Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753048AbXKVR4S (ORCPT ); Thu, 22 Nov 2007 12:56:18 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751960AbXKVR4H (ORCPT ); Thu, 22 Nov 2007 12:56:07 -0500 Received: from x346.tv-sign.ru ([89.108.83.215]:50312 "EHLO mail.screens.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751955AbXKVR4G (ORCPT ); Thu, 22 Nov 2007 12:56:06 -0500 Date: Thu, 22 Nov 2007 20:56:22 +0300 From: Oleg Nesterov To: Andrew Morton Cc: Roland McGrath , Scott James Remnant , linux-kernel@vger.kernel.org Subject: [PATCH -mm 4/2] ptrace_stop: fix racy nonstop_code setting Message-ID: <20071122175622.GA7051@tv-sign.ru> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.11 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2518 Lines: 64 If the tracer is gone and we are not going to stop, ptrace_stop() sets ->exit_code = nostop_code. However, the tracer could actually clear the exit code before detaching. In that case get_signal_to_deliver() "resends" the signal which was cancelled by the debugger. For example, it is possible that a quick PTRACE_ATTACH + PTRACE_DETACH can leave the tracee in STOPPED state. Change the behaviour of ptrace_stop(). If the caller is ptrace notify(), we should always clear ->exit_code. If the caller is get_signal_to_deliver(), we should not touch it at all. To do so, change the nonstop_code parameter to "bool clear_code" and change the callers accordingly. Signed-off-by: Oleg Nesterov --- PT/kernel/signal.c~4_non_stop 2007-11-22 16:59:35.000000000 +0300 +++ PT/kernel/signal.c 2007-11-22 20:29:11.000000000 +0300 @@ -1601,10 +1601,10 @@ static inline int may_ptrace_stop(void) * That makes it a way to test a stopped process for * being ptrace-stopped vs being job-control-stopped. * - * If we actually decide not to stop at all because the tracer is gone, - * we leave nostop_code in current->exit_code. + * If we actually decide not to stop at all because the tracer + * is gone, we keep current->exit_code unless clear_code. */ -static void ptrace_stop(int exit_code, int nostop_code, siginfo_t *info) +static void ptrace_stop(int exit_code, int clear_code, siginfo_t *info) { /* * If there is a group stop in progress, @@ -1631,7 +1631,8 @@ static void ptrace_stop(int exit_code, i * Don't drop the lock yet, another tracer may come. */ __set_current_state(TASK_RUNNING); - current->exit_code = nostop_code; + if (clear_code) + current->exit_code = 0; read_unlock(&tasklist_lock); } @@ -1665,7 +1666,7 @@ void ptrace_notify(int exit_code) /* Let the debugger run. */ spin_lock_irq(¤t->sighand->siglock); - ptrace_stop(exit_code, 0, &info); + ptrace_stop(exit_code, 1, &info); spin_unlock_irq(¤t->sighand->siglock); } @@ -1813,7 +1814,7 @@ relock: ptrace_signal_deliver(regs, cookie); /* Let the debugger run. */ - ptrace_stop(signr, signr, info); + ptrace_stop(signr, 0, info); /* We're back. Did the debugger cancel the sig? */ signr = current->exit_code; - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/