Re: [PATCH] HDLC patch for 2.5.5 (1/3)

[Francois Romieu]

Jeff Garzik <[e-mail]> :
> Francois Romieu wrote:
> > 
> > [1/3]:
> > - struct if_settings in struct ifreq becomes struct if_settings *
> 
> mostly ok... I'm wondering how this specific item (shown below in the
> patch portion quoted) affects binary compatibility...  how does this
> affect userland if ifru_settings is suddenly a pointer?

2.4.18-preX:include/linux/if.h
[...]
        union {
                struct  sockaddr ifru_addr;
                struct  sockaddr ifru_dstaddr;
                struct  sockaddr ifru_broadaddr;
                struct  sockaddr ifru_netmask;
                struct  sockaddr ifru_hwaddr;
                short   ifru_flags;
                int     ifru_ivalue;
                int     ifru_mtu;
                struct  ifmap ifru_map;
                char    ifru_slave[IFNAMSIZ];   /* Just fits the size */
                char    ifru_newname[IFNAMSIZ];
                char *  ifru_data; <-- Nothing after this field.
        } ifr_ifru;
};

The member ifru_settings is added to the union by Krzysztof's patch.
Existing userland (Krzysztof's new utility aside) shouldn't matter as
- it ignores this member;
- the member doesn't make the union bigger (Krzysztof noticed my previous
  patch choked on this).

If nobody does it before, the utility will be modified when I feel like it's
time to test the code. Shouldn't be long (TM).

-- 
Ueimor