tcpreplay/src/send_packets.c

/* $Id: send_packets.c 2008 2008-05-01 18:39:05Z aturner $ */

/*
 * Copyright (c) 2001-2008 Aaron Turner.
 * 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.
 * 3. Neither the names of the copyright owners nor the names of its
 *    contributors may be used to endorse or promote products derived from
 *    this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED ``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 AUTHORS OR COPYRIGHT HOLDERS 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.
 */

#include "config.h"
#include "defines.h"
#include "common.h"

#include <sys/time.h>
#include <sys/types.h>
#include <signal.h>
#include <string.h>
#include <netinet/in.h>
#include <errno.h>
#include <stdlib.h>
#include <unistd.h>
#include <fcntl.h>

#include "tcpreplay.h"

#ifdef TCPREPLAY
#include "tcpreplay_opts.h"

#ifdef TCPREPLAY_EDIT
#include "tcpedit/tcpedit.h"
extern tcpedit_t *tcpedit;
#endif

#endif /* TCPREPLAY */

#include "send_packets.h"
#include "sleep.h"

extern tcpreplay_opt_t options;
extern struct timeval begin, end;
extern COUNTER bytes_sent, failed, pkts_sent;
extern volatile int didsig;

#ifdef DEBUG
extern int debug;
#endif

static void do_sleep(struct timeval *time, struct timeval *last, int len, 
    int accurate, sendpacket_t *sp, COUNTER counter, delta_t *ctx);
static const u_char *get_next_packet(pcap_t *pcap, struct pcap_pkthdr *pkthdr, 
    int file_idx, packet_cache_t **prev_packet);
static u_int32_t get_user_count(sendpacket_t *sp, COUNTER counter);

/**
 * the main loop function for tcpreplay.  This is where we figure out
 * what to do with each packet
 */
void
send_packets(pcap_t *pcap, int cache_file_idx)
{
    struct timeval last = { 0, 0 };
    COUNTER packetnum = 0;
    struct pcap_pkthdr pkthdr;
    const u_char *pktdata = NULL;
    sendpacket_t *sp = options.intf1;
    u_int32_t pktlen;
	packet_cache_t *cached_packet = NULL;
	packet_cache_t **prev_packet = NULL;
#if defined TCPREPLAY && defined TCPREPLAY_EDIT
    struct pcap_pkthdr *pkthdr_ptr;
#endif
    delta_t delta_ctx;
    
    /* register signals */
    didsig = 0;
    if (options.speed.mode != SPEED_ONEATATIME) {
        (void)signal(SIGINT, catcher);
    } else {
        (void)signal(SIGINT, break_now);
    }

    if (options.enable_file_cache) {
        prev_packet = &cached_packet;
    } else {
        prev_packet = NULL;
    }

	
    /* MAIN LOOP 
     * Keep sending while we have packets or until
     * we've sent enough packets
     */
    while ((pktdata = get_next_packet(pcap, &pkthdr, cache_file_idx, prev_packet)) != NULL) {
        /* die? */
        if (didsig)
            break_now(0);

        /* stop sending based on the limit -L? */
        if (options.limit_send > 0 && pkts_sent >= options.limit_send)
            return;

        packetnum++;

#ifdef TCPREPLAY
        /* do we use the snaplen (caplen) or the "actual" packet len? */
        pktlen = HAVE_OPT(PKTLEN) ? pkthdr.len : pkthdr.caplen;
#elif TCPBRIDGE
        pktlen = pkthdr.caplen;
#else
#error WTF???  We should not be here!
#endif

        dbgx(2, "packet " COUNTER_SPEC " caplen %d", packetnum, pktlen);
        
        /* Dual nic processing */
        if (options.intf2 != NULL) {

            sp = (sendpacket_t *) cache_mode(options.cachedata, packetnum);
        
            /* sometimes we should not send the packet */
            if (sp == TCPR_DIR_NOSEND)
                continue;
        }
    
        /* do we need to print the packet via tcpdump? */
#ifdef ENABLE_VERBOSE
        if (options.verbose)
            tcpdump_print(options.tcpdump, &pkthdr, pktdata);
#endif

