freebsd-dev/usr.sbin/ppp/ccp.c

682 lines
19 KiB
C
Raw Normal View History

1995-01-31 06:29:58 +00:00
/*
* PPP Compression Control Protocol (CCP) Module
*
* Written by Toshiharu OHNO (tony-o@iij.ad.jp)
*
* Copyright (C) 1994, 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.
1995-05-30 03:57:47 +00:00
*
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:
* o Support other compression protocols
*/
#include <sys/param.h>
#include <netinet/in.h>
#include <netinet/in_systm.h>
#include <netinet/ip.h>
#include <sys/un.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h> /* memcpy() on some archs */
#include <termios.h>
#include "layer.h"
#include "defs.h"
#include "command.h"
#include "mbuf.h"
#include "log.h"
#include "timer.h"
1995-01-31 06:29:58 +00:00
#include "fsm.h"
#include "proto.h"
#include "pred.h"
#include "deflate.h"
1998-03-13 21:07:46 +00:00
#include "throughput.h"
#include "iplist.h"
#include "slcompress.h"
#include "lqr.h"
#include "hdlc.h"
#include "lcp.h"
#include "ccp.h"
1998-03-13 21:07:46 +00:00
#include "ipcp.h"
1998-03-16 22:52:54 +00:00
#include "filter.h"
#include "descriptor.h"
#include "prompt.h"
#include "link.h"
o Move struct lcp and struct ccp into struct link. o Remove bundle2lcp(), bundle2ccp() and bundle2link(). They're too resource-hungry and we have `owner pointers' to do their job. o Make our FSM understand LCPs that are always ST_OPENED (with a minimum code that != 1). o Send FSM code rejects for invalid codes. o Make our bundle fsm_parent deal with multiple links. o Make timer diagnostics pretty and allow access via ~t in `term' mode (not just when logging debug) and `show timers'. Only show timers every second in debug mode, otherwise we get too many diagnostics to be useful (we probably still do). Also, don't restrict ~m in term mode to depend on debug logging. o Rationalise our bundles' phases. o Create struct mp (multilink protocol). This is both an NCP and a type of struct link. It feeds off other NCPs for output, passing fragmented packets into the queues of available datalinks. It also gets PROTO_MP input, reassembles the fragments into ppp frames, and passes them back to the HDLC layer that the fragments were passed from. ** It's not yet possible to enter multilink mode :-( ** o Add `set weight' (requires context) for deciding on a links weighting in multilink mode. Weighting is simplistic (and probably badly implemented) for now. o Remove the function pointers in struct link. They ended up only applying to physical links. o Configure our tun device with an MTU equal to the MRU from struct mp's LCP and a speed equal to the sum of our link speeds. o `show {lcp,ccp,proto}' and `set deflate' now have optional context and use ChooseLink() to decide on which `struct link' to use. This allows behaviour as before when in non-multilink mode, and allows access to the MP logical link in multilink mode. o Ignore reconnect and redial values when in -direct mode and when cleaning up. Always redial when in -ddial or -dedicated mode (unless cleaning up). o Tell our links to `staydown' when we close them due to a signal. o Remove remaining `#ifdef SIGALRM's (ppp doesn't function without alarms). o Don't bother strdup()ing our physical link name. o Various other cosmetic changes.
