diff options
Diffstat (limited to 'kernel/trace/trace_kprobe.c')
| -rw-r--r-- | kernel/trace/trace_kprobe.c | 2402 |
1 files changed, 1589 insertions, 813 deletions
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 7ed6976493c8..9953506370a5 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -1,432 +1,855 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Kprobes-based tracing events * * Created by Masami Hiramatsu <mhiramat@redhat.com> * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ +#define pr_fmt(fmt) "trace_kprobe: " fmt +#include <linux/bpf-cgroup.h> +#include <linux/cleanup.h> +#include <linux/error-injection.h> #include <linux/module.h> +#include <linux/rculist.h> +#include <linux/security.h> #include <linux/uaccess.h> +#include <asm/setup.h> /* for COMMAND_LINE_SIZE */ + +#include "trace_dynevent.h" +#include "trace_kprobe_selftest.h" #include "trace_probe.h" +#include "trace_probe_kernel.h" +#include "trace_probe_tmpl.h" #define KPROBE_EVENT_SYSTEM "kprobes" +#define KRETPROBE_MAXACTIVE_MAX 4096 -/** +/* Kprobe early definition from command line */ +static char kprobe_boot_events_buf[COMMAND_LINE_SIZE] __initdata; + +static int __init set_kprobe_boot_events(char *str) +{ + strscpy(kprobe_boot_events_buf, str, COMMAND_LINE_SIZE); + disable_tracing_selftest("running kprobe events"); + + return 1; +} +__setup("kprobe_event=", set_kprobe_boot_events); + +static int trace_kprobe_create(const char *raw_command); +static int trace_kprobe_show(struct seq_file *m, struct dyn_event *ev); +static int trace_kprobe_release(struct dyn_event *ev); +static bool trace_kprobe_is_busy(struct dyn_event *ev); +static bool trace_kprobe_match(const char *system, const char *event, + int argc, const char **argv, struct dyn_event *ev); + +static struct dyn_event_operations trace_kprobe_ops = { + .create = trace_kprobe_create, + .show = trace_kprobe_show, + .is_busy = trace_kprobe_is_busy, + .free = trace_kprobe_release, + .match = trace_kprobe_match, +}; + +/* * Kprobe event core functions */ -struct trace_probe { - struct list_head list; +struct trace_kprobe { + struct dyn_event devent; struct kretprobe rp; /* Use rp.kp for kprobe use */ - unsigned long nhit; - unsigned int flags; /* For TP_FLAG_* */ + unsigned long __percpu *nhit; const char *symbol; /* symbol name */ - struct ftrace_event_class class; - struct ftrace_event_call call; - struct list_head files; - ssize_t size; /* trace entry size */ - unsigned int nr_args; - struct probe_arg args[]; + struct trace_probe tp; }; -struct event_file_link { - struct ftrace_event_file *file; - struct list_head list; -}; +static bool is_trace_kprobe(struct dyn_event *ev) +{ + return ev->ops == &trace_kprobe_ops; +} + +static struct trace_kprobe *to_trace_kprobe(struct dyn_event *ev) +{ + return container_of(ev, struct trace_kprobe, devent); +} -#define SIZEOF_TRACE_PROBE(n) \ - (offsetof(struct trace_probe, args) + \ - (sizeof(struct probe_arg) * (n))) +/** + * for_each_trace_kprobe - iterate over the trace_kprobe list + * @pos: the struct trace_kprobe * for each entry + * @dpos: the struct dyn_event * to use as a loop cursor + */ +#define for_each_trace_kprobe(pos, dpos) \ + for_each_dyn_event(dpos) \ + if (is_trace_kprobe(dpos) && (pos = to_trace_kprobe(dpos))) +static nokprobe_inline bool trace_kprobe_is_return(struct trace_kprobe *tk) +{ + return tk->rp.handler != NULL; +} + +static nokprobe_inline const char *trace_kprobe_symbol(struct trace_kprobe *tk) +{ + return tk->symbol ? tk->symbol : "unknown"; +} + +static nokprobe_inline unsigned long trace_kprobe_offset(struct trace_kprobe *tk) +{ + return tk->rp.kp.offset; +} + +static nokprobe_inline bool trace_kprobe_has_gone(struct trace_kprobe *tk) +{ + return kprobe_gone(&tk->rp.kp); +} + +static nokprobe_inline bool trace_kprobe_within_module(struct trace_kprobe *tk, + struct module *mod) +{ + int len = strlen(module_name(mod)); + const char *name = trace_kprobe_symbol(tk); + + return strncmp(module_name(mod), name, len) == 0 && name[len] == ':'; +} + +#ifdef CONFIG_MODULES +static nokprobe_inline bool trace_kprobe_module_exist(struct trace_kprobe *tk) +{ + char *p; + bool ret; + + if (!tk->symbol) + return false; + p = strchr(tk->symbol, ':'); + if (!p) + return true; + *p = '\0'; + scoped_guard(rcu) + ret = !!find_module(tk->symbol); + *p = ':'; + + return ret; +} +#else +static inline bool trace_kprobe_module_exist(struct trace_kprobe *tk) +{ + return false; +} +#endif -static __kprobes bool trace_probe_is_return(struct trace_probe *tp) +static bool trace_kprobe_is_busy(struct dyn_event *ev) { - return tp->rp.handler != NULL; + struct trace_kprobe *tk = to_trace_kprobe(ev); + + return trace_probe_is_enabled(&tk->tp); } -static __kprobes const char *trace_probe_symbol(struct trace_probe *tp) +static bool trace_kprobe_match_command_head(struct trace_kprobe *tk, + int argc, const char **argv) { - return tp->symbol ? tp->symbol : "unknown"; + char buf[MAX_ARGSTR_LEN + 1]; + + if (!argc) + return true; + + if (!tk->symbol) + snprintf(buf, sizeof(buf), "0x%p", tk->rp.kp.addr); + else if (tk->rp.kp.offset) + snprintf(buf, sizeof(buf), "%s+%u", + trace_kprobe_symbol(tk), tk->rp.kp.offset); + else + snprintf(buf, sizeof(buf), "%s", trace_kprobe_symbol(tk)); + if (strcmp(buf, argv[0])) + return false; + argc--; argv++; + + return trace_probe_match_command_args(&tk->tp, argc, argv); } -static __kprobes unsigned long trace_probe_offset(struct trace_probe *tp) +static bool trace_kprobe_match(const char *system, const char *event, + int argc, const char **argv, struct dyn_event *ev) { - return tp->rp.kp.offset; + struct trace_kprobe *tk = to_trace_kprobe(ev); + + return (event[0] == '\0' || + strcmp(trace_probe_name(&tk->tp), event) == 0) && + (!system || strcmp(trace_probe_group_name(&tk->tp), system) == 0) && + trace_kprobe_match_command_head(tk, argc, argv); } -static __kprobes bool trace_probe_is_enabled(struct trace_probe *tp) +static nokprobe_inline unsigned long trace_kprobe_nhit(struct trace_kprobe *tk) { - return !!(tp->flags & (TP_FLAG_TRACE | TP_FLAG_PROFILE)); + unsigned long nhit = 0; + int cpu; + + for_each_possible_cpu(cpu) + nhit += *per_cpu_ptr(tk->nhit, cpu); + + return nhit; } -static __kprobes bool trace_probe_is_registered(struct trace_probe *tp) +static nokprobe_inline bool trace_kprobe_is_registered(struct trace_kprobe *tk) { - return !!(tp->flags & TP_FLAG_REGISTERED); + return !(list_empty(&tk->rp.kp.list) && + hlist_unhashed(&tk->rp.kp.hlist)); } -static __kprobes bool trace_probe_has_gone(struct trace_probe *tp) +/* Return 0 if it fails to find the symbol address */ +static nokprobe_inline +unsigned long trace_kprobe_address(struct trace_kprobe *tk) { - return !!(kprobe_gone(&tp->rp.kp)); + unsigned long addr; + + if (tk->symbol) { + addr = (unsigned long) + kallsyms_lookup_name(trace_kprobe_symbol(tk)); + if (addr) + addr += tk->rp.kp.offset; + } else { + addr = (unsigned long)tk->rp.kp.addr; + } + return addr; } -static __kprobes bool trace_probe_within_module(struct trace_probe *tp, - struct module *mod) +static nokprobe_inline struct trace_kprobe * +trace_kprobe_primary_from_call(struct trace_event_call *call) { - int len = strlen(mod->name); - const char *name = trace_probe_symbol(tp); - return strncmp(mod->name, name, len) == 0 && name[len] == ':'; + struct trace_probe *tp; + + tp = trace_probe_primary_from_call(call); + if (WARN_ON_ONCE(!tp)) + return NULL; + + return container_of(tp, struct trace_kprobe, tp); } -static __kprobes bool trace_probe_is_on_module(struct trace_probe *tp) +bool trace_kprobe_on_func_entry(struct trace_event_call *call) { - return !!strchr(trace_probe_symbol(tp), ':'); + struct trace_kprobe *tk = trace_kprobe_primary_from_call(call); + + return tk ? (kprobe_on_func_entry(tk->rp.kp.addr, + tk->rp.kp.addr ? NULL : tk->rp.kp.symbol_name, + tk->rp.kp.addr ? 0 : tk->rp.kp.offset) == 0) : false; } -static int register_probe_event(struct trace_probe *tp); -static void unregister_probe_event(struct trace_probe *tp); +bool trace_kprobe_error_injectable(struct trace_event_call *call) +{ + struct trace_kprobe *tk = trace_kprobe_primary_from_call(call); -static DEFINE_MUTEX(probe_lock); -static LIST_HEAD(probe_list); + return tk ? within_error_injection_list(trace_kprobe_address(tk)) : + false; +} + +static int register_kprobe_event(struct trace_kprobe *tk); +static int unregister_kprobe_event(struct trace_kprobe *tk); static int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs); static int kretprobe_dispatcher(struct kretprobe_instance *ri, struct pt_regs *regs); +static void free_trace_kprobe(struct trace_kprobe *tk) +{ + if (tk) { + trace_probe_cleanup(&tk->tp); + kfree(tk->symbol); + free_percpu(tk->nhit); + kfree(tk); + } +} + +DEFINE_FREE(free_trace_kprobe, struct trace_kprobe *, + if (!IS_ERR_OR_NULL(_T)) free_trace_kprobe(_T)) + /* * Allocate new trace_probe and initialize it (including kprobes). */ -static struct trace_probe *alloc_trace_probe(const char *group, +static struct trace_kprobe *alloc_trace_kprobe(const char *group, const char *event, void *addr, const char *symbol, unsigned long offs, + int maxactive, int nargs, bool is_return) { - struct trace_probe *tp; + struct trace_kprobe *tk __free(free_trace_kprobe) = NULL; int ret = -ENOMEM; - tp = kzalloc(SIZEOF_TRACE_PROBE(nargs), GFP_KERNEL); - if (!tp) + tk = kzalloc(struct_size(tk, tp.args, nargs), GFP_KERNEL); + if (!tk) + return ERR_PTR(ret); + + tk->nhit = alloc_percpu(unsigned long); + if (!tk->nhit) return ERR_PTR(ret); if (symbol) { - tp->symbol = kstrdup(symbol, GFP_KERNEL); - if (!tp->symbol) - goto error; - tp->rp.kp.symbol_name = tp->symbol; - tp->rp.kp.offset = offs; + tk->symbol = kstrdup(symbol, GFP_KERNEL); + if (!tk->symbol) + return ERR_PTR(ret); + tk->rp.kp.symbol_name = tk->symbol; + tk->rp.kp.offset = offs; } else - tp->rp.kp.addr = addr; + tk->rp.kp.addr = addr; if (is_return) - tp->rp.handler = kretprobe_dispatcher; + tk->rp.handler = kretprobe_dispatcher; else - tp->rp.kp.pre_handler = kprobe_dispatcher; + tk->rp.kp.pre_handler = kprobe_dispatcher; - if (!event || !is_good_name(event)) { - ret = -EINVAL; - goto error; - } + tk->rp.maxactive = maxactive; + INIT_HLIST_NODE(&tk->rp.kp.hlist); + INIT_LIST_HEAD(&tk->rp.kp.list); - tp->call.class = &tp->class; - tp->call.name = kstrdup(event, GFP_KERNEL); - if (!tp->call.name) - goto error; + ret = trace_probe_init(&tk->tp, event, group, false, nargs); + if (ret < 0) + return ERR_PTR(ret); - if (!group || !is_good_name(group)) { - ret = -EINVAL; - goto error; - } + dyn_event_init(&tk->devent, &trace_kprobe_ops); + return_ptr(tk); +} - tp->class.system = kstrdup(group, GFP_KERNEL); - if (!tp->class.system) - goto error; +static struct trace_kprobe *find_trace_kprobe(const char *event, + const char *group) +{ + struct dyn_event *pos; + struct trace_kprobe *tk; - INIT_LIST_HEAD(&tp->list); - INIT_LIST_HEAD(&tp->files); - return tp; -error: - kfree(tp->call.name); - kfree(tp->symbol); - kfree(tp); - return ERR_PTR(ret); + for_each_trace_kprobe(tk, pos) + if (strcmp(trace_probe_name(&tk->tp), event) == 0 && + strcmp(trace_probe_group_name(&tk->tp), group) == 0) + return tk; + return NULL; } -static void free_trace_probe(struct trace_probe *tp) +static inline int __enable_trace_kprobe(struct trace_kprobe *tk) { - int i; + int ret = 0; - for (i = 0; i < tp->nr_args; i++) - traceprobe_free_probe_arg(&tp->args[i]); + if (trace_kprobe_is_registered(tk) && !trace_kprobe_has_gone(tk)) { + if (trace_kprobe_is_return(tk)) + ret = enable_kretprobe(&tk->rp); + else + ret = enable_kprobe(&tk->rp.kp); + } - kfree(tp->call.class->system); - kfree(tp->call.name); - kfree(tp->symbol); - kfree(tp); + return ret; } -static struct trace_probe *find_trace_probe(const char *event, - const char *group) +static void __disable_trace_kprobe(struct trace_probe *tp) { - struct trace_probe *tp; + struct trace_kprobe *tk; - list_for_each_entry(tp, &probe_list, list) - if (strcmp(tp->call.name, event) == 0 && - strcmp(tp->call.class->system, group) == 0) - return tp; - return NULL; + list_for_each_entry(tk, trace_probe_probe_list(tp), tp.list) { + if (!trace_kprobe_is_registered(tk)) + continue; + if (trace_kprobe_is_return(tk)) + disable_kretprobe(&tk->rp); + else + disable_kprobe(&tk->rp.kp); + } } /* * Enable trace_probe * if the file is NULL, enable "perf" handler, or enable "trace" handler. */ -static int -enable_trace_probe(struct trace_probe *tp, struct ftrace_event_file *file) +static int enable_trace_kprobe(struct trace_event_call *call, + struct trace_event_file *file) { + struct trace_probe *tp; + struct trace_kprobe *tk; + bool enabled; int ret = 0; - if (file) { - struct event_file_link *link; + tp = trace_probe_primary_from_call(call); + if (WARN_ON_ONCE(!tp)) + return -ENODEV; + enabled = trace_probe_is_enabled(tp); - link = kmalloc(sizeof(*link), GFP_KERNEL); - if (!link) { - ret = -ENOMEM; - goto out; - } + /* This also changes "enabled" state */ + if (file) { + ret = trace_probe_add_file(tp, file); + if (ret) + return ret; + } else + trace_probe_set_flag(tp, TP_FLAG_PROFILE); - link->file = file; - list_add_tail_rcu(&link->list, &tp->files); + if (enabled) + return 0; - tp->flags |= TP_FLAG_TRACE; - } else - tp->flags |= TP_FLAG_PROFILE; + list_for_each_entry(tk, trace_probe_probe_list(tp), tp.list) { + if (trace_kprobe_has_gone(tk)) + continue; + ret = __enable_trace_kprobe(tk); + if (ret) + break; + enabled = true; + } - if (trace_probe_is_registered(tp) && !trace_probe_has_gone(tp)) { - if (trace_probe_is_return(tp)) - ret = enable_kretprobe(&tp->rp); + if (ret) { + /* Failed to enable one of them. Roll back all */ + if (enabled) + __disable_trace_kprobe(tp); + if (file) + trace_probe_remove_file(tp, file); else - ret = enable_kprobe(&tp->rp.kp); + trace_probe_clear_flag(tp, TP_FLAG_PROFILE); } - out: - return ret; -} - -static struct event_file_link * -find_event_file_link(struct trace_probe *tp, struct ftrace_event_file *file) -{ - struct event_file_link *link; - - list_for_each_entry(link, &tp->files, list) - if (link->file == file) - return link; - return NULL; + return ret; } /* * Disable trace_probe * if the file is NULL, disable "perf" handler, or disable "trace" handler. */ -static int -disable_trace_probe(struct trace_probe *tp, struct ftrace_event_file *file) +static int disable_trace_kprobe(struct trace_event_call *call, + struct trace_event_file *file) { - int ret = 0; + struct trace_probe *tp; - if (file) { - struct event_file_link *link; + tp = trace_probe_primary_from_call(call); + if (WARN_ON_ONCE(!tp)) + return -ENODEV; - link = find_event_file_link(tp, file); - if (!link) { - ret = -EINVAL; + if (file) { + if (!trace_probe_get_file_link(tp, file)) + return -ENOENT; + if (!trace_probe_has_single_file(tp)) goto out; - } + trace_probe_clear_flag(tp, TP_FLAG_TRACE); + } else + trace_probe_clear_flag(tp, TP_FLAG_PROFILE); - list_del_rcu(&link->list); - /* synchronize with kprobe_trace_func/kretprobe_trace_func */ - synchronize_sched(); - kfree(link); + if (!trace_probe_is_enabled(tp)) + __disable_trace_kprobe(tp); - if (!list_empty(&tp->files)) - goto out; + out: + if (file) + /* + * Synchronization is done in below function. For perf event, + * file == NULL and perf_trace_event_unreg() calls + * tracepoint_synchronize_unregister() to ensure synchronize + * event. We don't need to care about it. + */ + trace_probe_remove_file(tp, file); - tp->flags &= ~TP_FLAG_TRACE; - } else - tp->flags &= ~TP_FLAG_PROFILE; + return 0; +} - if (!trace_probe_is_enabled(tp) && trace_probe_is_registered(tp)) { - if (trace_probe_is_return(tp)) - disable_kretprobe(&tp->rp); - else - disable_kprobe(&tp->rp.kp); +#if defined(CONFIG_DYNAMIC_FTRACE) && \ + !defined(CONFIG_KPROBE_EVENTS_ON_NOTRACE) +static bool __within_notrace_func(unsigned long addr) +{ + unsigned long offset, size; + + if (!addr || !kallsyms_lookup_size_offset(addr, &size, &offset)) + return false; + + /* Get the entry address of the target function */ + addr -= offset; + + /* + * Since ftrace_location_range() does inclusive range check, we need + * to subtract 1 byte from the end address. + */ + return !ftrace_location_range(addr, addr + size - 1); +} + +static bool within_notrace_func(struct trace_kprobe *tk) +{ + unsigned long addr = trace_kprobe_address(tk); + char symname[KSYM_NAME_LEN], *p; + + if (!__within_notrace_func(addr)) + return false; + + /* Check if the address is on a suffixed-symbol */ + if (!lookup_symbol_name(addr, symname)) { + p = strchr(symname, '.'); + if (!p) + return true; + *p = '\0'; + addr = (unsigned long)kprobe_lookup_name(symname, 0); + if (addr) + return __within_notrace_func(addr); } - out: - return ret; + + return true; } +#else +#define within_notrace_func(tk) (false) +#endif /* Internal register function - just handle k*probes and flags */ -static int __register_trace_probe(struct trace_probe *tp) +static int __register_trace_kprobe(struct trace_kprobe *tk) { int i, ret; - if (trace_probe_is_registered(tp)) + ret = security_locked_down(LOCKDOWN_KPROBES); + if (ret) + return ret; + + if (trace_kprobe_is_registered(tk)) + return -EINVAL; + + if (within_notrace_func(tk)) { + pr_warn("Could not probe notrace function %ps\n", + (void *)trace_kprobe_address(tk)); return -EINVAL; + } - for (i = 0; i < tp->nr_args; i++) - traceprobe_update_arg(&tp->args[i]); + for (i = 0; i < tk->tp.nr_args; i++) { + ret = traceprobe_update_arg(&tk->tp.args[i]); + if (ret) + return ret; + } /* Set/clear disabled flag according to tp->flag */ - if (trace_probe_is_enabled(tp)) - tp->rp.kp.flags &= ~KPROBE_FLAG_DISABLED; + if (trace_probe_is_enabled(&tk->tp)) + tk->rp.kp.flags &= ~KPROBE_FLAG_DISABLED; else - tp->rp.kp.flags |= KPROBE_FLAG_DISABLED; + tk->rp.kp.flags |= KPROBE_FLAG_DISABLED; - if (trace_probe_is_return(tp)) - ret = register_kretprobe(&tp->rp); + if (trace_kprobe_is_return(tk)) + ret = register_kretprobe(&tk->rp); else - ret = register_kprobe(&tp->rp.kp); - - if (ret == 0) - tp->flags |= TP_FLAG_REGISTERED; - else { - pr_warning("Could not insert probe at %s+%lu: %d\n", - trace_probe_symbol(tp), trace_probe_offset(tp), ret); - if (ret == -ENOENT && trace_probe_is_on_module(tp)) { - pr_warning("This probe might be able to register after" - "target module is loaded. Continue.\n"); - ret = 0; - } else if (ret == -EILSEQ) { - pr_warning("Probing address(0x%p) is not an " - "instruction boundary.\n", - tp->rp.kp.addr); - ret = -EINVAL; - } - } + ret = register_kprobe(&tk->rp.kp); return ret; } /* Internal unregister function - just handle k*probes and flags */ -static void __unregister_trace_probe(struct trace_probe *tp) +static void __unregister_trace_kprobe(struct trace_kprobe *tk) { - if (trace_probe_is_registered(tp)) { - if (trace_probe_is_return(tp)) - unregister_kretprobe(&tp->rp); + if (trace_kprobe_is_registered(tk)) { + if (trace_kprobe_is_return(tk)) + unregister_kretprobe(&tk->rp); else - unregister_kprobe(&tp->rp.kp); - tp->flags &= ~TP_FLAG_REGISTERED; - /* Cleanup kprobe for reuse */ - if (tp->rp.kp.symbol_name) - tp->rp.kp.addr = NULL; + unregister_kprobe(&tk->rp.kp); + /* Cleanup kprobe for reuse and mark it unregistered */ + INIT_HLIST_NODE(&tk->rp.kp.hlist); + INIT_LIST_HEAD(&tk->rp.kp.list); + if (tk->rp.kp.symbol_name) + tk->rp.kp.addr = NULL; } } -/* Unregister a trace_probe and probe_event: call with locking probe_lock */ -static int unregister_trace_probe(struct trace_probe *tp) +/* Unregister a trace_probe and probe_event */ +static int unregister_trace_kprobe(struct trace_kprobe *tk) { + /* If other probes are on the event, just unregister kprobe */ + if (trace_probe_has_sibling(&tk->tp)) + goto unreg; + /* Enabled event can not be unregistered */ - if (trace_probe_is_enabled(tp)) + if (trace_probe_is_enabled(&tk->tp)) return -EBUSY; - __unregister_trace_probe(tp); - list_del(&tp->list); - unregister_probe_event(tp); + /* If there's a reference to the dynamic event */ + if (trace_event_dyn_busy(trace_probe_event_call(&tk->tp))) + return -EBUSY; + + /* Will fail if probe is being used by ftrace or perf */ + if (unregister_kprobe_event(tk)) + return -EBUSY; + +unreg: + __unregister_trace_kprobe(tk); + dyn_event_remove(&tk->devent); + trace_probe_unlink(&tk->tp); return 0; } +static bool trace_kprobe_has_same_kprobe(struct trace_kprobe *orig, + struct trace_kprobe *comp) +{ + struct trace_probe_event *tpe = orig->tp.event; + int i; + + list_for_each_entry(orig, &tpe->probes, tp.list) { + if (strcmp(trace_kprobe_symbol(orig), + trace_kprobe_symbol(comp)) || + trace_kprobe_offset(orig) != trace_kprobe_offset(comp)) + continue; + + /* + * trace_probe_compare_arg_type() ensured that nr_args and + * each argument name and type are same. Let's compare comm. + */ + for (i = 0; i < orig->tp.nr_args; i++) { + if (strcmp(orig->tp.args[i].comm, + comp->tp.args[i].comm)) + break; + } + + if (i == orig->tp.nr_args) + return true; + } + + return false; +} + +static int append_trace_kprobe(struct trace_kprobe *tk, struct trace_kprobe *to) +{ + int ret; + + ret = trace_probe_compare_arg_type(&tk->tp, &to->tp); + if (ret) { + /* Note that argument starts index = 2 */ + trace_probe_log_set_index(ret + 1); + trace_probe_log_err(0, DIFF_ARG_TYPE); + return -EEXIST; + } + if (trace_kprobe_has_same_kprobe(to, tk)) { + trace_probe_log_set_index(0); + trace_probe_log_err(0, SAME_PROBE); + return -EEXIST; + } + + /* Append to existing event */ + ret = trace_probe_append(&tk->tp, &to->tp); + if (ret) + return ret; + + /* Register k*probe */ + ret = __register_trace_kprobe(tk); + if (ret == -ENOENT && !trace_kprobe_module_exist(tk)) { + pr_warn("This probe might be able to register after target module is loaded. Continue.