mirror of
https://github.com/compiler-explorer/compiler-explorer.git
synced 2025-12-27 09:23:52 -05:00
b88f255729
Previously, the "View IR Documentation" functionality for LLVM IR casting operations would not work at all, generating errors similar to the following: > There was an error fetching the documentation for this opcode (Error: Unknown opcode 'FPTRUNC'). This is because the generated documentation adds the `-to` suffix to all casting operations, causing a mismatch with the instruction's real name in the IR. Here are minimal examples for all casting instructions: - [`trunc`](https://godbolt.org/#g:!((g:!((g:!((h:codeEditor,i:(filename:'1',fontScale:14,fontUsePx:'0',j:1,lang:llvm,selection:(endColumn:1,endLineNumber:6,positionColumn:1,positionLineNumber:6,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:'define+dso_local+noundef+i8+@main()+%7B%0Aentry:%0A++%25X+%3D+trunc+i32+257+to+i8+%0A++ret+i8+%25X%0A%7D%0A'),l:'5',n:'0',o:'LLVM+IR+source+%231',t:'0')),k:35.25084847852495,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:compiler,i:(compiler:irclangtrunk,filters:(b:'0',binary:'1',binaryObject:'1',commentOnly:'0',debugCalls:'1',demangle:'0',directives:'0',execute:'1',intel:'0',libraryCode:'0',trim:'1',verboseDemangling:'0'),flagsViewOpen:'1',fontScale:14,fontUsePx:'0',j:2,lang:llvm,libs:!(),options:'',overrides:!(),selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:1),l:'5',n:'0',o:'+clang+(trunk)+(Editor+%231)',t:'0')),k:31.41581818814172,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:ir,i:('-fno-discard-value-names':'0',compilerName:'clang+(trunk)',demangle-symbols:'0',editorid:1,filter-attributes:'0',filter-comments:'0',filter-debug-info:'0',filter-instruction-metadata:'0',fontScale:14,fontUsePx:'0',j:2,selection:(endColumn:11,endLineNumber:3,positionColumn:11,positionLineNumber:3,selectionStartColumn:11,selectionStartLineNumber:3,startColumn:11,startLineNumber:3),show-optimized:'0',treeid:0,wrap:'1'),l:'5',n:'0',o:'LLVM+IR+Viewer+clang+(trunk)+(Editor+%231,+Compiler+%232)',t:'0')),k:33.33333333333333,l:'4',n:'0',o:'',s:0,t:'0')),l:'2',n:'0',o:'',t:'0')),version:4) - [`zext`](https://godbolt.org/#g:!((g:!((g:!((h:codeEditor,i:(filename:'1',fontScale:14,fontUsePx:'0',j:1,lang:llvm,selection:(endColumn:1,endLineNumber:6,positionColumn:1,positionLineNumber:6,selectionStartColumn:1,selectionStartLineNumber:6,startColumn:1,startLineNumber:6),source:'define+dso_local+noundef+i64+@main()+%7B%0Aentry:%0A++%25X+%3D+zext+i32+257+to+i64+%0A++ret+i64+%25X%0A%7D%0A'),l:'5',n:'0',o:'LLVM+IR+source+%231',t:'0')),k:35.25084847852495,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:compiler,i:(compiler:irclangtrunk,filters:(b:'0',binary:'1',binaryObject:'1',commentOnly:'0',debugCalls:'1',demangle:'0',directives:'0',execute:'1',intel:'0',libraryCode:'0',trim:'1',verboseDemangling:'0'),flagsViewOpen:'1',fontScale:14,fontUsePx:'0',j:2,lang:llvm,libs:!(),options:'',overrides:!(),selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:1),l:'5',n:'0',o:'+clang+(trunk)+(Editor+%231)',t:'0')),k:31.41581818814172,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:ir,i:('-fno-discard-value-names':'0',compilerName:'clang+(trunk)',demangle-symbols:'0',editorid:1,filter-attributes:'0',filter-comments:'0',filter-debug-info:'0',filter-instruction-metadata:'0',fontScale:14,fontUsePx:'0',j:2,selection:(endColumn:12,endLineNumber:3,positionColumn:12,positionLineNumber:3,selectionStartColumn:8,selectionStartLineNumber:3,startColumn:8,startLineNumber:3),show-optimized:'0',treeid:0,wrap:'1'),l:'5',n:'0',o:'LLVM+IR+Viewer+clang+(trunk)+(Editor+%231,+Compiler+%232)',t:'0')),k:33.33333333333333,l:'4',n:'0',o:'',s:0,t:'0')),l:'2',n:'0',o:'',t:'0')),version:4) - [`sext`](https://godbolt.org/#g:!((g:!((g:!((h:codeEditor,i:(filename:'1',fontScale:14,fontUsePx:'0',j:1,lang:llvm,selection:(endColumn:1,endLineNumber:6,positionColumn:1,positionLineNumber:6,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:'define+dso_local+noundef+i16+@main()+%7B%0Aentry:%0A++%25X+%3D+sext+i8++-1+to+i16+%0A++ret+i16+%25X%0A%7D%0A'),l:'5',n:'0',o:'LLVM+IR+source+%231',t:'0')),k:35.25084847852495,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:compiler,i:(compiler:irclangtrunk,filters:(b:'0',binary:'1',binaryObject:'1',commentOnly:'0',debugCalls:'1',demangle:'0',directives:'0',execute:'1',intel:'0',libraryCode:'0',trim:'1',verboseDemangling:'0'),flagsViewOpen:'1',fontScale:14,fontUsePx:'0',j:2,lang:llvm,libs:!(),options:'',overrides:!(),selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:1),l:'5',n:'0',o:'+clang+(trunk)+(Editor+%231)',t:'0')),k:31.41581818814172,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:ir,i:('-fno-discard-value-names':'0',compilerName:'clang+(trunk)',demangle-symbols:'0',editorid:1,filter-attributes:'0',filter-comments:'0',filter-debug-info:'0',filter-instruction-metadata:'0',fontScale:14,fontUsePx:'0',j:2,selection:(endColumn:12,endLineNumber:3,positionColumn:12,positionLineNumber:3,selectionStartColumn:8,selectionStartLineNumber:3,startColumn:8,startLineNumber:3),show-optimized:'0',treeid:0,wrap:'1'),l:'5',n:'0',o:'LLVM+IR+Viewer+clang+(trunk)+(Editor+%231,+Compiler+%232)',t:'0')),k:33.33333333333333,l:'4',n:'0',o:'',s:0,t:'0')),l:'2',n:'0',o:'',t:'0')),version:4) - [`fptrunc`](https://godbolt.org/#g:!((g:!((g:!((h:codeEditor,i:(filename:'1',fontScale:14,fontUsePx:'0',j:1,lang:llvm,selection:(endColumn:1,endLineNumber:6,positionColumn:1,positionLineNumber:6,selectionStartColumn:1,selectionStartLineNumber:6,startColumn:1,startLineNumber:6),source:'define+dso_local+noundef+float+@main()+%7B%0Aentry:%0A++%25X+%3D+fptrunc+double+16777217.0+to+float+%0A++ret+float+%25X%0A%7D%0A'),l:'5',n:'0',o:'LLVM+IR+source+%231',t:'0')),k:35.25084847852495,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:compiler,i:(compiler:irclangtrunk,filters:(b:'0',binary:'1',binaryObject:'1',commentOnly:'0',debugCalls:'1',demangle:'0',directives:'0',execute:'1',intel:'0',libraryCode:'0',trim:'1',verboseDemangling:'0'),flagsViewOpen:'1',fontScale:14,fontUsePx:'0',j:2,lang:llvm,libs:!(),options:'',overrides:!(),selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:1),l:'5',n:'0',o:'+clang+(trunk)+(Editor+%231)',t:'0')),k:31.41581818814172,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:ir,i:('-fno-discard-value-names':'0',compilerName:'clang+(trunk)',demangle-symbols:'0',editorid:1,filter-attributes:'0',filter-comments:'0',filter-debug-info:'0',filter-instruction-metadata:'0',fontScale:14,fontUsePx:'0',j:2,selection:(endColumn:15,endLineNumber:3,positionColumn:15,positionLineNumber:3,selectionStartColumn:8,selectionStartLineNumber:3,startColumn:8,startLineNumber:3),show-optimized:'0',treeid:0,wrap:'1'),l:'5',n:'0',o:'LLVM+IR+Viewer+clang+(trunk)+(Editor+%231,+Compiler+%232)',t:'0')),k:33.33333333333333,l:'4',n:'0',o:'',s:0,t:'0')),l:'2',n:'0',o:'',t:'0')),version:4) - [`fpext`](https://godbolt.org/#g:!((g:!((g:!((h:codeEditor,i:(filename:'1',fontScale:14,fontUsePx:'0',j:1,lang:llvm,selection:(endColumn:1,endLineNumber:6,positionColumn:1,positionLineNumber:6,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:'define+dso_local+noundef+double+@main()+%7B%0Aentry:%0A++%25X+%3D+fpext+float+3.125+to+double%0A++ret+double+%25X%0A%7D%0A'),l:'5',n:'0',o:'LLVM+IR+source+%231',t:'0')),k:35.25084847852495,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:compiler,i:(compiler:irclangtrunk,filters:(b:'0',binary:'1',binaryObject:'1',commentOnly:'0',debugCalls:'1',demangle:'0',directives:'0',execute:'1',intel:'0',libraryCode:'0',trim:'1',verboseDemangling:'0'),flagsViewOpen:'1',fontScale:14,fontUsePx:'0',j:2,lang:llvm,libs:!(),options:'',overrides:!(),selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:1),l:'5',n:'0',o:'+clang+(trunk)+(Editor+%231)',t:'0')),k:31.41581818814172,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:ir,i:('-fno-discard-value-names':'0',compilerName:'clang+(trunk)',demangle-symbols:'0',editorid:1,filter-attributes:'0',filter-comments:'0',filter-debug-info:'0',filter-instruction-metadata:'0',fontScale:14,fontUsePx:'0',j:2,selection:(endColumn:13,endLineNumber:3,positionColumn:13,positionLineNumber:3,selectionStartColumn:8,selectionStartLineNumber:3,startColumn:8,startLineNumber:3),show-optimized:'0',treeid:0,wrap:'1'),l:'5',n:'0',o:'LLVM+IR+Viewer+clang+(trunk)+(Editor+%231,+Compiler+%232)',t:'0')),k:33.33333333333333,l:'4',n:'0',o:'',s:0,t:'0')),l:'2',n:'0',o:'',t:'0')),version:4) - [`fptoui`](https://godbolt.org/#g:!((g:!((g:!((h:codeEditor,i:(filename:'1',fontScale:14,fontUsePx:'0',j:1,lang:llvm,selection:(endColumn:1,endLineNumber:6,positionColumn:1,positionLineNumber:6,selectionStartColumn:1,selectionStartLineNumber:6,startColumn:1,startLineNumber:6),source:'define+dso_local+noundef+i32+@main()+%7B%0Aentry:%0A++%25X+%3D+fptoui+double+123.0+to+i32%0A++ret+i32+%25X%0A%7D%0A'),l:'5',n:'0',o:'LLVM+IR+source+%231',t:'0')),k:35.25084847852495,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:compiler,i:(compiler:irclangtrunk,filters:(b:'0',binary:'1',binaryObject:'1',commentOnly:'0',debugCalls:'1',demangle:'0',directives:'0',execute:'1',intel:'0',libraryCode:'0',trim:'1',verboseDemangling:'0'),flagsViewOpen:'1',fontScale:14,fontUsePx:'0',j:2,lang:llvm,libs:!(),options:'',overrides:!(),selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:1),l:'5',n:'0',o:'+clang+(trunk)+(Editor+%231)',t:'0')),k:31.41581818814172,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:ir,i:('-fno-discard-value-names':'0',compilerName:'clang+(trunk)',demangle-symbols:'0',editorid:1,filter-attributes:'0',filter-comments:'0',filter-debug-info:'0',filter-instruction-metadata:'0',fontScale:14,fontUsePx:'0',j:2,selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),show-optimized:'0',treeid:0,wrap:'1'),l:'5',n:'0',o:'LLVM+IR+Viewer+clang+(trunk)+(Editor+%231,+Compiler+%232)',t:'0')),k:33.33333333333333,l:'4',n:'0',o:'',s:0,t:'0')),l:'2',n:'0',o:'',t:'0')),version:4) - [`fptosi`](https://godbolt.org/#g:!((g:!((g:!((h:codeEditor,i:(filename:'1',fontScale:14,fontUsePx:'0',j:1,lang:llvm,selection:(endColumn:2,endLineNumber:5,positionColumn:2,positionLineNumber:5,selectionStartColumn:2,selectionStartLineNumber:5,startColumn:2,startLineNumber:5),source:'define+dso_local+noundef+i32+@main()+%7B%0Aentry:%0A++%25X+%3D+fptosi+double+123.0+to+i32%0A++ret+i32+%25X%0A%7D%0A'),l:'5',n:'0',o:'LLVM+IR+source+%231',t:'0')),k:35.25084847852495,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:compiler,i:(compiler:irclangtrunk,filters:(b:'0',binary:'1',binaryObject:'1',commentOnly:'0',debugCalls:'1',demangle:'0',directives:'0',execute:'1',intel:'0',libraryCode:'0',trim:'1',verboseDemangling:'0'),flagsViewOpen:'1',fontScale:14,fontUsePx:'0',j:2,lang:llvm,libs:!(),options:'',overrides:!(),selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:1),l:'5',n:'0',o:'+clang+(trunk)+(Editor+%231)',t:'0')),k:31.41581818814172,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:ir,i:('-fno-discard-value-names':'0',compilerName:'clang+(trunk)',demangle-symbols:'0',editorid:1,filter-attributes:'0',filter-comments:'0',filter-debug-info:'0',filter-instruction-metadata:'0',fontScale:14,fontUsePx:'0',j:2,selection:(endColumn:14,endLineNumber:3,positionColumn:14,positionLineNumber:3,selectionStartColumn:8,selectionStartLineNumber:3,startColumn:8,startLineNumber:3),show-optimized:'0',treeid:0,wrap:'1'),l:'5',n:'0',o:'LLVM+IR+Viewer+clang+(trunk)+(Editor+%231,+Compiler+%232)',t:'0')),k:33.33333333333333,l:'4',n:'0',o:'',s:0,t:'0')),l:'2',n:'0',o:'',t:'0')),version:4) - [`uitofp`](https://godbolt.org/#g:!((g:!((g:!((h:codeEditor,i:(filename:'1',fontScale:14,fontUsePx:'0',j:1,lang:llvm,selection:(endColumn:10,endLineNumber:3,positionColumn:10,positionLineNumber:3,selectionStartColumn:10,selectionStartLineNumber:3,startColumn:10,startLineNumber:3),source:'define+dso_local+noundef+float+@main()+%7B%0Aentry:%0A++%25X+%3D+uitofp+i32+257+to+float%0A++ret+float+%25X%0A%7D%0A'),l:'5',n:'0',o:'LLVM+IR+source+%231',t:'0')),k:35.25084847852495,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:compiler,i:(compiler:irclangtrunk,filters:(b:'0',binary:'1',binaryObject:'1',commentOnly:'0',debugCalls:'1',demangle:'0',directives:'0',execute:'1',intel:'0',libraryCode:'0',trim:'1',verboseDemangling:'0'),flagsViewOpen:'1',fontScale:14,fontUsePx:'0',j:2,lang:llvm,libs:!(),options:'',overrides:!(),selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:1),l:'5',n:'0',o:'+clang+(trunk)+(Editor+%231)',t:'0')),k:31.41581818814172,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:ir,i:('-fno-discard-value-names':'0',compilerName:'clang+(trunk)',demangle-symbols:'0',editorid:1,filter-attributes:'0',filter-comments:'0',filter-debug-info:'0',filter-instruction-metadata:'0',fontScale:14,fontUsePx:'0',j:2,selection:(endColumn:14,endLineNumber:3,positionColumn:14,positionLineNumber:3,selectionStartColumn:8,selectionStartLineNumber:3,startColumn:8,startLineNumber:3),show-optimized:'0',treeid:0,wrap:'1'),l:'5',n:'0',o:'LLVM+IR+Viewer+clang+(trunk)+(Editor+%231,+Compiler+%232)',t:'0')),k:33.33333333333333,l:'4',n:'0',o:'',s:0,t:'0')),l:'2',n:'0',o:'',t:'0')),version:4) - [`sitofp`](https://godbolt.org/#g:!((g:!((g:!((h:codeEditor,i:(filename:'1',fontScale:14,fontUsePx:'0',j:1,lang:llvm,selection:(endColumn:15,endLineNumber:4,positionColumn:15,positionLineNumber:4,selectionStartColumn:15,selectionStartLineNumber:4,startColumn:15,startLineNumber:4),source:'define+dso_local+noundef+float+@main()+%7B%0Aentry:%0A++%25X+%3D+sitofp+i32+257+to+float%0A++ret+float+%25X%0A%7D%0A'),l:'5',n:'0',o:'LLVM+IR+source+%231',t:'0')),k:35.25084847852495,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:compiler,i:(compiler:irclangtrunk,filters:(b:'0',binary:'1',binaryObject:'1',commentOnly:'0',debugCalls:'1',demangle:'0',directives:'0',execute:'1',intel:'0',libraryCode:'0',trim:'1',verboseDemangling:'0'),flagsViewOpen:'1',fontScale:14,fontUsePx:'0',j:2,lang:llvm,libs:!(),options:'',overrides:!(),selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:1),l:'5',n:'0',o:'+clang+(trunk)+(Editor+%231)',t:'0')),k:31.41581818814172,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:ir,i:('-fno-discard-value-names':'0',compilerName:'clang+(trunk)',demangle-symbols:'0',editorid:1,filter-attributes:'0',filter-comments:'0',filter-debug-info:'0',filter-instruction-metadata:'0',fontScale:14,fontUsePx:'0',j:2,selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),show-optimized:'0',treeid:0,wrap:'1'),l:'5',n:'0',o:'LLVM+IR+Viewer+clang+(trunk)+(Editor+%231,+Compiler+%232)',t:'0')),k:33.33333333333333,l:'4',n:'0',o:'',s:0,t:'0')),l:'2',n:'0',o:'',t:'0')),version:4) - [`ptrtoint`](https://godbolt.org/#g:!((g:!((g:!((h:codeEditor,i:(filename:'1',fontScale:14,fontUsePx:'0',j:1,lang:llvm,selection:(endColumn:1,endLineNumber:7,positionColumn:1,positionLineNumber:7,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:'define+dso_local+noundef+i64+@main()+%7B%0Aentry:%0A++%25P+%3D+alloca+i32%0A++%25X+%3D+ptrtoint+ptr+%25P+to+i64%0A++ret+i64+%25X%0A%7D%0A'),l:'5',n:'0',o:'LLVM+IR+source+%231',t:'0')),k:35.25084847852495,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:compiler,i:(compiler:irclangtrunk,filters:(b:'0',binary:'1',binaryObject:'1',commentOnly:'0',debugCalls:'1',demangle:'0',directives:'0',execute:'1',intel:'0',libraryCode:'0',trim:'1',verboseDemangling:'0'),flagsViewOpen:'1',fontScale:14,fontUsePx:'0',j:2,lang:llvm,libs:!(),options:'',overrides:!(),selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:1),l:'5',n:'0',o:'+clang+(trunk)+(Editor+%231)',t:'0')),k:31.41581818814172,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:ir,i:('-fno-discard-value-names':'0',compilerName:'clang+(trunk)',demangle-symbols:'0',editorid:1,filter-attributes:'0',filter-comments:'0',filter-debug-info:'0',filter-instruction-metadata:'0',fontScale:14,fontUsePx:'0',j:2,selection:(endColumn:16,endLineNumber:4,positionColumn:16,positionLineNumber:4,selectionStartColumn:8,selectionStartLineNumber:4,startColumn:8,startLineNumber:4),show-optimized:'0',treeid:0,wrap:'1'),l:'5',n:'0',o:'LLVM+IR+Viewer+clang+(trunk)+(Editor+%231,+Compiler+%232)',t:'0')),k:33.33333333333333,l:'4',n:'0',o:'',s:0,t:'0')),l:'2',n:'0',o:'',t:'0')),version:4) - [`ptrtoaddr`](https://godbolt.org/#g:!((g:!((g:!((h:codeEditor,i:(filename:'1',fontScale:14,fontUsePx:'0',j:1,lang:llvm,selection:(endColumn:17,endLineNumber:4,positionColumn:17,positionLineNumber:4,selectionStartColumn:17,selectionStartLineNumber:4,startColumn:17,startLineNumber:4),source:'define+dso_local+noundef+i64+@main()+%7B%0Aentry:%0A++%25P+%3D+alloca+i32%0A++%25X+%3D+ptrtoaddr+ptr+%25P+to+i64%0A++ret+i64+%25X%0A%7D%0A'),l:'5',n:'0',o:'LLVM+IR+source+%231',t:'0')),k:35.25084847852495,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:compiler,i:(compiler:irclangtrunk,filters:(b:'0',binary:'1',binaryObject:'1',commentOnly:'0',debugCalls:'1',demangle:'0',directives:'0',execute:'1',intel:'0',libraryCode:'0',trim:'1',verboseDemangling:'0'),flagsViewOpen:'1',fontScale:14,fontUsePx:'0',j:2,lang:llvm,libs:!(),options:'',overrides:!(),selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:1),l:'5',n:'0',o:'+clang+(trunk)+(Editor+%231)',t:'0')),k:31.41581818814172,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:ir,i:('-fno-discard-value-names':'0',compilerName:'clang+(trunk)',demangle-symbols:'0',editorid:1,filter-attributes:'0',filter-comments:'0',filter-debug-info:'0',filter-instruction-metadata:'0',fontScale:14,fontUsePx:'0',j:2,selection:(endColumn:2,endLineNumber:6,positionColumn:2,positionLineNumber:6,selectionStartColumn:2,selectionStartLineNumber:6,startColumn:2,startLineNumber:6),show-optimized:'0',treeid:0,wrap:'1'),l:'5',n:'0',o:'LLVM+IR+Viewer+clang+(trunk)+(Editor+%231,+Compiler+%232)',t:'0')),k:33.33333333333333,l:'4',n:'0',o:'',s:0,t:'0')),l:'2',n:'0',o:'',t:'0')),version:4) - [`inttoptr`](https://godbolt.org/#g:!((g:!((g:!((h:codeEditor,i:(filename:'1',fontScale:14,fontUsePx:'0',j:1,lang:llvm,selection:(endColumn:13,endLineNumber:4,positionColumn:13,positionLineNumber:4,selectionStartColumn:13,selectionStartLineNumber:4,startColumn:13,startLineNumber:4),source:'define+dso_local+noundef+ptr+@main()+%7B%0Aentry:%0A++%25X+%3D+inttoptr+i32+255+to+ptr%0A++ret+ptr+%25X%0A%7D%0A'),l:'5',n:'0',o:'LLVM+IR+source+%231',t:'0')),k:35.25084847852495,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:compiler,i:(compiler:irclangtrunk,filters:(b:'0',binary:'1',binaryObject:'1',commentOnly:'0',debugCalls:'1',demangle:'0',directives:'0',execute:'1',intel:'0',libraryCode:'0',trim:'1',verboseDemangling:'0'),flagsViewOpen:'1',fontScale:14,fontUsePx:'0',j:2,lang:llvm,libs:!(),options:'',overrides:!(),selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:1),l:'5',n:'0',o:'+clang+(trunk)+(Editor+%231)',t:'0')),k:31.41581818814172,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:ir,i:('-fno-discard-value-names':'0',compilerName:'clang+(trunk)',demangle-symbols:'0',editorid:1,filter-attributes:'0',filter-comments:'0',filter-debug-info:'0',filter-instruction-metadata:'0',fontScale:14,fontUsePx:'0',j:2,selection:(endColumn:16,endLineNumber:3,positionColumn:16,positionLineNumber:3,selectionStartColumn:8,selectionStartLineNumber:3,startColumn:8,startLineNumber:3),show-optimized:'0',treeid:0,wrap:'1'),l:'5',n:'0',o:'LLVM+IR+Viewer+clang+(trunk)+(Editor+%231,+Compiler+%232)',t:'0')),k:33.33333333333333,l:'4',n:'0',o:'',s:0,t:'0')),l:'2',n:'0',o:'',t:'0')),version:4) - [`bitcast`](https://godbolt.org/#g:!((g:!((g:!((h:codeEditor,i:(filename:'1',fontScale:14,fontUsePx:'0',j:1,lang:llvm,selection:(endColumn:28,endLineNumber:3,positionColumn:28,positionLineNumber:3,selectionStartColumn:28,selectionStartLineNumber:3,startColumn:28,startLineNumber:3),source:'define+dso_local+noundef+i8+@main()+%7B%0Aentry:%0A++%25X+%3D+bitcast+i8+255+to+i8%0A++ret+i8+%25X%0A%7D%0A'),l:'5',n:'0',o:'LLVM+IR+source+%231',t:'0')),k:35.25084847852495,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:compiler,i:(compiler:irclangtrunk,filters:(b:'0',binary:'1',binaryObject:'1',commentOnly:'0',debugCalls:'1',demangle:'0',directives:'0',execute:'1',intel:'0',libraryCode:'0',trim:'1',verboseDemangling:'0'),flagsViewOpen:'1',fontScale:14,fontUsePx:'0',j:2,lang:llvm,libs:!(),options:'',overrides:!(),selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:1),l:'5',n:'0',o:'+clang+(trunk)+(Editor+%231)',t:'0')),k:31.41581818814172,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:ir,i:('-fno-discard-value-names':'0',compilerName:'clang+(trunk)',demangle-symbols:'0',editorid:1,filter-attributes:'0',filter-comments:'0',filter-debug-info:'0',filter-instruction-metadata:'0',fontScale:14,fontUsePx:'0',j:2,selection:(endColumn:15,endLineNumber:3,positionColumn:15,positionLineNumber:3,selectionStartColumn:8,selectionStartLineNumber:3,startColumn:8,startLineNumber:3),show-optimized:'0',treeid:0,wrap:'1'),l:'5',n:'0',o:'LLVM+IR+Viewer+clang+(trunk)+(Editor+%231,+Compiler+%232)',t:'0')),k:33.33333333333333,l:'4',n:'0',o:'',s:0,t:'0')),l:'2',n:'0',o:'',t:'0')),version:4) - [`addrspacecast`](https://godbolt.org/#g:!((g:!((g:!((h:codeEditor,i:(filename:'1',fontScale:14,fontUsePx:'0',j:1,lang:llvm,selection:(endColumn:7,endLineNumber:2,positionColumn:7,positionLineNumber:2,selectionStartColumn:7,selectionStartLineNumber:2,startColumn:7,startLineNumber:2),source:'define+dso_local+noundef+ptr+addrspace(1)+@main()+%7B%0Aentry:%0A++%25P+%3D+alloca+i32%0A++%25X+%3D+addrspacecast+ptr+%25P+to+ptr+addrspace(1)%0A++ret+ptr+addrspace(1)+%25X%0A%7D%0A'),l:'5',n:'0',o:'LLVM+IR+source+%231',t:'0')),k:35.25084847852495,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:compiler,i:(compiler:irclangtrunk,filters:(b:'0',binary:'1',binaryObject:'1',commentOnly:'0',debugCalls:'1',demangle:'0',directives:'0',execute:'1',intel:'0',libraryCode:'0',trim:'1',verboseDemangling:'0'),flagsViewOpen:'1',fontScale:14,fontUsePx:'0',j:2,lang:llvm,libs:!(),options:'',overrides:!(),selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:1),l:'5',n:'0',o:'+clang+(trunk)+(Editor+%231)',t:'0')),k:31.41581818814172,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:ir,i:('-fno-discard-value-names':'0',compilerName:'clang+(trunk)',demangle-symbols:'0',editorid:1,filter-attributes:'0',filter-comments:'0',filter-debug-info:'0',filter-instruction-metadata:'0',fontScale:14,fontUsePx:'0',j:2,selection:(endColumn:21,endLineNumber:4,positionColumn:21,positionLineNumber:4,selectionStartColumn:8,selectionStartLineNumber:4,startColumn:8,startLineNumber:4),show-optimized:'0',treeid:0,wrap:'1'),l:'5',n:'0',o:'LLVM+IR+Viewer+clang+(trunk)+(Editor+%231,+Compiler+%232)',t:'0')),k:33.33333333333333,l:'4',n:'0',o:'',s:0,t:'0')),l:'2',n:'0',o:'',t:'0')),version:4) The suggested fix is to: - Remove the trailing `-to` from any instruction parsed from the LLVM IR docs, as it is _probably_ a casting operation. - Add `ptrtoaddr` to the `llvmStatements` list, as it didn't seem to be recognized as an operation otherwise (and so failed the precondition to have a 'View IR Documentation' link). The diff for the generated documentation may seem larger than expected, that's because the IDs of the embedded HTML seem to have changed since last time.