1998-04-03 19:21:56 +00:00
#include "mp.h"
#include "async.h"
#include "physical.h"
#ifndef NORADIUS
#include "radius.h"
#endif
#ifdef HAVE_DES
#include "mppe.h"
#endif
o Move struct lcp and struct ccp into struct link. o Remove bundle2lcp(), bundle2ccp() and bundle2link(). They're too resource-hungry and we have `owner pointers' to do their job. o Make our FSM understand LCPs that are always ST_OPENED (with a minimum code that != 1). o Send FSM code rejects for invalid codes. o Make our bundle fsm_parent deal with multiple links. o Make timer diagnostics pretty and allow access via ~t in `term' mode (not just when logging debug) and `show timers'. Only show timers every second in debug mode, otherwise we get too many diagnostics to be useful (we probably still do). Also, don't restrict ~m in term mode to depend on debug logging. o Rationalise our bundles' phases. o Create struct mp (multilink protocol). This is both an NCP and a type of struct link. It feeds off other NCPs for output, passing fragmented packets into the queues of available datalinks. It also gets PROTO_MP input, reassembles the fragments into ppp frames, and passes them back to the HDLC layer that the fragments were passed from. ** It's not yet possible to enter multilink mode :-( ** o Add `set weight' (requires context) for deciding on a links weighting in multilink mode. Weighting is simplistic (and probably badly implemented) for now. o Remove the function pointers in struct link. They ended up only applying to physical links. o Configure our tun device with an MTU equal to the MRU from struct mp's LCP and a speed equal to the sum of our link speeds. o `show {lcp,ccp,proto}' and `set deflate' now have optional context and use ChooseLink() to decide on which `struct link' to use. This allows behaviour as before when in non-multilink mode, and allows access to the MP logical link in multilink mode. o Ignore reconnect and redial values when in -direct mode and when cleaning up. Always redial when in -ddial or -dedicated mode (unless cleaning up). o Tell our links to `staydown' when we close them due to a signal. o Remove remaining `#ifdef SIGALRM's (ppp doesn't function without alarms). o Don't bother strdup()ing our physical link name. o Various other cosmetic changes.
1998-04-03 19:21:56 +00:00
#include "bundle.h"
1995-01-31 06:29:58 +00:00
static void CcpSendConfigReq(struct fsm *);
static void CcpSentTerminateReq(struct fsm *);
static void CcpSendTerminateAck(struct fsm *, u_char);
1998-03-13 21:08:05 +00:00
static void CcpDecodeConfig(struct fsm *, u_char *, int, int,
struct fsm_decode *);
static void CcpLayerStart(struct fsm *);
static void CcpLayerFinish(struct fsm *);
static int CcpLayerUp(struct fsm *);
static void CcpLayerDown(struct fsm *);
static void CcpInitRestartCounter(struct fsm *, int);
static void CcpRecvResetReq(struct fsm *);
static void CcpRecvResetAck(struct fsm *, u_char);
1995-01-31 06:29:58 +00:00
static struct fsm_callbacks ccp_Callbacks = {
1995-01-31 06:29:58 +00:00
CcpLayerUp,
CcpLayerDown,
