Thread (1 message) 1 message, 1 author, 2021-10-22

Re: [PATCH 3/6] clone: fix a memory leak of the "git_dir" variable

From: Junio C Hamano <hidden>
Date: 2021-10-22 00:07:55

Ævar Arnfjörð Bjarmason  [off-list ref] writes:
quoted hunk
At this point in cmd_clone the "git_dir" is always either an
xstrdup()'d string, or something we got from mkpathdup(). Let's free()
it before we clobber it.

Signed-off-by: Ævar Arnfjörð Bjarmason <redacted>
---
 builtin/clone.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/builtin/clone.c b/builtin/clone.c
index 559acf9e036..fb377b27657 100644
--- a/builtin/clone.c
+++ b/builtin/clone.c
@@ -1040,8 +1040,10 @@ int cmd_clone(int argc, const char **argv, const char *prefix)
 	init_db(git_dir, real_git_dir, option_template, GIT_HASH_UNKNOWN, NULL,
 		INIT_DB_QUIET);
 
-	if (real_git_dir)
+	if (real_git_dir) {
+		free((char *)git_dir);
 		git_dir = real_git_dir;
+	}
 
 	/*
 	 * additional config can be injected with -c, make sure it's included
I had to wonder if the old git_dir can still be pointed at by
junk_git_dir.  Much earlier than this point there is this:

	if (real_git_dir) {
		if (real_dest_exists)
			junk_git_dir_flags |= REMOVE_DIR_KEEP_TOPLEVEL;
		junk_git_dir = real_git_dir;
	} else {
		if (dest_exists)
			junk_git_dir_flags |= REMOVE_DIR_KEEP_TOPLEVEL;
		junk_git_dir = git_dir;
	}
	if (safe_create_leading_directories_const(git_dir) < 0)
		die(_("could not create leading directories of '%s'"), git_dir);

Luckily, junk_git_dir gets git_dir only when !real_git_dir, so it is
safe.  real_git_dir can only be set via the --separate-git-dir
command line option, so we are safe here.

Thanks.
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help