linux-trace-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH 0/6] kprobes: jump label: Cleanup with guard and __free
@ 2025-01-03 23:01 Masami Hiramatsu (Google)
  2025-01-03 23:01 ` [PATCH 1/6] tracing/kprobes: Fix to free objects when failed to copy a symbol Masami Hiramatsu (Google)
                   ` (5 more replies)
  0 siblings, 6 replies; 13+ messages in thread
From: Masami Hiramatsu (Google) @ 2025-01-03 23:01 UTC (permalink / raw)
  To: Steven Rostedt, Peter Zijlstra
  Cc: Anil S Keshavamurthy, Masami Hiramatsu, David S . Miller,
	Mathieu Desnoyers, Oleg Nesterov, Tzvetomir Stoyanov,
	Naveen N Rao, Josh Poimboeuf, Jason Baron, Ard Biesheuvel,
	linux-kernel, linux-trace-kernel

Hi,

Here is a series of patches to fix and cleanup probe events in ftrace with __free().

Thanks,

---

Masami Hiramatsu (Google) (6):
      tracing/kprobes: Fix to free objects when failed to copy a symbol
      Provide __free(argv) for argv_split() users
      tracing: Use __free() for argv in dynevent
      tracing: Use __free() in trace_probe for cleanup
      tracing: Use __free() for kprobe events to cleanup
      tracing/kprobes: Simplify __trace_kprobe_create() by removing gotos


 include/linux/string.h        |    2 +
 kernel/trace/trace_dynevent.c |   47 ++++++----------
 kernel/trace/trace_kprobe.c   |  123 +++++++++++++++++++----------------------
 kernel/trace/trace_probe.c    |   52 ++++++-----------
 4 files changed, 97 insertions(+), 127 deletions(-)

--
Masami Hiramatsu (Google) <mhiramat@kernel.org>

^ permalink raw reply	[flat|nested] 13+ messages in thread

* [PATCH 1/6] tracing/kprobes: Fix to free objects when failed to copy a symbol
  2025-01-03 23:01 [PATCH 0/6] kprobes: jump label: Cleanup with guard and __free Masami Hiramatsu (Google)
@ 2025-01-03 23:01 ` Masami Hiramatsu (Google)
  2025-01-03 23:01 ` [PATCH 2/6] Provide __free(argv) for argv_split() users Masami Hiramatsu (Google)
                   ` (4 subsequent siblings)
  5 siblings, 0 replies; 13+ messages in thread
From: Masami Hiramatsu (Google) @ 2025-01-03 23:01 UTC (permalink / raw)
  To: Steven Rostedt, Peter Zijlstra
  Cc: Anil S Keshavamurthy, Masami Hiramatsu, David S . Miller,
	Mathieu Desnoyers, Oleg Nesterov, Tzvetomir Stoyanov,
	Naveen N Rao, Josh Poimboeuf, Jason Baron, Ard Biesheuvel,
	linux-kernel, linux-trace-kernel

From: Masami Hiramatsu (Google) <mhiramat@kernel.org>

In __trace_kprobe_create(), if something fails it must goto error block
to free objects. But when strdup() a symbol, it returns without that.
Fix it to goto the error block to free objects correctly.

Fixes: 6212dd29683e ("tracing/kprobes: Use dyn_event framework for kprobe events")
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
---
 kernel/trace/trace_kprobe.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index bae26eb14449..4c3e316454a0 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -935,8 +935,10 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 		}
 		/* a symbol specified */
 		symbol = kstrdup(argv[1], GFP_KERNEL);
-		if (!symbol)
-			return -ENOMEM;
+		if (!symbol) {
+			ret = -ENOMEM;
+			goto error;
+		}
 
 		tmp = strchr(symbol, '%');
 		if (tmp) {


^ permalink raw reply related	[flat|nested] 13+ messages in thread

* [PATCH 2/6] Provide __free(argv) for argv_split() users
  2025-01-03 23:01 [PATCH 0/6] kprobes: jump label: Cleanup with guard and __free Masami Hiramatsu (Google)
  2025-01-03 23:01 ` [PATCH 1/6] tracing/kprobes: Fix to free objects when failed to copy a symbol Masami Hiramatsu (Google)
@ 2025-01-03 23:01 ` Masami Hiramatsu (Google)
  2025-01-04  6:39   ` kernel test robot
                     ` (3 more replies)
  2025-01-03 23:01 ` [PATCH 3/6] tracing: Use __free() for argv in dynevent Masami Hiramatsu (Google)
                   ` (3 subsequent siblings)
  5 siblings, 4 replies; 13+ messages in thread
From: Masami Hiramatsu (Google) @ 2025-01-03 23:01 UTC (permalink / raw)
  To: Steven Rostedt, Peter Zijlstra
  Cc: Anil S Keshavamurthy, Masami Hiramatsu, David S . Miller,
	Mathieu Desnoyers, Oleg Nesterov, Tzvetomir Stoyanov,
	Naveen N Rao, Josh Poimboeuf, Jason Baron, Ard Biesheuvel,
	linux-kernel, linux-trace-kernel

From: Masami Hiramatsu (Google) <mhiramat@kernel.org>

Provide __free(argv) macro for argv_split() users so that they can
avoid gotos.

Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
---
 include/linux/string.h |    2 ++
 1 file changed, 2 insertions(+)

diff --git a/include/linux/string.h b/include/linux/string.h
index 493ac4862c77..7035a70e30be 100644
--- a/include/linux/string.h
+++ b/include/linux/string.h
@@ -312,6 +312,8 @@ extern void *kmemdup_array(const void *src, size_t count, size_t element_size, g
 extern char **argv_split(gfp_t gfp, const char *str, int *argcp);
 extern void argv_free(char **argv);
 
+DEFINE_FREE(argv, char **, argv_free(_T))
+
 /* lib/cmdline.c */
 extern int get_option(char **str, int *pint);
 extern char *get_options(const char *str, int nints, int *ints);


^ permalink raw reply related	[flat|nested] 13+ messages in thread

* [PATCH 3/6] tracing: Use __free() for argv in dynevent
  2025-01-03 23:01 [PATCH 0/6] kprobes: jump label: Cleanup with guard and __free Masami Hiramatsu (Google)
  2025-01-03 23:01 ` [PATCH 1/6] tracing/kprobes: Fix to free objects when failed to copy a symbol Masami Hiramatsu (Google)
  2025-01-03 23:01 ` [PATCH 2/6] Provide __free(argv) for argv_split() users Masami Hiramatsu (Google)
@ 2025-01-03 23:01 ` Masami Hiramatsu (Google)
  2025-01-03 23:02 ` [PATCH 4/6] tracing: Use __free() in trace_probe for cleanup Masami Hiramatsu (Google)
                   ` (2 subsequent siblings)
  5 siblings, 0 replies; 13+ messages in thread
From: Masami Hiramatsu (Google) @ 2025-01-03 23:01 UTC (permalink / raw)
  To: Steven Rostedt, Peter Zijlstra
  Cc: Anil S Keshavamurthy, Masami Hiramatsu, David S . Miller,
	Mathieu Desnoyers, Oleg Nesterov, Tzvetomir Stoyanov,
	Naveen N Rao, Josh Poimboeuf, Jason Baron, Ard Biesheuvel,
	linux-kernel, linux-trace-kernel

From: Masami Hiramatsu (Google) <mhiramat@kernel.org>

Use __free() for the args allocated by argv_split() in dynevent.

Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
---
 kernel/trace/trace_dynevent.c |   47 +++++++++++++++++------------------------
 1 file changed, 19 insertions(+), 28 deletions(-)

diff --git a/kernel/trace/trace_dynevent.c b/kernel/trace/trace_dynevent.c
index eb8f669c15e1..ec8fa7b143e2 100644
--- a/kernel/trace/trace_dynevent.c
+++ b/kernel/trace/trace_dynevent.c
@@ -73,24 +73,20 @@ int dyn_event_release(const char *raw_command, struct dyn_event_operations *type
 	struct dyn_event *pos, *n;
 	char *system = NULL, *event, *p;
 	int argc, ret = -ENOENT;
-	char **argv;
+	char **argv __free(argv) = NULL;
 
 	argv = argv_split(GFP_KERNEL, raw_command, &argc);
 	if (!argv)
 		return -ENOMEM;
 
 	if (argv[0][0] == '-') {
-		if (argv[0][1] != ':') {
-			ret = -EINVAL;
-			goto out;
-		}
+		if (argv[0][1] != ':')
+			return -EINVAL;
 		event = &argv[0][2];
 	} else {
 		event = strchr(argv[0], ':');
-		if (!event) {
-			ret = -EINVAL;
-			goto out;
-		}
+		if (!event)
+			return -EINVAL;
 		event++;
 	}
 
@@ -100,27 +96,22 @@ int dyn_event_release(const char *raw_command, struct dyn_event_operations *type
 		event = p + 1;
 		*p = '\0';
 	}
-	if (!system && event[0] == '\0') {
-		ret = -EINVAL;
-		goto out;
-	}
+	if (!system && event[0] == '\0')
+		return -EINVAL;
 
-	scoped_guard(mutex, &event_mutex) {
-		for_each_dyn_event_safe(pos, n) {
-			if (type && type != pos->ops)
-				continue;
-			if (!pos->ops->match(system, event,
-					argc - 1, (const char **)argv + 1, pos))
-				continue;
-
-			ret = pos->ops->free(pos);
-			if (ret)
-				break;
-		}
-		tracing_reset_all_online_cpus();
+	guard(mutex)(&event_mutex);
+	for_each_dyn_event_safe(pos, n) {
+		if (type && type != pos->ops)
+			continue;
+		if (!pos->ops->match(system, event,
+				argc - 1, (const char **)argv + 1, pos))
+			continue;
+
+		ret = pos->ops->free(pos);
+		if (ret)
+			break;
 	}
-out:
-	argv_free(argv);
+	tracing_reset_all_online_cpus();
 	return ret;
 }
 


^ permalink raw reply related	[flat|nested] 13+ messages in thread

* [PATCH 4/6] tracing: Use __free() in trace_probe for cleanup
  2025-01-03 23:01 [PATCH 0/6] kprobes: jump label: Cleanup with guard and __free Masami Hiramatsu (Google)
                   ` (2 preceding siblings ...)
  2025-01-03 23:01 ` [PATCH 3/6] tracing: Use __free() for argv in dynevent Masami Hiramatsu (Google)
@ 2025-01-03 23:02 ` Masami Hiramatsu (Google)
  2025-01-03 23:02 ` [PATCH 5/6] tracing: Use __free() for kprobe events to cleanup Masami Hiramatsu (Google)
  2025-01-03 23:02 ` [PATCH 6/6] tracing/kprobes: Simplify __trace_kprobe_create() by removing gotos Masami Hiramatsu (Google)
  5 siblings, 0 replies; 13+ messages in thread
From: Masami Hiramatsu (Google) @ 2025-01-03 23:02 UTC (permalink / raw)
  To: Steven Rostedt, Peter Zijlstra
  Cc: Anil S Keshavamurthy, Masami Hiramatsu, David S . Miller,
	Mathieu Desnoyers, Oleg Nesterov, Tzvetomir Stoyanov,
	Naveen N Rao, Josh Poimboeuf, Jason Baron, Ard Biesheuvel,
	linux-kernel, linux-trace-kernel

From: Masami Hiramatsu (Google) <mhiramat@kernel.org>

Use __free() in trace_probe to cleanup some gotos.

Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
---
 kernel/trace/trace_probe.c |   52 +++++++++++++++-----------------------------
 1 file changed, 18 insertions(+), 34 deletions(-)

diff --git a/kernel/trace/trace_probe.c b/kernel/trace/trace_probe.c
index 16a5e368e7b7..bf6a7b81ae95 100644
--- a/kernel/trace/trace_probe.c
+++ b/kernel/trace/trace_probe.c
@@ -1409,7 +1409,7 @@ static int traceprobe_parse_probe_arg_body(const char *argv, ssize_t *size,
 					   struct traceprobe_parse_context *ctx)
 {
 	struct fetch_insn *code, *tmp = NULL;
-	char *type, *arg;
+	char *type, *arg __free(kfree) = NULL;
 	int ret, len;
 
 	len = strlen(argv);
@@ -1426,22 +1426,16 @@ static int traceprobe_parse_probe_arg_body(const char *argv, ssize_t *size,
 		return -ENOMEM;
 
 	parg->comm = kstrdup(arg, GFP_KERNEL);
-	if (!parg->comm) {
-		ret = -ENOMEM;
-		goto out;
-	}
+	if (!parg->comm)
+		return -ENOMEM;
 
 	type = parse_probe_arg_type(arg, parg, ctx);
-	if (IS_ERR(type)) {
-		ret = PTR_ERR(type);
-		goto out;
-	}
+	if (IS_ERR(type))
+		return PTR_ERR(type);
 
 	code = tmp = kcalloc(FETCH_INSN_MAX, sizeof(*code), GFP_KERNEL);
-	if (!code) {
-		ret = -ENOMEM;
-		goto out;
-	}
+	if (!code)
+		return -ENOMEM;
 	code[FETCH_INSN_MAX - 1].op = FETCH_OP_END;
 