CcpLayerStart,
CcpLayerFinish,
CcpInitRestartCounter,
CcpSendConfigReq,
CcpSentTerminateReq,
1995-01-31 06:29:58 +00:00
CcpSendTerminateAck,
CcpDecodeConfig,
CcpRecvResetReq,
CcpRecvResetAck
1995-01-31 06:29:58 +00:00
};
static const char * const ccp_TimerNames[] =
{"CCP restart", "CCP openmode", "CCP stopped"};
static const char *
protoname(int proto)
{
static char const * const cftypes[] = {
/* Check out the latest ``Compression Control Protocol'' rfc (1962) */
"OUI", /* 0: OUI */
"PRED1", /* 1: Predictor type 1 */
"PRED2", /* 2: Predictor type 2 */
"PUDDLE", /* 3: Puddle Jumber */
NULL, NULL, NULL, NULL, NULL, NULL,
NULL, NULL, NULL, NULL, NULL, NULL,
"HWPPC", /* 16: Hewlett-Packard PPC */
"STAC", /* 17: Stac Electronics LZS (rfc1974) */
"MPPE", /* 18: Microsoft PPC (rfc2118) and */
/* Microsoft PPE (draft-ietf-pppext-mppe) */
"GAND", /* 19: Gandalf FZA (rfc1993) */
"V42BIS", /* 20: ARG->DATA.42bis compression */
"BSD", /* 21: BSD LZW Compress */
NULL,
"LZS-DCP", /* 23: LZS-DCP Compression Protocol (rfc1967) */
"MAGNALINK/DEFLATE",/* 24: Magnalink Variable Resource (rfc1975) */
/* 24: Deflate (according to pppd-2.3.*) */
"DCE", /* 25: Data Circuit-Terminating Equip (rfc1976) */
"DEFLATE", /* 26: Deflate (rfc1979) */
};
if (proto < 0 || proto > sizeof cftypes / sizeof *cftypes ||
cftypes[proto] == NULL)
return HexStr(proto, NULL, 0);
return cftypes[proto];
}
/* We support these algorithms, and Req them in the given order */
static const struct ccp_algorithm * const algorithm[] = {
&DeflateAlgorithm,
&Pred1Algorithm,
&PppdDeflateAlgorithm
#ifdef HAVE_DES
, &MPPEAlgorithm
#endif
};
#define NALGORITHMS (sizeof algorithm/sizeof algorithm[0])
int
ccp_ReportStatus(struct cmdargs const *arg)
1995-01-31 06:29:58 +00:00
{
struct link *l;
struct ccp *ccp;
l = command_ChooseLink(arg);
ccp = &l->ccp;
prompt_Printf(arg->prompt, "%s: %s [%s]\n", l->name, ccp->fsm.name,
State2Nam(ccp->fsm.state));
if (ccp->fsm.state == ST_OPENED) {
prompt_Printf(arg->prompt, " My protocol = %s, His protocol = %s\n",
protoname(ccp->my_proto), protoname(ccp->his_proto));
prompt_Printf(arg->prompt, " Output: %ld --> %ld, Input: %ld --> %ld\n",
ccp->uncompout, ccp->compout,
ccp->compin, ccp->uncompin);
}
prompt_Printf(arg->prompt, "\n Defaults: ");
prompt_Printf(arg->prompt, "FSM retry = %us, max %u Config"
" REQ%s, %u Term REQ%s\n", ccp->cfg.fsm.timeout,
ccp->cfg.fsm.maxreq, ccp->cfg.fsm.maxreq == 1 ? "" : "s",
ccp->cfg.fsm.maxtrm, ccp->cfg.fsm.maxtrm == 1 ? "" : "s");
prompt_Printf(arg->prompt, " deflate windows: ");
prompt_Printf(arg->prompt, "incoming = %d, ", ccp->cfg.deflate.in.winsize);
prompt_Printf(arg->prompt, "outgoing = %d\n", ccp->cfg.deflate.out.winsize);
prompt_Printf(arg->prompt, " DEFLATE: %s\n",
command_ShowNegval(ccp->cfg.neg[CCP_NEG_DEFLATE]));
prompt_Printf(arg->prompt, " PREDICTOR1: %s\n",
command_ShowNegval(ccp->cfg.neg[CCP_NEG_PRED1]));
prompt_Printf(arg->prompt, " DEFLATE24: %s\n",
command_ShowNegval(ccp->cfg.neg[CCP_NEG_DEFLATE24]));
#ifdef HAVE_DES
prompt_Printf(arg->prompt, " MPPE: %s\n",
