Searched refs:calls (Results 1 – 18 of 18) sorted by relevance
| /oneTBB/doc/main/tbb_userguide/ |
| H A D | communicate_with_nodes.rst | 12 operations, such as calls to graph::wait_for_all and exception handling.
13 To optimize performance, the library may make calls to a node's
18 instead use explicit calls to try_put. This will prevent the runtime
80 calls is important. In the (incorrect) code segment above, the first
89 If the calls to wait_for_all are swapped, the code works as expected:
|
| H A D | Mapping_Nodes2Tasks.rst | 10 three calls to try_put spawn three tasks; each one applies the lambda
15 functions in oneTBB, the thread that calls ``wait_for_all`` is not spinning
41 All execution in the flow graph happens asynchronously. The calls to
|
| H A D | use_make_edge.rst | 23 calls these node functions directly to implement optimizations on the
|
| H A D | automatically-replacing-malloc.rst | 8 replace all calls to standard functions for dynamic memory allocation
|
| H A D | work_isolation.rst | 12 particular, when a parallel construct calls another parallel
17 In the following example with two ``parallel_for`` calls, the call to
|
| H A D | concurrent_vector_ug.rst | 33 if it is shorter. Concurrent calls to the growth methods do not
|
| H A D | use_input_node.rst | 24 // use it in calls to make_edge…
|
| H A D | Guiding_Task_Scheduler_Execution.rst | 64 …ms with hybrid CPU architecture show reasonable performance without involving additional API calls.
|
| H A D | Nodes.rst | 100 before spawning a third task to apply the body to 3. The calls to
|
| H A D | create_token_based_system.rst | 128 number of explicit calls to ``try_put``, but there are other options. For
|
| H A D | Flow_Graph_Reservation.rst | 26 - The reserving input port calls ``try_reserve`` on each edge in pull
|
| /oneTBB/test/common/ |
| H A D | concurrent_lru_cache_common.h | 38 static int calls; member
41 ++calls; in call()
47 int call_counter<Tag, Type>::calls = 0;
|
| /oneTBB/examples/graph/cholesky/ |
| H A D | README.md | 4 **dpotrf**: An implementation that calls the oneAPI Math Kernel Library (oneMKL) `dpotrf` function …
8 …s used to create an unrolled version of the computation. Where the oneMKL calls would have been ma…
10 …r each call. So for example, there is only a single node that applies all calls to `dtrsm`. This n…
|
| /oneTBB/test/conformance/ |
| H A D | conformance_concurrent_lru_cache.cpp | 114 preset::counter_type::calls == 1,
|
| /oneTBB/doc/main/tbb_userguide/design_patterns/ |
| H A D | Divide_and_Conquer.rst | 178 // Wait for recursive calls to complete
192 recursive calls in parallel.
|
| H A D | GUI_Thread.rst | 83 each event, the GUI thread calls a user-defined function ``WndProc`` to process an event.
138 #. The GUI thread calls ``LongRunningWork``, which hands off the work
|
| /oneTBB/doc/main/reference/ |
| H A D | type_specified_message_keys.rst | 41 When constructed this way, a ``join_node`` calls the
|
| H A D | concurrent_lru_cache_cls.rst | 32 When no item is found for a given key, the container calls the user-specified
|