Changeset 415
- Timestamp:
- 08/31/10 21:16:24 (14 years ago)
- Location:
- trunk/examples/nardi/shallow_performanceMeasurement
- Files:
-
- 2 added
- 5 edited
Legend:
- Unmodified
- Added
- Removed
-
trunk/examples/nardi/shallow_performanceMeasurement/2shalw_100x100.h
r414 r415 487 487 } 488 488 } 489 489 490 endForward=my_gettimeofday(); 490 491 491 return(0); 492 492 } … … 509 509 Yvsmatt (YNBS_Vfil, YNBI_Vfil, YMAX_JAC_NBI, &YG1Y_Vfil, Yjac[0], Ytbeta); 510 510 YG1_Vfil( Yi, Yj, YTemps-1) += Ytbeta[0]; 511 #pragma omp atomic 511 512 YG1_Vphy( Yi, Yj, YTemps-1) += Ytbeta[1]; 513 #pragma omp atomic 512 514 YG1_Vphy( Yi, Yj, YTemps) += Ytbeta[2]; 513 515 Yting[0]=YS1_Ufil( Yi, Yj, YTemps-1); … … 518 520 Yvsmatt (YNBS_Ufil, YNBI_Ufil, YMAX_JAC_NBI, &YG1Y_Ufil, Yjac[0], Ytbeta); 519 521 YG1_Ufil( Yi, Yj, YTemps-1) += Ytbeta[0]; 522 #pragma omp atomic 520 523 YG1_Uphy( Yi, Yj, YTemps-1) += Ytbeta[1]; 524 #pragma omp atomic 521 525 YG1_Uphy( Yi, Yj, YTemps) += Ytbeta[2]; 522 526 Yting[0]=YS1_Hfil( Yi, Yj, YTemps-1); … … 527 531 Yvsmatt (YNBS_Hfil, YNBI_Hfil, YMAX_JAC_NBI, &YG1Y_Hfil, Yjac[0], Ytbeta); 528 532 YG1_Hfil( Yi, Yj, YTemps-1) += Ytbeta[0]; 533 #pragma omp atomic 529 534 YG1_Hphy( Yi, Yj, YTemps-1) += Ytbeta[1]; 535 #pragma omp atomic 530 536 YG1_Hphy( Yi, Yj, YTemps) += Ytbeta[2]; 531 537 Yting[0]=YS1_Vfil( Yi, Yj, YTemps-1); … … 552 558 Yvsmatt (YNBS_Vphy, YNBI_Vphy, YMAX_JAC_NBI, &YG1Y_Vphy, Yjac[0], Ytbeta); 553 559 YG1_Vfil( Yi, Yj, YTemps-1) += Ytbeta[0]; 560 #pragma omp atomic 554 561 YG1_Hphy( Yi, Yj, YTemps-1) += Ytbeta[1]; 555 562 if (1==0 || Yj-1<0){} 556 563 else 564 #pragma omp atomic 557 565 YG1_Hphy( Yi, Yj-1, YTemps-1) += Ytbeta[2]; 558 566 if (1==0 || Yi-1<0 || Yj-1<0){} … … 566 574 if (1==0 || Yj-1<0){} 567 575 else 576 #pragma omp atomic 568 577 YG1_Uphy( Yi, Yj-1, YTemps-1) += Ytbeta[5]; 578 #pragma omp atomic 569 579 YG1_Uphy( Yi, Yj, YTemps-1) += Ytbeta[6]; 570 580 Yting[0]=YS1_Ufil( Yi, Yj, YTemps-1); … … 595 605 #pragma omp atomic 596 606 YG1_Hphy( Yi+1, Yj, YTemps-1) += Ytbeta[1]; 607 #pragma omp atomic 597 608 YG1_Hphy( Yi, Yj, YTemps-1) += Ytbeta[2]; 609 #pragma omp atomic 598 610 YG1_Vphy( Yi, Yj, YTemps-1) += Ytbeta[3]; 599 611 if (1==0 || Yj+1>YA2_Soce-1){} 600 612 else 613 #pragma omp atomic 601 614 YG1_Vphy( Yi, Yj+1, YTemps-1) += Ytbeta[4]; 