404 lines
403 KiB
TypeScript
Generated
404 lines
403 KiB
TypeScript
Generated
import type {AssemblyInstructionInfo} from '../../../types/assembly-docs.interfaces.js';
|
||
|
||
export function getAsmOpcode(opcode: string | undefined): AssemblyInstructionInfo | undefined {
|
||
if (!opcode) return;
|
||
switch (opcode.toUpperCase()) {
|
||
case 'RET':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#ret-instruction`,
|
||
html: `<html><head></head><body><span id="i-ret"></span><h4><a class="toc-backref" href="#id2156" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">ret</span></code>’ Instruction</a></h4><section id="id32"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">ret</span> <span class="o"><</span><span class="nb">type</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="p">;</span> <span class="n">Return</span> <span class="n">a</span> <span class="n">value</span> <span class="kn">from</span> <span class="nn">a</span> <span class="n">non</span><span class="o">-</span><span class="n">void</span> <span class="n">function</span><span class="n">ret</span> <span class="n">void</span> <span class="p">;</span> <span class="n">Return</span> <span class="kn">from</span> <span class="nn">void</span> <span class="n">function</span></pre></div></div></section><section id="overview"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">ret</span></code>’ instruction is used to return control flow (and optionallya value) from a function back to the caller.</p><p>There are two forms of the ‘<code class="docutils literal notranslate"><span class="pre">ret</span></code>’ instruction: one that returns avalue and then causes control flow, and one that just causes controlflow to occur.</p></section><section id="arguments"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">ret</span></code>’ instruction optionally accepts a single argument, thereturn value. The type of the return value must be a ‘<a class="reference internal" href="#t-firstclass"><span class="std std-ref">firstclass</span></a>’ type.</p><p>A function is not <a class="reference internal" href="#wellformed"><span class="std std-ref">well formed</span></a> if it has a non-voidreturn type and contains a ‘<code class="docutils literal notranslate"><span class="pre">ret</span></code>’ instruction with no return value ora return value with a type that does not match its type, or if it has avoid return type and contains a ‘<code class="docutils literal notranslate"><span class="pre">ret</span></code>’ instruction with a returnvalue.</p></section><section id="id33"><h5>Semantics:</h5><p>When the ‘<code class="docutils literal notranslate"><span class="pre">ret</span></code>’ instruction is executed, control flow returns back tothe calling function’s context. If the caller is a“<a class="reference internal" href="#i-call"><span class="std std-ref">call</span></a>” instruction, execution continues at theinstruction after the call. If the caller was an“<a class="reference internal" href="#i-invoke"><span class="std std-ref">invoke</span></a>” instruction, execution continues at thebeginning of the “normal” destination block. If the instruction returnsa value, that value shall set the call or invoke instruction’s returnvalue.</p></section><section id="example"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="k">ret</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">5</span><span class="w"> </span><span class="c">; Return an integer value of 5</span><span class="k">ret</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="c">; Return from a void function</span><span class="k">ret</span><span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="kt">i32</span><span class="p">,</span><span class="w"> </span><span class="kt">i8</span><span class="w"> </span><span class="p">}</span><span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">4</span><span class="p">,</span><span class="w"> </span><span class="kt">i8</span><span class="w"> </span><span class="m">2</span><span class="w"> </span><span class="p">}</span><span class="w"> </span><span class="c">; Return a struct of values 4 and 2</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘ret’ instruction is used to return control flow (and optionallya value) from a function back to the caller.There are two forms of the ‘ret’ instruction: one that returns avalue and then causes control flow, and one that just causes controlflow to occur.`,
|
||
};
|
||
case 'BR':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#br-instruction`,
|
||
html: `<html><head></head><body><span id="i-br"></span><h4><a class="toc-backref" href="#id2157" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">br</span></code>’ Instruction</a></h4><section id="id34"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">br</span> <span class="n">i1</span> <span class="o"><</span><span class="n">cond</span><span class="o">></span><span class="p">,</span> <span class="n">label</span> <span class="o"><</span><span class="n">iftrue</span><span class="o">></span><span class="p">,</span> <span class="n">label</span> <span class="o"><</span><span class="n">iffalse</span><span class="o">></span><span class="n">br</span> <span class="n">label</span> <span class="o"><</span><span class="n">dest</span><span class="o">></span> <span class="p">;</span> <span class="n">Unconditional</span> <span class="n">branch</span></pre></div></div></section><section id="id35"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">br</span></code>’ instruction is used to cause control flow to transfer to adifferent basic block in the current function. There are two forms ofthis instruction, corresponding to a conditional branch and anunconditional branch.</p></section><section id="id36"><h5>Arguments:</h5><p>The conditional branch form of the ‘<code class="docutils literal notranslate"><span class="pre">br</span></code>’ instruction takes a single‘<code class="docutils literal notranslate"><span class="pre">i1</span></code>’ value and two ‘<code class="docutils literal notranslate"><span class="pre">label</span></code>’ values. The unconditional form of the‘<code class="docutils literal notranslate"><span class="pre">br</span></code>’ instruction takes a single ‘<code class="docutils literal notranslate"><span class="pre">label</span></code>’ value as a target.</p></section><section id="id37"><h5>Semantics:</h5><p>Upon execution of a conditional ‘<code class="docutils literal notranslate"><span class="pre">br</span></code>’ instruction, the ‘<code class="docutils literal notranslate"><span class="pre">i1</span></code>’argument is evaluated. If the value is <code class="docutils literal notranslate"><span class="pre">true</span></code>, control flows to the‘<code class="docutils literal notranslate"><span class="pre">iftrue</span></code>’ <code class="docutils literal notranslate"><span class="pre">label</span></code> argument. If “cond” is <code class="docutils literal notranslate"><span class="pre">false</span></code>, control flowsto the ‘<code class="docutils literal notranslate"><span class="pre">iffalse</span></code>’ <code class="docutils literal notranslate"><span class="pre">label</span></code> argument.If ‘<code class="docutils literal notranslate"><span class="pre">cond</span></code>’ is <code class="docutils literal notranslate"><span class="pre">poison</span></code> or <code class="docutils literal notranslate"><span class="pre">undef</span></code>, this instruction has undefinedbehavior.</p></section><section id="id38"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nl">Test:</span><span class="w"> </span><span class="nv">%cond</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">icmp</span><span class="w"> </span><span class="k">eq</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="nv">%a</span><span class="p">,</span><span class="w"> </span><span class="nv">%b</span><span class="w"> </span><span class="k">br</span><span class="w"> </span><span class="kt">i1</span><span class="w"> </span><span class="nv">%cond</span><span class="p">,</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%IfEqual</span><span class="p">,</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%IfUnequal</span><span class="nl">IfEqual:</span><span class="w"> </span><span class="k">ret</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">1</span><span class="nl">IfUnequal:</span><span class="w"> </span><span class="k">ret</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">0</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘br’ instruction is used to cause control flow to transfer to adifferent basic block in the current function. There are two forms ofthis instruction, corresponding to a conditional branch and anunconditional branch.`,
|
||
};
|
||
case 'SWITCH':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#switch-instruction`,
|
||
html: `<html><head></head><body><span id="i-switch"></span><h4><a class="toc-backref" href="#id2158" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">switch</span></code>’ Instruction</a></h4><section id="id39"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">switch</span> <span class="o"><</span><span class="n">intty</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span><span class="p">,</span> <span class="n">label</span> <span class="o"><</span><span class="n">defaultdest</span><span class="o">></span> <span class="p">[</span> <span class="o"><</span><span class="n">intty</span><span class="o">></span> <span class="o"><</span><span class="n">val</span><span class="o">></span><span class="p">,</span> <span class="n">label</span> <span class="o"><</span><span class="n">dest</span><span class="o">></span> <span class="o">...</span> <span class="p">]</span></pre></div></div></section><section id="id40"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">switch</span></code>’ instruction is used to transfer control flow to one ofseveral different places. It is a generalization of the ‘<code class="docutils literal notranslate"><span class="pre">br</span></code>’instruction, allowing a branch to occur to one of many possibledestinations.</p></section><section id="id41"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">switch</span></code>’ instruction uses three parameters: an integercomparison value ‘<code class="docutils literal notranslate"><span class="pre">value</span></code>’, a default ‘<code class="docutils literal notranslate"><span class="pre">label</span></code>’ destination, and anarray of pairs of comparison value constants and ‘<code class="docutils literal notranslate"><span class="pre">label</span></code>’s. The tableis not allowed to contain duplicate constant entries.</p></section><section id="id42"><h5>Semantics:</h5><p>The <code class="docutils literal notranslate"><span class="pre">switch</span></code> instruction specifies a table of values and destinations.When the ‘<code class="docutils literal notranslate"><span class="pre">switch</span></code>’ instruction is executed, this table is searchedfor the given value. If the value is found, control flow is transferredto the corresponding destination; otherwise, control flow is transferredto the default destination.If ‘<code class="docutils literal notranslate"><span class="pre">value</span></code>’ is <code class="docutils literal notranslate"><span class="pre">poison</span></code> or <code class="docutils literal notranslate"><span class="pre">undef</span></code>, this instruction has undefinedbehavior.</p></section><section id="implementation"><h5>Implementation:</h5><p>Depending on properties of the target machine and the particular<code class="docutils literal notranslate"><span class="pre">switch</span></code> instruction, this instruction may be code generated indifferent ways. For example, it could be generated as a series ofchained conditional branches or with a lookup table.</p></section><section id="id43"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="c">; Emulate a conditional br instruction</span><span class="nv">%Val</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">zext</span><span class="w"> </span><span class="kt">i1</span><span class="w"> </span><span class="nv">%value</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i32</span><span class="k">switch</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="nv">%Val</span><span class="p">,</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%truedest</span><span class="w"> </span><span class="p">[</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">0</span><span class="p">,</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%falsedest</span><span class="w"> </span><span class="p">]</span><span class="c">; Emulate an unconditional br instruction</span><span class="k">switch</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">0</span><span class="p">,</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%dest</span><span class="w"> </span><span class="p">[</span><span class="w"> </span><span class="p">]</span><span class="c">; Implement a jump table:</span><span class="k">switch</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="nv">%val</span><span class="p">,</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%otherwise</span><span class="w"> </span><span class="p">[</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">0</span><span class="p">,</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%onzero</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">1</span><span class="p">,</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%onone</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">2</span><span class="p">,</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%ontwo</span><span class="w"> </span><span class="p">]</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘switch’ instruction is used to transfer control flow to one ofseveral different places. It is a generalization of the ‘br’instruction, allowing a branch to occur to one of many possibledestinations.`,
|
||
};
|
||
case 'INDIRECTBR':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#indirectbr-instruction`,
|
||
html: `<html><head></head><body><span id="i-indirectbr"></span><h4><a class="toc-backref" href="#id2159" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">indirectbr</span></code>’ Instruction</a></h4><section id="id44"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">indirectbr</span> <span class="n">ptr</span> <span class="o"><</span><span class="n">address</span><span class="o">></span><span class="p">,</span> <span class="p">[</span> <span class="n">label</span> <span class="o"><</span><span class="n">dest1</span><span class="o">></span><span class="p">,</span> <span class="n">label</span> <span class="o"><</span><span class="n">dest2</span><span class="o">></span><span class="p">,</span> <span class="o">...</span> <span class="p">]</span></pre></div></div></section><section id="id45"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">indirectbr</span></code>’ instruction implements an indirect branch to alabel within the current function, whose address is specified by“<code class="docutils literal notranslate"><span class="pre">address</span></code>”. Address must be derived from a<a class="reference internal" href="#blockaddress"><span class="std std-ref">blockaddress</span></a> constant.</p></section><section id="id46"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">address</span></code>’ argument is the address of the label to jump to. Therest of the arguments indicate the full set of possible destinationsthat the address may point to. Blocks are allowed to occur multipletimes in the destination list, though this isn’t particularly useful.</p><p>This destination list is required so that dataflow analysis has anaccurate understanding of the CFG.</p></section><section id="id47"><h5>Semantics:</h5><p>Control transfers to the block specified in the address argument. Allpossible destination blocks must be listed in the label list, otherwisethis instruction has undefined behavior. This implies that jumps tolabels defined in other functions have undefined behavior as well.If ‘<code class="docutils literal notranslate"><span class="pre">address</span></code>’ is <code class="docutils literal notranslate"><span class="pre">poison</span></code> or <code class="docutils literal notranslate"><span class="pre">undef</span></code>, this instruction has undefinedbehavior.</p></section><section id="id48"><h5>Implementation:</h5><p>This is typically implemented with a jump through a register.</p></section><section id="id49"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="k">indirectbr</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%Addr</span><span class="p">,</span><span class="w"> </span><span class="p">[</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%bb1</span><span class="p">,</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%bb2</span><span class="p">,</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%bb3</span><span class="w"> </span><span class="p">]</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘indirectbr’ instruction implements an indirect branch to alabel within the current function, whose address is specified by“address”. Address must be derived from ablockaddress constant.`,
|
||
};
|
||
case 'INVOKE':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#invoke-instruction`,
|
||
html: `<html><head></head><body><span id="i-invoke"></span><h4><a class="toc-backref" href="#id2160" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">invoke</span></code>’ Instruction</a></h4><section id="id50"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">invoke</span> <span class="p">[</span><span class="n">cconv</span><span class="p">]</span> <span class="p">[</span><span class="n">ret</span> <span class="n">attrs</span><span class="p">]</span> <span class="p">[</span><span class="n">addrspace</span><span class="p">(</span><span class="o"><</span><span class="n">num</span><span class="o">></span><span class="p">)]</span> <span class="o"><</span><span class="n">ty</span><span class="o">>|<</span><span class="n">fnty</span><span class="o">></span> <span class="o"><</span><span class="n">fnptrval</span><span class="o">></span><span class="p">(</span><span class="o"><</span><span class="n">function</span> <span class="n">args</span><span class="o">></span><span class="p">)</span> <span class="p">[</span><span class="n">fn</span> <span class="n">attrs</span><span class="p">]</span> <span class="p">[</span><span class="n">operand</span> <span class="n">bundles</span><span class="p">]</span> <span class="n">to</span> <span class="n">label</span> <span class="o"><</span><span class="n">normal</span> <span class="n">label</span><span class="o">></span> <span class="n">unwind</span> <span class="n">label</span> <span class="o"><</span><span class="n">exception</span> <span class="n">label</span><span class="o">></span></pre></div></div></section><section id="id51"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">invoke</span></code>’ instruction causes control to transfer to a specifiedfunction, with the possibility of control flow transfer to either the‘<code class="docutils literal notranslate"><span class="pre">normal</span></code>’ label or the ‘<code class="docutils literal notranslate"><span class="pre">exception</span></code>’ label. If the callee functionreturns with the “<code class="docutils literal notranslate"><span class="pre">ret</span></code>” instruction, control flow will return to the“normal” label. If the callee (or any indirect callees) returns via the“<a class="reference internal" href="#i-resume"><span class="std std-ref">resume</span></a>” instruction or other exception handlingmechanism, control is interrupted and continued at the dynamicallynearest “exception” label.</p><p>The ‘<code class="docutils literal notranslate"><span class="pre">exception</span></code>’ label is a <a class="reference external" href="ExceptionHandling.html#overview">landingpad</a> for the exception. As such,‘<code class="docutils literal notranslate"><span class="pre">exception</span></code>’ label is required to have the“<a class="reference internal" href="#i-landingpad"><span class="std std-ref">landingpad</span></a>” instruction, which contains theinformation about the behavior of the program after unwinding happens,as its first non-PHI instruction. The restrictions on the“<code class="docutils literal notranslate"><span class="pre">landingpad</span></code>” instruction’s tightly couples it to the “<code class="docutils literal notranslate"><span class="pre">invoke</span></code>”instruction, so that the important information contained within the“<code class="docutils literal notranslate"><span class="pre">landingpad</span></code>” instruction can’t be lost through normal code motion.</p></section><section id="id52"><h5>Arguments:</h5><p>This instruction requires several arguments:</p><ol class="arabic simple"><li><p>The optional “cconv” marker indicates which <a class="reference internal" href="#callingconv"><span class="std std-ref">callingconvention</span></a> the call should use. If none isspecified, the call defaults to using C calling conventions.</p></li><li><p>The optional <a class="reference internal" href="#paramattrs"><span class="std std-ref">Parameter Attributes</span></a> list for returnvalues. Only ‘<code class="docutils literal notranslate"><span class="pre">zeroext</span></code>’, ‘<code class="docutils literal notranslate"><span class="pre">signext</span></code>’, ‘<code class="docutils literal notranslate"><span class="pre">noext</span></code>’, and ‘<code class="docutils literal notranslate"><span class="pre">inreg</span></code>’attributes are valid here.</p></li><li><p>The optional addrspace attribute can be used to indicate the address spaceof the called function. If it is not specified, the program address spacefrom the <a class="reference internal" href="#langref-datalayout"><span class="std std-ref">datalayout string</span></a> will be used.</p></li><li><p>‘<code class="docutils literal notranslate"><span class="pre">ty</span></code>’: the type of the call instruction itself which is also thetype of the return value. Functions that return no value are marked<code class="docutils literal notranslate"><span class="pre">void</span></code>. The signature is computed based on the return type and argumenttypes.</p></li><li><p>‘<code class="docutils literal notranslate"><span class="pre">fnty</span></code>’: shall be the signature of the function being invoked. Theargument types must match the types implied by this signature. Thisis only required if the signature specifies a varargs type.</p></li><li><p>‘<code class="docutils literal notranslate"><span class="pre">fnptrval</span></code>’: An LLVM value containing a pointer to a function tobe invoked. In most cases, this is a direct function invocation, butindirect <code class="docutils literal notranslate"><span class="pre">invoke</span></code>’s are just as possible, calling an arbitrary pointerto function value.</p></li><li><p>‘<code class="docutils literal notranslate"><span class="pre">function</span> <span class="pre">args</span></code>’: argument list whose types match the functionsignature argument types and parameter attributes. All arguments mustbe of <a class="reference internal" href="#t-firstclass"><span class="std std-ref">first class</span></a> type. If the function signatureindicates the function accepts a variable number of arguments, theextra arguments can be specified.</p></li><li><p>‘<code class="docutils literal notranslate"><span class="pre">normal</span> <span class="pre">label</span></code>’: the label reached when the called functionexecutes a ‘<code class="docutils literal notranslate"><span class="pre">ret</span></code>’ instruction.</p></li><li><p>‘<code class="docutils literal notranslate"><span class="pre">exception</span> <span class="pre">label</span></code>’: the label reached when a callee returns viathe <a class="reference internal" href="#i-resume"><span class="std std-ref">resume</span></a> instruction or other exception handlingmechanism.</p></li><li><p>The optional <a class="reference internal" href="#fnattrs"><span class="std std-ref">function attributes</span></a> list.</p></li><li><p>The optional <a class="reference internal" href="#opbundles"><span class="std std-ref">operand bundles</span></a> list.</p></li></ol></section><section id="id53"><h5>Semantics:</h5><p>This instruction is designed to operate as a standard ‘<code class="docutils literal notranslate"><span class="pre">call</span></code>’instruction in most regards. The primary difference is that itestablishes an association with a label, which is used by the runtimelibrary to unwind the stack.</p><p>This instruction is used in languages with destructors to ensure thatproper cleanup is performed in the case of either a <code class="docutils literal notranslate"><span class="pre">longjmp</span></code> or athrown exception. Additionally, this is important for implementation of‘<code class="docutils literal notranslate"><span class="pre">catch</span></code>’ clauses in high-level languages that support them.</p><p>For the purposes of the SSA form, the definition of the value returnedby the ‘<code class="docutils literal notranslate"><span class="pre">invoke</span></code>’ instruction is deemed to occur on the edge from thecurrent block to the “normal” label. If the callee unwinds then noreturn value is available.</p></section><section id="id54"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%retval</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">invoke</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="vg">@Test</span><span class="p">(</span><span class="kt">i32</span><span class="w"> </span><span class="m">15</span><span class="p">)</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%Continue</span><span class="w"> </span><span class="k">unwind</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%TestCleanup</span><span class="w"> </span><span class="c">; i32:retval set</span><span class="nv">%retval</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">invoke</span><span class="w"> </span><span class="k">coldcc</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="nv">%Testfnptr</span><span class="p">(</span><span class="kt">i32</span><span class="w"> </span><span class="m">15</span><span class="p">)</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%Continue</span><span class="w"> </span><span class="k">unwind</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%TestCleanup</span><span class="w"> </span><span class="c">; i32:retval set</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘invoke’ instruction causes control to transfer to a specifiedfunction, with the possibility of control flow transfer to either the‘normal’ label or the ‘exception’ label. If the callee functionreturns with the “ret” instruction, control flow will return to the“normal” label. If the callee (or any indirect callees) returns via the“resume” instruction or other exception handlingmechanism, control is interrupted and continued at the dynamicallynearest “exception” label.The ‘exception’ label is a landingpad for the exception. As such,‘exception’ label is required to have the“landingpad” instruction, which contains theinformation about the behavior of the program after unwinding happens,as its first non-PHI instruction. The restrictions on the“landingpad” instruction’s tightly couples it to the “invoke”instruction, so that the important information contained within the“landingpad” instruction can’t be lost through normal code motion.`,
|
||
};
|
||
case 'CALLBR':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#callbr-instruction`,
|
||
html: `<html><head></head><body><span id="i-callbr"></span><h4><a class="toc-backref" href="#id2161" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">callbr</span></code>’ Instruction</a></h4><section id="id55"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">callbr</span> <span class="p">[</span><span class="n">cconv</span><span class="p">]</span> <span class="p">[</span><span class="n">ret</span> <span class="n">attrs</span><span class="p">]</span> <span class="p">[</span><span class="n">addrspace</span><span class="p">(</span><span class="o"><</span><span class="n">num</span><span class="o">></span><span class="p">)]</span> <span class="o"><</span><span class="n">ty</span><span class="o">>|<</span><span class="n">fnty</span><span class="o">></span> <span class="o"><</span><span class="n">fnptrval</span><span class="o">></span><span class="p">(</span><span class="o"><</span><span class="n">function</span> <span class="n">args</span><span class="o">></span><span class="p">)</span> <span class="p">[</span><span class="n">fn</span> <span class="n">attrs</span><span class="p">]</span> <span class="p">[</span><span class="n">operand</span> <span class="n">bundles</span><span class="p">]</span> <span class="n">to</span> <span class="n">label</span> <span class="o"><</span><span class="n">fallthrough</span> <span class="n">label</span><span class="o">></span> <span class="p">[</span><span class="n">indirect</span> <span class="n">labels</span><span class="p">]</span></pre></div></div></section><section id="id56"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">callbr</span></code>’ instruction causes control to transfer to a specifiedfunction, with the possibility of control flow transfer to either the‘<code class="docutils literal notranslate"><span class="pre">fallthrough</span></code>’ label or one of the ‘<code class="docutils literal notranslate"><span class="pre">indirect</span></code>’ labels.</p><p>This instruction should only be used to implement the “goto” feature of gccstyle inline assembly. Any other usage is an error in the IR verifier.</p><p>Note that in order to support outputs along indirect edges, LLVM may need tosplit critical edges, which may require synthesizing a replacement block forthe <code class="docutils literal notranslate"><span class="pre">indirect</span> <span class="pre">labels</span></code>. Therefore, the address of a label as seen by another<code class="docutils literal notranslate"><span class="pre">callbr</span></code> instruction, or for a <a class="reference internal" href="#blockaddress"><span class="std std-ref">blockaddress</span></a> constant,may not be equal to the address provided for the same block to thisinstruction’s <code class="docutils literal notranslate"><span class="pre">indirect</span> <span class="pre">labels</span></code> operand. The assembly code may only transfercontrol to addresses provided via this instruction’s <code class="docutils literal notranslate"><span class="pre">indirect</span> <span class="pre">labels</span></code>.</p><p>On target architectures that implement branch target enforcement by requiringindirect (register-controlled) branch instructions to jump only to locationsmarked by a special instruction (such as AArch64 <code class="docutils literal notranslate"><span class="pre">bti</span></code>), the called code isexpected not to use such an indirect branch to transfer control to thelocations in <code class="docutils literal notranslate"><span class="pre">indirect</span> <span class="pre">labels</span></code>. Therefore, including a label in the<code class="docutils literal notranslate"><span class="pre">indirect</span> <span class="pre">labels</span></code> of a <code class="docutils literal notranslate"><span class="pre">callbr</span></code> does not require the compiler to put a<code class="docutils literal notranslate"><span class="pre">bti</span></code> or equivalent instruction at the label.</p></section><section id="id57"><h5>Arguments:</h5><p>This instruction requires several arguments:</p><ol class="arabic simple"><li><p>The optional “cconv” marker indicates which <a class="reference internal" href="#callingconv"><span class="std std-ref">callingconvention</span></a> the call should use. If none isspecified, the call defaults to using C calling conventions.</p></li><li><p>The optional <a class="reference internal" href="#paramattrs"><span class="std std-ref">Parameter Attributes</span></a> list for returnvalues. Only ‘<code class="docutils literal notranslate"><span class="pre">zeroext</span></code>’, ‘<code class="docutils literal notranslate"><span class="pre">signext</span></code>’, ‘<code class="docutils literal notranslate"><span class="pre">noext</span></code>’, and ‘<code class="docutils literal notranslate"><span class="pre">inreg</span></code>’attributes are valid here.</p></li><li><p>The optional addrspace attribute can be used to indicate the address spaceof the called function. If it is not specified, the program address spacefrom the <a class="reference internal" href="#langref-datalayout"><span class="std std-ref">datalayout string</span></a> will be used.</p></li><li><p>‘<code class="docutils literal notranslate"><span class="pre">ty</span></code>’: the type of the call instruction itself which is also thetype of the return value. Functions that return no value are marked<code class="docutils literal notranslate"><span class="pre">void</span></code>. The signature is computed based on the return type and argumenttypes.</p></li><li><p>‘<code class="docutils literal notranslate"><span class="pre">fnty</span></code>’: shall be the signature of the function being called. Theargument types must match the types implied by this signature. Thisis only required if the signature specifies a varargs type.</p></li><li><p>‘<code class="docutils literal notranslate"><span class="pre">fnptrval</span></code>’: An LLVM value containing a pointer to a function tobe called. In most cases, this is a direct function call, butother <code class="docutils literal notranslate"><span class="pre">callbr</span></code>’s are just as possible, calling an arbitrary pointerto function value.</p></li><li><p>‘<code class="docutils literal notranslate"><span class="pre">function</span> <span class="pre">args</span></code>’: argument list whose types match the functionsignature argument types and parameter attributes. All arguments mustbe of <a class="reference internal" href="#t-firstclass"><span class="std std-ref">first class</span></a> type. If the function signatureindicates the function accepts a variable number of arguments, theextra arguments can be specified.</p></li><li><p>‘<code class="docutils literal notranslate"><span class="pre">fallthrough</span> <span class="pre">label</span></code>’: the label reached when the inline assembly’sexecution exits the bottom.</p></li><li><p>‘<code class="docutils literal notranslate"><span class="pre">indirect</span> <span class="pre">labels</span></code>’: the labels reached when a callee transfers controlto a location other than the ‘<code class="docutils literal notranslate"><span class="pre">fallthrough</span> <span class="pre">label</span></code>’. Label constraintsrefer to these destinations.</p></li><li><p>The optional <a class="reference internal" href="#fnattrs"><span class="std std-ref">function attributes</span></a> list.</p></li><li><p>The optional <a class="reference internal" href="#opbundles"><span class="std std-ref">operand bundles</span></a> list.</p></li></ol></section><section id="id58"><h5>Semantics:</h5><p>This instruction is designed to operate as a standard ‘<code class="docutils literal notranslate"><span class="pre">call</span></code>’instruction in most regards. The primary difference is that itestablishes an association with additional labels to define where controlflow goes after the call.</p><p>The output values of a ‘<code class="docutils literal notranslate"><span class="pre">callbr</span></code>’ instruction are available both in thethe ‘<code class="docutils literal notranslate"><span class="pre">fallthrough</span></code>’ block, and any ‘<code class="docutils literal notranslate"><span class="pre">indirect</span></code>’ blocks(s).</p><p>The only use of this today is to implement the “goto” feature of gcc inlineassembly where additional labels can be provided as locations for the inlineassembly to jump to.</p></section><section id="id59"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="c">; "asm goto" without output constraints.</span><span class="k">callbr</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="k">asm</span><span class="w"> </span><span class="s">""</span><span class="p">,</span><span class="w"> </span><span class="s">"r,!i"</span><span class="p">(</span><span class="kt">i32</span><span class="w"> </span><span class="nv">%x</span><span class="p">)</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%fallthrough</span><span class="w"> </span><span class="p">[</span><span class="kt">label</span><span class="w"> </span><span class="nv">%indirect</span><span class="p">]</span><span class="c">; "asm goto" with output constraints.</span><span class="p"><</span><span class="err">res</span><span class="k">ult</span><span class="p">></span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">callbr</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="k">asm</span><span class="w"> </span><span class="s">""</span><span class="p">,</span><span class="w"> </span><span class="s">"=r,r,!i"</span><span class="p">(</span><span class="kt">i32</span><span class="w"> </span><span class="nv">%x</span><span class="p">)</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%fallthrough</span><span class="w"> </span><span class="p">[</span><span class="kt">label</span><span class="w"> </span><span class="nv">%indirect</span><span class="p">]</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘callbr’ instruction causes control to transfer to a specifiedfunction, with the possibility of control flow transfer to either the‘fallthrough’ label or one of the ‘indirect’ labels.This instruction should only be used to implement the “goto” feature of gccstyle inline assembly. Any other usage is an error in the IR verifier.Note that in order to support outputs along indirect edges, LLVM may need tosplit critical edges, which may require synthesizing a replacement block forthe indirect labels. Therefore, the address of a label as seen by anothercallbr instruction, or for a blockaddress constant,may not be equal to the address provided for the same block to thisinstruction’s indirect labels operand. The assembly code may only transfercontrol to addresses provided via this instruction’s indirect labels.On target architectures that implement branch target enforcement by requiringindirect (register-controlled) branch instructions to jump only to locationsmarked by a special instruction (such as AArch64 bti), the called code isexpected not to use such an indirect branch to transfer control to thelocations in indirect labels. Therefore, including a label in theindirect labels of a callbr does not require the compiler to put abti or equivalent instruction at the label.`,
|
||
};
|
||
case 'RESUME':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#resume-instruction`,
|
||
html: `<html><head></head><body><span id="i-resume"></span><h4><a class="toc-backref" href="#id2162" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">resume</span></code>’ Instruction</a></h4><section id="id60"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">resume</span> <span class="o"><</span><span class="nb">type</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span></pre></div></div></section><section id="id61"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">resume</span></code>’ instruction is a terminator instruction that has nosuccessors.</p></section><section id="id62"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">resume</span></code>’ instruction requires one argument, which must have thesame type as the result of any ‘<code class="docutils literal notranslate"><span class="pre">landingpad</span></code>’ instruction in the samefunction.</p></section><section id="id63"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">resume</span></code>’ instruction resumes propagation of an existing(in-flight) exception whose unwinding was interrupted with a<a class="reference internal" href="#i-landingpad"><span class="std std-ref">landingpad</span></a> instruction.</p></section><section id="id64"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="k">resume</span><span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="kt">ptr</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="p">}</span><span class="w"> </span><span class="nv">%exn</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘resume’ instruction is a terminator instruction that has nosuccessors.`,
|
||
};
|
||
case 'CATCHSWITCH':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#catchswitch-instruction`,
|
||
html: `<html><head></head><body><span id="i-catchswitch"></span><h4><a class="toc-backref" href="#id2163" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">catchswitch</span></code>’ Instruction</a></h4><section id="id65"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">resultval</span><span class="o">></span> <span class="o">=</span> <span class="n">catchswitch</span> <span class="n">within</span> <span class="o"><</span><span class="n">parent</span><span class="o">></span> <span class="p">[</span> <span class="n">label</span> <span class="o"><</span><span class="n">handler1</span><span class="o">></span><span class="p">,</span> <span class="n">label</span> <span class="o"><</span><span class="n">handler2</span><span class="o">></span><span class="p">,</span> <span class="o">...</span> <span class="p">]</span> <span class="n">unwind</span> <span class="n">to</span> <span class="n">caller</span><span class="o"><</span><span class="n">resultval</span><span class="o">></span> <span class="o">=</span> <span class="n">catchswitch</span> <span class="n">within</span> <span class="o"><</span><span class="n">parent</span><span class="o">></span> <span class="p">[</span> <span class="n">label</span> <span class="o"><</span><span class="n">handler1</span><span class="o">></span><span class="p">,</span> <span class="n">label</span> <span class="o"><</span><span class="n">handler2</span><span class="o">></span><span class="p">,</span> <span class="o">...</span> <span class="p">]</span> <span class="n">unwind</span> <span class="n">label</span> <span class="o"><</span><span class="n">default</span><span class="o">></span></pre></div></div></section><section id="id66"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">catchswitch</span></code>’ instruction is used by <a class="reference external" href="ExceptionHandling.html#overview">LLVM’s exception handling system</a> to describe the set of possible catch handlersthat may be executed by the <a class="reference internal" href="#personalityfn"><span class="std std-ref">EH personality routine</span></a>.</p></section><section id="id67"><h5>Arguments:</h5><p>The <code class="docutils literal notranslate"><span class="pre">parent</span></code> argument is the token of the funclet that contains the<code class="docutils literal notranslate"><span class="pre">catchswitch</span></code> instruction. If the <code class="docutils literal notranslate"><span class="pre">catchswitch</span></code> is not inside a funclet,this operand may be the token <code class="docutils literal notranslate"><span class="pre">none</span></code>.</p><p>The <code class="docutils literal notranslate"><span class="pre">default</span></code> argument is the label of another basic block beginning witheither a <code class="docutils literal notranslate"><span class="pre">cleanuppad</span></code> or <code class="docutils literal notranslate"><span class="pre">catchswitch</span></code> instruction. This unwind destinationmust be a legal target with respect to the <code class="docutils literal notranslate"><span class="pre">parent</span></code> links, as described inthe <a class="reference external" href="ExceptionHandling.html#wineh-constraints">exception handling documentation</a>.</p><p>The <code class="docutils literal notranslate"><span class="pre">handlers</span></code> are a nonempty list of successor blocks that each begin with a<a class="reference internal" href="#i-catchpad"><span class="std std-ref">catchpad</span></a> instruction.</p></section><section id="id68"><h5>Semantics:</h5><p>Executing this instruction transfers control to one of the successors in<code class="docutils literal notranslate"><span class="pre">handlers</span></code>, if appropriate, or continues to unwind via the unwind label ifpresent.</p><p>The <code class="docutils literal notranslate"><span class="pre">catchswitch</span></code> is both a terminator and a “pad” instruction, meaning thatit must be both the first non-phi instruction and last instruction in the basicblock. Therefore, it must be the only non-phi instruction in the block.</p></section><section id="id69"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span>dispatch1: %cs1 = catchswitch within none [label %handler0, label %handler1] unwind to callerdispatch2: %cs2 = catchswitch within %parenthandler [label %handler0] unwind label %cleanup</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘catchswitch’ instruction is used by LLVM’s exception handling system to describe the set of possible catch handlersthat may be executed by the EH personality routine.`,
|
||
};
|
||
case 'CATCHRET':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#catchret-instruction`,
|
||
html: `<html><head></head><body><span id="i-catchret"></span><h4><a class="toc-backref" href="#id2164" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">catchret</span></code>’ Instruction</a></h4><section id="id70"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">catchret</span> <span class="kn">from</span> <span class="o"><</span><span class="n">token</span><span class="o">></span> <span class="n">to</span> <span class="n">label</span> <span class="o"><</span><span class="n">normal</span><span class="o">></span></pre></div></div></section><section id="id71"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">catchret</span></code>’ instruction is a terminator instruction that has asingle successor.</p></section><section id="id72"><h5>Arguments:</h5><p>The first argument to a ‘<code class="docutils literal notranslate"><span class="pre">catchret</span></code>’ indicates which <code class="docutils literal notranslate"><span class="pre">catchpad</span></code> itexits. It must be a <a class="reference internal" href="#i-catchpad"><span class="std std-ref">catchpad</span></a>.The second argument to a ‘<code class="docutils literal notranslate"><span class="pre">catchret</span></code>’ specifies where control willtransfer to next.</p></section><section id="id73"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">catchret</span></code>’ instruction ends an existing (in-flight) exception whoseunwinding was interrupted with a <a class="reference internal" href="#i-catchpad"><span class="std std-ref">catchpad</span></a> instruction. The<a class="reference internal" href="#personalityfn"><span class="std std-ref">personality function</span></a> gets a chance to execute arbitrarycode to, for example, destroy the active exception. Control then transfers to<code class="docutils literal notranslate"><span class="pre">normal</span></code>.</p><p>The <code class="docutils literal notranslate"><span class="pre">token</span></code> argument must be a token produced by a <code class="docutils literal notranslate"><span class="pre">catchpad</span></code> instruction.If the specified <code class="docutils literal notranslate"><span class="pre">catchpad</span></code> is not the most-recently-entered not-yet-exitedfunclet pad (as described in the <a class="reference external" href="ExceptionHandling.html#wineh-constraints">EH documentation</a>),the <code class="docutils literal notranslate"><span class="pre">catchret</span></code>’s behavior is undefined.</p></section><section id="id74"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span>catchret from %catch to label %continue</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘catchret’ instruction is a terminator instruction that has asingle successor.`,
