>  #define BPF_DLTLISTSIZ 10
>  struct bpf_dltlist {
>          size_t	bdl_len;

i'd suggest that a u_int (as below) would suffice here.  what you're
storing isn't a size, per se, and i can't see any need for the count
to grow to a 64 bit on a 64 bit platform.  if a 64 platform ends up
with a list of dlt types that exceeds 32 bits...well...i think you see
my point.  :)

>          u_int		bdl_dlts[BPF_DLTLISTSIZ];
>  };
>
>BIOCSDLT is self-explanatory. BIOCGDLTLIST will store up to BPF_DLTLISTSIZ
>dlts in bdl_dlts and set bdl_len to the number it stores, unless the
>number of DLTs exceed BPF_DLTLISTSIZ. In that case, it writes to bdl_len
>the total number of DLTs for the interface.

simplify the interface, perhaps?  perhaps use the "trick" you have in
struct ifdltlist with the 1

>  struct ifdltlist {
>          u_int ifdl_count;                   /* length of ifdl_list */
>          struct ifdltlistelt ifdl_list[1];   /* records for eligible DLTs */
>  };

and *always* store the number of dlts in bdl_len.  require the caller
to preset bdl_len to the maximum number for which it has allocated
storage.  i could use that, for example, by passing you a pointer to a
u_int (or whatever the size of bdl_len ends up being) that's set to 0,
and get back a number that tells me how large my buffer needs to be.
then i malloc and call again.  :)

-- 
|-----< "CODE WARRIOR" >-----|
codewarrior@daemon.org             * "ah!  i see you have the internet
twofsonet@graffiti.com (Andrew Brown)                that goes *ping*!"
andrew@crossbar.com       * "information is power -- share the wealth."