typedef int (*filter_pred_fn_t) (struct filter_pred *pred, void *event, int val1, int val2); +typedef int (*regex_match_func)(char *str, struct regex *r, int len); + +enum regex_type { + MATCH_FULL = 0, + MATCH_FRONT_ONLY, + MATCH_MIDDLE_ONLY, + MATCH_END_ONLY, +}; + +struct regex { + char pattern[MAX_FILTER_STR_VAL]; + int len; + int field_len; + regex_match_func match; +}; + struct filter_pred { - filter_pred_fn_t fn; - u64 val; - char str_val[MAX_FILTER_STR_VAL]; - int str_len; - char *field_name; - int offset; - int not; - int op; - int pop_n; + filter_pred_fn_t fn; + u64 val; + struct regex regex; + char *field_name; + int offset; + int not; + int op; + int pop_n; }; +extern enum regex_type +filter_parse_regex(char *buff, int len, char **search, int *not); extern void print_event_filter(struct ftrace_event_call *call, struct trace_seq *s); extern int apply_event_filter(struct ftrace_event_call *call, @@ -730,7 +743,8 @@ filter_check_discard(struct ftrace_event_call *call, void *rec, struct ring_buffer *buffer, struct ring_buffer_event *event) { - if (unlikely(call->filter_active) && !filter_match_preds(call, rec)) { + if (unlikely(call->filter_active) && + !filter_match_preds(call->filter, rec)) { ring_buffer_discard_commit(buffer, event); return 1; } diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index d128f65778e6..7c18d154ea28 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -507,7 +507,7 @@ extern char *__bad_type_size(void); #define FIELD(type, name) \ sizeof(type) != sizeof(field.name) ? __bad_type_size() : \ #type, "common_" #name, offsetof(typeof(field), name), \ - sizeof(field.name) + sizeof(field.name), is_signed_type(type) static int trace_write_header(struct trace_seq *s) { @@ -515,17 +515,17 @@ static int trace_write_header(struct trace_seq *s) /* struct trace_entry */ return trace_seq_printf(s, - "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\n" - "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\n" - "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\n" - "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\n" - "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\n" - "\n", - FIELD(unsigned short, type), - FIELD(unsigned char, flags), - FIELD(unsigned char, preempt_count), - FIELD(int, pid), - FIELD(int, lock_depth)); + "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\tsigned:%u;\n" + "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\tsigned:%u;\n" + "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\tsigned:%u;\n" + "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\tsigned:%u;\n" + "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\tsigned:%u;\n" + "\n", + FIELD(unsigned short, type), + FIELD(unsigned char, flags), + FIELD(unsigned char, preempt_count), + FIELD(int, pid), + FIELD(int, lock_depth)); } static ssize_t @@ -878,9 +878,9 @@ event_subsystem_dir(const char *name, struct dentry *d_events) "'%s/filter' entry\n", name); } - entry = trace_create_file("enable", 0644, system->entry, - (void *)system->name, - &ftrace_system_enable_fops); + trace_create_file("enable", 0644, system->entry, + (void *)system->name, + &ftrace_system_enable_fops); return system->entry; } @@ -892,7 +892,6 @@ event_create_dir(struct ftrace_event_call *call, struct dentry *d_events, const struct file_operations *filter, const struct file_operations *format) { - struct dentry *entry; int ret; /* @@ -910,12 +909,12 @@ event_create_dir(struct ftrace_event_call *call, struct dentry *d_events, } if (call->regfunc) - entry = trace_create_file("enable", 0644, call->dir, call, - enable); + trace_create_file("enable", 0644, call->dir, call, + enable); if (call->id && call->profile_enable) - entry = trace_create_file("id", 0444, call->dir, call, - id); + trace_create_file("id", 