Re: net/* module-related oddities

From: David S. Miller (kpnzegny.xvkggiri@rockopnh.nl)
Date: Sat Feb 21 2004 - 23:02:34 EET

  • Next message: Jorge Matias: "Re: Kpsk doesn't work with ALSA/es1968 module"

    On Sat, 21 Feb 2004 04:41:44 +0000
    zfx@mail.dy.fi wrote:

    > What is intended there? Is it "netrom already speaks that protocol
    > and it would be unhappy if you try to interfere"? Then the first case above
    > is a bug - we can get netrom loaded and such socket opened. If it's not that,
    > WTF it is?

    It cannot be because of a protocol handling conflict. NETROM registers
    AF_NETROM, AX25 registers AF_AX25, and ROSE registers AF_ROSE.

    I think all of these tests are simply backwards. What it wants to do is
    say ESOCKTNOSUPPORT if the netrom/rose/etc. protocols aren't present or
    their module type can't be loaded, during ax25_create().

    This code has been like this for centuries, I'm so scared to touch it.

    Ralf, since you're the listed AX.25/ROSE/NETROM maintainer :) Take a look
    at the crap happening in ax25_create(), all thoses tests about SEQPACKET
    sockets and IP, ROSE, NETROM. They all look reversed. They have to be,
    if not neither I nor Al here have any idea what the code is trying to do.

    Thanks.
    -
    To unsubscribe from this list: send the line "unsubscribe linux-hams" in
    the body of a message to kamoohhe.vwzwa@basys-gmbh.de
    More majordomo info at http://vger.kernel.org/majordomo-info.html



    This archive was generated by hypermail 2b30 : Sat Feb 21 2004 - 23:22:24 EET