From 270f71442ea40e85efbf81b3a257f1cd0af43643 Mon Sep 17 00:00:00 2001 From: Mike Karels Date: Sat, 1 Jul 1989 01:16:01 -0800 Subject: [PATCH] merge new code from van, sundry cleanups SCCS-vsn: sys/net/if_sl.c 7.16 SCCS-vsn: sys/net/if_slvar.h 7.3 SCCS-vsn: sys/net/slcompress.h 7.2 --- usr/src/sys/net/if_sl.c | 528 ++++++++++++++++++----------------- usr/src/sys/net/if_slvar.h | 5 +- usr/src/sys/net/slcompress.h | 35 ++- 3 files changed, 295 insertions(+), 273 deletions(-) diff --git a/usr/src/sys/net/if_sl.c b/usr/src/sys/net/if_sl.c index 3374c7ff47..9174150370 100644 --- a/usr/src/sys/net/if_sl.c +++ b/usr/src/sys/net/if_sl.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 1987 Regents of the University of California. + * Copyright (c) 1987, 1989 Regents of the University of California. * All rights reserved. * * Redistribution and use in source and binary forms are permitted @@ -14,7 +14,7 @@ * IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED * WARRANTIES OF MERCHANTIBILITY AND FITNESS FOR A PARTICULAR PURPOSE. * - * @(#)if_sl.c 7.15 (Berkeley) %G% + * @(#)if_sl.c 7.16 (Berkeley) %G% */ /* @@ -35,11 +35,20 @@ * * Converted to 4.3BSD Beta by Chris Torek. * Other changes made at Berkeley, based in part on code by Kirk Smith. - * W. Jolitz, added slip abort & time domain window - * also added Van Jacobson's hdr compression code + * W. Jolitz added slip abort. + * + * Hacked almost beyond recognition by Van Jacobson (van@helios.ee.lbl.gov). + * Added priority queuing for "interactive" traffic; hooks for TCP + * header compression; ICMP filtering (at 2400 baud, some cretin + * pinging you can use up all your bandwidth). Made low clist behavior + * more robust and slightly less likely to hang serial line. + * Sped up a bunch of things. + * + * Note that splimp() is used throughout to block both (tty) input + * interrupts and network activity; thus, splimp must be >= spltty. */ -/* $Header: if_sl.c,v 1.12 85/12/20 21:54:55 chris Exp $ */ +/* $Header: if_sl.c,v 1.7 89/05/31 02:24:52 van Exp $ */ /* from if_sl.c,v 1.11 84/10/04 12:54:47 rick Exp */ #include "sl.h" @@ -67,62 +76,74 @@ #include "../netinet/in_systm.h" #include "../netinet/in_var.h" #include "../netinet/ip.h" -#include "slcompress.h" #endif -#include "if_slvar.h" #include "machine/mtpr.h" +#include "slcompress.h" +#include "if_slvar.h" + /* - * N.B.: SLMTU is now a hard limit on input packet size. - * SLMTU must be <= MCLBYTES - sizeof(struct ifnet *). + * SLMTU is a hard limit on input packet size. To simplify the code + * and improve performance, we require that packets fit in an mbuf + * cluster, that there be enough extra room for the ifnet pointer that + * IP input requires and, if we get a compressed packet, there's + * enough extra room to expand the header into a max length tcp/ip + * header (128 bytes). So, SLMTU can be at most + * MCLBYTES - sizeof(struct ifnet *) - 128 + * + * To insure we get good interactive response, the MTU wants to be + * the smallest size that amortizes the header cost. (Remember + * that even with type-of-service queuing, we have to wait for any + * in-progress packet to finish. I.e., we wait, on the average, + * 1/2 * mtu / cps, where cps is the line speed in characters per + * second. E.g., 533ms wait for a 1024 byte MTU on a 9600 baud + * line. The average compressed header size is 6-8 bytes so any + * MTU > 90 bytes will give us 90% of the line