| blob | 517a23e959206e01fe00625449b2acf2bc8aab7c |
1 User Guides\r
2 ===========\r
3 \r
4 NOTE: If you are a user who is only interested in using an LLVM-based compiler,\r
5 you should look into `Clang <https://clang.llvm.org>`_ instead. The\r
6 documentation here is intended for users who have a need to work with the\r
7 intermediate LLVM representation.\r
8 \r
9 .. contents::\r
10 :local:\r
11 \r
12 .. toctree::\r
13 :hidden:\r
14 \r
15 AArch64SME\r
16 AddingConstrainedIntrinsics\r
17 AdvancedBuilds\r
18 AliasAnalysis\r
19 AMDGPUUsage\r
20 Benchmarking\r
21 BigEndianNEON\r
22 BuildingADistribution\r
23 CFIVerify\r
24 CMake\r
25 CMakePrimer\r
26 CodeGenerator\r
27 CodeOfConduct\r
28 CommandLine\r
29 CompileCudaWithLLVM\r
30 CoverageMappingFormat\r
31 CycleTerminology\r
32 DebuggingJITedCode\r
33 DirectXUsage\r
34 Docker\r
35 ExtendingLLVM\r
36 GoldPlugin\r
37 HowToBuildOnARM\r
38 HowToBuildWithPGO\r
39 HowToBuildWindowsItaniumPrograms\r
40 HowToCrossCompileBuiltinsOnArm\r
41 HowToCrossCompileLLVM\r
42 HowToUpdateDebugInfo\r
43 InstrRefDebugInfo\r
44 LinkTimeOptimization\r
45 LoopTerminology\r
46 MarkdownQuickstartTemplate\r
47 MemorySSA\r
48 MergeFunctions\r
49 MCJITDesignAndImplementation\r
50 MisExpect\r
51 ORCv2\r
52 OpaquePointers\r
53 JITLink\r
54 NewPassManager\r
55 NVPTXUsage\r
56 Phabricator\r
57 Passes\r
58 ReportingGuide\r
59 ResponseGuide\r
60 Remarks\r
61 RISCVUsage\r
62 SourceLevelDebugging\r
63 SPIRVUsage\r
64 StackSafetyAnalysis\r
65 SupportLibrary\r
66 TableGen/index\r
67 TableGenFundamentals\r
68 Vectorizers\r
69 WritingAnLLVMPass\r
70 WritingAnLLVMNewPMPass\r
71 WritingAnLLVMBackend\r
72 yaml2obj\r
73 \r
74 Clang\r
75 -----\r
76 \r
77 :doc:`HowToBuildOnARM`\r
78 Notes on building and testing LLVM/Clang on ARM.\r
79 \r
80 :doc:`HowToBuildWithPGO`\r
81 Notes on building LLVM/Clang with PGO.\r
82 \r
83 :doc:`HowToCrossCompileLLVM`\r
84 Notes on cross-building and testing LLVM/Clang.\r
85 \r
86 `How to build the C, C++, ObjC, and ObjC++ front end`__\r
87 Instructions for building the clang front-end from source.\r
88 \r
89 .. __: https://clang.llvm.org/get_started.html\r
90 \r
91 :doc:`CoverageMappingFormat`\r
92 This describes the format and encoding used for LLVM’s code coverage mapping.\r
93 \r
94 :doc:`CFIVerify`\r
95 A description of the verification tool for Control Flow Integrity.\r
96 \r
97 LLVM Builds and Distributions\r
98 -----------------------------\r
99 \r
100 :doc:`BuildingADistribution`\r
101 A best-practices guide for using LLVM's CMake build system to package and\r
102 distribute LLVM-based tools.\r
103 \r
104 :doc:`CMake`\r
105 An addendum to the main Getting Started guide for those using the `CMake\r
106 build system <http://www.cmake.org>`_.\r
107 \r
108 :doc:`Docker`\r
109 A reference for using Dockerfiles provided with LLVM.\r
110 \r
111 :doc:`Support Library <SupportLibrary>`\r
112 This document describes the LLVM Support Library (``lib/Support``) and\r
113 how to keep LLVM source code portable.\r
114 \r
115 :doc:`AdvancedBuilds`\r
116 This document describes more advanced build configurations.\r
117 \r
118 Optimizations\r
119 -------------\r
120 \r
121 :doc:`WritingAnLLVMPass`\r
122 Information on how to write LLVM transformations and analyses.\r
123 \r
124 :doc:`WritingAnLLVMNewPMPass`\r
125 Information on how to write LLVM transformations under the new pass\r
126 manager.\r
127 \r
128 :doc:`Passes`\r
129 A list of optimizations and analyses implemented in LLVM.\r
130 \r
131 :doc:`StackSafetyAnalysis`\r
132 This document describes the design of the stack safety analysis of local\r
133 variables.\r
134 \r
135 :doc:`MergeFunctions`\r
136 Describes functions merging optimization.\r
137 \r
138 :doc:`AliasAnalysis`\r
139 Information on how to write a new alias analysis implementation or how to\r
140 use existing analyses.\r
141 \r
142 :doc:`MemorySSA`\r
143 Information about the MemorySSA utility in LLVM, as well as how to use it.\r
144 \r
145 :doc:`LoopTerminology`\r
146 A document describing Loops and associated terms as used in LLVM.\r
147 \r
148 :doc:`CycleTerminology`\r
149 A document describing cycles as a generalization of loops.\r
150 \r
151 :doc:`Vectorizers`\r
