From f87bd49b66a303706c4b13e6003fad105ed3aa38 Mon Sep 17 00:00:00 2001 From: dumerrill Date: Mon, 14 Dec 2015 13:12:25 -0500 Subject: [PATCH] 1.5.0 doxgen added bool sorting test Former-commit-id: 9607db2517fced1c3bda71dac951007be0a02300 --- CHANGE_LOG.TXT | 15 + cub/block/block_load.cuh | 3 +- docs/download_cub.html | 6 +- docs/html/CHANGE_LOG.TXT | 15 + docs/html/annotated.html | 121 ++-- .../arg__index__input__iterator_8cuh.html | 4 +- ...g__index__input__iterator_8cuh_source.html | 46 +- docs/html/block__discontinuity_8cuh.html | 2 +- .../block__discontinuity_8cuh_source.html | 4 +- docs/html/block__exchange_8cuh.html | 2 +- ...__exchange_8cuh_source.html.REMOVED.git-id | 2 +- docs/html/block__histogram_8cuh.html | 2 +- docs/html/block__histogram_8cuh_source.html | 4 +- docs/html/block__load_8cuh.html | 2 +- ...lock__load_8cuh_source.html.REMOVED.git-id | 2 +- docs/html/block__radix__sort_8cuh.html | 2 +- docs/html/block__radix__sort_8cuh_source.html | 60 +- docs/html/block__reduce_8cuh.html | 2 +- docs/html/block__reduce_8cuh_source.html | 4 +- docs/html/block__scan_8cuh.html | 2 +- ...lock__scan_8cuh_source.html.REMOVED.git-id | 2 +- docs/html/block__store_8cuh.html | 2 +- docs/html/block__store_8cuh_source.html | 16 +- ...cache__modified__input__iterator_8cuh.html | 2 +- ...modified__input__iterator_8cuh_source.html | 2 +- ...ache__modified__output__iterator_8cuh.html | 2 +- ...odified__output__iterator_8cuh_source.html | 2 +- ..._1_1_arg_index_input_iterator-members.html | 37 +- ...classcub_1_1_arg_index_input_iterator.html | 25 +- ...sscub_1_1_block_discontinuity-members.html | 2 +- .../classcub_1_1_block_discontinuity.html | 2 +- .../classcub_1_1_block_exchange-members.html | 2 +- docs/html/classcub_1_1_block_exchange.html | 2 +- .../classcub_1_1_block_histogram-members.html | 2 +- docs/html/classcub_1_1_block_histogram.html | 2 +- .../html/classcub_1_1_block_load-members.html | 2 +- docs/html/classcub_1_1_block_load.html | 12 +- ...classcub_1_1_block_radix_sort-members.html | 2 +- docs/html/classcub_1_1_block_radix_sort.html | 4 +- .../classcub_1_1_block_reduce-members.html | 2 +- docs/html/classcub_1_1_block_reduce.html | 14 +- .../html/classcub_1_1_block_scan-members.html | 2 +- ...lasscub_1_1_block_scan.html.REMOVED.git-id | 2 +- .../classcub_1_1_block_store-members.html | 2 +- docs/html/classcub_1_1_block_store.html | 2 +- ...cache_modified_input_iterator-members.html | 2 +- ...cub_1_1_cache_modified_input_iterator.html | 2 +- ...ache_modified_output_iterator-members.html | 2 +- ...ub_1_1_cache_modified_output_iterator.html | 2 +- ...b_1_1_constant_input_iterator-members.html | 2 +- .../classcub_1_1_constant_input_iterator.html | 2 +- ...b_1_1_counting_input_iterator-members.html | 2 +- .../classcub_1_1_counting_input_iterator.html | 2 +- .../classcub_1_1_swizzle_scan_op-members.html | 2 +- docs/html/classcub_1_1_swizzle_scan_op.html | 2 +- ...ub_1_1_tex_obj_input_iterator-members.html | 2 +- .../classcub_1_1_tex_obj_input_iterator.html | 2 +- ...ub_1_1_tex_ref_input_iterator-members.html | 2 +- .../classcub_1_1_tex_ref_input_iterator.html | 2 +- ..._1_1_transform_input_iterator-members.html | 2 +- ...classcub_1_1_transform_input_iterator.html | 2 +- .../classcub_1_1_warp_reduce-members.html | 2 +- docs/html/classcub_1_1_warp_reduce.html | 2 +- docs/html/classcub_1_1_warp_scan-members.html | 2 +- ...classcub_1_1_warp_scan.html.REMOVED.git-id | 2 +- docs/html/classes.html | 69 +- docs/html/constant__input__iterator_8cuh.html | 2 +- ...constant__input__iterator_8cuh_source.html | 2 +- docs/html/counting__input__iterator_8cuh.html | 2 +- ...counting__input__iterator_8cuh_source.html | 2 +- docs/html/cub_8cuh.html | 4 +- docs/html/cub_8cuh_source.html | 88 +-- docs/html/device__histogram_8cuh.html | 2 +- docs/html/device__histogram_8cuh_source.html | 638 +++++++++--------- docs/html/device__partition_8cuh.html | 2 +- docs/html/device__partition_8cuh_source.html | 152 ++--- docs/html/device__radix__sort_8cuh.html | 2 +- .../html/device__radix__sort_8cuh_source.html | 554 +++++++-------- docs/html/device__reduce_8cuh.html | 3 +- docs/html/device__reduce_8cuh_source.html | 488 +++++++------- .../device__run__length__encode_8cuh.html | 2 +- ...vice__run__length__encode_8cuh_source.html | 176 ++--- docs/html/device__scan_8cuh.html | 2 +- docs/html/device__scan_8cuh_source.html | 268 ++++---- .../device__segmented__radix__sort_8cuh.html | 2 +- ...e__segmented__radix__sort_8cuh_source.html | 633 +++++++++-------- docs/html/device__segmented__reduce_8cuh.html | 3 +- ...device__segmented__reduce_8cuh_source.html | 461 ++++++------- docs/html/device__select_8cuh.html | 4 +- docs/html/device__select_8cuh_source.html | 224 +++--- docs/html/device__spmv_8cuh.html | 2 +- docs/html/device__spmv_8cuh_source.html | 2 +- .../dir_011e1c944d88f71be72e1e24a5fda7cf.html | 2 +- .../dir_18fc672d63781b5a743137aee24ff656.html | 4 +- .../dir_80932b4cec52750ff92b1a1912314cf5.html | 2 +- .../dir_bb50a5ef59f19d030d06415663184d05.html | 2 +- .../dir_cb3a671affffe7eeb3fdf5ae58e42cc8.html | 2 +- .../dir_d583f216f1aafe19404e836b0c097ad2.html | 2 +- docs/html/download_cub.html | 6 +- .../example_block_radix_sort_8cu-example.html | 2 +- .../example_block_reduce_8cu-example.html | 2 +- docs/html/example_block_scan_8cu-example.html | 2 +- .../example_device_histogram_8cu-example.html | 2 +- ..._device_partition_flagged_8cu-example.html | 2 +- ...ample_device_partition_if_8cu-example.html | 2 +- ...example_device_radix_sort_8cu-example.html | 36 +- .../example_device_reduce_8cu-example.html | 6 +- .../html/example_device_scan_8cu-example.html | 2 +- ...ple_device_select_flagged_8cu-example.html | 2 +- .../example_device_select_if_8cu-example.html | 2 +- ...mple_device_select_unique_8cu-example.html | 2 +- docs/html/examples.html | 2 +- docs/html/files.html | 35 +- docs/html/functions.html | 14 +- docs/html/functions_0x62.html | 3 +- docs/html/functions_0x63.html | 6 +- docs/html/functions_0x64.html | 18 +- docs/html/functions_0x65.html | 3 +- docs/html/functions_0x66.html | 3 +- docs/html/functions_0x68.html | 3 +- docs/html/functions_0x69.html | 3 +- docs/html/functions_0x6c.html | 3 +- docs/html/functions_0x6d.html | 9 +- docs/html/functions_0x6e.html | 6 +- docs/html/functions_0x6f.html | 80 ++- docs/html/functions_0x70.html | 3 +- docs/html/functions_0x72.html | 15 +- docs/html/functions_0x73.html | 35 +- docs/html/functions_0x74.html | 3 +- docs/html/functions_0x75.html | 3 +- docs/html/functions_0x76.html | 15 +- docs/html/functions_0x77.html | 3 +- docs/html/functions_0x7e.html | 3 +- docs/html/functions_func.html | 10 +- docs/html/functions_func_0x62.html | 2 +- docs/html/functions_func_0x63.html | 5 +- docs/html/functions_func_0x64.html | 7 +- docs/html/functions_func_0x65.html | 2 +- docs/html/functions_func_0x66.html | 2 +- docs/html/functions_func_0x68.html | 2 +- docs/html/functions_func_0x69.html | 2 +- docs/html/functions_func_0x6c.html | 2 +- docs/html/functions_func_0x6d.html | 8 +- docs/html/functions_func_0x6e.html | 5 +- docs/html/functions_func_0x6f.html | 80 +-- docs/html/functions_func_0x72.html | 10 +- docs/html/functions_func_0x73.html | 17 +- docs/html/functions_func_0x74.html | 2 +- docs/html/functions_func_0x75.html | 2 +- docs/html/functions_func_0x77.html | 2 +- docs/html/functions_func_0x7e.html | 2 +- docs/html/functions_rela.html | 2 +- docs/html/functions_type.html | 24 +- docs/html/functions_vars.html | 20 +- docs/html/globals.html | 8 +- docs/html/globals_defs.html | 8 +- docs/html/group___block_module.html | 2 +- docs/html/group___collective_module.html | 2 +- docs/html/group___device_module.html | 2 +- docs/html/group___segmented_module.html | 10 +- docs/html/group___single_module.html | 2 +- docs/html/group___util_io.html | 20 +- docs/html/group___util_iterator.html | 4 +- docs/html/group___util_mgmt.html | 12 +- docs/html/group___util_module.html | 20 +- docs/html/group___util_ptx.html | 2 +- docs/html/group___warp_module.html | 2 +- docs/html/hierarchy.html | 105 ++- docs/html/index.html | 12 +- docs/html/modules.html | 2 +- docs/html/namespacecub.html.REMOVED.git-id | 2 +- docs/html/namespacemembers.html | 2 +- docs/html/namespacemembers_enum.html | 2 +- docs/html/namespacemembers_eval.html | 2 +- docs/html/namespacemembers_func.html | 2 +- docs/html/namespaces.html | 2 +- docs/html/search/all_61.js | 9 +- docs/html/search/all_63.js | 7 +- docs/html/search/all_64.js | 6 +- docs/html/search/all_69.js | 1 - docs/html/search/all_6d.js | 4 +- docs/html/search/all_6e.js | 2 +- docs/html/search/all_6f.js | 2 +- docs/html/search/all_72.js | 4 +- docs/html/search/all_73.js | 10 +- docs/html/search/all_74.js | 18 +- docs/html/search/all_75.js | 2 - docs/html/search/all_76.js | 1 - docs/html/search/classes_63.js | 3 +- docs/html/search/classes_64.js | 4 +- docs/html/search/classes_69.js | 1 - docs/html/search/classes_74.js | 20 +- docs/html/search/files_64.js | 1 + docs/html/search/functions_61.js | 6 +- docs/html/search/functions_63.js | 3 +- docs/html/search/functions_64.js | 3 +- docs/html/search/functions_6d.js | 4 +- docs/html/search/functions_6e.js | 3 +- docs/html/search/functions_6f.js | 2 +- docs/html/search/functions_72.js | 4 +- docs/html/search/functions_73.js | 4 +- docs/html/search/search.js | 8 +- docs/html/search/typedefs_64.js | 1 - docs/html/search/typedefs_76.js | 1 - docs/html/structcub_1_1_arg_max-members.html | 2 +- docs/html/structcub_1_1_arg_max.html | 6 +- docs/html/structcub_1_1_arg_min-members.html | 2 +- docs/html/structcub_1_1_arg_min.html | 6 +- ..._block_discontinuity_1_1_temp_storage.html | 28 +- ...1_block_discontinuity_1_1_temp_storage.png | Bin 1495 -> 1458 bytes ...b_1_1_block_exchange_1_1_temp_storage.html | 28 +- ...ub_1_1_block_exchange_1_1_temp_storage.png | Bin 1962 -> 1931 bytes ..._1_1_block_histogram_1_1_temp_storage.html | 28 +- ...b_1_1_block_histogram_1_1_temp_storage.png | Bin 1956 -> 1926 bytes ...00_01_d_u_m_m_y_01_4_1_1_temp_storage.html | 30 +- ..._00_01_d_u_m_m_y_01_4_1_1_temp_storage.png | Bin 2382 -> 2345 bytes ..._o_s_402c3164d23f1ec647db5dad06a54584.html | 30 +- ...p_o_s_402c3164d23f1ec647db5dad06a54584.png | Bin 2406 -> 2375 bytes ..._o_s_e4c36dfe8f549604998f6c46cc8fbd1d.html | 30 +- ...p_o_s_e4c36dfe8f549604998f6c46cc8fbd1d.png | Bin 2520 -> 2487 bytes ...ctcub_1_1_block_load_1_1_temp_storage.html | 30 +- ...uctcub_1_1_block_load_1_1_temp_storage.png | Bin 1919 -> 1890 bytes ...1_1_block_radix_sort_1_1_temp_storage.html | 28 +- ..._1_1_block_radix_sort_1_1_temp_storage.png | Bin 2567 -> 2538 bytes ...cub_1_1_block_reduce_1_1_temp_storage.html | 28 +- ...tcub_1_1_block_reduce_1_1_temp_storage.png | Bin 1576 -> 1538 bytes ...ctcub_1_1_block_scan_1_1_temp_storage.html | 28 +- ...uctcub_1_1_block_scan_1_1_temp_storage.png | Bin 1563 -> 1531 bytes ..._00_09dfae03f13932c7dbdb41be30a5767ba.html | 28 +- ...e_00_09dfae03f13932c7dbdb41be30a5767ba.png | Bin 2399 -> 2361 bytes ..._s_p_263becc1ca5b47586740c2f7bb0d0145.html | 28 +- ...n_s_p_263becc1ca5b47586740c2f7bb0d0145.png | Bin 2595 -> 2561 bytes ..._s_p_8d170856b7ed1df0ed565731a681b449.html | 28 +- ...n_s_p_8d170856b7ed1df0ed565731a681b449.png | Bin 2439 -> 2421 bytes ...tcub_1_1_block_store_1_1_temp_storage.html | 28 +- ...ctcub_1_1_block_store_1_1_temp_storage.png | Bin 1929 -> 1899 bytes ..._1_1_caching_device_allocator-members.html | 2 +- ...tructcub_1_1_caching_device_allocator.html | 2 +- docs/html/structcub_1_1_cast-members.html | 2 +- docs/html/structcub_1_1_cast.html | 2 +- ...tructcub_1_1_device_histogram-members.html | 2 +- ...b_1_1_device_histogram.html.REMOVED.git-id | 2 +- ...tructcub_1_1_device_partition-members.html | 2 +- docs/html/structcub_1_1_device_partition.html | 14 +- ...ructcub_1_1_device_radix_sort-members.html | 2 +- .../html/structcub_1_1_device_radix_sort.html | 94 ++- .../structcub_1_1_device_reduce-members.html | 6 +- docs/html/structcub_1_1_device_reduce.html | 190 +++--- ..._1_1_device_run_length_encode-members.html | 2 +- ...tructcub_1_1_device_run_length_encode.html | 14 +- .../structcub_1_1_device_scan-members.html | 2 +- docs/html/structcub_1_1_device_scan.html | 31 +- ...1_device_segmented_radix_sort-members.html | 4 +- ...e_segmented_radix_sort.html.REMOVED.git-id | 2 +- ...b_1_1_device_segmented_reduce-members.html | 14 +- ...structcub_1_1_device_segmented_reduce.html | 270 ++++---- .../structcub_1_1_device_select-members.html | 2 +- docs/html/structcub_1_1_device_select.html | 19 +- .../structcub_1_1_device_spmv-members.html | 2 +- docs/html/structcub_1_1_device_spmv.html | 4 +- docs/html/structcub_1_1_equality-members.html | 2 +- docs/html/structcub_1_1_equality.html | 2 +- docs/html/structcub_1_1_equals-members.html | 2 +- docs/html/structcub_1_1_equals.html | 4 +- docs/html/structcub_1_1_if-members.html | 2 +- docs/html/structcub_1_1_if.html | 4 +- .../structcub_1_1_inequality-members.html | 2 +- docs/html/structcub_1_1_inequality.html | 2 +- ...uctcub_1_1_inequality_wrapper-members.html | 2 +- .../structcub_1_1_inequality_wrapper.html | 2 +- .../structcub_1_1_is_pointer-members.html | 2 +- docs/html/structcub_1_1_is_pointer.html | 8 +- .../structcub_1_1_is_volatile-members.html | 2 +- docs/html/structcub_1_1_is_volatile.html | 8 +- docs/html/structcub_1_1_log2-members.html | 2 +- docs/html/structcub_1_1_log2.html | 10 +- docs/html/structcub_1_1_max-members.html | 2 +- docs/html/structcub_1_1_max.html | 2 +- docs/html/structcub_1_1_min-members.html | 2 +- docs/html/structcub_1_1_min.html | 2 +- .../structcub_1_1_power_of_two-members.html | 2 +- docs/html/structcub_1_1_power_of_two.html | 8 +- ...tructcub_1_1_reduce_by_key_op-members.html | 2 +- docs/html/structcub_1_1_reduce_by_key_op.html | 2 +- ...tcub_1_1_reduce_by_segment_op-members.html | 2 +- .../structcub_1_1_reduce_by_segment_op.html | 8 +- ...ructcub_1_1_remove_qualifiers-members.html | 2 +- .../html/structcub_1_1_remove_qualifiers.html | 4 +- docs/html/structcub_1_1_sum-members.html | 2 +- docs/html/structcub_1_1_sum.html | 2 +- ...tcub_1_1_warp_reduce_1_1_temp_storage.html | 28 +- ...ctcub_1_1_warp_reduce_1_1_temp_storage.png | Bin 1310 -> 1280 bytes ...uctcub_1_1_warp_scan_1_1_temp_storage.html | 28 +- ...ructcub_1_1_warp_scan_1_1_temp_storage.png | Bin 1287 -> 1259 bytes docs/html/tex__obj__input__iterator_8cuh.html | 2 +- ...tex__obj__input__iterator_8cuh_source.html | 2 +- docs/html/tex__ref__input__iterator_8cuh.html | 2 +- ...tex__ref__input__iterator_8cuh_source.html | 2 +- docs/html/thread__load_8cuh.html | 2 +- docs/html/thread__load_8cuh_source.html | 38 +- docs/html/thread__operators_8cuh.html | 2 +- docs/html/thread__operators_8cuh_source.html | 18 +- docs/html/thread__store_8cuh.html | 2 +- docs/html/thread__store_8cuh_source.html | 137 ++-- .../html/transform__input__iterator_8cuh.html | 2 +- ...ransform__input__iterator_8cuh_source.html | 2 +- docs/html/util__allocator_8cuh_source.html | 14 +- docs/html/util__arch_8cuh.html | 2 +- docs/html/util__arch_8cuh_source.html | 2 +- docs/html/util__debug_8cuh.html | 10 +- docs/html/util__debug_8cuh_source.html | 8 +- docs/html/util__device_8cuh.html | 5 +- docs/html/util__device_8cuh_source.html | 448 ++++++------ docs/html/util__ptx_8cuh.html | 2 +- docs/html/util__ptx_8cuh_source.html | 4 +- docs/html/util__type_8cuh.html | 21 +- ...util__type_8cuh_source.html.REMOVED.git-id | 2 +- docs/html/warp__reduce_8cuh.html | 2 +- docs/html/warp__reduce_8cuh_source.html | 4 +- docs/html/warp__scan_8cuh.html | 2 +- docs/html/warp__scan_8cuh_source.html | 4 +- docs/mainpage.dox | 6 +- test/test_device_radix_sort.cu | 3 + test/test_util.h | 25 + 324 files changed, 3345 insertions(+), 3878 deletions(-) diff --git a/CHANGE_LOG.TXT b/CHANGE_LOG.TXT index 202ec6f160..96973a5e90 100644 --- a/CHANGE_LOG.TXT +++ b/CHANGE_LOG.TXT @@ -1,5 +1,20 @@ //----------------------------------------------------------------------------- +1.5.0 12/14/2015 + - New Features: + - Added new segmented device-wide operations for device-wide sort and + reduction primitives. + - Bug fixes: + - Fix for Git Issue 36 (Compilation error with GCC 4.8.4 nvcc 7.0.27) and + Forums thread (ThreadLoad generates compiler errors when loading from + pointer-to-const) + - Fix for Git Issue 29 (DeviceRadixSort::SortKeys yields compiler + errors) + - Fix for Git Issue 26 (CUDA error: misaligned address after + cub::DeviceRadixSort::SortKeys()) + +//----------------------------------------------------------------------------- + 1.4.1 04/13/2015 - Bug fixes: - Fixes for CUDA 7.0 issues with SHFL-based warp-scan and warp-reduction diff --git a/cub/block/block_load.cuh b/cub/block/block_load.cuh index 2df8cfebb5..4803576922 100644 --- a/cub/block/block_load.cuh +++ b/cub/block/block_load.cuh @@ -200,7 +200,8 @@ __device__ __forceinline__ void InternalLoadDirectBlockedVectorized( #pragma unroll for (int ITEM = 0; ITEM < ITEMS_PER_THREAD; ITEM++) { - items[ITEM] = reinterpret_cast(vec_items)[ITEM]; +// items[ITEM] = reinterpret_cast(vec_items)[ITEM]; + items[ITEM] = *(reinterpret_cast(vec_items) + ITEM); } } diff --git a/docs/download_cub.html b/docs/download_cub.html index e900a57c53..a9f8ae9530 100644 --- a/docs/download_cub.html +++ b/docs/download_cub.html @@ -37,14 +37,14 @@
If your download doesn't start in 3s:

- -Download CUB! + +Download CUB!
diff --git a/docs/html/CHANGE_LOG.TXT b/docs/html/CHANGE_LOG.TXT index 202ec6f160..96973a5e90 100644 --- a/docs/html/CHANGE_LOG.TXT +++ b/docs/html/CHANGE_LOG.TXT @@ -1,5 +1,20 @@ //----------------------------------------------------------------------------- +1.5.0 12/14/2015 + - New Features: + - Added new segmented device-wide operations for device-wide sort and + reduction primitives. + - Bug fixes: + - Fix for Git Issue 36 (Compilation error with GCC 4.8.4 nvcc 7.0.27) and + Forums thread (ThreadLoad generates compiler errors when loading from + pointer-to-const) + - Fix for Git Issue 29 (DeviceRadixSort::SortKeys yields compiler + errors) + - Fix for Git Issue 26 (CUDA error: misaligned address after + cub::DeviceRadixSort::SortKeys()) + +//----------------------------------------------------------------------------- + 1.4.1 04/13/2015 - Bug fixes: - Fixes for CUDA 7.0 issues with SHFL-based warp-scan and warp-reduction diff --git a/docs/html/annotated.html b/docs/html/annotated.html index a31cdfc151..2398acf04e 100644 --- a/docs/html/annotated.html +++ b/docs/html/annotated.html @@ -103,145 +103,144 @@  oCCachingDeviceAllocatorA simple caching allocator for device memory allocations  oCIfType selection (IF ? ThenType : ElseType)  oCEqualsType equality test - oCNullTypeA simple "NULL" marker type - oCInt2TypeAllows for the treatment of an integral constant as a type at compile-time (e.g., to achieve static call dispatch based on constant integral values) - oCCubVectorExposes a member typedef Type that names the corresponding CUDA vector type if one exists. Otherwise Type refers to the CubVector structure itself, which will wrap the corresponding x, y, etc. vector fields - oCUninitializedA storage-backing wrapper that allows types with non-trivial constructors to be aliased in unions - oCKeyValuePairA key identifier paired with a corresponding value - oCDoubleBufferDouble-buffer storage wrapper for multi-pass stream transformations that require more than one storage array for streaming intermediate results back and forth - oCLog2Statically determine log2(N), rounded up - oCPowerOfTwoStatically determine if N is a power-of-two - oCIsPointerPointer vs. iterator - oCIsVolatileVolatile modifier test - oCRemoveQualifiersRemoves const and volatile qualifiers from type Tp - oCArgIndexInputIteratorA random-access input wrapper for pairing dereferenced values with their corresponding indices (forming KeyValuePair tuples) - oCCacheModifiedInputIteratorA random-access input wrapper for dereferencing array values using a PTX cache load modifier - oCCacheModifiedOutputIteratorA random-access output wrapper for storing array values using a PTX cache-modifier - oCConstantInputIteratorA random-access input generator for dereferencing a sequence of homogeneous values - oCCountingInputIteratorA random-access input generator for dereferencing a sequence of incrementing integer values - oCTexObjInputIteratorA random-access input wrapper for dereferencing array values through texture cache. Uses newer Kepler-style texture objects - oCTexRefInputIteratorA random-access input wrapper for dereferencing array values through texture cache. Uses older Tesla/Fermi-style texture references - oCTransformInputIteratorA random-access input wrapper for transforming dereferenced values - oCEqualityDefault equality functor - oCInequalityDefault inequality functor - oCInequalityWrapperInequality functor (wraps equality functor) - oCSumDefault sum functor - oCMaxDefault max functor - oCArgMaxArg max functor (keeps the value and offset of the first occurrence of the larger item) - oCMinDefault min functor - oCArgMinArg min functor (keeps the value and offset of the first occurrence of the smallest item) - oCCastDefault cast functor - oCSwizzleScanOpBinary operator wrapper for switching non-commutative scan arguments - oCReduceBySegmentOpReduce-by-segment functor - oCReduceByKeyOp< Binary reduction operator to apply to values - oCBlockDiscontinuityThe BlockDiscontinuity class provides collective methods for flagging discontinuities within an ordered set of items partitioned across a CUDA thread block.

+ oCLog2Statically determine log2(N), rounded up + oCPowerOfTwoStatically determine if N is a power-of-two + oCIsPointerPointer vs. iterator + oCIsVolatileVolatile modifier test + oCRemoveQualifiersRemoves const and volatile qualifiers from type Tp + oCArgIndexInputIteratorA random-access input wrapper for pairing dereferenced values with their corresponding indices (forming KeyValuePair tuples) + oCCacheModifiedInputIteratorA random-access input wrapper for dereferencing array values using a PTX cache load modifier + oCCacheModifiedOutputIteratorA random-access output wrapper for storing array values using a PTX cache-modifier + oCConstantInputIteratorA random-access input generator for dereferencing a sequence of homogeneous values + oCCountingInputIteratorA random-access input generator for dereferencing a sequence of incrementing integer values + oCTexObjInputIteratorA random-access input wrapper for dereferencing array values through texture cache. Uses newer Kepler-style texture objects + oCTexRefInputIteratorA random-access input wrapper for dereferencing array values through texture cache. Uses older Tesla/Fermi-style texture references + oCTransformInputIteratorA random-access input wrapper for transforming dereferenced values + oCEqualityDefault equality functor + oCInequalityDefault inequality functor + oCInequalityWrapperInequality functor (wraps equality functor) + oCSumDefault sum functor + oCMaxDefault max functor + oCArgMaxArg max functor (keeps the value and offset of the first occurrence of the larger item) + oCMinDefault min functor + oCArgMinArg min functor (keeps the value and offset of the first occurrence of the smallest item) + oCCastDefault cast functor + oCSwizzleScanOpBinary operator wrapper for switching non-commutative scan arguments + oCReduceBySegmentOpReduce-by-segment functor + oCReduceByKeyOp< Binary reduction operator to apply to values + oCBlockDiscontinuityThe BlockDiscontinuity class provides collective methods for flagging discontinuities within an ordered set of items partitioned across a CUDA thread block.

discont_logo.png
- |\CTempStorageThe operations exposed by BlockDiscontinuity require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse - oCBlockExchangeThe BlockExchange class provides collective methods for rearranging data partitioned across a CUDA thread block.

+ |\CTempStorageThe operations exposed by BlockDiscontinuity require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse + oCBlockExchangeThe BlockExchange class provides collective methods for rearranging data partitioned across a CUDA thread block.

transpose_logo.png
- |\CTempStorageThe operations exposed by BlockExchange require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse - oCBlockHistogramThe BlockHistogram class provides collective methods for constructing block-wide histograms from data samples partitioned across a CUDA thread block.

+ |\CTempStorageThe operations exposed by BlockExchange require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse + oCBlockHistogramThe BlockHistogram class provides collective methods for constructing block-wide histograms from data samples partitioned across a CUDA thread block.

histogram_logo.png
- |\CTempStorageThe operations exposed by BlockHistogram require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse - oCBlockLoadThe BlockLoad class provides collective data movement methods for loading a linear segment of items from memory into a blocked arrangement across a CUDA thread block.

+ |\CTempStorageThe operations exposed by BlockHistogram require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse + oCBlockLoadThe BlockLoad class provides collective data movement methods for loading a linear segment of items from memory into a blocked arrangement across a CUDA thread block.

block_load_logo.png
- |\CTempStorageThe operations exposed by BlockLoad require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse - oCBlockRadixSortThe BlockRadixSort class provides collective methods for sorting items partitioned across a CUDA thread block using a radix sorting method.

+ |\CTempStorageThe operations exposed by BlockLoad require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse + oCBlockRadixSortThe BlockRadixSort class provides collective methods for sorting items partitioned across a CUDA thread block using a radix sorting method.

sorting_logo.png
- |\CTempStorageThe operations exposed by BlockScan require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse - oCBlockReduceThe BlockReduce class provides collective methods for computing a parallel reduction of items partitioned across a CUDA thread block.

+ |\CTempStorageThe operations exposed by BlockScan require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse + oCBlockReduceThe BlockReduce class provides collective methods for computing a parallel reduction of items partitioned across a CUDA thread block.

reduce_logo.png
- |\CTempStorageThe operations exposed by BlockReduce require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse - oCBlockScanThe BlockScan class provides collective methods for computing a parallel prefix sum/scan of items partitioned across a CUDA thread block.

+ |\CTempStorageThe operations exposed by BlockReduce require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse + oCBlockScanThe BlockScan class provides collective methods for computing a parallel prefix sum/scan of items partitioned across a CUDA thread block.

block_scan_logo.png
- |\CTempStorageThe operations exposed by BlockScan require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse - oCBlockStoreThe BlockStore class provides collective data movement methods for writing a blocked arrangement of items partitioned across a CUDA thread block to a linear segment of memory.

+ |\CTempStorageThe operations exposed by BlockScan require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse + oCBlockStoreThe BlockStore class provides collective data movement methods for writing a blocked arrangement of items partitioned across a CUDA thread block to a linear segment of memory.

block_store_logo.png
- |\CTempStorageThe operations exposed by BlockStore require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse - oCDeviceHistogramDeviceHistogram provides device-wide parallel operations for constructing histogram(s) from a sequence of samples data residing within device-accessible memory.

+ |\CTempStorageThe operations exposed by BlockStore require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse + oCDeviceHistogramDeviceHistogram provides device-wide parallel operations for constructing histogram(s) from a sequence of samples data residing within device-accessible memory.

histogram_logo.png
- oCDevicePartitionDevicePartition provides device-wide, parallel operations for partitioning sequences of data items residing within device-accessible memory.

+ oCDevicePartitionDevicePartition provides device-wide, parallel operations for partitioning sequences of data items residing within device-accessible memory.

partition_logo.png
- oCDeviceRadixSortDeviceRadixSort provides device-wide, parallel operations for computing a radix sort across a sequence of data items residing within device-accessible memory.

+ oCDeviceRadixSortDeviceRadixSort provides device-wide, parallel operations for computing a radix sort across a sequence of data items residing within device-accessible memory.

sorting_logo.png
- oCDeviceReduceDeviceReduce provides device-wide, parallel operations for computing a reduction across a sequence of data items residing within device-accessible memory.

+ oCDeviceReduceDeviceReduce provides device-wide, parallel operations for computing a reduction across a sequence of data items residing within device-accessible memory.

reduce_logo.png
- oCDeviceRunLengthEncodeDeviceRunLengthEncode provides device-wide, parallel operations for demarcating "runs" of same-valued items within a sequence residing within device-accessible memory.

+ oCDeviceRunLengthEncodeDeviceRunLengthEncode provides device-wide, parallel operations for demarcating "runs" of same-valued items within a sequence residing within device-accessible memory.

run_length_encode_logo.png
- oCDeviceScanDeviceScan provides device-wide, parallel operations for computing a prefix scan across a sequence of data items residing within device-accessible memory.

+ oCDeviceScanDeviceScan provides device-wide, parallel operations for computing a prefix scan across a sequence of data items residing within device-accessible memory.

device_scan.png
- oCDeviceSegmentedRadixSortDeviceSegmentedRadixSort provides device-wide, parallel operations for computing a batched radix sort across multiple, non-overlapping sequences of data items residing within device-accessible memory.

+ oCDeviceSegmentedRadixSortDeviceSegmentedRadixSort provides device-wide, parallel operations for computing a batched radix sort across multiple, non-overlapping sequences of data items residing within device-accessible memory.

segmented_sorting_logo.png
- oCDeviceSelectDeviceSelect provides device-wide, parallel operations for compacting selected items from sequences of data items residing within device-accessible memory.

+ oCDeviceSegmentedReduceDeviceSegmentedReduce provides device-wide, parallel operations for computing a reduction across multiple sequences of data items residing within device-accessible memory.

+
+reduce_logo.png +
+ + oCDeviceSelectDeviceSelect provides device-wide, parallel operations for compacting selected items from sequences of data items residing within device-accessible memory.

select_logo.png
- oCDeviceSpmvDeviceSpmv provides device-wide parallel operations for performing sparse-matrix * dense-vector multiplication (SpMV) - oCWarpScanThe WarpScan class provides collective methods for computing a parallel prefix scan of items partitioned across a CUDA thread warp.

+ oCDeviceSpmvDeviceSpmv provides device-wide parallel operations for performing sparse-matrix * dense-vector multiplication (SpMV) + oCWarpScanThe WarpScan class provides collective methods for computing a parallel prefix scan of items partitioned across a CUDA thread warp.

warp_scan_logo.png
- |\CTempStorageThe operations exposed by WarpScan require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse - \CWarpReduceThe WarpReduce class provides collective methods for computing a parallel reduction of items partitioned across a CUDA thread warp.

+ |\CTempStorageThe operations exposed by WarpScan require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse + \CWarpReduceThe WarpReduce class provides collective methods for computing a parallel reduction of items partitioned across a CUDA thread warp.

warp_reduce_logo.png
-  \CTempStorageThe operations exposed by WarpReduce require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse +  \CTempStorageThe operations exposed by WarpReduce require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
-Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
diff --git a/docs/html/arg__index__input__iterator_8cuh.html b/docs/html/arg__index__input__iterator_8cuh.html index 74e1b88e89..2f1540b18c 100644 --- a/docs/html/arg__index__input__iterator_8cuh.html +++ b/docs/html/arg__index__input__iterator_8cuh.html @@ -117,7 +117,7 @@

Classes

class  cub::ArgIndexInputIterator< InputIteratorT, OffsetT > - A random-access input wrapper for pairing dereferenced values with their corresponding indices (forming KeyValuePair tuples). More...
+ A random-access input wrapper for pairing dereferenced values with their corresponding indices (forming KeyValuePair tuples). More...
  - - - - - - - - - - - - - - - - - + + + + + + + + + + + + + + + + + +

@@ -134,7 +134,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/arg__index__input__iterator_8cuh_source.html b/docs/html/arg__index__input__iterator_8cuh_source.html index a2c6a9f367..599e17c1c6 100644 --- a/docs/html/arg__index__input__iterator_8cuh_source.html +++ b/docs/html/arg__index__input__iterator_8cuh_source.html @@ -167,9 +167,9 @@
121  // Required iterator traits
122  typedef ArgIndexInputIterator self_type;
123  typedef OffsetT difference_type;
-
124  typedef KeyValuePair<difference_type, T> value_type;
-
125  typedef value_type* pointer;
-
126  typedef value_type reference;
+
124  typedef KeyValuePair<difference_type, T> value_type;
+
125  typedef value_type* pointer;
+
126  typedef value_type reference;
127 
128 #if (THRUST_VERSION >= 100700)
129  // Use Thrust's iterator categories so we can use these iterators in Thrust 1.7 (or newer) methods
@@ -177,7 +177,7 @@
131  thrust::any_system_tag,
132  thrust::random_access_traversal_tag,
133  value_type,
-
134  reference
+
134  reference
135  >::type iterator_category;
136 #else
137  typedef std::random_access_iterator_tag iterator_category;
@@ -211,11 +211,11 @@
168  return *this;
169  }
170 
-
172  __host__ __device__ __forceinline__ reference operator*() const
+
172  __host__ __device__ __forceinline__ reference operator*() const
173  {
-
174  value_type retval;
-
175  retval.value = itr[offset];
-
176  retval.key = offset;
+
174  value_type retval;
+
175  retval.value = itr[offset];
+
176  retval.key = offset;
177  return retval;
178  }
179 
@@ -253,13 +253,13 @@
216  }
217 
219  template <typename Distance>
-
220  __host__ __device__ __forceinline__ reference operator[](Distance n) const
+
220  __host__ __device__ __forceinline__ reference operator[](Distance n) const
221  {
222  self_type offset = (*this) + n;
223  return *offset;
224  }
225 
-
227  __host__ __device__ __forceinline__ pointer operator->()
+
227  __host__ __device__ __forceinline__ pointer operator->()
228  {
229  return &(*(*this));
230  }
@@ -274,22 +274,28 @@
241  return ((itr != rhs.itr) || (offset != rhs.offset));
242  }
243 
-
245  friend std::ostream& operator<<(std::ostream& os, const self_type& itr)
+
245  __host__ __device__ __forceinline__ void normalize()
246  {
-
247  return os;
-
248  }
-
249 };
+
247  itr += offset;
+
248  offset = 0;
+
249  }
250 
-
251 
-
252  // end group UtilIterator
-
254 
-
255 } // CUB namespace
-
256 CUB_NS_POSTFIX // Optional outer namespace(s)
+
252  friend std::ostream& operator<<(std::ostream& os, const self_type& itr)
+
253  {
+
254  return os;
+
255  }
+
256 };
+
257 
+
258 
+
259  // end group UtilIterator
+
261 
+
262 } // CUB namespace
+
263 CUB_NS_POSTFIX // Optional outer namespace(s)
-Generated on Tue Nov 10 2015 13:58:02 for CUB by   +Generated on Mon Dec 14 2015 13:11:13 for CUB by   doxygen 1.8.4
diff --git a/docs/html/block__discontinuity_8cuh.html b/docs/html/block__discontinuity_8cuh.html index 412f338a0a..924df6faa7 100644 --- a/docs/html/block__discontinuity_8cuh.html +++ b/docs/html/block__discontinuity_8cuh.html @@ -138,7 +138,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/block__discontinuity_8cuh_source.html b/docs/html/block__discontinuity_8cuh_source.html index 8f94d667e1..b4ee4efef1 100644 --- a/docs/html/block__discontinuity_8cuh_source.html +++ b/docs/html/block__discontinuity_8cuh_source.html @@ -286,7 +286,7 @@
256 
257 public:
258 
-
260  struct TempStorage : Uninitialized<_TempStorage> {};
+
260  struct TempStorage : Uninitialized<_TempStorage> {};
261 
262 
263  /******************************************************************/
@@ -679,7 +679,7 @@
-Generated on Tue Nov 10 2015 13:58:02 for CUB by   +Generated on Mon Dec 14 2015 13:11:13 for CUB by   doxygen 1.8.4
diff --git a/docs/html/block__exchange_8cuh.html b/docs/html/block__exchange_8cuh.html index 3f6ca145b8..80d3eae731 100644 --- a/docs/html/block__exchange_8cuh.html +++ b/docs/html/block__exchange_8cuh.html @@ -140,7 +140,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/block__exchange_8cuh_source.html.REMOVED.git-id b/docs/html/block__exchange_8cuh_source.html.REMOVED.git-id index b262a1b85f..97ec72db0e 100644 --- a/docs/html/block__exchange_8cuh_source.html.REMOVED.git-id +++ b/docs/html/block__exchange_8cuh_source.html.REMOVED.git-id @@ -1 +1 @@ -7b242fc630635a9576fbb98ed7104c98e20cf7c1 \ No newline at end of file +c806ddfb4031fade9e67fee00d92ec496a8f9b90 \ No newline at end of file diff --git a/docs/html/block__histogram_8cuh.html b/docs/html/block__histogram_8cuh.html index d4fbb5681d..ec4df08145 100644 --- a/docs/html/block__histogram_8cuh.html +++ b/docs/html/block__histogram_8cuh.html @@ -149,7 +149,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/block__histogram_8cuh_source.html b/docs/html/block__histogram_8cuh_source.html index 4408af40fb..8b5b248fe9 100644 --- a/docs/html/block__histogram_8cuh_source.html +++ b/docs/html/block__histogram_8cuh_source.html @@ -218,7 +218,7 @@
214 
215 public:
216 
-
218  struct TempStorage : Uninitialized<_TempStorage> {};
+
218  struct TempStorage : Uninitialized<_TempStorage> {};
219 
220 
221  /******************************************************************/
@@ -295,7 +295,7 @@
-Generated on Tue Nov 10 2015 13:58:02 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/block__load_8cuh.html b/docs/html/block__load_8cuh.html index 4182ba48ae..b80ab9674e 100644 --- a/docs/html/block__load_8cuh.html +++ b/docs/html/block__load_8cuh.html @@ -211,7 +211,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/block__load_8cuh_source.html.REMOVED.git-id b/docs/html/block__load_8cuh_source.html.REMOVED.git-id index a92f73336b..56a9ba86b2 100644 --- a/docs/html/block__load_8cuh_source.html.REMOVED.git-id +++ b/docs/html/block__load_8cuh_source.html.REMOVED.git-id @@ -1 +1 @@ -f16e3469ee2feac63eb12d1fd509886f96672f72 \ No newline at end of file +4f8a06e07d757fc2158c92b1d7f4a7bb7aa92332 \ No newline at end of file diff --git a/docs/html/block__radix__sort_8cuh.html b/docs/html/block__radix__sort_8cuh.html index b3fe1d860b..2d0a5cddef 100644 --- a/docs/html/block__radix__sort_8cuh.html +++ b/docs/html/block__radix__sort_8cuh.html @@ -141,7 +141,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/block__radix__sort_8cuh_source.html b/docs/html/block__radix__sort_8cuh_source.html index 3db5e3e2eb..fef42dcec2 100644 --- a/docs/html/block__radix__sort_8cuh_source.html +++ b/docs/html/block__radix__sort_8cuh_source.html @@ -172,7 +172,7 @@
146  };
147 
148  // KeyT traits and unsigned bits type
-
149  typedef NumericTraits<KeyT> KeyTraits;
+
149  typedef Traits<KeyT> KeyTraits;
150  typedef typename KeyTraits::UnsignedBits UnsignedBits;
151 
153  typedef BlockRadixRank<
@@ -238,7 +238,7 @@
221  int (&ranks)[ITEMS_PER_THREAD],
222  int begin_bit,
223  int pass_bits,
-
224  Int2Type<false> is_descending)
+
224  Int2Type<false> is_descending)
225  {
226  AscendingBlockRadixRank(temp_storage.asending_ranking_storage).RankKeys(
227  unsigned_keys,
@@ -252,7 +252,7 @@
236  int (&ranks)[ITEMS_PER_THREAD],
237  int begin_bit,
238  int pass_bits,
-
239  Int2Type<true> is_descending)
+
239  Int2Type<true> is_descending)
240  {
241  DescendingBlockRadixRank(temp_storage.descending_ranking_storage).RankKeys(
242  unsigned_keys,
@@ -264,8 +264,8 @@
249  __device__ __forceinline__ void ExchangeValues(
250  ValueT (&values)[ITEMS_PER_THREAD],
251  int (&ranks)[ITEMS_PER_THREAD],
-
252  Int2Type<false> is_keys_only,
-
253  Int2Type<true> is_blocked)
+
252  Int2Type<false> is_keys_only,
+
253  Int2Type<true> is_blocked)
254  {
255  __syncthreads();
256 
@@ -276,8 +276,8 @@
262  __device__ __forceinline__ void ExchangeValues(
263  ValueT (&values)[ITEMS_PER_THREAD],
264  int (&ranks)[ITEMS_PER_THREAD],
-
265  Int2Type<false> is_keys_only,
-
266  Int2Type<false> is_blocked)
+
265  Int2Type<false> is_keys_only,
+
266  Int2Type<false> is_blocked)
267  {
268  __syncthreads();
269 
@@ -289,8 +289,8 @@
276  __device__ __forceinline__ void ExchangeValues(
277  ValueT (&values)[ITEMS_PER_THREAD],
278  int (&ranks)[ITEMS_PER_THREAD],
-
279  Int2Type<true> is_keys_only,
-
280  Int2Type<IS_BLOCKED> is_blocked)
+
279  Int2Type<true> is_keys_only,
+
280  Int2Type<IS_BLOCKED> is_blocked)
281  {}
282 
284  template <int DESCENDING, int KEYS_ONLY>
@@ -299,8 +299,8 @@
287  ValueT (&values)[ITEMS_PER_THREAD],
288  int begin_bit,
289  int end_bit,
-
290  Int2Type<DESCENDING> is_descending,
-
291  Int2Type<KEYS_ONLY> is_keys_only)
+
290  Int2Type<DESCENDING> is_descending,
+
291  Int2Type<KEYS_ONLY> is_keys_only)
292  {
293  UnsignedBits (&unsigned_keys)[ITEMS_PER_THREAD] =
294  reinterpret_cast<UnsignedBits (&)[ITEMS_PER_THREAD]>(keys);
@@ -328,7 +328,7 @@
316  BlockExchangeKeys(temp_storage.exchange_keys).ScatterToBlocked(keys, ranks);
317 
318  // Exchange values through shared memory in blocked arrangement
-
319  ExchangeValues(values, ranks, is_keys_only, Int2Type<true>());
+
319  ExchangeValues(values, ranks, is_keys_only, Int2Type<true>());
320 
321  // Quit if done
322  if (begin_bit >= end_bit) break;
@@ -354,8 +354,8 @@
343  ValueT (&values)[ITEMS_PER_THREAD],
344  int begin_bit,
345  int end_bit,
-
346  Int2Type<DESCENDING> is_descending,
-
347  Int2Type<KEYS_ONLY> is_keys_only)
+
346  Int2Type<DESCENDING> is_descending,
+
347  Int2Type<KEYS_ONLY> is_keys_only)
348  {
349  UnsignedBits (&unsigned_keys)[ITEMS_PER_THREAD] =
350  reinterpret_cast<UnsignedBits (&)[ITEMS_PER_THREAD]>(keys);
@@ -386,7 +386,7 @@
375  BlockExchangeKeys(temp_storage.exchange_keys).ScatterToStriped(keys, ranks);
376 
377  // Last pass exchanges through shared memory in striped arrangement
-
378  ExchangeValues(values, ranks, is_keys_only, Int2Type<false>());
+
378  ExchangeValues(values, ranks, is_keys_only, Int2Type<false>());
379 
380  // Quit
381  break;
@@ -396,7 +396,7 @@
385  BlockExchangeKeys(temp_storage.exchange_keys).ScatterToBlocked(keys, ranks);
386 
387  // Exchange values through shared memory in blocked arrangement
-
388  ExchangeValues(values, ranks, is_keys_only, Int2Type<true>());
+
388  ExchangeValues(values, ranks, is_keys_only, Int2Type<true>());
389 
390  __syncthreads();
391  }
@@ -411,7 +411,7 @@
400 
401 #endif // DOXYGEN_SHOULD_SKIP_THIS
402 
-
404  struct TempStorage : Uninitialized<_TempStorage> {};
+
404  struct TempStorage : Uninitialized<_TempStorage> {};
405 
406 
407  /******************************************************************/
@@ -438,9 +438,9 @@
478  int begin_bit = 0,
479  int end_bit = sizeof(KeyT) * 8)
480  {
-
481  NullType values[ITEMS_PER_THREAD];
+
481  NullType values[ITEMS_PER_THREAD];
482 
-
483  SortBlocked(keys, values, begin_bit, end_bit, Int2Type<false>(), Int2Type<KEYS_ONLY>());
+
483  SortBlocked(keys, values, begin_bit, end_bit, Int2Type<false>(), Int2Type<KEYS_ONLY>());
484  }
485 
486 
@@ -450,7 +450,7 @@
534  int begin_bit = 0,
535  int end_bit = sizeof(KeyT) * 8)
536  {
-
537  SortBlocked(keys, values, begin_bit, end_bit, Int2Type<false>(), Int2Type<KEYS_ONLY>());
+
537  SortBlocked(keys, values, begin_bit, end_bit, Int2Type<false>(), Int2Type<KEYS_ONLY>());
538  }
539 
577  __device__ __forceinline__ void SortDescending(
@@ -458,9 +458,9 @@
579  int begin_bit = 0,
580  int end_bit = sizeof(KeyT) * 8)
581  {
-
582  NullType values[ITEMS_PER_THREAD];
+
582  NullType values[ITEMS_PER_THREAD];
583 
-
584  SortBlocked(keys, values, begin_bit, end_bit, Int2Type<true>(), Int2Type<KEYS_ONLY>());
+
584  SortBlocked(keys, values, begin_bit, end_bit, Int2Type<true>(), Int2Type<KEYS_ONLY>());
585  }
586 
587 
@@ -470,7 +470,7 @@
635  int begin_bit = 0,
636  int end_bit = sizeof(KeyT) * 8)
637  {
-
638  SortBlocked(keys, values, begin_bit, end_bit, Int2Type<true>(), Int2Type<KEYS_ONLY>());
+
638  SortBlocked(keys, values, begin_bit, end_bit, Int2Type<true>(), Int2Type<KEYS_ONLY>());
639  }
640 
641 
@@ -482,9 +482,9 @@
689  int begin_bit = 0,
690  int end_bit = sizeof(KeyT) * 8)
691  {
-
692  NullType values[ITEMS_PER_THREAD];
+
692  NullType values[ITEMS_PER_THREAD];
693 
-
694  SortBlockedToStriped(keys, values, begin_bit, end_bit, Int2Type<false>(), Int2Type<KEYS_ONLY>());
+
694  SortBlockedToStriped(keys, values, begin_bit, end_bit, Int2Type<false>(), Int2Type<KEYS_ONLY>());
695  }
696 
697 
@@ -494,7 +494,7 @@
745  int begin_bit = 0,
746  int end_bit = sizeof(KeyT) * 8)
747  {
-
748  SortBlockedToStriped(keys, values, begin_bit, end_bit, Int2Type<false>(), Int2Type<KEYS_ONLY>());
+
748  SortBlockedToStriped(keys, values, begin_bit, end_bit, Int2Type<false>(), Int2Type<KEYS_ONLY>());
749  }
750 
751 
@@ -503,9 +503,9 @@
792  int begin_bit = 0,
793  int end_bit = sizeof(KeyT) * 8)
794  {
-
795  NullType values[ITEMS_PER_THREAD];
+
795  NullType values[ITEMS_PER_THREAD];
796 
-
797  SortBlockedToStriped(keys, values, begin_bit, end_bit, Int2Type<true>(), Int2Type<KEYS_ONLY>());
+
797  SortBlockedToStriped(keys, values, begin_bit, end_bit, Int2Type<true>(), Int2Type<KEYS_ONLY>());
798  }
799 
800 
@@ -515,7 +515,7 @@
848  int begin_bit = 0,
849  int end_bit = sizeof(KeyT) * 8)
850  {
-
851  SortBlockedToStriped(keys, values, begin_bit, end_bit, Int2Type<true>(), Int2Type<KEYS_ONLY>());
+
851  SortBlockedToStriped(keys, values, begin_bit, end_bit, Int2Type<true>(), Int2Type<KEYS_ONLY>());
852  }
853 
854 
@@ -529,7 +529,7 @@
-Generated on Tue Nov 10 2015 13:58:02 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/block__reduce_8cuh.html b/docs/html/block__reduce_8cuh.html index cb6db45a82..00bf842208 100644 --- a/docs/html/block__reduce_8cuh.html +++ b/docs/html/block__reduce_8cuh.html @@ -151,7 +151,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/block__reduce_8cuh_source.html b/docs/html/block__reduce_8cuh_source.html index 9405d5d6b2..fd466c475d 100644 --- a/docs/html/block__reduce_8cuh_source.html +++ b/docs/html/block__reduce_8cuh_source.html @@ -221,7 +221,7 @@
273 
274 public:
275 
-
277  struct TempStorage : Uninitialized<_TempStorage> {};
+
277  struct TempStorage : Uninitialized<_TempStorage> {};
278 
279 
280  /******************************************************************/
@@ -328,7 +328,7 @@
-Generated on Tue Nov 10 2015 13:58:02 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/block__scan_8cuh.html b/docs/html/block__scan_8cuh.html index cfccdd27ca..561a96141a 100644 --- a/docs/html/block__scan_8cuh.html +++ b/docs/html/block__scan_8cuh.html @@ -151,7 +151,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/block__scan_8cuh_source.html.REMOVED.git-id b/docs/html/block__scan_8cuh_source.html.REMOVED.git-id index 3a8cc12acf..3716f488da 100644 --- a/docs/html/block__scan_8cuh_source.html.REMOVED.git-id +++ b/docs/html/block__scan_8cuh_source.html.REMOVED.git-id @@ -1 +1 @@ -5448840077c040640722b0624858306969fe6a45 \ No newline at end of file +753cec633a135d727295bfdeeac9c4426a237ab5 \ No newline at end of file diff --git a/docs/html/block__store_8cuh.html b/docs/html/block__store_8cuh.html index 40d87fa334..8d261660a5 100644 --- a/docs/html/block__store_8cuh.html +++ b/docs/html/block__store_8cuh.html @@ -199,7 +199,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/block__store_8cuh_source.html b/docs/html/block__store_8cuh_source.html index 78eb55a778..10223f9ebe 100644 --- a/docs/html/block__store_8cuh_source.html +++ b/docs/html/block__store_8cuh_source.html @@ -207,7 +207,7 @@
155  };
156 
157  // Vector type
-
158  typedef typename CubVector<T, VEC_SIZE>::Type Vector;
+
158  typedef typename CubVector<T, VEC_SIZE>::Type Vector;
159 
160  // Alias global pointer
161  Vector *block_ptr_vectors = reinterpret_cast<Vector*>(const_cast<T*>(block_ptr));
@@ -384,7 +384,7 @@
533  template <int DUMMY>
534  struct StoreInternal<BLOCK_STORE_DIRECT, DUMMY>
535  {
-
537  typedef NullType TempStorage;
+
537  typedef NullType TempStorage;
538 
540  int linear_tid;
541 
@@ -415,7 +415,7 @@
572  template <int DUMMY>
573  struct StoreInternal<BLOCK_STORE_VECTORIZE, DUMMY>
574  {
-
576  typedef NullType TempStorage;
+
576  typedef NullType TempStorage;
577 
579  int linear_tid;
580 
@@ -459,7 +459,7 @@
625 
627  typedef typename BlockExchange::TempStorage _TempStorage;
628 
-
630  struct TempStorage : Uninitialized<_TempStorage> {};
+
630  struct TempStorage : Uninitialized<_TempStorage> {};
631 
633  _TempStorage &temp_storage;
634 
@@ -508,7 +508,7 @@
684 
686  typedef typename BlockExchange::TempStorage _TempStorage;
687 
-
689  struct TempStorage : Uninitialized<_TempStorage> {};
+
689  struct TempStorage : Uninitialized<_TempStorage> {};
690 
692  _TempStorage &temp_storage;
693 
@@ -557,7 +557,7 @@
743 
745  typedef typename BlockExchange::TempStorage _TempStorage;
746 
-
748  struct TempStorage : Uninitialized<_TempStorage> {};
+
748  struct TempStorage : Uninitialized<_TempStorage> {};
749 
751  _TempStorage &temp_storage;
752 
@@ -621,7 +621,7 @@
819 public:
820 
821 
-
823  struct TempStorage : Uninitialized<_TempStorage> {};
+
823  struct TempStorage : Uninitialized<_TempStorage> {};
824 
825 
826  /******************************************************************/
@@ -668,7 +668,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/cache__modified__input__iterator_8cuh.html b/docs/html/cache__modified__input__iterator_8cuh.html index c06ec30f7c..b73c2ea373 100644 --- a/docs/html/cache__modified__input__iterator_8cuh.html +++ b/docs/html/cache__modified__input__iterator_8cuh.html @@ -133,7 +133,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/cache__modified__input__iterator_8cuh_source.html b/docs/html/cache__modified__input__iterator_8cuh_source.html index f682bcedeb..68a9808d19 100644 --- a/docs/html/cache__modified__input__iterator_8cuh_source.html +++ b/docs/html/cache__modified__input__iterator_8cuh_source.html @@ -278,7 +278,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/cache__modified__output__iterator_8cuh.html b/docs/html/cache__modified__output__iterator_8cuh.html index cef32060c6..2415b9b43f 100644 --- a/docs/html/cache__modified__output__iterator_8cuh.html +++ b/docs/html/cache__modified__output__iterator_8cuh.html @@ -133,7 +133,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/cache__modified__output__iterator_8cuh_source.html b/docs/html/cache__modified__output__iterator_8cuh_source.html index 6d1dd74225..028b1c710b 100644 --- a/docs/html/cache__modified__output__iterator_8cuh_source.html +++ b/docs/html/cache__modified__output__iterator_8cuh_source.html @@ -289,7 +289,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_arg_index_input_iterator-members.html b/docs/html/classcub_1_1_arg_index_input_iterator-members.html index f335777631..871c4cc828 100644 --- a/docs/html/classcub_1_1_arg_index_input_iterator-members.html +++ b/docs/html/classcub_1_1_arg_index_input_iterator-members.html @@ -107,28 +107,29 @@

ArgIndexInputIterator(InputIteratorT itr, difference_type offset=0)cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
difference_type typedefcub::ArgIndexInputIterator< InputIteratorT, OffsetT >
iterator_category typedefcub::ArgIndexInputIterator< InputIteratorT, OffsetT >
operator!=(const self_type &rhs)cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator*() const cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator+(Distance n) const cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator++(int)cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator++()cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator+=(Distance n)cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator-(Distance n) const cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator-(self_type other) const cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator-=(Distance n)cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator->()cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator<<(std::ostream &os, const self_type &itr)cub::ArgIndexInputIterator< InputIteratorT, OffsetT >friend
operator==(const self_type &rhs)cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator[](Distance n) const cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
pointer typedefcub::ArgIndexInputIterator< InputIteratorT, OffsetT >
reference typedefcub::ArgIndexInputIterator< InputIteratorT, OffsetT >
self_type typedefcub::ArgIndexInputIterator< InputIteratorT, OffsetT >
value_type typedefcub::ArgIndexInputIterator< InputIteratorT, OffsetT >
normalize()cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator!=(const self_type &rhs)cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator*() const cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator+(Distance n) const cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator++(int)cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator++()cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator+=(Distance n)cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator-(Distance n) const cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator-(self_type other) const cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator-=(Distance n)cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator->()cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator<<(std::ostream &os, const self_type &itr)cub::ArgIndexInputIterator< InputIteratorT, OffsetT >friend
operator==(const self_type &rhs)cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
operator[](Distance n) const cub::ArgIndexInputIterator< InputIteratorT, OffsetT >inline
pointer typedefcub::ArgIndexInputIterator< InputIteratorT, OffsetT >
reference typedefcub::ArgIndexInputIterator< InputIteratorT, OffsetT >
self_type typedefcub::ArgIndexInputIterator< InputIteratorT, OffsetT >
value_type typedefcub::ArgIndexInputIterator< InputIteratorT, OffsetT >
-Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_arg_index_input_iterator.html b/docs/html/classcub_1_1_arg_index_input_iterator.html index 7d19899df1..d555ec036d 100644 --- a/docs/html/classcub_1_1_arg_index_input_iterator.html +++ b/docs/html/classcub_1_1_arg_index_input_iterator.html @@ -112,9 +112,9 @@ typename OffsetT = ptrdiff_t>
class cub::ArgIndexInputIterator< InputIteratorT, OffsetT > -

A random-access input wrapper for pairing dereferenced values with their corresponding indices (forming KeyValuePair tuples).

+

A random-access input wrapper for pairing dereferenced values with their corresponding indices (forming KeyValuePair tuples).

Overview
    -
  • ArgIndexInputIteratorTwraps a random access input iterator itr of type InputIteratorT. Dereferencing an ArgIndexInputIteratorTat offset i produces a KeyValuePair value whose key field is i and whose value field is itr[i].
    • +
  • ArgIndexInputIteratorTwraps a random access input iterator itr of type InputIteratorT. Dereferencing an ArgIndexInputIteratorTat offset i produces a KeyValuePair value whose key field is i and whose value field is itr[i].
  • Can be used with any data type.
  • Can be constructed, manipulated, and exchanged within and between host and device functions. Wrapped host memory can only be dereferenced on the host, and wrapped device memory can only be dereferenced on the device.
  • Compatible with Thrust API v1.7 or newer.
    • @@ -130,17 +130,17 @@
    cub::ArgIndexInputIterator<double*> itr(d_in);
    // Within device code:
    -
    typedef typename cub::ArgIndexInputIterator<double*>::value_type Tuple;
    -
    Tuple item_offset_pair.key = *itr;
    +
    typedef typename cub::ArgIndexInputIterator<double*>::value_type Tuple;
    +
    Tuple item_offset_pair.key = *itr;
    printf("%f @ %d\n",
    -
    item_offset_pair.value,
    -
    item_offset_pair.key); // 8.0 @ 0
    +
    item_offset_pair.value,
    +
    item_offset_pair.key); // 8.0 @ 0
    itr = itr + 6;
    item_offset_pair.key = *itr;
    printf("%f @ %d\n",
    -
    item_offset_pair.value,
    -
    item_offset_pair.key); // 9.0 @ 6
    +
    item_offset_pair.value,
    +
    item_offset_pair.key); // 9.0 @ 6
Template Parameters
@@ -163,7 +163,7 @@ @@ -252,6 +252,11 @@ __forceinline__ bool  + + +
 Type to express the result of subtracting one iterator from another.
 
-typedef KeyValuePair
+typedef KeyValuePair
< difference_type, T > 		value_type
 The type of the element the iterator can point to.
 
operator!= (const self_type &rhs)
 Not equal to.
 
+__host__ __device__
+__forceinline__ void 		normalize ()
 Normalize.
 
@@ -315,7 +320,7 @@

Constructor & Destructor Documentation

-Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_block_discontinuity-members.html b/docs/html/classcub_1_1_block_discontinuity-members.html index 6b2fe38bda..e371714be5 100644 --- a/docs/html/classcub_1_1_block_discontinuity-members.html +++ b/docs/html/classcub_1_1_block_discontinuity-members.html @@ -118,7 +118,7 @@
-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_block_discontinuity.html b/docs/html/classcub_1_1_block_discontinuity.html index 958d9f9416..1f75cb43af 100644 --- a/docs/html/classcub_1_1_block_discontinuity.html +++ b/docs/html/classcub_1_1_block_discontinuity.html @@ -1116,7 +1116,7 @@

Member Function Documentation

-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_block_exchange-members.html b/docs/html/classcub_1_1_block_exchange-members.html index 6b921bcf36..cc58407824 100644 --- a/docs/html/classcub_1_1_block_exchange-members.html +++ b/docs/html/classcub_1_1_block_exchange-members.html @@ -118,7 +118,7 @@
-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_block_exchange.html b/docs/html/classcub_1_1_block_exchange.html index c46c77b5f7..9fb3e824de 100644 --- a/docs/html/classcub_1_1_block_exchange.html +++ b/docs/html/classcub_1_1_block_exchange.html @@ -749,7 +749,7 @@

Member Function Documentation

-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_block_histogram-members.html b/docs/html/classcub_1_1_block_histogram-members.html index 94b024c19e..2b21e183aa 100644 --- a/docs/html/classcub_1_1_block_histogram-members.html +++ b/docs/html/classcub_1_1_block_histogram-members.html @@ -113,7 +113,7 @@
-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_block_histogram.html b/docs/html/classcub_1_1_block_histogram.html index 87d9f37154..ad5fe798b8 100644 --- a/docs/html/classcub_1_1_block_histogram.html +++ b/docs/html/classcub_1_1_block_histogram.html @@ -479,7 +479,7 @@

Member Function Documentation

-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_block_load-members.html b/docs/html/classcub_1_1_block_load-members.html index f5ea4a4329..fe5d952392 100644 --- a/docs/html/classcub_1_1_block_load-members.html +++ b/docs/html/classcub_1_1_block_load-members.html @@ -113,7 +113,7 @@
-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_block_load.html b/docs/html/classcub_1_1_block_load.html index 1c6c94440a..e64f00570e 100644 --- a/docs/html/classcub_1_1_block_load.html +++ b/docs/html/classcub_1_1_block_load.html @@ -165,7 +165,7 @@
Suppose the input d_data is 0, 1, 2, 3, 4, 5, .... The set of thread_data across the block of threads in those threads will be { [0,1,2,3], [4,5,6,7], ..., [508,509,510,511] }.
Examples:
example_block_radix_sort.cu, and example_block_scan.cu.
-

Definition at line 659 of file block_load.cuh.

+

Definition at line 674 of file block_load.cuh.

Friends

@@ -227,7 +227,7 @@

Constructor & Destructor Documentation

-

Definition at line 1073 of file block_load.cuh.

+

Definition at line 1088 of file block_load.cuh.

@@ -296,7 +296,7 @@

Member Function Documentation

-

Definition at line 1126 of file block_load.cuh.

+

Definition at line 1141 of file block_load.cuh.

@@ -371,7 +371,7 @@

Member Function Documentation

-

Definition at line 1171 of file block_load.cuh.

+

Definition at line 1186 of file block_load.cuh.

@@ -453,7 +453,7 @@

Member Function Documentation

-

Definition at line 1218 of file block_load.cuh.

+

Definition at line 1233 of file block_load.cuh.

@@ -464,7 +464,7 @@

Member Function Documentation

-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_block_radix_sort-members.html b/docs/html/classcub_1_1_block_radix_sort-members.html index 0cf302a181..d694c8a9de 100644 --- a/docs/html/classcub_1_1_block_radix_sort-members.html +++ b/docs/html/classcub_1_1_block_radix_sort-members.html @@ -118,7 +118,7 @@
-Generated on Tue Nov 10 2015 13:58:06 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_block_radix_sort.html b/docs/html/classcub_1_1_block_radix_sort.html index 159f1b06ad..c7d1c408d0 100644 --- a/docs/html/classcub_1_1_block_radix_sort.html +++ b/docs/html/classcub_1_1_block_radix_sort.html @@ -129,7 +129,7 @@
- + @@ -924,7 +924,7 @@

Member Function Documentation

-Generated on Tue Nov 10 2015 13:58:06 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_block_reduce-members.html b/docs/html/classcub_1_1_block_reduce-members.html index de5ec20b00..0362e1f917 100644 --- a/docs/html/classcub_1_1_block_reduce-members.html +++ b/docs/html/classcub_1_1_block_reduce-members.html @@ -116,7 +116,7 @@
-Generated on Tue Nov 10 2015 13:58:06 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_block_reduce.html b/docs/html/classcub_1_1_block_reduce.html index 0602c4aec0..2b42dfbb0b 100644 --- a/docs/html/classcub_1_1_block_reduce.html +++ b/docs/html/classcub_1_1_block_reduce.html @@ -319,14 +319,14 @@

Member Function Documentation

Template Parameters

Classes

KeyTKeyT type
BLOCK_DIM_XThe thread block length in threads along the X dimension
ITEMS_PER_THREADThe number of items per thread
ValueT[optional] ValueT type (default: cub::NullType, which indicates a keys-only sort)
ValueT[optional] ValueT type (default: cub::NullType, which indicates a keys-only sort)
RADIX_BITS[optional] The number of radix bits per digit place (default: 4 bits)
MEMOIZE_OUTER_SCAN[optional] Whether or not to buffer outer raking scan partials to incur fewer shared memory reads at the expense of higher register pressure (default: true for architectures SM35 and newer, false otherwise).
INNER_SCAN_ALGORITHM[optional] The cub::BlockScanAlgorithm algorithm to use (default: cub::BLOCK_SCAN_WARP_SCANS)
- +
ReductionOp[inferred] Binary reduction functor (e.g., an instance of cub::Sum, cub::Min, cub::Max, etc.) type having member T operator()(const T &a, const T &b)
ReductionOp[inferred] Binary reduction functor type having member T operator()(const T &a, const T &b)
Parameters
- +
[in]inputCalling thread's input
[in]reduction_opBinary reduction functor (e.g., an instance of cub::Sum, cub::Min, cub::Max, etc.)
[in]reduction_opBinary reduction functor
@@ -399,14 +399,14 @@

Member Function Documentation

Template Parameters
- +
ITEMS_PER_THREAD[inferred] The number of consecutive items partitioned onto each thread.
ReductionOp[inferred] Binary reduction functor (e.g., an instance of cub::Sum, cub::Min, cub::Max, etc.) type having member T operator()(const T &a, const T &b)
ReductionOp[inferred] Binary reduction functor type having member T operator()(const T &a, const T &b)
Parameters
- +
[in]inputsCalling thread's input segment
[in]reduction_opBinary reduction functor (e.g., an instance of cub::Sum, cub::Min, cub::Max, etc.)
[in]reduction_opBinary reduction functor
@@ -484,14 +484,14 @@

Member Function Documentation

Template Parameters
- +
ReductionOp[inferred] Binary reduction functor (e.g., an instance of cub::Sum, cub::Min, cub::Max, etc.) type having member T operator()(const T &a, const T &b)
ReductionOp[inferred] Binary reduction functor type having member T operator()(const T &a, const T &b)
Parameters
- +
[in]inputCalling thread's input
[in]reduction_opBinary reduction functor (e.g., an instance of cub::Sum, cub::Min, cub::Max, etc.)
[in]reduction_opBinary reduction functor
[in]num_validNumber of threads containing valid elements (may be less than BLOCK_THREADS)
@@ -708,7 +708,7 @@

Member Function Documentation

-Generated on Tue Nov 10 2015 13:58:06 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_block_scan-members.html b/docs/html/classcub_1_1_block_scan-members.html index 3cc5f8b38a..48abbc03b1 100644 --- a/docs/html/classcub_1_1_block_scan-members.html +++ b/docs/html/classcub_1_1_block_scan-members.html @@ -134,7 +134,7 @@
-Generated on Tue Nov 10 2015 13:58:06 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_block_scan.html.REMOVED.git-id b/docs/html/classcub_1_1_block_scan.html.REMOVED.git-id index e04db5ddde..04b4fb437a 100644 --- a/docs/html/classcub_1_1_block_scan.html.REMOVED.git-id +++ b/docs/html/classcub_1_1_block_scan.html.REMOVED.git-id @@ -1 +1 @@ -05c53712eb7c477ef7baefa462ddc7be1ff916bf \ No newline at end of file +7756e9c499e97154bdf8b2fa4343354784e594d1 \ No newline at end of file diff --git a/docs/html/classcub_1_1_block_store-members.html b/docs/html/classcub_1_1_block_store-members.html index 1aa7e39eca..e93c53a5d4 100644 --- a/docs/html/classcub_1_1_block_store-members.html +++ b/docs/html/classcub_1_1_block_store-members.html @@ -112,7 +112,7 @@
-Generated on Tue Nov 10 2015 13:58:06 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_block_store.html b/docs/html/classcub_1_1_block_store.html index 3e651f7bbe..007a7c5a96 100644 --- a/docs/html/classcub_1_1_block_store.html +++ b/docs/html/classcub_1_1_block_store.html @@ -390,7 +390,7 @@

Member Function Documentation

-Generated on Tue Nov 10 2015 13:58:06 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_cache_modified_input_iterator-members.html b/docs/html/classcub_1_1_cache_modified_input_iterator-members.html index 7832d3cc3d..72d655ec03 100644 --- a/docs/html/classcub_1_1_cache_modified_input_iterator-members.html +++ b/docs/html/classcub_1_1_cache_modified_input_iterator-members.html @@ -129,7 +129,7 @@
-Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_cache_modified_input_iterator.html b/docs/html/classcub_1_1_cache_modified_input_iterator.html index 4fc544548d..bddefb784c 100644 --- a/docs/html/classcub_1_1_cache_modified_input_iterator.html +++ b/docs/html/classcub_1_1_cache_modified_input_iterator.html @@ -307,7 +307,7 @@

Constructor & Destructor Documentation

-Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_cache_modified_output_iterator-members.html b/docs/html/classcub_1_1_cache_modified_output_iterator-members.html index dc3b8fd0a6..21d5c0d821 100644 --- a/docs/html/classcub_1_1_cache_modified_output_iterator-members.html +++ b/docs/html/classcub_1_1_cache_modified_output_iterator-members.html @@ -127,7 +127,7 @@
-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_cache_modified_output_iterator.html b/docs/html/classcub_1_1_cache_modified_output_iterator.html index afb413e958..6cf697dbb0 100644 --- a/docs/html/classcub_1_1_cache_modified_output_iterator.html +++ b/docs/html/classcub_1_1_cache_modified_output_iterator.html @@ -300,7 +300,7 @@

Constructor & Destructor Documentation

-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_constant_input_iterator-members.html b/docs/html/classcub_1_1_constant_input_iterator-members.html index cf60a0d63e..e03e9a10bb 100644 --- a/docs/html/classcub_1_1_constant_input_iterator-members.html +++ b/docs/html/classcub_1_1_constant_input_iterator-members.html @@ -128,7 +128,7 @@
-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_constant_input_iterator.html b/docs/html/classcub_1_1_constant_input_iterator.html index b2cf7d8402..3b4839e81d 100644 --- a/docs/html/classcub_1_1_constant_input_iterator.html +++ b/docs/html/classcub_1_1_constant_input_iterator.html @@ -302,7 +302,7 @@

Constructor & Destructor Documentation

-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_counting_input_iterator-members.html b/docs/html/classcub_1_1_counting_input_iterator-members.html index 45bb323ee1..c936431ca3 100644 --- a/docs/html/classcub_1_1_counting_input_iterator-members.html +++ b/docs/html/classcub_1_1_counting_input_iterator-members.html @@ -128,7 +128,7 @@
-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_counting_input_iterator.html b/docs/html/classcub_1_1_counting_input_iterator.html index 1cf4f2e2ef..c5fa03dcd9 100644 --- a/docs/html/classcub_1_1_counting_input_iterator.html +++ b/docs/html/classcub_1_1_counting_input_iterator.html @@ -290,7 +290,7 @@

Constructor & Destructor Documentation

-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_swizzle_scan_op-members.html b/docs/html/classcub_1_1_swizzle_scan_op-members.html index 9b9d8f04eb..f2fb5ac0ff 100644 --- a/docs/html/classcub_1_1_swizzle_scan_op-members.html +++ b/docs/html/classcub_1_1_swizzle_scan_op-members.html @@ -110,7 +110,7 @@
-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_swizzle_scan_op.html b/docs/html/classcub_1_1_swizzle_scan_op.html index 0722a42a08..2f65a1b020 100644 --- a/docs/html/classcub_1_1_swizzle_scan_op.html +++ b/docs/html/classcub_1_1_swizzle_scan_op.html @@ -133,7 +133,7 @@
-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_tex_obj_input_iterator-members.html b/docs/html/classcub_1_1_tex_obj_input_iterator-members.html index ec000a8eaf..1d00600edf 100644 --- a/docs/html/classcub_1_1_tex_obj_input_iterator-members.html +++ b/docs/html/classcub_1_1_tex_obj_input_iterator-members.html @@ -130,7 +130,7 @@
-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_tex_obj_input_iterator.html b/docs/html/classcub_1_1_tex_obj_input_iterator.html index 8b76589fac..eb8fa18164 100644 --- a/docs/html/classcub_1_1_tex_obj_input_iterator.html +++ b/docs/html/classcub_1_1_tex_obj_input_iterator.html @@ -331,7 +331,7 @@

Member Function Documentation

-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_tex_ref_input_iterator-members.html b/docs/html/classcub_1_1_tex_ref_input_iterator-members.html index 939ea07c22..291fd23fea 100644 --- a/docs/html/classcub_1_1_tex_ref_input_iterator-members.html +++ b/docs/html/classcub_1_1_tex_ref_input_iterator-members.html @@ -129,7 +129,7 @@
-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_tex_ref_input_iterator.html b/docs/html/classcub_1_1_tex_ref_input_iterator.html index 2f96d62aa0..276762c641 100644 --- a/docs/html/classcub_1_1_tex_ref_input_iterator.html +++ b/docs/html/classcub_1_1_tex_ref_input_iterator.html @@ -332,7 +332,7 @@

Member Function Documentation

-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_transform_input_iterator-members.html b/docs/html/classcub_1_1_transform_input_iterator-members.html index ce4fbaf0c2..631e07b05e 100644 --- a/docs/html/classcub_1_1_transform_input_iterator-members.html +++ b/docs/html/classcub_1_1_transform_input_iterator-members.html @@ -128,7 +128,7 @@
-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_transform_input_iterator.html b/docs/html/classcub_1_1_transform_input_iterator.html index 7be29dfd33..6ef534526b 100644 --- a/docs/html/classcub_1_1_transform_input_iterator.html +++ b/docs/html/classcub_1_1_transform_input_iterator.html @@ -320,7 +320,7 @@

Constructor & Destructor Documentation

-Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_warp_reduce-members.html b/docs/html/classcub_1_1_warp_reduce-members.html index 53df6aa02e..5adb946e49 100644 --- a/docs/html/classcub_1_1_warp_reduce-members.html +++ b/docs/html/classcub_1_1_warp_reduce-members.html @@ -117,7 +117,7 @@
-Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_warp_reduce.html b/docs/html/classcub_1_1_warp_reduce.html index 23aedc27a2..f262d6ce47 100644 --- a/docs/html/classcub_1_1_warp_reduce.html +++ b/docs/html/classcub_1_1_warp_reduce.html @@ -885,7 +885,7 @@

