1 //====--- OMPGridValues.h - Language-specific address spaces --*- C++ -*-====//
2 //
3 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4 // See https://llvm.org/LICENSE.txt for license information.
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6 //
7 //===----------------------------------------------------------------------===//
8 ///
9 /// \file
10 /// \brief Provides definitions for Target specific Grid Values
11 ///
12 //===----------------------------------------------------------------------===//
13
14 #ifndef LLVM_FRONTEND_OPENMP_OMPGRIDVALUES_H
15 #define LLVM_FRONTEND_OPENMP_OMPGRIDVALUES_H
16
17 namespace llvm {
18
19 namespace omp {
20
21 /// \brief Defines various target-specific GPU grid values that must be
22 /// consistent between host RTL (plugin), device RTL, and clang.
23 /// We can change grid values for a "fat" binary so that different
24 /// passes get the correct values when generating code for a
25 /// multi-target binary. Both amdgcn and nvptx values are stored in
26 /// this file. In the future, should there be differences between GPUs
27 /// of the same architecture, then simply make a different array and
28 /// use the new array name.
29 ///
30 /// Example usage in clang:
31 /// const unsigned slot_size =
32 /// ctx.GetTargetInfo().getGridValue().GV_Warp_Size;
33 ///
34 /// Example usage in libomptarget/deviceRTLs:
35 /// #include "llvm/Frontend/OpenMP/OMPGridValues.h"
36 /// #ifdef __AMDGPU__
37 /// #define GRIDVAL AMDGPUGridValues
38 /// #else
39 /// #define GRIDVAL NVPTXGridValues
40 /// #endif
41 /// ... Then use this reference for GV_Warp_Size in the deviceRTL source.
42 /// llvm::omp::GRIDVAL().GV_Warp_Size
43 ///
44 /// Example usage in libomptarget hsa plugin:
45 /// #include "llvm/Frontend/OpenMP/OMPGridValues.h"
46 /// #define GRIDVAL AMDGPUGridValues
47 /// ... Then use this reference to access GV_Warp_Size in the hsa plugin.
48 /// llvm::omp::GRIDVAL().GV_Warp_Size
49 ///
50 /// Example usage in libomptarget cuda plugin:
51 /// #include "llvm/Frontend/OpenMP/OMPGridValues.h"
52 /// #define GRIDVAL NVPTXGridValues
53 /// ... Then use this reference to access GV_Warp_Size in the cuda plugin.
54 /// llvm::omp::GRIDVAL().GV_Warp_Size
55 ///
56
57 struct GV {
58 /// The size reserved for data in a shared memory slot.
59 const unsigned GV_Slot_Size;
60 /// The default value of maximum number of threads in a worker warp.
61 const unsigned GV_Warp_Size;
62
warpSlotSizeGV63 constexpr unsigned warpSlotSize() const {
64 return GV_Warp_Size * GV_Slot_Size;
65 }
66
67 /// the maximum number of teams.
68 const unsigned GV_Max_Teams;
69 // An alternative to the heavy data sharing infrastructure that uses global
70 // memory is one that uses device __shared__ memory. The amount of such space
71 // (in bytes) reserved by the OpenMP runtime is noted here.
72 const unsigned GV_SimpleBufferSize;
73 // The absolute maximum team size for a working group
74 const unsigned GV_Max_WG_Size;
75 // The default maximum team size for a working group
76 const unsigned GV_Default_WG_Size;
77
maxWarpNumberGV78 constexpr unsigned maxWarpNumber() const {
79 return GV_Max_WG_Size / GV_Warp_Size;
80 }
81 };
82
83 /// For AMDGPU GPUs
84 static constexpr GV AMDGPUGridValues64 = {
85 256, // GV_Slot_Size
86 64, // GV_Warp_Size
87 128, // GV_Max_Teams
88 896, // GV_SimpleBufferSize
89 1024, // GV_Max_WG_Size,
90 256, // GV_Default_WG_Size
91 };
92
93 static constexpr GV AMDGPUGridValues32 = {
94 256, // GV_Slot_Size
95 32, // GV_Warp_Size
96 128, // GV_Max_Teams
97 896, // GV_SimpleBufferSize
98 1024, // GV_Max_WG_Size,
99 256, // GV_Default_WG_Size
100 };
101
getAMDGPUGridValues()102 template <unsigned wavesize> constexpr const GV &getAMDGPUGridValues() {
103 static_assert(wavesize == 32 || wavesize == 64, "");
104 return wavesize == 32 ? AMDGPUGridValues32 : AMDGPUGridValues64;
105 }
106
107 /// For Nvidia GPUs
108 static constexpr GV NVPTXGridValues = {
109 256, // GV_Slot_Size
110 32, // GV_Warp_Size
111 1024, // GV_Max_Teams
112 896, // GV_SimpleBufferSize
113 1024, // GV_Max_WG_Size
114 128, // GV_Default_WG_Size
115 };
116
117 } // namespace omp
118 } // namespace llvm
119
120 #endif // LLVM_FRONTEND_OPENMP_OMPGRIDVALUES_H
121