Implement checkable "minimal complete definitions" (#7633)
[ghc.git] / docs / users_guide / using.xml
1 <?xml version="1.0" encoding="iso-8859-1"?>
2 <chapter id="using-ghc">
3 <title>Using GHC</title>
4
5 <indexterm><primary>GHC, using</primary></indexterm>
6 <indexterm><primary>using GHC</primary></indexterm>
7
8 <sect1>
9 <title>Getting started: compiling programs</title>
10
11 <para>
12 In this chapter you'll find a complete reference to the GHC
13 command-line syntax, including all 400+ flags. It's a large and
14 complex system, and there are lots of details, so it can be
15 quite hard to figure out how to get started. With that in mind,
16 this introductory section provides a quick introduction to the
17 basic usage of GHC for compiling a Haskell program, before the
18 following sections dive into the full syntax.
19 </para>
20
21 <para>
22 Let's create a Hello World program, and compile and run it.
23 First, create a file <filename>hello.hs</filename> containing
24 the Haskell code:
25 </para>
26
27 <programlisting>
28 main = putStrLn "Hello, World!"
29 </programlisting>
30
31 <para>To compile the program, use GHC like this:</para>
32
33 <screen>
34 $ ghc hello.hs
35 </screen>
36
37 <para>(where <literal>$</literal> represents the prompt: don't
38 type it). GHC will compile the source
39 file <filename>hello.hs</filename>, producing
40 an <firstterm>object
41 file</firstterm> <filename>hello.o</filename> and
42 an <firstterm>interface
43 file</firstterm> <filename>hello.hi</filename>, and then it
44 will link the object file to the libraries that come with GHC
45 to produce an executable called <filename>hello</filename> on
46 Unix/Linux/Mac, or <filename>hello.exe</filename> on
47 Windows.</para>
48
49 <para>
50 By default GHC will be very quiet about what it is doing, only
51 printing error messages. If you want to see in more detail
52 what's going on behind the scenes, add <option>-v</option> to
53 the command line.
54 </para>
55
56 <para>
57 Then we can run the program like this:
58 </para>
59
60 <screen>
61 $ ./hello
62 Hello World!
63 </screen>
64
65 <para>
66 If your program contains multiple modules, then you only need to
67 tell GHC the name of the source file containing
68 the <filename>Main</filename> module, and GHC will examine
69 the <literal>import</literal> declarations to find the other
70 modules that make up the program and find their source files.
71 This means that, with the exception of
72 the <literal>Main</literal> module, every source file should be
73 named after the module name that it contains (with dots replaced
74 by directory separators). For example, the
75 module <literal>Data.Person</literal> would be in the
76 file <filename>Data/Person.hs</filename> on Unix/Linux/Mac,
77 or <filename>Data\Person.hs</filename> on Windows.
78 </para>
79 </sect1>
80
81 <sect1>
82 <title>Options overview</title>
83
84 <para>GHC's behaviour is controlled by
85 <firstterm>options</firstterm>, which for historical reasons are
86 also sometimes referred to as command-line flags or arguments.
87 Options can be specified in three ways:</para>
88
89 <sect2>
90 <title>Command-line arguments</title>
91
92 <indexterm><primary>structure, command-line</primary></indexterm>
93 <indexterm><primary>command-line</primary><secondary>arguments</secondary></indexterm>
94 <indexterm><primary>arguments</primary><secondary>command-line</secondary></indexterm>
95
96 <para>An invocation of GHC takes the following form:</para>
97
98 <screen>
99 ghc [argument...]
100 </screen>
101
102 <para>Command-line arguments are either options or file names.</para>
103
104 <para>Command-line options begin with <literal>-</literal>.
105 They may <emphasis>not</emphasis> be grouped:
106 <option>-vO</option> is different from <option>-v -O</option>.
107 Options need not precede filenames: e.g., <literal>ghc *.o -o
108 foo</literal>. All options are processed and then applied to
109 all files; you cannot, for example, invoke <literal>ghc -c -O1
110 Foo.hs -O2 Bar.hs</literal> to apply different optimisation
111 levels to the files <filename>Foo.hs</filename> and
112 <filename>Bar.hs</filename>.</para>
113 </sect2>
114
115 <sect2 id="source-file-options">
116 <title>Command line options in source files</title>
117
118 <indexterm><primary>source-file options</primary></indexterm>
119
120 <para>Sometimes it is useful to make the connection between a
121 source file and the command-line options it requires quite
122 tight. For instance, if a Haskell source file deliberately
123 uses name shadowing, it should be compiled with the
124 <option>-fno-warn-name-shadowing</option> option. Rather than maintaining
125 the list of per-file options in a <filename>Makefile</filename>,
126 it is possible to do this directly in the source file using the
127 <literal>OPTIONS_GHC</literal> pragma <indexterm><primary>OPTIONS_GHC
128 pragma</primary></indexterm>:</para>
129
130 <programlisting>
131 {-# OPTIONS_GHC -fno-warn-name-shadowing #-}
132 module X where
133 ...
