| 1 | .\" Automatically generated by Pod::Man v1.37, Pod::Parser v1.32 |
| 2 | .\" |
| 3 | .\" Standard preamble: |
| 4 | .\" ======================================================================== |
| 5 | .de Sh \" Subsection heading |
| 6 | .br |
| 7 | .if t .Sp |
| 8 | .ne 5 |
| 9 | .PP |
| 10 | \fB\\$1\fR |
| 11 | .PP |
| 12 | .. |
| 13 | .de Sp \" Vertical space (when we can't use .PP) |
| 14 | .if t .sp .5v |
| 15 | .if n .sp |
| 16 | .. |
| 17 | .de Vb \" Begin verbatim text |
| 18 | .ft CW |
| 19 | .nf |
| 20 | .ne \\$1 |
| 21 | .. |
| 22 | .de Ve \" End verbatim text |
| 23 | .ft R |
| 24 | .fi |
| 25 | .. |
| 26 | .\" Set up some character translations and predefined strings. \*(-- will |
| 27 | .\" give an unbreakable dash, \*(PI will give pi, \*(L" will give a left |
| 28 | .\" double quote, and \*(R" will give a right double quote. | will give a |
| 29 | .\" real vertical bar. \*(C+ will give a nicer C++. Capital omega is used to |
| 30 | .\" do unbreakable dashes and therefore won't be available. \*(C` and \*(C' |
| 31 | .\" expand to `' in nroff, nothing in troff, for use with C<>. |
| 32 | .tr \(*W-|\(bv\*(Tr |
| 33 | .ds C+ C\v'-.1v'\h'-1p'\s-2+\h'-1p'+\s0\v'.1v'\h'-1p' |
| 34 | .ie n \{\ |
| 35 | . ds -- \(*W- |
| 36 | . ds PI pi |
| 37 | . if (\n(.H=4u)&(1m=24u) .ds -- \(*W\h'-12u'\(*W\h'-12u'-\" diablo 10 pitch |
| 38 | . if (\n(.H=4u)&(1m=20u) .ds -- \(*W\h'-12u'\(*W\h'-8u'-\" diablo 12 pitch |
| 39 | . ds L" "" |
| 40 | . ds R" "" |
| 41 | . ds C` "" |
| 42 | . ds C' "" |
| 43 | 'br\} |
| 44 | .el\{\ |
| 45 | . ds -- \|\(em\| |
| 46 | . ds PI \(*p |
| 47 | . ds L" `` |
| 48 | . ds R" '' |
| 49 | 'br\} |
| 50 | .\" |
| 51 | .\" If the F register is turned on, we'll generate index entries on stderr for |
| 52 | .\" titles (.TH), headers (.SH), subsections (.Sh), items (.Ip), and index |
| 53 | .\" entries marked with X<> in POD. Of course, you'll have to process the |
| 54 | .\" output yourself in some meaningful fashion. |
| 55 | .if \nF \{\ |
| 56 | . de IX |
| 57 | . tm Index:\\$1\t\\n%\t"\\$2" |
| 58 | .. |
| 59 | . nr % 0 |
| 60 | . rr F |
| 61 | .\} |
| 62 | .\" |
| 63 | .\" For nroff, turn off justification. Always turn off hyphenation; it makes |
| 64 | .\" way too many mistakes in technical documents. |
| 65 | .hy 0 |
| 66 | .if n .na |
| 67 | .\" |
| 68 | .\" Accent mark definitions (@(#)ms.acc 1.5 88/02/08 SMI; from UCB 4.2). |
| 69 | .\" Fear. Run. Save yourself. No user-serviceable parts. |
| 70 | . \" fudge factors for nroff and troff |
| 71 | .if n \{\ |
| 72 | . ds #H 0 |
| 73 | . ds #V .8m |
| 74 | . ds #F .3m |
| 75 | . ds #[ \f1 |
| 76 | . ds #] \fP |
| 77 | .\} |
| 78 | .if t \{\ |
| 79 | . ds #H ((1u-(\\\\n(.fu%2u))*.13m) |
| 80 | . ds #V .6m |
| 81 | . ds #F 0 |
| 82 | . ds #[ \& |
| 83 | . ds #] \& |
| 84 | .\} |
| 85 | . \" simple accents for nroff and troff |
| 86 | .if n \{\ |
| 87 | . ds ' \& |
| 88 | . ds ` \& |
| 89 | . ds ^ \& |
| 90 | . ds , \& |
| 91 | . ds ~ ~ |
| 92 | . ds / |
| 93 | .\} |
| 94 | .if t \{\ |
| 95 | . ds ' \\k:\h'-(\\n(.wu*8/10-\*(#H)'\'\h"|\\n:u" |
| 96 | . ds ` \\k:\h'-(\\n(.wu*8/10-\*(#H)'\`\h'|\\n:u' |
| 97 | . ds ^ \\k:\h'-(\\n(.wu*10/11-\*(#H)'^\h'|\\n:u' |
| 98 | . ds , \\k:\h'-(\\n(.wu*8/10)',\h'|\\n:u' |
| 99 | . ds ~ \\k:\h'-(\\n(.wu-\*(#H-.1m)'~\h'|\\n:u' |
| 100 | . ds / \\k:\h'-(\\n(.wu*8/10-\*(#H)'\z\(sl\h'|\\n:u' |