Member Function Documentation

-Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_warp_scan-members.html b/docs/html/classcub_1_1_warp_scan-members.html index 2b4965be29..c47f350506 100644 --- a/docs/html/classcub_1_1_warp_scan-members.html +++ b/docs/html/classcub_1_1_warp_scan-members.html @@ -123,7 +123,7 @@
-Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
diff --git a/docs/html/classcub_1_1_warp_scan.html.REMOVED.git-id b/docs/html/classcub_1_1_warp_scan.html.REMOVED.git-id index c66485791d..086844ed64 100644 --- a/docs/html/classcub_1_1_warp_scan.html.REMOVED.git-id +++ b/docs/html/classcub_1_1_warp_scan.html.REMOVED.git-id @@ -1 +1 @@ -1e2360acc25ce6d42a73bc80aa33477290d69d8f \ No newline at end of file +a1b3c49e0d85019e2659b4c116b40b52c77c02ff \ No newline at end of file diff --git a/docs/html/classes.html b/docs/html/classes.html index b90a5c7253..5f8a7463f9 100644 --- a/docs/html/classes.html +++ b/docs/html/classes.html @@ -97,54 +97,49 @@
Class Index
-
A | B | C | D | E | I | K | L | M | N | P | R | S | T | U | W
+
A | B | C | D | E | I | L | M | P | R | S | T | W
- - - - + + + + + - - - - - - - + + + + + + + - - + + - - - - - - + + + + +
  A  
-		ConstantInputIterator (cub)   Inequality (cub)   
  R  
-		WarpReduce::TempStorage (cub)   
CountingInputIterator (cub)   InequalityWrapper (cub)   BlockStore::StoreInternal< BLOCK_STORE_TRANSPOSE, DUMMY >::TempStorage (cub)   
ArgIndexInputIterator (cub)   CubVector (cub)   Int2Type (cub)   ReduceByKeyOp (cub)   BlockScan::TempStorage (cub)   
ArgMax (cub)   
  D  
-		IsPointer (cub)   ReduceBySegmentOp (cub)   WarpScan::TempStorage (cub)   
ArgMin (cub)   IsVolatile (cub)   RemoveQualifiers (cub)   BlockHistogram::TempStorage (cub)   
CacheModifiedOutputIterator (cub)   Equals (cub)   
  R  
+		BlockStore::TempStorage (cub)   
CachingDeviceAllocator (cub)   
  I  
+		BlockLoad::LoadInternal< BLOCK_LOAD_TRANSPOSE, DUMMY >::TempStorage (cub)   
ArgIndexInputIterator (cub)   Cast (cub)   ReduceByKeyOp (cub)   BlockRadixSort::TempStorage (cub)   
ArgMax (cub)   ConstantInputIterator (cub)   If (cub)   ReduceBySegmentOp (cub)   BlockScan::TempStorage (cub)   
ArgMin (cub)   CountingInputIterator (cub)   Inequality (cub)   RemoveQualifiers (cub)   BlockStore::StoreInternal< BLOCK_STORE_WARP_TRANSPOSE, DUMMY >::TempStorage (cub)   
  B  
-		DeviceHistogram (cub)   
  K  
-
  S  
-		BlockReduce::TempStorage (cub)   
DevicePartition (cub)   TexObjInputIterator (cub)   
BlockDiscontinuity (cub)   DeviceRadixSort (cub)   KeyValuePair (cub)   Sum (cub)   TexRefInputIterator (cub)   
BlockExchange (cub)   DeviceReduce (cub)   
  L  
-		SwizzleScanOp (cub)   TransformInputIterator (cub)   
BlockHistogram (cub)   DeviceRunLengthEncode (cub)   
  T  
-
  U  
-
BlockLoad (cub)   DeviceScan (cub)   Log2 (cub)   
BlockRadixSort (cub)   DeviceSegmentedRadixSort (cub)   
  M  
-		BlockLoad::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE_TIMESLICED, DUMMY >::TempStorage (cub)   Uninitialized (cub)   
BlockReduce (cub)   DeviceSelect (cub)   BlockStore::StoreInternal< BLOCK_STORE_WARP_TRANSPOSE_TIMESLICED, DUMMY >::TempStorage (cub)   
  W  
+
  D  
+		InequalityWrapper (cub)   
  S  
+		BlockLoad::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE_TIMESLICED, DUMMY >::TempStorage (cub)   
IsPointer (cub)   WarpReduce::TempStorage (cub)   
BlockDiscontinuity (cub)   DeviceHistogram (cub)   IsVolatile (cub)   Sum (cub)   BlockLoad::TempStorage (cub)   
BlockExchange (cub)   DevicePartition (cub)   
  L  
+		SwizzleScanOp (cub)   BlockStore::StoreInternal< BLOCK_STORE_WARP_TRANSPOSE_TIMESLICED, DUMMY >::TempStorage (cub)   
BlockHistogram (cub)   DeviceRadixSort (cub)   
  T  
+		TexObjInputIterator (cub)   
BlockLoad (cub)   DeviceReduce (cub)   Log2 (cub)   TexRefInputIterator (cub)   
BlockRadixSort (cub)   DeviceRunLengthEncode (cub)   
  M  
+		BlockStore::StoreInternal< BLOCK_STORE_TRANSPOSE, DUMMY >::TempStorage (cub)   TransformInputIterator (cub)   
BlockReduce (cub)   DeviceScan (cub)   BlockDiscontinuity::TempStorage (cub)   
  W  
BlockScan (cub)   DeviceSpmv (cub)   Max (cub)   BlockLoad::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE, DUMMY >::TempStorage (cub)   
BlockStore (cub)   DoubleBuffer (cub)   Min (cub)   BlockDiscontinuity::TempStorage (cub)   WarpReduce (cub)   
BlockScan (cub)   DeviceSegmentedRadixSort (cub)   Max (cub)   BlockLoad::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE, DUMMY >::TempStorage (cub)   
BlockStore (cub)   DeviceSegmentedReduce (cub)   Min (cub)   BlockReduce::TempStorage (cub)   WarpReduce (cub)   
  C  
-
  E  
-
  N  
-		BlockExchange::TempStorage (cub)   WarpScan (cub)   
BlockStore::TempStorage (cub)   
CacheModifiedInputIterator (cub)   Equality (cub)   NullType (cub)   BlockLoad::TempStorage (cub)   
CacheModifiedOutputIterator (cub)   Equals (cub)   
  P  
-		BlockRadixSort::TempStorage (cub)   
CachingDeviceAllocator (cub)   
  I  
-		BlockLoad::LoadInternal< BLOCK_LOAD_TRANSPOSE, DUMMY >::TempStorage (cub)   
Cast (cub)   PowerOfTwo (cub)   BlockStore::StoreInternal< BLOCK_STORE_WARP_TRANSPOSE, DUMMY >::TempStorage (cub)   
If (cub)   
DeviceSelect (cub)   
  P  
+		BlockHistogram::TempStorage (cub)   WarpScan (cub)   
DeviceSpmv (cub)   WarpScan::TempStorage (cub)   
CacheModifiedInputIterator (cub)   
  E  
+		PowerOfTwo (cub)   BlockExchange::TempStorage (cub)   
Equality (cub)   
-
A | B | C | D | E | I | K | L | M | N | P | R | S | T | U | W
+
A | B | C | D | E | I | L | M | P | R | S | T | W
-Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
diff --git a/docs/html/constant__input__iterator_8cuh.html b/docs/html/constant__input__iterator_8cuh.html index d9fd868f0f..08714d1cee 100644 --- a/docs/html/constant__input__iterator_8cuh.html +++ b/docs/html/constant__input__iterator_8cuh.html @@ -132,7 +132,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/constant__input__iterator_8cuh_source.html b/docs/html/constant__input__iterator_8cuh_source.html index 63be06f7b9..1c2bfb7d89 100644 --- a/docs/html/constant__input__iterator_8cuh_source.html +++ b/docs/html/constant__input__iterator_8cuh_source.html @@ -282,7 +282,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/counting__input__iterator_8cuh.html b/docs/html/counting__input__iterator_8cuh.html index 2fefba4608..64b2345c03 100644 --- a/docs/html/counting__input__iterator_8cuh.html +++ b/docs/html/counting__input__iterator_8cuh.html @@ -133,7 +133,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/counting__input__iterator_8cuh_source.html b/docs/html/counting__input__iterator_8cuh_source.html index c40ec44b26..0c6a140b11 100644 --- a/docs/html/counting__input__iterator_8cuh_source.html +++ b/docs/html/counting__input__iterator_8cuh_source.html @@ -277,7 +277,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/cub_8cuh.html b/docs/html/cub_8cuh.html index df805b753a..9ea76f4d4c 100644 --- a/docs/html/cub_8cuh.html +++ b/docs/html/cub_8cuh.html @@ -113,6 +113,8 @@ #include "device/device_reduce.cuh"
#include "device/device_run_length_encode.cuh"
#include "device/device_scan.cuh"
+#include "device/device_segmented_radix_sort.cuh"
+#include "device/device_segmented_reduce.cuh"
#include "device/device_select.cuh"
#include "device/device_spmv.cuh"
#include "grid/grid_even_share.cuh"
@@ -136,7 +138,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/cub_8cuh_source.html b/docs/html/cub_8cuh_source.html index bf9ae8fdef..d9c61fb3af 100644 --- a/docs/html/cub_8cuh_source.html +++ b/docs/html/cub_8cuh_source.html @@ -151,53 +151,55 @@
53 #include "device/device_reduce.cuh"
54 #include "device/device_run_length_encode.cuh"
55 #include "device/device_scan.cuh"
-
56 #include "device/device_select.cuh"
-
57 #include "device/device_spmv.cuh"
-
58 
-
59 // Grid
-
60 //#include "grid/grid_barrier.cuh"
-
61 #include "grid/grid_even_share.cuh"
-
62 #include "grid/grid_mapping.cuh"
-
63 #include "grid/grid_queue.cuh"
-
64 
-
65 // Host
-
66 #include "host/spinlock.cuh"
-
67 
-
68 // Thread
-
69 #include "thread/thread_load.cuh"
-
70 #include "thread/thread_operators.cuh"
-
71 #include "thread/thread_reduce.cuh"
-
72 #include "thread/thread_scan.cuh"
-
73 #include "thread/thread_store.cuh"
-
74 
-
75 // Warp
-
76 #include "warp/warp_reduce.cuh"
-
77 #include "warp/warp_scan.cuh"
-
78 
-
79 // Iterator
-
80 #include "iterator/arg_index_input_iterator.cuh"
-
81 #include "iterator/cache_modified_input_iterator.cuh"
-
82 #include "iterator/cache_modified_output_iterator.cuh"
-
83 #include "iterator/constant_input_iterator.cuh"
-
84 #include "iterator/counting_input_iterator.cuh"
-
85 #include "iterator/tex_obj_input_iterator.cuh"
-
86 #include "iterator/tex_ref_input_iterator.cuh"
-
87 #include "iterator/transform_input_iterator.cuh"
-
88 
-
89 // Util
-
90 #include "util_allocator.cuh"
-
91 #include "util_arch.cuh"
-
92 #include "util_debug.cuh"
-
93 #include "util_device.cuh"
-
94 #include "util_macro.cuh"
-
95 #include "util_ptx.cuh"
-
96 #include "util_type.cuh"
-
97 
+
56 #include "device/device_segmented_radix_sort.cuh"
+
57 #include "device/device_segmented_reduce.cuh"
+
58 #include "device/device_select.cuh"
+
59 #include "device/device_spmv.cuh"
+
60 
+
61 // Grid
+
62 //#include "grid/grid_barrier.cuh"
+
63 #include "grid/grid_even_share.cuh"
+
64 #include "grid/grid_mapping.cuh"
+
65 #include "grid/grid_queue.cuh"
+
66 
+
67 // Host
+
68 #include "host/spinlock.cuh"
+
69 
+
70 // Thread
+
71 #include "thread/thread_load.cuh"
+
72 #include "thread/thread_operators.cuh"
+
73 #include "thread/thread_reduce.cuh"
+
74 #include "thread/thread_scan.cuh"
+
75 #include "thread/thread_store.cuh"
+
76 
+
77 // Warp
+
78 #include "warp/warp_reduce.cuh"
+
79 #include "warp/warp_scan.cuh"
+
80 
+
81 // Iterator
+
82 #include "iterator/arg_index_input_iterator.cuh"
+
83 #include "iterator/cache_modified_input_iterator.cuh"
+
84 #include "iterator/cache_modified_output_iterator.cuh"
+
85 #include "iterator/constant_input_iterator.cuh"
+
86 #include "iterator/counting_input_iterator.cuh"
+
87 #include "iterator/tex_obj_input_iterator.cuh"
+
88 #include "iterator/tex_ref_input_iterator.cuh"
+
89 #include "iterator/transform_input_iterator.cuh"
+
90 
+
91 // Util
+
92 #include "util_allocator.cuh"
+
93 #include "util_arch.cuh"
+
94 #include "util_debug.cuh"
+
95 #include "util_device.cuh"
+
96 #include "util_macro.cuh"
+
97 #include "util_ptx.cuh"
+
98 #include "util_type.cuh"
+
99 
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__histogram_8cuh.html b/docs/html/device__histogram_8cuh.html index 5fa7a626b9..68250778fc 100644 --- a/docs/html/device__histogram_8cuh.html +++ b/docs/html/device__histogram_8cuh.html @@ -137,7 +137,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__histogram_8cuh_source.html b/docs/html/device__histogram_8cuh_source.html index 2ba5dfd544..25c2f364d5 100644 --- a/docs/html/device__histogram_8cuh_source.html +++ b/docs/html/device__histogram_8cuh_source.html @@ -148,333 +148,333 @@
64 {
65  /******************************************************************/
69 
-
119  template <
-
120  typename SampleIteratorT,
-
121  typename CounterT,
-
122  typename LevelT,
-
123  typename OffsetT>
-
124  CUB_RUNTIME_FUNCTION
-
125  static cudaError_t HistogramEven(
-
126  void* d_temp_storage,
-
127  size_t& temp_storage_bytes,
-
128  SampleIteratorT d_samples,
-
129  CounterT* d_histogram,
-
130  int num_levels,
-
131  LevelT lower_level,
-
132  LevelT upper_level,
-
133  OffsetT num_samples,
-
134  cudaStream_t stream = 0,
-
135  bool debug_synchronous = false)
-
136  {
-
138  typedef typename std::iterator_traits<SampleIteratorT>::value_type SampleT;
-
139 
-
140  CounterT* d_histogram1[1] = {d_histogram};
-
141  int num_levels1[1] = {num_levels};
-
142  LevelT lower_level1[1] = {lower_level};
-
143  LevelT upper_level1[1] = {upper_level};
-
144 
-
145  return MultiHistogramEven<1, 1>(
-
146  d_temp_storage,
-
147  temp_storage_bytes,
-
148  d_samples,
-
149  d_histogram1,
-
150  num_levels1,
-
151  lower_level1,
-
152  upper_level1,
-
153  num_samples,
-
154  1,
-
155  sizeof(SampleT) * num_samples,
-
156  stream,
-
157  debug_synchronous);
-
158  }
+
118  template <
+
119  typename SampleIteratorT,
+
120  typename CounterT,
+
121  typename LevelT,
+
122  typename OffsetT>
+
123  CUB_RUNTIME_FUNCTION
+
124  static cudaError_t HistogramEven(
+
125  void* d_temp_storage,
+
126  size_t& temp_storage_bytes,
+
127  SampleIteratorT d_samples,
+
128  CounterT* d_histogram,
+
129  int num_levels,
+
130  LevelT lower_level,
+
131  LevelT upper_level,
+
132  OffsetT num_samples,
+
133  cudaStream_t stream = 0,
+
134  bool debug_synchronous = false)
+
135  {
+
137  typedef typename std::iterator_traits<SampleIteratorT>::value_type SampleT;
+
138 
+
139  CounterT* d_histogram1[1] = {d_histogram};
+
140  int num_levels1[1] = {num_levels};
+
141  LevelT lower_level1[1] = {lower_level};
+
142  LevelT upper_level1[1] = {upper_level};
+
143 
+
144  return MultiHistogramEven<1, 1>(
+
145  d_temp_storage,
+
146  temp_storage_bytes,
+
147  d_samples,
+
148  d_histogram1,
+
149  num_levels1,
+
150  lower_level1,
+
151  upper_level1,
+
152  num_samples,
+
153  1,
+
154  sizeof(SampleT) * num_samples,
+
155  stream,
+
156  debug_synchronous);
+
157  }
+
158 
159 
-
160 
-
219  template <
-
220  typename SampleIteratorT,
-
221  typename CounterT,
-
222  typename LevelT,
-
223  typename OffsetT>
-
224  CUB_RUNTIME_FUNCTION
-
225  static cudaError_t HistogramEven(
-
226  void* d_temp_storage,
-
227  size_t& temp_storage_bytes,
-
228  SampleIteratorT d_samples,
-
229  CounterT* d_histogram,
-
230  int num_levels,
-
231  LevelT lower_level,
-
232  LevelT upper_level,
-
233  OffsetT num_row_samples,
-
234  OffsetT num_rows,
-
235  size_t row_stride_bytes,
-
236  cudaStream_t stream = 0,
-
237  bool debug_synchronous = false)
-
238  {
-
239  CounterT* d_histogram1[1] = {d_histogram};
-
240  int num_levels1[1] = {num_levels};
-
241  LevelT lower_level1[1] = {lower_level};
-
242  LevelT upper_level1[1] = {upper_level};
-
243 
-
244  return MultiHistogramEven<1, 1>(
-
245  d_temp_storage,
-
246  temp_storage_bytes,
-
247  d_samples,
-
248  d_histogram1,
-
249  num_levels1,
-
250  lower_level1,
-
251  upper_level1,
-
252  num_row_samples,
-
253  num_rows,
-
254  row_stride_bytes,
-
255  stream,
-
256  debug_synchronous);
-
257  }
-
258 
-
320  template <
-
321  int NUM_CHANNELS,
-
322  int NUM_ACTIVE_CHANNELS,
-
323  typename SampleIteratorT,
-
324  typename CounterT,
-
325  typename LevelT,
-
326  typename OffsetT>
-
327  CUB_RUNTIME_FUNCTION
-
328  static cudaError_t MultiHistogramEven(
-
329  void* d_temp_storage,
-
330  size_t& temp_storage_bytes,
-
331  SampleIteratorT d_samples,
-
332  CounterT* d_histogram[NUM_ACTIVE_CHANNELS],
-
333  int num_levels[NUM_ACTIVE_CHANNELS],
-
334  LevelT lower_level[NUM_ACTIVE_CHANNELS],
-
335  LevelT upper_level[NUM_ACTIVE_CHANNELS],
-
336  OffsetT num_pixels,
-
337  cudaStream_t stream = 0,
-
338  bool debug_synchronous = false)
-
339  {
-
341  typedef typename std::iterator_traits<SampleIteratorT>::value_type SampleT;
-
342 
-
343  return MultiHistogramEven<NUM_CHANNELS, NUM_ACTIVE_CHANNELS>(
-
344  d_temp_storage,
-
345  temp_storage_bytes,
-
346  d_samples,
-
347  d_histogram,
-
348  num_levels,
-
349  lower_level,
-
350  upper_level,
-
351  num_pixels,
-
352  1,
-
353  sizeof(SampleT) * NUM_CHANNELS * num_pixels,
-
354  stream,
-
355  debug_synchronous);
-
356  }
-
357 
-
358 
-
428  template <
-
429  int NUM_CHANNELS,
-
430  int NUM_ACTIVE_CHANNELS,
-
431  typename SampleIteratorT,
-
432  typename CounterT,
-
433  typename LevelT,
-
434  typename OffsetT>
-
435  CUB_RUNTIME_FUNCTION
-
436  static cudaError_t MultiHistogramEven(
-
437  void* d_temp_storage,
-
438  size_t& temp_storage_bytes,
-
439  SampleIteratorT d_samples,
-
440  CounterT* d_histogram[NUM_ACTIVE_CHANNELS],
-
441  int num_levels[NUM_ACTIVE_CHANNELS],
-
442  LevelT lower_level[NUM_ACTIVE_CHANNELS],
-
443  LevelT upper_level[NUM_ACTIVE_CHANNELS],
-
444  OffsetT num_row_pixels,
-
445  OffsetT num_rows,
-
446  size_t row_stride_bytes,
-
447  cudaStream_t stream = 0,
-
448  bool debug_synchronous = false)
-
449  {
-
451  typedef typename std::iterator_traits<SampleIteratorT>::value_type SampleT;
-
452  Int2Type<sizeof(SampleT) == 1> is_byte_sample;
+
217  template <
+
218  typename SampleIteratorT,
+
219  typename CounterT,
+
220  typename LevelT,
+
221  typename OffsetT>
+
222  CUB_RUNTIME_FUNCTION
+
223  static cudaError_t HistogramEven(
+
224  void* d_temp_storage,
+
225  size_t& temp_storage_bytes,
+
226  SampleIteratorT d_samples,
+
227  CounterT* d_histogram,
+
228  int num_levels,
+
229  LevelT lower_level,
+
230  LevelT upper_level,
+
231  OffsetT num_row_samples,
+
232  OffsetT num_rows,
+
233  size_t row_stride_bytes,
+
234  cudaStream_t stream = 0,
+
235  bool debug_synchronous = false)
+
236  {
+
237  CounterT* d_histogram1[1] = {d_histogram};
+
238  int num_levels1[1] = {num_levels};
+
239  LevelT lower_level1[1] = {lower_level};
+
240  LevelT upper_level1[1] = {upper_level};
+
241 
+
242  return MultiHistogramEven<1, 1>(
+
243  d_temp_storage,
+
244  temp_storage_bytes,
+
245  d_samples,
+
246  d_histogram1,
+
247  num_levels1,
+
248  lower_level1,
+
249  upper_level1,
+
250  num_row_samples,
+
251  num_rows,
+
252  row_stride_bytes,
+
253  stream,
+
254  debug_synchronous);
+
255  }
+
256 
+
317  template <
+
318  int NUM_CHANNELS,
+
319  int NUM_ACTIVE_CHANNELS,
+
320  typename SampleIteratorT,
+
321  typename CounterT,
+
322  typename LevelT,
+
323  typename OffsetT>
+
324  CUB_RUNTIME_FUNCTION
+
325  static cudaError_t MultiHistogramEven(
+
326  void* d_temp_storage,
+
327  size_t& temp_storage_bytes,
+
328  SampleIteratorT d_samples,
+
329  CounterT* d_histogram[NUM_ACTIVE_CHANNELS],
+
330  int num_levels[NUM_ACTIVE_CHANNELS],
+
331  LevelT lower_level[NUM_ACTIVE_CHANNELS],
+
332  LevelT upper_level[NUM_ACTIVE_CHANNELS],
+
333  OffsetT num_pixels,
+
334  cudaStream_t stream = 0,
+
335  bool debug_synchronous = false)
+
336  {
+
338  typedef typename std::iterator_traits<SampleIteratorT>::value_type SampleT;
+
339 
+
340  return MultiHistogramEven<NUM_CHANNELS, NUM_ACTIVE_CHANNELS>(
+
341  d_temp_storage,
+
342  temp_storage_bytes,
+
343  d_samples,
+
344  d_histogram,
+
345  num_levels,
+
346  lower_level,
+
347  upper_level,
+
348  num_pixels,
+
349  1,
+
350  sizeof(SampleT) * NUM_CHANNELS * num_pixels,
+
351  stream,
+
352  debug_synchronous);
+
353  }
+
354 
+
355 
+
424  template <
+
425  int NUM_CHANNELS,
+
426  int NUM_ACTIVE_CHANNELS,
+
427  typename SampleIteratorT,
+
428  typename CounterT,
+
429  typename LevelT,
+
430  typename OffsetT>
+
431  CUB_RUNTIME_FUNCTION
+
432  static cudaError_t MultiHistogramEven(
+
433  void* d_temp_storage,
+
434  size_t& temp_storage_bytes,
+
435  SampleIteratorT d_samples,
+
436  CounterT* d_histogram[NUM_ACTIVE_CHANNELS],
+
437  int num_levels[NUM_ACTIVE_CHANNELS],
+
438  LevelT lower_level[NUM_ACTIVE_CHANNELS],
+
439  LevelT upper_level[NUM_ACTIVE_CHANNELS],
+
440  OffsetT num_row_pixels,
+
441  OffsetT num_rows,
+
442  size_t row_stride_bytes,
+
443  cudaStream_t stream = 0,
+
444  bool debug_synchronous = false)
+
445  {
+
447  typedef typename std::iterator_traits<SampleIteratorT>::value_type SampleT;
+
448  Int2Type<sizeof(SampleT) == 1> is_byte_sample;
+
449 
+
450  if ((sizeof(OffsetT) > sizeof(int)) && (row_stride_bytes * num_rows < std::numeric_limits<int>::max()))
+
451  {
+
452  // Down-convert OffsetT data type
453 
-
454  if ((sizeof(OffsetT) > sizeof(int)) && (row_stride_bytes * num_rows < std::numeric_limits<int>::max()))
-
455  {
-
456  // Down-convert OffsetT data type
-
457 
-
458 
-
459  return DipatchHistogram<NUM_CHANNELS, NUM_ACTIVE_CHANNELS, SampleIteratorT, CounterT, LevelT, int>::DispatchEven(
-
460  d_temp_storage, temp_storage_bytes, d_samples, d_histogram, num_levels, lower_level, upper_level,
-
461  (int) num_row_pixels, (int) num_rows, (int) (row_stride_bytes / sizeof(SampleT)),
-
462  stream, debug_synchronous, is_byte_sample);
-
463  }
-
464 
-
465  return DipatchHistogram<NUM_CHANNELS, NUM_ACTIVE_CHANNELS, SampleIteratorT, CounterT, LevelT, OffsetT>::DispatchEven(
-
466  d_temp_storage, temp_storage_bytes, d_samples, d_histogram, num_levels, lower_level, upper_level,
-
467  num_row_pixels, num_rows, (OffsetT) (row_stride_bytes / sizeof(SampleT)),
-
468  stream, debug_synchronous, is_byte_sample);
-
469  }
-
470 
-
471 
-
473  /******************************************************************/
-
477 
-
526  template <
-
527  typename SampleIteratorT,
-
528  typename CounterT,
-
529  typename LevelT,
-
530  typename OffsetT>
-
531  CUB_RUNTIME_FUNCTION
-
532  static cudaError_t HistogramRange(
-
533  void* d_temp_storage,
-
534  size_t& temp_storage_bytes,
-
535  SampleIteratorT d_samples,
-
536  CounterT* d_histogram,
-
537  int num_levels,
-
538  LevelT* d_levels,
-
539  OffsetT num_samples,
-
540  cudaStream_t stream = 0,
-
541  bool debug_synchronous = false)
-
542  {
-
544  typedef typename std::iterator_traits<SampleIteratorT>::value_type SampleT;
-
545 
-
546  CounterT* d_histogram1[1] = {d_histogram};
-
547  int num_levels1[1] = {num_levels};
-
548  LevelT* d_levels1[1] = {d_levels};
-
549 
-
550  return MultiHistogramRange<1, 1>(
-
551  d_temp_storage,
-
552  temp_storage_bytes,
-
553  d_samples,
-
554  d_histogram1,
-
555  num_levels1,
-
556  d_levels1,
-
557  num_samples,
-
558  1,
-
559  sizeof(SampleT) * num_samples,
-
560  stream,
-
561  debug_synchronous);
-
562  }
-
563 
-
564 
-
622  template <
-
623  typename SampleIteratorT,
-
624  typename CounterT,
-
625  typename LevelT,
-
626  typename OffsetT>
-
627  CUB_RUNTIME_FUNCTION
-
628  static cudaError_t HistogramRange(
-
629  void* d_temp_storage,
-
630  size_t& temp_storage_bytes,
-
631  SampleIteratorT d_samples,
-
632  CounterT* d_histogram,
-
633  int num_levels,
-
634  LevelT* d_levels,
-
635  OffsetT num_row_samples,
-
636  OffsetT num_rows,
-
637  size_t row_stride_bytes,
-
638  cudaStream_t stream = 0,
-
639  bool debug_synchronous = false)
-
640  {
-
641  CounterT* d_histogram1[1] = {d_histogram};
-
642  int num_levels1[1] = {num_levels};
-
643  LevelT* d_levels1[1] = {d_levels};
-
644 
-
645  return MultiHistogramRange<1, 1>(
-
646  d_temp_storage,
-
647  temp_storage_bytes,
-
648  d_samples,
-
649  d_histogram1,
-
650  num_levels1,
-
651  d_levels1,
-
652  num_row_samples,
-
653  num_rows,
-
654  row_stride_bytes,
-
655  stream,
-
656  debug_synchronous);
-
657  }
-
658 
-
720  template <
-
721  int NUM_CHANNELS,
-
722  int NUM_ACTIVE_CHANNELS,
-
723  typename SampleIteratorT,
-
724  typename CounterT,
-
725  typename LevelT,
-
726  typename OffsetT>
-
727  CUB_RUNTIME_FUNCTION
-
728  static cudaError_t MultiHistogramRange(
-
729  void* d_temp_storage,
-
730  size_t& temp_storage_bytes,
-
731  SampleIteratorT d_samples,
-
732  CounterT* d_histogram[NUM_ACTIVE_CHANNELS],
-
733  int num_levels[NUM_ACTIVE_CHANNELS],
-
734  LevelT* d_levels[NUM_ACTIVE_CHANNELS],
-
735  OffsetT num_pixels,
-
736  cudaStream_t stream = 0,
-
737  bool debug_synchronous = false)
-
738  {
-
740  typedef typename std::iterator_traits<SampleIteratorT>::value_type SampleT;
-
741 
-
742  return MultiHistogramRange<NUM_CHANNELS, NUM_ACTIVE_CHANNELS>(
-
743  d_temp_storage,
-
744  temp_storage_bytes,
-
745  d_samples,
-
746  d_histogram,
-
747  num_levels,
-
748  d_levels,
-
749  num_pixels,
-
750  1,
-
751  sizeof(SampleT) * NUM_CHANNELS * num_pixels,
-
752  stream,
-
753  debug_synchronous);
-
754  }
-
755 
-
756 
-
824  template <
-
825  int NUM_CHANNELS,
-
826  int NUM_ACTIVE_CHANNELS,
-
827  typename SampleIteratorT,
-
828  typename CounterT,
-
829  typename LevelT,
-
830  typename OffsetT>
-
831  CUB_RUNTIME_FUNCTION
-
832  static cudaError_t MultiHistogramRange(
-
833  void* d_temp_storage,
-
834  size_t& temp_storage_bytes,
-
835  SampleIteratorT d_samples,
-
836  CounterT* d_histogram[NUM_ACTIVE_CHANNELS],
-
837  int num_levels[NUM_ACTIVE_CHANNELS],
-
838  LevelT* d_levels[NUM_ACTIVE_CHANNELS],
-
839  OffsetT num_row_pixels,
-
840  OffsetT num_rows,
-
841  size_t row_stride_bytes,
-
842  cudaStream_t stream = 0,
-
843  bool debug_synchronous = false)
-
844  {
-
846  typedef typename std::iterator_traits<SampleIteratorT>::value_type SampleT;
-
847  Int2Type<sizeof(SampleT) == 1> is_byte_sample;
-
848 
-
849  if ((sizeof(OffsetT) > sizeof(int)) && (row_stride_bytes * num_rows < std::numeric_limits<int>::max()))
-
850  {
-
851  // Down-convert OffsetT data type
-
852  return DipatchHistogram<NUM_CHANNELS, NUM_ACTIVE_CHANNELS, SampleIteratorT, CounterT, LevelT, int>::DispatchRange(
-
853  d_temp_storage, temp_storage_bytes, d_samples, d_histogram, num_levels, d_levels,
-
854  (int) num_row_pixels, (int) num_rows, (int) (row_stride_bytes / sizeof(SampleT)),
-
855  stream, debug_synchronous, is_byte_sample);
-
856  }
+
454 
+
455  return DipatchHistogram<NUM_CHANNELS, NUM_ACTIVE_CHANNELS, SampleIteratorT, CounterT, LevelT, int>::DispatchEven(
+
456  d_temp_storage, temp_storage_bytes, d_samples, d_histogram, num_levels, lower_level, upper_level,
+
457  (int) num_row_pixels, (int) num_rows, (int) (row_stride_bytes / sizeof(SampleT)),
+
458  stream, debug_synchronous, is_byte_sample);
+
459  }
+
460 
+
461  return DipatchHistogram<NUM_CHANNELS, NUM_ACTIVE_CHANNELS, SampleIteratorT, CounterT, LevelT, OffsetT>::DispatchEven(
+
462  d_temp_storage, temp_storage_bytes, d_samples, d_histogram, num_levels, lower_level, upper_level,
+
463  num_row_pixels, num_rows, (OffsetT) (row_stride_bytes / sizeof(SampleT)),
+
464  stream, debug_synchronous, is_byte_sample);
+
465  }
+
466 
+
467 
+
469  /******************************************************************/
+
473 
+
521  template <
+
522  typename SampleIteratorT,
+
523  typename CounterT,
+
524  typename LevelT,
+
525  typename OffsetT>
+
526  CUB_RUNTIME_FUNCTION
+
527  static cudaError_t HistogramRange(
+
528  void* d_temp_storage,
+
529  size_t& temp_storage_bytes,
+
530  SampleIteratorT d_samples,
+
531  CounterT* d_histogram,
+
532  int num_levels,
+
533  LevelT* d_levels,
+
534  OffsetT num_samples,
+
535  cudaStream_t stream = 0,
+
536  bool debug_synchronous = false)
+
537  {
+
539  typedef typename std::iterator_traits<SampleIteratorT>::value_type SampleT;
+
540 
+
541  CounterT* d_histogram1[1] = {d_histogram};
+
542  int num_levels1[1] = {num_levels};
+
543  LevelT* d_levels1[1] = {d_levels};
+
544 
+
545  return MultiHistogramRange<1, 1>(
+
546  d_temp_storage,
+
547  temp_storage_bytes,
+
548  d_samples,
+
549  d_histogram1,
+
550  num_levels1,
+
551  d_levels1,
+
552  num_samples,
+
553  1,
+
554  sizeof(SampleT) * num_samples,
+
555  stream,
+
556  debug_synchronous);
+
557  }
+
558 
+
559 
+
616  template <
+
617  typename SampleIteratorT,
+
618  typename CounterT,
+
619  typename LevelT,
+
620  typename OffsetT>
+
621  CUB_RUNTIME_FUNCTION
+
622  static cudaError_t HistogramRange(
+
623  void* d_temp_storage,
+
624  size_t& temp_storage_bytes,
+
625  SampleIteratorT d_samples,
+
626  CounterT* d_histogram,
+
627  int num_levels,
+
628  LevelT* d_levels,
+
629  OffsetT num_row_samples,
+
630  OffsetT num_rows,
+
631  size_t row_stride_bytes,
+
632  cudaStream_t stream = 0,
+
633  bool debug_synchronous = false)
+
634  {
+
635  CounterT* d_histogram1[1] = {d_histogram};
+
636  int num_levels1[1] = {num_levels};
+
637  LevelT* d_levels1[1] = {d_levels};
+
638 
+
639  return MultiHistogramRange<1, 1>(
+
640  d_temp_storage,
+
641  temp_storage_bytes,
+
642  d_samples,
+
643  d_histogram1,
+
644  num_levels1,
+
645  d_levels1,
+
646  num_row_samples,
+
647  num_rows,
+
648  row_stride_bytes,
+
649  stream,
+
650  debug_synchronous);
+
651  }
+
652 
+
713  template <
+
714  int NUM_CHANNELS,
+
715  int NUM_ACTIVE_CHANNELS,
+
716  typename SampleIteratorT,
+
717  typename CounterT,
+
718  typename LevelT,
+
719  typename OffsetT>
+
720  CUB_RUNTIME_FUNCTION
+
721  static cudaError_t MultiHistogramRange(
+
722  void* d_temp_storage,
+
723  size_t& temp_storage_bytes,
+
724  SampleIteratorT d_samples,
+
725  CounterT* d_histogram[NUM_ACTIVE_CHANNELS],
+
726  int num_levels[NUM_ACTIVE_CHANNELS],
+
727  LevelT* d_levels[NUM_ACTIVE_CHANNELS],
+
728  OffsetT num_pixels,
+
729  cudaStream_t stream = 0,
+
730  bool debug_synchronous = false)
+
731  {
+
733  typedef typename std::iterator_traits<SampleIteratorT>::value_type SampleT;
+
734 
+
735  return MultiHistogramRange<NUM_CHANNELS, NUM_ACTIVE_CHANNELS>(
+
736  d_temp_storage,
+
737  temp_storage_bytes,
+
738  d_samples,
+
739  d_histogram,
+
740  num_levels,
+
741  d_levels,
+
742  num_pixels,
+
743  1,
+
744  sizeof(SampleT) * NUM_CHANNELS * num_pixels,
+
745  stream,
+
746  debug_synchronous);
+
747  }
+
748 
+
749 
+
816  template <
+
817  int NUM_CHANNELS,
+
818  int NUM_ACTIVE_CHANNELS,
+
819  typename SampleIteratorT,
+
820  typename CounterT,
+
821  typename LevelT,
+
822  typename OffsetT>
+
823  CUB_RUNTIME_FUNCTION
+
824  static cudaError_t MultiHistogramRange(
+
825  void* d_temp_storage,
+
826  size_t& temp_storage_bytes,
+
827  SampleIteratorT d_samples,
+
828  CounterT* d_histogram[NUM_ACTIVE_CHANNELS],
+
829  int num_levels[NUM_ACTIVE_CHANNELS],
+
830  LevelT* d_levels[NUM_ACTIVE_CHANNELS],
+
831  OffsetT num_row_pixels,
+
832  OffsetT num_rows,
+
833  size_t row_stride_bytes,
+
834  cudaStream_t stream = 0,
+
835  bool debug_synchronous = false)
+
836  {
+
838  typedef typename std::iterator_traits<SampleIteratorT>::value_type SampleT;
+
839  Int2Type<sizeof(SampleT) == 1> is_byte_sample;
+
840 
+
841  if ((sizeof(OffsetT) > sizeof(int)) && (row_stride_bytes * num_rows < std::numeric_limits<int>::max()))
+
842  {
+
843  // Down-convert OffsetT data type
+
844  return DipatchHistogram<NUM_CHANNELS, NUM_ACTIVE_CHANNELS, SampleIteratorT, CounterT, LevelT, int>::DispatchRange(
+
845  d_temp_storage, temp_storage_bytes, d_samples, d_histogram, num_levels, d_levels,
+
846  (int) num_row_pixels, (int) num_rows, (int) (row_stride_bytes / sizeof(SampleT)),
+
847  stream, debug_synchronous, is_byte_sample);
+
848  }
+
849 
+
850  return DipatchHistogram<NUM_CHANNELS, NUM_ACTIVE_CHANNELS, SampleIteratorT, CounterT, LevelT, OffsetT>::DispatchRange(
+
851  d_temp_storage, temp_storage_bytes, d_samples, d_histogram, num_levels, d_levels,
+
852  num_row_pixels, num_rows, (OffsetT) (row_stride_bytes / sizeof(SampleT)),
+
853  stream, debug_synchronous, is_byte_sample);
+
854  }
+
855 
+
856 
857 
-
858  return DipatchHistogram<NUM_CHANNELS, NUM_ACTIVE_CHANNELS, SampleIteratorT, CounterT, LevelT, OffsetT>::DispatchRange(
-
859  d_temp_storage, temp_storage_bytes, d_samples, d_histogram, num_levels, d_levels,
-
860  num_row_pixels, num_rows, (OffsetT) (row_stride_bytes / sizeof(SampleT)),
-
861  stream, debug_synchronous, is_byte_sample);
-
862  }
-
863 
-
864 
-
865 
-
867 };
+
859 };
+
860 
+
865 } // CUB namespace
+
866 CUB_NS_POSTFIX // Optional outer namespace(s)
+
867 
868 
-
873 } // CUB namespace
-
874 CUB_NS_POSTFIX // Optional outer namespace(s)
-
875 
-
876 
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__partition_8cuh.html b/docs/html/device__partition_8cuh.html index 999d5ee6f6..7ee16ef467 100644 --- a/docs/html/device__partition_8cuh.html +++ b/docs/html/device__partition_8cuh.html @@ -136,7 +136,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__partition_8cuh_source.html b/docs/html/device__partition_8cuh_source.html index ae92126710..28ef8bb5ff 100644 --- a/docs/html/device__partition_8cuh_source.html +++ b/docs/html/device__partition_8cuh_source.html @@ -145,88 +145,88 @@
49 
73 struct DevicePartition
74 {
-
121  template <
-
122  typename InputIteratorT,
-
123  typename FlagIterator,
-
124  typename OutputIteratorT,
-
125  typename NumSelectedIteratorT>
-
126  CUB_RUNTIME_FUNCTION __forceinline__
-
127  static cudaError_t Flagged(
-
128  void* d_temp_storage,
-
129  size_t &temp_storage_bytes,
-
130  InputIteratorT d_in,
-
131  FlagIterator d_flags,
-
132  OutputIteratorT d_out,
-
133  NumSelectedIteratorT d_num_selected_out,
-
134  int num_items,
-
135  cudaStream_t stream = 0,
-
136  bool debug_synchronous = false)
-
137  {
-
138  typedef int OffsetT; // Signed integer type for global offsets
-
139  typedef NullType SelectOp; // Selection op (not used)
-
140  typedef NullType EqualityOp; // Equality operator (not used)
-
141 
-
142  return DispatchSelectIf<InputIteratorT, FlagIterator, OutputIteratorT, NumSelectedIteratorT, SelectOp, EqualityOp, OffsetT, true>::Dispatch(
-
143  d_temp_storage,
-
144  temp_storage_bytes,
-
145  d_in,
-
146  d_flags,
-
147  d_out,
-
148  d_num_selected_out,
-
149  SelectOp(),
-
150  EqualityOp(),
-
151  num_items,
-
152  stream,
-
153  debug_synchronous);
-
154  }
+
120  template <
+
121  typename InputIteratorT,
+
122  typename FlagIterator,
+
123  typename OutputIteratorT,
+
124  typename NumSelectedIteratorT>
+
125  CUB_RUNTIME_FUNCTION __forceinline__
+
126  static cudaError_t Flagged(
+
127  void* d_temp_storage,
+
128  size_t &temp_storage_bytes,
+
129  InputIteratorT d_in,
+
130  FlagIterator d_flags,
+
131  OutputIteratorT d_out,
+
132  NumSelectedIteratorT d_num_selected_out,
+
133  int num_items,
+
134  cudaStream_t stream = 0,
+
135  bool debug_synchronous = false)
+
136  {
+
137  typedef int OffsetT; // Signed integer type for global offsets
+
138  typedef NullType SelectOp; // Selection op (not used)
+
139  typedef NullType EqualityOp; // Equality operator (not used)
+
140 
+
141  return DispatchSelectIf<InputIteratorT, FlagIterator, OutputIteratorT, NumSelectedIteratorT, SelectOp, EqualityOp, OffsetT, true>::Dispatch(
+
142  d_temp_storage,
+
143  temp_storage_bytes,
+
144  d_in,
+
145  d_flags,
+
146  d_out,
+
147  d_num_selected_out,
+
148  SelectOp(),
+
149  EqualityOp(),
+
150  num_items,
+
151  stream,
+
152  debug_synchronous);
+
153  }
+
154 
155 
-
156 
-
230  template <
-
231  typename InputIteratorT,
-
232  typename OutputIteratorT,
-
233  typename NumSelectedIteratorT,
-
234  typename SelectOp>
-
235  CUB_RUNTIME_FUNCTION __forceinline__
-
236  static cudaError_t If(
-
237  void* d_temp_storage,
-
238  size_t &temp_storage_bytes,
-
239  InputIteratorT d_in,
-
240  OutputIteratorT d_out,
-
241  NumSelectedIteratorT d_num_selected_out,
-
242  int num_items,
-
243  SelectOp select_op,
-
244  cudaStream_t stream = 0,
-
245  bool debug_synchronous = false)
-
246  {
-
247  typedef int OffsetT; // Signed integer type for global offsets
-
248  typedef NullType* FlagIterator; // FlagT iterator type (not used)
-
249  typedef NullType EqualityOp; // Equality operator (not used)
-
250 
-
251  return DispatchSelectIf<InputIteratorT, FlagIterator, OutputIteratorT, NumSelectedIteratorT, SelectOp, EqualityOp, OffsetT, true>::Dispatch(
-
252  d_temp_storage,
-
253  temp_storage_bytes,
-
254  d_in,
-
255  NULL,
-
256  d_out,
-
257  d_num_selected_out,
-
258  select_op,
-
259  EqualityOp(),
-
260  num_items,
-
261  stream,
-
262  debug_synchronous);
-
263  }
+
228  template <
+
229  typename InputIteratorT,
+
230  typename OutputIteratorT,
+
231  typename NumSelectedIteratorT,
+
232  typename SelectOp>
+
233  CUB_RUNTIME_FUNCTION __forceinline__
+
234  static cudaError_t If(
+
235  void* d_temp_storage,
+
236  size_t &temp_storage_bytes,
+
237  InputIteratorT d_in,
+
238  OutputIteratorT d_out,
+
239  NumSelectedIteratorT d_num_selected_out,
+
240  int num_items,
+
241  SelectOp select_op,
+
242  cudaStream_t stream = 0,
+
243  bool debug_synchronous = false)
+
244  {
+
245  typedef int OffsetT; // Signed integer type for global offsets
+
246  typedef NullType* FlagIterator; // FlagT iterator type (not used)
+
247  typedef NullType EqualityOp; // Equality operator (not used)
+
248 
+
249  return DispatchSelectIf<InputIteratorT, FlagIterator, OutputIteratorT, NumSelectedIteratorT, SelectOp, EqualityOp, OffsetT, true>::Dispatch(
+
250  d_temp_storage,
+
251  temp_storage_bytes,
+
252  d_in,
+
253  NULL,
+
254  d_out,
+
255  d_num_selected_out,
+
256  select_op,
+
257  EqualityOp(),
+
258  num_items,
+
259  stream,
+
260  debug_synchronous);
+
261  }
+
262 
+
263 };
264 
-
265 };
-
266 
-
272 } // CUB namespace
-
273 CUB_NS_POSTFIX // Optional outer namespace(s)
-
274 
-
275 
+
270 } // CUB namespace
+
271 CUB_NS_POSTFIX // Optional outer namespace(s)
+
272 
+
273 
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__radix__sort_8cuh.html b/docs/html/device__radix__sort_8cuh.html index 95b986135e..a675af87a9 100644 --- a/docs/html/device__radix__sort_8cuh.html +++ b/docs/html/device__radix__sort_8cuh.html @@ -137,7 +137,7 @@
-Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__radix__sort_8cuh_source.html b/docs/html/device__radix__sort_8cuh_source.html index 4958207fda..b30b00fdf2 100644 --- a/docs/html/device__radix__sort_8cuh_source.html +++ b/docs/html/device__radix__sort_8cuh_source.html @@ -149,290 +149,290 @@
84 
85  /******************************************************************/
89 
-
144  template <
-
145  typename KeyT,
-
146  typename ValueT>
-
147  CUB_RUNTIME_FUNCTION
-
148  static cudaError_t SortPairs(
-
149  void *d_temp_storage,
-
150  size_t &temp_storage_bytes,
-
151  KeyT *d_keys_in,
-
152  KeyT *d_keys_out,
-
153  ValueT *d_values_in,
-
154  ValueT *d_values_out,
-
155  int num_items,
-
156  int begin_bit = 0,
-
157  int end_bit = sizeof(KeyT) * 8,
-
158  cudaStream_t stream = 0,
-
159  bool debug_synchronous = false)
-
160  {
-
161  // Signed integer type for global offsets
-
162  typedef int OffsetT;
-
163 
-
164  DoubleBuffer<KeyT> d_keys(d_keys_in, d_keys_out);
-
165  DoubleBuffer<ValueT> d_values(d_values_in, d_values_out);
-
166 
-
167  return DispatchRadixSort<false, KeyT, ValueT, OffsetT>::Dispatch(
-
168  d_temp_storage,
-
169  temp_storage_bytes,
-
170  d_keys,
-
171  d_values,
-
172  num_items,
-
173  begin_bit,
-
174  end_bit,
-
175  false,
-
176  stream,
-
177  debug_synchronous);
-
178  }
+
143  template <
+
144  typename KeyT,
+
145  typename ValueT>
+
146  CUB_RUNTIME_FUNCTION
+
147  static cudaError_t SortPairs(
+
148  void *d_temp_storage,
+
149  size_t &temp_storage_bytes,
+
150  KeyT *d_keys_in,
+
151  KeyT *d_keys_out,
+
152  ValueT *d_values_in,
+
153  ValueT *d_values_out,
+
154  int num_items,
+
155  int begin_bit = 0,
+
156  int end_bit = sizeof(KeyT) * 8,
+
157  cudaStream_t stream = 0,
+
158  bool debug_synchronous = false)
+
159  {
+
160  // Signed integer type for global offsets
+
161  typedef int OffsetT;
+
162 
+
163  DoubleBuffer<KeyT> d_keys(d_keys_in, d_keys_out);
+
164  DoubleBuffer<ValueT> d_values(d_values_in, d_values_out);
+
165 
+
166  return DispatchRadixSort<false, KeyT, ValueT, OffsetT>::Dispatch(
+
167  d_temp_storage,
+
168  temp_storage_bytes,
+
169  d_keys,
+
170  d_values,
+
171  num_items,
+
172  begin_bit,
+
173  end_bit,
+
174  false,
+
175  stream,
+
176  debug_synchronous);
+
177  }
+
178 
179 
-
180 
-
246  template <
-
247  typename KeyT,
-
248  typename ValueT>
-
249  CUB_RUNTIME_FUNCTION
-
250  static cudaError_t SortPairs(
-
251  void *d_temp_storage,
-
252  size_t &temp_storage_bytes,
-
253  DoubleBuffer<KeyT> &d_keys,
-
254  DoubleBuffer<ValueT> &d_values,
-
255  int num_items,
-
256  int begin_bit = 0,
-
257  int end_bit = sizeof(KeyT) * 8,
-
258  cudaStream_t stream = 0,
-
259  bool debug_synchronous = false)
-
260  {
-
261  // Signed integer type for global offsets
-
262  typedef int OffsetT;
-
263 
-
264  return DispatchRadixSort<false, KeyT, ValueT, OffsetT>::Dispatch(
-
265  d_temp_storage,
-
266  temp_storage_bytes,
-
267  d_keys,
-
268  d_values,
-
269  num_items,
-
270  begin_bit,
-
271  end_bit,
-
272  true,
-
273  stream,
-
274  debug_synchronous);
-
275  }
-
276 
-
277 
-
327  template <
-
328  typename KeyT,
-
329  typename ValueT>
-
330  CUB_RUNTIME_FUNCTION
-
331  static cudaError_t SortPairsDescending(
-
332  void *d_temp_storage,
-
333  size_t &temp_storage_bytes,
-
334  KeyT *d_keys_in,
-
335  KeyT *d_keys_out,
-
336  ValueT *d_values_in,
-
337  ValueT *d_values_out,
-
338  int num_items,
-
339  int begin_bit = 0,
-
340  int end_bit = sizeof(KeyT) * 8,
-
341  cudaStream_t stream = 0,
-
342  bool debug_synchronous = false)
-
343  {
-
344  // Signed integer type for global offsets
-
345  typedef int OffsetT;
+
244  template <
+
245  typename KeyT,
+
246  typename ValueT>
+
247  CUB_RUNTIME_FUNCTION
+
248  static cudaError_t SortPairs(
+
249  void *d_temp_storage,
+
250  size_t &temp_storage_bytes,
+
251  DoubleBuffer<KeyT> &d_keys,
+
252  DoubleBuffer<ValueT> &d_values,
+
253  int num_items,
+
254  int begin_bit = 0,
+
255  int end_bit = sizeof(KeyT) * 8,
+
256  cudaStream_t stream = 0,
+
257  bool debug_synchronous = false)
+
258  {
+
259  // Signed integer type for global offsets
+
260  typedef int OffsetT;
+
261 
+
262  return DispatchRadixSort<false, KeyT, ValueT, OffsetT>::Dispatch(
+
263  d_temp_storage,
+
264  temp_storage_bytes,
+
265  d_keys,
+
266  d_values,
+
267  num_items,
+
268  begin_bit,
+
269  end_bit,
+
270  true,
+
271  stream,
+
272  debug_synchronous);
+
273  }
+
274 
+
275 
+
324  template <
+
325  typename KeyT,
+
326  typename ValueT>
+
327  CUB_RUNTIME_FUNCTION
+
328  static cudaError_t SortPairsDescending(
+
329  void *d_temp_storage,
+
330  size_t &temp_storage_bytes,
+
331  KeyT *d_keys_in,
+
332  KeyT *d_keys_out,
+
333  ValueT *d_values_in,
+
334  ValueT *d_values_out,
+
335  int num_items,
+
336  int begin_bit = 0,
+
337  int end_bit = sizeof(KeyT) * 8,
+
338  cudaStream_t stream = 0,
+
339  bool debug_synchronous = false)
+
340  {
+
341  // Signed integer type for global offsets
+
342  typedef int OffsetT;
+
343 
+
344  DoubleBuffer<KeyT> d_keys(d_keys_in, d_keys_out);
+
345  DoubleBuffer<ValueT> d_values(d_values_in, d_values_out);
346 
-
347  DoubleBuffer<KeyT> d_keys(d_keys_in, d_keys_out);
-
348  DoubleBuffer<ValueT> d_values(d_values_in, d_values_out);
-
349 
-
350  return DispatchRadixSort<true, KeyT, ValueT, OffsetT>::Dispatch(
-
351  d_temp_storage,
-
352  temp_storage_bytes,
-
353  d_keys,
-
354  d_values,
-
355  num_items,
-
356  begin_bit,
-
357  end_bit,
-
358  false,
-
359  stream,
-
360  debug_synchronous);
-
361  }
-
362 
-
363 
-
424  template <
-
425  typename KeyT,
-
426  typename ValueT>
-
427  CUB_RUNTIME_FUNCTION
-
428  static cudaError_t SortPairsDescending(
-
429  void *d_temp_storage,
-
430  size_t &temp_storage_bytes,
-
431  DoubleBuffer<KeyT> &d_keys,
-
432  DoubleBuffer<ValueT> &d_values,
-
433  int num_items,
-
434  int begin_bit = 0,
-
435  int end_bit = sizeof(KeyT) * 8,
-
436  cudaStream_t stream = 0,
-
437  bool debug_synchronous = false)
-
438  {
-
439  // Signed integer type for global offsets
-
440  typedef int OffsetT;
-
441 
-
442  return DispatchRadixSort<true, KeyT, ValueT, OffsetT>::Dispatch(
-
443  d_temp_storage,
-
444  temp_storage_bytes,
-
445  d_keys,
-
446  d_values,
-
447  num_items,
-
448  begin_bit,
-
449  end_bit,
-
450  true,
-
451  stream,
-
452  debug_synchronous);
-
453  }
-
454 
-
455 
-
457  /******************************************************************/
-
461 
-
462 
-
509  template <typename KeyT>
-
510  CUB_RUNTIME_FUNCTION
-
511  static cudaError_t SortKeys(
-
512  void *d_temp_storage,
-
513  size_t &temp_storage_bytes,
-
514  KeyT *d_keys_in,
-
515  KeyT *d_keys_out,
-
516  int num_items,
-
517  int begin_bit = 0,
-
518  int end_bit = sizeof(KeyT) * 8,
-
519  cudaStream_t stream = 0,
-
520  bool debug_synchronous = false)
-
521  {
-
522  // Signed integer type for global offsets
-
523  typedef int OffsetT;
-
524 
-
525  // Null value type
-
526  DoubleBuffer<KeyT> d_keys(d_keys_in, d_keys_out);
-
527  DoubleBuffer<NullType> d_values;
-
528 
-
529  return DispatchRadixSort<false, KeyT, NullType, OffsetT>::Dispatch(
-
530  d_temp_storage,
-
531  temp_storage_bytes,
-
532  d_keys,
-
533  d_values,
-
534  num_items,
-
535  begin_bit,
-
536  end_bit,
-
537  false,
-
538  stream,
-
539  debug_synchronous);
-
540  }
-
541 
-
542 
-
599  template <typename KeyT>
-
600  CUB_RUNTIME_FUNCTION
-
601  static cudaError_t SortKeys(
-
602  void *d_temp_storage,
-
603  size_t &temp_storage_bytes,
-
604  DoubleBuffer<KeyT> &d_keys,
-
605  int num_items,
-
606  int begin_bit = 0,
-
607  int end_bit = sizeof(KeyT) * 8,
-
608  cudaStream_t stream = 0,
-
609  bool debug_synchronous = false)
-
610  {
-
611  // Signed integer type for global offsets
-
612  typedef int OffsetT;
-
613 
-
614  // Null value type
-
615  DoubleBuffer<NullType> d_values;
-
616 
-
617  return DispatchRadixSort<false, KeyT, NullType, OffsetT>::Dispatch(
-
618  d_temp_storage,
-
619  temp_storage_bytes,
-
620  d_keys,
-
621  d_values,
-
622  num_items,
-
623  begin_bit,
-
624  end_bit,
-
625  true,
-
626  stream,
-
627  debug_synchronous);
-
628  }
-
629 
-
675  template <typename KeyT>
-
676  CUB_RUNTIME_FUNCTION
-
677  static cudaError_t SortKeysDescending(
-
678  void *d_temp_storage,
-
679  size_t &temp_storage_bytes,
-
680  KeyT *d_keys_in,
-
681  KeyT *d_keys_out,
-
682  int num_items,
-
683  int begin_bit = 0,
-
684  int end_bit = sizeof(KeyT) * 8,
-
685  cudaStream_t stream = 0,
-
686  bool debug_synchronous = false)
-
687  {
-
688  // Signed integer type for global offsets
-
689  typedef int OffsetT;
-
690 
-
691  DoubleBuffer<KeyT> d_keys(d_keys_in, d_keys_out);
-
692  DoubleBuffer<NullType> d_values;
-
693 
-
694  return DispatchRadixSort<true, KeyT, NullType, OffsetT>::Dispatch(
-
695  d_temp_storage,
-
696  temp_storage_bytes,
-
697  d_keys,
-
698  d_values,
-
699  num_items,
-
700  begin_bit,
-
701  end_bit,
-
702  false,
-
703  stream,
-
704  debug_synchronous);
-
705  }
-
706 
-
707 
-
760  template <typename KeyT>
-
761  CUB_RUNTIME_FUNCTION
-
762  static cudaError_t SortKeysDescending(
-
763  void *d_temp_storage,
-
764  size_t &temp_storage_bytes,
-
765  DoubleBuffer<KeyT> &d_keys,
-
766  int num_items,
-
767  int begin_bit = 0,
-
768  int end_bit = sizeof(KeyT) * 8,
-
769  cudaStream_t stream = 0,
-
770  bool debug_synchronous = false)
-
771  {
-
772  // Signed integer type for global offsets
-
773  typedef int OffsetT;
-
774 
-
775  // Null value type
-
776  DoubleBuffer<NullType> d_values;
-
777 
-
778  return DispatchRadixSort<true, KeyT, NullType, OffsetT>::Dispatch(
-
779  d_temp_storage,
-
780  temp_storage_bytes,
-
781  d_keys,
-
782  d_values,
-
783  num_items,
-
784  begin_bit,
-
785  end_bit,
-
786  true,
-
787  stream,
-
788  debug_synchronous);
-
789  }
-
790 
-
791 
-
793 
-
794 
-
795 };
+
347  return DispatchRadixSort<true, KeyT, ValueT, OffsetT>::Dispatch(
+
348  d_temp_storage,
+
349  temp_storage_bytes,
+
350  d_keys,
+
351  d_values,
+
352  num_items,
+
353  begin_bit,
+
354  end_bit,
+
355  false,
+
356  stream,
+
357  debug_synchronous);
+
358  }
+
359 
+
360 
+
420  template <
+
421  typename KeyT,
+
422  typename ValueT>
+
423  CUB_RUNTIME_FUNCTION
+
424  static cudaError_t SortPairsDescending(
+
425  void *d_temp_storage,
+
426  size_t &temp_storage_bytes,
+
427  DoubleBuffer<KeyT> &d_keys,
+
428  DoubleBuffer<ValueT> &d_values,
+
429  int num_items,
+
430  int begin_bit = 0,
+
431  int end_bit = sizeof(KeyT) * 8,
+
432  cudaStream_t stream = 0,
+
433  bool debug_synchronous = false)
+
434  {
+
435  // Signed integer type for global offsets
+
436  typedef int OffsetT;
+
437 
+
438  return DispatchRadixSort<true, KeyT, ValueT, OffsetT>::Dispatch(
+
439  d_temp_storage,
+
440  temp_storage_bytes,
+
441  d_keys,
+
442  d_values,
+
443  num_items,
+
444  begin_bit,
+
445  end_bit,
+
446  true,
+
447  stream,
+
448  debug_synchronous);
+
449  }
+
450 
+
451 
+
453  /******************************************************************/
+
457 
+
458 
+
504  template <typename KeyT>
+
505  CUB_RUNTIME_FUNCTION
+
506  static cudaError_t SortKeys(
+
507  void *d_temp_storage,
+
508  size_t &temp_storage_bytes,
+
509  KeyT *d_keys_in,
+
510  KeyT *d_keys_out,
+
511  int num_items,
+
512  int begin_bit = 0,
+
513  int end_bit = sizeof(KeyT) * 8,
+
514  cudaStream_t stream = 0,
+
515  bool debug_synchronous = false)
+
516  {
+
517  // Signed integer type for global offsets
+
518  typedef int OffsetT;
+
519 
+
520  // Null value type
+
521  DoubleBuffer<KeyT> d_keys(d_keys_in, d_keys_out);
+
522  DoubleBuffer<NullType> d_values;
+
523 
+
524  return DispatchRadixSort<false, KeyT, NullType, OffsetT>::Dispatch(
+
525  d_temp_storage,
+
526  temp_storage_bytes,
+
527  d_keys,
+
528  d_values,
+
529  num_items,
+
530  begin_bit,
+
531  end_bit,
+
532  false,
+
533  stream,
+
534  debug_synchronous);
+
535  }
+
536 
+
537 
+
593  template <typename KeyT>
+
594  CUB_RUNTIME_FUNCTION
+
595  static cudaError_t SortKeys(
+
596  void *d_temp_storage,
+
597  size_t &temp_storage_bytes,
+
598  DoubleBuffer<KeyT> &d_keys,
+
599  int num_items,
+
600  int begin_bit = 0,
+
601  int end_bit = sizeof(KeyT) * 8,
+
602  cudaStream_t stream = 0,
+
603  bool debug_synchronous = false)
+
604  {
+
605  // Signed integer type for global offsets
+
606  typedef int OffsetT;
+
607 
+
608  // Null value type
+
609  DoubleBuffer<NullType> d_values;
+
610 
+
611  return DispatchRadixSort<false, KeyT, NullType, OffsetT>::Dispatch(
+
612  d_temp_storage,
+
613  temp_storage_bytes,
+
614  d_keys,
+
615  d_values,
+
616  num_items,
+
617  begin_bit,
+
618  end_bit,
+
619  true,
+
620  stream,
+
621  debug_synchronous);
+
622  }
+
623 
+
668  template <typename KeyT>
+
669  CUB_RUNTIME_FUNCTION
+
670  static cudaError_t SortKeysDescending(
+
671  void *d_temp_storage,
+
672  size_t &temp_storage_bytes,
+
673  KeyT *d_keys_in,
+
674  KeyT *d_keys_out,
+
675  int num_items,
+
676  int begin_bit = 0,
+
677  int end_bit = sizeof(KeyT) * 8,
+
678  cudaStream_t stream = 0,
+
679  bool debug_synchronous = false)
+
680  {
+
681  // Signed integer type for global offsets
+
682  typedef int OffsetT;
+
683 
+
684  DoubleBuffer<KeyT> d_keys(d_keys_in, d_keys_out);
+
685  DoubleBuffer<NullType> d_values;
+
686 
+
687  return DispatchRadixSort<true, KeyT, NullType, OffsetT>::Dispatch(
+
688  d_temp_storage,
+
689  temp_storage_bytes,
+
690  d_keys,
+
691  d_values,
+
692  num_items,
+
693  begin_bit,
+
694  end_bit,
+
695  false,
+
696  stream,
+
697  debug_synchronous);
+
698  }
+
699 
+
700 
+
752  template <typename KeyT>
+
753  CUB_RUNTIME_FUNCTION
+
754  static cudaError_t SortKeysDescending(
+
755  void *d_temp_storage,
+
756  size_t &temp_storage_bytes,
+
757  DoubleBuffer<KeyT> &d_keys,
+
758  int num_items,
+
759  int begin_bit = 0,
+
760  int end_bit = sizeof(KeyT) * 8,
+
761  cudaStream_t stream = 0,
+
762  bool debug_synchronous = false)
+
763  {
+
764  // Signed integer type for global offsets
+
765  typedef int OffsetT;
+
766 
+
767  // Null value type
+
768  DoubleBuffer<NullType> d_values;
+
769 
+
770  return DispatchRadixSort<true, KeyT, NullType, OffsetT>::Dispatch(
+
771  d_temp_storage,
+
772  temp_storage_bytes,
+
773  d_keys,
+
774  d_values,
+
775  num_items,
+
776  begin_bit,
+
777  end_bit,
+
778  true,
+
779  stream,
+
780  debug_synchronous);
+
781  }
+
782 
+
783 
+
785 
+
786 
+
787 };
+
788 
+
793 } // CUB namespace
+
794 CUB_NS_POSTFIX // Optional outer namespace(s)
+
795 
796 
-
801 } // CUB namespace
-
802 CUB_NS_POSTFIX // Optional outer namespace(s)
-
803 
-
804 
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__reduce_8cuh.html b/docs/html/device__reduce_8cuh.html index f3f384a9e6..621d856088 100644 --- a/docs/html/device__reduce_8cuh.html +++ b/docs/html/device__reduce_8cuh.html @@ -106,6 +106,7 @@
#include <stdio.h>
#include <iterator>
+#include <limits>
#include "dispatch/dispatch_reduce.cuh"
#include "dispatch/dispatch_reduce_by_key.cuh"
#include "../util_namespace.cuh"
@@ -137,7 +138,7 @@
-Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__reduce_8cuh_source.html b/docs/html/device__reduce_8cuh_source.html index 94f5ef601f..e6c2a32cb7 100644 --- a/docs/html/device__reduce_8cuh_source.html +++ b/docs/html/device__reduce_8cuh_source.html @@ -134,266 +134,256 @@
36 
37 #include <stdio.h>
38 #include <iterator>
-
39 
-
40 #include "dispatch/dispatch_reduce.cuh"
-
41 #include "dispatch/dispatch_reduce_by_key.cuh"
-
42 #include "../util_namespace.cuh"
-
43 
-
45 CUB_NS_PREFIX
-
46 
-
48 namespace cub {
-
49 
+
39 #include <limits>
+
40 
+
41 #include "dispatch/dispatch_reduce.cuh"
+
42 #include "dispatch/dispatch_reduce_by_key.cuh"
+
43 #include "../util_namespace.cuh"
+
44 
+
46 CUB_NS_PREFIX
+
47 
+
49 namespace cub {
50 
-
82 struct DeviceReduce
-
83 {
-
137  template <
-
138  typename InputIteratorT,
-
139  typename OutputIteratorT,
-
140  typename ReductionOp>
+
51 
+
83 struct DeviceReduce
+
84 {
+
136  template <
+
137  typename InputIteratorT,
+
138  typename OutputIteratorT,
+
139  typename ReductionOpT,
+
140  typename T>
141  CUB_RUNTIME_FUNCTION
-
142  static cudaError_t Reduce(
-
143  void* d_temp_storage,
+
142  static cudaError_t Reduce(
+
143  void *d_temp_storage,
144  size_t &temp_storage_bytes,
145  InputIteratorT d_in,
146  OutputIteratorT d_out,
147  int num_items,
-
148  ReductionOp reduction_op,
-
149  cudaStream_t stream = 0,
-
150  bool debug_synchronous = false)
-
151  {
-
152  // Signed integer type for global offsets
-
153  typedef int OffsetT;
+
148  ReductionOpT reduction_op,
+
149  T init,
+
150  cudaStream_t stream = 0,
+
151  bool debug_synchronous = false)
+
152  {
+
153  typedef int OffsetT; // Signed integer type for global offsets
154 
-
155  // Dispatch type
-
156  typedef DispatchReduce<InputIteratorT, OutputIteratorT, OffsetT, ReductionOp> DispatchReduce;
-
157 
-
158  return DispatchReduce::Dispatch(
-
159  d_temp_storage,
-
160  temp_storage_bytes,
-
161  d_in,
-
162  d_out,
-
163  num_items,
-
164  reduction_op,
-
165  stream,
-
166  debug_synchronous);
-
167  }
-
168 
-
169 
-
215  template <
-
216  typename InputIteratorT,
-
217  typename OutputIteratorT>
-
218  CUB_RUNTIME_FUNCTION
-
219  static cudaError_t Sum(
-
220  void* d_temp_storage,
-
221  size_t &temp_storage_bytes,
-
222  InputIteratorT d_in,
-
223  OutputIteratorT d_out,
-
224  int num_items,
-
225  cudaStream_t stream = 0,
-
226  bool debug_synchronous = false)
-
227  {
-
228  // Signed integer type for global offsets
-
229  typedef int OffsetT;
-
230 
-
231  // Dispatch type
-
232  typedef DispatchReduce<InputIteratorT, OutputIteratorT, OffsetT, cub::Sum> DispatchReduce;
-
233 
-
234  return DispatchReduce::Dispatch(
-
235  d_temp_storage,
-
236  temp_storage_bytes,
-
237  d_in,
-
238  d_out,
-
239  num_items,
-
240  cub::Sum(),
-
241  stream,
-
242  debug_synchronous);
-
243  }
-
244 
-
245 
-
287  template <
-
288  typename InputIteratorT,
-
289  typename OutputIteratorT>
-
290  CUB_RUNTIME_FUNCTION
-
291  static cudaError_t Min(
-
292  void* d_temp_storage,
-
293  size_t &temp_storage_bytes,
-
294  InputIteratorT d_in,
-
295  OutputIteratorT d_out,
-
296  int num_items,
-
297  cudaStream_t stream = 0,
-
298  bool debug_synchronous = false)
-
299  {
-
300  // Signed integer type for global offsets
-
301  typedef int OffsetT;
-
302 
-
303  // Dispatch type
-
304  typedef DispatchReduce<InputIteratorT, OutputIteratorT, OffsetT, cub::Min> DispatchReduce;
-
305 
-
306  return DispatchReduce::Dispatch(
-
307  d_temp_storage,
-
308  temp_storage_bytes,
-
309  d_in,
-
310  d_out,
-
311  num_items,
-
312  cub::Min(),
-
313  stream,
-
314  debug_synchronous);
-
315  }
-
316 
-
317 
-
364  template <
-
365  typename InputIteratorT,
-
366  typename OutputIteratorT>
-
367  CUB_RUNTIME_FUNCTION
-
368  static cudaError_t ArgMin(
-
369  void* d_temp_storage,
-
370  size_t &temp_storage_bytes,
-
371  InputIteratorT d_in,
-
372  OutputIteratorT d_out,
-
373  int num_items,
-
374  cudaStream_t stream = 0,
-
375  bool debug_synchronous = false)
-
376  {
-
377  // Signed integer type for global offsets
-
378  typedef int OffsetT;
-
379 
-
380  // Wrapped input iterator
-
381  typedef ArgIndexInputIterator<InputIteratorT, int> ArgIndexInputIteratorT;
-
382  ArgIndexInputIteratorT d_argmin_in(d_in, 0);
-
383 
-
384  // Dispatch type
-
385  typedef DispatchReduce<ArgIndexInputIteratorT, OutputIteratorT, OffsetT, cub::ArgMin> DispatchReduce;
-
386 
-
387  return DispatchReduce::Dispatch(
-
388  d_temp_storage,
-
389  temp_storage_bytes,
-
390  d_argmin_in,
-
391  d_out,
-
392  num_items,
-
393  cub::ArgMin(),
-
394  stream,
-
395  debug_synchronous);
-
396  }
-
397 
-
398 
-
440  template <
-
441  typename InputIteratorT,
-
442  typename OutputIteratorT>
-
443  CUB_RUNTIME_FUNCTION
-
444  static cudaError_t Max(
-
445  void* d_temp_storage,
-
446  size_t &temp_storage_bytes,
-
447  InputIteratorT d_in,
-
448  OutputIteratorT d_out,
-
449  int num_items,
-
450  cudaStream_t stream = 0,
-
451  bool debug_synchronous = false)
-
452  {
-
453  // Signed integer type for global offsets
-
454  typedef int OffsetT;
-
455 
-
456  // Dispatch type
-
457  typedef DispatchReduce<InputIteratorT, OutputIteratorT, OffsetT, cub::Max> DispatchReduce;
-
458 
-
459  return DispatchReduce::Dispatch(
-
460  d_temp_storage,
-
461  temp_storage_bytes,
-
462  d_in,
-
463  d_out,
-
464  num_items,
-
465  cub::Max(),
-
466  stream,
-
467  debug_synchronous);
-
468  }
-
469 
-
470 
-
517  template <
-
518  typename InputIteratorT,
-
519  typename OutputIteratorT>
-
520  CUB_RUNTIME_FUNCTION
-
521  static cudaError_t ArgMax(
-
522  void* d_temp_storage,
-
523  size_t &temp_storage_bytes,
-
524  InputIteratorT d_in,
-
525  OutputIteratorT d_out,
-
526  int num_items,
-
527  cudaStream_t stream = 0,
-
528  bool debug_synchronous = false)
-
529  {
-
530  // Signed integer type for global offsets
-
531  typedef int OffsetT;
-
532 
-
533  // Wrapped input iterator
-
534  typedef ArgIndexInputIterator<InputIteratorT, int> ArgIndexInputIteratorT;
-
535  ArgIndexInputIteratorT d_argmax_in(d_in, 0);
-
536 
-
537  // Dispatch type
-
538  typedef DispatchReduce<ArgIndexInputIteratorT, OutputIteratorT, OffsetT, cub::ArgMax> DispatchReduce;
-
539 
-
540  return DispatchReduce::Dispatch(
-
541  d_temp_storage,
-
542  temp_storage_bytes,
-
543  d_argmax_in,
-
544  d_out,
-
545  num_items,
-
546  cub::ArgMax(),
-
547  stream,
-
548  debug_synchronous);
-
549  }
-
550 
-
551 
-
634  template <
-
635  typename KeysInputIteratorT,
-
636  typename UniqueOutputIteratorT,
-
637  typename ValuesInputIteratorT,
-
638  typename AggregatesOutputIteratorT,
-
639  typename NumRunsOutputIteratorT,
-
640  typename ReductionOp>
-
641  CUB_RUNTIME_FUNCTION __forceinline__
-
642  static cudaError_t ReduceByKey(
-
643  void* d_temp_storage,
-
644  size_t &temp_storage_bytes,
-
645  KeysInputIteratorT d_keys_in,
-
646  UniqueOutputIteratorT d_unique_out,
-
647  ValuesInputIteratorT d_values_in,
-
648  AggregatesOutputIteratorT d_aggregates_out,
-
649  NumRunsOutputIteratorT d_num_runs_out,
-
650  ReductionOp reduction_op,
-
651  int num_items,
-
652  cudaStream_t stream = 0,
-
653  bool debug_synchronous = false)
-
654  {
-
655  typedef int OffsetT; // Signed integer type for global offsets
-
656  typedef NullType* FlagIterator; // FlagT iterator type (not used)
-
657  typedef NullType SelectOp; // Selection op (not used)
-
658  typedef Equality EqualityOp; // Default == operator
-
659 
-
660  return DispatchReduceByKey<KeysInputIteratorT, UniqueOutputIteratorT, ValuesInputIteratorT, AggregatesOutputIteratorT, NumRunsOutputIteratorT, EqualityOp, ReductionOp, OffsetT>::Dispatch(
-
661  d_temp_storage,
-
662  temp_storage_bytes,
-
663  d_keys_in,
-
664  d_unique_out,
-
665  d_values_in,
-
666  d_aggregates_out,
-
667  d_num_runs_out,
-
668  EqualityOp(),
-
669  reduction_op,
-
670  num_items,
-
671  stream,
-
672  debug_synchronous);
-
673  }
-
674 
-
675 };
-
676 
-
681 } // CUB namespace
-
682 CUB_NS_POSTFIX // Optional outer namespace(s)
-
683 
-
684 
+
155  return DispatchReduce<InputIteratorT, OutputIteratorT, OffsetT, ReductionOpT>::Dispatch(
+
156  d_temp_storage,
+
157  temp_storage_bytes,
+
158  d_in,
+
159  d_out,
+
160  num_items,
+
161  reduction_op,
+
162  init,
+
163  stream,
+
164  debug_synchronous);
+
165  }
+
166 
+
167 
+
213  template <
+
214  typename InputIteratorT,
+
215  typename OutputIteratorT>
+
216  CUB_RUNTIME_FUNCTION
+
217  static cudaError_t Sum(
+
218  void *d_temp_storage,
+
219  size_t &temp_storage_bytes,
+
220  InputIteratorT d_in,
+
221  OutputIteratorT d_out,
+
222  int num_items,
+
223  cudaStream_t stream = 0,
+
224  bool debug_synchronous = false)
+
225  {
+
226  typedef int OffsetT; // Signed integer type for global offsets
+
227  typedef typename std::iterator_traits<InputIteratorT>::value_type T; // Data element type
+
228 
+
229  return DispatchReduce<InputIteratorT, OutputIteratorT, OffsetT, cub::Sum>::Dispatch(
+
230  d_temp_storage,
+
231  temp_storage_bytes,
+
232  d_in,
+
233  d_out,
+
234  num_items,
+
235  cub::Sum(),
+
236  T(), // zero-initialize
+
237  stream,
+
238  debug_synchronous);
+
239  }
+
240 
+
241 
+
280  template <
+
281  typename InputIteratorT,
+
282  typename OutputIteratorT>
+
283  CUB_RUNTIME_FUNCTION
+
284  static cudaError_t Min(
+
285  void *d_temp_storage,
+
286  size_t &temp_storage_bytes,
+
287  InputIteratorT d_in,
+
288  OutputIteratorT d_out,
+
289  int num_items,
+
290  cudaStream_t stream = 0,
+
291  bool debug_synchronous = false)
+
292  {
+
293  typedef int OffsetT; // Signed integer type for global offsets
+
294  typedef typename std::iterator_traits<InputIteratorT>::value_type T; // Data element type
+
295 
+
296  return DispatchReduce<InputIteratorT, OutputIteratorT, OffsetT, cub::Min>::Dispatch(
+
297  d_temp_storage,
+
298  temp_storage_bytes,
+
299  d_in,
+
300  d_out,
+
301  num_items,
+
302  cub::Min(),
+
303  Traits<T>::Max(), // replace with std::numeric_limits<T>::max() when C++11 support is more prevalent
+
304  stream,
+
305  debug_synchronous);
+
306  }
+
307 
+
308 
+
349  template <
+
350  typename InputIteratorT,
+
351  typename OutputIteratorT>
+
352  CUB_RUNTIME_FUNCTION
+
353  static cudaError_t ArgMin(
+
354  void *d_temp_storage,
+
355  size_t &temp_storage_bytes,
+
356  InputIteratorT d_in,
+
357  OutputIteratorT d_out,
+
358  int num_items,
+
359  cudaStream_t stream = 0,
+
360  bool debug_synchronous = false)
+
361  {
+
362  typedef int OffsetT; // Signed integer type for global offsets
+
363  typedef typename std::iterator_traits<InputIteratorT>::value_type T; // Data element type
+
364  typedef ArgIndexInputIterator<InputIteratorT, int> ArgIndexInputIteratorT; // Wrapped input iterator type
+
365 
+
366  ArgIndexInputIteratorT d_argmin_in(d_in);
+
367  KeyValuePair<OffsetT, T> init = {1, Traits<T>::Max()}; // replace with std::numeric_limits<T>::max() when C++11 support is more prevalent
+
368 
+
369  return DispatchReduce<ArgIndexInputIteratorT, OutputIteratorT, OffsetT, cub::ArgMin>::Dispatch(
+
370  d_temp_storage,
+
371  temp_storage_bytes,
+
372  d_argmin_in,
+
373  d_out,
+
374  num_items,
+
375  cub::ArgMin(),
+
376  init,
+
377  stream,
+
378  debug_synchronous);
+
379  }
+
380 
+
381 
+
420  template <
+
421  typename InputIteratorT,
+
422  typename OutputIteratorT>
+
423  CUB_RUNTIME_FUNCTION
+
424  static cudaError_t Max(
+
425  void *d_temp_storage,
+
426  size_t &temp_storage_bytes,
+
427  InputIteratorT d_in,
+
428  OutputIteratorT d_out,
+
429  int num_items,
+
430  cudaStream_t stream = 0,
+
431  bool debug_synchronous = false)
+
432  {
+
433  typedef int OffsetT; // Signed integer type for global offsets
+
434  typedef typename std::iterator_traits<InputIteratorT>::value_type T; // Data element type
+
435 
+
436  return DispatchReduce<InputIteratorT, OutputIteratorT, OffsetT, cub::Max>::Dispatch(
+
437  d_temp_storage,
+
438  temp_storage_bytes,
+
439  d_in,
+
440  d_out,
+
441  num_items,
+
442  cub::Max(),
+
443  Traits<T>::Lowest(), // replace with std::numeric_limits<T>::lowest() when C++11 support is more prevalent
+
444  stream,
+
445  debug_synchronous);
+
446  }
+
447 
+
448 
+
489  template <
+
490  typename InputIteratorT,
+
491  typename OutputIteratorT>
+
492  CUB_RUNTIME_FUNCTION
+
493  static cudaError_t ArgMax(
+
494  void *d_temp_storage,
+
495  size_t &temp_storage_bytes,
+
496  InputIteratorT d_in,
+
497  OutputIteratorT d_out,
+
498  int num_items,
+
499  cudaStream_t stream = 0,
+
500  bool debug_synchronous = false)
+
501  {
+
502  typedef int OffsetT; // Signed integer type for global offsets
+
503  typedef typename std::iterator_traits<InputIteratorT>::value_type T; // Data element type
+
504  typedef ArgIndexInputIterator<InputIteratorT, int> ArgIndexInputIteratorT; // Wrapped input iterator
+
505 
+
506  ArgIndexInputIteratorT d_argmax_in(d_in);
+
507  KeyValuePair<OffsetT, T> init = {1, Traits<T>::Lowest()}; // replace with std::numeric_limits<T>::lowest() when C++11 support is more prevalent
+
508 
+
509  return DispatchReduce<ArgIndexInputIteratorT, OutputIteratorT, OffsetT, cub::ArgMax>::Dispatch(
+
510  d_temp_storage,
+
511  temp_storage_bytes,
+
512  d_argmax_in,
+
513  d_out,
+
514  num_items,
+
515  cub::ArgMax(),
+
516  init,
+
517  stream,
+
518  debug_synchronous);
+
519  }
+
520 
+
521 
+
603  template <
+
604  typename KeysInputIteratorT,
+
605  typename UniqueOutputIteratorT,
+
606  typename ValuesInputIteratorT,
+
607  typename AggregatesOutputIteratorT,
+
608  typename NumRunsOutputIteratorT,
+
609  typename ReductionOpT>
+
610  CUB_RUNTIME_FUNCTION __forceinline__
+
611  static cudaError_t ReduceByKey(
+
612  void *d_temp_storage,
+
613  size_t &temp_storage_bytes,
+
614  KeysInputIteratorT d_keys_in,
+
615  UniqueOutputIteratorT d_unique_out,
+
616  ValuesInputIteratorT d_values_in,
+
617  AggregatesOutputIteratorT d_aggregates_out,
+
618  NumRunsOutputIteratorT d_num_runs_out,
+
619  ReductionOpT reduction_op,
+
620  int num_items,
+
621  cudaStream_t stream = 0,
+
622  bool debug_synchronous = false)
+
623  {
+
624  typedef int OffsetT; // Signed integer type for global offsets
+
625  typedef NullType* FlagIterator; // FlagT iterator type (not used)
+
626  typedef NullType SelectOp; // Selection op (not used)
+
627  typedef Equality EqualityOp; // Default == operator
+
628 
+
629  return DispatchReduceByKey<KeysInputIteratorT, UniqueOutputIteratorT, ValuesInputIteratorT, AggregatesOutputIteratorT, NumRunsOutputIteratorT, EqualityOp, ReductionOpT, OffsetT>::Dispatch(
+
630  d_temp_storage,
+
631  temp_storage_bytes,
+
632  d_keys_in,
+
633  d_unique_out,
+
634  d_values_in,
+
635  d_aggregates_out,
+
636  d_num_runs_out,
+
637  EqualityOp(),
+
638  reduction_op,
+
639  num_items,
+
640  stream,
+
641  debug_synchronous);
+
642  }
+
643 
+
644 };
+
645 
+
650 } // CUB namespace
+
651 CUB_NS_POSTFIX // Optional outer namespace(s)
+
652 
+
653 
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__run__length__encode_8cuh.html b/docs/html/device__run__length__encode_8cuh.html index 1bd7ef3344..b45c9335cf 100644 --- a/docs/html/device__run__length__encode_8cuh.html +++ b/docs/html/device__run__length__encode_8cuh.html @@ -137,7 +137,7 @@
-Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__run__length__encode_8cuh_source.html b/docs/html/device__run__length__encode_8cuh_source.html index 3e79613958..1cd4f8eda7 100644 --- a/docs/html/device__run__length__encode_8cuh_source.html +++ b/docs/html/device__run__length__encode_8cuh_source.html @@ -147,100 +147,100 @@
78 struct DeviceRunLengthEncode
79 {
80 
-
143  template <
-
144  typename InputIteratorT,
-
145  typename UniqueOutputIteratorT,
-
146  typename LengthsOutputIteratorT,
-
147  typename NumRunsOutputIteratorT>
-
148  CUB_RUNTIME_FUNCTION __forceinline__
-
149  static cudaError_t Encode(
-
150  void* d_temp_storage,
-
151  size_t &temp_storage_bytes,
-
152  InputIteratorT d_in,
-
153  UniqueOutputIteratorT d_unique_out,
-
154  LengthsOutputIteratorT d_counts_out,
-
155  NumRunsOutputIteratorT d_num_runs_out,
-
156  int num_items,
-
157  cudaStream_t stream = 0,
-
158  bool debug_synchronous = false)
-
159  {
-
160  // Data type of value iterator
-
161  typedef typename std::iterator_traits<LengthsOutputIteratorT>::value_type Value;
-
162 
-
163  typedef int OffsetT; // Signed integer type for global offsets
-
164  typedef NullType* FlagIterator; // FlagT iterator type (not used)
-
165  typedef NullType SelectOp; // Selection op (not used)
-
166  typedef Equality EqualityOp; // Default == operator
-
167  typedef cub::Sum ReductionOp; // Value reduction operator
-
168 
-
169  // Generator type for providing 1s values for run-length reduction
-
170  typedef ConstantInputIterator<Value, OffsetT> LengthsInputIteratorT;
-
171 
-
172  Value one_val;
-
173  one_val = 1;
-
174 
-
175  return DispatchReduceByKey<InputIteratorT, UniqueOutputIteratorT, LengthsInputIteratorT, LengthsOutputIteratorT, NumRunsOutputIteratorT, EqualityOp, ReductionOp, OffsetT>::Dispatch(
-
176  d_temp_storage,
-
177  temp_storage_bytes,
-
178  d_in,
-
179  d_unique_out,
-
180  LengthsInputIteratorT(one_val),
-
181  d_counts_out,
-
182  d_num_runs_out,
-
183  EqualityOp(),
-
184  ReductionOp(),
-
185  num_items,
-
186  stream,
-
187  debug_synchronous);
-
188  }
+
142  template <
+
143  typename InputIteratorT,
+
144  typename UniqueOutputIteratorT,
+
145  typename LengthsOutputIteratorT,
+
146  typename NumRunsOutputIteratorT>
+
147  CUB_RUNTIME_FUNCTION __forceinline__
+
148  static cudaError_t Encode(
+
149  void* d_temp_storage,
+
150  size_t &temp_storage_bytes,
+
151  InputIteratorT d_in,
+
152  UniqueOutputIteratorT d_unique_out,
+
153  LengthsOutputIteratorT d_counts_out,
+
154  NumRunsOutputIteratorT d_num_runs_out,
+
155  int num_items,
+
156  cudaStream_t stream = 0,
+
157  bool debug_synchronous = false)
+
158  {
+
159  // Data type of value iterator
+
160  typedef typename std::iterator_traits<LengthsOutputIteratorT>::value_type Value;
+
161 
+
162  typedef int OffsetT; // Signed integer type for global offsets
+
163  typedef NullType* FlagIterator; // FlagT iterator type (not used)
+
164  typedef NullType SelectOp; // Selection op (not used)
+
165  typedef Equality EqualityOp; // Default == operator
+
166  typedef cub::Sum ReductionOp; // Value reduction operator
+
167 
+
168  // Generator type for providing 1s values for run-length reduction
+
169  typedef ConstantInputIterator<Value, OffsetT> LengthsInputIteratorT;
+
170 
+
171  Value one_val;
+
172  one_val = 1;
+
173 
+
174  return DispatchReduceByKey<InputIteratorT, UniqueOutputIteratorT, LengthsInputIteratorT, LengthsOutputIteratorT, NumRunsOutputIteratorT, EqualityOp, ReductionOp, OffsetT>::Dispatch(
+
175  d_temp_storage,
+
176  temp_storage_bytes,
+
177  d_in,
+
178  d_unique_out,
+
179  LengthsInputIteratorT(one_val),
+
180  d_counts_out,
+
181  d_num_runs_out,
+
182  EqualityOp(),
+
183  ReductionOp(),
+
184  num_items,
+
185  stream,
+
186  debug_synchronous);
+
187  }
+
188 
189 
-
190 
-
241  template <
-
242  typename InputIteratorT,
-
243  typename OffsetsOutputIteratorT,
-
244  typename LengthsOutputIteratorT,
-
245  typename NumRunsOutputIteratorT>
-
246  CUB_RUNTIME_FUNCTION __forceinline__
-
247  static cudaError_t NonTrivialRuns(
-
248  void* d_temp_storage,
-
249  size_t &temp_storage_bytes,
-
250  InputIteratorT d_in,
-
251  OffsetsOutputIteratorT d_offsets_out,
-
252  LengthsOutputIteratorT d_lengths_out,
-
253  NumRunsOutputIteratorT d_num_runs_out,
-
254  int num_items,
-
255  cudaStream_t stream = 0,
-
256  bool debug_synchronous = false)
-
257  {
-
258  typedef int OffsetT; // Signed integer type for global offsets
-
259  typedef Equality EqualityOp; // Default == operator
-
260 
-
261  return DeviceRleDispatch<InputIteratorT, OffsetsOutputIteratorT, LengthsOutputIteratorT, NumRunsOutputIteratorT, EqualityOp, OffsetT>::Dispatch(
-
262  d_temp_storage,
-
263  temp_storage_bytes,
-
264  d_in,
-
265  d_offsets_out,
-
266  d_lengths_out,
-
267  d_num_runs_out,
-
268  EqualityOp(),
-
269  num_items,
-
270  stream,
-
271  debug_synchronous);
-
272  }
-
273 
+
239  template <
+
240  typename InputIteratorT,
+
241  typename OffsetsOutputIteratorT,
+
242  typename LengthsOutputIteratorT,
+
243  typename NumRunsOutputIteratorT>
+
244  CUB_RUNTIME_FUNCTION __forceinline__
+
245  static cudaError_t NonTrivialRuns(
+
246  void* d_temp_storage,
+
247  size_t &temp_storage_bytes,
+
248  InputIteratorT d_in,
+
249  OffsetsOutputIteratorT d_offsets_out,
+
250  LengthsOutputIteratorT d_lengths_out,
+
251  NumRunsOutputIteratorT d_num_runs_out,
+
252  int num_items,
+
253  cudaStream_t stream = 0,
+
254  bool debug_synchronous = false)
+
255  {
+
256  typedef int OffsetT; // Signed integer type for global offsets
+
257  typedef Equality EqualityOp; // Default == operator
+
258 
+
259  return DeviceRleDispatch<InputIteratorT, OffsetsOutputIteratorT, LengthsOutputIteratorT, NumRunsOutputIteratorT, EqualityOp, OffsetT>::Dispatch(
+
260  d_temp_storage,
+
261  temp_storage_bytes,
+
262  d_in,
+
263  d_offsets_out,
+
264  d_lengths_out,
+
265  d_num_runs_out,
+
266  EqualityOp(),
+
267  num_items,
+
268  stream,
+
269  debug_synchronous);
+
270  }
+
271 
+
272 
+
273 };
274 
-
275 };
-
276 
-
277 
-
278 } // CUB namespace
-
279 CUB_NS_POSTFIX // Optional outer namespace(s)
-
280 
-
281 
+
275 
+
276 } // CUB namespace
+
277 CUB_NS_POSTFIX // Optional outer namespace(s)
+
278 
+
279 
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__scan_8cuh.html b/docs/html/device__scan_8cuh.html index f9ca7bc126..0aa8676a39 100644 --- a/docs/html/device__scan_8cuh.html +++ b/docs/html/device__scan_8cuh.html @@ -136,7 +136,7 @@
-Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__scan_8cuh_source.html b/docs/html/device__scan_8cuh_source.html index d661a0f0d5..6ab23988fd 100644 --- a/docs/html/device__scan_8cuh_source.html +++ b/docs/html/device__scan_8cuh_source.html @@ -147,145 +147,145 @@
78 {
79  /******************************************************************/
83 
-
129  template <
-
130  typename InputIteratorT,
-
131  typename OutputIteratorT>
-
132  CUB_RUNTIME_FUNCTION
-
133  static cudaError_t ExclusiveSum(
-
134  void *d_temp_storage,
-
135  size_t &temp_storage_bytes,
-
136  InputIteratorT d_in,
-
137  OutputIteratorT d_out,
-
138  int num_items,
-
139  cudaStream_t stream = 0,
-
140  bool debug_synchronous = false)
-
141  {
-
142  // Signed integer type for global offsets
-
143  typedef int OffsetT;
-
144 
-
145  // Scan data type
-
146  typedef typename std::iterator_traits<InputIteratorT>::value_type T;
-
147 
-
148  return DispatchScan<InputIteratorT, OutputIteratorT, Sum, T, OffsetT>::Dispatch(
-
149  d_temp_storage,
-
150  temp_storage_bytes,
-
151  d_in,
-
152  d_out,
-
153  Sum(),
-
154  T(),
-
155  num_items,
-
156  stream,
-
157  debug_synchronous);
-
158  }
+
128  template <
+
129  typename InputIteratorT,
+
130  typename OutputIteratorT>
+
131  CUB_RUNTIME_FUNCTION
+
132  static cudaError_t ExclusiveSum(
+
133  void *d_temp_storage,
+
134  size_t &temp_storage_bytes,
+
135  InputIteratorT d_in,
+
136  OutputIteratorT d_out,
+
137  int num_items,
+
138  cudaStream_t stream = 0,
+
139  bool debug_synchronous = false)
+
140  {
+
141  // Signed integer type for global offsets
+
142  typedef int OffsetT;
+
143 
+
144  // Scan data type
+
145  typedef typename std::iterator_traits<InputIteratorT>::value_type T;
+
146 
+
147  return DispatchScan<InputIteratorT, OutputIteratorT, Sum, T, OffsetT>::Dispatch(
+
148  d_temp_storage,
+
149  temp_storage_bytes,
+
150  d_in,
+
151  d_out,
+
152  Sum(),
+
153  T(),
+
154  num_items,
+
155  stream,
+
156  debug_synchronous);
+
157  }
+
158 
159 
-
160 
-
215  template <
-
216  typename InputIteratorT,
-
217  typename OutputIteratorT,
-
218  typename ScanOp,
-
219  typename Identity>
-
220  CUB_RUNTIME_FUNCTION
-
221  static cudaError_t ExclusiveScan(
-
222  void *d_temp_storage,
-
223  size_t &temp_storage_bytes,
-
224  InputIteratorT d_in,
-
225  OutputIteratorT d_out,
-
226  ScanOp scan_op,
-
227  Identity identity,
-
228  int num_items,
-
229  cudaStream_t stream = 0,
-
230  bool debug_synchronous = false)
-
231  {
-
232  // Signed integer type for global offsets
-
233  typedef int OffsetT;
-
234 
-
235  return DispatchScan<InputIteratorT, OutputIteratorT, ScanOp, Identity, OffsetT>::Dispatch(
-
236  d_temp_storage,
-
237  temp_storage_bytes,
-
238  d_in,
-
239  d_out,
-
240  scan_op,
-
241  identity,
-
242  num_items,
-
243  stream,
-
244  debug_synchronous);
-
245  }
-
246 
-
247 
-
249  /******************************************************************/
-
253 
-
254 
-
296  template <
-
297  typename InputIteratorT,
-
298  typename OutputIteratorT>
-
299  CUB_RUNTIME_FUNCTION
-
300  static cudaError_t InclusiveSum(
-
301  void* d_temp_storage,
-
302  size_t& temp_storage_bytes,
-
303  InputIteratorT d_in,
-
304  OutputIteratorT d_out,
-
305  int num_items,
-
306  cudaStream_t stream = 0,
-
307  bool debug_synchronous = false)
-
308  {
-
309  // Signed integer type for global offsets
-
310  typedef int OffsetT;
-
311 
-
312  return DispatchScan<InputIteratorT, OutputIteratorT, Sum, NullType, OffsetT>::Dispatch(
-
313  d_temp_storage,
-
314  temp_storage_bytes,
-
315  d_in,
-
316  d_out,
-
317  Sum(),
-
318  NullType(),
-
319  num_items,
-
320  stream,
-
321  debug_synchronous);
-
322  }
-
323 
-
324 
-
378  template <
-
379  typename InputIteratorT,
-
380  typename OutputIteratorT,
-
381  typename ScanOp>
-
382  CUB_RUNTIME_FUNCTION
-
383  static cudaError_t InclusiveScan(
-
384  void *d_temp_storage,
-
385  size_t &temp_storage_bytes,
-
386  InputIteratorT d_in,
-
387  OutputIteratorT d_out,
-
388  ScanOp scan_op,
-
389  int num_items,
-
390  cudaStream_t stream = 0,
-
391  bool debug_synchronous = false)
-
392  {
-
393  // Signed integer type for global offsets
-
394  typedef int OffsetT;
-
395 
-
396  return DispatchScan<InputIteratorT, OutputIteratorT, ScanOp, NullType, OffsetT>::Dispatch(
-
397  d_temp_storage,
-
398  temp_storage_bytes,
-
399  d_in,
-
400  d_out,
-
401  scan_op,
-
402  NullType(),
-
403  num_items,
-
404  stream,
-
405  debug_synchronous);
-
406  }
-
407 
-
409 
-
410 };
-
411 
-
416 } // CUB namespace
-
417 CUB_NS_POSTFIX // Optional outer namespace(s)
-
418 
-
419 
+
210  template <
+
211  typename InputIteratorT,
+
212  typename OutputIteratorT,
+
213  typename ScanOp,
+
214  typename Identity>
+
215  CUB_RUNTIME_FUNCTION
+
216  static cudaError_t ExclusiveScan(
+
217  void *d_temp_storage,
+
218  size_t &temp_storage_bytes,
+
219  InputIteratorT d_in,
+
220  OutputIteratorT d_out,
+
221  ScanOp scan_op,
+
222  Identity identity,
+
223  int num_items,
+
224  cudaStream_t stream = 0,
+
225  bool debug_synchronous = false)
+
226  {
+
227  // Signed integer type for global offsets
+
228  typedef int OffsetT;
+
229 
+
230  return DispatchScan<InputIteratorT, OutputIteratorT, ScanOp, Identity, OffsetT>::Dispatch(
+
231  d_temp_storage,
+
232  temp_storage_bytes,
+
233  d_in,
+
234  d_out,
+
235  scan_op,
+
236  identity,
+
237  num_items,
+
238  stream,
+
239  debug_synchronous);
+
240  }
+
241 
+
242 
+
244  /******************************************************************/
+
248 
+
249 
+
287  template <
+
288  typename InputIteratorT,
+
289  typename OutputIteratorT>
+
290  CUB_RUNTIME_FUNCTION
+
291  static cudaError_t InclusiveSum(
+
292  void* d_temp_storage,
+
293  size_t& temp_storage_bytes,
+
294  InputIteratorT d_in,
+
295  OutputIteratorT d_out,
+
296  int num_items,
+
297  cudaStream_t stream = 0,
+
298  bool debug_synchronous = false)
+
299  {
+
300  // Signed integer type for global offsets
+
301  typedef int OffsetT;
+
302 
+
303  return DispatchScan<InputIteratorT, OutputIteratorT, Sum, NullType, OffsetT>::Dispatch(
+
304  d_temp_storage,
+
305  temp_storage_bytes,
+
306  d_in,
+
307  d_out,
+
308  Sum(),
+
309  NullType(),
+
310  num_items,
+
311  stream,
+
312  debug_synchronous);
+
313  }
+
314 
+
315 
+
365  template <
+
366  typename InputIteratorT,
+
367  typename OutputIteratorT,
+
368  typename ScanOp>
+
369  CUB_RUNTIME_FUNCTION
+
370  static cudaError_t InclusiveScan(
+
371  void *d_temp_storage,
+
372  size_t &temp_storage_bytes,
+
373  InputIteratorT d_in,
+
374  OutputIteratorT d_out,
+
375  ScanOp scan_op,
+
376  int num_items,
+
377  cudaStream_t stream = 0,
+
378  bool debug_synchronous = false)
+
379  {
+
380  // Signed integer type for global offsets
+
381  typedef int OffsetT;
+
382 
+
383  return DispatchScan<InputIteratorT, OutputIteratorT, ScanOp, NullType, OffsetT>::Dispatch(
+
384  d_temp_storage,
+
385  temp_storage_bytes,
+
386  d_in,
+
387  d_out,
+
388  scan_op,
+
389  NullType(),
+
390  num_items,
+
391  stream,
+
392  debug_synchronous);
+
393  }
+
394 
+
396 
+
397 };
+
398 
+
403 } // CUB namespace
+
404 CUB_NS_POSTFIX // Optional outer namespace(s)
+
405 
+
406 
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__segmented__radix__sort_8cuh.html b/docs/html/device__segmented__radix__sort_8cuh.html index f1d9bb0a04..c3d8fecb38 100644 --- a/docs/html/device__segmented__radix__sort_8cuh.html +++ b/docs/html/device__segmented__radix__sort_8cuh.html @@ -137,7 +137,7 @@
-Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__segmented__radix__sort_8cuh_source.html b/docs/html/device__segmented__radix__sort_8cuh_source.html index be1c36dff3..af9d3e3d98 100644 --- a/docs/html/device__segmented__radix__sort_8cuh_source.html +++ b/docs/html/device__segmented__radix__sort_8cuh_source.html @@ -151,337 +151,336 @@
82 
136  template <
137  typename KeyT,
-
138  typename ValueT,
-
139  typename OffsetT>
-
140  CUB_RUNTIME_FUNCTION
-
141  static cudaError_t SortPairs(
-
142  void *d_temp_storage,
-
143  size_t &temp_storage_bytes,
-
144  KeyT *d_keys_in,
-
145  KeyT *d_keys_out,
-
146  ValueT *d_values_in,
-
147  ValueT *d_values_out,
-
148  int num_items,
-
149  OffsetT num_segments,
-
150  OffsetT *d_begin_offsets,
-
151  OffsetT *d_end_offsets,
-
152  int begin_bit = 0,
-
153  int end_bit = sizeof(KeyT) * 8,
-
154  cudaStream_t stream = 0,
-
155  bool debug_synchronous = false)
-
156  {
-
157  // Signed integer type for global offsets
-
158  typedef int OffsetT;
-
159 
-
160  DoubleBuffer<KeyT> d_keys(d_keys_in, d_keys_out);
-
161  DoubleBuffer<ValueT> d_values(d_values_in, d_values_out);
-
162 
-
163  return DispatchSegmentedRadixSort<false, KeyT, ValueT, OffsetT>::Dispatch(
-
164  d_temp_storage,
-
165  temp_storage_bytes,
-
166  d_keys,
-
167  d_values,
-
168  num_items,
-
169  num_segments,
-
170  d_begin_offsets,
-
171  d_end_offsets,
-
172  begin_bit,
-
173  end_bit,
-
174  false,
-
175  stream,
-
176  debug_synchronous);
-
177  }
+
138  typename ValueT>
+
139  CUB_RUNTIME_FUNCTION
+
140  static cudaError_t SortPairs(
+
141  void *d_temp_storage,
+
142  size_t &temp_storage_bytes,
+
143  KeyT *d_keys_in,
+
144  KeyT *d_keys_out,
+
145  ValueT *d_values_in,
+
146  ValueT *d_values_out,
+
147  int num_items,
+
148  int num_segments,
+
149  int *d_begin_offsets,
+
150  int *d_end_offsets,
+
151  int begin_bit = 0,
+
152  int end_bit = sizeof(KeyT) * 8,
+
153  cudaStream_t stream = 0,
+
154  bool debug_synchronous = false)
+
155  {
+
156  // Signed integer type for global offsets
+
157  typedef int OffsetT;
+
158 
+
159  DoubleBuffer<KeyT> d_keys(d_keys_in, d_keys_out);
+
160  DoubleBuffer<ValueT> d_values(d_values_in, d_values_out);
+
161 
+
162  return DispatchSegmentedRadixSort<false, KeyT, ValueT, OffsetT>::Dispatch(
+
163  d_temp_storage,
+
164  temp_storage_bytes,
+
165  d_keys,
+
166  d_values,
+
167  num_items,
+
168  num_segments,
+
169  d_begin_offsets,
+
170  d_end_offsets,
+
171  begin_bit,
+
172  end_bit,
+
173  false,
+
174  stream,
+
175  debug_synchronous);
+
176  }
+
177 
178 
-
179 
-
244  template <
-
245  typename KeyT,
-
246  typename ValueT>
-
247  CUB_RUNTIME_FUNCTION
-
248  static cudaError_t SortPairs(
-
249  void *d_temp_storage,
-
250  size_t &temp_storage_bytes,
-
251  DoubleBuffer<KeyT> &d_keys,
-
252  DoubleBuffer<ValueT> &d_values,
-
253  int num_items,
-
254  int num_segments,
-
255  int *d_begin_offsets,
-
256  int *d_end_offsets,
-
257  int begin_bit = 0,
-
258  int end_bit = sizeof(KeyT) * 8,
-
259  cudaStream_t stream = 0,
-
260  bool debug_synchronous = false)
-
261  {
-
262  // Signed integer type for global offsets
-
263  typedef int OffsetT;
-
264 
-
265  return DispatchSegmentedRadixSort<false, KeyT, ValueT, OffsetT>::Dispatch(
-
266  d_temp_storage,
-
267  temp_storage_bytes,
-
268  d_keys,
-
269  d_values,
-
270  num_items,
-
271  num_segments,
-
272  d_begin_offsets,
-
273  d_end_offsets,
-
274  begin_bit,
-
275  end_bit,
-
276  true,
-
277  stream,
-
278  debug_synchronous);
-
279  }
+
243  template <
+
244  typename KeyT,
+
245  typename ValueT>
+
246  CUB_RUNTIME_FUNCTION
+
247  static cudaError_t SortPairs(
+
248  void *d_temp_storage,
+
249  size_t &temp_storage_bytes,
+
250  DoubleBuffer<KeyT> &d_keys,
+
251  DoubleBuffer<ValueT> &d_values,
+
252  int num_items,
+
253  int num_segments,
+
254  int *d_begin_offsets,
+
255  int *d_end_offsets,
+
256  int begin_bit = 0,
+
257  int end_bit = sizeof(KeyT) * 8,
+
258  cudaStream_t stream = 0,
+
259  bool debug_synchronous = false)
+
260  {
+
261  // Signed integer type for global offsets
+
262  typedef int OffsetT;
+
263 
+
264  return DispatchSegmentedRadixSort<false, KeyT, ValueT, OffsetT>::Dispatch(
+
265  d_temp_storage,
+
266  temp_storage_bytes,
+
267  d_keys,
+
268  d_values,
+
269  num_items,
+
270  num_segments,
+
271  d_begin_offsets,
+
272  d_end_offsets,
+
273  begin_bit,
+
274  end_bit,
+
275  true,
+
276  stream,
+
277  debug_synchronous);
+
278  }
+
279 
280 
-
281 
-
335  template <
-
336  typename KeyT,
-
337  typename ValueT>
-
338  CUB_RUNTIME_FUNCTION
-
339  static cudaError_t SortPairsDescending(
-
340  void *d_temp_storage,
-
341  size_t &temp_storage_bytes,
-
342  KeyT *d_keys_in,
-
343  KeyT *d_keys_out,
-
344  ValueT *d_values_in,
-
345  ValueT *d_values_out,
-
346  int num_items,
-
347  int num_segments,
-
348  int *d_begin_offsets,
-
349  int *d_end_offsets,
-
350  int begin_bit = 0,
-
351  int end_bit = sizeof(KeyT) * 8,
-
352  cudaStream_t stream = 0,
-
353  bool debug_synchronous = false)
-
354  {
-
355  // Signed integer type for global offsets
-
356  typedef int OffsetT;
-
357 
-
358  DoubleBuffer<KeyT> d_keys(d_keys_in, d_keys_out);
-
359  DoubleBuffer<ValueT> d_values(d_values_in, d_values_out);
-
360 
-
361  return DispatchSegmentedRadixSort<true, KeyT, ValueT, OffsetT>::Dispatch(
-
362  d_temp_storage,
-
363  temp_storage_bytes,
-
364  d_keys,
-
365  d_values,
-
366  num_items,
-
367  num_segments,
-
368  d_begin_offsets,
-
369  d_end_offsets,
-
370  begin_bit,
-
371  end_bit,
-
372  false,
-
373  stream,
-
374  debug_synchronous);
-
375  }
+
334  template <
+
335  typename KeyT,
+
336  typename ValueT>
+
337  CUB_RUNTIME_FUNCTION
+
338  static cudaError_t SortPairsDescending(
+
339  void *d_temp_storage,
+
340  size_t &temp_storage_bytes,
+
341  KeyT *d_keys_in,
+
342  KeyT *d_keys_out,
+
343  ValueT *d_values_in,
+
344  ValueT *d_values_out,
+
345  int num_items,
+
346  int num_segments,
+
347  int *d_begin_offsets,
+
348  int *d_end_offsets,
+
349  int begin_bit = 0,
+
350  int end_bit = sizeof(KeyT) * 8,
+
351  cudaStream_t stream = 0,
+
352  bool debug_synchronous = false)
+
353  {
+
354  // Signed integer type for global offsets
+
355  typedef int OffsetT;
+
356 
+
357  DoubleBuffer<KeyT> d_keys(d_keys_in, d_keys_out);
+
358  DoubleBuffer<ValueT> d_values(d_values_in, d_values_out);
+
359 
+
360  return DispatchSegmentedRadixSort<true, KeyT, ValueT, OffsetT>::Dispatch(
+
361  d_temp_storage,
+
362  temp_storage_bytes,
+
363  d_keys,
+
364  d_values,
+
365  num_items,
+
366  num_segments,
+
367  d_begin_offsets,
+
368  d_end_offsets,
+
369  begin_bit,
+
370  end_bit,
+
371  false,
+
372  stream,
+
373  debug_synchronous);
+
374  }
+
375 
376 
-
377 
-
442  template <
-
443  typename KeyT,
-
444  typename ValueT>
-
445  CUB_RUNTIME_FUNCTION
-
446  static cudaError_t SortPairsDescending(
-
447  void *d_temp_storage,
-
448  size_t &temp_storage_bytes,
-
449  DoubleBuffer<KeyT> &d_keys,
-
450  DoubleBuffer<ValueT> &d_values,
-
451  int num_items,
-
452  int num_segments,
-
453  int *d_begin_offsets,
-
454  int *d_end_offsets,
-
455  int begin_bit = 0,
-
456  int end_bit = sizeof(KeyT) * 8,
-
457  cudaStream_t stream = 0,
-
458  bool debug_synchronous = false)
-
459  {
-
460  // Signed integer type for global offsets
-
461  typedef int OffsetT;
-
462 
-
463  return DispatchSegmentedRadixSort<true, KeyT, ValueT, OffsetT>::Dispatch(
-
464  d_temp_storage,
-
465  temp_storage_bytes,
-
466  d_keys,
-
467  d_values,
-
468  num_items,
-
469  num_segments,
-
470  d_begin_offsets,
-
471  d_end_offsets,
-
472  begin_bit,
-
473  end_bit,
-
474  true,
-
475  stream,
-
476  debug_synchronous);
-
477  }
+
441  template <
+
442  typename KeyT,
+
443  typename ValueT>
+
444  CUB_RUNTIME_FUNCTION
+
445  static cudaError_t SortPairsDescending(
+
446  void *d_temp_storage,
+
447  size_t &temp_storage_bytes,
+
448  DoubleBuffer<KeyT> &d_keys,
+
449  DoubleBuffer<ValueT> &d_values,
+
450  int num_items,
+
451  int num_segments,
+
452  int *d_begin_offsets,
+
453  int *d_end_offsets,
+
454  int begin_bit = 0,
+
455  int end_bit = sizeof(KeyT) * 8,
+
456  cudaStream_t stream = 0,
+
457  bool debug_synchronous = false)
+
458  {
+
459  // Signed integer type for global offsets
+
460  typedef int OffsetT;
+
461 
+
462  return DispatchSegmentedRadixSort<true, KeyT, ValueT, OffsetT>::Dispatch(
+
463  d_temp_storage,
+
464  temp_storage_bytes,
+
465  d_keys,
+
466  d_values,
+
467  num_items,
+
468  num_segments,
+
469  d_begin_offsets,
+
470  d_end_offsets,
+
471  begin_bit,
+
472  end_bit,
+
473  true,
+
474  stream,
+
475  debug_synchronous);
+
476  }
+
477 
478 
-
479 
-
481  /******************************************************************/
+
480  /******************************************************************/
+
484 
485 
-
486 
-
533  template <typename KeyT>
-
534  CUB_RUNTIME_FUNCTION
-
535  static cudaError_t SortKeys(
-
536  void *d_temp_storage,
-
537  size_t &temp_storage_bytes,
-
538  KeyT *d_keys_in,
-
539  KeyT *d_keys_out,
-
540  int num_items,
-
541  int num_segments,
-
542  int *d_begin_offsets,
-
543  int *d_end_offsets,
-
544  int begin_bit = 0,
-
545  int end_bit = sizeof(KeyT) * 8,
-
546  cudaStream_t stream = 0,
-
547  bool debug_synchronous = false)
-
548  {
-
549  // Signed integer type for global offsets
-
550  typedef int OffsetT;
-
551 
-
552  // Null value type
-
553  DoubleBuffer<KeyT> d_keys(d_keys_in, d_keys_out);
-
554  DoubleBuffer<NullType> d_values;
-
555 
-
556  return DispatchSegmentedRadixSort<false, KeyT, NullType, OffsetT>::Dispatch(
-
557  d_temp_storage,
-
558  temp_storage_bytes,
-
559  d_keys,
-
560  d_values,
-
561  num_items,
-
562  num_segments,
-
563  d_begin_offsets,
-
564  d_end_offsets,
-
565  begin_bit,
-
566  end_bit,
-
567  false,
-
568  stream,
-
569  debug_synchronous);
-
570  }
+
532  template <typename KeyT>
+
533  CUB_RUNTIME_FUNCTION
+
534  static cudaError_t SortKeys(
+
535  void *d_temp_storage,
+
536  size_t &temp_storage_bytes,
+
537  KeyT *d_keys_in,
+
538  KeyT *d_keys_out,
+
539  int num_items,
+
540  int num_segments,
+
541  int *d_begin_offsets,
+
542  int *d_end_offsets,
+
543  int begin_bit = 0,
+
544  int end_bit = sizeof(KeyT) * 8,
+
545  cudaStream_t stream = 0,
+
546  bool debug_synchronous = false)
+
547  {
+
548  // Signed integer type for global offsets
+
549  typedef int OffsetT;
+
550 
+
551  // Null value type
+
552  DoubleBuffer<KeyT> d_keys(d_keys_in, d_keys_out);
+
553  DoubleBuffer<NullType> d_values;
+
554 
+
555  return DispatchSegmentedRadixSort<false, KeyT, NullType, OffsetT>::Dispatch(
+
556  d_temp_storage,
+
557  temp_storage_bytes,
+
558  d_keys,
+
559  d_values,
+
560  num_items,
+
561  num_segments,
+
562  d_begin_offsets,
+
563  d_end_offsets,
+
564  begin_bit,
+
565  end_bit,
+
566  false,
+
567  stream,
+
568  debug_synchronous);
+
569  }
+
570 
571 
-
572 
-
629  template <typename KeyT>
-
630  CUB_RUNTIME_FUNCTION
-
631  static cudaError_t SortKeys(
-
632  void *d_temp_storage,
-
633  size_t &temp_storage_bytes,
-
634  DoubleBuffer<KeyT> &d_keys,
-
635  int num_items,
-
636  int num_segments,
-
637  int *d_begin_offsets,
-
638  int *d_end_offsets,
-
639  int begin_bit = 0,
-
640  int end_bit = sizeof(KeyT) * 8,
-
641  cudaStream_t stream = 0,
-
642  bool debug_synchronous = false)
-
643  {
-
644  // Signed integer type for global offsets
-
645  typedef int OffsetT;
-
646 
-
647  // Null value type
-
648  DoubleBuffer<NullType> d_values;
-
649 
-
650  return DispatchSegmentedRadixSort<false, KeyT, NullType, OffsetT>::Dispatch(
-
651  d_temp_storage,
-
652  temp_storage_bytes,
-
653  d_keys,
-
654  d_values,
-
655  num_items,
-
656  num_segments,
-
657  d_begin_offsets,
-
658  d_end_offsets,
-
659  begin_bit,
-
660  end_bit,
-
661  true,
-
662  stream,
-
663  debug_synchronous);
-
664  }
-
665 
-
715  template <typename KeyT>
-
716  CUB_RUNTIME_FUNCTION
-
717  static cudaError_t SortKeysDescending(
-
718  void *d_temp_storage,
-
719  size_t &temp_storage_bytes,
-
720  KeyT *d_keys_in,
-
721  KeyT *d_keys_out,
-
722  int num_items,
-
723  int num_segments,
-
724  int *d_begin_offsets,
-
725  int *d_end_offsets,
-
726  int begin_bit = 0,
-
727  int end_bit = sizeof(KeyT) * 8,
-
728  cudaStream_t stream = 0,
-
729  bool debug_synchronous = false)
-
730  {
-
731  // Signed integer type for global offsets
-
732  typedef int OffsetT;
-
733 
-
734  DoubleBuffer<KeyT> d_keys(d_keys_in, d_keys_out);
-
735  DoubleBuffer<NullType> d_values;
-
736 
-
737  return DispatchSegmentedRadixSort<true, KeyT, NullType, OffsetT>::Dispatch(
-
738  d_temp_storage,
-
739  temp_storage_bytes,
-
740  d_keys,
-
741  d_values,
-
742  num_items,
-
743  num_segments,
-
744  d_begin_offsets,
-
745  d_end_offsets,
-
746  begin_bit,
-
747  end_bit,
-
748  false,
-
749  stream,
-
750  debug_synchronous);
-
751  }
+
628  template <typename KeyT>
+
629  CUB_RUNTIME_FUNCTION
+
630  static cudaError_t SortKeys(
+
631  void *d_temp_storage,
+
632  size_t &temp_storage_bytes,
+
633  DoubleBuffer<KeyT> &d_keys,
+
634  int num_items,
+
635  int num_segments,
+
636  int *d_begin_offsets,
+
637  int *d_end_offsets,
+
638  int begin_bit = 0,
+
639  int end_bit = sizeof(KeyT) * 8,
+
640  cudaStream_t stream = 0,
+
641  bool debug_synchronous = false)
+
642  {
+
643  // Signed integer type for global offsets
+
644  typedef int OffsetT;
+
645 
+
646  // Null value type
+
647  DoubleBuffer<NullType> d_values;
+
648 
+
649  return DispatchSegmentedRadixSort<false, KeyT, NullType, OffsetT>::Dispatch(
+
650  d_temp_storage,
+
651  temp_storage_bytes,
+
652  d_keys,
+
653  d_values,
+
654  num_items,
+
655  num_segments,
+
656  d_begin_offsets,
+
657  d_end_offsets,
+
658  begin_bit,
+
659  end_bit,
+
660  true,
+
661  stream,
+
662  debug_synchronous);
+
663  }
+
664 
+
714  template <typename KeyT>
+
715  CUB_RUNTIME_FUNCTION
+
716  static cudaError_t SortKeysDescending(
+
717  void *d_temp_storage,
+
718  size_t &temp_storage_bytes,
+
719  KeyT *d_keys_in,
+
720  KeyT *d_keys_out,
+
721  int num_items,
+
722  int num_segments,
+
723  int *d_begin_offsets,
+
724  int *d_end_offsets,
+
725  int begin_bit = 0,
+
726  int end_bit = sizeof(KeyT) * 8,
+
727  cudaStream_t stream = 0,
+
728  bool debug_synchronous = false)
+
729  {
+
730  // Signed integer type for global offsets
+
731  typedef int OffsetT;
+
732 
+
733  DoubleBuffer<KeyT> d_keys(d_keys_in, d_keys_out);
+
734  DoubleBuffer<NullType> d_values;
+
735 
+
736  return DispatchSegmentedRadixSort<true, KeyT, NullType, OffsetT>::Dispatch(
+
737  d_temp_storage,
+
738  temp_storage_bytes,
+
739  d_keys,
+
740  d_values,
+
741  num_items,
+
742  num_segments,
+
743  d_begin_offsets,
+
744  d_end_offsets,
+
745  begin_bit,
+
746  end_bit,
+
747  false,
+
748  stream,
+
749  debug_synchronous);
+
750  }
+
751 
752 
-
753 
-
810  template <typename KeyT>
-
811  CUB_RUNTIME_FUNCTION
-
812  static cudaError_t SortKeysDescending(
-
813  void *d_temp_storage,
-
814  size_t &temp_storage_bytes,
-
815  DoubleBuffer<KeyT> &d_keys,
-
816  int num_items,
-
817  int num_segments,
-
818  int *d_begin_offsets,
-
819  int *d_end_offsets,
-
820  int begin_bit = 0,
-
821  int end_bit = sizeof(KeyT) * 8,
-
822  cudaStream_t stream = 0,
-
823  bool debug_synchronous = false)
-
824  {
-
825  // Signed integer type for global offsets
-
826  typedef int OffsetT;
-
827 
-
828  // Null value type
-
829  DoubleBuffer<NullType> d_values;
-
830 
-
831  return DispatchSegmentedRadixSort<true, KeyT, NullType, OffsetT>::Dispatch(
-
832  d_temp_storage,
-
833  temp_storage_bytes,
-
834  d_keys,
-
835  d_values,
-
836  num_items,
-
837  num_segments,
-
838  d_begin_offsets,
-
839  d_end_offsets,
-
840  begin_bit,
-
841  end_bit,
-
842  true,
-
843  stream,
-
844  debug_synchronous);
-
845  }
+
809  template <typename KeyT>
+
810  CUB_RUNTIME_FUNCTION
+
811  static cudaError_t SortKeysDescending(
+
812  void *d_temp_storage,
+
813  size_t &temp_storage_bytes,
+
814  DoubleBuffer<KeyT> &d_keys,
+
815  int num_items,
+
816  int num_segments,
+
817  int *d_begin_offsets,
+
818  int *d_end_offsets,
+
819  int begin_bit = 0,
+
820  int end_bit = sizeof(KeyT) * 8,
+
821  cudaStream_t stream = 0,
+
822  bool debug_synchronous = false)
+
823  {
+
824  // Signed integer type for global offsets
+
825  typedef int OffsetT;
+
826 
+
827  // Null value type
+
828  DoubleBuffer<NullType> d_values;
+
829 
+
830  return DispatchSegmentedRadixSort<true, KeyT, NullType, OffsetT>::Dispatch(
+
831  d_temp_storage,
+
832  temp_storage_bytes,
+
833  d_keys,
+
834  d_values,
+
835  num_items,
+
836  num_segments,
+
837  d_begin_offsets,
+
838  d_end_offsets,
+
839  begin_bit,
+
840  end_bit,
+
841  true,
+
842  stream,
+
843  debug_synchronous);
+
844  }
+
845 
846 
-
847 
+
848 
849 
-
850 
-
851 };
-
852 
-
853 } // CUB namespace
-
854 CUB_NS_POSTFIX // Optional outer namespace(s)
+
850 };
+
851 
+
852 } // CUB namespace
+
853 CUB_NS_POSTFIX // Optional outer namespace(s)
+
854 
855 
-
856 
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__segmented__reduce_8cuh.html b/docs/html/device__segmented__reduce_8cuh.html index 40c32eb081..fa843eb667 100644 --- a/docs/html/device__segmented__reduce_8cuh.html +++ b/docs/html/device__segmented__reduce_8cuh.html @@ -108,6 +108,7 @@ #include <iterator>
#include "dispatch/dispatch_reduce.cuh"
#include "dispatch/dispatch_reduce_by_key.cuh"
+#include "../util_type.cuh"
#include "../util_namespace.cuh"