0444, call->dir, call, + id); if (call->define_fields) { ret = call->define_fields(call); @@ -924,16 +923,16 @@ event_create_dir(struct ftrace_event_call *call, struct dentry *d_events, " events/%s\n", call->name); return ret; } - entry = trace_create_file("filter", 0644, call->dir, call, - filter); + trace_create_file("filter", 0644, call->dir, call, + filter); } /* A trace may not want to export its format */ if (!call->show_format) return 0; - entry = trace_create_file("format", 0444, call->dir, call, - format); + trace_create_file("format", 0444, call->dir, call, + format); return 0; } diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 98a6cc5c64ed..50504cb228de 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -18,11 +18,10 @@ * Copyright (C) 2009 Tom Zanussi <tzanussi@gmail.com> */ -#include <linux/debugfs.h> -#include <linux/uaccess.h> #include <linux/module.h> #include <linux/ctype.h> #include <linux/mutex.h> +#include <linux/perf_event.h> #include "trace.h" #include "trace_output.h" @@ -31,6 +30,7 @@ enum filter_op_ids { OP_OR, OP_AND, + OP_GLOB, OP_NE, OP_EQ, OP_LT, @@ -48,16 +48,17 @@ struct filter_op { }; static struct filter_op filter_ops[] = { - { OP_OR, "||", 1 }, - { OP_AND, "&&", 2 }, - { OP_NE, "!=", 4 }, - { OP_EQ, "==", 4 }, - { OP_LT, "<", 5 }, - { OP_LE, "<=", 5 }, - { OP_GT, ">", 5 }, - { OP_GE, ">=", 5 }, - { OP_NONE, "OP_NONE", 0 }, - { OP_OPEN_PAREN, "(", 0 }, + { OP_OR, "||", 1 }, + { OP_AND, "&&", 2 }, + { OP_GLOB, "~", 4 }, + { OP_NE, "!=", 4 }, + { OP_EQ, "==", 4 }, + { OP_LT, "<", 5 }, + { OP_LE, "<=", 5 }, + { OP_GT, ">", 5 }, + { OP_GE, ">=", 5 }, + { OP_NONE, "OP_NONE", 0 }, + { OP_OPEN_PAREN, "(", 0 }, }; enum { @@ -197,9 +198,9 @@ static int filter_pred_string(struct filter_pred *pred, void *event, char *addr = (char *)(event + pred->offset); int cmp, match; - cmp = strncmp(addr, pred->str_val, pred->str_len); + cmp = pred->regex.match(addr, &pred->regex, pred->regex.field_len); - match = (!cmp) ^ pred->not; + match = cmp ^ pred->not; return match; } @@ -211,9 +212,9 @@ static int filter_pred_pchar(struct filter_pred *pred, void *event, char **addr = (char **)(event + pred->offset); int cmp, match; - cmp = strncmp(*addr, pred->str_val, pred->str_len); + cmp = pred->regex.match(*addr, &pred->regex, pred->regex.field_len); - match = (!cmp) ^ pred->not; + match = cmp ^ pred->not; return match; } @@ -237,9 +238,9 @@ static int filter_pred_strloc(struct filter_pred *pred, void *event, char *addr = (char *)(event + str_loc); int cmp, match; - cmp = strncmp(addr, pred->str_val, str_len); + cmp = pred->regex.match(addr, &pred->regex, str_len); - match = (!cmp) ^ pred->not; + match = cmp ^ pred->not; return match; } @@ -250,10 +251,121 @@ static int filter_pred_none(struct filter_pred *pred, void *event, return 0; } +/* Basic regex callbacks */ +static int regex_match_full(char *str, struct regex *r, int len) +{ + if (strncmp(str, r->pattern, len) == 0) + return 1; + return 0; +} + +static int regex_match_front(char *str, struct regex *r, int len) +{ + if (strncmp(str, r->pattern, len) == 0) + return 1; + return 0; +} + +static int regex_match_middle(char *str, struct regex *r, int len) +{ + if (strstr(str, r->pattern)) + return 1; + return 0; +} + +static int regex_match_end(char *str, struct regex *r, int len) +{ + char *ptr = strstr(str, r->pattern); + + if (ptr && (ptr[r->len] == 0)) + return 1; + return 0; +} + +/** + * filter_parse_regex - parse a basic regex + * @buff: the raw regex + * @len: length of the regex + * @search: will point to the beginning of the string to compare + * @not: tell whether the match will have to be inverted + * + * This passes in a buffer containing a regex and this function will + * set search to point to the search part of the buffer and + * return the type of search it is (see enum above). + * This does modify buff. + * + * Returns enum type. + * search returns the pointer to use for comparison. + * not returns 1 if buff started with a '!' + * 0 otherwise. + */ +enum regex_type filter_parse_regex(char *buff, int len, char **search, int *not) +{ + int type = MATCH_FULL; + int i; + + if (buff[0] == '!') { + *not = 1; + buff++; + len--; + } else + *not = 0; + + *search = buff; + + for (i = 0; i < len; i++) { + if (buff[i] == '*') { + if (!i) { + *search = buff + 1; + type = MATCH_END_ONLY; + } else { + if (type == MATCH_END_ONLY) + type = MATCH_MIDDLE_ONLY; + else + type = MATCH_FRONT_ONLY; + buff[i] = 0; + break; + } + } + } + + return type; +} + +static void filter_build_regex(struct filter_pred *pred) +{ + struct regex *r = &pred->regex; + char *search; + enum regex_type type = MATCH_FULL; + int not = 0; + + if (pred->op == OP_GLOB) { + type = filter_parse_regex(r->pattern, r->len, &search, ¬); + r->len = strlen(search); + memmove(r->pattern, search, r->len+1); + } + + switch (type) { + case MATCH_FULL: + r->match = regex_match_full; + break; + case MATCH_FRONT_ONLY: + r->match = regex_match_front; + break; + case MATCH_MIDDLE_ONLY: + r->match = regex_match_middle; + break; + case MATCH_END_ONLY: + r->match = regex_match_end; + break; + } + + pred->not ^= not; +} + /* return 1 if event matches, 0 otherwise (discard) */ -int filter_match_preds(struct ftrace_event_call *call, void *rec) +int filter_match_preds(struct event_filter *filter, void *rec) { - struct event_filter *filter = call->filter; int match, top = 0, val1 = 0, val2 = 0; int stack[MAX_FILTER_PRED]; struct filter_pred *pred; @@ -396,7 +508,7 @@ static void filter_clear_pred(struct filter_pred *pred) { kfree(pred->field_name); pred->field_name = NULL; - pred->str_len = 0; + pred->regex.len = 0; } static int filter_set_pred(struct filter_pred *dest, @@ -426,9 +538,8 @@ static void filter_disable_preds(struct ftrace_event_call *call) filter->preds[i]->fn = filter_pred_none; } -void destroy_preds(struct ftrace_event_call *call) +static void __free_preds(struct event_filter *filter) { - struct event_filter *filter = call->filter; int i; if (!filter) @@ -441,21 +552,24 @@ void destroy_preds(struct ftrace_event_call *call) kfree(filter->preds); kfree(filter->filter_string); kfree(filter); +} + +void destroy_preds(struct ftrace_event_call *call) +{ + __free_preds(call->filter); call->filter = NULL; + call->filter_active = 0; } -static int init_preds(struct ftrace_event_call *call) +static struct event_filter *__alloc_preds(void) { struct event_filter *filter; struct filter_pred *pred; int i; - if (call->filter) - return 0; - - filter = call->filter = kzalloc(sizeof(*filter), GFP_KERNEL); - if (!call->filter) - return -ENOMEM; + filter = kzalloc(sizeof(*filter), GFP_KERNEL); + if (!filter) + return ERR_PTR(-ENOMEM); filter->n_preds = 0; @@ -471,12 +585,24 @@ static int init_preds(struct ftrace_event_call *call) filter->preds[i] = pred; } - return 0; + return filter; oom: - destroy_preds(call); + __free_preds(filter); + return ERR_PTR(-ENOMEM); +} - return -ENOMEM; +static int init_preds(struct ftrace_event_call *call) +{ + if (call->filter) + return 0; + + call->filter_active = 0; + call->filter = __alloc_preds(); + if (IS_ERR(call->filter)) + return PTR_ERR(call->filter); + + return 0; } static int init_subsystem_preds(struct event_subsystem *system) @@ -499,14 +625,7 @@ static int init_subsystem_preds(struct event_subsystem *system) return 0; } -enum { - FILTER_DISABLE_ALL, - FILTER_INIT_NO_RESET, - FILTER_SKIP_NO_RESET, -}; - -static void filter_free_subsystem_preds(struct event_subsystem *system, - int flag) +static void filter_free_subsystem_preds(struct event_subsystem *system) { struct