/////////////////////////////////////////////////////////////////////////////////////////////
// File   : main.c   (for router application)
// Date   : november 2014
// author : Alain Greiner
/////////////////////////////////////////////////////////////////////////////////////////////
// This multi-threaded application emulates a packet routing communication application,
// described as a TCG (Task and Communication Graph).
// It contains 2 + N tasks : one "producer", one "consumer" and N "router")
// It contains 2 MWMR channels : "fifo_in" and "fifo_out".
// - The "producer" task writes NMAX token into "fifo_in".
// - The N "router" tasks read token from "fifo_in" and write them into "fifo_out".
// - The "consumer" task read token from "fifo_out" and displays instrumentation results.
// Token are indexed (by the producer) from 0 to NMAX-1.
// The router task contain a random delay emulating a variable processing time.
//
// This application is intended to run on a multi-processors, multi-clusters architecture,
//  with one thread per processor. 
//
// It uses the he following hardware parameters, defined in the hard_config.h file:
// - X_SIZE       : number of clusters in a row
// - Y_SIZE       : number of clusters in a column
// - NB_PROCS_MAX : number of processors per cluster
// 
// There is two global arrays (indexed by the token index) for insrumentation:
// - The "router_tab" array is filled concurrently by all "router" tasks.
//   Each entry contains the processor index that routed the token.
// - The "consumer_tab" array is filled by the "consumer" task.
//   Each entry contain the arrival order to the consumer task.
/////////////////////////////////////////////////////////////////////////////////////////////

#include "stdio.h"
#include "mwmr_channel.h"
#include "mapping_info.h"
#include "hard_config.h"


#define NMAX   50			// total number of token
#define DEPTH  20           // MWMR channels depth

//////////////// MWMR channels and associated buffers  /////////////////////

__attribute__((section (".data_in")))  mwmr_channel_t fifo_in;
__attribute__((section (".data_in")))  unsigned int   buf_in[DEPTH];

__attribute__((section (".data_out"))) mwmr_channel_t fifo_out;
__attribute__((section (".data_out"))) unsigned int   buf_out[DEPTH];
 
//////////////// Instrumentation Counters //////////////////////////////////

__attribute__((section (".data_out")))  unsigned int consumer_tab[NMAX];
__attribute__((section (".data_out")))  unsigned int router_tab[NMAX];



/////////////////////////////////////////////
__attribute__ ((constructor)) void producer()
{

    unsigned int 	n;
    unsigned int 	buf;

    // get processor identifiers
    unsigned int    x;
    unsigned int    y;
    unsigned int    lpid;
    giet_proc_xyp( &x, &y, &lpid );

    giet_shr_printf("\n*** Starting task producer on P[%d,%d,%d] at cycle %d\n", 
                    x, y, lpid, giet_proctime() );

    // initializes fifo_in
    mwmr_init( &fifo_in, buf_in, 1 , DEPTH );

    // main loop : display token value = source index
    for(n = 0 ; n < NMAX ; n++) 
    { 
        buf = n;
        mwmr_write( &fifo_in , &buf , 1 );
    }

    giet_exit( "Producer task completed");

} // end producer()

/////////////////////////////////////////////
__attribute__ ((constructor)) void consumer()
{
    unsigned int 	n;
    unsigned int 	buf;

    // get processor identifiers
    unsigned int    x;
    unsigned int    y;
    unsigned int    lpid;
    giet_proc_xyp( &x, &y, &lpid );

    giet_shr_printf("\n*** Starting task consumer on P[%d,%d,%d] at cycle %d\n", 
                    x, y, lpid, giet_proctime() );

    // initializes fifo_out 
    mwmr_init( &fifo_out, buf_out, 1 , DEPTH );

    // main loop : register token arrival index and value
    for( n = 0 ; n < NMAX ; n++ ) 
    { 
        mwmr_read( &fifo_out , &buf , 1 );
        consumer_tab[n] = buf;
    }

    // instrumentation display
    giet_shr_printf("\n");
    for( n = 0 ; n < NMAX ; n++ )
    {
        giet_shr_printf("@@@ arrival = %d / value = %d / router = %x\n",
                        n, consumer_tab[n], router_tab[n] );
    }

    giet_exit( "Consumer task completed");

} // end consumer()

///////////////////////////////////////////
__attribute__ ((constructor)) void router()
{
    unsigned int 	buf;
    unsigned int 	n;
    unsigned int	tempo;

    // get processor identifiers
    unsigned int    x;
    unsigned int    y;
    unsigned int    lpid;
    giet_proc_xyp( &x, &y, &lpid );

    giet_shr_printf("\n*** Starting task router on P[%d,%d,%d] at cycle %d\n", 
                    x, y, lpid, giet_proctime() );

    // waiting fifo_in and fifo_out initialisation
    while( (fifo_in.depth == 0) || (fifo_out.depth == 0) ) asm volatile( "nop" ); 

    // main loop
    while(1)
    {
        mwmr_read( &fifo_in , &buf , 1 );

        tempo = giet_rand();
        for ( n = 0 ; n < tempo ; n++ ) asm volatile ( "nop" );

        router_tab[buf] = (x<<(Y_WIDTH + P_WIDTH)) + (y<<P_WIDTH) + lpid;

        mwmr_write( &fifo_out , &buf , 1 );
    }
} // end router