602 615 if (1==0 || Yi+1>YA1_Soce-1){} … … 623 636 Yvsmatt (YNBS_Hphy, YNBI_Hphy, YMAX_JAC_NBI, &YG1Y_Hphy, Yjac[0], Ytbeta); 624 637 YG1_Hfil( Yi, Yj, YTemps-1) += Ytbeta[0]; 638 #pragma omp atomic 625 639 YG1_Uphy( Yi, Yj, YTemps-1) += Ytbeta[1]; 626 640 if (1==0 || Yi-1<0){} … … 630 644 if (1==0 || Yj+1>YA2_Soce-1){} 631 645 else 646 #pragma omp atomic 632 647 YG1_Vphy( Yi, Yj+1, YTemps-1) += Ytbeta[3]; 648 #pragma omp atomic 633 649 YG1_Vphy( Yi, Yj, YTemps-1) += Ytbeta[4]; 634 650 … … 646 662 printf("Average %f\n", (differenceForward + differenceBackward) / counter ); 647 663 printf("\n############################################################################\n\n"); 664 648 665 649 666 -
trunk/examples/nardi/shallow_performanceMeasurement/2shalw_200x200.h
r414 r415 508 508 Yvsmatt (YNBS_Vfil, YNBI_Vfil, YMAX_JAC_NBI, &YG1Y_Vfil, Yjac[0], Ytbeta); 509 509 YG1_Vfil( Yi, Yj, YTemps-1) += Ytbeta[0]; 510 #pragma omp atomic 510 511 YG1_Vphy( Yi, Yj, YTemps-1) += Ytbeta[1]; 512 #pragma omp atomic 511 513 YG1_Vphy( Yi, Yj, YTemps) += Ytbeta[2]; 512 514 Yting[0]=YS1_Ufil( Yi, Yj, YTemps-1); … … 517 519 Yvsmatt (YNBS_Ufil, YNBI_Ufil, YMAX_JAC_NBI, &YG1Y_Ufil, Yjac[0], Ytbeta); 518 520 YG1_Ufil( Yi, Yj, YTemps-1) += Ytbeta[0]; 521 #pragma omp atomic 519 522 YG1_Uphy( Yi, Yj, YTemps-1) += Ytbeta[1]; 523 #pragma omp atomic 520 524 YG1_Uphy( Yi, Yj, YTemps) += Ytbeta[2]; 521 525 Yting[0]=YS1_Hfil( Yi, Yj, YTemps-1); … … 526 530 Yvsmatt (YNBS_Hfil, YNBI_Hfil, YMAX_JAC_NBI, &YG1Y_Hfil, Yjac[0], Ytbeta); 527 531 YG1_Hfil( Yi, Yj, YTemps-1) += Ytbeta[0]; 532 #pragma omp atomic 528 533 YG1_Hphy( Yi, Yj, YTemps-1) += Ytbeta[1]; 534 #pragma omp atomic 529 535 YG1_Hphy( Yi, Yj, YTemps) += Ytbeta[2]; 530 536 Yting[0]=YS1_Vfil( Yi, Yj, YTemps-1); … … 551 557 Yvsmatt (YNBS_Vphy, YNBI_Vphy, YMAX_JAC_NBI, &YG1Y_Vphy, Yjac[0], Ytbeta); 552 558 YG1_Vfil( Yi, Yj, YTemps-1) += Ytbeta[0]; 559 #pragma omp atomic 553 560 YG1_Hphy( Yi, Yj, YTemps-1) += Ytbeta[1]; 554 561 if (1==0 || Yj-1<0){} 555 562 else 563 #pragma omp atomic 556 564 YG1_Hphy( Yi, Yj-1, YTemps-1) += Ytbeta[2]; 557 565 if (1==0 || Yi-1<0 || Yj-1<0){} … … 565 573 if (1==0 || Yj-1<0){} 566 574 else 575 #pragma omp atomic 567 576 YG1_Uphy( Yi, Yj-1, YTemps-1) += Ytbeta[5]; 577 #pragma omp atomic 568 578 YG1_Uphy( Yi, Yj, YTemps-1) += Ytbeta[6]; 