Go to the source code of this file.

@@ -137,7 +138,7 @@
-Generated on Wed Nov 18 2015 10:24:43 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__segmented__reduce_8cuh_source.html b/docs/html/device__segmented__reduce_8cuh_source.html index 734ebc1f66..712c987b24 100644 --- a/docs/html/device__segmented__reduce_8cuh_source.html +++ b/docs/html/device__segmented__reduce_8cuh_source.html @@ -137,249 +137,236 @@
39 
40 #include "dispatch/dispatch_reduce.cuh"
41 #include "dispatch/dispatch_reduce_by_key.cuh"
-
42 #include "../util_namespace.cuh"
-
43 
-
45 CUB_NS_PREFIX
-
46 
-
48 namespace cub {
-
49 
+
42 #include "../util_type.cuh"
+
43 #include "../util_namespace.cuh"
+
44 
+
46 CUB_NS_PREFIX
+
47 
+
49 namespace cub {
50 
-
63 struct DeviceSegmentedReduce
-
64 {
-
126  template <
-
127  typename InputIteratorT,
-
128  typename OutputIteratorT,
-
129  typename ReductionOp,
-
130  typename T>
-
131  CUB_RUNTIME_FUNCTION
-
132  static cudaError_t Reduce(
-
133  void *d_temp_storage,
-
134  size_t &temp_storage_bytes,
-
135  InputIteratorT d_in,
-
136  OutputIteratorT d_out,
-
137  int num_segments,
-
138  int *d_begin_offsets,
-
139  int *d_end_offsets,
-
140  ReductionOp reduction_op,
-
141  T identity,
-
142  cudaStream_t stream = 0,
-
143  bool debug_synchronous = false)
-
144  {
-
145  // Signed integer type for global offsets
-
146  typedef int OffsetT;
-
147 
-
148  return DispatchSegmentedReduce<InputIteratorT, OutputIteratorT, OffsetT, ReductionOp>::Dispatch(
-
149  d_temp_storage,
-
150  temp_storage_bytes,
-
151  d_in,
-
152  d_out,
-
153  num_segments,
-
154  d_begin_offsets,
-
155  d_end_offsets,
-
156  reduction_op,
-
157  identity,
-
158  stream,
-
159  debug_synchronous);
-
160  }
-
161 
-
162 
-
212  template <
-
213  typename InputIteratorT,
-
214  typename OutputIteratorT,
-
215  typename T>
-
216  CUB_RUNTIME_FUNCTION
-
217  static cudaError_t Sum(
-
218  void *d_temp_storage,
-
219  size_t &temp_storage_bytes,
-
220  InputIteratorT d_in,
-
221  OutputIteratorT d_out,
-
222  int num_segments,
-
223  int *d_begin_offsets,
-
224  int *d_end_offsets,
-
225  T identity,
-
226  cudaStream_t stream = 0,
-
227  bool debug_synchronous = false)
-
228  {
-
229  // Signed integer type for global offsets
-
230  typedef int OffsetT;
-
231 
-
232  return DispatchSegmentedReduce<InputIteratorT, OutputIteratorT, OffsetT, cub::Sum>::Dispatch(
-
233  d_temp_storage,
-
234  temp_storage_bytes,
-
235  d_in,
-
236  d_out,
-
237  num_segments,
-
238  d_begin_offsets,
-
239  d_end_offsets,
-
240  cub::Sum(),
-
241  identity,
-
242  stream,
-
243  debug_synchronous);
-
244  }
-
245 
-
246 
-
296  template <
-
297  typename InputIteratorT,
-
298  typename OutputIteratorT,
-
299  typename T>
-
300  CUB_RUNTIME_FUNCTION
-
301  static cudaError_t Min(
-
302  void *d_temp_storage,
-
303  size_t &temp_storage_bytes,
-
304  InputIteratorT d_in,
-
305  OutputIteratorT d_out,
-
306  int num_segments,
-
307  int *d_begin_offsets,
-
308  int *d_end_offsets,
-
309  T identity,
-
310  cudaStream_t stream = 0,
-
311  bool debug_synchronous = false)
-
312  {
-
313  // Signed integer type for global offsets
-
314  typedef int OffsetT;
+
51 
+
64 struct DeviceSegmentedReduce
+
65 {
+
124  template <
+
125  typename InputIteratorT,
+
126  typename OutputIteratorT,
+
127  typename ReductionOp,
+
128  typename T>
+
129  CUB_RUNTIME_FUNCTION
+
130  static cudaError_t Reduce(
+
131  void *d_temp_storage,
+
132  size_t &temp_storage_bytes,
+
133  InputIteratorT d_in,
+
134  OutputIteratorT d_out,
+
135  int num_segments,
+
136  int *d_begin_offsets,
+
137  int *d_end_offsets,
+
138  ReductionOp reduction_op,
+
139  T init,
+
140  cudaStream_t stream = 0,
+
141  bool debug_synchronous = false)
+
142  {
+
143  // Signed integer type for global offsets
+
144  typedef int OffsetT;
+
145 
+
146  return DispatchSegmentedReduce<InputIteratorT, OutputIteratorT, OffsetT, ReductionOp>::Dispatch(
+
147  d_temp_storage,
+
148  temp_storage_bytes,
+
149  d_in,
+
150  d_out,
+
151  num_segments,
+
152  d_begin_offsets,
+
153  d_end_offsets,
+
154  reduction_op,
+
155  init,
+
156  stream,
+
157  debug_synchronous);
+
158  }
+
159 
+
160 
+
206  template <
+
207  typename InputIteratorT,
+
208  typename OutputIteratorT>
+
209  CUB_RUNTIME_FUNCTION
+
210  static cudaError_t Sum(
+
211  void *d_temp_storage,
+
212  size_t &temp_storage_bytes,
+
213  InputIteratorT d_in,
+
214  OutputIteratorT d_out,
+
215  int num_segments,
+
216  int *d_begin_offsets,
+
217  int *d_end_offsets,
+
218  cudaStream_t stream = 0,
+
219  bool debug_synchronous = false)
+
220  {
+
221  typedef int OffsetT; // Signed integer type for global offsets
+
222  typedef typename std::iterator_traits<InputIteratorT>::value_type T; // Data element type
+
223 
+
224  return DispatchSegmentedReduce<InputIteratorT, OutputIteratorT, OffsetT, cub::Sum>::Dispatch(
+
225  d_temp_storage,
+
226  temp_storage_bytes,
+
227  d_in,
+
228  d_out,
+
229  num_segments,
+
230  d_begin_offsets,
+
231  d_end_offsets,
+
232  cub::Sum(),
+
233  T(), // zero-initialize
+
234  stream,
+
235  debug_synchronous);
+
236  }
+
237 
+
238 
+
284  template <
+
285  typename InputIteratorT,
+
286  typename OutputIteratorT>
+
287  CUB_RUNTIME_FUNCTION
+
288  static cudaError_t Min(
+
289  void *d_temp_storage,
+
290  size_t &temp_storage_bytes,
+
291  InputIteratorT d_in,
+
292  OutputIteratorT d_out,
+
293  int num_segments,
+
294  int *d_begin_offsets,
+
295  int *d_end_offsets,
+
296  cudaStream_t stream = 0,
+
297  bool debug_synchronous = false)
+
298  {
+
299  typedef int OffsetT; // Signed integer type for global offsets
+
300  typedef typename std::iterator_traits<InputIteratorT>::value_type T; // Data element type
+
301 
+
302  return DispatchSegmentedReduce<InputIteratorT, OutputIteratorT, OffsetT, cub::Min>::Dispatch(
+
303  d_temp_storage,
+
304  temp_storage_bytes,
+
305  d_in,
+
306  d_out,
+
307  num_segments,
+
308  d_begin_offsets,
+
309  d_end_offsets,
+
310  cub::Min(),
+
311  Traits<T>::Max(), // replace with std::numeric_limits<T>::max() when C++11 support is more prevalent
+
312  stream,
+
313  debug_synchronous);
+
314  }
315 
-
316  return DispatchSegmentedReduce<InputIteratorT, OutputIteratorT, OffsetT, cub::Min>::Dispatch(
-
317  d_temp_storage,
-
318  temp_storage_bytes,
-
319  d_in,
-
320  d_out,
-
321  num_segments,
-
322  d_begin_offsets,
-
323  d_end_offsets,
-
324  cub::Min(),
-
325  identity,
-
326  stream,
-
327  debug_synchronous);
-
328  }
-
329 
-
330 
-
383  template <
-
384  typename InputIteratorT,
-
385  typename OutputIteratorT,
-
386  typename T>
-
387  CUB_RUNTIME_FUNCTION
-
388  static cudaError_t ArgMin(
-
389  void *d_temp_storage,
-
390  size_t &temp_storage_bytes,
-
391  InputIteratorT d_in,
-
392  OutputIteratorT d_out,
-
393  int num_segments,
-
394  int *d_begin_offsets,
-
395  int *d_end_offsets,
-
396  T identity,
-
397  cudaStream_t stream = 0,
-
398  bool debug_synchronous = false)
-
399  {
-
400  // Signed integer type for global offsets
-
401  typedef int OffsetT;
-
402 
-
403  // Wrapped input iterator
-
404  typedef ArgIndexInputIterator<InputIteratorT, int> ArgIndexInputIteratorT;
-
405  ArgIndexInputIteratorT d_argmin_in(d_in, 0);
-
406 
-
407  KeyValuePair<OffsetT, T> identity_pair = {-1, identity};
-
408 
-
409  return DispatchSegmentedReduce<ArgIndexInputIteratorT, OutputIteratorT, OffsetT, cub::ArgMin>::Dispatch(
-
410  d_temp_storage,
-
411  temp_storage_bytes,
-
412  d_argmin_in,
-
413  d_out,
-
414  num_segments,
-
415  d_begin_offsets,
-
416  d_end_offsets,
-
417  cub::ArgMin(),
-
418  identity_pair,
-
419  stream,
-
420  debug_synchronous);
-
421  }
-
422 
-
423 
-
473  template <
-
474  typename InputIteratorT,
-
475  typename OutputIteratorT,
-
476  typename T>
-
477  CUB_RUNTIME_FUNCTION
-
478  static cudaError_t Max(
-
479  void *d_temp_storage,
-
480  size_t &temp_storage_bytes,
-
481  InputIteratorT d_in,
-
482  OutputIteratorT d_out,
-
483  int num_segments,
-
484  int *d_begin_offsets,
-
485  int *d_end_offsets,
-
486  T identity,
-
487  cudaStream_t stream = 0,
-
488  bool debug_synchronous = false)
-
489  {
-
490  // Signed integer type for global offsets
-
491  typedef int OffsetT;
-
492 
-
493  return DispatchSegmentedReduce<InputIteratorT, OutputIteratorT, OffsetT, cub::Max>::Dispatch(
-
494  d_temp_storage,
-
495  temp_storage_bytes,
-
496  d_in,
-
497  d_out,
-
498  num_segments,
-
499  d_begin_offsets,
-
500  d_end_offsets,
-
501  cub::Max(),
-
502  identity,
-
503  stream,
-
504  debug_synchronous);
-
505  }
-
506 
-
507 
-
560  template <
-
561  typename InputIteratorT,
-
562  typename OutputIteratorT,
-
563  typename T>
-
564  CUB_RUNTIME_FUNCTION
-
565  static cudaError_t ArgMax(
-
566  void *d_temp_storage,
-
567  size_t &temp_storage_bytes,
-
568  InputIteratorT d_in,
-
569  OutputIteratorT d_out,
-
570  int num_segments,
-
571  int *d_begin_offsets,
-
572  int *d_end_offsets,
-
573  T identity,
-
574  cudaStream_t stream = 0,
-
575  bool debug_synchronous = false)
-
576  {
-
577  // Signed integer type for global offsets
-
578  typedef int OffsetT;
-
579 
-
580  // Wrapped input iterator
-
581  typedef ArgIndexInputIterator<InputIteratorT, OffsetT> ArgIndexInputIteratorT;
-
582  ArgIndexInputIteratorT d_argmax_in(d_in, 0);
-
583 
-
584  KeyValuePair<OffsetT, T> identity_pair = {-1, identity};
-
585 
-
586  return DispatchSegmentedReduce<ArgIndexInputIteratorT, OutputIteratorT, OffsetT, cub::ArgMax>::Dispatch(
-
587  d_temp_storage,
-
588  temp_storage_bytes,
-
589  d_argmax_in,
-
590  d_out,
-
591  num_segments,
-
592  d_begin_offsets,
-
593  d_end_offsets,
-
594  cub::ArgMax(),
-
595  identity_pair,
-
596  stream,
-
597  debug_synchronous);
-
598  }
-
599 
-
600 };
-
601 
-
602 } // CUB namespace
-
603 CUB_NS_POSTFIX // Optional outer namespace(s)
-
604 
-
605 
+
316 
+
364  template <
+
365  typename InputIteratorT,
+
366  typename OutputIteratorT>
+
367  CUB_RUNTIME_FUNCTION
+
368  static cudaError_t ArgMin(
+
369  void *d_temp_storage,
+
370  size_t &temp_storage_bytes,
+
371  InputIteratorT d_in,
+
372  OutputIteratorT d_out,
+
373  int num_segments,
+
374  int *d_begin_offsets,
+
375  int *d_end_offsets,
+
376  cudaStream_t stream = 0,
+
377  bool debug_synchronous = false)
+
378  {
+
379  typedef int OffsetT; // Signed integer type for global offsets
+
380  typedef typename std::iterator_traits<InputIteratorT>::value_type T; // Data element type
+
381  typedef ArgIndexInputIterator<InputIteratorT, int> ArgIndexInputIteratorT; // Wrapped input iterator type
+
382 
+
383  ArgIndexInputIteratorT d_argmin_in(d_in);
+
384  KeyValuePair<OffsetT, T> init = {1, Traits<T>::Max()}; // replace with std::numeric_limits<T>::max() when C++11 support is more prevalent
+
385 
+
386  return DispatchSegmentedReduce<ArgIndexInputIteratorT, OutputIteratorT, OffsetT, cub::ArgMin>::Dispatch(
+
387  d_temp_storage,
+
388  temp_storage_bytes,
+
389  d_argmin_in,
+
390  d_out,
+
391  num_segments,
+
392  d_begin_offsets,
+
393  d_end_offsets,
+
394  cub::ArgMin(),
+
395  init,
+
396  stream,
+
397  debug_synchronous);
+
398  }
+
399 
+
400 
+
446  template <
+
447  typename InputIteratorT,
+
448  typename OutputIteratorT>
+
449  CUB_RUNTIME_FUNCTION
+
450  static cudaError_t Max(
+
451  void *d_temp_storage,
+
452  size_t &temp_storage_bytes,
+
453  InputIteratorT d_in,
+
454  OutputIteratorT d_out,
+
455  int num_segments,
+
456  int *d_begin_offsets,
+
457  int *d_end_offsets,
+
458  cudaStream_t stream = 0,
+
459  bool debug_synchronous = false)
+
460  {
+
461  typedef int OffsetT; // Signed integer type for global offsets
+
462  typedef typename std::iterator_traits<InputIteratorT>::value_type T; // Data element type
+
463 
+
464  return DispatchSegmentedReduce<InputIteratorT, OutputIteratorT, OffsetT, cub::Max>::Dispatch(
+
465  d_temp_storage,
+
466  temp_storage_bytes,
+
467  d_in,
+
468  d_out,
+
469  num_segments,
+
470  d_begin_offsets,
+
471  d_end_offsets,
+
472  cub::Max(),
+
473  Traits<T>::Lowest(), // replace with std::numeric_limits<T>::lowest() when C++11 support is more prevalent
+
474  stream,
+
475  debug_synchronous);
+
476  }
+
477 
+
478 
+
526  template <
+
527  typename InputIteratorT,
+
528  typename OutputIteratorT>
+
529  CUB_RUNTIME_FUNCTION
+
530  static cudaError_t ArgMax(
+
531  void *d_temp_storage,
+
532  size_t &temp_storage_bytes,
+
533  InputIteratorT d_in,
+
534  OutputIteratorT d_out,
+
535  int num_segments,
+
536  int *d_begin_offsets,
+
537  int *d_end_offsets,
+
538  cudaStream_t stream = 0,
+
539  bool debug_synchronous = false)
+
540  {
+
541  typedef int OffsetT; // Signed integer type for global offsets
+
542  typedef typename std::iterator_traits<InputIteratorT>::value_type T; // Data element type
+
543  typedef ArgIndexInputIterator<InputIteratorT, int> ArgIndexInputIteratorT; // Wrapped input iterator
+
544 
+
545  ArgIndexInputIteratorT d_argmax_in(d_in);
+
546  KeyValuePair<OffsetT, T> init = {1, Traits<T>::Lowest()}; // replace with std::numeric_limits<T>::lowest() when C++11 support is more prevalent
+
547 
+
548  return DispatchSegmentedReduce<ArgIndexInputIteratorT, OutputIteratorT, OffsetT, cub::ArgMax>::Dispatch(
+
549  d_temp_storage,
+
550  temp_storage_bytes,
+
551  d_argmax_in,
+
552  d_out,
+
553  num_segments,
+
554  d_begin_offsets,
+
555  d_end_offsets,
+
556  cub::ArgMax(),
+
557  init,
+
558  stream,
+
559  debug_synchronous);
+
560  }
+
561 
+
562 };
+
563 
+
564 } // CUB namespace
+
565 CUB_NS_POSTFIX // Optional outer namespace(s)
+
566 
+
567 
-Generated on Wed Nov 18 2015 10:24:42 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__select_8cuh.html b/docs/html/device__select_8cuh.html index 49a594d978..9fb417afaa 100644 --- a/docs/html/device__select_8cuh.html +++ b/docs/html/device__select_8cuh.html @@ -129,14 +129,14 @@  

Detailed Description

-

cub::DeviceSelect provides device-wide, parallel operations for selecting items from sequences of data items residing within device-accessible memory.

+

cub::DeviceSelect provides device-wide, parallel operations for compacting selected items from sequences of data items residing within device-accessible memory.

Definition in file device_select.cuh.

-Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__select_8cuh_source.html b/docs/html/device__select_8cuh_source.html index f7eee08b35..bd3a25a25f 100644 --- a/docs/html/device__select_8cuh_source.html +++ b/docs/html/device__select_8cuh_source.html @@ -145,123 +145,123 @@
49 
82 struct DeviceSelect
83 {
-
128  template <
-
129  typename InputIteratorT,
-
130  typename FlagIterator,
-
131  typename OutputIteratorT,
-
132  typename NumSelectedIteratorT>
-
133  CUB_RUNTIME_FUNCTION __forceinline__
-
134  static cudaError_t Flagged(
-
135  void* d_temp_storage,
-
136  size_t &temp_storage_bytes,
-
137  InputIteratorT d_in,
-
138  FlagIterator d_flags,
-
139  OutputIteratorT d_out,
-
140  NumSelectedIteratorT d_num_selected_out,
-
141  int num_items,
-
142  cudaStream_t stream = 0,
-
143  bool debug_synchronous = false)
-
144  {
-
145  typedef int OffsetT; // Signed integer type for global offsets
-
146  typedef NullType SelectOp; // Selection op (not used)
-
147  typedef NullType EqualityOp; // Equality operator (not used)
-
148 
-
149  return DispatchSelectIf<InputIteratorT, FlagIterator, OutputIteratorT, NumSelectedIteratorT, SelectOp, EqualityOp, OffsetT, false>::Dispatch(
-
150  d_temp_storage,
-
151  temp_storage_bytes,
-
152  d_in,
-
153  d_flags,
-
154  d_out,
-
155  d_num_selected_out,
-
156  SelectOp(),
-
157  EqualityOp(),
-
158  num_items,
-
159  stream,
-
160  debug_synchronous);
-
161  }
+
127  template <
+
128  typename InputIteratorT,
+
129  typename FlagIterator,
+
130  typename OutputIteratorT,
+
131  typename NumSelectedIteratorT>
+
132  CUB_RUNTIME_FUNCTION __forceinline__
+
133  static cudaError_t Flagged(
+
134  void* d_temp_storage,
+
135  size_t &temp_storage_bytes,
+
136  InputIteratorT d_in,
+
137  FlagIterator d_flags,
+
138  OutputIteratorT d_out,
+
139  NumSelectedIteratorT d_num_selected_out,
+
140  int num_items,
+
141  cudaStream_t stream = 0,
+
142  bool debug_synchronous = false)
+
143  {
+
144  typedef int OffsetT; // Signed integer type for global offsets
+
145  typedef NullType SelectOp; // Selection op (not used)
+
146  typedef NullType EqualityOp; // Equality operator (not used)
+
147 
+
148  return DispatchSelectIf<InputIteratorT, FlagIterator, OutputIteratorT, NumSelectedIteratorT, SelectOp, EqualityOp, OffsetT, false>::Dispatch(
+
149  d_temp_storage,
+
150  temp_storage_bytes,
+
151  d_in,
+
152  d_flags,
+
153  d_out,
+
154  d_num_selected_out,
+
155  SelectOp(),
+
156  EqualityOp(),
+
157  num_items,
+
158  stream,
+
159  debug_synchronous);
+
160  }
+
161 
162 
-
163 
-
235  template <
-
236  typename InputIteratorT,
-
237  typename OutputIteratorT,
-
238  typename NumSelectedIteratorT,
-
239  typename SelectOp>
-
240  CUB_RUNTIME_FUNCTION __forceinline__
-
241  static cudaError_t If(
-
242  void* d_temp_storage,
-
243  size_t &temp_storage_bytes,
-
244  InputIteratorT d_in,
-
245  OutputIteratorT d_out,
-
246  NumSelectedIteratorT d_num_selected_out,
-
247  int num_items,
-
248  SelectOp select_op,
-
249  cudaStream_t stream = 0,
-
250  bool debug_synchronous = false)
-
251  {
-
252  typedef int OffsetT; // Signed integer type for global offsets
-
253  typedef NullType* FlagIterator; // FlagT iterator type (not used)
-
254  typedef NullType EqualityOp; // Equality operator (not used)
-
255 
-
256  return DispatchSelectIf<InputIteratorT, FlagIterator, OutputIteratorT, NumSelectedIteratorT, SelectOp, EqualityOp, OffsetT, false>::Dispatch(
-
257  d_temp_storage,
-
258  temp_storage_bytes,
-
259  d_in,
-
260  NULL,
-
261  d_out,
-
262  d_num_selected_out,
-
263  select_op,
-
264  EqualityOp(),
-
265  num_items,
-
266  stream,
-
267  debug_synchronous);
-
268  }
-
269 
-
270 
-
327  template <
-
328  typename InputIteratorT,
-
329  typename OutputIteratorT,
-
330  typename NumSelectedIteratorT>
-
331  CUB_RUNTIME_FUNCTION __forceinline__
-
332  static cudaError_t Unique(
-
333  void* d_temp_storage,
-
334  size_t &temp_storage_bytes,
-
335  InputIteratorT d_in,
-
336  OutputIteratorT d_out,
-
337  NumSelectedIteratorT d_num_selected_out,
-
338  int num_items,
-
339  cudaStream_t stream = 0,
-
340  bool debug_synchronous = false)
-
341  {
-
342  typedef int OffsetT; // Signed integer type for global offsets
-
343  typedef NullType* FlagIterator; // FlagT iterator type (not used)
-
344  typedef NullType SelectOp; // Selection op (not used)
-
345  typedef Equality EqualityOp; // Default == operator
-
346 
-
347  return DispatchSelectIf<InputIteratorT, FlagIterator, OutputIteratorT, NumSelectedIteratorT, SelectOp, EqualityOp, OffsetT, false>::Dispatch(
-
348  d_temp_storage,
-
349  temp_storage_bytes,
-
350  d_in,
-
351  NULL,
-
352  d_out,
-
353  d_num_selected_out,
-
354  SelectOp(),
-
355  EqualityOp(),
-
356  num_items,
-
357  stream,
-
358  debug_synchronous);
-
359  }
-
360 
-
361 };
-
362 
-
369 } // CUB namespace
-
370 CUB_NS_POSTFIX // Optional outer namespace(s)
-
371 
-
372 
+
233  template <
+
234  typename InputIteratorT,
+
235  typename OutputIteratorT,
+
236  typename NumSelectedIteratorT,
+
237  typename SelectOp>
+
238  CUB_RUNTIME_FUNCTION __forceinline__
+
239  static cudaError_t If(
+
240  void* d_temp_storage,
+
241  size_t &temp_storage_bytes,
+
242  InputIteratorT d_in,
+
243  OutputIteratorT d_out,
+
244  NumSelectedIteratorT d_num_selected_out,
+
245  int num_items,
+
246  SelectOp select_op,
+
247  cudaStream_t stream = 0,
+
248  bool debug_synchronous = false)
+
249  {
+
250  typedef int OffsetT; // Signed integer type for global offsets
+
251  typedef NullType* FlagIterator; // FlagT iterator type (not used)
+
252  typedef NullType EqualityOp; // Equality operator (not used)
+
253 
+
254  return DispatchSelectIf<InputIteratorT, FlagIterator, OutputIteratorT, NumSelectedIteratorT, SelectOp, EqualityOp, OffsetT, false>::Dispatch(
+
255  d_temp_storage,
+
256  temp_storage_bytes,
+
257  d_in,
+
258  NULL,
+
259  d_out,
+
260  d_num_selected_out,
+
261  select_op,
+
262  EqualityOp(),
+
263  num_items,
+
264  stream,
+
265  debug_synchronous);
+
266  }
+
267 
+
268 
+
324  template <
+
325  typename InputIteratorT,
+
326  typename OutputIteratorT,
+
327  typename NumSelectedIteratorT>
+
328  CUB_RUNTIME_FUNCTION __forceinline__
+
329  static cudaError_t Unique(
+
330  void* d_temp_storage,
+
331  size_t &temp_storage_bytes,
+
332  InputIteratorT d_in,
+
333  OutputIteratorT d_out,
+
334  NumSelectedIteratorT d_num_selected_out,
+
335  int num_items,
+
336  cudaStream_t stream = 0,
+
337  bool debug_synchronous = false)
+
338  {
+
339  typedef int OffsetT; // Signed integer type for global offsets
+
340  typedef NullType* FlagIterator; // FlagT iterator type (not used)
+
341  typedef NullType SelectOp; // Selection op (not used)
+
342  typedef Equality EqualityOp; // Default == operator
+
343 
+
344  return DispatchSelectIf<InputIteratorT, FlagIterator, OutputIteratorT, NumSelectedIteratorT, SelectOp, EqualityOp, OffsetT, false>::Dispatch(
+
345  d_temp_storage,
+
346  temp_storage_bytes,
+
347  d_in,
+
348  NULL,
+
349  d_out,
+
350  d_num_selected_out,
+
351  SelectOp(),
+
352  EqualityOp(),
+
353  num_items,
+
354  stream,
+
355  debug_synchronous);
+
356  }
+
357 
+
358 };
+
359 
+
366 } // CUB namespace
+
367 CUB_NS_POSTFIX // Optional outer namespace(s)
+
368 
+
369 
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__spmv_8cuh.html b/docs/html/device__spmv_8cuh.html index 394ed1584e..5c7cabbd56 100644 --- a/docs/html/device__spmv_8cuh.html +++ b/docs/html/device__spmv_8cuh.html @@ -132,7 +132,7 @@
-Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
diff --git a/docs/html/device__spmv_8cuh_source.html b/docs/html/device__spmv_8cuh_source.html index 14953addeb..4a5eec6214 100644 --- a/docs/html/device__spmv_8cuh_source.html +++ b/docs/html/device__spmv_8cuh_source.html @@ -197,7 +197,7 @@
-Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
diff --git a/docs/html/dir_011e1c944d88f71be72e1e24a5fda7cf.html b/docs/html/dir_011e1c944d88f71be72e1e24a5fda7cf.html index e94cd91c96..a9a93a9d0a 100644 --- a/docs/html/dir_011e1c944d88f71be72e1e24a5fda7cf.html +++ b/docs/html/dir_011e1c944d88f71be72e1e24a5fda7cf.html @@ -119,7 +119,7 @@
-Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
diff --git a/docs/html/dir_18fc672d63781b5a743137aee24ff656.html b/docs/html/dir_18fc672d63781b5a743137aee24ff656.html index 8e70f77bfa..7ab18175db 100644 --- a/docs/html/dir_18fc672d63781b5a743137aee24ff656.html +++ b/docs/html/dir_18fc672d63781b5a743137aee24ff656.html @@ -112,6 +112,8 @@   file  device_segmented_radix_sort.cuh [code]   +file  device_segmented_reduce.cuh [code] +  file  device_select.cuh [code]   file  device_spmv.cuh [code] @@ -121,7 +123,7 @@
-Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
diff --git a/docs/html/dir_80932b4cec52750ff92b1a1912314cf5.html b/docs/html/dir_80932b4cec52750ff92b1a1912314cf5.html index b870667d22..04e8ebf9bf 100644 --- a/docs/html/dir_80932b4cec52750ff92b1a1912314cf5.html +++ b/docs/html/dir_80932b4cec52750ff92b1a1912314cf5.html @@ -119,7 +119,7 @@
-Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
diff --git a/docs/html/dir_bb50a5ef59f19d030d06415663184d05.html b/docs/html/dir_bb50a5ef59f19d030d06415663184d05.html index 50ae279f23..bbc264bb96 100644 --- a/docs/html/dir_bb50a5ef59f19d030d06415663184d05.html +++ b/docs/html/dir_bb50a5ef59f19d030d06415663184d05.html @@ -109,7 +109,7 @@
-Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
diff --git a/docs/html/dir_cb3a671affffe7eeb3fdf5ae58e42cc8.html b/docs/html/dir_cb3a671affffe7eeb3fdf5ae58e42cc8.html index d7fb1baeac..cc4380b5fe 100644 --- a/docs/html/dir_cb3a671affffe7eeb3fdf5ae58e42cc8.html +++ b/docs/html/dir_cb3a671affffe7eeb3fdf5ae58e42cc8.html @@ -107,7 +107,7 @@
-Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
diff --git a/docs/html/dir_d583f216f1aafe19404e836b0c097ad2.html b/docs/html/dir_d583f216f1aafe19404e836b0c097ad2.html index bd0641455a..e8692a183f 100644 --- a/docs/html/dir_d583f216f1aafe19404e836b0c097ad2.html +++ b/docs/html/dir_d583f216f1aafe19404e836b0c097ad2.html @@ -130,7 +130,7 @@
-Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
diff --git a/docs/html/download_cub.html b/docs/html/download_cub.html index f2d78ad3e0..a9f8ae9530 100644 --- a/docs/html/download_cub.html +++ b/docs/html/download_cub.html @@ -37,14 +37,14 @@
If your download doesn't start in 3s:

- -Download CUB! + +Download CUB!
diff --git a/docs/html/example_block_radix_sort_8cu-example.html b/docs/html/example_block_radix_sort_8cu-example.html index 12f8eda1cb..f9bdc51bdd 100644 --- a/docs/html/example_block_radix_sort_8cu-example.html +++ b/docs/html/example_block_radix_sort_8cu-example.html @@ -398,7 +398,7 @@
-Generated on Tue Nov 10 2015 13:58:02 for CUB by   +Generated on Mon Dec 14 2015 13:11:13 for CUB by   doxygen 1.8.4
diff --git a/docs/html/example_block_reduce_8cu-example.html b/docs/html/example_block_reduce_8cu-example.html index 0bef27ff33..4ff2a1dbe7 100644 --- a/docs/html/example_block_reduce_8cu-example.html +++ b/docs/html/example_block_reduce_8cu-example.html @@ -365,7 +365,7 @@
-Generated on Tue Nov 10 2015 13:58:02 for CUB by   +Generated on Mon Dec 14 2015 13:11:13 for CUB by   doxygen 1.8.4
diff --git a/docs/html/example_block_scan_8cu-example.html b/docs/html/example_block_scan_8cu-example.html index f971046324..690a02a2e7 100644 --- a/docs/html/example_block_scan_8cu-example.html +++ b/docs/html/example_block_scan_8cu-example.html @@ -414,7 +414,7 @@
-Generated on Tue Nov 10 2015 13:58:02 for CUB by   +Generated on Mon Dec 14 2015 13:11:13 for CUB by   doxygen 1.8.4
diff --git a/docs/html/example_device_histogram_8cu-example.html b/docs/html/example_device_histogram_8cu-example.html index 71d3e4b786..773eebcb1c 100644 --- a/docs/html/example_device_histogram_8cu-example.html +++ b/docs/html/example_device_histogram_8cu-example.html @@ -312,7 +312,7 @@
-Generated on Tue Nov 10 2015 13:58:02 for CUB by   +Generated on Mon Dec 14 2015 13:11:13 for CUB by   doxygen 1.8.4
diff --git a/docs/html/example_device_partition_flagged_8cu-example.html b/docs/html/example_device_partition_flagged_8cu-example.html index 98d3b78d1a..8f6e6ce981 100644 --- a/docs/html/example_device_partition_flagged_8cu-example.html +++ b/docs/html/example_device_partition_flagged_8cu-example.html @@ -318,7 +318,7 @@
-Generated on Tue Nov 10 2015 13:58:02 for CUB by   +Generated on Mon Dec 14 2015 13:11:13 for CUB by   doxygen 1.8.4
diff --git a/docs/html/example_device_partition_if_8cu-example.html b/docs/html/example_device_partition_if_8cu-example.html index 8be1c5003e..f8fadfbd58 100644 --- a/docs/html/example_device_partition_if_8cu-example.html +++ b/docs/html/example_device_partition_if_8cu-example.html @@ -329,7 +329,7 @@
-Generated on Tue Nov 10 2015 13:58:02 for CUB by   +Generated on Mon Dec 14 2015 13:11:13 for CUB by   doxygen 1.8.4
diff --git a/docs/html/example_device_radix_sort_8cu-example.html b/docs/html/example_device_radix_sort_8cu-example.html index 6ab4c37e57..fa0df1c0c7 100644 --- a/docs/html/example_device_radix_sort_8cu-example.html +++ b/docs/html/example_device_radix_sort_8cu-example.html @@ -259,33 +259,33 @@
Initialize(h_keys, h_values, h_reference_keys, h_reference_values, num_items);
// Allocate device arrays
-
DoubleBuffer<float> d_keys;
-
DoubleBuffer<int> d_values;
-
CubDebugExit(g_allocator.DeviceAllocate((void**)&d_keys.d_buffers[0], sizeof(float) * num_items));
-
CubDebugExit(g_allocator.DeviceAllocate((void**)&d_keys.d_buffers[1], sizeof(float) * num_items));
-
CubDebugExit(g_allocator.DeviceAllocate((void**)&d_values.d_buffers[0], sizeof(int) * num_items));
-
CubDebugExit(g_allocator.DeviceAllocate((void**)&d_values.d_buffers[1], sizeof(int) * num_items));
+
DoubleBuffer<float> d_keys;
+
DoubleBuffer<int> d_values;
+
CubDebugExit(g_allocator.DeviceAllocate((void**)&d_keys.d_buffers[0], sizeof(float) * num_items));
+
CubDebugExit(g_allocator.DeviceAllocate((void**)&d_keys.d_buffers[1], sizeof(float) * num_items));
+
CubDebugExit(g_allocator.DeviceAllocate((void**)&d_values.d_buffers[0], sizeof(int) * num_items));
+
CubDebugExit(g_allocator.DeviceAllocate((void**)&d_values.d_buffers[1], sizeof(int) * num_items));
// Allocate temporary storage
size_t temp_storage_bytes = 0;
void *d_temp_storage = NULL;
-
CubDebugExit(DeviceRadixSort::SortPairs(d_temp_storage, temp_storage_bytes, d_keys, d_values, num_items));
+
CubDebugExit(DeviceRadixSort::SortPairs(d_temp_storage, temp_storage_bytes, d_keys, d_values, num_items));
CubDebugExit(g_allocator.DeviceAllocate(&d_temp_storage, temp_storage_bytes));
// Initialize device arrays
-
CubDebugExit(cudaMemcpy(d_keys.d_buffers[d_keys.selector], h_keys, sizeof(float) * num_items, cudaMemcpyHostToDevice));
-
CubDebugExit(cudaMemcpy(d_values.d_buffers[d_values.selector], h_values, sizeof(int) * num_items, cudaMemcpyHostToDevice));
+
CubDebugExit(cudaMemcpy(d_keys.d_buffers[d_keys.selector], h_keys, sizeof(float) * num_items, cudaMemcpyHostToDevice));
+
CubDebugExit(cudaMemcpy(d_values.d_buffers[d_values.selector], h_values, sizeof(int) * num_items, cudaMemcpyHostToDevice));
// Run
CubDebugExit(DeviceRadixSort::SortPairs(d_temp_storage, temp_storage_bytes, d_keys, d_values, num_items));
// Check for correctness (and display results, if specified)
-
int compare = CompareDeviceResults(h_reference_keys, d_keys.Current(), num_items, true, g_verbose);
-
printf("\t Compare keys (selector %d): %s\n", d_keys.selector, compare ? "FAIL" : "PASS");
+
int compare = CompareDeviceResults(h_reference_keys, d_keys.Current(), num_items, true, g_verbose);
+
printf("\t Compare keys (selector %d): %s\n", d_keys.selector, compare ? "FAIL" : "PASS");
AssertEquals(0, compare);
-
compare = CompareDeviceResults(h_reference_values, d_values.Current(), num_items, true, g_verbose);
-
printf("\t Compare values (selector %d): %s\n", d_values.selector, compare ? "FAIL" : "PASS");
+
compare = CompareDeviceResults(h_reference_values, d_values.Current(), num_items, true, g_verbose);
+
printf("\t Compare values (selector %d): %s\n", d_values.selector, compare ? "FAIL" : "PASS");
AssertEquals(0, compare);
// Cleanup
@@ -294,10 +294,10 @@
if (h_values) delete[] h_values;
if (h_reference_values) delete[] h_reference_values;
-
if (d_keys.d_buffers[0]) CubDebugExit(g_allocator.DeviceFree(d_keys.d_buffers[0]));
-
if (d_keys.d_buffers[1]) CubDebugExit(g_allocator.DeviceFree(d_keys.d_buffers[1]));
-
if (d_values.d_buffers[0]) CubDebugExit(g_allocator.DeviceFree(d_values.d_buffers[0]));
-
if (d_values.d_buffers[1]) CubDebugExit(g_allocator.DeviceFree(d_values.d_buffers[1]));
+
if (d_keys.d_buffers[0]) CubDebugExit(g_allocator.DeviceFree(d_keys.d_buffers[0]));
+
if (d_keys.d_buffers[1]) CubDebugExit(g_allocator.DeviceFree(d_keys.d_buffers[1]));
+
if (d_values.d_buffers[0]) CubDebugExit(g_allocator.DeviceFree(d_values.d_buffers[0]));
+
if (d_values.d_buffers[1]) CubDebugExit(g_allocator.DeviceFree(d_values.d_buffers[1]));
if (d_temp_storage) CubDebugExit(g_allocator.DeviceFree(d_temp_storage));
printf("\n\n");
@@ -311,7 +311,7 @@
-Generated on Tue Nov 10 2015 13:58:02 for CUB by   +Generated on Mon Dec 14 2015 13:11:13 for CUB by   doxygen 1.8.4
diff --git a/docs/html/example_device_reduce_8cu-example.html b/docs/html/example_device_reduce_8cu-example.html index a655cd646a..2b6b709725 100644 --- a/docs/html/example_device_reduce_8cu-example.html +++ b/docs/html/example_device_reduce_8cu-example.html @@ -238,11 +238,11 @@
// Request and allocate temporary storage
void *d_temp_storage = NULL;
size_t temp_storage_bytes = 0;
-
CubDebugExit(DeviceReduce::Sum(d_temp_storage, temp_storage_bytes, d_in, d_out, num_items));
+
CubDebugExit(DeviceReduce::Sum(d_temp_storage, temp_storage_bytes, d_in, d_out, num_items));
CubDebugExit(g_allocator.DeviceAllocate(&d_temp_storage, temp_storage_bytes));
// Run
-
CubDebugExit(DeviceReduce::Sum(d_temp_storage, temp_storage_bytes, d_in, d_out, num_items));
+
CubDebugExit(DeviceReduce::Sum(d_temp_storage, temp_storage_bytes, d_in, d_out, num_items));
// Check for correctness (and display results, if specified)
int compare = CompareDeviceResults(&h_reference, d_out, 1, g_verbose, g_verbose);
@@ -266,7 +266,7 @@
-Generated on Tue Nov 10 2015 13:58:02 for CUB by   +Generated on Mon Dec 14 2015 13:11:13 for CUB by   doxygen 1.8.4
diff --git a/docs/html/example_device_scan_8cu-example.html b/docs/html/example_device_scan_8cu-example.html index 360f8ee7b9..2ec49fed77 100644 --- a/docs/html/example_device_scan_8cu-example.html +++ b/docs/html/example_device_scan_8cu-example.html @@ -272,7 +272,7 @@
-Generated on Tue Nov 10 2015 13:58:02 for CUB by   +Generated on Mon Dec 14 2015 13:11:13 for CUB by   doxygen 1.8.4
diff --git a/docs/html/example_device_select_flagged_8cu-example.html b/docs/html/example_device_select_flagged_8cu-example.html index 7821efa0c4..85cdb1ba8d 100644 --- a/docs/html/example_device_select_flagged_8cu-example.html +++ b/docs/html/example_device_select_flagged_8cu-example.html @@ -318,7 +318,7 @@
-Generated on Tue Nov 10 2015 13:58:02 for CUB by   +Generated on Mon Dec 14 2015 13:11:13 for CUB by   doxygen 1.8.4
diff --git a/docs/html/example_device_select_if_8cu-example.html b/docs/html/example_device_select_if_8cu-example.html index 6225038057..9881612c35 100644 --- a/docs/html/example_device_select_if_8cu-example.html +++ b/docs/html/example_device_select_if_8cu-example.html @@ -327,7 +327,7 @@
-Generated on Tue Nov 10 2015 13:58:02 for CUB by   +Generated on Mon Dec 14 2015 13:11:13 for CUB by   doxygen 1.8.4
diff --git a/docs/html/example_device_select_unique_8cu-example.html b/docs/html/example_device_select_unique_8cu-example.html index 3da3edbc17..f883582e40 100644 --- a/docs/html/example_device_select_unique_8cu-example.html +++ b/docs/html/example_device_select_unique_8cu-example.html @@ -307,7 +307,7 @@
-Generated on Tue Nov 10 2015 13:58:02 for CUB by   +Generated on Mon Dec 14 2015 13:11:13 for CUB by   doxygen 1.8.4
diff --git a/docs/html/examples.html b/docs/html/examples.html index a85843055d..7605af1b12 100644 --- a/docs/html/examples.html +++ b/docs/html/examples.html @@ -121,7 +121,7 @@
-Generated on Tue Nov 10 2015 13:58:09 for CUB by   +Generated on Mon Dec 14 2015 13:11:20 for CUB by   doxygen 1.8.4
diff --git a/docs/html/files.html b/docs/html/files.html index ed789536df..0d610a43de 100644 --- a/docs/html/files.html +++ b/docs/html/files.html @@ -120,29 +120,30 @@ o*device_run_length_encode.cuh o*device_scan.cuh o*device_segmented_radix_sort.cuh -o*device_select.cuh -o*device_spmv.cuh -o*tex_obj_input_iterator.cuh -o*tex_ref_input_iterator.cuh -o*thread_load.cuh -o*thread_operators.cuh -o*thread_store.cuh -o*transform_input_iterator.cuh -o*util_allocator.cuh -o*util_arch.cuh -o*util_debug.cuh -o*util_device.cuh -o*util_ptx.cuh -o*util_type.cuh -o*warp_reduce.cuh -\*warp_scan.cuh +o*device_segmented_reduce.cuh +o*device_select.cuh +o*device_spmv.cuh +o*tex_obj_input_iterator.cuh +o*tex_ref_input_iterator.cuh +o*thread_load.cuh +o*thread_operators.cuh +o*thread_store.cuh +o*transform_input_iterator.cuh +o*util_allocator.cuh +o*util_arch.cuh +o*util_debug.cuh +o*util_device.cuh +o*util_ptx.cuh +o*util_type.cuh +o*warp_reduce.cuh +\*warp_scan.cuh
-Generated on Tue Nov 10 2015 13:58:09 for CUB by   +Generated on Mon Dec 14 2015 13:11:20 for CUB by   doxygen 1.8.4
diff --git a/docs/html/functions.html b/docs/html/functions.html index 2ad58e18b8..85b3a9f84b 100644 --- a/docs/html/functions.html +++ b/docs/html/functions.html @@ -96,7 +96,6 @@
  • f
  • h
  • i
    • -
  • k
  • l
  • m
  • n
    • @@ -130,30 +129,23 @@
    Here is a list of all documented class members with links to the class documentation for each member:

    - a -

    -Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/functions_0x62.html b/docs/html/functions_0x62.html index 4d01f0b978..417149c5ed 100644 --- a/docs/html/functions_0x62.html +++ b/docs/html/functions_0x62.html @@ -96,7 +96,6 @@
  • f
  • h
  • i
    • -
  • k
  • l
  • m
  • n
    • @@ -172,7 +171,7 @@

    - b -

    -

    Definition at line 103 of file block_load.cuh.

