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

490 lines
12 KiB
C
Raw Normal View History

/*-
* Copyright (c) 1999 Brian Somers <brian@Awfulhak.org>
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* $Id: tty.c,v 1.2 1999/05/12 09:49:07 brian Exp $
*/
#include <sys/param.h>
#include <sys/socket.h>
#include <netinet/in.h>
#include <arpa/inet.h>
#include <netdb.h>
#include <netinet/in_systm.h>
#include <netinet/ip.h>
#include <sys/un.h>
#if defined(__OpenBSD__) || defined(__NetBSD__)
#include <sys/ioctl.h>
#include <util.h>
#else
#include <libutil.h>
#endif
#include <errno.h>
#include <fcntl.h>
#include <paths.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/wait.h>
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
#include <sys/uio.h>
#include <termios.h>
#include <unistd.h>
#include "layer.h"
#include "defs.h"
#include "mbuf.h"
#include "log.h"
#include "id.h"
#include "sync.h"
#include "timer.h"
#include "lqr.h"
#include "hdlc.h"
#include "throughput.h"
#include "fsm.h"
#include "lcp.h"
#include "ccp.h"
#include "link.h"
#include "async.h"
#include "slcompress.h"
#include "iplist.h"
#include "ipcp.h"
#include "filter.h"
#include "descriptor.h"
#include "physical.h"
#include "mp.h"
#ifndef NORADIUS
#include "radius.h"
#endif
#include "chat.h"
#include "command.h"
#include "bundle.h"
#include "prompt.h"
#include "auth.h"
#include "chap.h"
#include "cbcp.h"
#include "datalink.h"
#include "tty.h"
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
#define Online(dev) ((dev)->mbits & TIOCM_CD)
struct ttydevice {
struct device dev; /* What struct physical knows about */
struct pppTimer Timer; /* CD checks */
int mbits; /* Current DCD status */
struct termios ios; /* To be able to reset from raw mode */
};
#define device2tty(d) ((d)->type == TTY_DEVICE ? (struct ttydevice *)d : NULL)
static int
tty_Lock(struct physical *p, int tunno)
{
int res;
FILE *lockfile;
char fn[MAXPATHLEN];
if (*p->name.full != '/')
return 0;
if (p->type != PHYS_DIRECT &&
(res = ID0uu_lock(p->name.base)) != UU_LOCK_OK) {
if (res == UU_LOCK_INUSE)
log_Printf(LogPHASE, "%s: %s is in use\n", p->link.name, p->name.full);
else
log_Printf(LogPHASE, "%s: %s is in use: uu_lock: %s\n",
p->link.name, p->name.full, uu_lockerr(res));
return (-1);
}
snprintf(fn, sizeof fn, "%s%s.if", _PATH_VARRUN, p->name.base);
lockfile = ID0fopen(fn, "w");
if (lockfile != NULL) {
fprintf(lockfile, "%s%d\n", TUN_NAME, tunno);
fclose(lockfile);
}
#ifndef RELEASE_CRUNCH
else
log_Printf(LogALERT, "%s: Can't create %s: %s\n",
p->link.name, fn, strerror(errno));
#endif
return 0;
}
static void
tty_Unlock(struct physical *p)
{
char fn[MAXPATHLEN];
if (*p->name.full != '/')
return;
snprintf(fn, sizeof fn, "%s%s.if", _PATH_VARRUN, p->name.base);
#ifndef RELEASE_CRUNCH
if (ID0unlink(fn) == -1)
log_Printf(LogALERT, "%s: Can't remove %s: %s\n",
p->link.name, fn, strerror(errno));
#else
ID0unlink(fn);
#endif
if (p->type != PHYS_DIRECT && ID0uu_unlock(p->name.base) == -1)
log_Printf(LogALERT, "%s: Can't uu_unlock %s\n", p->link.name, fn);
}
static struct device *
tty_SetupDevice(struct physical *p)
{
struct ttydevice *dev;
struct termios ios;
int oldflag;
if ((dev = malloc(sizeof *dev)) == NULL)
return NULL;
tcgetattr(p->fd, &ios);
dev->ios = ios;
log_Printf(LogDEBUG, "%s: tty_SetupDevice: physical (get): fd = %d,"
" iflag = %lx, oflag = %lx, cflag = %lx\n", p->link.name, p->fd,
(u_long)ios.c_iflag, (u_long)ios.c_oflag, (u_long)ios.c_cflag);
cfmakeraw(&ios);
if (p->cfg.rts_cts)
ios.c_cflag |= CLOCAL | CCTS_OFLOW | CRTS_IFLOW;
else {
ios.c_cflag |= CLOCAL;
ios.c_iflag |= IXOFF;
}
ios.c_iflag |= IXON;
if (p->type != PHYS_DEDICATED)
ios.c_cflag |= HUPCL;
if (p->type != PHYS_DIRECT) {
/* Change tty speed when we're not in -direct mode */
ios.c_cflag &= ~(CSIZE | PARODD | PARENB);
ios.c_cflag |= p->cfg.parity;
if (cfsetspeed(&ios, IntToSpeed(p->cfg.speed)) == -1)
log_Printf(LogWARN, "%s: %s: Unable to set speed to %d\n",
p->link.name, p->name.full, p->cfg.speed);
}
tcsetattr(p->fd, TCSADRAIN, &ios);
log_Printf(LogDEBUG, "%s: physical (put): iflag = %lx, oflag = %lx, "