| 101 | .\} |
| 102 | . \" troff and (daisy-wheel) nroff accents |
| 103 | .ds : \\k:\h'-(\\n(.wu*8/10-\*(#H+.1m+\*(#F)'\v'-\*(#V'\z.\h'.2m+\*(#F'.\h'|\\n:u'\v'\*(#V' |
| 104 | .ds 8 \h'\*(#H'\(*b\h'-\*(#H' |
| 105 | .ds o \\k:\h'-(\\n(.wu+\w'\(de'u-\*(#H)/2u'\v'-.3n'\*(#[\z\(de\v'.3n'\h'|\\n:u'\*(#] |
| 106 | .ds d- \h'\*(#H'\(pd\h'-\w'~'u'\v'-.25m'\f2\(hy\fP\v'.25m'\h'-\*(#H' |
| 107 | .ds D- D\\k:\h'-\w'D'u'\v'-.11m'\z\(hy\v'.11m'\h'|\\n:u' |
| 108 | .ds th \*(#[\v'.3m'\s+1I\s-1\v'-.3m'\h'-(\w'I'u*2/3)'\s-1o\s+1\*(#] |
| 109 | .ds Th \*(#[\s+2I\s-2\h'-\w'I'u*3/5'\v'-.3m'o\v'.3m'\*(#] |
| 110 | .ds ae a\h'-(\w'a'u*4/10)'e |
| 111 | .ds Ae A\h'-(\w'A'u*4/10)'E |
| 112 | . \" corrections for vroff |
| 113 | .if v .ds ~ \\k:\h'-(\\n(.wu*9/10-\*(#H)'\s-2\u~\d\s+2\h'|\\n:u' |
| 114 | .if v .ds ^ \\k:\h'-(\\n(.wu*10/11-\*(#H)'\v'-.4m'^\v'.4m'\h'|\\n:u' |
| 115 | . \" for low resolution devices (crt and lpr) |
| 116 | .if \n(.H>23 .if \n(.V>19 \ |
| 117 | \{\ |
| 118 | . ds : e |
| 119 | . ds 8 ss |
| 120 | . ds o a |
| 121 | . ds d- d\h'-1'\(ga |
| 122 | . ds D- D\h'-1'\(hy |
| 123 | . ds th \o'bp' |
| 124 | . ds Th \o'LP' |
| 125 | . ds ae ae |
| 126 | . ds Ae AE |
| 127 | .\} |
| 128 | .rm #[ #] #H #V #F C |
| 129 | .\" ======================================================================== |
| 130 | .\" |
| 131 | .IX Title "PERLFORK 1" |
| 132 | .TH PERLFORK 1 "2006-01-07" "perl v5.8.8" "Perl Programmers Reference Guide" |
| 133 | .SH "NAME" |
| 134 | perlfork \- Perl's fork() emulation |
| 135 | .SH "SYNOPSIS" |
| 136 | .IX Header "SYNOPSIS" |
| 137 | .Vb 4 |
| 138 | \& NOTE: As of the 5.8.0 release, fork() emulation has considerably |
| 139 | \& matured. However, there are still a few known bugs and differences |
| 140 | \& from real fork() that might affect you. See the "BUGS" and |
| 141 | \& "CAVEATS AND LIMITATIONS" sections below. |
| 142 | .Ve |
| 143 | .PP |
| 144 | Perl provides a \fIfork()\fR keyword that corresponds to the Unix system call |
| 145 | of the same name. On most Unix-like platforms where the \fIfork()\fR system |
| 146 | call is available, Perl's \fIfork()\fR simply calls it. |
| 147 | .PP |
| 148 | On some platforms such as Windows where the \fIfork()\fR system call is not |
| 149 | available, Perl can be built to emulate \fIfork()\fR at the interpreter level. |
| 150 | While the emulation is designed to be as compatible as possible with the |
| 151 | real \fIfork()\fR at the level of the Perl program, there are certain |
| 152 | important differences that stem from the fact that all the pseudo child |
| 153 | \&\*(L"processes\*(R" created this way live in the same real process as far as the |
| 154 | operating system is concerned. |
| 155 | .PP |
| 156 | This document provides a general overview of the capabilities and |
| 157 | limitations of the \fIfork()\fR emulation. Note that the issues discussed here |
| 158 | are not applicable to platforms where a real \fIfork()\fR is available and Perl |
| 159 | has been configured to use it. |
| 160 | .SH "DESCRIPTION" |
| 161 | .IX Header "DESCRIPTION" |
| 162 | The \fIfork()\fR emulation is implemented at the level of the Perl interpreter. |
| 163 | What this means in general is that running \fIfork()\fR will actually clone the |
| 164 | running interpreter and all its state, and run the cloned interpreter in |
