mjd@saul.cis.upenn.edu (Mark-Jason Dominus) (10/17/90)
The defininition of struct sockaddr_in in the <netinet/in.h> files on our suns has four members, one of which is char sin_zero[8]; What is it for? -- In some sense a stochastic process can do better; at least it has a chance. Mark-Jason Dominus mjd@central.cis.upenn.edu
nieters@phobos.crd.ge.com (coolbean) (10/17/90)
;
; The defininition of struct sockaddr_in in the <netinet/in.h>
; files on our suns has four members, one of which is
; char sin_zero[8];
;
; What is it for?
This is an unused member whose sole purpose is to pad the structure
out to 16 bytes. Similarly, the Xerox NS family has a "_zero" member
struct sockaddr_ns {
u_short sns_family; /* AF_NS */
struct ns_addr sns_addr; /* the 12-byte XNS address */
char sns_zero[2]; /* unused */
^^^^^^^^^^^^^^^^^
};
total sizeof(struct sockaddr_ns) == 16 bytes.
If you look in <sys/socket.h> you'll find the definition for struct
sockaddr which is 2 bytes of address family (sa_family) followed by
"up to 14 bytes of direct address" (sa_data[14] ... a.k.a. protocol
specific address.) Since system calls dealing with sockets
(e.g. accept() ) take the generic (struct sockaddr *) argument, protocol
specific structures (such as sockaddr_in, in this case) need to be
padded out to the same size as the generic.
hope this helps.
--ed
--
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Ed Nieters INTERnet: nieters@crd.ge.com
GE Corporate R&D UUCPnet: uunet!crd.ge.com!nieters
Schenectady, NY 12301 BELLnet: (518) 387-5187jim@cs.strath.ac.uk (Jim Reid) (10/17/90)
In article <MJD.90Oct16130739@saul.cis.upenn.edu> mjd@saul.cis.upenn.edu (Mark-Jason Dominus) writes:
The defininition of struct sockaddr_in in the <netinet/in.h>
files on our suns has four members, one of which is
char sin_zero[8];
What is it for?
Nothing. It's purely to pad out the sockaddr_in struct to 16 bytes,
the size of the generic sockaddr structure in <sys/socket.h>.
Jimchris@%griffon@mcc.oz (Chris Robertson) (10/22/90)
In article <JIM.90Oct17174526@baird.cs.strath.ac.uk> jim@cs.strath.ac.uk (Jim Reid) writes: >In article <MJD.90Oct16130739@saul.cis.upenn.edu> mjd@saul.cis.upenn.edu (Mark-Jason Dominus) writes: > > The defininition of struct sockaddr_in in the <netinet/in.h> > files on our suns has four members, one of which is > char sin_zero[8]; > > What is it for? > >Nothing. It's purely to pad out the sockaddr_in struct to 16 bytes, >the size of the generic sockaddr structure in <sys/socket.h>. > > Jim Actually, it's a little-known fact of Unix history that the "8" in "sin_zero[8]" is really an infinity symbol, placed vertically due to lack of an appropriate ascii character. The significance of all this is although original Unix direct from the Hand of the Creator was pure and without sin, the Berkeley variants took on the knowledge of evil as well as good, thus acquiring Infinite Original Sin... (;-), for the humour-impaired) -- "In the beginning was the Word -- no, just a sec, | Chris Robertson the Byte, no, the Bit... oh hell, forget it!" | chris%griffon@mcc.oz