#if defined TCPREPLAY && defined TCPREPLAY_EDIT        
        pkthdr_ptr = &pkthdr;
        if (tcpedit_packet(tcpedit, &pkthdr_ptr, &pktdata, sp->cache_dir) == -1) {
            errx(1, "Error editing packet #" COUNTER_SPEC ": %s", packetnum, tcpedit_geterr(tcpedit));
        }
        pktlen = HAVE_OPT(PKTLEN) ? pkthdr_ptr->len : pkthdr_ptr->caplen;
#endif

        /*
         * we have to cast the ts, since OpenBSD sucks
         * had to be special and use bpf_timeval.
         * Only sleep if we're not in top speed mode (-t)
         */
        if (options.speed.mode != SPEED_TOPSPEED)
            do_sleep((struct timeval *)&pkthdr.ts, &last, pktlen, options.accurate, sp, packetnum, &delta_ctx);
        
        /* mark the time when we send the packet */
        start_delta_time(&delta_ctx);
        
        /* write packet out on network */
        if (sendpacket(sp, pktdata, pktlen) < (int)pktlen)
            warnx("Unable to send packet: %s", sendpacket_geterr(sp));

        /* 
         * track the time of the "last packet sent".  Again, because of OpenBSD
         * we have to do a mempcy rather then assignment. 
         * 
         * A number of 3rd party tools generate bad timestamps which go backwards
         * in time.  Hence, don't update the "last" unless pkthdr.ts > last
         */
        if (timercmp(&last, &pkthdr.ts, <))
            memcpy(&last, &pkthdr.ts, sizeof(struct timeval));
        pkts_sent ++;
        bytes_sent += pktlen;
    } /* while */

	if (options.enable_file_cache) {
		options.file_cache[cache_file_idx].cached = TRUE;
	}
}

/**
 * Gets the next packet to be sent out. This will either read from the pcap file
 * or will retrieve the packet from the internal cache.
 *	
 * The parameter prev_packet is used as the parent of the new entry in the cache list.
 * This should be NULL on the first call to this function for each file and
 * will be updated as new entries are added (or retrieved) from the cache list.
 */
static const u_char *
get_next_packet(pcap_t *pcap, struct pcap_pkthdr *pkthdr, int file_idx, 
    packet_cache_t **prev_packet)
{
	u_char *pktdata = NULL;
    u_int32_t pktlen;

    /* pcap may be null in cache mode! */
    /* packet_cache_t may be null in file read mode! */
    assert(pkthdr);

	/*
	 * Check if we're caching files
	 */
	if (options.enable_file_cache && (prev_packet != NULL)) {
		/*
		 * Yes we are caching files - has this one been cached?
		 */
		if (options.file_cache[file_idx].cached) {
			if (*prev_packet == NULL) {
				/*
				 * Get the first packet in the cache list directly from the file
				 */
				*prev_packet = options.file_cache[file_idx].packet_cache;
			} else {
				/*
				 * Get the next packet in the cache list
				 */
				*prev_packet = (*prev_packet)->next;
			}
			
			if (*prev_packet != NULL) {
				pktdata = (*prev_packet)->pktdata;
				memcpy(pkthdr, &((*prev_packet)->pkthdr), sizeof(struct pcap_pkthdr));
			}
		} else {
			/*
			 * We should read the pcap file, and cache the results
			 */
			pktdata = (u_char *)pcap_next(pcap, pkthdr);
			if (pktdata != NULL) {
				if (*prev_packet == NULL) {
					/*
					 * Create the first packet in the list
					 */
					*prev_packet = safe_malloc(sizeof(packet_cache_t));
					options.file_cache[file_idx].packet_cache = *prev_packet;
				} else {
					/*
					 * Add a packet to the end of the list
					 */
					(*prev_packet)->next = safe_malloc(sizeof(packet_cache_t));
					*prev_packet = (*prev_packet)->next;
				}
				
				if (*prev_packet != NULL) {
					(*prev_packet)->next = NULL;
					pktlen = pkthdr->len;
					
					(*prev_packet)->pktdata = safe_malloc(pktlen);
					memcpy((*prev_packet)->pktdata, pktdata, pktlen);
					memcpy(&((*prev_packet)->pkthdr), pkthdr, sizeof(struct pcap_pkthdr));
				}
			}
		}
	} else {
		/*
		 * Read pcap file as normal
		 */
		pktdata = (u_char *)pcap_next(pcap, pkthdr);
	}

