Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Fri, 3 Aug 2018 11:52:45 +0200 (CEST)
From:   Julia Lawall <julia.lawall@lip6.fr>
To:     Andy Shevchenko <andy.shevchenko@gmail.com>
cc:     Finn Thain <fthain@telegraphics.com.au>,
        zhong jiang <zhongjiang@huawei.com>,
        Michael Schmitz <schmitzmic@gmail.com>,
        "James E . J . Bottomley" <jejb@linux.vnet.ibm.com>,
        "Martin K. Petersen" <martin.petersen@oracle.com>,
        John Garry <john.garry@huawei.com>,
        linux-scsi <linux-scsi@vger.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] scsi:NCR5380: remove same check condition in
 NCR5380_select
In-Reply-To: <CAHp75Vc8tWKUpKS8yY52CdJ=mU-iqcrUCz32ZSSJixLA1bM0tQ@mail.gmail.com>
Message-ID: <alpine.DEB.2.20.1808031146090.3871@hadrien>
References: <1533179408-20631-1-git-send-email-zhongjiang@huawei.com> <alpine.LNX.2.21.1808031156350.16@nippy.intranet> <CAHp75Vc8tWKUpKS8yY52CdJ=mU-iqcrUCz32ZSSJixLA1bM0tQ@mail.gmail.com>
User-Agent: Alpine 2.20 (DEB 67 2015-01-07)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On Fri, 3 Aug 2018, Andy Shevchenko wrote:

> On Fri, Aug 3, 2018 at 5:24 AM, Finn Thain <fthain@telegraphics.com.au> wrote:
> > On Thu, 2 Aug 2018, zhong jiang wrote:
> >
> >> The same check condition is redundant, so remove one of them.
> >>
> >
> > If you are trying to find redundant code, your coccinelle script is
> > dangerously flawed.
>
> These days too many coccinelle helpers make people think they are
> doing right "clean ups" when in the practice they bring the
> regressions.
>
> Julia, is possible by coccinelle to distinguish memory accesses versus
> I/O? At least it would increase robustness in some cases.

With make coccicheck, the semantic patch should already emit the warning:

//# A common source of false positives is when the argument performs a side
//# effect.

I can modify the rule so that it doesn't report on code that involves
function calls.  It could lose some desirable warnings, where the function
call is just a wrapper for eg extracting some field, but it is probably
safer in practice.

julia