1 files changed, 292 insertions, 258 deletions

diff --git a/kernel/trace/trace_eprobe.c b/kernel/trace/trace_eprobe.c
index 191db32dec46..3ee39715d5e4 100644
--- a/kernel/trace/trace_eprobe.c
+++ b/kernel/trace/trace_eprobe.c
@@ -9,12 +9,14 @@
  * Copyright (C) 2021, VMware Inc, Tzvetomir Stoyanov tz.stoyanov@gmail.com>
  *
  */
+#include <linux/cleanup.h>
+#include <linux/ftrace.h>
 #include <linux/module.h>
 #include <linux/mutex.h>
-#include <linux/ftrace.h>
 
 #include "trace_dynevent.h"
 #include "trace_probe.h"
+#include "trace_probe_kernel.h"
 #include "trace_probe_tmpl.h"
 
 #define EPROBE_EVENT_SYSTEM "eprobes"
@@ -26,6 +28,9 @@ struct trace_eprobe {
 	/* tracepoint event */
 	const char *event_name;
 
+	/* filter string for the tracepoint */
+	char *filter_str;
+
 	struct trace_event_call *event;
 
 	struct dyn_event	devent;
@@ -37,6 +42,10 @@ struct eprobe_data {
 	struct trace_eprobe	*ep;
 };
 
+
+#define for_each_trace_eprobe_tp(ep, _tp) \
+	list_for_each_entry(ep, trace_probe_probe_list(_tp), tp.list)
+
 static int __trace_eprobe_create(int argc, const char *argv[]);
 
 static void trace_event_probe_cleanup(struct trace_eprobe *ep)
@@ -48,9 +57,13 @@ static void trace_event_probe_cleanup(struct trace_eprobe *ep)
 	kfree(ep->event_system);
 	if (ep->event)
 		trace_event_put_ref(ep->event);
+	kfree(ep->filter_str);
 	kfree(ep);
 }
 
+DEFINE_FREE(trace_event_probe_cleanup, struct trace_eprobe *,
+		if (!IS_ERR_OR_NULL(_T)) trace_event_probe_cleanup(_T))
+
 static struct trace_eprobe *to_trace_eprobe(struct dyn_event *ev)
 {
 	return container_of(ev, struct trace_eprobe, devent);
@@ -125,6 +138,7 @@ static bool eprobe_dyn_event_match(const char *system, const char *event,
 	 * We match the following:
 	 *  event only			- match all eprobes with event name
 	 *  system and event only	- match all system/event probes
+	 *  system only			- match all system probes
 	 *
 	 * The below has the above satisfied with more arguments:
 	 *
@@ -143,7 +157,7 @@ static bool eprobe_dyn_event_match(const char *system, const char *event,
 		return false;
 
 	/* Must match the event name */
-	if (strcmp(trace_probe_name(&ep->tp), event) != 0)
+	if (event[0] != '\0' && strcmp(trace_probe_name(&ep->tp), event) != 0)
 		return false;
 
 	/* No arguments match all */
@@ -186,10 +200,10 @@ static struct trace_eprobe *alloc_event_probe(const char *group,
 					      struct trace_event_call *event,
 					      int nargs)
 {
-	struct trace_eprobe *ep;
+	struct trace_eprobe *ep __free(trace_event_probe_cleanup) = NULL;
 	const char *event_name;
 	const char *sys_name;
-	int ret = -ENOMEM;
+	int ret;
 
 	if (!event)
 		return ERR_PTR(-ENODEV);
@@ -200,49 +214,26 @@ static struct trace_eprobe *alloc_event_probe(const char *group,
 	ep = kzalloc(struct_size(ep, tp.args, nargs), GFP_KERNEL);
 	if (!ep) {
 		trace_event_put_ref(event);
-		goto error;
+		return ERR_PTR(-ENOMEM);
 	}
 	ep->event = event;
 	ep->event_name = kstrdup(event_name, GFP_KERNEL);
 	if (!ep->event_name)
-		goto error;
+		return ERR_PTR(-ENOMEM);
 	ep->event_system = kstrdup(sys_name, GFP_KERNEL);
 	if (!ep->event_system)
-		goto error;
+		return ERR_PTR(-ENOMEM);
 