ftrace_event_call *call; @@ -517,14 +636,6 @@ static void filter_free_subsystem_preds(struct event_subsystem *system, if (strcmp(call->system, system->name) != 0) continue; - if (flag == FILTER_INIT_NO_RESET) { - call->filter->no_reset = false; - continue; - } - - if (flag == FILTER_SKIP_NO_RESET && call->filter->no_reset) - continue; - filter_disable_preds(call); remove_filter_string(call->filter); } @@ -532,10 +643,10 @@ static void filter_free_subsystem_preds(struct event_subsystem *system, static int filter_add_pred_fn(struct filter_parse_state *ps, struct ftrace_event_call *call, + struct event_filter *filter, struct filter_pred *pred, filter_pred_fn_t fn) { - struct event_filter *filter = call->filter; int idx, err; if (filter->n_preds == MAX_FILTER_PRED) { @@ -550,7 +661,6 @@ static int filter_add_pred_fn(struct filter_parse_state *ps, return err; filter->n_preds++; - call->filter_active = 1; return 0; } @@ -575,7 +685,10 @@ static bool is_string_field(struct ftrace_event_field *field) static int is_legal_op(struct ftrace_event_field *field, int op) { - if (is_string_field(field) && (op != OP_EQ && op != OP_NE)) + if (is_string_field(field) && + (op != OP_EQ && op != OP_NE && op != OP_GLOB)) + return 0; + if (!is_string_field(field) && op == OP_GLOB) return 0; return 1; @@ -626,6 +739,7 @@ static filter_pred_fn_t select_comparison_fn(int op, int field_size, static int filter_add_pred(struct filter_parse_state *ps, struct ftrace_event_call *call, + struct event_filter *filter, struct filter_pred *pred, bool dry_run) { @@ -660,21 +774,22 @@ static int filter_add_pred(struct filter_parse_state *ps, } if (is_string_field(field)) { - pred->str_len = field->size; + filter_build_regex(pred); - if (field->filter_type == FILTER_STATIC_STRING) + if (field->filter_type == FILTER_STATIC_STRING) { fn = filter_pred_string; - else if (field->filter_type == FILTER_DYN_STRING) + pred->regex.field_len = field->size; + } else if (field->filter_type == FILTER_DYN_STRING) fn = filter_pred_strloc; else { fn = filter_pred_pchar; - pred->str_len = strlen(pred->str_val); + pred->regex.field_len = strlen(pred->regex.pattern); } } else { if (field->is_signed) - ret = strict_strtoll(pred->str_val, 0, &val); + ret = strict_strtoll(pred->regex.pattern, 0, &val); else - ret = strict_strtoull(pred->str_val, 0, &val); + ret = strict_strtoull(pred->regex.pattern, 0, &val); if (ret) { parse_error(ps, FILT_ERR_ILLEGAL_INTVAL, 0); return -EINVAL; @@ -694,45 +809,7 @@ static int filter_add_pred(struct filter_parse_state *ps, add_pred_fn: if (!dry_run) - return filter_add_pred_fn(ps, call, pred, fn); - return 0; -} - -static int filter_add_subsystem_pred(struct filter_parse_state *ps, - struct event_subsystem *system, - struct filter_pred *pred, - char *filter_string, - bool dry_run) -{ - struct ftrace_event_call *call; - int err = 0; - bool fail = true; - - list_for_each_entry(call, &ftrace_events, list) { - - if (!call->define_fields) - continue; - - if (strcmp(call->system, system->name)) - continue; - - if (call->filter->no_reset) - continue; - - err = filter_add_pred(ps, call, pred, dry_run); - if (err) - call->filter->no_reset = true; - else - fail = false; - - if (!dry_run) - replace_filter_string(call->filter, filter_string); - } - - if (fail) { - parse_error(ps, FILT_ERR_BAD_SUBSYS_FILTER, 0); - return err; - } + return filter_add_pred_fn(ps, call, filter, pred, fn); return 0; } @@ -1045,8 +1122,8 @@ static struct filter_pred *create_pred(int op, char *operand1, char *operand2) return NULL; } - strcpy(pred->str_val, operand2); - pred->str_len = strlen(operand2); + strcpy(pred->regex.pattern, operand2); + pred->regex.len = strlen(pred->regex.pattern); pred->op = op; @@ -1090,8 +1167,8 @@ static int check_preds(struct filter_parse_state *ps) return 