134 </programlisting>
135
136 <para><literal>OPTIONS_GHC</literal> is a <emphasis>file-header pragma</emphasis>
137 (see <xref linkend="pragmas"/>).</para>
138
139 <para>Only <emphasis>dynamic</emphasis> flags can be used in an <literal>OPTIONS_GHC</literal> pragma
140 (see <xref linkend="static-dynamic-flags"/>).</para>
141
142 <para>Note that your command shell does not
143 get to the source file options, they are just included literally
144 in the array of command-line arguments the compiler
145 maintains internally, so you'll be desperately disappointed if
146 you try to glob etc. inside <literal>OPTIONS_GHC</literal>.</para>
147
148 <para>NOTE: the contents of OPTIONS_GHC are appended to the
149 command-line options, so options given in the source file
150 override those given on the command-line.</para>
151
152 <para>It is not recommended to move all the contents of your
153 Makefiles into your source files, but in some circumstances, the
154 <literal>OPTIONS_GHC</literal> pragma is the Right Thing. (If you
155 use <option>-keep-hc-file</option> and have OPTION flags in
156 your module, the OPTIONS_GHC will get put into the generated .hc
157 file).</para>
158 </sect2>
159
160 <sect2>
161 <title>Setting options in GHCi</title>
162
163 <para>Options may also be modified from within GHCi, using the
164 <literal>:set</literal> command. See <xref linkend="ghci-set"/>
165 for more details.</para>
166 </sect2>
167 </sect1>
168
169 <sect1 id="static-dynamic-flags">
170 <title>Static, Dynamic, and Mode options</title>
171 <indexterm><primary>static</primary><secondary>options</secondary>
172 </indexterm>
173 <indexterm><primary>dynamic</primary><secondary>options</secondary>
174 </indexterm>
175 <indexterm><primary>mode</primary><secondary>options</secondary>
176 </indexterm>
177
178 <para>Each of GHC's command line options is classified as
179 <firstterm>static</firstterm>, <firstterm>dynamic</firstterm> or
180 <firstterm>mode</firstterm>:</para>
181
182 <variablelist>
183 <varlistentry>
184 <term>Mode flags</term>
185 <listitem>
186 <para>For example, <option>--make</option> or <option>-E</option>.
187 There may only be a single mode flag on the command line. The
188 available modes are listed in <xref linkend="modes"/>.</para>
189 </listitem>
190 </varlistentry>
191 <varlistentry>
192 <term>Dynamic Flags</term>
193 <listitem>
194 <para>Most non-mode flags fall into this category. A dynamic flag
195 may be used on the command line, in a
196 <literal>OPTIONS_GHC</literal> pragma in a source file, or set
197 using <literal>:set</literal> in GHCi.</para>
198 </listitem>
199 </varlistentry>
200 <varlistentry>
201 <term>Static Flags</term>
202 <listitem>
203 <para>A few flags are "static", which means they can only be used on
204 the command-line, and remain in force over the entire GHC/GHCi
205 run.</para>
206 </listitem>
207 </varlistentry>
208 </variablelist>
209
210 <para>The flag reference tables (<xref
211 linkend="flag-reference"/>) lists the status of each flag.</para>
212
213 <para>There are a few flags that are static except that they can
214 also be used with GHCi's <literal>:set</literal> command; these
215 are listed as &ldquo;static/<literal>:set</literal>&rdquo; in the
216 table.</para>
217 </sect1>
218
219 <sect1 id="file-suffixes">
220 <title>Meaningful file suffixes</title>
221
222 <indexterm><primary>suffixes, file</primary></indexterm>
223 <indexterm><primary>file suffixes for GHC</primary></indexterm>
224
225 <para>File names with &ldquo;meaningful&rdquo; suffixes (e.g.,
226 <filename>.lhs</filename> or <filename>.o</filename>) cause the
227 &ldquo;right thing&rdquo; to happen to those files.</para>
228
229 <variablelist>
230
231 <varlistentry>