-	ret = trace_probe_init(&ep->tp, this_event, group, false);
+	ret = trace_probe_init(&ep->tp, this_event, group, false, nargs);
 	if (ret < 0)
-		goto error;
+		return ERR_PTR(ret);
 
 	dyn_event_init(&ep->devent, &eprobe_dyn_event_ops);
-	return ep;
-error:
-	trace_event_probe_cleanup(ep);
-	return ERR_PTR(ret);
-}
-
-static int trace_eprobe_tp_arg_update(struct trace_eprobe *ep, int i)
-{
-	struct probe_arg *parg = &ep->tp.args[i];
-	struct ftrace_event_field *field;
-	struct list_head *head;
-
-	head = trace_get_fields(ep->event);
-	list_for_each_entry(field, head, link) {
-		if (!strcmp(parg->code->data, field->name)) {
-			kfree(parg->code->data);
-			parg->code->data = field;
-			return 0;
-		}
-	}
-	kfree(parg->code->data);
-	parg->code->data = NULL;
-	return -ENOENT;
+	return_ptr(ep);
 }
 
 static int eprobe_event_define_fields(struct trace_event_call *event_call)
 {
-	int ret;
 	struct eprobe_trace_entry_head field;
 	struct trace_probe *tp;
 
@@ -250,8 +241,6 @@ static int eprobe_event_define_fields(struct trace_event_call *event_call)
 	if (WARN_ON_ONCE(!tp))
 		return -ENOENT;
 
-	DEFINE_FIELD(unsigned int, type, FIELD_STRING_TYPE, 0);
-
 	return traceprobe_define_arg_fields(event_call, sizeof(field), tp);
 }
 
@@ -270,7 +259,9 @@ print_eprobe_event(struct trace_iterator *iter, int flags,
 	struct trace_event_call *pevent;
 	struct trace_event *probed_event;
 	struct trace_seq *s = &iter->seq;
+	struct trace_eprobe *ep;
 	struct trace_probe *tp;
+	unsigned int type;
 
 	field = (struct eprobe_trace_entry_head *)iter->ent;
 	tp = trace_probe_primary_from_call(
@@ -278,20 +269,23 @@ print_eprobe_event(struct trace_iterator *iter, int flags,
 	if (WARN_ON_ONCE(!tp))
 		goto out;
 
+	ep = container_of(tp, struct trace_eprobe, tp);
+	type = ep->event->event.type;
+
 	trace_seq_printf(s, "%s: (", trace_probe_name(tp));
 
-	probed_event = ftrace_find_event(field->type);
+	probed_event = ftrace_find_event(type);
 	if (probed_event) {
 		pevent = container_of(probed_event, struct trace_event_call, event);
 		trace_seq_printf(s, "%s.%s", pevent->class->system,
 				 trace_event_name(pevent));
 	} else {
-		trace_seq_printf(s, "%u", field->type);
+		trace_seq_printf(s, "%u", type);
 	}
 
 	trace_seq_putc(s, ')');
 
-	if (print_probe_args(s, tp->args, tp->nr_args,
+	if (trace_probe_print_args(s, tp->args, tp->nr_args,
 			     (u8 *)&field[1], field) < 0)
 		goto out;
 
@@ -300,7 +294,8 @@ print_eprobe_event(struct trace_iterator *iter, int flags,
 	return trace_handle_return(s);
 }
 
-static unsigned long get_event_field(struct fetch_insn *code, void *rec)
+static nokprobe_inline unsigned long
+get_event_field(struct fetch_insn *code, void *rec)
 {
 	struct ftrace_event_field *field = code->data;
 	unsigned long val;
@@ -308,6 +303,27 @@ static unsigned long get_event_field(struct fetch_insn *code, void *rec)
 
 	addr = rec + field->offset;
 
