Source file src/cmd/compile/doc.go
1 // Copyright 2009 The Go Authors. All rights reserved. 2 // Use of this source code is governed by a BSD-style 3 // license that can be found in the LICENSE file. 4 5 /* 6 Compile, typically invoked as ``go tool compile,'' compiles a single Go package 7 comprising the files named on the command line. It then writes a single 8 object file named for the basename of the first source file with a .o suffix. 9 The object file can then be combined with other objects into a package archive 10 or passed directly to the linker (``go tool link''). If invoked with -pack, the compiler 11 writes an archive directly, bypassing the intermediate object file. 12 13 The generated files contain type information about the symbols exported by 14 the package and about types used by symbols imported by the package from 15 other packages. It is therefore not necessary when compiling client C of 16 package P to read the files of P's dependencies, only the compiled output of P. 17 18 # Command Line 19 20 Usage: 21 22 go tool compile [flags] file... 23 24 The specified files must be Go source files and all part of the same package. 25 The same compiler is used for all target operating systems and architectures. 26 The GOOS and GOARCH environment variables set the desired target. 27 28 Flags: 29 30 -D path 31 Set relative path for local imports. 32 -I dir1 -I dir2 33 Search for imported packages in dir1, dir2, etc, 34 after consulting $GOROOT/pkg/$GOOS_$GOARCH. 35 -L 36 Show complete file path in error messages. 37 -N 38 Disable optimizations. 39 -S 40 Print assembly listing to standard output (code only). 41 -S -S 42 Print assembly listing to standard output (code and data). 43 -V 44 Print compiler version and exit. 45 -asmhdr file 46 Write assembly header to file. 47 -asan 48 Insert calls to C/C++ address sanitizer. 49 -buildid id 50 Record id as the build id in the export metadata. 51 -blockprofile file 52 Write block profile for the compilation to file. 53 -c int 54 Concurrency during compilation. Set 1 for no concurrency (default is 1). 55 -complete 56 Assume package has no non-Go components. 57 -cpuprofile file 58 Write a CPU profile for the compilation to file. 59 -dynlink 60 Allow references to Go symbols in shared libraries (experimental). 61 -e 62 Remove the limit on the number of errors reported (default limit is 10). 63 -goversion string 64 Specify required go tool version of the runtime. 65 Exits when the runtime go version does not match goversion. 66 -h 67 Halt with a stack trace at the first error detected. 68 -importcfg file 69 Read import configuration from file. 70 In the file, set importmap, packagefile to specify import resolution. 71 -installsuffix suffix 72 Look for packages in $GOROOT/pkg/$GOOS_$GOARCH_suffix 73 instead of $GOROOT/pkg/$GOOS_$GOARCH. 74 -l 75 Disable inlining. 76 -lang version 77 Set language version to compile, as in -lang=go1.12. 78 Default is current version. 79 -linkobj file 80 Write linker-specific object to file and compiler-specific 81 object to usual output file (as specified by -o). 82 Without this flag, the -o output is a combination of both 83 linker and compiler input. 84 -m 85 Print optimization decisions. Higher values or repetition 86 produce more detail. 87 -memprofile file 88 Write memory profile for the compilation to file. 89 -memprofilerate rate 90 Set runtime.MemProfileRate for the compilation to rate. 91 -msan 92 Insert calls to C/C++ memory sanitizer. 93 -mutexprofile file 94 Write mutex profile for the compilation to file. 95 -nolocalimports 96 Disallow local (relative) imports. 97 -o file 98 Write object to file (default file.o or, with -pack, file.a). 99 -p path 100 Set expected package import path for the code being compiled, 101 and diagnose imports that would cause a circular dependency. 102 -pack 103 Write a package (archive) file rather than an object file 104 -race 105 Compile with race detector enabled. 106 -s 107 Warn about composite literals that can be simplified. 108 -shared 109 Generate code that can be linked into a shared library. 110 -spectre list 111 Enable spectre mitigations in list (all, index, ret). 112 -traceprofile file 113 Write an execution trace to file. 114 -trimpath prefix 115 Remove prefix from recorded source file paths. 116 117 Flags related to debugging information: 118 119 -dwarf 120 Generate DWARF symbols. 121 -dwarflocationlists 122 Add location lists to DWARF in optimized mode. 123 -gendwarfinl int 124 Generate DWARF inline info records (default 2). 125 126 Flags to debug the compiler itself: 127 128 -E 129 Debug symbol export. 130 -K 131 Debug missing line numbers. 