232 <term><filename>.hs</filename></term>
233 <listitem>
234 <para>A Haskell module.</para>
235 </listitem>
236 </varlistentry>
237
238 <varlistentry>
239 <term>
240 <filename>.lhs</filename>
241 <indexterm><primary><literal>lhs</literal> suffix</primary></indexterm>
242 </term>
243 <listitem>
244 <para>A &ldquo;literate Haskell&rdquo; module.</para>
245 </listitem>
246 </varlistentry>
247
248 <varlistentry>
249 <term><filename>.hi</filename></term>
250 <listitem>
251 <para>A Haskell interface file, probably
252 compiler-generated.</para>
253 </listitem>
254 </varlistentry>
255
256 <varlistentry>
257 <term><filename>.hc</filename></term>
258 <listitem>
259 <para>Intermediate C file produced by the Haskell
260 compiler.</para>
261 </listitem>
262 </varlistentry>
263
264 <varlistentry>
265 <term><filename>.c</filename></term>
266 <listitem>
267 <para>A C&nbsp;file not produced by the Haskell
268 compiler.</para>
269 </listitem>
270 </varlistentry>
271
272 <varlistentry>
273 <term><filename>.ll</filename></term>
274 <listitem>
275 <para>An llvm-intermediate-language source file, usually
276 produced by the compiler.</para>
277 </listitem>
278 </varlistentry>
279
280 <varlistentry>
281 <term><filename>.bc</filename></term>
282 <listitem>
283 <para>An llvm-intermediate-language bitcode file, usually
284 produced by the compiler.</para>
285 </listitem>
286 </varlistentry>
287
288 <varlistentry>
289 <term><filename>.s</filename></term>
290 <listitem>
291 <para>An assembly-language source file, usually produced by
292 the compiler.</para>
293 </listitem>
294 </varlistentry>
295
296 <varlistentry>
297 <term><filename>.o</filename></term>
298 <listitem>
299 <para>An object file, produced by an assembler.</para>
300 </listitem>
301 </varlistentry>
302 </variablelist>
303
304 <para>Files with other suffixes (or without suffixes) are passed
305 straight to the linker.</para>
306
307 </sect1>
308
309 <sect1 id="modes">
310 <title>Modes of operation</title>
311
312 <para>
313 GHC's behaviour is firstly controlled by a mode flag. Only one
314 of these flags may be given, but it does not necessarily need to
315 be the first option on the command-line.
316 </para>
317
318 <para>
319 If no mode flag is present, then GHC will enter make mode
320 (<xref linkend="make-mode" />) if there are any Haskell source
321 files given on the command line, or else it will link the
322 objects named on the command line to produce an executable.
323 </para>
324
325 <para>The available mode flags are:</para>
326
327 <variablelist>
328 <varlistentry>
329 <term>
330 <cmdsynopsis><command>ghc --interactive</command>
331 </cmdsynopsis>
332 <indexterm><primary>interactive mode</primary></indexterm>
333 <indexterm><primary>ghci</primary></indexterm>
334 </term>
335 <listitem>
336 <para>Interactive mode, which is also available as
337 <command>ghci</command>. Interactive mode is described in
338 more detail in <xref linkend="ghci"/>.</para>
339 </listitem>
340 </varlistentry>
341
342 <varlistentry>
343 <term>
344 <cmdsynopsis><command>ghc --make</command>
345 </cmdsynopsis>
346 <indexterm><primary>make mode</primary></indexterm>
347 <indexterm><primary><option>--make</option></primary></indexterm>
348 </term>
349 <listitem>
350 <para>In this mode, GHC will build a multi-module Haskell
351 program automatically, figuring out dependencies for itself.
352 If you have a straightforward Haskell program, this is
353 likely to be much easier, and faster, than using
354 <command>make</command>. Make mode is described in <xref
355 linkend="make-mode"/>.</para>
356
357 <para>
358 This mode is the default if there are any Haskell
359 source files mentioned on the command line, and in this case
360 the <option>--make</option> option can be omitted.
