[PATCH v2] scripts/tracepoint-update: fix memory leak in add_string() on failure

[PATCH v2] scripts/tracepoint-update: fix memory leak in add_string() on failure

[Weigang He]

When realloc() fails in add_string(), the function returns -1 but leaves
*vals pointing to the previously allocated memory. This can cause memory
leaks in callers like make_trace_array() that return on error without
freeing the partially built array.

Fix this by freeing *vals and setting it to NULL when realloc() fails.
This makes the error handling self-contained in add_string() so callers
don't need to handle cleanup on failure.

This bug is found by my static analysis tool and my code review.

Signed-off-by: Weigang He <geoffreyhe2@gmail.com>
---
 scripts/tracepoint-update.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/scripts/tracepoint-update.c b/scripts/tracepoint-update.c
index 90046aedc97b9..5cf43c0aac891 100644
--- a/scripts/tracepoint-update.c
+++ b/scripts/tracepoint-update.c
@@ -49,6 +49,8 @@ static int add_string(const char *str, const char ***vals, int *count)
 		array = realloc(array, sizeof(char *) * size);
 		if (!array) {
 			fprintf(stderr, "Failed memory allocation\n");
+			free(*vals);
+			*vals = NULL;
 			return -1;
 		}
 		*vals = array;
-- 
2.34.1

Re: [PATCH v2] scripts/tracepoint-update: fix memory leak in add_string() on failure

[Markus Elfring]

…
> This bug is found by my static analysis tool …

Will any additional background information become more helpful here?


> ---
>  scripts/tracepoint-update.c | 2 ++
…

Some contributors would appreciate patch version descriptions.
https://lore.kernel.org/all/?q=%22This+looks+like+a+new+version+of+a+previously+submitted+patch%22
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/process/submitting-patches.rst?h=v6.19-rc6#n310

Regards,
Markus

Re: [PATCH v2] scripts/tracepoint-update: fix memory leak in add_string() on failure

[Masami Hiramatsu]

On Mon, 19 Jan 2026 11:45:42 +0000
Weigang He <geoffreyhe2@gmail.com> wrote:

> When realloc() fails in add_string(), the function returns -1 but leaves
> *vals pointing to the previously allocated memory. This can cause memory
> leaks in callers like make_trace_array() that return on error without
> freeing the partially built array.
> 
> Fix this by freeing *vals and setting it to NULL when realloc() fails.
> This makes the error handling self-contained in add_string() so callers
> don't need to handle cleanup on failure.

This looks not enough. If the memory allocation is failed, it should NOT
continue anything.

I think we need to make the command itself failure when it fails to
allocate memory, as below:

diff --git a/scripts/tracepoint-update.c b/scripts/tracepoint-update.c
index 90046aedc97b..1b4129a21942 100644
--- a/scripts/tracepoint-update.c
+++ b/scripts/tracepoint-update.c
@@ -94,7 +94,7 @@ static void make_trace_array(struct elf_tracepoint *etrace)
 		if (!len)
 			continue;
 		if (add_string(str, &vals, &count) < 0)
-			return;
+			exit(EXIT_FAILURE);
 	}
 
 	/* If CONFIG_TRACEPOINT_VERIFY_USED is not set, there's nothing to do */


Thank you,

> 
> This bug is found by my static analysis tool and my code review.
> 
> Signed-off-by: Weigang He <geoffreyhe2@gmail.com>
> ---
>  scripts/tracepoint-update.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/scripts/tracepoint-update.c b/scripts/tracepoint-update.c
> index 90046aedc97b9..5cf43c0aac891 100644
> --- a/scripts/tracepoint-update.c
> +++ b/scripts/tracepoint-update.c
> @@ -49,6 +49,8 @@ static int add_string(const char *str, const char ***vals, int *count)
>  		array = realloc(array, sizeof(char *) * size);
>  		if (!array) {
>  			fprintf(stderr, "Failed memory allocation\n");
> +			free(*vals);
> +			*vals = NULL;
>  			return -1;
>  		}
>  		*vals = array;
> -- 
> 2.34.1
> 


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

Re: [PATCH v2] scripts/tracepoint-update: fix memory leak in add_string() on failure

[Steven Rostedt]

On Wed, 21 Jan 2026 11:22:53 +0900
Masami Hiramatsu (Google) <mhiramat@kernel.org> wrote:

> > 
> > Fix this by freeing *vals and setting it to NULL when realloc() fails.
> > This makes the error handling self-contained in add_string() so callers
> > don't need to handle cleanup on failure.  
> 
> This looks not enough. If the memory allocation is failed, it should NOT
> continue anything.
> 
> I think we need to make the command itself failure when it fails to
> allocate memory, as below:

That's a separate bug. I'm taking this current patch as-is as a fix for the
leak. Not stopping immediately is a separate issue.

-- Steve