//************************************************************************** // Multi-threaded Matrix Multiply benchmark //-------------------------------------------------------------------------- // TA : Christopher Celio // Student: // // // This benchmark multiplies two 2-D arrays together and writes the results to // a third vector. The input data (and reference data) should be generated // using the matmul_gendata.pl perl script and dumped to a file named // dataset.h. // print out arrays, etc. //#define DEBUG //-------------------------------------------------------------------------- // Includes #include #include #include //-------------------------------------------------------------------------- // Input/Reference Data typedef float data_t; #include "dataset.h" //-------------------------------------------------------------------------- // Basic Utilities and Multi-thread Support __thread unsigned long coreid; unsigned long ncores; #include "util.h" #define stringify_1(s) #s #define stringify(s) stringify_1(s) #define stats(code) do { \ unsigned long _c = -rdcycle(), _i = -rdinstret(); \ code; \ _c += rdcycle(), _i += rdinstret(); \ if (coreid == 0) \ printf("%s: %ld cycles, %ld.%ld cycles/iter, %ld.%ld CPI\n", \ stringify(code), _c, _c/DIM_SIZE/DIM_SIZE/DIM_SIZE, 10*_c/DIM_SIZE/DIM_SIZE/DIM_SIZE%10, _c/_i, 10*_c/_i%10); \ } while(0) //-------------------------------------------------------------------------- // Helper functions void printArray( char name[], int n, data_t arr[] ) { int i; if (coreid != 0) return; printf( " %10s :", name ); for ( i = 0; i < n; i++ ) printf( " %3ld ", (long) arr[i] ); printf( "\n" ); } void __attribute__((noinline)) verify(size_t n, const data_t* test, const data_t* correct) { if (coreid != 0) return; size_t i; for (i = 0; i < n; i++) { if (test[i] != correct[i]) { printf("FAILED test[%d]= %3ld, correct[%d]= %3ld\n", i, (long)test[i], i, (long)correct[i]); exit(-1); } } return; } //-------------------------------------------------------------------------- // matmul function // single-thread, naive version void __attribute__((noinline)) matmul_naive(const int lda, const data_t A[], const data_t B[], data_t C[] ) { int i, j, k; if (coreid > 0) return; for ( i = 0; i < lda; i++ ) for ( j = 0; j < lda; j++ ) { for ( k = 0; k < lda; k++ ) { C[i + j*lda] += A[j*lda + k] * B[k*lda + i]; } } } void __attribute__((noinline)) matmul(const int lda, const data_t A[], const data_t B[], data_t C[] ) { // ***************************** // // **** ADD YOUR CODE HERE ***** // // ***************************** // // // feel free to make a separate function for MI and MSI versions. int i, j, k , jj , kk; int start_i = coreid*lda/2; int end_i = start_i + lda/2; int step_j, step_k; int start_k, end_k, start_j, end_j; int j_lda; int pos_A , pos_B, pos_C; data_t temp00, temp01,temp02,temp03,temp04,temp05,temp06,temp07; data_t temp10, temp11,temp12,temp13,temp14,temp15,temp16,temp17; data_t temp_A0, temp_A1, temp_A2, temp_A3, temp_A4, temp_A5, temp_A6, temp_A7; if (coreid == 0) { step_k = 1; start_k= 0; end_k = lda; step_j = 2; start_j= 0; end_j = lda; }else { step_k = -1; start_k = lda-1; end_k = -1; step_j = -2; start_j= lda-2; end_j = -2; } for( kk = start_k ; kk!