    +

    Definition at line 104 of file block_load.cuh.

    @@ -598,7 +598,7 @@

    Function Documentation

    -

    Definition at line 132 of file block_load.cuh.

    +

    Definition at line 134 of file block_load.cuh.

    @@ -658,7 +658,7 @@

    Function Documentation

    -

    Definition at line 222 of file block_load.cuh.

    +

    Definition at line 228 of file block_load.cuh.

    @@ -714,7 +714,7 @@

    Function Documentation

    -

    Definition at line 280 of file block_load.cuh.

    +

    Definition at line 286 of file block_load.cuh.

    @@ -777,7 +777,7 @@

    Function Documentation

    -

    Definition at line 312 of file block_load.cuh.

    +

    Definition at line 317 of file block_load.cuh.

    @@ -847,7 +847,7 @@

    Function Documentation

    -

    Definition at line 343 of file block_load.cuh.

    +

    Definition at line 349 of file block_load.cuh.

    @@ -903,7 +903,7 @@

    Function Documentation

    -

    Definition at line 383 of file block_load.cuh.

    +

    Definition at line 392 of file block_load.cuh.

    @@ -966,7 +966,7 @@

    Function Documentation

    -

    Definition at line 417 of file block_load.cuh.

    +

    Definition at line 427 of file block_load.cuh.

    @@ -1036,7 +1036,7 @@

    Function Documentation

    -

    Definition at line 454 of file block_load.cuh.

    +

    Definition at line 465 of file block_load.cuh.

    @@ -1459,7 +1459,7 @@

    Function Documentation

    -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/group___util_iterator.html b/docs/html/group___util_iterator.html index d5ee9f220a..ca7c0b5081 100644 --- a/docs/html/group___util_iterator.html +++ b/docs/html/group___util_iterator.html @@ -97,7 +97,7 @@

    Classes

    class  cub::ArgIndexInputIterator< InputIteratorT, OffsetT > - A random-access input wrapper for pairing dereferenced values with their corresponding indices (forming KeyValuePair tuples). More...
    + A random-access input wrapper for pairing dereferenced values with their corresponding indices (forming KeyValuePair tuples). More...
      class  cub::CacheModifiedInputIterator< MODIFIER, ValueType, OffsetT >  A random-access input wrapper for dereferencing array values using a PTX cache load modifier. More...
    @@ -125,7 +125,7 @@
    -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/group___util_mgmt.html b/docs/html/group___util_mgmt.html index 9b38f3c2c6..863486dab3 100644 --- a/docs/html/group___util_mgmt.html +++ b/docs/html/group___util_mgmt.html @@ -112,10 +112,10 @@ #define CubDebugExit(e)   if (cub::Debug((e), __FILE__, __LINE__)) { exit(1); }  Debug macro with exit.
      - -#define CubLog(format,...)   printf(format,__VA_ARGS__); - Log macro for printf statements.
    -  + +#define _CubLog(format,...)   printf(format,__VA_ARGS__); + Log macro for printf statements.
    +  @@ -190,7 +190,7 @@

    Function Documentation

    Type definition of the EmptyKernel kernel entry point

    Force EmptyKernel<void> to be generated if this class is used

    -

    Definition at line 115 of file util_device.cuh.

    +

    Definition at line 116 of file util_device.cuh.

    @@ -198,7 +198,7 @@

    Function Documentation

    -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/group___util_module.html b/docs/html/group___util_module.html index d867982cc9..8837412284 100644 --- a/docs/html/group___util_module.html +++ b/docs/html/group___util_module.html @@ -114,24 +114,6 @@
    - - - - - - - - - - - - - - - - - - @@ -188,7 +170,7 @@
    -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/group___util_ptx.html b/docs/html/group___util_ptx.html index 6196b73420..971b6951b8 100644 --- a/docs/html/group___util_ptx.html +++ b/docs/html/group___util_ptx.html @@ -216,7 +216,7 @@

    Function Documentation

    -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/group___warp_module.html b/docs/html/group___warp_module.html index 5844e780dc..9247457ef7 100644 --- a/docs/html/group___warp_module.html +++ b/docs/html/group___warp_module.html @@ -351,7 +351,7 @@

    Function Documentation

    -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/hierarchy.html b/docs/html/hierarchy.html index 210ad9ef40..4f8eda17e5 100644 --- a/docs/html/hierarchy.html +++ b/docs/html/hierarchy.html @@ -99,7 +99,7 @@
    This inheritance list is sorted roughly, but not completely, alphabetically:
    [detail level 12]

    Functions

    struct  cub::Equals< A, B >
     Type equality test. More...
     
    struct  cub::NullType
     A simple "NULL" marker type. More...
     
    struct  cub::Int2Type< A >
     Allows for the treatment of an integral constant as a type at compile-time (e.g., to achieve static call dispatch based on constant integral values) More...
     
    struct  cub::CubVector< T, vec_elements >
     Exposes a member typedef Type that names the corresponding CUDA vector type if one exists. Otherwise Type refers to the CubVector structure itself, which will wrap the corresponding x, y, etc. vector fields. More...
     
    struct  cub::Uninitialized< T >
     A storage-backing wrapper that allows types with non-trivial constructors to be aliased in unions. More...
     
    struct  cub::KeyValuePair< _Key, _Value >
     A key identifier paired with a corresponding value. More...
     
    struct  cub::DoubleBuffer< T >
     Double-buffer storage wrapper for multi-pass stream transformations that require more than one storage array for streaming intermediate results back and forth. More...
     
    struct  cub::Log2< N, CURRENT_VAL, COUNT >
     Statically determine log2(N), rounded up. More...
     
    - + - - - - - - - - + - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + - +__forceinline__ KeyValuePair
    +< OffsetT, T > 
    oCcub::ArgIndexInputIterator< InputIteratorT, OffsetT >A random-access input wrapper for pairing dereferenced values with their corresponding indices (forming KeyValuePair tuples)
    oCcub::ArgIndexInputIterator< InputIteratorT, OffsetT >A random-access input wrapper for pairing dereferenced values with their corresponding indices (forming KeyValuePair tuples)
    oCcub::ArgMaxArg max functor (keeps the value and offset of the first occurrence of the larger item)
    oCcub::ArgMinArg min functor (keeps the value and offset of the first occurrence of the smallest item)
    oCcub::BlockDiscontinuity< T, BLOCK_DIM_X, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >The BlockDiscontinuity class provides collective methods for flagging discontinuities within an ordered set of items partitioned across a CUDA thread block.

    @@ -148,40 +148,44 @@
    oCcub::Cast< B >Default cast functor
    oCcub::ConstantInputIterator< ValueType, OffsetT >A random-access input generator for dereferencing a sequence of homogeneous values
    oCcub::CountingInputIterator< ValueType, OffsetT >A random-access input generator for dereferencing a sequence of incrementing integer values
    oCcub::CubVector< T, vec_elements >Exposes a member typedef Type that names the corresponding CUDA vector type if one exists. Otherwise Type refers to the CubVector structure itself, which will wrap the corresponding x, y, etc. vector fields
    oCcub::DeviceHistogramDeviceHistogram provides device-wide parallel operations for constructing histogram(s) from a sequence of samples data residing within device-accessible memory.

    +
    oCcub::DeviceHistogramDeviceHistogram provides device-wide parallel operations for constructing histogram(s) from a sequence of samples data residing within device-accessible memory.

    histogram_logo.png
    oCcub::DevicePartitionDevicePartition provides device-wide, parallel operations for partitioning sequences of data items residing within device-accessible memory.

    +
    oCcub::DevicePartitionDevicePartition provides device-wide, parallel operations for partitioning sequences of data items residing within device-accessible memory.

    partition_logo.png
    oCcub::DeviceRadixSortDeviceRadixSort provides device-wide, parallel operations for computing a radix sort across a sequence of data items residing within device-accessible memory.

    +
    oCcub::DeviceRadixSortDeviceRadixSort provides device-wide, parallel operations for computing a radix sort across a sequence of data items residing within device-accessible memory.

    sorting_logo.png
    oCcub::DeviceReduceDeviceReduce provides device-wide, parallel operations for computing a reduction across a sequence of data items residing within device-accessible memory.

    +
    oCcub::DeviceReduceDeviceReduce provides device-wide, parallel operations for computing a reduction across a sequence of data items residing within device-accessible memory.

    reduce_logo.png
    oCcub::DeviceRunLengthEncodeDeviceRunLengthEncode provides device-wide, parallel operations for demarcating "runs" of same-valued items within a sequence residing within device-accessible memory.

    +
    oCcub::DeviceRunLengthEncodeDeviceRunLengthEncode provides device-wide, parallel operations for demarcating "runs" of same-valued items within a sequence residing within device-accessible memory.

    run_length_encode_logo.png
    oCcub::DeviceScanDeviceScan provides device-wide, parallel operations for computing a prefix scan across a sequence of data items residing within device-accessible memory.

    +
    oCcub::DeviceScanDeviceScan provides device-wide, parallel operations for computing a prefix scan across a sequence of data items residing within device-accessible memory.

    device_scan.png
    oCcub::DeviceSegmentedRadixSortDeviceSegmentedRadixSort provides device-wide, parallel operations for computing a batched radix sort across multiple, non-overlapping sequences of data items residing within device-accessible memory.

    +
    oCcub::DeviceSegmentedRadixSortDeviceSegmentedRadixSort provides device-wide, parallel operations for computing a batched radix sort across multiple, non-overlapping sequences of data items residing within device-accessible memory.

    segmented_sorting_logo.png +
    +
    oCcub::DeviceSegmentedReduceDeviceSegmentedReduce provides device-wide, parallel operations for computing a reduction across multiple sequences of data items residing within device-accessible memory.

    +
    +reduce_logo.png
    oCcub::DeviceSelectDeviceSelect provides device-wide, parallel operations for compacting selected items from sequences of data items residing within device-accessible memory.

    @@ -190,53 +194,48 @@
    oCcub::DeviceSpmvDeviceSpmv provides device-wide parallel operations for performing sparse-matrix * dense-vector multiplication (SpMV)
    oCcub::DoubleBuffer< T >Double-buffer storage wrapper for multi-pass stream transformations that require more than one storage array for streaming intermediate results back and forth
    oCcub::EqualityDefault equality functor
    oCcub::Equals< A, B >Type equality test
    oCcub::If< IF, ThenType, ElseType >Type selection (IF ? ThenType : ElseType)
    oCcub::InequalityDefault inequality functor
    oCcub::InequalityWrapper< EqualityOp >Inequality functor (wraps equality functor)
    oCcub::Int2Type< A >Allows for the treatment of an integral constant as a type at compile-time (e.g., to achieve static call dispatch based on constant integral values)
    oCcub::IsPointer< Tp >Pointer vs. iterator
    oCcub::IsVolatile< Tp >Volatile modifier test
    oCcub::KeyValuePair< _Key, _Value >A key identifier paired with a corresponding value
    oCcub::Log2< N, CURRENT_VAL, COUNT >Statically determine log2(N), rounded up
    oCcub::MaxDefault max functor
    oCcub::MinDefault min functor
    oCcub::NullTypeA simple "NULL" marker type
    oCcub::PowerOfTwo< N >Statically determine if N is a power-of-two
    oCcub::ReduceByKeyOp< ReductionOpT >< Binary reduction operator to apply to values
    oCcub::ReduceBySegmentOp< ReductionOpT >Reduce-by-segment functor
    oCcub::RemoveQualifiers< Tp, Up >Removes const and volatile qualifiers from type Tp
    oCcub::SumDefault sum functor
    oCcub::SwizzleScanOp< ScanOp >Binary operator wrapper for switching non-commutative scan arguments
    oCcub::TexObjInputIterator< T, OffsetT >A random-access input wrapper for dereferencing array values through texture cache. Uses newer Kepler-style texture objects
    oCcub::TexRefInputIterator< T, UNIQUE_ID, OffsetT >A random-access input wrapper for dereferencing array values through texture cache. Uses older Tesla/Fermi-style texture references
    oCcub::TransformInputIterator< ValueType, ConversionOp, InputIteratorT, OffsetT >A random-access input wrapper for transforming dereferenced values
    oCcub::Uninitialized< T >A storage-backing wrapper that allows types with non-trivial constructors to be aliased in unions
    oCcub::Uninitialized< _TempStorage >
    |oCcub::BlockDiscontinuity< T, BLOCK_DIM_X, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorageThe operations exposed by BlockDiscontinuity require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |oCcub::BlockExchange< T, BLOCK_DIM_X, ITEMS_PER_THREAD, WARP_TIME_SLICING, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorageThe operations exposed by BlockExchange require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |oCcub::BlockHistogram< T, BLOCK_DIM_X, ITEMS_PER_THREAD, BINS, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorageThe operations exposed by BlockHistogram require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |oCcub::BlockLoad< InputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::LoadInternal< BLOCK_LOAD_TRANSPOSE, DUMMY >::TempStorageAlias wrapper allowing storage to be unioned
    |oCcub::BlockLoad< InputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE, DUMMY >::TempStorageAlias wrapper allowing storage to be unioned
    |oCcub::BlockLoad< InputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE_TIMESLICED, DUMMY >::TempStorageAlias wrapper allowing storage to be unioned
    |oCcub::BlockLoad< InputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorageThe operations exposed by BlockLoad require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |oCcub::BlockRadixSort< KeyT, BLOCK_DIM_X, ITEMS_PER_THREAD, ValueT, RADIX_BITS, MEMOIZE_OUTER_SCAN, INNER_SCAN_ALGORITHM, SMEM_CONFIG, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorageThe operations exposed by BlockScan require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |oCcub::BlockReduce< T, BLOCK_DIM_X, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorageThe operations exposed by BlockReduce require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |oCcub::BlockScan< T, BLOCK_DIM_X, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorageThe operations exposed by BlockScan require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |oCcub::BlockStore< OutputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::StoreInternal< BLOCK_STORE_TRANSPOSE, DUMMY >::TempStorageAlias wrapper allowing storage to be unioned
    |oCcub::BlockStore< OutputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::StoreInternal< BLOCK_STORE_WARP_TRANSPOSE, DUMMY >::TempStorageAlias wrapper allowing storage to be unioned
    |oCcub::BlockStore< OutputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::StoreInternal< BLOCK_STORE_WARP_TRANSPOSE_TIMESLICED, DUMMY >::TempStorageAlias wrapper allowing storage to be unioned
    |oCcub::BlockStore< OutputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorageThe operations exposed by BlockStore require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |oCcub::WarpReduce< T, LOGICAL_WARP_THREADS, PTX_ARCH >::TempStorageThe operations exposed by WarpReduce require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |\Ccub::WarpScan< T, LOGICAL_WARP_THREADS, PTX_ARCH >::TempStorageThe operations exposed by WarpScan require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    oCcub::WarpReduce< T, LOGICAL_WARP_THREADS, PTX_ARCH >The WarpReduce class provides collective methods for computing a parallel reduction of items partitioned across a CUDA thread warp.

    +
    oCcub::EqualityDefault equality functor
    oCcub::Equals< A, B >Type equality test
    oCcub::If< IF, ThenType, ElseType >Type selection (IF ? ThenType : ElseType)
    oCcub::InequalityDefault inequality functor
    oCcub::InequalityWrapper< EqualityOp >Inequality functor (wraps equality functor)
    oCcub::IsPointer< Tp >Pointer vs. iterator
    oCcub::IsVolatile< Tp >Volatile modifier test
    oCcub::Log2< N, CURRENT_VAL, COUNT >Statically determine log2(N), rounded up
    oCcub::MaxDefault max functor
    oCcub::MinDefault min functor
    oCcub::PowerOfTwo< N >Statically determine if N is a power-of-two
    oCcub::ReduceByKeyOp< ReductionOpT >< Binary reduction operator to apply to values
    oCcub::ReduceBySegmentOp< ReductionOpT >Reduce-by-segment functor
    oCcub::RemoveQualifiers< Tp, Up >Removes const and volatile qualifiers from type Tp
    oCcub::SumDefault sum functor
    oCcub::SwizzleScanOp< ScanOp >Binary operator wrapper for switching non-commutative scan arguments
    oCcub::TexObjInputIterator< T, OffsetT >A random-access input wrapper for dereferencing array values through texture cache. Uses newer Kepler-style texture objects
    oCcub::TexRefInputIterator< T, UNIQUE_ID, OffsetT >A random-access input wrapper for dereferencing array values through texture cache. Uses older Tesla/Fermi-style texture references
    oCcub::TransformInputIterator< ValueType, ConversionOp, InputIteratorT, OffsetT >A random-access input wrapper for transforming dereferenced values
    oCUninitialized
    |oCcub::BlockDiscontinuity< T, BLOCK_DIM_X, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorageThe operations exposed by BlockDiscontinuity require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |oCcub::BlockExchange< T, BLOCK_DIM_X, ITEMS_PER_THREAD, WARP_TIME_SLICING, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorageThe operations exposed by BlockExchange require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |oCcub::BlockHistogram< T, BLOCK_DIM_X, ITEMS_PER_THREAD, BINS, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorageThe operations exposed by BlockHistogram require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |oCcub::BlockLoad< InputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::LoadInternal< BLOCK_LOAD_TRANSPOSE, DUMMY >::TempStorageAlias wrapper allowing storage to be unioned
    |oCcub::BlockLoad< InputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE, DUMMY >::TempStorageAlias wrapper allowing storage to be unioned
    |oCcub::BlockLoad< InputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE_TIMESLICED, DUMMY >::TempStorageAlias wrapper allowing storage to be unioned
    |oCcub::BlockLoad< InputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorageThe operations exposed by BlockLoad require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |oCcub::BlockRadixSort< KeyT, BLOCK_DIM_X, ITEMS_PER_THREAD, ValueT, RADIX_BITS, MEMOIZE_OUTER_SCAN, INNER_SCAN_ALGORITHM, SMEM_CONFIG, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorageThe operations exposed by BlockScan require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |oCcub::BlockReduce< T, BLOCK_DIM_X, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorageThe operations exposed by BlockReduce require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |oCcub::BlockScan< T, BLOCK_DIM_X, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorageThe operations exposed by BlockScan require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |oCcub::BlockStore< OutputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::StoreInternal< BLOCK_STORE_TRANSPOSE, DUMMY >::TempStorageAlias wrapper allowing storage to be unioned
    |oCcub::BlockStore< OutputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::StoreInternal< BLOCK_STORE_WARP_TRANSPOSE, DUMMY >::TempStorageAlias wrapper allowing storage to be unioned
    |oCcub::BlockStore< OutputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::StoreInternal< BLOCK_STORE_WARP_TRANSPOSE_TIMESLICED, DUMMY >::TempStorageAlias wrapper allowing storage to be unioned
    |oCcub::BlockStore< OutputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorageThe operations exposed by BlockStore require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |oCcub::WarpReduce< T, LOGICAL_WARP_THREADS, PTX_ARCH >::TempStorageThe operations exposed by WarpReduce require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    |\Ccub::WarpScan< T, LOGICAL_WARP_THREADS, PTX_ARCH >::TempStorageThe operations exposed by WarpScan require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse
    oCcub::WarpReduce< T, LOGICAL_WARP_THREADS, PTX_ARCH >The WarpReduce class provides collective methods for computing a parallel reduction of items partitioned across a CUDA thread warp.

    warp_reduce_logo.png
    \Ccub::WarpScan< T, LOGICAL_WARP_THREADS, PTX_ARCH >The WarpScan class provides collective methods for computing a parallel prefix scan of items partitioned across a CUDA thread warp.

    +
    \Ccub::WarpScan< T, LOGICAL_WARP_THREADS, PTX_ARCH >The WarpScan class provides collective methods for computing a parallel prefix scan of items partitioned across a CUDA thread warp.

    warp_scan_logo.png
    @@ -247,7 +246,7 @@
    -Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/index.html b/docs/html/index.html index c9b01439aa..67cb8b4a3e 100644 --- a/docs/html/index.html +++ b/docs/html/index.html @@ -115,7 +115,7 @@    -Download CUB v1.5.1 +Download CUB v1.5.0
    		@@ -394,10 +394,10 @@

    CUB releases are labeled using version identifiers having three fields: <epoch>.<feature>.<update>. The epoch field corresponds to support for a major change or update to the CUDA programming model. The feature field corresponds to a stable set of features, functionality, and interface. The update field corresponds to a bug-fix or performance update for that feature set. At the moment, we do not publicly provide non-stable releases such as development snapshots, beta releases or rolling releases. (Feel free to contact us if you would like access to such things.)
    The following table enumerates prior feature releases as well as update versions for the current feature release. - +__forceinline__ KeyValuePair
    +< OffsetT, T > 
    11/13/2015
    +    		12/15/2015
    CUB v1.5.0
    • New Features:
    • See the change-log for further details, including bug-fixes
      • @@ -410,7 +410,7 @@

    • New Features:
      • Updated cub::DeviceHistogram implementation that provides the same "histogram-even" and "histogram-range" functionality as IPP/NPP. Provides extremely fast and, perhaps more importantly, very uniform performance response across diverse real-world datasets, including pathological (homogeneous) sample distributions (resilience)
      • New cub::DeviceSpmv methods for multiplying sparse matrices by dense vectors, load-balanced using a merge-based parallel decomposition.
        • -
      • New cub::DeviceRadixSort sorting entry-points that always return the sorted output into the specified buffer (as opposed to the cub::DoubleBuffer in which it could end up in either buffer)
        • +
      • New cub::DeviceRadixSort sorting entry-points that always return the sorted output into the specified buffer (as opposed to the cub::DoubleBuffer in which it could end up in either buffer)
      • New cub::DeviceRunLengthEncode::NonTrivialRuns for finding the starting offsets and lengths of all non-trivial runs (i.e., length > 1) of keys in a given sequence. (Useful for top-down partitioning algorithms like MSD sorting of very-large keys.)
      • @@ -435,7 +435,7 @@

    04/01/2014
    CUB v1.2.3     		@@ -515,7 +515,7 @@

    -Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/modules.html b/docs/html/modules.html index 22dbfeb450..c63eae0437 100644 --- a/docs/html/modules.html +++ b/docs/html/modules.html @@ -110,7 +110,7 @@
    -Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/namespacecub.html.REMOVED.git-id b/docs/html/namespacecub.html.REMOVED.git-id index c1e9fdd955..808ec4c14b 100644 --- a/docs/html/namespacecub.html.REMOVED.git-id +++ b/docs/html/namespacecub.html.REMOVED.git-id @@ -1 +1 @@ -0d06c26bfa4facfa64eff829bb8b41956185e96e \ No newline at end of file +d46891c8728626f5a8154a296afcfab28abe9920 \ No newline at end of file diff --git a/docs/html/namespacemembers.html b/docs/html/namespacemembers.html index 19cd422a27..8448ca3356 100644 --- a/docs/html/namespacemembers.html +++ b/docs/html/namespacemembers.html @@ -369,7 +369,7 @@

    - w -

      -Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
      diff --git a/docs/html/namespacemembers_enum.html b/docs/html/namespacemembers_enum.html index 624f8ea53b..41a86e1d93 100644 --- a/docs/html/namespacemembers_enum.html +++ b/docs/html/namespacemembers_enum.html @@ -128,7 +128,7 @@
      -Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
      diff --git a/docs/html/namespacemembers_eval.html b/docs/html/namespacemembers_eval.html index c4114b43cd..d1ecfbff65 100644 --- a/docs/html/namespacemembers_eval.html +++ b/docs/html/namespacemembers_eval.html @@ -217,7 +217,7 @@

      - s -

        -Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
        diff --git a/docs/html/namespacemembers_func.html b/docs/html/namespacemembers_func.html index 81adfe8668..6edd277373 100644 --- a/docs/html/namespacemembers_func.html +++ b/docs/html/namespacemembers_func.html @@ -250,7 +250,7 @@

        - w -

          -Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
          diff --git a/docs/html/namespaces.html b/docs/html/namespaces.html index a111806707..53e4c66ae3 100644 --- a/docs/html/namespaces.html +++ b/docs/html/namespaces.html @@ -106,7 +106,7 @@
          -Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
          diff --git a/docs/html/search/all_61.js b/docs/html/search/all_61.js index 86015b2e45..28c4256365 100644 --- a/docs/html/search/all_61.js +++ b/docs/html/search/all_61.js @@ -1,13 +1,10 @@ var searchData= [ - ['alias',['Alias',['../structcub_1_1_uninitialized.html#a790b865325f19ac45cc84d3fed0d3038',1,'cub::Uninitialized']]], - ['align0',['align0',['../structcub_1_1_key_value_pair.html#a6babb4cdcde74f159b5bff9e4a569d83',1,'cub::KeyValuePair']]], - ['alternate',['Alternate',['../structcub_1_1_double_buffer.html#a3895f1d57aeb379bee79de56ace8e35a',1,'cub::DoubleBuffer']]], ['arg_5findex_5finput_5fiterator_2ecuh',['arg_index_input_iterator.cuh',['../arg__index__input__iterator_8cuh.html',1,'']]], - ['argindexinputiterator',['ArgIndexInputIterator',['../classcub_1_1_arg_index_input_iterator.html#a97a6c3755ab132c099e90616e5f67cb6',1,'cub::ArgIndexInputIterator']]], ['argindexinputiterator',['ArgIndexInputIterator',['../classcub_1_1_arg_index_input_iterator.html',1,'cub']]], + ['argindexinputiterator',['ArgIndexInputIterator',['../classcub_1_1_arg_index_input_iterator.html#a97a6c3755ab132c099e90616e5f67cb6',1,'cub::ArgIndexInputIterator']]], ['argmax',['ArgMax',['../structcub_1_1_arg_max.html',1,'cub']]], - ['argmax',['ArgMax',['../structcub_1_1_device_reduce.html#a07a9ecbf0b6db1882107f6adee1c4276',1,'cub::DeviceReduce']]], + ['argmax',['ArgMax',['../structcub_1_1_device_reduce.html#a07a9ecbf0b6db1882107f6adee1c4276',1,'cub::DeviceReduce::ArgMax()'],['../structcub_1_1_device_segmented_reduce.html#ad0e1f7eede9a0f93a379950eac7d8329',1,'cub::DeviceSegmentedReduce::ArgMax()']]], ['argmin',['ArgMin',['../structcub_1_1_arg_min.html',1,'cub']]], - ['argmin',['ArgMin',['../structcub_1_1_device_reduce.html#a6b35963e90120b6d2c76a6068b0340a9',1,'cub::DeviceReduce']]] + ['argmin',['ArgMin',['../structcub_1_1_device_reduce.html#a6b35963e90120b6d2c76a6068b0340a9',1,'cub::DeviceReduce::ArgMin()'],['../structcub_1_1_device_segmented_reduce.html#a085150ad8d55de8665b45a3f69f38bce',1,'cub::DeviceSegmentedReduce::ArgMin()']]] ]; diff --git a/docs/html/search/all_63.js b/docs/html/search/all_63.js index fd0c551ba7..39cdb9f58f 100644 --- a/docs/html/search/all_63.js +++ b/docs/html/search/all_63.js @@ -8,8 +8,8 @@ var searchData= ['cachemodifiedoutputiterator',['CacheModifiedOutputIterator',['../classcub_1_1_cache_modified_output_iterator.html',1,'cub']]], ['cachemodifiedoutputiterator',['CacheModifiedOutputIterator',['../classcub_1_1_cache_modified_output_iterator.html#a8d19a08f1fdc19bf71521beb5ccfd5c7',1,'cub::CacheModifiedOutputIterator']]], ['cachestoremodifier',['CacheStoreModifier',['../group___util_io.html#ga648d25a92a50ca41cf73e93a35f21f37',1,'cub']]], - ['cachingdeviceallocator',['CachingDeviceAllocator',['../structcub_1_1_caching_device_allocator.html#a8819cc293615f15d1f08f41140349b90',1,'cub::CachingDeviceAllocator::CachingDeviceAllocator(unsigned int bin_growth, unsigned int min_bin, unsigned int max_bin, size_t max_cached_bytes, bool skip_cleanup=false)'],['../structcub_1_1_caching_device_allocator.html#ab08a4c3d066ec2303d07363a25466bff',1,'cub::CachingDeviceAllocator::CachingDeviceAllocator(bool skip_cleanup=false)']]], ['cachingdeviceallocator',['CachingDeviceAllocator',['../structcub_1_1_caching_device_allocator.html',1,'cub']]], + ['cachingdeviceallocator',['CachingDeviceAllocator',['../structcub_1_1_caching_device_allocator.html#a8819cc293615f15d1f08f41140349b90',1,'cub::CachingDeviceAllocator::CachingDeviceAllocator(unsigned int bin_growth, unsigned int min_bin, unsigned int max_bin, size_t max_cached_bytes, bool skip_cleanup=false)'],['../structcub_1_1_caching_device_allocator.html#ab08a4c3d066ec2303d07363a25466bff',1,'cub::CachingDeviceAllocator::CachingDeviceAllocator(bool skip_cleanup=false)']]], ['cast',['Cast',['../structcub_1_1_cast.html',1,'cub']]], ['composite',['Composite',['../classcub_1_1_block_histogram.html#a90ef25b7af7ce819367eca662a47dd2d',1,'cub::BlockHistogram']]], ['constant_5finput_5fiterator_2ecuh',['constant_input_iterator.cuh',['../constant__input__iterator_8cuh.html',1,'']]], @@ -22,8 +22,5 @@ var searchData= ['cub',['cub',['../namespacecub.html',1,'']]], ['cub_2ecuh',['cub.cuh',['../cub_8cuh.html',1,'']]], ['cubdebug',['CubDebug',['../group___util_mgmt.html#ga84c3a4c178bf6593e0fad2b763606236',1,'util_debug.cuh']]], - ['cubdebugexit',['CubDebugExit',['../group___util_mgmt.html#ga26211db894893b3cec946e4e537536f8',1,'util_debug.cuh']]], - ['cublog',['CubLog',['../group___util_mgmt.html#ga6788287a780dc10c443aa1ab5ae9f0db',1,'util_debug.cuh']]], - ['cubvector',['CubVector',['../structcub_1_1_cub_vector.html',1,'cub']]], - ['current',['Current',['../structcub_1_1_double_buffer.html#a861d3dff1a70d5e5926057a44d9b8724',1,'cub::DoubleBuffer']]] + ['cubdebugexit',['CubDebugExit',['../group___util_mgmt.html#ga26211db894893b3cec946e4e537536f8',1,'util_debug.cuh']]] ]; diff --git a/docs/html/search/all_64.js b/docs/html/search/all_64.js index 6265c56487..bc22ef1f2c 100644 --- a/docs/html/search/all_64.js +++ b/docs/html/search/all_64.js @@ -1,6 +1,5 @@ var searchData= [ - ['d_5fbuffers',['d_buffers',['../structcub_1_1_double_buffer.html#a38a2d8a9d5a36e9e4b9132166717a0b4',1,'cub::DoubleBuffer']]], ['debug',['Debug',['../group___util_mgmt.html#ga5a175d2a88f63f7f1ab30e8b4f2cfa95',1,'cub']]], ['device_5fhistogram_2ecuh',['device_histogram.cuh',['../device__histogram_8cuh.html',1,'']]], ['device_5fpartition_2ecuh',['device_partition.cuh',['../device__partition_8cuh.html',1,'']]], @@ -9,6 +8,7 @@ var searchData= ['device_5frun_5flength_5fencode_2ecuh',['device_run_length_encode.cuh',['../device__run__length__encode_8cuh.html',1,'']]], ['device_5fscan_2ecuh',['device_scan.cuh',['../device__scan_8cuh.html',1,'']]], ['device_5fsegmented_5fradix_5fsort_2ecuh',['device_segmented_radix_sort.cuh',['../device__segmented__radix__sort_8cuh.html',1,'']]], + ['device_5fsegmented_5freduce_2ecuh',['device_segmented_reduce.cuh',['../device__segmented__reduce_8cuh.html',1,'']]], ['device_5fselect_2ecuh',['device_select.cuh',['../device__select_8cuh.html',1,'']]], ['device_5fspmv_2ecuh',['device_spmv.cuh',['../device__spmv_8cuh.html',1,'']]], ['deviceallocate',['DeviceAllocate',['../structcub_1_1_caching_device_allocator.html#ae1088ac6ba6e5d55832ffbc0b2a5d714',1,'cub::CachingDeviceAllocator::DeviceAllocate(int device, void **d_ptr, size_t bytes, cudaStream_t active_stream=0)'],['../structcub_1_1_caching_device_allocator.html#ab2fe4020cf8bc86ad886c797525ac8ea',1,'cub::CachingDeviceAllocator::DeviceAllocate(void **d_ptr, size_t bytes, cudaStream_t active_stream=0)']]], @@ -21,11 +21,9 @@ var searchData= ['devicerunlengthencode',['DeviceRunLengthEncode',['../structcub_1_1_device_run_length_encode.html',1,'cub']]], ['devicescan',['DeviceScan',['../structcub_1_1_device_scan.html',1,'cub']]], ['devicesegmentedradixsort',['DeviceSegmentedRadixSort',['../structcub_1_1_device_segmented_radix_sort.html',1,'cub']]], + ['devicesegmentedreduce',['DeviceSegmentedReduce',['../structcub_1_1_device_segmented_reduce.html',1,'cub']]], ['deviceselect',['DeviceSelect',['../structcub_1_1_device_select.html',1,'cub']]], ['devicespmv',['DeviceSpmv',['../structcub_1_1_device_spmv.html',1,'cub']]], - ['deviceword',['DeviceWord',['../structcub_1_1_uninitialized.html#a848f6233dbde22d8a42c022d3f0aa6f6',1,'cub::Uninitialized']]], ['difference_5ftype',['difference_type',['../classcub_1_1_arg_index_input_iterator.html#acd5e39570dd51f4883547ef33f9ca8c6',1,'cub::ArgIndexInputIterator::difference_type()'],['../classcub_1_1_cache_modified_input_iterator.html#a268f1e7a4b42a05c5b4d0bca2775c26e',1,'cub::CacheModifiedInputIterator::difference_type()'],['../classcub_1_1_cache_modified_output_iterator.html#ac8cdd3a29db7e398f14e44a7aa054750',1,'cub::CacheModifiedOutputIterator::difference_type()'],['../classcub_1_1_constant_input_iterator.html#ab45cc48afbfda7eaa3b4ea643e719c33',1,'cub::ConstantInputIterator::difference_type()'],['../classcub_1_1_counting_input_iterator.html#ac66dea2a687f0ad6ed0cbcd6217a3029',1,'cub::CountingInputIterator::difference_type()'],['../classcub_1_1_tex_obj_input_iterator.html#a710012acbff3dd0c35822951e28148ea',1,'cub::TexObjInputIterator::difference_type()'],['../classcub_1_1_tex_ref_input_iterator.html#aff1afce146f69adb655e2ff7366b869f',1,'cub::TexRefInputIterator::difference_type()'],['../classcub_1_1_transform_input_iterator.html#ac05064e9ad33dc032452e3d09e8768f6',1,'cub::TransformInputIterator::difference_type()']]], - ['doublebuffer',['DoubleBuffer',['../structcub_1_1_double_buffer.html',1,'cub']]], - ['doublebuffer',['DoubleBuffer',['../structcub_1_1_double_buffer.html#a8d51dcd30484a1f186e056eb7ab09979',1,'cub::DoubleBuffer::DoubleBuffer()'],['../structcub_1_1_double_buffer.html#a100c51f0e1aefdff4cdfe6480c89f59e',1,'cub::DoubleBuffer::DoubleBuffer(T *d_current, T *d_alternate)']]], ['device_2c_20kernel_2c_20and_20storage_20management',['Device, kernel, and storage management',['../group___util_mgmt.html',1,'']]] ]; diff --git a/docs/html/search/all_69.js b/docs/html/search/all_69.js index b0aaecc6f4..03cdac8f56 100644 --- a/docs/html/search/all_69.js +++ b/docs/html/search/all_69.js @@ -9,7 +9,6 @@ var searchData= ['inequalitywrapper',['InequalityWrapper',['../structcub_1_1_inequality_wrapper.html#a933c170bd76eef23017c5ff6ea010f49',1,'cub::InequalityWrapper']]], ['inequalitywrapper',['InequalityWrapper',['../structcub_1_1_inequality_wrapper.html',1,'cub']]], ['inithistogram',['InitHistogram',['../classcub_1_1_block_histogram.html#a080faf88a47e73e8bfa9964cd8f26feb',1,'cub::BlockHistogram']]], - ['int2type',['Int2Type',['../structcub_1_1_int2_type.html',1,'cub']]], ['ispointer',['IsPointer',['../structcub_1_1_is_pointer.html',1,'cub']]], ['isvolatile',['IsVolatile',['../structcub_1_1_is_volatile.html',1,'cub']]], ['iterator_5fcategory',['iterator_category',['../classcub_1_1_arg_index_input_iterator.html#a2cff9aacc1ba59ae9f74735c257261e5',1,'cub::ArgIndexInputIterator::iterator_category()'],['../classcub_1_1_cache_modified_input_iterator.html#ada1fa84e98ba648e275d11e6e8fd4970',1,'cub::CacheModifiedInputIterator::iterator_category()'],['../classcub_1_1_cache_modified_output_iterator.html#af8ebc2633b875aa62423c20f24143938',1,'cub::CacheModifiedOutputIterator::iterator_category()'],['../classcub_1_1_constant_input_iterator.html#ae0ec37dfa33ad77189ffb5153d923364',1,'cub::ConstantInputIterator::iterator_category()'],['../classcub_1_1_counting_input_iterator.html#a984b01f872ab978ad5db527083a4c254',1,'cub::CountingInputIterator::iterator_category()'],['../classcub_1_1_tex_obj_input_iterator.html#a53ae83bf8355764e5ef91a11d9a3c714',1,'cub::TexObjInputIterator::iterator_category()'],['../classcub_1_1_tex_ref_input_iterator.html#a1c95fea4acb0f52261acc21a0daa9056',1,'cub::TexRefInputIterator::iterator_category()'],['../classcub_1_1_transform_input_iterator.html#a6166ef08c614347c39fcba5f0fb776f6',1,'cub::TransformInputIterator::iterator_category()']]] diff --git a/docs/html/search/all_6d.js b/docs/html/search/all_6d.js index b2ef9e1339..4deaba7c06 100644 --- a/docs/html/search/all_6d.js +++ b/docs/html/search/all_6d.js @@ -1,9 +1,9 @@ var searchData= [ ['max',['Max',['../structcub_1_1_max.html',1,'cub']]], - ['max',['Max',['../structcub_1_1_device_reduce.html#a974a241463ca892c8f5e73b879065e48',1,'cub::DeviceReduce']]], + ['max',['Max',['../structcub_1_1_device_reduce.html#a974a241463ca892c8f5e73b879065e48',1,'cub::DeviceReduce::Max()'],['../structcub_1_1_device_segmented_reduce.html#aa1f982f913c95d9974412b8fc0995183',1,'cub::DeviceSegmentedReduce::Max()']]], ['min',['Min',['../structcub_1_1_min.html',1,'cub']]], - ['min',['Min',['../structcub_1_1_device_reduce.html#a14d9c12a1beb9a04f77e903d07fa596a',1,'cub::DeviceReduce']]], + ['min',['Min',['../structcub_1_1_device_reduce.html#a14d9c12a1beb9a04f77e903d07fa596a',1,'cub::DeviceReduce::Min()'],['../structcub_1_1_device_segmented_reduce.html#a2fb8a073bb504afd0e05cd06d008ec29',1,'cub::DeviceSegmentedReduce::Min()']]], ['multihistogrameven',['MultiHistogramEven',['../structcub_1_1_device_histogram.html#a917e507d773ef1b11424ed8bc49e4d95',1,'cub::DeviceHistogram::MultiHistogramEven(void *d_temp_storage, size_t &temp_storage_bytes, SampleIteratorT d_samples, CounterT *d_histogram[NUM_ACTIVE_CHANNELS], int num_levels[NUM_ACTIVE_CHANNELS], LevelT lower_level[NUM_ACTIVE_CHANNELS], LevelT upper_level[NUM_ACTIVE_CHANNELS], OffsetT num_pixels, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_histogram.html#a309c2e30bd62935e1cfcd623fb04b668',1,'cub::DeviceHistogram::MultiHistogramEven(void *d_temp_storage, size_t &temp_storage_bytes, SampleIteratorT d_samples, CounterT *d_histogram[NUM_ACTIVE_CHANNELS], int num_levels[NUM_ACTIVE_CHANNELS], LevelT lower_level[NUM_ACTIVE_CHANNELS], LevelT upper_level[NUM_ACTIVE_CHANNELS], OffsetT num_row_pixels, OffsetT num_rows, size_t row_stride_bytes, cudaStream_t stream=0, bool debug_synchronous=false)']]], ['multihistogramrange',['MultiHistogramRange',['../structcub_1_1_device_histogram.html#a77511d5ae7cc53f0b5c984fd32ad29fe',1,'cub::DeviceHistogram::MultiHistogramRange(void *d_temp_storage, size_t &temp_storage_bytes, SampleIteratorT d_samples, CounterT *d_histogram[NUM_ACTIVE_CHANNELS], int num_levels[NUM_ACTIVE_CHANNELS], LevelT *d_levels[NUM_ACTIVE_CHANNELS], OffsetT num_pixels, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_histogram.html#a5926ff12b29d9e8e67ecf2684071542f',1,'cub::DeviceHistogram::MultiHistogramRange(void *d_temp_storage, size_t &temp_storage_bytes, SampleIteratorT d_samples, CounterT *d_histogram[NUM_ACTIVE_CHANNELS], int num_levels[NUM_ACTIVE_CHANNELS], LevelT *d_levels[NUM_ACTIVE_CHANNELS], OffsetT num_row_pixels, OffsetT num_rows, size_t row_stride_bytes, cudaStream_t stream=0, bool debug_synchronous=false)']]] ]; diff --git a/docs/html/search/all_6e.js b/docs/html/search/all_6e.js index fffd0a9194..4312eb4584 100644 --- a/docs/html/search/all_6e.js +++ b/docs/html/search/all_6e.js @@ -1,5 +1,5 @@ var searchData= [ ['nontrivialruns',['NonTrivialRuns',['../structcub_1_1_device_run_length_encode.html#aa2318dc7a69f28a8c47d417aaf53db3a',1,'cub::DeviceRunLengthEncode']]], - ['nulltype',['NullType',['../structcub_1_1_null_type.html',1,'cub']]] + ['normalize',['normalize',['../classcub_1_1_arg_index_input_iterator.html#aa3dd1dfb19d87d8e0b5fc3c8773fd1dc',1,'cub::ArgIndexInputIterator']]] ]; diff --git a/docs/html/search/all_6f.js b/docs/html/search/all_6f.js index 1f8f9b5687..243751e651 100644 --- a/docs/html/search/all_6f.js +++ b/docs/html/search/all_6f.js @@ -1,7 +1,7 @@ var searchData= [ ['op',['op',['../structcub_1_1_inequality_wrapper.html#a4143eec319a0c231f0f5159fba11371d',1,'cub::InequalityWrapper::op()'],['../structcub_1_1_reduce_by_segment_op.html#a562b1870ca14ca9886ad22535ca673f3',1,'cub::ReduceBySegmentOp::op()'],['../structcub_1_1_reduce_by_key_op.html#a4c6624999354ccf78a94226b9762bdbf',1,'cub::ReduceByKeyOp::op()']]], - ['operator_21_3d',['operator!=',['../structcub_1_1_key_value_pair.html#aa7bcc39b09d285d41c9c0226a49790f2',1,'cub::KeyValuePair::operator!=()'],['../classcub_1_1_arg_index_input_iterator.html#a17b0be0179d38b73e8a612d4d0202772',1,'cub::ArgIndexInputIterator::operator!=()'],['../classcub_1_1_cache_modified_input_iterator.html#acbe543a259aa55230c8a7be052fe311c',1,'cub::CacheModifiedInputIterator::operator!=()'],['../classcub_1_1_cache_modified_output_iterator.html#a9bb801bcb260204ff8312cbb21e77bad',1,'cub::CacheModifiedOutputIterator::operator!=()'],['../classcub_1_1_constant_input_iterator.html#af6345ecd7617d8eb9913d63be52d4a7f',1,'cub::ConstantInputIterator::operator!=()'],['../classcub_1_1_counting_input_iterator.html#af51f1a3ae34158f4f67964e278649240',1,'cub::CountingInputIterator::operator!=()'],['../classcub_1_1_tex_obj_input_iterator.html#af500ff312438db4a22ce466968fd4c28',1,'cub::TexObjInputIterator::operator!=()'],['../classcub_1_1_tex_ref_input_iterator.html#a3be490820de8d3cc7968601fc0fe34ab',1,'cub::TexRefInputIterator::operator!=()'],['../classcub_1_1_transform_input_iterator.html#a5d3245a6a5829ed10ff39af7cfecb3be',1,'cub::TransformInputIterator::operator!=()']]], + ['operator_21_3d',['operator!=',['../classcub_1_1_arg_index_input_iterator.html#a17b0be0179d38b73e8a612d4d0202772',1,'cub::ArgIndexInputIterator::operator!=()'],['../classcub_1_1_cache_modified_input_iterator.html#acbe543a259aa55230c8a7be052fe311c',1,'cub::CacheModifiedInputIterator::operator!=()'],['../classcub_1_1_cache_modified_output_iterator.html#a9bb801bcb260204ff8312cbb21e77bad',1,'cub::CacheModifiedOutputIterator::operator!=()'],['../classcub_1_1_constant_input_iterator.html#af6345ecd7617d8eb9913d63be52d4a7f',1,'cub::ConstantInputIterator::operator!=()'],['../classcub_1_1_counting_input_iterator.html#af51f1a3ae34158f4f67964e278649240',1,'cub::CountingInputIterator::operator!=()'],['../classcub_1_1_tex_obj_input_iterator.html#af500ff312438db4a22ce466968fd4c28',1,'cub::TexObjInputIterator::operator!=()'],['../classcub_1_1_tex_ref_input_iterator.html#a3be490820de8d3cc7968601fc0fe34ab',1,'cub::TexRefInputIterator::operator!=()'],['../classcub_1_1_transform_input_iterator.html#a5d3245a6a5829ed10ff39af7cfecb3be',1,'cub::TransformInputIterator::operator!=()']]], ['operator_28_29',['operator()',['../structcub_1_1_equality.html#a9db81c4cbcf79dbb8087b3c59593cae0',1,'cub::Equality::operator()()'],['../structcub_1_1_inequality.html#ac2b51f35b929dc74ea766b012e89b552',1,'cub::Inequality::operator()()'],['../structcub_1_1_inequality_wrapper.html#a97067089c24f8d6a09c914ab6a163c65',1,'cub::InequalityWrapper::operator()()'],['../structcub_1_1_sum.html#a1edd85dbc039f93c8e45eb2096704a86',1,'cub::Sum::operator()()'],['../structcub_1_1_max.html#ab06fa8091c6aa396fe127f37e0a545d3',1,'cub::Max::operator()()'],['../structcub_1_1_arg_max.html#abc8619b45e188b364d96c0bdf0b29c2d',1,'cub::ArgMax::operator()()'],['../structcub_1_1_min.html#ade40f60337afc51da556ed65d5708136',1,'cub::Min::operator()()'],['../structcub_1_1_arg_min.html#a57bab80de70f6401ea6899ca2488f710',1,'cub::ArgMin::operator()()'],['../structcub_1_1_cast.html#a649573c7f5d02ab09f176b5cc7cf1e5c',1,'cub::Cast::operator()()'],['../classcub_1_1_swizzle_scan_op.html#a9548d484ccfb5161d56d3db1003693e2',1,'cub::SwizzleScanOp::operator()()'],['../structcub_1_1_reduce_by_segment_op.html#aff56752999857ef68029cfa150d88d5e',1,'cub::ReduceBySegmentOp::operator()()'],['../structcub_1_1_reduce_by_key_op.html#ad5e0d929c9cc2447df47c0f2c6409c9a',1,'cub::ReduceByKeyOp::operator()()']]], ['operator_2a',['operator*',['../classcub_1_1_arg_index_input_iterator.html#a04e18ece1d7438dd929665e92091241f',1,'cub::ArgIndexInputIterator::operator*()'],['../classcub_1_1_cache_modified_input_iterator.html#a803d29cb5212f1eaa5f4a34415987768',1,'cub::CacheModifiedInputIterator::operator*()'],['../classcub_1_1_cache_modified_output_iterator.html#a4548b15114d6a8a3f714dcbf0f22e854',1,'cub::CacheModifiedOutputIterator::operator*()'],['../classcub_1_1_constant_input_iterator.html#afd9a8d22abd868d9fb13c235bd800eae',1,'cub::ConstantInputIterator::operator*()'],['../classcub_1_1_counting_input_iterator.html#a7acf3db367c8485a5145d9dbf47be1f7',1,'cub::CountingInputIterator::operator*()'],['../classcub_1_1_tex_obj_input_iterator.html#a8052b0744d233b7f9647d0eb6ba71583',1,'cub::TexObjInputIterator::operator*()'],['../classcub_1_1_tex_ref_input_iterator.html#a78417e7a9f78ded354cc152997c2d7d1',1,'cub::TexRefInputIterator::operator*()'],['../classcub_1_1_transform_input_iterator.html#ab84a2e02eeec401af33903dcb183d5ca',1,'cub::TransformInputIterator::operator*()']]], ['operator_2b',['operator+',['../classcub_1_1_arg_index_input_iterator.html#a1bf6ce760cfa6764c43dd13699601030',1,'cub::ArgIndexInputIterator::operator+()'],['../classcub_1_1_cache_modified_input_iterator.html#a455c2df363449c1b92628de54fb2bad5',1,'cub::CacheModifiedInputIterator::operator+()'],['../classcub_1_1_cache_modified_output_iterator.html#a487a5a434527081b425e5a36ea0d4dd3',1,'cub::CacheModifiedOutputIterator::operator+()'],['../classcub_1_1_constant_input_iterator.html#a1e37cefa0808ae8b8b0853e30013bfba',1,'cub::ConstantInputIterator::operator+()'],['../classcub_1_1_counting_input_iterator.html#a4491f4d5635571964f4d98cf1c3cc753',1,'cub::CountingInputIterator::operator+()'],['../classcub_1_1_tex_obj_input_iterator.html#a1284c246857d30e96e5bc1ea10d68116',1,'cub::TexObjInputIterator::operator+()'],['../classcub_1_1_tex_ref_input_iterator.html#a85fcecba8f3d7b4941b3fa446cddb50a',1,'cub::TexRefInputIterator::operator+()'],['../classcub_1_1_transform_input_iterator.html#a36a3eb7ede1ef22b2316fc281aed52ab',1,'cub::TransformInputIterator::operator+()']]], diff --git a/docs/html/search/all_72.js b/docs/html/search/all_72.js index be8bbdf5b2..9680b9e491 100644 --- a/docs/html/search/all_72.js +++ b/docs/html/search/all_72.js @@ -1,7 +1,7 @@ var searchData= [ - ['reduce',['Reduce',['../classcub_1_1_block_reduce.html#a089953b3bdfe7c48208632d0cc2ac1fb',1,'cub::BlockReduce::Reduce(T input, ReductionOp reduction_op)'],['../classcub_1_1_block_reduce.html#a81878a614ef3b39de654918fc1f6144d',1,'cub::BlockReduce::Reduce(T(&inputs)[ITEMS_PER_THREAD], ReductionOp reduction_op)'],['../classcub_1_1_block_reduce.html#a0e947f6a1d812d21839632b87aaf32e5',1,'cub::BlockReduce::Reduce(T input, ReductionOp reduction_op, int num_valid)'],['../structcub_1_1_device_reduce.html#a326ef5ae888b92442295c26190a6c39c',1,'cub::DeviceReduce::Reduce()'],['../classcub_1_1_warp_reduce.html#a0dd72fc4cf7e1ecf59e8b15bd6819185',1,'cub::WarpReduce::Reduce(T input, ReductionOp reduction_op)'],['../classcub_1_1_warp_reduce.html#ad1ecfeddf0e7fb3f359cf61b60f4745a',1,'cub::WarpReduce::Reduce(T input, ReductionOp reduction_op, int valid_items)']]], - ['reducebykey',['ReduceByKey',['../structcub_1_1_device_reduce.html#a3206c7c11b4d894ca176482e86215b02',1,'cub::DeviceReduce']]], + ['reduce',['Reduce',['../classcub_1_1_block_reduce.html#a089953b3bdfe7c48208632d0cc2ac1fb',1,'cub::BlockReduce::Reduce(T input, ReductionOp reduction_op)'],['../classcub_1_1_block_reduce.html#a81878a614ef3b39de654918fc1f6144d',1,'cub::BlockReduce::Reduce(T(&inputs)[ITEMS_PER_THREAD], ReductionOp reduction_op)'],['../classcub_1_1_block_reduce.html#a0e947f6a1d812d21839632b87aaf32e5',1,'cub::BlockReduce::Reduce(T input, ReductionOp reduction_op, int num_valid)'],['../structcub_1_1_device_reduce.html#aa4adabeb841b852a7a5ecf4f99a2daeb',1,'cub::DeviceReduce::Reduce()'],['../structcub_1_1_device_segmented_reduce.html#ad9b73f245930740c4d8786fc1a812364',1,'cub::DeviceSegmentedReduce::Reduce()'],['../classcub_1_1_warp_reduce.html#a0dd72fc4cf7e1ecf59e8b15bd6819185',1,'cub::WarpReduce::Reduce(T input, ReductionOp reduction_op)'],['../classcub_1_1_warp_reduce.html#ad1ecfeddf0e7fb3f359cf61b60f4745a',1,'cub::WarpReduce::Reduce(T input, ReductionOp reduction_op, int valid_items)']]], + ['reducebykey',['ReduceByKey',['../structcub_1_1_device_reduce.html#a303ae673ac32825f95912b4bfff8bef1',1,'cub::DeviceReduce']]], ['reducebykeyop',['ReduceByKeyOp',['../structcub_1_1_reduce_by_key_op.html',1,'cub']]], ['reducebykeyop',['ReduceByKeyOp',['../structcub_1_1_reduce_by_key_op.html#a88feb445a30081d205a7e1560c07bc6e',1,'cub::ReduceByKeyOp::ReduceByKeyOp()'],['../structcub_1_1_reduce_by_key_op.html#aa9e777450d365effbfc54e4e7700dd24',1,'cub::ReduceByKeyOp::ReduceByKeyOp(ReductionOpT op)']]], ['reducebysegmentop',['ReduceBySegmentOp',['../structcub_1_1_reduce_by_segment_op.html',1,'cub']]], diff --git a/docs/html/search/all_73.js b/docs/html/search/all_73.js index 36fb5b74f9..fbfd9969f9 100644 --- a/docs/html/search/all_73.js +++ b/docs/html/search/all_73.js @@ -6,7 +6,6 @@ var searchData= ['scattertostriped',['ScatterToStriped',['../classcub_1_1_block_exchange.html#a024c9172cda50cd0e2ab364bbe9be63d',1,'cub::BlockExchange::ScatterToStriped(T items[ITEMS_PER_THREAD], OffsetT ranks[ITEMS_PER_THREAD])'],['../classcub_1_1_block_exchange.html#a46a32397168592af2c1050c4061690c7',1,'cub::BlockExchange::ScatterToStriped(T items[ITEMS_PER_THREAD], OffsetT ranks[ITEMS_PER_THREAD], ValidFlag is_valid[ITEMS_PER_THREAD])']]], ['scattertostripedguarded',['ScatterToStripedGuarded',['../classcub_1_1_block_exchange.html#a83cdc3266e21fb11d5c37ac79dc66598',1,'cub::BlockExchange']]], ['segmented_2dproblem_20_28batch_29',['Segmented-problem (batch)',['../group___segmented_module.html',1,'']]], - ['selector',['selector',['../structcub_1_1_double_buffer.html#a9641172c847169904c4054856d7c26f4',1,'cub::DoubleBuffer']]], ['self_5ftype',['self_type',['../classcub_1_1_arg_index_input_iterator.html#a6ad07b9b511ecbd6219756ebeffd6be6',1,'cub::ArgIndexInputIterator::self_type()'],['../classcub_1_1_cache_modified_input_iterator.html#af26facb4f00568a940d5529047bd90d7',1,'cub::CacheModifiedInputIterator::self_type()'],['../classcub_1_1_cache_modified_output_iterator.html#abca48cb8f0172b355ba465ff3c6308c2',1,'cub::CacheModifiedOutputIterator::self_type()'],['../classcub_1_1_constant_input_iterator.html#a0d85b740ebf81e48973e2c6cf9d0e812',1,'cub::ConstantInputIterator::self_type()'],['../classcub_1_1_counting_input_iterator.html#a4e89463441c1c008e7610473ae8d7435',1,'cub::CountingInputIterator::self_type()'],['../classcub_1_1_tex_obj_input_iterator.html#aa7e85f2b2f4d580e3ded497028604a5c',1,'cub::TexObjInputIterator::self_type()'],['../classcub_1_1_tex_ref_input_iterator.html#a03d63027c130152d31623eee579f8506',1,'cub::TexRefInputIterator::self_type()'],['../classcub_1_1_transform_input_iterator.html#ad9e475cc2aae6601422445fff16de9b4',1,'cub::TransformInputIterator::self_type()']]], ['setmaxcachedbytes',['SetMaxCachedBytes',['../structcub_1_1_caching_device_allocator.html#a2216ab13fdeb11ce61f04f69899fff33',1,'cub::CachingDeviceAllocator']]], ['shl_5fadd',['SHL_ADD',['../group___util_ptx.html#ga79b9963a4b033d545e42159e5f0b1621',1,'cub']]], @@ -22,9 +21,8 @@ var searchData= ['sortdescendingblockedtostriped',['SortDescendingBlockedToStriped',['../classcub_1_1_block_radix_sort.html#ae89db4d439f0996b0dd7ad4b105e4f1d',1,'cub::BlockRadixSort::SortDescendingBlockedToStriped(KeyT(&keys)[ITEMS_PER_THREAD], int begin_bit=0, int end_bit=sizeof(KeyT)*8)'],['../classcub_1_1_block_radix_sort.html#ab56d4d4c3a8da9ff718906bfc5d815c2',1,'cub::BlockRadixSort::SortDescendingBlockedToStriped(KeyT(&keys)[ITEMS_PER_THREAD], ValueT(&values)[ITEMS_PER_THREAD], int begin_bit=0, int end_bit=sizeof(KeyT)*8)']]], ['sortkeys',['SortKeys',['../structcub_1_1_device_radix_sort.html#a4f555afa8ac2949d9fef49fad52a50d6',1,'cub::DeviceRadixSort::SortKeys(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_radix_sort.html#ad7b5dbc2e3fd44c21193f2b792706191',1,'cub::DeviceRadixSort::SortKeys(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#ab6a917a29d441021949e197d3a639fba',1,'cub::DeviceSegmentedRadixSort::SortKeys(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#ab8b433b55358ac507a7fcbba933cdbac',1,'cub::DeviceSegmentedRadixSort::SortKeys(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)']]], ['sortkeysdescending',['SortKeysDescending',['../structcub_1_1_device_radix_sort.html#a24761009c4cc15fd2e54cb72663af0ef',1,'cub::DeviceRadixSort::SortKeysDescending(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_radix_sort.html#a95a8939332405efec4527442c26c2628',1,'cub::DeviceRadixSort::SortKeysDescending(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#a49ccbb6ca7a1d26e4b01e68e8da8a701',1,'cub::DeviceSegmentedRadixSort::SortKeysDescending(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#a5c21fba171c718aaf9b3b3c27bda6a94',1,'cub::DeviceSegmentedRadixSort::SortKeysDescending(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)']]], - ['sortpairs',['SortPairs',['../structcub_1_1_device_radix_sort.html#a31d7224d3f6c6a9309a0b84571f25eb9',1,'cub::DeviceRadixSort::SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, ValueT *d_values_in, ValueT *d_values_out, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_radix_sort.html#a0e0f38bafdc30403a68b3ff5c7b80027',1,'cub::DeviceRadixSort::SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#a4c291958575f14acc6a9e6d3e2ea6597',1,'cub::DeviceSegmentedRadixSort::SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, ValueT *d_values_in, ValueT *d_values_out, int num_items, OffsetT num_segments, OffsetT *d_begin_offsets, OffsetT *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#a4cffada9bba553f9871a34d926bbb148',1,'cub::DeviceSegmentedRadixSort::SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)']]], + ['sortpairs',['SortPairs',['../structcub_1_1_device_radix_sort.html#a31d7224d3f6c6a9309a0b84571f25eb9',1,'cub::DeviceRadixSort::SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, ValueT *d_values_in, ValueT *d_values_out, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_radix_sort.html#a0e0f38bafdc30403a68b3ff5c7b80027',1,'cub::DeviceRadixSort::SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#a6770adfa8e5a99c8015d9e6ab5ed8ca0',1,'cub::DeviceSegmentedRadixSort::SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, ValueT *d_values_in, ValueT *d_values_out, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#a4cffada9bba553f9871a34d926bbb148',1,'cub::DeviceSegmentedRadixSort::SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)']]], ['sortpairsdescending',['SortPairsDescending',['../structcub_1_1_device_radix_sort.html#add6a87f54c8058edba4b9e875bb0626a',1,'cub::DeviceRadixSort::SortPairsDescending(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, ValueT *d_values_in, ValueT *d_values_out, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_radix_sort.html#aea53a40e665f2d5ed683a6655a3d188e',1,'cub::DeviceRadixSort::SortPairsDescending(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#a39de3d48166582a802cc7df1481d3ff4',1,'cub::DeviceSegmentedRadixSort::SortPairsDescending(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, ValueT *d_values_in, ValueT *d_values_out, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#aafce5852dfbfbeef027493c3791d6347',1,'cub::DeviceSegmentedRadixSort::SortPairsDescending(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)']]], - ['storage',['storage',['../structcub_1_1_uninitialized.html#a5fa7311d943222333e8c87497ff8e782',1,'cub::Uninitialized']]], ['store',['Store',['../classcub_1_1_block_store.html#a86fd777fd9bef8264787d756b16303ed',1,'cub::BlockStore::Store(OutputIteratorT block_itr, T(&items)[ITEMS_PER_THREAD])'],['../classcub_1_1_block_store.html#a67d0aa8fcf37f92b2074e986581ffdf5',1,'cub::BlockStore::Store(OutputIteratorT block_itr, T(&items)[ITEMS_PER_THREAD], int valid_items)']]], ['store_5fcg',['STORE_CG',['../group___util_io.html#gga648d25a92a50ca41cf73e93a35f21f37aacea07ea298b89dd1962a40b4823652d',1,'cub']]], ['store_5fcs',['STORE_CS',['../group___util_io.html#gga648d25a92a50ca41cf73e93a35f21f37a00ae8891d1acad179d134fdd60d7839b',1,'cub']]], @@ -37,8 +35,8 @@ var searchData= ['storedirectstriped',['StoreDirectStriped',['../group___util_io.html#gac688ffdb1ecacc06375295947144d233',1,'cub::StoreDirectStriped(int linear_tid, OutputIteratorT block_itr, T(&items)[ITEMS_PER_THREAD])'],['../group___util_io.html#ga5af198f11043a66ebfaab51c8f4f6fc9',1,'cub::StoreDirectStriped(int linear_tid, OutputIteratorT block_itr, T(&items)[ITEMS_PER_THREAD], int valid_items)']]], ['storedirectwarpstriped',['StoreDirectWarpStriped',['../group___util_io.html#ga9b2bb8f452cd26b5e297c9d5924fcf03',1,'cub::StoreDirectWarpStriped(int linear_tid, OutputIteratorT block_itr, T(&items)[ITEMS_PER_THREAD])'],['../group___util_io.html#ga3978707feae6ad3e1b6467a6e525bc09',1,'cub::StoreDirectWarpStriped(int linear_tid, OutputIteratorT block_itr, T(&items)[ITEMS_PER_THREAD], int valid_items)']]], ['stripedtoblocked',['StripedToBlocked',['../classcub_1_1_block_exchange.html#a2855471bbbcc4d66ac6a29d35a040e0c',1,'cub::BlockExchange']]], - ['sum',['Sum',['../classcub_1_1_block_reduce.html#a7632bd9c8950dd6a3528ca99fa3f0890',1,'cub::BlockReduce::Sum(T input)'],['../classcub_1_1_block_reduce.html#ac5d4591d9513f08b180d4112cb0c4c51',1,'cub::BlockReduce::Sum(T(&inputs)[ITEMS_PER_THREAD])'],['../classcub_1_1_block_reduce.html#a33ddffdde07275ab0c4e1bf61b0d9409',1,'cub::BlockReduce::Sum(T input, int num_valid)'],['../structcub_1_1_device_reduce.html#ab7f21e8255eb842aaf74305975ae607f',1,'cub::DeviceReduce::Sum()'],['../classcub_1_1_warp_scan.html#a25bd83f795e88b9260ec2bcbf846fb20',1,'cub::WarpScan::Sum()'],['../classcub_1_1_warp_reduce.html#abe4aeeabf8859a7582a0b5858b84ee7a',1,'cub::WarpReduce::Sum(T input)'],['../classcub_1_1_warp_reduce.html#ad9c4a8d85a7795cf220713f362c36f30',1,'cub::WarpReduce::Sum(T input, int valid_items)']]], ['sum',['Sum',['../structcub_1_1_sum.html',1,'cub']]], - ['swizzlescanop',['SwizzleScanOp',['../classcub_1_1_swizzle_scan_op.html',1,'cub']]], - ['swizzlescanop',['SwizzleScanOp',['../classcub_1_1_swizzle_scan_op.html#ae81a38aa9d94025da72544b9e0dc611b',1,'cub::SwizzleScanOp']]] + ['sum',['Sum',['../classcub_1_1_block_reduce.html#a7632bd9c8950dd6a3528ca99fa3f0890',1,'cub::BlockReduce::Sum(T input)'],['../classcub_1_1_block_reduce.html#ac5d4591d9513f08b180d4112cb0c4c51',1,'cub::BlockReduce::Sum(T(&inputs)[ITEMS_PER_THREAD])'],['../classcub_1_1_block_reduce.html#a33ddffdde07275ab0c4e1bf61b0d9409',1,'cub::BlockReduce::Sum(T input, int num_valid)'],['../structcub_1_1_device_reduce.html#ab7f21e8255eb842aaf74305975ae607f',1,'cub::DeviceReduce::Sum()'],['../structcub_1_1_device_segmented_reduce.html#aefdf8fcdfb5e5d76459ee222360924eb',1,'cub::DeviceSegmentedReduce::Sum()'],['../classcub_1_1_warp_scan.html#a25bd83f795e88b9260ec2bcbf846fb20',1,'cub::WarpScan::Sum()'],['../classcub_1_1_warp_reduce.html#abe4aeeabf8859a7582a0b5858b84ee7a',1,'cub::WarpReduce::Sum(T input)'],['../classcub_1_1_warp_reduce.html#ad9c4a8d85a7795cf220713f362c36f30',1,'cub::WarpReduce::Sum(T input, int valid_items)']]], + ['swizzlescanop',['SwizzleScanOp',['../classcub_1_1_swizzle_scan_op.html#ae81a38aa9d94025da72544b9e0dc611b',1,'cub::SwizzleScanOp']]], + ['swizzlescanop',['SwizzleScanOp',['../classcub_1_1_swizzle_scan_op.html',1,'cub']]] ]; diff --git a/docs/html/search/all_74.js b/docs/html/search/all_74.js index b6d6e32a8b..10ec064243 100644 --- a/docs/html/search/all_74.js +++ b/docs/html/search/all_74.js @@ -2,22 +2,22 @@ var searchData= [ ['tailsegmentedreduce',['TailSegmentedReduce',['../classcub_1_1_warp_reduce.html#ab89e501348f1ae4ade9a21c4e88d2d6c',1,'cub::WarpReduce']]], ['tailsegmentedsum',['TailSegmentedSum',['../classcub_1_1_warp_reduce.html#a4c73c5bb3636224abb3fe3e1e5e05705',1,'cub::WarpReduce']]], - ['tempstorage',['TempStorage',['../structcub_1_1_block_histogram_1_1_temp_storage.html',1,'cub::BlockHistogram']]], - ['tempstorage',['TempStorage',['../structcub_1_1_block_scan_1_1_temp_storage.html',1,'cub::BlockScan']]], - ['tempstorage',['TempStorage',['../structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___t_r_a_n_s_p_o_s_e_00_01_d_u_m_m_y_01_4_1_1_temp_storage.html',1,'cub::BlockLoad::LoadInternal< BLOCK_LOAD_TRANSPOSE, DUMMY >']]], + ['tempstorage',['TempStorage',['../structcub_1_1_block_load_1_1_temp_storage.html',1,'cub::BlockLoad']]], ['tempstorage',['TempStorage',['../structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_e4c36dfe8f549604998f6c46cc8fbd1d.html',1,'cub::BlockLoad::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE_TIMESLICED, DUMMY >']]], - ['tempstorage',['TempStorage',['../structcub_1_1_block_radix_sort_1_1_temp_storage.html',1,'cub::BlockRadixSort']]], + ['tempstorage',['TempStorage',['../structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___t_r_a_n_s_p_o_s_e_00_01_d_u_m_m_y_01_4_1_1_temp_storage.html',1,'cub::BlockLoad::LoadInternal< BLOCK_LOAD_TRANSPOSE, DUMMY >']]], ['tempstorage',['TempStorage',['../structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___t_r_a_n_s_p_o_s_e_00_09dfae03f13932c7dbdb41be30a5767ba.html',1,'cub::BlockStore::StoreInternal< BLOCK_STORE_TRANSPOSE, DUMMY >']]], - ['tempstorage',['TempStorage',['../structcub_1_1_warp_reduce_1_1_temp_storage.html',1,'cub::WarpReduce']]], - ['tempstorage',['TempStorage',['../structcub_1_1_block_discontinuity_1_1_temp_storage.html',1,'cub::BlockDiscontinuity']]], - ['tempstorage',['TempStorage',['../structcub_1_1_block_load_1_1_temp_storage.html',1,'cub::BlockLoad']]], ['tempstorage',['TempStorage',['../structcub_1_1_block_store_1_1_temp_storage.html',1,'cub::BlockStore']]], + ['tempstorage',['TempStorage',['../structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___w_a_r_p___t_r_a_n_s_p_8d170856b7ed1df0ed565731a681b449.html',1,'cub::BlockStore::StoreInternal< BLOCK_STORE_WARP_TRANSPOSE, DUMMY >']]], + ['tempstorage',['TempStorage',['../structcub_1_1_block_scan_1_1_temp_storage.html',1,'cub::BlockScan']]], ['tempstorage',['TempStorage',['../structcub_1_1_block_reduce_1_1_temp_storage.html',1,'cub::BlockReduce']]], ['tempstorage',['TempStorage',['../structcub_1_1_block_exchange_1_1_temp_storage.html',1,'cub::BlockExchange']]], ['tempstorage',['TempStorage',['../structcub_1_1_warp_scan_1_1_temp_storage.html',1,'cub::WarpScan']]], - ['tempstorage',['TempStorage',['../structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___w_a_r_p___t_r_a_n_s_p_8d170856b7ed1df0ed565731a681b449.html',1,'cub::BlockStore::StoreInternal< BLOCK_STORE_WARP_TRANSPOSE, DUMMY >']]], - ['tempstorage',['TempStorage',['../structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_402c3164d23f1ec647db5dad06a54584.html',1,'cub::BlockLoad::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE, DUMMY >']]], ['tempstorage',['TempStorage',['../structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___w_a_r_p___t_r_a_n_s_p_263becc1ca5b47586740c2f7bb0d0145.html',1,'cub::BlockStore::StoreInternal< BLOCK_STORE_WARP_TRANSPOSE_TIMESLICED, DUMMY >']]], + ['tempstorage',['TempStorage',['../structcub_1_1_block_histogram_1_1_temp_storage.html',1,'cub::BlockHistogram']]], + ['tempstorage',['TempStorage',['../structcub_1_1_warp_reduce_1_1_temp_storage.html',1,'cub::WarpReduce']]], + ['tempstorage',['TempStorage',['../structcub_1_1_block_radix_sort_1_1_temp_storage.html',1,'cub::BlockRadixSort']]], + ['tempstorage',['TempStorage',['../structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_402c3164d23f1ec647db5dad06a54584.html',1,'cub::BlockLoad::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE, DUMMY >']]], + ['tempstorage',['TempStorage',['../structcub_1_1_block_discontinuity_1_1_temp_storage.html',1,'cub::BlockDiscontinuity']]], ['tex_5fobj_5finput_5fiterator_2ecuh',['tex_obj_input_iterator.cuh',['../tex__obj__input__iterator_8cuh.html',1,'']]], ['tex_5fref_5finput_5fiterator_2ecuh',['tex_ref_input_iterator.cuh',['../tex__ref__input__iterator_8cuh.html',1,'']]], ['texobjinputiterator',['TexObjInputIterator',['../classcub_1_1_tex_obj_input_iterator.html',1,'cub']]], diff --git a/docs/html/search/all_75.js b/docs/html/search/all_75.js index ad7083c3d9..786e910f7a 100644 --- a/docs/html/search/all_75.js +++ b/docs/html/search/all_75.js @@ -1,8 +1,6 @@ var searchData= [ ['unbindtexture',['UnbindTexture',['../classcub_1_1_tex_obj_input_iterator.html#a35b98dd35b014a052f996ee37afe1b67',1,'cub::TexObjInputIterator::UnbindTexture()'],['../classcub_1_1_tex_ref_input_iterator.html#a5f451683e1f40e5f702251b707e87956',1,'cub::TexRefInputIterator::UnbindTexture()']]], - ['uninitialized',['Uninitialized',['../structcub_1_1_uninitialized.html',1,'cub']]], - ['uninitialized_3c_20_5ftempstorage_20_3e',['Uninitialized< _TempStorage >',['../structcub_1_1_uninitialized.html',1,'cub']]], ['unique',['Unique',['../structcub_1_1_device_select.html#a5e9ea8f2581326391ea63e1808ba514d',1,'cub::DeviceSelect']]], ['util_5farch_2ecuh',['util_arch.cuh',['../util__arch_8cuh.html',1,'']]], ['util_5fdebug_2ecuh',['util_debug.cuh',['../util__debug_8cuh.html',1,'']]], diff --git a/docs/html/search/all_76.js b/docs/html/search/all_76.js index 8cf51fc0d2..bce599db9a 100644 --- a/docs/html/search/all_76.js +++ b/docs/html/search/all_76.js @@ -1,5 +1,4 @@ var searchData= [ - ['value',['Value',['../structcub_1_1_key_value_pair.html#a9fd385872c09fd3757e9ba59b2754955',1,'cub::KeyValuePair::Value()'],['../structcub_1_1_key_value_pair.html#a468bef1440a66f45bd9b5193594bf1a4',1,'cub::KeyValuePair::value()']]], ['value_5ftype',['value_type',['../classcub_1_1_arg_index_input_iterator.html#a6c40ea1dc7c0923b9010f1e938e07d22',1,'cub::ArgIndexInputIterator::value_type()'],['../classcub_1_1_cache_modified_input_iterator.html#a60689c564e2ade39722947eeaf40156a',1,'cub::CacheModifiedInputIterator::value_type()'],['../classcub_1_1_cache_modified_output_iterator.html#aba997f95620d692cbad05c74ef169fa8',1,'cub::CacheModifiedOutputIterator::value_type()'],['../classcub_1_1_constant_input_iterator.html#a72a3996ca30a1b2eb6f676923a2ee3ce',1,'cub::ConstantInputIterator::value_type()'],['../classcub_1_1_counting_input_iterator.html#a3afabb6a47c8cf7526220eb817e2a97f',1,'cub::CountingInputIterator::value_type()'],['../classcub_1_1_tex_obj_input_iterator.html#a4405a9e3d39593b7c468629dff098144',1,'cub::TexObjInputIterator::value_type()'],['../classcub_1_1_tex_ref_input_iterator.html#a1de28970e3874202646e11984a0d9026',1,'cub::TexRefInputIterator::value_type()'],['../classcub_1_1_transform_input_iterator.html#a8230414e069b99a279e2afabed83fe52',1,'cub::TransformInputIterator::value_type()']]] ]; diff --git a/docs/html/search/classes_63.js b/docs/html/search/classes_63.js index 31c37749ac..fdd21ccef6 100644 --- a/docs/html/search/classes_63.js +++ b/docs/html/search/classes_63.js @@ -5,6 +5,5 @@ var searchData= ['cachingdeviceallocator',['CachingDeviceAllocator',['../structcub_1_1_caching_device_allocator.html',1,'cub']]], ['cast',['Cast',['../structcub_1_1_cast.html',1,'cub']]], ['constantinputiterator',['ConstantInputIterator',['../classcub_1_1_constant_input_iterator.html',1,'cub']]], - ['countinginputiterator',['CountingInputIterator',['../classcub_1_1_counting_input_iterator.html',1,'cub']]], - ['cubvector',['CubVector',['../structcub_1_1_cub_vector.html',1,'cub']]] + ['countinginputiterator',['CountingInputIterator',['../classcub_1_1_counting_input_iterator.html',1,'cub']]] ]; diff --git a/docs/html/search/classes_64.js b/docs/html/search/classes_64.js index 0bf30727e5..7bd3da01db 100644 --- a/docs/html/search/classes_64.js +++ b/docs/html/search/classes_64.js @@ -7,7 +7,7 @@ var searchData= ['devicerunlengthencode',['DeviceRunLengthEncode',['../structcub_1_1_device_run_length_encode.html',1,'cub']]], ['devicescan',['DeviceScan',['../structcub_1_1_device_scan.html',1,'cub']]], ['devicesegmentedradixsort',['DeviceSegmentedRadixSort',['../structcub_1_1_device_segmented_radix_sort.html',1,'cub']]], + ['devicesegmentedreduce',['DeviceSegmentedReduce',['../structcub_1_1_device_segmented_reduce.html',1,'cub']]], ['deviceselect',['DeviceSelect',['../structcub_1_1_device_select.html',1,'cub']]], - ['devicespmv',['DeviceSpmv',['../structcub_1_1_device_spmv.html',1,'cub']]], - ['doublebuffer',['DoubleBuffer',['../structcub_1_1_double_buffer.html',1,'cub']]] + ['devicespmv',['DeviceSpmv',['../structcub_1_1_device_spmv.html',1,'cub']]] ]; diff --git a/docs/html/search/classes_69.js b/docs/html/search/classes_69.js index 446b93123f..53e48188e6 100644 --- a/docs/html/search/classes_69.js +++ b/docs/html/search/classes_69.js @@ -3,7 +3,6 @@ var searchData= ['if',['If',['../structcub_1_1_if.html',1,'cub']]], ['inequality',['Inequality',['../structcub_1_1_inequality.html',1,'cub']]], ['inequalitywrapper',['InequalityWrapper',['../structcub_1_1_inequality_wrapper.html',1,'cub']]], - ['int2type',['Int2Type',['../structcub_1_1_int2_type.html',1,'cub']]], ['ispointer',['IsPointer',['../structcub_1_1_is_pointer.html',1,'cub']]], ['isvolatile',['IsVolatile',['../structcub_1_1_is_volatile.html',1,'cub']]] ]; diff --git a/docs/html/search/classes_74.js b/docs/html/search/classes_74.js index cc574a1735..9cc662147d 100644 --- a/docs/html/search/classes_74.js +++ b/docs/html/search/classes_74.js @@ -1,21 +1,21 @@ var searchData= [ - ['tempstorage',['TempStorage',['../structcub_1_1_block_reduce_1_1_temp_storage.html',1,'cub::BlockReduce']]], - ['tempstorage',['TempStorage',['../structcub_1_1_block_histogram_1_1_temp_storage.html',1,'cub::BlockHistogram']]], - ['tempstorage',['TempStorage',['../structcub_1_1_block_scan_1_1_temp_storage.html',1,'cub::BlockScan']]], - ['tempstorage',['TempStorage',['../structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___t_r_a_n_s_p_o_s_e_00_01_d_u_m_m_y_01_4_1_1_temp_storage.html',1,'cub::BlockLoad::LoadInternal< BLOCK_LOAD_TRANSPOSE, DUMMY >']]], - ['tempstorage',['TempStorage',['../structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_e4c36dfe8f549604998f6c46cc8fbd1d.html',1,'cub::BlockLoad::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE_TIMESLICED, DUMMY >']]], ['tempstorage',['TempStorage',['../structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___w_a_r_p___t_r_a_n_s_p_263becc1ca5b47586740c2f7bb0d0145.html',1,'cub::BlockStore::StoreInternal< BLOCK_STORE_WARP_TRANSPOSE_TIMESLICED, DUMMY >']]], - ['tempstorage',['TempStorage',['../structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_402c3164d23f1ec647db5dad06a54584.html',1,'cub::BlockLoad::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE, DUMMY >']]], + ['tempstorage',['TempStorage',['../structcub_1_1_block_load_1_1_temp_storage.html',1,'cub::BlockLoad']]], + ['tempstorage',['TempStorage',['../structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_e4c36dfe8f549604998f6c46cc8fbd1d.html',1,'cub::BlockLoad::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE_TIMESLICED, DUMMY >']]], + ['tempstorage',['TempStorage',['../structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___t_r_a_n_s_p_o_s_e_00_01_d_u_m_m_y_01_4_1_1_temp_storage.html',1,'cub::BlockLoad::LoadInternal< BLOCK_LOAD_TRANSPOSE, DUMMY >']]], + ['tempstorage',['TempStorage',['../structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___t_r_a_n_s_p_o_s_e_00_09dfae03f13932c7dbdb41be30a5767ba.html',1,'cub::BlockStore::StoreInternal< BLOCK_STORE_TRANSPOSE, DUMMY >']]], ['tempstorage',['TempStorage',['../structcub_1_1_block_discontinuity_1_1_temp_storage.html',1,'cub::BlockDiscontinuity']]], + ['tempstorage',['TempStorage',['../structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_402c3164d23f1ec647db5dad06a54584.html',1,'cub::BlockLoad::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE, DUMMY >']]], + ['tempstorage',['TempStorage',['../structcub_1_1_block_reduce_1_1_temp_storage.html',1,'cub::BlockReduce']]], + ['tempstorage',['TempStorage',['../structcub_1_1_block_histogram_1_1_temp_storage.html',1,'cub::BlockHistogram']]], + ['tempstorage',['TempStorage',['../structcub_1_1_warp_scan_1_1_temp_storage.html',1,'cub::WarpScan']]], ['tempstorage',['TempStorage',['../structcub_1_1_block_exchange_1_1_temp_storage.html',1,'cub::BlockExchange']]], ['tempstorage',['TempStorage',['../structcub_1_1_block_store_1_1_temp_storage.html',1,'cub::BlockStore']]], - ['tempstorage',['TempStorage',['../structcub_1_1_block_load_1_1_temp_storage.html',1,'cub::BlockLoad']]], - ['tempstorage',['TempStorage',['../structcub_1_1_block_radix_sort_1_1_temp_storage.html',1,'cub::BlockRadixSort']]], - ['tempstorage',['TempStorage',['../structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___t_r_a_n_s_p_o_s_e_00_09dfae03f13932c7dbdb41be30a5767ba.html',1,'cub::BlockStore::StoreInternal< BLOCK_STORE_TRANSPOSE, DUMMY >']]], ['tempstorage',['TempStorage',['../structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___w_a_r_p___t_r_a_n_s_p_8d170856b7ed1df0ed565731a681b449.html',1,'cub::BlockStore::StoreInternal< BLOCK_STORE_WARP_TRANSPOSE, DUMMY >']]], + ['tempstorage',['TempStorage',['../structcub_1_1_block_radix_sort_1_1_temp_storage.html',1,'cub::BlockRadixSort']]], + ['tempstorage',['TempStorage',['../structcub_1_1_block_scan_1_1_temp_storage.html',1,'cub::BlockScan']]], ['tempstorage',['TempStorage',['../structcub_1_1_warp_reduce_1_1_temp_storage.html',1,'cub::WarpReduce']]], - ['tempstorage',['TempStorage',['../structcub_1_1_warp_scan_1_1_temp_storage.html',1,'cub::WarpScan']]], ['texobjinputiterator',['TexObjInputIterator',['../classcub_1_1_tex_obj_input_iterator.html',1,'cub']]], ['texrefinputiterator',['TexRefInputIterator',['../classcub_1_1_tex_ref_input_iterator.html',1,'cub']]], ['transforminputiterator',['TransformInputIterator',['../classcub_1_1_transform_input_iterator.html',1,'cub']]] diff --git a/docs/html/search/files_64.js b/docs/html/search/files_64.js index 7ca198e4a7..31d4e1a095 100644 --- a/docs/html/search/files_64.js +++ b/docs/html/search/files_64.js @@ -7,6 +7,7 @@ var searchData= ['device_5frun_5flength_5fencode_2ecuh',['device_run_length_encode.cuh',['../device__run__length__encode_8cuh.html',1,'']]], ['device_5fscan_2ecuh',['device_scan.cuh',['../device__scan_8cuh.html',1,'']]], ['device_5fsegmented_5fradix_5fsort_2ecuh',['device_segmented_radix_sort.cuh',['../device__segmented__radix__sort_8cuh.html',1,'']]], + ['device_5fsegmented_5freduce_2ecuh',['device_segmented_reduce.cuh',['../device__segmented__reduce_8cuh.html',1,'']]], ['device_5fselect_2ecuh',['device_select.cuh',['../device__select_8cuh.html',1,'']]], ['device_5fspmv_2ecuh',['device_spmv.cuh',['../device__spmv_8cuh.html',1,'']]] ]; diff --git a/docs/html/search/functions_61.js b/docs/html/search/functions_61.js index 79643ee95d..5e92c1d5da 100644 --- a/docs/html/search/functions_61.js +++ b/docs/html/search/functions_61.js @@ -1,8 +1,6 @@ var searchData= [ - ['alias',['Alias',['../structcub_1_1_uninitialized.html#a790b865325f19ac45cc84d3fed0d3038',1,'cub::Uninitialized']]], - ['alternate',['Alternate',['../structcub_1_1_double_buffer.html#a3895f1d57aeb379bee79de56ace8e35a',1,'cub::DoubleBuffer']]], ['argindexinputiterator',['ArgIndexInputIterator',['../classcub_1_1_arg_index_input_iterator.html#a97a6c3755ab132c099e90616e5f67cb6',1,'cub::ArgIndexInputIterator']]], - ['argmax',['ArgMax',['../structcub_1_1_device_reduce.html#a07a9ecbf0b6db1882107f6adee1c4276',1,'cub::DeviceReduce']]], - ['argmin',['ArgMin',['../structcub_1_1_device_reduce.html#a6b35963e90120b6d2c76a6068b0340a9',1,'cub::DeviceReduce']]] + ['argmax',['ArgMax',['../structcub_1_1_device_reduce.html#a07a9ecbf0b6db1882107f6adee1c4276',1,'cub::DeviceReduce::ArgMax()'],['../structcub_1_1_device_segmented_reduce.html#ad0e1f7eede9a0f93a379950eac7d8329',1,'cub::DeviceSegmentedReduce::ArgMax()']]], + ['argmin',['ArgMin',['../structcub_1_1_device_reduce.html#a6b35963e90120b6d2c76a6068b0340a9',1,'cub::DeviceReduce::ArgMin()'],['../structcub_1_1_device_segmented_reduce.html#a085150ad8d55de8665b45a3f69f38bce',1,'cub::DeviceSegmentedReduce::ArgMin()']]] ]; diff --git a/docs/html/search/functions_63.js b/docs/html/search/functions_63.js index 66860b1520..9232b52ce1 100644 --- a/docs/html/search/functions_63.js +++ b/docs/html/search/functions_63.js @@ -6,6 +6,5 @@ var searchData= ['composite',['Composite',['../classcub_1_1_block_histogram.html#a90ef25b7af7ce819367eca662a47dd2d',1,'cub::BlockHistogram']]], ['constantinputiterator',['ConstantInputIterator',['../classcub_1_1_constant_input_iterator.html#a1237efdec602999d589afd3c898b7494',1,'cub::ConstantInputIterator']]], ['countinginputiterator',['CountingInputIterator',['../classcub_1_1_counting_input_iterator.html#ac51f0117505e3b13f5cf8771d99ea623',1,'cub::CountingInputIterator']]], - ['csrmv',['CsrMV',['../structcub_1_1_device_spmv.html#abbcd4c04d8bbbcdfabc1eb62f860b8b2',1,'cub::DeviceSpmv']]], - ['current',['Current',['../structcub_1_1_double_buffer.html#a861d3dff1a70d5e5926057a44d9b8724',1,'cub::DoubleBuffer']]] + ['csrmv',['CsrMV',['../structcub_1_1_device_spmv.html#abbcd4c04d8bbbcdfabc1eb62f860b8b2',1,'cub::DeviceSpmv']]] ]; diff --git a/docs/html/search/functions_64.js b/docs/html/search/functions_64.js index 1c026b445d..c9f9ea3d6e 100644 --- a/docs/html/search/functions_64.js +++ b/docs/html/search/functions_64.js @@ -2,6 +2,5 @@ var searchData= [ ['debug',['Debug',['../group___util_mgmt.html#ga5a175d2a88f63f7f1ab30e8b4f2cfa95',1,'cub']]], ['deviceallocate',['DeviceAllocate',['../structcub_1_1_caching_device_allocator.html#ae1088ac6ba6e5d55832ffbc0b2a5d714',1,'cub::CachingDeviceAllocator::DeviceAllocate(int device, void **d_ptr, size_t bytes, cudaStream_t active_stream=0)'],['../structcub_1_1_caching_device_allocator.html#ab2fe4020cf8bc86ad886c797525ac8ea',1,'cub::CachingDeviceAllocator::DeviceAllocate(void **d_ptr, size_t bytes, cudaStream_t active_stream=0)']]], - ['devicefree',['DeviceFree',['../structcub_1_1_caching_device_allocator.html#a66e72cc3cc7d0dbd59148ac7c4ba0de6',1,'cub::CachingDeviceAllocator::DeviceFree(int device, void *d_ptr)'],['../structcub_1_1_caching_device_allocator.html#adbf65c59172b140420636e150325deeb',1,'cub::CachingDeviceAllocator::DeviceFree(void *d_ptr)']]], - ['doublebuffer',['DoubleBuffer',['../structcub_1_1_double_buffer.html#a8d51dcd30484a1f186e056eb7ab09979',1,'cub::DoubleBuffer::DoubleBuffer()'],['../structcub_1_1_double_buffer.html#a100c51f0e1aefdff4cdfe6480c89f59e',1,'cub::DoubleBuffer::DoubleBuffer(T *d_current, T *d_alternate)']]] + ['devicefree',['DeviceFree',['../structcub_1_1_caching_device_allocator.html#a66e72cc3cc7d0dbd59148ac7c4ba0de6',1,'cub::CachingDeviceAllocator::DeviceFree(int device, void *d_ptr)'],['../structcub_1_1_caching_device_allocator.html#adbf65c59172b140420636e150325deeb',1,'cub::CachingDeviceAllocator::DeviceFree(void *d_ptr)']]] ]; diff --git a/docs/html/search/functions_6d.js b/docs/html/search/functions_6d.js index 3a0a2471bd..967c2ce048 100644 --- a/docs/html/search/functions_6d.js +++ b/docs/html/search/functions_6d.js @@ -1,7 +1,7 @@ var searchData= [ - ['max',['Max',['../structcub_1_1_device_reduce.html#a974a241463ca892c8f5e73b879065e48',1,'cub::DeviceReduce']]], - ['min',['Min',['../structcub_1_1_device_reduce.html#a14d9c12a1beb9a04f77e903d07fa596a',1,'cub::DeviceReduce']]], + ['max',['Max',['../structcub_1_1_device_reduce.html#a974a241463ca892c8f5e73b879065e48',1,'cub::DeviceReduce::Max()'],['../structcub_1_1_device_segmented_reduce.html#aa1f982f913c95d9974412b8fc0995183',1,'cub::DeviceSegmentedReduce::Max()']]], + ['min',['Min',['../structcub_1_1_device_reduce.html#a14d9c12a1beb9a04f77e903d07fa596a',1,'cub::DeviceReduce::Min()'],['../structcub_1_1_device_segmented_reduce.html#a2fb8a073bb504afd0e05cd06d008ec29',1,'cub::DeviceSegmentedReduce::Min()']]], ['multihistogrameven',['MultiHistogramEven',['../structcub_1_1_device_histogram.html#a917e507d773ef1b11424ed8bc49e4d95',1,'cub::DeviceHistogram::MultiHistogramEven(void *d_temp_storage, size_t &temp_storage_bytes, SampleIteratorT d_samples, CounterT *d_histogram[NUM_ACTIVE_CHANNELS], int num_levels[NUM_ACTIVE_CHANNELS], LevelT lower_level[NUM_ACTIVE_CHANNELS], LevelT upper_level[NUM_ACTIVE_CHANNELS], OffsetT num_pixels, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_histogram.html#a309c2e30bd62935e1cfcd623fb04b668',1,'cub::DeviceHistogram::MultiHistogramEven(void *d_temp_storage, size_t &temp_storage_bytes, SampleIteratorT d_samples, CounterT *d_histogram[NUM_ACTIVE_CHANNELS], int num_levels[NUM_ACTIVE_CHANNELS], LevelT lower_level[NUM_ACTIVE_CHANNELS], LevelT upper_level[NUM_ACTIVE_CHANNELS], OffsetT num_row_pixels, OffsetT num_rows, size_t row_stride_bytes, cudaStream_t stream=0, bool debug_synchronous=false)']]], ['multihistogramrange',['MultiHistogramRange',['../structcub_1_1_device_histogram.html#a77511d5ae7cc53f0b5c984fd32ad29fe',1,'cub::DeviceHistogram::MultiHistogramRange(void *d_temp_storage, size_t &temp_storage_bytes, SampleIteratorT d_samples, CounterT *d_histogram[NUM_ACTIVE_CHANNELS], int num_levels[NUM_ACTIVE_CHANNELS], LevelT *d_levels[NUM_ACTIVE_CHANNELS], OffsetT num_pixels, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_histogram.html#a5926ff12b29d9e8e67ecf2684071542f',1,'cub::DeviceHistogram::MultiHistogramRange(void *d_temp_storage, size_t &temp_storage_bytes, SampleIteratorT d_samples, CounterT *d_histogram[NUM_ACTIVE_CHANNELS], int num_levels[NUM_ACTIVE_CHANNELS], LevelT *d_levels[NUM_ACTIVE_CHANNELS], OffsetT num_row_pixels, OffsetT num_rows, size_t row_stride_bytes, cudaStream_t stream=0, bool debug_synchronous=false)']]] ]; diff --git a/docs/html/search/functions_6e.js b/docs/html/search/functions_6e.js index 159a543681..4312eb4584 100644 --- a/docs/html/search/functions_6e.js +++ b/docs/html/search/functions_6e.js @@ -1,4 +1,5 @@ var searchData= [ - ['nontrivialruns',['NonTrivialRuns',['../structcub_1_1_device_run_length_encode.html#aa2318dc7a69f28a8c47d417aaf53db3a',1,'cub::DeviceRunLengthEncode']]] + ['nontrivialruns',['NonTrivialRuns',['../structcub_1_1_device_run_length_encode.html#aa2318dc7a69f28a8c47d417aaf53db3a',1,'cub::DeviceRunLengthEncode']]], + ['normalize',['normalize',['../classcub_1_1_arg_index_input_iterator.html#aa3dd1dfb19d87d8e0b5fc3c8773fd1dc',1,'cub::ArgIndexInputIterator']]] ]; diff --git a/docs/html/search/functions_6f.js b/docs/html/search/functions_6f.js index 41bc282ea6..c1e9c8366c 100644 --- a/docs/html/search/functions_6f.js +++ b/docs/html/search/functions_6f.js @@ -1,6 +1,6 @@ var searchData= [ - ['operator_21_3d',['operator!=',['../structcub_1_1_key_value_pair.html#aa7bcc39b09d285d41c9c0226a49790f2',1,'cub::KeyValuePair::operator!=()'],['../classcub_1_1_arg_index_input_iterator.html#a17b0be0179d38b73e8a612d4d0202772',1,'cub::ArgIndexInputIterator::operator!=()'],['../classcub_1_1_cache_modified_input_iterator.html#acbe543a259aa55230c8a7be052fe311c',1,'cub::CacheModifiedInputIterator::operator!=()'],['../classcub_1_1_cache_modified_output_iterator.html#a9bb801bcb260204ff8312cbb21e77bad',1,'cub::CacheModifiedOutputIterator::operator!=()'],['../classcub_1_1_constant_input_iterator.html#af6345ecd7617d8eb9913d63be52d4a7f',1,'cub::ConstantInputIterator::operator!=()'],['../classcub_1_1_counting_input_iterator.html#af51f1a3ae34158f4f67964e278649240',1,'cub::CountingInputIterator::operator!=()'],['../classcub_1_1_tex_obj_input_iterator.html#af500ff312438db4a22ce466968fd4c28',1,'cub::TexObjInputIterator::operator!=()'],['../classcub_1_1_tex_ref_input_iterator.html#a3be490820de8d3cc7968601fc0fe34ab',1,'cub::TexRefInputIterator::operator!=()'],['../classcub_1_1_transform_input_iterator.html#a5d3245a6a5829ed10ff39af7cfecb3be',1,'cub::TransformInputIterator::operator!=()']]], + ['operator_21_3d',['operator!=',['../classcub_1_1_arg_index_input_iterator.html#a17b0be0179d38b73e8a612d4d0202772',1,'cub::ArgIndexInputIterator::operator!=()'],['../classcub_1_1_cache_modified_input_iterator.html#acbe543a259aa55230c8a7be052fe311c',1,'cub::CacheModifiedInputIterator::operator!=()'],['../classcub_1_1_cache_modified_output_iterator.html#a9bb801bcb260204ff8312cbb21e77bad',1,'cub::CacheModifiedOutputIterator::operator!=()'],['../classcub_1_1_constant_input_iterator.html#af6345ecd7617d8eb9913d63be52d4a7f',1,'cub::ConstantInputIterator::operator!=()'],['../classcub_1_1_counting_input_iterator.html#af51f1a3ae34158f4f67964e278649240',1,'cub::CountingInputIterator::operator!=()'],['../classcub_1_1_tex_obj_input_iterator.html#af500ff312438db4a22ce466968fd4c28',1,'cub::TexObjInputIterator::operator!=()'],['../classcub_1_1_tex_ref_input_iterator.html#a3be490820de8d3cc7968601fc0fe34ab',1,'cub::TexRefInputIterator::operator!=()'],['../classcub_1_1_transform_input_iterator.html#a5d3245a6a5829ed10ff39af7cfecb3be',1,'cub::TransformInputIterator::operator!=()']]], ['operator_28_29',['operator()',['../structcub_1_1_equality.html#a9db81c4cbcf79dbb8087b3c59593cae0',1,'cub::Equality::operator()()'],['../structcub_1_1_inequality.html#ac2b51f35b929dc74ea766b012e89b552',1,'cub::Inequality::operator()()'],['../structcub_1_1_inequality_wrapper.html#a97067089c24f8d6a09c914ab6a163c65',1,'cub::InequalityWrapper::operator()()'],['../structcub_1_1_sum.html#a1edd85dbc039f93c8e45eb2096704a86',1,'cub::Sum::operator()()'],['../structcub_1_1_max.html#ab06fa8091c6aa396fe127f37e0a545d3',1,'cub::Max::operator()()'],['../structcub_1_1_arg_max.html#abc8619b45e188b364d96c0bdf0b29c2d',1,'cub::ArgMax::operator()()'],['../structcub_1_1_min.html#ade40f60337afc51da556ed65d5708136',1,'cub::Min::operator()()'],['../structcub_1_1_arg_min.html#a57bab80de70f6401ea6899ca2488f710',1,'cub::ArgMin::operator()()'],['../structcub_1_1_cast.html#a649573c7f5d02ab09f176b5cc7cf1e5c',1,'cub::Cast::operator()()'],['../classcub_1_1_swizzle_scan_op.html#a9548d484ccfb5161d56d3db1003693e2',1,'cub::SwizzleScanOp::operator()()'],['../structcub_1_1_reduce_by_segment_op.html#aff56752999857ef68029cfa150d88d5e',1,'cub::ReduceBySegmentOp::operator()()'],['../structcub_1_1_reduce_by_key_op.html#ad5e0d929c9cc2447df47c0f2c6409c9a',1,'cub::ReduceByKeyOp::operator()()']]], ['operator_2a',['operator*',['../classcub_1_1_arg_index_input_iterator.html#a04e18ece1d7438dd929665e92091241f',1,'cub::ArgIndexInputIterator::operator*()'],['../classcub_1_1_cache_modified_input_iterator.html#a803d29cb5212f1eaa5f4a34415987768',1,'cub::CacheModifiedInputIterator::operator*()'],['../classcub_1_1_cache_modified_output_iterator.html#a4548b15114d6a8a3f714dcbf0f22e854',1,'cub::CacheModifiedOutputIterator::operator*()'],['../classcub_1_1_constant_input_iterator.html#afd9a8d22abd868d9fb13c235bd800eae',1,'cub::ConstantInputIterator::operator*()'],['../classcub_1_1_counting_input_iterator.html#a7acf3db367c8485a5145d9dbf47be1f7',1,'cub::CountingInputIterator::operator*()'],['../classcub_1_1_tex_obj_input_iterator.html#a8052b0744d233b7f9647d0eb6ba71583',1,'cub::TexObjInputIterator::operator*()'],['../classcub_1_1_tex_ref_input_iterator.html#a78417e7a9f78ded354cc152997c2d7d1',1,'cub::TexRefInputIterator::operator*()'],['../classcub_1_1_transform_input_iterator.html#ab84a2e02eeec401af33903dcb183d5ca',1,'cub::TransformInputIterator::operator*()']]], ['operator_2b',['operator+',['../classcub_1_1_arg_index_input_iterator.html#a1bf6ce760cfa6764c43dd13699601030',1,'cub::ArgIndexInputIterator::operator+()'],['../classcub_1_1_cache_modified_input_iterator.html#a455c2df363449c1b92628de54fb2bad5',1,'cub::CacheModifiedInputIterator::operator+()'],['../classcub_1_1_cache_modified_output_iterator.html#a487a5a434527081b425e5a36ea0d4dd3',1,'cub::CacheModifiedOutputIterator::operator+()'],['../classcub_1_1_constant_input_iterator.html#a1e37cefa0808ae8b8b0853e30013bfba',1,'cub::ConstantInputIterator::operator+()'],['../classcub_1_1_counting_input_iterator.html#a4491f4d5635571964f4d98cf1c3cc753',1,'cub::CountingInputIterator::operator+()'],['../classcub_1_1_tex_obj_input_iterator.html#a1284c246857d30e96e5bc1ea10d68116',1,'cub::TexObjInputIterator::operator+()'],['../classcub_1_1_tex_ref_input_iterator.html#a85fcecba8f3d7b4941b3fa446cddb50a',1,'cub::TexRefInputIterator::operator+()'],['../classcub_1_1_transform_input_iterator.html#a36a3eb7ede1ef22b2316fc281aed52ab',1,'cub::TransformInputIterator::operator+()']]], diff --git a/docs/html/search/functions_72.js b/docs/html/search/functions_72.js index 5859a95e43..6cf024ba18 100644 --- a/docs/html/search/functions_72.js +++ b/docs/html/search/functions_72.js @@ -1,7 +1,7 @@ var searchData= [ - ['reduce',['Reduce',['../classcub_1_1_block_reduce.html#a089953b3bdfe7c48208632d0cc2ac1fb',1,'cub::BlockReduce::Reduce(T input, ReductionOp reduction_op)'],['../classcub_1_1_block_reduce.html#a81878a614ef3b39de654918fc1f6144d',1,'cub::BlockReduce::Reduce(T(&inputs)[ITEMS_PER_THREAD], ReductionOp reduction_op)'],['../classcub_1_1_block_reduce.html#a0e947f6a1d812d21839632b87aaf32e5',1,'cub::BlockReduce::Reduce(T input, ReductionOp reduction_op, int num_valid)'],['../structcub_1_1_device_reduce.html#a326ef5ae888b92442295c26190a6c39c',1,'cub::DeviceReduce::Reduce()'],['../classcub_1_1_warp_reduce.html#a0dd72fc4cf7e1ecf59e8b15bd6819185',1,'cub::WarpReduce::Reduce(T input, ReductionOp reduction_op)'],['../classcub_1_1_warp_reduce.html#ad1ecfeddf0e7fb3f359cf61b60f4745a',1,'cub::WarpReduce::Reduce(T input, ReductionOp reduction_op, int valid_items)']]], - ['reducebykey',['ReduceByKey',['../structcub_1_1_device_reduce.html#a3206c7c11b4d894ca176482e86215b02',1,'cub::DeviceReduce']]], + ['reduce',['Reduce',['../classcub_1_1_block_reduce.html#a089953b3bdfe7c48208632d0cc2ac1fb',1,'cub::BlockReduce::Reduce(T input, ReductionOp reduction_op)'],['../classcub_1_1_block_reduce.html#a81878a614ef3b39de654918fc1f6144d',1,'cub::BlockReduce::Reduce(T(&inputs)[ITEMS_PER_THREAD], ReductionOp reduction_op)'],['../classcub_1_1_block_reduce.html#a0e947f6a1d812d21839632b87aaf32e5',1,'cub::BlockReduce::Reduce(T input, ReductionOp reduction_op, int num_valid)'],['../structcub_1_1_device_reduce.html#aa4adabeb841b852a7a5ecf4f99a2daeb',1,'cub::DeviceReduce::Reduce()'],['../structcub_1_1_device_segmented_reduce.html#ad9b73f245930740c4d8786fc1a812364',1,'cub::DeviceSegmentedReduce::Reduce()'],['../classcub_1_1_warp_reduce.html#a0dd72fc4cf7e1ecf59e8b15bd6819185',1,'cub::WarpReduce::Reduce(T input, ReductionOp reduction_op)'],['../classcub_1_1_warp_reduce.html#ad1ecfeddf0e7fb3f359cf61b60f4745a',1,'cub::WarpReduce::Reduce(T input, ReductionOp reduction_op, int valid_items)']]], + ['reducebykey',['ReduceByKey',['../structcub_1_1_device_reduce.html#a303ae673ac32825f95912b4bfff8bef1',1,'cub::DeviceReduce']]], ['reducebykeyop',['ReduceByKeyOp',['../structcub_1_1_reduce_by_key_op.html#a88feb445a30081d205a7e1560c07bc6e',1,'cub::ReduceByKeyOp::ReduceByKeyOp()'],['../structcub_1_1_reduce_by_key_op.html#aa9e777450d365effbfc54e4e7700dd24',1,'cub::ReduceByKeyOp::ReduceByKeyOp(ReductionOpT op)']]], ['reducebysegmentop',['ReduceBySegmentOp',['../structcub_1_1_reduce_by_segment_op.html#a6f6fe38f76f8f50eff19ea20105cb1b3',1,'cub::ReduceBySegmentOp::ReduceBySegmentOp()'],['../structcub_1_1_reduce_by_segment_op.html#aa01e5bafddc8cf5b6df1601ba783b17e',1,'cub::ReduceBySegmentOp::ReduceBySegmentOp(ReductionOpT op)']]], ['rowmajortid',['RowMajorTid',['../group___util_ptx.html#gaa3f839b109cc6dc9d9ece4f1acf7d2ce',1,'cub']]] diff --git a/docs/html/search/functions_73.js b/docs/html/search/functions_73.js index 9a364f3591..b63a6587cf 100644 --- a/docs/html/search/functions_73.js +++ b/docs/html/search/functions_73.js @@ -17,7 +17,7 @@ var searchData= ['sortdescendingblockedtostriped',['SortDescendingBlockedToStriped',['../classcub_1_1_block_radix_sort.html#ae89db4d439f0996b0dd7ad4b105e4f1d',1,'cub::BlockRadixSort::SortDescendingBlockedToStriped(KeyT(&keys)[ITEMS_PER_THREAD], int begin_bit=0, int end_bit=sizeof(KeyT)*8)'],['../classcub_1_1_block_radix_sort.html#ab56d4d4c3a8da9ff718906bfc5d815c2',1,'cub::BlockRadixSort::SortDescendingBlockedToStriped(KeyT(&keys)[ITEMS_PER_THREAD], ValueT(&values)[ITEMS_PER_THREAD], int begin_bit=0, int end_bit=sizeof(KeyT)*8)']]], ['sortkeys',['SortKeys',['../structcub_1_1_device_radix_sort.html#a4f555afa8ac2949d9fef49fad52a50d6',1,'cub::DeviceRadixSort::SortKeys(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_radix_sort.html#ad7b5dbc2e3fd44c21193f2b792706191',1,'cub::DeviceRadixSort::SortKeys(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#ab6a917a29d441021949e197d3a639fba',1,'cub::DeviceSegmentedRadixSort::SortKeys(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#ab8b433b55358ac507a7fcbba933cdbac',1,'cub::DeviceSegmentedRadixSort::SortKeys(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)']]], ['sortkeysdescending',['SortKeysDescending',['../structcub_1_1_device_radix_sort.html#a24761009c4cc15fd2e54cb72663af0ef',1,'cub::DeviceRadixSort::SortKeysDescending(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_radix_sort.html#a95a8939332405efec4527442c26c2628',1,'cub::DeviceRadixSort::SortKeysDescending(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#a49ccbb6ca7a1d26e4b01e68e8da8a701',1,'cub::DeviceSegmentedRadixSort::SortKeysDescending(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#a5c21fba171c718aaf9b3b3c27bda6a94',1,'cub::DeviceSegmentedRadixSort::SortKeysDescending(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)']]], - ['sortpairs',['SortPairs',['../structcub_1_1_device_radix_sort.html#a31d7224d3f6c6a9309a0b84571f25eb9',1,'cub::DeviceRadixSort::SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, ValueT *d_values_in, ValueT *d_values_out, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_radix_sort.html#a0e0f38bafdc30403a68b3ff5c7b80027',1,'cub::DeviceRadixSort::SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#a4c291958575f14acc6a9e6d3e2ea6597',1,'cub::DeviceSegmentedRadixSort::SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, ValueT *d_values_in, ValueT *d_values_out, int num_items, OffsetT num_segments, OffsetT *d_begin_offsets, OffsetT *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#a4cffada9bba553f9871a34d926bbb148',1,'cub::DeviceSegmentedRadixSort::SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)']]], + ['sortpairs',['SortPairs',['../structcub_1_1_device_radix_sort.html#a31d7224d3f6c6a9309a0b84571f25eb9',1,'cub::DeviceRadixSort::SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, ValueT *d_values_in, ValueT *d_values_out, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_radix_sort.html#a0e0f38bafdc30403a68b3ff5c7b80027',1,'cub::DeviceRadixSort::SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#a6770adfa8e5a99c8015d9e6ab5ed8ca0',1,'cub::DeviceSegmentedRadixSort::SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, ValueT *d_values_in, ValueT *d_values_out, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#a4cffada9bba553f9871a34d926bbb148',1,'cub::DeviceSegmentedRadixSort::SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)']]], ['sortpairsdescending',['SortPairsDescending',['../structcub_1_1_device_radix_sort.html#add6a87f54c8058edba4b9e875bb0626a',1,'cub::DeviceRadixSort::SortPairsDescending(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, ValueT *d_values_in, ValueT *d_values_out, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_radix_sort.html#aea53a40e665f2d5ed683a6655a3d188e',1,'cub::DeviceRadixSort::SortPairsDescending(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#a39de3d48166582a802cc7df1481d3ff4',1,'cub::DeviceSegmentedRadixSort::SortPairsDescending(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, ValueT *d_values_in, ValueT *d_values_out, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)'],['../structcub_1_1_device_segmented_radix_sort.html#aafce5852dfbfbeef027493c3791d6347',1,'cub::DeviceSegmentedRadixSort::SortPairsDescending(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)']]], ['store',['Store',['../classcub_1_1_block_store.html#a86fd777fd9bef8264787d756b16303ed',1,'cub::BlockStore::Store(OutputIteratorT block_itr, T(&items)[ITEMS_PER_THREAD])'],['../classcub_1_1_block_store.html#a67d0aa8fcf37f92b2074e986581ffdf5',1,'cub::BlockStore::Store(OutputIteratorT block_itr, T(&items)[ITEMS_PER_THREAD], int valid_items)']]], ['storedirectblocked',['StoreDirectBlocked',['../group___util_io.html#ga58460e08f9c9c2560ab311c30ab58aca',1,'cub::StoreDirectBlocked(int linear_tid, OutputIteratorT block_itr, T(&items)[ITEMS_PER_THREAD])'],['../group___util_io.html#ga732f5998fb8698032d3c598944eb28a8',1,'cub::StoreDirectBlocked(int linear_tid, OutputIteratorT block_itr, T(&items)[ITEMS_PER_THREAD], int valid_items)']]], @@ -25,6 +25,6 @@ var searchData= ['storedirectstriped',['StoreDirectStriped',['../group___util_io.html#gac688ffdb1ecacc06375295947144d233',1,'cub::StoreDirectStriped(int linear_tid, OutputIteratorT block_itr, T(&items)[ITEMS_PER_THREAD])'],['../group___util_io.html#ga5af198f11043a66ebfaab51c8f4f6fc9',1,'cub::StoreDirectStriped(int linear_tid, OutputIteratorT block_itr, T(&items)[ITEMS_PER_THREAD], int valid_items)']]], ['storedirectwarpstriped',['StoreDirectWarpStriped',['../group___util_io.html#ga9b2bb8f452cd26b5e297c9d5924fcf03',1,'cub::StoreDirectWarpStriped(int linear_tid, OutputIteratorT block_itr, T(&items)[ITEMS_PER_THREAD])'],['../group___util_io.html#ga3978707feae6ad3e1b6467a6e525bc09',1,'cub::StoreDirectWarpStriped(int linear_tid, OutputIteratorT block_itr, T(&items)[ITEMS_PER_THREAD], int valid_items)']]], ['stripedtoblocked',['StripedToBlocked',['../classcub_1_1_block_exchange.html#a2855471bbbcc4d66ac6a29d35a040e0c',1,'cub::BlockExchange']]], - ['sum',['Sum',['../classcub_1_1_block_reduce.html#a7632bd9c8950dd6a3528ca99fa3f0890',1,'cub::BlockReduce::Sum(T input)'],['../classcub_1_1_block_reduce.html#ac5d4591d9513f08b180d4112cb0c4c51',1,'cub::BlockReduce::Sum(T(&inputs)[ITEMS_PER_THREAD])'],['../classcub_1_1_block_reduce.html#a33ddffdde07275ab0c4e1bf61b0d9409',1,'cub::BlockReduce::Sum(T input, int num_valid)'],['../structcub_1_1_device_reduce.html#ab7f21e8255eb842aaf74305975ae607f',1,'cub::DeviceReduce::Sum()'],['../classcub_1_1_warp_scan.html#a25bd83f795e88b9260ec2bcbf846fb20',1,'cub::WarpScan::Sum()'],['../classcub_1_1_warp_reduce.html#abe4aeeabf8859a7582a0b5858b84ee7a',1,'cub::WarpReduce::Sum(T input)'],['../classcub_1_1_warp_reduce.html#ad9c4a8d85a7795cf220713f362c36f30',1,'cub::WarpReduce::Sum(T input, int valid_items)']]], + ['sum',['Sum',['../classcub_1_1_block_reduce.html#a7632bd9c8950dd6a3528ca99fa3f0890',1,'cub::BlockReduce::Sum(T input)'],['../classcub_1_1_block_reduce.html#ac5d4591d9513f08b180d4112cb0c4c51',1,'cub::BlockReduce::Sum(T(&inputs)[ITEMS_PER_THREAD])'],['../classcub_1_1_block_reduce.html#a33ddffdde07275ab0c4e1bf61b0d9409',1,'cub::BlockReduce::Sum(T input, int num_valid)'],['../structcub_1_1_device_reduce.html#ab7f21e8255eb842aaf74305975ae607f',1,'cub::DeviceReduce::Sum()'],['../structcub_1_1_device_segmented_reduce.html#aefdf8fcdfb5e5d76459ee222360924eb',1,'cub::DeviceSegmentedReduce::Sum()'],['../classcub_1_1_warp_scan.html#a25bd83f795e88b9260ec2bcbf846fb20',1,'cub::WarpScan::Sum()'],['../classcub_1_1_warp_reduce.html#abe4aeeabf8859a7582a0b5858b84ee7a',1,'cub::WarpReduce::Sum(T input)'],['../classcub_1_1_warp_reduce.html#ad9c4a8d85a7795cf220713f362c36f30',1,'cub::WarpReduce::Sum(T input, int valid_items)']]], ['swizzlescanop',['SwizzleScanOp',['../classcub_1_1_swizzle_scan_op.html#ae81a38aa9d94025da72544b9e0dc611b',1,'cub::SwizzleScanOp']]] ]; diff --git a/docs/html/search/search.js b/docs/html/search/search.js index ccd325d797..13a3875f4a 100644 --- a/docs/html/search/search.js +++ b/docs/html/search/search.js @@ -7,13 +7,13 @@ var indexSectionsWithContent = { - 0: "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000111111011011111101111110000001000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000", - 1: "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000111110001011110101111010000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000", + 0: "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000010111111011001111101111110000001000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000", + 1: "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000111110001001100101110010000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000", 2: "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000", 3: "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000111100000000000000011010000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000", 4: "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000111111011001111101111010000001000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000", - 5: "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000100100000010001100100100000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000", - 6: "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000100001010000101110100000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000", + 5: "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001100000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000", + 6: "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000100001000000101110100000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000", 7: "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000011000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000", 8: "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000010000000001000000100000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000", 9: "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000", diff --git a/docs/html/search/typedefs_64.js b/docs/html/search/typedefs_64.js index ca6b90d973..6c9c1cb7ba 100644 --- a/docs/html/search/typedefs_64.js +++ b/docs/html/search/typedefs_64.js @@ -1,5 +1,4 @@ var searchData= [ - ['deviceword',['DeviceWord',['../structcub_1_1_uninitialized.html#a848f6233dbde22d8a42c022d3f0aa6f6',1,'cub::Uninitialized']]], ['difference_5ftype',['difference_type',['../classcub_1_1_arg_index_input_iterator.html#acd5e39570dd51f4883547ef33f9ca8c6',1,'cub::ArgIndexInputIterator::difference_type()'],['../classcub_1_1_cache_modified_input_iterator.html#a268f1e7a4b42a05c5b4d0bca2775c26e',1,'cub::CacheModifiedInputIterator::difference_type()'],['../classcub_1_1_cache_modified_output_iterator.html#ac8cdd3a29db7e398f14e44a7aa054750',1,'cub::CacheModifiedOutputIterator::difference_type()'],['../classcub_1_1_constant_input_iterator.html#ab45cc48afbfda7eaa3b4ea643e719c33',1,'cub::ConstantInputIterator::difference_type()'],['../classcub_1_1_counting_input_iterator.html#ac66dea2a687f0ad6ed0cbcd6217a3029',1,'cub::CountingInputIterator::difference_type()'],['../classcub_1_1_tex_obj_input_iterator.html#a710012acbff3dd0c35822951e28148ea',1,'cub::TexObjInputIterator::difference_type()'],['../classcub_1_1_tex_ref_input_iterator.html#aff1afce146f69adb655e2ff7366b869f',1,'cub::TexRefInputIterator::difference_type()'],['../classcub_1_1_transform_input_iterator.html#ac05064e9ad33dc032452e3d09e8768f6',1,'cub::TransformInputIterator::difference_type()']]] ]; diff --git a/docs/html/search/typedefs_76.js b/docs/html/search/typedefs_76.js index 2967e89efe..bce599db9a 100644 --- a/docs/html/search/typedefs_76.js +++ b/docs/html/search/typedefs_76.js @@ -1,5 +1,4 @@ var searchData= [ - ['value',['Value',['../structcub_1_1_key_value_pair.html#a9fd385872c09fd3757e9ba59b2754955',1,'cub::KeyValuePair']]], ['value_5ftype',['value_type',['../classcub_1_1_arg_index_input_iterator.html#a6c40ea1dc7c0923b9010f1e938e07d22',1,'cub::ArgIndexInputIterator::value_type()'],['../classcub_1_1_cache_modified_input_iterator.html#a60689c564e2ade39722947eeaf40156a',1,'cub::CacheModifiedInputIterator::value_type()'],['../classcub_1_1_cache_modified_output_iterator.html#aba997f95620d692cbad05c74ef169fa8',1,'cub::CacheModifiedOutputIterator::value_type()'],['../classcub_1_1_constant_input_iterator.html#a72a3996ca30a1b2eb6f676923a2ee3ce',1,'cub::ConstantInputIterator::value_type()'],['../classcub_1_1_counting_input_iterator.html#a3afabb6a47c8cf7526220eb817e2a97f',1,'cub::CountingInputIterator::value_type()'],['../classcub_1_1_tex_obj_input_iterator.html#a4405a9e3d39593b7c468629dff098144',1,'cub::TexObjInputIterator::value_type()'],['../classcub_1_1_tex_ref_input_iterator.html#a1de28970e3874202646e11984a0d9026',1,'cub::TexRefInputIterator::value_type()'],['../classcub_1_1_transform_input_iterator.html#a8230414e069b99a279e2afabed83fe52',1,'cub::TransformInputIterator::value_type()']]] ]; diff --git a/docs/html/structcub_1_1_arg_max-members.html b/docs/html/structcub_1_1_arg_max-members.html index 259c9060fa..9e74bac7ae 100644 --- a/docs/html/structcub_1_1_arg_max-members.html +++ b/docs/html/structcub_1_1_arg_max-members.html @@ -109,7 +109,7 @@
          -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
          diff --git a/docs/html/structcub_1_1_arg_max.html b/docs/html/structcub_1_1_arg_max.html index e102fefb54..246fe90e57 100644 --- a/docs/html/structcub_1_1_arg_max.html +++ b/docs/html/structcub_1_1_arg_max.html @@ -114,8 +114,8 @@