command_ShowNegval(ccp->cfg.neg[CCP_NEG_MPPE]));
prompt_Printf(arg->prompt, "Key Size = %d-bits\n", ccp->cfg.mppe.keybits);
#endif
return 0;
1995-01-31 06:29:58 +00:00
}
void
ccp_SetupCallbacks(struct ccp *ccp)
{
ccp->fsm.fn = &ccp_Callbacks;
ccp->fsm.FsmTimer.name = ccp_TimerNames[0];
ccp->fsm.OpenTimer.name = ccp_TimerNames[1];
ccp->fsm.StoppedTimer.name = ccp_TimerNames[2];
}
void
ccp_Init(struct ccp *ccp, struct bundle *bundle, struct link *l,
const struct fsm_parent *parent)
1995-01-31 06:29:58 +00:00
{
/* Initialise ourselves */
o Move struct lcp and struct ccp into struct link. o Remove bundle2lcp(), bundle2ccp() and bundle2link(). They're too resource-hungry and we have `owner pointers' to do their job. o Make our FSM understand LCPs that are always ST_OPENED (with a minimum code that != 1). o Send FSM code rejects for invalid codes. o Make our bundle fsm_parent deal with multiple links. o Make timer diagnostics pretty and allow access via ~t in `term' mode (not just when logging debug) and `show timers'. Only show timers every second in debug mode, otherwise we get too many diagnostics to be useful (we probably still do). Also, don't restrict ~m in term mode to depend on debug logging. o Rationalise our bundles' phases. o Create struct mp (multilink protocol). This is both an NCP and a type of struct link. It feeds off other NCPs for output, passing fragmented packets into the queues of available datalinks. It also gets PROTO_MP input, reassembles the fragments into ppp frames, and passes them back to the HDLC layer that the fragments were passed from. ** It's not yet possible to enter multilink mode :-( ** o Add `set weight' (requires context) for deciding on a links weighting in multilink mode. Weighting is simplistic (and probably badly implemented) for now. o Remove the function pointers in struct link. They ended up only applying to physical links. o Configure our tun device with an MTU equal to the MRU from struct mp's LCP and a speed equal to the sum of our link speeds. o `show {lcp,ccp,proto}' and `set deflate' now have optional context and use ChooseLink() to decide on which `struct link' to use. This allows behaviour as before when in non-multilink mode, and allows access to the MP logical link in multilink mode. o Ignore reconnect and redial values when in -direct mode and when cleaning up. Always redial when in -ddial or -dedicated mode (unless cleaning up). o Tell our links to `staydown' when we close them due to a signal. o Remove remaining `#ifdef SIGALRM's (ppp doesn't function without alarms). o Don't bother strdup()ing our physical link name. o Various other cosmetic changes.
1998-04-03 19:21:56 +00:00
fsm_Init(&ccp->fsm, "CCP", PROTO_CCP, 1, CCP_MAXCODE, LogCCP,
bundle, l, parent, &ccp_Callbacks, ccp_TimerNames);
ccp->cfg.deflate.in.winsize = 0;
ccp->cfg.deflate.out.winsize = 15;
ccp->cfg.fsm.timeout = DEF_FSMRETRY;
ccp->cfg.fsm.maxreq = DEF_FSMTRIES;
ccp->cfg.fsm.maxtrm = DEF_FSMTRIES;
ccp->cfg.neg[CCP_NEG_DEFLATE] = NEG_ENABLED|NEG_ACCEPTED;