= end_k ; kk+=(step_k*16) ) { for( jj = start_j ; jj!= end_j ; jj+=(step_j*8) ) { for ( i = start_i; i < end_i; i+=8 ) { //pos_C = i + jj*lda; for ( j = jj; j != (jj+(step_j*8)) ; j+=step_j ) { pos_C = i + j*lda; temp00 = C[(pos_C + 0)]; temp01 = C[(pos_C + 1)]; temp02 = C[(pos_C + 2)]; temp03 = C[(pos_C + 3)]; temp04 = C[(pos_C + 4)]; temp05 = C[(pos_C + 5)]; temp06 = C[(pos_C + 6)]; temp07 = C[(pos_C + 7)]; //pos_C += lda; pos_C = i + (j+1)*lda; temp10 = C[(pos_C + 0)]; temp11 = C[(pos_C + 1)]; temp12 = C[(pos_C + 2)]; temp13 = C[(pos_C + 3)]; temp14 = C[(pos_C + 4)]; temp15 = C[(pos_C + 5)]; temp16 = C[(pos_C + 6)]; temp17 = C[(pos_C + 7)]; pos_B = kk*lda + i; pos_A = j*lda + kk; for ( k = kk; k != (kk+(step_k*16)) ; k+=step_k ) { temp_A0 = A[ pos_A ] ; temp_A1 = A[pos_A +lda]; temp00 += temp_A0 * B[(pos_B + 0)]; temp01 += temp_A0 * B[(pos_B + 1)]; temp02 += temp_A0 * B[(pos_B + 2)]; temp03 += temp_A0 * B[(pos_B + 3)]; temp04 += temp_A0 * B[(pos_B + 4)]; temp05 += temp_A0 * B[(pos_B + 5)]; temp06 += temp_A0 * B[(pos_B + 6)]; temp07 += temp_A0 * B[(pos_B + 7)]; temp10 += temp_A1 * B[(pos_B + 0)]; temp11 += temp_A1 * B[(pos_B + 1)]; temp12 += temp_A1 * B[(pos_B + 2)]; temp13 += temp_A1 * B[(pos_B + 3)]; temp14 += temp_A1 * B[(pos_B + 4)]; temp15 += temp_A1 * B[(pos_B + 5)]; temp16 += temp_A1 * B[(pos_B + 6)]; temp17 += temp_A1 * B[(pos_B + 7)]; pos_B += (lda*step_k) ; pos_A += step_k; } //barrier(); C[(pos_C + 0)] = temp10; C[(pos_C + 1)] = temp11; C[(pos_C + 2)] = temp12; C[(pos_C + 3)] = temp13; C[(pos_C + 4)] = temp14; C[(pos_C + 5)] = temp15; C[(pos_C + 6)] = temp16; C[(pos_C + 7)] = temp17; //barrier(); pos_C = i + j*lda; //pos_C -= lda; C[(pos_C + 0)] = temp00; C[(pos_C + 1)] = temp01; C[(pos_C + 2)] = temp02; C[(pos_C + 3)] = temp03; C[(pos_C + 4)] = temp04; C[(pos_C + 5)] = temp05; C[(pos_C + 6)] = temp06; C[(pos_C + 7)] = temp07; //barrier(); //pos_C += step_j * lda; } //barrier(); } //barrier(); } //barrier(); } } //-------------------------------------------------------------------------- // Main // // all threads start executing thread_entry(). Use their "coreid" to // differentiate between threads (each thread is running on a separate core). void thread_entry(int cid, int nc) { coreid = cid; ncores = nc; // static allocates data in the binary, which is visible to both threads static data_t results_data[ARRAY_SIZE]; /* // Execute the provided, naive matmul barrier(); stats(matmul_naive(DIM_SIZE, input1_data, input2_data, results_data); barrier()); // verify verify(ARRAY_SIZE, results_data, verify_data); // clear results from the first trial size_t i; if (coreid == 0) for (i=0; i < ARRAY_SIZE; i++) results_data[i] = 0; barrier(); */ // Execute your faster matmul barrier(); stats(matmul(DIM_SIZE, input1_data, input2_data, results_data); barrier()); #ifdef DEBUG printArray("results:", ARRAY_SIZE, results_data); printArray("verify :", ARRAY_SIZE, verify_data); #endif // verify verify(ARRAY_SIZE, results_data, verify_data); barrier(); //printf("input1_data"); exit(0); }