From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932815Ab1JXOuT (ORCPT ); Mon, 24 Oct 2011 10:50:19 -0400 Received: from 95-31-19-74.broadband.corbina.ru ([95.31.19.74]:45149 "EHLO dnet.ilyx.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932589Ab1JXOuT (ORCPT ); Mon, 24 Oct 2011 10:50:19 -0400 Message-ID: <4EA57B25.1010100@ilyx.ru> Date: Mon, 24 Oct 2011 18:50:13 +0400 From: Ilya Zykov User-Agent: Thunderbird 2.0.0.24 (Windows/20100228) MIME-Version: 1.0 To: Jiri Slaby CC: Greg Kroah-Hartman , Alan Cox , linux-kernel@vger.kernel.org, ilya@ilyx.ru Subject: Re: [PATCH] TTY: pty, fix pty counting References: <4EA48094.8020508@ilyx.ru> <4EA563E6.3040205@suse.cz> In-Reply-To: <4EA563E6.3040205@suse.cz> Content-Type: text/plain; charset=windows-1251 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Jiri Slaby wrote: > On 10/23/2011 11:01 PM, Ilya Zykov wrote: >> New version for commit: 24d406a6bf736f7aebdc8fa0f0ec86e0890c6d24 > > Although it will work, as ptms are not allowed to be reopen, it doesn't > look correct. We should decrement the count in ->remove, because we are > incrementing in install. > > Now, when I understand ptm+devpts layer a bit more, instead of the > current hackish approach introduced by 24d406a6b (TTY: pty, fix pty > counting), I think we may introduce a ->remove hook specific to ptms. In > that one we could decrement the count and don't bother with the > pty_count macros anymore. Right? > > BTW you cannot remove ->remove hook of pty layer. It would cause an OOPS > because driver->ttys is not allocated for ptys. > >> diff -uprN a/drivers/tty/pty.c b/drivers/tty/pty.c >> --- a/drivers/tty/pty.c 2011-05-19 08:06:34.000000000 +0400 >> +++ b/drivers/tty/pty.c 2011-10-23 18:01:20.000000000 +0400 >> @@ -36,13 +36,15 @@ >> static struct tty_driver *ptm_driver; >> static struct tty_driver *pts_driver; >> #endif >> +static int pty_count; >> >> static void pty_close(struct tty_struct *tty, struct file *filp) >> { >> BUG_ON(!tty); >> - if (tty->driver->subtype == PTY_TYPE_MASTER) >> + if (tty->driver->subtype == PTY_TYPE_MASTER) { >> WARN_ON(tty->count > 1); >> - else { >> + pty_count--; >> + } else { >> if (tty->count > 2) >> return; >> } >> @@ -446,7 +448,6 @@ static inline void legacy_pty_init(void) >> int pty_limit = NR_UNIX98_PTY_DEFAULT; >> static int pty_limit_min; >> static int pty_limit_max = NR_UNIX98_PTY_MAX; >> -static int pty_count; >> >> static struct cdev ptmx_cdev; >> >> @@ -599,15 +600,9 @@ free_mem_out: >> return -ENOMEM; >> } >> >> -static void pty_unix98_remove(struct tty_driver *driver, struct tty_struct *tty) >> -{ >> - pty_count--; >> -} >> - >> static const struct tty_operations ptm_unix98_ops = { >> .lookup = ptm_unix98_lookup, >> .install = pty_unix98_install, >> - .remove = pty_unix98_remove, >> .open = pty_open, >> .close = pty_close, >> .write = pty_write, >> @@ -624,7 +619,6 @@ static const struct tty_operations ptm_u >> static const struct tty_operations pty_unix98_ops = { >> .lookup = pts_unix98_lookup, >> .install = pty_unix98_install, >> - .remove = pty_unix98_remove, >> .open = pty_open, >> .close = pty_close, >> .write = pty_write, > > thanks, Invoke tty_driver_remove_tty() from driver, not good idea IMHO.