|
||
};
|
||
case 'CLEANUPRET':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#cleanupret-instruction`,
|
||
html: `<html><head></head><body><span id="i-cleanupret"></span><h4><a class="toc-backref" href="#id2165" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">cleanupret</span></code>’ Instruction</a></h4><section id="id75"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">cleanupret</span> <span class="kn">from</span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="n">unwind</span> <span class="n">label</span> <span class="o"><</span><span class="k">continue</span><span class="o">></span><span class="n">cleanupret</span> <span class="kn">from</span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="n">unwind</span> <span class="n">to</span> <span class="n">caller</span></pre></div></div></section><section id="id76"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">cleanupret</span></code>’ instruction is a terminator instruction that hasan optional successor.</p></section><section id="id77"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">cleanupret</span></code>’ instruction requires one argument, which indicateswhich <code class="docutils literal notranslate"><span class="pre">cleanuppad</span></code> it exits, and must be a <a class="reference internal" href="#i-cleanuppad"><span class="std std-ref">cleanuppad</span></a>.If the specified <code class="docutils literal notranslate"><span class="pre">cleanuppad</span></code> is not the most-recently-entered not-yet-exitedfunclet pad (as described in the <a class="reference external" href="ExceptionHandling.html#wineh-constraints">EH documentation</a>),the <code class="docutils literal notranslate"><span class="pre">cleanupret</span></code>’s behavior is undefined.</p><p>The ‘<code class="docutils literal notranslate"><span class="pre">cleanupret</span></code>’ instruction also has an optional successor, <code class="docutils literal notranslate"><span class="pre">continue</span></code>,which must be the label of another basic block beginning with either a<code class="docutils literal notranslate"><span class="pre">cleanuppad</span></code> or <code class="docutils literal notranslate"><span class="pre">catchswitch</span></code> instruction. This unwind destination mustbe a legal target with respect to the <code class="docutils literal notranslate"><span class="pre">parent</span></code> links, as described in the<a class="reference external" href="ExceptionHandling.html#wineh-constraints">exception handling documentation</a>.</p></section><section id="id80"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">cleanupret</span></code>’ instruction indicates to the<a class="reference internal" href="#personalityfn"><span class="std std-ref">personality function</span></a> that one<a class="reference internal" href="#i-cleanuppad"><span class="std std-ref">cleanuppad</span></a> it transferred control to has ended.It transfers control to <code class="docutils literal notranslate"><span class="pre">continue</span></code> or unwinds out of the function.</p></section><section id="id81"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span>cleanupret from %cleanup unwind to callercleanupret from %cleanup unwind label %continue</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘cleanupret’ instruction is a terminator instruction that hasan optional successor.`,
|
||
};
|
||
case 'UNREACHABLE':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#unreachable-instruction`,
|
||
html: `<html><head></head><body><span id="i-unreachable"></span><h4><a class="toc-backref" href="#id2166" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">unreachable</span></code>’ Instruction</a></h4><section id="id82"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">unreachable</span></pre></div></div></section><section id="id83"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">unreachable</span></code>’ instruction has no defined semantics. Thisinstruction is used to inform the optimizer that a particular portion ofthe code is not reachable. This can be used to indicate that the codeafter a no-return function cannot be reached, and other facts.</p></section><section id="id84"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">unreachable</span></code>’ instruction has no defined semantics.</p></section></body></html>`,
|
||
tooltip: `The ‘unreachable’ instruction has no defined semantics. Thisinstruction is used to inform the optimizer that a particular portion ofthe code is not reachable. This can be used to indicate that the codeafter a no-return function cannot be reached, and other facts.`,
|
||
};
|
||
case 'FNEG':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#fneg-instruction`,
|
||
html: `<html><head></head><body><span id="i-fneg"></span><h4><a class="toc-backref" href="#id2168" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">fneg</span></code>’ Instruction</a></h4><section id="id85"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">fneg</span> <span class="p">[</span><span class="n">fast</span><span class="o">-</span><span class="n">math</span> <span class="n">flags</span><span class="p">]</span><span class="o">*</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id86"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fneg</span></code>’ instruction returns the negation of its operand.</p></section><section id="id87"><h5>Arguments:</h5><p>The argument to the ‘<code class="docutils literal notranslate"><span class="pre">fneg</span></code>’ instruction must be a<a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> offloating-point values.</p></section><section id="id88"><h5>Semantics:</h5><p>The value produced is a copy of the operand with its sign bit flipped.The value is otherwise completely identical; in particular, if the input is aNaN, then the quiet/signaling bit and payload are perfectly preserved.</p><p>This instruction can also take any number of <a class="reference internal" href="#fastmath"><span class="std std-ref">fast-mathflags</span></a>, which are optimization hints to enable otherwiseunsafe floating-point optimizations:</p></section><section id="id89"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = fneg float %val ; yields float:result = -%var</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘fneg’ instruction returns the negation of its operand.`,
|
||
};
|
||
case 'ADD':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#add-instruction`,
|
||
html: `<html><head></head><body><span id="i-add"></span><h4><a class="toc-backref" href="#id2170" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">add</span></code>’ Instruction</a></h4><section id="id90"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">add</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">add</span> <span class="n">nuw</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">add</span> <span class="n">nsw</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">add</span> <span class="n">nuw</span> <span class="n">nsw</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id91"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">add</span></code>’ instruction returns the sum of its two operands.</p></section><section id="id92"><h5>Arguments:</h5><p>The two arguments to the ‘<code class="docutils literal notranslate"><span class="pre">add</span></code>’ instruction must be<a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> of integer values. Botharguments must have identical types.</p></section><section id="id93"><h5>Semantics:</h5><p>The value produced is the integer sum of the two operands.</p><p>If the sum has unsigned overflow, the result returned is themathematical result modulo 2<sup>n</sup>, where n is the bit width ofthe result.</p><p>Because LLVM integers use a two’s complement representation, thisinstruction is appropriate for both signed and unsigned integers.</p><p><code class="docutils literal notranslate"><span class="pre">nuw</span></code> and <code class="docutils literal notranslate"><span class="pre">nsw</span></code> stand for “No Unsigned Wrap” and “No Signed Wrap”,respectively. If the <code class="docutils literal notranslate"><span class="pre">nuw</span></code> and/or <code class="docutils literal notranslate"><span class="pre">nsw</span></code> keywords are present, theresult value of the <code class="docutils literal notranslate"><span class="pre">add</span></code> is a <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison value</span></a> ifunsigned and/or signed overflow, respectively, occurs.</p></section><section id="id94"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = add i32 4, %var ; yields i32:result = 4 + %var</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘add’ instruction returns the sum of its two operands.`,
|
||
};
|
||
case 'FADD':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#fadd-instruction`,
|
||
html: `<html><head></head><body><span id="i-fadd"></span><h4><a class="toc-backref" href="#id2171" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">fadd</span></code>’ Instruction</a></h4><section id="id95"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">fadd</span> <span class="p">[</span><span class="n">fast</span><span class="o">-</span><span class="n">math</span> <span class="n">flags</span><span class="p">]</span><span class="o">*</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id96"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fadd</span></code>’ instruction returns the sum of its two operands.</p></section><section id="id97"><h5>Arguments:</h5><p>The two arguments to the ‘<code class="docutils literal notranslate"><span class="pre">fadd</span></code>’ instruction must be<a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> offloating-point values. Both arguments must have identical types.</p></section><section id="id98"><h5>Semantics:</h5><p>The value produced is the floating-point sum of the two operands.This instruction is assumed to execute in the default <a class="reference internal" href="#floatenv"><span class="std std-ref">floating-pointenvironment</span></a>.This instruction can also take any number of <a class="reference internal" href="#fastmath"><span class="std std-ref">fast-mathflags</span></a>, which are optimization hints to enable otherwiseunsafe floating-point optimizations:</p></section><section id="id99"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = fadd float 4.0, %var ; yields float:result = 4.0 + %var</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘fadd’ instruction returns the sum of its two operands.`,
|
||
};
|
||
case 'SUB':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#sub-instruction`,
|
||
html: `<html><head></head><body><span id="i-sub"></span><h4><a class="toc-backref" href="#id2172" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">sub</span></code>’ Instruction</a></h4><section id="id100"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">sub</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">sub</span> <span class="n">nuw</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">sub</span> <span class="n">nsw</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">sub</span> <span class="n">nuw</span> <span class="n">nsw</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id101"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">sub</span></code>’ instruction returns the difference of its two operands.</p><p>Note that the ‘<code class="docutils literal notranslate"><span class="pre">sub</span></code>’ instruction is used to represent the ‘<code class="docutils literal notranslate"><span class="pre">neg</span></code>’instruction present in most other intermediate representations.</p></section><section id="id102"><h5>Arguments:</h5><p>The two arguments to the ‘<code class="docutils literal notranslate"><span class="pre">sub</span></code>’ instruction must be<a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> of integer values. Botharguments must have identical types.</p></section><section id="id103"><h5>Semantics:</h5><p>The value produced is the integer difference of the two operands.</p><p>If the difference has unsigned overflow, the result returned is themathematical result modulo 2<sup>n</sup>, where n is the bit width ofthe result.</p><p>Because LLVM integers use a two’s complement representation, thisinstruction is appropriate for both signed and unsigned integers.</p><p><code class="docutils literal notranslate"><span class="pre">nuw</span></code> and <code class="docutils literal notranslate"><span class="pre">nsw</span></code> stand for “No Unsigned Wrap” and “No Signed Wrap”,respectively. If the <code class="docutils literal notranslate"><span class="pre">nuw</span></code> and/or <code class="docutils literal notranslate"><span class="pre">nsw</span></code> keywords are present, theresult value of the <code class="docutils literal notranslate"><span class="pre">sub</span></code> is a <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison value</span></a> ifunsigned and/or signed overflow, respectively, occurs.</p></section><section id="id104"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = sub i32 4, %var ; yields i32:result = 4 - %var<result> = sub i32 0, %val ; yields i32:result = -%var</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘sub’ instruction returns the difference of its two operands.Note that the ‘sub’ instruction is used to represent the ‘neg’instruction present in most other intermediate representations.`,
|
||
};
|
||
case 'FSUB':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#fsub-instruction`,
|
||
html: `<html><head></head><body><span id="i-fsub"></span><h4><a class="toc-backref" href="#id2173" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">fsub</span></code>’ Instruction</a></h4><section id="id105"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">fsub</span> <span class="p">[</span><span class="n">fast</span><span class="o">-</span><span class="n">math</span> <span class="n">flags</span><span class="p">]</span><span class="o">*</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id106"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fsub</span></code>’ instruction returns the difference of its two operands.</p></section><section id="id107"><h5>Arguments:</h5><p>The two arguments to the ‘<code class="docutils literal notranslate"><span class="pre">fsub</span></code>’ instruction must be<a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> offloating-point values. Both arguments must have identical types.</p></section><section id="id108"><h5>Semantics:</h5><p>The value produced is the floating-point difference of the two operands.This instruction is assumed to execute in the default <a class="reference internal" href="#floatenv"><span class="std std-ref">floating-pointenvironment</span></a>.This instruction can also take any number of <a class="reference internal" href="#fastmath"><span class="std std-ref">fast-mathflags</span></a>, which are optimization hints to enable otherwiseunsafe floating-point optimizations:</p></section><section id="id109"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = fsub float 4.0, %var ; yields float:result = 4.0 - %var<result> = fsub float -0.0, %val ; yields float:result = -%var</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘fsub’ instruction returns the difference of its two operands.`,
|
||
};
|
||
case 'MUL':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#mul-instruction`,
|
||
html: `<html><head></head><body><span id="i-mul"></span><h4><a class="toc-backref" href="#id2174" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">mul</span></code>’ Instruction</a></h4><section id="id110"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">mul</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">mul</span> <span class="n">nuw</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">mul</span> <span class="n">nsw</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">mul</span> <span class="n">nuw</span> <span class="n">nsw</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id111"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">mul</span></code>’ instruction returns the product of its two operands.</p></section><section id="id112"><h5>Arguments:</h5><p>The two arguments to the ‘<code class="docutils literal notranslate"><span class="pre">mul</span></code>’ instruction must be<a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> of integer values. Botharguments must have identical types.</p></section><section id="id113"><h5>Semantics:</h5><p>The value produced is the integer product of the two operands.</p><p>If the result of the multiplication has unsigned overflow, the resultreturned is the mathematical result modulo 2<sup>n</sup>, where n is thebit width of the result.</p><p>Because LLVM integers use a two’s complement representation, and theresult is the same width as the operands, this instruction returns thecorrect result for both signed and unsigned integers. If a full product(e.g., <code class="docutils literal notranslate"><span class="pre">i32</span></code> * <code class="docutils literal notranslate"><span class="pre">i32</span></code> -> <code class="docutils literal notranslate"><span class="pre">i64</span></code>) is needed, the operands should besign-extended or zero-extended as appropriate to the width of the fullproduct.</p><p><code class="docutils literal notranslate"><span class="pre">nuw</span></code> and <code class="docutils literal notranslate"><span class="pre">nsw</span></code> stand for “No Unsigned Wrap” and “No Signed Wrap”,respectively. If the <code class="docutils literal notranslate"><span class="pre">nuw</span></code> and/or <code class="docutils literal notranslate"><span class="pre">nsw</span></code> keywords are present, theresult value of the <code class="docutils literal notranslate"><span class="pre">mul</span></code> is a <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison value</span></a> ifunsigned and/or signed overflow, respectively, occurs.</p></section><section id="id114"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = mul i32 4, %var ; yields i32:result = 4 * %var</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘mul’ instruction returns the product of its two operands.`,
|
||
};
|
||
case 'FMUL':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#fmul-instruction`,
|
||
html: `<html><head></head><body><span id="i-fmul"></span><h4><a class="toc-backref" href="#id2175" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">fmul</span></code>’ Instruction</a></h4><section id="id115"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">fmul</span> <span class="p">[</span><span class="n">fast</span><span class="o">-</span><span class="n">math</span> <span class="n">flags</span><span class="p">]</span><span class="o">*</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id116"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fmul</span></code>’ instruction returns the product of its two operands.</p></section><section id="id117"><h5>Arguments:</h5><p>The two arguments to the ‘<code class="docutils literal notranslate"><span class="pre">fmul</span></code>’ instruction must be<a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> offloating-point values. Both arguments must have identical types.</p></section><section id="id118"><h5>Semantics:</h5><p>The value produced is the floating-point product of the two operands.This instruction is assumed to execute in the default <a class="reference internal" href="#floatenv"><span class="std std-ref">floating-pointenvironment</span></a>.This instruction can also take any number of <a class="reference internal" href="#fastmath"><span class="std std-ref">fast-mathflags</span></a>, which are optimization hints to enable otherwiseunsafe floating-point optimizations:</p></section><section id="id119"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = fmul float 4.0, %var ; yields float:result = 4.0 * %var</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘fmul’ instruction returns the product of its two operands.`,
|
||
};
|
||
case 'UDIV':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#udiv-instruction`,
|
||
html: `<html><head></head><body><span id="i-udiv"></span><h4><a class="toc-backref" href="#id2176" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">udiv</span></code>’ Instruction</a></h4><section id="id120"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">udiv</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">udiv</span> <span class="n">exact</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id121"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">udiv</span></code>’ instruction returns the quotient of its two operands.</p></section><section id="id122"><h5>Arguments:</h5><p>The two arguments to the ‘<code class="docutils literal notranslate"><span class="pre">udiv</span></code>’ instruction must be<a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> of integer values. Botharguments must have identical types.</p></section><section id="id123"><h5>Semantics:</h5><p>The value produced is the unsigned integer quotient of the two operands.</p><p>Note that unsigned integer division and signed integer division aredistinct operations; for signed integer division, use ‘<code class="docutils literal notranslate"><span class="pre">sdiv</span></code>’.</p><p>Division by zero is undefined behavior. For vectors, if any elementof the divisor is zero, the operation has undefined behavior.</p><p>If the <code class="docutils literal notranslate"><span class="pre">exact</span></code> keyword is present, the result value of the <code class="docutils literal notranslate"><span class="pre">udiv</span></code> isa <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison value</span></a> if %op1 is not a multiple of %op2 (assuch, “((a udiv exact b) mul b) == a”).</p></section><section id="id124"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = udiv i32 4, %var ; yields i32:result = 4 / %var</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘udiv’ instruction returns the quotient of its two operands.`,
|
||
};
|
||
case 'SDIV':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#sdiv-instruction`,
|
||
html: `<html><head></head><body><span id="i-sdiv"></span><h4><a class="toc-backref" href="#id2177" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">sdiv</span></code>’ Instruction</a></h4><section id="id125"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">sdiv</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">sdiv</span> <span class="n">exact</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id126"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">sdiv</span></code>’ instruction returns the quotient of its two operands.</p></section><section id="id127"><h5>Arguments:</h5><p>The two arguments to the ‘<code class="docutils literal notranslate"><span class="pre">sdiv</span></code>’ instruction must be<a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> of integer values. Botharguments must have identical types.</p></section><section id="id128"><h5>Semantics:</h5><p>The value produced is the signed integer quotient of the two operandsrounded towards zero.</p><p>Note that signed integer division and unsigned integer division aredistinct operations; for unsigned integer division, use ‘<code class="docutils literal notranslate"><span class="pre">udiv</span></code>’.</p><p>Division by zero is undefined behavior. For vectors, if any elementof the divisor is zero, the operation has undefined behavior.Overflow also leads to undefined behavior; this is a rare case, but canoccur, for example, by doing a 32-bit division of -2147483648 by -1.</p><p>If the <code class="docutils literal notranslate"><span class="pre">exact</span></code> keyword is present, the result value of the <code class="docutils literal notranslate"><span class="pre">sdiv</span></code> isa <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison value</span></a> if the result would be rounded.</p></section><section id="id129"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = sdiv i32 4, %var ; yields i32:result = 4 / %var</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘sdiv’ instruction returns the quotient of its two operands.`,
|
||
};
|
||
case 'FDIV':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#fdiv-instruction`,
|
||
html: `<html><head></head><body><span id="i-fdiv"></span><h4><a class="toc-backref" href="#id2178" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">fdiv</span></code>’ Instruction</a></h4><section id="id130"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">fdiv</span> <span class="p">[</span><span class="n">fast</span><span class="o">-</span><span class="n">math</span> <span class="n">flags</span><span class="p">]</span><span class="o">*</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id131"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fdiv</span></code>’ instruction returns the quotient of its two operands.</p></section><section id="id132"><h5>Arguments:</h5><p>The two arguments to the ‘<code class="docutils literal notranslate"><span class="pre">fdiv</span></code>’ instruction must be<a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> offloating-point values. Both arguments must have identical types.</p></section><section id="id133"><h5>Semantics:</h5><p>The value produced is the floating-point quotient of the two operands.This instruction is assumed to execute in the default <a class="reference internal" href="#floatenv"><span class="std std-ref">floating-pointenvironment</span></a>.This instruction can also take any number of <a class="reference internal" href="#fastmath"><span class="std std-ref">fast-mathflags</span></a>, which are optimization hints to enable otherwiseunsafe floating-point optimizations:</p></section><section id="id134"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = fdiv float 4.0, %var ; yields float:result = 4.0 / %var</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘fdiv’ instruction returns the quotient of its two operands.`,
|
||
};
|
||
case 'UREM':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#urem-instruction`,
|
||
html: `<html><head></head><body><span id="i-urem"></span><h4><a class="toc-backref" href="#id2179" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">urem</span></code>’ Instruction</a></h4><section id="id135"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">urem</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id136"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">urem</span></code>’ instruction returns the remainder from the unsigneddivision of its two arguments.</p></section><section id="id137"><h5>Arguments:</h5><p>The two arguments to the ‘<code class="docutils literal notranslate"><span class="pre">urem</span></code>’ instruction must be<a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> of integer values. Botharguments must have identical types.</p></section><section id="id138"><h5>Semantics:</h5><p>This instruction returns the unsigned integer <em>remainder</em> of a division.This instruction always performs an unsigned division to get theremainder.</p><p>Note that unsigned integer remainder and signed integer remainder aredistinct operations; for signed integer remainder, use ‘<code class="docutils literal notranslate"><span class="pre">srem</span></code>’.</p><p>Taking the remainder of a division by zero is undefined behavior.For vectors, if any element of the divisor is zero, the operation hasundefined behavior.</p></section><section id="id139"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = urem i32 4, %var ; yields i32:result = 4 % %var</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘urem’ instruction returns the remainder from the unsigneddivision of its two arguments.`,
|
||
};
|
||
case 'SREM':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#srem-instruction`,
|
||
html: `<html><head></head><body><span id="i-srem"></span><h4><a class="toc-backref" href="#id2180" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">srem</span></code>’ Instruction</a></h4><section id="id140"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">srem</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id141"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">srem</span></code>’ instruction returns the remainder from the signeddivision of its two operands. This instruction can also take<a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> versions of the values in which case the elementsmust be integers.</p></section><section id="id142"><h5>Arguments:</h5><p>The two arguments to the ‘<code class="docutils literal notranslate"><span class="pre">srem</span></code>’ instruction must be<a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> of integer values. Botharguments must have identical types.</p></section><section id="id143"><h5>Semantics:</h5><p>This instruction returns the <em>remainder</em> of a division (where the resultis either zero or has the same sign as the dividend, <code class="docutils literal notranslate"><span class="pre">op1</span></code>), not the<em>modulo</em> operator (where the result is either zero or has the same signas the divisor, <code class="docutils literal notranslate"><span class="pre">op2</span></code>) of a value. For more information about thedifference, see <a class="reference external" href="http://mathforum.org/dr.math/problems/anne.4.28.99.html">The MathForum</a>. For atable of how this is implemented in various languages, please see<a class="reference external" href="http://en.wikipedia.org/wiki/Modulo_operation">Wikipedia: modulooperation</a>.</p><p>Note that signed integer remainder and unsigned integer remainder aredistinct operations; for unsigned integer remainder, use ‘<code class="docutils literal notranslate"><span class="pre">urem</span></code>’.</p><p>Taking the remainder of a division by zero is undefined behavior.For vectors, if any element of the divisor is zero, the operation hasundefined behavior.Overflow also leads to undefined behavior; this is a rare case, but canoccur, for example, by taking the remainder of a 32-bit division of-2147483648 by -1. (The remainder doesn’t actually overflow, but thisrule lets srem be implemented using instructions that return both theresult of the division and the remainder.)</p></section><section id="id144"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = srem i32 4, %var ; yields i32:result = 4 % %var</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘srem’ instruction returns the remainder from the signeddivision of its two operands. This instruction can also takevector versions of the values in which case the elementsmust be integers.`,
|
||
};
|
||
case 'FREM':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#frem-instruction`,
|
||
html: `<html><head></head><body><span id="i-frem"></span><h4><a class="toc-backref" href="#id2181" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">frem</span></code>’ Instruction</a></h4><section id="id145"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">frem</span> <span class="p">[</span><span class="n">fast</span><span class="o">-</span><span class="n">math</span> <span class="n">flags</span><span class="p">]</span><span class="o">*</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id146"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">frem</span></code>’ instruction returns the remainder from the division ofits two operands.</p><div class="admonition note"><p class="admonition-title">Note</p><p>The instruction is implemented as a call to libm’s ‘<code class="docutils literal notranslate"><span class="pre">fmod</span></code>’for some targets, and using the instruction may thus require linking libm.</p></div></section><section id="id147"><h5>Arguments:</h5><p>The two arguments to the ‘<code class="docutils literal notranslate"><span class="pre">frem</span></code>’ instruction must be<a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> offloating-point values. Both arguments must have identical types.</p></section><section id="id148"><h5>Semantics:</h5><p>The value produced is the floating-point remainder of the two operands.This is the same output as a libm ‘<code class="docutils literal notranslate"><span class="pre">fmod</span></code>’ function, but without anypossibility of setting <code class="docutils literal notranslate"><span class="pre">errno</span></code>. The remainder has the same sign as thedividend.This instruction is assumed to execute in the default <a class="reference internal" href="#floatenv"><span class="std std-ref">floating-pointenvironment</span></a>.This instruction can also take any number of <a class="reference internal" href="#fastmath"><span class="std std-ref">fast-mathflags</span></a>, which are optimization hints to enable otherwiseunsafe floating-point optimizations:</p></section><section id="id149"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = frem float 4.0, %var ; yields float:result = 4.0 % %var</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘frem’ instruction returns the remainder from the division ofits two operands.`,
|
||
};
|
||
case 'SHL':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#shl-instruction`,
|
||
html: `<html><head></head><body><span id="i-shl"></span><h4><a class="toc-backref" href="#id2183" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">shl</span></code>’ Instruction</a></h4><section id="id150"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">shl</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">shl</span> <span class="n">nuw</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">shl</span> <span class="n">nsw</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">shl</span> <span class="n">nuw</span> <span class="n">nsw</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id151"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">shl</span></code>’ instruction returns the first operand shifted to the lefta specified number of bits.</p></section><section id="id152"><h5>Arguments:</h5><p>Both arguments to the ‘<code class="docutils literal notranslate"><span class="pre">shl</span></code>’ instruction must be the same<a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> of integer type.‘<code class="docutils literal notranslate"><span class="pre">op2</span></code>’ is treated as an unsigned value.</p></section><section id="id153"><h5>Semantics:</h5><p>The value produced is <code class="docutils literal notranslate"><span class="pre">op1</span></code> * 2<sup>op2</sup> mod 2<sup>n</sup>,where <code class="docutils literal notranslate"><span class="pre">n</span></code> is the width of the result. If <code class="docutils literal notranslate"><span class="pre">op2</span></code> is (statically ordynamically) equal to or larger than the number of bits in<code class="docutils literal notranslate"><span class="pre">op1</span></code>, this instruction returns a <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison value</span></a>.If the arguments are vectors, each vector element of <code class="docutils literal notranslate"><span class="pre">op1</span></code> is shiftedby the corresponding shift amount in <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p><p>If the <code class="docutils literal notranslate"><span class="pre">nuw</span></code> keyword is present, then the shift produces a poisonvalue if it shifts out any non-zero bits.If the <code class="docutils literal notranslate"><span class="pre">nsw</span></code> keyword is present, then the shift produces a poisonvalue if it shifts out any bits that disagree with the resultant sign bit.</p></section><section id="id154"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = shl i32 4, %var ; yields i32: 4 << %var<result> = shl i32 4, 2 ; yields i32: 16<result> = shl i32 1, 10 ; yields i32: 1024<result> = shl i32 1, 32 ; undefined<result> = shl <2 x i32> < i32 1, i32 1>, < i32 1, i32 2> ; yields: result=<2 x i32> < i32 2, i32 4></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘shl’ instruction returns the first operand shifted to the lefta specified number of bits.`,
|
||
};
|
||
case 'LSHR':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#lshr-instruction`,
|
||
html: `<html><head></head><body><span id="i-lshr"></span><h4><a class="toc-backref" href="#id2184" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">lshr</span></code>’ Instruction</a></h4><section id="id155"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">lshr</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">lshr</span> <span class="n">exact</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id156"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">lshr</span></code>’ instruction (logical shift right) returns the firstoperand shifted to the right a specified number of bits with zero fill.</p></section><section id="id157"><h5>Arguments:</h5><p>Both arguments to the ‘<code class="docutils literal notranslate"><span class="pre">lshr</span></code>’ instruction must be the same<a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> of integer type.‘<code class="docutils literal notranslate"><span class="pre">op2</span></code>’ is treated as an unsigned value.</p></section><section id="id158"><h5>Semantics:</h5><p>This instruction always performs a logical shift right operation. Themost significant bits of the result will be filled with zero bits afterthe shift. If <code class="docutils literal notranslate"><span class="pre">op2</span></code> is (statically or dynamically) equal to or largerthan the number of bits in <code class="docutils literal notranslate"><span class="pre">op1</span></code>, this instruction returns a <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poisonvalue</span></a>. If the arguments are vectors, each vector elementof <code class="docutils literal notranslate"><span class="pre">op1</span></code> is shifted by the corresponding shift amount in <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p><p>If the <code class="docutils literal notranslate"><span class="pre">exact</span></code> keyword is present, the result value of the <code class="docutils literal notranslate"><span class="pre">lshr</span></code> isa poison value if any of the bits shifted out are non-zero.</p></section><section id="id159"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = lshr i32 4, 1 ; yields i32:result = 2<result> = lshr i32 4, 2 ; yields i32:result = 1<result> = lshr i8 4, 3 ; yields i8:result = 0<result> = lshr i8 -2, 1 ; yields i8:result = 0x7F<result> = lshr i32 1, 32 ; undefined<result> = lshr <2 x i32> < i32 -2, i32 4>, < i32 1, i32 2> ; yields: result=<2 x i32> < i32 0x7FFFFFFF, i32 1></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘lshr’ instruction (logical shift right) returns the firstoperand shifted to the right a specified number of bits with zero fill.`,
|
||
};
|
||
case 'ASHR':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#ashr-instruction`,
|
||
html: `<html><head></head><body><span id="i-ashr"></span><h4><a class="toc-backref" href="#id2185" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">ashr</span></code>’ Instruction</a></h4><section id="id160"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">ashr</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">ashr</span> <span class="n">exact</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id161"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">ashr</span></code>’ instruction (arithmetic shift right) returns the firstoperand shifted to the right a specified number of bits with signextension.</p></section><section id="id162"><h5>Arguments:</h5><p>Both arguments to the ‘<code class="docutils literal notranslate"><span class="pre">ashr</span></code>’ instruction must be the same<a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> of integer type.‘<code class="docutils literal notranslate"><span class="pre">op2</span></code>’ is treated as an unsigned value.</p></section><section id="id163"><h5>Semantics:</h5><p>This instruction always performs an arithmetic shift right operation,The most significant bits of the result will be filled with the sign bitof <code class="docutils literal notranslate"><span class="pre">op1</span></code>. If <code class="docutils literal notranslate"><span class="pre">op2</span></code> is (statically or dynamically) equal to or largerthan the number of bits in <code class="docutils literal notranslate"><span class="pre">op1</span></code>, this instruction returns a <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poisonvalue</span></a>. If the arguments are vectors, each vector elementof <code class="docutils literal notranslate"><span class="pre">op1</span></code> is shifted by the corresponding shift amount in <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p><p>If the <code class="docutils literal notranslate"><span class="pre">exact</span></code> keyword is present, the result value of the <code class="docutils literal notranslate"><span class="pre">ashr</span></code> isa poison value if any of the bits shifted out are non-zero.</p></section><section id="id164"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = ashr i32 4, 1 ; yields i32:result = 2<result> = ashr i32 4, 2 ; yields i32:result = 1<result> = ashr i8 4, 3 ; yields i8:result = 0<result> = ashr i8 -2, 1 ; yields i8:result = -1<result> = ashr i32 1, 32 ; undefined<result> = ashr <2 x i32> < i32 -2, i32 4>, < i32 1, i32 3> ; yields: result=<2 x i32> < i32 -1, i32 0></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘ashr’ instruction (arithmetic shift right) returns the firstoperand shifted to the right a specified number of bits with signextension.`,
|
||
};
|
||
case 'AND':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#and-instruction`,
|
||
html: `<html><head></head><body><span id="i-and"></span><h4><a class="toc-backref" href="#id2186" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">and</span></code>’ Instruction</a></h4><section id="id165"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="ow">and</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id166"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">and</span></code>’ instruction returns the bitwise logical and of its twooperands.</p></section><section id="id167"><h5>Arguments:</h5><p>The two arguments to the ‘<code class="docutils literal notranslate"><span class="pre">and</span></code>’ instruction must be<a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> of integer values. Botharguments must have identical types.</p></section><section id="id168"><h5>Semantics:</h5><p>The truth table used for the ‘<code class="docutils literal notranslate"><span class="pre">and</span></code>’ instruction is:</p><table class="docutils align-default"><tbody><tr class="row-odd"><td><p>In0</p></td><td><p>In1</p></td><td><p>Out</p></td></tr><tr class="row-even"><td><p>0</p></td><td><p>0</p></td><td><p>0</p></td></tr><tr class="row-odd"><td><p>0</p></td><td><p>1</p></td><td><p>0</p></td></tr><tr class="row-even"><td><p>1</p></td><td><p>0</p></td><td><p>0</p></td></tr><tr class="row-odd"><td><p>1</p></td><td><p>1</p></td><td><p>1</p></td></tr></tbody></table></section><section id="id169"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = and i32 4, %var ; yields i32:result = 4 & %var<result> = and i32 15, 40 ; yields i32:result = 8<result> = and i32 4, 8 ; yields i32:result = 0</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘and’ instruction returns the bitwise logical and of its twooperands.`,
|
||
};
|
||
case 'OR':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#or-instruction`,
|
||
html: `<html><head></head><body><span id="i-or"></span><h4><a class="toc-backref" href="#id2187" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">or</span></code>’ Instruction</a></h4><section id="id170"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="ow">or</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="ow">or</span> <span class="n">disjoint</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id171"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">or</span></code>’ instruction returns the bitwise logical inclusive or of itstwo operands.</p></section><section id="id172"><h5>Arguments:</h5><p>The two arguments to the ‘<code class="docutils literal notranslate"><span class="pre">or</span></code>’ instruction must be<a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> of integer values. Botharguments must have identical types.</p></section><section id="id173"><h5>Semantics:</h5><p>The truth table used for the ‘<code class="docutils literal notranslate"><span class="pre">or</span></code>’ instruction is:</p><table class="docutils align-default"><tbody><tr class="row-odd"><td><p>In0</p></td><td><p>In1</p></td><td><p>Out</p></td></tr><tr class="row-even"><td><p>0</p></td><td><p>0</p></td><td><p>0</p></td></tr><tr class="row-odd"><td><p>0</p></td><td><p>1</p></td><td><p>1</p></td></tr><tr class="row-even"><td><p>1</p></td><td><p>0</p></td><td><p>1</p></td></tr><tr class="row-odd"><td><p>1</p></td><td><p>1</p></td><td><p>1</p></td></tr></tbody></table><p><code class="docutils literal notranslate"><span class="pre">disjoint</span></code> means that for each bit, that bit is zero in at least one of theinputs. This allows the Or to be treated as an Add since no carry can occur fromany bit. If the disjoint keyword is present, the result value of the <code class="docutils literal notranslate"><span class="pre">or</span></code> is a<a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison value</span></a> if both inputs have a one in the same bitposition. For vectors, only the element containing the bit is poison.</p></section><section id="id174"><h5>Example:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="ow">or</span> <span class="n">i32</span> <span class="mi">4</span><span class="p">,</span> <span class="o">%</span><span class="n">var</span> <span class="p">;</span> <span class="n">yields</span> <span class="n">i32</span><span class="p">:</span><span class="n">result</span> <span class="o">=</span> <span class="mi">4</span> <span class="o">|</span> <span class="o">%</span><span class="n">var</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="ow">or</span> <span class="n">i32</span> <span class="mi">15</span><span class="p">,</span> <span class="mi">40</span> <span class="p">;</span> <span class="n">yields</span> <span class="n">i32</span><span class="p">:</span><span class="n">result</span> <span class="o">=</span> <span class="mi">47</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="ow">or</span> <span class="n">i32</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">8</span> <span class="p">;</span> <span class="n">yields</span> <span class="n">i32</span><span class="p">:</span><span class="n">result</span> <span class="o">=</span> <span class="mi">12</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘or’ instruction returns the bitwise logical inclusive or of itstwo operands.`,