    /* this get's casted to a const on the way out */
	return pktdata;
}

/**
 * determines based upon the cachedata which interface the given packet 
 * should go out.  Also rewrites any layer 2 data we might need to adjust.
 * Returns a void cased pointer to the options.intfX of the corresponding 
 * interface.
 */
void *
cache_mode(char *cachedata, COUNTER packet_num)
{
    void *sp = NULL;
    int result;

    if (packet_num > options.cache_packets)
        err(1, "Exceeded number of packets in cache file.");

    result = check_cache(cachedata, packet_num);
    if (result == TCPR_DIR_NOSEND) {
        dbgx(2, "Cache: Not sending packet " COUNTER_SPEC ".", packet_num);
        return TCPR_DIR_NOSEND;
    }
    else if (result == TCPR_DIR_C2S) {
        dbgx(2, "Cache: Sending packet " COUNTER_SPEC " out primary interface.", packet_num);
        sp = options.intf1;
    }
    else if (result == TCPR_DIR_S2C) {
        dbgx(2, "Cache: Sending packet " COUNTER_SPEC " out secondary interface.", packet_num);
        sp = options.intf2;
    }
    else {
        err(1, "check_cache() returned an error.  Aborting...");
    }

    return sp;
}


/**
 * Given the timestamp on the current packet and the last packet sent,
 * calculate the appropriate amount of time to sleep and do so.
 */
static void
do_sleep(struct timeval *time, struct timeval *last, int len, int accurate, 
    sendpacket_t *sp, COUNTER counter, delta_t *delta_ctx)
{
    static struct timeval didsleep = { 0, 0 };
    static struct timeval start = { 0, 0 };
#ifdef DEBUG
    static struct timeval totalsleep = { 0, 0 };
#endif
    struct timespec adjuster = { 0, 0 };
    static struct timespec nap = { 0, 0 }, delta_time = {0, 0};
    struct timeval nap_for, now, sleep_until;
    struct timespec nap_this_time;
    static int32_t nsec_adjuster = -1, nsec_times = -1;
    float n;
    static u_int32_t send = 0;      /* accellerator.   # of packets to send w/o sleeping */
    u_int64_t ppnsec; /* packets per usec */


#ifdef TCPREPLAY
    adjuster.tv_nsec = OPT_VALUE_SLEEP_ACCEL * 1000;
    dbgx(2, "Adjuster: " TIMEVAL_FORMAT, adjuster.tv_sec, adjuster.tv_nsec);
#else
    adjuster.tv_nsec = 0;
#endif
    
    /* acclerator time? */
    if (send > 0) {
        send --;
        return;
    }
        
    dbgx(4, "Last time: " TIMEVAL_FORMAT, last->tv_sec, last->tv_usec);

    if (gettimeofday(&now, NULL) < 0)
        errx(1, "Error gettimeofday: %s", strerror(errno));

    dbgx(4, "Now time: " TIMEVAL_FORMAT, now.tv_sec, now.tv_usec);

    /* First time through for this file */
    if (pkts_sent == 0 || ((options.speed.mode != SPEED_MBPSRATE) && (counter == 0))) {
        start = now;
        timerclear(&sleep_until);
        timerclear(&didsleep);
    }
    else {
        timersub(&now, &start, &sleep_until);
    }

    switch(options.speed.mode) {
    /* 
     * If top speed, you shouldn't even be here, but handle it anyways
     */
    case SPEED_TOPSPEED:
        notice("you shouldn't call do_sleep() in top speed mode.");
        return;
        break;
        
    case SPEED_MULTIPLIER:
        /* 
         * Replay packets a factor of the time they were originally sent.
         */
        if (timerisset(last) && timercmp(time, last, >)) {
            timersub(time, last, &nap_for);
            TIMEVAL_TO_TIMESPEC(&nap_for, &nap);
            timesdiv(&nap, options.speed.speed);
        }
        else {
            /* 
             * Don't sleep if this is our first packet, or if the
             * this packet appears to have been sent before the 
             * last packet.
             */
            timesclear(&nap);
        }        
        break;

