1Support, Getting Involved, and FAQ
2==================================
3
4Please do not hesitate to reach out to us via [email protected] or join
5one of our :ref:`regular calls <calls>`. Some common questions are answered in
6the :ref:`faq`.
7
8.. _calls:
9
10Calls
11-----
12
13OpenMP in LLVM Technical Call
14^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
15
16-   Development updates on OpenMP (and OpenACC) in the LLVM Project, including Clang, optimization, and runtime work.
17-   Join `OpenMP in LLVM Technical Call <https://bluejeans.com/544112769//webrtc>`__.
18-   Time: Weekly call on every Wednesday 7:00 AM Pacific time.
19-   Meeting minutes are `here <https://docs.google.com/document/d/1Tz8WFN13n7yJ-SCE0Qjqf9LmjGUw0dWO9Ts1ss4YOdg/edit>`__.
20-   Status tracking `page <https://openmp.llvm.org/docs>`__.
21
22
23OpenMP in Flang Technical Call
24^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
25-   Development updates on OpenMP and OpenACC in the Flang Project.
26-   Join `OpenMP in Flang Technical Call <https://bit.ly/39eQW3o>`_
27-   Time: Weekly call on every Thursdays 8:00 AM Pacific time.
28-   Meeting minutes are `here <https://docs.google.com/document/d/1yA-MeJf6RYY-ZXpdol0t7YoDoqtwAyBhFLr5thu5pFI>`__.
29-   Status tracking `page <https://docs.google.com/spreadsheets/d/1FvHPuSkGbl4mQZRAwCIndvQx9dQboffiD-xD0oqxgU0/edit#gid=0>`__.
30
31
32.. _faq:
33
34FAQ
35---
36
37.. note::
38   The FAQ is a work in progress and most of the expected content is not
39   yet available. While you can expect changes, we always welcome feedback and
40   additions. Please contact, e.g., through ``[email protected]``.
41
42
43Q: How to contribute a patch to the webpage or any other part?
44^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
45
46All patches go through the regular `LLVM review process
47<https://llvm.org/docs/Contributing.html#how-to-submit-a-patch>`_.
48
49
50.. _build_offload_capable_compiler:
51
52Q: How to build an OpenMP offload capable compiler?
53^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
54
55To build an *effective* OpenMP offload capable compiler, only one extra CMake
56option, `LLVM_ENABLE_RUNTIMES="openmp"`, is needed when building LLVM (Generic
57information about building LLVM is available `here <https://llvm.org/docs/GettingStarted.html>`__.).
58Make sure all backends that are targeted by OpenMP to be enabled. By default,
59Clang will be built with all backends enabled.
60
61If your build machine is not the target machine or automatic detection of the
62available GPUs failed, you should also set:
63
64- `CLANG_OPENMP_NVPTX_DEFAULT_ARCH=sm_XX` where `XX` is the architecture of your GPU, e.g, 80.
65- `LIBOMPTARGET_NVPTX_COMPUTE_CAPABILITIES=YY` where `YY` is the numeric compute capacity of your GPU, e.g., 75.
66
67.. note::
68  The compiler that generates the offload code should be the same (version) as
69  the compiler that builds the OpenMP device runtimes. The OpenMP host runtime
70  can be built by a different compiler.
71
72.. _advanced_builds: https://llvm.org//docs/AdvancedBuilds.html
73
74
75
76Q: Does OpenMP offloading support work in pre-packaged LLVM releases?
77^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
78For now, the answer is most likely *no*. Please see :ref:`build_offload_capable_compiler`.
79
80Q: Does OpenMP offloading support work in packages distributed as part of my OS?
81^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
82For now, the answer is most likely *no*. Please see :ref:`build_offload_capable_compiler`.
83
84
85.. _math_and_complex_in_target_regions:
86
87Q: Does Clang support `<math.h>` and `<complex.h>` operations in OpenMP target on GPUs?
88^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
89
90Yes, LLVM/Clang allows math functions and complex arithmetic inside of OpenMP target regions
91that are compiled for GPUs.
92
93Clang provides a set of wrapper headers that are found first when `math.h` and
94`complex.h`, for C, `cmath` and `complex`, for C++, or similar headers are
95included by the application. These wrappers will eventually include the system
96version of the corresponding header file after setting up a target device
97specific environment. The fact that the system header is included is important
98because they differ based on the architecture and operating system and may
99contain preprocessor, variable, and function definitions that need to be
100available in the target region regardless of the targeted device architecture.
101However, various functions may require specialized device versions, e.g.,
102`sin`, and others are only available on certain devices, e.g., `__umul64hi`. To
103provide "native" support for math and complex on the respective architecture,
104Clang will wrap the "native" math functions, e.g., as provided by the device
105vendor, in an OpenMP begin/end declare variant. These functions will then be
106picked up instead of the host versions while host only variables and function
107definitions are still available. Complex arithmetic and functions are support
108through a similar mechanism. It is worth noting that this support requires
109`extensions to the OpenMP begin/end declare variant context selector
110<https://clang.llvm.org/docs/AttributeReference.html#pragma-omp-declare-variant>`__
111that are exposed through LLVM/Clang to the user as well.
112
113Q: What is a way to debug errors from mapping memory to a target device?
114^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
115
116An experimental way to debug these errors is to use :ref:`remote process
117offloading <remote_offloading_plugin>`.
118By using ``libomptarget.rtl.rpc.so`` and ``openmp-offloading-server``, it is
119possible to explicitly perform memory transfers between processes on the host
120CPU and run sanitizers while doing so in order to catch these errors.
121
122Q: Why does my application say "Named symbol not found" and abort when I run it?
123^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
124
125This is most likely caused by trying to use OpenMP offloading with static
126libraries. Static libraries do not contain any device code, so when the runtime
127attempts to execute the target region it will not be found and you will get an
128an error like this.
129
130.. code-block:: text
131
132   CUDA error: Loading '__omp_offloading_fd02_3231c15__Z3foov_l2' Failed
133   CUDA error: named symbol not found
134   Libomptarget error: Unable to generate entries table for device id 0.
135
136Currently, the only solution is to change how the application is built and avoid
137the use of static libraries.
138