| 165 | a separate thread, beginning execution in the new thread just after the |
| 166 | point where the \fIfork()\fR was called in the parent. We will refer to the |
| 167 | thread that implements this child \*(L"process\*(R" as the pseudo\-process. |
| 168 | .PP |
| 169 | To the Perl program that called \fIfork()\fR, all this is designed to be |
| 170 | transparent. The parent returns from the \fIfork()\fR with a pseudo-process |
| 171 | \&\s-1ID\s0 that can be subsequently used in any process manipulation functions; |
| 172 | the child returns from the \fIfork()\fR with a value of \f(CW0\fR to signify that |
| 173 | it is the child pseudo\-process. |
| 174 | .Sh "Behavior of other Perl features in forked pseudo-processes" |
| 175 | .IX Subsection "Behavior of other Perl features in forked pseudo-processes" |
| 176 | Most Perl features behave in a natural way within pseudo\-processes. |
| 177 | .ie n .IP "$$ or $PROCESS_ID" 8 |
| 178 | .el .IP "$$ or \f(CW$PROCESS_ID\fR" 8 |
| 179 | .IX Item "$$ or $PROCESS_ID" |
| 180 | This special variable is correctly set to the pseudo-process \s-1ID\s0. |
| 181 | It can be used to identify pseudo-processes within a particular |
| 182 | session. Note that this value is subject to recycling if any |
| 183 | pseudo-processes are launched after others have been \fIwait()\fR\-ed on. |
| 184 | .IP "%ENV" 8 |
| 185 | .IX Item "%ENV" |
| 186 | Each pseudo-process maintains its own virtual environment. Modifications |
| 187 | to \f(CW%ENV\fR affect the virtual environment, and are only visible within that |
| 188 | pseudo\-process, and in any processes (or pseudo\-processes) launched from |
| 189 | it. |
| 190 | .IP "\fIchdir()\fR and all other builtins that accept filenames" 8 |
| 191 | .IX Item "chdir() and all other builtins that accept filenames" |
| 192 | Each pseudo-process maintains its own virtual idea of the current directory. |
| 193 | Modifications to the current directory using \fIchdir()\fR are only visible within |
| 194 | that pseudo\-process, and in any processes (or pseudo\-processes) launched from |
| 195 | it. All file and directory accesses from the pseudo-process will correctly |
| 196 | map the virtual working directory to the real working directory appropriately. |
| 197 | .IP "\fIwait()\fR and \fIwaitpid()\fR" 8 |
| 198 | .IX Item "wait() and waitpid()" |
| 199 | \&\fIwait()\fR and \fIwaitpid()\fR can be passed a pseudo-process \s-1ID\s0 returned by \fIfork()\fR. |
| 200 | These calls will properly wait for the termination of the pseudo-process |
| 201 | and return its status. |
| 202 | .IP "\fIkill()\fR" 8 |
| 203 | .IX Item "kill()" |
| 204 | \&\fIkill()\fR can be used to terminate a pseudo-process by passing it the \s-1ID\s0 returned |
| 205 | by \fIfork()\fR. This should not be used except under dire circumstances, because |
| 206 | the operating system may not guarantee integrity of the process resources |
| 207 | when a running thread is terminated. Note that using \fIkill()\fR on a |
| 208 | pseudo\-\fIprocess()\fR may typically cause memory leaks, because the thread that |
| 209 | implements the pseudo-process does not get a chance to clean up its resources. |
| 210 | .IP "\fIexec()\fR" 8 |
| 211 | .IX Item "exec()" |
| 212 | Calling \fIexec()\fR within a pseudo-process actually spawns the requested |
