The patch titled
containers: implement namespace tracking subsystem (v3)
has been added to the -mm tree. Its filename is
containers-implement-namespace-tracking-subsystem-v3.patch
*** Remember to use Documentation/SubmitChecklist when testing your code ***
See http://www.zip.com.au/~akpm/linux/patches/stuff/added-to-mm.txt to find
out what to do about this
------------------------------------------------------
Subject: containers: implement namespace tracking subsystem (v3)
From: "Serge E. Hallyn" <serue@xxxxxxxxxx>
When a task enters a new namespace via a clone() or unshare(), a new container
is created and the task moves into it.
This version names containers which are automatically created using
container_clone() as "node_<pid>" where pid is the pid of the unsharing or
cloned process. (Thanks Pavel for the idea) This is safe because if the
process unshares again, it will create
/containers/(...)/node_<pid>/node_<pid>
The only possibilities (AFAICT) for a -EEXIST on unshare are
1. pid wraparound
2. a process fails an unshare, then tries again.
Case 1 is unlikely enough that I ignore it (at least for now). In case 2, the
node_<pid> will be empty and can be rmdir'ed to make the subsequent unshare()
succeed.
Changelog:
Name cloned containers as "node_<pid>".
Signed-off-by: Serge E. Hallyn <serue@xxxxxxxxxx>
Cc: Paul Menage <menage@xxxxxxxxxx>
Cc: "Eric W. Biederman" <ebiederm@xxxxxxxxxxxx>
Cc: Dave Hansen <haveblue@xxxxxxxxxx>
Cc: Balbir Singh <balbir@xxxxxxxxxx>
Cc: Paul Jackson <pj@xxxxxxx>
Cc: Kirill Korotaev <dev@xxxxxxxxxx>
Cc: Herbert Poetzl <herbert@xxxxxxxxxxxx>
Cc: Srivatsa Vaddagiri <vatsa@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/container_subsys.h | 6 +
include/linux/nsproxy.h | 7 ++
init/Kconfig | 9 ++
kernel/Makefile | 1
kernel/container.c | 23 ++++--
kernel/ns_container.c | 99 +++++++++++++++++++++++++++++
kernel/nsproxy.c | 15 ++++
7 files changed, 151 insertions(+), 9 deletions(-)
diff -puN include/linux/container_subsys.h~containers-implement-namespace-tracking-subsystem-v3 include/linux/container_subsys.h
--- a/include/linux/container_subsys.h~containers-implement-namespace-tracking-subsystem-v3
+++ a/include/linux/container_subsys.h
@@ -24,3 +24,9 @@ SUBSYS(debug)
#endif
/* */
+
+#ifdef CONFIG_CONTAINER_NS
+SUBSYS(ns)
+#endif
+
+/* */
diff -puN include/linux/nsproxy.h~containers-implement-namespace-tracking-subsystem-v3 include/linux/nsproxy.h
--- a/include/linux/nsproxy.h~containers-implement-namespace-tracking-subsystem-v3
+++ a/include/linux/nsproxy.h
@@ -55,4 +55,11 @@ static inline void exit_task_namespaces(
put_nsproxy(ns);
}
}
+
+#ifdef CONFIG_CONTAINER_NS
+int ns_container_clone(struct task_struct *tsk);
+#else
+static inline int ns_container_clone(struct task_struct *tsk) { return 0; }
+#endif
+
#endif
diff -puN init/Kconfig~containers-implement-namespace-tracking-subsystem-v3 init/Kconfig
--- a/init/Kconfig~containers-implement-namespace-tracking-subsystem-v3
+++ a/init/Kconfig
@@ -347,6 +347,15 @@ config CONTAINER_CPUACCT
Provides a simple Resource Controller for monitoring the
total CPU consumed by the tasks in a container
+config CONTAINER_NS
+ bool "Namespace container subsystem"
+ select CONTAINERS
+ help
+ Provides a simple namespace container subsystem to
+ provide hierarchical naming of sets of namespaces,
+ for instance virtual servers and checkpoint/restart
+ jobs.