bandwidth. A 100ms + * wait is tolerable (500ms is not), so want an MTU around 256. + * (Since TCP will send 212 byte segments (to allow for 40 byte + * headers), the typical packet size on the wire will be around 220 + * bytes). In 4.3tahoe+ systems, we can set an MTU in a route + * so we do that & leave the interface MTU relatively high (so we + * don't IP fragment when acting as a gateway to someone using a + * stupid MTU). */ -#define SLMTU 1006 -#define SLIP_HIWAT 1000 /* don't start a new packet if HIWAT on queue */ -#define CLISTRESERVE 1000 /* Can't let clists get too low */ - +#define SLMTU 576 +#define BUFOFFSET (128+sizeof(struct ifnet **)) +#define SLIP_HIWAT 1024 /* don't start a new packet if HIWAT on queue */ +#define CLISTRESERVE 1024 /* Can't let clists get too low */ /* * SLIP ABORT ESCAPE MECHANISM: * (inspired by HAYES modem escape arrangement) * 1sec escape 1sec escape 1sec escape { 1sec escape 1sec escape } * signals a "soft" exit from slip mode by usermode process - * (hard exit unimplemented -- currently system dependant) */ #define ABT_ESC '\033' /* can't be t_intr - distant host must know it*/ #define ABT_WAIT 1 /* in seconds - idle before an escape & after */ #define ABT_RECYCLE (5*2+2) /* in seconds - time window processing abort */ -/* a "soft" abort means to pass a suggestion to user code to abort slip */ #define ABT_SOFT 3 /* count of escapes */ -/* a "hard" abort means to force abort slip within the kernel -- process jam? */ -#define ABT_HARD 5 /* count of escapes */ - /* - * SLIP TIME WINDOW: - * Only accept packets with octets that come at least this often. - * With non-reliable but fast modems (FAX, Packet Radio), we assume that - * packets come in groups (time domain), and that fractional groups that - * come erratically are just noise that will foul subsequent packets. - * We reject them on a time filter basis. - * - * This is a very coarse filter, because error correcting modems like the - * telebit take there own sweet time encoding/decoding packets. If you - * are using an MNP,PEP or other such arrangement, this won't help much. - * If you are using packet radio, use the millisecond time val with - * as small a resolution as possible. In any case, the coarse filter - * saves noisey lines about 50 % of the time. + * The following disgusting hack gets around the problem that IP TOS + * can't be set in BSD/Sun OS yet. We want to put "interactive" + * traffic on a high priority queue. To decide if traffic is + * interactive, we check that a) it is TCP and b) one of it's ports + * if telnet, rlogin or ftp control. */ +static u_short interactive_ports[8] = { + 0, 513, 0, 0, + 0, 21, 0, 23, +}; +#define INTERACTIVE(p) (interactive_ports[(p) & 7] == (p)) -#define TIME_WINDOW 2 /* max seconds between valid packet chars */ +struct sl_softc sl_softc[NSL]; -struct sl_softc sl_softc[NSL]; - -#define FRAME_END 0300 /* Frame End */ -#define FRAME_ESCAPE 0333 /* Frame Esc */ -#define TRANS_FRAME_END 0334 /* transposed frame end */ -#define TRANS_FRAME_ESCAPE 0335 /* transposed frame esc */ +#define FRAME_END 0xc0 /* Frame End */ +#define FRAME_ESCAPE 0xdb /* Frame Esc */ +#define TRANS_FRAME_END 0xdc /* transposed frame end */ +#define TRANS_FRAME_ESCAPE 0xdd /* transposed frame esc */ #define t_sc T_LINEP @@ -143,11 +164,34 @@ slattach() sc->sc_if.if_flags = IFF_POINTOPOINT; sc->sc_if.if_ioctl = slioctl; sc->sc_if.if_output = sloutput; - sc->sc_if.if_snd.ifq_maxlen = IFQ_MAXLEN; + sc->sc_if.if_snd.ifq_maxlen = 50; + sc->sc_fastq.ifq_maxlen = 32; if_attach(&sc->sc_if); } } +static int +slinit(sc) + register struct sl_softc *sc; +{ + register caddr_t p; + + if (sc->sc_ep == (u_char *) 0) { + MCLALLOC(p, M_WAIT); + if (p) + sc->sc_ep = (u_char *)p + (BUFOFFSET + SLMTU); + else { + printf("sl%d: can't allocate buffer\n", sc - sl_softc); + sc->sc_if.if_flags &= ~IFF_UP; + return (0); + } + } + sc->sc_buf = sc->sc_ep - SLMTU; + sc->sc_mp = sc->sc_buf; + sl_compress_init(&sc->sc_comp); + return (1); +} + /* * Line specific open routine. * Attach the given tty to the first available sl unit. @@ -163,24 +207,19 @@ slopen(dev, tp) if (error = suser(u.u_cred, &u.u_acflag)) return (error); + if (tp->t_line == SLIPDISC) return (EBUSY); - for (nsl = 0, sc = sl_softc; nsl < NSL; nsl++, sc++) + for (nsl = NSL, sc = sl_softc; --nsl >= 0; sc++) if (sc->sc_ttyp == NULL) { - sc->sc_flags = 0; - sc->sc_ilen = 0; if (slinit(sc) == 0) return (ENOBUFS); -#ifdef INET - sl_compress_init(&sc->sc_comp); -#endif tp->t_sc = (caddr_t)sc; sc->sc_ttyp = tp; ttyflush(tp, FREAD | FWRITE); return (0); } - return (ENXIO); } @@ -196,16 +235,17 @@ slclose(tp) int s; ttywflush(tp); + s = splimp(); /* actually, max(spltty, splnet) */ tp->t_line = 0; - s = splimp(); /* paranoid; splnet probably ok */ sc = (struct sl_softc *)tp->t_sc; if (sc != NULL) { if_down(&sc->sc_if); sc->sc_ttyp = NULL; tp->t_sc = NULL; - MCLFREE((struct mbuf *)sc->sc_buf); + MCLFREE((struct mbuf *)(sc->sc_ep - (SLMTU + BUFOFFSET))); + sc->sc_ep = 0; + sc->sc_mp = 0; sc->sc_buf = 0; - sc->sc_mp = (char *) 4; /*XXX!?! */ } splx(s); } @@ -223,20 +263,15 @@ sltioctl(tp, cmd, data, flag) int s; switch (cmd) { - case TIOCGETD: + case TIOCGETD: /* XXX */ + case SLIOGUNIT: *(int *)data = sc->sc_if.if_unit; break; - case TIOCMGET: - if (tp->t_state&TS_CARR_ON) - *(int *)data = TIOCM_CAR ; - else *(int *)data = 0 ; - if (sc->sc_flags&SC_ABORT) - *(int *)data |= TIOCM_DTR ; - break; case SLIOCGFLAGS: *(int *)data = sc->sc_flags; break; + case SLIOCSFLAGS: #define SC_MASK (SC_COMPRESS|SC_NOICMP) s = splimp(); @@ -244,6 +279,7 @@ sltioctl(tp, cmd, data, flag) (sc->sc_flags &~ SC_MASK) | ((*(int *)data) & SC_MASK); splx(s); break; + default: return (-1); } @@ -259,6 +295,8 @@ sloutput(ifp, m, dst) struct sockaddr *dst; { register struct sl_softc *sc; + register struct ip *ip; + register struct ifqueue *ifq; int s; /* @@ -281,30 +319,37 @@ sloutput(ifp, m, dst) m_freem(m); return (EHOSTUNREACH); } - s = splimp(); -#ifdef INET - if (sc->sc_flags & (SC_COMPRESS|SC_NOICMP)) { - register struct ip *ip = mtod(m, struct ip *); - if (ip->ip_p == IPPROTO_TCP) { - /* add stuff to TOS routing */ - if (sc->sc_flags & SC_COMPRESS) - (void) sl_compress_tcp(m, ip, &sc->sc_comp); - } else if ((sc->sc_flags & SC_NOICMP) && - ip->ip_p == IPPROTO_ICMP) { - m_freem(m); - splx(s); - return (0); + ifq = &ifp->if_snd; + if ((ip = mtod(m, struct ip *))->ip_p == IPPROTO_TCP) { + register int p = ((int *)ip)[ip->ip_hl]; + + if (INTERACTIVE(p & 0xffff) || INTERACTIVE(p >> 16)) + ifq = &sc->sc_fastq; + + if (sc->sc_flags & SC_COMPRESS) { + /* if two copies of sl_compress_tcp are running + * for the same line, the compression state can + * get screwed up. We're assuming that sloutput + * was invoked at splnet so this isn't possible + * (this assumption is correct for 4.xbsd, x<=4). + * In a multi-threaded