\n"); + ret = 0; + } + + if (ret) + trace_probe_unlink(&tk->tp); + else + dyn_event_add(&tk->devent, trace_probe_event_call(&tk->tp)); + + return ret; +} + /* Register a trace_probe and probe_event */ -static int register_trace_probe(struct trace_probe *tp) +static int register_trace_kprobe(struct trace_kprobe *tk) { - struct trace_probe *old_tp; + struct trace_kprobe *old_tk; int ret; - mutex_lock(&probe_lock); + guard(mutex)(&event_mutex); - /* Delete old (same name) event if exist */ - old_tp = find_trace_probe(tp->call.name, tp->call.class->system); - if (old_tp) { - ret = unregister_trace_probe(old_tp); - if (ret < 0) - goto end; - free_trace_probe(old_tp); + old_tk = find_trace_kprobe(trace_probe_name(&tk->tp), + trace_probe_group_name(&tk->tp)); + if (old_tk) { + if (trace_kprobe_is_return(tk) != trace_kprobe_is_return(old_tk)) { + trace_probe_log_set_index(0); + trace_probe_log_err(0, DIFF_PROBE_TYPE); + return -EEXIST; + } + return append_trace_kprobe(tk, old_tk); } /* Register new event */ - ret = register_probe_event(tp); + ret = register_kprobe_event(tk); if (ret) { - pr_warning("Failed to register probe event(%d)\n", ret); - goto end; + if (ret == -EEXIST) { + trace_probe_log_set_index(0); + trace_probe_log_err(0, EVENT_EXIST); + } else + pr_warn("Failed to register probe event(%d)\n", ret); + return ret; } /* Register k*probe */ - ret = __register_trace_probe(tp); + ret = __register_trace_kprobe(tk); + if (ret == -ENOENT && !trace_kprobe_module_exist(tk)) { + pr_warn("This probe might be able to register after target module is loaded. Continue.\n"); + ret = 0; + } + if (ret < 0) - unregister_probe_event(tp); + unregister_kprobe_event(tk); else - list_add_tail(&tp->list, &probe_list); + dyn_event_add(&tk->devent, trace_probe_event_call(&tk->tp)); -end: - mutex_unlock(&probe_lock); + return ret; +} + +#ifdef CONFIG_MODULES +static int validate_module_probe_symbol(const char *modname, const char *symbol); + +static int register_module_trace_kprobe(struct module *mod, struct trace_kprobe *tk) +{ + const char *p; + int ret = 0; + + p = strchr(trace_kprobe_symbol(tk), ':'); + if (p) + ret = validate_module_probe_symbol(module_name(mod), p + 1); + if (!ret) + ret = __register_trace_kprobe(tk); return ret; } /* Module notifier call back, checking event on the module */ -static int trace_probe_module_callback(struct notifier_block *nb, +static int trace_kprobe_module_callback(struct notifier_block *nb, unsigned long val, void *data) { struct module *mod = data; - struct trace_probe *tp; + struct dyn_event *pos; + struct trace_kprobe *tk; int ret; if (val != MODULE_STATE_COMING) return NOTIFY_DONE; /* Update probes on coming module */ - mutex_lock(&probe_lock); - list_for_each_entry(tp, &probe_list, list) { - if (trace_probe_within_module(tp, mod)) { + guard(mutex)(&event_mutex); + for_each_trace_kprobe(tk, pos) { + if (trace_kprobe_within_module(tk, mod)) { /* Don't need to check busy - this should have gone. */ - __unregister_trace_probe(tp); - ret = __register_trace_probe(tp); + __unregister_trace_kprobe(tk); + ret = register_module_trace_kprobe(mod, tk); if (ret) - pr_warning("Failed to re-register probe %s on" - "%s: %d\n", - tp->call.name, mod->name, ret); + pr_warn("Failed to re-register probe %s on %s: %d\n", + trace_probe_name(&tk->tp), + module_name(mod), ret); } } - mutex_unlock(&probe_lock); return NOTIFY_DONE; } -static struct notifier_block trace_probe_module_nb = { - .notifier_call = trace_probe_module_callback, - .priority = 1 /* Invoked after kprobe module callback */ +static struct notifier_block trace_kprobe_module_nb = { + .notifier_call = trace_kprobe_module_callback, + .priority = 2 /* Invoked after kprobe and jump_label module callback */ +}; +static int trace_kprobe_register_module_notifier(void) +{ + return register_module_notifier(&trace_kprobe_module_nb); +} +#else +static int trace_kprobe_register_module_notifier(void) +{ + return 0; +} +#endif /* CONFIG_MODULES */ + +static int count_symbols(void *data, unsigned long unused) +{ + unsigned int *count = data; + + (*count)++; + + return 0; +} + +struct sym_count_ctx { + unsigned int count; + const char *name; }; -static int create_trace_probe(int argc, char **argv) +static int count_mod_symbols(void *data, const char *name, unsigned long unused) +{ + struct sym_count_ctx *ctx = data; + + if (strcmp(name, ctx->name) == 0) + ctx->count++; + + return 0; +} + +static unsigned int number_of_same_symbols(const char *mod, const char *func_name) +{ + struct sym_count_ctx ctx = { .count = 0, .name = func_name }; + + if (!mod) + kallsyms_on_each_match_symbol(count_symbols, func_name, &ctx.count); + + module_kallsyms_on_each_symbol(mod, count_mod_symbols, &ctx); + + return ctx.count; +} + +static int validate_module_probe_symbol(const char *modname, const char *symbol) +{ + unsigned int count = number_of_same_symbols(modname, symbol); + + if (count > 1) { + /* + * Users should use ADDR to remove the ambiguity of + * using KSYM only. + */ + return -EADDRNOTAVAIL; + } else if (count == 0) { + /* + * We can return ENOENT earlier than when register the + * kprobe. + */ + return -ENOENT; + } + return 0; +} + +#ifdef CONFIG_MODULES +/* Return NULL if the module is not loaded or under unloading. */ +static struct module *try_module_get_by_name(const char *name) +{ + struct module *mod; + + guard(rcu)(); + mod = find_module(name); + if (mod && !try_module_get(mod)) + mod = NULL; + return mod; +} +#else +#define try_module_get_by_name(name) (NULL) +#endif + +static int validate_probe_symbol(char *symbol) +{ + struct module *mod = NULL; + char *modname = NULL, *p; + int ret = 0; + + p = strchr(symbol, ':'); + if (p) { + modname = symbol; + symbol = p + 1; + *p = '\0'; + mod = try_module_get_by_name(modname); + if (!mod) + goto out; + } + + ret = validate_module_probe_symbol(modname, symbol); +out: + if (p) + *p = ':'; + if (mod) + module_put(mod); + return ret; +} + +static int trace_kprobe_entry_handler(struct kretprobe_instance *ri, + struct pt_regs *regs); + +static int trace_kprobe_create_internal(int argc, const char *argv[], + struct traceprobe_parse_context *ctx) { /* * Argument syntax: - * - Add kprobe: p[:[GRP/]EVENT] [MOD:]KSYM[+OFFS]|KADDR [FETCHARGS] - * - Add kretprobe: r[:[GRP/]EVENT] [MOD:]KSYM[+0] [FETCHARGS] + * - Add kprobe: + * p[:[GRP/][EVENT]] [MOD:]KSYM[+OFFS]|KADDR [FETCHARGS] + * - Add kretprobe: + * r[MAXACTIVE][:[GRP/][EVENT]] [MOD:]KSYM[+0] [FETCHARGS] + * Or + * p[:[GRP/][EVENT]] [MOD:]KSYM[+0]%return [FETCHARGS] + * * Fetch args: * $retval : fetch return value * $stack : fetch stack address * $stackN : fetch Nth of stack (N:0-) + * $comm : fetch current task comm * @ADDR : fetch memory at ADDR (ADDR should be in kernel) * @SYM[+|-offs] : fetch memory at SYM +|- offs (SYM is a data symbol) * %REG : fetch register REG @@ -437,244 +860,473 @@ static int create_trace_probe(int argc, char **argv) * Type of args: * FETCHARG:TYPE : use TYPE instead of unsigned long. */ - struct trace_probe *tp; - int i, ret = 0; - bool is_return = false, is_delete = false; - char *symbol = NULL, *event = NULL, *group = NULL; - char *arg; - unsigned long offset = 0; + struct trace_kprobe *tk __free(free_trace_kprobe) = NULL; + const char *event = NULL, *group = KPROBE_EVENT_SYSTEM; + const char **new_argv __free(kfree) = NULL; + int i, len, new_argc = 0, ret = 0; + char *symbol __free(kfree) = NULL; + char *ebuf __free(kfree) = NULL; + char *gbuf __free(kfree) = NULL; + char *abuf __free(kfree) = NULL; + char *dbuf __free(kfree) = NULL; + enum probe_print_type ptype; + bool is_return = false; + int maxactive = 0; void *addr = NULL; - char buf[MAX_EVENT_NAME_LEN]; + char *tmp = NULL; + long offset = 0; - /* argc must be >= 1 */ - if (argv[0][0] == 'p') - is_return = false; - else if (argv[0][0] == 'r') + switch (argv[0][0]) { + case 'r': is_return = true; - else if (argv[0][0] == '-') - is_delete = true; - else { - pr_info("Probe definition must be started with 'p', 'r' or" - " '-'.\n"); - return -EINVAL; + break; + case 'p': + break; + default: + return -ECANCELED; } + if (argc < 2) + return -ECANCELED; - if (argv[0][1] == ':') { - event = &argv[0][2]; - if (strchr(event, '/')) { - group = event; - event = strchr(group, '/') + 1; - event[-1] = '\0'; - if (strlen(group) == 0) { - pr_info("Group name is not specified\n"); - return -EINVAL; - } + event = strchr(&argv[0][1], ':'); + if (event) + event++; + + if (isdigit(argv[0][1])) { + char *buf __free(kfree) = NULL; + + if (!is_return) { + trace_probe_log_err(1, BAD_MAXACT_TYPE); + return -EINVAL; } - if (strlen(event) == 0) { - pr_info("Event name is not specified\n"); + if (event) + len = event - &argv[0][1] - 1; + else + len = strlen(&argv[0][1]); + if (len > MAX_EVENT_NAME_LEN - 1) { + trace_probe_log_err(1, BAD_MAXACT); return -EINVAL; } - } - if (!group) - group = KPROBE_EVENT_SYSTEM; - - if (is_delete) { - if (!event) { - pr_info("Delete command needs an event name.