+	if (is_string_field(field)) {
+		switch (field->filter_type) {
+		case FILTER_DYN_STRING:
+			val = (unsigned long)(rec + (*(unsigned int *)addr & 0xffff));
+			break;
+		case FILTER_RDYN_STRING:
+			val = (unsigned long)(addr + (*(unsigned int *)addr & 0xffff));
+			break;
+		case FILTER_STATIC_STRING:
+			val = (unsigned long)addr;
+			break;
+		case FILTER_PTR_STRING:
+			val = (unsigned long)(*(char *)addr);
+			break;
+		default:
+			WARN_ON_ONCE(1);
+			return 0;
+		}
+		return val;
+	}
+
 	switch (field->size) {
 	case 1:
 		if (field->is_signed)
@@ -328,10 +344,15 @@ static unsigned long get_event_field(struct fetch_insn *code, void *rec)
 			val = *(unsigned int *)addr;
 		break;
 	default:
-		if (field->is_signed)
-			val = *(long *)addr;
-		else
-			val = *(unsigned long *)addr;
+		if (field->size == sizeof(long)) {
+			if (field->is_signed)
+				val = *(long *)addr;
+			else
+				val = *(unsigned long *)addr;
+			break;
+		}
+		/* This is an array, point to the addr itself */
+		val = (unsigned long)addr;
 		break;
 	}
 	return val;
@@ -339,16 +360,30 @@ static unsigned long get_event_field(struct fetch_insn *code, void *rec)
 
 static int get_eprobe_size(struct trace_probe *tp, void *rec)
 {
+	struct fetch_insn *code;
 	struct probe_arg *arg;
 	int i, len, ret = 0;
 
 	for (i = 0; i < tp->nr_args; i++) {
 		arg = tp->args + i;
-		if (unlikely(arg->dynamic)) {
+		if (arg->dynamic) {
 			unsigned long val;
 
-			val = get_event_field(arg->code, rec);
-			len = process_fetch_insn_bottom(arg->code + 1, val, NULL, NULL);
+			code = arg->code;
+ retry:
+			switch (code->op) {
+			case FETCH_OP_TP_ARG:
+				val = get_event_field(code, rec);
+				break;
+			case FETCH_NOP_SYMBOL:	/* Ignore a place holder */
+				code++;
+				goto retry;
+			default:
+				if (process_common_fetch_insn(code, &val) < 0)
+					continue;
+			}
+			code++;
+			len = process_fetch_insn_bottom(code, val, NULL, NULL);
 			if (len > 0)
 				ret += len;
 		}
@@ -361,119 +396,30 @@ static int get_eprobe_size(struct trace_probe *tp, void *rec)
 
 /* 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 *dest,
-		   void *base)
+process_fetch_insn(struct fetch_insn *code, void *rec, void *edata,
+		   void *dest, void *base)
 {
 	unsigned long val;
+	int ret;
 
-	val = get_event_field(code, rec);
-	return process_fetch_insn_bottom(code + 1, val, dest, base);
+ retry:
+	switch (code->op) {
+	case FETCH_OP_TP_ARG:
+		val = get_event_field(code, rec);
+		break;
+	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)
 
