FAILED: patch "[PATCH] pidfd: hold exec_update_lock around namespace ioctl" failed to apply to 6.18-stable tree
Subsystems (2)
[email protected]
maintainerThe patch below does not apply to the 6.18-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.
To reproduce the conflict and resubmit, you may use the following commands:
git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.18.y
git checkout FETCH_HEAD
git cherry-pick -x 9688a46802939da28f00cb40e8129615d5d4af39
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<[email protected]>' --in-reply-to '2026090328-fleshed-scrambler-e1f2@gregkh' --subject-prefix 'PATCH 6.18.y' 'HEAD^..'
Possible dependencies:
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
From 9688a46802939da28f00cb40e8129615d5d4af39 Mon Sep 17 00:00:00 2001
From: Chen Linxuan <[email protected]>
Date: Fri, 31 Jul 2026 22:50:08 +0800
Subject: [PATCH] pidfd: hold exec_update_lock around namespace ioctl
The PIDFD_GET_*_NAMESPACE ioctls in pidfd_ioctl() perform a filesystem credentials ptrace access check before handing out a namespace file descriptor. The accompanying comment states that the code "mirrors nsfs behavior", but, unlike the corresponding procfs paths, it does so without holding the target task's exec_update_lock.
proc_ns_get_link() and proc_ns_readlink() both take exec_update_lock for reading around the ptrace check and the namespace lookup, so that the credentials used for the access decision match those of the task when its namespace is read. Without it, a caller can pass the check against the target's old credentials and then read the namespace after the target has execve()'d a setuid binary and committed new credentials -- accessing namespace information it should have been denied.
Hold exec_update_lock for reading around the ptrace check and the namespace lookup so that pidfd truly mirrors nsfs behavior, as the comment already claims. open_namespace() itself runs outside the lock: once a namespace reference is obtained it carries its own refcount and is opened with the caller's own credentials, so a concurrent execve() on the target can no longer affect the outcome.
fs/pidfs.c +31 -12
@@ -531,6 +531,7 @@ static long pidfd_ioctl(struct file *file, unsigned int cmd, unsigned long arg) struct task_struct *task __free(put_task) = NULL; struct nsproxy *nsp __free(put_nsproxy) = NULL;struct ns_common *ns_common =NULL ; +int error;if (!pidfs_ioctl_valid (cmd))return -ENOIOCTLCMD ;
@@ -554,20 +555,33 @@ static long pidfd_ioctl(struct file *file, unsigned int cmd, unsigned long arg)if (arg)return -EINVAL ; +/* +* We're trying to open a file descriptor to the namespace so perform a +* filesystem cred ptrace check. Hold @task's exec_update_lock for the +* duration of the ptrace check and the namespace lookup so that the +* credentials used for the access decision match those of @task at the +* time its namespace is read, preventing a concurrent execve() from +* swapping the task's credentials in between the check and the use. We +* mirror nsfs behavior. +*/ + error =down_read_killable (&task->signal->exec_update_lock); +if (error) +return error; + +if (!ptrace_may_access (task,PTRACE_MODE_READ_FSCREDS )) { + error = -EACCES ; + goto out_unlock; + } + scoped_guard(task_lock, task) { nsp = task->nsproxy;if (nsp)get_nsproxy (nsp); } -if (!nsp) -return -ESRCH ;/* just pretend it didn't exist */ - -/* -* We're trying to open a file descriptor to the namespace so perform a -* filesystem cred ptrace check. Also, we mirror nsfs behavior. -*/ -if (!ptrace_may_access (task,PTRACE_MODE_READ_FSCREDS )) -return -EACCES ; +if (!nsp) { + error = -ESRCH ;/* just pretend it didn't exist */ + goto out_unlock; + }switch (cmd) {/* Namespaces that hang of nsproxy. */
@@ -649,11 +663,16 @@ static long pidfd_ioctl(struct file *file, unsigned int cmd, unsigned long arg)#endif break ;default : -return -ENOIOCTLCMD ; + error = -ENOIOCTLCMD ; } -if (!ns_common) -return -EOPNOTSUPP ; +if (!error && !ns_common) + error = -EOPNOTSUPP ; + +out_unlock: +up_read (&task->signal->exec_update_lock); +if (error) +return error;/* open_namespace() unconditionally consumes the reference */ return open_namespace (ns_common);
