kernel/fork.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
If unshare_nsproxy_namespaces() successfully creates the new_nsproxy,
but then set_cred_ucounts() fails, on its error path there is no cleanup
for new_nsproxy, so it is leaked. Let's fix that by freeing new_nsproxy
if it's not NULL on this error path.
Fixes: 905ae01c4ae2a ("Add a reference to ucounts for each cred")
Signed-off-by: Pavel Tikhomirov <ptikhomirov@virtuozzo.com>
---
kernel/fork.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/kernel/fork.c b/kernel/fork.c
index 3da0f08615a95..6f7332e3e0c8c 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -3133,8 +3133,11 @@ int ksys_unshare(unsigned long unshare_flags)
if (new_cred) {
err = set_cred_ucounts(new_cred);
- if (err)
+ if (err) {
+ if (new_nsproxy)
+ free_nsproxy(new_nsproxy);
goto bad_unshare_cleanup_cred;
+ }
}
if (new_fs || new_fd || do_sysvsem || new_cred || new_nsproxy) {
--
2.51.1
* Pavel Tikhomirov <ptikhomirov@virtuozzo.com> [251118 01:46]:
> If unshare_nsproxy_namespaces() successfully creates the new_nsproxy,
> but then set_cred_ucounts() fails, on its error path there is no cleanup
> for new_nsproxy, so it is leaked. Let's fix that by freeing new_nsproxy
> if it's not NULL on this error path.
new_nsproxy may be set to an error pointer, but that case is handled
earlier in this function. That's a pretty subtle detail.
unshare_nsproxy_namespaces() should probably set it to NULL if it's an
error.. everywhere else looks fine.
This fix looks good!
Reviewed-by: Liam R. Howlett <Liam.Howlett@oracle.com>
>
> Fixes: 905ae01c4ae2a ("Add a reference to ucounts for each cred")
> Signed-off-by: Pavel Tikhomirov <ptikhomirov@virtuozzo.com>
> ---
> kernel/fork.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/kernel/fork.c b/kernel/fork.c
> index 3da0f08615a95..6f7332e3e0c8c 100644
> --- a/kernel/fork.c
> +++ b/kernel/fork.c
> @@ -3133,8 +3133,11 @@ int ksys_unshare(unsigned long unshare_flags)
>
> if (new_cred) {
> err = set_cred_ucounts(new_cred);
> - if (err)
> + if (err) {
> + if (new_nsproxy)
> + free_nsproxy(new_nsproxy);
> goto bad_unshare_cleanup_cred;
> + }
> }
>
> if (new_fs || new_fd || do_sysvsem || new_cred || new_nsproxy) {
> --
> 2.51.1
>
On Tue, Nov 18, 2025 at 02:45:50PM +0800, Pavel Tikhomirov wrote:
> If unshare_nsproxy_namespaces() successfully creates the new_nsproxy,
> but then set_cred_ucounts() fails, on its error path there is no cleanup
> for new_nsproxy, so it is leaked. Let's fix that by freeing new_nsproxy
> if it's not NULL on this error path.
>
> Fixes: 905ae01c4ae2a ("Add a reference to ucounts for each cred")
> Signed-off-by: Pavel Tikhomirov <ptikhomirov@virtuozzo.com>
Cc: stable@vger.kernel.org
Acked-by: Alexey Gladkov <legion@kernel.org>
> ---
> kernel/fork.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/kernel/fork.c b/kernel/fork.c
> index 3da0f08615a95..6f7332e3e0c8c 100644
> --- a/kernel/fork.c
> +++ b/kernel/fork.c
> @@ -3133,8 +3133,11 @@ int ksys_unshare(unsigned long unshare_flags)
>
> if (new_cred) {
> err = set_cred_ucounts(new_cred);
> - if (err)
> + if (err) {
> + if (new_nsproxy)
> + free_nsproxy(new_nsproxy);
> goto bad_unshare_cleanup_cred;
> + }
> }
>
> if (new_fs || new_fd || do_sysvsem || new_cred || new_nsproxy) {
> --
> 2.51.1
>
--
Rgrds, legion
© 2016 - 2025 Red Hat, Inc.