Lines Matching refs:range_j
149 const struct fxdiv_divisor_size_t range_j = threadpool->params.parallelize_2d.range_j; in thread_parallelize_2d() local
150 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(range_start, range_j); in thread_parallelize_2d()
156 if (++j == range_j.value) { in thread_parallelize_2d()
172 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(linear_index, range_j); in thread_parallelize_2d()
196 const size_t range_j = threadpool->params.parallelize_2d_tile_1d.range_j; in thread_parallelize_2d_tile_1d() local
198 task(argument, i, start_j, min(range_j - start_j, tile_j)); in thread_parallelize_2d_tile_1d()
200 if (start_j >= range_j) { in thread_parallelize_2d_tile_1d()
218 task(argument, tile_index_i_j.quotient, start_j, min(range_j - start_j, tile_j)); in thread_parallelize_2d_tile_1d()
243 const size_t range_j = threadpool->params.parallelize_2d_tile_2d.range_j; in thread_parallelize_2d_tile_2d() local
245 task(argument, start_i, start_j, min(range_i - start_i, tile_i), min(range_j - start_j, tile_j)); in thread_parallelize_2d_tile_2d()
247 if (start_j >= range_j) { in thread_parallelize_2d_tile_2d()
266 task(argument, start_i, start_j, min(range_i - start_i, tile_i), min(range_j - start_j, tile_j)); in thread_parallelize_2d_tile_2d()
296 const size_t range_j = threadpool->params.parallelize_2d_tile_2d_with_uarch.range_j; in thread_parallelize_2d_tile_2d_with_uarch() local
302 …task(argument, uarch_index, start_i, start_j, min(range_i - start_i, tile_i), min(range_j - start_… in thread_parallelize_2d_tile_2d_with_uarch()
304 if (start_j >= range_j) { in thread_parallelize_2d_tile_2d_with_uarch()
323 …task(argument, uarch_index, start_i, start_j, min(range_i - start_i, tile_i), min(range_j - start_… in thread_parallelize_2d_tile_2d_with_uarch()
342 const struct fxdiv_divisor_size_t range_j = threadpool->params.parallelize_3d.range_j; in thread_parallelize_3d() local
343 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(index_ij_k.quotient, range_j); in thread_parallelize_3d()
352 if (++j == range_j.value) { in thread_parallelize_3d()
370 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(index_ij_k.quotient, range_j); in thread_parallelize_3d()
390 const struct fxdiv_divisor_size_t range_j = threadpool->params.parallelize_3d_tile_1d.range_j; in thread_parallelize_3d_tile_1d() local
391 …const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(tile_index_ij_k.quotient, range_j… in thread_parallelize_3d_tile_1d()
403 if (++j == range_j.value) { in thread_parallelize_3d_tile_1d()
421 …onst struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(tile_index_ij_k.quotient, range_j); in thread_parallelize_3d_tile_1d()
451 const size_t range_j = threadpool->params.parallelize_3d_tile_2d.range_j; in thread_parallelize_3d_tile_2d() local
453 …task(argument, i, start_j, start_k, min(range_j - start_j, tile_j), min(range_k - start_k, tile_k)… in thread_parallelize_3d_tile_2d()
458 if (start_j >= range_j) { in thread_parallelize_3d_tile_2d()
479 …task(argument, tile_index_i_j.quotient, start_j, start_k, min(range_j - start_j, tile_j), min(rang… in thread_parallelize_3d_tile_2d()
516 const size_t range_j = threadpool->params.parallelize_3d_tile_2d_with_uarch.range_j; in thread_parallelize_3d_tile_2d_with_uarch() local
518 …task(argument, uarch_index, i, start_j, start_k, min(range_j - start_j, tile_j), min(range_k - sta… in thread_parallelize_3d_tile_2d_with_uarch()
523 if (start_j >= range_j) { in thread_parallelize_3d_tile_2d_with_uarch()
