freebsd-skq/usr.sbin/ppp/vjcomp.c

190 lines
4.9 KiB
C
Raw Normal View History

1995-01-31 06:29:58 +00:00
/*
* Input/Output VJ Compressed packets
*
* Written by Toshiharu OHNO (tony-o@iij.ad.jp)
*
* Copyright (C) 1993, Internet Initiative Japan, Inc. All rights reserverd.
*
* Redistribution and use in source and binary forms are permitted
* provided that the above copyright notice and this paragraph are
* duplicated in all such forms and that any documentation,
* advertising materials, and other materials related to such
* distribution and use acknowledge that the software was developed
* by the Internet Initiative Japan, Inc. The name of the
* IIJ may not be used to endorse or promote products derived
* from this software without specific prior written permission.
* THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED
* WARRANTIES OF MERCHANTIBILITY AND FITNESS FOR A PARTICULAR PURPOSE.
*
1999-08-28 01:35:59 +00:00
* $FreeBSD$
1995-05-30 03:57:47 +00:00
*
1995-01-31 06:29:58 +00:00
* TODO:
*/
#include <sys/param.h>
#include <netinet/in.h>
1995-01-31 06:29:58 +00:00
#include <netinet/in_systm.h>
#include <netinet/ip.h>
#include <sys/un.h>
#include <stdio.h>
#include <string.h> /* strlen/memcpy */
#include <termios.h>
#include "layer.h"
#include "mbuf.h"
#include "log.h"
#include "timer.h"
#include "fsm.h"
#include "proto.h"
1995-01-31 06:29:58 +00:00
#include "slcompress.h"
#include "lqr.h"
1995-01-31 06:29:58 +00:00
#include "hdlc.h"
#include "defs.h"
#include "iplist.h"
#include "throughput.h"
1995-01-31 06:29:58 +00:00
#include "ipcp.h"
#include "lcp.h"
#include "ccp.h"
#include "link.h"
#include "filter.h"
#include "descriptor.h"
#include "mp.h"
#ifndef NORADIUS
#include "radius.h"
#endif
#include "bundle.h"
#include "vjcomp.h"
1995-01-31 06:29:58 +00:00
#define MAX_VJHEADER 16 /* Maximum size of compressed header */
1995-01-31 06:29:58 +00:00
static struct mbuf *
vj_LayerPush(struct bundle *bundle, struct link *l, struct mbuf *bp, int pri,
u_short *proto)
1995-01-31 06:29:58 +00:00
{
int type;
struct ip *pip;
u_short cproto = bundle->ncp.ipcp.peer_compproto >> 16;
1995-01-31 06:29:58 +00:00
bp = m_pullup(bp);
pip = (struct ip *)MBUF_CTOP(bp);
if (*proto == PROTO_IP && pip->ip_p == IPPROTO_TCP &&
cproto == PROTO_VJCOMP) {
type = sl_compress_tcp(bp, pip, &bundle->ncp.ipcp.vj.cslc,
&bundle->ncp.ipcp.vj.slstat,
bundle->ncp.ipcp.peer_compproto & 0xff);
log_Printf(LogDEBUG, "vj_LayerWrite: type = %x\n", type);
1995-01-31 06:29:58 +00:00
switch (type) {
case TYPE_IP:
break;
1995-01-31 06:29:58 +00:00
case TYPE_UNCOMPRESSED_TCP:
*proto = PROTO_VJUNCOMP;
Allow ``host:port/udp'' devices and support ``host:port/tcp'' as being the same as the previous (still supported) ``host:port'' syntax for tcp socket devices. A udp device uses synchronous ppp rather than async, and avoids the double-retransmit overhead that comes with ppp over tcp (it's usually a bad idea to transport IP over a reliable transport that itself is using an unreliable transport). PPP over UDP provides througput of ** 1.5Mb per second ** with all compression disabled, maxing out a PPro/200 when running ppp twice, back-to-back. This proves that PPPoE is plausable in userland.... This change adds a few more handler functions to struct device and allows derivations of struct device (which may contain their own data etc) to pass themselves through the unix domain socket for MP. ** At last **, struct physical has lost all the tty crud ! iov2physical() is now smart enough to restore the correct stack of layers so that MP servers will work again. The version number has bumped as our MP link transfer contents have changed (they now may contain a `struct device'). Don't extract the protocol twice in MP mode (resulting in protocol rejects for every MP packet). This was broken with my original layering changes. Add ``Physical'' and ``Sync'' log levels for logging the relevent raw packets and add protocol-tracking LogDEBUG stuff in various LayerPush & LayerPull functions. Assign our physical device name for incoming tcp connections by calling getpeername(). Assign our physical device name for incoming udp connections from the address retrieved by the first recvfrom().