kernel, a lockout might + * be needed here. */ + p = sl_compress_tcp(m, ip, &sc->sc_comp); + *mtod(m, u_char *) |= p; } + } else if (sc->sc_flags & SC_NOICMP && ip->ip_p == IPPROTO_ICMP) { + m_freem(m); + return (0); } -#endif - if (IF_QFULL(&ifp->if_snd)) { - IF_DROP(&ifp->if_snd); - splx(s); + s = splimp(); + if (IF_QFULL(ifq)) { + IF_DROP(ifq); m_freem(m); + splx(s); sc->sc_if.if_oerrors++; return (ENOBUFS); } - IF_ENQUEUE(&ifp->if_snd, m); + IF_ENQUEUE(ifq, m); if (sc->sc_ttyp->t_outq.c_cc == 0) { splx(s); slstart(sc->sc_ttyp); @@ -323,9 +368,8 @@ slstart(tp) { register struct sl_softc *sc = (struct sl_softc *)tp->t_sc; register struct mbuf *m; - register int len; register u_char *cp; - int nd, np, n, s; + int s; struct mbuf *m2; extern int cfreecount; @@ -335,78 +379,92 @@ slstart(tp) * We are being called in lieu of ttstart and must do what * it would. */ - if (tp->t_outq.c_cc > 0) - ttstart(tp); - if (tp->t_outq.c_cc > SLIP_HIWAT) - return; - + if (tp->t_outq.c_cc != 0) { + (*tp->t_oproc)(tp); + if (tp->t_outq.c_cc > SLIP_HIWAT) + return; + } /* * This happens briefly when the line shuts down. */ if (sc == NULL) return; - /* - * If system is getting low on clists - * and we have something running already, stop here. - */ - if (cfreecount < CLISTRESERVE + SLMTU && tp->t_outq.c_cc) - return; - /* * Get a packet and send it to the interface. */ s = splimp(); - IF_DEQUEUE(&sc->sc_if.if_snd, m); + IF_DEQUEUE(&sc->sc_fastq, m); + if (m == NULL) + IF_DEQUEUE(&sc->sc_if.if_snd, m); splx(s); if (m == NULL) return; + /* + * If system is getting low on clists, just flush our + * output queue (if the stuff was important, it'll get + * retransmitted). + */ + if (cfreecount < CLISTRESERVE + SLMTU) { + m_freem(m); + sc->sc_if.if_collisions++; + continue; + } /* * The extra FRAME_END will start up a new packet, and thus * will flush any accumulated garbage. We do this whenever * the line may have been idle for some time. */ - if (tp->t_outq.c_cc == 0) + if (tp->t_outq.c_cc == 0) { + ++sc->sc_bytessent; (void) putc(FRAME_END, &tp->t_outq); + } while (m) { - cp = mtod(m, u_char *); - len = m->m_len; - while (len > 0) { + register u_char *ep; + + cp = mtod(m, u_char *); ep = cp + m->m_len; + while (cp < ep) { /* * Find out how many bytes in the string we can * handle without doing something special. */ - nd = locc(FRAME_ESCAPE, len, cp); - np = locc(FRAME_END, len, cp); - n = len - MAX(nd, np); - if (n) { + register u_char *bp = cp; + + while (cp < ep) { + switch (*cp++) { + case FRAME_ESCAPE: + case FRAME_END: + --cp; + goto out; + } + } + out: + if (cp > bp) { /* * Put n characters at once * into the tty output queue. */ - if (b_to_q((char *)cp, n, &tp->t_outq)) + if (b_to_q((char *)bp, cp - bp, &tp->t_outq)) break; - len -= n; - cp += n; + sc->sc_bytessent += cp - bp; } /* * If there are characters left in the mbuf, * the first one must be special.. * Put it out in a different form. */ - if (len) { + if (cp < ep) { if (putc(FRAME_ESCAPE, &tp->t_outq)) break; - if (putc(*cp == FRAME_ESCAPE ? + if (putc(*cp++ == FRAME_ESCAPE ? TRANS_FRAME_ESCAPE : TRANS_FRAME_END, &tp->t_outq)) { (void) unputc(&tp->t_outq); break; } - cp++; - len--; + sc->sc_bytessent += 2; } } MFREE(m, m2); @@ -424,88 +482,53 @@ slstart(tp) (void) unputc(&tp->t_outq); (void) putc(FRAME_END, &tp->t_outq); sc->sc_if.if_collisions++; - } else - sc->sc_if.if_opackets++; - } -} - -slinit(sc) - register struct