aboutsummaryrefslogtreecommitdiffstats
path: root/fs
diff options
context:
space:
mode:
authorEric W. Biederman <ebiederm@xmission.com>2007-11-28 19:21:26 -0500
committerLinus Torvalds <torvalds@woody.linux-foundation.org>2007-11-29 12:24:52 -0500
commit19fd4bb2a0cfede054e4904e0b167e0ca4f36cc7 (patch)
treee37d0ee2c7ea08f6d8721a4e9e807f85994b608a /fs
parentc895078355b6b6e05c60aa205892526dd3390f0a (diff)
proc: remove races from proc_id_readdir()
Oleg noticed that the call of task_pid_nr_ns() in proc_pid_readdir is racy with respect to tasks exiting. After a bit of examination it also appears that the call itself is completely unnecessary. So to fix the problem this patch modifies next_tgid() to return both a tgid and the task struct in question. A structure is introduced to return these values because it is slightly cleaner and easier to optimize, and the resulting code is a little shorter. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Oleg Nesterov <oleg@tv-sign.ru> Cc: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'fs')
-rw-r--r--fs/proc/base.c51
1 files changed, 28 insertions, 23 deletions
diff --git a/fs/proc/base.c b/fs/proc/base.c
index a17c26859074..02a63ac04178 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -2411,19 +2411,23 @@ out:
2411 * Find the first task with tgid >= tgid 2411 * Find the first task with tgid >= tgid
2412 * 2412 *
2413 */ 2413 */
2414static struct task_struct *next_tgid(unsigned int tgid, 2414struct tgid_iter {
2415 struct pid_namespace *ns) 2415 unsigned int tgid;
2416{
2417 struct task_struct *task; 2416 struct task_struct *task;
2417};
2418static struct tgid_iter next_tgid(struct pid_namespace *ns, struct tgid_iter iter)
2419{
2418 struct pid *pid; 2420 struct pid *pid;
2419 2421
2422 if (iter.task)
2423 put_task_struct(iter.task);
2420 rcu_read_lock(); 2424 rcu_read_lock();
2421retry: 2425retry:
2422 task = NULL; 2426 iter.task = NULL;
2423 pid = find_ge_pid(tgid, ns); 2427 pid = find_ge_pid(iter.tgid, ns);
2424 if (pid) { 2428 if (pid) {
2425 tgid = pid_nr_ns(pid, ns) + 1; 2429 iter.tgid = pid_nr_ns(pid, ns);
2426 task = pid_task(pid, PIDTYPE_PID); 2430 iter.task = pid_task(pid, PIDTYPE_PID);
2427 /* What we to know is if the pid we have find is the 2431 /* What we to know is if the pid we have find is the
2428 * pid of a thread_group_leader. Testing for task 2432 * pid of a thread_group_leader. Testing for task
2429 * being a thread_group_leader is the obvious thing 2433 * being a thread_group_leader is the obvious thing
@@ -2436,23 +2440,25 @@ retry:
2436 * found doesn't happen to be a thread group leader. 2440 * found doesn't happen to be a thread group leader.
2437 * As we don't care in the case of readdir. 2441 * As we don't care in the case of readdir.
2438 */ 2442 */
2439 if (!task || !has_group_leader_pid(task)) 2443 if (!iter.task || !has_group_leader_pid(iter.task)) {
2444 iter.tgid += 1;
2440 goto retry; 2445 goto retry;
2441 get_task_struct(task); 2446 }
2447 get_task_struct(iter.task);
2442 } 2448 }
2443 rcu_read_unlock(); 2449 rcu_read_unlock();
2444 return task; 2450 return iter;
2445} 2451}
2446 2452
2447#define TGID_OFFSET (FIRST_PROCESS_ENTRY + ARRAY_SIZE(proc_base_stuff)) 2453#define TGID_OFFSET (FIRST_PROCESS_ENTRY + ARRAY_SIZE(proc_base_stuff))
2448 2454
2449static int proc_pid_fill_cache(struct file *filp, void *dirent, filldir_t filldir, 2455static int proc_pid_fill_cache(struct file *filp, void *dirent, filldir_t filldir,
2450 struct task_struct *task, int tgid) 2456 struct tgid_iter iter)
2451{ 2457{
2452 char name[PROC_NUMBUF]; 2458 char name[PROC_NUMBUF];
2453 int len = snprintf(name, sizeof(name), "%d", tgid); 2459 int len = snprintf(name, sizeof(name), "%d", iter.tgid);
2454 return proc_fill_cache(filp, dirent, filldir, name, len, 2460 return proc_fill_cache(filp, dirent, filldir, name, len,
2455 proc_pid_instantiate, task, NULL); 2461 proc_pid_instantiate, iter.task, NULL);
2456} 2462}
2457 2463
2458/* for the /proc/ directory itself, after non-process stuff has been done */ 2464/* for the /proc/ directory itself, after non-process stuff has been done */
@@ -2460,8 +2466,7 @@ int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir)
2460{ 2466{
2461 unsigned int nr = filp->f_pos - FIRST_PROCESS_ENTRY; 2467 unsigned int nr = filp->f_pos - FIRST_PROCESS_ENTRY;
2462 struct task_struct *reaper = get_proc_task(filp->f_path.dentry->d_inode); 2468 struct task_struct *reaper = get_proc_task(filp->f_path.dentry->d_inode);
2463 struct task_struct *task; 2469 struct tgid_iter iter;
2464 int tgid;
2465 struct pid_namespace *ns; 2470 struct pid_namespace *ns;
2466 2471
2467 if (!reaper) 2472 if (!reaper)
@@ -2474,14 +2479,14 @@ int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir)
2474 } 2479 }
2475 2480
2476 ns = filp->f_dentry->d_sb->s_fs_info; 2481 ns = filp->f_dentry->d_sb->s_fs_info;
2477 tgid = filp->f_pos - TGID_OFFSET; 2482 iter.task = NULL;
2478 for (task = next_tgid(tgid, ns); 2483 iter.tgid = filp->f_pos - TGID_OFFSET;
2479 task; 2484 for (iter = next_tgid(ns, iter);
2480 put_task_struct(task), task = next_tgid(tgid + 1, ns)) { 2485 iter.task;
2481 tgid = task_pid_nr_ns(task, ns); 2486 iter.tgid += 1, iter = next_tgid(ns, iter)) {
2482 filp->f_pos = tgid + TGID_OFFSET; 2487 filp->f_pos = iter.tgid + TGID_OFFSET;
2483 if (proc_pid_fill_cache(filp, dirent, filldir, task, tgid) < 0) { 2488 if (proc_pid_fill_cache(filp, dirent, filldir, iter) < 0) {
2484 put_task_struct(task); 2489 put_task_struct(iter.task);
2485 goto out; 2490 goto out;
2486 } 2491 }
2487 } 2492 }