From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Message-ID: <4F38DA79.2040803@suse.cz> Date: Mon, 13 Feb 2012 10:40:09 +0100 From: Jiri Slaby MIME-Version: 1.0 References: <1329006070-4275-1-git-send-email-richard@nod.at> <20120212010211.GN23916@ZenIV.linux.org.uk> <4F37B34F.1040307@suse.cz> <20120212190611.GP23916@ZenIV.linux.org.uk> In-Reply-To: <20120212190611.GP23916@ZenIV.linux.org.uk> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org Subject: Re: your mail To: Al Viro Cc: Jiri Slaby , Richard Weinberger , linux-kernel@vger.kernel.org, user-mode-linux-devel@lists.sourceforge.net, akpm@linux-foundation.org, alan@lxorguk.ukuu.org.uk, gregkh@linuxfoundation.org List-ID: On 02/12/2012 08:06 PM, Al Viro wrote: > Yecchhh... If I'm reading (and grepping) it right, there are only two > non-default instance of tty_operations ->shutdown() - pty and vt ones. > Lovely... And while we are at it, vt instance is definitely not safe > from interrupts - calls console_lock(). Not that it was relevant in > this case... Thanks for looking into that. I was too lazy to do that on Sunday. You're right that it may cause problems. Fortunately vt doesn't refcount ttys. Hence con_shutdown can be called only from release_tty (close path) in the user context. Adding to my TODO list, unless somebody beats me to fix it. thanks, -- js suse labs