4.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kirill Tkhai <ktkhai@xxxxxxxxxxxxx>
[ Upstream commit 8896c23d2ef803f1883fea73117a435925c2b4c4 ]
alloc_pidmap() advances pid_namespace::last_pid. When first pid
allocation fails, then next created process will have pid 2 and
pid_ns_prepare_proc() won't be called. So, pid_namespace::proc_mnt will
never be initialized (not to mention that there won't be a child
reaper).
I saw crash stack of such case on kernel 3.10:
BUG: unable to handle kernel NULL pointer dereference at (null)
IP: proc_flush_task+0x8f/0x1b0
Call Trace:
release_task+0x3f/0x490
wait_consider_task.part.10+0x7ff/0xb00
do_wait+0x11f/0x280
SyS_wait4+0x7d/0x110
We may fix this by restore of last_pid in 0 or by prohibiting of futher
allocations. Since there was a similar issue in Oleg Nesterov's commit
314a8ad0f18a ("pidns: fix free_pid() to handle the first fork failure").
and it was fixed via prohibiting allocation, let's follow this way, and
do the same.
Link: http://lkml.kernel.org/r/149201021004.4863.6762095011554287922.stgit@localhost.localdomain
Signed-off-by: Kirill Tkhai <ktkhai@xxxxxxxxxxxxx>
Acked-by: Cyrill Gorcunov <gorcunov@xxxxxxxxxx>
Cc: Andrei Vagin <avagin@xxxxxxxxxxxxx>
Cc: Andreas Gruenbacher <agruenba@xxxxxxxxxx>
Cc: Kees Cook <keescook@xxxxxxxxxxxx>
Cc: Michael Kerrisk <mtk.manpages@xxxxxxxxxxxxxx>
Cc: Al Viro <viro@xxxxxxxxxxxxxxxxxx>
Cc: Oleg Nesterov <oleg@xxxxxxxxxx>
Cc: Paul Moore <paul@xxxxxxxxxxxxxx>
Cc: Eric Biederman <ebiederm@xxxxxxxxxxxx>
Cc: Andy Lutomirski <luto@xxxxxxxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxxxxx>
Cc: Serge Hallyn <serge@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Sasha Levin <alexander.levin@xxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
kernel/pid.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
--- a/kernel/pid.c
+++ b/kernel/pid.c
@@ -322,8 +322,10 @@ struct pid *alloc_pid(struct pid_namespa
}
if (unlikely(is_child_reaper(pid))) {
- if (pid_ns_prepare_proc(ns))
+ if (pid_ns_prepare_proc(ns)) {
+ disable_pid_allocation(ns);
goto out_free;
+ }
}
get_pid_ns(ns);