Sasha Levin
maintainerPATCH 1/2 pidfs: simplify PIDFD_GET_<type>_NAMESPACE ioctls
From: Christian Brauner <[email protected]>
[ Upstream commit a71e4f103aed69e7a11ea913312726bb194c76ee ]
We have reworked namespaces sufficiently that all this special-casing shouldn't be needed anymore
Stable-dep-of: 9688a4680293 ("pidfd: hold exec_update_lock around namespace ioctl")
--- fs/pidfs.c | 75 ++++++++++++++++++++++++++---------------------------- 1 file changed, 36 insertions(+), 39 deletions(-)
fs/pidfs.c +36 -39
@@ -456,7 +456,6 @@ static long pidfd_ioctl(struct file *file, unsigned int cmd, unsigned long arg) struct task_struct *task __free(put_task) = NULL; struct nsproxy *nsp __free(put_nsproxy) = NULL;struct ns_common *ns_common =NULL ; -struct pid_namespace *pid_ns;if (!pidfs_ioctl_valid (cmd))return -ENOIOCTLCMD ;
@@ -498,66 +497,64 @@ static long pidfd_ioctl(struct file *file, unsigned int cmd, unsigned long arg)switch (cmd) {/* Namespaces that hang of nsproxy. */ case PIDFD_GET_CGROUP_NAMESPACE : -if (IS_ENABLED (CONFIG_CGROUPS )) { -get_cgroup_ns (nsp->cgroup_ns); - ns_common =to_ns_common (nsp->cgroup_ns); - } +if (!ns_ref_get (nsp->cgroup_ns)) +break ; + ns_common =to_ns_common (nsp->cgroup_ns);break ;case PIDFD_GET_IPC_NAMESPACE : -if (IS_ENABLED (CONFIG_IPC_NS )) { -get_ipc_ns (nsp->ipc_ns); - ns_common =to_ns_common (nsp->ipc_ns); - } +if (!ns_ref_get (nsp->ipc_ns)) +break ; + ns_common =to_ns_common (nsp->ipc_ns);break ;case PIDFD_GET_MNT_NAMESPACE : -get_mnt_ns (nsp->mnt_ns); +if (!ns_ref_get (nsp->mnt_ns)) +break ; ns_common =to_ns_common (nsp->mnt_ns);break ;case PIDFD_GET_NET_NAMESPACE : -if (IS_ENABLED (CONFIG_NET_NS )) { - ns_common =to_ns_common (nsp->net_ns); -get_net_ns (ns_common); - } +if (!ns_ref_get (nsp->net_ns)) +break ; + ns_common =to_ns_common (nsp->net_ns);break ;case PIDFD_GET_PID_FOR_CHILDREN_NAMESPACE : -if (IS_ENABLED (CONFIG_PID_NS )) { -get_pid_ns (nsp->pid_ns_for_children); - ns_common =to_ns_common (nsp->pid_ns_for_children); - } +if (!ns_ref_get (nsp->pid_ns_for_children)) +break ; + ns_common =to_ns_common (nsp->pid_ns_for_children);break ;case PIDFD_GET_TIME_NAMESPACE : -if (IS_ENABLED (CONFIG_TIME_NS )) { -get_time_ns (nsp->time_ns); - ns_common =to_ns_common (nsp->time_ns); - } +if (!ns_ref_get (nsp->time_ns)) +break ; + ns_common =to_ns_common (nsp->time_ns);break ;case PIDFD_GET_TIME_FOR_CHILDREN_NAMESPACE : -if (IS_ENABLED (CONFIG_TIME_NS )) { -get_time_ns (nsp->time_ns_for_children); - ns_common =to_ns_common (nsp->time_ns_for_children); - } +if (!ns_ref_get (nsp->time_ns_for_children)) +break ; + ns_common =to_ns_common (nsp->time_ns_for_children);break ;case PIDFD_GET_UTS_NAMESPACE : -if (IS_ENABLED (CONFIG_UTS_NS )) { -get_uts_ns (nsp->uts_ns); - ns_common =to_ns_common (nsp->uts_ns); - } +if (!ns_ref_get (nsp->uts_ns)) +break ; + ns_common =to_ns_common (nsp->uts_ns);break ;/* Namespaces that don't hang of nsproxy. */ case PIDFD_GET_USER_NAMESPACE : -if (IS_ENABLED (CONFIG_USER_NS )) { -rcu_read_lock (); - ns_common =to_ns_common (get_user_ns (task_cred_xxx (task, user_ns))); -rcu_read_unlock (); + scoped_guard(rcu) { +struct user_namespace *user_ns; + + user_ns =task_cred_xxx (task, user_ns); +if (!ns_ref_get (user_ns)) +break ; + ns_common =to_ns_common (user_ns); }break ;case PIDFD_GET_PID_NAMESPACE : -if (IS_ENABLED (CONFIG_PID_NS )) { -rcu_read_lock (); + scoped_guard(rcu) { +struct pid_namespace *pid_ns; + pid_ns =task_active_pid_ns (task); -if (pid_ns) - ns_common =to_ns_common (get_pid_ns (pid_ns)); -rcu_read_unlock (); +if (!ns_ref_get (pid_ns)) +break ; + ns_common =to_ns_common (pid_ns); }break ;default :
Signature · 1 line
2.53.0
Sasha Levin
maintainerPATCH 2/2 pidfd: hold exec_update_lock around namespace ioctl
From: Chen Linxuan <[email protected]>
[ Upstream commit 9688a46802939da28f00cb40e8129615d5d4af39 ]
The PIDFD_GET_*_NAMESPACE ioctls in pidfd_ioctl() perform a filesystem credentials ptrace access check before handing out a namespace file descriptor. The accompanying comment states that the code "mirrors nsfs behavior", but, unlike the corresponding procfs paths, it does so without holding the target task's exec_update_lock.