132 -d list 133 Print debug information about items in list. Try -d help for further information. 134 -live 135 Debug liveness analysis. 136 -v 137 Increase debug verbosity. 138 -% 139 Debug non-static initializers. 140 -W 141 Debug parse tree after type checking. 142 -f 143 Debug stack frames. 144 -i 145 Debug line number stack. 146 -j 147 Debug runtime-initialized variables. 148 -r 149 Debug generated wrappers. 150 -w 151 Debug type checking. 152 153 # Compiler Directives 154 155 The compiler accepts directives in the form of comments. 156 Each directive must be placed its own line, with only leading spaces and tabs 157 allowed before the comment, and there must be no space between the comment 158 opening and the name of the directive, to distinguish it from a regular comment. 159 Tools unaware of the directive convention or of a particular 160 directive can skip over a directive like any other comment. 161 162 Other than the line directive, which is a historical special case; 163 all other compiler directives are of the form 164 //go:name, indicating that they are defined by the Go toolchain. 165 */ 166 // # Line Directives 167 // 168 // Line directives come in several forms: 169 // 170 // //line :line 171 // //line :line:col 172 // //line filename:line 173 // //line filename:line:col 174 // /*line :line*/ 175 // /*line :line:col*/ 176 // /*line filename:line*/ 177 // /*line filename:line:col*/ 178 // 179 // In order to be recognized as a line directive, the comment must start with 180 // //line or /*line followed by a space, and must contain at least one colon. 181 // The //line form must start at the beginning of a line. 182 // A line directive specifies the source position for the character immediately following 183 // the comment as having come from the specified file, line and column: 184 // For a //line comment, this is the first character of the next line, and 185 // for a /*line comment this is the character position immediately following the closing */. 186 // If no filename is given, the recorded filename is empty if there is also no column number; 187 // otherwise it is the most recently recorded filename (actual filename or filename specified 188 // by previous line directive). 189 // If a line directive doesn't specify a column number, the column is "unknown" until 190 // the next directive and the compiler does not report column numbers for that range. 191 // The line directive text is interpreted from the back: First the trailing :ddd is peeled 192 // off from the directive text if ddd is a valid number > 0. Then the second :ddd 193 // is peeled off the same way if it is valid. Anything before that is considered the filename 194 // (possibly including blanks and colons). Invalid line or column values are reported as errors. 195 // 196 // Examples: 197 // 198 // //line foo.go:10 the filename is foo.go, and the line number is 10 for the next line 199 // //line C:foo.go:10 colons are permitted in filenames, here the filename is C:foo.go, and the line is 10 200 // //line a:100 :10 blanks are permitted in filenames, here the filename is " a:100 " (excluding quotes) 201 // /*line :10:20*/x the position of x is in the current file with line number 10 and column number 20 202 // /*line foo: 10 */ this comment is recognized as invalid line directive (extra blanks around line number) 203 // 204 // Line directives typically appear in machine-generated code, so that compilers and debuggers 205 // will report positions in the original input to the generator. 206 /* 207 # Function Directives 208 209 A function directive applies to the Go function that immediately follows it. 210 211 //go:noescape 212 213 The //go:noescape directive must be followed by a function declaration without 214 a body (meaning that the function has an implementation not written in Go). 215 It specifies that the function does not allow any of the pointers passed as 216 arguments to escape into the heap or into the values returned from the function. 217 This information can be used during the compiler's escape analysis of Go code 218 calling the function. 219 220 //go:uintptrescapes 221 222 The //go:uintptrescapes directive must be followed by a function declaration. 223 It specifies that the function's uintptr arguments may be pointer values that 224 have been converted to uintptr and must be on the heap and kept alive for the 225 duration of the call, even though from the types alone it would appear that the 226 object is no longer needed during the call. The conversion from pointer to 227 uintptr must appear in the argument list of any call to this function. This 228 directive is necessary for some low-level system call implementations and 229 should be avoided otherwise. 