| /oneTBB/.github/workflows/ |
| H A D | ci.yml | 17 on:
37 runs-on: [ubuntu-20.04]
47 runs-on: [ubuntu-20.04]
62 runs-on: [ubuntu-22.04]
85 runs-on: ubuntu-latest
114 runs-on: [ubuntu-20.04]
133 runs-on: [ubuntu-latest]
146 runs-on: ['${{ matrix.os }}']
184 runs-on: ['${{ matrix.os }}']
210 runs-on: ['${{ matrix.os }}']
[all …]
|
| H A D | labeler.yml | 10 # distributed under the License is distributed on an "AS IS" BASIS,
15 on:
23 runs-on: ubuntu-latest
|
| H A D | issue_labeler.yml | 10 # distributed under the License is distributed on an "AS IS" BASIS,
16 on:
24 runs-on: ubuntu-latest
|
| /oneTBB/doc/main/tbb_userguide/ |
| H A D | Parallelizing_Flow_Graph.rst | 15 execute its body object on the incoming message. Messages flow through
50 peanut butter or jelly on the bread. You must spread on the peanut
51 butter before you put away the peanut butter jar, and likewise spread on
52 the jelly before you put away the jelly jar. And, you need to spread on
55 matter if you spread on the peanut butter first or the jelly first. It
76 graph. For example, you must "Put jelly on 1 slice" **before** you "Put
88 the peanut butter might be spread on one slice of bread in parallel with
89 the jelly being spread on the other slice. The interface expresses what
|
| H A D | Containers.rst | 17 thread operate on the container at a time, but that approach eliminates
25 - **Fine-grained locking:** Multiple threads operate on the container
36 have higher overheads than regular STL containers. Operations on
46 operation on the same object. Otherwise the resulting race may cause
47 the operation to be executed on an undefined object.
|
| H A D | Controlling_Chunking_os.rst | 42 ``parallel_for`` in the example invokes ``ApplyFoo::operator()`` on
62 Because of the impact of grainsize on parallel loops, it is worth
63 reading the following material even if you rely on ``auto_partitioner``
87 of useful work depends upon the grainsize, not on the number of grains.
117 on the side of being a little too high instead of a little too low,
128 versus grainsize, based on the floating point ``a[i]=b[i]*c``
130 The times were collected on a four-socket machine with eight hardware
142 The scale is logarithmic. The downward slope on the left side indicates
147 large grainsize. At the end on the right, the curve turns up because the
|
| H A D | Mutual_Exclusion.rst | 10 on which a thread can acquire a lock. Only one thread at a time can have
11 a lock on a mutex; other threads have to wait their turn.
15 on a ``spin_mutex`` busy waits until it can acquire the lock. A
50 on ``FreeListMutex``. The destructor releases the lock. The braces
89 Method ``acquire`` waits until it can acquire a lock on the mutex;
98 why there are not simply acquire and release methods on the mutex object
|
| H A D | Guiding_Task_Scheduler_Execution.rst | 30 if the process affinity mask excludes execution on some of the NUMA nodes, then these NUMA nodes are
37 The execution on systems with non-uniform memory access (`NUMA <https://en.wikipedia.org/wiki/Non-u…
38 may cause a performance penalty if threads from one NUMA node access the memory allocated on
84 allow more than one thread to run on each core simultaneously. However, there might be situations
88 The example shows how to allow only one thread to run on each core at a time:
97 A more composable way to limit the number of threads executing on cores is by setting the maximal
|
| H A D | Flow_Graph_Buffering_in_Nodes.rst | 10 the message by that node. As noted in the section on Single-push vs.
12 successors, depending on the type of the node, how many successors are
17 successor. In this case what happens to the message depends on the type
|
| H A D | appendix_A.rst | 32 and subsequently thread B runs on the same physical thread, B will tend
36 worse yet, the next time slice for thread A may be on a different
41 lock on a resource, and its time slice runs out before it releases the
44 turn at a time slice to come up. Any other threads waiting on the lock
|
| H A D | Lock_Pathologies.rst | 25 - Each thread holds at least one lock on a mutex, and is waiting on
45 acquire a lock on one of each, you could always acquire the "outer
|
| /oneTBB/doc/main/intro/ |
| H A D | Benefits.rst | 38 provide a foundation, not a solution. Instead, oneTBB focuses on the
53 multiple threads to work on different parts of a collection.
59 - **oneTBB relies on generic programming**. Traditional libraries
65 interfaces are specified by *requirements* on types. For example, C++
67 defined in terms of iterators on the sequence. The requirements on
82 Specification in terms of requirements on types enables the template to
84 deques. Similarly, the oneTBB templates specify requirements on types,
|
| /oneTBB/cmake/toolchains/ |
| H A D | mips.cmake | 10 # distributed under the License is distributed on an "AS IS" BASIS,
30 # In old CMake versions (checked on 3.5) there is invocation of try_run command in FindThreads.cmak…
31 # It can't be executed on host system in case of cross-compilation for MIPS architecture.