-/* Return the length of string -- including null terminal byte */
-static nokprobe_inline int
-fetch_store_strlen_user(unsigned long addr)
-{
-	const void __user *uaddr =  (__force const void __user *)addr;
-
-	return strnlen_user_nofault(uaddr, MAX_STRING_SIZE);
-}
-
-/* Return the length of string -- including null terminal byte */
-static nokprobe_inline int
-fetch_store_strlen(unsigned long addr)
-{
-	int ret, len = 0;
-	u8 c;
-
-#ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE
-	if (addr < TASK_SIZE)
-		return fetch_store_strlen_user(addr);
-#endif
-
-	do {
-		ret = copy_from_kernel_nofault(&c, (u8 *)addr + len, 1);
-		len++;
-	} while (c && ret == 0 && len < MAX_STRING_SIZE);
-
-	return (ret < 0) ? ret : len;
-}
-
-/*
- * Fetch a null-terminated string from user. Caller MUST set *(u32 *)buf
- * with max length and relative data location.
- */
-static nokprobe_inline int
-fetch_store_string_user(unsigned long addr, void *dest, void *base)
-{
-	const void __user *uaddr =  (__force const void __user *)addr;
-	int maxlen = get_loc_len(*(u32 *)dest);
-	void *__dest;
-	long ret;
-
-	if (unlikely(!maxlen))
-		return -ENOMEM;
-
-	__dest = get_loc_data(dest, base);
-
-	ret = strncpy_from_user_nofault(__dest, uaddr, maxlen);
-	if (ret >= 0)
-		*(u32 *)dest = make_data_loc(ret, __dest - base);
-
-	return ret;
-}
-
-/*
- * Fetch a null-terminated string. Caller MUST set *(u32 *)buf with max
- * length and relative data location.
- */
-static nokprobe_inline int
-fetch_store_string(unsigned long addr, void *dest, void *base)
-{
-	int maxlen = get_loc_len(*(u32 *)dest);
-	void *__dest;
-	long ret;
-
-#ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE
-	if ((unsigned long)addr < TASK_SIZE)
-		return fetch_store_string_user(addr, dest, base);
-#endif
-
-	if (unlikely(!maxlen))
-		return -ENOMEM;
-
-	__dest = get_loc_data(dest, base);
-
-	/*
-	 * Try to get string again, since the string can be changed while
-	 * probing.
-	 */
-	ret = strncpy_from_kernel_nofault(__dest, (void *)addr, maxlen);
-	if (ret >= 0)
-		*(u32 *)dest = make_data_loc(ret, __dest - base);
-
-	return ret;
-}
-
-static nokprobe_inline int
-probe_mem_read_user(void *dest, void *src, size_t size)
-{
-	const void __user *uaddr =  (__force const void __user *)src;
-
-	return copy_from_user_nofault(dest, uaddr, size);
-}
-
-static nokprobe_inline int
-probe_mem_read(void *dest, void *src, size_t size)
-{
-#ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE
-	if ((unsigned long)src < TASK_SIZE)
-		return probe_mem_read_user(dest, src, size);
-#endif
-	return copy_from_kernel_nofault(dest, src, size);
-}
-
 /* eprobe handler */
 static inline void
 __eprobe_trace_func(struct eprobe_data *edata, void *rec)
@@ -498,11 +444,7 @@ __eprobe_trace_func(struct eprobe_data *edata, void *rec)
 		return;
 
 	entry = fbuffer.entry = ring_buffer_event_data(fbuffer.event);
-	if (edata->ep->event)
-		entry->type = edata->ep->event->event.type;
-	else
-		entry->type = 0;
-	store_trace_args(&entry[1], &edata->ep->tp, rec, sizeof(*entry), dsize);
+	store_trace_args(&entry[1], &edata->ep->tp, rec, NULL, sizeof(*entry), dsize);
 
 	trace_event_buffer_commit(&fbuffer);
 }
@@ -513,20 +455,17 @@ __eprobe_trace_func(struct eprobe_data *edata, void *rec)
  * functions are just stubs to fulfill what is needed to use the trigger
  * infrastructure.
  */
-static int eprobe_trigger_init(struct event_trigger_ops *ops,
-			       struct event_trigger_data *data)
+static int eprobe_trigger_init(struct event_trigger_data *data)
 {
 	return 0;
 }
 
-static void eprobe_trigger_free(struct event_trigger_ops *ops,
-				struct event_trigger_data *data)
+static void eprobe_trigger_free(struct event_trigger_data *data)
 {
 
 }
 
 static int eprobe_trigger_print(struct seq_file *m,
-				struct event_trigger_ops *ops,
 				struct event_trigger_data *data)
 {
 	/* Do not print eprobe event triggers */
@@ -539,19 +478,16 @@ static void eprobe_trigger_func(struct event_trigger_data *data,
 {
 	struct eprobe_data *edata = data->private_data;
 
+	if (unlikely(!rec))
+		return;
+
 	__eprobe_trace_func(edata, rec);
 }
 
-static struct event_trigger_ops eprobe_trigger_ops = {
-	.trigger		= eprobe_trigger_func,
-	.print			= eprobe_trigger_print,
-	.init			= eprobe_trigger_init,
-	.free			= eprobe_trigger_free,
-};
-
 static int eprobe_trigger_cmd_parse(struct event_command *cmd_ops,
 				    struct trace_event_file *file,
-				    char *glob, char *cmd, char *param)
+				    char *glob, char *cmd,
+				    char *param_and_filter)
 {
 	return -1;
 }
@@ -570,12 +506,6 @@ static void eprobe_trigger_unreg_func(char *glob,
 