proc_ns_get_link() and proc_ns_readlink() both take exec_update_lock for reading around the ptrace check and the namespace lookup, so that the credentials used for the access decision match those of the task when its namespace is read. Without it, a caller can pass the check against the target's old credentials and then read the namespace after the target has execve()'d a setuid binary and committed new credentials -- accessing namespace information it should have been denied.
Hold exec_update_lock for reading around the ptrace check and the namespace lookup so that pidfd truly mirrors nsfs behavior, as the comment already claims. open_namespace() itself runs outside the lock: once a namespace reference is obtained it carries its own refcount and is opened with the caller's own credentials, so a concurrent execve() on the target can no longer affect the outcome.
--- fs/pidfs.c | 43 +++++++++++++++++++++++++++++++------------ 1 file changed, 31 insertions(+), 12 deletions(-)
fs/pidfs.c +31 -12
@@ -456,6 +456,7 @@ static long pidfd_ioctl(struct file *file, unsigned int cmd, unsigned long arg) struct task_struct *task __free(put_task) = NULL; struct nsproxy *nsp __free(put_nsproxy) = NULL;struct ns_common *ns_common =NULL ; +int error;if (!pidfs_ioctl_valid (cmd))return -ENOIOCTLCMD ;
@@ -479,20 +480,33 @@ static long pidfd_ioctl(struct file *file, unsigned int cmd, unsigned long arg)if (arg)return -EINVAL ; +/* +* We're trying to open a file descriptor to the namespace so perform a +* filesystem cred ptrace check. Hold @task's exec_update_lock for the +* duration of the ptrace check and the namespace lookup so that the +* credentials used for the access decision match those of @task at the +* time its namespace is read, preventing a concurrent execve() from +* swapping the task's credentials in between the check and the use. We +* mirror nsfs behavior. +*/ + error =down_read_killable (&task->signal->exec_update_lock); +if (error) +return error; + +if (!ptrace_may_access (task,PTRACE_MODE_READ_FSCREDS )) { + error = -EACCES ; + goto out_unlock; + } + scoped_guard(task_lock, task) { nsp = task->nsproxy;if (nsp)get_nsproxy (nsp); } -if (!nsp) -return -ESRCH ;/* just pretend it didn't exist */ - -/* -* We're trying to open a file descriptor to the namespace so perform a -* filesystem cred ptrace check. Also, we mirror nsfs behavior. -*/ -if (!ptrace_may_access (task,PTRACE_MODE_READ_FSCREDS )) -return -EACCES ; +if (!nsp) { + error = -ESRCH ;/* just pretend it didn't exist */ + goto out_unlock; + }switch (cmd) {/* Namespaces that hang of nsproxy. */
@@ -558,11 +572,16 @@ static long pidfd_ioctl(struct file *file, unsigned int cmd, unsigned long arg) }break ;default : -return -ENOIOCTLCMD ; + error = -ENOIOCTLCMD ; } -if (!ns_common) -return -EOPNOTSUPP ; +if (!error && !ns_common) + error = -EOPNOTSUPP ; + +out_unlock: +up_read (&task->signal->exec_update_lock); +if (error) +return error;/* open_namespace() unconditionally consumes the reference */ return open_namespace (ns_common);
Signature · 1 line
2.53.0