sl_softc *sc; -{ - register caddr_t p; - - if (sc->sc_buf == (char *) 0) { - MCLALLOC(p, M_WAIT); - if (p) { - sc->sc_buf = p; - sc->sc_mp = p; } else { - printf("sl%d: can't allocate buffer\n", sc - sl_softc); - sc->sc_if.if_flags &= ~IFF_UP; - return (0); + ++sc->sc_bytessent; + sc->sc_if.if_opackets++; } } - return (1); } /* - * Copy data buffer to mbuf chain; add ifnet pointer ifp. + * Copy data buffer to mbuf chain; add ifnet pointer. */ -struct mbuf * -sl_btom(sc, len, ifp) - struct sl_softc *sc; +static struct mbuf * +sl_btom(sc, len) + register struct sl_softc *sc; register int len; - struct ifnet *ifp; { - register caddr_t cp; - register struct mbuf *m, **mp; - register unsigned count; - struct mbuf *top = NULL; + register u_char *cp; + register struct mbuf *m; - cp = sc->sc_buf; - mp = ⊤ + MGETHDR(m, M_DONTWAIT, MT_DATA); + if (m == NULL) + return (NULL); - while (len > 0) { - if (top == NULL) { - MGETHDR(m, M_DONTWAIT, MT_DATA); - } else { - MGET(m, M_DONTWAIT, MT_DATA); - } - if (m == NULL) { - m_freem(top); + /* + * If we have more than MLEN bytes, it's cheaper to + * queue the cluster we just filled & allocate a new one + * for the input buffer. Otherwise, fill the mbuf we + * allocated above. Note that code in the input routine + * guarantees that packet will fit in a cluster. + */ + cp = sc->sc_buf; + if (len >= MHLEN) { + MCLGET(m, M_DONTWAIT); + if ((m->m_flags & M_EXT) == 0) { + /* we couldn't get a cluster - if memory's this + * low, it's time to start dropping packets. */ + m_freem(m); return (NULL); } - if (top == NULL) { - m->m_pkthdr.rcvif = ifp; - m->m_pkthdr.len = len; - m->m_len = MHLEN; - } else - m->m_len = MLEN; - *mp = m; - /* - * If we have at least MINCLSIZE bytes, - * allocate a new page. Swap the current - * buffer page with the new one. - */ - if (len >= MINCLSIZE) { - MCLGET(m, M_DONTWAIT); - if (m->m_flags & M_EXT) { - cp = mtod(m, char *); - m->m_data = sc->sc_buf; - sc->sc_buf = cp; - count = MIN(len, MCLBYTES); - goto nocopy; - } - } - count = MIN(len, m->m_len); - bcopy(cp, mtod(m, caddr_t), count); -nocopy: - m->m_len = count; - cp += count; - len -= count; - mp = &m->m_next; - } - return (top); + sc->sc_ep = mtod(m, u_char *) + (BUFOFFSET + SLMTU); + m->m_data = (caddr_t)cp; + } else + bcopy((caddr_t)cp, mtod(m, caddr_t), len); + + m->m_len = len; + m->m_pkthdr.len = len; + m->m_pkthdr.rcvif = &sc->sc_if; + return (m); } /* @@ -517,106 +540,105 @@ slinput(c, tp) { register struct sl_softc *sc; register struct mbuf *m; + register int len; int s; tk_nin++; sc = (struct sl_softc *)tp->t_sc; if (sc == NULL) return; - if (!(tp->t_state&TS_CARR_ON)) /*XXX*/ + if (!(tp->t_state&TS_CARR_ON)) /* XXX */ return; - c &= 0xff; - - /* if we see an abort after "idle" time, count it */ - if ((c&0x7f) == ABT_ESC && time.tv_sec >= sc->sc_lasttime + ABT_WAIT) { - sc->sc_abortcount++; - /* record when the first abort escape arrived */ - if (sc->sc_abortcount == 1) sc->sc_starttime = time.tv_sec; - } - - /* if we have an abort, see that we have not run out of time, or - that we have an "idle" time after the complete escape sequence */ - if (sc->sc_abortcount) { - if (time.tv_sec >= sc->sc_starttime + ABT_RECYCLE) - sc->sc_abortcount = 0; - if (sc->sc_abortcount >= ABT_SOFT - && time.tv_sec >= sc->sc_lasttime + ABT_WAIT) - sc->sc_flags |= SC_ABORT; - } - - if (sc->sc_ilen && time.tv_sec >= sc->sc_lasttime + TIME_WINDOW) { - sc->sc_flags &= ~SC_ESCAPED; - sc->sc_mp = sc->sc_buf; - sc->sc_ilen = 0; - sc->sc_if.if_ierrors++; - return; + ++sc->sc_bytesrcvd; + c &= 0xff; /* XXX */ + +#ifdef ABT_ESC + if (sc->sc_flags & SC_ABORT) { + /* if we see an abort after "idle" time, count it */ + if (c == ABT_ESC && time.tv_sec >= sc->sc_lasttime + ABT_WAIT) { + sc->sc_abortcount++; + /* record when the first abort escape arrived */ + if (sc->sc_abortcount == 1) + sc->sc_starttime = time.tv_sec; + } + /* + * if we have an abort, see that we have not run out of time, + * or that we have an "idle" time after the complete escape + * sequence + */ + if (sc->sc_abortcount) { + if (time.tv_sec >= sc->sc_starttime + ABT_RECYCLE) + sc->sc_abortcount = 0; + if (sc->sc_abortcount >= ABT_SOFT && + time.tv_sec >= sc->sc_lasttime + ABT_WAIT) { + slclose(tp); + return; + } + } + sc->sc_lasttime = time.tv_sec; } +#endif - sc->sc_lasttime = time.tv_sec; - - if (sc->sc_flags & SC_ESCAPED) { - sc->sc_flags &= ~SC_ESCAPED; - switch (c) { + switch (c) { - case TRANS_FRAME_ESCAPE: + case TRANS_FRAME_ESCAPE: + if (sc->sc_escape) c = FRAME_ESCAPE; - break; + break; - case TRANS_FRAME_END: + case TRANS_FRAME_END: + if (sc->sc_escape) c = FRAME_END; - break; + break; - default: - sc->sc_if.if_ierrors++; - sc->sc_mp = sc->sc_buf; - sc->sc_ilen = 0; - return; - } - } else { - switch (c) { + case FRAME_ESCAPE: + sc->sc_escape = 1; + return; - case FRAME_END: - if (sc->sc_ilen == 0) /* ignore */ - return; - m = sl_btom(sc, sc->sc_ilen, &sc->sc_if); - sc->sc_mp = sc->sc_buf; - sc->sc_ilen = 0; - if (m == NULL) { - sc->sc_if.if_ierrors++; - return; - } - sc->sc_if.if_ipackets++; -#ifdef INET - { u_char type = *mtod(m, u_char *); - if (!(m = sl_uncompress_tcp(m, type&0xf0, &sc->sc_comp))) - return; - } -#endif - s = splimp(); - if (IF_QFULL(&ipintrq)) { - IF_DROP(&ipintrq); - sc->sc_if.if_ierrors++; - m_freem(m); - } else { - IF_ENQUEUE(&ipintrq, m); - schednetisr(NETISR_IP); - } - splx(s); - return; + case FRAME_END: + len = sc->sc_mp - sc->sc_buf; + if (len < 3) + /* less than min length packet - ignore */ + goto newpack; + + if ((c = (*sc->sc_buf & 0xf0)) != (IPVERSION << 4)) { + if (c & 0x80) + c = TYPE_COMPRESSED_TCP; + else if (c == TYPE_UNCOMPRESSED_TCP) + *sc->sc_buf &= 0x4f; /* XXX */ + len = sl_uncompress_tcp(&sc->sc_buf, len, (u_int)c, + &sc->sc_comp); + if (len <= 0) + goto error; + } + m = sl_btom(sc, len); + if (m == NULL) + goto error; - case FRAME_ESCAPE: - sc->sc_flags |= SC_ESCAPED; - return; + sc->sc_if.if_ipackets++; + s = splimp(); + if (IF_QFULL(&ipintrq)) { + IF_DROP(&ipintrq); + sc->sc_if.if_ierrors++; + m_freem(m); + } else { + IF_ENQUEUE(&ipintrq, m); + schednetisr(NETISR_IP); } + splx(s); + goto newpack; } - if (++sc->sc_ilen > SLMTU) { - sc->sc_if.if_ierrors++; - sc->sc_mp = sc->sc_buf; - sc->sc_ilen = 0; + if (sc->sc_mp < sc->sc_ep) { + *sc->sc_mp++ = c; + sc->sc_escape = 0; return; } - *sc->sc_mp++ = c; +error: + sc->sc_if.if_ierrors++; +newpack: + sc->sc_mp = sc->sc_buf = sc->sc_ep - SLMTU; + sc->sc_escape = 0; } /* diff --git a/usr/src/sys/net/if_slvar.h b/usr/src/sys/net/if_slvar.h index 3a058e79a5..28eba35893 100644 --- a/usr/src/sys/net/if_slvar.h +++ b/usr/src/sys/net/if_slvar.h @@ -1,4 +1,4 @@ -/* @(#)if_slvar.h 7.2 (Berkeley) %G% */ +/* @(#)if_slvar.h 7.3 (Berkeley) %G% */ /* * Definitions for SLIP interface data structures @@ -22,7 +22,9 @@ struct sl_softc { long sc_lasttime; /* last time a char arrived */ long sc_starttime; /* last time a char arrived */ long sc_abortcount; /* number of abort esacpe chars */ +#ifdef INET /* XXX */ struct slcompress sc_comp; /* tcp compression data */ +#endif }; /* flags */ @@ -33,3 +35,4 @@ struct sl_softc { /* this stuff doesn't belong here... */ #define SLIOCGFLAGS _IOR('t', 90, int) /* get configuration flags */ #define SLIOCSFLAGS _IOW('t', 89, int) /* set configuration flags */ +#define SLIOGUNIT _IOW('t', 88, int) /* get slip unit number */ diff --git a/usr/src/sys/net/slcompress.h b/usr/src/sys/net/slcompress.h index 8bc214fd76..6624ec2148 100644 --- a/usr/src/sys/net/slcompress.h +++ b/usr/src/sys/net/slcompress.h @@ -1,16 +1,10 @@ -/* - * THIS CODE IS NOT FOR DISTRIBUTION! - * KEEP YOUR GRUBBY HANDS OFF UNLESS AUTHORIZED BY VAN JACOBSON TO COPY! - * ASK SAM, MIKE, OR BILL ABOUT IT. - */ - /* * Definitions for tcp compression routines. * * Copyright (c) 1988, 1989 by Van Jacobson, Lawrence Berkeley Laboratory * All rights reserved. * - * $Header: slcompress.h,v 1.3 89/03/19 18:10:38 van Locked $ + * $Header: slcompress.h,v 1.6 89/06/05 08:29:13 van Exp $ */ #define MAX_STATES 16 /* must be > 2 and < 256 */ @@ -28,13 +22,6 @@ * sequence number changes, one change per bit set in the header * (there may be no changes and there are two special cases where * the receiver implicitly knows what changed -- see below). - * - * { Note that the ip version number field this overlays is 4 bits - * wide, and that we use type 4 to pass thru unaltered packets, - * type 5 to pass thru uncompressed packets that will be state - * information indexed by conversation. If msb (e.g type 8) is - * set, the other type bits are stolen to encode the difference - * information of a compressed TCP packet. -wfj } * * There are 5 numbers which can change (they are always inserted * in the following order): TCP urgent pointer, window, @@ -68,7 +55,7 @@ /* packet types */ #define TYPE_IP 0x40 -#define TYPE_UNCOMPRESSED_TCP 0x50 +#define TYPE_UNCOMPRESSED_TCP 0x70 #define TYPE_COMPRESSED_TCP 0x80 #define TYPE_ERROR 0x00 @@ -111,10 +98,20 @@ struct cstate { * per line). */ struct slcompress { - struct cstate *last_cs; /* most recently used tstate */ - u_char last_recv; /* last rcvd conn. id */ - u_char last_xmit; /* last sent conn. id */ + struct cstate *last_cs; /* most recently used tstate */ + u_char last_recv; /* last rcvd conn. id */ + u_char last_xmit; /* last sent conn. id */ u_short flags; +#ifndef NO_SL_STATS + int sls_packets; /* outbound packets */ + int sls_compressed; /* outbound compressed packets */ + int sls_searches; /* searches for connection state */ + int sls_misses; /* times couldn't find conn. state */ + int sls_uncompressedin; /* inbound uncompressed packets */ + int sls_compressedin; /* inbound compressed packets */ + int sls_errorin; /* inbound unknown type packets */ + int sls_tossed; /* inbound packets tossed because of error */ +#endif struct cstate tstate[MAX_STATES]; /* xmit connection states */ struct cstate rstate[MAX_STATES]; /* receive connection states */ }; @@ -123,4 +120,4 @@ struct slcompress { extern void sl_compress_init(/* struct slcompress * */); extern u_char sl_compress_tcp(/* struct mbuf *, struct ip *, struct slcompress * */); -extern struct mbuf *sl_uncompress_tcp(/* struct mbuf *, u_char, struct slcompress * */); +extern int sl_uncompress_tcp(/* u_char **, int, u_char, struct slcompress * */); -- 2.20.1