544 …task(argument, uarch_index, tile_index_i_j.quotient, start_j, start_k, min(range_j - start_j, tile… in thread_parallelize_3d_tile_2d_with_uarch()
563 const struct fxdiv_divisor_size_t range_j = threadpool->params.parallelize_4d.range_j; in thread_parallelize_4d() local
564 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(index_ij_kl.quotient, range_j); in thread_parallelize_4d()
579 if (++j == range_j.value) { in thread_parallelize_4d()
598 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(index_ij_kl.quotient, range_j); in thread_parallelize_4d()
619 const struct fxdiv_divisor_size_t range_j = threadpool->params.parallelize_4d_tile_1d.range_j; in thread_parallelize_4d_tile_1d() local
620 …nst struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(tile_index_ij_kl.quotient, range_j); in thread_parallelize_4d_tile_1d()
638 if (++j == range_j.value) { in thread_parallelize_4d_tile_1d()
657 …nst struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(tile_index_ij_kl.quotient, range_j); in thread_parallelize_4d_tile_1d()
679 const struct fxdiv_divisor_size_t range_j = threadpool->params.parallelize_4d_tile_2d.range_j; in thread_parallelize_4d_tile_2d() local
680 …nst struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(tile_index_ij_kl.quotient, range_j); in thread_parallelize_4d_tile_2d()
700 if (++j == range_j.value) { in thread_parallelize_4d_tile_2d()
719 …nst struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(tile_index_ij_kl.quotient, range_j); in thread_parallelize_4d_tile_2d()
751 …const struct fxdiv_divisor_size_t range_j = threadpool->params.parallelize_4d_tile_2d_with_uarch.r… in thread_parallelize_4d_tile_2d_with_uarch() local
752 …nst struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(tile_index_ij_kl.quotient, range_j); in thread_parallelize_4d_tile_2d_with_uarch()
772 if (++j == range_j.value) { in thread_parallelize_4d_tile_2d_with_uarch()
791 …nst struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(tile_index_ij_kl.quotient, range_j); in thread_parallelize_4d_tile_2d_with_uarch()
818 const struct fxdiv_divisor_size_t range_j = threadpool->params.parallelize_5d.range_j; in thread_parallelize_5d() local
819 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(index_ij_k.quotient, range_j); in thread_parallelize_5d()
835 if (++j == range_j.value) { in thread_parallelize_5d()
857 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(index_ij_k.quotient, range_j); in thread_parallelize_5d()
879 const struct fxdiv_divisor_size_t range_j = threadpool->params.parallelize_5d_tile_1d.range_j; in thread_parallelize_5d_tile_1d() local
880 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(index_ij_kl.quotient, range_j); in thread_parallelize_5d_tile_1d()
901 if (++j == range_j.value) { in thread_parallelize_5d_tile_1d()
922 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(index_ij_kl.quotient, range_j); in thread_parallelize_5d_tile_1d()
949 const struct fxdiv_divisor_size_t range_j = threadpool->params.parallelize_5d_tile_2d.range_j; in thread_parallelize_5d_tile_2d() local
950 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(index_ij_k.quotient, range_j); in thread_parallelize_5d_tile_2d()
971 if (++j == range_j.value) { in thread_parallelize_5d_tile_2d()
993 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(index_ij_k.quotient, range_j); in thread_parallelize_5d_tile_2d()
1020 const struct fxdiv_divisor_size_t range_j = threadpool->params.parallelize_6d.range_j; in thread_parallelize_6d() local
