Date: Tue, 14 Apr 2015 11:00:11 +0300
From: Dan Carpenter <dan.carpenter@oracle.com>
To: Mateusz Kulikowski <mateusz.kulikowski@gmail.com>
Cc: gregkh@linuxfoundation.org, devel@driverdev.osuosl.org,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2 10/21] staging: rtl8192e: replace memcpy() ->
 ether_addr_copy_unaligned()
Message-ID: <20150414080011.GD10964@mwanda>
References: <1428961664-4241-1-git-send-email-mateusz.kulikowski@gmail.com>
 <1428961664-4241-11-git-send-email-mateusz.kulikowski@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1428961664-4241-11-git-send-email-mateusz.kulikowski@gmail.com>
User-Agent: Mutt/1.5.21 (2010-09-15)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1190
Lines: 27

On Mon, Apr 13, 2015 at 11:47:33PM +0200, Mateusz Kulikowski wrote:
> rtl8192e driver uses memcpy() to copy hw addresses in several places.
> checkpatch.pl suggests to use ether_addr_copy(), but most of
> addresses in driver may be unaligned.

The pointer to the struct is normally aligned (or you are in trouble
anyway and can't use it like a normal pointer).  Then you can calculate
the alignment of the struct member by looking at the struct.

BAReq->addr1 is part of struct rtllib_hdr_2addr.  It's 4 bytes into
the struct so it's fine.

I sort of like the ether_addr_copy_unaligned() macro because it would
let us silence some checkpatch false positives because otherwise people
will eventually introduce bugs like a dripping roof leak will eventually
destroy a building.  But it should be in the main kernel header with a
name like that.  And also this is misleading that we are using it for
data which is aligned.

regards,
dan carpenter

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/