| 213 | executable in a separate process and waits for it to complete before |
| 214 | exiting with the same exit status as that process. This means that the |
| 215 | process \s-1ID\s0 reported within the running executable will be different from |
| 216 | what the earlier Perl \fIfork()\fR might have returned. Similarly, any process |
| 217 | manipulation functions applied to the \s-1ID\s0 returned by \fIfork()\fR will affect the |
| 218 | waiting pseudo-process that called \fIexec()\fR, not the real process it is |
| 219 | waiting for after the \fIexec()\fR. |
| 220 | .IP "\fIexit()\fR" 8 |
| 221 | .IX Item "exit()" |
| 222 | \&\fIexit()\fR always exits just the executing pseudo\-process, after automatically |
| 223 | \&\fIwait()\fR\-ing for any outstanding child pseudo\-processes. Note that this means |
| 224 | that the process as a whole will not exit unless all running pseudo-processes |
| 225 | have exited. |
| 226 | .IP "Open handles to files, directories and network sockets" 8 |
| 227 | .IX Item "Open handles to files, directories and network sockets" |
| 228 | All open handles are \fIdup()\fR\-ed in pseudo\-processes, so that closing |
| 229 | any handles in one process does not affect the others. See below for |
| 230 | some limitations. |
| 231 | .Sh "Resource limits" |
| 232 | .IX Subsection "Resource limits" |
| 233 | In the eyes of the operating system, pseudo-processes created via the \fIfork()\fR |
| 234 | emulation are simply threads in the same process. This means that any |
| 235 | process-level limits imposed by the operating system apply to all |
| 236 | pseudo-processes taken together. This includes any limits imposed by the |
| 237 | operating system on the number of open file, directory and socket handles, |
| 238 | limits on disk space usage, limits on memory size, limits on \s-1CPU\s0 utilization |
| 239 | etc. |
| 240 | .Sh "Killing the parent process" |
| 241 | .IX Subsection "Killing the parent process" |
| 242 | If the parent process is killed (either using Perl's \fIkill()\fR builtin, or |
| 243 | using some external means) all the pseudo-processes are killed as well, |
| 244 | and the whole process exits. |
| 245 | .Sh "Lifetime of the parent process and pseudo-processes" |
| 246 | .IX Subsection "Lifetime of the parent process and pseudo-processes" |
| 247 | During the normal course of events, the parent process and every |
| 248 | pseudo-process started by it will wait for their respective pseudo-children |
| 249 | to complete before they exit. This means that the parent and every |
| 250 | pseudo-child created by it that is also a pseudo-parent will only exit |
| 251 | after their pseudo-children have exited. |
| 252 | .PP |
| 253 | A way to mark a pseudo-processes as running detached from their parent (so |
| 254 | that the parent would not have to \fIwait()\fR for them if it doesn't want to) |
| 255 | will be provided in future. |
| 256 | .Sh "\s-1CAVEATS\s0 \s-1AND\s0 \s-1LIMITATIONS\s0" |
| 257 | .IX Subsection "CAVEATS AND LIMITATIONS" |
| 258 | .IP "\s-1BEGIN\s0 blocks" 8 |
| 259 | .IX Item "BEGIN blocks" |
| 260 | The \fIfork()\fR emulation will not work entirely correctly when called from |
| 261 | within a \s-1BEGIN\s0 block. The forked copy will run the contents of the |
| 262 | \&\s-1BEGIN\s0 block, but will not continue parsing the source stream after the |