1999-05-12 09:49:12 +00:00
log_Printf(LogDEBUG, "vj_LayerPush: PROTO_IP -> PROTO_VJUNCOMP\n");
m_settype(bp, MB_VJOUT);
1995-01-31 06:29:58 +00:00
break;
1995-01-31 06:29:58 +00:00
case TYPE_COMPRESSED_TCP:
*proto = PROTO_VJCOMP;
Allow ``host:port/udp'' devices and support ``host:port/tcp'' as being the same as the previous (still supported) ``host:port'' syntax for tcp socket devices. A udp device uses synchronous ppp rather than async, and avoids the double-retransmit overhead that comes with ppp over tcp (it's usually a bad idea to transport IP over a reliable transport that itself is using an unreliable transport). PPP over UDP provides througput of ** 1.5Mb per second ** with all compression disabled, maxing out a PPro/200 when running ppp twice, back-to-back. This proves that PPPoE is plausable in userland.... This change adds a few more handler functions to struct device and allows derivations of struct device (which may contain their own data etc) to pass themselves through the unix domain socket for MP. ** At last **, struct physical has lost all the tty crud ! iov2physical() is now smart enough to restore the correct stack of layers so that MP servers will work again. The version number has bumped as our MP link transfer contents have changed (they now may contain a `struct device'). Don't extract the protocol twice in MP mode (resulting in protocol rejects for every MP packet). This was broken with my original layering changes. Add ``Physical'' and ``Sync'' log levels for logging the relevent raw packets and add protocol-tracking LogDEBUG stuff in various LayerPush & LayerPull functions. Assign our physical device name for incoming tcp connections by calling getpeername(). Assign our physical device name for incoming udp connections from the address retrieved by the first recvfrom().
1999-05-12 09:49:12 +00:00
log_Printf(LogDEBUG, "vj_LayerPush: PROTO_IP -> PROTO_VJUNCOMP\n");
m_settype(bp, MB_VJOUT);
1995-01-31 06:29:58 +00:00
break;
1995-01-31 06:29:58 +00:00
default:
Allow ``host:port/udp'' devices and support ``host:port/tcp'' as being the same as the previous (still supported) ``host:port'' syntax for tcp socket devices. A udp device uses synchronous ppp rather than async, and avoids the double-retransmit overhead that comes with ppp over tcp (it's usually a bad idea to transport IP over a reliable transport that itself is using an unreliable transport). PPP over UDP provides througput of ** 1.5Mb per second ** with all compression disabled, maxing out a PPro/200 when running ppp twice, back-to-back. This proves that PPPoE is plausable in userland.... This change adds a few more handler functions to struct device and allows derivations of struct device (which may contain their own data etc) to pass themselves through the unix domain socket for MP. ** At last **, struct physical has lost all the tty crud ! iov2physical() is now smart enough to restore the correct stack of layers so that MP servers will work again. The version number has bumped as our MP link transfer contents have changed (they now may contain a `struct device'). Don't extract the protocol twice in MP mode (resulting in protocol rejects for every MP packet). This was broken with my original layering changes. Add ``Physical'' and ``Sync'' log levels for logging the relevent raw packets and add protocol-tracking LogDEBUG stuff in various LayerPush & LayerPull functions. Assign our physical device name for incoming tcp connections by calling getpeername(). Assign our physical device name for incoming udp connections from the address retrieved by the first recvfrom().
1999-05-12 09:49:12 +00:00
log_Printf(LogERROR, "vj_LayerPush: Unknown frame type %x\n", type);
m_freem(bp);
return NULL;
1995-01-31 06:29:58 +00:00
}
}
return bp;
1995-01-31 06:29:58 +00:00
}
static struct mbuf *
VjUncompressTcp(struct ipcp *ipcp, struct mbuf *bp, u_char type)
1995-01-31 06:29:58 +00:00
{
u_char *bufp;
int len, olen, rlen;
u_char work[MAX_HDR + MAX_VJHEADER]; /* enough to hold TCP/IP header */
1995-01-31 06:29:58 +00:00
bp = m_pullup(bp);
olen = len = m_length(bp);
1995-01-31 06:29:58 +00:00
if (type == TYPE_UNCOMPRESSED_TCP) {
/*
* Uncompressed packet does NOT change its size, so that we can use mbuf
* space for uncompression job.
1995-01-31 06:29:58 +00:00
*/
bufp = MBUF_CTOP(bp);
len = sl_uncompress_tcp(&bufp, len, type, &ipcp->vj.cslc, &ipcp->vj.slstat,
(ipcp->my_compproto >> 8) & 255);
if (len <= 0) {
m_freem(bp);
bp = NULL;
} else
m_settype(bp, MB_VJIN);
return bp;
1995-01-31 06:29:58 +00:00
}
1995-01-31 06:29:58 +00:00
/*
* Handle compressed packet. 1) Read upto MAX_VJHEADER bytes into work
* space. 2) Try to uncompress it. 3) Compute amount of necesary space. 4)
* Copy unread data info there.
