linux-kernel - Re: I need help with a sparse warning

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <200809121725.06870.mb@bu3sch.de>
Date:	Fri, 12 Sep 2008 17:25:06 +0200
From:	Michael Buesch <mb@...sch.de>
To:	Larry Finger <Larry.Finger@...inger.net>
Cc:	Steven Noonan <steven@...inklabs.net>,
	LKML <linux-kernel@...r.kernel.org>,
	wireless <linux-wireless@...r.kernel.org>
Subject: Re: I need help with a sparse warning

On Friday 12 September 2008, Larry Finger wrote:
> Steven Noonan wrote:
> > On Thu, Sep 11, 2008 at 7:13 PM, Larry Finger <Larry.Finger@...inger.net> wrote:
> >> In file drivers/net/wireless/p54/p54common.c, the statement
> >>
> >>       priv->rx_mtu = (size_t) le16_to_cpu((__le16)bootrec->data[10]);
> >>
> >> generates the sparse warning
> >>
> >> .../p54common.c:185:29: warning: cast to restricted __le16
> >>
> >> where bootrec->data is u32, and priv->rx_mtu is u16.
> >>
> > 
> > (Whoops, didn't CC the mailing lists. Sorry about the double-message, Larry.)
> > 
> > If priv->rx_mtu is u16, I'm surprised it doesn't get noisy about the
> > size_t cast. Unless the machine it's being compiled on is 16-bit, that
> > should throw a truncation warning, because size_t should be a 32-bit
> > integer on 32-bit machines (typically).
> > 
> > I think if you change the (__le16) cast to (__le16 __force) it will
> > stop warning you about that particular issue.
> 
> This one gets rid of the sparse warning.
> 
>         priv->rx_mtu = le16_to_cpu((__le16 __force)
>                                    bootrec->data[10]);

I think you should rather fix the type of bootrec->data.
If it is u32, but the data in fact is __le16, please fix the struct instead
of adding dangerous casts.
Note that this is especially dangerous, because you cast between different
typesizes and endianesses at the same time.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/