|
||
};
|
||
case 'XOR':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#xor-instruction`,
|
||
html: `<html><head></head><body><span id="i-xor"></span><h4><a class="toc-backref" href="#id2188" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">xor</span></code>’ Instruction</a></h4><section id="id175"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">xor</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id176"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">xor</span></code>’ instruction returns the bitwise logical exclusive or ofits two operands. The <code class="docutils literal notranslate"><span class="pre">xor</span></code> is used to implement the “one’scomplement” operation, which is the “~” operator in C.</p></section><section id="id177"><h5>Arguments:</h5><p>The two arguments to the ‘<code class="docutils literal notranslate"><span class="pre">xor</span></code>’ instruction must be<a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> or <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> of integer values. Botharguments must have identical types.</p></section><section id="id178"><h5>Semantics:</h5><p>The truth table used for the ‘<code class="docutils literal notranslate"><span class="pre">xor</span></code>’ instruction is:</p><table class="docutils align-default"><tbody><tr class="row-odd"><td><p>In0</p></td><td><p>In1</p></td><td><p>Out</p></td></tr><tr class="row-even"><td><p>0</p></td><td><p>0</p></td><td><p>0</p></td></tr><tr class="row-odd"><td><p>0</p></td><td><p>1</p></td><td><p>1</p></td></tr><tr class="row-even"><td><p>1</p></td><td><p>0</p></td><td><p>1</p></td></tr><tr class="row-odd"><td><p>1</p></td><td><p>1</p></td><td><p>0</p></td></tr></tbody></table></section><section id="id179"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = xor i32 4, %var ; yields i32:result = 4 ^ %var<result> = xor i32 15, 40 ; yields i32:result = 39<result> = xor i32 4, 8 ; yields i32:result = 12<result> = xor i32 %V, -1 ; yields i32:result = ~%V</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘xor’ instruction returns the bitwise logical exclusive or ofits two operands. The xor is used to implement the “one’scomplement” operation, which is the “~” operator in C.`,
|
||
};
|
||
case 'EXTRACTELEMENT':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#extractelement-instruction`,
|
||
html: `<html><head></head><body><span id="i-extractelement"></span><h4><a class="toc-backref" href="#id2190" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">extractelement</span></code>’ Instruction</a></h4><section id="id180"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">extractelement</span> <span class="o"><</span><span class="n">n</span> <span class="n">x</span> <span class="o"><</span><span class="n">ty</span><span class="o">>></span> <span class="o"><</span><span class="n">val</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="o"><</span><span class="n">idx</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">extractelement</span> <span class="o"><</span><span class="n">vscale</span> <span class="n">x</span> <span class="n">n</span> <span class="n">x</span> <span class="o"><</span><span class="n">ty</span><span class="o">>></span> <span class="o"><</span><span class="n">val</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="o"><</span><span class="n">idx</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span></pre></div></div></section><section id="id181"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">extractelement</span></code>’ instruction extracts a single scalar elementfrom a vector at a specified index.</p></section><section id="id182"><h5>Arguments:</h5><p>The first operand of an ‘<code class="docutils literal notranslate"><span class="pre">extractelement</span></code>’ instruction is a value of<a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> type. The second operand is an index indicatingthe position from which to extract the element. The index may be avariable of any integer type, and will be treated as an unsigned integer.</p></section><section id="id183"><h5>Semantics:</h5><p>The result is a scalar of the same type as the element type of <code class="docutils literal notranslate"><span class="pre">val</span></code>.Its value is the value at position <code class="docutils literal notranslate"><span class="pre">idx</span></code> of <code class="docutils literal notranslate"><span class="pre">val</span></code>. If <code class="docutils literal notranslate"><span class="pre">idx</span></code>exceeds the length of <code class="docutils literal notranslate"><span class="pre">val</span></code> for a fixed-length vector, the result is a<a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison value</span></a>. For a scalable vector, if the valueof <code class="docutils literal notranslate"><span class="pre">idx</span></code> exceeds the runtime length of the vector, the result is a<a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison value</span></a>.</p></section><section id="id184"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = extractelement <4 x i32> %vec, i32 0 ; yields i32</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘extractelement’ instruction extracts a single scalar elementfrom a vector at a specified index.`,
|
||
};
|
||
case 'INSERTELEMENT':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#insertelement-instruction`,
|
||
html: `<html><head></head><body><span id="i-insertelement"></span><h4><a class="toc-backref" href="#id2191" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">insertelement</span></code>’ Instruction</a></h4><section id="id185"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">insertelement</span> <span class="o"><</span><span class="n">n</span> <span class="n">x</span> <span class="o"><</span><span class="n">ty</span><span class="o">>></span> <span class="o"><</span><span class="n">val</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">elt</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="o"><</span><span class="n">idx</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="o"><</span><span class="n">n</span> <span class="n">x</span> <span class="o"><</span><span class="n">ty</span><span class="o">>></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">insertelement</span> <span class="o"><</span><span class="n">vscale</span> <span class="n">x</span> <span class="n">n</span> <span class="n">x</span> <span class="o"><</span><span class="n">ty</span><span class="o">>></span> <span class="o"><</span><span class="n">val</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">elt</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="o"><</span><span class="n">idx</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="o"><</span><span class="n">vscale</span> <span class="n">x</span> <span class="n">n</span> <span class="n">x</span> <span class="o"><</span><span class="n">ty</span><span class="o">>></span></pre></div></div></section><section id="id186"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">insertelement</span></code>’ instruction inserts a scalar element into avector at a specified index.</p></section><section id="id187"><h5>Arguments:</h5><p>The first operand of an ‘<code class="docutils literal notranslate"><span class="pre">insertelement</span></code>’ instruction is a value of<a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> type. The second operand is a scalar value whosetype must equal the element type of the first operand. The third operandis an index indicating the position at which to insert the value. Theindex may be a variable of any integer type, and will be treated as anunsigned integer.</p></section><section id="id188"><h5>Semantics:</h5><p>The result is a vector of the same type as <code class="docutils literal notranslate"><span class="pre">val</span></code>. Its element valuesare those of <code class="docutils literal notranslate"><span class="pre">val</span></code> except at position <code class="docutils literal notranslate"><span class="pre">idx</span></code>, where it gets the value<code class="docutils literal notranslate"><span class="pre">elt</span></code>. If <code class="docutils literal notranslate"><span class="pre">idx</span></code> exceeds the length of <code class="docutils literal notranslate"><span class="pre">val</span></code> for a fixed-length vector,the result is a <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison value</span></a>. For a scalable vector,if the value of <code class="docutils literal notranslate"><span class="pre">idx</span></code> exceeds the runtime length of the vector, the resultis a <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison value</span></a>.</p></section><section id="id189"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = insertelement <4 x i32> %vec, i32 1, i32 0 ; yields <4 x i32></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘insertelement’ instruction inserts a scalar element into avector at a specified index.`,
|
||
};
|
||
case 'SHUFFLEVECTOR':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#shufflevector-instruction`,
|
||
html: `<html><head></head><body><span id="i-shufflevector"></span><h4><a class="toc-backref" href="#id2192" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">shufflevector</span></code>’ Instruction</a></h4><section id="id190"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">shufflevector</span> <span class="o"><</span><span class="n">n</span> <span class="n">x</span> <span class="o"><</span><span class="n">ty</span><span class="o">>></span> <span class="o"><</span><span class="n">v1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">n</span> <span class="n">x</span> <span class="o"><</span><span class="n">ty</span><span class="o">>></span> <span class="o"><</span><span class="n">v2</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">m</span> <span class="n">x</span> <span class="n">i32</span><span class="o">></span> <span class="o"><</span><span class="n">mask</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="o"><</span><span class="n">m</span> <span class="n">x</span> <span class="o"><</span><span class="n">ty</span><span class="o">>></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">shufflevector</span> <span class="o"><</span><span class="n">vscale</span> <span class="n">x</span> <span class="n">n</span> <span class="n">x</span> <span class="o"><</span><span class="n">ty</span><span class="o">>></span> <span class="o"><</span><span class="n">v1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">vscale</span> <span class="n">x</span> <span class="n">n</span> <span class="n">x</span> <span class="o"><</span><span class="n">ty</span><span class="o">>></span> <span class="n">v2</span><span class="p">,</span> <span class="o"><</span><span class="n">vscale</span> <span class="n">x</span> <span class="n">m</span> <span class="n">x</span> <span class="n">i32</span><span class="o">></span> <span class="o"><</span><span class="n">mask</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="o"><</span><span class="n">vscale</span> <span class="n">x</span> <span class="n">m</span> <span class="n">x</span> <span class="o"><</span><span class="n">ty</span><span class="o">>></span></pre></div></div></section><section id="id191"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">shufflevector</span></code>’ instruction constructs a permutation of elementsfrom two input vectors, returning a vector with the same element type asthe input and length that is the same as the shuffle mask.</p></section><section id="id192"><h5>Arguments:</h5><p>The first two operands of a ‘<code class="docutils literal notranslate"><span class="pre">shufflevector</span></code>’ instruction are vectorswith the same type. The third argument is a shuffle mask vector constantwhose element type is <code class="docutils literal notranslate"><span class="pre">i32</span></code>. The mask vector elements must be constantintegers or <code class="docutils literal notranslate"><span class="pre">poison</span></code> values. The result of the instruction is a vectorwhose length is the same as the shuffle mask and whose element type is thesame as the element type of the first two operands.</p></section><section id="id193"><h5>Semantics:</h5><p>The elements of the two input vectors are numbered from left to rightacross both of the vectors. For each element of the result vector, theshuffle mask selects an element from one of the input vectors to copyto the result. Non-negative elements in the mask represent an indexinto the concatenated pair of input vectors.</p><p>A <code class="docutils literal notranslate"><span class="pre">poison</span></code> element in the mask vector specifies that the resulting elementis <code class="docutils literal notranslate"><span class="pre">poison</span></code>.For backwards-compatibility reasons, LLVM temporarily also accepts <code class="docutils literal notranslate"><span class="pre">undef</span></code>mask elements, which will be interpreted the same way as <code class="docutils literal notranslate"><span class="pre">poison</span></code> elements.If the shuffle mask selects an <code class="docutils literal notranslate"><span class="pre">undef</span></code> element from one of the inputvectors, the resulting element is <code class="docutils literal notranslate"><span class="pre">undef</span></code>.</p><p>For scalable vectors, the only valid mask values at present are<code class="docutils literal notranslate"><span class="pre">zeroinitializer</span></code>, <code class="docutils literal notranslate"><span class="pre">undef</span></code> and <code class="docutils literal notranslate"><span class="pre">poison</span></code>, since we cannot write all indices asliterals for a vector with a length unknown at compile time.</p></section><section id="id194"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = shufflevector <4 x i32> %v1, <4 x i32> %v2, <4 x i32> <i32 0, i32 4, i32 1, i32 5> ; yields <4 x i32><result> = shufflevector <4 x i32> %v1, <4 x i32> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 3> ; yields <4 x i32> - Identity shuffle.<result> = shufflevector <8 x i32> %v1, <8 x i32> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 3> ; yields <4 x i32><result> = shufflevector <4 x i32> %v1, <4 x i32> %v2, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7 > ; yields <8 x i32></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘shufflevector’ instruction constructs a permutation of elementsfrom two input vectors, returning a vector with the same element type asthe input and length that is the same as the shuffle mask.`,
|
||
};
|
||
case 'EXTRACTVALUE':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#extractvalue-instruction`,
|
||
html: `<html><head></head><body><span id="i-extractvalue"></span><h4><a class="toc-backref" href="#id2194" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">extractvalue</span></code>’ Instruction</a></h4><section id="id195"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">extractvalue</span> <span class="o"><</span><span class="n">aggregate</span> <span class="nb">type</span><span class="o">></span> <span class="o"><</span><span class="n">val</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">idx</span><span class="o">></span><span class="p">{,</span> <span class="o"><</span><span class="n">idx</span><span class="o">></span><span class="p">}</span><span class="o">*</span></pre></div></div></section><section id="id196"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">extractvalue</span></code>’ instruction extracts the value of a member fieldfrom an <a class="reference internal" href="#t-aggregate"><span class="std std-ref">aggregate</span></a> value.</p></section><section id="id197"><h5>Arguments:</h5><p>The first operand of an ‘<code class="docutils literal notranslate"><span class="pre">extractvalue</span></code>’ instruction is a value of<a class="reference internal" href="#t-struct"><span class="std std-ref">struct</span></a> or <a class="reference internal" href="#t-array"><span class="std std-ref">array</span></a> type. The other operands areconstant indices to specify which value to extract in a similar manneras indices in a ‘<code class="docutils literal notranslate"><span class="pre">getelementptr</span></code>’ instruction.</p><p>The major differences to <code class="docutils literal notranslate"><span class="pre">getelementptr</span></code> indexing are:</p><ul class="simple"><li><p>Since the value being indexed is not a pointer, the first index isomitted and assumed to be zero.</p></li><li><p>At least one index must be specified.</p></li><li><p>Not only struct indices but also array indices must be in bounds.</p></li></ul></section><section id="id198"><h5>Semantics:</h5><p>The result is the value at the position in the aggregate specified bythe index operands.</p></section><section id="id199"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = extractvalue {i32, float} %agg, 0 ; yields i32</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘extractvalue’ instruction extracts the value of a member fieldfrom an aggregate value.`,
|
||
};
|
||
case 'INSERTVALUE':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#insertvalue-instruction`,
|
||
html: `<html><head></head><body><span id="i-insertvalue"></span><h4><a class="toc-backref" href="#id2195" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">insertvalue</span></code>’ Instruction</a></h4><section id="id200"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">insertvalue</span> <span class="o"><</span><span class="n">aggregate</span> <span class="nb">type</span><span class="o">></span> <span class="o"><</span><span class="n">val</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">elt</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">idx</span><span class="o">></span><span class="p">{,</span> <span class="o"><</span><span class="n">idx</span><span class="o">></span><span class="p">}</span><span class="o">*</span> <span class="p">;</span> <span class="n">yields</span> <span class="o"><</span><span class="n">aggregate</span> <span class="nb">type</span><span class="o">></span></pre></div></div></section><section id="id201"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">insertvalue</span></code>’ instruction inserts a value into a member field inan <a class="reference internal" href="#t-aggregate"><span class="std std-ref">aggregate</span></a> value.</p></section><section id="id202"><h5>Arguments:</h5><p>The first operand of an ‘<code class="docutils literal notranslate"><span class="pre">insertvalue</span></code>’ instruction is a value of<a class="reference internal" href="#t-struct"><span class="std std-ref">struct</span></a> or <a class="reference internal" href="#t-array"><span class="std std-ref">array</span></a> type. The second operand isa first-class value to insert. The following operands are constantindices indicating the position at which to insert the value in asimilar manner as indices in a ‘<code class="docutils literal notranslate"><span class="pre">extractvalue</span></code>’ instruction. The valueto insert must have the same type as the value identified by theindices.</p></section><section id="id203"><h5>Semantics:</h5><p>The result is an aggregate of the same type as <code class="docutils literal notranslate"><span class="pre">val</span></code>. Its value isthat of <code class="docutils literal notranslate"><span class="pre">val</span></code> except that the value at the position specified by theindices is that of <code class="docutils literal notranslate"><span class="pre">elt</span></code>.</p></section><section id="id204"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%agg1</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">insertvalue</span><span class="w"> </span><span class="p">{</span><span class="kt">i32</span><span class="p">,</span><span class="w"> </span><span class="kt">float</span><span class="p">}</span><span class="w"> </span><span class="k">poison</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">1</span><span class="p">,</span><span class="w"> </span><span class="m">0</span><span class="w"> </span><span class="c">; yields {i32 1, float poison}</span><span class="nv">%agg2</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">insertvalue</span><span class="w"> </span><span class="p">{</span><span class="kt">i32</span><span class="p">,</span><span class="w"> </span><span class="kt">float</span><span class="p">}</span><span class="w"> </span><span class="nv">%agg1</span><span class="p">,</span><span class="w"> </span><span class="kt">float</span><span class="w"> </span><span class="nv">%val</span><span class="p">,</span><span class="w"> </span><span class="m">1</span><span class="w"> </span><span class="c">; yields {i32 1, float %val}</span><span class="nv">%agg3</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">insertvalue</span><span class="w"> </span><span class="p">{</span><span class="kt">i32</span><span class="p">,</span><span class="w"> </span><span class="p">{</span><span class="kt">float</span><span class="p">}}</span><span class="w"> </span><span class="k">poison</span><span class="p">,</span><span class="w"> </span><span class="kt">float</span><span class="w"> </span><span class="nv">%val</span><span class="p">,</span><span class="w"> </span><span class="m">1</span><span class="p">,</span><span class="w"> </span><span class="m">0</span><span class="w"> </span><span class="c">; yields {i32 poison, {float %val}}</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘insertvalue’ instruction inserts a value into a member field inan aggregate value.`,
|
||
};
|
||
case 'ALLOCA':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#alloca-instruction`,
|
||
html: `<html><head></head><body><span id="i-alloca"></span><h4><a class="toc-backref" href="#id2197" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">alloca</span></code>’ Instruction</a></h4><section id="id205"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">alloca</span> <span class="p">[</span><span class="n">inalloca</span><span class="p">]</span> <span class="o"><</span><span class="nb">type</span><span class="o">></span> <span class="p">[,</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">NumElements</span><span class="o">></span><span class="p">]</span> <span class="p">[,</span> <span class="n">align</span> <span class="o"><</span><span class="n">alignment</span><span class="o">></span><span class="p">]</span> <span class="p">[,</span> <span class="n">addrspace</span><span class="p">(</span><span class="o"><</span><span class="n">num</span><span class="o">></span><span class="p">)]</span> <span class="p">;</span> <span class="n">yields</span> <span class="nb">type</span> <span class="n">addrspace</span><span class="p">(</span><span class="n">num</span><span class="p">)</span><span class="o">*</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id206"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">alloca</span></code>’ instruction allocates memory on the stack frame of thecurrently executing function, to be automatically released when thisfunction returns to its caller. If the address space is not explicitlyspecified, the default address space 0 is used.</p></section><section id="id207"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">alloca</span></code>’ instruction allocates <code class="docutils literal notranslate"><span class="pre">sizeof(<type>)*NumElements</span></code>bytes of memory on the runtime stack, returning a pointer of theappropriate type to the program. If “NumElements” is specified, it isthe number of elements allocated, otherwise “NumElements” is defaultedto be one.</p><p>If a constant alignment is specified, the value result of theallocation is guaranteed to be aligned to at least that boundary. Thealignment may not be greater than <code class="docutils literal notranslate"><span class="pre">1</span> <span class="pre"><<</span> <span class="pre">32</span></code>.</p><p>The alignment is only optional when parsing textual IR; for in-memory IR,it is always present. If not specified, the target can choose to align theallocation on any convenient boundary compatible with the type.</p><p>‘<code class="docutils literal notranslate"><span class="pre">type</span></code>’ may be any sized type.</p><p>Structs containing scalable vectors cannot be used in allocas unless allfields are the same scalable vector type (e.g., <code class="docutils literal notranslate"><span class="pre">{<vscale</span> <span class="pre">x</span> <span class="pre">2</span> <span class="pre">x</span> <span class="pre">i32>,</span><span class="pre"><vscale</span> <span class="pre">x</span> <span class="pre">2</span> <span class="pre">x</span> <span class="pre">i32>}</span></code> contains the same type while <code class="docutils literal notranslate"><span class="pre">{<vscale</span> <span class="pre">x</span> <span class="pre">2</span> <span class="pre">x</span> <span class="pre">i32>,</span><span class="pre"><vscale</span> <span class="pre">x</span> <span class="pre">2</span> <span class="pre">x</span> <span class="pre">i64>}</span></code> doesn’t).</p></section><section id="id208"><h5>Semantics:</h5><p>Memory is allocated; a pointer is returned. The allocated memory isuninitialized, and loading from uninitialized memory produces an undefinedvalue. The operation itself is undefined if there is insufficient stackspace for the allocation.’<code class="docutils literal notranslate"><span class="pre">alloca</span></code>’d memory is automatically releasedwhen the function returns. The ‘<code class="docutils literal notranslate"><span class="pre">alloca</span></code>’ instruction is commonly usedto represent automatic variables that must have an address available. Whenthe function returns (either with the <code class="docutils literal notranslate"><span class="pre">ret</span></code> or <code class="docutils literal notranslate"><span class="pre">resume</span></code> instructions),the memory is reclaimed. Allocating zero bytes is legal, but the returnedpointer may not be unique. The order in which memory is allocated (ie.,which way the stack grows) is not specified.</p><p>Note that ‘<code class="docutils literal notranslate"><span class="pre">alloca</span></code>’ outside of the alloca address space from the<a class="reference internal" href="#langref-datalayout"><span class="std std-ref">datalayout string</span></a> is meaningful only if thetarget has assigned it a semantics. For targets that specify a non-zero allocaaddress space in the <a class="reference internal" href="#langref-datalayout"><span class="std std-ref">datalayout string</span></a>, the allocaaddress space needs to be explicitly specified in the instruction if it is to beused.</p><p>If the returned pointer is used by <a class="reference internal" href="#int-lifestart"><span class="std std-ref">llvm.lifetime.start</span></a>,the returned object is initially dead.See <a class="reference internal" href="#int-lifestart"><span class="std std-ref">llvm.lifetime.start</span></a> and<a class="reference internal" href="#int-lifeend"><span class="std std-ref">llvm.lifetime.end</span></a> for the precise semantics oflifetime-manipulating intrinsics.</p></section><section id="id209"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%ptr</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">alloca</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="c">; yields ptr</span><span class="nv">%ptr</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">alloca</span><span class="w"> </span><span class="kt">i32</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">4</span><span class="w"> </span><span class="c">; yields ptr</span><span class="nv">%ptr</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">alloca</span><span class="w"> </span><span class="kt">i32</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">4</span><span class="p">,</span><span class="w"> </span><span class="k">align</span><span class="w"> </span><span class="m">1024</span><span class="w"> </span><span class="c">; yields ptr</span><span class="nv">%ptr</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">alloca</span><span class="w"> </span><span class="kt">i32</span><span class="p">,</span><span class="w"> </span><span class="k">align</span><span class="w"> </span><span class="m">1024</span><span class="w"> </span><span class="c">; yields ptr</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘alloca’ instruction allocates memory on the stack frame of thecurrently executing function, to be automatically released when thisfunction returns to its caller. If the address space is not explicitlyspecified, the default address space 0 is used.`,
|
||
};
|
||
case 'LOAD':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#load-instruction`,
|
||
html: `<html><head></head><body><span id="i-load"></span><h4><a class="toc-backref" href="#id2198" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">load</span></code>’ Instruction</a></h4><section id="id210"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><result> = load [volatile] <ty>, ptr <pointer>[, align <alignment>][, !nontemporal !<nontemp_node>][, !invariant.load !<empty_node>][, !invariant.group !<empty_node>][, !nonnull !<empty_node>][, !dereferenceable !<deref_bytes_node>][, !dereferenceable_or_null !<deref_bytes_node>][, !align !<align_node>][, !noundef !<empty_node>]<result> = load atomic [volatile] <ty>, ptr <pointer> [syncscope("<target-scope>")] <ordering>, align <alignment> [, !invariant.group !<empty_node>]!<nontemp_node> = !{ i32 1 }!<empty_node> = !{}!<deref_bytes_node> = !{ i64 <dereferenceable_bytes> }!<align_node> = !{ i64 <value_alignment> }</pre></div></div></section><section id="id211"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">load</span></code>’ instruction is used to read from memory.</p></section><section id="id212"><h5>Arguments:</h5><p>The argument to the <code class="docutils literal notranslate"><span class="pre">load</span></code> instruction specifies the memory address from whichto load. The type specified must be a <a class="reference internal" href="#t-firstclass"><span class="std std-ref">first class</span></a> type ofknown size (i.e., not containing an <a class="reference internal" href="#t-opaque"><span class="std std-ref">opaque structural type</span></a>). Ifthe <code class="docutils literal notranslate"><span class="pre">load</span></code> is marked as <code class="docutils literal notranslate"><span class="pre">volatile</span></code>, then the optimizer is not allowed tomodify the number or order of execution of this <code class="docutils literal notranslate"><span class="pre">load</span></code> with other<a class="reference internal" href="#volatile"><span class="std std-ref">volatile operations</span></a>.</p><p>If the <code class="docutils literal notranslate"><span class="pre">load</span></code> is marked as <code class="docutils literal notranslate"><span class="pre">atomic</span></code>, it takes an extra <a class="reference internal" href="#ordering"><span class="std std-ref">ordering</span></a> and optional <code class="docutils literal notranslate"><span class="pre">syncscope("<target-scope>")</span></code> argument. The<code class="docutils literal notranslate"><span class="pre">release</span></code> and <code class="docutils literal notranslate"><span class="pre">acq_rel</span></code> orderings are not valid on <code class="docutils literal notranslate"><span class="pre">load</span></code> instructions.Atomic loads produce <a class="reference internal" href="#memmodel"><span class="std std-ref">defined</span></a> results when they may seemultiple atomic stores. The type of the pointee must be an integer, pointer,floating-point, or vector type whose bit width is a power of two greater thanor equal to eight. <code class="docutils literal notranslate"><span class="pre">align</span></code> must beexplicitly specified on atomic loads. Note: if the alignment is not greater orequal to the size of the <cite><value></cite> type, the atomic operation is likely torequire a lock and have poor performance. <code class="docutils literal notranslate"><span class="pre">!nontemporal</span></code> does not have anydefined semantics for atomic loads.</p><p>The optional constant <code class="docutils literal notranslate"><span class="pre">align</span></code> argument specifies the alignment of theoperation (that is, the alignment of the memory address). It is theresponsibility of the code emitter to ensure that the alignment information iscorrect. Overestimating the alignment results in undefined behavior.Underestimating the alignment may produce less efficient code. An alignment of1 is always safe. The maximum possible alignment is <code class="docutils literal notranslate"><span class="pre">1</span> <span class="pre"><<</span> <span class="pre">32</span></code>. An alignmentvalue higher than the size of the loaded type does <em>not</em> imply (without targetspecific knowledge) that memory up to the alignment value bytes can be safelyloaded without trapping.</p><p>The alignment is only optional when parsing textual IR; for in-memory IR, it isalways present. An omitted <code class="docutils literal notranslate"><span class="pre">align</span></code> argument means that the operation has theABI alignment for the target.</p><p>The optional <code class="docutils literal notranslate"><span class="pre">!nontemporal</span></code> metadata must reference a singlemetadata name <code class="docutils literal notranslate"><span class="pre"><nontemp_node></span></code> corresponding to a metadata node with one<code class="docutils literal notranslate"><span class="pre">i32</span></code> entry of value 1. The existence of the <code class="docutils literal notranslate"><span class="pre">!nontemporal</span></code>metadata on the instruction tells the optimizer and code generatorthat this load is not expected to be reused in the cache. The codegenerator may select special instructions to save cache bandwidth, suchas the <code class="docutils literal notranslate"><span class="pre">MOVNT</span></code> instruction on x86.</p><p>The optional <code class="docutils literal notranslate"><span class="pre">!invariant.load</span></code> metadata must reference a singlemetadata name <code class="docutils literal notranslate"><span class="pre"><empty_node></span></code> corresponding to a metadata node with noentries. If a load instruction tagged with the <code class="docutils literal notranslate"><span class="pre">!invariant.load</span></code>metadata is executed, the memory location referenced by the load hasto contain the same value at all points in the program where thememory location is dereferenceable; otherwise, the behavior isundefined.</p><dl class="simple"><dt>The optional <code class="docutils literal notranslate"><span class="pre">!invariant.group</span></code> metadata must reference a single metadata name</dt><dd><p><code class="docutils literal notranslate"><span class="pre"><empty_node></span></code> corresponding to a metadata node with no entries.See <code class="docutils literal notranslate"><span class="pre">invariant.group</span></code> metadata <a class="reference internal" href="#md-invariant-group"><span class="std std-ref">invariant.group</span></a>.</p></dd></dl><p>The optional <code class="docutils literal notranslate"><span class="pre">!nonnull</span></code> metadata must reference a singlemetadata name <code class="docutils literal notranslate"><span class="pre"><empty_node></span></code> corresponding to a metadata node with noentries. The existence of the <code class="docutils literal notranslate"><span class="pre">!nonnull</span></code> metadata on theinstruction tells the optimizer that the value loaded is known tonever be null. If the value is null at runtime, a poison value is returnedinstead. This is analogous to the <code class="docutils literal notranslate"><span class="pre">nonnull</span></code> attribute on parameters andreturn values. This metadata can only be applied to loads of a pointer type.</p><p>The optional <code class="docutils literal notranslate"><span class="pre">!dereferenceable</span></code> metadata must reference a single metadataname <code class="docutils literal notranslate"><span class="pre"><deref_bytes_node></span></code> corresponding to a metadata node with one <code class="docutils literal notranslate"><span class="pre">i64</span></code>entry.See <code class="docutils literal notranslate"><span class="pre">dereferenceable</span></code> metadata <a class="reference internal" href="#md-dereferenceable"><span class="std std-ref">dereferenceable</span></a>.</p><p>The optional <code class="docutils literal notranslate"><span class="pre">!dereferenceable_or_null</span></code> metadata must reference a singlemetadata name <code class="docutils literal notranslate"><span class="pre"><deref_bytes_node></span></code> corresponding to a metadata node with one<code class="docutils literal notranslate"><span class="pre">i64</span></code> entry.See <code class="docutils literal notranslate"><span class="pre">dereferenceable_or_null</span></code> metadata <a class="reference internal" href="#md-dereferenceable-or-null"><span class="std std-ref">dereferenceable_or_null</span></a>.</p><p>The optional <code class="docutils literal notranslate"><span class="pre">!align</span></code> metadata must reference a single metadata name<code class="docutils literal notranslate"><span class="pre"><align_node></span></code> corresponding to a metadata node with one <code class="docutils literal notranslate"><span class="pre">i64</span></code> entry.The existence of the <code class="docutils literal notranslate"><span class="pre">!align</span></code> metadata on the instruction tells theoptimizer that the value loaded is known to be aligned to a boundary specifiedby the integer value in the metadata node. The alignment must be a power of 2.This is analogous to the ‘’align’’ attribute on parameters and return values.This metadata can only be applied to loads of a pointer type. If the returnedvalue is not appropriately aligned at runtime, a poison value is returnedinstead.</p><p>The optional <code class="docutils literal notranslate"><span class="pre">!noundef</span></code> metadata must reference a single metadata name<code class="docutils literal notranslate"><span class="pre"><empty_node></span></code> corresponding to a node with no entries. The existence of<code class="docutils literal notranslate"><span class="pre">!noundef</span></code> metadata on the instruction tells the optimizer that the valueloaded is known to be <a class="reference internal" href="#welldefinedvalues"><span class="std std-ref">well defined</span></a>.If the value isn’t well defined, the behavior is undefined. If the <code class="docutils literal notranslate"><span class="pre">!noundef</span></code>metadata is combined with poison-generating metadata like <code class="docutils literal notranslate"><span class="pre">!nonnull</span></code>,violation of that metadata constraint will also result in undefined behavior.</p></section><section id="id213"><h5>Semantics:</h5><p>The location of memory pointed to is loaded. If the value being loadedis of scalar type then the number of bytes read does not exceed theminimum number of bytes needed to hold all bits of the type. Forexample, loading an <code class="docutils literal notranslate"><span class="pre">i24</span></code> reads at most three bytes. When loading avalue of a type like <code class="docutils literal notranslate"><span class="pre">i20</span></code> with a size that is not an integral numberof bytes, the result is undefined if the value was not originallywritten using a store of the same type.If the value being loaded is of aggregate type, the bytes that correspond topadding may be accessed but are ignored, because it is impossible to observepadding from the loaded aggregate value.If <code class="docutils literal notranslate"><span class="pre"><pointer></span></code> is not a well-defined value, the behavior is undefined.</p></section><section id="id214"><h5>Examples:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%ptr</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">alloca</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="c">; yields ptr</span><span class="k">store</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">3</span><span class="p">,</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%ptr</span><span class="w"> </span><span class="c">; yields void</span><span class="nv">%val</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">load</span><span class="w"> </span><span class="kt">i32</span><span class="p">,</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%ptr</span><span class="w"> </span><span class="c">; yields i32:val = i32 3</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘load’ instruction is used to read from memory.`,
|
||
};
|
||
case 'STORE':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#store-instruction`,
|
||
html: `<html><head></head><body><span id="i-store"></span><h4><a class="toc-backref" href="#id2199" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">store</span></code>’ Instruction</a></h4><section id="id215"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span>store [volatile] <ty> <value>, ptr <pointer>[, align <alignment>][, !nontemporal !<nontemp_node>][, !invariant.group !<empty_node>] ; yields voidstore atomic [volatile] <ty> <value>, ptr <pointer> [syncscope("<target-scope>")] <ordering>, align <alignment> [, !invariant.group !<empty_node>] ; yields void!<nontemp_node> = !{ i32 1 }!<empty_node> = !{}</pre></div></div></section><section id="id216"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">store</span></code>’ instruction is used to write to memory.</p></section><section id="id217"><h5>Arguments:</h5><p>There are two arguments to the <code class="docutils literal notranslate"><span class="pre">store</span></code> instruction: a value to store and anaddress at which to store it. The type of the <code class="docutils literal notranslate"><span class="pre"><pointer></span></code> operand must be apointer to the <a class="reference internal" href="#t-firstclass"><span class="std std-ref">first class</span></a> type of the <code class="docutils literal notranslate"><span class="pre"><value></span></code>operand. If the <code class="docutils literal notranslate"><span class="pre">store</span></code> is marked as <code class="docutils literal notranslate"><span class="pre">volatile</span></code>, then the optimizer is notallowed to modify the number or order of execution of this <code class="docutils literal notranslate"><span class="pre">store</span></code> with other<a class="reference internal" href="#volatile"><span class="std std-ref">volatile operations</span></a>. Only values of <a class="reference internal" href="#t-firstclass"><span class="std std-ref">first class</span></a> types of known size (i.e., not containing an <a class="reference internal" href="#t-opaque"><span class="std std-ref">opaquestructural type</span></a>) can be stored.</p><p>If the <code class="docutils literal notranslate"><span class="pre">store</span></code> is marked as <code class="docutils literal notranslate"><span class="pre">atomic</span></code>, it takes an extra <a class="reference internal" href="#ordering"><span class="std std-ref">ordering</span></a> and optional <code class="docutils literal notranslate"><span class="pre">syncscope("<target-scope>")</span></code> argument. The<code class="docutils literal notranslate"><span class="pre">acquire</span></code> and <code class="docutils literal notranslate"><span class="pre">acq_rel</span></code> orderings aren’t valid on <code class="docutils literal notranslate"><span class="pre">store</span></code> instructions.Atomic loads produce <a class="reference internal" href="#memmodel"><span class="std std-ref">defined</span></a> results when they may seemultiple atomic stores. The type of the pointee must be an integer, pointer,floating-point, or vector type whose bit width is a power of two greater thanor equal to eight. <code class="docutils literal notranslate"><span class="pre">align</span></code> must beexplicitly specified on atomic stores. Note: if the alignment is not greater orequal to the size of the <cite><value></cite> type, the atomic operation is likely torequire a lock and have poor performance. <code class="docutils literal notranslate"><span class="pre">!nontemporal</span></code> does not have anydefined semantics for atomic stores.</p><p>The optional constant <code class="docutils literal notranslate"><span class="pre">align</span></code> argument specifies the alignment of theoperation (that is, the alignment of the memory address). It is theresponsibility of the code emitter to ensure that the alignment information iscorrect. Overestimating the alignment results in undefined behavior.Underestimating the alignment may produce less efficient code. An alignment of1 is always safe. The maximum possible alignment is <code class="docutils literal notranslate"><span class="pre">1</span> <span class="pre"><<</span> <span class="pre">32</span></code>. An alignmentvalue higher than the size of the stored type does <em>not</em> imply (without targetspecific knowledge) that memory up to the alignment value bytes can be safelyloaded without trapping.</p><p>The alignment is only optional when parsing textual IR; for in-memory IR, it isalways present. An omitted <code class="docutils literal notranslate"><span class="pre">align</span></code> argument means that the operation has theABI alignment for the target.</p><p>The optional <code class="docutils literal notranslate"><span class="pre">!nontemporal</span></code> metadata must reference a single metadataname <code class="docutils literal notranslate"><span class="pre"><nontemp_node></span></code> corresponding to a metadata node with one <code class="docutils literal notranslate"><span class="pre">i32</span></code> entryof value 1. The existence of the <code class="docutils literal notranslate"><span class="pre">!nontemporal</span></code> metadata on the instructiontells the optimizer and code generator that this load is not expected tobe reused in the cache. The code generator may select specialinstructions to save cache bandwidth, such as the <code class="docutils literal notranslate"><span class="pre">MOVNT</span></code> instruction onx86.</p><p>The optional <code class="docutils literal notranslate"><span class="pre">!invariant.group</span></code> metadata must reference asingle metadata name <code class="docutils literal notranslate"><span class="pre"><empty_node></span></code>. See <code class="docutils literal notranslate"><span class="pre">invariant.group</span></code> metadata.</p></section><section id="id218"><h5>Semantics:</h5><p>The contents of memory are updated to contain <code class="docutils literal notranslate"><span class="pre"><value></span></code> at thelocation specified by the <code class="docutils literal notranslate"><span class="pre"><pointer></span></code> operand. If <code class="docutils literal notranslate"><span class="pre"><value></span></code> isof scalar type then the number of bytes written does not exceed theminimum number of bytes needed to hold all bits of the type. Forexample, storing an <code class="docutils literal notranslate"><span class="pre">i24</span></code> writes at most three bytes. When writing avalue of a type like <code class="docutils literal notranslate"><span class="pre">i20</span></code> with a size that is not an integral numberof bytes, it is unspecified what happens to the extra bits that do notbelong to the type, but they will typically be overwritten.If <code class="docutils literal notranslate"><span class="pre"><value></span></code> is of aggregate type, padding is filled with<a class="reference internal" href="#undefvalues"><span class="std std-ref">undef</span></a>.If <code class="docutils literal notranslate"><span class="pre"><pointer></span></code> is not a well-defined value, the behavior is undefined.</p></section><section id="id219"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%ptr</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">alloca</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="c">; yields ptr</span><span class="k">store</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">3</span><span class="p">,</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%ptr</span><span class="w"> </span><span class="c">; yields void</span><span class="nv">%val</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">load</span><span class="w"> </span><span class="kt">i32</span><span class="p">,</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%ptr</span><span class="w"> </span><span class="c">; yields i32:val = i32 3</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘store’ instruction is used to write to memory.`,