    template<typename T , typename OffsetT >
    __host__ __device__
    -__forceinline__ KeyValuePair
    -< OffsetT, T >     		operator() (const KeyValuePair< OffsetT, T > &a, const KeyValuePair< OffsetT, T > &b) const
    operator() (const KeyValuePair< OffsetT, T > &a, const KeyValuePair< OffsetT, T > &b) const
     Boolean max operator, preferring the item having the smaller offset in case of ties.
     
    @@ -126,7 +126,7 @@
    -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_arg_min-members.html b/docs/html/structcub_1_1_arg_min-members.html index 6b59d53c38..ef1a744330 100644 --- a/docs/html/structcub_1_1_arg_min-members.html +++ b/docs/html/structcub_1_1_arg_min-members.html @@ -109,7 +109,7 @@
    -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_arg_min.html b/docs/html/structcub_1_1_arg_min.html index 38c731d8bf..4bf81fe630 100644 --- a/docs/html/structcub_1_1_arg_min.html +++ b/docs/html/structcub_1_1_arg_min.html @@ -114,8 +114,8 @@

    template<typename T , typename OffsetT >
    __host__ __device__
    -__forceinline__ KeyValuePair
    -< OffsetT, T >     		operator() (const KeyValuePair< OffsetT, T > &a, const KeyValuePair< OffsetT, T > &b) const
    operator() (const KeyValuePair< OffsetT, T > &a, const KeyValuePair< OffsetT, T > &b) const
     Boolean min operator, preferring the item having the smaller offset in case of ties.
     
    @@ -126,7 +126,7 @@
    -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_block_discontinuity_1_1_temp_storage.html b/docs/html/structcub_1_1_block_discontinuity_1_1_temp_storage.html index 854cdb4501..0614937371 100644 --- a/docs/html/structcub_1_1_block_discontinuity_1_1_temp_storage.html +++ b/docs/html/structcub_1_1_block_discontinuity_1_1_temp_storage.html @@ -97,8 +97,6 @@
    -
    -List of all members
    cub::BlockDiscontinuity< T, BLOCK_DIM_X, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorage Struct Reference
    @@ -121,32 +119,8 @@
    -cub::Uninitialized< _TempStorage >
    - - - - - - - - - - - - - - - - -

    -Additional Inherited Members

    - Public Types inherited from cub::Uninitialized< _TempStorage >
    enum  
     
    -typedef UnitWord< _TempStorage >
    -::DeviceWord     		DeviceWord
     Biggest memory-access word that T is a whole multiple of and is not larger than the alignment of T.
     
    - Public Methods inherited from cub::Uninitialized< _TempStorage >
    -__host__ __device__
    -__forceinline__ _TempStorage &     		Alias ()
     Alias.
     
    - Public Members inherited from cub::Uninitialized< _TempStorage >
    -DeviceWord storage [WORDS]
     Backing storage.
     
    The documentation for this struct was generated from the following file: @@ -154,7 +128,7 @@
    -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_block_discontinuity_1_1_temp_storage.png b/docs/html/structcub_1_1_block_discontinuity_1_1_temp_storage.png index 850b02cc145ac04dee4b352fc22f1154e1e0ce85..9bea73c53d6e81e67fd1d3f5d54794ff18e32e0d 100644 GIT binary patch delta 1337 zcmchWTUgQu7{>p`f^BA-Iv*`dI)KfmJfw-DG-ruQWiB&O1a;<z5c$}3}RLld25;(ovE_S=`=KH9twHa$3Mu<2qotAaZrg@=pmmp5G^}RZB&z=*0*Kg$nUe7Nm z+|AO1EquOHonA~Drs7oL>R<{SN2v2yJ^WlmzN30G{Jih4mj={h(lai3J`mMbU;l2= z$P*Xl3X!~)Ylz#F<-7p8AThs>kt_USBzN_|!DD^VZ~eidQU!Crg6mji;I%p6hR-a# zmnSU7!<#T?GGxW+pJU%ehSRyt;#AgjpsyxCP-%?~p{IdnN8x$Gqq2_d;=zG){o+_u z)j~SgnSDogaLJZrJ6EUWEmu0pW^8`na;x0=Y<4#Wh{S> zoN^kHxZ_qGoTw8!>BAtdFKxs^&S9n|xDAdw;OH;bo%-Y?x4AIPY$;+JZguy}#_DPZ zad!$HQvN0EK&)R$&Zm~<1+Xj=#tjw#l>ms?eqhVDu4~2!6adL*lOQfYCnl8jlmb9^ zSv`Zs01JYIb+p~UIsy5ALWlo4_rcZN+h;?Bl6V(8yfyve)|4+FH1;$D z;j~WCzMAgLPo9iF>`+`Y7~Vgov$C;w#mlp!^vBf8P3_Lm@CL1Pzo*$ND7<)pJ{7+- zK9A9Cit$b3d94~*x`yPAb2D?w*Nx@V@~E=oe8VL&^ukMuo9DbPEWXUe~vDY<` zTYV%6{tYUE?egCKE?>Jg@ybjf7wL~S$V=oRRga)So7gH|YuYh%n1+%MRAyFiyFYZ{ zcq*)i)QCudFuO!|!)Rixbl?w0idPxYTWALx>bbxQwjsqP4z3IfoHkRU5HzY?j$>JE(z5ctpKTe(e+c(cEJimUvX2Q4X_4-rp6~( zBZ#)~`F>(q;-v6Cp6G2qe}GRyR|~RzWwk!l!9@-DP0SvVH6D*MgqPuc=|y-l$aKgZ~oex>wa*+`Ge&TqA!(W#WAuhR=nPyx#1=__!@ zlv6MroaS|NsS^!I*u<^qiQl5flpZx?0MIe#vm2j?nGuTa%U zS0^_>ap}wI?;u(LfQmo3Z!ht@(p0`6Tx6@hGPkIFQI26et6fu1sLJ$lcE_4hd{fW_ zy2(hEpU3*gJ34)T^YcKQqN5GL0~x)QUns5@R;4uJCuHO1uqEGeNF!CdyI*H##pPCXDBgB;4+ zPFA0|YyG@7Wwz8L41WFjZq_uLsnKy$8Z_pcigpmo5G_8(~Gr&&Z;m1`zfchSTD2*ln{b|mBP$8z@gqh=~ zyb~yUy$euIPdlQ(!LN9e^grX3+s4J~+zuMwyXA@~1hf_1CH!YgoFp=bGb4E)GT;^S z2{X`G&0X=b#M>TusFYJxoSK-zhY|)xH_L0`^BegKkcLQYUKBWPs(2r z?^T+vice5x&FhDsT(o3UgP{$)S~hRIh&R`Gt?5(_MFoRm;M|vv^*uW?Aya9b=0l7T zD?{9e6L~UAUTXYMYx@OuN#E&I(QJ`<4@=Bl4wWQ(i=J+U?N;XcO6_uxwCCy%^IqNe zvg!?a4Eg-01v2s2VNHpuV8fCM z?_PAU_(ZNJgZf|~>4vUuubeu=KcuC;ZE;Zx7}Y)K@X$Xr2=~T15=qvx4GSh9V-%TK zo%VEs@8WoU=|Vf};@V@;Q`wnfvi2<+3nMK?$U0VfM@+S00C9g^)Y|Ftgo1ahOvk(@ z)+Vv>)*8x3EY${M;XnxM_V}sug zF(6kiArrjwFu-|wNy%iyTeHw9*wG3IyQiz0CWkXymFl@}J?%89p67x__RmcKt~ml` zt-Xv5G`gbn#PUn3&Yb2c0%rXKLd-1J?PEIl#Qh6< CjK@C! diff --git a/docs/html/structcub_1_1_block_exchange_1_1_temp_storage.html b/docs/html/structcub_1_1_block_exchange_1_1_temp_storage.html index 6f6d1dd63a..4ecca4c0f8 100644 --- a/docs/html/structcub_1_1_block_exchange_1_1_temp_storage.html +++ b/docs/html/structcub_1_1_block_exchange_1_1_temp_storage.html @@ -97,8 +97,6 @@
    -
    -List of all members
    cub::BlockExchange< T, BLOCK_DIM_X, ITEMS_PER_THREAD, WARP_TIME_SLICING, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorage Struct Reference
    @@ -123,32 +121,8 @@
    -cub::Uninitialized< _TempStorage >
    - - - - - - - - - - - - - - - - -

    -Additional Inherited Members

    - Public Types inherited from cub::Uninitialized< _TempStorage >
    enum  
     
    -typedef UnitWord< _TempStorage >
    -::DeviceWord     		DeviceWord
     Biggest memory-access word that T is a whole multiple of and is not larger than the alignment of T.
     
    - Public Methods inherited from cub::Uninitialized< _TempStorage >
    -__host__ __device__
    -__forceinline__ _TempStorage &     		Alias ()
     Alias.
     
    - Public Members inherited from cub::Uninitialized< _TempStorage >
    -DeviceWord storage [WORDS]
     Backing storage.
     
    The documentation for this struct was generated from the following file: @@ -156,7 +130,7 @@
    -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_block_exchange_1_1_temp_storage.png b/docs/html/structcub_1_1_block_exchange_1_1_temp_storage.png index c32b5f7dcd9851bfe2b62eb4d0cd6b08b18e6a8b..f0d7eabdcdca0e1ae92f383137edc3fc4e8dbc2b 100644 GIT binary patch literal 1931 zcmc&#eN@s}8vYrk9+Vs#b;_J&j**@$lSK_v25Z9)6v7-GR~!6TO@~BbGVl|%cC4)N z3&&84awjD1NLB*o2g2Hv<5-#E$BLS3On#7n<|u=R>`!O^+CS#sz31F}pZlKYJ@0eh z_n!Bbgr5qAL7kug0Ko7exCj7%P(U1R1p!xK|4S;k?FlDD2AWJJ5Y8{IGRk`@K(sw& zGTF|ONjP}4x)5>d3jic-jgg-tQ2=1+gvXtT%r;YVV=hgE?lOl2yfRE@cN}0B$2K+) z=f`hTy$+X%b~k=jP;wP%XXonDgTvV$4`Ej#_Tz)tM|%1AZ5Z`0Z#;^fjA(-1bOr6- zA)*rHh(3_UbEnoJ9L`(Qu79Ib7cis;nv!a7KQZBl*n+{PYF>aj$S=0xwpuk@cNluJA@z@yR5AH@pbc>dqSxHpJxV zDFM6CxkLmbI#q#O#Ue<9mth`EB3t!{!jAmOn*VGpzVJ7x)>%5X+r}6qf9@Zwi?pIi z7wZr#U`sd#oqn#bT9ZS(*`*$}!9&`sKCj}HZ;MA0J5%>hDKZE#egx|#GKC$?Zx~7$ zKrBGO@b?kjn?VN^7p0tL?SOGqqBAbiK1EQH(e#RRc-zaN6h)q{KNQT~dGK1sLtVGR z+{kR~HDVP{dRhCo)p|$C_x+NdEy7l?h4>lKSJ?08~G-oz{*~^GC4;ia<0IcIs>Dg~B?Tr7R$cxTabeb+-!6-6j z-njf3zR+HH=7&}Tfh=>-ntA; zRj!leamQQ(km*j@B8qL#h(ntRF&XVakA=ngc(fQ^2$eQk`E0D1Gn7>Nv|^?a&TlwO zwdJ&I_!RzPsmi;xp&01x%Uj~Ucu|QJ({ZyxrY9ln?mBPt0fkGFi(4b*{Y>&F9Y1Et z8DU?gDg0+PD=D;hD~mE-`#t?^#L)u{JL&g}rd3pTb)cJWO3Q-|N2^{5@lQ@pME0DQ ztvwcrvR2BLZ(-M0Fy;?Da&FWX!^%;a5%&~phSO8~*0Z|>b8C4uRBFaIImrrRyr%j} z3n|SjUE^)p9lh@BN?vN&rQ-$WV9Sr&6tlbq(;=Z}v1NRb#+hvj)l6%KG$*_rvj0e+sn|)$%^c`d93rQHNiiIQI{HH2OkwiqpEr_ zeJ>!#9QmuB6%CaBzt@R<4|ZZ^j-5z-I*m*(Q&jUMcA7i2Ybn@^%+m2((gaq!gIz0Z6f&G2fZRK=bjp>oIIC6?T=1aoZQ zq();WU6u)_cj;QWM$L{)kHRrwVsJmlB)qP%3h)sGsG5GtIIDSYGWa8G@zq({kras! zZDlfVE|BJNP(s;VPgZ_RlP`{-qh`d%3-&pGI^gAH|L;h3X879Kxx=%!TWW4}&iR)4 z8(?z9_hjk^Y&pH5L6`4P+QLR$kUc`7*QDP`jowo5!C5T5mh>D9g6@p5po;9Yby|mge zC}NFOK7%HTP=W;YsE$g9QWc@0C5_h*Z_54Xy6gTof8Vp#+2?om{_V5ZKEJj0T{k>d z8L5c`06-aZK1%=qIWmmh_RGPR^ZP3bd}z42c%IwY*@2;CW|@}J_5enEyQQT?{_cn_ zeBB>Sz4|0(CDDXTZU7ytkfsBXxdSJy< zGEC@p?lXodEPZ81DRWXtx8@s^k?5}EO+(OQrBkK?Q^?2AIEHsra9(ljJ#B-sk3HEY zX)p6&U!oRl!lA(j*yPkGj~ZUIEiXYOHooFG)_OQ(=#(=-K<6uoYRl7(-V25j7R{%< z+_T`qV=WNa*X-$0cPpDvzmX40*+b77z0w(P{IIMMyRelR58cQ2b!rX5#jl5Y!9LND z70AsYoE@H-yEgphiJj%8cdk}yMHZ}luF_qPH>4Wkurz5;A;79QCjLSh9flwWuA{=dT*hc52GsmUfnNPf zT4u=!=HS7w?%$;zjVj~^hw8M7W1Rop=d4n!A~!}4$u ze<{wcl>gIj{5fY&WJ*-;yAK7j&wxe}|ITGVp#=u=Lso!X0B~qwI>$@S98d^QMl0|7 z&&A&f832@$5SNm1dw%T|rCq<*7rr;Ce=nHwC8+)->Hjg9#J`tCurw+OB7EOnoYl9_ z`AgrzD)KIP(gqYtaP1@6B`{XQPyN(C}e6 zLD~p8WTL3KXr(40CoRIdWA@ptO}0F~qO}I4YA=9bJ}~dk9au zS%0IWa=|tJ*SfCmN`|iCeCtJgNcvJ$ik@X@ZCs#_#(<9WM_g#UW}SA4rhlD``}%17 zNn%ZVGt{g)_sk{klf6I{)#K1(lU*%#y9T5p*UEa?1S7jDYJ-}wn{Sy`zk*mL?7IVL z2$SGYDs^&p!gh-+Yf$tfF3tp@(0#~GwP{= zn%X=xT~lt%IH4F++iHC7aJ6d7KI3$BO0x}1jV6E8&)pipU=pw1Z@JYiQ$g3ZZ zU%ydZ6_7Ye(c3(a)4IH7q50kBTWglKb+&NrM`?@woaj?w9(p-SC$svGDm$9{^mFyX z2!Aswa`j!>Y^UsRZPo?eCnkT0Hf)tl+KsmVO#aM#9G4@t$)#6I%>|r)`VVrh;8FM2 z$Aip1V+RuvpJngwk`ie-t(`UP!EV>V!*dZOpA#aj>@})&D=7Q%YuE(XeA1hkS_NNVE zWf;NWDYK2M2fDOS2)5%qF>Hp%_ zP;LMF}F-iiM^rj_4#L>ez4E@YJMy^!AdYjm(hfQzEUScU)g?N(zQvEL#Rh{ zp2DAUekY#ApLE>f0BYOm#4Sbfw(`UH*lr$X`@MgvjqYo|%J~wW2#CEsm~*JUaI+#L zNzMO=tjBnM?t=ii`fvGBN4+Fl*H>t0Lomo`)YaUbPP*cTgF1xISN5jT`<+{VHJv(A i$^V@S4A7`#bN%a|O6#4bzee~1@B+`_&$7=1B>x9JoRUZY diff --git a/docs/html/structcub_1_1_block_histogram_1_1_temp_storage.html b/docs/html/structcub_1_1_block_histogram_1_1_temp_storage.html index 061a05f1b8..fae635575d 100644 --- a/docs/html/structcub_1_1_block_histogram_1_1_temp_storage.html +++ b/docs/html/structcub_1_1_block_histogram_1_1_temp_storage.html @@ -97,8 +97,6 @@
    -
    -List of all members
    cub::BlockHistogram< T, BLOCK_DIM_X, ITEMS_PER_THREAD, BINS, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorage Struct Reference
    @@ -124,32 +122,8 @@
    -cub::Uninitialized< _TempStorage >
    - - - - - - - - - - - - - - - - -

    -Additional Inherited Members

    - Public Types inherited from cub::Uninitialized< _TempStorage >
    enum  
     
    -typedef UnitWord< _TempStorage >
    -::DeviceWord     		DeviceWord
     Biggest memory-access word that T is a whole multiple of and is not larger than the alignment of T.
     
    - Public Methods inherited from cub::Uninitialized< _TempStorage >
    -__host__ __device__
    -__forceinline__ _TempStorage &     		Alias ()
     Alias.
     
    - Public Members inherited from cub::Uninitialized< _TempStorage >
    -DeviceWord storage [WORDS]
     Backing storage.
     
    The documentation for this struct was generated from the following file: @@ -157,7 +131,7 @@
    -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_block_histogram_1_1_temp_storage.png b/docs/html/structcub_1_1_block_histogram_1_1_temp_storage.png index 31b02c03353d566f7a6b84097bf829fcedff1598..0a65457754e6b5bb8c99e862a28123439ba41323 100644 GIT binary patch literal 1926 zcmd5+`#YQ27XH*oO$npw(u*lsE{c} z+Q2G=`wA=W9%Fxs)Go!^O4pd z^DY4*Ji!ehAj+^fnGhFWlEkW9M1gQ=lSE++hBPz%Ymu7pHS;0cLN0Azb>k3uEUjE> zSHNv&cCxwKYcP-8^FPegH>5G@Px>^bHdW5RKxJy_sO7Lj-H$h!OE~%K*q8SUZK0>n zbP~Y<_!fcN1x`{`-alUZ1NI@iD3B*4vMG9|zU;(X7kRa~ zOXS*v*_#?n*qlnE`t39Z(lE$!nomH1A~+O~JLxd*({ht=g?eYX{MWSPvrc%%ouTsI zhZdabZNP;{0dfkKZW{M9d+E16i@5%#VizR!O$o06Ih?iTeo8P8YdTOnBgrd6;Bj-f zE2UWl+lqv2?h>R8W#jwBDT|sp&G)#os|z8lHp{9#H{5?Q{R%GIbYo`O`S zm_go-l=Yw~_iUyDBA_=bChASfC>&q(&{8#|zQ`DIb@sV%NBK{7kPvvtttU8kJ{Lo9 zSWi-{5p+QgL_Br1lMQ9&7~Awel(cLUqeuk8&--dtX3ADT`%hRJU=Ch|h=qFO5k@ z)`H$^WV8$M2UT}1W_@uz!C?(R`T+VFT7;{e8}O5W+efvdo6$oR5IqU=O}6}Pfa ztRu=U(P>?jK8$Lew@{%r&x}tUsIpUr&<738T*`3)6IxwU>U_CKS^0wBlv=r5RS~l@ zNA|*VcgEkR4wT+{m6v5W4sTZ_k?r1^xAW#7)$_LT`Ref+gslc(gEt z)d;|=i84m3eXirgHu?#kee?%CLGC6JOX@&wAfUgvZ&zVD{b-WUc}>EK_TbI#hX_Q@ z!=W|smcffCW$;9_48yxQI5Ykszx^%acryH-*j(uuK5JB%x*#8MDK4h2#DpcheQO*Q zt_(>CK5Js1KroYVl4R(XzARUG*YA@nKD_FIBv^$9 zxwpdvu=eb#+QZ`+GsDvt9``@I#<8mDjxuX#SKh=H7jIb&i!l)OKU!X?jX64Ec7L>0 zqk>I4&9?VR&Ld7M-0?wXIi4d`s4CJNSez;J5M9w}scz)ndNtFPC)5Kk#LAS}2rEl8 z>N;jPmJIS@yX%Oql`Pm`j;*E{wj|A^G>_jXA?0?FG!wZpFIP9|GDCE@Pcag`aQECe zrulK-(gn@TL3yTXRzh>8 zAM#YcDb;s>F8?+?4||zLvDWE-LHj0C=-EPc#JjjiNZhfv#2am$K{4uV^zM}_czPJm zD>FIMH5}8Lg5inm#*4g+;!>7mFEk+Uq^Wo$PeDZgTJvX|Gr#Q4k4gKe*@sDpC z_%gQGkCrR;r}oBV%^er*0e`89t0}Wg9$}XT;ul-1~9YVE*ayfCmls}C|{Lhm9 icfG&2!~gd)h+feS+h=gOVH#5iKn=wBd-IP(T>cmJ5_Nk3 literal 1956 zcmc&!dpML?9R93ZC9*`bC1W4Mx+Ym74N{D6rqJXv+1zc`$gNDSgY8kdWf-2wG?h?d zn7%NRh8^SjbKdWq_jk^F-rx89Qk@+!^1Jr$ z0sufBbU1|v07(Ljop(sWqsbbI2u}#-(=K*mu^5IQXGA1QQx=T2dqYEmbn06q{JA3% z?|2S?Nn358!{rD7$eMtsPPjx%%+mRng~t@6X*kz)v31O|kW?un{0!oliA5Ej*mgTw zIYrxRm=$8eXN6qkvv`;IEV2R(^&NPG8i1NQXK;HP<|emLQJSDVmHLzbufKA@E2-kB z8Soyq$q($4&z-_>-jL(GM#I2q!sxk}FJK>ab0S6(J|$8TW`@4Q8i2C}2e9~_J8JHI zYF9iAz_m7rK+Qc_u1JYjXerkE+*_Qm&YxPWKtHMZ)1qn2hH8Irl~T#E3!$zpAcBy` zGh$tr;G6sQz!Gw8l?8XW_?<)1RSqvz_zG-=Ur{9YisT-9r}Ha?!OW3|=u}+kpocNo z0qeu>!lZu8i76%EvORh8%ma*sB2oExve0FqB=_6NrP> zGtq(_k%iT7rAv2bx)z+H43w5H1Um&2QUA){C#B%8S-um(H>(QauNVU|@AfUoHUJ1f za=}7X4hBhTUpAogfJ_5W>;nP2lO)!(WuV`Mr;Rsc-eP_Uww+{3}zk>&-3A z)fiLzBJZG8`^8Rk-Z8{*FB(j0p0%Vuu*~YVHM9F@aDutC;_^+B^cjZsA4Thk=b8vn z$|c=zO^DVQ+n40_Sm9@t2Le1}#mKE?crw4}ewqkduwlcfk~;prdZgBifw(r183$c*`%Q_1u0R+?;%`!fj3TBpE6)5sx)J?{j}z zRHxKOUA^e~ne(&Al6_BOx-V8lTYEy7Xox+*5S=}^x`KvoH@4rJqcq0`uXitLW}t5! z-B%X0N8x6hgnz+c|EyelR{tmyWV~V$gLIsYeK4X}v;4B5T;;tZHRg7FrrG3hw*gg5 z40J4e4Q)W~Q5wi}3y(SO&BoTYwt0Q%MLX*g@BEKaf|Mvj-1ip*j|`eSGUoysXZh~z z2bQycW{Ko>MobOtr#`f)d(XP7EqcJi`5EfFDVHB0n?0S4ejR4?be!(sjv0&J(e}&f z9xjtWX4dAniaRA^53c$~IqADy2(aGJI}-=pkc`cKGVjJWUDcw|8M~;Tq5LfJc6fMPNP?@c*ctfSvOLrE1?km;2;C^RiJ zs%!ii8Kd*qBs8ACTy+UO@3QP4Iy5~juXx#VrX~A>X;=XPU@vq_iO;1=JinRemd))S zM~;e{R_c#W5;j9Ug^|?~cbC6-jQGkNK7u9(UlT&yxZUAR%*E;Mk7LG-J!~l-UMEdN zNmC(eZdDfU^`WJxj%Chv@p&hf`LF8pNvS*+fBK)Auw*Qjs57lDLs?Ld0&oNE98Xo+ H`ds@Px7dhJ diff --git a/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___t_r_a_n_s_p_o_s_e_00_01_d_u_m_m_y_01_4_1_1_temp_storage.html b/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___t_r_a_n_s_p_o_s_e_00_01_d_u_m_m_y_01_4_1_1_temp_storage.html index 94e61c9337..aff70026b4 100644 --- a/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___t_r_a_n_s_p_o_s_e_00_01_d_u_m_m_y_01_4_1_1_temp_storage.html +++ b/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___t_r_a_n_s_p_o_s_e_00_01_d_u_m_m_y_01_4_1_1_temp_storage.html @@ -97,8 +97,6 @@
    -
    -List of all members
    cub::BlockLoad< InputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::LoadInternal< BLOCK_LOAD_TRANSPOSE, DUMMY >::TempStorage Struct Reference
    @@ -118,39 +116,15 @@

    Alias wrapper allowing storage to be unioned.

    -

    Definition at line 823 of file block_load.cuh.

    +

    Definition at line 838 of file block_load.cuh.

    Inheritance diagram for cub::BlockLoad< InputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::LoadInternal< BLOCK_LOAD_TRANSPOSE, DUMMY >::TempStorage:
    -cub::Uninitialized< _TempStorage >
    - - - - - - - - - - - - - - - - -

    -Additional Inherited Members

    - Public Types inherited from cub::Uninitialized< _TempStorage >
    enum  
     
    -typedef UnitWord< _TempStorage >
    -::DeviceWord     		DeviceWord
     Biggest memory-access word that T is a whole multiple of and is not larger than the alignment of T.
     
    - Public Methods inherited from cub::Uninitialized< _TempStorage >
    -__host__ __device__
    -__forceinline__ _TempStorage &     		Alias ()
     Alias.
     
    - Public Members inherited from cub::Uninitialized< _TempStorage >
    -DeviceWord storage [WORDS]
     Backing storage.
     
    The documentation for this struct was generated from the following file: @@ -158,7 +132,7 @@
    -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___t_r_a_n_s_p_o_s_e_00_01_d_u_m_m_y_01_4_1_1_temp_storage.png b/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___t_r_a_n_s_p_o_s_e_00_01_d_u_m_m_y_01_4_1_1_temp_storage.png index febc4db9a1ea47fc76fcfa660f1d8f7d479e05a2..052ce695d4f5f685d73b0efd5f25ae580f7816eb 100644 GIT binary patch delta 1998 zcmZuydpy(YA78Cs7e@)dN-BzJ(d#6XFqcB=+sUX|4nq4SjO4ocg_+C2>E^O!(rim& zjSO>Hs3AGBWysuiiQLDQI!3wn8?SSIuiqcP=kZlyfyB7CSc;m{Yzre0`Vs<;n6^3HOJVug0gJQW{CWZ`D*s>FHf`{O!)~->rUw z>;+5-evr2FA95-v%8if((FcOB!GLwbiO}1KB`76Mg!_=vL^))g*b8`qMF1h5!24rD zQpWM7O|cAR7Zpktz#TfBdC0%*6{E7uzi(rN!^uENudB%C_~^xv|+EMAa>@nsevVl zk@tg248^lhyY^Y0{bibs0Ul_$s|?T%U?IkyTyU3c|A&<7tWQhWtg^J_Ycro7%wlK;HNB`DM4WszCwJa z9(`k=me75PaDTn-AX1>{_6N~b@jt~ z!y=Ft2>eA44^mCoa2&LGXu2N`+S~}zN-+37+ZwY`DQX10NRV&3wszpxBx2qRCvLz4mqWVi37n`S{Ssd@uC== zhR;}wYO$ZMi)W}^=Rp!pO-$Yr7#&|WSX%Rw$G+m^XExkXYy}7`%)?u#;#$Hz-EI%e z*k@6_H%$?`n1y(cU5H6e`f%rrrmM9X5v2N(2bXNdwZUIAouuLvwz#`0a^if*hLU?} zzZj>!D|NbaGVkJ#r-9zsPPJ16Mun^~FJ%!c9w< z?fl{C^v!y@(>8M#5-Nk?lqGi_?iddHCso=A0<$Qr9{jO(kyr%IWlr&rBD-pF4G1r` z^m_d?`v*gM@>ov;t|$%5)m>209#z=Ru=}Odi*aw@5_{f9ZcqE=uZLi`V_`GeL%m(E z(DcLpX;|$W5!{dF&~F8Vj*e;6bK4@cNd>~Txzf_j00-hmX$#R|8uhmG$<&P76)!%a zT=@9@_-w2Zsmi%9j$1Z%&e9Jd82hy=MC36Cqb#q=&9GCER!}eZeB@Lk{A|hM_g6vm z7*Rcw>}q$7{>f@^-%Gko(aC>b!OjP3L<*!fwpg zyGtA^lLE_!2CZ1G_mw-o2B^Mb!MpK!yp26l(u6X;;0eRPh;D3hS#N~F{YAwI_2R*K zkuRH}y??(j!oRXzS8H@q_2Oqv$a5xgAolTlpU5RiRPXMY2HmP|^IX}Ao<{p=eb{SQ z80CU3$IP%Ud)s1WkEY@9WnHFMt~JM`wVKDLsWj_jg?kW!lAkT}94IjvLI$2!)RndE z2rX!602L~Z%%NBz_Ya`$G;%m(^k6vx^Jf0m4tZ2-k03PAGSEGKB=xf=rq&y`Lnc^= z&1=K(P)XZElRuqK#&9gd%t{-=0}_V!THMfP*^*n+rVkRyaR;KP%%oFqYim;vqb~uB zTtwvL60?4LUcaH-@wK4@?`;w_-U!!kI+exe#G%gN24+k( zqb7~`-A{E6Ir}}n*8J#IAU!{mESJ@%Jb6m`C4j#2uYQXf#8G@UC(!Gh{DwaDD9hDy1&nPt!KQtALm;tWVb2wW&*;xQhM>KJmB?i31eXQx+ z{-7}*MEEg$U3xe4=2WG2@DEzdJ4sb&z4(Fyz3E%Zm?3H3LnD62=>_KFjzaXVt`nZ8 z7?-S^Wo}oDn{71QEBMx>;Vd<|BEJ!nvnq=A&$t}DrK*HdA8&mN)28be5y>8Zo&zoD zSeR(;EW*L>!DMbl0S8)wWXGO+KTtBdGouquuK1qfKhki~PCfy9ToPPn|IE~-&F)Fs z*!gAEV|gXf>v{la^-%9Xb#DVcpP~~=oAF5?F%|h6(w8_$)#IptSmI2>Xj(%Hf#A4| zrHPfqUGcV8qaTDZ6>9z4l&}c`mz#d0e23ThR%osRY9q+)Z=h7dvl2v!4DfDfQyY8@ jecw1bK3e;PzlN-W7xV@ujBna(1}O({7VgNjzn1hLSXK83 delta 2014 zcmZuydo+~W8=k09o#>)Y5>3_-N=1eg#|V|aFp}HJrQ{M5V^VxVGvhcGCBlrfFkf_1 zpP?p~ammz(u5%I@Glnrt(be2!GLHKhPG_y(AHTiUUi;nee%|MO*M8Q1H+@~ly1XFr z0C3RB=i(9>)JGZKv1S?7<=M=(;)OGNssO#&o27JxNLt{s?_Y~`eS?_rf6`I}k zOmA3Vbd-{B{0nfHl5I@f)DNhaO@I}JiEN-Q8sWw`?HE(FC)}8Ggsb0+T{I~9K_GA>kV^0c=gf0CKHZ>`6W4LaE<^Mw+MT51|C)V+s+WFq}=CSyl5GZ4e*tJ z3r?xjm9ARI+~x+392L%G3`LF92j@tAJRiurV=3izA@%_Ggc*5how6Ka$VTQe#*Tn1 zkCU>n*QCtOa|*s1qg=q?mJ5`#%K7thixyBb=ye9z?9*aBsDD=&L8kC(0;|^Y24f}L zCYS<7=5_TI!EK8-Mb2dLOYDEj-@dZvux?nk82JRDm%4FMGX$|Jb%`b7)Jssq5}|hPY%O){ z(OqyMLzcyuA{GEo*3G=)B>y;1l`UP2`E41CiqTA>#%coUxy|sRmQi|$^YYg8HEAo1 zmg+h$Kb492kNgGIT+^6r0#=^P5Vxoz7&iPsZ_BC)<=hm)nn$qEedr#sw=O8C@Wan^ z;+{}1Va}xPO?*4?OD#X#D&i$q zpi#6-LLSq#NVNOskOL*k)tR>*?89E&VKE(S#;v=0AlOP+PqwZKT{$4|GMFN2WJR7? zm%1dekKSKq(MK2X%^=ojVM;M5w>=*CWMt?)_o~tPDz8v&)9x0tfE$M2pK50nNAOMw zx4e*CN_=-XfUM|n*Q&TSyfZvFeXba~xzdGI3k1YN_z9j_>nDztbPwOG_AxMX!r)pP z+E6d#K!bM!S$HR@Bqe2Qx9>pGogGtnv4YowpX!xQ$;fAubKJ%WrC)DBW3KEpP@5Z;ajpq~Ud@jAhmi{DSA)6Ac2}1gnzFW}YB3fu*k(o3greZqdf z_B<$)!to|7|a79eNe@anDAEj;+i@pkn-*J4or02eT%7^ znK7}j1JKU_i$q(pX!HS2#f7;dOKWsd`60u!A@&1odz4NDrv^HI!*|FZ=&j#jQ~Qui z%su9tc&_&X#Qx~n$g&LuB_EgY|vEHBW^+;__m5> zrG29G&G6HNJdOuj$IjEgb@PxY$c4(RDCf1)_kSk}29Q{30r=-&0a?z_v%3%i=f5jC z5*fih1IwKib`>!3DOOfLvBmy6K9=^o=B-sypU0?v9L)o-kzIgp;ob!+KYwqAwk#K2 z0F%!}rp_?q9KSG7jb|~VB5SwRXd5Eab^N_${D#jsuiQKnE05~a?kAbu$6mbwRRhjt ze-^q$!n2{h`x?B-=GWO#fv&$uL@Be{8iu)jbcYh%KbRUs9u|Mk@ig4ZA$T5QCuKS>TCV75SR>CJ-4wV=MuU^Jr zeHXLQDFw~;Wx+Dh_;6g6+uT-a3CqUy@vobR9ge7sy|N1dqSLUIgT%})EYV4JR=t_J zXzFN7POJ2@!AM!?F+0{pjZ4KhGH3p#wX9j6nlaHiG>>Ys0g>-ep-^!2?6f1ZdswC% z_~|Y8I>!1*oVYP-iOIdKuqRTktNAMmpQ?zPf8RDS}#}skePJ3|`}G`9Ga9ouEWTf14ya6NKdm^#BfG4>FuX(*6Oq CYx|i1 diff --git a/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_402c3164d23f1ec647db5dad06a54584.html b/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_402c3164d23f1ec647db5dad06a54584.html index 0a9d06ba7d..cbe30858df 100644 --- a/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_402c3164d23f1ec647db5dad06a54584.html +++ b/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_402c3164d23f1ec647db5dad06a54584.html @@ -97,8 +97,6 @@
    -
    -List of all members
    cub::BlockLoad< InputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE, DUMMY >::TempStorage Struct Reference
    @@ -118,39 +116,15 @@

    Alias wrapper allowing storage to be unioned.

    -

    Definition at line 894 of file block_load.cuh.

    +

    Definition at line 909 of file block_load.cuh.

    Inheritance diagram for cub::BlockLoad< InputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE, DUMMY >::TempStorage:
    -cub::Uninitialized< _TempStorage >
    - - - - - - - - - - - - - - - - -

    -Additional Inherited Members

    - Public Types inherited from cub::Uninitialized< _TempStorage >
    enum  
     
    -typedef UnitWord< _TempStorage >
    -::DeviceWord     		DeviceWord
     Biggest memory-access word that T is a whole multiple of and is not larger than the alignment of T.
     
    - Public Methods inherited from cub::Uninitialized< _TempStorage >
    -__host__ __device__
    -__forceinline__ _TempStorage &     		Alias ()
     Alias.
     
    - Public Members inherited from cub::Uninitialized< _TempStorage >
    -DeviceWord storage [WORDS]
     Backing storage.
     
    The documentation for this struct was generated from the following file: @@ -158,7 +132,7 @@
    -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_402c3164d23f1ec647db5dad06a54584.png b/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_402c3164d23f1ec647db5dad06a54584.png index edd6f65fbe6190252facb2652c49f94ea659c23b..9123be8bc827abc636653d5e8260f61be99008a6 100644 GIT binary patch literal 2375 zcmd5;X;hQP8vYPOw8RJ~SeB*)gyLFJQ3#7lXbM8uB1&XYNfl5~kbuap0tOTspooMN zqO5_C8iXKwuLYuPR+bP$h^#@#V#*RBgxp~D&;8}kJLk-tcV?dR%)HM#^UmPhkO~wQRImqR5g*n4t#Ag2ISB3Sdwzaiw#En1{ zU9|`gq$i*Nt(U=9D09WokR$vA>ZX!_$i6sUuf6fq0Z+NS({%IjP|sH5*EabzZ>m+R zq(ffsfAetKT#}>dWVS2E!|8{ToZpJdRR@c7QFmso;283WBZ}VCJqGGdup5ZnrGc;b zH2Kg7>Iqt(+>d~l>So*>mc!yhxuLG}=V~wHBd3@t4`Aiwc)_1x0>Z1 zAhO~5S~iUg;p5FI0N#JCif7^oLoP(7!}Zw1=Q+ zH6un5eQ7;LGS@;gy`S<8nrwN~0OJ)gRz%j#$&8faanh`0ugY;i4Y)Q2Nd&e58;%2Kare}}2UHS)KXm1* zTG>hpYa5_#1XWf8l)eY5I#N3?nty_}q2ZYIA(h09xDH^$I*Z~+{EN^1g5QgPtNu%y z@r71jgM2fTZ|4l@9|03JnPdA08x&4c5sbAyYF%9o|TI|~5flPFn z34crPY+O^T4PYQ(Q*p5_Q&%q`!hFM@4YSV}d6m4#gN!}Td=B!ZTRrObsxcdtvX~>M z?K-YC5|M&H8s)OT=-6xHToHIAebY=-e5GAgq?$K`oL=1gYBK&QptVwvV5J_=p8afC zx6gxTyh#4DQJxxUKY$!h*jtn@_<+@>g`G;8wanhyb~hDn!JqQiWChlFPlLxOzj&6k zwRi@E-(So$%S z-An>XC$0{1ba7O$7A9TszR~C>u{1a^OJ_o1wi)RmG!nbXA4)@HEp3A8X`>}xRN>wo zg_`J^SYbSRNjtN+SL|dX} z1TYQnL9$Iy>kjWmfj!f=yl@*Uf=C68WWtvQdJ9701|`vqUWd9}L3S*|Y^i-1X7;-I z+XA`c&W1zWYjL^wC6jbIq(J)M>BgE0nE2%BH|bZ0 z|Kg+(jL+RU8X5#f>yVZNd1fJm3uVaOD4UuZ>qm^~+4_Q!atSHL3oDXh>hdLA-^+30 zC8;BxV|v3!Efr>v;HQEN|rwEls{ zZe?BU^Q6wzVMP%QA-T4DXoliTWtGQehm{MXN&N%%P`#?~T{OW$>>|fSs)s+48a6lW zfdTv6oajM;t#T4>h`^MTfY-4|T9BK}`aL_hUxnbO-=AbM7aR7jH#AX4sP1N?B`Zj| z+C9BI*1(wD?{7*J(L25wAK{#xe^%}gSvUO1d}Wph>F`#MiG$MO`rpd*LW?T=OrZo{ zJx-~9q>RXDmB{Q7nIQaw&X1GliI{(!gUS$%<+XAX!5_T5WGPQB0wS8?wdXboxla!P^1{^^i+cLX zZD>r3EbNK$f_jN7=sHId)ee}h>RB)KwlxPN79#ZCiNqh93eS$|-4?GgZ0l(9l+USM yuJK*=i+8-%&&mHd@X_pR9t5!=?SK9z$sZv}_6u*$weO|_isI;igjb(DfBU~gkXc>; literal 2406 zcmd5-c{rO{7k`y%2MK+&8b!4kHEp$`s>o=Vu{MZ7$=9fDDz%i-N`lg$3;C>RVpnNR zhoZJNbOayqRV`_2Us6l79aM>>NGw4!Z%pTz@6Y-3yU+99_jk`d_uO-T=iGA>935;R zatGu90Dz#+SvmuNOgo4jcT0o)$dy{K2Vjo&E>>c(7=)i@g(T{Wdm!5Wnwpw+CX8x= z!`(z@hw}iaBxwV$T=YSa5N-LRON7)MVa$1;s7PNpvI%S|yYA zc>1f;*z|(Toa9ce=hOfinocV)dD^ZuU4@>8r=$pkr|-S;oLmbFAZ}r8b>F=%wQC{J z<}gNq89%1JXecUQe!`v00zD8B!QDmviT@66yyq+wAkz9#B;Ndh?g?~9)oQ(K9VdV< z@+b8lJtEu-wQ~!g3#LI21U*3q{mFxJMFXzD2${DLsDlhZ;l{JK_Wn29(GbK4buf%ajDZYLJuo=U=epE zAO+p)=R-sH+MKWTpF-1k!_T+k9w03<3zl3Mc?m%eKa7>00hgvq)??~Q-t)z!@CtSq zZuFHo&7;&yEd)XxCdJSVp2}AchQ@;Vg@S5sFeL+wM=LxH?AaYWU*-#t?&H+$(2QZ~ zw$RQmJ)-MenC*XyqjNSkiI>d5&rpH=qzJbs@zpCP*KU1_w=- zF5o3|u-JjMrEJXCdHEy-Py|RnJv0d0UhEC#&7b;U*8HgtIw!HE2Y{|gfx6xR=$Q2O zMkgqLZ3O+xkh+TM|FxmdhR(K}f^<6i*pW16TkXPh>}{ECuJi!D_dhwGGVKhaZ?ZB5 zcLjrayYI~mxq*!WH0DP5mKEo7I?DEPj}*4_G#+?%T%&bUuFLSFDUo+YKc~U>{7nI& zthUs%HKZO=gw37hBL_{OdtWYX-HKWSwmKJtlmX~rz6FV1?Uyr(P&t&}wN7c)!{!yV zj2YW9&@--g!o`Yn@x_?my;K}K!qyJN1&uoANBZ_k?Pl5Gn?1kK+pVjta?I}5WMVRQ zv^k`3P7NX~(y>m^_%u~7mvTO|^~3E)!K!mvjg$yRCDffL{hu(u$I=FE*HxH@&BgcI zAIMp-nR+L|OGk2Wt}GejinS|Wq+RbXhFXVP`yMt)njm_|#=@EKrod%W%xZQ>(mNgQ zeV6?n!_@nAq^O&tjhfr~#mFMCeH)oif>nOXt|EtRQPOGeV=I55|A=p{rCX;gIuH(}qWU%4XikTdAlI z|3c+`)3V!D#vGv@ITSqrFU_pR6h_Zf3T;sMe;E|`3|xJOTi&?$a&1MGmNMDFt>8kQd(F015E)#{`;6QUmH52J2%^ZKi zmmEFf>4d&VsJ!t@Q*fPXicW5@N2o5d_cFmc_ojy1K@5s^2{%~| z`n9jQ`OD*j5kKbDtLZo-x#6o?tY3X94UdNTothO|;@ciS-!K~e;TKP7nibc&=2%Mw zSeC;zc2)!xZKbq}UJK5^1C7K2Rvzm+CET*93eKN;SPE{^5yCEMA4YDu;}vD*ku((R z^ba>gtcOIS)DZ*4bmNZRK6{j@=vUD-SbSn%wQ*O8F}1#)n49!&b&b_--VttM7ABt% zKWK5MeTDBk?v8JWX?=86gTb|1m@_)3a8{f71@;rZuYSdeCVUo6Y2Ses>?-Dr6KD|m zDAM~z<)NQ;!8&`I;cC>y32%%((s@|mKY$cl!s1gHMJ-e-Y-%KNV&T)XiaqmQ4LXa9 zN1RX@t_}^38U65$<8Qen{l4KIl(m}nzRP1;(=#(9*0*i0{1bhNT~iw`9&5f_`&7Sa7b1$_dB0A>r>6$MPzgmH?gd9!ti~E0cB0`%)AmV@OHC^ z@j4fH>G@XZY_5Oz;CtR?M0-uw`u^KHx)(*bGEElS#}6<~WBc8SDn%S+wvA+|Y2^DI zCHu!NH?nnPvp*X}%rb68wEH+8<2 sdswXBvw!dwfDcsR{}fnz+hZqR#`JXWC`m@Q)c{~aTRB)(n0w#;14yn^`~Uy| diff --git a/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_e4c36dfe8f549604998f6c46cc8fbd1d.html b/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_e4c36dfe8f549604998f6c46cc8fbd1d.html index 7134ca36ef..9c4256feef 100644 --- a/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_e4c36dfe8f549604998f6c46cc8fbd1d.html +++ b/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_e4c36dfe8f549604998f6c46cc8fbd1d.html @@ -97,8 +97,6 @@
    -
    -List of all members
    cub::BlockLoad< InputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE_TIMESLICED, DUMMY >::TempStorage Struct Reference
    @@ -118,39 +116,15 @@

    Alias wrapper allowing storage to be unioned.