1021 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(index_ij_k.quotient, range_j); in thread_parallelize_6d()
1042 if (++j == range_j.value) { in thread_parallelize_6d()
1066 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(index_ij_k.quotient, range_j); in thread_parallelize_6d()
1091 const struct fxdiv_divisor_size_t range_j = threadpool->params.parallelize_6d_tile_1d.range_j; in thread_parallelize_6d_tile_1d() local
1092 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(index_ij_k.quotient, range_j); in thread_parallelize_6d_tile_1d()
1116 if (++j == range_j.value) { in thread_parallelize_6d_tile_1d()
1140 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(index_ij_k.quotient, range_j); in thread_parallelize_6d_tile_1d()
1167 const struct fxdiv_divisor_size_t range_j = threadpool->params.parallelize_6d_tile_2d.range_j; in thread_parallelize_6d_tile_2d() local
1168 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(index_ij_kl.quotient, range_j); in thread_parallelize_6d_tile_2d()
1195 if (++j == range_j.value) { in thread_parallelize_6d_tile_2d()
1218 const struct fxdiv_result_size_t index_i_j = fxdiv_divide_size_t(index_ij_kl.quotient, range_j); in thread_parallelize_6d_tile_2d()
1362 size_t range_j, in pthreadpool_parallelize_2d() argument
1366 …ol == NULL || (threads_count = threadpool->threads_count.value) <= 1 || (range_i | range_j) <= 1) { in pthreadpool_parallelize_2d()
1374 for (size_t j = 0; j < range_j; j++) { in pthreadpool_parallelize_2d()
1382 const size_t range = range_i * range_j; in pthreadpool_parallelize_2d()
1384 .range_j = fxdiv_init_size_t(range_j), in pthreadpool_parallelize_2d()
1404 size_t range_j, in pthreadpool_parallelize_2d_tile_1d() argument
1409 …|| (threads_count = threadpool->threads_count.value) <= 1 || (range_i <= 1 && range_j <= tile_j)) { in pthreadpool_parallelize_2d_tile_1d()
1417 for (size_t j = 0; j < range_j; j += tile_j) { in pthreadpool_parallelize_2d_tile_1d()
1418 task(argument, i, j, min(range_j - j, tile_j)); in pthreadpool_parallelize_2d_tile_1d()
1425 const size_t tile_range_j = divide_round_up(range_j, tile_j); in pthreadpool_parallelize_2d_tile_1d()
1428 .range_j = range_j, in pthreadpool_parallelize_2d_tile_1d()
1450 size_t range_j, in pthreadpool_parallelize_2d_tile_2d() argument
1456 …hreads_count = threadpool->threads_count.value) <= 1 || (range_i <= tile_i && range_j <= tile_j)) { in pthreadpool_parallelize_2d_tile_2d()
1464 for (size_t j = 0; j < range_j; j += tile_j) { in pthreadpool_parallelize_2d_tile_2d()
1465 task(argument, i, j, min(range_i - i, tile_i), min(range_j - j, tile_j)); in pthreadpool_parallelize_2d_tile_2d()
1473 const size_t tile_range_j = divide_round_up(range_j, tile_j); in pthreadpool_parallelize_2d_tile_2d()
1478 .range_j = range_j, in pthreadpool_parallelize_2d_tile_2d()
1502 size_t range_j, in pthreadpool_parallelize_2d_tile_2d_with_uarch() argument
1508 …hreads_count = threadpool->threads_count.value) <= 1 || (range_i <= tile_i && range_j <= tile_j)) { in pthreadpool_parallelize_2d_tile_2d_with_uarch()
1525 for (size_t j = 0; j < range_j; j += tile_j) { in pthreadpool_parallelize_2d_tile_2d_with_uarch()
1526 task(argument, uarch_index, i, j, min(range_i - i, tile_i), min(range_j - j, tile_j)); in pthreadpool_parallelize_2d_tile_2d_with_uarch()
1534 const size_t tile_range_j = divide_round_up(range_j, tile_j); in pthreadpool_parallelize_2d_tile_2d_with_uarch()
1541 .range_j = range_j, in pthreadpool_parallelize_2d_tile_2d_with_uarch()