361 </para>
362 </listitem>
363 </varlistentry>
364
365 <varlistentry>
366 <term>
367 <cmdsynopsis><command>ghc -e</command>
368 <arg choice='plain'><replaceable>expr</replaceable></arg>
369 </cmdsynopsis>
370 <indexterm><primary>eval mode</primary></indexterm>
371 </term>
372 <listitem>
373 <para>Expression-evaluation mode. This is very similar to
374 interactive mode, except that there is a single expression
375 to evaluate (<replaceable>expr</replaceable>) which is given
376 on the command line. See <xref linkend="eval-mode"/> for
377 more details.</para>
378 </listitem>
379 </varlistentry>
380
381 <varlistentry>
382 <term>
383 <cmdsynopsis>
384 <command>ghc -E</command>
385 <command>ghc -c</command>
386 <command>ghc -S</command>
387 <command>ghc -c</command>
388 </cmdsynopsis>
389 <indexterm><primary><option>-E</option></primary></indexterm>
390 <indexterm><primary><option>-C</option></primary></indexterm>
391 <indexterm><primary><option>-S</option></primary></indexterm>
392 <indexterm><primary><option>-c</option></primary></indexterm>
393 </term>
394 <listitem>
395 <para>This is the traditional batch-compiler mode, in which
396 GHC can compile source files one at a time, or link objects
397 together into an executable. This mode also applies if
398 there is no other mode flag specified on the command line,
399 in which case it means that the specified files should be
400 compiled and then linked to form a program. See <xref
401 linkend="options-order"/>.</para>
402 </listitem>
403 </varlistentry>
404
405 <varlistentry>
406 <term>
407 <cmdsynopsis>
408 <command>ghc -M</command>
409 </cmdsynopsis>
410 <indexterm><primary>dependency-generation mode</primary></indexterm>
411 </term>
412 <listitem>
413 <para>Dependency-generation mode. In this mode, GHC can be
414 used to generate dependency information suitable for use in
415 a <literal>Makefile</literal>. See <xref
416 linkend="makefile-dependencies"/>.</para>
417 </listitem>
418 </varlistentry>
419
420 <varlistentry>
421 <term>
422 <cmdsynopsis>
423 <command>ghc --mk-dll</command>
424 </cmdsynopsis>
425 <indexterm><primary>DLL-creation mode</primary></indexterm>
426 </term>
427 <listitem>
428 <para>DLL-creation mode (Windows only). See <xref
429 linkend="win32-dlls-create"/>.</para>
430 </listitem>
431 </varlistentry>
432
433 <varlistentry>
434 <term>
435 <cmdsynopsis>
436 <command>ghc --help</command> <command>ghc -?</command>
437 </cmdsynopsis>
438 <indexterm><primary><option>--help</option></primary></indexterm>
439 </term>
440 <listitem>
441 <para>Cause GHC to spew a long usage message to standard
442 output and then exit.</para>
443 </listitem>
444 </varlistentry>
445
446 <varlistentry>
447 <term>
448 <cmdsynopsis>
449 <command>ghc --show-iface <replaceable>file</replaceable></command>
450 </cmdsynopsis>
451 <indexterm><primary><option>--show-iface</option></primary></indexterm>
452 </term>
453 <listitem>
454 <para>Read the interface in
455 <replaceable>file</replaceable> and dump it as text to
456 <literal>stdout</literal>. For example <literal>ghc --show-iface M.hi</literal>.</para>
457 </listitem>
458 </varlistentry>
459
460 <varlistentry>
461 <term>
462 <cmdsynopsis>
463 <command>ghc --supported-extensions</command>
464 <command>ghc --supported-languages</command>
465 </cmdsynopsis>
466 <indexterm><primary><option>--supported-extensions</option></primary><primary><option>--supported-languages</option></primary></indexterm>
467 </term>
468 <listitem>
469 <para>Print the supported language extensions.</para>
470 </listitem>
471 </varlistentry>
472
473 <varlistentry>
474 <term>
475 <cmdsynopsis>
476 <command>ghc --show-options</command>
477 </cmdsynopsis>
478 <indexterm><primary><option>--show-options</option></primary></indexterm>
479 </term>
480 <listitem>
481 <para>Print the supported command line options. This flag can be used for autocompletion in a shell.</para>
482 </listitem>
483 </varlistentry>
484
485 <varlistentry>
486 <term>
487 <cmdsynopsis>
488 <command>ghc --info</command>
489 </cmdsynopsis>
490 <indexterm><primary><option>--info</option></primary></indexterm>
491 </term>
492 <listitem>
493 <para>Print information about the compiler.</para>
494 </listitem>
495 </varlistentry>
496
497 <varlistentry>