"cflag = %lx\n", p->link.name, (u_long)ios.c_iflag,
(u_long)ios.c_oflag, (u_long)ios.c_cflag);
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
if (ioctl(p->fd, TIOCMGET, &dev->mbits) == -1) {
if (p->type != PHYS_DIRECT) {
log_Printf(LogWARN, "%s: Open: Cannot get physical status: %s\n",
p->link.name, strerror(errno));
physical_Close(p);
return NULL;
} else
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
dev->mbits = TIOCM_CD;
}
log_Printf(LogDEBUG, "%s: Open: physical control = %o\n",
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
p->link.name, dev->mbits);
oldflag = fcntl(p->fd, F_GETFL, 0);
if (oldflag < 0) {
log_Printf(LogWARN, "%s: Open: Cannot get physical flags: %s\n",
p->link.name, strerror(errno));
physical_Close(p);
return NULL;
} else
fcntl(p->fd, F_SETFL, oldflag & ~O_NONBLOCK);
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
physical_SetupStack(p, PHYSICAL_NOFORCE);
return &dev->dev;
}
/*
* tty_Timeout() watches the DCD signal and mentions it if it's status
* changes.
*/
static void
tty_Timeout(void *data)
{
struct physical *p = data;
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
struct ttydevice *dev = device2tty(p->handler);
int ombits, change;
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
timer_Stop(&dev->Timer);
dev->Timer.load = SECTICKS; /* Once a second please */
timer_Start(&dev->Timer);
ombits = dev->mbits;
if (p->fd >= 0) {
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
if (ioctl(p->fd, TIOCMGET, &dev->mbits) < 0) {
log_Printf(LogPHASE, "%s: ioctl error (%s)!\n", p->link.name,
strerror(errno));
datalink_Down(p->dl, CLOSE_NORMAL);
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
timer_Stop(&dev->Timer);
return;
}
} else
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
dev->mbits = 0;
if (ombits == -1) {
/* First time looking for carrier */
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
if (Online(dev))
log_Printf(LogDEBUG, "%s: %s: CD detected\n", p->link.name, p->name.full);
else if (p->cfg.cd.required) {
log_Printf(LogPHASE, "%s: %s: Required CD not detected\n",
p->link.name, p->name.full);
datalink_Down(p->dl, CLOSE_NORMAL);
} else {
log_Printf(LogPHASE, "%s: %s doesn't support CD\n",
p->link.name, p->name.full);
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
timer_Stop(&dev->Timer);
dev->mbits = TIOCM_CD;
}
} else {
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
change = ombits ^ dev->mbits;
if (change & TIOCM_CD) {
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
if (dev->mbits & TIOCM_CD)
log_Printf(LogDEBUG, "%s: offline -> online\n", p->link.name);
else {
log_Printf(LogDEBUG, "%s: online -> offline\n", p->link.name);
log_Printf(LogPHASE, "%s: Carrier lost\n", p->link.name);
datalink_Down(p->dl, CLOSE_NORMAL);
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
timer_Stop(&dev->Timer);
}
} else
log_Printf(LogDEBUG, "%s: Still %sline\n", p->link.name,
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
Online(dev) ? "on" : "off");
}
}
static void
tty_StartTimer(struct physical *p)
{
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
struct ttydevice *dev = device2tty(p->handler);
timer_Stop(&dev->Timer);
dev->Timer.load = SECTICKS * p->cfg.cd.delay;
dev->Timer.func = tty_Timeout;
dev->Timer.name = "tty CD";
dev->Timer.arg = p;
log_Printf(LogDEBUG, "%s: Using tty_Timeout [%p]\n",
p->link.name, tty_Timeout);
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
dev->mbits = -1; /* So we know it's the first time */
timer_Start(&dev->Timer);
}
static int
tty_Raw(struct physical *p)
{
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
struct ttydevice *dev = device2tty(p->handler);
struct termios ios;
int oldflag;
if (physical_IsSync(p))
return 1;
log_Printf(LogDEBUG, "%s: Entering physical_Raw\n", p->link.name);
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
if (p->type != PHYS_DIRECT && p->fd >= 0 && !Online(dev))
log_Printf(LogDEBUG, "%s: Raw: descriptor = %d, mbits = %x\n",
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
p->link.name, p->fd, dev->mbits);
tcgetattr(p->fd, &ios);
cfmakeraw(&ios);