|
||
};
|
||
case 'FENCE':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#fence-instruction`,
|
||
html: `<html><head></head><body><span id="i-fence"></span><h4><a class="toc-backref" href="#id2200" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">fence</span></code>’ Instruction</a></h4><section id="id220"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">fence</span> <span class="p">[</span><span class="n">syncscope</span><span class="p">(</span><span class="s2">"<target-scope>"</span><span class="p">)]</span> <span class="o"><</span><span class="n">ordering</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">void</span></pre></div></div></section><section id="id221"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fence</span></code>’ instruction is used to introduce happens-before edgesbetween operations.</p></section><section id="id222"><h5>Arguments:</h5><p>‘<code class="docutils literal notranslate"><span class="pre">fence</span></code>’ instructions take an <a class="reference internal" href="#ordering"><span class="std std-ref">ordering</span></a> argument whichdefines what <em>synchronizes-with</em> edges they add. They can only be given<code class="docutils literal notranslate"><span class="pre">acquire</span></code>, <code class="docutils literal notranslate"><span class="pre">release</span></code>, <code class="docutils literal notranslate"><span class="pre">acq_rel</span></code>, and <code class="docutils literal notranslate"><span class="pre">seq_cst</span></code> orderings.</p></section><section id="id223"><h5>Semantics:</h5><p>A fence A which has (at least) <code class="docutils literal notranslate"><span class="pre">release</span></code> ordering semantics<em>synchronizes with</em> a fence B with (at least) <code class="docutils literal notranslate"><span class="pre">acquire</span></code> orderingsemantics if and only if there exist atomic operations X and Y, bothoperating on some atomic object M, such that A is sequenced before X, Xmodifies M (either directly or through some side effect of a sequenceheaded by X), Y is sequenced before B, and Y observes M. This provides a<em>happens-before</em> dependency between A and B. Rather than an explicit<code class="docutils literal notranslate"><span class="pre">fence</span></code>, one (but not both) of the atomic operations X or Y mightprovide a <code class="docutils literal notranslate"><span class="pre">release</span></code> or <code class="docutils literal notranslate"><span class="pre">acquire</span></code> (resp.) ordering constraint andstill <em>synchronize-with</em> the explicit <code class="docutils literal notranslate"><span class="pre">fence</span></code> and establish the<em>happens-before</em> edge.</p><p>A <code class="docutils literal notranslate"><span class="pre">fence</span></code> which has <code class="docutils literal notranslate"><span class="pre">seq_cst</span></code> ordering, in addition to having both<code class="docutils literal notranslate"><span class="pre">acquire</span></code> and <code class="docutils literal notranslate"><span class="pre">release</span></code> semantics specified above, participates inthe global program order of other <code class="docutils literal notranslate"><span class="pre">seq_cst</span></code> operations and/orfences. Furthermore, the global ordering created by a <code class="docutils literal notranslate"><span class="pre">seq_cst</span></code>fence must be compatible with the individual total orders of<code class="docutils literal notranslate"><span class="pre">monotonic</span></code> (or stronger) memory accesses occurring before and aftersuch a fence. The exact semantics of this interaction are somewhatcomplicated, see the C++ standard’s <a class="reference external" href="https://wg21.link/atomics.order">[atomics.order]</a> section for more details.</p><p>A <code class="docutils literal notranslate"><span class="pre">fence</span></code> instruction can also take an optional“<a class="reference internal" href="#syncscope"><span class="std std-ref">syncscope</span></a>” argument.</p></section><section id="id225"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span>fence acquire ; yields voidfence syncscope("singlethread") seq_cst ; yields voidfence syncscope("agent") seq_cst ; yields void</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘fence’ instruction is used to introduce happens-before edgesbetween operations.`,
|
||
};
|
||
case 'CMPXCHG':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#cmpxchg-instruction`,
|
||
html: `<html><head></head><body><span id="i-cmpxchg"></span><h4><a class="toc-backref" href="#id2201" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">cmpxchg</span></code>’ Instruction</a></h4><section id="id226"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">cmpxchg</span> <span class="p">[</span><span class="n">weak</span><span class="p">]</span> <span class="p">[</span><span class="n">volatile</span><span class="p">]</span> <span class="n">ptr</span> <span class="o"><</span><span class="n">pointer</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">cmp</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">new</span><span class="o">></span> <span class="p">[</span><span class="n">syncscope</span><span class="p">(</span><span class="s2">"<target-scope>"</span><span class="p">)]</span> <span class="o"><</span><span class="n">success</span> <span class="n">ordering</span><span class="o">></span> <span class="o"><</span><span class="n">failure</span> <span class="n">ordering</span><span class="o">></span><span class="p">[,</span> <span class="n">align</span> <span class="o"><</span><span class="n">alignment</span><span class="o">></span><span class="p">]</span> <span class="p">;</span> <span class="n">yields</span> <span class="p">{</span> <span class="n">ty</span><span class="p">,</span> <span class="n">i1</span> <span class="p">}</span></pre></div></div></section><section id="id227"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">cmpxchg</span></code>’ instruction is used to atomically modify memory. Itloads a value in memory and compares it to a given value. If they areequal, it tries to store a new value into the memory.</p></section><section id="id228"><h5>Arguments:</h5><p>There are three arguments to the ‘<code class="docutils literal notranslate"><span class="pre">cmpxchg</span></code>’ instruction: an addressto operate on, a value to compare to the value currently be at thataddress, and a new value to place at that address if the compared valuesare equal. The type of ‘<cmp>’ must be an integer or pointer type whosebit width is a power of two greater than or equal to eight.‘<cmp>’ and ‘<new>’ musthave the same type, and the type of ‘<pointer>’ must be a pointer tothat type. If the <code class="docutils literal notranslate"><span class="pre">cmpxchg</span></code> is marked as <code class="docutils literal notranslate"><span class="pre">volatile</span></code>, then theoptimizer is not allowed to modify the number or order of execution ofthis <code class="docutils literal notranslate"><span class="pre">cmpxchg</span></code> with other <a class="reference internal" href="#volatile"><span class="std std-ref">volatile operations</span></a>.</p><p>The success and failure <a class="reference internal" href="#ordering"><span class="std std-ref">ordering</span></a> arguments specify how this<code class="docutils literal notranslate"><span class="pre">cmpxchg</span></code> synchronizes with other atomic operations. Both ordering parametersmust be at least <code class="docutils literal notranslate"><span class="pre">monotonic</span></code>, the failure ordering cannot be either<code class="docutils literal notranslate"><span class="pre">release</span></code> or <code class="docutils literal notranslate"><span class="pre">acq_rel</span></code>.</p><p>A <code class="docutils literal notranslate"><span class="pre">cmpxchg</span></code> instruction can also take an optional“<a class="reference internal" href="#syncscope"><span class="std std-ref">syncscope</span></a>” argument.</p><p>Note: if the alignment is not greater or equal to the size of the <cite><value></cite>type, the atomic operation is likely to require a lock and have poorperformance.</p><p>The alignment is only optional when parsing textual IR; for in-memory IR, it isalways present. If unspecified, the alignment is assumed to be equal to thesize of the ‘<value>’ type. Note that this default alignment assumption isdifferent from the alignment used for the load/store instructions when alignisn’t specified.</p><p>The pointer passed into cmpxchg must have alignment greater than orequal to the size in memory of the operand.</p></section><section id="id229"><h5>Semantics:</h5><p>The contents of memory at the location specified by the ‘<code class="docutils literal notranslate"><span class="pre"><pointer></span></code>’ operandis read and compared to ‘<code class="docutils literal notranslate"><span class="pre"><cmp></span></code>’; if the values are equal, ‘<code class="docutils literal notranslate"><span class="pre"><new></span></code>’ iswritten to the location. The original value at the location is returned,together with a flag indicating success (true) or failure (false).</p><p>If the cmpxchg operation is marked as <code class="docutils literal notranslate"><span class="pre">weak</span></code> then a spurious failure ispermitted: the operation may not write <code class="docutils literal notranslate"><span class="pre"><new></span></code> even if the comparisonmatched.</p><p>If the cmpxchg operation is strong (the default), the i1 value is 1 if and onlyif the value loaded equals <code class="docutils literal notranslate"><span class="pre">cmp</span></code>.</p><p>A successful <code class="docutils literal notranslate"><span class="pre">cmpxchg</span></code> is a read-modify-write instruction for the purpose ofidentifying release sequences. A failed <code class="docutils literal notranslate"><span class="pre">cmpxchg</span></code> is equivalent to an atomicload with an ordering parameter determined the second ordering parameter.</p></section><section id="id230"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nl">entry:</span><span class="w"> </span><span class="nv">%orig</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">load</span><span class="w"> </span><span class="k">atomic</span><span class="w"> </span><span class="kt">i32</span><span class="p">,</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%ptr</span><span class="w"> </span><span class="k">unordered</span><span class="p">,</span><span class="w"> </span><span class="k">align</span><span class="w"> </span><span class="m">4</span><span class="w"> </span><span class="c">; yields i32</span><span class="w"> </span><span class="k">br</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%loop</span><span class="nl">loop:</span><span class="w"> </span><span class="nv">%cmp</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">phi</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="p">[</span><span class="w"> </span><span class="nv">%orig</span><span class="p">,</span><span class="w"> </span><span class="nv">%entry</span><span class="w"> </span><span class="p">],</span><span class="w"> </span><span class="p">[</span><span class="nv">%value_loaded</span><span class="p">,</span><span class="w"> </span><span class="nv">%loop</span><span class="p">]</span><span class="w"> </span><span class="nv">%squared</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">mul</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="nv">%cmp</span><span class="p">,</span><span class="w"> </span><span class="nv">%cmp</span><span class="w"> </span><span class="nv">%val_success</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">cmpxchg</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%ptr</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="nv">%cmp</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="nv">%squared</span><span class="w"> </span><span class="k">acq_rel</span><span class="w"> </span><span class="k">monotonic</span><span class="w"> </span><span class="c">; yields { i32, i1 }</span><span class="w"> </span><span class="nv">%value_loaded</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">extractvalue</span><span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="kt">i32</span><span class="p">,</span><span class="w"> </span><span class="kt">i1</span><span class="w"> </span><span class="p">}</span><span class="w"> </span><span class="nv">%val_success</span><span class="p">,</span><span class="w"> </span><span class="m">0</span><span class="w"> </span><span class="nv">%success</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">extractvalue</span><span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="kt">i32</span><span class="p">,</span><span class="w"> </span><span class="kt">i1</span><span class="w"> </span><span class="p">}</span><span class="w"> </span><span class="nv">%val_success</span><span class="p">,</span><span class="w"> </span><span class="m">1</span><span class="w"> </span><span class="k">br</span><span class="w"> </span><span class="kt">i1</span><span class="w"> </span><span class="nv">%success</span><span class="p">,</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%done</span><span class="p">,</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%loop</span><span class="nl">done:</span><span class="w"> </span><span class="p">...</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘cmpxchg’ instruction is used to atomically modify memory. Itloads a value in memory and compares it to a given value. If they areequal, it tries to store a new value into the memory.`,
|
||
};
|
||
case 'ATOMICRMW':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#atomicrmw-instruction`,
|
||
html: `<html><head></head><body><span id="i-atomicrmw"></span><h4><a class="toc-backref" href="#id2202" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">atomicrmw</span></code>’ Instruction</a></h4><section id="id231"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">atomicrmw</span> <span class="p">[</span><span class="n">volatile</span><span class="p">]</span> <span class="o"><</span><span class="n">operation</span><span class="o">></span> <span class="n">ptr</span> <span class="o"><</span><span class="n">pointer</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="p">[</span><span class="n">syncscope</span><span class="p">(</span><span class="s2">"<target-scope>"</span><span class="p">)]</span> <span class="o"><</span><span class="n">ordering</span><span class="o">></span><span class="p">[,</span> <span class="n">align</span> <span class="o"><</span><span class="n">alignment</span><span class="o">></span><span class="p">]</span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span></pre></div></div></section><section id="id232"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">atomicrmw</span></code>’ instruction is used to atomically modify memory.</p></section><section id="id233"><h5>Arguments:</h5><p>There are three arguments to the ‘<code class="docutils literal notranslate"><span class="pre">atomicrmw</span></code>’ instruction: anoperation to apply, an address whose value to modify, an argument to theoperation. The operation must be one of the following keywords:</p><ul class="simple"><li><p>xchg</p></li><li><p>add</p></li><li><p>sub</p></li><li><p>and</p></li><li><p>nand</p></li><li><p>or</p></li><li><p>xor</p></li><li><p>max</p></li><li><p>min</p></li><li><p>umax</p></li><li><p>umin</p></li><li><p>fadd</p></li><li><p>fsub</p></li><li><p>fmax</p></li><li><p>fmin</p></li><li><p>fmaximum</p></li><li><p>fminimum</p></li><li><p>uinc_wrap</p></li><li><p>udec_wrap</p></li><li><p>usub_cond</p></li><li><p>usub_sat</p></li></ul><p>For most of these operations, the type of ‘<value>’ must be an integertype whose bit width is a power of two greater than or equal to eight.For xchg, thismay also be a floating point or a pointer type with the same size constraintsas integers. For fadd/fsub/fmax/fmin/fmaximum/fminimum, this must be a floating-pointor fixed vector of floating-point type. The type of the ‘<code class="docutils literal notranslate"><span class="pre"><pointer></span></code>’operand must be a pointer to that type. If the <code class="docutils literal notranslate"><span class="pre">atomicrmw</span></code> is markedas <code class="docutils literal notranslate"><span class="pre">volatile</span></code>, then the optimizer is not allowed to modify thenumber or order of execution of this <code class="docutils literal notranslate"><span class="pre">atomicrmw</span></code> with other<a class="reference internal" href="#volatile"><span class="std std-ref">volatile operations</span></a>.</p><p>Note: if the alignment is not greater or equal to the size of the <cite><value></cite>type, the atomic operation is likely to require a lock and have poorperformance.</p><p>The alignment is only optional when parsing textual IR; for in-memory IR, it isalways present. If unspecified, the alignment is assumed to be equal to thesize of the ‘<value>’ type. Note that this default alignment assumption isdifferent from the alignment used for the load/store instructions when alignisn’t specified.</p><p>An <code class="docutils literal notranslate"><span class="pre">atomicrmw</span></code> instruction can also take an optional“<a class="reference internal" href="#syncscope"><span class="std std-ref">syncscope</span></a>” argument.</p></section><section id="id234"><h5>Semantics:</h5><p>The contents of memory at the location specified by the ‘<code class="docutils literal notranslate"><span class="pre"><pointer></span></code>’operand are atomically read, modified, and written back. The originalvalue at the location is returned. The modification is specified by theoperation argument:</p><ul class="simple"><li><p>xchg: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">val</span></code></p></li><li><p>add: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">*ptr</span> <span class="pre">+</span> <span class="pre">val</span></code></p></li><li><p>sub: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">*ptr</span> <span class="pre">-</span> <span class="pre">val</span></code></p></li><li><p>and: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">*ptr</span> <span class="pre">&</span> <span class="pre">val</span></code></p></li><li><p>nand: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">~(*ptr</span> <span class="pre">&</span> <span class="pre">val)</span></code></p></li><li><p>or: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">*ptr</span> <span class="pre">|</span> <span class="pre">val</span></code></p></li><li><p>xor: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">*ptr</span> <span class="pre">^</span> <span class="pre">val</span></code></p></li><li><p>max: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">*ptr</span> <span class="pre">></span> <span class="pre">val</span> <span class="pre">?</span> <span class="pre">*ptr</span> <span class="pre">:</span> <span class="pre">val</span></code> (using a signed comparison)</p></li><li><p>min: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">*ptr</span> <span class="pre"><</span> <span class="pre">val</span> <span class="pre">?</span> <span class="pre">*ptr</span> <span class="pre">:</span> <span class="pre">val</span></code> (using a signed comparison)</p></li><li><p>umax: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">*ptr</span> <span class="pre">></span> <span class="pre">val</span> <span class="pre">?</span> <span class="pre">*ptr</span> <span class="pre">:</span> <span class="pre">val</span></code> (using an unsigned comparison)</p></li><li><p>umin: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">*ptr</span> <span class="pre"><</span> <span class="pre">val</span> <span class="pre">?</span> <span class="pre">*ptr</span> <span class="pre">:</span> <span class="pre">val</span></code> (using an unsigned comparison)</p></li><li><p>fadd: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">*ptr</span> <span class="pre">+</span> <span class="pre">val</span></code> (using floating point arithmetic)</p></li><li><p>fsub: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">*ptr</span> <span class="pre">-</span> <span class="pre">val</span></code> (using floating point arithmetic)</p></li><li><p>fmax: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">maxnum(*ptr,</span> <span class="pre">val)</span></code> (match the <cite>llvm.maxnum.*</cite> intrinsic)</p></li><li><p>fmin: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">minnum(*ptr,</span> <span class="pre">val)</span></code> (match the <cite>llvm.minnum.*</cite> intrinsic)</p></li><li><p>fmaximum: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">maximum(*ptr,</span> <span class="pre">val)</span></code> (match the <cite>llvm.maximum.*</cite> intrinsic)</p></li><li><p>fminimum: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">minimum(*ptr,</span> <span class="pre">val)</span></code> (match the <cite>llvm.minimum.*</cite> intrinsic)</p></li><li><p>uinc_wrap: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">(*ptr</span> <span class="pre">u>=</span> <span class="pre">val)</span> <span class="pre">?</span> <span class="pre">0</span> <span class="pre">:</span> <span class="pre">(*ptr</span> <span class="pre">+</span> <span class="pre">1)</span></code> (increment value with wraparound to zero when incremented above input value)</p></li><li><p>udec_wrap: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">((*ptr</span> <span class="pre">==</span> <span class="pre">0)</span> <span class="pre">||</span> <span class="pre">(*ptr</span> <span class="pre">u></span> <span class="pre">val))</span> <span class="pre">?</span> <span class="pre">val</span> <span class="pre">:</span> <span class="pre">(*ptr</span> <span class="pre">-</span> <span class="pre">1)</span></code> (decrement with wraparound to input value when decremented below zero).</p></li><li><p>usub_cond: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">(*ptr</span> <span class="pre">u>=</span> <span class="pre">val)</span> <span class="pre">?</span> <span class="pre">*ptr</span> <span class="pre">-</span> <span class="pre">val</span> <span class="pre">:</span> <span class="pre">*ptr</span></code> (subtract only if no unsigned overflow).</p></li><li><p>usub_sat: <code class="docutils literal notranslate"><span class="pre">*ptr</span> <span class="pre">=</span> <span class="pre">(*ptr</span> <span class="pre">u>=</span> <span class="pre">val)</span> <span class="pre">?</span> <span class="pre">*ptr</span> <span class="pre">-</span> <span class="pre">val</span> <span class="pre">:</span> <span class="pre">0</span></code> (subtract with unsigned clamping to zero).</p></li></ul></section><section id="id235"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%old</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">atomicrmw</span><span class="w"> </span><span class="k">add</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%ptr</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">1</span><span class="w"> </span><span class="k">acquire</span><span class="w"> </span><span class="c">; yields i32</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘atomicrmw’ instruction is used to atomically modify memory.`,
|
||
};
|
||
case 'GETELEMENTPTR':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#getelementptr-instruction`,
|
||
html: `<html><head></head><body><span id="i-getelementptr"></span><h4><a class="toc-backref" href="#id2203" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">getelementptr</span></code>’ Instruction</a></h4><section id="id236"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">getelementptr</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span><span class="p">,</span> <span class="n">ptr</span> <span class="o"><</span><span class="n">ptrval</span><span class="o">></span><span class="p">{,</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">idx</span><span class="o">></span><span class="p">}</span><span class="o">*</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">getelementptr</span> <span class="n">inbounds</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span><span class="p">,</span> <span class="n">ptr</span> <span class="o"><</span><span class="n">ptrval</span><span class="o">></span><span class="p">{,</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">idx</span><span class="o">></span><span class="p">}</span><span class="o">*</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">getelementptr</span> <span class="n">nusw</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span><span class="p">,</span> <span class="n">ptr</span> <span class="o"><</span><span class="n">ptrval</span><span class="o">></span><span class="p">{,</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">idx</span><span class="o">></span><span class="p">}</span><span class="o">*</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">getelementptr</span> <span class="n">nuw</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span><span class="p">,</span> <span class="n">ptr</span> <span class="o"><</span><span class="n">ptrval</span><span class="o">></span><span class="p">{,</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">idx</span><span class="o">></span><span class="p">}</span><span class="o">*</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">getelementptr</span> <span class="n">inrange</span><span class="p">(</span><span class="n">S</span><span class="p">,</span><span class="n">E</span><span class="p">)</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span><span class="p">,</span> <span class="n">ptr</span> <span class="o"><</span><span class="n">ptrval</span><span class="o">></span><span class="p">{,</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">idx</span><span class="o">></span><span class="p">}</span><span class="o">*</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">getelementptr</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">N</span> <span class="n">x</span> <span class="n">ptr</span><span class="o">></span> <span class="o"><</span><span class="n">ptrval</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">vector</span> <span class="n">index</span> <span class="nb">type</span><span class="o">></span> <span class="o"><</span><span class="n">idx</span><span class="o">></span></pre></div></div></section><section id="id237"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">getelementptr</span></code>’ instruction is used to get the address of asubelement of an <a class="reference internal" href="#t-aggregate"><span class="std std-ref">aggregate</span></a> data structure. It performsaddress calculation only and does not access memory. The instruction can alsobe used to calculate a vector of such addresses.</p></section><section id="id238"><h5>Arguments:</h5><p>The first argument is always a type used as the basis for the calculations.The second argument is always a pointer or a vector of pointers, and is thebase address to start from. The remaining arguments are indicesthat indicate which of the elements of the aggregate object are indexed.The interpretation of each index is dependent on the type being indexedinto. The first index always indexes the pointer value given as thesecond argument, the second index indexes a value of the type pointed to(not necessarily the value directly pointed to, since the first indexcan be non-zero), etc. The first type indexed into must be a pointervalue, subsequent types can be arrays, vectors, and structs. Note thatsubsequent types being indexed into can never be pointers, since thatwould require loading the pointer before continuing calculation.</p><p>The type of each index argument depends on the type it is indexing into.When indexing into a (optionally packed) structure, only <code class="docutils literal notranslate"><span class="pre">i32</span></code> integer<strong>constants</strong> are allowed (when using a vector of indices they must allbe the <strong>same</strong> <code class="docutils literal notranslate"><span class="pre">i32</span></code> integer constant). When indexing into an array,pointer or vector, integers of any width are allowed, and they are notrequired to be constant. These integers are treated as signed valueswhere relevant.</p><p>For example, let’s consider a C code fragment and how it gets compiledto LLVM:</p><div class="highlight-c notranslate"><div class="highlight"><pre><span></span><span class="k">struct</span><span class="w"> </span><span class="nc">RT</span><span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="kt">char</span><span class="w"> </span><span class="n">A</span><span class="p">;</span><span class="w"> </span><span class="kt">int</span><span class="w"> </span><span class="n">B</span><span class="p">[</span><span class="mi">10</span><span class="p">][</span><span class="mi">20</span><span class="p">];</span><span class="w"> </span><span class="kt">char</span><span class="w"> </span><span class="n">C</span><span class="p">;</span><span class="p">};</span><span class="k">struct</span><span class="w"> </span><span class="nc">ST</span><span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="kt">int</span><span class="w"> </span><span class="n">X</span><span class="p">;</span><span class="w"> </span><span class="kt">double</span><span class="w"> </span><span class="n">Y</span><span class="p">;</span><span class="w"> </span><span class="k">struct</span><span class="w"> </span><span class="nc">RT</span><span class="w"> </span><span class="n">Z</span><span class="p">;</span><span class="p">};</span><span class="kt">int</span><span class="w"> </span><span class="o">*</span><span class="nf">foo</span><span class="p">(</span><span class="k">struct</span><span class="w"> </span><span class="nc">ST</span><span class="w"> </span><span class="o">*</span><span class="n">s</span><span class="p">)</span><span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="k">return</span><span class="w"> </span><span class="o">&</span><span class="n">s</span><span class="p">[</span><span class="mi">1</span><span class="p">].</span><span class="n">Z</span><span class="p">.</span><span class="n">B</span><span class="p">[</span><span class="mi">5</span><span class="p">][</span><span class="mi">13</span><span class="p">];</span><span class="p">}</span></pre></div></div><p>The LLVM code generated by Clang is approximately:</p><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%struct.RT</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">type</span><span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="kt">i8</span><span class="p">,</span><span class="w"> </span><span class="p">[</span><span class="m">10</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="p">[</span><span class="m">20</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i32</span><span class="p">]],</span><span class="w"> </span><span class="kt">i8</span><span class="w"> </span><span class="p">}</span><span class="nv">%struct.ST</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">type</span><span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="kt">i32</span><span class="p">,</span><span class="w"> </span><span class="kt">double</span><span class="p">,</span><span class="w"> </span><span class="nv">%struct.RT</span><span class="w"> </span><span class="p">}</span><span class="k">define</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="vg">@foo</span><span class="p">(</span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%s</span><span class="p">)</span><span class="w"> </span><span class="p">{</span><span class="nl">entry:</span><span class="w"> </span><span class="nv">%arrayidx</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">getelementptr</span><span class="w"> </span><span class="k">inbounds</span><span class="w"> </span><span class="nv">%struct.ST</span><span class="p">,</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%s</span><span class="p">,</span><span class="w"> </span><span class="kt">i64</span><span class="w"> </span><span class="m">1</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">2</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">1</span><span class="p">,</span><span class="w"> </span><span class="kt">i64</span><span class="w"> </span><span class="m">5</span><span class="p">,</span><span class="w"> </span><span class="kt">i64</span><span class="w"> </span><span class="m">13</span><span class="w"> </span><span class="k">ret</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%arrayidx</span><span class="p">}</span></pre></div></div></section><section id="id239"><h5>Semantics:</h5><p>In the example above, the first index is indexing into the‘<code class="docutils literal notranslate"><span class="pre">%struct.ST*</span></code>’ type, which is a pointer, yielding a ‘<code class="docutils literal notranslate"><span class="pre">%struct.ST</span></code>’= ‘<code class="docutils literal notranslate"><span class="pre">{</span> <span class="pre">i32,</span> <span class="pre">double,</span> <span class="pre">%struct.RT</span> <span class="pre">}</span></code>’ type, a structure. The second indexindexes into the third element of the structure, yielding a‘<code class="docutils literal notranslate"><span class="pre">%struct.RT</span></code>’ = ‘<code class="docutils literal notranslate"><span class="pre">{</span> <span class="pre">i8</span> <span class="pre">,</span> <span class="pre">[10</span> <span class="pre">x</span> <span class="pre">[20</span> <span class="pre">x</span> <span class="pre">i32]],</span> <span class="pre">i8</span> <span class="pre">}</span></code>’ type, anotherstructure. The third index indexes into the second element of thestructure, yielding a ‘<code class="docutils literal notranslate"><span class="pre">[10</span> <span class="pre">x</span> <span class="pre">[20</span> <span class="pre">x</span> <span class="pre">i32]]</span></code>’ type, an array. The twodimensions of the array are subscripted into, yielding an ‘<code class="docutils literal notranslate"><span class="pre">i32</span></code>’type. The ‘<code class="docutils literal notranslate"><span class="pre">getelementptr</span></code>’ instruction returns a pointer to thiselement.</p><p>Note that it is perfectly legal to index partially through a structure,returning a pointer to an inner element. Because of this, the LLVM codefor the given testcase is equivalent to:</p><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="k">define</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="vg">@foo</span><span class="p">(</span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%s</span><span class="p">)</span><span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="nv">%t1</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">getelementptr</span><span class="w"> </span><span class="nv">%struct.ST</span><span class="p">,</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%s</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">1</span><span class="w"> </span><span class="nv">%t2</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">getelementptr</span><span class="w"> </span><span class="nv">%struct.ST</span><span class="p">,</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%t1</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">0</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">2</span><span class="w"> </span><span class="nv">%t3</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">getelementptr</span><span class="w"> </span><span class="nv">%struct.RT</span><span class="p">,</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%t2</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">0</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">1</span><span class="w"> </span><span class="nv">%t4</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">getelementptr</span><span class="w"> </span><span class="p">[</span><span class="m">10</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="p">[</span><span class="m">20</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i32</span><span class="p">]],</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%t3</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">0</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">5</span><span class="w"> </span><span class="nv">%t5</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">getelementptr</span><span class="w"> </span><span class="p">[</span><span class="m">20</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i32</span><span class="p">],</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%t4</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">0</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">13</span><span class="w"> </span><span class="k">ret</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%t5</span><span class="p">}</span></pre></div></div><p>The indices are first converted to offsets in the pointer’s index type. If thecurrently indexed type is a struct type, the struct offset corresponding to theindex is sign-extended or truncated to the pointer index type. Otherwise, theindex itself is sign-extended or truncated, and then multiplied by the typeallocation size (that is, the size rounded up to the ABI alignment) of thecurrently indexed type.</p><p>The offsets are then added to the low bits of the base address up to the indextype width, with silently-wrapping two’s complement arithmetic. If the pointersize is larger than the index size, this means that the bits outside the indextype width will not be affected.</p><p>The result value of the <code class="docutils literal notranslate"><span class="pre">getelementptr</span></code> may be outside the object pointedto by the base pointer. The result value may not necessarily be used to accessmemory though, even if it happens to point into allocated storage. See the<a class="reference internal" href="#pointeraliasing"><span class="std std-ref">Pointer Aliasing Rules</span></a> section for moreinformation.</p><p>The <code class="docutils literal notranslate"><span class="pre">getelementptr</span></code> instruction may have a number of attributes that imposeadditional rules. If any of the rules are violated, the result value is a<a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison value</span></a>. In cases where the base is a vector ofpointers, the attributes apply to each computation element-wise.</p><p>For <code class="docutils literal notranslate"><span class="pre">nusw</span></code> (no unsigned signed wrap):</p><blockquote><div><ul class="simple"><li><p>If the type of an index is larger than the pointer index type, thetruncation to the pointer index type preserves the signed value(<code class="docutils literal notranslate"><span class="pre">trunc</span> <span class="pre">nsw</span></code>).</p></li><li><p>The multiplication of an index by the type size does not wrap the pointerindex type in a signed sense (<code class="docutils literal notranslate"><span class="pre">mul</span> <span class="pre">nsw</span></code>).</p></li><li><p>The successive addition of each offset (without adding the base address)does not wrap the pointer index type in a signed sense (<code class="docutils literal notranslate"><span class="pre">add</span> <span class="pre">nsw</span></code>).</p></li><li><p>The successive addition of the current address, truncated to the pointerindex type and interpreted as an unsigned number, and each offset,interpreted as a signed number, does not wrap the pointer index type.</p></li></ul></div></blockquote><p>For <code class="docutils literal notranslate"><span class="pre">nuw</span></code> (no unsigned wrap):</p><blockquote><div><ul class="simple"><li><p>If the type of an index is larger than the pointer index type, thetruncation to the pointer index type preserves the unsigned value(<code class="docutils literal notranslate"><span class="pre">trunc</span> <span class="pre">nuw</span></code>).</p></li><li><p>The multiplication of an index by the type size does not wrap the pointerindex type in an unsigned sense (<code class="docutils literal notranslate"><span class="pre">mul</span> <span class="pre">nuw</span></code>).</p></li><li><p>The successive addition of each offset (without adding the base address)does not wrap the pointer index type in an unsigned sense (<code class="docutils literal notranslate"><span class="pre">add</span> <span class="pre">nuw</span></code>).</p></li><li><p>The successive addition of the current address, truncated to the pointerindex type and interpreted as an unsigned number, and each offset, alsointerpreted as an unsigned number, does not wrap the pointer index type(<code class="docutils literal notranslate"><span class="pre">add</span> <span class="pre">nuw</span></code>).</p></li></ul></div></blockquote><p>For <code class="docutils literal notranslate"><span class="pre">inbounds</span></code> all rules of the <code class="docutils literal notranslate"><span class="pre">nusw</span></code> attribute apply. Additionally,if the <code class="docutils literal notranslate"><span class="pre">getelementptr</span></code> has any non-zero indices, the following rules apply:</p><blockquote><div><ul class="simple"><li><p>The base pointer has an <em>in bounds</em> address of the<a class="reference internal" href="#allocatedobjects"><span class="std std-ref">allocated object</span></a> that it is<a class="reference internal" href="#pointeraliasing"><span class="std std-ref">based</span></a> on. This means that it points into thatallocated object, or to its end. Note that the object does not have to belive anymore; being in-bounds of a deallocated object is sufficient.If the allocated object can grow, then the relevant size for being <em>inbounds</em> is the maximal size the object could have while satisfying theallocated object rules, not its current size.</p></li><li><p>During the successive addition of offsets to the address, the resultingpointer must remain <em>in bounds</em> of the allocated object at each step.</p></li></ul></div></blockquote><p>Note that <code class="docutils literal notranslate"><span class="pre">getelementptr</span></code> with all-zero indices is always considered to be<code class="docutils literal notranslate"><span class="pre">inbounds</span></code>, even if the base pointer does not point to an allocated object.As a corollary, the only pointer in bounds of the null pointer in the defaultaddress space is the null pointer itself.</p><p>If <code class="docutils literal notranslate"><span class="pre">inbounds</span></code> is present on a <code class="docutils literal notranslate"><span class="pre">getelementptr</span></code> instruction, the <code class="docutils literal notranslate"><span class="pre">nusw</span></code>attribute will be automatically set as well. For this reason, the <code class="docutils literal notranslate"><span class="pre">nusw</span></code>will also not be printed in textual IR if <code class="docutils literal notranslate"><span class="pre">inbounds</span></code> is already present.</p><p>If the <code class="docutils literal notranslate"><span class="pre">inrange(Start,</span> <span class="pre">End)</span></code> attribute is present, loading from orstoring to any pointer derived from the <code class="docutils literal notranslate"><span class="pre">getelementptr</span></code> has undefinedbehavior if the load or store would access memory outside the half-open range<code class="docutils literal notranslate"><span class="pre">[Start,</span> <span class="pre">End)</span></code> from the <code class="docutils literal notranslate"><span class="pre">getelementptr</span></code> expression result. The result ofa pointer comparison or <code class="docutils literal notranslate"><span class="pre">ptrtoint</span></code> (including <code class="docutils literal notranslate"><span class="pre">ptrtoint</span></code>-like operationsinvolving memory) involving a pointer derived from a <code class="docutils literal notranslate"><span class="pre">getelementptr</span></code> withthe <code class="docutils literal notranslate"><span class="pre">inrange</span></code> keyword is undefined, with the exception of comparisonsin the case where both operands are in the closed range <code class="docutils literal notranslate"><span class="pre">[Start,</span> <span class="pre">End]</span></code>.Note that the <code class="docutils literal notranslate"><span class="pre">inrange</span></code> keyword is currently only allowedin constant <code class="docutils literal notranslate"><span class="pre">getelementptr</span></code> expressions.</p><p>The getelementptr instruction is often confusing. For some more insightinto how it works, see <a class="reference internal" href="GetElementPtr.html"><span class="doc">the getelementptr FAQ</span></a>.