From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner+w=401wt.eu-S1751408AbXCEBkP@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1751408AbXCEBkP (ORCPT <rfc822;w@1wt.eu>);
	Sun, 4 Mar 2007 20:40:15 -0500
Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751418AbXCEBkO
	(ORCPT <rfc822;linux-kernel-outgoing>);
	Sun, 4 Mar 2007 20:40:14 -0500
Received: from nwd2mail10.analog.com ([137.71.25.55]:34875 "EHLO
	nwd2mail10.analog.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1751408AbXCEBkM (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Sun, 4 Mar 2007 20:40:12 -0500
X-IronPort-AV: i="4.14,247,1170651600"; 
   d="scan'208"; a="29902408:sNHT26174204"
Subject: Re: Questions about the SYSVIPC share memory on NOMMU
	uClinuxarchitecture
From: "Wu, Bryan" <bryan.wu@analog.com>
Reply-To: bryan.wu@analog.com
To: bryan.wu@analog.com, ebiederm@xmission.com
Cc: linux-kernel@vger.kernel.org
In-Reply-To: <1172831630.5264.139.camel@roc-desktop>
References: <1172831630.5264.139.camel@roc-desktop>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Organization: Analog Devices, Inc.
Date: Mon, 05 Mar 2007 09:40:01 +0800
Message-Id: <1173058801.5264.143.camel@roc-desktop>
Mime-Version: 1.0
X-Mailer: Evolution 2.6.1 
Sender: linux-kernel-owner@vger.kernel.org
X-Mailing-List: linux-kernel@vger.kernel.org

On Fri, 2007-03-02 at 05:33 -0500, Wu, Bryan wrote:
> Hi folks,
> 
> Recently, I was struggling in a bug about the shm->nattch. Actually,
> the 
> test case is from LTP kernel/syscall/ipc/shmctl/shmctl01.c code. We 
> ported it to the uClinux-blackfin platform.
> 

Sorry for dropping the kernel version inforamtion, I found this in
2.6.19 kernel and 2.6.20-mm2 kernel.

> The algorithm is very simple.  
> a) the parent process will create a share memory  
> b) parent will vfork/execlp 4 children process 
> c) children will call shmat() attach to the share memory (shm->nattch 
> should be increased), then children will pause() 
> d) parent call shmclt() to get the share memory nattch, if nattch !=
> 4, 
> then the testcase will fail.
> 
> In our uClinux-blackfin platform, nattch = 1.
> 
> So I dig into the source code ipc/shm.c, then there are some
> questions 
> about the code.
> 
> a) 
> in function do_shmat(), after nattch++ why nattch-- as following: 
> ================================================================================ 
>  user_addr = (void*) do_mmap (file, addr, size, prot, flags, 0);
> 
>         // here no return or goto valid place.
> 
> invalid: 
>         up_write(&current->mm->mmap_sem);
> 
>         mutex_lock(&shm_ids(ns).mutex); 
>         shp = shm_lock(ns, shmid); 
>         BUG_ON(!shp); 
>         shp->shm_nattch--; /* Why??? */ 
>         if(shp->shm_nattch == 0 && 
>            shp->shm_perm.mode & SHM_DEST) 
>                 shm_destroy(ns, shp); 
>         else 
>                 shm_unlock(shp); 
>         mutex_unlock(&shm_ids(ns).mutex);
> 
>         *raddr = (unsigned long) user_addr; 
>         err = 0; 
>         if (IS_ERR(user_addr)) 
>                 err = PTR_ERR(user_addr); 
> out: 
>         return err; 
> ================================================================================
> 
> b) do_mmap() -> mm/nommu.c do_mmap_pgoff() 
> When create a new vma structure, shm_open(), shm_inc() will be called.
> Then nattch++. 
> So the nattch counting is disordered.
> 
> Please give me some hint about it. Actually, this case can pass on X86
> platform
> 

Is there any help available? 
Thanks a lot
-Bryan