[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <1398548627.16049.8.camel@deadeye.wl.decadent.org.uk>
Date: Sat, 26 Apr 2014 22:43:47 +0100
From: Ben Hutchings <ben@...adent.org.uk>
To: Tom Herbert <therbert@...gle.com>
Cc: davem@...emloft.net, netdev@...r.kernel.org
Subject: Re: [PATCH 5/9 v2] net: Generalize checksum_init functions
On Sat, 2014-04-26 at 22:42 +0100, Ben Hutchings wrote:
> On Sat, 2014-04-26 at 14:25 -0700, Tom Herbert wrote:
[...]
> > +/*
> > + * Perform checksum validate (init). Note that this is a macro since we only
> > + * want to calculate the pseudo header which is an input function if necessary.
> > + * First we try to validate without any computation (checksum unnecessary) and
> > + * then calculate based on checksum complete calling the function to compute
> > + * pseudo header.
> > + */
> > +#define __skb_checksum_validate(skb, proto, complete, \
> > + zero_okay, check, compute_pseudo) \
> > +({ \
> > + __sum16 __ret = __skb_checksum_validate_unnecessary(skb, \
> > + zero_okay, check); \
> > + if (__ret) \
> > + __ret = __skb_checksum_validate_complete(skb, \
> > + complete, compute_pseudo(skb, proto)); \
> > + __ret; \
> > +})
>
> This may or may not yield an actual checksum, which is not obvious from
> the comment.
[...]
Sorry, this is incorrect; it does yield an actual checksum.
Ben.
--
Ben Hutchings
The most exhausting thing in life is being insincere. - Anne Morrow Lindberg
Download attachment "signature.asc" of type "application/pgp-signature" (812 bytes)
Powered by blists - more mailing lists