    -

    Definition at line 965 of file block_load.cuh.

    +

    Definition at line 980 of file block_load.cuh.

    Inheritance diagram for cub::BlockLoad< InputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::LoadInternal< BLOCK_LOAD_WARP_TRANSPOSE_TIMESLICED, DUMMY >::TempStorage:
    -cub::Uninitialized< _TempStorage >
    - - - - - - - - - - - - - - - - -

    -Additional Inherited Members

    - Public Types inherited from cub::Uninitialized< _TempStorage >
    enum  
     
    -typedef UnitWord< _TempStorage >
    -::DeviceWord     		DeviceWord
     Biggest memory-access word that T is a whole multiple of and is not larger than the alignment of T.
     
    - Public Methods inherited from cub::Uninitialized< _TempStorage >
    -__host__ __device__
    -__forceinline__ _TempStorage &     		Alias ()
     Alias.
     
    - Public Members inherited from cub::Uninitialized< _TempStorage >
    -DeviceWord storage [WORDS]
     Backing storage.
     
    The documentation for this struct was generated from the following file: @@ -158,7 +132,7 @@
    -Generated on Tue Nov 10 2015 13:58:06 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_e4c36dfe8f549604998f6c46cc8fbd1d.png b/docs/html/structcub_1_1_block_load_1_1_load_internal_3_01_b_l_o_c_k___l_o_a_d___w_a_r_p___t_r_a_n_s_p_o_s_e4c36dfe8f549604998f6c46cc8fbd1d.png index 33990c504d935e706f33ba1f58b17ebdc41bc6fb..2ba54b3f39cb58052dc676ba700c14e9df40d0e5 100644 GIT binary patch delta 2110 zcmZWqdmz*68=sWw{8Ct0QmamfbW)bgUCJ$z$Y@+58$7SwcT@gjiEexfSd(6yj zRwpL+Si`JYwIbGx<1(|2Gx`1g`n`X=e>~6U`8@CYJkR@iN5Ld;#wEbo*7A%?lnf8g zK$Vz~ww%{T^4F+<(%$!*VGrUT@m~$OFse(f(4+^LC$f{`?8JuNKBg8 zEhzT|vPlAzIW(ziTyecM-I5HE?pj8$awe@c;aR^dp9{m)NjZ?~9E-d2pLI`;tRRHT z;+2N1xCFLWyqk#%XJ{e#O;83gWNK=LWu6B&GIj1|j>xriaXT`zk~;rj7ck^B1x4-+ts zl0ECCmsoZ4{)nJj7_N9~bycdl0M_fBSv#Ko?fVT+S80bNKcl4K56cf-T4#wiYxl6z zdXnjMR_pogpRNAO2eio?wCuVnvpEr@0NV6AD+MAWwQK{qPqHbwiU&Yadr0UZ>kThm zIE(7tCdHY98sUn+uH$45C}`sj$hP8wWRLGl!2jO>$*e;_n-jT8IH>Glfbc7I;)+nr z^$sRR_Khk3po^k_Sdh^e$r~)p>mKiI_|ohfP}@*rU1^sDJN#mMcaLLq%nSM*w=HQc|5e(_81%5sJiPtU+e^U@D%FIT(;WAW_C$XG8zkpfYAr^WS1>BL@9NFxwecHI!6Sq?Z%xi>EC; zO^7{4<)K@c8r$o7MZclqzABg;r20IGxV|{Opd>>cSEuO zI;!$)c#)mtgEu!l=3VvGbNq5<8(~7A_H~Y@i&9UgWlvh%8J_^d+^WWKB4=0w(=42I zXXQ4elp7cHsZfbihG#HW%K4HEUTUcAP(q!B1|Ud6^?dm3TW`2#4{3hfOM7Zsz(EN* z8VC1RGs3x#gmU_7(>GVZu;G1J!oIC)D$CcJTZAijenh=EI@MutxC6tmYUat)8p}-^ zk?l+pdV&~q7tfxvG9{4lx$n5&mw6wT1yAm}Mew=4*?9ss(m#IgAJH=iAIGRJII0A& z%ynUHMj^L4#@loQ!kKU3w=41IlyjCs;&Zy@dt$s=X2fMWCP%1D3Iio@uHiOB2*aC$ zviGOdj^SWOU1xzq)PoQGONr)^Q~zPi+n17)+6_FW&!sWDDFrNchz%-mr(me>2?W}h zfqhG{6wk(y)?5+r{hg~9z;COIfUc!*Pewz5#*c+Vc{@*ePNvF|uqCY_{)y9%zX$63 zqTvI(M$Twpg+^0xQhaz}sWpgUiz(~lbEp8197lLTjB1aP0-XjW*} zuD9}r^fk*gLO>PsYv1Z=FdB<=_{ek+?S(DnNCMhNHE5sbj6)cfX?=jG2~rAIQ@ZTe zB6-G8MjMwFrOsJ?Dg<-7@G=dh>7p}O0>M+S=t!jg`CYmZlJDu$Q#w18v9vD+r4!K~ zS$!={%Z}+_*n`xBB1q|oV~#DFCh|cxqv5ouO(cN?m1X`gOb}wP>eKK@Sb3PJ9*iD)MCXi+8LN*GVzh z2Lg{gt(_p9^p^t<#1^O@#FI_(OPmmcFQ=f>$U<)7*+rLZqXm0cr_{NU%PF)A?;~&s z2G3wTAc$Puffx)bn5(RehoiOEYD@!J3)hqCy^_4PdIKsA7i!06iCI3WTApZ@tfcbH$`W5v50L} zWV1c!{?pLAW;{}TR2774A1pg=Pk<&&)Wi^(7xO-y;7eb(b)0!xmFF;`EDQ~P-{h}? z3QO23?`A& zM)_TC^z(_RIsA>-PSAkO-cN$4xN%Ll<(12=t(^k?2lT_0wSyZ1Z4QD&*~4$~0IReO z)dkJEC;wN_b2L$#x8b&IhG+>D(Ero_jjz7@ju3i+cVDVIj`6A{p0)nygrxj`}6*OUf(xOhoy5r)E(vF z>J*TyBqR#_r|oE)LWH%T%pyf4_Zo4|P_r99p3$S0n-C6%X^l_&F8!eOq$Q($Zljf} z+9+fIh^nh$-!PnPzCOxZwSx`BN8aI^y1SKG37)2Qm&8LW#3k2ewwgmra4 z#X~2)w4}S`uM^U$j`mAG^{KTuHcvE05ytZCy|Jf3GJnPkE7<-%y!!Ot9sR}`Srmb1 zPa=4ME1cgSBiCwWqu;sLpN4^C*t4zF_-Xb$x1T!!la=kao2Bbe6^&NHjMho5j|=CgYz)`1H2=SG_BNUI z|4V5zL}{vU(clzBVeEl~;&Zh2Kz_)CTu+cmEFypS%USlf)1eb;rK|{RLMYzPH)}$a z6?LxpxJxPsoS90#o}-T`Da+q;3#mxk)6}TQ|M%i5$VRquPeLv(dW#`&QZqu_lqUOT z0m8E*g75LFh@NnwAvewY(=U;jxUeP~iygE-kkt3k;6r<1yLuQ6_V5jyYRX+5 zALE?oy^N(3eXhkeQBWZZCx#457zS-N8bb?faNvSav&kXA!~;|ET8dOM{&rG+1Io-( zjcT+mC%tvzpTCm?0NIHyBcAOwMJov^_MdoL_}Y4mm(2dHyv6qqqmtC@CDCAlobhX%f_&2SHpjpGEQhvG=P)#`=7M!Hdw5Fn&yA9my zVDm3eux?X{Cqp;)WQAerevDEuiZTDe{$b3g)W(E&D2b)V{3KeAHgJrgfUm5{($muA zP^^R@MH38P!4_yaBD~92HB0qv^_ZtYNkwdnToI@YQj`ib{=gyUkckJG!N#%U6SFg^ z0n1`BAU<+TuKveIJtU`4^6n<#*P!RgZEPx``Sr{bsEIao^@1K%_i^c)5i0^^o)>Ky zxVOWXl)*1tVx+#J{BmZYcA}g5S+a9RBusjKLZ56un7y@Yzyw(bZ!rPM$g8yUKRdQ9 zZ9|q(#9g-}B~Sz#UtQm@>YzT}2z_rTG_eIP-vACQd+GVO5yZSLE;hIncyo4q=v*T- ziQRXcfDSv#n{smr^8@xVJEqaC)2I5B9leum5RCgOg8@ce(CkMnGrA|O>(W1hT2yKI z$&$o3ovj1U#H!sjsKjO+ofM)TMX(;i~~%`<>f~yd7hIrpkJ5=y8bfoiP?!XG>YU z)V7zrs%R{+K7r6VAx9zSDE+yZi6}?S2wa-P;p`oNKwh zr-3`!kk?qa1jOdV?IX%gSYeK9V@}3{J5rLH`DJo9 z=-qqvKKTauagKcIet8iTQ}2boGkv10;KV?7>VsKWrJsE<$;bYm#UseI0~e-m>fD^1 z2~4yt0O6R15K;I|U6kFSkcxo|0$zP~7eOYxKQ=K0OiT%SmrciXyU*J>WGvAm9T47x z9CWoBuJp>HllJ>j!S%ti;QIKAv9&Vq7vIm`iXkoP+EU%3+A>y0^$}L$I0475e`X%e zs}8TgTFx>I?m7s;=Hv#bzzeOy^`aVNd~Qf>k-$r*8Wt2E+%3%+TOZsS@+3Y&XQ&+& z2!sh^cDHIz(N;wF84k0KnmaNJz`>6^4Gyo-N}Kh{#-U*PR*^t#u+Ue30A`E$agSxjcVFu3`C?GAsA=#5etXRqp;j zfrbFvwDxZZ{1on_ik_(f(u`E(=V@A18-s=JnTo>pqC3_%pj
    -
    -List of all members
    cub::BlockLoad< InputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorage Struct Reference
    @@ -116,39 +114,15 @@

    The operations exposed by BlockLoad require a temporary memory allocation of this nested type for thread communication. This opaque storage can be allocated directly using the __shared__ keyword. Alternatively, it can be aliased to externally allocated memory (shared or global) or union'd with other storage allocation types to facilitate memory reuse.

    -

    Definition at line 1052 of file block_load.cuh.

    +

    Definition at line 1067 of file block_load.cuh.

    Inheritance diagram for cub::BlockLoad< InputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorage:
    -cub::Uninitialized< _TempStorage >
    - - - - - - - - - - - - - - - - -

    -Additional Inherited Members

    - Public Types inherited from cub::Uninitialized< _TempStorage >
    enum  
     
    -typedef UnitWord< _TempStorage >
    -::DeviceWord     		DeviceWord
     Biggest memory-access word that T is a whole multiple of and is not larger than the alignment of T.
     
    - Public Methods inherited from cub::Uninitialized< _TempStorage >
    -__host__ __device__
    -__forceinline__ _TempStorage &     		Alias ()
     Alias.
     
    - Public Members inherited from cub::Uninitialized< _TempStorage >
    -DeviceWord storage [WORDS]
     Backing storage.
     
    The documentation for this struct was generated from the following file: @@ -156,7 +130,7 @@
    -Generated on Tue Nov 10 2015 13:58:06 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_block_load_1_1_temp_storage.png b/docs/html/structcub_1_1_block_load_1_1_temp_storage.png index 3b3d3f627a67942f01d9d7daf89890fb7dffdcc5..419de89d6e0278a59e7b9f5cf69972911fb11043 100644 GIT binary patch literal 1890 zcmc&!dpMg}9{yCREvZ$rJH<|wVmehe>dp>3Gf`5SsG#Xbia}jcQYxW>V0KleThWY5 zQR%i~OcKeAOQa}G%Q97$HWQbLOOzpT38jRIOXj2d@6Nw_p68tNp67Yr_xWAU^P~Es zykUop4+8)IL;4^B06-Iju)mHbbUN#=U4pJ-{(ga8YPA}I^2K!`?Rg=D_P05OZ@s#6PI{a03IQKhmQtUii@OGM6Ly?}0;YiG>mJEWIdikZy4jd5$jn0o*4A z7zg{(mnKj0Y#w(Ay6v}1m^q5}&S^Bg=`Y0-7^Er!5{^G(mN1oi|G6XyJ+&-bW1JW1 zL%wrIBzAwLGPB5N8&fHGq0gCCl;TjxEc3oOm!2lJViiX-Bi|;9W8%>Ce_3Rja4eCj z{)28}qxQiuQG1`4ml?7H)9|M1^%td)atWs?)^X=3TTtRhUwRgg6n)fK%MW5%+*4Nd zz|+m*(zN;CKJ+ck${$K#*M-tkO!tG%1y_{L4CMx0MW)QxkQ%@n$b^aN7Ft+|FFmny z&n1I56!4R*exPB$U-Q!Y2Td;Ro zm}*@}ndP}WCr36lH&QKcorhWwQU}u4WX(-f^jHy8`8cc%)P^XFPxgf8pvqK7{LdDp zVXV!cBM|rUbqo#e0@)gyy4#v?KwAhzh4TTuY>gkMq@Jx>;eg)Xj|S@sfn(WcCVJ~b z_6WbcC-`)Jsc{U@EX*4`0O?M^IQwMd_hg6KKWA&e0Vp6qGaP{a!=8{CvG45D5&!=a zZd(60=LmE2vnTcvg#*&=-wOyOHa1xy)I@ zgEYz0iaHK%HbZy7E~czynOJ{Ml>8lUTAa!#7yQ~CA@IfHdbY$HxQ`{BtGhx=G<-QD z?&%Z7(R$ZF*)3A#K{>-71~`QV;M1o~F?eEv;KPpY4{n1Sb=Ka(u6YF~71Ma--5VEN zFZgPG=@7PNH1NXrc3EZKBTr*UlcdyurT+V>7cD@D|^29lK0R7G8IWBMb@rE>r)A}<>7JGPIH$FjX?|eNhV4=FwbyStO z_^z;$6Le(mZgkdSQsk&NO3gZn6HmtX#hk<~;oY9+zi zeN@=DLTSqRxJ-U8e$H`hEZVMmW`%5gjc5BbEN}tq*V@?P<7cQ4_Z%6UdZuv4RJn&n zXG)UMOu5qi-xQU?<5Si*E~R40Rjxh1Cu<0ij^n;bD~e#SiEk1^2@6@G>fLv> zj!I)*!HimIB_#}`=yWk!zj9+c;zSh;|-mePD$j5gbw#BYW~7G zu7W5v-Q4xqVSq1#TJ+w=50x*;_4q1}2iGPIO`PIkGh}{~tT|+AD%QruorD$}53gnN zujq^oi_?>0vX#9_l(S=Q%dd8#ZtF{rC(ZFIdVnm>z`J{+1I=`S59WHLDXCs@*?2WFww@XXb=+M&irExer^b2($h z>QhfvBq}~`%_oZRfaY7ZESeb-vG0JKkikfXqlDNS!CSrT9ox}c?HfKXCa$i2nLypK zxoE;5goioR)}Xg7k5$~}6K;BO+i#jKRT9U^P0NaRmW=bE}+k;uK)`D`#^nXREK9*q=@D=MHZPLiEG0 zvNGdxWs#I%*!AZG_(H#M-0sF@DAw0lU3GbX7Ll7tp&ve^Aoi!rZpH6SnXU@Z gz`*MLsLMaNd#Kn{zi;-6-VFd+NG}wE_j!2kZx`>0x&QzG literal 1919 zcmc&!X;4#H7Je9rs00;s8c;zL5mX4M7y%J`0>T5?!X8Y^qSAy0EJ#=qS%y)%WNpJL zi)priga9J1fkYsULI!D|Wl4k<(CP*xm>`k>4TQ`KR#(lBsrfs%>fZa^I_Il<&Uene z58PcHP`h<@0|0;m9qn)cfWX7pT@?YZn7tBTc++yf=4CIJ%VGH8qm-P}{R51)Czs1r zdGJ;bKB|&%uATtwq!?qbyv_o^PIb`kYp-}EQQ1PY@VSOcfV208Oz+QqP2y%m(8Z1Z zhWu#FUneWW=FygO-$f6O|55s^&vH)nU$mPBeU7w;RjW_&+YqPa<4nsm&<1zEAcC}q z2FJl5KNIjQY%^A4o&f1bO>FUfpF~S)T_?f^Ky?E$W0@2l;Z3x$M$P>r_K|I^1&Y9yU|S+N-gr|my&DciZShHd zWO4|vk+x2yOFcRMtoEu4iw*1zbpHwPDV%TmdlRUjIN|w1K=^)g(K)dA!y*4OV0y2D zTXy8WUY+yHE^gwmE8FI9fh?xPVrQS^YKZnLTD+O+F1^DkFWoklc-4ZG0tDk&oB3%!LiXvxnb_NEY=|KO{nfELo5_U4A zwK9%2XX4giKzz1k|7c=kZcknTmVBh9b9wanfUP=*fM38()pRaey*4bM@4fy>d3msK zrT;fm{Sqwsam#|imDSQPpy|Ca#K3HY6Il(SmEg6aV2=ecCtCTt`l(I|#(u%#ByT>L zpV|;-%e9(h5Gh$p{g=A8Y$_40*15rPv!v9d&tVI_hQFt~(w3Y`JW;@f^H4EuRUGrq z$xNul2xw@u_>*XYx#PU|tA44Q9Fhy%EWJ&XglT0v z4bJ7 z!*U|REwUn{u3B&>x)9czw;0yc&fZM&V1<~Aeia@wcWLoFZFMJOr{44owEjk!xiy6P#Eg7MbBQK16wVxsm^a?yVwc=<&rvf)O^I^hCbJ*P)^u zc%PH0w8hp?r_`r&rXCIu153$eHcsKuD&mjVcE8pBky_9;WEi=(J|dp?dspmaEo39! zxVJbovO0>-9=?_08^xN>%ywrkE)O8TdDG+(ZZyq=2Ie9kWK=@$&WKD$bp(=5JX%>C zO6+JmcZE9Jc0;Xo(3)gBeoRC3=&BR<<+r^RKJAy(8^(&W{FrQK5{}8KJxU(B{WH4O z;v^qjr@s}tF)euOn#73V*vog*oP~JSOiu-3wG6-L+EgTM#*d^6*(l1rXE z>nDUto_*}b-NQPGh2lsP)k0bmy?oHN7%eg2AnL{kYF3gfU+iGK5yRvM$ znna(JB99>7+$Mr+1sWLuK%W&E+jh=&?K_b2V9k4Tiz_P}A$KYzm*X2(qiS@yzu-28 zl?5TmzNkCh$v@Y-A;p@So0ld4>SLXV_TJ=qmf|
    -
    -List of all members
    cub::BlockRadixSort< KeyT, BLOCK_DIM_X, ITEMS_PER_THREAD, ValueT, RADIX_BITS, MEMOIZE_OUTER_SCAN, INNER_SCAN_ALGORITHM, SMEM_CONFIG, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorage Struct Reference
    @@ -127,32 +125,8 @@
    -cub::Uninitialized< _TempStorage >
    - - - - - - - - - - - - - - - - -

    -Additional Inherited Members

    - Public Types inherited from cub::Uninitialized< _TempStorage >
    enum  
     
    -typedef UnitWord< _TempStorage >
    -::DeviceWord     		DeviceWord
     Biggest memory-access word that T is a whole multiple of and is not larger than the alignment of T.
     
    - Public Methods inherited from cub::Uninitialized< _TempStorage >
    -__host__ __device__
    -__forceinline__ _TempStorage &     		Alias ()
     Alias.
     
    - Public Members inherited from cub::Uninitialized< _TempStorage >
    -DeviceWord storage [WORDS]
     Backing storage.
     
    The documentation for this struct was generated from the following file: @@ -160,7 +134,7 @@
    -Generated on Tue Nov 10 2015 13:58:06 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_block_radix_sort_1_1_temp_storage.png b/docs/html/structcub_1_1_block_radix_sort_1_1_temp_storage.png index e49210f10b6cc6e98f0341edbd203b3388dab1f9..3230402b4b67496c2b42be20d024ca37bcf9a196 100644 GIT binary patch literal 2538 zcmb_deK?crA0Ijrc}Y$aW7UrMoie2q(o`oc5h|&SNmMk%*cQWzwwp6~s=Klk_Zxj*;yJl_-#H@LQ@ zktP5D&~`q!|0n?P14I#fXs9Wk?4h&+ikG2>tEZD(E?0!Zqm!7_7io&<+uhL6@Iy)u zSTWRyI_h=|pa3Z^_Me{C0Kkf9=ly#)BKL{d&DaNa49E6911M=C0aTPr z^xz9%)jej5h5xDbbHqr^@Tkk%)nUu%ERCwINFJD4jO8G&N_b2iS#)Ki0EFgm`O%)w3{4Z%ow3+=(% z*~wqVhg8xt-xs}~dDWp|>7A>i8pX+h_nK53ktrQL^%lQ7v3B@T&}dxUDnXl-8l5Gt zh0pJrT-%V)NJ>W3BP*|TCp_va#h;>)5wpE97CPL@jMwAL1gTvtubJ1^z zKGB*uFhqS@JS)dqaBUi_d-Ea!SFnG9hnWXCYB4Pdl@5iTGAg*H)t?vfOfqcoyrMQF zpwyVlRb%(2nniX0bX^}A&IirKZNOKrnQknF&}px3&6h&pah4`_+YdZNp*fz>eWeEn;xCnDZha z6^p_=PBlAxQF~gvBa?C#c(tO+n=w~6*j$x0vKeYlEMHF3I`g_RNl#$g@lXQ9*3U=6 zbCXa{1~M|5`VSiMCV25c;xnr_Fh;ZY5yJ%-65yyQfWW6vB3-uLEgbl|b5MP}aLUpc zS{ohkDkbWj-`Vm0W;C(3+=5Oe_`Nbj7oNB`cM(TVyf@jsb2=Cad!_u+uF z2eTK$sz)exrBZo|6vqIbWr`eh##bJZoP825><}=G1klcN4q`4}b z-D@oQ&FS%|)H<{cH3#+GLZR6{7Ut{t3R~X`-;mlr4uwQZ3r>dpg?Be;+kGQz@{3m< zbvxE~)+qL9`Hu9>GR)WM^F$E(TrB2Z=V){AUY^Vk%TJq1#fh$PT6;cql3MsC$G6sD zW7fW*^;fr6J?XF_-^-?pPQfcG=TXG+ciAU8vj%G+enQ+p(_IMAC1yCGexTW0=Ov05 z`?*7fS=E_XHRd*j-ocN$!V&~w`TR<>Y4Yl)qhLQX14f^ap7pu8mkNrrRK(=1XKDQNg|DuTgYIj=F*m_CQ65nJ*V*qUKd9QJ<#LJH(md&x($4Z0R z3%x3KMFjReOBD+%7%wMGE1_R)pzSkOB2%kwsAuEU| z91`A#8PCPR=l7!2zWuD{Q9i3%0*C5{m-^M$olDK+d;j~l&I9|G{wY`M+haLW^8AJv Pptzi!-1b-OJDu=1Bz&wAkOK6`It(qEiRAMlto!F@c#b9)5sYq3XXizO3s;y`)r9*Ak zSR-iCB#5o{j-s`c*ixh|35hMyNXaBd=RWuTb^pHa^L*cT&U=36obUY3`+XT78vX-j1OUiq+Fm#hikKrD~z{$g- z_N?Qf(%5g@{&q;d>g7f`19u$Nd$E39E4ZYl6X!EWwgp}dGf@xF7gtn$P*R_LY(MV9 z?4Ze)Skh)58;>6`BgT{57{X@J18s3I!kwD~i{wX0AupK21`H+?dM(8x}0UYz>_ zNm}?0CVRu_1YW|F;s4x1NuSSRfBN@H(C;DM3-HWb5=hAbq~3E^wSEAIg=2twriYFJ zcHaRUxWf`~n{9TK85{8qK>7zj@Mo86C*-$9x$qM^KB)G`04a&r0jTuuzktpf{|#De z$NfKm{Gg|b@44E^7iS!@JEfiBYF8U~@~^r82gIM$4kce}1-wz(p~UJdApmn(Y^ZN$ zzq1HgO+G@E5#~IvO+yL6Lv`^^NzYYAv=LowJP+XpdBg;%q*;tqvE5}8J>e!D-p{^K z_u-{8d9(G;vJ5|(BpT`pe@f<+>zMbnxyPTb)>C}A_1X9YzW~!%l8lEHM;@JHhYLNd z2Mr@9O?A3HT0!>Hawuv?*j@WOa?GA`GD1!R4IJ_dZF}3C?y$fssx9~Iu^KPV1)`$* z(qv2p;M~(SmV@o*;_FSl=?HmEI(VMEuhxTsp8NIXBZG=}k$0Zc?6Kd!?m0>dG~o#3 zHf*WYV{@~qrX{ztM{k+vM*OBtL9PjmzE?q8v`*Vw<_I`r`25n{zqI zu*=Ps+oo=zk9GQ()3)wzf>#-Z{Q#W>?8GDnKBarZu`{al2Die^hsrRw`t*JgqpH#D zT&vlpf}Icnf$zp#xZ%tvgw+I4m;XO(tCV72;T&lOK>8FhC>i4*WG1ood?Kz{z6r5w z1mO`IWRO7@L4719gp#BJ6!4xz3{09*mtmsk*2`6jCF7f_(5P}qBomWBA=NNe(pr7& za+@h_?%+yuY??Ac`-(EUXY|Q;1nLF1^w) zaD=83Y-m_v^{P?lLRfh@NY|{Eer45T5)wOz<7v*PF9D$z2dOmYJQVX9dND4)c-7f- zozyVT|+DMz6KYV`;QX+zT+72@-~a24L?&K?mu=HM$ccI>BLQfE21t^S){h8x46NZD6?CHzv!$q=#~*O}g<2nw8++lnaUhuKAz?pyRA=T-!wF zKP9!ebppBEZ_PM)**^E%B`25fSSi&7Ic>a!BroEyZsOM z9^V+i_4?zv14!oRpQBhV`y6fxJGc$=pF0_j!!DsXRF>i8#$StuDKb$efsih4_o0H_*`In34MhfT-TyfS z0hq6&?>Ujo)2pkNt$IgxJ-t|G4iGQfi%u7+tbR`T6JpT7 Aj{pDw diff --git a/docs/html/structcub_1_1_block_reduce_1_1_temp_storage.html b/docs/html/structcub_1_1_block_reduce_1_1_temp_storage.html index 586dbbed1d..1eaab02852 100644 --- a/docs/html/structcub_1_1_block_reduce_1_1_temp_storage.html +++ b/docs/html/structcub_1_1_block_reduce_1_1_temp_storage.html @@ -97,8 +97,6 @@
    -
    -List of all members
    cub::BlockReduce< T, BLOCK_DIM_X, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorage Struct Reference
    @@ -122,32 +120,8 @@
    -cub::Uninitialized< _TempStorage >
    - - - - - - - - - - - - - - - - -

    -Additional Inherited Members

    - Public Types inherited from cub::Uninitialized< _TempStorage >
    enum  
     
    -typedef UnitWord< _TempStorage >
    -::DeviceWord     		DeviceWord
     Biggest memory-access word that T is a whole multiple of and is not larger than the alignment of T.
     
    - Public Methods inherited from cub::Uninitialized< _TempStorage >
    -__host__ __device__
    -__forceinline__ _TempStorage &     		Alias ()
     Alias.
     
    - Public Members inherited from cub::Uninitialized< _TempStorage >
    -DeviceWord storage [WORDS]
     Backing storage.
     
    The documentation for this struct was generated from the following file: @@ -155,7 +129,7 @@
    -Generated on Tue Nov 10 2015 13:58:06 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_block_reduce_1_1_temp_storage.png b/docs/html/structcub_1_1_block_reduce_1_1_temp_storage.png index 4e2cc421920ca52ec510589db0335660d12e1e9c..a02391625872dfae4245b24648cf66189724aacf 100644 GIT binary patch literal 1538 zcmcIkdpO%y82*K2E#?G(@-*aI)eB^<)IkrOqwpCDpGI0~MB?;oA5T0*%<;6wl9Q*=j zyOxMarnNNiW&OvS=?bb24$0{F`$_G{xQAN!Q%R+?qKm^D-3YF&hCbC^T|&-{ux&)>b&D{MJ-z9*RZybOP_s)*oqwZIv%TeV zc)l6>7^1nnBVeexZV8;+=@tAZafM8deH+{iI~xy5BfD0_ z9{UVbXP%&6vy6~F#SgYJy=aYe+Nx!O_)Z|-XeR7$>CCHTW@KlHuFw#H=KBXhbyDxnKd0%yy@;ShCI4e9DRR20V5vZk<_8W~D2R#+*4!i&Vv-F7$zUq*9i)!-|4knE`0$CWG&DvG z0hLrp^^gCa{f;6p?ss9#Vq^ug2R97KJ&2o0%rp^SCUL?RrOvx!NyEcq?pFsLyI36k zaQ>u|Q1_KV{{qw&x4Z)$pf&};Hxi~tpJPIDHZd>n=wP}Zz2v#N_-dETUC~7#u zH=-aaGxW;rxaPWk#ZBY+M=)LnKggoxy&akA{0qTW*vk}msN+whw3V$E*|nCWxJ#4{ zT)!l_)qxb9#l;dt;o@gWJ>7}w`0agEn1w~H-VZ*P6BkEq3g-+(zYb5lE3lHLMfElM zuo(H_mN;DDEmSMjNiQWykHNFYoFet^^PhB0ZVes647InUG|acNLtr-om0VK8Wy5QurF>Ek) zcK`uTz4|L+M;_wJ3jZT|Eth7BS?EaUOqZ$9hUn_er{$q;Y aHsCH%rB?32V`#8p16Vh2Ox>Y#l)nJY71g!? literal 1576 zcmcIkdpOit82-78tr@bZFfolXwCuR$Qai=?g&1;a%!U$$$slaw8VzGo?WDrwmYDKj zDMhY>VNj}-Vpwt+gGvms7T@1?a?dnV<6~*5Ri&87h>wg``=pv)~?qB6)iZ}XWz@D2~%JhgIIZf=l zF&S&BX)1c+xHzzM>-mnPJf@4tB&@Pjw+uUQ;8~MJSNU!%WZj(W;+D7zX&Ts_mUkn_U5_+WE-&xLdN#X+@1k+^1T%Sw5lrw0{sZHL zyt_yKgcBGW6q>h+=k8D)u6^BO?3@CUmbHDa!nn>tUG}5Is2Y#?xrv7&W5`uqLCA85 z5OW!c_>Ajp)C?LEm;~R!s|V)th#KQ*1NKfh+eY4NzN@`hx}jI<#!Md$t1hAP93zMh zNzKEyMGx|8tIlqz3;(XJ7pT!#O37g*`5Pbl?GF~7jjw{a}0WxY@A9ghYdCfFQWvciM#asm# z^gMv~0wrY4ZPA-a^vxu#(oM34;_m;O>~5o#pMkeE$WLL&L8%h@K+V=Q|8+7$3VhHe zTpmCAwQy>-Bb1sv_Vh@wYrffC$BEmNjdsto!I2oAO;1h9bL)PW{wh|xvdgm6Sx!{k z#yi&ROWwYr%jLO!rilqa4x&X`|Dz7bKFrJ}txeW<9}HOy35*GwuEa(&XI6z%pPSk* zmKd0?S+f?Kpf15z4hUZPlB><#>)31IHtgJM(Pr!oi~E2k0W9imA75_@yjfb1Szi(q z6Rqf2d>@Dp*CwqOj(d}9By;{m$;H?1htG;L?a={6P8bhWmQcRCyyuRp5){(9`?hH4!UX zhSSpxiURr25yhPOy@^;DSud)eh@0lu834XiR<;cdy7`p zJq)8c7?So>Q0P*r$o5+Q5F%Z^a^%Rhwqo|43OnU{DhwD@s%4tpeDAl>YwKjGopD*Fo+}1(6kq?G^S& zug{#M_D6w^FS1jYb7lKAI&|?8m@@$~V9ynmF`CKG5!vtGpc)$Xlk{-8@1G{U+p$F8 z8c#(eJ4hprd(?mN+`h_&$;QN>NvvGI!C$mzujxPpq-TE4tC<93>T(~VApTgxohPTh zhI_@Y?$iWw7W8Ul_?^)>Br1#st9m64mw@gw}q~T&eedE=Q_03fN_c#CUMxirz%MxC;S{iVX1akdO V`f%2ulDMn`I4cLs;^XIO{{WX5+i3s* diff --git a/docs/html/structcub_1_1_block_scan_1_1_temp_storage.html b/docs/html/structcub_1_1_block_scan_1_1_temp_storage.html index 24191d9f7c..1321899f8e 100644 --- a/docs/html/structcub_1_1_block_scan_1_1_temp_storage.html +++ b/docs/html/structcub_1_1_block_scan_1_1_temp_storage.html @@ -97,8 +97,6 @@
    -
    -List of all members
    cub::BlockScan< T, BLOCK_DIM_X, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorage Struct Reference
    @@ -122,32 +120,8 @@
    -cub::Uninitialized< _TempStorage >
    - - - - - - - - - - - - - - - - -

    -Additional Inherited Members

    - Public Types inherited from cub::Uninitialized< _TempStorage >
    enum  
     
    -typedef UnitWord< _TempStorage >
    -::DeviceWord     		DeviceWord
     Biggest memory-access word that T is a whole multiple of and is not larger than the alignment of T.
     
    - Public Methods inherited from cub::Uninitialized< _TempStorage >
    -__host__ __device__
    -__forceinline__ _TempStorage &     		Alias ()
     Alias.
     
    - Public Members inherited from cub::Uninitialized< _TempStorage >
    -DeviceWord storage [WORDS]
     Backing storage.
     
    The documentation for this struct was generated from the following file: @@ -155,7 +129,7 @@
    -Generated on Tue Nov 10 2015 13:58:06 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_block_scan_1_1_temp_storage.png b/docs/html/structcub_1_1_block_scan_1_1_temp_storage.png index 63cb1a34ccf18be53f3765f70c38517058e1fccf..8cbd7d3bed612b9f82bb67289c2cfae458b83647 100644 GIT binary patch literal 1531 zcmc&!dpML?9R6}A%(`qynlO{fC67{Bw_K;0hjFjmZ&PleMahhzhc1*`pBZeCWb7mv zvaE(;*sa|UOJQRev}+JCZjCV+GyCPQ{cHc-^E~Ig?|Fac`905h&+m6~J@FV7rM*f3 z0H|QkIC}$t6bN%q1!;J#n7_~sgSMxeuZu(?f$7IZ{*}V+BAD$=7Kx^;rOG4>`^KakGHEuPIY36Zr zZhkJ!QP5BeH!zq)=wn^MJk%>)GSAV@&Ps6C^hbg%(8NJoVHiceaJ zcJY|?PL!ov+@A45lzUoG%I+eX#xojiHX3jL(`@d#6`e|CFt@P?Yi37vBbo z_~#bZZ&KI=cK>p#ov04m^V)XM@Y)^X`-|v#IZhB{@nncT*t@2g=&`hbjV%}?x4JyI zSon5^(eyY33Ov*?<=5Yy3e{$joa!W+xdD|ZOH4IEi`a?2{iOBEIi5)k)y6YP_dOdN zIJRU`Lc5g^+(5|u(Zr*^rGk@68N~Hd_c6GWLV=l;!X#H2tCx1Korapo)1s#5O%RX# z5su}TA$XcxH)z--(S#dJO?T^Mf=HjVz*Vct-c#iAF<+~vY({LNBN61_{AwO1sJG=fq6JQk2{lta3X zeMPCW>MXz{`=*RJ0H=b(a3)w**paiS|DVh_V9@Oh-z7)j1ZV@&D+$j~*LR(lF;)x% z;G&Y!z7E$EAQz^lO1Xr8Gv|?MQBMq#J2h?W7Gm{led#)th+#^mM>0+X>!sS2o9N(Z zV9ZnqGqP`GcVNeN4x#Jh?Unf3Aiw5*iQs@@QGp>DIA~Kt5adUFOEwE_j}3rk&k5Xn zS{m-#btI6|%$wN3s&6wsCLswgh`k!fvMUHcE#e!L1=nLJooRU27MFpMA!M|QUoVH!kfWJH%`p~bbj$I4RK+ECD zshkJNkXH&6aRFTyu~piO5tm689T0|nZQaH*mH8A(d$4)4=rc%mf?{sPwTev9p-&;N zcRq{ylcW3O(VvbXi3{{d#N5V-k`Z2%w?`Z)vzq(CZ`IoJbx`BQ?D(4Gjp~Kr7nqsE z(9O=0%oESsl*fAmd}eGH-7~CD`mHUPBE+H;_YE1y?9qNc(^40EC8lQ48Sy)Yhx5PF zReX+Geif>c_}SNOOeiyxF4Gmi=b)nyFZ`O~l!~L$Cq6usDd+R=#!1m4wf$dp)e@>@ zC$`w}?KZ*?NbKhSFd**T_7tVOWmA0f)I2yq=|m=f!;kK@ugr3~NiZTO;JQ6GTHZz8nc(46gWQGzuslLsdgFz#z&7Az7ejHNbEJ!+S28^q~Tfa zNae6Sy`cG+f$V52RX%}yvZs4`Ep_Oz?(XqP@25soQI%=y(7aJ>)3&muPUlh;D4%xQ zs<{4;Ub|wWZO%S{K0&?COZ`LY=4#%E(I%2(QB*jU**8F6wB|J|^Z7UU!SWw<7>qD( zVWRMFnYq7!kh$9E>EmQfDcn_(DVwPGEC1bW|7vEjv_Uf%p`^fJz&|yBb-_C`9K*8y E0sigI@&Et; literal 1563 zcmc&!dpML?82{Xg%4&;Ptwnq?RBn~9YAu%eTxJZ0A(t4rJ)@G0a;b-8#j-PIhzKoh zMwD@xt>>vjG5fa6qyk#GCSITd-mTw&vVZEJMa6R=Q-#1p5L2}c6CtPp}qqE z07ayu9R>iT$r6l~myxVRZ$q9+j2hb6!(J>FOW>P#A1`G;{au2-%=Y$nxpc0kWGf$o zadihINn2y2*TWP5}t_eBDX$a1%=@00;gOj7Dntva$YCuL63jqd<2N z>IcNh0k5StU6ti$lvz#WJe|ue)t*FehF0n{SzoSyvxMMt=EuH2ldpw5Ci?qH%#zVTPE<81_qF!CTRyd{4fdn-Ljx#(Xo9vq z$HUKF)hG@7QCH<9HgOJFd)Da;=h*54?a)f^s-kR>BR9^p%Yfc)5?x5+7?o6v{@^7#0 z2X!kBxJFYcWo?d+B4+HtyawMcfja9PSG$KqF7@U)^{nf!jmYIWtjDpl#Wx$cUP6=4 zHF3Dd16fPD6=yu_>Ct`+JD4=e3GqQbQDwMc$lG4m1rsUHA!v9ef!dZbAKP2Ut^wcr zONjksJ(wRmltuW$wD;GT4*a(W2?iD}Ed0Ijg8ri}tT8ELfIlPC4yavQ4F;ulwW}p0 zu#td_u`G5^JCND&Cny2_*OLESi-m`&82;9Z*?2b523BKWmX3iB`N9!}SV12StTeOoh>jOV zvb{B4PBzc5W731{!P+1}UWV6(n)rbYQD>R<)QriGL}(K&N)c-lV7v_+bt96oK_;V$ zU|wk&+oieTghV-^ycaKwnEU2NP4~@J>y+N3*0-VQOA#U~yGVPSR^Kxe><2#NMNbW(is1q+w%oMgDMpaa#Ayg&_p2C3VoOwfX{4dS(7occ}FB%5YqL$3&0M zPTFeitCf}MNKil`T+y$(y});;drb;dCrh9^7Mo87Y zyAU>$5_%e5-o0@D!lEZkLt{yM&~EA@`rrA)*XHXvks9NTu#25AL%e(H=I(XA9 zqk~zDKGl>(u|6Q{%~07Nm1N$KzRnD|ltDY(AHTE0)nLum_ec{9RS0je3X$ssB9#k| zE$4@IV{0-1QOd6uRJBvQQ^`PEXLe}w$qw=PzMqp=EdFql* O4j}Da?W%tcO#TNs!`n&# diff --git a/docs/html/structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___t_r_a_n_s_p_o_s_e_00_09dfae03f13932c7dbdb41be30a5767ba.html b/docs/html/structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___t_r_a_n_s_p_o_s_e_00_09dfae03f13932c7dbdb41be30a5767ba.html index 0684e58a1a..f5ba061848 100644 --- a/docs/html/structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___t_r_a_n_s_p_o_s_e_00_09dfae03f13932c7dbdb41be30a5767ba.html +++ b/docs/html/structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___t_r_a_n_s_p_o_s_e_00_09dfae03f13932c7dbdb41be30a5767ba.html @@ -97,8 +97,6 @@
    -
    -List of all members
    cub::BlockStore< OutputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::StoreInternal< BLOCK_STORE_TRANSPOSE, DUMMY >::TempStorage Struct Reference
    @@ -125,32 +123,8 @@
    -cub::Uninitialized< _TempStorage >
    - - - - - - - - - - - - - - - - -

    -Additional Inherited Members

    - Public Types inherited from cub::Uninitialized< _TempStorage >
    enum  
     
    -typedef UnitWord< _TempStorage >
    -::DeviceWord     		DeviceWord
     Biggest memory-access word that T is a whole multiple of and is not larger than the alignment of T.
     
    - Public Methods inherited from cub::Uninitialized< _TempStorage >
    -__host__ __device__
    -__forceinline__ _TempStorage &     		Alias ()
     Alias.
     
    - Public Members inherited from cub::Uninitialized< _TempStorage >
    -DeviceWord storage [WORDS]
     Backing storage.
     
    The documentation for this struct was generated from the following file: @@ -158,7 +132,7 @@
    -Generated on Tue Nov 10 2015 13:58:06 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___t_r_a_n_s_p_o_s_e_00_09dfae03f13932c7dbdb41be30a5767ba.png b/docs/html/structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___t_r_a_n_s_p_o_s_e_00_09dfae03f13932c7dbdb41be30a5767ba.png index e18b9e3ff5ac5a7d37aae30589a2b6acb1435657..a45827361d4fe1a1aa7a174b7f5c013d807c8b90 100644 GIT binary patch literal 2361 zcmd5-d00|u7st%B(rK|$Bt;E#933T1$x*Q;Qqi!oL^8Lea7)QC)Lf#~T&Khl%@i}Y zDM%wV6HORPTyYI7m1xFXA)F>xOqnY+-}8LmU-RepKF_`Pyyu+voO^!1bI(mV>h7$u zL1zO91kymepgcgJwM;-hs-_Iwdu_A;1xWko;S-1CaycMQNM=d5nr;J<;^J^PDk;NI zV5@e)!~Hl2a9Umb_7j$Xsuhhoa3WUeL%~qo2xF@Xok(~thfKq$_iSCLTjMVgf<}+? zTGuFrveHf&8XiVnPf3Pv*|OU(5Uqx`q@01Zxll4oRG3VK0NaHY2jbC`Oh9m=2%5PR zg^&(=l$K!xXwqR|w6_+-jdID>-{k10dv)3M`~uFfo$_9LRahgiYr3 zgWWuy=e2mrd}3+i7i2)t61Uel%1pdAG> zsPM*bZ>eTwzVgXnvWn)t{@OfSQ_H--H3t>ioEtY(yb(Dw)Aat?^YhV$| zmYAU}ZK$w2rd%EGR8nfh?u)Nx#W(Z7rZv(6iRl*`QCpn3XR>!LptU5BaeFFzovoEX zP>JhkRK)Hh8Y5TY>P|!^EHB8&2iUjgD3J9-ts=-RO<0n=fB6e+mx({{K47%W&ZnbRb95Cv|CIlxV3i9?O0~Jd zZeE*ut@&;7Gab=7WJ_R{L87 zq#t;5?;@2hR3BmQ$X115Pu`7kODt|EBpiP9A|F~bI5u3~MbtGTOuyNi6WNZ45sE+i zatLnpN{Q%rV@!VV){c2qX|v{x9eWXKZ5JoN_xxTYfp{b#uDFR{g1pWu zOo+!uKMJQM8DNfyroyD>_l#VdiPj6qkLY`q=oH_73oh^SQjI8+4M;nGUyBLRyKyeL z2d(1M+I)&Di`Hj(#CuQ{3o+-bP7JPR*|@@M@%z$7 z;@Oh&17Olj=dhESdSjE*1XBqP< zQcumDdJnuZ_{|bYC$e?Q!!U8lZ(mK`CB0 z>q@LKnmvxMF@dWW#K>sOGST87C$R%p8$A{*?tk54U}uBPX$d+h%y9{aPL*`a?XGD1 zKM!KKp0cgVeou|U^9}B<38qBGn$ky47j5kdyP`K2{awX^Fq8>z(GGhNV-%)N@(yzS)TU+@+-|b(yso2J8D~S$%|`!4sfbgQu}EU^{T*&drXg zGxpPj^;S^_*-zZp`tl3iMLKkeOTFz2Tg%yrm$<6^1NXZPR+!AZrrdm+svG`$v1I#3 zE^e{Fm=tMvCxw|e9)-tP?}5EIe6Yp&Z4qaYdU@W)3-iN@N=DaYu z+dQrrFdwdKtnZ)IEP@Np)COTE)f0zsiNm7Pj1&#o4{gRRns^Q9+!V*bN^rzjJ9O$P zlS@IGKMAUfE$V51CfbDZn_N)JN%stk!wzXVbvl&Y%{qvz#9Y1L+w1((u`Qm$%H?{! zvCc~SLA@I0Yc72+~5%V6&BJSD0TP7Y#uwr1{<`U zN&b!_df(E|PF$8pUVi&DmF##vOrp+_8;B)-Lmvwfg5`NJSfj_P`fc931W6DNq{`4| z#H^F*Y9=@ig5*XqUMoo!iR9Pjyg%Pe$~(VKeDamBY{%pUN5q#=&H)|%H@|?KP548o k++;P)EvHvU?90^8n*@eqS25mw$3ehCA96=k{1|ZcZ+_ZK*#H0l literal 2399 zcmd5-eLRzU8(%Fcr_pniS*!L4NnT@=h{VhmvUz_wOp}+q9Wk3nh1T+zsdvEw^s;Ib|&(lhN1O$0c;ROctv!AOQ; zb~p2h-P^pvC}Wckp%+*q?+R<5n|#nzIh~WOri!u2NmsK^%~0PkV3<2l_-Vic!x-cs z^^Eg341cv?pHOYUUa4fzro5TvO>klSkave&P&m1%IY0Iz^+M$zhO)jmQP47HDv zFSi-&9^bJ~i_a4feWf8oa9U!`QN2aI*qD5^HXR@XZ3qhk?m{_2FBdaZg*a^B+R_?h z_Z6GNR{%1_&%z#uPawJM>V0~$X_U3h8Mhow%z82|Xw%tzY&+YpRA-js>f)rw>pIWNKovCVBQS#$clcWJvJJ z&C=A<2gITNKot0J09gIAy|qT~O`wJm$^T>h)(Z)o5x1i6$}%bVkIq<32PuQV3&^$g z2LZIBGk4e>w6znYd^vb!v8Q*l{`%&R>lrdY?DD36sq7KFE;hyJ_R) zsb|dZ_#1!Wq3|C7E)ZB>&sb?4yOjVm76`n45YSfNh~oYtT0Beq|0a5&UY2Wz)x4dH zH_$Be$76%Bwi^-veJX!P-vt7F%1^v1K)6VQuUul~>rn>n>P~wtm_r%IkZLfz8RFgHAEsQE<#4G&ev-zjdsQ940Q3SZ$_8hdm zyy{)Chs?P@e`Cb6N;FQ82H6He%of z_kDWii$T9Y@oEP=6?a6UBuaB>Xz9!27Qea!Ksc@H1v58gnlm32b5`)^M&YG*rEb;##*;M<> zf@$&IMpceQWZ>l)IaYVqFB!Kf-a#oI=`#6|jMR7CHH}{)r6z{b%k`6A9`yWzA};s5jY@YB#W_MXa<->)%(60epFe9$ zwYBO&^Y1Z6jU+txqnkx1h~qriuig~i87llkShENflgqJ^%W7(tLNnu3ccPhO`eUKB znO$9vD*veKgPCxz9zT`A0l7eZZj3*@5KhbMd!5_=MxaPo41;M{QeKu)?L7$^9@vh2jn)(wI9DgMg0Q_il9z`r0_5VBN<41lEPo zUPU{PL2c+>R|}j{%&P2vLlr-d&z^_Lu2+^b9r_en9& z@mMZ8=MgDRVUlr^cU~^3o7SR&k3pF;=ARSI=J~Co1IK7un)UO7I7)q}BaYsq>P%CK(# z!pJnj@Y{@^jF|L%uW(?E^nIM~IGOr){s-~OTK_d(ZPvg%Psu>O-i2-uurMeG>#9?P G
    -
    -List of all members
    cub::BlockStore< OutputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::StoreInternal< BLOCK_STORE_WARP_TRANSPOSE_TIMESLICED, DUMMY >::TempStorage Struct Reference
    @@ -125,32 +123,8 @@
    -cub::Uninitialized< _TempStorage >
    - - - - - - - - - - - - - - - - -

    -Additional Inherited Members

    - Public Types inherited from cub::Uninitialized< _TempStorage >
    enum  
     
    -typedef UnitWord< _TempStorage >
    -::DeviceWord     		DeviceWord
     Biggest memory-access word that T is a whole multiple of and is not larger than the alignment of T.
     
    - Public Methods inherited from cub::Uninitialized< _TempStorage >
    -__host__ __device__
    -__forceinline__ _TempStorage &     		Alias ()
     Alias.
     
    - Public Members inherited from cub::Uninitialized< _TempStorage >
    -DeviceWord storage [WORDS]
     Backing storage.
     
    The documentation for this struct was generated from the following file: @@ -158,7 +132,7 @@
    -Generated on Tue Nov 10 2015 13:58:06 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___w_a_r_p___t_r_a_n_s_p_263becc1ca5b47586740c2f7bb0d0145.png b/docs/html/structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___w_a_r_p___t_r_a_n_s_p_263becc1ca5b47586740c2f7bb0d0145.png index bcc4ffbafff41ff4b256005b2e0b3c08815c7802..17488343334073928b37f4809d0e0497ba9f86db 100644 GIT binary patch delta 2079 zcmY*a3s}<07RRvMG;`}#rfZtludH2yZl*>ln&CCycE$9cirZeyTPiB8Z-uPP9-EIP z!Od6sRuozkK}d9aY#I65Lc zM6oaOI4Z+PS~-q$IMBYHSPm2FZ>!$SgST}+`W9^m?fi$&`yYN*a}s;LjWKj4{yt;G zE50p%%l&lJ zVB9`@-2hmmE%Qb5@2M3?Qp@pW@1PV$5lzcfiuGcqqW_tyeYSQw7mx+II#peJscr4&wI4@pd#ta)kw6=dh(PH7<@T=)%*pmEuwDiIro=l3~--0r4Aqe-- z4Fa}U2T58T*`ByQ)Yg`skcEhLRAIXK$F5q18A3q5v33%r!=rkh@d~K+K@dSMaGTIIL?vpAeVWUB}wg&&7Iwkfmb?-+cY0cF566oNTuyqEpW^PY6 z{vE*hk{8j#cpGsFe1OOXpZc?#63vWH6FOUXXFD&CgF|-udpPpbUHTVuCVSwKHd4+O z4)awR!HaRb%VLtrE_;c-n3215v9$~3@eF1Z87NJl)5H;Sx1oQMJnUGm79r_)nnzoj zDPYd8ITy&M;TEtXWOA^{H(w#b-4auLTMOB>Z0KVAzXEXH?nanKnT3U(%{+ULVj>VKmWo^)#sB)mzD7?1MNC1Ls^ zOVBc=9RGDT^|acmIZ`t2{rWb?vNh|jpPYxYiC3dh%eDXl2SFE$I)=DRd}{c4-5m%4 znk=_|s}6DB$J^Q-51`L2ry%Dw+P4*rbDF%SjU54H%)Uhes+V6^;5ztx5!bK63eI-< z15nJt;q}~|Q6AlF*rilA*Q3<4c}+b#RR>FvG`fvX&A;3jjsH>Kgoqu| zzRfYdm{Jt=f*$W&H6wq43&?Ah;2W!nB3z~1e8Qieb0Ln}$(3l1Vn*qZ%p4d+vIQ;- z7T3)X^iFwYq>qNF#)meA0tQi3UECPL)JOjccUIU&JTyRrr8&v@R{Q zl0T}EJH~be&O`*;-8eMlN0M)z$eAW&5#kDGN%39R5-LtwlII&}^sL|DFFAFD5wxSRws=4O_6TWDTbJK!8zLOvEI8ONHwO8D0Lt_^qcZd zZrm6{RrKd^49|w;q3f5^=pg0{0UA8Iu3uK# zhg%vG6+eEj;U$D?1;JNzK!9!|0Nt9nrvnFfJHWG6RTqoA2*~B@o8#VdA&vRn$C6|E z7CGR6xTW*Jz)cnc6C;`in@DF>%brV4m6ji|K-DuseO>_u2y16KQfpJvpTweM~B$w!) z4}3Uvms$rv!ovr%CTg>o_a&{eoFiQMU|(h@+8mjV{#odJb_&9?2T>kpL*LDhNhgv~ z2r?q!5!ywpK0FPk+oUhBXg6U6E418pLSK{OO~-wZzB(s+#jaW(vw5hfe_CiUA%A0K zNXSrp@bq`v^_c)ds7oV8Gs>qB&b?Z0PnLP|Pf5w5;0-dz4H@oZAObNiPKVU0Ohi(5 z-KRqL9nE@o;LkMNyQj!Ej%}o4LyXq_p3&bjxBRu|Njz+9_ay-%_&9N~l z><}_G<}#fmm#qymVlt|^wk^5L?!S4??{}U)f4tw%=k>f_ug~lAyx-3|{re~1KL~QA zc$_)qi`Eepiel$%>huZzJch>8d4H|--)RuX7`C>a55X0Z8{sUZX^NQ!9u_(!QRIr z=?<-gdB{gJU3yc}GPk7mv?uHH{^4DfCsoyi|lRTzA99J#fjWNAa$Adocs@}z7F@&TCHvv1AIYiNfMRp7Rt$e{)izgo7RM!w2>Qc-(hw2Kni-QE zM)F=|7j<2-Cp({;YA%fK`7#*wDSC!-8h4IUKKC&PD%~n>b<5NKeY4f-b0?74Ll%-}MzPJ659xRung`%h&oXmaoluY71zak%5}PLK`CaFKLm8 zw_o=&!5O5p1EjQ8>KcL8oCJLX(wQ#m+PDLxmj*Hdxv6D)XnI;~^`8C}+<55!@God9 z&(+d3CqemlZFICrTV8AgU!(UAYDv+{Z8n7D(sUhiTMf5;m2GoHy8eLJ58pltj{5H8 zWvGhiD;Pe)T!`09UXQ7lxP2NsAEq|sVOW@H2CpN0xuAXS2`TJNsT`JS_!fn_l8bTb zOrA%prb;-LT$q>YHajR!FpmEqURzmI*565ibQUm7 z{aW2|97tAlS8QN}0MXV|c?9slwV5_1uD^jH7d=AW8>T|->g30)mQ<-X*xbd1lLLBz zyIK5jAjxJN)!$AC2v4Fr922*XJ*k8+jARyXR)qj zBrmhqe0z5XwNuM$)Mr9I2qRPM)k$l|UuAutBbS$-4T7!e(6u(h49L`*UZL zFtFOvY)^kjY12&QvyjjUE_*-;*c;VRX0>x8zsq&<=a$s?Jb%iIkO`mSGwccnMQ%zcjo%4w%G>%jCc=e#F;D8%c2NV*q+W5n77T*9@d3o>LEaTkL4ctowcX5!Bp z4@YW2b>@V|%BMUf67vk`F%GeoST%Bgw;Pg8z6gGID0*~jaf->%JhE$&#c+PTRD1j9S`gMOS6o0=$knI*cI>^kS5Rf|ifmufe_r7I#-e<36o2`($q66u(iKvfeLKR0tA>9V4h_2YZq0HE;WMJP`fBly8#h&o&I5`~} z`pu{`T=t#~n)3OeE^g+*LO&8Uo05T z_S87R57?0(ddD^NZrKv-&>h>hS8@U;oTf4@?{rJ?!`1RSlD?`gUkRoJ?Hc(hLKusN zCjPyyQx_z9RAupPd zN#>e7{i26Rzc=rT8na)w8{?DuB+G%|BTZ9Vjb1y|m!mi%PL(&}iZRCZ9WX#N?S;Y}H0261ODCbA3z31g;myzwaoCzEHw&NVV zP}Z$kzl$rSk3*B{5mT_LrEWye7yKHBppR{o!zqu-_Fpw&@_2RTQu0{j^zlkZi)@J^ zF1DyjqF>@@_d!x{Fd8Q`O5;s9;bY=O2~o#vNS+o)FWV8~Fx%!0;n~gT(SQ#8X(cek zpfkw}Z|fg}G$bW@C6uITgA|7nIslg&KqnidmzFp$-NpWS-o=hN#1}+SDigSTygqtV z8c?HiV8tv*Mo3rup~J92mH(~b7uFn)D@{7;_WDIT2f5aZjZe(LE7W-
    -
    -List of all members
    cub::BlockStore< OutputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::StoreInternal< BLOCK_STORE_WARP_TRANSPOSE, DUMMY >::TempStorage Struct Reference
    @@ -125,32 +123,8 @@
    -cub::Uninitialized< _TempStorage >
    - - - - - - - - - - - - - - - - -

    -Additional Inherited Members

    - Public Types inherited from cub::Uninitialized< _TempStorage >
    enum  
     
    -typedef UnitWord< _TempStorage >
    -::DeviceWord     		DeviceWord
     Biggest memory-access word that T is a whole multiple of and is not larger than the alignment of T.
     
    - Public Methods inherited from cub::Uninitialized< _TempStorage >
    -__host__ __device__
    -__forceinline__ _TempStorage &     		Alias ()
     Alias.
     
    - Public Members inherited from cub::Uninitialized< _TempStorage >
    -DeviceWord storage [WORDS]
     Backing storage.
     
    The documentation for this struct was generated from the following file: @@ -158,7 +132,7 @@
    -Generated on Tue Nov 10 2015 13:58:06 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___w_a_r_p___t_r_a_n_s_p_8d170856b7ed1df0ed565731a681b449.png b/docs/html/structcub_1_1_block_store_1_1_store_internal_3_01_b_l_o_c_k___s_t_o_r_e___w_a_r_p___t_r_a_n_s_p_8d170856b7ed1df0ed565731a681b449.png index 882b42e32ccc2e4f240ae340b7cc01c918d478a5..29e6fe7c925e6a4fb0021dcd219d9d63f0f6eb49 100644 GIT binary patch delta 2080 zcmZ8i2~d;Q77h(UwFOivD3GA+AYu^#n-t=LbwDFX0fnfjXprQECG1Q1aYfn!+VB7g z5QMUrkVIt(2@TXz%Vq$DKx4upOMrv`f*~eAXhPq-dGqefoqOk;bH8)W_syJ}X(-$I z@FJk??R6|5NlRK1oFHHuZTSN6RMGp1Z)6&mXNO{EDH@IUKY5@UD_cBmUwPNxXKQ8N zP9J;gJpZn!%yN@}I~YGbN-L=Y=~|=1D@)mG;m#{lTgBE7g#Y*o_Ts<=!C3)G|JYA6?@?3oDOp6VBib^dPyU28lyvjW7};H35(H8~n;G*X@3!BUs3mg*pI zY_e)oM~=3p0ku}Lo&Swm+{gSM$$ji9^FsBHf<6aDi3~J{)?UzBf`xkqXglb$-*$ak zwoL$U8ZTya=5Ef>Ml9~b;V*gH zQ>m&X=vnldUjrS!JCpQPeD}>zB~&CyiyuQXf5#?@qBBG{q4)mvps&O*+QOLpvw#VH z18Jz>yW%BA1luhDNlaMRRk(WWLuHFI;=-rwqUGa?yAd#!)PD^#QY|r2NJTS&^&`$G zM#XLU{_#O&0yk)aZjScQ3ti}@L9j^+14|~X67Pwq4%~&C!zd2BqCU+ha^VXhRp-dB zgx)UNk9>|~eJ_=1`L=+e8;(FC47|uk`|l<{7A9-~%#dvRJz+tl?op1*4cS^3B%!KzBOL7PP&GR4^xSiz{twKBk1^X4 zB6uOmu-VzM>M1v?vAMimHDvUdB_d7?vu3_@;s~n*YYj`C5lge+uH2h+&iI85HYJ?KASEf9mL!JgS^e5ytC zC{aL!gXetBua!Nmkgs&qymvIKiUedj2D^ zL4RqaOw2feq4)na-QT@!R zFPqd+PvfJYrrpz4&PP@m-TzEt_3-o|9spZ8yv!7C{@Hf_Ax${MbU34F zbPQv&D7hWCifxB&QOso*5;L2bGv{}HkKaGPzdn!0>-~AZU!TwW{dqh;(#@$`9{c<0 zJ3F5Bx~U;4oI;JB+^n^wGc9rD{HMBB&F#URIUz&5zBgk1bcXu{Ww39KiCI^5ih6xz z*0nh&*cS&eBcHV{*J2eb80kE(y9jq(a_%0Kr|zm6cqglPJ8*tyV1{{56}P{!!lwJR zFA^Yovfs|>tZs;RM3T?eju!C-Mt2@1*FDn~7S-2T><6o>ZOtk`_@F)P9IHK4jc$%q z??E8RN})t&-%{GyC3!WP(Y&HW8=`PYqY#~>w%n`s70&KP(N8&El+GJJh=&Hy zmVg@PpQDN>!?hn>eIhv63TE3S59n2Js8yN4KDCiipAgTEp``fTtDP1({h44D3*Aqv z6j7Y{tsf$7x+Yz%)o1*SMjZuILwIBOf~+Bdj$!f?=2&Tvk_($B84Qf7FW%B8u*sfZ zc4hXnM2!4SN?{;3bbz$#0KEStK20A;~Ls-&VUOjUKxS zbBZ9JMV}iZ8F;PO{vm;)+Z=ioIcxj?>d4`^*NXd?&*$=Y=Wv>v1}qNALHinY|yGoX6V z|HK&BKjPxDBNC*sn)YC0En$ylEHkIn)mzzJLfk5LWkevz}Sm9xOBs;WT^4;gxQTz_8Qh) zO1M!D4BX^vL_DH>vgqne@_3>@vP2S6c6B}4eo+xMTw8b30FOv3@di#q4_e~N{WyjX zTU_tcgKO*+M&qJe=#dcG+lYkmxa9G$qI;2P?+dqlJ#*a7gciN-R|q1M8ZpK`rf>sv zY*7Jp%unn?+e>BCa*; z5hE#2YITcgPK$r}IiW0v(;>;SQsH^{du%4F0&=mdY;56mGZT^A&y`(jKwB^pn8fQ+ zoTK8KWMp^Q)Wq${En)dTniIId^JJ;QznI=c4fZ%G;7qIE`tg#*CL?9Pyw6N{2tBh z@9T`B_j0tRVJIvjw72fHaOw2*36$vDV&G-j`LRRBz0SmyzO~mP0Sp|;ziqfZ={W7~ z!s8y(fLvZK_d{^(<@`oNKbAgSJ0epVggsIEW*~-fI8#LS4I}7FA6xUKmUM9?4<7(I zH77OeI}y6C7<+zl>{CFhfgg)GLl{I<&IE)e%`T`)dF)QH=*9jIx>f{kg(kS@eZow) z5r5i1;=pz*y$@WRc~Vu4!3j1%`X^SXw|8vokB-UpF^=&IvNltk{*FmSsj}u}pAnyiX8eD1eSOKBc-jp_;j4YE?@{H>B+Y2g;;tmn@-aFMG ztiu}Rz!TpVTPu4lepZ1A_dtD13X~)N(p<>bG3V_BiZP~~iSUlLZae$1GGBhMxO8dx zxcpM-L?ZZNUOKN=-?sk(oEiJm!8C?tl%Os;kdfIUvGSgFL{e=}Dcxt2L(CTx`{qolFrMYiEVIwbmBlux6cG^^aXF|)UrPkx^^0i!lFWxN2f+{ zWdI|FM@j@bdaE%PK3iOot&o!JHl==aV6kdPsr%lfSw#~)Jge-h7mHzDtERaYTjZ6Y>WF>|VFQp{@&dUdNZ}wEVmj5^=xG987qkeL&wh<2t^m7>X+2-mG@AKyXYxw4s^}5&s0!0y zJFKdgYhdsn%e3l;0oj#S&_Sgx5*s@5$)Z_SqWcy8P25c^U%?kQ+P}@Dh%qQlHGlVj z{2#U-usFG7pZgym?*S>ZJ9PiX^bPHO;J=~TY?8)XMeBNy`X~M;>yAC@=zrmR?(lCZ f9xif$QqS5+O9`$wO$U7s0)d>J+#D;<{rk>eR_j4} diff --git a/docs/html/structcub_1_1_block_store_1_1_temp_storage.html b/docs/html/structcub_1_1_block_store_1_1_temp_storage.html index 57daaccc5a..0c2fffd3ec 100644 --- a/docs/html/structcub_1_1_block_store_1_1_temp_storage.html +++ b/docs/html/structcub_1_1_block_store_1_1_temp_storage.html @@ -97,8 +97,6 @@
    -
    -List of all members
    cub::BlockStore< OutputIteratorT, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::TempStorage Struct Reference
    @@ -123,32 +121,8 @@
    -cub::Uninitialized< _TempStorage >
    - - - - - - - - - - - - - - - - -

    -Additional Inherited Members

    - Public Types inherited from cub::Uninitialized< _TempStorage >
    enum  
     
    -typedef UnitWord< _TempStorage >
    -::DeviceWord     		DeviceWord
     Biggest memory-access word that T is a whole multiple of and is not larger than the alignment of T.
     
    - Public Methods inherited from cub::Uninitialized< _TempStorage >
    -__host__ __device__
    -__forceinline__ _TempStorage &     		Alias ()
     Alias.
     
    - Public Members inherited from cub::Uninitialized< _TempStorage >
    -DeviceWord storage [WORDS]
     Backing storage.
     
    The documentation for this struct was generated from the following file: @@ -156,7 +130,7 @@
    -Generated on Tue Nov 10 2015 13:58:06 for CUB by   +Generated on Mon Dec 14 2015 13:11:17 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_block_store_1_1_temp_storage.png b/docs/html/structcub_1_1_block_store_1_1_temp_storage.png index 9c40a5bdf7dd37f952ade3069eb531aaa101fc67..e08c8fc3531653dc35510163da2881442ed2e3f4 100644 GIT binary patch literal 1899 zcmc&#dpMhC8-FcjEn>wMha%I(w58P|R8?)0G~N<*r5PnwzKv)a+Gy*T6?3(tpI)u< z&4^P=(9WThV381%evw30hgig+bxwlN7DdTk`hWl3b6wB#{+{Q)@85GDuIGMl`}*ME z+NRn70Kh>{3;_UOWC;82hCwTqm;DO*82fq?v1+v%g0pkW(f8UjAha`iJf7z5VI;KO zeVyQQ4uF)l$3QpH9ssn~LCm+rIF0!NSW!OCfT z$oH*{zW&$y+cB7&w6tbR6yG!RKI$uQ^+(YTWExcFqD(;hOkxT;3pziYd8US(84Ae= zdnDeOig#%X`s1%;6DEmXy}YZj<4gCMRx&SDC2G?S(j($}y?d8*JH`F+#T;axs|~2u zf%pcZ%+|6)R<5WH&P>Rb$YPIsMAr8iOkoKJJ(7PyOB&oz*wE8_;Ysg9=xX*J>epcO z93#>cKDLs7?PmcNZw+?Th4XQxsV7K-^a^g*Qdv>^Ld~r#=9wMx(7IB;!%JrFF~?5h zP;l{s00kQ@Z$8c_bw=Pb|EPuBdTz+*l!c#MQ(0}OSQ?KnNNj(f8)*Lg-IAafB)IIv z6;vn37fqZJle6kOax4$N@P8Vb`POJVrswGlH>VPtA?-5!5L!nb$M&gqRAw@)vY1z) z>%TC#QWWtlY6!_!Y$n+x^e39VuS!9MF%tRplFQ1e*~bGGslP#q>^$!Pu3Met{x*6I zTu3^5@L;p0WUo zCmYVVbnXr3A6H{c-y9~&aw2D22Kb?YqF2kK;+ARRrm?$b1e#R>lEf{e@ei=JN?7b0 zAby6lNnxG)|J6qb^x01}*fX~r)b(UaXGL0Jw}#nS?U`7|JT7TouxQ;Gmb6;yL0aq@ z6db+oVOI6}TbBZ1&|jP{_L?H7g3sda0tW+A@_qW2{8#V6D_) z;C1HdO8-X1e3(e_N$em{9{(~xL=YZ9Q?0qK^&_S}J@2y#pTS~8dXYop>*eYpHMO9O z^w^y9d2yxmxIx?EWFbNB@0jNtu8R9?no=BmQNl76x5c8BmpKEuir96uC|k*KrUm!T zl@_?j)~i)-G76|pg%kPbkczpPC53Qx8Nt9e%=g6bsh%O5yEZK8guwhNc7H^Z>WELU zyP!@K^oYjFwKlo5@6sjo@NA@g)%1#Mbars#nBAoY{IqAl52)U6Y(xus!;_e`p1Buz zqXa$a@%u-Plw72=!q&^j7W%RnsDv|hVNQ9v-HaQyA%*X<7;s!^L5Y{Bu*qSD#6ez9 zP{%v%BjJs@7n9;Uo-`gZPph3${B@U6DxgPQ4i=H^G#BO)*-_!i3jsls<(7G8W7yXa z&X$++1 z(rSA~WlVX*6t{!ZNr81{tJ)$2?szc9<273n5%~AWFB{%blE|$|s|#iD4m)JwF91JDU3|570cc4@h$n zrE+^gg2O-Ry6tHoswGwjX|A(BqGCP;2PGjr?A{hx-7~0I@!p>hD5Q{sH;$dw&1` literal 1929 zcmc&!eOQuL8-J90NVbWA>bR$FB%6wmi{LK(ORt*G-nj=E>pVlL2KjT^Z5q1 zB)h=r<|M3N0027a#!zpdBLEm$Vmyxprsyq{&tK%Ym~9I27+Tg)(*Kin^gE+-g4jtI z!hZCZW60pK(Z}$;2znhW~65X>hN&@cn#*0hEZ_?@YrkU19Y*0_g3m#Lz&X?H?*vgTqv-aJxm z7ZfBLai4DYqY}Gy?xMf?Ym#dT0j#`=Tx#OIy9VWi@R?DD&8wNiAq;duRoX8EYSO+p zQNeDS)jT>$CA$PU&aK7qILLS(`9v}3VWO+HCxm8t>Q&0{P;7o{DaYHmITwBxHVHB? z>$fA|3z;Z(VVHt!!4Efh3x-TCGH?MR5RcH|!M(KFU3AK(jJ5hYo?B_O>Vjs5cUcnB zHh~#9I&`2USRHrY+IC31-VBmEn*lDij;>I3-@st^gA!!yrAN$F8lxgku~<6t4MFm+ z{XDH`Ukn5#ctk-IfW&-5#XcZ_XpcA6_~5`gjn2?YT;2taom`EJ0*A7G0HD_u0$ z8vvLMtU~8nh2O~K{e-4x2|)61d=`9kpZ1-`x}u@-uz#z5 z+>@gutgPiAOM@dRdio)QvxmlV-|25&BJIkFHryGUdW1-~MyWKy+rq){-zolam&A2r zVf%9GzTyhh&T7}}DZ$zvJcnU(c22E$S; zy3M6nii$4N-pxc!v5*8?Q&J;hetJh|Htew%C(?0#C(ZQ0Xr8!J8rxiq=-VPWYhEVK zj6`g4;EEauR|=Yiu4S*+g0b5m$-=F1bA2sm)Gh003{7jWCwXR-wK#WkvhpWeGFR3g zD=E7wN0ro&Gb(X|%okZCM2sH~=hx!8P9qF1^N#ER>Q8jhMGd}pXkNuMR3ag-CdGAjacw?Yyu4#;p;Ei_13f;-`^U~{ zcvPCOYaG=xc7{3Po9XIGCrYgDwc8o4L_O<%zBTgq*RUT3LeF?z`+YmX76xDuN zBVk<8>A`7z@kwcFcsJNknov#@5$S84E{!i^$jO9t+RI@DnSb>Z=MxgtnQgh8Gv1xb zRqhM%9Dms5ZLoQ%$b>d@yeW}|i~n`DQ2L7NB6+_3@v(3@LU@i^UYmej0ehz#<-x%i z3$kjz3>yX&1h?KfppY6t`@+QVh6`9TJLZp=RdltgW$yFFkWBUa&~5pxz5$nFn8~7r z;nfya8(@=WC9PgrQ%tewPrYNfDy3Of$L;gE@AgtATWbU_7t@`%O%nEWe@hN=xkFk> z=qS4&J{<4i?(`t_#d+M@*X7Yg77y8*suU-32NFttYeTs=PrX$|wAw`AL=#e`yTsH}0DDBgO12i935^a?SkTZk6|cH2ibs9T
    -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_caching_device_allocator.html b/docs/html/structcub_1_1_caching_device_allocator.html index 93c2ac1fa8..56722d4440 100644 --- a/docs/html/structcub_1_1_caching_device_allocator.html +++ b/docs/html/structcub_1_1_caching_device_allocator.html @@ -460,7 +460,7 @@

    Member Function Documentation

    -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_cast-members.html b/docs/html/structcub_1_1_cast-members.html index 4907311cff..0c1b5083af 100644 --- a/docs/html/structcub_1_1_cast-members.html +++ b/docs/html/structcub_1_1_cast-members.html @@ -109,7 +109,7 @@
    -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_cast.html b/docs/html/structcub_1_1_cast.html index eafcc6a567..e3d8e193fa 100644 --- a/docs/html/structcub_1_1_cast.html +++ b/docs/html/structcub_1_1_cast.html @@ -129,7 +129,7 @@
    -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_histogram-members.html b/docs/html/structcub_1_1_device_histogram-members.html index b04d9cfa2b..d2b48e3885 100644 --- a/docs/html/structcub_1_1_device_histogram-members.html +++ b/docs/html/structcub_1_1_device_histogram-members.html @@ -116,7 +116,7 @@
    -Generated on Tue Nov 10 2015 13:58:07 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_histogram.html.REMOVED.git-id b/docs/html/structcub_1_1_device_histogram.html.REMOVED.git-id index 1db5f6715b..debf4f81a9 100644 --- a/docs/html/structcub_1_1_device_histogram.html.REMOVED.git-id +++ b/docs/html/structcub_1_1_device_histogram.html.REMOVED.git-id @@ -1 +1 @@ -30d089be1000e90e09af754a87c5ae99340e0143 \ No newline at end of file +085e20f5050ace16d0c8ff8b312cd0608597ef5c \ No newline at end of file diff --git a/docs/html/structcub_1_1_device_partition-members.html b/docs/html/structcub_1_1_device_partition-members.html index 4990ea347e..192302646a 100644 --- a/docs/html/structcub_1_1_device_partition-members.html +++ b/docs/html/structcub_1_1_device_partition-members.html @@ -110,7 +110,7 @@
    -Generated on Tue Nov 10 2015 13:58:07 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_partition.html b/docs/html/structcub_1_1_device_partition.html index 9b1720b59b..55566d2548 100644 --- a/docs/html/structcub_1_1_device_partition.html +++ b/docs/html/structcub_1_1_device_partition.html @@ -112,7 +112,7 @@ .
    Overview
    These operations apply a selection criterion to construct a partitioned output sequence from items selected/unselected from a specified input sequence.
    Usage Considerations
      -
    • Dynamic parallelism. DevicePartition methods can be called within kernel code on devices in which CUDA dynamic parallelism is supported. When calling these methods from kernel code, be sure to define the CUB_CDP macro in your compiler's macro definitions.
      • +
    • Dynamic parallelism. DevicePartition methods can be called within kernel code on devices in which CUDA dynamic parallelism is supported.
    Performance
    The work-complexity of partition as a function of input size is linear, resulting in performance throughput that plateaus with problem sizes large enough to saturate the GPU.
    @@ -232,8 +232,7 @@

    Member Function Documentation

    • The value type of d_flags must be castable to bool (e.g., bool, char, int, etc.).
    • Copies of the selected items are compacted into d_out and maintain their original relative ordering, however copies of the unselected items are compacted into the rear of d_out in reverse order.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    Snippet
    The code snippet below illustrates the compaction of items selected from an int device vector.
    @@ -286,7 +285,7 @@

    Member Function Documentation

    Examples:
    example_device_partition_flagged.cu.
    -

    Definition at line 127 of file device_partition.cuh.

    +

    Definition at line 126 of file device_partition.cuh.

    @@ -374,8 +373,7 @@

    Member Function Documentation

    • Copies of the selected items are compacted into d_out and maintain their original relative ordering, however copies of the unselected items are compacted into the rear of d_out in reverse order.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    Performance
    The following charts illustrate saturated partition-if performance across different CUDA architectures for int32 and int64 items, respectively. Items are selected for the first partition with 50% probability.
    @@ -456,7 +454,7 @@

    Member Function Documentation

    Examples:
    example_device_partition_if.cu.
    -

    Definition at line 236 of file device_partition.cuh.

    +

    Definition at line 234 of file device_partition.cuh.

    @@ -467,7 +465,7 @@

    Member Function Documentation

    -Generated on Tue Nov 10 2015 13:58:07 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_radix_sort-members.html b/docs/html/structcub_1_1_device_radix_sort-members.html index ff31b7998a..c546e19fae 100644 --- a/docs/html/structcub_1_1_device_radix_sort-members.html +++ b/docs/html/structcub_1_1_device_radix_sort-members.html @@ -116,7 +116,7 @@
    -Generated on Tue Nov 10 2015 13:58:07 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_radix_sort.html b/docs/html/structcub_1_1_device_radix_sort.html index af5a8a78df..b609c47c22 100644 --- a/docs/html/structcub_1_1_device_radix_sort.html +++ b/docs/html/structcub_1_1_device_radix_sort.html @@ -112,7 +112,7 @@
    Overview
    The radix sorting method arranges items into ascending (or descending) order. The algorithm relies upon a positional representation for keys, i.e., each key is comprised of an ordered sequence of symbols (e.g., digits, characters, etc.) specified from least-significant to most-significant. For a given input sequence of keys and a set of rules specifying a total ordering of the symbolic alphabet, the radix sorting method produces a lexicographic ordering of those keys.
    DeviceRadixSort can sort all of the built-in C++ numeric primitive types, e.g.: unsigned char, int, double, etc. Although the direct radix sorting method can only be applied to unsigned integral types, DeviceRadixSort is able to sort signed and floating-point types via simple bit-wise transformations that ensure lexicographic key ordering.
    Usage Considerations
      -
    • Dynamic parallelism. DeviceRadixSort methods can be called within kernel code on devices in which CUDA dynamic parallelism is supported. When calling these methods from kernel code, be sure to define the CUB_CDP macro in your compiler's macro definitions.
      • +
    • Dynamic parallelism. DeviceRadixSort methods can be called within kernel code on devices in which CUDA dynamic parallelism is supported.
    Performance
    The work-complexity of radix sort as a function of input size is linear, resulting in performance throughput that plateaus with problem sizes large enough to saturate the GPU. The following chart illustrates DeviceRadixSort::SortKeys performance across different CUDA architectures for uniform-random uint32 keys. Performance plots for other scenarios can be found in the detailed method descriptions below.
    @@ -132,7 +132,7 @@   template<typename KeyT , typename ValueT > static CUB_RUNTIME_FUNCTION
    -cudaError_t SortPairs (void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false) +cudaError_t SortPairs (void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)  Sorts key-value pairs into ascending order. (~N auxiliary storage required) More...
      template<typename KeyT , typename ValueT > @@ -142,7 +142,7 @@   template<typename KeyT , typename ValueT > static CUB_RUNTIME_FUNCTION
    -cudaError_t SortPairsDescending (void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false) +cudaError_t SortPairsDescending (void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)  Sorts key-value pairs into descending order. (~N auxiliary storage required). More...
     
