On Wed, 21 Jan 2026 11:30:35 +0900 Masami Hiramatsu (Google) <[email protected]> wrote:
> > It would make much more sense to have add_string() free vals, and set > > vals to NULL on error. > > I think it should be failed if it fails to add string. Can it > continue checking tracepoints even after the error? This patch is simply fixing a memory leak on failure (which isn't really a big deal since this is just a user space tool that runs for a short time during build). Returning a failure here is out of scope of this patch. Feel free to send an RFC patch that returns a failure built on top of this patch, and we can discuss if that should be done or not for that change. -- Steve