152 This document describes the current status of vectorization in LLVM.\r
153 \r
154 :doc:`LinkTimeOptimization`\r
155 This document describes the interface between LLVM intermodular optimizer\r
156 and the linker and its design\r
157 \r
158 :doc:`GoldPlugin`\r
159 How to build your programs with link-time optimization on Linux.\r
160 \r
161 :doc:`Remarks`\r
162 A reference on the implementation of remarks in LLVM.\r
163 \r
164 :doc:`Source Level Debugging with LLVM <SourceLevelDebugging>`\r
165 This document describes the design and philosophy behind the LLVM\r
166 source-level debugger.\r
167 \r
168 :doc:`How to Update Debug Info <HowToUpdateDebugInfo>`\r
169 This document specifies how to correctly update debug info in various kinds\r
170 of code transformations.\r
171 \r
172 :doc:`InstrRefDebugInfo`\r
173 This document explains how LLVM uses value tracking, or instruction\r
174 referencing, to determine variable locations for debug info in the final\r
175 stages of compilation.\r
176 \r
177 Code Generation\r
178 ---------------\r
179 \r
180 :doc:`WritingAnLLVMBackend`\r
181 Information on how to write LLVM backends for machine targets.\r
182 \r
183 :doc:`CodeGenerator`\r
184 The design and implementation of the LLVM code generator. Useful if you are\r
185 working on retargetting LLVM to a new architecture, designing a new codegen\r
186 pass, or enhancing existing components.\r
187 \r
188 :doc:`TableGen <TableGen/index>`\r
189 Describes the TableGen tool, which is used heavily by the LLVM code\r
190 generator.\r
191 \r
192 ===\r
193 JIT\r
194 ===\r
195 \r
196 :doc:`MCJITDesignAndImplementation`\r
197 Describes the inner workings of MCJIT execution engine.\r
198 \r
199 :doc:`ORCv2`\r
200 Describes the design and implementation of the ORC APIs, including some\r
201 usage examples, and a guide for users transitioning from ORCv1 to ORCv2.\r
202 \r
203 :doc:`JITLink`\r
204 Describes the design and APIs for the JITLink library, ORC's new JIT\r
205 linker.\r
206 \r
207 :doc:`DebuggingJITedCode`\r
208 How to debug JITed code with GDB.\r
209 \r
210 Additional Topics\r
211 -----------------\r
212 \r
213 :doc:`CommandLine`\r
214 Provides information on using the command line parsing library.\r
215 \r
216 :doc:`ExtendingLLVM`\r
217 Look here to see how to add instructions and intrinsics to LLVM.\r
218 \r
219 :doc:`AddingConstrainedIntrinsics`\r
220 Gives the steps necessary when adding a new constrained math intrinsic\r
221 to LLVM.\r
222 \r
223 :doc:`HowToBuildWindowsItaniumPrograms`\r
224 Notes on assembling a Windows Itanium environment.\r
225 \r
226 :doc:`HowToCrossCompileBuiltinsOnArm`\r
227 Notes on cross-building and testing the compiler-rt builtins for Arm.\r
228 \r
229 :doc:`BigEndianNEON`\r
230 LLVM's support for generating NEON instructions on big endian ARM targets is\r
231 somewhat nonintuitive. This document explains the implementation and rationale.\r
232 \r
233 :doc:`AArch64SME`\r
234 LLVM's support for AArch64 SME ACLE and ABI.\r
235 \r
236 :doc:`CompileCudaWithLLVM`\r
237 LLVM support for CUDA.\r
238 \r
239 :doc:`NVPTXUsage`\r
240 This document describes using the NVPTX backend to compile GPU kernels.\r
241 \r
242 :doc:`AMDGPUUsage`\r
243 This document describes using the AMDGPU backend to compile GPU kernels.\r
244 \r
245 :doc:`AMDGPUDwarfExtensionsForHeterogeneousDebugging`\r
246 This document describes DWARF extensions to support heterogeneous debugging\r
247 for targets such as the AMDGPU backend.\r
248 \r
249 :doc:`AMDGPUDwarfExtensionAllowLocationDescriptionOnTheDwarfExpressionStack/AMDGPUDwarfExtensionAllowLocationDescriptionOnTheDwarfExpressionStack`\r
250 This document describes a DWARF extension to allow location descriptions on\r
251 the DWARF expression stack. It is part of\r
252 :doc:`AMDGPUDwarfExtensionsForHeterogeneousDebugging`.\r
253 \r
254 :doc:`SPIRVUsage`\r
255 This document describes using the SPIR-V target to compile GPU kernels.\r
256 \r
257 :doc:`DirectXUsage`\r
258 This document describes using the DirectX target to compile GPU code for the\r
259 DirectX runtime.\r
260 \r
261 :doc:`RISCVUsage`\r
262 This document describes using the RISCV-V target.\r
263 \r