    Keys-only
    @@ -153,7 +153,7 @@   template<typename KeyT > static CUB_RUNTIME_FUNCTION
    -cudaError_t SortKeys (void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false) +cudaError_t SortKeys (void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)  Sorts keys into ascending order. (~N auxiliary storage required). More...
      template<typename KeyT > @@ -163,7 +163,7 @@   template<typename KeyT > static CUB_RUNTIME_FUNCTION
    -cudaError_t SortKeysDescending (void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false) +cudaError_t SortKeysDescending (void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, int num_items, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)  Sorts keys into descending order. (~N auxiliary storage required). More...
      @@ -260,9 +260,8 @@

    Member Function Documentation

    • The contents of the input data are not altered by the sorting operation
    • An optional bit subrange [begin_bit, end_bit) of differentiating key bits can be specified. This can reduce overall sorting overhead and yield a corresponding performance improvement.
      • -
    • This operation requires an allocation of temporary device storage that is O(N+P), where N is the length of the input and P is the number of streaming multiprocessors on the device. For sorting using only O(P) temporary storage, see the sorting interface using DoubleBuffer wrappers below.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • This operation requires an allocation of temporary device storage that is O(N+P), where N is the length of the input and P is the number of streaming multiprocessors on the device. For sorting using only O(P) temporary storage, see the sorting interface using DoubleBuffer wrappers below.
      • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    Performance
    The following charts illustrate saturated sorting performance across different CUDA architectures for uniform-random uint32,uint32 and uint64,uint64 pairs, respectively.
    @@ -322,9 +321,9 @@

    Member Function Documentation

    -
    Examples:
    example_device_radix_sort.cu.
    +
    Examples:
    example_device_radix_sort.cu.
    -

    Definition at line 148 of file device_radix_sort.cuh.

    +

    Definition at line 147 of file device_radix_sort.cuh.

    @@ -352,13 +351,13 @@

    Member Function Documentation

    - DoubleBuffer< KeyT > &  + DoubleBuffer< KeyT > &  d_keys, - DoubleBuffer< ValueT > &  + DoubleBuffer< ValueT > &  d_values, @@ -406,13 +405,12 @@

    Member Function Documentation

    Sorts key-value pairs into ascending order. (~N auxiliary storage required)

      -
    • The sorting operation is given a pair of key buffers and a corresponding pair of associated value buffers. Each pair is managed by a DoubleBuffer structure that indicates which of the two buffers is "current" (and thus contains the input data to be sorted).
      • +
    • The sorting operation is given a pair of key buffers and a corresponding pair of associated value buffers. Each pair is managed by a DoubleBuffer structure that indicates which of the two buffers is "current" (and thus contains the input data to be sorted).
    • The contents of both buffers within each pair may be altered by the sorting operation.
      • -
    • Upon completion, the sorting operation will update the "current" indicator within each DoubleBuffer wrapper to reference which of the two buffers now contains the sorted output sequence (a function of the number of key bits specified and the targeted device architecture).
      • +
    • Upon completion, the sorting operation will update the "current" indicator within each DoubleBuffer wrapper to reference which of the two buffers now contains the sorted output sequence (a function of the number of key bits specified and the targeted device architecture).
    • An optional bit subrange [begin_bit, end_bit) of differentiating key bits can be specified. This can reduce overall sorting overhead and yield a corresponding performance improvement.
    • This operation requires a relatively small allocation of temporary device storage that is O(P), where P is the number of streaming multiprocessors on the device (and is typically a small constant relative to the input size N).
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    Performance
    The following charts illustrate saturated sorting performance across different CUDA architectures for uniform-random uint32,uint32 and uint64,uint64 pairs, respectively.
    @@ -435,7 +433,7 @@

    Member Function Documentation

    // Create a set of DoubleBuffers to wrap pairs of device pointers
    cub::DoubleBuffer<int> d_keys(d_key_buf, d_key_alt_buf);
    -
    cub::DoubleBuffer<int> d_values(d_value_buf, d_value_alt_buf);
    +
    cub::DoubleBuffer<int> d_values(d_value_buf, d_value_alt_buf);
    // Determine temporary device storage requirements
    void *d_temp_storage = NULL;
    @@ -473,7 +471,7 @@

    Member Function Documentation

    -

    Definition at line 250 of file device_radix_sort.cuh.

    +

    Definition at line 248 of file device_radix_sort.cuh.

    @@ -569,9 +567,8 @@

    Member Function Documentation

    • The contents of the input data are not altered by the sorting operation
    • An optional bit subrange [begin_bit, end_bit) of differentiating key bits can be specified. This can reduce overall sorting overhead and yield a corresponding performance improvement.
      • -
    • This operation requires an allocation of temporary device storage that is O(N+P), where N is the length of the input and P is the number of streaming multiprocessors on the device. For sorting using only O(P) temporary storage, see the sorting interface using DoubleBuffer wrappers below.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • This operation requires an allocation of temporary device storage that is O(N+P), where N is the length of the input and P is the number of streaming multiprocessors on the device. For sorting using only O(P) temporary storage, see the sorting interface using DoubleBuffer wrappers below.
      • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    Performance
    Performance is similar to DeviceRadixSort::SortPairs.
    @@ -626,7 +623,7 @@

    Member Function Documentation

    -

    Definition at line 331 of file device_radix_sort.cuh.

    +

    Definition at line 328 of file device_radix_sort.cuh.

    @@ -654,13 +651,13 @@

    Member Function Documentation

    - DoubleBuffer< KeyT > &  + DoubleBuffer< KeyT > &  d_keys, - DoubleBuffer< ValueT > &  + DoubleBuffer< ValueT > &  d_values, @@ -708,13 +705,12 @@

    Member Function Documentation

    Sorts key-value pairs into descending order. (~N auxiliary storage required).

      -
    • The sorting operation is given a pair of key buffers and a corresponding pair of associated value buffers. Each pair is managed by a DoubleBuffer structure that indicates which of the two buffers is "current" (and thus contains the input data to be sorted).
      • +
    • The sorting operation is given a pair of key buffers and a corresponding pair of associated value buffers. Each pair is managed by a DoubleBuffer structure that indicates which of the two buffers is "current" (and thus contains the input data to be sorted).
    • The contents of both buffers within each pair may be altered by the sorting operation.
      • -
    • Upon completion, the sorting operation will update the "current" indicator within each DoubleBuffer wrapper to reference which of the two buffers now contains the sorted output sequence (a function of the number of key bits specified and the targeted device architecture).
      • +
    • Upon completion, the sorting operation will update the "current" indicator within each DoubleBuffer wrapper to reference which of the two buffers now contains the sorted output sequence (a function of the number of key bits specified and the targeted device architecture).
    • An optional bit subrange [begin_bit, end_bit) of differentiating key bits can be specified. This can reduce overall sorting overhead and yield a corresponding performance improvement.
    • This operation requires a relatively small allocation of temporary device storage that is O(P), where P is the number of streaming multiprocessors on the device (and is typically a small constant relative to the input size N).
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    Performance
    Performance is similar to DeviceRadixSort::SortPairs.
    @@ -731,7 +727,7 @@

    Member Function Documentation

    // Create a set of DoubleBuffers to wrap pairs of device pointers
    cub::DoubleBuffer<int> d_keys(d_key_buf, d_key_alt_buf);
    -
    cub::DoubleBuffer<int> d_values(d_value_buf, d_value_alt_buf);
    +
    cub::DoubleBuffer<int> d_values(d_value_buf, d_value_alt_buf);
    // Determine temporary device storage requirements
    void *d_temp_storage = NULL;
    @@ -769,7 +765,7 @@

    Member Function Documentation

    -

    Definition at line 428 of file device_radix_sort.cuh.

    +

    Definition at line 424 of file device_radix_sort.cuh.

    @@ -853,9 +849,8 @@

    Member Function Documentation

    • The contents of the input data are not altered by the sorting operation
    • An optional bit subrange [begin_bit, end_bit) of differentiating key bits can be specified. This can reduce overall sorting overhead and yield a corresponding performance improvement.
      • -
    • This operation requires an allocation of temporary device storage that is O(N+P), where N is the length of the input and P is the number of streaming multiprocessors on the device. For sorting using only O(P) temporary storage, see the sorting interface using DoubleBuffer wrappers below.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • This operation requires an allocation of temporary device storage that is O(N+P), where N is the length of the input and P is the number of streaming multiprocessors on the device. For sorting using only O(P) temporary storage, see the sorting interface using DoubleBuffer wrappers below.
      • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    Performance
    The following charts illustrate saturated sorting performance across different CUDA architectures for uniform-random uint32 and uint64 keys, respectively.
    @@ -908,7 +903,7 @@

    Member Function Documentation

    -

    Definition at line 511 of file device_radix_sort.cuh.

    +

    Definition at line 506 of file device_radix_sort.cuh.

    @@ -936,7 +931,7 @@

    Member Function Documentation

    - DoubleBuffer< KeyT > &  + DoubleBuffer< KeyT > &  d_keys, @@ -984,13 +979,12 @@

    Member Function Documentation

    Sorts keys into ascending order. (~N auxiliary storage required).

      -
    • The sorting operation is given a pair of key buffers managed by a DoubleBuffer structure that indicates which of the two buffers is "current" (and thus contains the input data to be sorted).
      • +
    • The sorting operation is given a pair of key buffers managed by a DoubleBuffer structure that indicates which of the two buffers is "current" (and thus contains the input data to be sorted).
    • The contents of both buffers may be altered by the sorting operation.
      • -
    • Upon completion, the sorting operation will update the "current" indicator within the DoubleBuffer wrapper to reference which of the two buffers now contains the sorted output sequence (a function of the number of key bits specified and the targeted device architecture).
      • +
    • Upon completion, the sorting operation will update the "current" indicator within the DoubleBuffer wrapper to reference which of the two buffers now contains the sorted output sequence (a function of the number of key bits specified and the targeted device architecture).
    • An optional bit subrange [begin_bit, end_bit) of differentiating key bits can be specified. This can reduce overall sorting overhead and yield a corresponding performance improvement.
    • This operation requires a relatively small allocation of temporary device storage that is O(P), where P is the number of streaming multiprocessors on the device (and is typically a small constant relative to the input size N).
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    Performance
    The following charts illustrate saturated sorting performance across different CUDA architectures for uniform-random uint32 and uint64 keys, respectively.
    @@ -1045,7 +1039,7 @@

    Member Function Documentation

    -

    Definition at line 601 of file device_radix_sort.cuh.

    +

    Definition at line 595 of file device_radix_sort.cuh.

    @@ -1129,9 +1123,8 @@

    Member Function Documentation

    • The contents of the input data are not altered by the sorting operation
    • An optional bit subrange [begin_bit, end_bit) of differentiating key bits can be specified. This can reduce overall sorting overhead and yield a corresponding performance improvement.
      • -
    • This operation requires an allocation of temporary device storage that is O(N+P), where N is the length of the input and P is the number of streaming multiprocessors on the device. For sorting using only O(P) temporary storage, see the sorting interface using DoubleBuffer wrappers below.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • This operation requires an allocation of temporary device storage that is O(N+P), where N is the length of the input and P is the number of streaming multiprocessors on the device. For sorting using only O(P) temporary storage, see the sorting interface using DoubleBuffer wrappers below.
      • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    Performance
    Performance is similar to DeviceRadixSort::SortKeys.
    @@ -1181,7 +1174,7 @@

    Member Function Documentation

    -

    Definition at line 677 of file device_radix_sort.cuh.

    +

    Definition at line 670 of file device_radix_sort.cuh.

    @@ -1209,7 +1202,7 @@

    Member Function Documentation

    - DoubleBuffer< KeyT > &  + DoubleBuffer< KeyT > &  d_keys, @@ -1257,13 +1250,12 @@

    Member Function Documentation

    Sorts keys into descending order. (~N auxiliary storage required).

      -
    • The sorting operation is given a pair of key buffers managed by a DoubleBuffer structure that indicates which of the two buffers is "current" (and thus contains the input data to be sorted).
      • +
    • The sorting operation is given a pair of key buffers managed by a DoubleBuffer structure that indicates which of the two buffers is "current" (and thus contains the input data to be sorted).
    • The contents of both buffers may be altered by the sorting operation.
      • -
    • Upon completion, the sorting operation will update the "current" indicator within the DoubleBuffer wrapper to reference which of the two buffers now contains the sorted output sequence (a function of the number of key bits specified and the targeted device architecture).
      • +
    • Upon completion, the sorting operation will update the "current" indicator within the DoubleBuffer wrapper to reference which of the two buffers now contains the sorted output sequence (a function of the number of key bits specified and the targeted device architecture).
    • An optional bit subrange [begin_bit, end_bit) of differentiating key bits can be specified. This can reduce overall sorting overhead and yield a corresponding performance improvement.
    • This operation requires a relatively small allocation of temporary device storage that is O(P), where P is the number of streaming multiprocessors on the device (and is typically a small constant relative to the input size N).
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    Performance
    Performance is similar to DeviceRadixSort::SortKeys.
    @@ -1312,7 +1304,7 @@

    Member Function Documentation

    -

    Definition at line 762 of file device_radix_sort.cuh.

    +

    Definition at line 754 of file device_radix_sort.cuh.

    @@ -1323,7 +1315,7 @@

    Member Function Documentation

    -Generated on Tue Nov 10 2015 13:58:07 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_reduce-members.html b/docs/html/structcub_1_1_device_reduce-members.html index 0af9b92329..cd65af67a7 100644 --- a/docs/html/structcub_1_1_device_reduce-members.html +++ b/docs/html/structcub_1_1_device_reduce-members.html @@ -108,14 +108,14 @@ ArgMin(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_items, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceReduceinlinestatic Max(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_items, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceReduceinlinestatic Min(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_items, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceReduceinlinestatic - Reduce(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_items, ReductionOp reduction_op, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceReduceinlinestatic - ReduceByKey(void *d_temp_storage, size_t &temp_storage_bytes, KeysInputIteratorT d_keys_in, UniqueOutputIteratorT d_unique_out, ValuesInputIteratorT d_values_in, AggregatesOutputIteratorT d_aggregates_out, NumRunsOutputIteratorT d_num_runs_out, ReductionOp reduction_op, int num_items, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceReduceinlinestatic + Reduce(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_items, ReductionOpT reduction_op, T init, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceReduceinlinestatic + ReduceByKey(void *d_temp_storage, size_t &temp_storage_bytes, KeysInputIteratorT d_keys_in, UniqueOutputIteratorT d_unique_out, ValuesInputIteratorT d_values_in, AggregatesOutputIteratorT d_aggregates_out, NumRunsOutputIteratorT d_num_runs_out, ReductionOpT reduction_op, int num_items, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceReduceinlinestatic Sum(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_items, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceReduceinlinestatic
    -Generated on Tue Nov 10 2015 13:58:07 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_reduce.html b/docs/html/structcub_1_1_device_reduce.html index add5d22631..b48c7e12f2 100644 --- a/docs/html/structcub_1_1_device_reduce.html +++ b/docs/html/structcub_1_1_device_reduce.html @@ -112,33 +112,33 @@ .
    Overview
    A reduction (or fold) uses a binary combining operator to compute a single aggregate from a sequence of input elements.
    Usage Considerations
      -
    • Dynamic parallelism. DeviceReduce methods can be called within kernel code on devices in which CUDA dynamic parallelism is supported. When calling these methods from kernel code, be sure to define the CUB_CDP macro in your compiler's macro definitions.
      • +
    • Dynamic parallelism. DeviceReduce methods can be called within kernel code on devices in which CUDA dynamic parallelism is supported.
    Performance
    The work-complexity of reduction, reduce-by-key, and run-length encode as a function of input size is linear, resulting in performance throughput that plateaus with problem sizes large enough to saturate the GPU.
    -
    The following chart illustrates DeviceReduce::Sum performance across different CUDA architectures for int32 keys.
    +
    The following chart illustrates DeviceReduce::Sum performance across different CUDA architectures for int32 keys.
    reduce_int32.png
    -
    The following chart illustrates DeviceReduce::ReduceByKey (summation) performance across different CUDA architectures for fp32 values. Segments are identified by int32 keys, and have lengths uniformly sampled from [1,1000].
    +
    The following chart illustrates DeviceReduce::ReduceByKey (summation) performance across different CUDA architectures for fp32 values. Segments are identified by int32 keys, and have lengths uniformly sampled from [1,1000].
    reduce_by_key_fp32_len_500.png
    Performance plots for other scenarios can be found in the detailed method descriptions below.
    -

    Definition at line 82 of file device_reduce.cuh.

    +

    Definition at line 83 of file device_reduce.cuh.

    - - - - + + + + - + - - - - + + + +

    Static Public Methods

    template<typename InputIteratorT , typename OutputIteratorT , typename ReductionOp >
    static CUB_RUNTIME_FUNCTION
    -cudaError_t     		Reduce (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_items, ReductionOp reduction_op, cudaStream_t stream=0, bool debug_synchronous=false)
     Computes a device-wide reduction using the specified binary reduction_op functor. More...
     
    template<typename InputIteratorT , typename OutputIteratorT , typename ReductionOpT , typename T >
    static CUB_RUNTIME_FUNCTION
    +cudaError_t     		Reduce (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_items, ReductionOpT reduction_op, T init, cudaStream_t stream=0, bool debug_synchronous=false)
     Computes a device-wide reduction using the specified binary reduction_op functor and initial value init. More...
     
    template<typename InputIteratorT , typename OutputIteratorT >
    static CUB_RUNTIME_FUNCTION
    cudaError_t     		Sum (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_items, cudaStream_t stream=0, bool debug_synchronous=false)
     Computes a device-wide sum using the addition ('+') operator. More...
     Computes a device-wide sum using the addition (+) operator. More...
     
    template<typename InputIteratorT , typename OutputIteratorT >
    static CUB_RUNTIME_FUNCTION
    @@ -160,18 +160,18 @@ cudaError_t     		ArgMax (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_items, cudaStream_t stream=0, bool debug_synchronous=false)
     Finds the first device-wide maximum using the greater-than ('>') operator, also returning the index of that item. More...
     
    template<typename KeysInputIteratorT , typename UniqueOutputIteratorT , typename ValuesInputIteratorT , typename AggregatesOutputIteratorT , typename NumRunsOutputIteratorT , typename ReductionOp >
    CUB_RUNTIME_FUNCTION static
    -__forceinline__ cudaError_t     		ReduceByKey (void *d_temp_storage, size_t &temp_storage_bytes, KeysInputIteratorT d_keys_in, UniqueOutputIteratorT d_unique_out, ValuesInputIteratorT d_values_in, AggregatesOutputIteratorT d_aggregates_out, NumRunsOutputIteratorT d_num_runs_out, ReductionOp reduction_op, int num_items, cudaStream_t stream=0, bool debug_synchronous=false)
     Reduces segments of values, where segments are demarcated by corresponding runs of identical keys. More...
     
    template<typename KeysInputIteratorT , typename UniqueOutputIteratorT , typename ValuesInputIteratorT , typename AggregatesOutputIteratorT , typename NumRunsOutputIteratorT , typename ReductionOpT >
    CUB_RUNTIME_FUNCTION static
    +__forceinline__ cudaError_t     		ReduceByKey (void *d_temp_storage, size_t &temp_storage_bytes, KeysInputIteratorT d_keys_in, UniqueOutputIteratorT d_unique_out, ValuesInputIteratorT d_values_in, AggregatesOutputIteratorT d_aggregates_out, NumRunsOutputIteratorT d_num_runs_out, ReductionOpT reduction_op, int num_items, cudaStream_t stream=0, bool debug_synchronous=false)
     Reduces segments of values, where segments are demarcated by corresponding runs of identical keys. More...
     

    Member Function Documentation

    - +
    -template<typename InputIteratorT , typename OutputIteratorT , typename ReductionOp >
    +template<typename InputIteratorT , typename OutputIteratorT , typename ReductionOpT , typename T >
    - + + + + + + + @@ -237,22 +243,20 @@

    Member Function Documentation

    @@ -209,9 +209,15 @@

    Member Function Documentation

    ReductionOp ReductionOpT  reduction_op,
    T init,
    -

    Computes a device-wide reduction using the specified binary reduction_op functor.

    +

    Computes a device-wide reduction using the specified binary reduction_op functor and initial value init.

      -
    • Does not support non-commutative reduction operators.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • Does not support binary reduction operators that are non-commutative.
      • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    -
    Performance
    Performance is typically similar to DeviceReduce::Sum.
    -
    Snippet
    The code snippet below illustrates a custom min reduction of a device vector of int items.
    +
    Snippet
    The code snippet below illustrates a user-defined min-reduction of a device vector of int data elements.
    #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
    // CustomMin functor
    struct CustomMin
    {
    template <typename T>
    -
    CUB_RUNTIME_FUNCTION __forceinline__
    +
    __device__ __forceinline__
    T operator()(const T &a, const T &b) const {
    return (b < a) ? b : a;
    }
    @@ -261,20 +265,21 @@

    Member Function Documentation

    // Declare, allocate, and initialize device-accessible pointers for input and output
    int num_items; // e.g., 7
    int *d_in; // e.g., [8, 6, 7, 5, 3, 0, 9]
    -
    int *d_out; // e.g., [ ]
    +
    int *d_out; // e.g., [-]
    CustomMin min_op;
    +
    int init; // e.g., INT_MAX
    ...
    // Determine temporary device storage requirements
    void *d_temp_storage = NULL;
    size_t temp_storage_bytes = 0;
    -
    cub::DeviceReduce::Reduce(d_temp_storage, temp_storage_bytes, d_in, d_out, num_items, min_op);
    +
    cub::DeviceReduce::Reduce(d_temp_storage, temp_storage_bytes, d_in, d_out, num_items, min_op, init);
    // Allocate temporary storage
    cudaMalloc(&d_temp_storage, temp_storage_bytes);
    // Run reduction
    -
    cub::DeviceReduce::Reduce(d_temp_storage, temp_storage_bytes, d_in, d_out, num_items, min_op);
    +
    cub::DeviceReduce::Reduce(d_temp_storage, temp_storage_bytes, d_in, d_out, num_items, min_op, init);
    // d_out <-- [0]
    @@ -282,7 +287,8 @@

    Member Function Documentation

    - + +
    InputIteratorT[inferred] Random-access input iterator type for reading input items (may be a simple pointer type)
    OutputIteratorT[inferred] Output iterator type for recording the reduced aggregate (may be a simple pointer type)
    ReductionOp[inferred] Binary reduction functor type having member T operator()(const T &a, const T &b) (e.g., cub::Sum, cub::Min, cub::Max, etc.)
    ReductionOpT[inferred] Binary reduction functor type having member T operator()(const T &a, const T &b)
    T[inferred] Data element type that is convertible to the value type of InputIteratorT
    @@ -293,7 +299,8 @@

    Member Function Documentation

    [in]d_inPointer to the input sequence of data items [out]d_outPointer to the output aggregate [in]num_itemsTotal number of input items (i.e., length of d_in) - [in]reduction_opBinary reduction functor (e.g., an instance of cub::Sum, cub::Min, cub::Max, etc.) + [in]reduction_opBinary reduction functor + [in]initInitial value of the reduction [in]stream[optional] CUDA stream to launch kernels within. Default is stream0. [in]debug_synchronous[optional] Whether or not to synchronize the stream after every kernel launch to check for errors. Also causes launch configurations to be printed to the console. Default is false. @@ -368,39 +375,39 @@

    Member Function Documentation

    -

    Computes a device-wide sum using the addition ('+') operator.

    +

    Computes a device-wide sum using the addition (+) operator.

      -
    • Does not support non-commutative reduction operators.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • Uses 0 as the initial value of the reduction.
      • +
    • Does not support + operators that are non-commutative..
      • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    -
    Performance
    The following charts illustrate saturated reduction (sum) performance across different CUDA architectures for int32 and int64 items, respectively.
    +
    Performance
    The following charts illustrate saturated sum-reduction performance across different CUDA architectures for int32 and int64 items, respectively.
    reduce_int32.png
    reduce_int64.png
    -
    Snippet
    The code snippet below illustrates the sum reduction of a device vector of int items.
    +
    Snippet
    The code snippet below illustrates the sum-reduction of a device vector of int data elements.
    #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
    // Declare, allocate, and initialize device-accessible pointers for input and output
    int num_items; // e.g., 7
    int *d_in; // e.g., [8, 6, 7, 5, 3, 0, 9]
    -
    int *d_out; // e.g., [ ]
    +
    int *d_out; // e.g., [-]
    ...
    // Determine temporary device storage requirements
    void *d_temp_storage = NULL;
    size_t temp_storage_bytes = 0;
    -
    cub::DeviceReduce::Sum(d_temp_storage, temp_storage_bytes, d_in, d_sum, num_items);
    +
    cub::DeviceReduce::Sum(d_temp_storage, temp_storage_bytes, d_in, d_out, num_items);
    // Allocate temporary storage
    cudaMalloc(&d_temp_storage, temp_storage_bytes);
    // Run sum-reduction
    -
    cub::DeviceReduce::Sum(d_temp_storage, temp_storage_bytes, d_in, d_sum, num_items);
    +
    cub::DeviceReduce::Sum(d_temp_storage, temp_storage_bytes, d_in, d_out, num_items);
    // d_out <-- [38]
    @@ -425,7 +432,7 @@

    Member Function Documentation

    Examples:
    example_device_reduce.cu.
    -

    Definition at line 219 of file device_reduce.cuh.

    +

    Definition at line 217 of file device_reduce.cuh.

    @@ -495,31 +502,30 @@

    Member Function Documentation

    Computes a device-wide minimum using the less-than ('<') operator.

      -
    • Does not support non-commutative minimum operators.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • Uses std::numeric_limits<T>::max() as the initial value of the reduction.
      • +
    • Does not support < operators that are non-commutative.
      • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    -
    Performance
    Performance is typically similar to DeviceReduce::Sum.
    -
    Snippet
    The code snippet below illustrates the min-reduction of a device vector of int items.
    +
    Snippet
    The code snippet below illustrates the min-reduction of a device vector of int data elements.
    #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
    // Declare, allocate, and initialize device-accessible pointers for input and output
    int num_items; // e.g., 7
    int *d_in; // e.g., [8, 6, 7, 5, 3, 0, 9]
    -
    int *d_out; // e.g., [ ]
    +
    int *d_out; // e.g., [-]
    ...
    // Determine temporary device storage requirements
    void *d_temp_storage = NULL;
    size_t temp_storage_bytes = 0;
    -
    cub::DeviceReduce::Min(d_temp_storage, temp_storage_bytes, d_in, d_min, num_items);
    +
    cub::DeviceReduce::Min(d_temp_storage, temp_storage_bytes, d_in, d_out, num_items);
    // Allocate temporary storage
    cudaMalloc(&d_temp_storage, temp_storage_bytes);
    // Run min-reduction
    -
    cub::DeviceReduce::Min(d_temp_storage, temp_storage_bytes, d_in, d_min, num_items);
    +
    cub::DeviceReduce::Min(d_temp_storage, temp_storage_bytes, d_in, d_out, num_items);
    // d_out <-- [0]
    @@ -543,7 +549,7 @@

    Member Function Documentation

    -

    Definition at line 291 of file device_reduce.cuh.

    +

    Definition at line 284 of file device_reduce.cuh.

    @@ -612,21 +618,23 @@

    Member Function Documentation

    Finds the first device-wide minimum using the less-than ('<') operator, also returning the index of that item.

    -
    Assuming the input d_in has value type T, the output d_out must have value type KeyValuePair<int, T>. The minimum value is written to d_out.value and its location in the input array is written to d_out.key.
      -
    • Does not support non-commutative minimum operators.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • The output value type of d_out is cub::KeyValuePair <int, T> (assuming the value type of d_in is T)
        +
      • The minimum is written to d_out.value and its offset in the input array is written to d_out.key.
        • +
      • The {1, std::numeric_limits<T>::max()} tuple is produced for zero-length inputs
        • +
      +
      • +
    • Does not support < operators that are non-commutative.
      • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    -
    Performance
    Performance is typically similar to DeviceReduce::Sum.
    -
    Snippet
    The code snippet below illustrates the argmin-reduction of a device vector of int items.
    +
    Snippet
    The code snippet below illustrates the argmin-reduction of a device vector of int data elements.
    #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
    // Declare, allocate, and initialize device-accessible pointers for input and output
    int num_items; // e.g., 7
    int *d_in; // e.g., [8, 6, 7, 5, 3, 0, 9]
    -
    KeyValuePair<int, int> *d_out; // e.g., [{ , }]
    +
    KeyValuePair<int, int> *d_out; // e.g., [{-,-}]
    ...
    // Determine temporary device storage requirements
    @@ -640,12 +648,12 @@

    Member Function Documentation

    // Run argmin-reduction
    cub::DeviceReduce::ArgMin(d_temp_storage, temp_storage_bytes, d_in, d_argmin, num_items);
    -
    // d_out <-- [{0, 5}]
    +
    // d_out <-- [{5, 0}]
    Template Parameters
    - +
    InputIteratorT[inferred] Random-access input iterator type for reading input items (of some type T) (may be a simple pointer type)
    OutputIteratorT[inferred] Output iterator type for recording the reduced aggregate (having value type KeyValuePair<int, T>) (may be a simple pointer type)
    OutputIteratorT[inferred] Output iterator type for recording the reduced aggregate (having value type cub::KeyValuePair<int, T>) (may be a simple pointer type)
    @@ -662,7 +670,7 @@

    Member Function Documentation

    -

    Definition at line 368 of file device_reduce.cuh.

    +

    Definition at line 353 of file device_reduce.cuh.

    @@ -732,19 +740,18 @@

    Member Function Documentation

    Computes a device-wide maximum using the greater-than ('>') operator.

      -
    • Does not support non-commutative maximum operators.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • Uses std::numeric_limits<T>::lowest() as the initial value of the reduction.
      • +
    • Does not support > operators that are non-commutative.
      • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    -
    Performance
    Performance is typically similar to DeviceReduce::Sum.
    -
    Snippet
    The code snippet below illustrates the max-reduction of a device vector of int items.
    +
    Snippet
    The code snippet below illustrates the max-reduction of a device vector of int data elements.
    #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
    // Declare, allocate, and initialize device-accessible pointers for input and output
    int num_items; // e.g., 7
    int *d_in; // e.g., [8, 6, 7, 5, 3, 0, 9]
    -
    int *d_out; // e.g., [ ]
    +
    int *d_out; // e.g., [-]
    ...
    // Determine temporary device storage requirements
    @@ -780,7 +787,7 @@

    Member Function Documentation

    -

    Definition at line 444 of file device_reduce.cuh.

    +

    Definition at line 424 of file device_reduce.cuh.

    @@ -849,21 +856,23 @@

    Member Function Documentation

    Finds the first device-wide maximum using the greater-than ('>') operator, also returning the index of that item.

    -
    Assuming the input d_in has value type T, the output d_out must have value type KeyValuePair<int, T>. The maximum value is written to d_out.value and its location in the input array is written to d_out.key.
      -
    • Does not support non-commutative maximum operators.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • The output value type of d_out is cub::KeyValuePair <int, T> (assuming the value type of d_in is T)
        +
      • The maximum is written to d_out.value and its offset in the input array is written to d_out.key.
        • +
      • The {1, std::numeric_limits<T>::lowest()} tuple is produced for zero-length inputs
        • +
      +
      • +
    • Does not support > operators that are non-commutative.
      • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    -
    Performance
    Performance is typically similar to DeviceReduce::Sum.
    -
    Snippet
    The code snippet below illustrates the argmax-reduction of a device vector of int items.
    +
    Snippet
    The code snippet below illustrates the argmax-reduction of a device vector of int data elements.
    #include <cub/cub.cuh> // or equivalently <cub/device/device_reduce.cuh>
    // Declare, allocate, and initialize device-accessible pointers for input and output
    int num_items; // e.g., 7
    int *d_in; // e.g., [8, 6, 7, 5, 3, 0, 9]
    -
    KeyValuePair<int, int> *d_out; // e.g., [{ , }]
    +
    KeyValuePair<int, int> *d_out; // e.g., [{-,-}]
    ...
    // Determine temporary device storage requirements
    @@ -877,12 +886,12 @@

    Member Function Documentation

    // Run argmax-reduction
    cub::DeviceReduce::ArgMax(d_temp_storage, temp_storage_bytes, d_in, d_argmax, num_items);
    -
    // d_out <-- [{9, 6}]
    +
    // d_out <-- [{6, 9}]
    Template Parameters
    - +
    InputIteratorT[inferred] Random-access input iterator type for reading input items (of some type T) (may be a simple pointer type)
    OutputIteratorT[inferred] Output iterator type for recording the reduced aggregate (having value type KeyValuePair<int, T>) (may be a simple pointer type)
    OutputIteratorT[inferred] Output iterator type for recording the reduced aggregate (having value type cub::KeyValuePair<int, T>) (may be a simple pointer type)
    @@ -899,15 +908,15 @@

    Member Function Documentation

    -

    Definition at line 521 of file device_reduce.cuh.

    +

    Definition at line 493 of file device_reduce.cuh.

    - +
    -template<typename KeysInputIteratorT , typename UniqueOutputIteratorT , typename ValuesInputIteratorT , typename AggregatesOutputIteratorT , typename NumRunsOutputIteratorT , typename ReductionOp >
    +template<typename KeysInputIteratorT , typename UniqueOutputIteratorT , typename ValuesInputIteratorT , typename AggregatesOutputIteratorT , typename NumRunsOutputIteratorT , typename ReductionOpT >
    - + @@ -995,8 +1004,7 @@

    Member Function Documentation

    This operation computes segmented reductions within d_values_in using the specified binary reduction_op functor. The segments are identified by "runs" of corresponding keys in d_keys_in, where runs are maximal ranges of consecutive, identical keys. For the ith run encountered, the first key of the run and the corresponding value aggregate of that run are written to d_unique_out[i] and d_aggregates_out[i], respectively. The total number of runs encountered is written to d_num_runs_out.
    • The == equality operator is used to determine whether keys are equivalent
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    Performance
    The following chart illustrates reduction-by-key (sum) performance across different CUDA architectures for fp32 and fp64 values, respectively. Segments are identified by int32 keys, and have lengths uniformly sampled from [1,1000].
    @@ -1030,26 +1038,26 @@

    Member Function Documentation

    int num_items; // e.g., 8
    int *d_keys_in; // e.g., [0, 2, 2, 9, 5, 5, 5, 8]
    int *d_values_in; // e.g., [0, 7, 1, 6, 2, 5, 3, 4]
    -
    int *d_unique_out; // e.g., [ , , , , , , , ]
    -
    int *d_aggregates_out; // e.g., [ , , , , , , , ]
    -
    int *d_num_runs_out; // e.g., [ ]
    +
    int *d_unique_out; // e.g., [-, -, -, -, -, -, -, -]
    +
    int *d_aggregates_out; // e.g., [-, -, -, -, -, -, -, -]
    +
    int *d_num_runs_out; // e.g., [-]
    CustomMin reduction_op;
    ...
    // Determine temporary device storage requirements
    void *d_temp_storage = NULL;
    size_t temp_storage_bytes = 0;
    -
    cub::DeviceReduce::ReduceByKey(d_temp_storage, temp_storage_bytes, d_keys_in, d_unique_out, d_values_in, d_aggregates_out, d_num_runs_out, reduction_op, num_items);
    +
    cub::DeviceReduce::ReduceByKey(d_temp_storage, temp_storage_bytes, d_keys_in, d_unique_out, d_values_in, d_aggregates_out, d_num_runs_out, reduction_op, num_items);
    // Allocate temporary storage
    cudaMalloc(&d_temp_storage, temp_storage_bytes);
    // Run reduce-by-key
    -
    cub::DeviceReduce::ReduceByKey(d_temp_storage, temp_storage_bytes, d_keys_in, d_unique_out, d_values_in, d_aggregates_out, d_num_runs_out, reduction_op, num_items);
    +
    cub::DeviceReduce::ReduceByKey(d_temp_storage, temp_storage_bytes, d_keys_in, d_unique_out, d_values_in, d_aggregates_out, d_num_runs_out, reduction_op, num_items);
    // d_unique_out <-- [0, 2, 9, 5, 8]
    // d_aggregates_out <-- [0, 1, 6, 2, 4]
    -
    // d_num_runs_out <-- [5]
    +
    // d_num_runs_out <-- [5]
    Template Parameters
    @@ -957,7 +966,7 @@

    Member Function Documentation

    ReductionOp ReductionOpT  reduction_op,
    @@ -1058,7 +1066,7 @@

    Member Function Documentation

    - +
    ValuesInputIteratorT[inferred] Random-access input iterator type for reading input values (may be a simple pointer type)
    AggregatesOutputIterator[inferred] Random-access output iterator type for writing output value aggregates (may be a simple pointer type)
    NumRunsOutputIteratorT[inferred] Output iterator type for recording the number of runs encountered (may be a simple pointer type)
    ReductionOp[inferred] Binary reduction functor type having member T operator()(const T &a, const T &b) (e.g., cub::Sum, cub::Min, cub::Max, etc.)
    ReductionOpT[inferred] Binary reduction functor type having member T operator()(const T &a, const T &b)
    @@ -1071,7 +1079,7 @@

    Member Function Documentation

    [in]d_values_inPointer to the input sequence of corresponding values [out]d_aggregates_outPointer to the output sequence of value aggregates (one aggregate per run) [out]d_num_runs_outPointer to total number of runs encountered (i.e., the length of d_unique_out) - [in]reduction_opBinary reduction functor (e.g., an instance of cub::Sum, cub::Min, cub::Max, etc.) + [in]reduction_opBinary reduction functor [in]num_itemsTotal number of associated key+value pairs (i.e., the length of d_in_keys and d_in_values) [in]stream[optional] CUDA stream to launch kernels within. Default is stream0. [in]debug_synchronous[optional] Whether or not to synchronize the stream after every kernel launch to check for errors. May cause significant slowdown. Default is false. @@ -1079,7 +1087,7 @@

    Member Function Documentation

    -

    Definition at line 642 of file device_reduce.cuh.

    +

    Definition at line 611 of file device_reduce.cuh.

    @@ -1090,7 +1098,7 @@

    Member Function Documentation

    -Generated on Tue Nov 10 2015 13:58:07 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_run_length_encode-members.html b/docs/html/structcub_1_1_device_run_length_encode-members.html index 50b1209581..1d579a5e15 100644 --- a/docs/html/structcub_1_1_device_run_length_encode-members.html +++ b/docs/html/structcub_1_1_device_run_length_encode-members.html @@ -110,7 +110,7 @@
    -Generated on Tue Nov 10 2015 13:58:07 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_run_length_encode.html b/docs/html/structcub_1_1_device_run_length_encode.html index d8dcadd9b3..7b745bb0ea 100644 --- a/docs/html/structcub_1_1_device_run_length_encode.html +++ b/docs/html/structcub_1_1_device_run_length_encode.html @@ -112,7 +112,7 @@ .
    Overview
    A run-length encoding computes a simple compressed representation of a sequence of input elements such that each maximal "run" of consecutive same-valued data items is encoded as a single data value along with a count of the elements in that run.
    Usage Considerations
      -
    • Dynamic parallelism. DeviceRunLengthEncode methods can be called within kernel code on devices in which CUDA dynamic parallelism is supported. When calling these methods from kernel code, be sure to define the CUB_CDP macro in your compiler's macro definitions.
      • +
    • Dynamic parallelism. DeviceRunLengthEncode methods can be called within kernel code on devices in which CUDA dynamic parallelism is supported.
    Performance
    The work-complexity of run-length encode as a function of input size is linear, resulting in performance throughput that plateaus with problem sizes large enough to saturate the GPU.
    @@ -219,8 +219,7 @@

    Member Function Documentation

  • For the ith run encountered, the first key of the run and its length are written to d_unique_out[i] and d_counts_out[i], respectively.
  • The total number of runs encountered is written to d_num_runs_out.
  • The == equality operator is used to determine whether values are equivalent
    • -
  • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    • -
  • +
  • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    • Performance
    The following charts illustrate saturated encode performance across different CUDA architectures for int32 and int64 items, respectively. Segments have lengths uniformly sampled from [1,1000].
    @@ -287,7 +286,7 @@

    Member Function Documentation

    -

    Definition at line 149 of file device_run_length_encode.cuh.

    +

    Definition at line 148 of file device_run_length_encode.cuh.

    @@ -372,8 +371,7 @@

    Member Function Documentation

  • For the ith non-trivial run, the run's starting offset and its length are written to d_offsets_out[i] and d_lengths_out[i], respectively.
  • The total number of runs encountered is written to d_num_runs_out.
  • The == equality operator is used to determine whether values are equivalent
    • -
  • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    • -
  • +
  • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    • Performance
    @@ -427,7 +425,7 @@

    Member Function Documentation

    -

    Definition at line 247 of file device_run_length_encode.cuh.

    +

    Definition at line 245 of file device_run_length_encode.cuh.

    @@ -438,7 +436,7 @@

    Member Function Documentation

    -Generated on Tue Nov 10 2015 13:58:07 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_scan-members.html b/docs/html/structcub_1_1_device_scan-members.html index 6b49a89bf9..bb357f7a99 100644 --- a/docs/html/structcub_1_1_device_scan-members.html +++ b/docs/html/structcub_1_1_device_scan-members.html @@ -112,7 +112,7 @@
    -Generated on Tue Nov 10 2015 13:58:07 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_scan.html b/docs/html/structcub_1_1_device_scan.html index daf18ef634..c8cc035d00 100644 --- a/docs/html/structcub_1_1_device_scan.html +++ b/docs/html/structcub_1_1_device_scan.html @@ -111,7 +111,7 @@ .
    Overview
    Given a sequence of input elements and a binary reduction operator, a prefix scan produces an output sequence where each element is computed to be the reduction of the elements occurring earlier in the input sequence. Prefix sum connotes a prefix scan with the addition operator. The term inclusive indicates that the ith output reduction incorporates the ith input. The term exclusive indicates the ith input is not incorporated into the ith output reduction.
    Usage Considerations
      -
    • Dynamic parallelism. DeviceScan methods can be called within kernel code on devices in which CUDA dynamic parallelism is supported. When calling these methods from kernel code, be sure to define the CUB_CDP macro in your compiler's macro definitions.
      • +
    • Dynamic parallelism. DeviceScan methods can be called within kernel code on devices in which CUDA dynamic parallelism is supported.
    Performance
    The work-complexity of prefix scan as a function of input size is linear, resulting in performance throughput that plateaus with problem sizes large enough to saturate the GPU.
    @@ -215,8 +215,7 @@

    Member Function Documentation

    Computes a device-wide exclusive prefix sum.

    • Supports non-commutative sum operators.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    Performance
    The following charts illustrate saturated exclusive sum performance across different CUDA architectures for int32 and int64 items, respectively.
    @@ -269,7 +268,7 @@

    Member Function Documentation

    Examples:
    example_device_scan.cu.
    -

    Definition at line 133 of file device_scan.cuh.

    +

    Definition at line 132 of file device_scan.cuh.

    @@ -352,11 +351,9 @@

    Member Function Documentation

    Computes a device-wide exclusive prefix scan using the specified binary scan_op functor.

    • Supports non-commutative scan operators.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    -
    Performance
    Performance is typically similar to DeviceScan::ExclusiveSum.
    Snippet
    The code snippet below illustrates the exclusive prefix min-scan of an int device vector
    #include <cub/cub.cuh> // or equivalently <cub/device/device_scan.cuh>
    @@ -405,7 +402,7 @@

    Member Function Documentation

    [in,out]temp_storage_bytesReference to size in bytes of d_temp_storage allocation [in]d_inPointer to the input sequence of data items [out]d_outPointer to the output sequence of data items - [in]scan_opBinary scan functor (e.g., an instance of cub::Sum, cub::Min, cub::Max, etc.) + [in]scan_opBinary scan functor [in]identityIdentity element [in]num_itemsTotal number of input items (i.e., the length of d_in) [in]stream[optional] CUDA stream to launch kernels within. Default is stream0. @@ -414,7 +411,7 @@

    Member Function Documentation

    -

    Definition at line 221 of file device_scan.cuh.

    +

    Definition at line 216 of file device_scan.cuh.

    @@ -485,11 +482,9 @@

    Member Function Documentation

    Computes a device-wide inclusive prefix sum.

    • Supports non-commutative sum operators.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    -
    Performance
    Performance is typically similar to DeviceScan::ExclusiveSum.
    Snippet
    The code snippet below illustrates the inclusive prefix sum of an int device vector.
    #include <cub/cub.cuh> // or equivalently <cub/device/device_scan.cuh>
    @@ -532,7 +527,7 @@

    Member Function Documentation

    -

    Definition at line 300 of file device_scan.cuh.

    +

    Definition at line 291 of file device_scan.cuh.

    @@ -609,11 +604,9 @@

    Member Function Documentation

    Computes a device-wide inclusive prefix scan using the specified binary scan_op functor.

    • Supports non-commutative scan operators.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    -
    Performance
    Performance is typically similar to DeviceScan::ExclusiveSum.
    Snippet
    The code snippet below illustrates the inclusive prefix min-scan of an int device vector.
    #include <cub/cub.cuh> // or equivalently <cub/device/device_scan.cuh>
    @@ -661,7 +654,7 @@

    Member Function Documentation

    [in,out]temp_storage_bytesReference to size in bytes of d_temp_storage allocation [in]d_inPointer to the input sequence of data items [out]d_outPointer to the output sequence of data items - [in]scan_opBinary scan functor (e.g., an instance of cub::Sum, cub::Min, cub::Max, etc.) + [in]scan_opBinary scan functor [in]num_itemsTotal number of input items (i.e., the length of d_in) [in]stream[optional] CUDA stream to launch kernels within. Default is stream0. [in]debug_synchronous[optional] Whether or not to synchronize the stream after every kernel launch to check for errors. May cause significant slowdown. Default is false. @@ -669,7 +662,7 @@

    Member Function Documentation

    -

    Definition at line 383 of file device_scan.cuh.

    +

    Definition at line 370 of file device_scan.cuh.

    @@ -680,7 +673,7 @@

    Member Function Documentation

    -Generated on Tue Nov 10 2015 13:58:07 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_segmented_radix_sort-members.html b/docs/html/structcub_1_1_device_segmented_radix_sort-members.html index 6d2dec7e84..c156d08b73 100644 --- a/docs/html/structcub_1_1_device_segmented_radix_sort-members.html +++ b/docs/html/structcub_1_1_device_segmented_radix_sort-members.html @@ -108,7 +108,7 @@ SortKeys(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedRadixSortinlinestatic SortKeysDescending(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedRadixSortinlinestatic SortKeysDescending(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedRadixSortinlinestatic - SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, ValueT *d_values_in, ValueT *d_values_out, int num_items, OffsetT num_segments, OffsetT *d_begin_offsets, OffsetT *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedRadixSortinlinestatic + SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, ValueT *d_values_in, ValueT *d_values_out, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedRadixSortinlinestatic SortPairs(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedRadixSortinlinestatic SortPairsDescending(void *d_temp_storage, size_t &temp_storage_bytes, KeyT *d_keys_in, KeyT *d_keys_out, ValueT *d_values_in, ValueT *d_values_out, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedRadixSortinlinestatic SortPairsDescending(void *d_temp_storage, size_t &temp_storage_bytes, DoubleBuffer< KeyT > &d_keys, DoubleBuffer< ValueT > &d_values, int num_items, int num_segments, int *d_begin_offsets, int *d_end_offsets, int begin_bit=0, int end_bit=sizeof(KeyT)*8, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedRadixSortinlinestatic @@ -116,7 +116,7 @@
    -Generated on Tue Nov 10 2015 13:58:07 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_segmented_radix_sort.html.REMOVED.git-id b/docs/html/structcub_1_1_device_segmented_radix_sort.html.REMOVED.git-id index 1e74b0c016..2d6e20cc76 100644 --- a/docs/html/structcub_1_1_device_segmented_radix_sort.html.REMOVED.git-id +++ b/docs/html/structcub_1_1_device_segmented_radix_sort.html.REMOVED.git-id @@ -1 +1 @@ -d62b06a2976f2d1c4abecbacc5c2f2563c95464c \ No newline at end of file +818721631567b517b0ee143268be8ba2c0a10915 \ No newline at end of file diff --git a/docs/html/structcub_1_1_device_segmented_reduce-members.html b/docs/html/structcub_1_1_device_segmented_reduce-members.html index 1dccab5b9c..9ef3d39214 100644 --- a/docs/html/structcub_1_1_device_segmented_reduce-members.html +++ b/docs/html/structcub_1_1_device_segmented_reduce-members.html @@ -104,17 +104,17 @@

    This is the complete list of members for cub::DeviceSegmentedReduce, including all inherited members.

    - - - - - - + + + + + +
    ArgMax(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, T identity, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedReduceinlinestatic
    ArgMin(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, T identity, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedReduceinlinestatic
    Max(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, T identity, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedReduceinlinestatic
    Min(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, T identity, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedReduceinlinestatic
    Reduce(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, ReductionOp reduction_op, T identity, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedReduceinlinestatic
    Sum(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, T identity, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedReduceinlinestatic
    ArgMax(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedReduceinlinestatic
    ArgMin(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedReduceinlinestatic
    Max(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedReduceinlinestatic
    Min(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedReduceinlinestatic
    Reduce(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, ReductionOp reduction_op, T init, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedReduceinlinestatic
    Sum(void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, cudaStream_t stream=0, bool debug_synchronous=false)cub::DeviceSegmentedReduceinlinestatic
    -Generated on Wed Nov 18 2015 10:24:47 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_segmented_reduce.html b/docs/html/structcub_1_1_device_segmented_reduce.html index ca06c7afda..81d0aa132e 100644 --- a/docs/html/structcub_1_1_device_segmented_reduce.html +++ b/docs/html/structcub_1_1_device_segmented_reduce.html @@ -116,43 +116,43 @@ -

    Definition at line 63 of file device_segmented_reduce.cuh.

    +

    Definition at line 64 of file device_segmented_reduce.cuh.

    - - - - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + + + + + + + + + + + + + + +

    Static Public Methods

    template<typename InputIteratorT , typename OutputIteratorT , typename ReductionOp , typename T >
    static CUB_RUNTIME_FUNCTION
    -cudaError_t     		Reduce (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, ReductionOp reduction_op, T identity, cudaStream_t stream=0, bool debug_synchronous=false)
     Computes a device-wide segmented reduction using the specified binary reduction_op functor. More...
     
    template<typename InputIteratorT , typename OutputIteratorT , typename T >
    static CUB_RUNTIME_FUNCTION
    -cudaError_t     		Sum (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, T identity, cudaStream_t stream=0, bool debug_synchronous=false)
     Computes a device-wide segmented sum using the addition ('+') operator. More...
     
    template<typename InputIteratorT , typename OutputIteratorT , typename T >
    static CUB_RUNTIME_FUNCTION
    -cudaError_t     		Min (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, T identity, cudaStream_t stream=0, bool debug_synchronous=false)
     Computes a device-wide segmented minimum using the less-than ('<') operator. More...
     
    template<typename InputIteratorT , typename OutputIteratorT , typename T >
    static CUB_RUNTIME_FUNCTION
    -cudaError_t     		ArgMin (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, T identity, cudaStream_t stream=0, bool debug_synchronous=false)
     Finds the first device-wide minimum in each segment using the less-than ('<') operator, also returning the in-segment index of that item. More...
     
    template<typename InputIteratorT , typename OutputIteratorT , typename T >
    static CUB_RUNTIME_FUNCTION
    -cudaError_t     		Max (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, T identity, cudaStream_t stream=0, bool debug_synchronous=false)
     Computes a device-wide segmented maximum using the greater-than ('>') operator. More...
     
    template<typename InputIteratorT , typename OutputIteratorT , typename T >
    static CUB_RUNTIME_FUNCTION
    -cudaError_t     		ArgMax (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, T identity, cudaStream_t stream=0, bool debug_synchronous=false)
     Finds the first device-wide maximum in each segment using the greater-than ('>') operator, also returning the in-segment index of that item. More...
     
    template<typename InputIteratorT , typename OutputIteratorT , typename ReductionOp , typename T >
    static CUB_RUNTIME_FUNCTION
    +cudaError_t     		Reduce (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, ReductionOp reduction_op, T init, cudaStream_t stream=0, bool debug_synchronous=false)
     Computes a device-wide segmented reduction using the specified binary reduction_op functor. More...
     
    template<typename InputIteratorT , typename OutputIteratorT >
    static CUB_RUNTIME_FUNCTION
    +cudaError_t     		Sum (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, cudaStream_t stream=0, bool debug_synchronous=false)
     Computes a device-wide segmented sum using the addition ('+') operator. More...
     
    template<typename InputIteratorT , typename OutputIteratorT >
    static CUB_RUNTIME_FUNCTION
    +cudaError_t     		Min (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, cudaStream_t stream=0, bool debug_synchronous=false)
     Computes a device-wide segmented minimum using the less-than ('<') operator. More...
     
    template<typename InputIteratorT , typename OutputIteratorT >
    static CUB_RUNTIME_FUNCTION
    +cudaError_t     		ArgMin (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, cudaStream_t stream=0, bool debug_synchronous=false)
     Finds the first device-wide minimum in each segment using the less-than ('<') operator, also returning the in-segment index of that item. More...
     
    template<typename InputIteratorT , typename OutputIteratorT >
    static CUB_RUNTIME_FUNCTION
    +cudaError_t     		Max (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, cudaStream_t stream=0, bool debug_synchronous=false)
     Computes a device-wide segmented maximum using the greater-than ('>') operator. More...
     
    template<typename InputIteratorT , typename OutputIteratorT >
    static CUB_RUNTIME_FUNCTION
    +cudaError_t     		ArgMax (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, int *d_begin_offsets, int *d_end_offsets, cudaStream_t stream=0, bool debug_synchronous=false)
     Finds the first device-wide maximum in each segment using the greater-than ('>') operator, also returning the in-segment index of that item. More...
     

    Member Function Documentation

    - +
    @@ -213,7 +213,7 @@

    Member Function Documentation

    T  - identity, + init, @@ -242,13 +242,12 @@

    Member Function Documentation

    Computes a device-wide segmented reduction using the specified binary reduction_op functor.

      -
    • The specified identity value is produced for zero-length segments. identity is assumed to be arithmetically neutral for the given operation, i.e., reduction_op(XXX, identity) will return XXX.
      • -
    • Does not support non-commutative reduction operators.
      • +
    • Does not support binary reduction operators that are non-commutative.
    • When input a contiguous sequence of segments, a single sequence segment_offsets (of length num_segments+1) can be aliased for both the d_begin_offsets and d_end_offsets parameters (where the latter is specified as segment_offsets+1).
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    -
    Snippet
    The code snippet below illustrates a custom min reduction of a device vector of int items.
    +
    Snippet
    The code snippet below illustrates a custom min-reduction of a device vector of int data elements.
    #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
    // CustomMin functor
    @@ -267,30 +266,30 @@

    Member Function Documentation

    int *d_in; // e.g., [8, 6, 7, 5, 3, 0, 9]
    int *d_out; // e.g., [-, -, -]
    CustomMin min_op;
    -
    int identity; // e.g., MAX_INT
    +
    int init; // e.g., INT_MAX
    ...
    // Determine temporary device storage requirements
    void *d_temp_storage = NULL;
    size_t temp_storage_bytes = 0;
    -
    cub::DeviceSegmentedReduce::Reduce(d_temp_storage, temp_storage_bytes, d_in, d_out,
    -
    num_segments, d_offsets, d_offsets + 1, min_op, identity);
    +
    cub::DeviceSegmentedReduce::Reduce(d_temp_storage, temp_storage_bytes, d_in, d_out,
    +
    num_segments, d_offsets, d_offsets + 1, min_op, init);
    // Allocate temporary storage
    cudaMalloc(&d_temp_storage, temp_storage_bytes);
    // Run reduction
    -
    cub::DeviceSegmentedReduce::Reduce(d_temp_storage, temp_storage_bytes, d_in, d_out,
    -
    num_segments, d_offsets, d_offsets + 1, min_op, identity);
    +
    cub::DeviceSegmentedReduce::Reduce(d_temp_storage, temp_storage_bytes, d_in, d_out,
    +
    num_segments, d_offsets, d_offsets + 1, min_op, init);
    -
    // d_out <-- [6, MAX_INT, 0]
    +
    // d_out <-- [6, INT_MAX, 0]
    Template Parameters
    - - + +
    InputIteratorT[inferred] Random-access input iterator type for reading input items (may be a simple pointer type)
    OutputIteratorT[inferred] Output iterator type for recording the reduced aggregate (may be a simple pointer type)
    ReductionOp[inferred] Binary reduction functor type having member T operator()(const T &a, const T &b) (e.g., cub::Sum, cub::Min, cub::Max, etc.)
    T[inferred] Data type being reduced
    ReductionOp[inferred] Binary reduction functor type having member T operator()(const T &a, const T &b)
    T[inferred] Data element type that is convertible to the value type of InputIteratorT
    @@ -303,23 +302,23 @@

    Member Function Documentation

    [in]num_segmentsThe number of segments that comprise the sorting data [in]d_begin_offsetsDevice-accessible pointer to the sequence of beginning offsets of length num_segments, such that d_begin_offsets[i] is the first element of the ith data segment in d_keys_* and d_values_* [in]d_end_offsetsDevice-accessible pointer to the sequence of ending offsets of length num_segments, such that d_end_offsets[i]-1 is the last element of the ith data segment in d_keys_* and d_values_*. If d_end_offsets[i]-1 <= d_begin_offsets[i], the ith is considered empty. - [in]reduction_opBinary reduction functor (e.g., an instance of cub::Sum, cub::Min, cub::Max, etc.) - [in]identityThe identity value to be returned for zero-length segments + [in]reduction_opBinary reduction functor + [in]initInitial value of the reduction for each segment [in]stream[optional] CUDA stream to launch kernels within. Default is stream0. [in]debug_synchronous[optional] Whether or not to synchronize the stream after every kernel launch to check for errors. Also causes launch configurations to be printed to the console. Default is false. -

    Definition at line 132 of file device_segmented_reduce.cuh.

    +

    Definition at line 130 of file device_segmented_reduce.cuh.

    - +
    -template<typename InputIteratorT , typename OutputIteratorT , typename T >
    +template<typename InputIteratorT , typename OutputIteratorT >
    - - - - - - @@ -399,13 +392,13 @@

    Member Function Documentation

    Computes a device-wide segmented sum using the addition ('+') operator.

      -
    • The specified identity value is produced for zero-length segments. identity is assumed to be arithmetically neutral for the input type's addition operation, i.e., XXX + identity = XXX
      • +
    • Uses 0 as the initial value of the reduction for each segment.
    • When input a contiguous sequence of segments, a single sequence segment_offsets (of length num_segments+1) can be aliased for both the d_begin_offsets and d_end_offsets parameters (where the latter is specified as segment_offsets+1).
      • -
    • Does not support non-commutative reduction operators.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • +
    • Does not support + operators that are non-commutative..
      • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    -
    Snippet
    The code snippet below illustrates the sum reduction of a device vector of int items.
    +
    Snippet
    The code snippet below illustrates the sum reduction of a device vector of int data elements.
    #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
    // Declare, allocate, and initialize device-accessible pointers for input and output
    @@ -413,21 +406,20 @@

    Member Function Documentation

    int *d_offsets; // e.g., [0, 3, 3, 7]
    int *d_in; // e.g., [8, 6, 7, 5, 3, 0, 9]
    int *d_out; // e.g., [-, -, -]
    -
    int identity // e.g., 0
    ...
    // Determine temporary device storage requirements
    void *d_temp_storage = NULL;
    size_t temp_storage_bytes = 0;
    -
    cub::DeviceSegmentedReduce::Sum(d_temp_storage, temp_storage_bytes, d_in, d_out,
    -
    num_segments, d_offsets, d_offsets + 1, identity);
    +
    cub::DeviceSegmentedReduce::Sum(d_temp_storage, temp_storage_bytes, d_in, d_out,
    +
    num_segments, d_offsets, d_offsets + 1);
    // Allocate temporary storage
    cudaMalloc(&d_temp_storage, temp_storage_bytes);
    // Run sum-reduction
    -
    cub::DeviceSegmentedReduce::Sum(d_temp_storage, temp_storage_bytes, d_in, d_out,
    -
    num_segments, d_offsets, d_offsets + 1, identity);
    +
    cub::DeviceSegmentedReduce::Sum(d_temp_storage, temp_storage_bytes, d_in, d_out,
    +
    num_segments, d_offsets, d_offsets + 1);
    // d_out <-- [21, 0, 17]
    @@ -435,7 +427,6 @@

    Member Function Documentation

    @@ -366,12 +365,6 @@

    Member Function Documentation

    		int *  d_end_offsets,
    T identity,
    -
    InputIteratorT[inferred] Random-access input iterator type for reading input items (may be a simple pointer type)
    OutputIteratorT[inferred] Output iterator type for recording the reduced aggregate (may be a simple pointer type)
    T[inferred] Data type being reduced
    @@ -448,22 +439,21 @@

    Member Function Documentation

    [in]num_segmentsThe number of segments that comprise the sorting data [in]d_begin_offsetsDevice-accessible pointer to the sequence of beginning offsets of length num_segments, such that d_begin_offsets[i] is the first element of the ith data segment in d_keys_* and d_values_* [in]d_end_offsetsDevice-accessible pointer to the sequence of ending offsets of length num_segments, such that d_end_offsets[i]-1 is the last element of the ith data segment in d_keys_* and d_values_*. If d_end_offsets[i]-1 <= d_begin_offsets[i], the ith is considered empty. - [in]identityThe identity value to be returned for zero-length segments [in]stream[optional] CUDA stream to launch kernels within. Default is stream0. [in]debug_synchronous[optional] Whether or not to synchronize the stream after every kernel launch to check for errors. Also causes launch configurations to be printed to the console. Default is false. -

    Definition at line 217 of file device_segmented_reduce.cuh.

    +

    Definition at line 210 of file device_segmented_reduce.cuh.

    - +
    -template<typename InputIteratorT , typename OutputIteratorT , typename T >
    +template<typename InputIteratorT , typename OutputIteratorT >
    - - - - - - @@ -543,13 +527,13 @@

    Member Function Documentation

    Computes a device-wide segmented minimum using the less-than ('<') operator.

      -
    • The specified identity value is produced for zero-length segments. identity is assumed to be arithmetically neutral for the input type's < operation, i.e., (identity < XXX) = false
      • +
    • Uses std::numeric_limits<T>::max() as the initial value of the reduction for each segment.
    • When input a contiguous sequence of segments, a single sequence segment_offsets (of length num_segments+1) can be aliased for both the d_begin_offsets and d_end_offsets parameters (where the latter is specified as segment_offsets+1).
      • -
    • Does not support non-commutative minimum operators.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • +
    • Does not support < operators that are non-commutative.
      • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    -
    Snippet
    The code snippet below illustrates the min-reduction of a device vector of int items.
    +
    Snippet
    The code snippet below illustrates the min-reduction of a device vector of int data elements.
    #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
    // Declare, allocate, and initialize device-accessible pointers for input and output
    @@ -557,29 +541,27 @@

    Member Function Documentation

    int *d_offsets; // e.g., [0, 3, 3, 7]
    int *d_in; // e.g., [8, 6, 7, 5, 3, 0, 9]
    int *d_out; // e.g., [-, -, -]
    -
    int identity; // e.g., MAX_INT
    ...
    // Determine temporary device storage requirements
    void *d_temp_storage = NULL;
    size_t temp_storage_bytes = 0;
    -
    cub::DeviceSegmentedReduce::Min(d_temp_storage, temp_storage_bytes, d_in, d_out,
    -
    num_segments, d_offsets, d_offsets + 1, identity);
    +
    cub::DeviceSegmentedReduce::Min(d_temp_storage, temp_storage_bytes, d_in, d_out,
    +
    num_segments, d_offsets, d_offsets + 1);
    // Allocate temporary storage
    cudaMalloc(&d_temp_storage, temp_storage_bytes);
    // Run min-reduction
    -
    cub::DeviceSegmentedReduce::Min(d_temp_storage, temp_storage_bytes, d_in, d_out,
    -
    num_segments, d_offsets, d_offsets + 1, identity);
    +
    cub::DeviceSegmentedReduce::Min(d_temp_storage, temp_storage_bytes, d_in, d_out,
    +
    num_segments, d_offsets, d_offsets + 1);
    -
    // d_out <-- [6, MAX_INT, 0]
    +
    // d_out <-- [6, INT_MAX, 0]
    Template Parameters
    @@ -510,12 +500,6 @@

    Member Function Documentation

    		int *  d_end_offsets,
    T identity,
    -
    InputIteratorT[inferred] Random-access input iterator type for reading input items (may be a simple pointer type)
    OutputIteratorT[inferred] Output iterator type for recording the reduced aggregate (may be a simple pointer type)
    T[inferred] Data type being reduced
    @@ -592,22 +574,21 @@

    Member Function Documentation

    [in]num_segmentsThe number of segments that comprise the sorting data [in]d_begin_offsetsDevice-accessible pointer to the sequence of beginning offsets of length num_segments, such that d_begin_offsets[i] is the first element of the ith data segment in d_keys_* and d_values_* [in]d_end_offsetsDevice-accessible pointer to the sequence of ending offsets of length num_segments, such that d_end_offsets[i]-1 is the last element of the ith data segment in d_keys_* and d_values_*. If d_end_offsets[i]-1 <= d_begin_offsets[i], the ith is considered empty. - [in]identityThe identity value to be returned for zero-length segments [in]stream[optional] CUDA stream to launch kernels within. Default is stream0. [in]debug_synchronous[optional] Whether or not to synchronize the stream after every kernel launch to check for errors. Also causes launch configurations to be printed to the console. Default is false. -

    Definition at line 301 of file device_segmented_reduce.cuh.

    +

    Definition at line 288 of file device_segmented_reduce.cuh.

    - +
    -template<typename InputIteratorT , typename OutputIteratorT , typename T >
    +template<typename InputIteratorT , typename OutputIteratorT >
    - - - - - - @@ -687,14 +662,17 @@

    Member Function Documentation

    Finds the first device-wide minimum in each segment using the less-than ('<') operator, also returning the in-segment index of that item.

      -
    • Assuming the input d_in has value type T, the output d_out must have value type KeyValuePair<int, T>. The minimum value is written to d_out.value and its location in the input array is written to d_out.key.
      • -
    • The specified identity value is produced for zero-length segments. identity is assumed to be arithmetically neutral for the input type's < operation, i.e., (identity < XXX) = false
      • +
    • The output value type of d_out is cub::KeyValuePair <int, T> (assuming the value type of d_in is T)
        +
      • The minimum of the ith segment is written to d_out[i].value and its offset in that segment is written to d_out[i].key.
        • +
      • The {1, std::numeric_limits<T>::max()} tuple is produced for zero-length inputs
        • +
      +
    • When input a contiguous sequence of segments, a single sequence segment_offsets (of length num_segments+1) can be aliased for both the d_begin_offsets and d_end_offsets parameters (where the latter is specified as segment_offsets+1).
      • -
    • Does not support non-commutative minimum operators.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • +
    • Does not support < operators that are non-commutative.
      • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    -
    Snippet
    The code snippet below illustrates the argmin-reduction of a device vector of int items.
    +
    Snippet
    The code snippet below illustrates the argmin-reduction of a device vector of int data elements.
    #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
    // Declare, allocate, and initialize device-accessible pointers for input and output
    @@ -702,29 +680,27 @@

    Member Function Documentation

    int *d_offsets; // e.g., [0, 3, 3, 7]
    int *d_in; // e.g., [8, 6, 7, 5, 3, 0, 9]
    KeyValuePair<int, int> *d_out; // e.g., [{-,-}, {-,-}, {-,-}]
    -
    int identity; // e.g., MAX_INT
    ...
    // Determine temporary device storage requirements
    void *d_temp_storage = NULL;
    size_t temp_storage_bytes = 0;
    -
    cub::DeviceSegmentedReduce::ArgMin(d_temp_storage, temp_storage_bytes, d_in, d_out,
    -
    num_segments, d_offsets, d_offsets + 1, identity);
    +
    cub::DeviceSegmentedReduce::ArgMin(d_temp_storage, temp_storage_bytes, d_in, d_out,
    +
    num_segments, d_offsets, d_offsets + 1);
    // Allocate temporary storage
    cudaMalloc(&d_temp_storage, temp_storage_bytes);
    // Run argmin-reduction
    -
    cub::DeviceSegmentedReduce::ArgMin(d_temp_storage, temp_storage_bytes, d_in, d_out,
    -
    num_segments, d_offsets, d_offsets + 1, identity);
    +
    cub::DeviceSegmentedReduce::ArgMin(d_temp_storage, temp_storage_bytes, d_in, d_out,
    +
    num_segments, d_offsets, d_offsets + 1);
    -
    // d_out <-- [{1,6}, {-1,MAX_INT}, {2,0}]
    +
    // d_out <-- [{1,6}, {1,INT_MAX}, {2,0}]
    Template Parameters
    @@ -654,12 +635,6 @@

    Member Function Documentation

    		int *  d_end_offsets,
    T identity,
    -
    InputIteratorT[inferred] Random-access input iterator type for reading input items (of some type T) (may be a simple pointer type)
    OutputIteratorT[inferred] Output iterator type for recording the reduced aggregate (having value type KeyValuePair<int, T>) (may be a simple pointer type)
    T[inferred] Data type being reduced
    @@ -737,22 +713,21 @@

    Member Function Documentation

    [in]num_segmentsThe number of segments that comprise the sorting data [in]d_begin_offsetsDevice-accessible pointer to the sequence of beginning offsets of length num_segments, such that d_begin_offsets[i] is the first element of the ith data segment in d_keys_* and d_values_* [in]d_end_offsetsDevice-accessible pointer to the sequence of ending offsets of length num_segments, such that d_end_offsets[i]-1 is the last element of the ith data segment in d_keys_* and d_values_*. If d_end_offsets[i]-1 <= d_begin_offsets[i], the ith is considered empty. - [in]identityThe identity value to be returned for zero-length segments [in]stream[optional] CUDA stream to launch kernels within. Default is stream0. [in]debug_synchronous[optional] Whether or not to synchronize the stream after every kernel launch to check for errors. Also causes launch configurations to be printed to the console. Default is false. -

    Definition at line 388 of file device_segmented_reduce.cuh.

    +

    Definition at line 368 of file device_segmented_reduce.cuh.

    - +
    -template<typename InputIteratorT , typename OutputIteratorT , typename T >
    +template<typename InputIteratorT , typename OutputIteratorT >
    - - - - - - @@ -832,13 +801,13 @@

    Member Function Documentation

    Computes a device-wide segmented maximum using the greater-than ('>') operator.

      -
    • The specified identity value is produced for zero-length segments. identity is assumed to be arithmetically neutral for the input type's > operation, i.e., (identity > XXX) = false
      • +
    • Uses std::numeric_limits<T>::lowest() as the initial value of the reduction.
    • When input a contiguous sequence of segments, a single sequence segment_offsets (of length num_segments+1) can be aliased for both the d_begin_offsets and d_end_offsets parameters (where the latter is specified as segment_offsets+1).
      • -
    • Does not support non-commutative maximum operators.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • +
    • Does not support > operators that are non-commutative.
      • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    -
    Snippet
    The code snippet below illustrates the max-reduction of a device vector of int items.
    +
    Snippet
    The code snippet below illustrates the max-reduction of a device vector of int data elements.
    #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
    // Declare, allocate, and initialize device-accessible pointers for input and output
    @@ -846,29 +815,27 @@

    Member Function Documentation

    int *d_offsets; // e.g., [0, 3, 3, 7]
    int *d_in; // e.g., [8, 6, 7, 5, 3, 0, 9]
    int *d_out; // e.g., [-, -, -]
    -
    int identity; // e.g., MIN_INT
    ...
    // Determine temporary device storage requirements
    void *d_temp_storage = NULL;
    size_t temp_storage_bytes = 0;
    -
    cub::DeviceSegmentedReduce::Max(d_temp_storage, temp_storage_bytes, d_in, d_out,
    -
    num_segments, d_offsets, d_offsets + 1, identity);
    +
    cub::DeviceSegmentedReduce::Max(d_temp_storage, temp_storage_bytes, d_in, d_out,
    +
    num_segments, d_offsets, d_offsets + 1);
    // Allocate temporary storage
    cudaMalloc(&d_temp_storage, temp_storage_bytes);
    // Run max-reduction
    -
    cub::DeviceSegmentedReduce::Max(d_temp_storage, temp_storage_bytes, d_in, d_out,
    -
    num_segments, d_offsets, d_offsets + 1, identity);
    +
    cub::DeviceSegmentedReduce::Max(d_temp_storage, temp_storage_bytes, d_in, d_out,
    +
    num_segments, d_offsets, d_offsets + 1);
    -
    // d_out <-- [8, MIN_INT, 9]
    +
    // d_out <-- [8, INT_MIN, 9]
    Template Parameters
    @@ -799,12 +774,6 @@

    Member Function Documentation

    		int *  d_end_offsets,
    T identity,
    -
    InputIteratorT[inferred] Random-access input iterator type for reading input items (may be a simple pointer type)
    OutputIteratorT[inferred] Output iterator type for recording the reduced aggregate (may be a simple pointer type)
    T[inferred] Data type being reduced
    @@ -881,22 +848,21 @@

    Member Function Documentation

    [in]num_segmentsThe number of segments that comprise the sorting data [in]d_begin_offsetsDevice-accessible pointer to the sequence of beginning offsets of length num_segments, such that d_begin_offsets[i] is the first element of the ith data segment in d_keys_* and d_values_* [in]d_end_offsetsDevice-accessible pointer to the sequence of ending offsets of length num_segments, such that d_end_offsets[i]-1 is the last element of the ith data segment in d_keys_* and d_values_*. If d_end_offsets[i]-1 <= d_begin_offsets[i], the ith is considered empty. - [in]identityThe identity value to be returned for zero-length segments [in]stream[optional] CUDA stream to launch kernels within. Default is stream0. [in]debug_synchronous[optional] Whether or not to synchronize the stream after every kernel launch to check for errors. Also causes launch configurations to be printed to the console. Default is false. -

    Definition at line 478 of file device_segmented_reduce.cuh.

    +

    Definition at line 450 of file device_segmented_reduce.cuh.

    - +
    -template<typename InputIteratorT , typename OutputIteratorT , typename T >
    +template<typename InputIteratorT , typename OutputIteratorT >
    - - - - - - @@ -976,14 +936,17 @@

    Member Function Documentation

    Finds the first device-wide maximum in each segment using the greater-than ('>') operator, also returning the in-segment index of that item.

      -
    • Assuming the input d_in has value type T, the output d_out must have value type KeyValuePair<int, T>. The maximum value is written to d_out.value and its location in the input array is written to d_out.key.
      • -
    • The specified identity value is produced for zero-length segments. identity is assumed to be arithmetically neutral for the input type's > operation, i.e., (identity > XXX) = false
      • +
    • The output value type of d_out is cub::KeyValuePair <int, T> (assuming the value type of d_in is T)
        +
      • The maximum of the ith segment is written to d_out[i].value and its offset in that segment is written to d_out[i].key.
        • +
      • The {1, std::numeric_limits<T>::lowest()} tuple is produced for zero-length inputs
        • +
      +
    • When input a contiguous sequence of segments, a single sequence segment_offsets (of length num_segments+1) can be aliased for both the d_begin_offsets and d_end_offsets parameters (where the latter is specified as segment_offsets+1).
      • -
    • Does not support non-commutative maximum operators.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • +
    • Does not support > operators that are non-commutative.
      • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    -
    Snippet
    The code snippet below illustrates the argmax-reduction of a device vector of int items.
    +
    Snippet
    The code snippet below illustrates the argmax-reduction of a device vector of int data elements.
    #include <cub/cub.cuh> // or equivalently <cub/device/device_reduce.cuh>
    // Declare, allocate, and initialize device-accessible pointers for input and output
    @@ -991,29 +954,27 @@

    Member Function Documentation

    int *d_offsets; // e.g., [0, 3, 3, 7]
    int *d_in; // e.g., [8, 6, 7, 5, 3, 0, 9]
    KeyValuePair<int, int> *d_out; // e.g., [{-,-}, {-,-}, {-,-}]
    -
    int identity; // e.g., MIN_INT
    ...
    // Determine temporary device storage requirements
    void *d_temp_storage = NULL;
    size_t temp_storage_bytes = 0;
    -
    cub::DeviceSegmentedReduce::ArgMax(d_temp_storage, temp_storage_bytes, d_in, d_out,
    -
    num_segments, d_offsets, d_offsets + 1, identity);
    +
    cub::DeviceSegmentedReduce::ArgMax(d_temp_storage, temp_storage_bytes, d_in, d_out,
    +
    num_segments, d_offsets, d_offsets + 1);
    // Allocate temporary storage
    cudaMalloc(&d_temp_storage, temp_storage_bytes);
    // Run argmax-reduction
    -
    cub::DeviceSegmentedReduce::ArgMax(d_temp_storage, temp_storage_bytes, d_in, d_out,
    -
    num_segments, d_offsets, d_offsets + 1, identity);
    +
    cub::DeviceSegmentedReduce::ArgMax(d_temp_storage, temp_storage_bytes, d_in, d_out,
    +
    num_segments, d_offsets, d_offsets + 1);
    -
    // d_out <-- [{0,8}, {-1,MIN_INT}, {3,9}]
    +
    // d_out <-- [{0,8}, {1,INT_MIN}, {3,9}]
    Template Parameters
    @@ -943,12 +909,6 @@

    Member Function Documentation

    		int *  d_end_offsets,
    T identity,
    -
    InputIteratorT[inferred] Random-access input iterator type for reading input items (of some type T) (may be a simple pointer type)
    OutputIteratorT[inferred] Output iterator type for recording the reduced aggregate (having value type KeyValuePair<int, T>) (may be a simple pointer type)
    T[inferred] Data type being reduced
    @@ -1026,14 +987,13 @@

    Member Function Documentation

    [in]num_segmentsThe number of segments that comprise the sorting data [in]d_begin_offsetsDevice-accessible pointer to the sequence of beginning offsets of length num_segments, such that d_begin_offsets[i] is the first element of the ith data segment in d_keys_* and d_values_* [in]d_end_offsetsDevice-accessible pointer to the sequence of ending offsets of length num_segments, such that d_end_offsets[i]-1 is the last element of the ith data segment in d_keys_* and d_values_*. If d_end_offsets[i]-1 <= d_begin_offsets[i], the ith is considered empty. - [in]identityThe identity value to be returned for zero-length segments [in]stream[optional] CUDA stream to launch kernels within. Default is stream0. [in]debug_synchronous[optional] Whether or not to synchronize the stream after every kernel launch to check for errors. Also causes launch configurations to be printed to the console. Default is false. -

    Definition at line 565 of file device_segmented_reduce.cuh.

    +

    Definition at line 530 of file device_segmented_reduce.cuh.

    @@ -1044,7 +1004,7 @@

    Member Function Documentation

    -Generated on Wed Nov 18 2015 10:24:47 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_select-members.html b/docs/html/structcub_1_1_device_select-members.html index 3bd4ef0552..1ddef29561 100644 --- a/docs/html/structcub_1_1_device_select-members.html +++ b/docs/html/structcub_1_1_device_select-members.html @@ -111,7 +111,7 @@
    -Generated on Tue Nov 10 2015 13:58:07 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_select.html b/docs/html/structcub_1_1_device_select.html index fb90b07ec1..aa12eb3825 100644 --- a/docs/html/structcub_1_1_device_select.html +++ b/docs/html/structcub_1_1_device_select.html @@ -112,7 +112,7 @@ .
    Overview
    These operations apply a selection criterion to selectively copy items from a specified input sequence to a compact output sequence.
    Usage Considerations
      -
    • Dynamic parallelism. DeviceSelect methods can be called within kernel code on devices in which CUDA dynamic parallelism is supported. When calling these methods from kernel code, be sure to define the CUB_CDP macro in your compiler's macro definitions.
      • +
    • Dynamic parallelism. DeviceSelect methods can be called within kernel code on devices in which CUDA dynamic parallelism is supported.
    Performance
    The work-complexity of select-flagged, select-if, and select-unique as a function of input size is linear, resulting in performance throughput that plateaus with problem sizes large enough to saturate the GPU.
    @@ -247,8 +247,7 @@

    Member Function Documentation

    • The value type of d_flags must be castable to bool (e.g., bool, char, int, etc.).
    • Copies of the selected items are compacted into d_out and maintain their original relative ordering.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    Snippet
    The code snippet below illustrates the compaction of items selected from an int device vector.
    @@ -301,7 +300,7 @@

    Member Function Documentation

    Examples:
    example_device_select_flagged.cu.
    -

    Definition at line 134 of file device_select.cuh.

    +

    Definition at line 133 of file device_select.cuh.

    @@ -389,8 +388,7 @@

    Member Function Documentation

    • Copies of the selected items are compacted into d_out and maintain their original relative ordering.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    Performance
    The following charts illustrate saturated select-if performance across different CUDA architectures for int32 and int64 items, respectively. Items are selected with 50% probability.
    @@ -471,7 +469,7 @@

    Member Function Documentation

    Examples:
    example_device_select_if.cu.
    -

    Definition at line 241 of file device_select.cuh.

    +

    Definition at line 239 of file device_select.cuh.

    @@ -554,8 +552,7 @@

    Member Function Documentation

    • The == equality operator is used to determine whether keys are equivalent
    • Copies of the selected items are compacted into d_out and maintain their original relative ordering.
      • -
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
      • -
    • +
    • When d_temp_storage is NULL, no work is done and the required allocation size is returned in temp_storage_bytes.
    Performance
    The following charts illustrate saturated select-unique performance across different CUDA architectures for int32 and int64 items, respectively. Segments have lengths uniformly sampled from [1,1000].
    @@ -619,7 +616,7 @@

    Member Function Documentation

    Examples:
    example_device_select_unique.cu.
    -

    Definition at line 332 of file device_select.cuh.