1563 size_t range_j, in pthreadpool_parallelize_3d() argument
1568 … || (threads_count = threadpool->threads_count.value) <= 1 || (range_i | range_j | range_k) <= 1) { in pthreadpool_parallelize_3d()
1576 for (size_t j = 0; j < range_j; j++) { in pthreadpool_parallelize_3d()
1586 const size_t range = range_i * range_j * range_k; in pthreadpool_parallelize_3d()
1588 .range_j = fxdiv_init_size_t(range_j), in pthreadpool_parallelize_3d()
1609 size_t range_j, in pthreadpool_parallelize_3d_tile_1d() argument
1615 …|| (threads_count = threadpool->threads_count.value) <= 1 || ((range_i | range_j) <= 1 && range_k … in pthreadpool_parallelize_3d_tile_1d()
1623 for (size_t j = 0; j < range_j; j++) { in pthreadpool_parallelize_3d_tile_1d()
1634 const size_t tile_range = range_i * range_j * tile_range_k; in pthreadpool_parallelize_3d_tile_1d()
1638 .range_j = fxdiv_init_size_t(range_j), in pthreadpool_parallelize_3d_tile_1d()
1659 size_t range_j, in pthreadpool_parallelize_3d_tile_2d() argument
1666 …hreads_count = threadpool->threads_count.value) <= 1 || (range_i <= 1 && range_j <= tile_j && rang… in pthreadpool_parallelize_3d_tile_2d()
1674 for (size_t j = 0; j < range_j; j += tile_j) { in pthreadpool_parallelize_3d_tile_2d()
1676 task(argument, i, j, k, min(range_j - j, tile_j), min(range_k - k, tile_k)); in pthreadpool_parallelize_3d_tile_2d()
1684 const size_t tile_range_j = divide_round_up(range_j, tile_j); in pthreadpool_parallelize_3d_tile_2d()
1688 .range_j = range_j, in pthreadpool_parallelize_3d_tile_2d()
1715 size_t range_j, in pthreadpool_parallelize_3d_tile_2d_with_uarch() argument
1722 …hreads_count = threadpool->threads_count.value) <= 1 || (range_i <= 1 && range_j <= tile_j && rang… in pthreadpool_parallelize_3d_tile_2d_with_uarch()
1739 for (size_t j = 0; j < range_j; j += tile_j) { in pthreadpool_parallelize_3d_tile_2d_with_uarch()
1741 task(argument, uarch_index, i, j, k, min(range_j - j, tile_j), min(range_k - k, tile_k)); in pthreadpool_parallelize_3d_tile_2d_with_uarch()
1749 const size_t tile_range_j = divide_round_up(range_j, tile_j); in pthreadpool_parallelize_3d_tile_2d_with_uarch()
1755 .range_j = range_j, in pthreadpool_parallelize_3d_tile_2d_with_uarch()
1780 size_t range_j, in pthreadpool_parallelize_4d() argument
1786 … || (threads_count = threadpool->threads_count.value) <= 1 || (range_i | range_j | range_k | range… in pthreadpool_parallelize_4d()
1794 for (size_t j = 0; j < range_j; j++) { in pthreadpool_parallelize_4d()
1807 const size_t range = range_i * range_j * range_kl; in pthreadpool_parallelize_4d()
1810 .range_j = fxdiv_init_size_t(range_j), in pthreadpool_parallelize_4d()
1832 size_t range_j, in pthreadpool_parallelize_4d_tile_1d() argument
1839 …|| (threads_count = threadpool->threads_count.value) <= 1 || ((range_i | range_j | range_k) <= 1 &… in pthreadpool_parallelize_4d_tile_1d()
1847 for (size_t j = 0; j < range_j; j++) { in pthreadpool_parallelize_4d_tile_1d()
1861 const size_t tile_range = range_i * range_j * tile_range_kl; in pthreadpool_parallelize_4d_tile_1d()
1866 .range_j = fxdiv_init_size_t(range_j), in pthreadpool_parallelize_4d_tile_1d()
1888 size_t range_j, in pthreadpool_parallelize_4d_tile_2d() argument
1896 …|| (threads_count = threadpool->threads_count.value) <= 1 || ((range_i | range_j) <= 1 && range_k … in pthreadpool_parallelize_4d_tile_2d()
1904 for (size_t j = 0; j < range_j; j++) { in pthreadpool_parallelize_4d_tile_2d()
1919 const size_t tile_range = range_i * range_j * tile_range_kl; in pthreadpool_parallelize_4d_tile_2d()