498 <term>
499 <cmdsynopsis>
500 <command>ghc --version</command>
501 <command>ghc -V</command>
502 </cmdsynopsis>
503 <indexterm><primary><option>-V</option></primary></indexterm>
504 <indexterm><primary><option>--version</option></primary></indexterm>
505 </term>
506 <listitem>
507 <para>Print a one-line string including GHC's version number.</para>
508 </listitem>
509 </varlistentry>
510
511 <varlistentry>
512 <term>
513 <cmdsynopsis>
514 <command>ghc --numeric-version</command>
515 </cmdsynopsis>
516 <indexterm><primary><option>--numeric-version</option></primary></indexterm>
517 </term>
518 <listitem>
519 <para>Print GHC's numeric version number only.</para>
520 </listitem>
521 </varlistentry>
522
523 <varlistentry>
524 <term>
525 <cmdsynopsis>
526 <command>ghc --print-libdir</command>
527 </cmdsynopsis>
528 <indexterm><primary><option>--print-libdir</option></primary></indexterm>
529 </term>
530 <listitem>
531 <para>Print the path to GHC's library directory. This is
532 the top of the directory tree containing GHC's libraries,
533 interfaces, and include files (usually something like
534 <literal>/usr/local/lib/ghc-5.04</literal> on Unix). This
535 is the value of
536 <literal>$libdir</literal><indexterm><primary><literal>libdir</literal></primary></indexterm>
537 in the package configuration file
538 (see <xref linkend="packages"/>).</para>
539 </listitem>
540 </varlistentry>
541
542 </variablelist>
543
544 <sect2 id="make-mode">
545 <title>Using <command>ghc</command> <option>--make</option></title>
546 <indexterm><primary><option>--make</option></primary></indexterm>
547 <indexterm><primary>separate compilation</primary></indexterm>
548
549 <para>In this mode, GHC will build a multi-module Haskell program by following
550 dependencies from one or more root modules (usually just
551 <literal>Main</literal>). For example, if your
552 <literal>Main</literal> module is in a file called
553 <filename>Main.hs</filename>, you could compile and link the
554 program like this:</para>
555
556 <screen>
557 ghc --make Main.hs
558 </screen>
559
560 <para>
561 In fact, GHC enters make mode automatically if there are any
562 Haskell source files on the command line and no other mode is
563 specified, so in this case we could just type
564 </para>
565
566 <screen>
567 ghc Main.hs
568 </screen>
569
570 <para>Any number of source file names or module names may be
571 specified; GHC will figure out all the modules in the program by
572 following the imports from these initial modules. It will then
573 attempt to compile each module which is out of date, and
574 finally, if there is a <literal>Main</literal> module, the
575 program will also be linked into an executable.</para>
576
577 <para>The main advantages to using <literal>ghc
578 --make</literal> over traditional
579 <literal>Makefile</literal>s are:</para>
580
581 <itemizedlist>
582 <listitem>
583 <para>GHC doesn't have to be restarted for each compilation,
584 which means it can cache information between compilations.
585 Compiling a multi-module program with <literal>ghc
586 --make</literal> can be up to twice as fast as
587 running <literal>ghc</literal> individually on each source
588 file.</para>
589 </listitem>
590 <listitem>
591 <para>You don't have to write a <literal>Makefile</literal>.</para>
592 <indexterm><primary><literal>Makefile</literal>s</primary><secondary>avoiding</secondary></indexterm>
593 </listitem>
594 <listitem>
595 <para>GHC re-calculates the dependencies each time it is
596 invoked, so the dependencies never get out of sync with the
597 source.</para>
598 </listitem>
599 <listitem>
600 <para>Using the <literal>-j</literal> flag, you can compile
601 modules in parallel. Specify <literal>-jN</literal> to
602 compile <replaceable>N</replaceable> jobs in parallel.</para>
603 </listitem>
604 </itemizedlist>
605
606 <para>Any of the command-line options described in the rest of
607 this chapter can be used with
608 <option>--make</option>, but note that any options
609 you give on the command line will apply to all the source files
610 compiled, so if you want any options to apply to a single source
611 file only, you'll need to use an <literal>OPTIONS_GHC</literal>
