linux-kernel - [PATCH 4/8] coredump_wait: don't call complete_vfork

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20110727163311.GD23793@redhat.com>
Date:	Wed, 27 Jul 2011 18:33:11 +0200
From:	Oleg Nesterov <oleg@...hat.com>
To:	Linus Torvalds <torvalds@...ux-foundation.org>,
	Roland McGrath <roland@...k.frob.com>,
	Tejun Heo <tj@...nel.org>
Cc:	Denys Vlasenko <dvlasenk@...hat.com>,
	KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>,
	Matt Fleming <matt.fleming@...ux.intel.com>,
	linux-kernel@...r.kernel.org
Subject: [PATCH 4/8] coredump_wait: don't call complete_vfork_done()

Now that CLONE_VFORK is killable, coredump_wait() no longer needs
complete_vfork_done(). zap_threads() should find and kill all tasks
with the same ->mm, this includes our parent if ->vfork_done is set.

mm_release() becomes the only caller, unexport complete_vfork_done().

Signed-off-by: Oleg Nesterov <oleg@...hat.com>
---

 include/linux/sched.h |    1 -
 fs/exec.c             |    7 -------
 kernel/fork.c         |    3 ++-
 3 files changed, 2 insertions(+), 9 deletions(-)

--- 3.1/include/linux/sched.h~4_coredump_no_vfork_done	2011-07-26 20:23:28.000000000 +0200
+++ 3.1/include/linux/sched.h	2011-07-26 20:28:24.000000000 +0200
@@ -2253,7 +2253,6 @@ extern int do_execve(const char *,
 		     const char __user * const __user *,
 		     const char __user * const __user *, struct pt_regs *);
 extern long do_fork(unsigned long, unsigned long, struct pt_regs *, unsigned long, int __user *, int __user *);
-extern void complete_vfork_done(struct task_struct *tsk);
 struct task_struct *fork_idle(int);
 
 extern void set_task_comm(struct task_struct *tsk, char *from);
--- 3.1/fs/exec.c~4_coredump_no_vfork_done	2011-07-26 20:23:28.000000000 +0200
+++ 3.1/fs/exec.c	2011-07-26 20:28:24.000000000 +0200
@@ -1899,13 +1899,6 @@ static int coredump_wait(int exit_code, 
 	if (unlikely(core_waiters < 0))
 		goto fail;
 
-	/*
-	 * Make sure nobody is waiting for us to release the VM,
-	 * otherwise we can deadlock when we wait on each other
-	 */
-	if (tsk->vfork_done)
-		complete_vfork_done(tsk);
-
 	if (core_waiters)
 		wait_for_completion(&core_state->startup);
 fail:
--- 3.1/kernel/fork.c~4_coredump_no_vfork_done	2011-07-26 20:23:28.000000000 +0200
+++ 3.1/kernel/fork.c	2011-07-26 20:28:24.000000000 +0200
@@ -647,6 +647,7 @@ struct mm_struct *get_task_mm(struct tas
 }
 EXPORT_SYMBOL_GPL(get_task_mm);
 
+static void complete_vfork_done(struct task_struct *tsk);
 /* Please note the differences between mmput and mm_release.
  * mmput is called whenever we stop holding onto a mm_struct,
  * error success whatever.
@@ -1442,7 +1443,7 @@ struct task_struct * __cpuinit fork_idle
 	return task;
 }
 
-void complete_vfork_done(struct task_struct *tsk)
+static void complete_vfork_done(struct task_struct *tsk)
 {
 	struct completion *vfork_done = xchg(&tsk->vfork_done, NULL);
 

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/