linux-kernel - Re: [PATCHv3 10/14] HSI: Introduce driver for SSI Protocol

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]

Date:	Fri, 25 Apr 2014 20:49:01 +0200
From:	Sebastian Reichel <sre@...nel.org>
To:	Pavel Machek <pavel@....cz>
Cc:	Linus Walleij <linus.walleij@...aro.org>,
	Shubhrajyoti Datta <omaplinuxkernel@...il.com>,
	Carlos Chinea <cch.devel@...il.com>,
	Tony Lindgren <tony@...mide.com>,
	Rob Herring <robh+dt@...nel.org>,
	Pawel Moll <pawel.moll@....com>,
	Mark Rutland <mark.rutland@....com>,
	Kumar Gala <galak@...eaurora.org>, devicetree@...r.kernel.org,
	linux-kernel@...r.kernel.org, linux-omap@...r.kernel.org,
	Pali Rohár <pali.rohar@...il.com>,
	Ивайло Димитров 
	<freemangordon@....bg>,
	Joni Lapilainen <joni.lapilainen@...il.com>,
	Aaro Koskinen <aaro.koskinen@....fi>,
	Carlos Chinea <carlos.chinea@...ia.com>
Subject: Re: [PATCHv3 10/14] HSI: Introduce driver for SSI Protocol

On Sat, Apr 19, 2014 at 09:49:36PM +0200, Pavel Machek wrote:
> Hi!
> 
> > This adds a driver for the SSI McSAAB protocol as used in
> > the Nokia N900.
> > 
> > Signed-off-by: Carlos Chinea <carlos.chinea@...ia.com>
> > Signed-off-by: Sebastian Reichel <sre@...nel.org>
> 
> > +#define SSIP_MIN_PN_HDR		6	/* FIXME: Revisit */
> > +#define SSIP_WDTOUT		2000	/* FIXME: has to be 500 msecs> */
> 
> Can they be fixed now, or do they have to wait?

I would prefer to wait. The timing above works and having it in the
git history doesn't hurt.

> > +#ifdef __LITTLE_ENDIAN
> > +	((u16 *)skb->data)[2] = swab16(((u16 *)skb->data)[2]);
> > +	dev_dbg(&dev->dev, "RX length fixed (%04x -> %u)\n",
> > +			((u16 *)skb->data)[2], ntohs(((u16 *)skb->data)[2]));
> > +#endif
> 
> Uh. Can this be replaced by
>     ((u16 *)skb->data)[2] = htons(((u16 *)skb->data)[2]);
> 
> (without the ifdef?)

Yes.

> > +	/*
> > +	 * Modem sends Phonet messages over SSI with its own endianess...
> > +	 * Assume that modem has the same endianess as we do.
> > +	 */
> > +	if (skb_cow_head(skb, 0))
> > +		goto drop;
> > +#ifdef __LITTLE_ENDIAN
> > +	((u16 *)skb->data)[2] = swab16(((u16 *)skb->data)[2]);
> > +#endif
> 
> Here, too?

Fixed in PATCHv4.

> But it looks like the comment does not match the code, because we
> swap.

The swap only converts the package length field from/to network byte
order. I think the comment still applies for the other actual message.

I added a comment for the length field and kept the other comment as
is for PATCHv4.

-- Sebastian

Download attachment "signature.asc" of type "application/pgp-signature" (820 bytes)