612 pragma (see <xref linkend="source-file-options"/>).</para>
613
614 <para>If the program needs to be linked with additional objects
615 (say, some auxiliary C code), then the object files can be
616 given on the command line and GHC will include them when linking
617 the executable.</para>
618
619 <para>Note that GHC can only follow dependencies if it has the
620 source file available, so if your program includes a module for
621 which there is no source file, even if you have an object and an
622 interface file for the module, then GHC will complain. The
623 exception to this rule is for package modules, which may or may
624 not have source files.</para>
625
626 <para>The source files for the program don't all need to be in
627 the same directory; the <option>-i</option> option can be used
628 to add directories to the search path (see <xref
629 linkend="search-path"/>).</para>
630 </sect2>
631
632 <sect2 id="eval-mode">
633 <title>Expression evaluation mode</title>
634
635 <para>This mode is very similar to interactive mode, except that
636 there is a single expression to evaluate which is specified on
637 the command line as an argument to the <option>-e</option>
638 option:</para>
639
640 <screen>
641 ghc -e <replaceable>expr</replaceable>
642 </screen>
643
644 <para>Haskell source files may be named on the command line, and
645 they will be loaded exactly as in interactive mode. The
646 expression is evaluated in the context of the loaded
647 modules.</para>
648
649 <para>For example, to load and run a Haskell program containing
650 a module <literal>Main</literal>, we might say</para>
651
652 <screen>
653 ghc -e Main.main Main.hs
654 </screen>
655
656 <para>or we can just use this mode to evaluate expressions in
657 the context of the <literal>Prelude</literal>:</para>
658
659 <screen>
660 $ ghc -e "interact (unlines.map reverse.lines)"
661 hello
662 olleh
663 </screen>
664 </sect2>
665
666 <sect2 id="options-order">
667 <title>Batch compiler mode</title>
668
669 <para>In <emphasis>batch mode</emphasis>, GHC will compile one or more source files
670 given on the command line.</para>
671
672 <para>The first phase to run is determined by each input-file
673 suffix, and the last phase is determined by a flag. If no
674 relevant flag is present, then go all the way through to linking.
675 This table summarises:</para>
676
677 <informaltable>
678 <tgroup cols="4">
679 <colspec align="left"/>
680 <colspec align="left"/>
681 <colspec align="left"/>
682 <colspec align="left"/>
683
684 <thead>
685 <row>
686 <entry>Phase of the compilation system</entry>
687 <entry>Suffix saying &ldquo;start here&rdquo;</entry>
688 <entry>Flag saying &ldquo;stop after&rdquo;</entry>
689 <entry>(suffix of) output file</entry>
690 </row>
691 </thead>
692 <tbody>
693 <row>
694 <entry>literate pre-processor</entry>
695 <entry><literal>.lhs</literal></entry>
696 <entry>-</entry>
697 <entry><literal>.hs</literal></entry>
698 </row>
699
700 <row>
701 <entry>C pre-processor (opt.) </entry>
702 <entry><literal>.hs</literal> (with
703 <option>-cpp</option>)</entry>
704 <entry><option>-E</option></entry>
705 <entry><literal>.hspp</literal></entry>
706 </row>
707
708 <row>
709 <entry>Haskell compiler</entry>
710 <entry><literal>.hs</literal></entry>
711 <entry><option>-C</option>, <option>-S</option></entry>
712 <entry><literal>.hc</literal>, <literal>.s</literal></entry>
713 </row>
714
715 <row>
716 <entry>C compiler (opt.)</entry>
717 <entry><literal>.hc</literal> or <literal>.c</literal></entry>
718 <entry><option>-S</option></entry>
719 <entry><literal>.s</literal></entry>
720 </row>
721
722 <row>
723 <entry>assembler</entry>
724 <entry><literal>.s</literal></entry>
725 <entry><option>-c</option></entry>
726 <entry><literal>.o</literal></entry>
727 </row>
728
729 <row>
730 <entry>linker</entry>
731 <entry><replaceable>other</replaceable></entry>
732 <entry>-</entry>
733 <entry><filename>a.out</filename></entry>
734 </row>
735 </tbody>
736 </tgroup>
737 </informaltable>
738
739 <indexterm><primary><option>-C</option></primary></indexterm>
740 <indexterm><primary><option>-E</option></primary></indexterm>
741 <indexterm><primary><option>-S</option></primary></indexterm>
742 <indexterm><primary><option>-c</option></primary></indexterm>
