Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Message-ID: <94bff1bedd0dfa957822a6a303b48eca787f9a21.camel@linux.ibm.com>
Subject: Re: [PATCH][next] scsi: aacraid: Replace one-element array with
 flexible-array member
From:   James Bottomley <jejb@linux.ibm.com>
Reply-To: jejb@linux.ibm.com
To:     "Gustavo A. R. Silva" <gustavo@embeddedor.com>,
        "Martin K. Petersen" <martin.petersen@oracle.com>,
        Kees Cook <keescook@chromium.org>
Cc:     "Gustavo A. R. Silva" <gustavoars@kernel.org>,
        Adaptec OEM Raid Solutions <aacraid@microsemi.com>,
        linux-scsi@vger.kernel.org, linux-kernel@vger.kernel.org,
        linux-hardening@vger.kernel.org
Date:   Tue, 13 Apr 2021 07:04:11 -0700
In-Reply-To: <aba7d5cb-79be-088d-d1f8-9309109e9afc@embeddedor.com>
References: <20210304203822.GA102218@embeddedor>
         <202104071216.5BEA350@keescook> <yq1h7ka7q68.fsf@ca-mkp.ca.oracle.com>
         <aba7d5cb-79be-088d-d1f8-9309109e9afc@embeddedor.com>
Content-Type: text/plain; charset="UTF-8"
User-Agent: Evolution 3.34.4 
Content-Transfer-Encoding: 7bit
MIME-Version: 1.0
Precedence: bulk

On Tue, 2021-04-13 at 00:45 -0500, Gustavo A. R. Silva wrote:
> Hi Martin,
> 
> On 4/12/21 23:52, Martin K. Petersen wrote:
> 
> > Silencing analyzer warnings shouldn't be done at the expense of
> > human
> > readers. If it is imperative to switch to flex_array_size() to
> > quiesce
> > checker warnings, please add a comment in the code explaining that
> > the
> > size evaluates to nseg_new-1 sge_ieee1212 structs.
> 
> Done:
> 	
> https://lore.kernel.org/lkml/20210413054032.GA276102@embeddedor/

I think the reason everyone gets confused is that they think the first
argument should do something.  If flex_array_size had been defined

#define flex_array_size(p, count)			\
	array_size(count,				\
		    sizeof(*(p)) + __must_be_array(p))

Then we could have used

flex_array_size(sge, nseg_new - 1)

or

flex_array_size(rio->sge, nseg_new - 1)

and everyone would have understood either expression.  This would also
have been useful, as the first example demonstrates, when we have a
pointer rather than a flexible member ... although that means the macro
likely needs a new name.

However, perhaps just do

array_size(nseg_new - 1, sizeof(*sge));

And lose the comment?

James