ccp->cfg.neg[CCP_NEG_PRED1] = NEG_ENABLED|NEG_ACCEPTED;
ccp->cfg.neg[CCP_NEG_DEFLATE24] = 0;
#ifdef HAVE_DES
ccp->cfg.mppe.keybits = 128;
ccp->cfg.neg[CCP_NEG_MPPE] = 0;
#endif
ccp_Setup(ccp);
}
void
ccp_Setup(struct ccp *ccp)
{
/* Set ourselves up for a startup */
ccp->fsm.open_mode = 0;
ccp->his_proto = ccp->my_proto = -1;
ccp->reset_sent = ccp->last_reset = -1;
ccp->in.algorithm = ccp->out.algorithm = -1;
ccp->in.state = ccp->out.state = NULL;
ccp->in.opt.id = -1;
ccp->out.opt = NULL;
ccp->his_reject = ccp->my_reject = 0;
ccp->uncompout = ccp->compout = 0;
ccp->uncompin = ccp->compin = 0;
1995-01-31 06:29:58 +00:00
}
static void
CcpInitRestartCounter(struct fsm *fp, int what)
1995-01-31 06:29:58 +00:00
{
/* Set fsm timer load */
struct ccp *ccp = fsm2ccp(fp);
fp->FsmTimer.load = ccp->cfg.fsm.timeout * SECTICKS;
switch (what) {
case FSM_REQ_TIMER:
fp->restart = ccp->cfg.fsm.maxreq;
break;
case FSM_TRM_TIMER:
fp->restart = ccp->cfg.fsm.maxtrm;
break;
default:
fp->restart = 1;
break;
}
1995-01-31 06:29:58 +00:00
}
static void
CcpSendConfigReq(struct fsm *fp)
1995-01-31 06:29:58 +00:00
{
/* Send config REQ please */
struct ccp *ccp = fsm2ccp(fp);
struct ccp_opt **o;
1998-03-13 21:08:05 +00:00
u_char *cp, buff[100];
int f, alloc;
1998-03-13 21:08:05 +00:00
cp = buff;
o = &ccp->out.opt;
alloc = ccp->his_reject == 0 && ccp->out.opt == NULL;
ccp->my_proto = -1;
ccp->out.algorithm = -1;
for (f = 0; f < NALGORITHMS; f++)
if (IsEnabled(ccp->cfg.neg[algorithm[f]->Neg]) &&
!REJECTED(ccp, algorithm[f]->id)) {
if (!alloc)
for (o = &ccp->out.opt; *o != NULL; o = &(*o)->next)
if ((*o)->val.id == algorithm[f]->id && (*o)->algorithm == f)
break;
if (alloc || *o == NULL) {
*o = (struct ccp_opt *)malloc(sizeof(struct ccp_opt));
(*o)->val.id = algorithm[f]->id;
(*o)->val.len = 2;
(*o)->next = NULL;
(*o)->algorithm = f;
(*algorithm[f]->o.OptInit)(&(*o)->val, &ccp->cfg);
}
if (cp + (*o)->val.len > buff + sizeof buff) {
log_Printf(LogERROR, "%s: CCP REQ buffer overrun !\n", fp->link->name);
1998-03-13 21:08:05 +00:00
break;
}
memcpy(cp, &(*o)->val, (*o)->val.len);
cp += (*o)->val.len;
ccp->my_proto = (*o)->val.id;
ccp->out.algorithm = f;
if (alloc)
o = &(*o)->next;
}
fsm_Output(fp, CODE_CONFIGREQ, fp->reqid, buff, cp - buff, MB_CCPOUT);
1995-01-31 06:29:58 +00:00
}
void
ccp_SendResetReq(struct fsm *fp)
1995-01-31 06:29:58 +00:00
{
/* We can't read our input - ask peer to reset */
struct ccp *ccp = fsm2ccp(fp);
ccp->reset_sent = fp->reqid;
ccp->last_reset = -1;
fsm_Output(fp, CODE_RESETREQ, fp->reqid, NULL, 0, MB_CCPOUT);
1995-01-31 06:29:58 +00:00
}
static void
CcpSentTerminateReq(struct fsm *fp)
1995-01-31 06:29:58 +00:00
{
/* Term REQ just sent by FSM */
1995-01-31 06:29:58 +00:00
}
static void
CcpSendTerminateAck(struct fsm *fp, u_char id)
1995-01-31 06:29:58 +00:00
{
/* Send Term ACK please */
fsm_Output(fp, CODE_TERMACK, id, NULL, 0, MB_CCPOUT);
1995-01-31 06:29:58 +00:00
}
static void
CcpRecvResetReq(struct fsm *fp)
1995-01-31 06:29:58 +00:00
{
/* Got a reset REQ, reset outgoing dictionary */
struct ccp *ccp = fsm2ccp(fp);
