shm: handle separate PID namespaces case
Vasiliy Kulikov [Thu, 28 Jul 2011 23:55:31 +0000 (03:55 +0400)]
shm_try_destroy_orphaned() and shm_try_destroy_current() didn't handle
the case of separate PID namespaces, but a single IPC namespace.  If
there are tasks with the same PID values using the same shmem object,
the wrong destroy decision could be reached.

On shm segment creation store the pointer to the creator task in
shmid_kernel->shm_creator field and zero it on task exit.  Then
use the ->shm_creator insread of shm_cprid in both functions.  As
shmid_kernel object is already locked at this stage, no additional
locking is needed.

Signed-off-by: Vasiliy Kulikov <segoon@openwall.com>
Acked-by: Serge Hallyn <serge.hallyn@canonical.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

include/linux/shm.h
ipc/shm.c

index 7d27ffd..92808b8 100644 (file)
@@ -95,6 +95,9 @@ struct shmid_kernel /* private to the kernel */
        pid_t                   shm_cprid;
        pid_t                   shm_lprid;
        struct user_struct      *mlock_user;
+
+       /* The task created the shm object.  NULL if the task is dead. */
+       struct task_struct      *shm_creator;
 };
 
 /* shm_mode upper byte flags */
index 3f5b143..fdaf8be 100644 (file)
--- a/ipc/shm.c
+++ b/ipc/shm.c
@@ -239,7 +239,23 @@ static int shm_try_destroy_current(int id, void *p, void *data)
        if (IS_ERR(shp))
                return 0;
 
-       if (shp->shm_cprid != task_tgid_vnr(current)) {
+       if (shp->shm_creator != current) {
+               shm_unlock(shp);
+               return 0;
+       }
+
+       /*
+        * Mark it as orphaned to destroy the segment when
+        * kernel.shm_rmid_forced is changed.
+        * It is noop if the following shm_may_destroy() returns true.
+        */
+       shp->shm_creator = NULL;
+
+       /*
+        * Don't even try to destroy it.  If shm_rmid_forced=0 and IPC_RMID
+        * is not set, it shouldn't be deleted here.
+        */
+       if (!ns->shm_rmid_forced) {
                shm_unlock(shp);
                return 0;
        }
@@ -255,7 +271,6 @@ static int shm_try_destroy_orphaned(int id, void *p, void *data)
 {
        struct ipc_namespace *ns = data;
        struct shmid_kernel *shp = shm_lock(ns, id);
-       struct task_struct *task;
 
        if (IS_ERR(shp))
                return 0;
@@ -263,11 +278,8 @@ static int shm_try_destroy_orphaned(int id, void *p, void *data)
        /*
         * We want to destroy segments without users and with already
         * exit'ed originating process.
-        *
-        * XXX: the originating process may exist in another pid namespace.
         */
-       task = find_task_by_vpid(shp->shm_cprid);
-       if (task != NULL) {
+       if (shp->shm_creator != NULL) {
                shm_unlock(shp);
                return 0;
        }
@@ -295,7 +307,7 @@ void exit_shm(struct task_struct *task)
        if (!nsp)
                return;
        ns = nsp->ipc_ns;
-       if (!ns || !ns->shm_rmid_forced)
+       if (!ns)
                return;
 
        /* Destroy all already created segments, but not mapped yet */
@@ -494,6 +506,7 @@ static int newseg(struct ipc_namespace *ns, struct ipc_params *params)
        shp->shm_segsz = size;
        shp->shm_nattch = 0;
        shp->shm_file = file;
+       shp->shm_creator = current;
        /*
         * shmid gets reported as "inode#" in /proc/pid/maps.
         * proc-ps tools use this. Changing this will break them.