[En-Nut-Discussion] MSS not calculated correctly

[Thiago A. Corrêa]

Hi,

2010/4/25 Henrik Maier <hmnews at proconx.com>:
>
>            /* Read MAXSEG option */
>            case TCPOPT_MAXSEG:
>                s = ((u_short)cp[2] << 8) | cp[3];
>                if (s < sock->so_mss)
>                    sock->so_mss = s;
>
> Any thoughts/agreements/disagreements?
>

I think this would break any future little endian platform. I believe
the proper way should be to keep the ntohs call and remove the bit
shift:

            /* Read MAXSEG option */
            case TCPOPT_MAXSEG:
                s = ntohs(*((u_short*)&cp[2]));
                if (s < sock->so_mss)
                    sock->so_mss = s;


You could also make sure someone doesn't inject this bogus MSS value,
adding a check to make sure it's above 4.

> PS2: I am also not sure if topics like this should go to the mailing list or
> rather be lodged as bug report on SourceForge instead (or both)?

Both is cool, gives us a warning and an oportunity to discuss, the
bugtracker makes sure it's not forgotten :)

Kind Regards,
    Thiago A. Correa