2
2
README for the LLVM* OpenMP* Offloading Runtime Library (libomptarget)
3
3
======================================================================
4
4
5
- How to Build the LLVM* OpenMP* Offloading Runtime Library (libomptarget)
6
- ========================================================================
7
- In-tree build:
8
-
9
- $ cd where-you-want-to-live
10
- Check out openmp (libomptarget lives under ./libomptarget) into llvm/projects
11
- $ cd where-you-want-to-build
12
- $ mkdir build && cd build
13
- $ cmake path/to/llvm -DCMAKE_C_COMPILER=<C compiler> -DCMAKE_CXX_COMPILER=<C++ compiler>
14
- $ make omptarget
15
-
16
- Out-of-tree build:
17
-
18
- $ cd where-you-want-to-live
19
- Check out openmp (libomptarget lives under ./libomptarget)
20
- $ cd where-you-want-to-live/openmp/libomptarget
21
- $ mkdir build && cd build
22
- $ cmake path/to/openmp -DCMAKE_C_COMPILER=<C compiler> -DCMAKE_CXX_COMPILER=<C++ compiler>
23
- $ make
24
-
25
- For details about building, please look at README.rst in the parent directory.
5
+ For details about building, please look at README.rst in the parent directory
6
+ and the build instructions as well as FAQ at https://openmp.llvm.org.
26
7
27
8
Architectures Supported
28
9
=======================
@@ -34,40 +15,13 @@ following host architectures:
34
15
* ARM(R) AArch64 architecture (little endian)
35
16
36
17
The currently supported offloading device architectures are:
37
- * Intel(R) 64 architecture (generic 64-bit plugin - mostly for testing purposes )
38
- * IBM(R) Power architecture (big endian) (generic 64-bit plugin - mostly for testing purposes)
39
- * IBM(R) Power architecture (little endian) (generic 64-bit plugin - mostly for testing purposes)
40
- * ARM(R) AArch64 architecture (little endian) (generic 64-bit plugin - mostly for testing purposes)
18
+ * Intel(R) or AMD(R) 64-bit architecture (x86_64 )
19
+ * IBM(R) Power architecture (big endian)
20
+ * IBM(R) Power architecture (little endian)
21
+ * ARM(R) AArch64 architecture (little endian)
41
22
* CUDA(R) enabled 64-bit NVIDIA(R) GPU architectures
42
-
43
- Supported RTL Build Configurations
44
- ==================================
45
- Supported Architectures: Intel(R) 64, IBM(R) Power 7 and Power 8
46
-
47
- ---------------------------
48
- | gcc | clang |
49
- --------------|------------|------------|
50
- | Linux* OS | Yes(1) | Yes(2) |
51
- -----------------------------------------
52
-
53
- (1) gcc version 4.8.2 or later is supported.
54
- (2) clang version 3.7 or later is supported.
55
-
56
-
57
- Front-end Compilers that work with this RTL
58
- ===========================================
59
-
60
- The following compilers are known to do compatible code generation for
61
- this RTL:
62
- - clang (from https://github.com/clang-ykt )
63
- - clang (development branch at http://clang.llvm.org - several features still
64
- under development)
23
+ * AMD(R) enabled 64-bit AMD(R) GPU architectures
65
24
66
25
-----------------------------------------------------------------------
67
26
68
- Notices
69
- =======
70
- This library and related compiler support is still under development, so the
71
- employed interface is likely to change in the future.
72
-
73
27
*Other names and brands may be claimed as the property of others.
0 commit comments