0; } -static int replace_preds(struct event_subsystem *system, - struct ftrace_event_call *call, +static int replace_preds(struct ftrace_event_call *call, + struct event_filter *filter, struct filter_parse_state *ps, char *filter_string, bool dry_run) @@ -1138,11 +1215,7 @@ static int replace_preds(struct event_subsystem *system, add_pred: if (!pred) return -ENOMEM; - if (call) - err = filter_add_pred(ps, call, pred, false); - else - err = filter_add_subsystem_pred(ps, system, pred, - filter_string, dry_run); + err = filter_add_pred(ps, call, filter, pred, dry_run); filter_free_pred(pred); if (err) return err; @@ -1153,10 +1226,50 @@ add_pred: return 0; } -int apply_event_filter(struct ftrace_event_call *call, char *filter_string) +static int replace_system_preds(struct event_subsystem *system, + struct filter_parse_state *ps, + char *filter_string) { + struct ftrace_event_call *call; + bool fail = true; int err; + list_for_each_entry(call, &ftrace_events, list) { + struct event_filter *filter = call->filter; + + if (!call->define_fields) + continue; + + if (strcmp(call->system, system->name) != 0) + continue; + + /* try to see if the filter can be applied */ + err = replace_preds(call, filter, ps, filter_string, true); + if (err) + continue; + + /* really apply the filter */ + filter_disable_preds(call); + err = replace_preds(call, filter, ps, filter_string, false); + if (err) + filter_disable_preds(call); + else { + call->filter_active = 1; + replace_filter_string(filter, filter_string); + } + fail = false; + } + + if (fail) { + parse_error(ps, FILT_ERR_BAD_SUBSYS_FILTER, 0); + return -EINVAL; + } + return 0; +} + +int apply_event_filter(struct ftrace_event_call *call, char *filter_string) +{ + int err; struct filter_parse_state *ps; mutex_lock(&event_mutex); @@ -1168,8 +1281,7 @@ int apply_event_filter(struct ftrace_event_call *call, char *filter_string) if (!strcmp(strstrip(filter_string), "0")) { filter_disable_preds(call); remove_filter_string(call->filter); - mutex_unlock(&event_mutex); - return 0; + goto out_unlock; } err = -ENOMEM; @@ -1187,10 +1299,11 @@ int apply_event_filter(struct ftrace_event_call *call, char *filter_string) goto out; } - err = replace_preds(NULL, call, ps, filter_string, false); + err = replace_preds(call, call->filter, ps, filter_string, false); if (err) append_filter_err(ps, call->filter); - + else + call->filter_active = 1; out: filter_opstack_clear(ps); postfix_clear(ps); @@ -1205,7 +1318,6 @@ int apply_subsystem_event_filter(struct event_subsystem *system, char *filter_string) { int err; - struct filter_parse_state *ps; mutex_lock(&event_mutex); @@ -1215,10 +1327,9 @@ int apply_subsystem_event_filter(struct event_subsystem *system, goto out_unlock; if (!strcmp(strstrip(filter_string), "0")) { - filter_free_subsystem_preds(system, FILTER_DISABLE_ALL); + filter_free_subsystem_preds(system); remove_filter_string(system->filter); - mutex_unlock(&event_mutex); - return 0; + goto out_unlock; } err = -ENOMEM; @@ -1235,31 +1346,87 @@ int apply_subsystem_event_filter(struct event_subsystem *system, goto out; } - filter_free_subsystem_preds(system, FILTER_INIT_NO_RESET); - - /* try to see the filter can be applied to which events */ - err = replace_preds(system, NULL, ps, filter_string, true); - if (err) { + err = replace_system_preds(system, ps, filter_string); + if (err) append_filter_err(ps, system->filter); - goto out; + +out: + filter_opstack_clear(ps); + postfix_clear(ps); + kfree(ps); +out_unlock: + mutex_unlock(&event_mutex); + + return err; +} + +#ifdef CONFIG_EVENT_PROFILE + +void ftrace_profile_free_filter(struct perf_event *event) +{ + struct event_filter *filter = event->filter; + + event->filter = NULL; + __free_preds(filter); +} + +int ftrace_profile_set_filter(struct