    case SPEED_MBPSRATE:
        /* 
         * Ignore the time supplied by the capture file and send data at
         * a constant 'rate' (bytes per second).
         */
        if (pkts_sent != 0) {
            n = (float)len / (options.speed.speed * 1024 * 1024 / 8); /* convert Mbps to bps */
            nap.tv_sec = n;
            nap.tv_nsec = (n - nap.tv_sec)  * 1000000000;
            
            dbgx(3, "packet size %d\t\tequals %f bps\t\tnap " TIMEVAL_FORMAT, len, n, 
                nap.tv_sec, nap.tv_nsec);
        }
        else {
            /* don't sleep at all for the first packet */
            timesclear(&nap);
        }
        break;

    case SPEED_PACKETRATE:
        /* only need to calculate this the first time */
        if (! timesisset(&nap)) {
            /* run in packets/sec */
            ppnsec = 1000000000 / options.speed.speed;
            NANOSEC_TO_TIMESPEC(ppnsec, &nap);
            dbgx(1, "sending 1 packet per %lu nsec", nap.tv_nsec);
        }
        break;
        
    case SPEED_ONEATATIME:
        /* do we skip prompting for a key press? */
        if (send == 0) {
            send = get_user_count(sp, counter);
        }

        /* decrement our send counter */
        printf("Sending packet " COUNTER_SPEC " out: %s\n", counter,
               sp == options.intf1 ? options.intf1_name : options.intf2_name);
        send --;

        /* leave do_sleep() */
        return;

        break;

    default:
        errx(1, "Unknown/supported speed mode: %d", options.speed.mode);
        break;
    }

    /* 
     * since we apply the adjuster to the sleep time, we can't modify nap
     */
    nap_this_time.tv_sec = nap.tv_sec;
    nap_this_time.tv_nsec = nap.tv_nsec;

    if (accurate != ACCURATE_ABS_TIME) {

        switch (options.speed.mode) {
            /* Mbps & Multipler are dynamic timings, so we round to the nearest usec */
            case SPEED_MBPSRATE:
            case SPEED_MULTIPLIER:
                ROUND_TIMESPEC_TO_MICROSEC(&nap_this_time);
                break;

            /* Packets/sec is static, so we weight packets for .1usec accuracy */
            case SPEED_PACKETRATE:
                if (nsec_adjuster < 0)
                    nsec_adjuster = (nap_this_time.tv_nsec % 10000) / 1000;
        
                /* update in the range of 0-9 */
                nsec_times = (nsec_times + 1) % 10;
        
                if (nsec_times < nsec_adjuster) {
                    /* sorta looks like a no-op, but gives us a nice round usec number */
                    nap_this_time.tv_nsec = (nap_this_time.tv_nsec / 1000 * 1000) + 1000;
                } else {
                    nap_this_time.tv_nsec -= (nap_this_time.tv_nsec % 1000);
                }
    
                dbgx(3, "(%ld)\tnsec_times = %ld\tnap adjust: %lu -> %lu", nsec_adjuster, nsec_times, nap.tv_nsec, nap_this_time.tv_nsec);            
                break;
                
            default:
                errx(1, "Unknown/supported speed mode: %d", options.speed.mode);
        }
    }


    get_delta_time(delta_ctx, &delta_time);
    dbgx(2, "delta: " TIMESPEC_FORMAT, delta_time.tv_sec, delta_time.tv_nsec);

    if (timesisset(&delta_time)) {
        if (timescmp(&nap_this_time, &delta_time, >)) {
            timessub(&nap_this_time, &delta_time, &nap_this_time);
            dbgx(3, "timesub: %lu %lu", delta_time.tv_sec, delta_time.tv_nsec);
        } else { 
            timesclear(&nap_this_time);
            dbgx(3, "timesclear: " TIMESPEC_FORMAT, delta_time.tv_sec, delta_time.tv_nsec);
        }
    }

    /* apply the adjuster... */
    if (timesisset(&adjuster)) {
        if (timescmp(&nap_this_time, &adjuster, >)) {
            timessub(&nap_this_time, &adjuster, &nap_this_time);
        } else { 
            timesclear(&nap_this_time);
        }
    }
    
    dbgx(2, "Sleeping: " TIMESPEC_FORMAT, nap_this_time.tv_sec, nap_this_time.tv_nsec);