 }
 
-static struct event_trigger_ops *eprobe_trigger_get_ops(char *cmd,
-							char *param)
-{
-	return &eprobe_trigger_ops;
-}
-
 static struct event_command event_trigger_cmd = {
 	.name			= "eprobe",
 	.trigger_type		= ETT_EVENT_EPROBE,
@@ -584,27 +514,30 @@ static struct event_command event_trigger_cmd = {
 	.reg			= eprobe_trigger_reg_func,
 	.unreg			= eprobe_trigger_unreg_func,
 	.unreg_all		= NULL,
-	.get_trigger_ops	= eprobe_trigger_get_ops,
 	.set_filter		= NULL,
+	.trigger		= eprobe_trigger_func,
+	.print			= eprobe_trigger_print,
+	.init			= eprobe_trigger_init,
+	.free			= eprobe_trigger_free,
 };
 
 static struct event_trigger_data *
 new_eprobe_trigger(struct trace_eprobe *ep, struct trace_event_file *file)
 {
 	struct event_trigger_data *trigger;
+	struct event_filter *filter = NULL;
 	struct eprobe_data *edata;
+	int ret;
 
 	edata = kzalloc(sizeof(*edata), GFP_KERNEL);
 	trigger = kzalloc(sizeof(*trigger), GFP_KERNEL);
 	if (!trigger || !edata) {
-		kfree(edata);
-		kfree(trigger);
-		return ERR_PTR(-ENOMEM);
+		ret = -ENOMEM;
+		goto error;
 	}
 
 	trigger->flags = EVENT_TRIGGER_FL_PROBE;
 	trigger->count = -1;
-	trigger->ops = &eprobe_trigger_ops;
 
 	/*
 	 * EVENT PROBE triggers are not registered as commands with
@@ -614,13 +547,25 @@ new_eprobe_trigger(struct trace_eprobe *ep, struct trace_event_file *file)
 	trigger->cmd_ops = &event_trigger_cmd;
 
 	INIT_LIST_HEAD(&trigger->list);
-	RCU_INIT_POINTER(trigger->filter, NULL);
+
+	if (ep->filter_str) {
+		ret = create_event_filter(file->tr, ep->event,
+					ep->filter_str, false, &filter);
+		if (ret)
+			goto error;
+	}
+	RCU_INIT_POINTER(trigger->filter, filter);
 
 	edata->file = file;
 	edata->ep = ep;
 	trigger->private_data = edata;
 
 	return trigger;
+error:
+	free_event_filter(filter);
+	kfree(edata);
+	kfree(trigger);
+	return ERR_PTR(ret);
 }
 
 static int enable_eprobe(struct trace_eprobe *ep,
@@ -652,22 +597,25 @@ static struct trace_event_functions eprobe_funcs = {
 static int disable_eprobe(struct trace_eprobe *ep,
 			  struct trace_array *tr)
 {
-	struct event_trigger_data *trigger;
+	struct event_trigger_data *trigger = NULL, *iter;
 	struct trace_event_file *file;
+	struct event_filter *filter;
 	struct eprobe_data *edata;
 
 	file = find_event_file(tr, ep->event_system, ep->event_name);
 	if (!file)
 		return -ENOENT;
 
-	list_for_each_entry(trigger, &file->triggers, list) {
-		if (!(trigger->flags & EVENT_TRIGGER_FL_PROBE))
+	list_for_each_entry(iter, &file->triggers, list) {
+		if (!(iter->flags & EVENT_TRIGGER_FL_PROBE))
 			continue;
-		edata = trigger->private_data;
-		if (edata->ep == ep)
+		edata = iter->private_data;
+		if (edata->ep == ep) {
+			trigger = iter;
 			break;
+		}
 	}
-	if (list_entry_is_head(trigger, &file->triggers, list))
+	if (!trigger)
 		return -ENODEV;
 
 	list_del_rcu(&trigger->list);
@@ -678,6 +626,10 @@ static int disable_eprobe(struct trace_eprobe *ep,
 	/* Make sure nothing is using the edata or trigger */
 	tracepoint_synchronize_unregister();
 
