Re: [PATCH] public/io/netif.h: make control ring hash protocol more general

[Paul Durrant <Paul.Durrant@citrix.com>]

> -----Original Message-----
> From: Jan Beulich [mailto:JBeulich@suse.com]
> Sent: 15 February 2016 11:13
> To: Paul Durrant
> Cc: Ian Campbell; Ian Jackson; xen-devel@lists.xenproject.org; Keir
> (Xen.org); Tim (Xen.org)
> Subject: Re: [PATCH] public/io/netif.h: make control ring hash protocol more
> general
> 
> >>> On 15.02.16 at 11:17, <paul.durrant@citrix.com> wrote:
> > +#define NETIF_CTRL_HASH_ALGORITHM_TOEPLITZ 1
> > +
> > +/*
> > + * This algorithm uses a 'key' as well as the data buffer itself.
> > + * (Buffer[] and Key[] are treated as shift-registers where the MSB of
> > + * Buffer/Key[0] is considered 'left-most' and the LSB of Buffer/Key[N-1]
> > + * is the 'right-most').
> > + *
> > + * Value = 0
> > + * For number of bits in Buffer[]
> > + *    If (left-most bit of Buffer[] is 1)
> > + *        Value ^= left-most 32 bits of Key[]
> > + *    Key[] << 1
> > + *    Buffer[] << 1
> > + *
> > + * The code below is provided for convenience where an operating
> system
> > + * does not already provide an implementation.
> > + */
> > +
> > +static inline uint32_t netif_toeplitz_hash(const uint8_t *key,
> > +                                           unsigned int keylen,
> > +                                           const uint8_t *buf,
> > +                                           unsigned int buflen)
> > +{
> > +        unsigned int keyi, bufi;
> > +        uint64_t prefix = 0;
> > +        uint64_t hash = 0;
> > +
> > +        /* Pre-load prefix with the first 8 bytes of the key */
> > +        for (keyi = 0; keyi < 8; keyi++) {
> > +                prefix <<= 8;
> > +                prefix |= (keyi < keylen) ? key[keyi] : 0;
> > +        }
> > +
> > +        for (bufi = 0; bufi < buflen; bufi++) {
> > +                uint8_t byte = buf[bufi];
> > +                unsigned int bit;
> > +
> > +                for (bit = 0; bit < 8; bit++) {
> > +                        if (byte & 0x80)
> > +                                hash ^= prefix;
> > +                        prefix <<= 1;
> > +                        byte <<=1;
> > +                }
> > +
> > +                /*
> > +                 * 'prefix' has now been left-shifted by 8, so
> > +                 * OR in the next byte.
> > +                 */
> > +                prefix |= (keyi < keylen) ? key[keyi] : 0;
> > +                keyi++;
> > +        }
> > +
> > +        /* The valid part of the hash is in the upper 32 bits. */
> > +        return hash >> 32;
> > +}
> 
> "inline" is not a C89 keyword and hence can't be used without
> suitable guarding in a public header. I'd suggest making this
> reference implementation an opt-in thing, controllable by the
> consumer needing to #define some tbd identifier.
> 

Ok. That's a good idea. I'll wrap it with something.

  Paul

> Jan