lkml.org 
[lkml]   [2014]   [Nov]   [24]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
Subject[PATCH 2/2] exit: pidns: alloc_pid() leaks pid_namespace if child_reaper is exiting
alloc_pid() does get_pid_ns() beforehand but forgets to put_pid_ns()
if it fails because disable_pid_allocation() was called by the exiting
child_reaper. We can move get_pid_ns() down to successful return.

While at it, simplify/cleanup the "goto out" mess, no need to confuse
the success/error return paths.

Signed-off-by: Oleg Nesterov <oleg@redhat.com>
---
kernel/pid.c | 7 +++----
1 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/kernel/pid.c b/kernel/pid.c
index 9b9a266..dfc2f3b 100644
--- a/kernel/pid.c
+++ b/kernel/pid.c
@@ -320,7 +320,6 @@ struct pid *alloc_pid(struct pid_namespace *ns)
goto out_free;
}

- get_pid_ns(ns);
atomic_set(&pid->count, 1);
for (type = 0; type < PIDTYPE_MAX; ++type)
INIT_HLIST_HEAD(&pid->tasks[type]);
@@ -336,7 +335,7 @@ struct pid *alloc_pid(struct pid_namespace *ns)
}
spin_unlock_irq(&pidmap_lock);

-out:
+ get_pid_ns(ns);
return pid;

out_unlock:
@@ -346,8 +345,8 @@ out_free:
free_pidmap(pid->numbers + i);

kmem_cache_free(ns->pid_cachep, pid);
- pid = NULL;
- goto out;
+out:
+ return NULL;
}

void disable_pid_allocation(struct pid_namespace *ns)
--
1.5.5.1


\
 
 \ /
  Last update: 2014-11-24 21:41    [W:0.121 / U:3.088 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site