+	filter = rcu_access_pointer(trigger->filter);
+
+	if (filter)
+		free_event_filter(filter);
 	kfree(edata);
 	kfree(trigger);
 
@@ -687,10 +639,11 @@ static int disable_eprobe(struct trace_eprobe *ep,
 static int enable_trace_eprobe(struct trace_event_call *call,
 			       struct trace_event_file *file)
 {
-	struct trace_probe *pos, *tp;
+	struct trace_probe *tp;
 	struct trace_eprobe *ep;
 	bool enabled;
 	int ret = 0;
+	int cnt = 0;
 
 	tp = trace_probe_primary_from_call(call);
 	if (WARN_ON_ONCE(!tp))
@@ -708,18 +661,29 @@ static int enable_trace_eprobe(struct trace_event_call *call,
 	if (enabled)
 		return 0;
 
-	list_for_each_entry(pos, trace_probe_probe_list(tp), list) {
-		ep = container_of(pos, struct trace_eprobe, tp);
+	for_each_trace_eprobe_tp(ep, tp) {
 		ret = enable_eprobe(ep, file);
 		if (ret)
 			break;
 		enabled = true;
+		cnt++;
 	}
 
 	if (ret) {
 		/* Failed to enable one of them. Roll back all */
-		if (enabled)
-			disable_eprobe(ep, file->tr);
+		if (enabled) {
+			/*
+			 * It's a bug if one failed for something other than memory
+			 * not being available but another eprobe succeeded.
+			 */
+			WARN_ON_ONCE(ret != -ENOMEM);
+
+			for_each_trace_eprobe_tp(ep, tp) {
+				disable_eprobe(ep, file->tr);
+				if (!--cnt)
+					break;
+			}
+		}
 		if (file)
 			trace_probe_remove_file(tp, file);
 		else
@@ -732,7 +696,7 @@ static int enable_trace_eprobe(struct trace_event_call *call,
 static int disable_trace_eprobe(struct trace_event_call *call,
 				struct trace_event_file *file)
 {
-	struct trace_probe *pos, *tp;
+	struct trace_probe *tp;
 	struct trace_eprobe *ep;
 
 	tp = trace_probe_primary_from_call(call);
@@ -749,10 +713,8 @@ static int disable_trace_eprobe(struct trace_event_call *call,
 		trace_probe_clear_flag(tp, TP_FLAG_PROFILE);
 
 	if (!trace_probe_is_enabled(tp)) {
-		list_for_each_entry(pos, trace_probe_probe_list(tp), list) {
-			ep = container_of(pos, struct trace_eprobe, tp);
+		for_each_trace_eprobe_tp(ep, tp)
 			disable_eprobe(ep, file->tr);
-		}
 	}
 
  out:
@@ -821,28 +783,53 @@ find_and_get_event(const char *system, const char *event_name)
 		name = trace_event_name(tp_event);
 		if (!name || strcmp(event_name, name))
 			continue;
-		if (!trace_event_try_get_ref(tp_event)) {
+		if (!trace_event_try_get_ref(tp_event))
 			return NULL;
-			break;
-		}
 		return tp_event;
-		break;
 	}
 	return NULL;
 }
 
-static int trace_eprobe_tp_update_arg(struct trace_eprobe *ep, const char *argv[], int i)
+static int trace_eprobe_parse_filter(struct trace_eprobe *ep, int argc, const char *argv[])
 {
-	unsigned int flags = TPARG_FL_KERNEL | TPARG_FL_TPOINT;
-	int ret;
+	struct event_filter *dummy = NULL;
+	int i, ret, len = 0;
+	char *p;
 
-	ret = traceprobe_parse_probe_arg(&ep->tp, i, argv[i], flags);
-	if (ret)
-		return ret;
+	if (argc == 0) {
+		trace_probe_log_err(0, NO_EP_FILTER);
+		return -EINVAL;
+	}
 