if (p->cfg.rts_cts)
ios.c_cflag |= CLOCAL | CCTS_OFLOW | CRTS_IFLOW;
else
ios.c_cflag |= CLOCAL;
if (p->type != PHYS_DEDICATED)
ios.c_cflag |= HUPCL;
tcsetattr(p->fd, TCSANOW, &ios);
oldflag = fcntl(p->fd, F_GETFL, 0);
if (oldflag < 0)
return 0;
fcntl(p->fd, F_SETFL, oldflag | O_NONBLOCK);
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
if (ioctl(p->fd, TIOCMGET, &dev->mbits) == 0)
tty_StartTimer(p);
return 1;
}
static void
tty_Offline(struct physical *p)
{
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
struct ttydevice *dev = device2tty(p->handler);
if (p->fd >= 0) {
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
timer_Stop(&dev->Timer);
dev->mbits &= ~TIOCM_DTR;
if (Online(dev)) {
struct termios tio;
tcgetattr(p->fd, &tio);
if (cfsetspeed(&tio, B0) == -1)
log_Printf(LogWARN, "%s: Unable to set physical to speed 0\n",
p->link.name);
else
tcsetattr(p->fd, TCSANOW, &tio);
}
}
}
static void
tty_Cooked(struct physical *p)
{
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
struct ttydevice *dev = device2tty(p->handler);
int oldflag;
tcflush(p->fd, TCIOFLUSH);
if (!physical_IsSync(p)) {
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
tcsetattr(p->fd, TCSAFLUSH, &dev->ios);
oldflag = fcntl(p->fd, F_GETFL, 0);
if (oldflag == 0)
fcntl(p->fd, F_SETFL, oldflag & ~O_NONBLOCK);
}
}
static void
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
tty_StopTimer(struct physical *p)
{
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
struct ttydevice *dev = device2tty(p->handler);
timer_Stop(&dev->Timer);
}
static void
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
tty_Free(struct physical *p)
{
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
struct ttydevice *dev = device2tty(p->handler);
tty_Unlock(p);
free(dev);
}
static int
tty_Speed(struct physical *p)
{
struct termios ios;
if (tcgetattr(p->fd, &ios) == -1)
return 0;
return SpeedToInt(cfgetispeed(&ios));
}
static const char *
tty_OpenInfo(struct physical *p)
{
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
struct ttydevice *dev = device2tty(p->handler);
static char buf[13];
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
if (Online(dev))
strcpy(buf, "with");
else
strcpy(buf, "no");
strcat(buf, " carrier");
return buf;
}
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
static void
tty_device2iov(struct physical *p, struct iovec *iov, int *niov,
int maxiov, pid_t newpid)
{
struct ttydevice *dev = p ? device2tty(p->handler) : NULL;
iov[*niov].iov_base = p ? p->handler : malloc(sizeof(struct ttydevice));
iov[*niov].iov_len = sizeof(struct ttydevice);
(*niov)++;
if (dev->Timer.state != TIMER_STOPPED) {
timer_Stop(&dev->Timer);
dev->Timer.state = TIMER_RUNNING;
}
}
static struct device basettydevice = {
TTY_DEVICE,
"tty",
tty_Raw,
tty_Offline,
tty_Cooked,
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
tty_StopTimer,
tty_Free,
NULL,
NULL,
tty_device2iov,
tty_Speed,
tty_OpenInfo
};
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
struct device *
tty_iov2device(int type, struct physical *p, struct iovec *iov, int *niov,
int maxiov)
{
if (type == TTY_DEVICE) {
struct ttydevice *dev;
/* It's one of ours ! Let's create the device */
dev = (struct ttydevice *)iov[(*niov)++].iov_base;
/* Refresh function pointers etc */
memcpy(&dev->dev, &basettydevice, sizeof dev->dev);
physical_SetupStack(p, PHYSICAL_NOFORCE);
if (dev->Timer.state != TIMER_STOPPED) {
dev->Timer.state = TIMER_STOPPED;
tty_StartTimer(p);
}
return &dev->dev;
}
return NULL;
}
struct device *
tty_Create(struct physical *p)
{
if (p->fd >= 0 && isatty(p->fd)) {
if (*p->name.full == '\0') {
log_Printf(LogDEBUG, "%s: Input is a tty\n", p->link.name);
physical_SetDevice(p, ttyname(p->fd));
if (tty_Lock(p, p->dl->bundle->unit) == -1) {
close(p->fd);
p->fd = -1;
} else
return tty_SetupDevice(p);
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 (tty_Lock(p, p->dl->bundle->unit) != -1) {
log_Printf(LogDEBUG, "%s: Opened %s\n", p->link.name, p->name.full);
return tty_SetupDevice(p);
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
}
}
return NULL;
}