 	ctx->last_type = NULL;
@@ -1497,8 +1491,6 @@ static int traceprobe_parse_probe_arg_body(const char *argv, ssize_t *size,
 				kfree(code->data);
 	}
 	kfree(tmp);
-out:
-	kfree(arg);
 
 	return ret;
 }
@@ -1668,7 +1660,7 @@ const char **traceprobe_expand_meta_args(int argc, const char *argv[],
 {
 	const struct btf_param *params = NULL;
 	int i, j, n, used, ret, args_idx = -1;
-	const char **new_argv = NULL;
+	const char **new_argv __free(kfree) = NULL;
 
 	ret = argv_has_var_arg(argc, argv, &args_idx, ctx);
 	if (ret < 0)
@@ -1707,7 +1699,7 @@ const char **traceprobe_expand_meta_args(int argc, const char *argv[],
 				ret = sprint_nth_btf_arg(n, "", buf + used,
 							 bufsize - used, ctx);
 				if (ret < 0)
-					goto error;
+					return ERR_PTR(ret);
 
 				new_argv[j++] = buf + used;
 				used += ret + 1;
@@ -1721,25 +1713,20 @@ const char **traceprobe_expand_meta_args(int argc, const char *argv[],
 			n = simple_strtoul(argv[i] + 4, &type, 10);
 			if (type && !(*type == ':' || *type == '\0')) {
 				trace_probe_log_err(0, BAD_VAR);
-				ret = -ENOENT;
-				goto error;
+				return ERR_PTR(-ENOENT);
 			}
 			/* Note: $argN starts from $arg1 */
 			ret = sprint_nth_btf_arg(n - 1, type, buf + used,
 						 bufsize - used, ctx);
 			if (ret < 0)
-				goto error;
+				return ERR_PTR(ret);
 			new_argv[j++] = buf + used;
 			used += ret + 1;
 		} else
 			new_argv[j++] = argv[i];
 	}
 
-	return new_argv;
-
-error:
-	kfree(new_argv);
-	return ERR_PTR(ret);
+	return_ptr(new_argv);
 }
 
 /* @buf: *buf must be equal to NULL. Caller must to free *buf */
@@ -1747,14 +1734,14 @@ int traceprobe_expand_dentry_args(int argc, const char *argv[], char **buf)
 {
 	int i, used, ret;
 	const int bufsize = MAX_DENTRY_ARGS_LEN;
-	char *tmpbuf = NULL;
+	char *tmpbuf __free(kfree) = NULL;
 
 	if (*buf)
 		return -EINVAL;
 
 	used = 0;
 	for (i = 0; i < argc; i++) {
-		char *tmp;
+		char *tmp __free(kfree) = NULL;
 		char *equal;
 		size_t arg_len;
 
@@ -1769,7 +1756,7 @@ int traceprobe_expand_dentry_args(int argc, const char *argv[], char **buf)
 
 		tmp = kstrdup(argv[i], GFP_KERNEL);
 		if (!tmp)
-			goto nomem;
+			return -ENOMEM;
 
 		equal = strchr(tmp, '=');
 		if (equal)
@@ -1790,18 +1777,15 @@ int traceprobe_expand_dentry_args(int argc, const char *argv[], char **buf)
 				       offsetof(struct file, f_path.dentry),
 				       equal ? equal + 1 : tmp);
 
-		kfree(tmp);
+		kfree(no_free_ptr(tmp));
 		if (ret >= bufsize - used)
-			goto nomem;
+			return -ENOMEM;
 		argv[i] = tmpbuf + used;
 		used += ret + 1;
 	}
 
-	*buf = tmpbuf;
+	*buf = no_free_ptr(tmpbuf);
 	return 0;
-nomem:
-	kfree(tmpbuf);
-	return -ENOMEM;
 }
 
 void traceprobe_finish_parse(struct traceprobe_parse_context *ctx)


^ permalink raw reply related	[flat|nested] 13+ messages in thread

* [PATCH 5/6] tracing: Use __free() for kprobe events to cleanup
  2025-01-03 23:01 [PATCH 0/6] kprobes: jump label: Cleanup with guard and __free Masami Hiramatsu (Google)
                   ` (3 preceding siblings ...)
  2025-01-03 23:02 ` [PATCH 4/6] tracing: Use __free() in trace_probe for cleanup Masami Hiramatsu (Google)
@ 2025-01-03 23:02 ` Masami Hiramatsu (Google)
  2025-01-03 23:02 ` [PATCH 6/6] tracing/kprobes: Simplify __trace_kprobe_create() by removing gotos Masami Hiramatsu (Google)
  5 siblings, 0 replies; 13+ messages in thread
From: Masami Hiramatsu (Google) @ 2025-01-03 23:02 UTC (permalink / raw)
  To: Steven Rostedt, Peter Zijlstra
  Cc: Anil S Keshavamurthy, Masami Hiramatsu, David S . Miller,
	Mathieu Desnoyers, Oleg Nesterov, Tzvetomir Stoyanov,
	Naveen N Rao, Josh Poimboeuf, Jason Baron, Ard Biesheuvel,
	linux-kernel, linux-trace-kernel

From: Masami Hiramatsu (Google) <mhiramat@kernel.org>

Use __free() in trace_kprobe.c to cleanup code.

Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
---
 kernel/trace/trace_kprobe.c |   53 ++++++++++++++++++-------------------------
 1 file changed, 22 insertions(+), 31 deletions(-)

diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index 4c3e316454a0..c5fd4471c137 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -8,6 +8,7 @@
 #define pr_fmt(fmt)	"trace_kprobe: " fmt
 
 #include <linux/bpf-cgroup.h>
+#include <linux/cleanup.h>
 #include <linux/security.h>
 #include <linux/module.h>
 #include <linux/uaccess.h>
@@ -257,6 +258,8 @@ static void free_trace_kprobe(struct trace_kprobe *tk)
 	}
 }
 