743
744 <para>Thus, a common invocation would be: </para>
745
746 <screen>
747 ghc -c Foo.hs
748 </screen>
749
750 <para>to compile the Haskell source file
751 <filename>Foo.hs</filename> to an object file
752 <filename>Foo.o</filename>.</para>
753
754 <para>Note: What the Haskell compiler proper produces depends on what
755 backend code generator is used. See <xref linkend="code-generators"/>
756 for more details.</para>
757
758 <para>Note: C pre-processing is optional, the
759 <option>-cpp</option><indexterm><primary><option>-cpp</option></primary></indexterm>
760 flag turns it on. See <xref linkend="c-pre-processor"/> for more
761 details.</para>
762
763 <para>Note: The option <option>-E</option><indexterm><primary>-E
764 option</primary></indexterm> runs just the pre-processing passes
765 of the compiler, dumping the result in a file.</para>
766
767 <sect3 id="overriding-suffixes">
768 <title>Overriding the default behaviour for a file</title>
769
770 <para>As described above, the way in which a file is processed by GHC
771 depends on its suffix. This behaviour can be overridden using the
772 <option>-x</option> option:</para>
773
774 <variablelist>
775 <varlistentry>
776 <term><option>-x</option> <replaceable>suffix</replaceable>
777 <indexterm><primary><option>-x</option></primary>
778 </indexterm></term>
779 <listitem>
780 <para>Causes all files following this option on the command
781 line to be processed as if they had the suffix
782 <replaceable>suffix</replaceable>. For example, to compile a
783 Haskell module in the file <literal>M.my-hs</literal>,
784 use <literal>ghc -c -x hs M.my-hs</literal>.</para>
785 </listitem>
786 </varlistentry>
787 </variablelist>
788 </sect3>
789
790 </sect2>
791 </sect1>
792
793 <sect1 id="options-help">
794 <title>Help and verbosity options</title>
795
796 <indexterm><primary>help options</primary></indexterm>
797 <indexterm><primary>verbosity options</primary></indexterm>
798
799 <para>See also the <option>--help</option>, <option>--version</option>, <option>--numeric-version</option>,
800 and <option>--print-libdir</option> modes in <xref linkend="modes"/>.</para>
801 <variablelist>
802 <varlistentry>
803 <term>
804 <option>-v</option>
805 <indexterm><primary><option>-v</option></primary></indexterm>
806 </term>
807 <listitem>
808 <para>The <option>-v</option> option makes GHC
809 <emphasis>verbose</emphasis>: it reports its version number
810 and shows (on stderr) exactly how it invokes each phase of
811 the compilation system. Moreover, it passes the
812 <option>-v</option> flag to most phases; each reports its
813 version number (and possibly some other information).</para>
814
815 <para>Please, oh please, use the <option>-v</option> option
816 when reporting bugs! Knowing that you ran the right bits in
817 the right order is always the first thing we want to
818 verify.</para>
819 </listitem>
820 </varlistentry>
821
822 <varlistentry>
823 <term>
824 <option>-v</option><replaceable>n</replaceable>
825 <indexterm><primary><option>-v</option></primary></indexterm>
826 </term>
827 <listitem>
828 <para>To provide more control over the compiler's verbosity,
829 the <option>-v</option> flag takes an optional numeric
830 argument. Specifying <option>-v</option> on its own is
831 equivalent to <option>-v3</option>, and the other levels
832 have the following meanings:</para>
833
834 <variablelist>
835 <varlistentry>
836 <term><option>-v0</option></term>
837 <listitem>
838 <para>Disable all non-essential messages (this is the
839 default).</para>
840 </listitem>
841 </varlistentry>
842
843 <varlistentry>
844 <term><option>-v1</option></term>
845 <listitem>
846 <para>Minimal verbosity: print one line per
847 compilation (this is the default when
848 <option>--make</option> or
849 <option>--interactive</option> is on).</para>
850 </listitem>
851 </varlistentry>
852
853 <varlistentry>
854 <term><option>-v2</option></term>
855 <listitem>
856 <para>Print the name of each compilation phase as it
857 is executed. (equivalent to
858 <option>-dshow-passes</option>).</para>
859 </listitem>
860 </varlistentry>
861
862 <varlistentry>
863 <term><option>-v3</option></term>
864 <listitem>