1925 .range_j = fxdiv_init_size_t(range_j), in pthreadpool_parallelize_4d_tile_2d()
1949 size_t range_j, in pthreadpool_parallelize_4d_tile_2d_with_uarch() argument
1957 …|| (threads_count = threadpool->threads_count.value) <= 1 || ((range_i | range_j) <= 1 && range_k … in pthreadpool_parallelize_4d_tile_2d_with_uarch()
1974 for (size_t j = 0; j < range_j; j++) { in pthreadpool_parallelize_4d_tile_2d_with_uarch()
1989 const size_t tile_range = range_i * range_j * tile_range_kl; in pthreadpool_parallelize_4d_tile_2d_with_uarch()
1997 .range_j = fxdiv_init_size_t(range_j), in pthreadpool_parallelize_4d_tile_2d_with_uarch()
2019 size_t range_j, in pthreadpool_parallelize_5d() argument
2026 … || (threads_count = threadpool->threads_count.value) <= 1 || (range_i | range_j | range_k | range… in pthreadpool_parallelize_5d()
2034 for (size_t j = 0; j < range_j; j++) { in pthreadpool_parallelize_5d()
2049 const size_t range = range_i * range_j * range_k * range_lm; in pthreadpool_parallelize_5d()
2052 .range_j = fxdiv_init_size_t(range_j), in pthreadpool_parallelize_5d()
2075 size_t range_j, in pthreadpool_parallelize_5d_tile_1d() argument
2083 …|| (threads_count = threadpool->threads_count.value) <= 1 || ((range_i | range_j | range_k | range… in pthreadpool_parallelize_5d_tile_1d()
2091 for (size_t j = 0; j < range_j; j++) { in pthreadpool_parallelize_5d_tile_1d()
2107 const size_t tile_range = range_i * range_j * range_kl * tile_range_m; in pthreadpool_parallelize_5d_tile_1d()
2112 .range_j = fxdiv_init_size_t(range_j), in pthreadpool_parallelize_5d_tile_1d()
2135 size_t range_j, in pthreadpool_parallelize_5d_tile_2d() argument
2144 …|| (threads_count = threadpool->threads_count.value) <= 1 || ((range_i | range_j | range_k) <= 1 &… in pthreadpool_parallelize_5d_tile_2d()
2152 for (size_t j = 0; j < range_j; j++) { in pthreadpool_parallelize_5d_tile_2d()
2169 const size_t tile_range = range_i * range_j * range_k * tile_range_lm; in pthreadpool_parallelize_5d_tile_2d()
2175 .range_j = fxdiv_init_size_t(range_j), in pthreadpool_parallelize_5d_tile_2d()
2198 size_t range_j, in pthreadpool_parallelize_6d() argument
2206 … || (threads_count = threadpool->threads_count.value) <= 1 || (range_i | range_j | range_k | range… in pthreadpool_parallelize_6d()
2214 for (size_t j = 0; j < range_j; j++) { in pthreadpool_parallelize_6d()
2231 const size_t range = range_i * range_j * range_k * range_lmn; in pthreadpool_parallelize_6d()
2234 .range_j = fxdiv_init_size_t(range_j), in pthreadpool_parallelize_6d()
2258 size_t range_j, in pthreadpool_parallelize_6d_tile_1d() argument
2267 …|| (threads_count = threadpool->threads_count.value) <= 1 || ((range_i | range_j | range_k | range… in pthreadpool_parallelize_6d_tile_1d()
2275 for (size_t j = 0; j < range_j; j++) { in pthreadpool_parallelize_6d_tile_1d()
2293 const size_t tile_range = range_i * range_j * range_k * tile_range_lmn; in pthreadpool_parallelize_6d_tile_1d()
2298 .range_j = fxdiv_init_size_t(range_j), in pthreadpool_parallelize_6d_tile_1d()
2322 size_t range_j, in pthreadpool_parallelize_6d_tile_2d() argument
2332 …|| (threads_count = threadpool->threads_count.value) <= 1 || ((range_i | range_j | range_k | range… in pthreadpool_parallelize_6d_tile_2d()
2340 for (size_t j = 0; j < range_j; j++) { in pthreadpool_parallelize_6d_tile_2d()
2360 const size_t tile_range = range_i * range_j * range_kl * tile_range_mn; in pthreadpool_parallelize_6d_tile_2d()
2367 .range_j = fxdiv_init_size_t(range_j), in pthreadpool_parallelize_6d_tile_2d()