+
config PROC_PID_CPUSET
bool "Include legacy /proc/<pid>/cpuset file"
depends on CPUSETS
diff -puN kernel/Makefile~containers-implement-namespace-tracking-subsystem-v3 kernel/Makefile
--- a/kernel/Makefile~containers-implement-namespace-tracking-subsystem-v3
+++ a/kernel/Makefile
@@ -41,6 +41,7 @@ obj-$(CONFIG_CONTAINERS) += container.o
obj-$(CONFIG_CONTAINER_DEBUG) += container_debug.o
obj-$(CONFIG_CPUSETS) += cpuset.o
obj-$(CONFIG_CONTAINER_CPUACCT) += cpu_acct.o
+obj-$(CONFIG_CONTAINER_NS) += ns_container.o
obj-$(CONFIG_IKCONFIG) += configs.o
obj-$(CONFIG_STOP_MACHINE) += stop_machine.o
obj-$(CONFIG_AUDIT) += audit.o auditfilter.o
diff -puN kernel/container.c~containers-implement-namespace-tracking-subsystem-v3 kernel/container.c
--- a/kernel/container.c~containers-implement-namespace-tracking-subsystem-v3
+++ a/kernel/container.c
@@ -2355,12 +2355,6 @@ void container_exit(struct task_struct *
put_css_group_taskexit(cg);
}
-static atomic_t namecnt;
-static void get_unused_name(char *buf)
-{
- sprintf(buf, "node%d", atomic_inc_return(&namecnt));
-}
-
/**
* container_clone - duplicate the current container in the hierarchy
* that the given subsystem is attached to, and move this task into
@@ -2370,7 +2364,7 @@ int container_clone(struct task_struct *
{
struct dentry *dentry;
int ret = 0;
- char nodename[32];
+ char nodename[MAX_CONTAINER_TYPE_NAMELEN];
struct container *parent, *child;
struct inode *inode;
struct css_group *cg;
@@ -2394,6 +2388,9 @@ int container_clone(struct task_struct *
}
cg = tsk->containers;
parent = task_container(tsk, subsys->subsys_id);
+
+ snprintf(nodename, MAX_CONTAINER_TYPE_NAMELEN, "node_%d", tsk->pid);
+
/* Pin the hierarchy */
atomic_inc(&parent->root->sb->s_active);
@@ -2402,7 +2399,6 @@ int container_clone(struct task_struct *
mutex_unlock(&container_mutex);
/* Now do the VFS work to create a container */
- get_unused_name(nodename);
inode = parent->dentry->d_inode;
/* Hold the parent directory mutex across this operation to
@@ -2475,8 +2471,14 @@ int container_clone(struct task_struct *
return ret;
}
-/* See if "cont" is a descendant of the current task's container in
+/*
+ * See if "cont" is a descendant of the current task's container in
* the appropriate hierarchy
+ *
+ * If we are sending in dummytop, then presumably we are creating
+ * the top container in the subsystem.
+ *
+ * Called only by the ns (nsproxy) container.
*/
int container_is_descendant(const struct container *cont)
{
@@ -2484,6 +2486,9 @@ int container_is_descendant(const struct
struct container *target;
int subsys_id;
+ if (cont == dummytop)
+ return 1;
+
get_first_subsys(cont, NULL, &subsys_id);
target = task_container(current, subsys_id);
while (cont != target && cont!= cont->top_container)
diff -puN /dev/null kernel/ns_container.c
--- /dev/null
+++ a/kernel/ns_container.c
@@ -0,0 +1,99 @@
+/*
+ * ns_container.c - namespace container subsystem
+ *
+ * Copyright 2006, 2007 IBM Corp
+ */
+
+#include <linux/module.h>
+#include <linux/container.h>
+#include <linux/fs.h>
+
+struct ns_container {
+ struct container_subsys_state css;
+ spinlock_t lock;
+};
+
+struct container_subsys ns_subsys;
+
+static inline struct ns_container *container_to_ns(
+ struct container *container)
+{
+ return container_of(container_subsys_state(container, ns_subsys_id),
+ struct ns_container, css);
+}
+
+int ns_container_clone(struct task_struct *task)
+{
+ return container_clone(task, &ns_subsys);
+}
+
+/*
+ * Rules:
+ * 1. you can only enter a container which is a child of your current
+ * container
+ * 2. you can only place another process into a container if
+ * a. you have CAP_SYS_ADMIN
+ * b. your container is an ancestor of task's destination container
+ * (hence either you are in the same container as task, or in an
+ * ancestor container thereof)
+ */