| 263 | \&\s-1BEGIN\s0 block. For example, consider the following code: |
| 264 | .Sp |
| 265 | .Vb 5 |
| 266 | \& BEGIN { |
| 267 | \& fork and exit; # fork child and exit the parent |
| 268 | \& print "inner\en"; |
| 269 | \& } |
| 270 | \& print "outer\en"; |
| 271 | .Ve |
| 272 | .Sp |
| 273 | This will print: |
| 274 | .Sp |
| 275 | .Vb 1 |
| 276 | \& inner |
| 277 | .Ve |
| 278 | .Sp |
| 279 | rather than the expected: |
| 280 | .Sp |
| 281 | .Vb 2 |
| 282 | \& inner |
| 283 | \& outer |
| 284 | .Ve |
| 285 | .Sp |
| 286 | This limitation arises from fundamental technical difficulties in |
| 287 | cloning and restarting the stacks used by the Perl parser in the |
| 288 | middle of a parse. |
| 289 | .IP "Open filehandles" 8 |
| 290 | .IX Item "Open filehandles" |
| 291 | Any filehandles open at the time of the \fIfork()\fR will be \fIdup()\fR\-ed. Thus, |
| 292 | the files can be closed independently in the parent and child, but beware |
| 293 | that the \fIdup()\fR\-ed handles will still share the same seek pointer. Changing |
| 294 | the seek position in the parent will change it in the child and vice\-versa. |
| 295 | One can avoid this by opening files that need distinct seek pointers |
| 296 | separately in the child. |
| 297 | .IP "Forking pipe \fIopen()\fR not yet implemented" 8 |
| 298 | .IX Item "Forking pipe open() not yet implemented" |
| 299 | The \f(CW\*(C`open(FOO, "|\-")\*(C'\fR and \f(CW\*(C`open(BAR, "\-|")\*(C'\fR constructs are not yet |
| 300 | implemented. This limitation can be easily worked around in new code |
| 301 | by creating a pipe explicitly. The following example shows how to |
| 302 | write to a forked child: |
| 303 | .Sp |
| 304 | .Vb 15 |
| 305 | \& # simulate open(FOO, "|-") |
| 306 | \& sub pipe_to_fork ($) { |
| 307 | \& my $parent = shift; |
| 308 | \& pipe my $child, $parent or die; |
| 309 | \& my $pid = fork(); |
| 310 | \& die "fork() failed: $!" unless defined $pid; |
| 311 | \& if ($pid) { |
| 312 | \& close $child; |
| 313 | \& } |
| 314 | \& else { |
| 315 | \& close $parent; |
| 316 | \& open(STDIN, "<&=" . fileno($child)) or die; |
| 317 | \& } |
| 318 | \& $pid; |
| 319 | \& } |
| 320 | .Ve |
| 321 | .Sp |
| 322 | .Vb 10 |
| 323 | \& if (pipe_to_fork('FOO')) { |
| 324 | \& # parent |
| 325 | \& print FOO "pipe_to_fork\en"; |
| 326 | \& close FOO; |
| 327 | \& } |
| 328 | \& else { |
| 329 | \& # child |
| 330 | \& while (<STDIN>) { print; } |
| 331 | \& exit(0); |
| 332 | \& } |
| 333 | .Ve |
| 334 | .Sp |
| 335 | And this one reads from the child: |
| 336 | .Sp |
| 337 | .Vb 15 |
| 338 | \& # simulate open(FOO, "-|") |
| 339 | \& sub pipe_from_fork ($) { |
| 340 | \& my $parent = shift; |
| 341 | \& pipe $parent, my $child or die; |
| 342 | \& my $pid = fork(); |
| 343 | \& die "fork() failed: $!" unless defined $pid; |
| 344 | \& if ($pid) { |
| 345 | \& close $child; |
| 346 | \& } |
| 347 | \& else { |
| 348 | \& close $parent; |
| 349 | \& open(STDOUT, ">&=" . fileno($child)) or die; |
| 350 | \& } |
| 351 | \& $pid; |
| 352 | \& } |
| 353 | .Ve |
| 354 | .Sp |
| 355 | .Vb 10 |
| 356 | \& if (pipe_from_fork('BAR')) { |
| 357 | \& # parent |
| 358 | \& while (<BAR>) { print; } |
| 359 | \& close BAR; |
| 360 | \& } |
| 361 | \& else { |
| 362 | \& # child |
| 363 | \& print "pipe_from_fork\en"; |
| 364 | \& exit(0); |
