From 5234ffb9f74cfa8993d174782bc861dd9b7b5bfb Mon Sep 17 00:00:00 2001 From: Alexander Gordeev Date: Fri, 9 Mar 2012 14:59:59 +0100 Subject: genirq: Get rid of unnecessary IRQTF_DIED flag Currently IRQTF_DIED flag is set when a IRQ thread handler calls do_exit() But also PF_EXITING per process flag gets set when a thread exits. This fix eliminates the duplicate by using PF_EXITING flag. Also, there is a race condition in exit_irq_thread(). In case a thread's bit is cleared in desc->threads_oneshot (and the IRQ line gets unmasked), but before IRQTF_DIED flag is set, a new interrupt might come in and set just cleared bit again, this time forever. This fix throws IRQTF_DIED flag away, eliminating the race as a result. [ tglx: Test THREAD_EXITING first as suggested by Oleg ] Reported-by: Oleg Nesterov Signed-off-by: Alexander Gordeev Link: http://lkml.kernel.org/r/20120309135958.GD2114@dhcp-26-207.brq.redhat.com Signed-off-by: Thomas Gleixner --- kernel/exit.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/exit.c') diff --git a/kernel/exit.c b/kernel/exit.c index 4b4042f9bc6..752d2c0abd1 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -935,8 +935,6 @@ void do_exit(long code) schedule(); } - exit_irq_thread(); - exit_signals(tsk); /* sets PF_EXITING */ /* * tsk->flags are checked in the futex code to protect against @@ -945,6 +943,8 @@ void do_exit(long code) smp_mb(); raw_spin_unlock_wait(&tsk->pi_lock); + exit_irq_thread(); + if (unlikely(in_atomic())) printk(KERN_INFO "note: %s[%d] exited with preempt_count %d\n", current->comm, task_pid_nr(current), -- cgit v1.2.3-70-g09d2 From e636825346b36a07ccfc8e30946d52855e21f681 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Mon, 19 Mar 2012 17:03:22 +0100 Subject: exit_signal: simplify the "we have changed execution domain" logic exit_notify() checks "tsk->self_exec_id != tsk->parent_exec_id" to handle the "we have changed execution domain" case. We can change do_thread() to always set ->exit_signal = SIGCHLD and remove this check to simplify the code. We could change setup_new_exec() instead, this looks more logical because it increments ->self_exec_id. But note that de_thread() already resets ->exit_signal if it changes the leader, let's keep both changes close to each other. Note that we change ->exit_signal lockless, this changes the rules. Thereafter ->exit_signal is not stable under tasklist but this is fine, the only possible change is OLDSIG -> SIGCHLD. This can race with eligible_child() but the race is harmless. We can race with reparent_leader() which changes our ->exit_signal in parallel, but it does the same change to SIGCHLD. The noticeable user-visible change is that the execing task is not "visible" to do_wait()->eligible_child(__WCLONE) right after exec. To me this looks more logical, and this is consistent with mt case. Signed-off-by: Oleg Nesterov Signed-off-by: Linus Torvalds --- fs/exec.c | 3 +++ kernel/exit.c | 7 +------ 2 files changed, 4 insertions(+), 6 deletions(-) (limited to 'kernel/exit.c') diff --git a/fs/exec.c b/fs/exec.c index b0695a9900e..1e94d2263ae 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -977,6 +977,9 @@ static int de_thread(struct task_struct *tsk) sig->notify_count = 0; no_thread_group: + /* we have changed execution domain */ + tsk->exit_signal = SIGCHLD; + if (current->mm) setmax_mm_hiwater_rss(&sig->maxrss, current->mm); diff --git a/kernel/exit.c b/kernel/exit.c index 752d2c0abd1..51ac4ced131 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -827,14 +827,9 @@ static void exit_notify(struct task_struct *tsk, int group_dead) * If the parent exec id doesn't match the exec id we saved * when we started then we know the parent has changed security * domain. - * - * If our self_exec id doesn't match our parent_exec_id then - * we have changed execution domain as these two values started - * the same after a fork. */ if (thread_group_leader(tsk) && tsk->exit_signal != SIGCHLD && - (tsk->parent_exec_id != tsk->real_parent->self_exec_id || - tsk->self_exec_id != tsk->parent_exec_id)) + tsk->parent_exec_id != tsk->real_parent->self_exec_id) tsk->exit_signal = SIGCHLD; if (unlikely(tsk->ptrace)) { -- cgit v1.2.3-70-g09d2 From b6e238dceed36891cc633167afe7151f1f3d83c5 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Mon, 19 Mar 2012 17:03:41 +0100 Subject: exit_signal: fix the "parent has changed security domain" logic exit_notify() changes ->exit_signal if the parent already did exec. This doesn't really work, we are not going to send the signal now if there is another live thread or the exiting task is traced. The parent can exec before the last dies or the tracer detaches. Move this check into do_notify_parent() which actually sends the signal. The user-visible change is that we do not change ->exit_signal, and thus the exiting task is still "clone children" for do_wait()->eligible_child(__WCLONE). Hopefully this is fine, the current logic is racy anyway. Signed-off-by: Oleg Nesterov Signed-off-by: Linus Torvalds --- kernel/exit.c | 14 -------------- kernel/signal.c | 9 +++++++++ 2 files changed, 9 insertions(+), 14 deletions(-) (limited to 'kernel/exit.c') diff --git a/kernel/exit.c b/kernel/exit.c index 51ac4ced131..ce5f758f40b 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -818,20 +818,6 @@ static void exit_notify(struct task_struct *tsk, int group_dead) if (group_dead) kill_orphaned_pgrp(tsk->group_leader, NULL); - /* Let father know we died - * - * Thread signals are configurable, but you aren't going to use - * that to send signals to arbitrary processes. - * That stops right now. - * - * If the parent exec id doesn't match the exec id we saved - * when we started then we know the parent has changed security - * domain. - */ - if (thread_group_leader(tsk) && tsk->exit_signal != SIGCHLD && - tsk->parent_exec_id != tsk->real_parent->self_exec_id) - tsk->exit_signal = SIGCHLD; - if (unlikely(tsk->ptrace)) { int sig = thread_group_leader(tsk) && thread_group_empty(tsk) && diff --git a/kernel/signal.c b/kernel/signal.c index 8511e39813c..e76001ccf5c 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1652,6 +1652,15 @@ bool do_notify_parent(struct task_struct *tsk, int sig) BUG_ON(!tsk->ptrace && (tsk->group_leader != tsk || !thread_group_empty(tsk))); + if (sig != SIGCHLD) { + /* + * This is only possible if parent == real_parent. + * Check if it has changed security domain. + */ + if (tsk->parent_exec_id != tsk->parent->self_exec_id) + sig = SIGCHLD; + } + info.si_signo = sig; info.si_errno = 0; /* -- cgit v1.2.3-70-g09d2