perf_event *event, int event_id, + char *filter_str) +{ + int err; + struct event_filter *filter; + struct filter_parse_state *ps; + struct ftrace_event_call *call = NULL; + + mutex_lock(&event_mutex); + + list_for_each_entry(call, &ftrace_events, list) { + if (call->id == event_id) + break; } - filter_free_subsystem_preds(system, FILTER_SKIP_NO_RESET); + err = -EINVAL; + if (!call) + goto out_unlock; - /* really apply the filter to the events */ - err = replace_preds(system, NULL, ps, filter_string, false); - if (err) { - append_filter_err(ps, system->filter); - filter_free_subsystem_preds(system, 2); + err = -EEXIST; + if (event->filter) + goto out_unlock; + + filter = __alloc_preds(); + if (IS_ERR(filter)) { + err = PTR_ERR(filter); + goto out_unlock; } -out: + err = -ENOMEM; + ps = kzalloc(sizeof(*ps), GFP_KERNEL); + if (!ps) + goto free_preds; + + parse_init(ps, filter_ops, filter_str); + err = filter_parse(ps); + if (err) + goto free_ps; + + err = replace_preds(call, filter, ps, filter_str, false); + if (!err) + event->filter = filter; + +free_ps: filter_opstack_clear(ps); postfix_clear(ps); kfree(ps); + +free_preds: + if (err) + __free_preds(filter); + out_unlock: mutex_unlock(&event_mutex); return err; } +#endif /* CONFIG_EVENT_PROFILE */ + diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c index 9753fcc61bc5..31da218ee10f 100644 --- a/kernel/trace/trace_export.c +++ b/kernel/trace/trace_export.c @@ -66,44 +66,47 @@ static void __used ____ftrace_check_##name(void) \ #undef __field #define __field(type, item) \ ret = trace_seq_printf(s, "\tfield:" #type " " #item ";\t" \ - "offset:%zu;\tsize:%zu;\n", \ + "offset:%zu;\tsize:%zu;\tsigned:%u;\n", \ offsetof(typeof(field), item), \ - sizeof(field.item)); \ + sizeof(field.item), is_signed_type(type)); \ if (!ret) \ return 0; #undef __field_desc #define __field_desc(type, container, item) \ ret = trace_seq_printf(s, "\tfield:" #type " " #item ";\t" \ - "offset:%zu;\tsize:%zu;\n", \ + "offset:%zu;\tsize:%zu;\tsigned:%u;\n", \ offsetof(typeof(field), container.item), \ - sizeof(field.container.item)); \ + sizeof(field.container.item), \ + is_signed_type(type)); \ if (!ret) \ return 0; #undef __array #define __array(type, item, len) \ ret = trace_seq_printf(s, "\tfield:" #type " " #item "[" #len "];\t" \ - "offset:%zu;\tsize:%zu;\n", \ - offsetof(typeof(field), item), \ - sizeof(field.item)); \ + "offset:%zu;\tsize:%zu;\tsigned:%u;\n", \ + offsetof(typeof(field), item), \ + sizeof(field.item), is_signed_type(type)); \ if (!ret) \ return 0; #undef __array_desc #define __array_desc(type, container, item, len) \ ret = trace_seq_printf(s, "\tfield:" #type " " #item "[" #len "];\t" \ - "offset:%zu;\tsize:%zu;\n", \ + "offset:%zu;\tsize:%zu;\tsigned:%u;\n", \ offsetof(typeof(field), container.item), \ - sizeof(field.container.item)); \ + sizeof(field.container.item), \ + is_signed_type(type)); \ if (!ret) \ return 0; #undef __dynamic_array #define __dynamic_array(type, item) \ ret = trace_seq_printf(s, "\tfield:" #type " " #item ";\t" \ - "offset:%zu;\tsize:0;\n", \ - offsetof(typeof(field), item)); \ + "offset:%zu;\tsize:0;\tsigned:%u;\n", \ + offsetof(typeof(field), item), \ + is_signed_type(type)); \ if (!ret) \ return 0; diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 527e17eae575..d00d1a8f1f26 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -14,6 +14,69 @@ static int sys_refcount_exit; static DECLARE_BITMAP(enabled_enter_syscalls, NR_syscalls); static DECLARE_BITMAP(enabled_exit_syscalls, NR_syscalls); +extern unsigned long __start_syscalls_metadata[]; +extern unsigned long __stop_syscalls_metadata[]; + +static struct syscall_metadata **syscalls_metadata; + +static struct syscall_metadata *find_syscall_meta(unsigned long syscall) +{ + struct syscall_metadata *start; + struct syscall_metadata *stop; + char str[KSYM_SYMBOL_LEN]; + + + start = (struct syscall_metadata *)__start_syscalls_metadata; + stop = (struct syscall_metadata *)__stop_syscalls_metadata; + kallsyms_lookup(syscall, NULL, NULL, NULL, str); + + for ( ; start < stop; start++) { + /* + * Only compare after the "sys" prefix. Archs that use + * syscall wrappers may have syscalls symbols aliases prefixed + * with "SyS" instead of "sys", leading to an unwanted + * mismatch. + */ + if (start->name && !strcmp(start->name + 3, str + 3)) + return start; + } + return NULL; +} + +static struct syscall_metadata *syscall_nr_to_meta(int nr) +{ + if (!syscalls_metadata || nr >= NR_syscalls || nr < 0) + return NULL; + + return syscalls_metadata[nr]; +} + +int syscall_name_to_nr(char *name) +{ + int i; + + if (!syscalls_metadata) + return -1; + + for (i = 0; i < NR_syscalls; i++) { + if (syscalls_metadata[i]) { + if (!strcmp(syscalls_metadata[i]->name, name)) + return i; + } + } + return -1; +} + +void set_syscall_enter_id(int num, int id) +{ + syscalls_metadata[num]->enter_id = id; +} + +void set_syscall_exit_id(int num, int id) +{ + syscalls_metadata[num]->exit_id = id; +} + enum print_line_t print_syscall_enter(struct trace_iterator *iter, int flags) { @@ -103,7 +166,8 @@ extern char *__bad_type_size(void); #define SYSCALL_FIELD(type, name) \ sizeof(type) != sizeof(trace.name) ? \ __bad_type_size() : \ - #type, #name, offsetof(typeof(trace), name), sizeof(trace.name) + #type, #name, offsetof(typeof(trace), name), \ + sizeof(trace.name), is_signed_type(type) int syscall_enter_format(struct ftrace_event_call *call, struct trace_seq *s) { @@ -120,7 +184,8 @@ int syscall_enter_format(struct ftrace_event_call *call, struct trace_seq *s) if (!entry) return 0; - ret = trace_seq_printf(s, "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\n", + ret = trace_seq_printf(s, "\tfield:%s %s;\toffset:%zu;\tsize:%zu;" + "\tsigned:%u;\n", SYSCALL_FIELD(int, nr)); if (!ret) return 0; @@ -130,8 +195,10 @@ int syscall_enter_format(struct ftrace_event_call *call, struct trace_seq *s) entry->args[i]); if (!ret) return 0; - ret = trace_seq_printf(s, "\toffset:%d;\tsize:%zu;\n", offset, - sizeof(unsigned long)); + ret = trace_seq_printf(s, "\toffset:%d;\tsize:%zu;" + "\tsigned:%u;\n", offset, + sizeof(unsigned long), + is_signed_type(unsigned long)); if (!ret) return 0; offset += sizeof(unsigned long); @@ -163,8 +230,10 @@ int syscall_exit_format(struct ftrace_event_call *call, struct trace_seq *s) struct syscall_trace_exit trace; ret = trace_seq_printf(s, - "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\n" - "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\n", + "\tfield:%s %s;\toffset:%zu;\tsize:%zu;" + "\tsigned:%u;\n" + "\tfield:%s %s;\toffset:%zu;\tsize:%zu;" + "\tsigned:%u;\n", SYSCALL_FIELD(int, nr), SYSCALL_FIELD(long, ret)); if (!ret) @@ -212,7 +281,7 @@ int syscall_exit_define_fields(struct ftrace_event_call *call) if (ret) return ret; - ret = trace_define_field(call, SYSCALL_FIELD(long, ret), 0, + ret = trace_define_field(call, SYSCALL_FIELD(long, ret), FILTER_OTHER); return ret; @@ -375,6 +444,29 @@ struct trace_event event_syscall_exit = { .trace = print_syscall_exit, }; +int __init init_ftrace_syscalls(void) +{ + struct syscall_metadata *meta; + unsigned long addr; + int i; + + syscalls_metadata = kzalloc(sizeof(*syscalls_metadata) * + NR_syscalls, GFP_KERNEL); + if (!syscalls_metadata) { + WARN_ON(1); + return -ENOMEM; + } + + for (i = 0; i < NR_syscalls; i++) { + addr = arch_syscall_addr(i); + meta = find_syscall_meta(addr); + syscalls_metadata[i] = meta; + } + + return 0; +} +core_initcall(init_ftrace_syscalls); + #ifdef CONFIG_EVENT_PROFILE static DECLARE_BITMAP(enabled_prof_enter_syscalls, NR_syscalls); |