230 231 //go:noinline 232 233 The //go:noinline directive must be followed by a function declaration. 234 It specifies that calls to the function should not be inlined, overriding 235 the compiler's usual optimization rules. This is typically only needed 236 for special runtime functions or when debugging the compiler. 237 238 //go:norace 239 240 The //go:norace directive must be followed by a function declaration. 241 It specifies that the function's memory accesses must be ignored by the 242 race detector. This is most commonly used in low-level code invoked 243 at times when it is unsafe to call into the race detector runtime. 244 245 //go:nosplit 246 247 The //go:nosplit directive must be followed by a function declaration. 248 It specifies that the function must omit its usual stack overflow check. 249 This is most commonly used by low-level runtime code invoked 250 at times when it is unsafe for the calling goroutine to be preempted. 251 252 # Linkname Directive 253 254 //go:linkname localname [importpath.name] 255 256 The //go:linkname directive conventionally precedes the var or func 257 declaration named by ``localname``, though its position does not 258 change its effect. 259 This directive determines the object-file symbol used for a Go var or 260 func declaration, allowing two Go symbols to alias the same 261 object-file symbol, thereby enabling one package to access a symbol in 262 another package even when this would violate the usual encapsulation 263 of unexported declarations, or even type safety. 264 For that reason, it is only enabled in files that have imported "unsafe". 265 266 It may be used in two scenarios. Let's assume that package upper 267 imports package lower, perhaps indirectly. In the first scenario, 268 package lower defines a symbol whose object file name belongs to 269 package upper. Both packages contain a linkname directive: package 270 lower uses the two-argument form and package upper uses the 271 one-argument form. In the example below, lower.f is an alias for the 272 function upper.g: 273 274 package upper 275 import _ "unsafe" 276 //go:linkname g 277 func g() 278 279 package lower 280 import _ "unsafe" 281 //go:linkname f upper.g 282 func f() { ... } 283 284 The linkname directive in package upper suppresses the usual error for 285 a function that lacks a body. (That check may alternatively be 286 suppressed by including a .s file, even an empty one, in the package.) 287 288 In the second scenario, package upper unilaterally creates an alias 289 for a symbol in package lower. In the example below, upper.g is an alias 290 for the function lower.f. 291 292 package upper 293 import _ "unsafe" 294 //go:linkname g lower.f 295 func g() 296 297 package lower 298 func f() { ... } 299 300 The declaration of lower.f may also have a linkname directive with a 301 single argument, f. This is optional, but helps alert the reader that 302 the function is accessed from outside the package. 303 304 # WebAssembly Directives 305 306 //go:wasmimport importmodule importname 307 308 The //go:wasmimport directive is wasm-only and must be followed by a 309 function declaration with no body. 310 It specifies that the function is provided by a wasm module identified 311 by ``importmodule'' and ``importname''. For example, 312 313 //go:wasmimport a_module f 314 func g() 315 316 causes g to refer to the WebAssembly function f from module a_module. 317 318 //go:wasmexport exportname 319 320 The //go:wasmexport directive is wasm-only and must be followed by a 321 function definition. 322 It specifies that the function is exported to the wasm host as ``exportname''. 323 For example, 324 325 //go:wasmexport h 326 func hWasm() { ... } 327 328 make Go function hWasm available outside this WebAssembly module as h. 329 330 For both go:wasmimport and go:wasmexport, 331 the types of parameters and return values to the Go function are translated to 332 Wasm according to the following table: 333 334 Go types Wasm types 335 bool i32 336 int32, uint32 i32 337 int64, uint64 i64 338 float32 f32 339 float64 f64 340 unsafe.Pointer i32 341 pointer i32 (more restrictions below) 342 string (i32, i32) (only permitted as a parameters, not a result) 343 344 Any other parameter types are disallowed by the compiler. 345 346 For a pointer type, its element type must be a bool, int8, uint8, int16, uint16, 347 int32, uint32, int64, uint64, float32, float64, an array whose element type is 348 a permitted pointer element type, or a struct, which, if non-empty, embeds 349 [structs.HostLayout], and contains only fields whose types are permitted pointer 350 element types. 351 */ 352 package main 353