-	if (ep->tp.args[i].code->op == FETCH_OP_TP_ARG)
-		ret = trace_eprobe_tp_arg_update(ep, i);
+	/* Recover the filter string */
+	for (i = 0; i < argc; i++)
+		len += strlen(argv[i]) + 1;
+
+	ep->filter_str = kzalloc(len, GFP_KERNEL);
+	if (!ep->filter_str)
+		return -ENOMEM;
+
+	p = ep->filter_str;
+	for (i = 0; i < argc; i++) {
+		if (i)
+			ret = snprintf(p, len, " %s", argv[i]);
+		else
+			ret = snprintf(p, len, "%s", argv[i]);
+		p += ret;
+		len -= ret;
+	}
 
+	/*
+	 * Ensure the filter string can be parsed correctly. Note, this
+	 * filter string is for the original event, not for the eprobe.
+	 */
+	ret = create_event_filter(top_trace_array(), ep->event, ep->filter_str,
+				  true, &dummy);
+	free_event_filter(dummy);
+	if (ret) {
+		kfree(ep->filter_str);
+		ep->filter_str = NULL;
+	}
 	return ret;
 }
 
@@ -850,89 +837,136 @@ static int __trace_eprobe_create(int argc, const char *argv[])
 {
 	/*
 	 * Argument syntax:
-	 *      e[:[GRP/]ENAME] SYSTEM.EVENT [FETCHARGS]
-	 * Fetch args:
+	 *      e[:[GRP/][ENAME]] SYSTEM.EVENT [FETCHARGS] [if FILTER]
+	 * Fetch args (no space):
 	 *  <name>=$<field>[:TYPE]
 	 */
+	struct traceprobe_parse_context *ctx __free(traceprobe_parse_context) = NULL;
+	struct trace_eprobe *ep __free(trace_event_probe_cleanup) = NULL;
+	const char *trlog __free(trace_probe_log_clear) = NULL;
 	const char *event = NULL, *group = EPROBE_EVENT_SYSTEM;
 	const char *sys_event = NULL, *sys_name = NULL;
 	struct trace_event_call *event_call;
-	struct trace_eprobe *ep = NULL;
-	char buf1[MAX_EVENT_NAME_LEN];
-	char buf2[MAX_EVENT_NAME_LEN];
-	int ret = 0;
-	int i;
+	char *buf1 __free(kfree) = NULL;
+	char *buf2 __free(kfree) = NULL;
+	char *gbuf __free(kfree) = NULL;
+	int ret = 0, filter_idx = 0;
+	int i, filter_cnt;
 
 	if (argc < 2 || argv[0][0] != 'e')
 		return -ECANCELED;
 
-	trace_probe_log_init("event_probe", argc, argv);
+	trlog = trace_probe_log_init("event_probe", argc, argv);
 
 	event = strchr(&argv[0][1], ':');
 	if (event) {
+		gbuf = kmalloc(MAX_EVENT_NAME_LEN, GFP_KERNEL);
+		if (!gbuf)
+			return -ENOMEM;
 		event++;
-		ret = traceprobe_parse_event_name(&event, &group, buf1,
+		ret = traceprobe_parse_event_name(&event, &group, gbuf,
 						  event - argv[0]);
 		if (ret)
-			goto parse_error;
-	} else {
-		strscpy(buf1, argv[1], MAX_EVENT_NAME_LEN);
-		sanitize_event_name(buf1);
-		event = buf1;
+			return -EINVAL;
 	}
-	if (!is_good_name(event) || !is_good_name(group))
-		goto parse_error;
 