if (ccp->out.state != NULL)
(*algorithm[ccp->out.algorithm]->o.Reset)(ccp->out.state);
1995-01-31 06:29:58 +00:00
}
static void
CcpLayerStart(struct fsm *fp)
1995-01-31 06:29:58 +00:00
{
/* We're about to start up ! */
struct ccp *ccp = fsm2ccp(fp);
log_Printf(LogCCP, "%s: LayerStart.\n", fp->link->name);
fp->more.reqs = fp->more.naks = fp->more.rejs = ccp->cfg.fsm.maxreq * 3;
1995-01-31 06:29:58 +00:00
}
static void
CcpLayerDown(struct fsm *fp)
1995-01-31 06:29:58 +00:00
{
/* About to come down */
struct ccp *ccp = fsm2ccp(fp);
struct ccp_opt *next;
log_Printf(LogCCP, "%s: LayerDown.\n", fp->link->name);
if (ccp->in.state != NULL) {
(*algorithm[ccp->in.algorithm]->i.Term)(ccp->in.state);
ccp->in.state = NULL;
ccp->in.algorithm = -1;
}
if (ccp->out.state != NULL) {
(*algorithm[ccp->out.algorithm]->o.Term)(ccp->out.state);
ccp->out.state = NULL;
ccp->out.algorithm = -1;
}
ccp->his_reject = ccp->my_reject = 0;
while (ccp->out.opt) {
next = ccp->out.opt->next;
free(ccp->out.opt);
ccp->out.opt = next;
}
ccp_Setup(ccp);
1995-01-31 06:29:58 +00:00
}
static void
CcpLayerFinish(struct fsm *fp)
1995-01-31 06:29:58 +00:00
{
/* We're now down */
log_Printf(LogCCP, "%s: LayerFinish.\n", fp->link->name);
1995-01-31 06:29:58 +00:00
}
/* Called when CCP has reached the OPEN state */
static int
CcpLayerUp(struct fsm *fp)
1995-01-31 06:29:58 +00:00
{
/* We're now up */
struct ccp *ccp = fsm2ccp(fp);
log_Printf(LogCCP, "%s: LayerUp.\n", fp->link->name);
if (ccp->in.state == NULL && ccp->in.algorithm >= 0 &&
ccp->in.algorithm < NALGORITHMS) {
ccp->in.state = (*algorithm[ccp->in.algorithm]->i.Init)(&ccp->in.opt);
if (ccp->in.state == NULL) {
log_Printf(LogERROR, "%s: %s (in) initialisation failure\n",
fp->link->name, protoname(ccp->his_proto));
ccp->his_proto = ccp->my_proto = -1;
fsm_Close(fp);
return 0;
}
}
1995-01-31 06:29:58 +00:00
if (ccp->out.state == NULL && ccp->out.algorithm >= 0 &&
ccp->out.algorithm < NALGORITHMS) {
ccp->out.state = (*algorithm[ccp->out.algorithm]->o.Init)
(&ccp->out.opt->val);
if (ccp->out.state == NULL) {
log_Printf(LogERROR, "%s: %s (out) initialisation failure\n",
fp->link->name, protoname(ccp->my_proto));
ccp->his_proto = ccp->my_proto = -1;
fsm_Close(fp);
return 0;
}
}
fp->more.reqs = fp->more.naks = fp->more.rejs = ccp->cfg.fsm.maxreq * 3;
log_Printf(LogCCP, "%s: Out = %s[%d], In = %s[%d]\n",
fp->link->name, protoname(ccp->my_proto), ccp->my_proto,
protoname(ccp->his_proto), ccp->his_proto);
return 1;
1995-01-31 06:29:58 +00:00
}
static void
1998-03-13 21:08:05 +00:00
CcpDecodeConfig(struct fsm *fp, u_char *cp, int plen, int mode_type,
struct fsm_decode *dec)
1995-01-31 06:29:58 +00:00
{
/* Deal with incoming data */
struct ccp *ccp = fsm2ccp(fp);
int type, length, f;
const char *end;
1995-01-31 06:29:58 +00:00
if (mode_type == MODE_REQ)
ccp->in.algorithm = -1; /* In case we've received two REQs in a row */
1995-01-31 06:29:58 +00:00
while (plen >= sizeof(struct fsmconfig)) {
type = *cp;
length = cp[1];
if (length == 0) {
log_Printf(LogCCP, "%s: CCP size zero\n", fp->link->name);