+static int ns_can_attach(struct container_subsys *ss,
+ struct container *new_container, struct task_struct *task)
+{
+ struct container *orig;
+
+ if (current != task) {
+ if (!capable(CAP_SYS_ADMIN))
+ return -EPERM;
+
+ if (!container_is_descendant(new_container))
+ return -EPERM;
+ }
+
+ if (atomic_read(&new_container->count) != 0)
+ return -EPERM;
+
+ orig = task_container(task, ns_subsys_id);
+ if (orig && orig != new_container->parent)
+ return -EPERM;
+
+ return 0;
+}
+
+/*
+ * Rules: you can only create a container if
+ * 1. you are capable(CAP_SYS_ADMIN)
+ * 2. the target container is a descendant of your own container
+ */
+static int ns_create(struct container_subsys *ss, struct container *container)
+{
+ struct ns_container *ns_container;
+
+ if (!capable(CAP_SYS_ADMIN))
+ return -EPERM;
+ if (!container_is_descendant(container))
+ return -EPERM;
+
+ ns_container = kzalloc(sizeof(*ns_container), GFP_KERNEL);
+ if (!ns_container) return -ENOMEM;
+ spin_lock_init(&ns_container->lock);
+ container->subsys[ns_subsys.subsys_id] = &ns_container->css;
+ return 0;
+}
+
+static void ns_destroy(struct container_subsys *ss,
+ struct container *container)
+{
+ struct ns_container *ns_container;
+
+ ns_container = container_to_ns(container);
+ kfree(ns_container);
+}
+
+struct container_subsys ns_subsys = {
+ .name = "ns",
+ .can_attach = ns_can_attach,
+ .create = ns_create,
+ .destroy = ns_destroy,
+ .subsys_id = ns_subsys_id,
+};
diff -puN kernel/nsproxy.c~containers-implement-namespace-tracking-subsystem-v3 kernel/nsproxy.c
--- a/kernel/nsproxy.c~containers-implement-namespace-tracking-subsystem-v3
+++ a/kernel/nsproxy.c
@@ -146,7 +146,14 @@ int copy_namespaces(unsigned long flags,
goto out;
}
+ err = ns_container_clone(tsk);
+ if (err) {
+ put_nsproxy(new_ns);
+ goto out;
+ }
+
tsk->nsproxy = new_ns;
+
out:
put_nsproxy(old_ns);
return err;
@@ -187,6 +194,14 @@ int unshare_nsproxy_namespaces(unsigned
new_fs ? new_fs : current->fs);
if (IS_ERR(*new_nsp))
err = PTR_ERR(*new_nsp);
+ goto out;
+ }
+
+ err = ns_container_clone(current);
+ if (err)
+ put_nsproxy(*new_nsp);
+
+out:
return err;
}
_
Patches currently in -mm which might be from serue@xxxxxxxxxx are
implement-file-posix-capabilities.patch
implement-file-posix-capabilities-fix.patch
remove-config_uts_ns-and-config_ipc_ns.patch
user-namespace-add-the-framework.patch
user-namespace-add-unshare.patch
mm-fix-create_new_namespaces-return-value.patch
cpuset-zero-malloc-revert-the-old-cpuset-fix.patch
containersv10-basic-container-framework.patch
containersv10-basic-container-framework-fix.patch
containersv10-example-cpu-accounting-subsystem.patch
containersv10-example-cpu-accounting-subsystem-fix.patch
containersv10-add-tasks-file-interface.patch
containersv10-add-tasks-file-interface-fix.patch
containersv10-add-fork-exit-hooks.patch
containersv10-add-fork-exit-hooks-fix.patch
containersv10-add-container_clone-interface.patch
containersv10-add-container_clone-interface-fix.patch
containersv10-add-procfs-interface.patch
containersv10-add-procfs-interface-fix.patch
containersv10-make-cpusets-a-client-of-containers.patch
containersv10-share-css_group-arrays-between-tasks-with-same-container-memberships.patch
containersv10-share-css_group-arrays-between-tasks-with-same-container-memberships-fix.patch
containersv10-share-css_group-arrays-between-tasks-with-same-container-memberships-cpuset-zero-malloc-fix-for-new-containers.patch
containersv10-simple-debug-info-subsystem.patch
containersv10-simple-debug-info-subsystem-fix.patch
containersv10-simple-debug-info-subsystem-fix-2.patch
containersv10-support-for-automatic-userspace-release-agents.patch
containers-implement-subsys-post_clone.patch
containers-implement-namespace-tracking-subsystem-v3.patch
-
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
