| /dports/math/dbcsr/dbcsr-2.1.0/src/acc/libsmm_acc/ |
| H A D | libsmm_acc.cpp | 84 double sumGPU = checkSum(h->mat_c, h->n_c, m, n); in validate_kernel() local
86 if(sumGPU != sumCPU){ in validate_kernel()
87 …nchecksum CPU: %g, checksum GPU: %g\nchecksum_diff: %g\n", m, n, k, sumCPU, sumGPU, sumGPU-sumCPU); in validate_kernel()
339 double sumGPU = checkSumTransp(h->mat_trs_a, h->n_stack_trs_a, m, n); in validate_transpose_kernel() local
341 if(sumGPU != sumCPU){ in validate_transpose_kernel()
342 …%i\nchecksum CPU: %g, checksum GPU: %g\nchecksum_diff: %g\n", m, n, sumCPU, sumGPU, sumGPU-sumCPU); in validate_transpose_kernel()
|
| H A D | libsmm_acc_benchmark.cpp | 291 double sumCPU, sumGPU; in libsmm_acc_benchmark() local
339 sumGPU = checkSum(h->mat_c, h->n_c, mat_m, mat_n); in libsmm_acc_benchmark()
340 if(sumGPU != sumCPU){ in libsmm_acc_benchmark()
341 printf("%sERROR %s checksum_diff: %g\n",msg_prefix, descr, sumGPU-sumCPU); in libsmm_acc_benchmark()
391 double sumCPU, sumGPU; in libsmm_acc_benchmark_transpose_() local
426 sumGPU = checkSumTransp(mat_trs, n_stack, mat_m, mat_n); in libsmm_acc_benchmark_transpose_()
427 if(sumGPU != sumCPU){ in libsmm_acc_benchmark_transpose_()
428 printf("%sERROR %s checksum_diff: %g\n", msg_prefix, descr, sumGPU-sumCPU); in libsmm_acc_benchmark_transpose_()
|
| /dports/science/cp2k/cp2k-2e995eec7fd208c8a72d9544807bd8b8ba8cd1cc/exts/dbcsr/src/acc/libsmm_acc/ |
| H A D | libsmm_acc_benchmark.cpp | 286 double sumCPU, sumGPU; in libsmm_acc_benchmark() local
334 sumGPU = checkSum(h->mat_c, h->n_c, mat_m, mat_n); in libsmm_acc_benchmark()
335 if(sumGPU != sumCPU){ in libsmm_acc_benchmark()
336 printf("%sERROR %s checksum_diff: %g\n",msg_prefix, descr, sumGPU-sumCPU); in libsmm_acc_benchmark()
386 double sumCPU, sumGPU; in libsmm_acc_benchmark_transpose_() local
421 sumGPU = checkSumTransp(mat_trs, n, n_stack, mat_m, mat_n); in libsmm_acc_benchmark_transpose_()
422 if(sumGPU != sumCPU){ in libsmm_acc_benchmark_transpose_()
423 printf("%sERROR %s checksum_diff: %g\n", msg_prefix, descr, sumGPU-sumCPU); in libsmm_acc_benchmark_transpose_()
|
| H A D | libsmm_acc.cpp | 89 double sumGPU = checkSum(h->mat_c, h->n_c, m, n); in validate_kernel() local
90 if(sumGPU != sumCPU){ in validate_kernel()
91 …rnel %ix%ix%i\nchecksum_diff: %g\nthreads: %i, grouping: %i\n", m, n, k, sumGPU-sumCPU, threads, g… in validate_kernel()
|
| /dports/devel/bullet/bullet3-3.21/test/OpenCL/ParallelPrimitives/ |
| H A D | main.cpp | 277 unsigned int sumHost, sumGPU; in prefixScanTest() local
280 scan->execute(buf2CL, buf3CL, size, &sumGPU); in prefixScanTest()
284 TEST_ASSERT(sumHost == sumGPU); in prefixScanTest()
|
| /dports/devel/py-bullet3/bullet3-3.21/test/OpenCL/ParallelPrimitives/ |
| H A D | main.cpp | 277 unsigned int sumHost, sumGPU; in prefixScanTest() local
280 scan->execute(buf2CL, buf3CL, size, &sumGPU); in prefixScanTest()
284 TEST_ASSERT(sumHost == sumGPU); in prefixScanTest()
|