break;
}
if (length > sizeof(struct lcp_opt)) {
length = sizeof(struct lcp_opt);
log_Printf(LogCCP, "%s: Warning: Truncating length to %d\n",
fp->link->name, length);
}
for (f = NALGORITHMS-1; f > -1; f--)
if (algorithm[f]->id == type)
break;
end = f == -1 ? "" : (*algorithm[f]->Disp)((struct lcp_opt *)cp);
if (end == NULL)
end = "";
log_Printf(LogCCP, " %s[%d] %s\n", protoname(type), length, end);
if (f == -1) {
/* Don't understand that :-( */
if (mode_type == MODE_REQ) {
ccp->my_reject |= (1 << type);
1998-03-13 21:08:05 +00:00
memcpy(dec->rejend, cp, length);
dec->rejend += length;
}
} else {
struct ccp_opt *o;
1995-01-31 06:29:58 +00:00
switch (mode_type) {
1995-01-31 06:29:58 +00:00
case MODE_REQ:
if (IsAccepted(ccp->cfg.neg[algorithm[f]->Neg]) &&
ccp->in.algorithm == -1) {
memcpy(&ccp->in.opt, cp, length);
switch ((*algorithm[f]->i.Set)(&ccp->in.opt, &ccp->cfg)) {
case MODE_REJ:
memcpy(dec->rejend, &ccp->in.opt, ccp->in.opt.len);
dec->rejend += ccp->in.opt.len;
break;
case MODE_NAK:
memcpy(dec->nakend, &ccp->in.opt, ccp->in.opt.len);
dec->nakend += ccp->in.opt.len;
break;
case MODE_ACK:
1998-03-13 21:08:05 +00:00
memcpy(dec->ackend, cp, length);
dec->ackend += length;
ccp->his_proto = type;
ccp->in.algorithm = f; /* This one'll do :-) */
break;
}
1995-01-31 06:29:58 +00:00
} else {
1998-03-13 21:08:05 +00:00
memcpy(dec->rejend, cp, length);
dec->rejend += length;
1995-01-31 06:29:58 +00:00
}
break;
case MODE_NAK:
for (o = ccp->out.opt; o != NULL; o = o->next)
if (o->val.id == cp[0])
break;
if (o == NULL)
log_Printf(LogCCP, "%s: Warning: Ignoring peer NAK of unsent"
" option\n", fp->link->name);
else {
memcpy(&o->val, cp, length);
if ((*algorithm[f]->o.Set)(&o->val) == MODE_ACK)
ccp->my_proto = algorithm[f]->id;
else {
ccp->his_reject |= (1 << type);
ccp->my_proto = -1;
}
}
break;
1995-01-31 06:29:58 +00:00
case MODE_REJ:
ccp->his_reject |= (1 << type);
ccp->my_proto = -1;
1995-01-31 06:29:58 +00:00
break;
}
}
plen -= cp[1];
cp += cp[1];
1995-01-31 06:29:58 +00:00
}
if (mode_type != MODE_NOP) {
if (dec->rejend != dec->rej) {
/* rejects are preferred */
dec->ackend = dec->ack;
dec->nakend = dec->nak;
if (ccp->in.state == NULL) {
ccp->his_proto = -1;
ccp->in.algorithm = -1;
}
} else if (dec->nakend != dec->nak) {
/* then NAKs */
dec->ackend = dec->ack;
if (ccp->in.state == NULL) {
ccp->his_proto = -1;
ccp->in.algorithm = -1;
}
}
}
1995-01-31 06:29:58 +00:00
}
extern struct mbuf *
ccp_Input(struct bundle *bundle, struct link *l, struct mbuf *bp)
1995-01-31 06:29:58 +00:00
{
/* Got PROTO_CCP from link */
m_settype(bp, MB_CCPIN);
if (bundle_Phase(bundle) == PHASE_NETWORK)
fsm_Input(&l->ccp.fsm, bp);
1995-01-31 06:29:58 +00:00
else {
if (bundle_Phase(bundle) < PHASE_NETWORK)
log_Printf(LogCCP, "%s: Error: Unexpected CCP in phase %s (ignored)\n",
l->ccp.fsm.link->name, bundle_PhaseName(bundle));
m_freem(bp);
1995-01-31 06:29:58 +00:00
}
return NULL;
1995-01-31 06:29:58 +00:00
}
static void
CcpRecvResetAck(struct fsm *fp, u_char id)
{
/* Got a reset ACK, reset incoming dictionary */