+DEFINE_FREE(trace_kprobe, struct trace_kprobe *, free_trace_kprobe(_T))
+
 /*
  * Allocate new trace_probe and initialize it (including kprobes).
  */
@@ -268,7 +271,7 @@ static struct trace_kprobe *alloc_trace_kprobe(const char *group,
 					     int maxactive,
 					     int nargs, bool is_return)
 {
-	struct trace_kprobe *tk;
+	struct trace_kprobe *tk __free(trace_kprobe) = NULL;
 	int ret = -ENOMEM;
 
 	tk = kzalloc(struct_size(tk, tp.args, nargs), GFP_KERNEL);
@@ -277,12 +280,12 @@ static struct trace_kprobe *alloc_trace_kprobe(const char *group,
 
 	tk->nhit = alloc_percpu(unsigned long);
 	if (!tk->nhit)
-		goto error;
+		return ERR_PTR(ret);
 
 	if (symbol) {
 		tk->symbol = kstrdup(symbol, GFP_KERNEL);
 		if (!tk->symbol)
-			goto error;
+			return ERR_PTR(ret);
 		tk->rp.kp.symbol_name = tk->symbol;
 		tk->rp.kp.offset = offs;
 	} else
@@ -299,13 +302,10 @@ static struct trace_kprobe *alloc_trace_kprobe(const char *group,
 
 	ret = trace_probe_init(&tk->tp, event, group, false, nargs);
 	if (ret < 0)
-		goto error;
+		return ERR_PTR(ret);
 
 	dyn_event_init(&tk->devent, &trace_kprobe_ops);
-	return tk;
-error:
-	free_trace_kprobe(tk);
-	return ERR_PTR(ret);
+	return_ptr(tk);
 }
 
 static struct trace_kprobe *find_trace_kprobe(const char *event,
@@ -861,11 +861,12 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 	 * Type of args:
 	 *  FETCHARG:TYPE : use TYPE instead of unsigned long.
 	 */
-	struct trace_kprobe *tk = NULL;
+	struct trace_kprobe *tk __free(trace_kprobe) = NULL;
 	int i, len, new_argc = 0, ret = 0;
 	bool is_return = false;
-	char *symbol = NULL, *tmp = NULL;
-	const char **new_argv = NULL;
+	char *symbol __free(kfree) = NULL;
+	char *tmp = NULL;
+	const char **new_argv __free(kfree) = NULL;
 	const char *event = NULL, *group = KPROBE_EVENT_SYSTEM;
 	enum probe_print_type ptype;
 	int maxactive = 0;
@@ -874,7 +875,7 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 	char buf[MAX_EVENT_NAME_LEN];
 	char gbuf[MAX_EVENT_NAME_LEN];
 	char abuf[MAX_BTF_ARGS_LEN];
-	char *dbuf = NULL;
+	char *dbuf __free(kfree) = NULL;
 	struct traceprobe_parse_context ctx = { .flags = TPARG_FL_KERNEL };
 
 	switch (argv[0][0]) {
@@ -931,7 +932,7 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 		/* Check whether uprobe event specified */
 		if (strchr(argv[1], '/') && strchr(argv[1], ':')) {
 			ret = -ECANCELED;
-			goto error;
+			goto out;
 		}
 		/* a symbol specified */
 		symbol = kstrdup(argv[1], GFP_KERNEL);
@@ -1035,7 +1036,7 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 		ctx.offset = 0;
 		ret = traceprobe_parse_probe_arg(&tk->tp, i, argv[i], &ctx);
 		if (ret)
-			goto error;	/* This can be -ENOMEM */
+			goto out;	/* This can be -ENOMEM */
 	}
 	/* entry handler for kretprobe */
 	if (is_return && tk->tp.entry_arg) {
@@ -1046,7 +1047,7 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 	ptype = is_return ? PROBE_PRINT_RETURN : PROBE_PRINT_NORMAL;
 	ret = traceprobe_set_print_fmt(&tk->tp, ptype);
 	if (ret < 0)
-		goto error;
+		goto out;
 
 	ret = register_trace_kprobe(tk);
 	if (ret) {
@@ -1057,21 +1058,16 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 			trace_probe_log_err(0, BAD_PROBE_ADDR);
 		else if (ret != -ENOMEM && ret != -EEXIST)
 			trace_probe_log_err(0, FAIL_REG_PROBE);
-		goto error;
-	}
+	} else
+		no_free_ptr(tk);
 
 out:
 	traceprobe_finish_parse(&ctx);
 	trace_probe_log_clear();
-	kfree(new_argv);
-	kfree(symbol);
-	kfree(dbuf);
 	return ret;
 
 parse_error:
 	ret = -EINVAL;
-error:
-	free_trace_kprobe(tk);
 	goto out;
 }
 
@@ -1893,7 +1889,7 @@ create_local_trace_kprobe(char *func, void *addr, unsigned long offs,
 			  bool is_return)
 {
 	enum probe_print_type ptype;
-	struct trace_kprobe *tk;
+	struct trace_kprobe *tk __free(trace_kprobe) = NULL;
 	int ret;
 	char *event;
 
@@ -1924,19 +1920,14 @@ create_local_trace_kprobe(char *func, void *addr, unsigned long offs,
 
 	ptype = trace_kprobe_is_return(tk) ?
 		PROBE_PRINT_RETURN : PROBE_PRINT_NORMAL;
-	if (traceprobe_set_print_fmt(&tk->tp, ptype) < 0) {
-		ret = -ENOMEM;
-		goto error;
-	}
+	if (traceprobe_set_print_fmt(&tk->tp, ptype) < 0)
+		return ERR_PTR(-ENOMEM);
 
 	ret = __register_trace_kprobe(tk);
 	if (ret < 0)
-		goto error;
+		return ERR_PTR(ret);
 
 	return trace_probe_event_call(&tk->tp);
-error:
-	free_trace_kprobe(tk);
-	return ERR_PTR(ret);
 }
 
 void destroy_local_trace_kprobe(struct trace_event_call *event_call)


^ permalink raw reply related	[flat|nested] 13+ messages in thread

* [PATCH 6/6] tracing/kprobes: Simplify __trace_kprobe_create() by removing gotos
  2025-01-03 23:01 [PATCH 0/6] kprobes: jump label: Cleanup with guard and __free Masami Hiramatsu (Google)
                   ` (4 preceding siblings ...)
  2025-01-03 23:02 ` [PATCH 5/6] tracing: Use __free() for kprobe events to cleanup Masami Hiramatsu (Google)
@ 2025-01-03 23:02 ` Masami Hiramatsu (Google)
  5 siblings, 0 replies; 13+ messages in thread
From: Masami Hiramatsu (Google) @ 2025-01-03 23:02 UTC (permalink / raw)
  To: Steven Rostedt, Peter Zijlstra
  Cc: Anil S Keshavamurthy, Masami Hiramatsu, David S . Miller,
	Mathieu Desnoyers, Oleg Nesterov, Tzvetomir Stoyanov,
	Naveen N Rao, Josh Poimboeuf, Jason Baron, Ard Biesheuvel,
	linux-kernel, linux-trace-kernel

From: Masami Hiramatsu (Google) <mhiramat@kernel.org>

Simplify __trace_kprobe_create() by removing gotos.

Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
---
 kernel/trace/trace_kprobe.c |   82 ++++++++++++++++++++++---------------------
 1 file changed, 41 insertions(+), 41 deletions(-)

diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index c5fd4471c137..604eea2ed2fa 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -835,7 +835,8 @@ static int validate_probe_symbol(char *symbol)
 static int trace_kprobe_entry_handler(struct kretprobe_instance *ri,
 				      struct pt_regs *regs);
 
-static int __trace_kprobe_create(int argc, const char *argv[])
+static int ___trace_kprobe_create(int argc, const char *argv[],
+				  struct traceprobe_parse_context *ctx)
 {
 	/*
 	 * Argument syntax:
@@ -876,7 +877,6 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 	char gbuf[MAX_EVENT_NAME_LEN];
 	char abuf[MAX_BTF_ARGS_LEN];
 	char *dbuf __free(kfree) = NULL;
-	struct traceprobe_parse_context ctx = { .flags = TPARG_FL_KERNEL };
 
 	switch (argv[0][0]) {
 	case 'r':
@@ -890,8 +890,6 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 	if (argc < 2)
 		return -ECANCELED;
 
-	trace_probe_log_init("trace_kprobe", argc, argv);
-
 	event = strchr(&argv[0][1], ':');
 	if (event)
 		event++;
@@ -899,7 +897,7 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 	if (isdigit(argv[0][1])) {
 		if (!is_return) {
 			trace_probe_log_err(1, BAD_MAXACT_TYPE);
-			goto parse_error;
+			return -EINVAL;
 		}
 		if (event)
 			len = event - &argv[0][1] - 1;
@@ -907,21 +905,21 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 			len = strlen(&argv[0][1]);
 		if (len > MAX_EVENT_NAME_LEN - 1) {
 			trace_probe_log_err(1, BAD_MAXACT);
-			goto parse_error;
+			return -EINVAL;
 		}
 		memcpy(buf, &argv[0][1], len);
 		buf[len] = '\0';
 		ret = kstrtouint(buf, 0, &maxactive);
 		if (ret || !maxactive) {
 			trace_probe_log_err(1, BAD_MAXACT);
-			goto parse_error;
+			return -EINVAL;
 		}
 		/* 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);
-			goto parse_error;
+			return -EINVAL;
 		}
 	}
 
@@ -930,16 +928,13 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 	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], ':')) {
-			ret = -ECANCELED;
-			goto out;
-		}
+		if (strchr(argv[1], '/') && strchr(argv[1], ':'))
+			return -ECANCELED;
+
 		/* a symbol specified */
 		symbol = kstrdup(argv[1], GFP_KERNEL);
-		if (!symbol) {
-			ret = -ENOMEM;
-			goto error;
-		}
+		if (!symbol)
+			return -ENOMEM;
 
 		tmp = strchr(symbol, '%');
 		if (tmp) {
@@ -948,7 +943,7 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 				is_return = true;
 			} else {
 				trace_probe_log_err(tmp - symbol, BAD_ADDR_SUFFIX);
-				goto parse_error;
+				return -EINVAL;
 			}
 		}
 
@@ -956,7 +951,7 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 		ret = traceprobe_split_symbol_offset(symbol, &offset);
 		if (ret || offset < 0 || offset > UINT_MAX) {
 			trace_probe_log_err(0, BAD_PROBE_ADDR);
-			goto parse_error;
+			return -EINVAL;
 		}
 		ret = validate_probe_symbol(symbol);
 		if (ret) {
@@ -964,17 +959,17 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 				trace_probe_log_err(0, NON_UNIQ_SYMBOL);
 			else
 				trace_probe_log_err(0, BAD_PROBE_ADDR);
-			goto parse_error;
+			return -EINVAL;
 		}
 		if (is_return)
-			ctx.flags |= TPARG_FL_RETURN;
+			ctx->flags |= TPARG_FL_RETURN;
 		ret = kprobe_on_func_entry(NULL, symbol, offset);
 		if (ret == 0 && !is_return)
-			ctx.flags |= TPARG_FL_FENTRY;
+			ctx->flags |= TPARG_FL_FENTRY;
 		/* Defer the ENOENT case until register kprobe */
 		if (ret == -EINVAL && is_return) {
 			trace_probe_log_err(0, BAD_RETPROBE);
-			goto parse_error;
+			return -EINVAL;
 		}
 	}
 
@@ -983,7 +978,7 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 		ret = traceprobe_parse_event_name(&event, &group, gbuf,
 						  event - argv[0]);
 		if (ret)
-			goto parse_error;
+			return ret;
 	}
 
 	if (!event) {
@@ -999,26 +994,24 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 	}
 
 	argc -= 2; argv += 2;
-	ctx.funcname = symbol;
+	ctx->funcname = symbol;
 	new_argv = traceprobe_expand_meta_args(argc, argv, &new_argc,
-					       abuf, MAX_BTF_ARGS_LEN, &ctx);
+					       abuf, MAX_BTF_ARGS_LEN, ctx);
 	if (IS_ERR(new_argv)) {
 		ret = PTR_ERR(new_argv);
 		new_argv = NULL;
-		goto out;
+		return ret;
 	}
 	if (new_argv) {
 		argc = new_argc;
 		argv = new_argv;
 	}
-	if (argc > MAX_TRACE_ARGS) {
-		ret = -E2BIG;
-		goto out;
-	}
+	if (argc > MAX_TRACE_ARGS)
+		return -E2BIG;
 
 	ret = traceprobe_expand_dentry_args(argc, argv, &dbuf);
 	if (ret)
-		goto out;
+		return ret;
 
 	/* setup a probe */
 	tk = alloc_trace_kprobe(group, event, addr, symbol, offset, maxactive,
@@ -1027,16 +1020,16 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 		ret = PTR_ERR(tk);
 		/* This must return -ENOMEM, else there is a bug */
 		WARN_ON_ONCE(ret != -ENOMEM);
-		goto out;	/* We know tk is not allocated */
+		return ret;	/* We know tk is not allocated */
 	}
 
 	/* 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);
+		ctx->offset = 0;
+		ret = traceprobe_parse_probe_arg(&tk->tp, i, argv[i], ctx);
 		if (ret)
-			goto out;	/* This can be -ENOMEM */
+			return ret;	/* This can be -ENOMEM */
 	}
 	/* entry handler for kretprobe */
 	if (is_return && tk->tp.entry_arg) {
@@ -1047,7 +1040,7 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 	ptype = is_return ? PROBE_PRINT_RETURN : PROBE_PRINT_NORMAL;
 	ret = traceprobe_set_print_fmt(&tk->tp, ptype);
 	if (ret < 0)
-		goto out;
+		return ret;
 
 	ret = register_trace_kprobe(tk);
 	if (ret) {
@@ -1061,14 +1054,21 @@ static int __trace_kprobe_create(int argc, const char *argv[])
 	} else
 		no_free_ptr(tk);
 