</p></section><section id="id240"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%aptr</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">getelementptr</span><span class="w"> </span><span class="p">{</span><span class="kt">i32</span><span class="p">,</span><span class="w"> </span><span class="p">[</span><span class="m">12</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i8</span><span class="p">]},</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%saptr</span><span class="p">,</span><span class="w"> </span><span class="kt">i64</span><span class="w"> </span><span class="m">0</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">1</span><span class="nv">%vptr</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">getelementptr</span><span class="w"> </span><span class="p">{</span><span class="kt">i32</span><span class="p">,</span><span class="w"> </span><span class="p"><</span><span class="m">2</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i8</span><span class="p">>},</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%svptr</span><span class="p">,</span><span class="w"> </span><span class="kt">i64</span><span class="w"> </span><span class="m">0</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">1</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">1</span><span class="nv">%eptr</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">getelementptr</span><span class="w"> </span><span class="p">[</span><span class="m">12</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i8</span><span class="p">],</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%aptr</span><span class="p">,</span><span class="w"> </span><span class="kt">i64</span><span class="w"> </span><span class="m">0</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">1</span><span class="nv">%iptr</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">getelementptr</span><span class="w"> </span><span class="p">[</span><span class="m">10</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i32</span><span class="p">],</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="vg">@arr</span><span class="p">,</span><span class="w"> </span><span class="kt">i16</span><span class="w"> </span><span class="m">0</span><span class="p">,</span><span class="w"> </span><span class="kt">i16</span><span class="w"> </span><span class="m">0</span></pre></div></div></section><section id="vector-of-pointers"><h5>Vector of pointers:</h5><p>The <code class="docutils literal notranslate"><span class="pre">getelementptr</span></code> returns a vector of pointers, instead of a single address,when one or more of its arguments is a vector. In such cases, all vectorarguments should have the same number of elements, and every scalar argumentwill be effectively broadcast into a vector during address calculation.</p><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="c">; All arguments are vectors:</span><span class="c">; A[i] = ptrs[i] + offsets[i]*sizeof(i8)</span><span class="nv">%A</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">getelementptr</span><span class="w"> </span><span class="kt">i8</span><span class="p">,</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i8</span><span class="p">*></span><span class="w"> </span><span class="nv">%ptrs</span><span class="p">,</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i64</span><span class="p">></span><span class="w"> </span><span class="nv">%offsets</span><span class="c">; Add the same scalar offset to each pointer of a vector:</span><span class="c">; A[i] = ptrs[i] + offset*sizeof(i8)</span><span class="nv">%A</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">getelementptr</span><span class="w"> </span><span class="kt">i8</span><span class="p">,</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">ptr</span><span class="p">></span><span class="w"> </span><span class="nv">%ptrs</span><span class="p">,</span><span class="w"> </span><span class="kt">i64</span><span class="w"> </span><span class="nv">%offset</span><span class="c">; Add distinct offsets to the same pointer:</span><span class="c">; A[i] = ptr + offsets[i]*sizeof(i8)</span><span class="nv">%A</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">getelementptr</span><span class="w"> </span><span class="kt">i8</span><span class="p">,</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%ptr</span><span class="p">,</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i64</span><span class="p">></span><span class="w"> </span><span class="nv">%offsets</span><span class="c">; In all cases described above the type of the result is <4 x ptr></span></pre></div></div><p>The two following instructions are equivalent:</p><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="k">getelementptr</span><span class="w"> </span><span class="nv">%struct.ST</span><span class="p">,</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">ptr</span><span class="p">></span><span class="w"> </span><span class="nv">%s</span><span class="p">,</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i64</span><span class="p">></span><span class="w"> </span><span class="nv">%ind1</span><span class="p">,</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i32</span><span class="p">></span><span class="w"> </span><span class="p"><</span><span class="kt">i32</span><span class="w"> </span><span class="m">2</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">2</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">2</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">2</span><span class="p">>,</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i32</span><span class="p">></span><span class="w"> </span><span class="p"><</span><span class="kt">i32</span><span class="w"> </span><span class="m">1</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">1</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">1</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">1</span><span class="p">>,</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i32</span><span class="p">></span><span class="w"> </span><span class="nv">%ind4</span><span class="p">,</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i64</span><span class="p">></span><span class="w"> </span><span class="p"><</span><span class="kt">i64</span><span class="w"> </span><span class="m">13</span><span class="p">,</span><span class="w"> </span><span class="kt">i64</span><span class="w"> </span><span class="m">13</span><span class="p">,</span><span class="w"> </span><span class="kt">i64</span><span class="w"> </span><span class="m">13</span><span class="p">,</span><span class="w"> </span><span class="kt">i64</span><span class="w"> </span><span class="m">13</span><span class="p">></span><span class="k">getelementptr</span><span class="w"> </span><span class="nv">%struct.ST</span><span class="p">,</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">ptr</span><span class="p">></span><span class="w"> </span><span class="nv">%s</span><span class="p">,</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i64</span><span class="p">></span><span class="w"> </span><span class="nv">%ind1</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">2</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">1</span><span class="p">,</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i32</span><span class="p">></span><span class="w"> </span><span class="nv">%ind4</span><span class="p">,</span><span class="w"> </span><span class="kt">i64</span><span class="w"> </span><span class="m">13</span></pre></div></div><p>Let’s look at the C code, where the vector version of <code class="docutils literal notranslate"><span class="pre">getelementptr</span></code>makes sense:</p><div class="highlight-c notranslate"><div class="highlight"><pre><span></span><span class="c1">// Let's assume that we vectorize the following loop:</span><span class="kt">double</span><span class="w"> </span><span class="o">*</span><span class="n">A</span><span class="p">,</span><span class="w"> </span><span class="o">*</span><span class="n">B</span><span class="p">;</span><span class="w"> </span><span class="kt">int</span><span class="w"> </span><span class="o">*</span><span class="n">C</span><span class="p">;</span><span class="k">for</span><span class="w"> </span><span class="p">(</span><span class="kt">int</span><span class="w"> </span><span class="n">i</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="mi">0</span><span class="p">;</span><span class="w"> </span><span class="n">i</span><span class="w"> </span><span class="o"><</span><span class="w"> </span><span class="n">size</span><span class="p">;</span><span class="w"> </span><span class="o">++</span><span class="n">i</span><span class="p">)</span><span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="n">A</span><span class="p">[</span><span class="n">i</span><span class="p">]</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">B</span><span class="p">[</span><span class="n">C</span><span class="p">[</span><span class="n">i</span><span class="p">]];</span><span class="p">}</span></pre></div></div><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="c">; get pointers for 8 elements from array B</span><span class="nv">%ptrs</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">getelementptr</span><span class="w"> </span><span class="kt">double</span><span class="p">,</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%B</span><span class="p">,</span><span class="w"> </span><span class="p"><</span><span class="m">8</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i32</span><span class="p">></span><span class="w"> </span><span class="nv">%C</span><span class="c">; load 8 elements from array B into A</span><span class="nv">%A</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">call</span><span class="w"> </span><span class="p"><</span><span class="m">8</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">double</span><span class="p">></span><span class="w"> </span><span class="vg">@llvm.masked.gather.v8f64.v8p0f64</span><span class="p">(</span><span class="w"> </span><span class="p"><</span><span class="m">8</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">ptr</span><span class="p">></span><span class="w"> </span><span class="k">align</span><span class="w"> </span><span class="m">8</span><span class="w"> </span><span class="nv">%ptrs</span><span class="p">,</span><span class="w"> </span><span class="p"><</span><span class="m">8</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i1</span><span class="p">></span><span class="w"> </span><span class="nv">%mask</span><span class="p">,</span><span class="w"> </span><span class="p"><</span><span class="m">8</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">double</span><span class="p">></span><span class="w"> </span><span class="nv">%passthru</span><span class="p">)</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘getelementptr’ instruction is used to get the address of asubelement of an aggregate data structure. It performsaddress calculation only and does not access memory. The instruction can alsobe used to calculate a vector of such addresses.`,
|
||
};
|
||
case 'TRUNC':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#trunc-to-instruction`,
|
||
html: `<html><head></head><body><span id="i-trunc"></span><h4><a class="toc-backref" href="#id2205" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">trunc</span> <span class="pre">..</span> <span class="pre">to</span></code>’ Instruction</a></h4><section id="id241"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">trunc</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="n">to</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty2</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">trunc</span> <span class="n">nsw</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="n">to</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty2</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">trunc</span> <span class="n">nuw</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="n">to</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty2</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">trunc</span> <span class="n">nuw</span> <span class="n">nsw</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="n">to</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty2</span></pre></div></div></section><section id="id242"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">trunc</span></code>’ instruction truncates its operand to the type <code class="docutils literal notranslate"><span class="pre">ty2</span></code>.</p></section><section id="id243"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">trunc</span></code>’ instruction takes a value to trunc, and a type to truncit to. Both types must be of <a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> types, or vectorsof the same number of integers. The bit size of the <code class="docutils literal notranslate"><span class="pre">value</span></code> must belarger than the bit size of the destination type, <code class="docutils literal notranslate"><span class="pre">ty2</span></code>. Equal sizedtypes are not allowed.</p></section><section id="id244"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">trunc</span></code>’ instruction truncates the high order bits in <code class="docutils literal notranslate"><span class="pre">value</span></code>and converts the remaining bits to <code class="docutils literal notranslate"><span class="pre">ty2</span></code>. Since the source size mustbe larger than the destination size, <code class="docutils literal notranslate"><span class="pre">trunc</span></code> cannot be a <em>no-op cast</em>.It will always truncate bits.</p><p>If the <code class="docutils literal notranslate"><span class="pre">nuw</span></code> keyword is present, and any of the truncated bits are non-zero,the result is a <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison value</span></a>. If the <code class="docutils literal notranslate"><span class="pre">nsw</span></code> keywordis present, and any of the truncated bits are not the same as the top bitof the truncation result, the result is a <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison value</span></a>.</p></section><section id="id245"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%X</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">trunc</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">257</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i8</span><span class="w"> </span><span class="c">; yields i8:1</span><span class="nv">%Y</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">trunc</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">123</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i1</span><span class="w"> </span><span class="c">; yields i1:true</span><span class="nv">%Z</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">trunc</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">122</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i1</span><span class="w"> </span><span class="c">; yields i1:false</span><span class="nv">%W</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">trunc</span><span class="w"> </span><span class="p"><</span><span class="m">2</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i16</span><span class="p">></span><span class="w"> </span><span class="p"><</span><span class="kt">i16</span><span class="w"> </span><span class="m">8</span><span class="p">,</span><span class="w"> </span><span class="kt">i16</span><span class="w"> </span><span class="m">7</span><span class="p">></span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="p"><</span><span class="m">2</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i8</span><span class="p">></span><span class="w"> </span><span class="c">; yields <i8 8, i8 7></span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘trunc’ instruction truncates its operand to the type ty2.`,
|
||
};
|
||
case 'ZEXT':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#zext-to-instruction`,
|
||
html: `<html><head></head><body><span id="i-zext"></span><h4><a class="toc-backref" href="#id2206" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">zext</span> <span class="pre">..</span> <span class="pre">to</span></code>’ Instruction</a></h4><section id="id246"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">zext</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="n">to</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty2</span></pre></div></div></section><section id="id247"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">zext</span></code>’ instruction zero extends its operand to type <code class="docutils literal notranslate"><span class="pre">ty2</span></code>.</p><p>The <code class="docutils literal notranslate"><span class="pre">nneg</span></code> (non-negative) flag, if present, specifies that the operand isnon-negative. This property may be used by optimization passes to laterconvert the <code class="docutils literal notranslate"><span class="pre">zext</span></code> into a <code class="docutils literal notranslate"><span class="pre">sext</span></code>.</p></section><section id="id248"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">zext</span></code>’ instruction takes a value to cast, and a type to cast itto. Both types must be of <a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> types, or vectors ofthe same number of integers. The bit size of the <code class="docutils literal notranslate"><span class="pre">value</span></code> must besmaller than the bit size of the destination type, <code class="docutils literal notranslate"><span class="pre">ty2</span></code>.</p></section><section id="id249"><h5>Semantics:</h5><p>The <code class="docutils literal notranslate"><span class="pre">zext</span></code> fills the high order bits of the <code class="docutils literal notranslate"><span class="pre">value</span></code> with zero bitsuntil it reaches the size of the destination type, <code class="docutils literal notranslate"><span class="pre">ty2</span></code>.</p><p>When zero extending from i1, the result will always be either 0 or 1.</p><p>If the <code class="docutils literal notranslate"><span class="pre">nneg</span></code> flag is set, and the <code class="docutils literal notranslate"><span class="pre">zext</span></code> argument is negative, the resultis a poison value.</p></section><section id="id250"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%X</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">zext</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">257</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i64</span><span class="w"> </span><span class="c">; yields i64:257</span><span class="nv">%Y</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">zext</span><span class="w"> </span><span class="kt">i1</span><span class="w"> </span><span class="k">true</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="c">; yields i32:1</span><span class="nv">%Z</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">zext</span><span class="w"> </span><span class="p"><</span><span class="m">2</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i16</span><span class="p">></span><span class="w"> </span><span class="p"><</span><span class="kt">i16</span><span class="w"> </span><span class="m">8</span><span class="p">,</span><span class="w"> </span><span class="kt">i16</span><span class="w"> </span><span class="m">7</span><span class="p">></span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="p"><</span><span class="m">2</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i32</span><span class="p">></span><span class="w"> </span><span class="c">; yields <i32 8, i32 7></span><span class="nv">%a</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">zext</span><span class="w"> </span><span class="err">nneg</span><span class="w"> </span><span class="kt">i8</span><span class="w"> </span><span class="m">127</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i16</span><span class="w"> </span><span class="c">; yields i16 127</span><span class="nv">%b</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">zext</span><span class="w"> </span><span class="err">nneg</span><span class="w"> </span><span class="kt">i8</span><span class="w"> </span><span class="m">-1</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i16</span><span class="w"> </span><span class="c">; yields i16 poison</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘zext’ instruction zero extends its operand to type ty2.The nneg (non-negative) flag, if present, specifies that the operand isnon-negative. This property may be used by optimization passes to laterconvert the zext into a sext.`,
|
||
};
|
||
case 'SEXT':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#sext-to-instruction`,
|
||
html: `<html><head></head><body><span id="i-sext"></span><h4><a class="toc-backref" href="#id2207" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">sext</span> <span class="pre">..</span> <span class="pre">to</span></code>’ Instruction</a></h4><section id="id251"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">sext</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="n">to</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty2</span></pre></div></div></section><section id="id252"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">sext</span></code>’ sign extends <code class="docutils literal notranslate"><span class="pre">value</span></code> to the type <code class="docutils literal notranslate"><span class="pre">ty2</span></code>.</p></section><section id="id253"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">sext</span></code>’ instruction takes a value to cast, and a type to cast itto. Both types must be of <a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> types, or vectors ofthe same number of integers. The bit size of the <code class="docutils literal notranslate"><span class="pre">value</span></code> must besmaller than the bit size of the destination type, <code class="docutils literal notranslate"><span class="pre">ty2</span></code>.</p></section><section id="id254"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">sext</span></code>’ instruction performs a sign extension by copying the signbit (highest order bit) of the <code class="docutils literal notranslate"><span class="pre">value</span></code> until it reaches the bit sizeof the type <code class="docutils literal notranslate"><span class="pre">ty2</span></code>.</p><p>When sign extending from i1, the extension always results in -1 or 0.</p></section><section id="id255"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%X</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">sext</span><span class="w"> </span><span class="kt">i8</span><span class="w"> </span><span class="m">-1</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i16</span><span class="w"> </span><span class="c">; yields i16 :65535</span><span class="nv">%Y</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">sext</span><span class="w"> </span><span class="kt">i1</span><span class="w"> </span><span class="k">true</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="c">; yields i32:-1</span><span class="nv">%Z</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">sext</span><span class="w"> </span><span class="p"><</span><span class="m">2</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i16</span><span class="p">></span><span class="w"> </span><span class="p"><</span><span class="kt">i16</span><span class="w"> </span><span class="m">8</span><span class="p">,</span><span class="w"> </span><span class="kt">i16</span><span class="w"> </span><span class="m">7</span><span class="p">></span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="p"><</span><span class="m">2</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i32</span><span class="p">></span><span class="w"> </span><span class="c">; yields <i32 8, i32 7></span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘sext’ sign extends value to the type ty2.`,
|
||
};
|
||
case 'FPTRUNC':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#fptrunc-to-instruction`,
|
||
html: `<html><head></head><body><span id="i-fptrunc"></span><h4><a class="toc-backref" href="#id2208" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">fptrunc</span> <span class="pre">..</span> <span class="pre">to</span></code>’ Instruction</a></h4><section id="id256"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">fptrunc</span> <span class="p">[</span><span class="n">fast</span><span class="o">-</span><span class="n">math</span> <span class="n">flags</span><span class="p">]</span><span class="o">*</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="n">to</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty2</span></pre></div></div></section><section id="id257"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fptrunc</span></code>’ instruction truncates <code class="docutils literal notranslate"><span class="pre">value</span></code> to type <code class="docutils literal notranslate"><span class="pre">ty2</span></code>.</p></section><section id="id258"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fptrunc</span></code>’ instruction takes a <a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a>value to cast and a <a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> type to cast it to.The size of <code class="docutils literal notranslate"><span class="pre">value</span></code> must be larger than the size of <code class="docutils literal notranslate"><span class="pre">ty2</span></code>. Thisimplies that <code class="docutils literal notranslate"><span class="pre">fptrunc</span></code> cannot be used to make a <em>no-op cast</em>.</p></section><section id="id259"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fptrunc</span></code>’ instruction casts a <code class="docutils literal notranslate"><span class="pre">value</span></code> from a larger<a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> type to a smaller <a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> type.This instruction is assumed to execute in the default <a class="reference internal" href="#floatenv"><span class="std std-ref">floating-pointenvironment</span></a>.</p><p>NaN values follow the usual <a class="reference internal" href="#floatnan"><span class="std std-ref">NaN behaviors</span></a>, except that _if_ aNaN payload is propagated from the input (“Quieting NaN propagation” or“Unchanged NaN propagation” cases), then the low order bits of the NaN payloadwhich cannot fit in the resulting type are discarded. Note that if discardingthe low order bits leads to an all-0 payload, this cannot be represented as asignaling NaN (it would represent an infinity instead), so in that case“Unchanged NaN propagation” is not possible.</p><p>This instruction can also take any number of <a class="reference internal" href="#fastmath"><span class="std std-ref">fast-mathflags</span></a>, which are optimization hints to enable otherwiseunsafe floating-point optimizations.</p></section><section id="id260"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%X</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">fptrunc</span><span class="w"> </span><span class="kt">double</span><span class="w"> </span><span class="m">16777217.0</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">float</span><span class="w"> </span><span class="c">; yields float:16777216.0</span><span class="nv">%Y</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">fptrunc</span><span class="w"> </span><span class="kt">double</span><span class="w"> </span><span class="m">1.0E+300</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">half</span><span class="w"> </span><span class="c">; yields half:+infinity</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘fptrunc’ instruction truncates value to type ty2.`,
|
||
};
|
||
case 'FPEXT':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#fpext-to-instruction`,
|
||
html: `<html><head></head><body><span id="i-fpext"></span><h4><a class="toc-backref" href="#id2209" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">fpext</span> <span class="pre">..</span> <span class="pre">to</span></code>’ Instruction</a></h4><section id="id261"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">fpext</span> <span class="p">[</span><span class="n">fast</span><span class="o">-</span><span class="n">math</span> <span class="n">flags</span><span class="p">]</span><span class="o">*</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="n">to</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty2</span></pre></div></div></section><section id="id262"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fpext</span></code>’ extends a floating-point <code class="docutils literal notranslate"><span class="pre">value</span></code> to a larger floating-pointvalue.</p></section><section id="id263"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fpext</span></code>’ instruction takes a <a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a><code class="docutils literal notranslate"><span class="pre">value</span></code> to cast, and a <a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> type to cast itto. The source type must be smaller than the destination type.</p></section><section id="id264"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fpext</span></code>’ instruction extends the <code class="docutils literal notranslate"><span class="pre">value</span></code> from a smaller<a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> type to a larger <a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> type. The <code class="docutils literal notranslate"><span class="pre">fpext</span></code> cannot be used to make a<em>no-op cast</em> because it always changes bits. Use <code class="docutils literal notranslate"><span class="pre">bitcast</span></code> to make a<em>no-op cast</em> for a floating-point cast.</p><p>NaN values follow the usual <a class="reference internal" href="#floatnan"><span class="std std-ref">NaN behaviors</span></a>, except that _if_ aNaN payload is propagated from the input (“Quieting NaN propagation” or“Unchanged NaN propagation” cases), then it is copied to the high order bits ofthe resulting payload, and the remaining low order bits are zero.</p><p>This instruction can also take any number of <a class="reference internal" href="#fastmath"><span class="std std-ref">fast-mathflags</span></a>, which are optimization hints to enable otherwiseunsafe floating-point optimizations.</p></section><section id="id265"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%X</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">fpext</span><span class="w"> </span><span class="kt">float</span><span class="w"> </span><span class="m">3.125</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">double</span><span class="w"> </span><span class="c">; yields double:3.125000e+00</span><span class="nv">%Y</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">fpext</span><span class="w"> </span><span class="kt">double</span><span class="w"> </span><span class="nv">%X</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">fp128</span><span class="w"> </span><span class="c">; yields fp128:0xL00000000000000004000900000000000</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘fpext’ extends a floating-point value to a larger floating-pointvalue.`,
|
||
};
|
||
case 'FPTOUI':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#fptoui-to-instruction`,
|
||
html: `<html><head></head><body><h4><a class="toc-backref" href="#id2210" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">fptoui</span> <span class="pre">..</span> <span class="pre">to</span></code>’ Instruction</a></h4><section id="id266"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">fptoui</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="n">to</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty2</span></pre></div></div></section><section id="id267"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fptoui</span></code>’ converts a floating-point <code class="docutils literal notranslate"><span class="pre">value</span></code> to its unsignedinteger equivalent of type <code class="docutils literal notranslate"><span class="pre">ty2</span></code>.</p></section><section id="id268"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fptoui</span></code>’ instruction takes a value to cast, which must be ascalar or vector <a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> value, and a type tocast it to <code class="docutils literal notranslate"><span class="pre">ty2</span></code>, which must be an <a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> type. If<code class="docutils literal notranslate"><span class="pre">ty</span></code> is a vector floating-point type, <code class="docutils literal notranslate"><span class="pre">ty2</span></code> must be a vector integertype with the same number of elements as <code class="docutils literal notranslate"><span class="pre">ty</span></code></p></section><section id="id269"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fptoui</span></code>’ instruction converts its <a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> operand into the nearest (rounding towards zero)unsigned integer value. If the value cannot fit in <code class="docutils literal notranslate"><span class="pre">ty2</span></code>, the resultis a <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison value</span></a>.</p></section><section id="id270"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%X</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">fptoui</span><span class="w"> </span><span class="kt">double</span><span class="w"> </span><span class="m">123.0</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="c">; yields i32:123</span><span class="nv">%Y</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">fptoui</span><span class="w"> </span><span class="kt">float</span><span class="w"> </span><span class="m">1.0E+300</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i1</span><span class="w"> </span><span class="c">; yields undefined:1</span><span class="nv">%Z</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">fptoui</span><span class="w"> </span><span class="kt">float</span><span class="w"> </span><span class="m">1.04E+17</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i8</span><span class="w"> </span><span class="c">; yields undefined:1</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘fptoui’ converts a floating-point value to its unsignedinteger equivalent of type ty2.`,
|
||
};
|
||
case 'FPTOSI':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#fptosi-to-instruction`,
|
||
html: `<html><head></head><body><h4><a class="toc-backref" href="#id2211" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">fptosi</span> <span class="pre">..</span> <span class="pre">to</span></code>’ Instruction</a></h4><section id="id271"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">fptosi</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="n">to</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty2</span></pre></div></div></section><section id="id272"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fptosi</span></code>’ instruction converts <a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a><code class="docutils literal notranslate"><span class="pre">value</span></code> to type <code class="docutils literal notranslate"><span class="pre">ty2</span></code>.</p></section><section id="id273"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fptosi</span></code>’ instruction takes a value to cast, which must be ascalar or vector <a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> value, and a type tocast it to <code class="docutils literal notranslate"><span class="pre">ty2</span></code>, which must be an <a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> type. If<code class="docutils literal notranslate"><span class="pre">ty</span></code> is a vector floating-point type, <code class="docutils literal notranslate"><span class="pre">ty2</span></code> must be a vector integertype with the same number of elements as <code class="docutils literal notranslate"><span class="pre">ty</span></code></p></section><section id="id274"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fptosi</span></code>’ instruction converts its <a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> operand into the nearest (rounding towards zero)signed integer value. If the value cannot fit in <code class="docutils literal notranslate"><span class="pre">ty2</span></code>, the resultis a <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison value</span></a>.</p></section><section id="id275"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%X</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">fptosi</span><span class="w"> </span><span class="kt">double</span><span class="w"> </span><span class="m">-123.0</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="c">; yields i32:-123</span><span class="nv">%Y</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">fptosi</span><span class="w"> </span><span class="kt">float</span><span class="w"> </span><span class="m">1.0E-247</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i1</span><span class="w"> </span><span class="c">; yields undefined:1</span><span class="nv">%Z</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">fptosi</span><span class="w"> </span><span class="kt">float</span><span class="w"> </span><span class="m">1.04E+17</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i8</span><span class="w"> </span><span class="c">; yields undefined:1</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘fptosi’ instruction converts floating-pointvalue to type ty2.`,
|
||
};
|
||
case 'UITOFP':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#uitofp-to-instruction`,
|
||
html: `<html><head></head><body><h4><a class="toc-backref" href="#id2212" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">uitofp</span> <span class="pre">..</span> <span class="pre">to</span></code>’ Instruction</a></h4><section id="id276"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">uitofp</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="n">to</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty2</span></pre></div></div></section><section id="id277"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">uitofp</span></code>’ instruction regards <code class="docutils literal notranslate"><span class="pre">value</span></code> as an unsigned integerand converts that value to the <code class="docutils literal notranslate"><span class="pre">ty2</span></code> type.</p><p>The <code class="docutils literal notranslate"><span class="pre">nneg</span></code> (non-negative) flag, if present, specifies that theoperand is non-negative. This property may be used by optimizationpasses to later convert the <code class="docutils literal notranslate"><span class="pre">uitofp</span></code> into a <code class="docutils literal notranslate"><span class="pre">sitofp</span></code>.</p></section><section id="id278"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">uitofp</span></code>’ instruction takes a value to cast, which must be ascalar or vector <a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> value, and a type to cast it to<code class="docutils literal notranslate"><span class="pre">ty2</span></code>, which must be an <a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> type. If<code class="docutils literal notranslate"><span class="pre">ty</span></code> is a vector integer type, <code class="docutils literal notranslate"><span class="pre">ty2</span></code> must be a vector floating-pointtype with the same number of elements as <code class="docutils literal notranslate"><span class="pre">ty</span></code></p></section><section id="id279"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">uitofp</span></code>’ instruction interprets its operand as an unsignedinteger quantity and converts it to the corresponding floating-pointvalue. If the value cannot be exactly represented, it is rounded usingthe default rounding mode.</p><p>If the <code class="docutils literal notranslate"><span class="pre">nneg</span></code> flag is set, and the <code class="docutils literal notranslate"><span class="pre">uitofp</span></code> argument is negative,the result is a poison value.</p></section><section id="id280"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%X</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">uitofp</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">257</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">float</span><span class="w"> </span><span class="c">; yields float:257.0</span><span class="nv">%Y</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">uitofp</span><span class="w"> </span><span class="kt">i8</span><span class="w"> </span><span class="m">-1</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">double</span><span class="w"> </span><span class="c">; yields double:255.0</span><span class="nv">%a</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">uitofp</span><span class="w"> </span><span class="err">nneg</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">256</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="c">; yields float:256.0</span><span class="nv">%b</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">uitofp</span><span class="w"> </span><span class="err">nneg</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">-256</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="c">; yields i32 poison</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘uitofp’ instruction regards value as an unsigned integerand converts that value to the ty2 type.The nneg (non-negative) flag, if present, specifies that theoperand is non-negative. This property may be used by optimizationpasses to later convert the uitofp into a sitofp.`,
|
||
};
|
||
case 'SITOFP':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#sitofp-to-instruction`,
|
||
html: `<html><head></head><body><h4><a class="toc-backref" href="#id2213" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">sitofp</span> <span class="pre">..</span> <span class="pre">to</span></code>’ Instruction</a></h4><section id="id281"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">sitofp</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="n">to</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty2</span></pre></div></div></section><section id="id282"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">sitofp</span></code>’ instruction regards <code class="docutils literal notranslate"><span class="pre">value</span></code> as a signed integer andconverts that value to the <code class="docutils literal notranslate"><span class="pre">ty2</span></code> type.</p></section><section id="id283"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">sitofp</span></code>’ instruction takes a value to cast, which must be ascalar or vector <a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> value, and a type to cast it to<code class="docutils literal notranslate"><span class="pre">ty2</span></code>, which must be an <a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> type. If<code class="docutils literal notranslate"><span class="pre">ty</span></code> is a vector integer type, <code class="docutils literal notranslate"><span class="pre">ty2</span></code> must be a vector floating-pointtype with the same number of elements as <code class="docutils literal notranslate"><span class="pre">ty</span></code></p></section><section id="id284"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">sitofp</span></code>’ instruction interprets its operand as a signed integerquantity and converts it to the corresponding floating-point value. If thevalue cannot be exactly represented, it is rounded using the default roundingmode.</p></section><section id="id285"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%X</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">sitofp</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">257</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">float</span><span class="w"> </span><span class="c">; yields float:257.0</span><span class="nv">%Y</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">sitofp</span><span class="w"> </span><span class="kt">i8</span><span class="w"> </span><span class="m">-1</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">double</span><span class="w"> </span><span class="c">; yields double:-1.0</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘sitofp’ instruction regards value as a signed integer andconverts that value to the ty2 type.`,
|
||
};
|
||
case 'PTRTOINT':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#ptrtoint-to-instruction`,
|
||
html: `<html><head></head><body><span id="i-ptrtoint"></span><h4><a class="toc-backref" href="#id2214" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">ptrtoint</span> <span class="pre">..</span> <span class="pre">to</span></code>’ Instruction</a></h4><section id="id286"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">ptrtoint</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="n">to</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty2</span></pre></div></div></section><section id="id287"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">ptrtoint</span></code>’ instruction converts the pointer or a vector ofpointers <code class="docutils literal notranslate"><span class="pre">value</span></code> to the integer (or vector of integers) type <code class="docutils literal notranslate"><span class="pre">ty2</span></code>.</p></section><section id="id288"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">ptrtoint</span></code>’ instruction takes a <code class="docutils literal notranslate"><span class="pre">value</span></code> to cast, which must bea value of type <a class="reference internal" href="#t-pointer"><span class="std std-ref">pointer</span></a> or a vector of pointers, and atype to cast it to <code class="docutils literal notranslate"><span class="pre">ty2</span></code>, which must be an <a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> ora vector of integers type.</p></section><section id="id289"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">ptrtoint</span></code>’ instruction converts <code class="docutils literal notranslate"><span class="pre">value</span></code> to integer type<code class="docutils literal notranslate"><span class="pre">ty2</span></code> by interpreting all the pointer representation bits as an integer(equivalent to a <code class="docutils literal notranslate"><span class="pre">bitcast</span></code>) and either truncating or zero extending that valueto the size of the integer type.If <code class="docutils literal notranslate"><span class="pre">value</span></code> is smaller than <code class="docutils literal notranslate"><span class="pre">ty2</span></code> then a zero extension is done. If<code class="docutils literal notranslate"><span class="pre">value</span></code> is larger than <code class="docutils literal notranslate"><span class="pre">ty2</span></code> then a truncation is done. If they arethe same size, then nothing is done (<em>no-op cast</em>) other than a typechange.The <code class="docutils literal notranslate"><span class="pre">ptrtoint</span></code> always <a class="reference internal" href="#pointercapture"><span class="std std-ref">captures address and provenance</span></a>of the pointer argument.</p></section><section id="id290"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%X</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">ptrtoint</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%P</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i8</span><span class="w"> </span><span class="c">; yields truncation on 32-bit architecture</span><span class="nv">%Y</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">ptrtoint</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%P</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i64</span><span class="w"> </span><span class="c">; yields zero extension on 32-bit architecture</span><span class="nv">%Z</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">ptrtoint</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">ptr</span><span class="p">></span><span class="w"> </span><span class="nv">%P</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i64</span><span class="p">></span><span class="c">; yields vector zero extension for a vector of addresses on 32-bit architecture</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘ptrtoint’ instruction converts the pointer or a vector ofpointers value to the integer (or vector of integers) type ty2.`,
