From owner-freebsd-drivers@freebsd.org  Sun Jun 18 10:11:10 2017
Return-Path: <owner-freebsd-drivers@freebsd.org>
Delivered-To: freebsd-drivers@mailman.ysv.freebsd.org
Received: from mx1.freebsd.org (mx1.freebsd.org
 [IPv6:2001:1900:2254:206a::19:1])
 by mailman.ysv.freebsd.org (Postfix) with ESMTP id 96D89D8DC48
 for <freebsd-drivers@mailman.ysv.freebsd.org>;
 Sun, 18 Jun 2017 10:11:10 +0000 (UTC) (envelope-from njm@njm.me.uk)
Received: from smtp002.apm-internet.net (smtp002.apm-internet.net
 [85.119.248.221])
 (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
 (Client did not present a certificate)
 by mx1.freebsd.org (Postfix) with ESMTPS id F2E65669EE
 for <freebsd-drivers@freebsd.org>; Sun, 18 Jun 2017 10:11:09 +0000 (UTC)
 (envelope-from njm@njm.me.uk)
Received: (qmail 80154 invoked from network); 18 Jun 2017 10:04:26 -0000
X-APM-Authkey: 18389/1 2
Received: from unknown (HELO meld.njm.me.uk) (90.194.167.105)
 by smtp002.apm-internet.net with SMTP; 18 Jun 2017 10:04:26 -0000
Received: from triton.njm.me.uk (triton.njm.me.uk [192.168.144.133])
 by meld.njm.me.uk (8.15.2/8.15.2) with ESMTP id v5IA4P3c002078;
 Sun, 18 Jun 2017 11:04:26 +0100 (BST) (envelope-from njm@njm.me.uk)
Received: from localhost (localhost [127.0.0.1])
 by triton.njm.me.uk (8.15.2/8.15.2) with ESMTP id v5IA4P6b072619;
 Sun, 18 Jun 2017 11:04:25 +0100 (BST) (envelope-from njm@njm.me.uk)
Date: Sun, 18 Jun 2017 11:04:25 +0100
From: "N.J. Mann" <njm@njm.me.uk>
To: Jia-Ju Bai <baijiaju1990@163.com>
cc: freebsd-drivers@freebsd.org
Subject: Re: [Bug 220095][PATCH] dpt_scsi: Fix a possible sleep-under-mutex
 bug in dpt_init
Message-ID: <7D8CD40C6A8CDF922B1E15A8@triton.njm.me.uk>
In-Reply-To: <20170618094601.40636-1-baijiaju1990@163.com>
References: <20170618094601.40636-1-baijiaju1990@163.com>
X-Mailer: Mulberry/4.0.8 (Linux/x86)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
X-BeenThere: freebsd-drivers@freebsd.org
X-Mailman-Version: 2.1.23
Precedence: list
List-Id: Writing device drivers for FreeBSD <freebsd-drivers.freebsd.org>
List-Unsubscribe: <https://lists.freebsd.org/mailman/options/freebsd-drivers>, 
 <mailto:freebsd-drivers-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-drivers/>
List-Post: <mailto:freebsd-drivers@freebsd.org>
List-Help: <mailto:freebsd-drivers-request@freebsd.org?subject=help>
List-Subscribe: <https://lists.freebsd.org/mailman/listinfo/freebsd-drivers>, 
 <mailto:freebsd-drivers-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Sun, 18 Jun 2017 10:11:10 -0000

Hi,


On Sunday, June 18, 2017 17:46:01 +0800 Jia-Ju Bai <baijiaju1990@163.com> wrote:
> The driver may sleep under a mutex, and the function call path is:
> dpt_init [acquire the mutex]
>    dptallocsgmap
>      bus_dmamap_load(BUS_DMA_WAITOK) --> may sleep
> 
> The possible fix of this bug is to set the last parameter in 
> bus_dmamap_load to "BUS_DMA_NOWAIT".

Have you read the manual page for bus_dmamap_load() ?

    bus_dmamap_load(dmat, map, buf, buflen, *callback, callback_arg, flags)
            Creates a mapping in device visible address space of buflen bytes
            of buf, associated with the DMA map map.  This call will always
            return immediately and will not block for any reason.

I am not saying that your change should not be made, just that your analysis
may be wrong.


Cheers,
       Nick.
--