    +

    Definition at line 329 of file device_select.cuh.

    @@ -630,7 +627,7 @@

    Member Function Documentation

    -Generated on Tue Nov 10 2015 13:58:07 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_spmv-members.html b/docs/html/structcub_1_1_device_spmv-members.html index 34680d3006..1d18350566 100644 --- a/docs/html/structcub_1_1_device_spmv-members.html +++ b/docs/html/structcub_1_1_device_spmv-members.html @@ -109,7 +109,7 @@
    -Generated on Tue Nov 10 2015 13:58:07 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_device_spmv.html b/docs/html/structcub_1_1_device_spmv.html index 28f6892cbc..2866fbd2cf 100644 --- a/docs/html/structcub_1_1_device_spmv.html +++ b/docs/html/structcub_1_1_device_spmv.html @@ -112,7 +112,7 @@
    Usage Considerations
      -
    • Dynamic parallelism. DeviceSpmv methods can be called within kernel code on devices in which CUDA dynamic parallelism is supported. When calling these methods from kernel code, be sure to define the CUB_CDP macro in your compiler's macro definitions.
      • +
    • Dynamic parallelism. DeviceSpmv methods can be called within kernel code on devices in which CUDA dynamic parallelism is supported.
    @@ -298,7 +298,7 @@

    Member Function Documentation

    -Generated on Tue Nov 10 2015 13:58:07 for CUB by   +Generated on Mon Dec 14 2015 13:11:18 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_equality-members.html b/docs/html/structcub_1_1_equality-members.html index 539519d703..3aaceca78c 100644 --- a/docs/html/structcub_1_1_equality-members.html +++ b/docs/html/structcub_1_1_equality-members.html @@ -109,7 +109,7 @@
    -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_equality.html b/docs/html/structcub_1_1_equality.html index 4850362ca7..27c3217fe6 100644 --- a/docs/html/structcub_1_1_equality.html +++ b/docs/html/structcub_1_1_equality.html @@ -125,7 +125,7 @@
    -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_equals-members.html b/docs/html/structcub_1_1_equals-members.html index 9e48e992a0..54b06ccee3 100644 --- a/docs/html/structcub_1_1_equals-members.html +++ b/docs/html/structcub_1_1_equals-members.html @@ -110,7 +110,7 @@
    -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_equals.html b/docs/html/structcub_1_1_equals.html index 3e14827c84..13073f5f8c 100644 --- a/docs/html/structcub_1_1_equals.html +++ b/docs/html/structcub_1_1_equals.html @@ -112,7 +112,7 @@

    Type equality test.

    -

    Definition at line 91 of file util_type.cuh.

    +

    Definition at line 92 of file util_type.cuh.

    @@ -128,7 +128,7 @@
    -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_if-members.html b/docs/html/structcub_1_1_if-members.html index c2b46812df..e59a3b22e8 100644 --- a/docs/html/structcub_1_1_if-members.html +++ b/docs/html/structcub_1_1_if-members.html @@ -109,7 +109,7 @@
    -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_if.html b/docs/html/structcub_1_1_if.html index 3e0a0ba60e..29a09f5540 100644 --- a/docs/html/structcub_1_1_if.html +++ b/docs/html/structcub_1_1_if.html @@ -113,7 +113,7 @@

    Type selection (IF ? ThenType : ElseType)

    -

    Definition at line 65 of file util_type.cuh.

    +

    Definition at line 66 of file util_type.cuh.

    Public Types

    @@ -129,7 +129,7 @@
    -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_inequality-members.html b/docs/html/structcub_1_1_inequality-members.html index df9c7f55e0..393d7c474e 100644 --- a/docs/html/structcub_1_1_inequality-members.html +++ b/docs/html/structcub_1_1_inequality-members.html @@ -109,7 +109,7 @@
    -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_inequality.html b/docs/html/structcub_1_1_inequality.html index 599096c20e..d0f71940e7 100644 --- a/docs/html/structcub_1_1_inequality.html +++ b/docs/html/structcub_1_1_inequality.html @@ -125,7 +125,7 @@
    -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_inequality_wrapper-members.html b/docs/html/structcub_1_1_inequality_wrapper-members.html index 0d5f1b7451..757b97aa6a 100644 --- a/docs/html/structcub_1_1_inequality_wrapper-members.html +++ b/docs/html/structcub_1_1_inequality_wrapper-members.html @@ -111,7 +111,7 @@
    -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_inequality_wrapper.html b/docs/html/structcub_1_1_inequality_wrapper.html index 02e471a28a..3fca37405a 100644 --- a/docs/html/structcub_1_1_inequality_wrapper.html +++ b/docs/html/structcub_1_1_inequality_wrapper.html @@ -141,7 +141,7 @@
    -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_is_pointer-members.html b/docs/html/structcub_1_1_is_pointer-members.html index 9fb5c05dea..4684e50c80 100644 --- a/docs/html/structcub_1_1_is_pointer-members.html +++ b/docs/html/structcub_1_1_is_pointer-members.html @@ -109,7 +109,7 @@
    -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
    diff --git a/docs/html/structcub_1_1_is_pointer.html b/docs/html/structcub_1_1_is_pointer.html index 6b6c93921a..f79191831d 100644 --- a/docs/html/structcub_1_1_is_pointer.html +++ b/docs/html/structcub_1_1_is_pointer.html @@ -111,13 +111,13 @@

    Pointer vs. iterator.

    -

    Definition at line 668 of file util_type.cuh.

    +

    Definition at line 164 of file util_type.cuh.

    Public Types

    - - +

    Public Types

    enum  { VALUE = 0 +
    enum  { VALUE = 0 }
     
     
    The documentation for this struct was generated from the following file:
    • util_type.cuh
      • @@ -126,7 +126,7 @@
      -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
      diff --git a/docs/html/structcub_1_1_is_volatile-members.html b/docs/html/structcub_1_1_is_volatile-members.html index c1a8ba9cc4..07573ac7d6 100644 --- a/docs/html/structcub_1_1_is_volatile-members.html +++ b/docs/html/structcub_1_1_is_volatile-members.html @@ -109,7 +109,7 @@
      -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
      diff --git a/docs/html/structcub_1_1_is_volatile.html b/docs/html/structcub_1_1_is_volatile.html index f3ab3b9306..66e7bd0708 100644 --- a/docs/html/structcub_1_1_is_volatile.html +++ b/docs/html/structcub_1_1_is_volatile.html @@ -111,13 +111,13 @@

      Volatile modifier test.

      -

      Definition at line 693 of file util_type.cuh.

      +

      Definition at line 189 of file util_type.cuh.

      - - +

      Public Types

      enum  { VALUE = 0 +
      enum  { VALUE = 0 }
       
       
      The documentation for this struct was generated from the following file:
      • util_type.cuh
        • @@ -126,7 +126,7 @@
        -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
        diff --git a/docs/html/structcub_1_1_log2-members.html b/docs/html/structcub_1_1_log2-members.html index 5992ffac7d..6d928025b5 100644 --- a/docs/html/structcub_1_1_log2-members.html +++ b/docs/html/structcub_1_1_log2-members.html @@ -109,7 +109,7 @@
        -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
        diff --git a/docs/html/structcub_1_1_log2.html b/docs/html/structcub_1_1_log2.html index 905952bc43..b0f0dd1abe 100644 --- a/docs/html/structcub_1_1_log2.html +++ b/docs/html/structcub_1_1_log2.html @@ -114,14 +114,14 @@

        Statically determine log2(N), rounded up.

        For example: Log2<8>::VALUE // 3 Log2<3>::VALUE // 2

        -

        Definition at line 630 of file util_type.cuh.

        +

        Definition at line 126 of file util_type.cuh.

        - - - + +

        Public Types

        enum  { VALUE = Log2<N, (CURRENT_VAL >> 1), COUNT + 1>::VALUE +
        enum  { VALUE = Log2<N, (CURRENT_VAL >> 1), COUNT + 1>::VALUE }
         Static logarithm value.
         
         Static logarithm value.
         
        The documentation for this struct was generated from the following file:
        • util_type.cuh
          • @@ -130,7 +130,7 @@
          -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
          diff --git a/docs/html/structcub_1_1_max-members.html b/docs/html/structcub_1_1_max-members.html index 1c4c4a611e..ced27421ee 100644 --- a/docs/html/structcub_1_1_max-members.html +++ b/docs/html/structcub_1_1_max-members.html @@ -109,7 +109,7 @@
          -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
          diff --git a/docs/html/structcub_1_1_max.html b/docs/html/structcub_1_1_max.html index 258670f6c4..e4a0ea37cb 100644 --- a/docs/html/structcub_1_1_max.html +++ b/docs/html/structcub_1_1_max.html @@ -125,7 +125,7 @@
          -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
          diff --git a/docs/html/structcub_1_1_min-members.html b/docs/html/structcub_1_1_min-members.html index b7718bcc04..9ca53347e1 100644 --- a/docs/html/structcub_1_1_min-members.html +++ b/docs/html/structcub_1_1_min-members.html @@ -109,7 +109,7 @@
          -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
          diff --git a/docs/html/structcub_1_1_min.html b/docs/html/structcub_1_1_min.html index 6f0ef3d04a..40032c1acd 100644 --- a/docs/html/structcub_1_1_min.html +++ b/docs/html/structcub_1_1_min.html @@ -125,7 +125,7 @@
          -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
          diff --git a/docs/html/structcub_1_1_power_of_two-members.html b/docs/html/structcub_1_1_power_of_two-members.html index 47eff51d7a..cb734fc323 100644 --- a/docs/html/structcub_1_1_power_of_two-members.html +++ b/docs/html/structcub_1_1_power_of_two-members.html @@ -109,7 +109,7 @@
          -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
          diff --git a/docs/html/structcub_1_1_power_of_two.html b/docs/html/structcub_1_1_power_of_two.html index 8411145afd..300208f4d4 100644 --- a/docs/html/structcub_1_1_power_of_two.html +++ b/docs/html/structcub_1_1_power_of_two.html @@ -111,13 +111,13 @@

          Statically determine if N is a power-of-two.

          -

          Definition at line 653 of file util_type.cuh.

          +

          Definition at line 149 of file util_type.cuh.

          - - +

          Public Types

          enum  { VALUE = ((N & (N - 1)) == 0) +
          enum  { VALUE = ((N & (N - 1)) == 0) }
           
           
          The documentation for this struct was generated from the following file:
          • util_type.cuh
            • @@ -126,7 +126,7 @@
            -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/structcub_1_1_reduce_by_key_op-members.html b/docs/html/structcub_1_1_reduce_by_key_op-members.html index dc11ac52af..09b914d123 100644 --- a/docs/html/structcub_1_1_reduce_by_key_op-members.html +++ b/docs/html/structcub_1_1_reduce_by_key_op-members.html @@ -112,7 +112,7 @@
            -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/structcub_1_1_reduce_by_key_op.html b/docs/html/structcub_1_1_reduce_by_key_op.html index 00642ca142..e26c572052 100644 --- a/docs/html/structcub_1_1_reduce_by_key_op.html +++ b/docs/html/structcub_1_1_reduce_by_key_op.html @@ -194,7 +194,7 @@

            Member Function Documentation

            -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/structcub_1_1_reduce_by_segment_op-members.html b/docs/html/structcub_1_1_reduce_by_segment_op-members.html index 96b5f76b3e..e950e40b68 100644 --- a/docs/html/structcub_1_1_reduce_by_segment_op-members.html +++ b/docs/html/structcub_1_1_reduce_by_segment_op-members.html @@ -112,7 +112,7 @@
            -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/structcub_1_1_reduce_by_segment_op.html b/docs/html/structcub_1_1_reduce_by_segment_op.html index 5e2b2bfc43..4dee525b6d 100644 --- a/docs/html/structcub_1_1_reduce_by_segment_op.html +++ b/docs/html/structcub_1_1_reduce_by_segment_op.html @@ -111,8 +111,8 @@ struct cub::ReduceBySegmentOp< ReductionOpT >

            Reduce-by-segment functor.

            -

            Given two cub::KeyValuePair inputs a and b and a binary associative combining operator f(const T &x, const T &y), an instance of this functor returns a cub::KeyValuePair whose key field is a.key + a.key, and whose value field is either b.value if b.key is non-zero, or f(a.value, b.value) otherwise.

            -

            ReduceBySegmentOp is an associative, non-commutative binary combining operator for input sequences of cub::KeyValuePair pairings. Such sequences are typically used to represent a segmented set of values to be reduced and a corresponding set of {0,1}-valued integer "head flags" demarcating the first value of each segment.< Binary reduction operator to apply to values

            +

            Given two cub::KeyValuePair inputs a and b and a binary associative combining operator f(const T &x, const T &y), an instance of this functor returns a cub::KeyValuePair whose key field is a.key + a.key, and whose value field is either b.value if b.key is non-zero, or f(a.value, b.value) otherwise.

            +

            ReduceBySegmentOp is an associative, non-commutative binary combining operator for input sequences of cub::KeyValuePair pairings. Such sequences are typically used to represent a segmented set of values to be reduced and a corresponding set of {0,1}-valued integer "head flags" demarcating the first value of each segment.< Binary reduction operator to apply to values

            Definition at line 249 of file thread_operators.cuh.

            @@ -177,7 +177,7 @@

            Member Function Documentation

            Scan operator.

            -

            < KeyValuePair pairing of T (value) and OffsetT (head flag)

            +

            < KeyValuePair pairing of T (value) and OffsetT (head flag)

            Parameters
            @@ -197,7 +197,7 @@

            Member Function Documentation

            -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/structcub_1_1_remove_qualifiers-members.html b/docs/html/structcub_1_1_remove_qualifiers-members.html index a8e364bb4e..dcb95d6355 100644 --- a/docs/html/structcub_1_1_remove_qualifiers-members.html +++ b/docs/html/structcub_1_1_remove_qualifiers-members.html @@ -109,7 +109,7 @@
            -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/structcub_1_1_remove_qualifiers.html b/docs/html/structcub_1_1_remove_qualifiers.html index bc97b30f50..2e4b3ae532 100644 --- a/docs/html/structcub_1_1_remove_qualifiers.html +++ b/docs/html/structcub_1_1_remove_qualifiers.html @@ -113,7 +113,7 @@

            Removes const and volatile qualifiers from type Tp.

            For example: typename RemoveQualifiers<volatile int>::Type // int;

            -

            Definition at line 720 of file util_type.cuh.

            +

            Definition at line 216 of file util_type.cuh.

            firstFirst partial reduction
            @@ -129,7 +129,7 @@
            -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/structcub_1_1_sum-members.html b/docs/html/structcub_1_1_sum-members.html index 798c56f67b..b59d184379 100644 --- a/docs/html/structcub_1_1_sum-members.html +++ b/docs/html/structcub_1_1_sum-members.html @@ -109,7 +109,7 @@
            -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/structcub_1_1_sum.html b/docs/html/structcub_1_1_sum.html index 8bd3c1f3b9..f0ca92271d 100644 --- a/docs/html/structcub_1_1_sum.html +++ b/docs/html/structcub_1_1_sum.html @@ -125,7 +125,7 @@
            -Generated on Tue Nov 10 2015 13:58:05 for CUB by   +Generated on Mon Dec 14 2015 13:11:16 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/structcub_1_1_warp_reduce_1_1_temp_storage.html b/docs/html/structcub_1_1_warp_reduce_1_1_temp_storage.html index 60c5dd1d34..1dda92ed99 100644 --- a/docs/html/structcub_1_1_warp_reduce_1_1_temp_storage.html +++ b/docs/html/structcub_1_1_warp_reduce_1_1_temp_storage.html @@ -97,8 +97,6 @@
            -
            -List of all members
            cub::WarpReduce< T, LOGICAL_WARP_THREADS, PTX_ARCH >::TempStorage Struct Reference
            @@ -119,32 +117,8 @@
            -cub::Uninitialized< _TempStorage >
            -

            Public Types

            - - - - - - - - - - - - - - - -

            -Additional Inherited Members

            - Public Types inherited from cub::Uninitialized< _TempStorage >
            enum  
             
            -typedef UnitWord< _TempStorage >
            -::DeviceWord             		DeviceWord
             Biggest memory-access word that T is a whole multiple of and is not larger than the alignment of T.
             
            - Public Methods inherited from cub::Uninitialized< _TempStorage >
            -__host__ __device__
            -__forceinline__ _TempStorage &             		Alias ()
             Alias.
             
            - Public Members inherited from cub::Uninitialized< _TempStorage >
            -DeviceWord storage [WORDS]
             Backing storage.
             
            The documentation for this struct was generated from the following file: @@ -152,7 +126,7 @@
            -Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/structcub_1_1_warp_reduce_1_1_temp_storage.png b/docs/html/structcub_1_1_warp_reduce_1_1_temp_storage.png index 4ba7bae455d8c38077bce3d8d39323b865d2cad2..13024937c7f9a4baf099313766da4d7df6168220 100644 GIT binary patch delta 1114 zcmV-g1f~0)3V;fbTz`*AL_t(|0qvd9nxh~Lh5y{ldH*-wwki-nYo|NAHj{py8AS*& zKS^#eGfSD71rjDnDM<>Wq$)|#lvE`tnv$v{MN?9hq-eS*)rV8nGk3RK?pczYrXOX? zPm-ouUj{hlpFvepz1;AbIVZ;3J(H>={e)_SxAZA@_k?=f)qgi7%a%l3CrDDSWqez@ zBCMN~mz$;4cw_Y_*(K=`x;oWTahto6QEfT)LmNFw?kY84nY55vO_x<=Mt#$|P>#%X z`CdL(rxdrest<*OMsBk((l*;~D@m8uRjEqSztZUld`zf*1=V{J9^}}m z2(+bFk}j>x)w8fZqVy(JNeZf@DoN3lR3#~zlBy&{Q!i4Lq-eTPH8V?@nZ*&nFVPo3 z3Lub@DnKxOeI-_aSW2n@!E}R~SOG#QsRG1PGLu{aFq3NnA(LwY6@TAE)yynqW)?^o zKnfs?k}5znB~^fEN~!?SlvDwtX@{yCygzIL@Ur_4Xs#0n<*}+=S8nU@diQv0MkiZ% z57)oja<|X6Z<*<)&DOVjyDuJ0^&c}>n7O_a4xqi90f1xu1*-EP0~~9vA`}8E9TN;D0+G4h&RGUBKoR_NhJz+UO*}zADvv8Xn8r^WAD0zb{+oRNvQE z50m};$vK=-t^aEBx~cz`>NJ$EQfsK7wJvQ>b=<>k z=kycZlS{Y$u)Iulr*3#x#sqnR>R8RE4W>2Q6}+hjBh^~(hkwkzV_LB}7hJmRgnXbn zz-`bsN$ph{(~P6H>f>{Q>T)+VeV6L^>D@QXsXpFyXZ}>K4pjR{O+6gx9Omj6`Z2Mlj_>5^o(E$E?n_q< zVo!BeF;^O0dVigz>gvgmtG>?c=IZ=#zDRYq**On)8s@1A-0iSi!aGb|u=b0)KcaKD zFzVG=GN6->^-f=Hdgd|JCw+Cy+!KAZ?=kJ-A=E!lwKMxYs-yRlvc6M+hZq~s>7cJJ zJ#1;zkKyi^6e{ccc23g-Hv{$GRw@9T`fsdLYu&CMbbn>B@0RaVZ8}J|L-A}*b;!Um z$)`o#H^wHuVD)lASrX3KU70qq9$NyyS$%sUP_MsZ1>nc^A6z}XWuN!|m+E^)mIOOx zl6n*g>Y8$#OMEmu{ClY)_D zqtBB5vMkGzT9#!ABo;|ZNm3XkRY{7bq$)|#lvE`tnv$v{MbllWmf@j|n7j2d&yr*` z8&g^TBx$Pc<*|Vz-BkZas*>6@8cLGY5t%b$`aQa4;IN?rX1U$ot`9f z)f%uGt)w>Bbyl@#Z~86NBXcLd_o6z$9joDlVw>5W0PkoPbVJp--ly~~Z1?uNyb;$Z zRM)E8PkfBnqJJt$M$XPt^&{21OK@9)@4__*)!~izcb==m88K@XNlv{{=Bgwe$aSnf zW^Yoh4e_nMI=1X{s@Yd1>7}+}_WApam+5NjyD8PD`E&9%v0C^3Bx%r2+kxHZZC;wE z-P$fINjKAts7lfebQ#kvdH!0ek0d9{?$UKuW3r!SwNwSOH=wsR9JkC)C6W z5K2iEAfA#blQsf4lT894lT88)fBhz^rIeB?rIbLT0HgrID5(NOQ&I(prlblGO-U6X znhvO%!N8qf5_|yrVX3ff=QQykPlP`xE=Z~sjX^bnsKyVZG0|JT@U9a zs&$85m6-I^zE8LA*!R_WkDl~Z-!{fQ`fljE=NJx?Lt{F)+#2%ft5nBNci%9l`h4fk z{H0tSsQO4vcR12H&DAmVV`5D=evqq2I~XglFI_E&p6aY(ZZuqae_f?&>dBC+w$2>p z>ilrNN%gSVnH?T9%u5y6t#DYv9VQp7^ zg}&;0Oh0)D_1~xJ+6+l~p#9rR z1%Q$N#%9{;cH4m~e~Z3bexGXBLEH*;XLG7U298NSE$X&0cJYPME)$eB;f&vnX&39U zB>-I2mnQ=4`ZHbt{nqzVvCNfjWPk}5zneNru@lvF9D7sSzj=K5m7TIwW+00000 LNkvXXu0mjf`T
            -
            -List of all members
            cub::WarpScan< T, LOGICAL_WARP_THREADS, PTX_ARCH >::TempStorage Struct Reference
            @@ -119,32 +117,8 @@
            -cub::Uninitialized< _TempStorage >
            - - - - - - - - - - - - - - - - -

            -Additional Inherited Members

            - Public Types inherited from cub::Uninitialized< _TempStorage >
            enum  
             
            -typedef UnitWord< _TempStorage >
            -::DeviceWord             		DeviceWord
             Biggest memory-access word that T is a whole multiple of and is not larger than the alignment of T.
             
            - Public Methods inherited from cub::Uninitialized< _TempStorage >
            -__host__ __device__
            -__forceinline__ _TempStorage &             		Alias ()
             Alias.
             
            - Public Members inherited from cub::Uninitialized< _TempStorage >
            -DeviceWord storage [WORDS]
             Backing storage.
             
            The documentation for this struct was generated from the following file: @@ -152,7 +126,7 @@
            -Generated on Tue Nov 10 2015 13:58:08 for CUB by   +Generated on Mon Dec 14 2015 13:11:19 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/structcub_1_1_warp_scan_1_1_temp_storage.png b/docs/html/structcub_1_1_warp_scan_1_1_temp_storage.png index c76c1b9624416c63d711cf3c9ee90baff701319f..c1582c4722eb94d8684a95f37581fbc0bd8c9a11 100644 GIT binary patch delta 1183 zcmZqYdd)c@roPJ4#WAFU@$KBZRgbMWT*b}rzW;x0`O3BSD%O5h%ag61e9jY&bm~xj zAZEOI%FM`R>oEji%TH%TP^WZis^?a{ZbNBpZoMou18MNPa(zDm~(bd7mncmxKn^Y=u zAE&wC7n}sVrOfs=CJ-^*kdiLFt*wFMX z=6kunPP#MUMeXOplWUjw{SH*NcbZ^g!|a~EOXsC;Pwb7tJ&GqAWVT1$YX4?vG|kIX z>dCXpzPcrT?l&wpCn?Mn2Y{N#%NshTNEw5uj2=jP~tJ-+kn zyy7$eG(6X5>c@Nc&37%H{ZV87)?06#qig#8f`3>|Nxhr5G*WBx@m$rH^LEJsy`~v! z_v>%+m)(|jLT^?2^xqWv$Eh-^n)Ul8Emk>t zUELiVx+#o*%4!*wFgR&jGPPVk?%u@GvF-_nK-_!@VHJl0RriMb3zl#Q#Ge#Y*f&Yp z0T}KLKfL-FfCe!FBZBEjco9p-dZ2m|6x8D~;y(+-Ng0U@nhn7CMze*v@9=}i?t%@I zYwH;1y)ew35xIZc!@sw$ug@zfJNa|Vf8EpVwWs!#J~PPOT4#6ZUz*WVe&fi9opl0y z3+maJ8J_>LJFfeJ!QZrgu>|iF``*;&k2sIl9ottk;~sCxdh09a+J%{Ud$=#&I%IfU z-EsC@7K^aQH-6OXzRxdNpZjNH&iu>EL-W%MuTPV?y+18;#;?sX(+^fpO|GAmE|a~d z>!#hRr94+`vnEta*&gRASa16@Tw1>HnkDOvOAgh4&5k!`U8wow6aUQOdPaRcyRCct zv1xK=WTKZWdsEwR)A8pDKEr(V_phftwz&8rp}f;t{_fG6x&OG1zh#Ph6jA#s>*yl; zZ-2Ly?%Zd;uKMjF`(J4WzhCU0^}eatqRmE9DPHF%1CPc1r#ozZ3H-E;SbeKXQ(nFA zb>I7%D9h4G{ly!eKZ&%eo6zqrcc%2}^!nvRJJb4BTUoel-@jqHMc>)`C$9?CcgCGL z-uXE3vQ&*#M$PnBt8PEsU|?;aXLDS&de4;X?FP3V>qtJh;&1h)zxKWLQ;TgYD!a4) zUi{uw_fh1||M`oJf6E=e`P+4)-HO%0A-n}vFZ-j@Z%O6pc23TY-yCyO=j7eLx7VzD zY$@{KVEuuFz3U(B>-eMCw|2=-*;VP;eZkYtcJ9%+x{h;eN7L4Hj#Kw8RPVMvx>x>p z4NHaZd%c=!=Z5)L;|soinr64Xe%9mfNt;zKs!HqoutvPNdF%aq27_0(F6JE&RAAk3 yW$OyR?a!rU7=swh1JKiEZIOww_e_bIuGi{&s)g%MdY0W|00K`}KbLh*2~7Z@SVnaK delta 1211 zcmaiydsNZ~9K}(y*UaTk?UZ8nkc#Q-VJT%=cv6IcM%H{LEyI%|5H%zp404*sS!h4h z@_|dS1EqX#CO<;XGt@IE2TA|{NuHjFraRfH0Ki1r+)JjrzYd> zZD4><_`->kl|U}8@o1P|qkGUWPX{U!SSDkK-qJJZ~V1bY5Pucg|%7 zcR-Y5PBiBU0%39m&%PZ>ltQvS34_>xRjBBXBu6FZ4QhQxtsJ<;XoaN$;tC`^H+_ZCu`h^?X z7=ZNn30+ozPZQpD21E422mf)+?O;fkWEc28J}R)g&9lmWL1nSuYD;d?Y}CZza5i5^ zMcMlpnf`~@hk&LwGeHtK+(|02TGl;4YgS`*JJ`lLknOt&^120{wT0{a4}ku^#3#V& zCML9Y8wv^;-fNQs0)LXH!O{v|yG4P5^@|%8g2VVdZCOp-D&a6Mf#T5hn}DV>s!#ki zc12mHSSl=}3)DMfm3cc)$pimHZuldPZO4eGRcp&{Foe$7w~i(RUYT=K+WPZIJ5{F$ zOclY2S779Goe#Dt{AW|MSb`%V4BACz$r>T^L$EMp57%fWEAp8u3iq`U|EuZsn6hdyN-(G z$+Nvnt^zR8r6Txn3b)L;VR=)69iBfr`8*(9C3KM83bqqL#}<1rRf>rSSo0wblnJS; zi$Lq*{gP&i2~*mvp->&ib$$Guv<4*x^%d6|_g%$YoMII|xye^de?48?>E(zVj$ zfJF?N|IpEpj9!X7)i0$Qk`HCK@vzSoeoLcK;g zVs!X5oHkh5qI$4p2diqOR4pksw}cFI-$XI61c9HMN=bdy>W0-iP3bu7?MB_6jd7wV5O;vTc|xTXg_B{pi>^FJcM zgm3r z*N(~NClkJY93J5Hz!v?MG`}1Mnx@S4lB7<%*Bu`0;5sI~_Jtj@%>j`f{`)&K$Y^No cITSQ~$jx{VC*5rj7(rGcgF??X2VTzp8!XXsoB#j- diff --git a/docs/html/tex__obj__input__iterator_8cuh.html b/docs/html/tex__obj__input__iterator_8cuh.html index 8bc88a3b34..a19cb3c494 100644 --- a/docs/html/tex__obj__input__iterator_8cuh.html +++ b/docs/html/tex__obj__input__iterator_8cuh.html @@ -134,7 +134,7 @@
            -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/tex__obj__input__iterator_8cuh_source.html b/docs/html/tex__obj__input__iterator_8cuh_source.html index 27c08c85da..4530b87bfa 100644 --- a/docs/html/tex__obj__input__iterator_8cuh_source.html +++ b/docs/html/tex__obj__input__iterator_8cuh_source.html @@ -341,7 +341,7 @@
            -Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/tex__ref__input__iterator_8cuh.html b/docs/html/tex__ref__input__iterator_8cuh.html index 0b53c1734a..1f316660ac 100644 --- a/docs/html/tex__ref__input__iterator_8cuh.html +++ b/docs/html/tex__ref__input__iterator_8cuh.html @@ -134,7 +134,7 @@
            -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/tex__ref__input__iterator_8cuh_source.html b/docs/html/tex__ref__input__iterator_8cuh_source.html index f32e8d9e7e..60bfd43322 100644 --- a/docs/html/tex__ref__input__iterator_8cuh_source.html +++ b/docs/html/tex__ref__input__iterator_8cuh_source.html @@ -393,7 +393,7 @@
            -Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/thread__load_8cuh.html b/docs/html/thread__load_8cuh.html index 9c85579d3e..0b69ad7b2a 100644 --- a/docs/html/thread__load_8cuh.html +++ b/docs/html/thread__load_8cuh.html @@ -152,7 +152,7 @@
            -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/thread__load_8cuh_source.html b/docs/html/thread__load_8cuh_source.html index 6600666ba1..1fee5c667d 100644 --- a/docs/html/thread__load_8cuh_source.html +++ b/docs/html/thread__load_8cuh_source.html @@ -174,17 +174,17 @@
            121 struct IterateThreadLoad
            122 {
            123  template <CacheLoadModifier MODIFIER, typename T>
            -
            124  static __device__ __forceinline__ void Load(T *ptr, T *vals)
            +
            124  static __device__ __forceinline__ void Load(T const *ptr, T *vals)
            125  {
            126  vals[COUNT] = ThreadLoad<MODIFIER>(ptr + COUNT);
            127  IterateThreadLoad<COUNT + 1, MAX>::template Load<MODIFIER>(ptr, vals);
            128  }
            129 
            130  template <typename InputIteratorT, typename T>
            -
            131  static __device__ __forceinline__ void Dereference(InputIteratorT ptr, T *vals)
            +
            131  static __device__ __forceinline__ void Dereference(InputIteratorT itr, T *vals)
            132  {
            -
            133  vals[COUNT] = ptr[COUNT];
            -
            134  IterateThreadLoad<COUNT + 1, MAX>::Dereference(ptr, vals);
            +
            133  vals[COUNT] = itr[COUNT];
            +
            134  IterateThreadLoad<COUNT + 1, MAX>::Dereference(itr, vals);
            135  }
            136 };
            137 
            @@ -193,16 +193,16 @@
            141 struct IterateThreadLoad<MAX, MAX>
            142 {
            143  template <CacheLoadModifier MODIFIER, typename T>
            -
            144  static __device__ __forceinline__ void Load(T *ptr, T *vals) {}
            +
            144  static __device__ __forceinline__ void Load(T const *ptr, T *vals) {}
            145 
            146  template <typename InputIteratorT, typename T>
            -
            147  static __device__ __forceinline__ void Dereference(InputIteratorT ptr, T *vals) {}
            +
            147  static __device__ __forceinline__ void Dereference(InputIteratorT itr, T *vals) {}
            148 };
            149 
            150 
            154 #define CUB_LOAD_16(cub_modifier, ptx_modifier) \
            155  template<> \
            -
            156  __device__ __forceinline__ uint4 ThreadLoad<cub_modifier, uint4*>(uint4* ptr) \
            +
            156  __device__ __forceinline__ uint4 ThreadLoad<cub_modifier, uint4 const *>(uint4 const *ptr) \
            157  { \
            158  uint4 retval; \
            159  asm volatile ("ld."#ptx_modifier".v4.u32 {%0, %1, %2, %3}, [%4];" : \
            @@ -214,7 +214,7 @@
            165  return retval; \
            166  } \
            167  template<> \
            -
            168  __device__ __forceinline__ ulonglong2 ThreadLoad<cub_modifier, ulonglong2*>(ulonglong2* ptr) \
            +
            168  __device__ __forceinline__ ulonglong2 ThreadLoad<cub_modifier, ulonglong2 const *>(ulonglong2 const *ptr) \
            169  { \
            170  ulonglong2 retval; \
            171  asm volatile ("ld."#ptx_modifier".v2.u64 {%0, %1}, [%2];" : \
            @@ -226,7 +226,7 @@
            177 
            181 #define CUB_LOAD_8(cub_modifier, ptx_modifier) \
            182  template<> \
            -
            183  __device__ __forceinline__ ushort4 ThreadLoad<cub_modifier, ushort4*>(ushort4* ptr) \
            +
            183  __device__ __forceinline__ ushort4 ThreadLoad<cub_modifier, ushort4 const *>(ushort4 const *ptr) \
            184  { \
            185  ushort4 retval; \
            186  asm volatile ("ld."#ptx_modifier".v4.u16 {%0, %1, %2, %3}, [%4];" : \
            @@ -238,7 +238,7 @@
            192  return retval; \
            193  } \
            194  template<> \
            -
            195  __device__ __forceinline__ uint2 ThreadLoad<cub_modifier, uint2*>(uint2* ptr) \
            +
            195  __device__ __forceinline__ uint2 ThreadLoad<cub_modifier, uint2 const *>(uint2 const *ptr) \
            196  { \
            197  uint2 retval; \
            198  asm volatile ("ld."#ptx_modifier".v2.u32 {%0, %1}, [%2];" : \
            @@ -248,7 +248,7 @@
            202  return retval; \
            203  } \
            204  template<> \
            -
            205  __device__ __forceinline__ unsigned long long ThreadLoad<cub_modifier, unsigned long long*>(unsigned long long* ptr) \
            +
            205  __device__ __forceinline__ unsigned long long ThreadLoad<cub_modifier, unsigned long long const *>(unsigned long long const *ptr) \
            206  { \
            207  unsigned long long retval; \
            208  asm volatile ("ld."#ptx_modifier".u64 %0, [%1];" : \
            @@ -259,7 +259,7 @@
            213 
            217 #define CUB_LOAD_4(cub_modifier, ptx_modifier) \
            218  template<> \
            -
            219  __device__ __forceinline__ unsigned int ThreadLoad<cub_modifier, unsigned int*>(unsigned int* ptr) \
            +
            219  __device__ __forceinline__ unsigned int ThreadLoad<cub_modifier, unsigned int const *>(unsigned int const *ptr) \
            220  { \
            221  unsigned int retval; \
            222  asm volatile ("ld."#ptx_modifier".u32 %0, [%1];" : \
            @@ -271,7 +271,7 @@
            228 
            232 #define CUB_LOAD_2(cub_modifier, ptx_modifier) \
            233  template<> \
            -
            234  __device__ __forceinline__ unsigned short ThreadLoad<cub_modifier, unsigned short*>(unsigned short* ptr) \
            +
            234  __device__ __forceinline__ unsigned short ThreadLoad<cub_modifier, unsigned short const *>(unsigned short const *ptr) \
            235  { \
            236  unsigned short retval; \
            237  asm volatile ("ld."#ptx_modifier".u16 %0, [%1];" : \
            @@ -283,7 +283,7 @@
            243 
            247 #define CUB_LOAD_1(cub_modifier, ptx_modifier) \
            248  template<> \
            -
            249  __device__ __forceinline__ unsigned char ThreadLoad<cub_modifier, unsigned char*>(unsigned char* ptr) \
            +
            249  __device__ __forceinline__ unsigned char ThreadLoad<cub_modifier, unsigned char const *>(unsigned char const *ptr) \
            250  { \
            251  unsigned short retval; \
            252  asm volatile ( \
            @@ -294,7 +294,7 @@
            257  "}" : \
            258  "=h"(retval) : \
            259  _CUB_ASM_PTR_(ptr)); \
            -
            260  return (unsigned char) retval; \
            +
            260  return (unsigned char) retval; \
            261  }
            262 
            263 
            @@ -374,7 +374,7 @@
            355 template <typename T>
            356 __device__ __forceinline__ T ThreadLoadVolatilePointer(
            357  T *ptr,
            -
            358  Int2Type<false> is_primitive)
            +
            358  Int2Type<false> is_primitive)
            359 {
            360 
            361 #if CUB_PTX_ARCH <= 130
            @@ -422,7 +422,7 @@
            406 
            410 template <typename T, int MODIFIER>
            411 __device__ __forceinline__ T ThreadLoad(
            -
            412  T *ptr,
            +
            412  T const *ptr,
            413  Int2Type<MODIFIER> modifier,
            414  Int2Type<true> is_pointer)
            415 {
            @@ -433,7 +433,7 @@
            420  DeviceWord words[DEVICE_MULTIPLE];
            421 
            422  IterateThreadLoad<0, DEVICE_MULTIPLE>::template Load<CacheLoadModifier(MODIFIER)>(
            -
            423  reinterpret_cast<DeviceWord*>(ptr),
            +
            423  reinterpret_cast<DeviceWord*>(const_cast<T*>(ptr)),
            424  words);
            425 
            426  return *reinterpret_cast<T*>(words);
            @@ -465,7 +465,7 @@
            -Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/thread__operators_8cuh.html b/docs/html/thread__operators_8cuh.html index a194e63460..afc65607de 100644 --- a/docs/html/thread__operators_8cuh.html +++ b/docs/html/thread__operators_8cuh.html @@ -163,7 +163,7 @@
            -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/thread__operators_8cuh_source.html b/docs/html/thread__operators_8cuh_source.html index 2caf4c57dc..1b75d13672 100644 --- a/docs/html/thread__operators_8cuh_source.html +++ b/docs/html/thread__operators_8cuh_source.html @@ -203,14 +203,14 @@
            137 struct ArgMax
            138 {
            140  template <typename T, typename OffsetT>
            -
            141  __host__ __device__ __forceinline__ KeyValuePair<OffsetT, T> operator()(
            -
            142  const KeyValuePair<OffsetT, T> &a,
            -
            143  const KeyValuePair<OffsetT, T> &b) const
            +
            141  __host__ __device__ __forceinline__ KeyValuePair<OffsetT, T> operator()(
            +
            142  const KeyValuePair<OffsetT, T> &a,
            +
            143  const KeyValuePair<OffsetT, T> &b) const
            144  {
            145 // Mooch BUG (device reduce argmax gk110 3.2 million random fp32)
            146 // return ((b.value > a.value) || ((a.value == b.value) && (b.key < a.key))) ? b : a;
            147 
            -
            148  if ((b.value > a.value) || ((a.value == b.value) && (b.key < a.key)))
            +
            148  if ((b.value > a.value) || ((a.value == b.value) && (b.key < a.key)))
            149  return b;
            150  return a;
            151  }
            @@ -230,14 +230,14 @@
            172 struct ArgMin
            173 {
            175  template <typename T, typename OffsetT>
            -
            176  __host__ __device__ __forceinline__ KeyValuePair<OffsetT, T> operator()(
            -
            177  const KeyValuePair<OffsetT, T> &a,
            -
            178  const KeyValuePair<OffsetT, T> &b) const
            +
            176  __host__ __device__ __forceinline__ KeyValuePair<OffsetT, T> operator()(
            +
            177  const KeyValuePair<OffsetT, T> &a,
            +
            178  const KeyValuePair<OffsetT, T> &b) const
            179  {
            180 // Mooch BUG (device reduce argmax gk110 3.2 million random fp32)
            181 // return ((b.value < a.value) || ((a.value == b.value) && (b.key < a.key))) ? b : a;
            182 
            -
            183  if ((b.value < a.value) || ((a.value == b.value) && (b.key < a.key)))
            +
            183  if ((b.value < a.value) || ((a.value == b.value) && (b.key < a.key)))
            184  return b;
            185  return a;
            186  }
            @@ -338,7 +338,7 @@
            -Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/thread__store_8cuh.html b/docs/html/thread__store_8cuh.html index 6c5c4e2e57..75f02b71c8 100644 --- a/docs/html/thread__store_8cuh.html +++ b/docs/html/thread__store_8cuh.html @@ -148,7 +148,7 @@
            -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/thread__store_8cuh_source.html b/docs/html/thread__store_8cuh_source.html index 1c0469b7bf..f2c9261768 100644 --- a/docs/html/thread__store_8cuh_source.html +++ b/docs/html/thread__store_8cuh_source.html @@ -355,80 +355,89 @@
            342 
            343 #else
            344 
            -
            345  typedef typename UnitWord<T>::VolatileWord VolatileWord; // Word type for memcopying
            -
            346 
            -
            347  const int VOLATILE_MULTIPLE = sizeof(T) / sizeof(VolatileWord);
            +
            345  // Create a temporary using shuffle-words, then store using volatile-words
            +
            346  typedef typename UnitWord<T>::VolatileWord VolatileWord;
            +
            347  typedef typename UnitWord<T>::ShuffleWord ShuffleWord;
            348 
            -
            349  VolatileWord words[VOLATILE_MULTIPLE];
            -
            350  *reinterpret_cast<T*>(words) = val;
            -
            351 
            -
            352 // VolatileWord *words = reinterpret_cast<VolatileWord*>(&val);
            +
            349  const int VOLATILE_MULTIPLE = sizeof(T) / sizeof(VolatileWord);
            +
            350  const int SHUFFLE_MULTIPLE = sizeof(T) / sizeof(ShuffleWord);
            +
            351 
            +
            352  VolatileWord words[VOLATILE_MULTIPLE];
            353 
            -
            354  IterateThreadStore<0, VOLATILE_MULTIPLE>::template Dereference(
            -
            355  reinterpret_cast<volatile VolatileWord*>(ptr),
            -
            356  words);
            +
            354  #pragma unroll
            +
            355  for (int i = 0; i < SHUFFLE_MULTIPLE; ++i)
            +
            356  reinterpret_cast<ShuffleWord*>(words)[i] = reinterpret_cast<ShuffleWord*>(&val)[i];
            357 
            -
            358 #endif // CUB_PTX_ARCH <= 130
            -
            359 
            -
            360 }
            +
            358  IterateThreadStore<0, VOLATILE_MULTIPLE>::template Dereference(
            +
            359  reinterpret_cast<volatile VolatileWord*>(ptr),
            +
            360  words);
            361 
            -
            362 
            -
            366 template <typename T>
            -
            367 __device__ __forceinline__ void ThreadStore(
            -
            368  T *ptr,
            -
            369  T val,
            -
            370  Int2Type<STORE_VOLATILE> modifier,
            -
            371  Int2Type<true> is_pointer)
            -
            372 {
            -
            373  ThreadStoreVolatilePtr(ptr, val, Int2Type<Traits<T>::PRIMITIVE>());
            -
            374 }
            -
            375 
            -
            376 
            -
            380 template <typename T, int MODIFIER>
            -
            381 __device__ __forceinline__ void ThreadStore(
            -
            382  T *ptr,
            -
            383  T val,
            -
            384  Int2Type<MODIFIER> modifier,
            -
            385  Int2Type<true> is_pointer)
            -
            386 {
            -
            387  typedef typename UnitWord<T>::DeviceWord DeviceWord; // Word type for memcopying
            -
            388 
            -
            389  const int DEVICE_MULTIPLE = sizeof(T) / sizeof(DeviceWord);
            -
            390 
            -
            391  DeviceWord words[DEVICE_MULTIPLE];
            -
            392 
            -
            393  *reinterpret_cast<T*>(words) = val;
            +
            362 #endif // CUB_PTX_ARCH <= 130
            +
            363 
            +
            364 }
            +
            365 
            +
            366 
            +
            370 template <typename T>
            +
            371 __device__ __forceinline__ void ThreadStore(
            +
            372  T *ptr,
            +
            373  T val,
            +
            374  Int2Type<STORE_VOLATILE> modifier,
            +
            375  Int2Type<true> is_pointer)
            +
            376 {
            +
            377  ThreadStoreVolatilePtr(ptr, val, Int2Type<Traits<T>::PRIMITIVE>());
            +
            378 }
            +
            379 
            +
            380 
            +
            384 template <typename T, int MODIFIER>
            +
            385 __device__ __forceinline__ void ThreadStore(
            +
            386  T *ptr,
            +
            387  T val,
            +
            388  Int2Type<MODIFIER> modifier,
            +
            389  Int2Type<true> is_pointer)
            +
            390 {
            +
            391  // Create a temporary using shuffle-words, then store using device-words
            +
            392  typedef typename UnitWord<T>::DeviceWord DeviceWord;
            +
            393  typedef typename UnitWord<T>::ShuffleWord ShuffleWord;
            394 
            -
            395  IterateThreadStore<0, DEVICE_MULTIPLE>::template Store<CacheStoreModifier(MODIFIER)>(
            -
            396  reinterpret_cast<DeviceWord*>(ptr),
            -
            397  words);
            -
            398 }
            +
            395  const int DEVICE_MULTIPLE = sizeof(T) / sizeof(DeviceWord);
            +
            396  const int SHUFFLE_MULTIPLE = sizeof(T) / sizeof(ShuffleWord);
            +
            397 
            +
            398  DeviceWord words[DEVICE_MULTIPLE];
            399 
            -
            400 
            -
            404 template <CacheStoreModifier MODIFIER, typename OutputIteratorT, typename T>
            -
            405 __device__ __forceinline__ void ThreadStore(OutputIteratorT itr, T val)
            -
            406 {
            -
            407  ThreadStore(
            -
            408  itr,
            -
            409  val,
            -
            410  Int2Type<MODIFIER>(),
            -
            411  Int2Type<IsPointer<OutputIteratorT>::VALUE>());
            -
            412 }
            -
            413 
            -
            414 
            -
            415 
            -
            416 #endif // DOXYGEN_SHOULD_SKIP_THIS
            -
            417 
            -
            418  // end group UtilIo
            -
            420 
            -
            421 
            -
            422 } // CUB namespace
            -
            423 CUB_NS_POSTFIX // Optional outer namespace(s)
            +
            400  #pragma unroll
            +
            401  for (int i = 0; i < SHUFFLE_MULTIPLE; ++i)
            +
            402  reinterpret_cast<ShuffleWord*>(words)[i] = reinterpret_cast<ShuffleWord*>(&val)[i];
            +
            403 
            +
            404  IterateThreadStore<0, DEVICE_MULTIPLE>::template Store<CacheStoreModifier(MODIFIER)>(
            +
            405  reinterpret_cast<DeviceWord*>(ptr),
            +
            406  words);
            +
            407 }
            +
            408 
            +
            409 
            +
            413 template <CacheStoreModifier MODIFIER, typename OutputIteratorT, typename T>
            +
            414 __device__ __forceinline__ void ThreadStore(OutputIteratorT itr, T val)
            +
            415 {
            +
            416  ThreadStore(
            +
            417  itr,
            +
            418  val,
            +
            419  Int2Type<MODIFIER>(),
            +
            420  Int2Type<IsPointer<OutputIteratorT>::VALUE>());
            +
            421 }
            +
            422 
            +
            423 
            +
            424 
            +
            425 #endif // DOXYGEN_SHOULD_SKIP_THIS
            +
            426 
            +
            427  // end group UtilIo
            +
            429 
            +
            430 
            +
            431 } // CUB namespace
            +
            432 CUB_NS_POSTFIX // Optional outer namespace(s)
            -Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/transform__input__iterator_8cuh.html b/docs/html/transform__input__iterator_8cuh.html index 3047305ec1..4f0177932a 100644 --- a/docs/html/transform__input__iterator_8cuh.html +++ b/docs/html/transform__input__iterator_8cuh.html @@ -133,7 +133,7 @@
            -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/transform__input__iterator_8cuh_source.html b/docs/html/transform__input__iterator_8cuh_source.html index b14ea97eba..77cffaccbe 100644 --- a/docs/html/transform__input__iterator_8cuh_source.html +++ b/docs/html/transform__input__iterator_8cuh_source.html @@ -280,7 +280,7 @@
            -Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/util__allocator_8cuh_source.html b/docs/html/util__allocator_8cuh_source.html index 5c8015a293..ad2d2c0300 100644 --- a/docs/html/util__allocator_8cuh_source.html +++ b/docs/html/util__allocator_8cuh_source.html @@ -349,7 +349,7 @@
            326 
            327  this->max_cached_bytes = max_cached_bytes;
            328 
            -
            329  if (debug) CubLog("New max_cached_bytes(%lld)\n", (long long) max_cached_bytes);
            +
            329  if (debug) _CubLog("New max_cached_bytes(%lld)\n", (long long) max_cached_bytes);
            330 
            331  // Unlock
            332  Unlock(&spin_lock);
            @@ -429,7 +429,7 @@
            413  cached_blocks.erase(block_itr);
            414  cached_bytes[device] -= search_key.bytes;
            415 
            -
            416  if (debug) CubLog("\tdevice %d reused cached block for stream %lld (%lld bytes, previously associated with stream %lld).\n\t\t %lld available blocks cached (%lld bytes), %lld live blocks outstanding.\n",
            +
            416  if (debug) _CubLog("\tdevice %d reused cached block for stream %lld (%lld bytes, previously associated with stream %lld).\n\t\t %lld available blocks cached (%lld bytes), %lld live blocks outstanding.\n",
            417  device, (long long) active_stream, (long long) search_key.bytes, (long long) prev_stream, (long long) cached_blocks.size(), (long long) cached_bytes[device], (long long) live_blocks.size());
            418 
            419  break;
            @@ -464,7 +464,7 @@
            448  // Insert into live blocks
            449  live_blocks.insert(search_key);
            450 
            -
            451  if (debug) CubLog("\tdevice %d allocating new device block %lld bytes associated with stream %lld.\n\t\t %lld available blocks cached (%lld bytes), %lld live blocks outstanding.\n",
            +
            451  if (debug) _CubLog("\tdevice %d allocating new device block %lld bytes associated with stream %lld.\n\t\t %lld available blocks cached (%lld bytes), %lld live blocks outstanding.\n",
            452  device, (long long) search_key.bytes, (long long) search_key.associated_stream, (long long) cached_blocks.size(), (long long) cached_bytes[device], (long long) live_blocks.size());
            453  }
            454 
            @@ -558,7 +558,7 @@
            556  cached_blocks.insert(search_key);
            557  cached_bytes[device] += search_key.bytes;
            558 
            -
            559  if (debug) CubLog("\tdevice %d returned %lld bytes from associated stream %lld.\n\t\t %lld available blocks cached (%lld bytes), %lld live blocks outstanding.\n",
            +
            559  if (debug) _CubLog("\tdevice %d returned %lld bytes from associated stream %lld.\n\t\t %lld available blocks cached (%lld bytes), %lld live blocks outstanding.\n",
            560  device, (long long) search_key.bytes, (long long) search_key.associated_stream, (long long) cached_blocks.size(), (long long) cached_bytes[device], (long long) live_blocks.size());
            561  }
            562  else
            @@ -573,7 +573,7 @@
            571  if (CubDebug(error = cudaFree(d_ptr))) break;
            572  if (CubDebug(error = cudaEventDestroy(search_key.ready_event))) break;
            573 
            -
            574  if (debug) CubLog("\tdevice %d freed %lld bytes from associated stream %lld.\n\t\t %lld available blocks cached (%lld bytes), %lld live blocks outstanding.\n",
            +
            574  if (debug) _CubLog("\tdevice %d freed %lld bytes from associated stream %lld.\n\t\t %lld available blocks cached (%lld bytes), %lld live blocks outstanding.\n",
            575  device, (long long) search_key.bytes, (long long) search_key.associated_stream, (long long) cached_blocks.size(), (long long) cached_bytes[device], (long long) live_blocks.size());
            576  }
            577  }
            @@ -652,7 +652,7 @@
            660  cached_bytes[current_device] -= begin->bytes;
            661  cached_blocks.erase(begin);
            662 
            -
            663  if (debug) CubLog("\tdevice %d freed %lld bytes.\n\t\t %lld available blocks cached (%lld bytes), %lld live blocks outstanding.\n",
            +
            663  if (debug) _CubLog("\tdevice %d freed %lld bytes.\n\t\t %lld available blocks cached (%lld bytes), %lld live blocks outstanding.\n",
            664  current_device, (long long) begin->bytes, (long long) cached_blocks.size(), (long long) cached_bytes[current_device], (long long) live_blocks.size());
            665  }
            666 
            @@ -692,7 +692,7 @@
            -Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/util__arch_8cuh.html b/docs/html/util__arch_8cuh.html index 8ef6316989..341691a911 100644 --- a/docs/html/util__arch_8cuh.html +++ b/docs/html/util__arch_8cuh.html @@ -121,7 +121,7 @@
            -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/util__arch_8cuh_source.html b/docs/html/util__arch_8cuh_source.html index 9191fc47df..f277d2c991 100644 --- a/docs/html/util__arch_8cuh_source.html +++ b/docs/html/util__arch_8cuh_source.html @@ -211,7 +211,7 @@
            -Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/util__debug_8cuh.html b/docs/html/util__debug_8cuh.html index b0233a5a0c..ee3aac79eb 100644 --- a/docs/html/util__debug_8cuh.html +++ b/docs/html/util__debug_8cuh.html @@ -127,10 +127,10 @@ #define CubDebugExit(e)   if (cub::Debug((e), __FILE__, __LINE__)) { exit(1); }  Debug macro with exit.
              - -#define CubLog(format,...)   printf(format,__VA_ARGS__); - Log macro for printf statements.
            -  + +#define _CubLog(format,...)   printf(format,__VA_ARGS__); + Log macro for printf statements.
            +  @@ -151,7 +151,7 @@
            -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/util__debug_8cuh_source.html b/docs/html/util__debug_8cuh_source.html index 21ff2909ef..8b99dced13 100644 --- a/docs/html/util__debug_8cuh_source.html +++ b/docs/html/util__debug_8cuh_source.html @@ -176,11 +176,11 @@
            101 #endif
            102 
            103 
            -
            107 #if !defined(CubLog)
            +
            107 #if !defined(_CubLog)
            108  #if (CUB_PTX_ARCH == 0)
            -
            109  #define CubLog(format, ...) printf(format,__VA_ARGS__);
            +
            109  #define _CubLog(format, ...) printf(format,__VA_ARGS__);
            110  #elif (CUB_PTX_ARCH >= 200)
            -
            111  #define CubLog(format, ...) printf("[block (%d,%d,%d), thread (%d,%d,%d)]: " format, blockIdx.z, blockIdx.y, blockIdx.x, threadIdx.z, threadIdx.y, threadIdx.x, __VA_ARGS__);
            +
            111  #define _CubLog(format, ...) printf("[block (%d,%d,%d), thread (%d,%d,%d)]: " format, blockIdx.z, blockIdx.y, blockIdx.x, threadIdx.z, threadIdx.y, threadIdx.x, __VA_ARGS__);
            112  #endif
            113 #endif
            114 
            @@ -194,7 +194,7 @@
            -Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/util__device_8cuh.html b/docs/html/util__device_8cuh.html index 1fbba7fc5b..870deb4966 100644 --- a/docs/html/util__device_8cuh.html +++ b/docs/html/util__device_8cuh.html @@ -104,7 +104,8 @@
            util_device.cuh File Reference
            -
            #include "util_arch.cuh"
            +
            #include "util_type.cuh"
            +#include "util_arch.cuh"
            #include "util_debug.cuh"
            #include "util_namespace.cuh"
            #include "util_macro.cuh"
            @@ -137,7 +138,7 @@
            -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/util__device_8cuh_source.html b/docs/html/util__device_8cuh_source.html index c9918da73c..02950e2b23 100644 --- a/docs/html/util__device_8cuh_source.html +++ b/docs/html/util__device_8cuh_source.html @@ -131,244 +131,248 @@
            28 
            34 #pragma once
            35 
            -
            36 #include "util_arch.cuh"
            -
            37 #include "util_debug.cuh"
            -
            38 #include "util_namespace.cuh"
            -
            39 #include "util_macro.cuh"
            -
            40 
            -
            42 CUB_NS_PREFIX
            -
            43 
            -
            45 namespace cub {
            -
            46 
            +
            36 #include "util_type.cuh"
            +
            37 #include "util_arch.cuh"
            +
            38 #include "util_debug.cuh"
            +
            39 #include "util_namespace.cuh"
            +
            40 #include "util_macro.cuh"
            +
            41 
            +
            43 CUB_NS_PREFIX
            +
            44 
            +
            46 namespace cub {
            47 
            -
            53 #ifndef DOXYGEN_SHOULD_SKIP_THIS // Do not document
            -
            54 
            -
            55 
            -
            59 template <int ALLOCATIONS>
            -
            60 CUB_RUNTIME_FUNCTION __forceinline__
            -
            61 cudaError_t AliasTemporaries(
            -
            62  void *d_temp_storage,
            -
            63  size_t &temp_storage_bytes,
            -
            64  void* (&allocations)[ALLOCATIONS],
            -
            65  size_t (&allocation_sizes)[ALLOCATIONS])
            -
            66 {
            -
            67  const int ALIGN_BYTES = 256;
            -
            68  const int ALIGN_MASK = ~(ALIGN_BYTES - 1);
            -
            69 
            -
            70  // Compute exclusive prefix sum over allocation requests
            -
            71  size_t allocation_offsets[ALLOCATIONS];
            -
            72  size_t bytes_needed = 0;
            -
            73  for (int i = 0; i < ALLOCATIONS; ++i)
            -
            74  {
            -
            75  size_t allocation_bytes = (allocation_sizes[i] + ALIGN_BYTES - 1) & ALIGN_MASK;
            -
            76  allocation_offsets[i] = bytes_needed;
            -
            77  bytes_needed += allocation_bytes;
            -
            78  }
            -
            79 
            -
            80  // Check if the caller is simply requesting the size of the storage allocation
            -
            81  if (!d_temp_storage)
            -
            82  {
            -
            83  temp_storage_bytes = bytes_needed;
            -
            84  return cudaSuccess;
            -
            85  }
            -
            86 
            -
            87  // Check if enough storage provided
            -
            88  if (temp_storage_bytes < bytes_needed)
            -
            89  {
            -
            90  return CubDebug(cudaErrorInvalidValue);
            -
            91  }
            -
            92 
            -
            93  // Alias
            -
            94  for (int i = 0; i < ALLOCATIONS; ++i)
            -
            95  {
            -
            96  allocations[i] = static_cast<char*>(d_temp_storage) + allocation_offsets[i];
            -
            97  }
            -
            98 
            -
            99  return cudaSuccess;
            -
            100 }
            -
            101 
            +
            48 
            +
            54 #ifndef DOXYGEN_SHOULD_SKIP_THIS // Do not document
            +
            55 
            +
            56 
            +
            60 template <int ALLOCATIONS>
            +
            61 CUB_RUNTIME_FUNCTION __forceinline__
            +
            62 cudaError_t AliasTemporaries(
            +
            63  void *d_temp_storage,
            +
            64  size_t &temp_storage_bytes,
            +
            65  void* (&allocations)[ALLOCATIONS],
            +
            66  size_t (&allocation_sizes)[ALLOCATIONS])
            +
            67 {
            +
            68  const int ALIGN_BYTES = 256;
            +
            69  const int ALIGN_MASK = ~(ALIGN_BYTES - 1);
            +
            70 
            +
            71  // Compute exclusive prefix sum over allocation requests
            +
            72  size_t allocation_offsets[ALLOCATIONS];
            +
            73  size_t bytes_needed = 0;
            +
            74  for (int i = 0; i < ALLOCATIONS; ++i)
            +
            75  {
            +
            76  size_t allocation_bytes = (allocation_sizes[i] + ALIGN_BYTES - 1) & ALIGN_MASK;
            +
            77  allocation_offsets[i] = bytes_needed;
            +
            78  bytes_needed += allocation_bytes;
            +
            79  }
            +
            80 
            +
            81  // Check if the caller is simply requesting the size of the storage allocation
            +
            82  if (!d_temp_storage)
            +
            83  {
            +
            84  temp_storage_bytes = bytes_needed;
            +
            85  return cudaSuccess;
            +
            86  }
            +
            87 
            +
            88  // Check if enough storage provided
            +
            89  if (temp_storage_bytes < bytes_needed)
            +
            90  {
            +
            91  return CubDebug(cudaErrorInvalidValue);
            +
            92  }
            +
            93 
            +
            94  // Alias
            +
            95  for (int i = 0; i < ALLOCATIONS; ++i)
            +
            96  {
            +
            97  allocations[i] = static_cast<char*>(d_temp_storage) + allocation_offsets[i];
            +
            98  }
            +
            99 
            +
            100  return cudaSuccess;
            +
            101 }
            102 
            -
            106 template <typename T>
            -
            107 __global__ void EmptyKernel(void) { }
            -
            108 
            +
            103 
            +
            107 template <typename T>
            +
            108 __global__ void EmptyKernel(void) { }
            109 
            -
            110 #endif // DOXYGEN_SHOULD_SKIP_THIS
            -
            111 
            -
            115 CUB_RUNTIME_FUNCTION __forceinline__ cudaError_t PtxVersion(int &ptx_version)
            -
            116 {
            -
            117  struct Dummy
            -
            118  {
            -
            120  typedef void (*EmptyKernelPtr)();
            -
            121 
            -
            123  CUB_RUNTIME_FUNCTION __forceinline__
            -
            124  EmptyKernelPtr Empty()
            -
            125  {
            -
            126  return EmptyKernel<void>;
            -
            127  }
            -
            128  };
            -
            129 
            +
            110 
            +
            111 #endif // DOXYGEN_SHOULD_SKIP_THIS
            +
            112 
            +
            116 CUB_RUNTIME_FUNCTION __forceinline__ cudaError_t PtxVersion(int &ptx_version)
            +
            117 {
            +
            118  struct Dummy
            +
            119  {
            +
            121  typedef void (*EmptyKernelPtr)();
            +
            122 
            +
            124  CUB_RUNTIME_FUNCTION __forceinline__
            +
            125  EmptyKernelPtr Empty()
            +
            126  {
            +
            127  return EmptyKernel<void>;
            +
            128  }
            +
            129  };
            130 
            -
            131 #ifndef CUB_RUNTIME_ENABLED
            -
            132 
            -
            133  // CUDA API calls not supported from this device
            -
            134  return cudaErrorInvalidConfiguration;
            -
            135 
            -
            136 #elif (CUB_PTX_ARCH > 0)
            -
            137 
            -
            138  ptx_version = CUB_PTX_ARCH;
            -
            139  return cudaSuccess;
            -
            140 
            -
            141 #else
            -
            142 
            -
            143  cudaError_t error = cudaSuccess;
            -
            144  do
            -
            145  {
            -
            146  cudaFuncAttributes empty_kernel_attrs;
            -
            147  if (CubDebug(error = cudaFuncGetAttributes(&empty_kernel_attrs, EmptyKernel<void>))) break;
            -
            148  ptx_version = empty_kernel_attrs.ptxVersion * 10;
            -
            149  }
            -
            150  while (0);
            -
            151 
            -
            152  return error;
            -
            153 
            -
            154 #endif
            -
            155 }
            -
            156 
            +
            131 
            +
            132 #ifndef CUB_RUNTIME_ENABLED
            +
            133 
            +
            134  // CUDA API calls not supported from this device
            +
            135  return cudaErrorInvalidConfiguration;
            +
            136 
            +
            137 #elif (CUB_PTX_ARCH > 0)
            +
            138 
            +
            139  ptx_version = CUB_PTX_ARCH;
            +
            140  return cudaSuccess;
            +
            141 
            +
            142 #else
            +
            143 
            +
            144  cudaError_t error = cudaSuccess;
            +
            145  do
            +
            146  {
            +
            147  cudaFuncAttributes empty_kernel_attrs;
            +
            148  if (CubDebug(error = cudaFuncGetAttributes(&empty_kernel_attrs, EmptyKernel<void>))) break;
            +
            149  ptx_version = empty_kernel_attrs.ptxVersion * 10;
            +
            150  }
            +
            151  while (0);
            +
            152 
            +
            153  return error;
            +
            154 
            +
            155 #endif
            +
            156 }
            157 
            -
            161 CUB_RUNTIME_FUNCTION __forceinline__ cudaError_t SmVersion(int &sm_version, int device_ordinal)
            -
            162 {
            -
            163 #ifndef CUB_RUNTIME_ENABLED
            -
            164 
            -
            165  // CUDA API calls not supported from this device
            -
            166  return cudaErrorInvalidConfiguration;
            -
            167 
            -
            168 #else
            -
            169 
            -
            170  cudaError_t error = cudaSuccess;
            -
            171  do
            -
            172  {
            -
            173  // Fill in SM version
            -
            174  int major, minor;
            -
            175  if (CubDebug(error = cudaDeviceGetAttribute(&major, cudaDevAttrComputeCapabilityMajor, device_ordinal))) break;
            -
            176  if (CubDebug(error = cudaDeviceGetAttribute(&minor, cudaDevAttrComputeCapabilityMinor, device_ordinal))) break;
            -
            177  sm_version = major * 100 + minor * 10;
            -
            178  }
            -
            179  while (0);
            -
            180 
            -
            181  return error;
            -
            182 
            -
            183 #endif
            -
            184 }
            -
            185 
            +
            158 
            +
            162 CUB_RUNTIME_FUNCTION __forceinline__ cudaError_t SmVersion(int &sm_version, int device_ordinal)
            +
            163 {
            +
            164 #ifndef CUB_RUNTIME_ENABLED
            +
            165 
            +
            166  // CUDA API calls not supported from this device
            +
            167  return cudaErrorInvalidConfiguration;
            +
            168 
            +
            169 #else
            +
            170 
            +
            171  cudaError_t error = cudaSuccess;
            +
            172  do
            +
            173  {
            +
            174  // Fill in SM version
            +
            175  int major, minor;
            +
            176  if (CubDebug(error = cudaDeviceGetAttribute(&major, cudaDevAttrComputeCapabilityMajor, device_ordinal))) break;
            +
            177  if (CubDebug(error = cudaDeviceGetAttribute(&minor, cudaDevAttrComputeCapabilityMinor, device_ordinal))) break;
            +
            178  sm_version = major * 100 + minor * 10;
            +
            179  }
            +
            180  while (0);
            +
            181 
            +
            182  return error;
            +
            183 
            +
            184 #endif
            +
            185 }
            186 
            -
            187 #ifndef DOXYGEN_SHOULD_SKIP_THIS // Do not document
            -
            188 
            -
            192 CUB_RUNTIME_FUNCTION __forceinline__
            -
            193 static cudaError_t SyncStream(cudaStream_t stream)
            -
            194 {
            -
            195 #if (CUB_PTX_ARCH == 0)
            -
            196  return cudaStreamSynchronize(stream);
            -
            197 #else
            -
            198  // Device can't yet sync on a specific stream
            -
            199  return cudaDeviceSynchronize();
            -
            200 #endif
            -
            201 }
            -
            202 
            +
            187 
            +
            188 #ifndef DOXYGEN_SHOULD_SKIP_THIS // Do not document
            +
            189 
            +
            193 CUB_RUNTIME_FUNCTION __forceinline__
            +
            194 static cudaError_t SyncStream(cudaStream_t stream)
            +
            195 {
            +
            196 #if (CUB_PTX_ARCH == 0)
            +
            197  return cudaStreamSynchronize(stream);
            +
            198 #else
            +
            199  // Device can't yet sync on a specific stream
            +
            200  return cudaDeviceSynchronize();
            +
            201 #endif
            +
            202 }
            203 
            -
            235 template <typename KernelPtr>
            -
            236 CUB_RUNTIME_FUNCTION __forceinline__
            -
            237 cudaError_t MaxSmOccupancy(
            -
            238  int &max_sm_occupancy,
            -
            239  KernelPtr kernel_ptr,
            -
            240  int block_threads,
            -
            241  int dynamic_smem_bytes = 0)
            -
            242 {
            -
            243 #ifndef CUB_RUNTIME_ENABLED
            -
            244 
            -
            245  // CUDA API calls not supported from this device
            -
            246  return CubDebug(cudaErrorInvalidConfiguration);
            -
            247 
            -
            248 #else
            -
            249 
            -
            250  return cudaOccupancyMaxActiveBlocksPerMultiprocessor (
            -
            251  &max_sm_occupancy,
            -
            252  kernel_ptr,
            -
            253  block_threads,
            -
            254  dynamic_smem_bytes);
            -
            255 
            -
            256 #endif // CUB_RUNTIME_ENABLED
            -
            257 }
            -
            258 
            +
            204 
            +
            236 template <typename KernelPtr>
            +
            237 CUB_RUNTIME_FUNCTION __forceinline__
            +
            238 cudaError_t MaxSmOccupancy(
            +
            239  int &max_sm_occupancy,
            +
            240  KernelPtr kernel_ptr,
            +
            241  int block_threads,
            +
            242  int dynamic_smem_bytes = 0)
            +
            243 {
            +
            244 #ifndef CUB_RUNTIME_ENABLED
            +
            245 
            +
            246  // CUDA API calls not supported from this device
            +
            247  return CubDebug(cudaErrorInvalidConfiguration);
            +
            248 
            +
            249 #else
            +
            250 
            +
            251  return cudaOccupancyMaxActiveBlocksPerMultiprocessor (
            +
            252  &max_sm_occupancy,
            +
            253  kernel_ptr,
            +
            254  block_threads,
            +
            255  dynamic_smem_bytes);
            +
            256 
            +
            257 #endif // CUB_RUNTIME_ENABLED
            +
            258 }
            259 
            -
            260 /******************************************************************************
            -
            261  * Policy management
            -
            262  ******************************************************************************/
            -
            263 
            -
            267 struct KernelConfig
            -
            268 {
            -
            269  int block_threads;
            -
            270  int items_per_thread;
            -
            271  int tile_size;
            -
            272  int sm_occupancy;
            -
            273 
            -
            274  template <typename AgentPolicyT, typename KernelPtrT>
            +
            260 
            +
            261 /******************************************************************************
            +
            262  * Policy management
            +
            263  ******************************************************************************/
            +
            264 
            +
            268 struct KernelConfig
            +
            269 {
            +
            270  int block_threads;
            +
            271  int items_per_thread;
            +
            272  int tile_size;
            +
            273  int sm_occupancy;
            +
            274 
            275  CUB_RUNTIME_FUNCTION __forceinline__
            -
            276  cudaError_t Init(KernelPtrT kernel_ptr)
            -
            277  {
            -
            278  block_threads = AgentPolicyT::BLOCK_THREADS;
            -
            279  items_per_thread = AgentPolicyT::ITEMS_PER_THREAD;
            -
            280  tile_size = block_threads * items_per_thread;
            -
            281  cudaError_t retval = MaxSmOccupancy(sm_occupancy, kernel_ptr, block_threads);
            -
            282  return retval;
            -
            283  }
            -
            284 };
            -
            285 
            -
            286 
            -
            287 
            -
            289 template <int PTX_VERSION, typename PolicyT, typename PrevPolicyT>
            -
            290 struct ChainedPolicy
            -
            291 {
            -
            293  typedef typename If<(CUB_PTX_ARCH < PTX_VERSION), PrevPolicyT::ActivePolicy, PolicyT>::Type ActivePolicy;
            -
            294 
            -
            296  template <typename FunctorT>
            -
            297  CUB_RUNTIME_FUNCTION __forceinline__
            -
            298  static cudaError_t Invoke(int ptx_version, FunctorT &op)
            -
            299  {
            -
            300  if (ptx_version < PTX_VERSION) {
            -
            301  return PrevPolicyT::Invoke(ptx_version, op);
            -
            302  }
            -
            303  return op.template Invoke<PolicyT>();
            -
            304  }
            -
            305 };
            -
            306 
            -
            308 template <int PTX_VERSION, typename PolicyT>
            -
            309 struct ChainedPolicy<PTX_VERSION, PolicyT, PolicyT>
            -
            310 {
            -
            312  typedef PolicyT ActivePolicy;
            -
            313 
            -
            315  template <typename FunctorT>
            -
            316  CUB_RUNTIME_FUNCTION __forceinline__
            -
            317  static cudaError_t Invoke(int ptx_version, FunctorT &op) {
            -
            318  return op.template Invoke<PolicyT>();
            -
            319  }
            -
            320 };
            -
            321 
            -
            322 
            -
            323 
            -
            324 
            -
            325 #endif // Do not document
            -
            326 
            +
            276  KernelConfig() : block_threads(0), items_per_thread(0), tile_size(0), sm_occupancy(0) {}
            +
            277 
            +
            278  template <typename AgentPolicyT, typename KernelPtrT>
            +
            279  CUB_RUNTIME_FUNCTION __forceinline__
            +
            280  cudaError_t Init(KernelPtrT kernel_ptr)
            +
            281  {
            +
            282  block_threads = AgentPolicyT::BLOCK_THREADS;
            +
            283  items_per_thread = AgentPolicyT::ITEMS_PER_THREAD;
            +
            284  tile_size = block_threads * items_per_thread;
            +
            285  cudaError_t retval = MaxSmOccupancy(sm_occupancy, kernel_ptr, block_threads);
            +
            286  return retval;
            +
            287  }
            +
            288 };
            +
            289 
            +
            290 
            +
            291 
            +
            293 template <int PTX_VERSION, typename PolicyT, typename PrevPolicyT>
            +
            294 struct ChainedPolicy
            +
            295 {
            +
            297  typedef typename If<(CUB_PTX_ARCH < PTX_VERSION), typename PrevPolicyT::ActivePolicy, PolicyT>::Type ActivePolicy;
            +
            298 
            +
            300  template <typename FunctorT>
            +
            301  CUB_RUNTIME_FUNCTION __forceinline__
            +
            302  static cudaError_t Invoke(int ptx_version, FunctorT &op)
            +
            303  {
            +
            304  if (ptx_version < PTX_VERSION) {
            +
            305  return PrevPolicyT::Invoke(ptx_version, op);
            +
            306  }
            +
            307  return op.template Invoke<PolicyT>();
            +
            308  }
            +
            309 };
            +
            310 
            +
            312 template <int PTX_VERSION, typename PolicyT>
            +
            313 struct ChainedPolicy<PTX_VERSION, PolicyT, PolicyT>
            +
            314 {
            +
            316  typedef PolicyT ActivePolicy;
            +
            317 
            +
            319  template <typename FunctorT>
            +
            320  CUB_RUNTIME_FUNCTION __forceinline__
            +
            321  static cudaError_t Invoke(int ptx_version, FunctorT &op) {
            +
            322  return op.template Invoke<PolicyT>();
            +
            323  }
            +
            324 };
            +
            325 
            +
            326 
            327 
            328 
            -
            329  // end group UtilMgmt
            +
            329 #endif // Do not document
            +
            330 
            331 
            -
            332 } // CUB namespace
            -
            333 CUB_NS_POSTFIX // Optional outer namespace(s)
            +
            332 
            +
            333  // end group UtilMgmt
            +
            335 
            +
            336 } // CUB namespace
            +
            337 CUB_NS_POSTFIX // Optional outer namespace(s)
            -Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/util__ptx_8cuh.html b/docs/html/util__ptx_8cuh.html index 2da638fb01..5e798e6cbf 100644 --- a/docs/html/util__ptx_8cuh.html +++ b/docs/html/util__ptx_8cuh.html @@ -229,7 +229,7 @@
            -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/util__ptx_8cuh_source.html b/docs/html/util__ptx_8cuh_source.html index fe1b6cae21..7612174173 100644 --- a/docs/html/util__ptx_8cuh_source.html +++ b/docs/html/util__ptx_8cuh_source.html @@ -237,7 +237,7 @@
            166  unsigned int bit_start,
            167  unsigned int num_bits)
            168 {
            -
            169  return BFE(source, bit_start, num_bits, Int2Type<sizeof(UnsignedBits)>());
            +
            169  return BFE(source, bit_start, num_bits, Int2Type<sizeof(UnsignedBits)>());
            170 }
            171 
            172 
            @@ -617,7 +617,7 @@
            -Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/util__type_8cuh.html b/docs/html/util__type_8cuh.html index cabebdaf8b..ef388725d5 100644 --- a/docs/html/util__type_8cuh.html +++ b/docs/html/util__type_8cuh.html @@ -106,6 +106,7 @@
            #include <iostream>
            #include <limits>
            +#include <cfloat>
            #include "util_macro.cuh"
            #include "util_arch.cuh"
            #include "util_namespace.cuh"
            @@ -120,24 +121,6 @@
            - - - - - - - - - - - - - - - - - - @@ -168,7 +151,7 @@
            -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/util__type_8cuh_source.html.REMOVED.git-id b/docs/html/util__type_8cuh_source.html.REMOVED.git-id index ca40c1f51f..05eb304320 100644 --- a/docs/html/util__type_8cuh_source.html.REMOVED.git-id +++ b/docs/html/util__type_8cuh_source.html.REMOVED.git-id @@ -1 +1 @@ -72d4fabf6d2b327ccbd3b99922d2c0e9bcaf53b2 \ No newline at end of file +c8dbf5bea258732932a8a61fe92a028b93b2f5f4 \ No newline at end of file diff --git a/docs/html/warp__reduce_8cuh.html b/docs/html/warp__reduce_8cuh.html index 1f1de2a3d0..8e5002d6d9 100644 --- a/docs/html/warp__reduce_8cuh.html +++ b/docs/html/warp__reduce_8cuh.html @@ -141,7 +141,7 @@
            -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/warp__reduce_8cuh_source.html b/docs/html/warp__reduce_8cuh_source.html index 56d73da582..cda9be286f 100644 --- a/docs/html/warp__reduce_8cuh_source.html +++ b/docs/html/warp__reduce_8cuh_source.html @@ -191,7 +191,7 @@
            187 
            188 public:
            189 
            -
            191  struct TempStorage : Uninitialized<_TempStorage> {};
            +
            191  struct TempStorage : Uninitialized<_TempStorage> {};
            192 
            193 
            194  /******************************************************************/
            @@ -297,7 +297,7 @@
            -Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:14 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/warp__scan_8cuh.html b/docs/html/warp__scan_8cuh.html index c1cccc5d98..6d82b618a3 100644 --- a/docs/html/warp__scan_8cuh.html +++ b/docs/html/warp__scan_8cuh.html @@ -141,7 +141,7 @@
            -Generated on Tue Nov 10 2015 13:58:04 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
            diff --git a/docs/html/warp__scan_8cuh_source.html b/docs/html/warp__scan_8cuh_source.html index dc760bc441..071623cc03 100644 --- a/docs/html/warp__scan_8cuh_source.html +++ b/docs/html/warp__scan_8cuh_source.html @@ -185,7 +185,7 @@
            188 
            189 public:
            190 
            -
            192  struct TempStorage : Uninitialized<_TempStorage> {};
            +
            192  struct TempStorage : Uninitialized<_TempStorage> {};
            193 
            194 
            195  /******************************************************************/
            @@ -365,7 +365,7 @@
            -Generated on Tue Nov 10 2015 13:58:03 for CUB by   +Generated on Mon Dec 14 2015 13:11:15 for CUB by   doxygen 1.8.4
            diff --git a/docs/mainpage.dox b/docs/mainpage.dox index e2daa2b44a..03b9526cc7 100644 --- a/docs/mainpage.dox +++ b/docs/mainpage.dox @@ -62,7 +62,7 @@    -Download CUB v1.5.1 +Download CUB v1.5.0

            Functions

            struct  cub::Equals< A, B >
             Type equality test. More...
             
            struct  cub::NullType
             A simple "NULL" marker type. More...
             
            struct  cub::Int2Type< A >
             Allows for the treatment of an integral constant as a type at compile-time (e.g., to achieve static call dispatch based on constant integral values) More...
             
            struct  cub::CubVector< T, vec_elements >
             Exposes a member typedef Type that names the corresponding CUDA vector type if one exists. Otherwise Type refers to the CubVector structure itself, which will wrap the corresponding x, y, etc. vector fields. More...
             
            struct  cub::Uninitialized< T >
             A storage-backing wrapper that allows types with non-trivial constructors to be aliased in unions. More...
             
            struct  cub::KeyValuePair< _Key, _Value >
             A key identifier paired with a corresponding value. More...
             
            struct  cub::DoubleBuffer< T >
             Double-buffer storage wrapper for multi-pass stream transformations that require more than one storage array for streaming intermediate results back and forth. More...
             
            struct  cub::Log2< N, CURRENT_VAL, COUNT >
             Statically determine log2(N), rounded up. More...
             
            @@ -527,11 +527,11 @@ for the current feature release. diff --git a/test/test_device_radix_sort.cu b/test/test_device_radix_sort.cu index 9876916fb5..ef0b7f4af1 100644 --- a/test/test_device_radix_sort.cu +++ b/test/test_device_radix_sort.cu @@ -1217,6 +1217,8 @@ int main(int argc, char** argv) // Compile/run thorough tests for (int i = 0; i <= g_repeat; ++i) { + TestGen (num_items, num_segments); +/* TestGen (num_items, num_segments); TestGen (num_items, num_segments); TestGen (num_items, num_segments); @@ -1237,6 +1239,7 @@ int main(int argc, char** argv) if (ptx_version > 120) // Don't check doubles on PTX120 or below because they're down-converted TestGen (num_items, num_segments); +*/ } diff --git a/test/test_util.h b/test/test_util.h index cb1d3381bc..944431a417 100644 --- a/test/test_util.h +++ b/test/test_util.h @@ -526,6 +526,31 @@ __host__ __device__ __forceinline__ void InitValue(GenMode gen_mode, T &value, i } +/** + * Initialize value (bool) + */ +__host__ __device__ __forceinline__ void InitValue(GenMode gen_mode, bool &value, int index = 0) +{ + switch (gen_mode) + { +#if (CUB_PTX_ARCH == 0) + case RANDOM: + char c; + RandomBits(c, 0, 0, 1); + value = (bool) c; + break; +#endif + case UNIFORM: + value = true; + break; + case INTEGER_SEED: + default: + value = (bool) index; + break; + } +} + + /** * cub::NullType test initialization */
            -11/13/2015
            +12/15/2015
            [CUB v1.5.0](download_cub.html)             		- New Features: - - Added new [segmented device-wide operations](group___segmented_module.html) for sort, scan, and reduction primitives. + - Added new [segmented device-wide operations](group___segmented_module.html) for sort and reduction primitives. - See the [change-log](CHANGE_LOG.TXT) for further details, including bug-fixes