\n"); + buf = kmemdup(&argv[0][1], len + 1, GFP_KERNEL); + if (!buf) + return -ENOMEM; + buf[len] = '\0'; + ret = kstrtouint(buf, 0, &maxactive); + if (ret || !maxactive) { + trace_probe_log_err(1, BAD_MAXACT); return -EINVAL; } - mutex_lock(&probe_lock); - tp = find_trace_probe(event, group); - if (!tp) { - mutex_unlock(&probe_lock); - pr_info("Event %s/%s doesn't exist.\n", group, event); - return -ENOENT; + /* kretprobes instances are iterated over via a list. The + * maximum should stay reasonable. + */ + if (maxactive > KRETPROBE_MAXACTIVE_MAX) { + trace_probe_log_err(1, MAXACT_TOO_BIG); + return -EINVAL; } - /* delete an event */ - ret = unregister_trace_probe(tp); - if (ret == 0) - free_trace_probe(tp); - mutex_unlock(&probe_lock); - return ret; } - if (argc < 2) { - pr_info("Probe point is not specified.\n"); - return -EINVAL; - } - if (isdigit(argv[1][0])) { - if (is_return) { - pr_info("Return probe point must be a symbol.\n"); - return -EINVAL; - } - /* an address specified */ - ret = kstrtoul(&argv[1][0], 0, (unsigned long *)&addr); - if (ret) { - pr_info("Failed to parse address.\n"); - return ret; - } - } else { + /* try to parse an address. if that fails, try to read the + * input as a symbol. */ + if (kstrtoul(argv[1], 0, (unsigned long *)&addr)) { + trace_probe_log_set_index(1); + /* Check whether uprobe event specified */ + if (strchr(argv[1], '/') && strchr(argv[1], ':')) + return -ECANCELED; + /* a symbol specified */ - symbol = argv[1]; + symbol = kstrdup(argv[1], GFP_KERNEL); + if (!symbol) + return -ENOMEM; + + tmp = strchr(symbol, '%'); + if (tmp) { + if (!strcmp(tmp, "%return")) { + *tmp = '\0'; + is_return = true; + } else { + trace_probe_log_err(tmp - symbol, BAD_ADDR_SUFFIX); + return -EINVAL; + } + } + /* TODO: support .init module functions */ ret = traceprobe_split_symbol_offset(symbol, &offset); + if (ret || offset < 0 || offset > UINT_MAX) { + trace_probe_log_err(0, BAD_PROBE_ADDR); + return -EINVAL; + } + ret = validate_probe_symbol(symbol); if (ret) { - pr_info("Failed to parse symbol.\n"); - return ret; + if (ret == -EADDRNOTAVAIL) + trace_probe_log_err(0, NON_UNIQ_SYMBOL); + else + trace_probe_log_err(0, BAD_PROBE_ADDR); + return -EINVAL; } - if (offset && is_return) { - pr_info("Return probe must be used without offset.\n"); + if (is_return) + ctx->flags |= TPARG_FL_RETURN; + ret = kprobe_on_func_entry(NULL, symbol, offset); + if (ret == 0 && !is_return) + ctx->flags |= TPARG_FL_FENTRY; + /* Defer the ENOENT case until register kprobe */ + if (ret == -EINVAL && is_return) { + trace_probe_log_err(0, BAD_RETPROBE); return -EINVAL; } } - argc -= 2; argv += 2; - /* setup a probe */ + trace_probe_log_set_index(0); + if (event) { + gbuf = kmalloc(MAX_EVENT_NAME_LEN, GFP_KERNEL); + if (!gbuf) + return -ENOMEM; + ret = traceprobe_parse_event_name(&event, &group, gbuf, + event - argv[0]); + if (ret) + return ret; + } + if (!event) { /* Make a new event name */ + ebuf = kmalloc(MAX_EVENT_NAME_LEN, GFP_KERNEL); + if (!ebuf) + return -ENOMEM; if (symbol) - snprintf(buf, MAX_EVENT_NAME_LEN, "%c_%s_%ld", + snprintf(ebuf, MAX_EVENT_NAME_LEN, "%c_%s_%ld", is_return ? 'r' : 'p', symbol, offset); else - snprintf(buf, MAX_EVENT_NAME_LEN, "%c_0x%p", + snprintf(ebuf, MAX_EVENT_NAME_LEN, "%c_0x%p", is_return ? 'r' : 'p', addr); - event = buf; + sanitize_event_name(ebuf); + event = ebuf; } - tp = alloc_trace_probe(group, event, addr, symbol, offset, argc, - is_return); - if (IS_ERR(tp)) { - pr_info("Failed to allocate trace_probe.(%d)\n", - (int)PTR_ERR(tp)); - return PTR_ERR(tp); + + abuf = kmalloc(MAX_BTF_ARGS_LEN, GFP_KERNEL); + if (!abuf) + return -ENOMEM; + argc -= 2; argv += 2; + ctx->funcname = symbol; + new_argv = traceprobe_expand_meta_args(argc, argv, &new_argc, + abuf, MAX_BTF_ARGS_LEN, ctx); + if (IS_ERR(new_argv)) { + ret = PTR_ERR(new_argv); + new_argv = NULL; + return ret; + } + if (new_argv) { + argc = new_argc; + argv = new_argv; + } + if (argc > MAX_TRACE_ARGS) { + trace_probe_log_set_index(2); + trace_probe_log_err(0, TOO_MANY_ARGS); + return -E2BIG; } - /* parse arguments */ - ret = 0; - for (i = 0; i < argc && i < MAX_TRACE_ARGS; i++) { - /* Increment count for freeing args in error case */ - tp->nr_args++; - - /* Parse argument name */ - arg = strchr(argv[i], '='); - if (arg) { - *arg++ = '\0'; - tp->args[i].name = kstrdup(argv[i], GFP_KERNEL); - } else { - arg = argv[i]; - /* If argument name is omitted, set "argN" */ - snprintf(buf, MAX_EVENT_NAME_LEN, "arg%d", i + 1); - tp->args[i].name = kstrdup(buf, GFP_KERNEL); - } + ret = traceprobe_expand_dentry_args(argc, argv, &dbuf); + if (ret) + return ret; - if (!tp->args[i].name) { - pr_info("Failed to allocate argument[%d] name.\n", i); - ret = -ENOMEM; - goto error; - } + /* setup a probe */ + tk = alloc_trace_kprobe(group, event, addr, symbol, offset, maxactive, + argc, is_return); + if (IS_ERR(tk)) { + ret = PTR_ERR(tk); + /* This must return -ENOMEM, else there is a bug */ + WARN_ON_ONCE(ret != -ENOMEM); + return ret; /* We know tk is not allocated */ + } - if (!is_good_name(tp->args[i].name)) { - pr_info("Invalid argument[%d] name: %s\n", - i, tp->args[i].name); - ret = -EINVAL; - goto error; - } + /* parse arguments */ + for (i = 0; i < argc; i++) { + trace_probe_log_set_index(i + 2); + ctx->offset = 0; + ret = traceprobe_parse_probe_arg(&tk->tp, i, argv[i], ctx); + if (ret) + return ret; /* This can be -ENOMEM */ + } + /* entry handler for kretprobe */ + if (is_return && tk->tp.entry_arg) { + tk->rp.entry_handler = trace_kprobe_entry_handler; + tk->rp.data_size = traceprobe_get_entry_data_size(&tk->tp); + } - if (traceprobe_conflict_field_name(tp->args[i].name, - tp->args, i)) { - pr_info("Argument[%d] name '%s' conflicts with " - "another field.\n", i, argv[i]); - ret = -EINVAL; - goto error; - } + ptype = is_return ? PROBE_PRINT_RETURN : PROBE_PRINT_NORMAL; + ret = traceprobe_set_print_fmt(&tk->tp, ptype); + if (ret < 0) + return ret; - /* Parse fetch argument */ - ret = traceprobe_parse_probe_arg(arg, &tp->size, &tp->args[i], - is_return, true); - if (ret) { - pr_info("Parse error at argument[%d]. (%d)\n", i, ret); - goto error; - } + ret = register_trace_kprobe(tk); + if (ret) { + trace_probe_log_set_index(1); + if (ret == -EILSEQ) + trace_probe_log_err(0, BAD_INSN_BNDRY); + else if (ret == -ENOENT) + trace_probe_log_err(0, BAD_PROBE_ADDR); + else if (ret != -ENOMEM && ret != -EEXIST) + trace_probe_log_err(0, FAIL_REG_PROBE); + return ret; } + /* + * Here, 'tk' has been registered to the list successfully, + * so we don't need to free it. + */ + tk = NULL; - ret = register_trace_probe(tp); - if (ret) - goto error; return 0; +} + +static int trace_kprobe_create_cb(int argc, const char *argv[]) +{ + struct traceprobe_parse_context *ctx __free(traceprobe_parse_context) = NULL; + int ret; + + ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + ctx->flags = TPARG_FL_KERNEL; + + trace_probe_log_init("trace_kprobe", argc, argv); + + ret = trace_kprobe_create_internal(argc, argv, ctx); -error: - free_trace_probe(tp); + trace_probe_log_clear(); return ret; } -static int release_all_trace_probes(void) +static int trace_kprobe_create(const char *raw_command) { - struct trace_probe *tp; - int ret = 0; + return trace_probe_create(raw_command, trace_kprobe_create_cb); +} + +static int create_or_delete_trace_kprobe(const char *raw_command) +{ + int ret; + + if (raw_command[0] == '-') + return dyn_event_release(raw_command, &trace_kprobe_ops); + + ret = dyn_event_create(raw_command, &trace_kprobe_ops); + return ret == -ECANCELED ? -EINVAL : ret; +} + +static int trace_kprobe_run_command(struct dynevent_cmd *cmd) +{ + return create_or_delete_trace_kprobe(cmd->seq.buffer); +} + +/** + * kprobe_event_cmd_init - Initialize a kprobe event command object + * @cmd: A pointer to the dynevent_cmd struct representing the new event + * @buf: A pointer to the buffer used to build the command + * @maxlen: The length of the buffer passed in @buf + * + * Initialize a synthetic event command object. Use this before + * calling any of the other kprobe_event functions. + */ +void kprobe_event_cmd_init(struct dynevent_cmd *cmd, char *buf, int maxlen) +{ + dynevent_cmd_init(cmd, buf, maxlen, DYNEVENT_TYPE_KPROBE, + trace_kprobe_run_command); +} +EXPORT_SYMBOL_GPL(kprobe_event_cmd_init); + +/** + * __kprobe_event_gen_cmd_start - Generate a kprobe event command from arg list + * @cmd: A pointer to the dynevent_cmd struct representing the new event + * @kretprobe: Is this a return probe? + * @name: The name of the kprobe event + * @loc: The location of the kprobe event + * @...: Variable number of arg (pairs), one pair for each field + * + * NOTE: Users normally won't want to call this function directly, but + * rather use the kprobe_event_gen_cmd_start() wrapper, which automatically + * adds a NULL to the end of the arg list. If this function is used + * directly, make sure the last arg in the variable arg list is NULL. + * + * Generate a kprobe event command to be executed by + * kprobe_event_gen_cmd_end(). This function can be used to generate the + * complete command or only the first part of it; in the latter case, + * kprobe_event_add_fields() can be used to add more fields following this. + * + * Unlikely the synth_event_gen_cmd_start(), @loc must be specified. This + * returns -EINVAL if @loc == NULL. + * + * Return: 0 if successful, error otherwise. + */ +int __kprobe_event_gen_cmd_start(struct dynevent_cmd *cmd, bool kretprobe, + const char *name, const char *loc, ...) +{ + char buf[MAX_EVENT_NAME_LEN]; + struct dynevent_arg arg; + va_list args; + int ret; + + if (cmd->type != DYNEVENT_TYPE_KPROBE) + return -EINVAL; + + if (!loc) + return -EINVAL; + + if (kretprobe) + snprintf(buf, MAX_EVENT_NAME_LEN, "r:kprobes/%s", name); + else + snprintf(buf, MAX_EVENT_NAME_LEN, "p:kprobes/%s", name); + + ret = dynevent_str_add(cmd, buf); + if (ret) + return ret; + + dynevent_arg_init(&arg, 0); + arg.str = loc; + ret = dynevent_arg_add(cmd, &arg, NULL); + if (ret) + return ret; + + va_start(args, loc); + for (;;) { + const char *field; - mutex_lock(&probe_lock); - /* Ensure no probe is in use. */ - list_for_each_entry(tp, &probe_list, list) - if (trace_probe_is_enabled(tp)) { - ret = -EBUSY; - goto end; + field = va_arg(args, const char *); + if (!field) + break; + + if (++cmd->n_fields > MAX_TRACE_ARGS) { + ret = -EINVAL; + break; } - /* TODO: Use batch unregistration */ - while (!list_empty(&probe_list)) { - tp = list_entry(probe_list.next, struct trace_probe, list); - unregister_trace_probe(tp); - free_trace_probe(tp); - } -end: - mutex_unlock(&probe_lock); + arg.str = field; + ret = dynevent_arg_add(cmd, &arg, NULL); + if (ret) + break; + } + va_end(args); return ret; } +EXPORT_SYMBOL_GPL(__kprobe_event_gen_cmd_start); -/* Probes listing interfaces */ -static void *probes_seq_start(struct seq_file *m, loff_t *pos) +/** + * __kprobe_event_add_fields - Add probe fields to a kprobe command from arg list + * @cmd: A pointer to the dynevent_cmd struct representing the new event + * @...: Variable number of arg (pairs), one pair for each field + * + * NOTE: Users normally won't want to call this function directly, but + * rather use the kprobe_event_add_fields() wrapper, which + * automatically adds a NULL to the end of the arg list. If this + * function is used directly, make sure the last arg in the variable + * arg list is NULL. + * + * Add probe fields to an existing kprobe command using a variable + * list of args. Fields are added in the same order they're listed. + * + * Return: 0 if successful, error otherwise. + */ +int __kprobe_event_add_fields(struct dynevent_cmd *cmd, ...) { - mutex_lock(&probe_lock); - return seq_list_start(&probe_list, *pos); + struct dynevent_arg arg; + va_list args; + int ret = 0; + + if (cmd->type != DYNEVENT_TYPE_KPROBE) + return -EINVAL; + + dynevent_arg_init(&arg, 0); + + va_start(args, cmd); + for (;;) { + const char *field; + + field = va_arg(args, const char *); + if (!field) + break; + + if (++cmd->n_fields > MAX_TRACE_ARGS) { + ret = -EINVAL; + break; + } + + arg.str = field; + ret = dynevent_arg_add(cmd, &arg, NULL); + if (ret) + break; + } + va_end(args); + + return ret; } +EXPORT_SYMBOL_GPL(__kprobe_event_add_fields); -static void *probes_seq_next(struct seq_file *m, void *v, loff_t *pos) +/** + * kprobe_event_delete - Delete a kprobe event + * @name: The name of the kprobe event to delete + * + * Delete a kprobe event with the give @name from kernel code rather + * than directly from the command line. + * + * Return: 0 if successful, error otherwise. + */ +int kprobe_event_delete(const char *name) { - return seq_list_next(v, &probe_list, pos); + char buf[MAX_EVENT_NAME_LEN]; + + snprintf(buf, MAX_EVENT_NAME_LEN, "-:%s", name); + + return create_or_delete_trace_kprobe(buf); } +EXPORT_SYMBOL_GPL(kprobe_event_delete); -static void probes_seq_stop(struct seq_file *m, void *v) +static int trace_kprobe_release(struct dyn_event *ev) { - mutex_unlock(&probe_lock); + struct trace_kprobe *tk = to_trace_kprobe(ev); + int ret = unregister_trace_kprobe(tk); + + if (!ret) + free_trace_kprobe(tk); + return ret; } -static int probes_seq_show(struct seq_file *m, void *v) +static int trace_kprobe_show(struct seq_file *m, struct dyn_event *ev) { - struct trace_probe *tp = v; + struct trace_kprobe *tk = to_trace_kprobe(ev); int i; - seq_printf(m, "%c", trace_probe_is_return(tp) ? 'r' : 'p'); - seq_printf(m, ":%s/%s", tp->call.class->system, tp->call.name); - - if (!tp->symbol) - seq_printf(m, " 0x%p", tp->rp.kp.addr); - else if (tp->rp.kp.offset) - seq_printf(m, " %s+%u", trace_probe_symbol(tp), - tp->rp.kp.offset); + seq_putc(m, trace_kprobe_is_return(tk) ? 'r' : 'p'); + if (trace_kprobe_is_return(tk) && tk->rp.maxactive) + seq_printf(m, "%d", tk->rp.maxactive); + seq_printf(m, ":%s/%s", trace_probe_group_name(&tk->tp), + trace_probe_name(&tk->tp)); + + if (!tk->symbol) + seq_printf(m, " 0x%p", tk->rp.kp.addr); + else if (tk->rp.kp.offset) + seq_printf(m, " %s+%u", trace_kprobe_symbol(tk), + tk->rp.kp.offset); else - seq_printf(m, " %s", trace_probe_symbol(tp)); + seq_printf(m, " %s", trace_kprobe_symbol(tk)); - for (i = 0; i < tp->nr_args; i++) - seq_printf(m, " %s=%s", tp->args[i].name, tp->args[i].comm); - seq_printf(m, "\n"); + for (i = 0; i < tk->tp.nr_args; i++) + seq_printf(m, " %s=%s", tk->tp.args[i].name, tk->tp.args[i].comm); + seq_putc(m, '\n'); return 0; } +static int probes_seq_show(struct seq_file *m, void *v) +{ + struct dyn_event *ev = v; + + if (!is_trace_kprobe(ev)) + return 0; + + return trace_kprobe_show(m, ev); +} + static const struct seq_operations probes_seq_op = { - .start = probes_seq_start, - .next = probes_seq_next, - .stop = probes_seq_stop, + .start = dyn_event_seq_start, + .next = dyn_event_seq_next, + .stop = dyn_event_seq_stop, .show = probes_seq_show }; @@ -682,8 +1334,12 @@ static int probes_open(struct inode *inode, struct file *file) { int ret; + ret = security_locked_down(LOCKDOWN_TRACEFS); + if (ret) + return ret; + if ((file->f_mode & FMODE_WRITE) && (file->f_flags & O_TRUNC)) { - ret = release_all_trace_probes(); + ret = dyn_events_release_all(&trace_kprobe_ops); if (ret < 0) return ret; } @@ -694,8 +1350,8 @@ static int probes_open(struct inode *inode, struct file *file) static ssize_t probes_write(struct file *file, const char __user *buffer, size_t count, loff_t *ppos) { - return traceprobe_probes_write(file, buffer, count, ppos, - create_trace_probe); + return trace_parse_run_command(file, buffer, count, ppos, + create_or_delete_trace_kprobe); } static const struct file_operations kprobe_events_ops = { @@ -707,26 +1363,47 @@ static const struct file_operations kprobe_events_ops = { .write = probes_write, }; +static unsigned long trace_kprobe_missed(struct trace_kprobe *tk) +{ + return trace_kprobe_is_return(tk) ? + tk->rp.kp.nmissed + tk->rp.nmissed : tk->rp.kp.nmissed; +} + /* Probes profiling interfaces */ static int probes_profile_seq_show(struct seq_file *m, void *v) { - struct trace_probe *tp = v; + struct dyn_event *ev = v; + struct trace_kprobe *tk; + unsigned long nmissed; - seq_printf(m, " %-44s %15lu %15lu\n", tp->call.name, tp->nhit, - tp->rp.kp.nmissed); + if (!is_trace_kprobe(ev)) + return 0; + + tk = to_trace_kprobe(ev); + nmissed = trace_kprobe_missed(tk); + seq_printf(m, " %-44s %15lu %15lu\n", + trace_probe_name(&tk->tp), + trace_kprobe_nhit(tk), + nmissed); return 0; } static const struct seq_operations profile_seq_op = { - .start = probes_seq_start, - .next = probes_seq_next, - .stop = probes_seq_stop, + .start = dyn_event_seq_start, + .next = dyn_event_seq_next, + .stop = dyn_event_seq_stop, .show = probes_profile_seq_show }; static int profile_open(struct inode *inode, struct file *file) { + int ret; + + ret = security_locked_down(LOCKDOWN_TRACEFS); + if (ret) + return ret; + return seq_open(file, &profile_seq_op); } @@ -738,150 +1415,157 @@ static const struct file_operations kprobe_profile_ops = { .release = seq_release, }; -/* Sum up total data length for dynamic arraies (strings) */ -static __kprobes int __get_data_size(struct trace_probe *tp, - struct pt_regs *regs) +/* Note that we don't verify it, since the code does not come from user space */ +static int +process_fetch_insn(struct fetch_insn *code, void *rec, void *edata, + void *dest, void *base) { - int i, ret = 0; - u32 len; - - for (i = 0; i < tp->nr_args; i++) - if (unlikely(tp->args[i].fetch_size.fn)) { - call_fetch(&tp->args[i].fetch_size, regs, &len); - ret += len; - } - - return ret; -} + struct pt_regs *regs = rec; + unsigned long val; + int ret; -/* Store the value of each argument */ -static __kprobes void store_trace_args(int ent_size, struct trace_probe *tp, - struct pt_regs *regs, - u8 *data, int maxlen) -{ - int i; - u32 end = tp->size; - u32 *dl; /* Data (relative) location */ - - for (i = 0; i < tp->nr_args; i++) { - if (unlikely(tp->args[i].fetch_size.fn)) { - /* - * First, we set the relative location and - * maximum data length to *dl - */ - dl = (u32 *)(data + tp->args[i].offset); - *dl = make_data_rloc(maxlen, end - tp->args[i].offset); - /* Then try to fetch string or dynamic array data */ - call_fetch(&tp->args[i].fetch, regs, dl); - /* Reduce maximum length */ - end += get_rloc_len(*dl); - maxlen -= get_rloc_len(*dl); - /* Trick here, convert data_rloc to data_loc */ - *dl = convert_rloc_to_loc(*dl, - ent_size + tp->args[i].offset); - } else - /* Just fetching data normally */ - call_fetch(&tp->args[i].fetch, regs, - data + tp->args[i].offset); +retry: + /* 1st stage: get value from context */ + switch (code->op) { + case FETCH_OP_REG: + val = regs_get_register(regs, code->param); + break; + case FETCH_OP_STACK: + val = regs_get_kernel_stack_nth(regs, code->param); + break; + case FETCH_OP_STACKP: + val = kernel_stack_pointer(regs); + break; + case FETCH_OP_RETVAL: + val = regs_return_value(regs); + break; +#ifdef CONFIG_HAVE_FUNCTION_ARG_ACCESS_API + case FETCH_OP_ARG: + val = regs_get_kernel_argument(regs, code->param); + break; + case FETCH_OP_EDATA: + val = *(unsigned long *)((unsigned long)edata + code->offset); + break; +#endif + case FETCH_NOP_SYMBOL: /* Ignore a place holder */ + code++; + goto retry; + default: + ret = process_common_fetch_insn(code, &val); + if (ret < 0) + return ret; } + code++; + + return process_fetch_insn_bottom(code, val, dest, base); } +NOKPROBE_SYMBOL(process_fetch_insn) /* Kprobe handler */ -static __kprobes void -__kprobe_trace_func(struct trace_probe *tp, struct pt_regs *regs, - struct ftrace_event_file *ftrace_file) +static nokprobe_inline void +__kprobe_trace_func(struct trace_kprobe *tk, struct pt_regs *regs, + struct trace_event_file *trace_file) { struct kprobe_trace_entry_head *entry; - struct ring_buffer_event *event; - struct ring_buffer *buffer; - int size, dsize, pc; - unsigned long irq_flags; - struct ftrace_event_call *call = &tp->call; + struct trace_event_call *call = trace_probe_event_call(&tk->tp); + struct trace_event_buffer fbuffer; + int dsize; - WARN_ON(call != ftrace_file->event_call); + WARN_ON(call != trace_file->event_call); - if (test_bit(FTRACE_EVENT_FL_SOFT_DISABLED_BIT, &ftrace_file->flags)) + if (trace_trigger_soft_disabled(trace_file)) return; - local_save_flags(irq_flags); - pc = preempt_count(); - - dsize = __get_data_size(tp, regs); - size = sizeof(*entry) + tp->size + dsize; + dsize = __get_data_size(&tk->tp, regs, NULL); - event = trace_event_buffer_lock_reserve(&buffer, ftrace_file, - call->event.type, - size, irq_flags, pc); - if (!event) + entry = trace_event_buffer_reserve(&fbuffer, trace_file, + sizeof(*entry) + tk->tp.size + dsize); + if (!entry) return; - entry = ring_buffer_event_data(event); - entry->ip = (unsigned long)tp->rp.kp.addr; - store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize); + fbuffer.regs = regs; + entry->ip = (unsigned long)tk->rp.kp.addr; + store_trace_args(&entry[1], &tk->tp, regs, NULL, sizeof(*entry), dsize); - if (!filter_current_check_discard(buffer, call, entry, event)) - trace_buffer_unlock_commit_regs(buffer, event, - irq_flags, pc, regs); + trace_event_buffer_commit(&fbuffer); } -static __kprobes void -kprobe_trace_func(struct trace_probe *tp, struct pt_regs *regs) +static void +kprobe_trace_func(struct trace_kprobe *tk, struct pt_regs *regs) { struct event_file_link *link; - list_for_each_entry_rcu(link, &tp->files, list) - __kprobe_trace_func(tp, regs, link->file); + trace_probe_for_each_link_rcu(link, &tk->tp) + __kprobe_trace_func(tk, regs, link->file); } +NOKPROBE_SYMBOL(kprobe_trace_func); /* Kretprobe handler */ -static __kprobes void -__kretprobe_trace_func(struct trace_probe *tp, struct kretprobe_instance *ri, + +static int trace_kprobe_entry_handler(struct kretprobe_instance *ri, + struct pt_regs *regs) +{ + struct kretprobe *rp = get_kretprobe(ri); + struct trace_kprobe *tk; + + /* + * There is a small chance that get_kretprobe(ri) returns NULL when + * the kretprobe is unregister on another CPU between kretprobe's + * trampoline_handler and this function. + */ + if (unlikely(!rp)) + return -ENOENT; + + tk = container_of(rp, struct trace_kprobe, rp); + + /* store argument values into ri->data as entry data */ + if (tk->tp.entry_arg) + store_trace_entry_data(ri->data, &tk->tp, regs); + + return 0; +} + + +static nokprobe_inline void +__kretprobe_trace_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, struct pt_regs *regs, - struct ftrace_event_file *ftrace_file) + struct trace_event_file *trace_file) { struct kretprobe_trace_entry_head *entry; - struct ring_buffer_event *event; - struct ring_buffer *buffer; - int size, pc, dsize; - unsigned long irq_flags; - struct ftrace_event_call *call = &tp->call; + struct trace_event_buffer fbuffer; + struct trace_event_call *call = trace_probe_event_call(&tk->tp); + int dsize; - WARN_ON(call != ftrace_file->event_call); + WARN_ON(call != trace_file->event_call); - if (test_bit(FTRACE_EVENT_FL_SOFT_DISABLED_BIT, &ftrace_file->flags)) + if (trace_trigger_soft_disabled(trace_file)) return; - local_save_flags(irq_flags); - pc = preempt_count(); - - dsize = __get_data_size(tp, regs); - size = sizeof(*entry) + tp->size + dsize; + dsize = __get_data_size(&tk->tp, regs, ri->data); - event = trace_event_buffer_lock_reserve(&buffer, ftrace_file, - call->event.type, - size, irq_flags, pc); - if (!event) + entry = trace_event_buffer_reserve(&fbuffer, trace_file, + sizeof(*entry) + tk->tp.size + dsize); + if (!entry) return; - entry = ring_buffer_event_data(event); - entry->func = (unsigned long)tp->rp.kp.addr; - entry->ret_ip = (unsigned long)ri->ret_addr; - store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize); + fbuffer.regs = regs; + entry->func = (unsigned long)tk->rp.kp.addr; + entry->ret_ip = get_kretprobe_retaddr(ri); + store_trace_args(&entry[1], &tk->tp, regs, ri->data, sizeof(*entry), dsize); - if (!filter_current_check_discard(buffer, call, entry, event)) - trace_buffer_unlock_commit_regs(buffer, event, - irq_flags, pc, regs); + trace_event_buffer_commit(&fbuffer); } -static __kprobes void -kretprobe_trace_func(struct trace_probe *tp, struct kretprobe_instance *ri, +static void +kretprobe_trace_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, struct pt_regs *regs) { struct event_file_link *link; - list_for_each_entry_rcu(link, &tp->files, list) - __kretprobe_trace_func(tp, ri, regs, link->file); + trace_probe_for_each_link_rcu(link, &tk->tp) + __kretprobe_trace_func(tk, ri, regs, link->file); } +NOKPROBE_SYMBOL(kretprobe_trace_func); /* Event entry printers */ static enum print_line_t @@ -891,33 +1575,27 @@ print_kprobe_event(struct trace_iterator *iter, int flags, struct kprobe_trace_entry_head *field; struct trace_seq *s = &iter->seq; struct trace_probe *tp; - u8 *data; - int i; field = (struct kprobe_trace_entry_head *)iter->ent; - tp = container_of(event, struct trace_probe, call.event); - - if (!trace_seq_printf(s, "%s: (", tp->call.name)) - goto partial; + tp = trace_probe_primary_from_call( + container_of(event, struct trace_event_call, event)); + if (WARN_ON_ONCE(!tp)) + goto out; - if (!seq_print_ip_sym(s, field->ip, flags | TRACE_ITER_SYM_OFFSET)) - goto partial; + trace_seq_printf(s, "%s: (", trace_probe_name(tp)); - if (!trace_seq_puts(s, ")")) - goto partial; + if (!seq_print_ip_sym_offset(s, field->ip, flags)) + goto out; - data = (u8 *)&field[1]; - for (i = 0; i < tp->nr_args; i++) - if (!tp->args[i].type->print(s, tp->args[i].name, - data + tp->args[i].offset, field)) - goto partial; + trace_seq_putc(s, ')'); - if (!trace_seq_puts(s, "\n")) - goto partial; + if (trace_probe_print_args(s, tp->args, tp->nr_args, + (u8 *)&field[1], field) < 0) + goto out; - return TRACE_TYPE_HANDLED; -partial: - return TRACE_TYPE_PARTIAL_LINE; + trace_seq_putc(s, '\n'); + out: + return trace_handle_return(s); } static enum print_line_t @@ -927,211 +1605,178 @@ print_kretprobe_event(struct trace_iterator *iter, int flags, struct kretprobe_trace_entry_head *field; struct trace_seq *s = &iter->seq; struct trace_probe *tp; - u8 *data; - int i; field = (struct kretprobe_trace_entry_head *)iter->ent; - tp = container_of(event, struct trace_probe, call.event); + tp = trace_probe_primary_from_call( + container_of(event, struct trace_event_call, event)); + if (WARN_ON_ONCE(!tp)) + goto out; - if (!trace_seq_printf(s, "%s: (", tp->call.name)) - goto partial; + trace_seq_printf(s, "%s: (", trace_probe_name(tp)); - if (!seq_print_ip_sym(s, field->ret_ip, flags | TRACE_ITER_SYM_OFFSET)) - goto partial; + if (!seq_print_ip_sym_offset(s, field->ret_ip, flags)) + goto out; - if (!trace_seq_puts(s, " <- ")) - goto partial; + trace_seq_puts(s, " <- "); - if (!seq_print_ip_sym(s, field->func, flags & ~TRACE_ITER_SYM_OFFSET)) - goto partial; + if (!seq_print_ip_sym_no_offset(s, field->func, flags)) + goto out; - if (!trace_seq_puts(s, ")")) - goto partial; + trace_seq_putc(s, ')'); - data = (u8 *)&field[1]; - for (i = 0; i < tp->nr_args; i++) - if (!tp->args[i].type->print(s, tp->args[i].name, - data + tp->args[i].offset, field)) - goto partial; + if (trace_probe_print_args(s, tp->args, tp->nr_args, + (u8 *)&field[1], field) < 0) + goto out; - if (!trace_seq_puts(s, "\n")) - goto partial; + trace_seq_putc(s, '\n'); - return TRACE_TYPE_HANDLED; -partial: - return TRACE_TYPE_PARTIAL_LINE; + out: + return trace_handle_return(s); } -static int kprobe_event_define_fields(struct ftrace_event_call *event_call) +static int kprobe_event_define_fields(struct trace_event_call *event_call) { - int ret, i; + int ret; struct kprobe_trace_entry_head field; - struct trace_probe *tp = (struct trace_probe *)event_call->data; + struct trace_probe *tp; + + tp = trace_probe_primary_from_call(event_call); + if (WARN_ON_ONCE(!tp)) + return -ENOENT; DEFINE_FIELD(unsigned long, ip, FIELD_STRING_IP, 0); - /* Set argument names as fields */ - for (i = 0; i < tp->nr_args; i++) { - ret = trace_define_field(event_call, tp->args[i].type->fmttype, - tp->args[i].name, - sizeof(field) + tp->args[i].offset, - tp->args[i].type->size, - tp->args[i].type->is_signed, - FILTER_OTHER); - if (ret) - return ret; - } - return 0; + + return traceprobe_define_arg_fields(event_call, sizeof(field), tp); } -static int kretprobe_event_define_fields(struct ftrace_event_call *event_call) +static int kretprobe_event_define_fields(struct trace_event_call *event_call) { - int ret, i; + int ret; struct kretprobe_trace_entry_head field; - struct trace_probe *tp = (struct trace_probe *)event_call->data; + struct trace_probe *tp; + + tp = trace_probe_primary_from_call(event_call); + if (WARN_ON_ONCE(!tp)) + return -ENOENT; DEFINE_FIELD(unsigned long, func, FIELD_STRING_FUNC, 0); DEFINE_FIELD(unsigned long, ret_ip, FIELD_STRING_RETIP, 0); - /* Set argument names as fields */ - for (i = 0; i < tp->nr_args; i++) { - ret = trace_define_field(event_call, tp->args[i].type->fmttype, - tp->args[i].name, - sizeof(field) + tp->args[i].offset, - tp->args[i].type->size, - tp->args[i].type->is_signed, - FILTER_OTHER); - if (ret) - return ret; - } - return 0; -} -static int __set_print_fmt(struct trace_probe *tp, char *buf, int len) -{ - int i; - int pos = 0; - - const char *fmt, *arg; - - if (!trace_probe_is_return(tp)) { - fmt = "(%lx)"; - arg = "REC->" FIELD_STRING_IP; - } else { - fmt = "(%lx <- %lx)"; - arg = "REC->" FIELD_STRING_FUNC ", REC->" FIELD_STRING_RETIP; - } - - /* When len=0, we just calculate the needed length */ -#define LEN_OR_ZERO (len ? len - pos : 0) - - pos += snprintf(buf + pos, LEN_OR_ZERO, "\"%s", fmt); - - for (i = 0; i < tp->nr_args; i++) { - pos += snprintf(buf + pos, LEN_OR_ZERO, " %s=%s", - tp->args[i].name, tp->args[i].type->fmt); - } - - pos += snprintf(buf + pos, LEN_OR_ZERO, "\", %s", arg); - - for (i = 0; i < tp->nr_args; i++) { - if (strcmp(tp->args[i].type->name, "string") == 0) - pos += snprintf(buf + pos, LEN_OR_ZERO, - ", __get_str(%s)", - tp->args[i].name); - else - pos += snprintf(buf + pos, LEN_OR_ZERO, ", REC->%s", - tp->args[i].name); - } - -#undef