-out:
+	return ret;
+}
+
+static int __trace_kprobe_create(int argc, const char *argv[])
+{
+	struct traceprobe_parse_context ctx = { .flags = TPARG_FL_KERNEL };
+	int ret;
+
+	trace_probe_log_init("trace_kprobe", argc, argv);
+
+	ret = ___trace_kprobe_create(argc, argv, &ctx);
+
 	traceprobe_finish_parse(&ctx);
 	trace_probe_log_clear();
 	return ret;
-
-parse_error:
-	ret = -EINVAL;
-	goto out;
 }
 
 static int trace_kprobe_create(const char *raw_command)


^ permalink raw reply related	[flat|nested] 13+ messages in thread

* Re: [PATCH 2/6] Provide __free(argv) for argv_split() users
  2025-01-03 23:01 ` [PATCH 2/6] Provide __free(argv) for argv_split() users Masami Hiramatsu (Google)
@ 2025-01-04  6:39   ` kernel test robot
  2025-01-04  6:51   ` kernel test robot
                     ` (2 subsequent siblings)
  3 siblings, 0 replies; 13+ messages in thread
From: kernel test robot @ 2025-01-04  6:39 UTC (permalink / raw)
  To: Masami Hiramatsu (Google), Steven Rostedt, Peter Zijlstra
  Cc: oe-kbuild-all, Anil S Keshavamurthy, Masami Hiramatsu,
	David S . Miller, Mathieu Desnoyers, Oleg Nesterov,
	Tzvetomir Stoyanov, Naveen N Rao, Josh Poimboeuf, Jason Baron,
	Ard Biesheuvel, linux-kernel, linux-trace-kernel

Hi Masami,

kernel test robot noticed the following build errors:

[auto build test ERROR on next-20241220]
[also build test ERROR on v6.13-rc5]
[cannot apply to kees/for-next/hardening linus/master rostedt-trace/for-next rostedt-trace/for-next-urgent v6.13-rc5 v6.13-rc4 v6.13-rc3]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Masami-Hiramatsu-Google/tracing-kprobes-Fix-to-free-objects-when-failed-to-copy-a-symbol/20250104-070535
base:   next-20241220
patch link:    https://lore.kernel.org/r/173594530753.1055889.17844868397124331132.stgit%40devnote2
patch subject: [PATCH 2/6] Provide __free(argv) for argv_split() users
config: i386-buildonly-randconfig-002-20250104 (https://download.01.org/0day-ci/archive/20250104/202501041433.GkehPwga-lkp@intel.com/config)
compiler: gcc-12 (Debian 12.2.0-14) 12.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20250104/202501041433.GkehPwga-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202501041433.GkehPwga-lkp@intel.com/

All errors (new ones prefixed by >>):

   In file included from include/linux/uuid.h:11,
                    from include/linux/mod_devicetable.h:14,
                    from scripts/mod/devicetable-offsets.c:3:
>> include/linux/string.h:315:18: error: expected ')' before 'char'
     315 | DEFINE_FREE(argv, char **, argv_free(_T))
         |                  ^~~~~
         |                  )
   make[3]: *** [scripts/Makefile.build:102: scripts/mod/devicetable-offsets.s] Error 1 shuffle=943069964
   make[3]: Target 'scripts/mod/' not remade because of errors.
   make[2]: *** [Makefile:1262: prepare0] Error 2 shuffle=943069964
   make[2]: Target 'prepare' not remade because of errors.
   make[1]: *** [Makefile:251: __sub-make] Error 2 shuffle=943069964
   make[1]: Target 'prepare' not remade because of errors.
   make: *** [Makefile:251: __sub-make] Error 2 shuffle=943069964
   make: Target 'prepare' not remade because of errors.


vim +315 include/linux/string.h

   314	
 > 315	DEFINE_FREE(argv, char **, argv_free(_T))
   316	

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki

^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [PATCH 2/6] Provide __free(argv) for argv_split() users
  2025-01-03 23:01 ` [PATCH 2/6] Provide __free(argv) for argv_split() users Masami Hiramatsu (Google)
  2025-01-04  6:39   ` kernel test robot
@ 2025-01-04  6:51   ` kernel test robot
  2025-01-05  9:12     ` Masami Hiramatsu
  2025-01-04 11:09   ` Oleg Nesterov
  2025-01-06 19:30   ` Steven Rostedt
  3 siblings, 1 reply; 13+ messages in thread
From: kernel test robot @ 2025-01-04  6:51 UTC (permalink / raw)
  To: Masami Hiramatsu (Google), Steven Rostedt, Peter Zijlstra
  Cc: llvm, oe-kbuild-all, Anil S Keshavamurthy, Masami Hiramatsu,
	David S . Miller, Mathieu Desnoyers, Oleg Nesterov,
	Tzvetomir Stoyanov, Naveen N Rao, Josh Poimboeuf, Jason Baron,
	Ard Biesheuvel, linux-kernel, linux-trace-kernel

Hi Masami,

kernel test robot noticed the following build errors:

[auto build test ERROR on next-20241220]
[also build test ERROR on v6.13-rc5]
[cannot apply to kees/for-next/hardening linus/master rostedt-trace/for-next rostedt-trace/for-next-urgent v6.13-rc5 v6.13-rc4 v6.13-rc3]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Masami-Hiramatsu-Google/tracing-kprobes-Fix-to-free-objects-when-failed-to-copy-a-symbol/20250104-070535
base:   next-20241220
patch link:    https://lore.kernel.org/r/173594530753.1055889.17844868397124331132.stgit%40devnote2
patch subject: [PATCH 2/6] Provide __free(argv) for argv_split() users
config: i386-buildonly-randconfig-001-20250104 (https://download.01.org/0day-ci/archive/20250104/202501041451.yG7LhbEv-lkp@intel.com/config)
compiler: clang version 19.1.3 (https://github.com/llvm/llvm-project ab51eccf88f5321e7c60591c5546b254b6afab99)
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20250104/202501041451.yG7LhbEv-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202501041451.yG7LhbEv-lkp@intel.com/

All errors (new ones prefixed by >>):