+	trace_probe_log_set_index(1);
 	sys_event = argv[1];
-	ret = traceprobe_parse_event_name(&sys_event, &sys_name, buf2,
-					  sys_event - argv[1]);
-	if (ret || !sys_name)
-		goto parse_error;
-	if (!is_good_name(sys_event) || !is_good_name(sys_name))
-		goto parse_error;
-
-	mutex_lock(&event_mutex);
-	event_call = find_and_get_event(sys_name, sys_event);
-	ep = alloc_event_probe(group, event, event_call, argc - 2);
-	mutex_unlock(&event_mutex);
+
+	buf2 = kmalloc(MAX_EVENT_NAME_LEN, GFP_KERNEL);
+	if (!buf2)
+		return -ENOMEM;
+
+	ret = traceprobe_parse_event_name(&sys_event, &sys_name, buf2, 0);
+	if (ret || !sys_event || !sys_name) {
+		trace_probe_log_err(0, NO_EVENT_INFO);
+		return -EINVAL;
+	}
+
+	if (!event) {
+		buf1 = kstrdup(sys_event, GFP_KERNEL);
+		if (!buf1)
+			return -ENOMEM;
+		event = buf1;
+	}
+
+	for (i = 2; i < argc; i++) {
+		if (!strcmp(argv[i], "if")) {
+			filter_idx = i + 1;
+			filter_cnt = argc - filter_idx;
+			argc = i;
+			break;
+		}
+	}
+
+	if (argc - 2 > MAX_TRACE_ARGS) {
+		trace_probe_log_set_index(2);
+		trace_probe_log_err(0, TOO_MANY_ARGS);
+		return -E2BIG;
+	}
+
+	scoped_guard(mutex, &event_mutex) {
+		event_call = find_and_get_event(sys_name, sys_event);
+		ep = alloc_event_probe(group, event, event_call, argc - 2);
+	}
 
 	if (IS_ERR(ep)) {
 		ret = PTR_ERR(ep);
+		if (ret == -ENODEV)
+			trace_probe_log_err(0, BAD_ATTACH_EVENT);
 		/* This must return -ENOMEM or missing event, else there is a bug */
 		WARN_ON_ONCE(ret != -ENOMEM && ret != -ENODEV);
-		ep = NULL;
-		goto error;
+		return ret;
 	}
 
+	if (filter_idx) {
+		trace_probe_log_set_index(filter_idx);
+		ret = trace_eprobe_parse_filter(ep, filter_cnt, argv + filter_idx);
+		if (ret)
+			return -EINVAL;
+	} else
+		ep->filter_str = NULL;
+
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+	ctx->event = ep->event;
+	ctx->flags = TPARG_FL_KERNEL | TPARG_FL_TEVENT;
+
 	argc -= 2; argv += 2;
 	/* parse arguments */
-	for (i = 0; i < argc && i < MAX_TRACE_ARGS; i++) {
+	for (i = 0; i < argc; i++) {
 		trace_probe_log_set_index(i + 2);
-		ret = trace_eprobe_tp_update_arg(ep, argv, i);
+
+		ret = traceprobe_parse_probe_arg(&ep->tp, i, argv[i], ctx);
+		/* Handle symbols "@" */
+		if (!ret)
+			ret = traceprobe_update_arg(&ep->tp.args[i]);
 		if (ret)
-			goto error;
+			return ret;
 	}
 	ret = traceprobe_set_print_fmt(&ep->tp, PROBE_PRINT_EVENT);
 	if (ret < 0)
-		goto error;
+		return ret;
+
 	init_trace_eprobe_call(ep);
-	mutex_lock(&event_mutex);
-	ret = trace_probe_register_event_call(&ep->tp);
-	if (ret) {
-		if (ret == -EEXIST) {
-			trace_probe_log_set_index(0);
-			trace_probe_log_err(0, EVENT_EXIST);
+	scoped_guard(mutex, &event_mutex) {
+		ret = trace_probe_register_event_call(&ep->tp);
+		if (ret) {
+			if (ret == -EEXIST) {
+				trace_probe_log_set_index(0);
+				trace_probe_log_err(0, EVENT_EXIST);
+			}
+			return ret;
 		}
-		mutex_unlock(&event_mutex);
-		goto error;
+		ret = dyn_event_add(&ep->devent, &ep->tp.event->call);
+		if (ret < 0) {
+			trace_probe_unregister_event_call(&ep->tp);
+			return ret;
+		}
+		/* To avoid freeing registered eprobe event, clear ep. */
+		ep = NULL;
 	}
-	ret = dyn_event_add(&ep->devent, &ep->tp.event->call);
-	mutex_unlock(&event_mutex);
-	return ret;
-parse_error:
-	ret = -EINVAL;
-error:
-	trace_event_probe_cleanup(ep);
 	return ret;
 }