|
||
};
|
||
case 'PTRTOADDR':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#ptrtoaddr-to-instruction`,
|
||
html: `<html><head></head><body><span id="i-ptrtoaddr"></span><h4><a class="toc-backref" href="#id2215" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">ptrtoaddr</span> <span class="pre">..</span> <span class="pre">to</span></code>’ Instruction</a></h4><section id="id291"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">ptrtoaddr</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="n">to</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty2</span></pre></div></div></section><section id="id292"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">ptrtoaddr</span></code>’ instruction converts the pointer or a vector ofpointers <code class="docutils literal notranslate"><span class="pre">value</span></code> to the underlying integer address (or vector of addresses) oftype <code class="docutils literal notranslate"><span class="pre">ty2</span></code>. This is different from <a class="reference internal" href="#i-ptrtoint"><span class="std std-ref">ptrtoint</span></a> in that itonly operates on the index bits of the pointer and ignores all other bits, anddoes not capture the provenance of the pointer.</p></section><section id="id293"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">ptrtoaddr</span></code>’ instruction takes a <code class="docutils literal notranslate"><span class="pre">value</span></code> to cast, which must bea value of type <a class="reference internal" href="#t-pointer"><span class="std std-ref">pointer</span></a> or a vector of pointers, and atype to cast it to <code class="docutils literal notranslate"><span class="pre">ty2</span></code>, which must be must be the <a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a>type (or vector of integers) matching the pointer index width of the addressspace of <code class="docutils literal notranslate"><span class="pre">ty</span></code>.</p></section><section id="id294"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">ptrtoaddr</span></code>’ instruction converts <code class="docutils literal notranslate"><span class="pre">value</span></code> to integer type <code class="docutils literal notranslate"><span class="pre">ty2</span></code> byinterpreting the lowest index-width pointer representation bits as an integer.If the address size and the pointer representation size are the same and<code class="docutils literal notranslate"><span class="pre">value</span></code> and <code class="docutils literal notranslate"><span class="pre">ty2</span></code> are the same size, then nothing is done (<em>no-op cast</em>)other than a type change.</p><p>The <code class="docutils literal notranslate"><span class="pre">ptrtoaddr</span></code> instruction always <a class="reference internal" href="#pointercapture"><span class="std std-ref">captures the address but not the provenance</span></a>of the pointer argument.</p></section><section id="id295"><h5>Example:</h5><p>This example assumes pointers in address space 1 are 64 bits in size with anaddress width of 32 bits (<code class="docutils literal notranslate"><span class="pre">p1:64:64:64:32</span></code> <a class="reference internal" href="#langref-datalayout"><span class="std std-ref">datalayout string</span></a>)</p><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%X</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="kt">ptr</span><span class="err">toaddr</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="k">addrspace</span><span class="p">(</span><span class="m">1</span><span class="p">)</span><span class="w"> </span><span class="nv">%P</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="c">; extracts low 32 bits of pointer</span><span class="nv">%Y</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="kt">ptr</span><span class="err">toaddr</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="k">addrspace</span><span class="p">(</span><span class="m">1</span><span class="p">)></span><span class="w"> </span><span class="nv">%P</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i32</span><span class="p">></span><span class="w"> </span><span class="c">; yields vector of low 32 bits for each pointer</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘ptrtoaddr’ instruction converts the pointer or a vector ofpointers value to the underlying integer address (or vector of addresses) oftype ty2. This is different from ptrtoint in that itonly operates on the index bits of the pointer and ignores all other bits, anddoes not capture the provenance of the pointer.`,
|
||
};
|
||
case 'INTTOPTR':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#inttoptr-to-instruction`,
|
||
html: `<html><head></head><body><span id="i-inttoptr"></span><h4><a class="toc-backref" href="#id2216" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">inttoptr</span> <span class="pre">..</span> <span class="pre">to</span></code>’ Instruction</a></h4><section id="id296"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><result> = inttoptr <ty> <value> to <ty2>[, !dereferenceable !<deref_bytes_node>][, !dereferenceable_or_null !<deref_bytes_node>][, !nofree !<empty_node>] ; yields ty2</pre></div></div></section><section id="id297"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">inttoptr</span></code>’ instruction converts an integer <code class="docutils literal notranslate"><span class="pre">value</span></code> to apointer type, <code class="docutils literal notranslate"><span class="pre">ty2</span></code>.</p></section><section id="id298"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">inttoptr</span></code>’ instruction takes an <a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> value tocast, and a type to cast it to, which must be a <a class="reference internal" href="#t-pointer"><span class="std std-ref">pointer</span></a>type.</p><p>The optional <code class="docutils literal notranslate"><span class="pre">!dereferenceable</span></code> metadata must reference a single metadataname <code class="docutils literal notranslate"><span class="pre"><deref_bytes_node></span></code> corresponding to a metadata node with one <code class="docutils literal notranslate"><span class="pre">i64</span></code>entry.See <code class="docutils literal notranslate"><span class="pre">dereferenceable</span></code> metadata.</p><p>The optional <code class="docutils literal notranslate"><span class="pre">!dereferenceable_or_null</span></code> metadata must reference a singlemetadata name <code class="docutils literal notranslate"><span class="pre"><deref_bytes_node></span></code> corresponding to a metadata node with one<code class="docutils literal notranslate"><span class="pre">i64</span></code> entry.See <code class="docutils literal notranslate"><span class="pre">dereferenceable_or_null</span></code> metadata.</p><p>The optional <code class="docutils literal notranslate"><span class="pre">!nofree</span></code> metadata must reference a single metadata name<code class="docutils literal notranslate"><span class="pre"><empty_node></span></code> corresponding to a metadata node with no entries.The existence of the <code class="docutils literal notranslate"><span class="pre">!nofree</span></code> metadata on the instruction tells the optimizerthat the memory pointed by the pointer will not be freed after this point.</p></section><section id="id299"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">inttoptr</span></code>’ instruction converts <code class="docutils literal notranslate"><span class="pre">value</span></code> to type <code class="docutils literal notranslate"><span class="pre">ty2</span></code> byapplying either a zero extension or a truncation depending on the sizeof the integer <code class="docutils literal notranslate"><span class="pre">value</span></code>. If <code class="docutils literal notranslate"><span class="pre">value</span></code> is larger than the size of apointer then a truncation is done. If <code class="docutils literal notranslate"><span class="pre">value</span></code> is smaller than the sizeof a pointer then a zero extension is done. If they are the same size,nothing is done (<em>no-op cast</em>).The behavior is equivalent to a <code class="docutils literal notranslate"><span class="pre">bitcast</span></code>, however, the resulting value is notguaranteed to be dereferenceable (e.g., if the result type is a<a class="reference internal" href="#nointptrtype"><span class="std std-ref">non-integral pointers</span></a>).</p></section><section id="id300"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%X</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">inttoptr</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">255</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="c">; yields zero extension on 64-bit architecture</span><span class="nv">%Y</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">inttoptr</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">255</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="c">; yields no-op on 32-bit architecture</span><span class="nv">%Z</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">inttoptr</span><span class="w"> </span><span class="kt">i64</span><span class="w"> </span><span class="m">0</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="c">; yields truncation on 32-bit architecture</span><span class="nv">%Z</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">inttoptr</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">i32</span><span class="p">></span><span class="w"> </span><span class="nv">%G</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">ptr</span><span class="p">></span><span class="c">; yields truncation of vector G to four pointers</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘inttoptr’ instruction converts an integer value to apointer type, ty2.`,
|
||
};
|
||
case 'BITCAST':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#bitcast-to-instruction`,
|
||
html: `<html><head></head><body><span id="i-bitcast"></span><h4><a class="toc-backref" href="#id2217" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">bitcast</span> <span class="pre">..</span> <span class="pre">to</span></code>’ Instruction</a></h4><section id="id301"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">bitcast</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span> <span class="n">to</span> <span class="o"><</span><span class="n">ty2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty2</span></pre></div></div></section><section id="id302"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">bitcast</span></code>’ instruction converts <code class="docutils literal notranslate"><span class="pre">value</span></code> to type <code class="docutils literal notranslate"><span class="pre">ty2</span></code> withoutchanging any bits.</p></section><section id="id303"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">bitcast</span></code>’ instruction takes a value to cast, which must be anon-aggregate first class value, and a type to cast it to, which mustalso be a non-aggregate <a class="reference internal" href="#t-firstclass"><span class="std std-ref">first class</span></a> type. Thebit sizes of <code class="docutils literal notranslate"><span class="pre">value</span></code> and the destination type, <code class="docutils literal notranslate"><span class="pre">ty2</span></code>, must beidentical. If the source type is a pointer, the destination type mustalso be a pointer of the same size. This instruction supports bitwiseconversion of vectors to integers and to vectors of other types (aslong as they have the same size).</p></section><section id="id304"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">bitcast</span></code>’ instruction converts <code class="docutils literal notranslate"><span class="pre">value</span></code> to type <code class="docutils literal notranslate"><span class="pre">ty2</span></code>. Itis always a <em>no-op cast</em> because no bits change with thisconversion. The conversion is done as if the <code class="docutils literal notranslate"><span class="pre">value</span></code> had been storedto memory and read back as type <code class="docutils literal notranslate"><span class="pre">ty2</span></code>. Pointer (or vector ofpointers) types may only be converted to other pointer (or vector ofpointers) types with the same address space through this instruction.To convert pointers to other types, use the <a class="reference internal" href="#i-inttoptr"><span class="std std-ref">inttoptr</span></a>or <a class="reference internal" href="#i-ptrtoint"><span class="std std-ref">ptrtoint</span></a> instructions first.</p><p>There is a caveat for bitcasts involving vector types in relation toendianness. For example <code class="docutils literal notranslate"><span class="pre">bitcast</span> <span class="pre"><2</span> <span class="pre">x</span> <span class="pre">i8></span> <span class="pre"><value></span> <span class="pre">to</span> <span class="pre">i16</span></code> puts element zeroof the vector in the least significant bits of the i16 for little-endian whileelement zero ends up in the most significant bits for big-endian.</p></section><section id="id305"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span>%X = bitcast i8 255 to i8 ; yields i8 :-1%Y = bitcast i32* %x to i16* ; yields i16*:%x%Z = bitcast <2 x i32> %V to i64; ; yields i64: %V (depends on endianness)%Z = bitcast <2 x i32*> %V to <2 x i64*> ; yields <2 x i64*></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘bitcast’ instruction converts value to type ty2 withoutchanging any bits.`,
|
||
};
|
||
case 'ADDRSPACECAST':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#addrspacecast-to-instruction`,
|
||
html: `<html><head></head><body><span id="i-addrspacecast"></span><h4><a class="toc-backref" href="#id2218" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">addrspacecast</span> <span class="pre">..</span> <span class="pre">to</span></code>’ Instruction</a></h4><section id="id306"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">addrspacecast</span> <span class="o"><</span><span class="n">pty</span><span class="o">></span> <span class="o"><</span><span class="n">ptrval</span><span class="o">></span> <span class="n">to</span> <span class="o"><</span><span class="n">pty2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">pty2</span></pre></div></div></section><section id="id307"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">addrspacecast</span></code>’ instruction converts <code class="docutils literal notranslate"><span class="pre">ptrval</span></code> from <code class="docutils literal notranslate"><span class="pre">pty</span></code> inaddress space <code class="docutils literal notranslate"><span class="pre">n</span></code> to type <code class="docutils literal notranslate"><span class="pre">pty2</span></code> in address space <code class="docutils literal notranslate"><span class="pre">m</span></code>.</p></section><section id="id308"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">addrspacecast</span></code>’ instruction takes a pointer or vector of pointer valueto cast and a pointer type to cast it to, which must have a differentaddress space.</p></section><section id="id309"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">addrspacecast</span></code>’ instruction converts the pointer value<code class="docutils literal notranslate"><span class="pre">ptrval</span></code> to type <code class="docutils literal notranslate"><span class="pre">pty2</span></code>. It can be a <em>no-op cast</em> or a complexvalue modification, depending on the target and the address spacepair. Pointer conversions within the same address space must beperformed with the <code class="docutils literal notranslate"><span class="pre">bitcast</span></code> instruction. Note that if the addressspace conversion produces a dereferenceable result then both resultand operand refer to the same memory location. The conversion musthave no side effects, and must not capture the value of the pointer.</p><p>If the source is <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison</span></a>, the result is<a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison</span></a>.</p><p>If the source is not <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison</span></a>, and both source anddestination are <a class="reference internal" href="#nointptrtype"><span class="std std-ref">integral pointers</span></a>, and theresult pointer is dereferenceable, the cast is assumed to bereversible (i.e., casting the result back to the original address spaceshould yield the original bit pattern).</p><p>Which address space casts are supported depends on the target. Unsupportedaddress space casts return <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison</span></a>.</p></section><section id="id310"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%X</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">addrspacecast</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%x</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="k">addrspace</span><span class="p">(</span><span class="m">1</span><span class="p">)</span><span class="nv">%Y</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">addrspacecast</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="k">addrspace</span><span class="p">(</span><span class="m">1</span><span class="p">)</span><span class="w"> </span><span class="nv">%y</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="k">addrspace</span><span class="p">(</span><span class="m">2</span><span class="p">)</span><span class="nv">%Z</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">addrspacecast</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">ptr</span><span class="p">></span><span class="w"> </span><span class="nv">%z</span><span class="w"> </span><span class="k">to</span><span class="w"> </span><span class="p"><</span><span class="m">4</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="k">addrspace</span><span class="p">(</span><span class="m">3</span><span class="p">)></span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘addrspacecast’ instruction converts ptrval from pty inaddress space n to type pty2 in address space m.`,
|
||
};
|
||
case 'ICMP':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#icmp-instruction`,
|
||
html: `<html><head></head><body><span id="i-icmp"></span><h4><a class="toc-backref" href="#id2220" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">icmp</span></code>’ Instruction</a></h4><section id="id311"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">icmp</span> <span class="o"><</span><span class="n">cond</span><span class="o">></span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">i1</span> <span class="ow">or</span> <span class="o"><</span><span class="n">N</span> <span class="n">x</span> <span class="n">i1</span><span class="o">></span><span class="p">:</span><span class="n">result</span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">icmp</span> <span class="n">samesign</span> <span class="o"><</span><span class="n">cond</span><span class="o">></span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">i1</span> <span class="ow">or</span> <span class="o"><</span><span class="n">N</span> <span class="n">x</span> <span class="n">i1</span><span class="o">></span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id312"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">icmp</span></code>’ instruction returns a boolean value or a vector ofboolean values based on comparison of its two integer, integer vector,pointer, or pointer vector operands.</p></section><section id="id313"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">icmp</span></code>’ instruction takes three operands. The first operand isthe condition code indicating the kind of comparison to perform. It isnot a value, just a keyword. The possible condition codes are:</p><ol class="arabic simple" id="icmp-md-cc"><li><p><code class="docutils literal notranslate"><span class="pre">eq</span></code>: equal</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ne</span></code>: not equal</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ugt</span></code>: unsigned greater than</p></li><li><p><code class="docutils literal notranslate"><span class="pre">uge</span></code>: unsigned greater or equal</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ult</span></code>: unsigned less than</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ule</span></code>: unsigned less or equal</p></li><li><p><code class="docutils literal notranslate"><span class="pre">sgt</span></code>: signed greater than</p></li><li><p><code class="docutils literal notranslate"><span class="pre">sge</span></code>: signed greater or equal</p></li><li><p><code class="docutils literal notranslate"><span class="pre">slt</span></code>: signed less than</p></li><li><p><code class="docutils literal notranslate"><span class="pre">sle</span></code>: signed less or equal</p></li></ol><p>The remaining two arguments must be <a class="reference internal" href="#t-integer"><span class="std std-ref">integer</span></a> or<a class="reference internal" href="#t-pointer"><span class="std std-ref">pointer</span></a> or integer <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> typed. Theymust also be identical types.</p></section><section id="id314"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">icmp</span></code>’ compares <code class="docutils literal notranslate"><span class="pre">op1</span></code> and <code class="docutils literal notranslate"><span class="pre">op2</span></code> according to the conditioncode given as <code class="docutils literal notranslate"><span class="pre">cond</span></code>. The comparison performed always yields either an<a class="reference internal" href="#t-integer"><span class="std std-ref">i1</span></a> or vector of <code class="docutils literal notranslate"><span class="pre">i1</span></code> result, as follows:</p><ol class="arabic simple" id="icmp-md-cc-sem"><li><p><code class="docutils literal notranslate"><span class="pre">eq</span></code>: yields <code class="docutils literal notranslate"><span class="pre">true</span></code> if the operands are equal, <code class="docutils literal notranslate"><span class="pre">false</span></code>otherwise. No sign interpretation is necessary or performed.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ne</span></code>: yields <code class="docutils literal notranslate"><span class="pre">true</span></code> if the operands are unequal, <code class="docutils literal notranslate"><span class="pre">false</span></code>otherwise. No sign interpretation is necessary or performed.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ugt</span></code>: interprets the operands as unsigned values and yields<code class="docutils literal notranslate"><span class="pre">true</span></code> if <code class="docutils literal notranslate"><span class="pre">op1</span></code> is greater than <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">uge</span></code>: interprets the operands as unsigned values and yields<code class="docutils literal notranslate"><span class="pre">true</span></code> if <code class="docutils literal notranslate"><span class="pre">op1</span></code> is greater than or equal to <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ult</span></code>: interprets the operands as unsigned values and yields<code class="docutils literal notranslate"><span class="pre">true</span></code> if <code class="docutils literal notranslate"><span class="pre">op1</span></code> is less than <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ule</span></code>: interprets the operands as unsigned values and yields<code class="docutils literal notranslate"><span class="pre">true</span></code> if <code class="docutils literal notranslate"><span class="pre">op1</span></code> is less than or equal to <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">sgt</span></code>: interprets the operands as signed values and yields <code class="docutils literal notranslate"><span class="pre">true</span></code>if <code class="docutils literal notranslate"><span class="pre">op1</span></code> is greater than <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">sge</span></code>: interprets the operands as signed values and yields <code class="docutils literal notranslate"><span class="pre">true</span></code>if <code class="docutils literal notranslate"><span class="pre">op1</span></code> is greater than or equal to <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">slt</span></code>: interprets the operands as signed values and yields <code class="docutils literal notranslate"><span class="pre">true</span></code>if <code class="docutils literal notranslate"><span class="pre">op1</span></code> is less than <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">sle</span></code>: interprets the operands as signed values and yields <code class="docutils literal notranslate"><span class="pre">true</span></code>if <code class="docutils literal notranslate"><span class="pre">op1</span></code> is less than or equal to <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li></ol><p>If the operands are <a class="reference internal" href="#t-pointer"><span class="std std-ref">pointer</span></a> typed, the pointer valuesare compared as if they were integers.</p><p>If the operands are integer vectors, then they are compared element byelement. The result is an <code class="docutils literal notranslate"><span class="pre">i1</span></code> vector with the same number of elementsas the values being compared. Otherwise, the result is an <code class="docutils literal notranslate"><span class="pre">i1</span></code>.</p><p>If the <code class="docutils literal notranslate"><span class="pre">samesign</span></code> keyword is present and the operands are not of thesame sign then the result is a <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison value</span></a>.</p></section><section id="id315"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = icmp eq i32 4, 5 ; yields: result=false<result> = icmp ne ptr %X, %X ; yields: result=false<result> = icmp ult i16 4, 5 ; yields: result=true<result> = icmp sgt i16 4, 5 ; yields: result=false<result> = icmp ule i16 -4, 5 ; yields: result=false<result> = icmp sge i16 4, 5 ; yields: result=false</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘icmp’ instruction returns a boolean value or a vector ofboolean values based on comparison of its two integer, integer vector,pointer, or pointer vector operands.`,
|
||
};
|
||
case 'FCMP':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#fcmp-instruction`,
|
||
html: `<html><head></head><body><span id="i-fcmp"></span><h4><a class="toc-backref" href="#id2221" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">fcmp</span></code>’ Instruction</a></h4><section id="id316"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">fcmp</span> <span class="p">[</span><span class="n">fast</span><span class="o">-</span><span class="n">math</span> <span class="n">flags</span><span class="p">]</span><span class="o">*</span> <span class="o"><</span><span class="n">cond</span><span class="o">></span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">op1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">op2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">i1</span> <span class="ow">or</span> <span class="o"><</span><span class="n">N</span> <span class="n">x</span> <span class="n">i1</span><span class="o">></span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id317"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fcmp</span></code>’ instruction returns a boolean value or vector of booleanvalues based on comparison of its operands.</p><p>If the operands are floating-point scalars, then the result type is aboolean (<a class="reference internal" href="#t-integer"><span class="std std-ref">i1</span></a>).</p><p>If the operands are floating-point vectors, then the result type is avector of boolean with the same number of elements as the operands beingcompared.</p></section><section id="id318"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fcmp</span></code>’ instruction takes three operands. The first operand isthe condition code indicating the kind of comparison to perform. It isnot a value, just a keyword. The possible condition codes are:</p><ol class="arabic simple"><li><p><code class="docutils literal notranslate"><span class="pre">false</span></code>: no comparison, always returns false</p></li><li><p><code class="docutils literal notranslate"><span class="pre">oeq</span></code>: ordered and equal</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ogt</span></code>: ordered and greater than</p></li><li><p><code class="docutils literal notranslate"><span class="pre">oge</span></code>: ordered and greater than or equal</p></li><li><p><code class="docutils literal notranslate"><span class="pre">olt</span></code>: ordered and less than</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ole</span></code>: ordered and less than or equal</p></li><li><p><code class="docutils literal notranslate"><span class="pre">one</span></code>: ordered and not equal</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ord</span></code>: ordered (no nans)</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ueq</span></code>: unordered or equal</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ugt</span></code>: unordered or greater than</p></li><li><p><code class="docutils literal notranslate"><span class="pre">uge</span></code>: unordered or greater than or equal</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ult</span></code>: unordered or less than</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ule</span></code>: unordered or less than or equal</p></li><li><p><code class="docutils literal notranslate"><span class="pre">une</span></code>: unordered or not equal</p></li><li><p><code class="docutils literal notranslate"><span class="pre">uno</span></code>: unordered (either nans)</p></li><li><p><code class="docutils literal notranslate"><span class="pre">true</span></code>: no comparison, always returns true</p></li></ol><p><em>Ordered</em> means that neither operand is a QNAN while <em>unordered</em> meansthat either operand may be a QNAN.</p><p>Each of <code class="docutils literal notranslate"><span class="pre">val1</span></code> and <code class="docutils literal notranslate"><span class="pre">val2</span></code> arguments must be either a <a class="reference internal" href="#t-floating"><span class="std std-ref">floating-point</span></a> type or a <a class="reference internal" href="#t-vector"><span class="std std-ref">vector</span></a> of floating-point type.They must have identical types.</p></section><section id="id319"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">fcmp</span></code>’ instruction compares <code class="docutils literal notranslate"><span class="pre">op1</span></code> and <code class="docutils literal notranslate"><span class="pre">op2</span></code> according to thecondition code given as <code class="docutils literal notranslate"><span class="pre">cond</span></code>. If the operands are vectors, then thevectors are compared element by element. Each comparison performedalways yields an <a class="reference internal" href="#t-integer"><span class="std std-ref">i1</span></a> result, as follows:</p><ol class="arabic simple"><li><p><code class="docutils literal notranslate"><span class="pre">false</span></code>: always yields <code class="docutils literal notranslate"><span class="pre">false</span></code>, regardless of operands.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">oeq</span></code>: yields <code class="docutils literal notranslate"><span class="pre">true</span></code> if both operands are not a QNAN and <code class="docutils literal notranslate"><span class="pre">op1</span></code>is equal to <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ogt</span></code>: yields <code class="docutils literal notranslate"><span class="pre">true</span></code> if both operands are not a QNAN and <code class="docutils literal notranslate"><span class="pre">op1</span></code>is greater than <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">oge</span></code>: yields <code class="docutils literal notranslate"><span class="pre">true</span></code> if both operands are not a QNAN and <code class="docutils literal notranslate"><span class="pre">op1</span></code>is greater than or equal to <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">olt</span></code>: yields <code class="docutils literal notranslate"><span class="pre">true</span></code> if both operands are not a QNAN and <code class="docutils literal notranslate"><span class="pre">op1</span></code>is less than <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ole</span></code>: yields <code class="docutils literal notranslate"><span class="pre">true</span></code> if both operands are not a QNAN and <code class="docutils literal notranslate"><span class="pre">op1</span></code>is less than or equal to <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">one</span></code>: yields <code class="docutils literal notranslate"><span class="pre">true</span></code> if both operands are not a QNAN and <code class="docutils literal notranslate"><span class="pre">op1</span></code>is not equal to <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ord</span></code>: yields <code class="docutils literal notranslate"><span class="pre">true</span></code> if both operands are not a QNAN.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ueq</span></code>: yields <code class="docutils literal notranslate"><span class="pre">true</span></code> if either operand is a QNAN or <code class="docutils literal notranslate"><span class="pre">op1</span></code> isequal to <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ugt</span></code>: yields <code class="docutils literal notranslate"><span class="pre">true</span></code> if either operand is a QNAN or <code class="docutils literal notranslate"><span class="pre">op1</span></code> isgreater than <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">uge</span></code>: yields <code class="docutils literal notranslate"><span class="pre">true</span></code> if either operand is a QNAN or <code class="docutils literal notranslate"><span class="pre">op1</span></code> isgreater than or equal to <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ult</span></code>: yields <code class="docutils literal notranslate"><span class="pre">true</span></code> if either operand is a QNAN or <code class="docutils literal notranslate"><span class="pre">op1</span></code> isless than <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">ule</span></code>: yields <code class="docutils literal notranslate"><span class="pre">true</span></code> if either operand is a QNAN or <code class="docutils literal notranslate"><span class="pre">op1</span></code> isless than or equal to <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">une</span></code>: yields <code class="docutils literal notranslate"><span class="pre">true</span></code> if either operand is a QNAN or <code class="docutils literal notranslate"><span class="pre">op1</span></code> isnot equal to <code class="docutils literal notranslate"><span class="pre">op2</span></code>.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">uno</span></code>: yields <code class="docutils literal notranslate"><span class="pre">true</span></code> if either operand is a QNAN.</p></li><li><p><code class="docutils literal notranslate"><span class="pre">true</span></code>: always yields <code class="docutils literal notranslate"><span class="pre">true</span></code>, regardless of operands.</p></li></ol><p>The <code class="docutils literal notranslate"><span class="pre">fcmp</span></code> instruction can also optionally take any number of<a class="reference internal" href="#fastmath"><span class="std std-ref">fast-math flags</span></a>, which are optimization hints to enableotherwise unsafe floating-point optimizations.</p><p>Any set of fast-math flags are legal on an <code class="docutils literal notranslate"><span class="pre">fcmp</span></code> instruction, but theonly flags that have any effect on its semantics are those that allowassumptions to be made about the values of input arguments; namely<code class="docutils literal notranslate"><span class="pre">nnan</span></code>, <code class="docutils literal notranslate"><span class="pre">ninf</span></code>, and <code class="docutils literal notranslate"><span class="pre">reassoc</span></code>. See <a class="reference internal" href="#fastmath"><span class="std std-ref">Fast-Math Flags</span></a> for more information.</p></section><section id="id320"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span><result> = fcmp oeq float 4.0, 5.0 ; yields: result=false<result> = fcmp one float 4.0, 5.0 ; yields: result=true<result> = fcmp olt float 4.0, 5.0 ; yields: result=true<result> = fcmp ueq double 1.0, 2.0 ; yields: result=false</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘fcmp’ instruction returns a boolean value or vector of booleanvalues based on comparison of its operands.If the operands are floating-point scalars, then the result type is aboolean (i1).If the operands are floating-point vectors, then the result type is avector of boolean with the same number of elements as the operands beingcompared.`,
|
||
};
|
||
case 'PHI':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#phi-instruction`,
|
||
html: `<html><head></head><body><span id="i-phi"></span><h4><a class="toc-backref" href="#id2222" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">phi</span></code>’ Instruction</a></h4><section id="id321"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">phi</span> <span class="p">[</span><span class="n">fast</span><span class="o">-</span><span class="n">math</span><span class="o">-</span><span class="n">flags</span><span class="p">]</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="p">[</span> <span class="o"><</span><span class="n">val0</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">label0</span><span class="o">></span><span class="p">],</span> <span class="o">...</span></pre></div></div></section><section id="id322"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">phi</span></code>’ instruction is used to implement the φ node in the SSAgraph representing the function.</p></section><section id="id323"><h5>Arguments:</h5><p>The type of the incoming values is specified with the first type field.After this, the ‘<code class="docutils literal notranslate"><span class="pre">phi</span></code>’ instruction takes a list of pairs asarguments, with one pair for each predecessor basic block of the currentblock. Only values of <a class="reference internal" href="#t-firstclass"><span class="std std-ref">first class</span></a> type may be used asthe value arguments to the PHI node. Only labels may be used as thelabel arguments.</p><p>There must be no non-phi instructions between the start of a basic blockand the PHI instructions: i.e., PHI instructions must be first in a basicblock.</p><p>For the purposes of the SSA form, the use of each incoming value isdeemed to occur on the edge from the corresponding predecessor block tothe current block (but after any definition of an ‘<code class="docutils literal notranslate"><span class="pre">invoke</span></code>’instruction’s return value on the same edge).</p><p>The optional <code class="docutils literal notranslate"><span class="pre">fast-math-flags</span></code> marker indicates that the phi has oneor more <a class="reference internal" href="#fastmath"><span class="std std-ref">fast-math-flags</span></a>. These are optimization hintsto enable otherwise unsafe floating-point optimizations. Fast-math-flagsare only valid for phis that return <a class="reference internal" href="#fastmath-return-types"><span class="std std-ref">supported floating-point types</span></a>.</p></section><section id="id324"><h5>Semantics:</h5><p>At runtime, the ‘<code class="docutils literal notranslate"><span class="pre">phi</span></code>’ instruction logically takes on the valuespecified by the pair corresponding to the predecessor basic block thatexecuted just prior to the current block.</p></section><section id="id325"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nl">Loop:</span><span class="w"> </span><span class="c">; Infinite loop that counts from 0 on up...</span><span class="w"> </span><span class="nv">%indvar</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">phi</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="p">[</span><span class="w"> </span><span class="m">0</span><span class="p">,</span><span class="w"> </span><span class="nv">%LoopHeader</span><span class="w"> </span><span class="p">],</span><span class="w"> </span><span class="p">[</span><span class="w"> </span><span class="nv">%nextindvar</span><span class="p">,</span><span class="w"> </span><span class="nv">%Loop</span><span class="w"> </span><span class="p">]</span><span class="w"> </span><span class="nv">%nextindvar</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">add</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="nv">%indvar</span><span class="p">,</span><span class="w"> </span><span class="m">1</span><span class="w"> </span><span class="k">br</span><span class="w"> </span><span class="kt">label</span><span class="w"> </span><span class="nv">%Loop</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘phi’ instruction is used to implement the φ node in the SSAgraph representing the function.`,
|
||
};
|
||
case 'SELECT':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#select-instruction`,
|
||
html: `<html><head></head><body><span id="i-select"></span><h4><a class="toc-backref" href="#id2223" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">select</span></code>’ Instruction</a></h4><section id="id326"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">select</span> <span class="p">[</span><span class="n">fast</span><span class="o">-</span><span class="n">math</span> <span class="n">flags</span><span class="p">]</span> <span class="n">selty</span> <span class="o"><</span><span class="n">cond</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">val1</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">ty</span><span class="o">></span> <span class="o"><</span><span class="n">val2</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="n">selty</span> <span class="ow">is</span> <span class="n">either</span> <span class="n">i1</span> <span class="ow">or</span> <span class="p">{</span><span class="o"><</span><span class="n">N</span> <span class="n">x</span> <span class="n">i1</span><span class="o">></span><span class="p">}</span></pre></div></div></section><section id="id327"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">select</span></code>’ instruction is used to choose one value based on acondition, without IR-level branching.</p></section><section id="id328"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">select</span></code>’ instruction requires an ‘i1’ value or a vector of ‘i1’values indicating the condition, and two values of the same <a class="reference internal" href="#t-firstclass"><span class="std std-ref">firstclass</span></a> type.</p><ol class="arabic simple"><li><p>The optional <code class="docutils literal notranslate"><span class="pre">fast-math</span> <span class="pre">flags</span></code> marker indicates that the select has one or more<a class="reference internal" href="#fastmath"><span class="std std-ref">fast-math flags</span></a>. These are optimization hints to enableotherwise unsafe floating-point optimizations. Fast-math flags are only validfor selects that return <a class="reference internal" href="#fastmath-return-types"><span class="std std-ref">supported floating-point types</span></a>. Note that the presence of value which would otherwise resultin poison does not cause the result to be poison if the value is on the non-selected arm.If <a class="reference internal" href="#fastmath"><span class="std std-ref">fast-math flags</span></a> are present, they are only applied to the result,not both arms.</p></li></ol></section><section id="id329"><h5>Semantics:</h5><p>If the condition is an i1 and it evaluates to 1, the instruction returnsthe first value argument; otherwise, it returns the second valueargument.</p><p>If the condition is a vector of i1, then the value arguments must bevectors of the same size, and the selection is done element by element.</p><p>If the condition is an i1 and the value arguments are vectors of thesame size, then an entire vector is selected.</p></section><section id="id330"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%X</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">select</span><span class="w"> </span><span class="kt">i1</span><span class="w"> </span><span class="k">true</span><span class="p">,</span><span class="w"> </span><span class="kt">i8</span><span class="w"> </span><span class="m">17</span><span class="p">,</span><span class="w"> </span><span class="kt">i8</span><span class="w"> </span><span class="m">42</span><span class="w"> </span><span class="c">; yields i8:17</span><span class="nv">%Y</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">select</span><span class="w"> </span><span class="k">nnan</span><span class="w"> </span><span class="kt">i1</span><span class="w"> </span><span class="k">true</span><span class="p">,</span><span class="w"> </span><span class="kt">float</span><span class="w"> </span><span class="m">0.0</span><span class="p">,</span><span class="w"> </span><span class="kt">float</span><span class="w"> </span><span class="err">NaN</span><span class="w"> </span><span class="c">; yields float:0.0</span><span class="nv">%Z</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">select</span><span class="w"> </span><span class="k">nnan</span><span class="w"> </span><span class="kt">i1</span><span class="w"> </span><span class="k">false</span><span class="p">,</span><span class="w"> </span><span class="kt">float</span><span class="w"> </span><span class="m">0.0</span><span class="p">,</span><span class="w"> </span><span class="kt">float</span><span class="w"> </span><span class="err">NaN</span><span class="w"> </span><span class="c">; yields float:poison</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘select’ instruction is used to choose one value based on acondition, without IR-level branching.`,
|
||
};
|
||
case 'FREEZE':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#freeze-instruction`,
|
||