569 579 Yting[0]=YS1_Ufil( Yi, Yj, YTemps-1); … … 594 604 #pragma omp atomic 595 605 YG1_Hphy( Yi+1, Yj, YTemps-1) += Ytbeta[1]; 606 #pragma omp atomic 596 607 YG1_Hphy( Yi, Yj, YTemps-1) += Ytbeta[2]; 608 #pragma omp atomic 597 609 YG1_Vphy( Yi, Yj, YTemps-1) += Ytbeta[3]; 598 610 if (1==0 || Yj+1>YA2_Soce-1){} 599 611 else 612 #pragma omp atomic 600 613 YG1_Vphy( Yi, Yj+1, YTemps-1) += Ytbeta[4]; 601 614 if (1==0 || Yi+1>YA1_Soce-1){} … … 622 635 Yvsmatt (YNBS_Hphy, YNBI_Hphy, YMAX_JAC_NBI, &YG1Y_Hphy, Yjac[0], Ytbeta); 623 636 YG1_Hfil( Yi, Yj, YTemps-1) += Ytbeta[0]; 637 #pragma omp atomic 624 638 YG1_Uphy( Yi, Yj, YTemps-1) += Ytbeta[1]; 625 639 if (1==0 || Yi-1<0){} … … 629 643 if (1==0 || Yj+1>YA2_Soce-1){} 630 644 else 645 #pragma omp atomic 631 646 YG1_Vphy( Yi, Yj+1, YTemps-1) += Ytbeta[3]; 647 #pragma omp atomic 632 648 YG1_Vphy( Yi, Yj, YTemps-1) += Ytbeta[4]; 633 649 -
trunk/examples/nardi/shallow_performanceMeasurement/2shalw_50x50.h
r414 r415 509 509 Yvsmatt (YNBS_Vfil, YNBI_Vfil, YMAX_JAC_NBI, &YG1Y_Vfil, Yjac[0], Ytbeta); 510 510 YG1_Vfil( Yi, Yj, YTemps-1) += Ytbeta[0]; 511 #pragma omp atomic 511 512 YG1_Vphy( Yi, Yj, YTemps-1) += Ytbeta[1]; 513 #pragma omp atomic 512 514 YG1_Vphy( Yi, Yj, YTemps) += Ytbeta[2]; 513 515 Yting[0]=YS1_Ufil( Yi, Yj, YTemps-1); … … 518 520 Yvsmatt (YNBS_Ufil, YNBI_Ufil, YMAX_JAC_NBI, &YG1Y_Ufil, Yjac[0], Ytbeta); 519 521 YG1_Ufil( Yi, Yj, YTemps-1) += Ytbeta[0]; 522 #pragma omp atomic 520 523 YG1_Uphy( Yi, Yj, YTemps-1) += Ytbeta[1]; 524 #pragma omp atomic 521 525 YG1_Uphy( Yi, Yj, YTemps) += Ytbeta[2]; 522 526 Yting[0]=YS1_Hfil( Yi, Yj, YTemps-1); … … 527 531 Yvsmatt (YNBS_Hfil, YNBI_Hfil, YMAX_JAC_NBI, &YG1Y_Hfil, Yjac[0], Ytbeta); 528 532 YG1_Hfil( Yi, Yj, YTemps-1) += Ytbeta[0]; 533 #pragma omp atomic 529 534 YG1_Hphy( Yi, Yj, YTemps-1) += Ytbeta[1]; 535 #pragma omp atomic 530 536 YG1_Hphy( Yi, Yj, YTemps) += Ytbeta[2]; 531 537 Yting[0]=YS1_Vfil( Yi, Yj, YTemps-1); … … 552 558 Yvsmatt (YNBS_Vphy, YNBI_Vphy, YMAX_JAC_NBI, &YG1Y_Vphy, Yjac[0], Ytbeta); 553 559 YG1_Vfil( Yi, Yj, YTemps-1) += Ytbeta[0]; 560 #pragma omp atomic 554 561 YG1_Hphy( Yi, Yj, YTemps-1) += Ytbeta[1]; 555 562 if (1==0 || Yj-1<0){} 556 563 else 564 #pragma omp atomic 557 565 YG1_Hphy( Yi, Yj-1, YTemps-1) += Ytbeta[2]; 558 566 if (1==0 || Yi-1<0 || Yj-1<0){} … … 