LEN_OR_ZERO - - /* return the length of print_fmt */ - return pos; -} - -static int set_print_fmt(struct trace_probe *tp) -{ - int len; - char *print_fmt; - - /* First: called with 0 length to calculate the needed length */ - len = __set_print_fmt(tp, NULL, 0); - print_fmt = kmalloc(len + 1, GFP_KERNEL); - if (!print_fmt) - return -ENOMEM; - - /* Second: actually write the @print_fmt */ - __set_print_fmt(tp, print_fmt, len + 1); - tp->call.print_fmt = print_fmt; - - return 0; + return traceprobe_define_arg_fields(event_call, sizeof(field), tp); } #ifdef CONFIG_PERF_EVENTS /* Kprobe profile handler */ -static __kprobes void -kprobe_perf_func(struct trace_probe *tp, struct pt_regs *regs) +static int +kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs) { - struct ftrace_event_call *call = &tp->call; + struct trace_event_call *call = trace_probe_event_call(&tk->tp); struct kprobe_trace_entry_head *entry; struct hlist_head *head; int size, __size, dsize; int rctx; + if (bpf_prog_array_valid(call)) { + unsigned long orig_ip = instruction_pointer(regs); + int ret; + + ret = trace_call_bpf(call, regs); + + /* + * We need to check and see if we modified the pc of the + * pt_regs, and if so return 1 so that we don't do the + * single stepping. + */ + if (orig_ip != instruction_pointer(regs)) + return 1; + if (!ret) + return 0; + } + head = this_cpu_ptr(call->perf_events); if (hlist_empty(head)) - return; + return 0; - dsize = __get_data_size(tp, regs); - __size = sizeof(*entry) + tp->size + dsize; + dsize = __get_data_size(&tk->tp, regs, NULL); + __size = sizeof(*entry) + tk->tp.size + dsize; size = ALIGN(__size + sizeof(u32), sizeof(u64)); size -= sizeof(u32); - if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE, - "profile buffer not large enough")) - return; - entry = perf_trace_buf_prepare(size, call->event.type, regs, &rctx); + entry = perf_trace_buf_alloc(size, NULL, &rctx); if (!entry) - return; + return 0; - entry->ip = (unsigned long)tp->rp.kp.addr; + entry->ip = (unsigned long)tk->rp.kp.addr; memset(&entry[1], 0, dsize); - store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize); - perf_trace_buf_submit(entry, size, rctx, 0, 1, regs, head, NULL); + store_trace_args(&entry[1], &tk->tp, regs, NULL, sizeof(*entry), dsize); + perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, + head, NULL); + return 0; } +NOKPROBE_SYMBOL(kprobe_perf_func); /* Kretprobe profile handler */ -static __kprobes void -kretprobe_perf_func(struct trace_probe *tp, struct kretprobe_instance *ri, +static void +kretprobe_perf_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, struct pt_regs *regs) { - struct ftrace_event_call *call = &tp->call; + struct trace_event_call *call = trace_probe_event_call(&tk->tp); struct kretprobe_trace_entry_head *entry; struct hlist_head *head; int size, __size, dsize; int rctx; + if (bpf_prog_array_valid(call) && !trace_call_bpf(call, regs)) + return; + head = this_cpu_ptr(call->perf_events); if (hlist_empty(head)) return; - dsize = __get_data_size(tp, regs); - __size = sizeof(*entry) + tp->size + dsize; + dsize = __get_data_size(&tk->tp, regs, ri->data); + __size = sizeof(*entry) + tk->tp.size + dsize; size = ALIGN(__size + sizeof(u32), sizeof(u64)); size -= sizeof(u32); - if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE, - "profile buffer not large enough")) - return; - entry = perf_trace_buf_prepare(size, call->event.type, regs, &rctx); + entry = perf_trace_buf_alloc(size, NULL, &rctx); if (!entry) return; - entry->func = (unsigned long)tp->rp.kp.addr; - entry->ret_ip = (unsigned long)ri->ret_addr; - store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize); - perf_trace_buf_submit(entry, size, rctx, 0, 1, regs, head, NULL); + entry->func = (unsigned long)tk->rp.kp.addr; + entry->ret_ip = get_kretprobe_retaddr(ri); + store_trace_args(&entry[1], &tk->tp, regs, ri->data, sizeof(*entry), dsize); + perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, + head, NULL); +} +NOKPROBE_SYMBOL(kretprobe_perf_func); + +int bpf_get_kprobe_info(const struct perf_event *event, u32 *fd_type, + const char **symbol, u64 *probe_offset, + u64 *probe_addr, unsigned long *missed, + bool perf_type_tracepoint) +{ + const char *pevent = trace_event_name(event->tp_event); + const char *group = event->tp_event->class->system; + struct trace_kprobe *tk; + + if (perf_type_tracepoint) + tk = find_trace_kprobe(pevent, group); + else + tk = trace_kprobe_primary_from_call(event->tp_event); + if (!tk) + return -EINVAL; + + *fd_type = trace_kprobe_is_return(tk) ? BPF_FD_TYPE_KRETPROBE + : BPF_FD_TYPE_KPROBE; + *probe_offset = tk->rp.kp.offset; + *probe_addr = kallsyms_show_value(current_cred()) ? + (unsigned long)tk->rp.kp.addr : 0; + *symbol = tk->symbol; + if (missed) + *missed = trace_kprobe_missed(tk); + return 0; } #endif /* CONFIG_PERF_EVENTS */ @@ -1141,24 +1786,22 @@ kretprobe_perf_func(struct trace_probe *tp, struct kretprobe_instance *ri, * kprobe_trace_self_tests_init() does enable_trace_probe/disable_trace_probe * lockless, but we can't race with this __init function. */ -static __kprobes -int kprobe_register(struct ftrace_event_call *event, - enum trace_reg type, void *data) +static int kprobe_register(struct trace_event_call *event, + enum trace_reg type, void *data) { - struct trace_probe *tp = (struct trace_probe *)event->data; - struct ftrace_event_file *file = data; + struct trace_event_file *file = data; switch (type) { case TRACE_REG_REGISTER: - return enable_trace_probe(tp, file); + return enable_trace_kprobe(event, file); case TRACE_REG_UNREGISTER: - return disable_trace_probe(tp, file); + return disable_trace_kprobe(event, file); #ifdef CONFIG_PERF_EVENTS case TRACE_REG_PERF_REGISTER: - return enable_trace_probe(tp, NULL); + return enable_trace_kprobe(event, NULL); case TRACE_REG_PERF_UNREGISTER: - return disable_trace_probe(tp, NULL); + return disable_trace_kprobe(event, NULL); case TRACE_REG_PERF_OPEN: case TRACE_REG_PERF_CLOSE: case TRACE_REG_PERF_ADD: @@ -1169,37 +1812,52 @@ int kprobe_register(struct ftrace_event_call *event, return 0; } -static __kprobes -int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs) +static int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs) { - struct trace_probe *tp = container_of(kp, struct trace_probe, rp.kp); + struct trace_kprobe *tk = container_of(kp, struct trace_kprobe, rp.kp); + unsigned int flags = trace_probe_load_flag(&tk->tp); + int ret = 0; - tp->nhit++; + raw_cpu_inc(*tk->nhit); - if (tp->flags & TP_FLAG_TRACE) - kprobe_trace_func(tp, regs); + if (flags & TP_FLAG_TRACE) + kprobe_trace_func(tk, regs); #ifdef CONFIG_PERF_EVENTS - if (tp->flags & TP_FLAG_PROFILE) - kprobe_perf_func(tp, regs); + if (flags & TP_FLAG_PROFILE) + ret = kprobe_perf_func(tk, regs); #endif - return 0; /* We don't tweek kernel, so just return 0 */ + return ret; } +NOKPROBE_SYMBOL(kprobe_dispatcher); -static __kprobes -int kretprobe_dispatcher(struct kretprobe_instance *ri, struct pt_regs *regs) +static int +kretprobe_dispatcher(struct kretprobe_instance *ri, struct pt_regs *regs) { - struct trace_probe *tp = container_of(ri->rp, struct trace_probe, rp); + struct kretprobe *rp = get_kretprobe(ri); + struct trace_kprobe *tk; + unsigned int flags; + + /* + * There is a small chance that get_kretprobe(ri) returns NULL when + * the kretprobe is unregister on another CPU between kretprobe's + * trampoline_handler and this function. + */ + if (unlikely(!rp)) + return 0; - tp->nhit++; + tk = container_of(rp, struct trace_kprobe, rp); + raw_cpu_inc(*tk->nhit); - if (tp->flags & TP_FLAG_TRACE) - kretprobe_trace_func(tp, ri, regs); + flags = trace_probe_load_flag(&tk->tp); + if (flags & TP_FLAG_TRACE) + kretprobe_trace_func(tk, ri, regs); #ifdef CONFIG_PERF_EVENTS - if (tp->flags & TP_FLAG_PROFILE) - kretprobe_perf_func(tp, ri, regs); + if (flags & TP_FLAG_PROFILE) + kretprobe_perf_func(tk, ri, regs); #endif - return 0; /* We don't tweek kernel, so just return 0 */ + return 0; /* We don't tweak kernel, so just return 0 */ } +NOKPROBE_SYMBOL(kretprobe_dispatcher); static struct trace_event_functions kretprobe_funcs = { .trace = print_kretprobe_event @@ -1209,158 +1867,261 @@ static struct trace_event_functions kprobe_funcs = { .trace = print_kprobe_event }; -static int register_probe_event(struct trace_probe *tp) +static struct trace_event_fields kretprobe_fields_array[] = { + { .type = TRACE_FUNCTION_TYPE, + .define_fields = kretprobe_event_define_fields }, + {} +}; + +static struct trace_event_fields kprobe_fields_array[] = { + { .type = TRACE_FUNCTION_TYPE, + .define_fields = kprobe_event_define_fields }, + {} +}; + +static inline void init_trace_event_call(struct trace_kprobe *tk) { - struct ftrace_event_call *call = &tp->call; - int ret; + struct trace_event_call *call = trace_probe_event_call(&tk->tp); - /* Initialize ftrace_event_call */ - INIT_LIST_HEAD(&call->class->fields); - if (trace_probe_is_return(tp)) { + if (trace_kprobe_is_return(tk)) { call->event.funcs = &kretprobe_funcs; - call->class->define_fields = kretprobe_event_define_fields; + call->class->fields_array = kretprobe_fields_array; } else { call->event.funcs = &kprobe_funcs; - call->class->define_fields = kprobe_event_define_fields; - } - if (set_print_fmt(tp) < 0) - return -ENOMEM; - ret = register_ftrace_event(&call->event); - if (!ret) { - kfree(call->print_fmt); - return -ENODEV; + call->class->fields_array = kprobe_fields_array; } - call->flags = 0; + + call->flags = TRACE_EVENT_FL_KPROBE; call->class->reg = kprobe_register; - call->data = tp; - ret = trace_add_event_call(call); - if (ret) { - pr_info("Failed to register kprobe event: %s\n", call->name); - kfree(call->print_fmt); - unregister_ftrace_event(&call->event); +} + +static int register_kprobe_event(struct trace_kprobe *tk) +{ + init_trace_event_call(tk); + + return trace_probe_register_event_call(&tk->tp); +} + +static int unregister_kprobe_event(struct trace_kprobe *tk) +{ + return trace_probe_unregister_event_call(&tk->tp); +} + +#ifdef CONFIG_PERF_EVENTS + +/* create a trace_kprobe, but don't add it to global lists */ +struct trace_event_call * +create_local_trace_kprobe(char *func, void *addr, unsigned long offs, + bool is_return) +{ + enum probe_print_type ptype; + struct trace_kprobe *tk __free(free_trace_kprobe) = NULL; + int ret; + char *event; + + if (func) { + ret = validate_probe_symbol(func); + if (ret) + return ERR_PTR(ret); } - return ret; + + /* + * local trace_kprobes are not added to dyn_event, so they are never + * searched in find_trace_kprobe(). Therefore, there is no concern of + * duplicated name here. + */ + event = func ? func : "DUMMY_EVENT"; + + tk = alloc_trace_kprobe(KPROBE_EVENT_SYSTEM, event, (void *)addr, func, + offs, 0 /* maxactive */, 0 /* nargs */, + is_return); + + if (IS_ERR(tk)) { + pr_info("Failed to allocate trace_probe.