html: `<html><head></head><body><span id="i-freeze"></span><h4><a class="toc-backref" href="#id2224" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">freeze</span></code>’ Instruction</a></h4><section id="id331"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="n">freeze</span> <span class="n">ty</span> <span class="o"><</span><span class="n">val</span><span class="o">></span> <span class="p">;</span> <span class="n">yields</span> <span class="n">ty</span><span class="p">:</span><span class="n">result</span></pre></div></div></section><section id="id332"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">freeze</span></code>’ instruction is used to stop propagation of<a class="reference internal" href="#undefvalues"><span class="std std-ref">undef</span></a> and <a class="reference internal" href="#poisonvalues"><span class="std std-ref">poison</span></a> values.</p></section><section id="id333"><h5>Arguments:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">freeze</span></code>’ instruction takes a single argument.</p></section><section id="id334"><h5>Semantics:</h5><p>If the argument is <code class="docutils literal notranslate"><span class="pre">undef</span></code> or <code class="docutils literal notranslate"><span class="pre">poison</span></code>, ‘<code class="docutils literal notranslate"><span class="pre">freeze</span></code>’ returns anarbitrary, but fixed, value of type ‘<code class="docutils literal notranslate"><span class="pre">ty</span></code>’.Otherwise, this instruction is a no-op and returns the input argument.All uses of a value returned by the same ‘<code class="docutils literal notranslate"><span class="pre">freeze</span></code>’ instruction areguaranteed to always observe the same value, while different ‘<code class="docutils literal notranslate"><span class="pre">freeze</span></code>’instructions may yield different values.</p><p>While <code class="docutils literal notranslate"><span class="pre">undef</span></code> and <code class="docutils literal notranslate"><span class="pre">poison</span></code> pointers can be frozen, the result is anon-dereferenceable pointer. See the<a class="reference internal" href="#pointeraliasing"><span class="std std-ref">Pointer Aliasing Rules</span></a> section for more information.If an aggregate value or vector is frozen, the operand is frozen element-wise.The padding of an aggregate isn’t considered, since it isn’t visiblewithout storing it into memory and loading it with a different type.</p></section><section id="id335"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span>%w = i32 undef%x = freeze i32 %w%y = add i32 %w, %w ; undef%z = add i32 %x, %x ; even number because all uses of %x observe ; the same value%x2 = freeze i32 %w%cmp = icmp eq i32 %x, %x2 ; can be true or false; example with vectors%v = <2 x i32> <i32 undef, i32 poison>%a = extractelement <2 x i32> %v, i32 0 ; undef%b = extractelement <2 x i32> %v, i32 1 ; poison%add = add i32 %a, %a ; undef%v.fr = freeze <2 x i32> %v ; element-wise freeze%d = extractelement <2 x i32> %v.fr, i32 0 ; not undef%add.f = add i32 %d, %d ; even number; branching on frozen value%poison = add nsw i1 %k, undef ; poison%c = freeze i1 %poisonbr i1 %c, label %foo, label %bar ; non-deterministic branch to %foo or %bar</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘freeze’ instruction is used to stop propagation ofundef and poison values.`,
|
||
};
|
||
case 'CALL':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#call-instruction`,
|
||
html: `<html><head></head><body><span id="i-call"></span><h4><a class="toc-backref" href="#id2225" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">call</span></code>’ Instruction</a></h4><section id="id336"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">result</span><span class="o">></span> <span class="o">=</span> <span class="p">[</span><span class="n">tail</span> <span class="o">|</span> <span class="n">musttail</span> <span class="o">|</span> <span class="n">notail</span> <span class="p">]</span> <span class="n">call</span> <span class="p">[</span><span class="n">fast</span><span class="o">-</span><span class="n">math</span> <span class="n">flags</span><span class="p">]</span> <span class="p">[</span><span class="n">cconv</span><span class="p">]</span> <span class="p">[</span><span class="n">ret</span> <span class="n">attrs</span><span class="p">]</span> <span class="p">[</span><span class="n">addrspace</span><span class="p">(</span><span class="o"><</span><span class="n">num</span><span class="o">></span><span class="p">)]</span> <span class="o"><</span><span class="n">ty</span><span class="o">>|<</span><span class="n">fnty</span><span class="o">></span> <span class="o"><</span><span class="n">fnptrval</span><span class="o">></span><span class="p">(</span><span class="o"><</span><span class="n">function</span> <span class="n">args</span><span class="o">></span><span class="p">)</span> <span class="p">[</span><span class="n">fn</span> <span class="n">attrs</span><span class="p">]</span> <span class="p">[</span> <span class="n">operand</span> <span class="n">bundles</span> <span class="p">]</span></pre></div></div></section><section id="id337"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">call</span></code>’ instruction represents a simple function call.</p></section><section id="id338"><h5>Arguments:</h5><p>This instruction requires several arguments:</p><ol class="arabic"><li><p>The optional <code class="docutils literal notranslate"><span class="pre">tail</span></code> and <code class="docutils literal notranslate"><span class="pre">musttail</span></code> markers indicate that the optimizersshould perform tail call optimization. The <code class="docutils literal notranslate"><span class="pre">tail</span></code> marker is a hint that<a class="reference external" href="CodeGenerator.html#tail-call-optimization">can be ignored</a>. The<code class="docutils literal notranslate"><span class="pre">musttail</span></code> marker means that the call must be tail call optimized in orderfor the program to be correct. This is true even in the presence ofattributes like “disable-tail-calls”. The <code class="docutils literal notranslate"><span class="pre">musttail</span></code> marker provides theseguarantees:</p><ul class="simple"><li><p>The call will not cause unbounded stack growth if it is part of arecursive cycle in the call graph.</p></li><li><p>Arguments with the <a class="reference internal" href="#attr-inalloca"><span class="std std-ref">inalloca</span></a> or<a class="reference internal" href="#attr-preallocated"><span class="std std-ref">preallocated</span></a> attribute are forwarded in place.</p></li><li><p>If the musttail call appears in a function with the <code class="docutils literal notranslate"><span class="pre">"thunk"</span></code> attributeand the caller and callee both have varargs, then any unprototypedarguments in register or memory are forwarded to the callee. Similarly,the return value of the callee is returned to the caller’s caller, evenif a void return type is in use.</p></li></ul><p>Both markers imply that the callee does not access allocas, va_args, orbyval arguments from the caller. As an exception to that, an alloca or byvalargument may be passed to the callee as a byval argument, which can bedereferenced inside the callee. For example:</p><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="k">declare</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="vg">@take_byval</span><span class="p">(</span><span class="kt">ptr</span><span class="w"> </span><span class="k">byval</span><span class="p">(</span><span class="kt">i64</span><span class="p">))</span><span class="k">declare</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="vg">@take_ptr</span><span class="p">(</span><span class="kt">ptr</span><span class="p">)</span><span class="c">; Invalid (assuming @take_ptr dereferences the pointer), because %local</span><span class="c">; may be de-allocated before the call to @take_ptr.</span><span class="k">define</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="vg">@invalid_alloca</span><span class="p">()</span><span class="w"> </span><span class="p">{</span><span class="nl">entry:</span><span class="w"> </span><span class="nv">%local</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">alloca</span><span class="w"> </span><span class="kt">i64</span><span class="w"> </span><span class="k">tail</span><span class="w"> </span><span class="k">call</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="vg">@take_ptr</span><span class="p">(</span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%local</span><span class="p">)</span><span class="w"> </span><span class="k">ret</span><span class="w"> </span><span class="k">void</span><span class="p">}</span><span class="c">; Valid, the byval attribute causes the memory allocated by %local to be</span><span class="c">; copied into @take_byval's stack frame.</span><span class="k">define</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="vg">@byval_alloca</span><span class="p">()</span><span class="w"> </span><span class="p">{</span><span class="nl">entry:</span><span class="w"> </span><span class="nv">%local</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">alloca</span><span class="w"> </span><span class="kt">i64</span><span class="w"> </span><span class="k">tail</span><span class="w"> </span><span class="k">call</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="vg">@take_byval</span><span class="p">(</span><span class="kt">ptr</span><span class="w"> </span><span class="k">byval</span><span class="p">(</span><span class="kt">i64</span><span class="p">)</span><span class="w"> </span><span class="nv">%local</span><span class="p">)</span><span class="w"> </span><span class="k">ret</span><span class="w"> </span><span class="k">void</span><span class="p">}</span><span class="c">; Invalid, because @use_global_va_list uses the variadic arguments from</span><span class="c">; @invalid_va_list.</span><span class="nv">%struct.va_list</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">type</span><span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="p">}</span><span class="vg">@va_list</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">external</span><span class="w"> </span><span class="k">global</span><span class="w"> </span><span class="nv">%struct.va_list</span><span class="k">define</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="vg">@use_global_va_list</span><span class="p">()</span><span class="w"> </span><span class="p">{</span><span class="nl">entry:</span><span class="w"> </span><span class="nv">%arg</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">va_arg</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="vg">@va_list</span><span class="p">,</span><span class="w"> </span><span class="kt">i64</span><span class="w"> </span><span class="k">ret</span><span class="w"> </span><span class="k">void</span><span class="p">}</span><span class="k">define</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="vg">@invalid_va_list</span><span class="p">(</span><span class="kt">i32</span><span class="w"> </span><span class="nv">%a</span><span class="p">,</span><span class="w"> </span><span class="p">...)</span><span class="w"> </span><span class="p">{</span><span class="nl">entry:</span><span class="w"> </span><span class="k">call</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="vg">@llvm.va_start.p0</span><span class="p">(</span><span class="kt">ptr</span><span class="w"> </span><span class="vg">@va_list</span><span class="p">)</span><span class="w"> </span><span class="k">tail</span><span class="w"> </span><span class="k">call</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="vg">@use_global_va_list</span><span class="p">()</span><span class="w"> </span><span class="k">ret</span><span class="w"> </span><span class="k">void</span><span class="p">}</span><span class="c">; Valid, byval argument forwarded to tail call as another byval argument.</span><span class="k">define</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="vg">@forward_byval</span><span class="p">(</span><span class="kt">ptr</span><span class="w"> </span><span class="k">byval</span><span class="p">(</span><span class="kt">i64</span><span class="p">)</span><span class="w"> </span><span class="nv">%x</span><span class="p">)</span><span class="w"> </span><span class="p">{</span><span class="nl">entry:</span><span class="w"> </span><span class="k">tail</span><span class="w"> </span><span class="k">call</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="vg">@take_byval</span><span class="p">(</span><span class="kt">ptr</span><span class="w"> </span><span class="k">byval</span><span class="p">(</span><span class="kt">i64</span><span class="p">)</span><span class="w"> </span><span class="nv">%x</span><span class="p">)</span><span class="w"> </span><span class="k">ret</span><span class="w"> </span><span class="k">void</span><span class="p">}</span><span class="c">; Invalid (assuming @take_ptr dereferences the pointer), byval argument</span><span class="c">; passed to tail callee as non-byval ptr.</span><span class="k">define</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="vg">@invalid_byval</span><span class="p">(</span><span class="kt">ptr</span><span class="w"> </span><span class="k">byval</span><span class="p">(</span><span class="kt">i64</span><span class="p">)</span><span class="w"> </span><span class="nv">%x</span><span class="p">)</span><span class="w"> </span><span class="p">{</span><span class="nl">entry:</span><span class="w"> </span><span class="k">tail</span><span class="w"> </span><span class="k">call</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="vg">@take_ptr</span><span class="p">(</span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%x</span><span class="p">)</span><span class="w"> </span><span class="k">ret</span><span class="w"> </span><span class="k">void</span><span class="p">}</span></pre></div></div><p>Calls marked <code class="docutils literal notranslate"><span class="pre">musttail</span></code> must obey the following additional rules:</p><ul class="simple"><li><p>The call must immediately precede a <a class="reference internal" href="#i-ret"><span class="std std-ref">ret</span></a> instruction,or a pointer bitcast followed by a ret instruction.</p></li><li><p>The ret instruction must return the (possibly bitcasted) valueproduced by the call, undef, or void.</p></li><li><p>The calling conventions of the caller and callee must match.</p></li><li><p>The callee must be varargs iff the caller is varargs. Bitcasting anon-varargs function to the appropriate varargs type is legal solong as the non-varargs prefixes obey the other rules.</p></li><li><p>The return type must not undergo automatic conversion to an <cite>sret</cite> pointer.</p></li></ul><p>In addition, if the calling convention is not <cite>swifttailcc</cite> or <cite>tailcc</cite>:</p><ul class="simple"><li><p>All ABI-impacting function attributes, such as sret, byval, inreg,returned, and inalloca, must match.</p></li><li><p>The caller and callee prototypes must match. Pointer types of parametersor return types do not differ in address space.</p></li></ul><p>On the other hand, if the calling convention is <cite>swifttailcc</cite> or <cite>tailcc</cite>:</p><ul class="simple"><li><p>Only these ABI-impacting attributes attributes are allowed: sret, byval,swiftself, and swiftasync.</p></li><li><p>Prototypes are not required to match.</p></li></ul><p>Tail call optimization for calls marked <code class="docutils literal notranslate"><span class="pre">tail</span></code> is guaranteed to occur ifthe following conditions are met:</p><ul class="simple"><li><p>Caller and callee both have the calling convention <code class="docutils literal notranslate"><span class="pre">fastcc</span></code> or <code class="docutils literal notranslate"><span class="pre">tailcc</span></code>.</p></li><li><p>The call is in tail position (ret immediately follows call and retuses value of call or is void).</p></li><li><p>Option <code class="docutils literal notranslate"><span class="pre">-tailcallopt</span></code> is enabled, <code class="docutils literal notranslate"><span class="pre">llvm::GuaranteedTailCallOpt</span></code> is<code class="docutils literal notranslate"><span class="pre">true</span></code>, or the calling convention is <code class="docutils literal notranslate"><span class="pre">tailcc</span></code>.</p></li><li><p><a class="reference external" href="CodeGenerator.html#tail-call-optimization">Platform-specific constraints are met.</a></p></li></ul></li><li><p>The optional <code class="docutils literal notranslate"><span class="pre">notail</span></code> marker indicates that the optimizers should not add<code class="docutils literal notranslate"><span class="pre">tail</span></code> or <code class="docutils literal notranslate"><span class="pre">musttail</span></code> markers to the call. It is used to prevent tailcall optimization from being performed on the call.</p></li><li><p>The optional <code class="docutils literal notranslate"><span class="pre">fast-math</span> <span class="pre">flags</span></code> marker indicates that the call has one or more<a class="reference internal" href="#fastmath"><span class="std std-ref">fast-math flags</span></a>, which are optimization hints to enableotherwise unsafe floating-point optimizations. Fast-math flags are only validfor calls that return <a class="reference internal" href="#fastmath-return-types"><span class="std std-ref">supported floating-point types</span></a>.</p></li><li><p>The optional “cconv” marker indicates which <a class="reference internal" href="#callingconv"><span class="std std-ref">callingconvention</span></a> the call should use. If none isspecified, the call defaults to using C calling conventions. Thecalling convention of the call must match the calling convention ofthe target function, or else the behavior is undefined.</p></li><li><p>The optional <a class="reference internal" href="#paramattrs"><span class="std std-ref">Parameter Attributes</span></a> list for returnvalues. Only ‘<code class="docutils literal notranslate"><span class="pre">zeroext</span></code>’, ‘<code class="docutils literal notranslate"><span class="pre">signext</span></code>’, ‘<code class="docutils literal notranslate"><span class="pre">noext</span></code>’, and ‘<code class="docutils literal notranslate"><span class="pre">inreg</span></code>’attributes are valid here.</p></li><li><p>The optional addrspace attribute can be used to indicate the address spaceof the called function. If it is not specified, the program address spacefrom the <a class="reference internal" href="#langref-datalayout"><span class="std std-ref">datalayout string</span></a> will be used.</p></li><li><p>‘<code class="docutils literal notranslate"><span class="pre">ty</span></code>’: the type of the call instruction itself which is also thetype of the return value. Functions that return no value are marked<code class="docutils literal notranslate"><span class="pre">void</span></code>. The signature is computed based on the return type and argumenttypes.</p></li><li><p>‘<code class="docutils literal notranslate"><span class="pre">fnty</span></code>’: shall be the signature of the function being called. Theargument types must match the types implied by this signature. Thisis only required if the signature specifies a varargs type.</p></li><li><p>‘<code class="docutils literal notranslate"><span class="pre">fnptrval</span></code>’: An LLVM value containing a pointer to a function tobe called. In most cases, this is a direct function call, butindirect <code class="docutils literal notranslate"><span class="pre">call</span></code>’s are just as possible, calling an arbitrary pointerto function value.</p></li><li><p>‘<code class="docutils literal notranslate"><span class="pre">function</span> <span class="pre">args</span></code>’: argument list whose types match the functionsignature argument types and parameter attributes. All arguments mustbe of <a class="reference internal" href="#t-firstclass"><span class="std std-ref">first class</span></a> type. If the function signatureindicates the function accepts a variable number of arguments, theextra arguments can be specified.</p></li><li><p>The optional <a class="reference internal" href="#fnattrs"><span class="std std-ref">function attributes</span></a> list.</p></li><li><p>The optional <a class="reference internal" href="#opbundles"><span class="std std-ref">operand bundles</span></a> list.</p></li></ol></section><section id="id339"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">call</span></code>’ instruction is used to cause control flow to transfer toa specified function, with its incoming arguments bound to the specifiedvalues. Upon a ‘<code class="docutils literal notranslate"><span class="pre">ret</span></code>’ instruction in the called function, controlflow continues with the instruction after the function call, and thereturn value of the function is bound to the result argument.</p><p>If the callee refers to an intrinsic function, the signature of the call mustmatch the signature of the callee. Otherwise, if the signature of the calldoes not match the signature of the called function, the behavior istarget-specific. For a significant mismatch, this likely results in undefinedbehavior. LLVM interprocedural optimizations generally only optimize callswhere the signature of the caller matches the signature of the callee.</p><p>Note that it is possible for the signatures to mismatch even if a call appearsto be a “direct” call, like <code class="docutils literal notranslate"><span class="pre">call</span> <span class="pre">void</span> <span class="pre">@f()</span></code>.</p></section><section id="id340"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="nv">%retval</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">call</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="vg">@test</span><span class="p">(</span><span class="kt">i32</span><span class="w"> </span><span class="nv">%argc</span><span class="p">)</span><span class="k">call</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="p">(</span><span class="kt">ptr</span><span class="p">,</span><span class="w"> </span><span class="p">...)</span><span class="w"> </span><span class="vg">@printf</span><span class="p">(</span><span class="kt">ptr</span><span class="w"> </span><span class="nv">%msg</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="m">12</span><span class="p">,</span><span class="w"> </span><span class="kt">i8</span><span class="w"> </span><span class="m">42</span><span class="p">)</span><span class="w"> </span><span class="c">; yields i32</span><span class="nv">%X</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">tail</span><span class="w"> </span><span class="k">call</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="vg">@foo</span><span class="p">()</span><span class="w"> </span><span class="c">; yields i32</span><span class="nv">%Y</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">tail</span><span class="w"> </span><span class="k">call</span><span class="w"> </span><span class="k">fastcc</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="vg">@foo</span><span class="p">()</span><span class="w"> </span><span class="c">; yields i32</span><span class="k">call</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="nv">%foo</span><span class="p">(</span><span class="kt">i8</span><span class="w"> </span><span class="k">signext</span><span class="w"> </span><span class="m">97</span><span class="p">)</span><span class="nv">%struct.A</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">type</span><span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="kt">i32</span><span class="p">,</span><span class="w"> </span><span class="kt">i8</span><span class="w"> </span><span class="p">}</span><span class="nv">%r</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">call</span><span class="w"> </span><span class="nv">%struct.A</span><span class="w"> </span><span class="vg">@foo</span><span class="p">()</span><span class="w"> </span><span class="c">; yields { i32, i8 }</span><span class="nv">%gr</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">extractvalue</span><span class="w"> </span><span class="nv">%struct.A</span><span class="w"> </span><span class="nv">%r</span><span class="p">,</span><span class="w"> </span><span class="m">0</span><span class="w"> </span><span class="c">; yields i32</span><span class="nv">%gr1</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">extractvalue</span><span class="w"> </span><span class="nv">%struct.A</span><span class="w"> </span><span class="nv">%r</span><span class="p">,</span><span class="w"> </span><span class="m">1</span><span class="w"> </span><span class="c">; yields i8</span><span class="nv">%Z</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">call</span><span class="w"> </span><span class="k">void</span><span class="w"> </span><span class="vg">@foo</span><span class="p">()</span><span class="w"> </span><span class="k">noreturn</span><span class="w"> </span><span class="c">; indicates that %foo never returns normally</span><span class="nv">%ZZ</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">call</span><span class="w"> </span><span class="k">zeroext</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="vg">@bar</span><span class="p">()</span><span class="w"> </span><span class="c">; Return value is %zero extended</span></pre></div></div><p>llvm treats calls to some functions with names and arguments that matchthe standard C99 library as being the C99 library functions, and mayperform optimizations or generate code for them under that assumption.This is something we’d like to change in the future to provide bettersupport for freestanding environments and non-C-based languages.</p></section></body></html>`,
|
||
tooltip: `The ‘call’ instruction represents a simple function call.`,
|
||
};
|
||
case 'VA-ARG':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#va-arg-instruction`,
|
||
html: `<html><head></head><body><span id="i-va-arg"></span><h4><a class="toc-backref" href="#id2226" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">va_arg</span></code>’ Instruction</a></h4><section id="id341"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">resultval</span><span class="o">></span> <span class="o">=</span> <span class="n">va_arg</span> <span class="o"><</span><span class="n">va_list</span><span class="o">*></span> <span class="o"><</span><span class="n">arglist</span><span class="o">></span><span class="p">,</span> <span class="o"><</span><span class="n">argty</span><span class="o">></span></pre></div></div></section><section id="id342"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">va_arg</span></code>’ instruction is used to access arguments passed throughthe “variable argument” area of a function call. It is used to implementthe <code class="docutils literal notranslate"><span class="pre">va_arg</span></code> macro in C.</p></section><section id="id343"><h5>Arguments:</h5><p>This instruction takes a <code class="docutils literal notranslate"><span class="pre">va_list*</span></code> value and the type of theargument. It returns a value of the specified argument type andincrements the <code class="docutils literal notranslate"><span class="pre">va_list</span></code> to point to the next argument. The actualtype of <code class="docutils literal notranslate"><span class="pre">va_list</span></code> is target specific.</p></section><section id="id344"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">va_arg</span></code>’ instruction loads an argument of the specified typefrom the specified <code class="docutils literal notranslate"><span class="pre">va_list</span></code> and causes the <code class="docutils literal notranslate"><span class="pre">va_list</span></code> to point tothe next argument. For more information, see the variable argumenthandling <a class="reference internal" href="#int-varargs"><span class="std std-ref">Intrinsic Functions</span></a>.</p><p>It is legal for this instruction to be called in a function which doesnot take a variable number of arguments, for example, the <code class="docutils literal notranslate"><span class="pre">vfprintf</span></code>function.</p><p><code class="docutils literal notranslate"><span class="pre">va_arg</span></code> is an LLVM instruction instead of an <a class="reference internal" href="#intrinsics"><span class="std std-ref">intrinsicfunction</span></a> because it takes a type as an argument.</p></section><section id="id345"><h5>Example:</h5><p>See the <a class="reference internal" href="#int-varargs"><span class="std std-ref">variable argument processing</span></a> section.</p><p>Note that the code generator does not yet fully support va_arg on manytargets. Also, it does not currently support va_arg with aggregatetypes on any target.</p></section></body></html>`,
|
||
tooltip: `The ‘va_arg’ instruction is used to access arguments passed throughthe “variable argument” area of a function call. It is used to implementthe va_arg macro in C.`,
|
||
};
|
||
case 'LANDINGPAD':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#landingpad-instruction`,
|
||
html: `<html><head></head><body><span id="i-landingpad"></span><h4><a class="toc-backref" href="#id2227" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">landingpad</span></code>’ Instruction</a></h4><section id="id346"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">resultval</span><span class="o">></span> <span class="o">=</span> <span class="n">landingpad</span> <span class="o"><</span><span class="n">resultty</span><span class="o">></span> <span class="o"><</span><span class="n">clause</span><span class="o">>+</span><span class="o"><</span><span class="n">resultval</span><span class="o">></span> <span class="o">=</span> <span class="n">landingpad</span> <span class="o"><</span><span class="n">resultty</span><span class="o">></span> <span class="n">cleanup</span> <span class="o"><</span><span class="n">clause</span><span class="o">>*</span><span class="o"><</span><span class="n">clause</span><span class="o">></span> <span class="o">:=</span> <span class="n">catch</span> <span class="o"><</span><span class="nb">type</span><span class="o">></span> <span class="o"><</span><span class="n">value</span><span class="o">></span><span class="o"><</span><span class="n">clause</span><span class="o">></span> <span class="o">:=</span> <span class="nb">filter</span> <span class="o"><</span><span class="n">array</span> <span class="n">constant</span> <span class="nb">type</span><span class="o">></span> <span class="o"><</span><span class="n">array</span> <span class="n">constant</span><span class="o">></span></pre></div></div></section><section id="id347"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">landingpad</span></code>’ instruction is used by <a class="reference external" href="ExceptionHandling.html#overview">LLVM’s exception handlingsystem</a> to specify that a basic blockis a landing pad — one where the exception lands, and corresponds to thecode found in the <code class="docutils literal notranslate"><span class="pre">catch</span></code> portion of a <code class="docutils literal notranslate"><span class="pre">try</span></code>/<code class="docutils literal notranslate"><span class="pre">catch</span></code> sequence. Itdefines values supplied by the <a class="reference internal" href="#personalityfn"><span class="std std-ref">personality function</span></a> uponre-entry to the function. The <code class="docutils literal notranslate"><span class="pre">resultval</span></code> has the type <code class="docutils literal notranslate"><span class="pre">resultty</span></code>.</p></section><section id="id349"><h5>Arguments:</h5><p>The optional<code class="docutils literal notranslate"><span class="pre">cleanup</span></code> flag indicates that the landing pad block is a cleanup.</p><p>A <code class="docutils literal notranslate"><span class="pre">clause</span></code> begins with the clause type — <code class="docutils literal notranslate"><span class="pre">catch</span></code> or <code class="docutils literal notranslate"><span class="pre">filter</span></code> — andcontains the global variable representing the “type” that may be caughtor filtered respectively. Unlike the <code class="docutils literal notranslate"><span class="pre">catch</span></code> clause, the <code class="docutils literal notranslate"><span class="pre">filter</span></code>clause takes an array constant as its argument. Use“<code class="docutils literal notranslate"><span class="pre">[0</span> <span class="pre">x</span> <span class="pre">ptr]</span> <span class="pre">undef</span></code>” for a filter which cannot throw. The‘<code class="docutils literal notranslate"><span class="pre">landingpad</span></code>’ instruction must contain <em>at least</em> one <code class="docutils literal notranslate"><span class="pre">clause</span></code> orthe <code class="docutils literal notranslate"><span class="pre">cleanup</span></code> flag.</p></section><section id="id350"><h5>Semantics:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">landingpad</span></code>’ instruction defines the values which are set by the<a class="reference internal" href="#personalityfn"><span class="std std-ref">personality function</span></a> upon re-entry to the function, andtherefore the “result type” of the <code class="docutils literal notranslate"><span class="pre">landingpad</span></code> instruction. As withcalling conventions, how the personality function results arerepresented in LLVM IR is target specific.</p><p>The clauses are applied in order from top to bottom. If two<code class="docutils literal notranslate"><span class="pre">landingpad</span></code> instructions are merged together through inlining, theclauses from the calling function are appended to the list of clauses.When the call stack is being unwound due to an exception being thrown,the exception is compared against each <code class="docutils literal notranslate"><span class="pre">clause</span></code> in turn. If it doesn’tmatch any of the clauses, and the <code class="docutils literal notranslate"><span class="pre">cleanup</span></code> flag is not set, thenunwinding continues further up the call stack.</p><p>The <code class="docutils literal notranslate"><span class="pre">landingpad</span></code> instruction has several restrictions:</p><ul class="simple"><li><p>A landing pad block is a basic block which is the unwind destinationof an ‘<code class="docutils literal notranslate"><span class="pre">invoke</span></code>’ instruction.</p></li><li><p>A landing pad block must have a ‘<code class="docutils literal notranslate"><span class="pre">landingpad</span></code>’ instruction as itsfirst non-PHI instruction.</p></li><li><p>There can be only one ‘<code class="docutils literal notranslate"><span class="pre">landingpad</span></code>’ instruction within the landingpad block.</p></li><li><p>A basic block that is not a landing pad block may not include a‘<code class="docutils literal notranslate"><span class="pre">landingpad</span></code>’ instruction.</p></li></ul></section><section id="id351"><h5>Example:</h5><div class="highlight-llvm notranslate"><div class="highlight"><pre><span></span><span class="c">;; A landing pad which can catch an integer.</span><span class="nv">%res</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">landingpad</span><span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="kt">ptr</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="p">}</span><span class="w"> </span><span class="k">catch</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="vg">@_ZTIi</span><span class="c">;; A landing pad that is a cleanup.</span><span class="nv">%res</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">landingpad</span><span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="kt">ptr</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="p">}</span><span class="w"> </span><span class="k">cleanup</span><span class="c">;; A landing pad which can catch an integer and can only throw a double.</span><span class="nv">%res</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="k">landingpad</span><span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="kt">ptr</span><span class="p">,</span><span class="w"> </span><span class="kt">i32</span><span class="w"> </span><span class="p">}</span><span class="w"> </span><span class="k">catch</span><span class="w"> </span><span class="kt">ptr</span><span class="w"> </span><span class="vg">@_ZTIi</span><span class="w"> </span><span class="k">filter</span><span class="w"> </span><span class="p">[</span><span class="m">1</span><span class="w"> </span><span class="k">x</span><span class="w"> </span><span class="kt">ptr</span><span class="p">]</span><span class="w"> </span><span class="p">[</span><span class="kt">ptr</span><span class="w"> </span><span class="vg">@_ZTId</span><span class="p">]</span></pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘landingpad’ instruction is used by LLVM’s exception handlingsystem to specify that a basic blockis a landing pad — one where the exception lands, and corresponds to thecode found in the catch portion of a try/catch sequence. Itdefines values supplied by the personality function uponre-entry to the function. The resultval has the type resultty.`,
|
||
};
|
||
case 'CATCHPAD':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#catchpad-instruction`,
|
||
html: `<html><head></head><body><span id="i-catchpad"></span><h4><a class="toc-backref" href="#id2228" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">catchpad</span></code>’ Instruction</a></h4><section id="id352"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">resultval</span><span class="o">></span> <span class="o">=</span> <span class="n">catchpad</span> <span class="n">within</span> <span class="o"><</span><span class="n">catchswitch</span><span class="o">></span> <span class="p">[</span><span class="o"><</span><span class="n">args</span><span class="o">>*</span><span class="p">]</span></pre></div></div></section><section id="id353"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">catchpad</span></code>’ instruction is used by <a class="reference external" href="ExceptionHandling.html#overview">LLVM’s exception handlingsystem</a> to specify that a basic blockbegins a catch handler — one where a personality routine attempts to transfercontrol to catch an exception.</p></section><section id="id355"><h5>Arguments:</h5><p>The <code class="docutils literal notranslate"><span class="pre">catchswitch</span></code> operand must always be a token produced by a<a class="reference internal" href="#i-catchswitch"><span class="std std-ref">catchswitch</span></a> instruction in a predecessor block. Thisensures that each <code class="docutils literal notranslate"><span class="pre">catchpad</span></code> has exactly one predecessor block, and it alwaysterminates in a <code class="docutils literal notranslate"><span class="pre">catchswitch</span></code>.</p><p>The <code class="docutils literal notranslate"><span class="pre">args</span></code> correspond to whatever information the personality routinerequires to determine if this is an appropriate handler for the exception. Controlwill transfer to the <code class="docutils literal notranslate"><span class="pre">catchpad</span></code> if this is the first appropriate handler forthe exception.</p><p>The <code class="docutils literal notranslate"><span class="pre">resultval</span></code> has the type <a class="reference internal" href="#t-token"><span class="std std-ref">token</span></a> and is used to match the<code class="docutils literal notranslate"><span class="pre">catchpad</span></code> to corresponding <a class="reference internal" href="#i-catchret"><span class="std std-ref">catchrets</span></a> and other nested EHpads.</p></section><section id="id356"><h5>Semantics:</h5><p>When the call stack is being unwound due to an exception being thrown, theexception is compared against the <code class="docutils literal notranslate"><span class="pre">args</span></code>. If it doesn’t match, control willnot reach the <code class="docutils literal notranslate"><span class="pre">catchpad</span></code> instruction. The representation of <code class="docutils literal notranslate"><span class="pre">args</span></code> isentirely target and personality function-specific.</p><p>Like the <a class="reference internal" href="#i-landingpad"><span class="std std-ref">landingpad</span></a> instruction, the <code class="docutils literal notranslate"><span class="pre">catchpad</span></code>instruction must be the first non-phi of its parent basic block.</p><p>The meaning of the tokens produced and consumed by <code class="docutils literal notranslate"><span class="pre">catchpad</span></code> and other “pad”instructions is described in the<a class="reference external" href="ExceptionHandling.html#wineh">Windows exception handling documentation</a>.</p><p>When a <code class="docutils literal notranslate"><span class="pre">catchpad</span></code> has been “entered” but not yet “exited” (asdescribed in the <a class="reference external" href="ExceptionHandling.html#wineh-constraints">EH documentation</a>),it is undefined behavior to execute a <a class="reference internal" href="#i-call"><span class="std std-ref">call</span></a> or <a class="reference internal" href="#i-invoke"><span class="std std-ref">invoke</span></a>that does not carry an appropriate <a class="reference internal" href="#ob-funclet"><span class="std std-ref">“funclet” bundle</span></a>.</p></section><section id="id358"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span>dispatch: %cs = catchswitch within none [label %handler0] unwind to caller ;; A catch block which can catch an integer.handler0: %tok = catchpad within %cs [ptr @_ZTIi]</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘catchpad’ instruction is used by LLVM’s exception handlingsystem to specify that a basic blockbegins a catch handler — one where a personality routine attempts to transfercontrol to catch an exception.`,
|
||
};
|
||
case 'CLEANUPPAD':
|
||
return {
|
||
url: `https://llvm.org/docs/LangRef.html#cleanuppad-instruction`,
|
||
html: `<html><head></head><body><span id="i-cleanuppad"></span><h4><a class="toc-backref" href="#id2229" role="doc-backlink">‘<code class="docutils literal notranslate"><span class="pre">cleanuppad</span></code>’ Instruction</a></h4><section id="id359"><h5>Syntax:</h5><div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o"><</span><span class="n">resultval</span><span class="o">></span> <span class="o">=</span> <span class="n">cleanuppad</span> <span class="n">within</span> <span class="o"><</span><span class="n">parent</span><span class="o">></span> <span class="p">[</span><span class="o"><</span><span class="n">args</span><span class="o">>*</span><span class="p">]</span></pre></div></div></section><section id="id360"><h5>Overview:</h5><p>The ‘<code class="docutils literal notranslate"><span class="pre">cleanuppad</span></code>’ instruction is used by <a class="reference external" href="ExceptionHandling.html#overview">LLVM’s exception handlingsystem</a> to specify that a basic blockis a cleanup block — one where a personality routine attempts totransfer control to run cleanup actions.The <code class="docutils literal notranslate"><span class="pre">args</span></code> correspond to whatever additionalinformation the <a class="reference internal" href="#personalityfn"><span class="std std-ref">personality function</span></a> requires toexecute the cleanup.The <code class="docutils literal notranslate"><span class="pre">resultval</span></code> has the type <a class="reference internal" href="#t-token"><span class="std std-ref">token</span></a> and is used tomatch the <code class="docutils literal notranslate"><span class="pre">cleanuppad</span></code> to corresponding <a class="reference internal" href="#i-cleanupret"><span class="std std-ref">cleanuprets</span></a>.The <code class="docutils literal notranslate"><span class="pre">parent</span></code> argument is the token of the funclet that contains the<code class="docutils literal notranslate"><span class="pre">cleanuppad</span></code> instruction. If the <code class="docutils literal notranslate"><span class="pre">cleanuppad</span></code> is not inside a funclet,this operand may be the token <code class="docutils literal notranslate"><span class="pre">none</span></code>.</p></section><section id="id362"><h5>Arguments:</h5><p>The instruction takes a list of arbitrary values which are interpretedby the <a class="reference internal" href="#personalityfn"><span class="std std-ref">personality function</span></a>.</p></section><section id="id363"><h5>Semantics:</h5><p>When the call stack is being unwound due to an exception being thrown,the <a class="reference internal" href="#personalityfn"><span class="std std-ref">personality function</span></a> transfers control to the<code class="docutils literal notranslate"><span class="pre">cleanuppad</span></code> with the aid of the personality-specific arguments.As with calling conventions, how the personality function results arerepresented in LLVM IR is target specific.</p><p>The <code class="docutils literal notranslate"><span class="pre">cleanuppad</span></code> instruction has several restrictions:</p><ul class="simple"><li><p>A cleanup block is a basic block which is the unwind destination ofan exceptional instruction.</p></li><li><p>A cleanup block must have a ‘<code class="docutils literal notranslate"><span class="pre">cleanuppad</span></code>’ instruction as itsfirst non-PHI instruction.</p></li><li><p>There can be only one ‘<code class="docutils literal notranslate"><span class="pre">cleanuppad</span></code>’ instruction within thecleanup block.</p></li><li><p>A basic block that is not a cleanup block may not include a‘<code class="docutils literal notranslate"><span class="pre">cleanuppad</span></code>’ instruction.</p></li></ul><p>When a <code class="docutils literal notranslate"><span class="pre">cleanuppad</span></code> has been “entered” but not yet “exited” (asdescribed in the <a class="reference external" href="ExceptionHandling.html#wineh-constraints">EH documentation</a>),it is undefined behavior to execute a <a class="reference internal" href="#i-call"><span class="std std-ref">call</span></a> or <a class="reference internal" href="#i-invoke"><span class="std std-ref">invoke</span></a>that does not carry an appropriate <a class="reference internal" href="#ob-funclet"><span class="std std-ref">“funclet” bundle</span></a>.</p></section><section id="id365"><h5>Example:</h5><div class="highlight-text notranslate"><div class="highlight"><pre><span></span>%tok = cleanuppad within %cs []</pre></div></div></section></body></html>`,
|
||
tooltip: `The ‘cleanuppad’ instruction is used by LLVM’s exception handlingsystem to specify that a basic blockis a cleanup block — one where a personality routine attempts totransfer control to run cleanup actions.The args correspond to whatever additionalinformation the personality function requires toexecute the cleanup.The resultval has the type token and is used tomatch the cleanuppad to corresponding cleanuprets.The parent argument is the token of the funclet that contains thecleanuppad instruction. If the cleanuppad is not inside a funclet,this operand may be the token none.`,
|
||
};
|
||
}
|
||
}
|