1995-01-31 06:29:58 +00:00
*/
if (len > MAX_VJHEADER)
len = MAX_VJHEADER;
1995-01-31 06:29:58 +00:00
rlen = len;
bufp = work + MAX_HDR;
bp = mbuf_Read(bp, bufp, rlen);
len = sl_uncompress_tcp(&bufp, olen, type, &ipcp->vj.cslc, &ipcp->vj.slstat,
(ipcp->my_compproto >> 8) & 255);
if (len <= 0) {
m_freem(bp);
return NULL;
}
1995-01-31 06:29:58 +00:00
len -= olen;
len += rlen;
bp = m_prepend(bp, bufp, len, 0);
m_settype(bp, MB_VJIN);
return bp;
1995-01-31 06:29:58 +00:00
}
static struct mbuf *
vj_LayerPull(struct bundle *bundle, struct link *l, struct mbuf *bp,
u_short *proto)
1995-01-31 06:29:58 +00:00
{
u_char type;
switch (*proto) {
1995-01-31 06:29:58 +00:00
case PROTO_VJCOMP:
type = TYPE_COMPRESSED_TCP;
Allow ``host:port/udp'' devices and support ``host:port/tcp'' as being the same as the previous (still supported) ``host:port'' syntax for tcp socket devices. A udp device uses synchronous ppp rather than async, and avoids the double-retransmit overhead that comes with ppp over tcp (it's usually a bad idea to transport IP over a reliable transport that itself is using an unreliable transport). PPP over UDP provides througput of ** 1.5Mb per second ** with all compression disabled, maxing out a PPro/200 when running ppp twice, back-to-back. This proves that PPPoE is plausable in userland.... This change adds a few more handler functions to struct device and allows derivations of struct device (which may contain their own data etc) to pass themselves through the unix domain socket for MP. ** At last **, struct physical has lost all the tty crud ! iov2physical() is now smart enough to restore the correct stack of layers so that MP servers will work again. The version number has bumped as our MP link transfer contents have changed (they now may contain a `struct device'). Don't extract the protocol twice in MP mode (resulting in protocol rejects for every MP packet). This was broken with my original layering changes. Add ``Physical'' and ``Sync'' log levels for logging the relevent raw packets and add protocol-tracking LogDEBUG stuff in various LayerPush & LayerPull functions. Assign our physical device name for incoming tcp connections by calling getpeername(). Assign our physical device name for incoming udp connections from the address retrieved by the first recvfrom().
1999-05-12 09:49:12 +00:00
log_Printf(LogDEBUG, "vj_LayerPull: PROTO_VJCOMP -> PROTO_IP\n");
1995-01-31 06:29:58 +00:00
break;
case PROTO_VJUNCOMP:
type = TYPE_UNCOMPRESSED_TCP;
Allow ``host:port/udp'' devices and support ``host:port/tcp'' as being the same as the previous (still supported) ``host:port'' syntax for tcp socket devices. A udp device uses synchronous ppp rather than async, and avoids the double-retransmit overhead that comes with ppp over tcp (it's usually a bad idea to transport IP over a reliable transport that itself is using an unreliable transport). PPP over UDP provides througput of ** 1.5Mb per second ** with all compression disabled, maxing out a PPro/200 when running ppp twice, back-to-back. This proves that PPPoE is plausable in userland.... This change adds a few more handler functions to struct device and allows derivations of struct device (which may contain their own data etc) to pass themselves through the unix domain socket for MP. ** At last **, struct physical has lost all the tty crud ! iov2physical() is now smart enough to restore the correct stack of layers so that MP servers will work again. The version number has bumped as our MP link transfer contents have changed (they now may contain a `struct device'). Don't extract the protocol twice in MP mode (resulting in protocol rejects for every MP packet). This was broken with my original layering changes. Add ``Physical'' and ``Sync'' log levels for logging the relevent raw packets and add protocol-tracking LogDEBUG stuff in various LayerPush & LayerPull functions. Assign our physical device name for incoming tcp connections by calling getpeername(). Assign our physical device name for incoming udp connections from the address retrieved by the first recvfrom().
1999-05-12 09:49:12 +00:00
log_Printf(LogDEBUG, "vj_LayerPull: PROTO_VJUNCOMP -> PROTO_IP\n");
1995-01-31 06:29:58 +00:00
break;
default:
return bp;
1995-01-31 06:29:58 +00:00
}
*proto = PROTO_IP;
return VjUncompressTcp(&bundle->ncp.ipcp, bp, type);
1995-01-31 06:29:58 +00:00
}
const char *
vj2asc(u_int32_t val)
{
static char asc[50]; /* The return value is used immediately */
if (val)
snprintf(asc, sizeof asc, "%d VJ slots with%s slot compression",
(int)((val>>8)&15)+1, val & 1 ? "" : "out");
else
strcpy(asc, "VJ disabled");
return asc;
}
struct layer vjlayer = { LAYER_VJ, "vj", vj_LayerPush, vj_LayerPull };