| 365 | \& } |
| 366 | .Ve |
| 367 | .Sp |
| 368 | Forking pipe \fIopen()\fR constructs will be supported in future. |
| 369 | .IP "Global state maintained by XSUBs" 8 |
| 370 | .IX Item "Global state maintained by XSUBs" |
| 371 | External subroutines (XSUBs) that maintain their own global state may |
| 372 | not work correctly. Such XSUBs will either need to maintain locks to |
| 373 | protect simultaneous access to global data from different pseudo\-processes, |
| 374 | or maintain all their state on the Perl symbol table, which is copied |
| 375 | naturally when \fIfork()\fR is called. A callback mechanism that provides |
| 376 | extensions an opportunity to clone their state will be provided in the |
| 377 | near future. |
| 378 | .IP "Interpreter embedded in larger application" 8 |
| 379 | .IX Item "Interpreter embedded in larger application" |
| 380 | The \fIfork()\fR emulation may not behave as expected when it is executed in an |
| 381 | application which embeds a Perl interpreter and calls Perl APIs that can |
| 382 | evaluate bits of Perl code. This stems from the fact that the emulation |
| 383 | only has knowledge about the Perl interpreter's own data structures and |
| 384 | knows nothing about the containing application's state. For example, any |
| 385 | state carried on the application's own call stack is out of reach. |
| 386 | .IP "Thread-safety of extensions" 8 |
| 387 | .IX Item "Thread-safety of extensions" |
| 388 | Since the \fIfork()\fR emulation runs code in multiple threads, extensions |
| 389 | calling into non-thread-safe libraries may not work reliably when |
| 390 | calling \fIfork()\fR. As Perl's threading support gradually becomes more |
| 391 | widely adopted even on platforms with a native \fIfork()\fR, such extensions |
| 392 | are expected to be fixed for thread\-safety. |
| 393 | .SH "BUGS" |
| 394 | .IX Header "BUGS" |
| 395 | .IP "\(bu" 8 |
| 396 | Having pseudo-process IDs be negative integers breaks down for the integer |
| 397 | \&\f(CW\*(C`\-1\*(C'\fR because the \fIwait()\fR and \fIwaitpid()\fR functions treat this number as |
| 398 | being special. The tacit assumption in the current implementation is that |
| 399 | the system never allocates a thread \s-1ID\s0 of \f(CW1\fR for user threads. A better |
| 400 | representation for pseudo-process IDs will be implemented in future. |
| 401 | .IP "\(bu" 8 |
| 402 | In certain cases, the OS-level handles created by the \fIpipe()\fR, \fIsocket()\fR, |
| 403 | and \fIaccept()\fR operators are apparently not duplicated accurately in |
| 404 | pseudo\-processes. This only happens in some situations, but where it |
| 405 | does happen, it may result in deadlocks between the read and write ends |
| 406 | of pipe handles, or inability to send or receive data across socket |
| 407 | handles. |
| 408 | .IP "\(bu" 8 |
| 409 | This document may be incomplete in some respects. |
| 410 | .SH "AUTHOR" |
| 411 | .IX Header "AUTHOR" |
| 412 | Support for concurrent interpreters and the \fIfork()\fR emulation was implemented |
| 413 | by ActiveState, with funding from Microsoft Corporation. |
| 414 | .PP |
| 415 | This document is authored and maintained by Gurusamy Sarathy |
| 416 | <gsar@activestate.com>. |
| 417 | .SH "SEE ALSO" |
| 418 | .IX Header "SEE ALSO" |
| 419 | \&\*(L"fork\*(R" in perlfunc, perlipc |