From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932953AbXDAMn0 (ORCPT ); Sun, 1 Apr 2007 08:43:26 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932969AbXDAMn0 (ORCPT ); Sun, 1 Apr 2007 08:43:26 -0400 Received: from ipn26-148.piekary.net ([83.238.26.148]:42989 "EHLO ipn26-148.piekary.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932953AbXDAMn0 (ORCPT ); Sun, 1 Apr 2007 08:43:26 -0400 Date: Sun, 1 Apr 2007 14:43:23 +0200 From: Michal Januszewski To: linux-kernel@vger.kernel.org Cc: Andrew Morton , "Antonino A. Daplas" , linux-fbdev-devel@lists.sourceforge.net Subject: [PATCH resend] vt: fix potential race in VT_WAITACTIVE handler Message-ID: <20070401124323.GA4040@spock.one.pl> Reply-To: spock@gentoo.org MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Disposition: inline X-PGP-Key: http://dev.gentoo.org/~spock/spock.gpg User-Agent: Mutt/1.5.14 (2007-02-12) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org From: Michal Januszewski On a multiprocessor machine the VT_WAITACTIVE ioctl call may return 0 if fg_console has already been updated in redraw_screen() but the console switch itself hasn't been completed. Fix this by checking fg_console in vt_waitactive() with the console sem held. Signed-off-by: Michal Januszewski --- This is the 2nd version of this patch. It incorporates Andrew's suggestions, ie. calls set_current_state() after down() and adds a comment explaining why acquiring the console sem is necessary. drivers/char/vt_ioctl.c | 14 ++++++++++++-- 1 files changed, 12 insertions(+), 2 deletions(-) diff --git a/drivers/char/vt_ioctl.c b/drivers/char/vt_ioctl.c index 1fa2da8..0508293 100644 --- a/drivers/char/vt_ioctl.c +++ b/drivers/char/vt_ioctl.c @@ -1039,10 +1039,20 @@ int vt_waitactive(int vt) add_wait_queue(&vt_activate_queue, &wait); for (;;) { - set_current_state(TASK_INTERRUPTIBLE); retval = 0; - if (vt == fg_console) + + /* Synchronize with redraw_screen(). By acquiring the console + * semaphore we make sure that the console switch is completed + * before we return. If we didn't wait for the semaphore, we + * could return at a point where fg_console has already been + * updated, but the console switch hasn't been completed. */ + acquire_console_sem(); + set_current_state(TASK_INTERRUPTIBLE); + if (vt == fg_console) { + release_console_sem(); break; + } + release_console_sem(); retval = -EINTR; if (signal_pending(current)) break;