32 # Define return code for this try_run as 0 since threads are expected to be available on target mac…
|
| /oneTBB/doc/main/tbb_userguide/design_patterns/ |
| H A D | Fenced_Data_Transfer.rst | 13 Write a message to memory and have another processor read it on
24 concurrently act on a memory location, or are using reads and writes
65 These assumptions are trivially true on uniprocessor hardware.
66 However, they may break on multiprocessor hardware. Reordering by the
131 thread that acquires a lock on a mutex always sees any memory writes
132 done by another thread before it released a lock on that mutex.
148 effect on the visible ordering of that write with respect to other
|
| H A D | Local_Serializer.rst | 18 order of "select" and "delete" operations on the same buffer would be
19 bad. However, commuting operations on different buffers might be
65 The following example builds on the Non-Preemptive Priorities example
107 Constraints on a given functor are resolved from top to bottom in the
251 If priorities are not necessary, there are two variations on method
259 thread operating on it will keep servicing those tasks to the
273 The pattern generalizes to constraints on work items more general
|
| /oneTBB/ |
| H A D | INSTALL.md | 6 …- Make sure you have installed CMake version 3.1 (or newer) on your system. oneTBB uses CMake buil…
42 …out installing prefix. It defaults to `/usr/local` on UNIX* and `c:/Program Files/${PROJECT_NAME}`…
62 - `runtime` - oneTBB runtime package (core shared libraries and `.dll` files on Windows* OS).
63 …e (header files, CMake integration files, library symbolic links, and `.lib` files on Windows* OS).
91 …ntributors. If the version is out of date, create an issue or pull request on the [vcpkg repositor…
|
| H A D | LICENSE.txt | 41 form, that is based on (or derived from) the Work and for which the
52 or by an individual or Legal Entity authorized to submit on behalf of
56 communication on electronic mailing lists, source code control systems,
57 and issue tracking systems that are managed by, or on behalf of, the
63 on behalf of whom a Contribution has been received by Licensor and
145 Contributor provides its Contributions) on an "AS IS" BASIS,
170 on Your own behalf and on Your sole responsibility, not on behalf
185 file or class name and description of purpose be included on the
198 distributed under the License is distributed on an "AS IS" BASIS,
|
| H A D | .gitattributes | 5 # to native line endings on checkout.
37 # Declare files that will always have CRLF line endings on checkout.
|
| /oneTBB/doc/GSG/ |
| H A D | next_steps.rst | 22 oneTBB can coordinate with Intel(R) OpenMP on CPU resources usage
130 If you need NUMA/Hybrid CPU support in oneTBB, you need to make sure that HWLOC* is installed on yo…
132 …ogy of modern architectures (NUMA, hybrid CPU systems, etc). oneTBB relies on HWLOC* to identify t…
134 …herefore, it's important to make sure that HWLOC* is installed before using oneTBB on such systems.
136 Check HWLOC* on the System
138 To check if HWLOC* is already installed on your system, run ``hwloc-ls``:
|
| /oneTBB/src/tbbmalloc_proxy/ |
| H A D | CMakeLists.txt | 10 # distributed under the License is distributed on an "AS IS" BASIS,
62 # Avoid use of target_link_libraries here as it changes /DEF option to \DEF on Windows.
70 # target_link_libraries may incorrectly handle some options (on Windows, for example).
|
| /oneTBB/examples/parallel_for/seismic/ |
| H A D | README.md | 10 … GUI mode, supported values are `gdi`, `d2d`, `con` on Windows, `x`,`con` on Linux and `mac`,`con`…
|
| /oneTBB/src/tbbbind/ |
| H A D | CMakeLists.txt | 10 # distributed under the License is distributed on an "AS IS" BASIS,
53 # Avoid use of target_link_libraries here as it changes /DEF option to \DEF on Windows.
70 # target_link_libraries may incorrectly handle some options (on Windows, for example).
|
| /oneTBB/src/tbbmalloc/ |
| H A D | CMakeLists.txt | 10 # distributed under the License is distributed on an "AS IS" BASIS,
70 # Avoid use of target_link_libraries here as it changes /DEF option to \DEF on Windows.
90 # target_link_libraries may incorrectly handle some options (on Windows, for example).
|
| /oneTBB/integration/pkg-config/ |
| H A D | tbb.pc.in | 10 # distributed under the License is distributed on an "AS IS" BASIS,
20 Description: C++ library for parallel programming on multi-core processors.
|
| /oneTBB/examples/test_all/fibonacci/ |
| H A D | CMakeLists.txt | 10 # distributed under the License is distributed on an "AS IS" BASIS,
27 $<$<PLATFORM_ID:Linux>:rt>) # Link "rt" library on Linux
|