Home
last modified time | relevance | path

Searched refs:space (Results 1 – 16 of 16) sorted by relevance

/oneTBB/include/oneapi/tbb/detail/
H A D_flow_graph_types_impl.h142 pointer_type space;
143 _unwind_space(pointer_type p) : space(p) {}
145 if(space) (void) new (space) Wrapper<default_constructed>(default_constructed());
155 guard.space = nullptr;
179 pointer_type space;
181 _unwind_class(pointer_type p) : space(p), already_built(0) {}
183 if(space) {
184 for(size_t i = already_built; i > 0 ; --i ) space[i-1].~value_type();
185 (void) new(space) Wrapper<default_constructed>(default_constructed());
206 guard.space = nullptr;
[all …]
/oneTBB/doc/main/tbb_userguide/
H A Dparallel_for_os.rst21 The iteration space here is of type ``size_t``, and goes from ``0`` to
23 space into chunks, and runs each chunk on a separate thread. The first
57 iteration space over type ``T``. Class ``parallel_for`` works with other
113 space from 0 to n-1, which ``parallel_for`` divides into subspaces for
116 type. The arguments ``begin`` and ``end`` specify the iteration space
H A DCook_Until_Done_parallel_do.rst7 For some loops, the end of the iteration space is not known in advance,
12 A linked list is an example of an iteration space that is not known in
H A DAdvanced_Topic_Other_Kinds_of_Iteration_Spaces.rst9 You can use |full_name| to define your own iteration space objects. The object
68 The iteration space does not have to be linear. Look at
H A DMutex_Flavors.rst47 user space. It would seem to be the worst of all possible worlds,
62 space. Use it when scalability and fairness are important.
H A DMemory_Allocation.rst55 cost in space, because it must allocate at least one cache line’s
H A DHow_Task_Scheduler_Works.rst27 - **Minimize space**. Execution of the shallowest task leads to the breadth-first unfolding of a gr…
H A DWorking_on_the_Assembly_Line_pipeline.rst312 // Read characters into space that is available in the next slice.
/oneTBB/include/oneapi/tbb/
H A Dcache_aligned_allocator.h124 std::size_t space = correct_size(bytes) + cache_line_alignment; in do_allocate() local
125 std::uintptr_t base = reinterpret_cast<std::uintptr_t>(m_upstream->allocate(space)); in do_allocate()
131 __TBB_ASSERT(space - (result - base) >= bytes, "Not enough space for the storage"); in do_allocate()
/oneTBB/src/tbb/
H A Dallocator.cpp218 std::size_t space = alignment + bytes; in std_cache_aligned_allocate()
219 std::uintptr_t base = reinterpret_cast<std::uintptr_t>(std::malloc(space)); in std_cache_aligned_allocate()
226 __TBB_ASSERT(space - (result - base) >= bytes, "Not enough space for the storage"); in std_cache_aligned_allocate()
/oneTBB/doc/_static/
H A Dtheme_overrides.css7 white-space: normal !important;
/oneTBB/test/tbbmalloc/
H A Dtest_malloc_pools.cpp44 char *space; member
50 bufSize(bufSz), space(new char[bufSize]) { in PoolSpace()
51 memset(space, 0, bufSize); in PoolSpace()
54 delete []space; in ~PoolSpace()
208 void *ret = poolSpace[pool_id].space + poolSpace[pool_id].pos; in CrossThreadGetMem()
H A Dtest_malloc_whitebox.cpp378 static char space[BUF_SIZE]; in getMem() local
384 void *ret = space + pos; in getMem()
1229 const int space = 2*16*1024; // fill at least 2 slabs in TestSlabAlignment() local
1230 void *pointers[space / min_sz]; // the worst case is min_sz byte object in TestSlabAlignment()
1233 for (size_t i = 0; i < space/sz; i++) { in TestSlabAlignment()
1241 for (size_t i = 0; i < space/sz; i++) in TestSlabAlignment()
/oneTBB/examples/task_group/sudoku/
H A DREADME.md4 It uses a straightforward state-space search algorithm that exhibits OR-parallelism. It can be opti…
/oneTBB/examples/parallel_for/seismic/
H A DREADME.md33 * `space` - toggles between parallel and serial execution modes.
/oneTBB/examples/parallel_for/tachyon/
H A DREADME.md48 * `space` - toggle run-time display updating mode while rendering (see no-display-updating above).