struct ccp *ccp = fsm2ccp(fp);
if (ccp->reset_sent != -1) {
if (id != ccp->reset_sent) {
log_Printf(LogCCP, "%s: Incorrect ResetAck (id %d, not %d)"
" ignored\n", fp->link->name, id, ccp->reset_sent);
return;
}
/* Whaddaya know - a correct reset ack */
} else if (id == ccp->last_reset)
log_Printf(LogCCP, "%s: Duplicate ResetAck (resetting again)\n",
fp->link->name);
else {
log_Printf(LogCCP, "%s: Unexpected ResetAck (id %d) ignored\n",
fp->link->name, id);
return;
}
ccp->last_reset = ccp->reset_sent;
ccp->reset_sent = -1;
if (ccp->in.state != NULL)
(*algorithm[ccp->in.algorithm]->i.Reset)(ccp->in.state);
}
static struct mbuf *
ccp_LayerPush(struct bundle *b, struct link *l, struct mbuf *bp,
int pri, u_short *proto)
{
if (PROTO_COMPRESSIBLE(*proto) && l->ccp.fsm.state == ST_OPENED &&
l->ccp.out.state != NULL) {
bp = (*algorithm[l->ccp.out.algorithm]->o.Write)
(l->ccp.out.state, &l->ccp, l, pri, proto, bp);
switch (*proto) {
case PROTO_ICOMPD:
m_settype(bp, MB_ICOMPDOUT);
break;
case PROTO_COMPD:
m_settype(bp, MB_COMPDOUT);
break;
}
}
return bp;
}
static struct mbuf *
ccp_LayerPull(struct bundle *b, struct link *l, struct mbuf *bp, u_short *proto)
{
/*
* If proto isn't PROTO_[I]COMPD, we still want to pass it to the
* decompression routines so that the dictionary's updated
*/
if (l->ccp.fsm.state == ST_OPENED) {
if (*proto == PROTO_COMPD || *proto == PROTO_ICOMPD) {
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, "ccp_LayerPull: PROTO_%sCOMPDP -> PROTO_IP\n",
*proto == PROTO_ICOMPD ? "I" : "");
/* Decompress incoming data */
if (l->ccp.reset_sent != -1)
/* Send another REQ and put the packet in the bit bucket */
fsm_Output(&l->ccp.fsm, CODE_RESETREQ, l->ccp.reset_sent, NULL, 0,
MB_CCPOUT);
else if (l->ccp.in.state != NULL) {
bp = (*algorithm[l->ccp.in.algorithm]->i.Read)
(l->ccp.in.state, &l->ccp, proto, bp);
switch (*proto) {
case PROTO_ICOMPD:
m_settype(bp, MB_ICOMPDIN);
break;
case PROTO_COMPD:
m_settype(bp, MB_COMPDIN);
break;
}
return bp;
}
m_freem(bp);
bp = NULL;
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
} else if (PROTO_COMPRESSIBLE(*proto) && l->ccp.in.state != NULL) {
log_Printf(LogDEBUG, "ccp_LayerPull: Ignore packet (dict only)\n");
/* Add incoming Network Layer traffic to our dictionary */
(*algorithm[l->ccp.in.algorithm]->i.DictSetup)
(l->ccp.in.state, &l->ccp, *proto, bp);
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
} else
log_Printf(LogDEBUG, "ccp_LayerPull: Ignore packet\n");
}
return bp;
}
u_short
ccp_Proto(struct ccp *ccp)
{
return !link2physical(ccp->fsm.link) || !ccp->fsm.bundle->ncp.mp.active ?
PROTO_COMPD : PROTO_ICOMPD;
}
int
ccp_SetOpenMode(struct ccp *ccp)
{
int f;
for (f = 0; f < CCP_NEG_TOTAL; f++)
if (IsEnabled(ccp->cfg.neg[f])) {
ccp->fsm.open_mode = 0;
return 1;
}
ccp->fsm.open_mode = OPEN_PASSIVE; /* Go straight to ST_STOPPED ? */
for (f = 0; f < CCP_NEG_TOTAL; f++)
if (IsAccepted(ccp->cfg.neg[f]))
return 1;
return 0; /* No CCP at all */
}
struct layer ccplayer = { LAYER_CCP, "ccp", ccp_LayerPush, ccp_LayerPull };