   In file included from scripts/mod/devicetable-offsets.c:3:
   In file included from include/linux/mod_devicetable.h:14:
   In file included from include/linux/uuid.h:11:
>> include/linux/string.h:315:19: error: expected identifier
     315 | DEFINE_FREE(argv, char **, argv_free(_T))
         |                   ^
>> include/linux/string.h:315:1: error: type specifier missing, defaults to 'int'; ISO C99 and later do not support implicit int [-Wimplicit-int]
     315 | DEFINE_FREE(argv, char **, argv_free(_T))
         | ^
         | int
>> include/linux/string.h:315:12: error: a function declaration without a prototype is deprecated in all versions of C [-Werror,-Wstrict-prototypes]
     315 | DEFINE_FREE(argv, char **, argv_free(_T))
         |            ^                            
         |                                         void
>> include/linux/string.h:315:42: error: expected ';' after top level declarator
     315 | DEFINE_FREE(argv, char **, argv_free(_T))
         |                                          ^
         |                                          ;
   4 errors generated.
   make[3]: *** [scripts/Makefile.build:102: scripts/mod/devicetable-offsets.s] Error 1 shuffle=808271366
   make[3]: Target 'scripts/mod/' not remade because of errors.
   make[2]: *** [Makefile:1262: prepare0] Error 2 shuffle=808271366
   make[2]: Target 'prepare' not remade because of errors.
   make[1]: *** [Makefile:251: __sub-make] Error 2 shuffle=808271366
   make[1]: Target 'prepare' not remade because of errors.
   make: *** [Makefile:251: __sub-make] Error 2 shuffle=808271366
   make: Target 'prepare' not remade because of errors.


vim +315 include/linux/string.h

   314	
 > 315	DEFINE_FREE(argv, char **, argv_free(_T))
   316	

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki

^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [PATCH 2/6] Provide __free(argv) for argv_split() users
  2025-01-03 23:01 ` [PATCH 2/6] Provide __free(argv) for argv_split() users Masami Hiramatsu (Google)
  2025-01-04  6:39   ` kernel test robot
  2025-01-04  6:51   ` kernel test robot
@ 2025-01-04 11:09   ` Oleg Nesterov
  2025-01-05  9:00     ` Masami Hiramatsu
  2025-01-06 19:30   ` Steven Rostedt
  3 siblings, 1 reply; 13+ messages in thread
From: Oleg Nesterov @ 2025-01-04 11:09 UTC (permalink / raw)
  To: Masami Hiramatsu (Google)
  Cc: Steven Rostedt, Peter Zijlstra, Anil S Keshavamurthy,
	David S . Miller, Mathieu Desnoyers, Tzvetomir Stoyanov,
	Naveen N Rao, Josh Poimboeuf, Jason Baron, Ard Biesheuvel,
	linux-kernel, linux-trace-kernel

On 01/04, Masami Hiramatsu (Google) wrote:
>
> +DEFINE_FREE(argv, char **, argv_free(_T))

This doesn't look right, I think we need

	DEFINE_FREE(argv, char **, if (_T) argv_free(_T))

?

From the next patch

	@@ -73,24 +73,20 @@ int dyn_event_release(const char *raw_command, struct dyn_event_operations *type
		struct dyn_event *pos, *n;
		char *system = NULL, *event, *p;
		int argc, ret = -ENOENT;
	-	char **argv;
	+	char **argv __free(argv) = NULL;
	 
		argv = argv_split(GFP_KERNEL, raw_command, &argc);
		if (!argv)
			return -ENOMEM;

if argv_split() returns NULL, then __free_argv() will call argv_free(NULL) ?

Oleg.


^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [PATCH 2/6] Provide __free(argv) for argv_split() users
  2025-01-04 11:09   ` Oleg Nesterov
@ 2025-01-05  9:00     ` Masami Hiramatsu
  0 siblings, 0 replies; 13+ messages in thread
From: Masami Hiramatsu @ 2025-01-05  9:00 UTC (permalink / raw)
  To: Oleg Nesterov
  Cc: Steven Rostedt, Peter Zijlstra, Anil S Keshavamurthy,
	David S . Miller, Mathieu Desnoyers, Tzvetomir Stoyanov,
	Naveen N Rao, Josh Poimboeuf, Jason Baron, Ard Biesheuvel,
	linux-kernel, linux-trace-kernel

On Sat, 4 Jan 2025 12:09:56 +0100
Oleg Nesterov <oleg@redhat.com> wrote:

> On 01/04, Masami Hiramatsu (Google) wrote:
> >
> > +DEFINE_FREE(argv, char **, argv_free(_T))
> 
> This doesn't look right, I think we need
> 
> 	DEFINE_FREE(argv, char **, if (_T) argv_free(_T))

Ah, yes. I misunderstood argv_free() can get NULL.

Thanks!

> 
> ?
> 
> From the next patch
> 
> 	@@ -73,24 +73,20 @@ int dyn_event_release(const char *raw_command, struct dyn_event_operations *type
> 		struct dyn_event *pos, *n;
> 		char *system = NULL, *event, *p;
> 		int argc, ret = -ENOENT;
> 	-	char **argv;
> 	+	char **argv __free(argv) = NULL;
> 	 
> 		argv = argv_split(GFP_KERNEL, raw_command, &argc);
> 		if (!argv)
> 			return -ENOMEM;
> 
> if argv_split() returns NULL, then __free_argv() will call argv_free(NULL) ?
> 
> Oleg.
> 
> 


-- 
Masami Hiramatsu (Google) <mhiramat@kernel.org>

^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [PATCH 2/6] Provide __free(argv) for argv_split() users
  2025-01-04  6:51   ` kernel test robot
@ 2025-01-05  9:12     ` Masami Hiramatsu
  0 siblings, 0 replies; 13+ messages in thread
From: Masami Hiramatsu @ 2025-01-05  9:12 UTC (permalink / raw)
  To: kernel test robot
  Cc: Steven Rostedt, Peter Zijlstra, llvm, oe-kbuild-all,
	Anil S Keshavamurthy, David S . Miller, Mathieu Desnoyers,
	Oleg Nesterov, Tzvetomir Stoyanov, Naveen N Rao, Josh Poimboeuf,
	Jason Baron, Ard Biesheuvel, linux-kernel, linux-trace-kernel

On Sat, 4 Jan 2025 14:51:12 +0800
kernel test robot <lkp@intel.com> wrote:

> Hi Masami,
> 
> kernel test robot noticed the following build errors:
> 
> [auto build test ERROR on next-20241220]
> [also build test ERROR on v6.13-rc5]
> [cannot apply to kees/for-next/hardening linus/master rostedt-trace/for-next rostedt-trace/for-next-urgent v6.13-rc5 v6.13-rc4 v6.13-rc3]
> [If your patch is applied to the wrong git tree, kindly drop us a note.
> And when submitting patch, we suggest to use '--base' as documented in
> https://git-scm.com/docs/git-format-patch#_base_tree_information]
> 
> url:    https://github.com/intel-lab-lkp/linux/commits/Masami-Hiramatsu-Google/tracing-kprobes-Fix-to-free-objects-when-failed-to-copy-a-symbol/20250104-070535
> base:   next-20241220
> patch link:    https://lore.kernel.org/r/173594530753.1055889.17844868397124331132.stgit%40devnote2
> patch subject: [PATCH 2/6] Provide __free(argv) for argv_split() users
> config: i386-buildonly-randconfig-001-20250104 (https://download.01.org/0day-ci/archive/20250104/202501041451.yG7LhbEv-lkp@intel.com/config)
> compiler: clang version 19.1.3 (https://github.com/llvm/llvm-project ab51eccf88f5321e7c60591c5546b254b6afab99)
> reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20250104/202501041451.yG7LhbEv-lkp@intel.com/reproduce)
> 
> If you fix the issue in a separate patch/commit (i.e. not just a new version of
> the same patch/commit), kindly add following tags
> | Reported-by: kernel test robot <lkp@intel.com>
> | Closes: https://lore.kernel.org/oe-kbuild-all/202501041451.yG7LhbEv-lkp@intel.com/
> 
> All errors (new ones prefixed by >>):
> 
>    In file included from scripts/mod/devicetable-offsets.c:3:
>    In file included from include/linux/mod_devicetable.h:14:
>    In file included from include/linux/uuid.h:11:
> >> include/linux/string.h:315:19: error: expected identifier
>      315 | DEFINE_FREE(argv, char **, argv_free(_T))
>          |                   ^
> >> include/linux/string.h:315:1: error: type specifier missing, defaults to 'int'; ISO C99 and later do not support implicit int [-Wimplicit-int]
>      315 | DEFINE_FREE(argv, char **, argv_free(_T))
>          | ^
>          | int
> >> include/linux/string.h:315:12: error: a function declaration without a prototype is deprecated in all versions of C [-Werror,-Wstrict-prototypes]
>      315 | DEFINE_FREE(argv, char **, argv_free(_T))
>          |            ^                            
>          |                                         void
> >> include/linux/string.h:315:42: error: expected ';' after top level declarator
>      315 | DEFINE_FREE(argv, char **, argv_free(_T))
>          |                                          ^
>          |                                          ;
>    4 errors generated.
>    make[3]: *** [scripts/Makefile.build:102: scripts/mod/devicetable-offsets.s] Error 1 shuffle=808271366
>    make[3]: Target 'scripts/mod/' not remade because of errors.
>    make[2]: *** [Makefile:1262: prepare0] Error 2 shuffle=808271366
>    make[2]: Target 'prepare' not remade because of errors.
>    make[1]: *** [Makefile:251: __sub-make] Error 2 shuffle=808271366
>    make[1]: Target 'prepare' not remade because of errors.
>    make: *** [Makefile:251: __sub-make] Error 2 shuffle=808271366
>    make: Target 'prepare' not remade because of errors.

Oops, I forgot to include cleanup.h!

Thanks, let me update it.

> 
> 
> vim +315 include/linux/string.h
> 
>    314	
>  > 315	DEFINE_FREE(argv, char **, argv_free(_T))
>    316	
> 
> -- 
> 0-DAY CI Kernel Test Service
> https://github.com/intel/lkp-tests/wiki


-- 
Masami Hiramatsu (Google) <mhiramat@kernel.org>

^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [PATCH 2/6] Provide __free(argv) for argv_split() users
  2025-01-03 23:01 ` [PATCH 2/6] Provide __free(argv) for argv_split() users Masami Hiramatsu (Google)
                     ` (2 preceding siblings ...)
  2025-01-04 11:09   ` Oleg Nesterov
@ 2025-01-06 19:30   ` Steven Rostedt
  3 siblings, 0 replies; 13+ messages in thread
From: Steven Rostedt @ 2025-01-06 19:30 UTC (permalink / raw)
  To: Masami Hiramatsu (Google)
  Cc: Peter Zijlstra, Anil S Keshavamurthy, David S . Miller,
	Mathieu Desnoyers, Oleg Nesterov, Tzvetomir Stoyanov,
	Naveen N Rao, Josh Poimboeuf, Jason Baron, Ard Biesheuvel,
	linux-kernel, linux-trace-kernel

On Sat,  4 Jan 2025 08:01:47 +0900
"Masami Hiramatsu (Google)" <mhiramat@kernel.org> wrote:

> From: Masami Hiramatsu (Google) <mhiramat@kernel.org>
> 
> Provide __free(argv) macro for argv_split() users so that they can
> avoid gotos.
> 
> Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
> ---
>  include/linux/string.h |    2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/include/linux/string.h b/include/linux/string.h
> index 493ac4862c77..7035a70e30be 100644
> --- a/include/linux/string.h
> +++ b/include/linux/string.h
> @@ -312,6 +312,8 @@ extern void *kmemdup_array(const void *src, size_t count, size_t element_size, g
>  extern char **argv_split(gfp_t gfp, const char *str, int *argcp);
>  extern void argv_free(char **argv);
>  
> +DEFINE_FREE(argv, char **, argv_free(_T))
> +
>  /* lib/cmdline.c */
>  extern int get_option(char **str, int *pint);
>  extern char *get_options(const char *str, int nints, int *ints);

FYI, I already have this change in linux-next:

  https://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace.git/commit/?h=for-next&id=9e49ca756d207f4313fb7af48648a67da8e4e250
  https://lore.kernel.org/20241220103313.4a74ec8e@gandalf.local.home

-- Steve

^ permalink raw reply	[flat|nested] 13+ messages in thread

end of thread, other threads:[~2025-01-06 19:28 UTC | newest]

Thread overview: 13+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-01-03 23:01 [PATCH 0/6] kprobes: jump label: Cleanup with guard and __free Masami Hiramatsu (Google)
2025-01-03 23:01 ` [PATCH 1/6] tracing/kprobes: Fix to free objects when failed to copy a symbol Masami Hiramatsu (Google)
2025-01-03 23:01 ` [PATCH 2/6] Provide __free(argv) for argv_split() users Masami Hiramatsu (Google)
2025-01-04  6:39   ` kernel test robot
2025-01-04  6:51   ` kernel test robot
2025-01-05  9:12     ` Masami Hiramatsu
2025-01-04 11:09   ` Oleg Nesterov
2025-01-05  9:00     ` Masami Hiramatsu
2025-01-06 19:30   ` Steven Rostedt
2025-01-03 23:01 ` [PATCH 3/6] tracing: Use __free() for argv in dynevent Masami Hiramatsu (Google)
2025-01-03 23:02 ` [PATCH 4/6] tracing: Use __free() in trace_probe for cleanup Masami Hiramatsu (Google)
2025-01-03 23:02 ` [PATCH 5/6] tracing: Use __free() for kprobe events to cleanup Masami Hiramatsu (Google)
2025-01-03 23:02 ` [PATCH 6/6] tracing/kprobes: Simplify __trace_kprobe_create() by removing gotos Masami Hiramatsu (Google)

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).