566 574 if (1==0 || Yj-1<0){} 567 575 else 576 #pragma omp atomic 568 577 YG1_Uphy( Yi, Yj-1, YTemps-1) += Ytbeta[5]; 578 #pragma omp atomic 569 579 YG1_Uphy( Yi, Yj, YTemps-1) += Ytbeta[6]; 570 580 Yting[0]=YS1_Ufil( Yi, Yj, YTemps-1); … … 595 605 #pragma omp atomic 596 606 YG1_Hphy( Yi+1, Yj, YTemps-1) += Ytbeta[1]; 607 #pragma omp atomic 597 608 YG1_Hphy( Yi, Yj, YTemps-1) += Ytbeta[2]; 609 #pragma omp atomic 598 610 YG1_Vphy( Yi, Yj, YTemps-1) += Ytbeta[3]; 599 611 if (1==0 || Yj+1>YA2_Soce-1){} 600 612 else 613 #pragma omp atomic 601 614 YG1_Vphy( Yi, Yj+1, YTemps-1) += Ytbeta[4]; 602 615 if (1==0 || Yi+1>YA1_Soce-1){} … … 623 636 Yvsmatt (YNBS_Hphy, YNBI_Hphy, YMAX_JAC_NBI, &YG1Y_Hphy, Yjac[0], Ytbeta); 624 637 YG1_Hfil( Yi, Yj, YTemps-1) += Ytbeta[0]; 638 #pragma omp atomic 625 639 YG1_Uphy( Yi, Yj, YTemps-1) += Ytbeta[1]; 626 640 if (1==0 || Yi-1<0){} … … 630 644 if (1==0 || Yj+1>YA2_Soce-1){} 631 645 else 646 #pragma omp atomic 632 647 YG1_Vphy( Yi, Yj+1, YTemps-1) += Ytbeta[3]; 648 #pragma omp atomic 633 649 YG1_Vphy( Yi, Yj, YTemps-1) += Ytbeta[4]; 634 650 -
trunk/examples/nardi/shallow_performanceMeasurement/scriptPerformance.sh
r414 r415 13 13 # Serial execution 14 14 echo num threads: serial... 15 ./shalw_50x50 > resultNotParallelFile_50x50.txt15 time ./shalw_50x50 > resultNotParallelFile_50x50.txt 16 16 17 17 YaoI -p +O3 shalw_50x50 18 18 # 1 thread execution 19 19 echo num threads: 1 20 OMP_NUM_THREADS=1 ./shalw_50x50 > resultParallelFile_1thread_50x50.txt20 OMP_NUM_THREADS=1 time ./shalw_50x50 > resultParallelFile_1thread_50x50.txt 21 21 22 22 # 2 thread execution 23 23 echo num threads: 2 24 OMP_NUM_THREADS=2 ./shalw_50x50 > resultParallelFile_2thread_50x50.txt24 OMP_NUM_THREADS=2 time ./shalw_50x50 > resultParallelFile_2thread_50x50.txt 25 25 26 26 # 4 thread execution 27 27 echo num threads: 4 28 OMP_NUM_THREADS=4 ./shalw_50x50 > resultParallelFile_4thread_50x50.txt28 OMP_NUM_THREADS=4 time ./shalw_50x50 > resultParallelFile_4thread_50x50.txt 29 29 30 30 # 8 thread execution 31 31 echo num threads: 8 32 OMP_NUM_THREADS=8 ./shalw_50x50 > resultParallelFile_8thread_50x50.txt32 OMP_NUM_THREADS=8 time ./shalw_50x50 > resultParallelFile_8thread_50x50.txt 33 33 34 34 … … 40 40 # Serial execution 41 41 echo num threads: serial... 