(%d)\n", + (int)PTR_ERR(tk)); + return ERR_CAST(tk); + } + + init_trace_event_call(tk); + + ptype = trace_kprobe_is_return(tk) ? + PROBE_PRINT_RETURN : PROBE_PRINT_NORMAL; + if (traceprobe_set_print_fmt(&tk->tp, ptype) < 0) + return ERR_PTR(-ENOMEM); + + ret = __register_trace_kprobe(tk); + if (ret < 0) + return ERR_PTR(ret); + + return trace_probe_event_call(&(no_free_ptr(tk)->tp)); } -static void unregister_probe_event(struct trace_probe *tp) +void destroy_local_trace_kprobe(struct trace_event_call *event_call) { - /* tp->event is unregistered in trace_remove_event_call() */ - trace_remove_event_call(&tp->call); - kfree(tp->call.print_fmt); + struct trace_kprobe *tk; + + tk = trace_kprobe_primary_from_call(event_call); + if (unlikely(!tk)) + return; + + if (trace_probe_is_enabled(&tk->tp)) { + WARN_ON(1); + return; + } + + __unregister_trace_kprobe(tk); + + free_trace_kprobe(tk); } +#endif /* CONFIG_PERF_EVENTS */ -/* Make a debugfs interface for controlling probe points */ -static __init int init_kprobe_trace(void) +static __init void enable_boot_kprobe_events(void) { - struct dentry *d_tracer; - struct dentry *entry; + struct trace_array *tr = top_trace_array(); + struct trace_event_file *file; + struct trace_kprobe *tk; + struct dyn_event *pos; + + guard(mutex)(&event_mutex); + for_each_trace_kprobe(tk, pos) { + list_for_each_entry(file, &tr->events, list) + if (file->event_call == trace_probe_event_call(&tk->tp)) + trace_event_enable_disable(file, 1, 0); + } +} + +static __init void setup_boot_kprobe_events(void) +{ + char *p, *cmd = kprobe_boot_events_buf; + int ret; - if (register_module_notifier(&trace_probe_module_nb)) + strreplace(kprobe_boot_events_buf, ',', ' '); + + while (cmd && *cmd != '\0') { + p = strchr(cmd, ';'); + if (p) + *p++ = '\0'; + + ret = create_or_delete_trace_kprobe(cmd); + if (ret) + pr_warn("Failed to add event(%d): %s\n", ret, cmd); + + cmd = p; + } + + enable_boot_kprobe_events(); +} + +/* + * Register dynevent at core_initcall. This allows kernel to setup kprobe + * events in postcore_initcall without tracefs. + */ +static __init int init_kprobe_trace_early(void) +{ + int ret; + + ret = dyn_event_register(&trace_kprobe_ops); + if (ret) + return ret; + + if (trace_kprobe_register_module_notifier()) return -EINVAL; - d_tracer = tracing_init_dentry(); - if (!d_tracer) - return 0; + return 0; +} +core_initcall(init_kprobe_trace_early); - entry = debugfs_create_file("kprobe_events", 0644, d_tracer, - NULL, &kprobe_events_ops); +/* Make a tracefs interface for controlling probe points */ +static __init int init_kprobe_trace(void) +{ + int ret; + + ret = tracing_init_dentry(); + if (ret) + return 0; /* Event list interface */ - if (!entry) - pr_warning("Could not create debugfs " - "'kprobe_events' entry\n"); + trace_create_file("kprobe_events", TRACE_MODE_WRITE, + NULL, NULL, &kprobe_events_ops); /* Profile interface */ - entry = debugfs_create_file("kprobe_profile", 0444, d_tracer, - NULL, &kprobe_profile_ops); + trace_create_file("kprobe_profile", TRACE_MODE_READ, + NULL, NULL, &kprobe_profile_ops); + + setup_boot_kprobe_events(); - if (!entry) - pr_warning("Could not create debugfs " - "'kprobe_profile' entry\n"); return 0; } fs_initcall(init_kprobe_trace); #ifdef CONFIG_FTRACE_STARTUP_TEST - -/* - * The "__used" keeps gcc from removing the function symbol - * from the kallsyms table. - */ -static __used int kprobe_trace_selftest_target(int a1, int a2, int a3, - int a4, int a5, int a6) -{ - return a1 + a2 + a3 + a4 + a5 + a6; -} - -static struct ftrace_event_file * -find_trace_probe_file(struct trace_probe *tp, struct trace_array *tr) +static __init struct trace_event_file * +find_trace_probe_file(struct trace_kprobe *tk, struct trace_array *tr) { - struct ftrace_event_file *file; + struct trace_event_file *file; list_for_each_entry(file, &tr->events, list) - if (file->event_call == &tp->call) + if (file->event_call == trace_probe_event_call(&tk->tp)) return file; return NULL; } /* - * Nobody but us can call enable_trace_probe/disable_trace_probe at this + * Nobody but us can call enable_trace_kprobe/disable_trace_kprobe at this * stage, we can do this lockless. */ static __init int kprobe_trace_self_tests_init(void) { int ret, warn = 0; int (*target)(int, int, int, int, int, int); - struct trace_probe *tp; - struct ftrace_event_file *file; + struct trace_kprobe *tk; + struct trace_event_file *file; + + if (tracing_is_disabled()) + return -ENODEV; + + if (tracing_selftest_disabled) + return 0; target = kprobe_trace_selftest_target; pr_info("Testing kprobe tracing: "); - ret = traceprobe_command("p:testprobe kprobe_trace_selftest_target " - "$stack $stack0 +0($stack)", - create_trace_probe); - if (WARN_ON_ONCE(ret)) { - pr_warn("error on probing function entry.\n"); + ret = create_or_delete_trace_kprobe("p:testprobe kprobe_trace_selftest_target $stack $stack0 +0($stack)"); + if (WARN_ONCE(ret, "error on probing function entry.")) { warn++; } else { /* Enable trace point */ - tp = find_trace_probe("testprobe", KPROBE_EVENT_SYSTEM); - if (WARN_ON_ONCE(tp == NULL)) { - pr_warn("error on getting new probe.\n"); + tk = find_trace_kprobe("testprobe", KPROBE_EVENT_SYSTEM); + if (WARN_ONCE(tk == NULL, "error on probing function entry.")) { warn++; } else { - file = find_trace_probe_file(tp, top_trace_array()); - if (WARN_ON_ONCE(file == NULL)) { - pr_warn("error on getting probe file.\n"); + file = find_trace_probe_file(tk, top_trace_array()); + if (WARN_ONCE(file == NULL, "error on getting probe file.")) { warn++; } else - enable_trace_probe(tp, file); + enable_trace_kprobe( + trace_probe_event_call(&tk->tp), file); } } - ret = traceprobe_command("r:testprobe2 kprobe_trace_selftest_target " - "$retval", create_trace_probe); - if (WARN_ON_ONCE(ret)) { - pr_warn("error on probing function return.\n"); + ret = create_or_delete_trace_kprobe("r:testprobe2 kprobe_trace_selftest_target $retval"); + if (WARN_ONCE(ret, "error on probing function return.")) { warn++; } else { /* Enable trace point */ - tp = find_trace_probe("testprobe2", KPROBE_EVENT_SYSTEM); - if (WARN_ON_ONCE(tp == NULL)) { - pr_warn("error on getting 2nd new probe.\n"); + tk = find_trace_kprobe("testprobe2", KPROBE_EVENT_SYSTEM); + if (WARN_ONCE(tk == NULL, "error on getting 2nd new probe.")) { warn++; } else { - file = find_trace_probe_file(tp, top_trace_array()); - if (WARN_ON_ONCE(file == NULL)) { - pr_warn("error on getting probe file.\n"); + file = find_trace_probe_file(tk, top_trace_array()); + if (WARN_ONCE(file == NULL, "error on getting probe file.")) { warn++; } else - enable_trace_probe(tp, file); + enable_trace_kprobe( + trace_probe_event_call(&tk->tp), file); } } @@ -1369,47 +2130,62 @@ static __init int kprobe_trace_self_tests_init(void) ret = target(1, 2, 3, 4, 5, 6); + /* + * Not expecting an error here, the check is only to prevent the + * optimizer from removing the call to target() as otherwise there + * are no side-effects and the call is never performed. + */ + if (ret != 21) + warn++; + /* Disable trace points before removing it */ - tp = find_trace_probe("testprobe", KPROBE_EVENT_SYSTEM); - if (WARN_ON_ONCE(tp == NULL)) { - pr_warn("error on getting test probe.\n"); + tk = find_trace_kprobe("testprobe", KPROBE_EVENT_SYSTEM); + if (WARN_ONCE(tk == NULL, "error on getting test probe.")) { warn++; } else { - file = find_trace_probe_file(tp, top_trace_array()); - if (WARN_ON_ONCE(file == NULL)) { - pr_warn("error on getting probe file.\n"); + if (WARN_ONCE(trace_kprobe_nhit(tk) != 1, + "incorrect number of testprobe hits.")) + warn++; + + file = find_trace_probe_file(tk, top_trace_array()); + if (WARN_ONCE(file == NULL, "error on getting probe file.")) { warn++; } else - disable_trace_probe(tp, file); + disable_trace_kprobe( + trace_probe_event_call(&tk->tp), file); } - tp = find_trace_probe("testprobe2", KPROBE_EVENT_SYSTEM); - if (WARN_ON_ONCE(tp == NULL)) { - pr_warn("error on getting 2nd test probe.\n"); + tk = find_trace_kprobe("testprobe2", KPROBE_EVENT_SYSTEM); + if (WARN_ONCE(tk == NULL, "error on getting 2nd test probe.")) { warn++; } else { - file = find_trace_probe_file(tp, top_trace_array()); - if (WARN_ON_ONCE(file == NULL)) { - pr_warn("error on getting probe file.\n"); + if (WARN_ONCE(trace_kprobe_nhit(tk) != 1, + "incorrect number of testprobe2 hits.")) + warn++; + + file = find_trace_probe_file(tk, top_trace_array()); + if (WARN_ONCE(file == NULL, "error on getting probe file.")) { warn++; } else - disable_trace_probe(tp, file); + disable_trace_kprobe( + trace_probe_event_call(&tk->tp), file); } - ret = traceprobe_command("-:testprobe", create_trace_probe); - if (WARN_ON_ONCE(ret)) { - pr_warn("error on deleting a probe.\n"); + ret = create_or_delete_trace_kprobe("-:testprobe"); + if (WARN_ONCE(ret, "error on deleting a probe.")) warn++; - } - ret = traceprobe_command("-:testprobe2", create_trace_probe); - if (WARN_ON_ONCE(ret)) { - pr_warn("error on deleting a probe.\n"); + ret = create_or_delete_trace_kprobe("-:testprobe2"); + if (WARN_ONCE(ret, "error on deleting a probe.")) warn++; - } + end: - release_all_trace_probes(); + /* + * Wait for the optimizer work to finish. Otherwise it might fiddle + * with probes in already freed __init text. + */ + wait_for_kprobe_optimizer(); if (warn) pr_cont("NG: Some tests are failed. Please check them.\n"); else |