    /* don't sleep if nap = {0, 0} */
    if (!timesisset(&nap_this_time))
        return;
        
    /*
     * Depending on the accurate method & packet rate computation method
     * We have multiple methods of sleeping, pick the right one...
     */
    switch (accurate) {
#ifdef HAVE_SELECT
    case ACCURATE_SELECT:
        select_sleep(nap_this_time);
        break;
#endif

#ifdef HAVE_IOPERM
    case ACCURATE_IOPORT:
        ioport_sleep(nap_this_time);
        break;
#endif

#ifdef HAVE_RDTSC        
    case ACCURATE_RDTSC:
        rdtsc_sleep(nap_this_time);
        break;
#endif

#ifdef HAVE_ABSOLUTE_TIME
    case ACCURATE_ABS_TIME:
        absolute_time_sleep(nap_this_time);
        break;
#endif

    case ACCURATE_GTOD:
        gettimeofday_sleep(nap_this_time);
        break;

    case ACCURATE_NANOSLEEP:
        nanosleep_sleep(nap_this_time);
        break;
        /*
        timeradd(&didsleep, &nap_this_time, &didsleep);

        dbgx(4, "I will sleep " TIMEVAL_FORMAT, nap_this_time.tv_sec, nap_this_time.tv_usec);

        if (timercmp(&didsleep, &sleep_until, >)) {
            timersub(&didsleep, &sleep_until, &nap_this_time);
            
            TIMEVAL_TO_TIMESPEC(&nap_this_time, &sleep);
            dbgx(4, "Sleeping " TIMEVAL_FORMAT, nap_this_time.tv_sec, nap_this_time.tv_usec);
#ifdef DEBUG
            timeradd(&totalsleep, &nap_this_time, &totalsleep);
#endif
            if (nanosleep(&sleep, &ignore) == -1) {
                warnx("nanosleep error: %s", strerror(errno));
            }
        }
        break;
        */
    default:
        errx(1, "Unknown timer mode %d", accurate);
    }

#ifdef DEBUG
    dbgx(4, "Total sleep time: " TIMEVAL_FORMAT, totalsleep.tv_sec, totalsleep.tv_usec);
#endif

    dbgx(2, "sleep delta: " TIMESPEC_FORMAT, delta_time.tv_sec, delta_time.tv_nsec);

}

/**
 * this function will keep calling gettimeofday() until it returns
 * >= time.  This should be a lot more accurate then using nanosleep(),
 * but at the cost of being more CPU intensive.
 */
static u_int32_t
get_user_count(sendpacket_t *sp, COUNTER counter) 
{
    struct pollfd poller[1];        /* use poll to read from the keyboard */
    char input[EBUF_SIZE];
    u_int32_t send = 0;
    
    printf("**** Next packet #" COUNTER_SPEC " out %s.  How many packets do you wish to send? ",
        counter, (sp == options.intf1 ? options.intf1_name : options.intf2_name));
    fflush(NULL);
    poller[0].fd = STDIN_FILENO;
    poller[0].events = POLLIN | POLLPRI | POLLNVAL;
    poller[0].revents = 0;

    if (fcntl(0, F_SETFL, fcntl(0, F_GETFL) & ~O_NONBLOCK)) 
           errx(1, "Unable to clear non-blocking flag on stdin: %s", strerror(errno));

    /* wait for the input */
    if (poll(poller, 1, -1) < 0)
        errx(1, "Error reading user input from stdin: %s", strerror(errno));
    
    /*
     * read to the end of the line or EBUF_SIZE,
     * Note, if people are stupid, and type in more text then EBUF_SIZE
     * then the next fgets() will pull in that data, which will have poor 
     * results.  fuck them.
     */
    if (fgets(input, sizeof(input), stdin) == NULL) {
        errx(1, "Unable to process user input for fd %d: %s", fileno(stdin), strerror(errno));
    } else if (strlen(input) > 1) {
        send = strtoul(input, NULL, 0);
    }

    /* how many packets should we send? */
    if (send == 0) {
        dbg(1, "Input was less then 1 or non-numeric, assuming 1");

        /* assume send only one packet */
        send = 1;
    }
    
    return send;
}

/*
 Local Variables:
 mode:c
 indent-tabs-mode:nil
 c-basic-offset:4
 End:
*/