42 ./shalw_100x100 > resultNotParallelFile_100x100.txt42 time ./shalw_100x100 > resultNotParallelFile_100x100.txt 43 43 44 44 YaoI -p +O3 shalw_100x100 45 45 # 1 thread execution 46 46 echo num threads: 1 47 OMP_NUM_THREADS=1 ./shalw_100x100 > resultParallelFile_1thread_100x100.txt47 OMP_NUM_THREADS=1 time ./shalw_100x100 > resultParallelFile_1thread_100x100.txt 48 48 49 49 # 2 thread execution 50 50 echo num threads: 2 51 OMP_NUM_THREADS=2 ./shalw_100x100 > resultParallelFile_2thread_100x100.txt51 OMP_NUM_THREADS=2 time ./shalw_100x100 > resultParallelFile_2thread_100x100.txt 52 52 53 53 # 4 thread execution 54 54 echo num threads: 4 55 OMP_NUM_THREADS=4 ./shalw_100x100 > resultParallelFile_4thread_100x100.txt55 OMP_NUM_THREADS=4 time ./shalw_100x100 > resultParallelFile_4thread_100x100.txt 56 56 57 57 # 8 thread execution 58 58 echo num threads: 8 59 OMP_NUM_THREADS=8 ./shalw_100x100 > resultParallelFile_8thread_100x100.txt59 OMP_NUM_THREADS=8 time ./shalw_100x100 > resultParallelFile_8thread_100x100.txt 60 60 61 61 … … 67 67 # Serial execution 68 68 echo num threads: serial... 69 ./shalw_200x200 > resultNotParallelFile_200x200.txt69 time ./shalw_200x200 > resultNotParallelFile_200x200.txt 70 70 71 71 YaoI -p +O3 shalw_200x200 72 72 # 1 thread execution 73 73 echo num threads: 1 74 OMP_NUM_THREADS=1 ./shalw_200x200 > resultParallelFile_1thread_200x200.txt74 OMP_NUM_THREADS=1 time ./shalw_200x200 > resultParallelFile_1thread_200x200.txt 75 75 76 76 # 2 thread execution 77 77 echo num threads: 2 78 OMP_NUM_THREADS=2 ./shalw_200x200 > resultParallelFile_2thread_200x200.txt78 OMP_NUM_THREADS=2 time ./shalw_200x200 > resultParallelFile_2thread_200x200.txt 79 79 80 80 # 4 thread execution 81 81 echo num threads: 4 82 OMP_NUM_THREADS=4 ./shalw_200x200 > resultParallelFile_4thread_200x200.txt82 OMP_NUM_THREADS=4 time ./shalw_200x200 > resultParallelFile_4thread_200x200.txt 83 83 84 84 # 8 thread execution 85 85 echo num threads: 8 86 OMP_NUM_THREADS=8 ./shalw_200x200 > resultParallelFile_8thread_200x200.txt86 OMP_NUM_THREADS=8 time ./shalw_200x200 > resultParallelFile_8thread_200x200.txt 87 87 88 88 -
trunk/examples/nardi/shallow_performanceMeasurement/shalw.d
r33 r415 8 8 9 9 //option o_gradtest 10 //option o_parallel10 option o_parallel 11 11 option o_m1qn3 12 12
Note: See TracChangeset
for help on using the changeset viewer.