865 <para>The same as <option>-v2</option>, except that in
866 addition the full command line (if appropriate) for
867 each compilation phase is also printed.</para>
868 </listitem>
869 </varlistentry>
870
871 <varlistentry>
872 <term><option>-v4</option></term>
873 <listitem>
874 <para>The same as <option>-v3</option> except that the
875 intermediate program representation after each
876 compilation phase is also printed (excluding
877 preprocessed and C/assembly files).</para>
878 </listitem>
879 </varlistentry>
880 </variablelist>
881 </listitem>
882 </varlistentry>
883
884 <varlistentry>
885 <term><option>-ferror-spans</option>
886 <indexterm><primary><option>-ferror-spans</option></primary>
887 </indexterm>
888 </term>
889 <listitem>
890 <para>Causes GHC to emit the full source span of the
891 syntactic entity relating to an error message. Normally, GHC
892 emits the source location of the start of the syntactic
893 entity only.</para>
894
895 <para>For example:</para>
896
897 <screen>
898 test.hs:3:6: parse error on input `where'
899 </screen>
900
901 <para>becomes:</para>
902
903 <screen>
904 test296.hs:3:6-10: parse error on input `where'
905 </screen>
906
907 <para>And multi-line spans are possible too:</para>
908
909 <screen>
910 test.hs:(5,4)-(6,7):
911 Conflicting definitions for `a'
912 Bound at: test.hs:5:4
913 test.hs:6:7
914 In the binding group for: a, b, a
915 </screen>
916
917 <para>Note that line numbers start counting at one, but
918 column numbers start at zero. This choice was made to
919 follow existing convention (i.e. this is how Emacs does
920 it).</para>
921 </listitem>
922 </varlistentry>
923
924 <varlistentry>
925 <term><option>-H</option><replaceable>size</replaceable>
926 <indexterm><primary><option>-H</option></primary></indexterm>
927 </term>
928 <listitem>
929 <para>Set the minimum size of the heap to
930 <replaceable>size</replaceable>.
931 This option is equivalent to
932 <literal>+RTS&nbsp;-H<replaceable>size</replaceable></literal>,
933 see <xref linkend="rts-options-gc" />.
934 </para>
935 </listitem>
936 </varlistentry>
937
938 <varlistentry>
939 <term><option>-Rghc-timing</option>
940 <indexterm><primary><option>-Rghc-timing</option></primary></indexterm>
941 </term>
942 <listitem>
943 <para>Prints a one-line summary of timing statistics for the
944 GHC run. This option is equivalent to
945 <literal>+RTS&nbsp;-tstderr</literal>, see <xref
946 linkend="rts-options-gc" />.
947 </para>
948 </listitem>
949 </varlistentry>
950 </variablelist>
951 </sect1>
952
953 &separate;
954
955 <sect1 id="options-sanity">
956 <title>Warnings and sanity-checking</title>
957
958 <indexterm><primary>sanity-checking options</primary></indexterm>
959 <indexterm><primary>warnings</primary></indexterm>
960
961
962 <para>GHC has a number of options that select which types of
963 non-fatal error messages, otherwise known as warnings, can be
964 generated during compilation. By default, you get a standard set
965 of warnings which are generally likely to indicate bugs in your
966 program. These are:
967 <option>-fwarn-overlapping-patterns</option>,
968 <option>-fwarn-warnings-deprecations</option>,
969 <option>-fwarn-amp</option>,
970 <option>-fwarn-deprecated-flags</option>,
971 <option>-fwarn-unrecognised-pragmas</option>,
972 <option>-fwarn-pointless-pragmas</option>,
973 <option>-fwarn-duplicate-constraints</option>,
974 <option>-fwarn-duplicate-exports</option>,
975 <option>-fwarn-overflowed-literals</option>,
976 <option>-fwarn-empty-enumerations</option>,
977 <option>-fwarn-missing-fields</option>,
978 <option>-fwarn-missing-methods</option>,
979 <option>-fwarn-lazy-unlifted-bindings</option>,
980 <option>-fwarn-wrong-do-bind</option>,
981 <option>-fwarn-unsupported-calling-conventions</option>,
982 <option>-fwarn-dodgy-foreign-imports</option>,
983 <option>-fwarn-inline-rule-shadowing</option>, and
984 <option>-fwarn-unsupported-llvm-version</option>.
985 The following flags are simple ways to select standard
986 &ldquo;packages&rdquo; of warnings:
987 </para>
988
989 <variablelist>
990
991 <varlistentry>
992 <term><option>-W</option>:</term>
993 <listitem>
994 <indexterm><primary>-W option</primary></indexterm>
995 <para>Provides the standard warnings plus
996 <option>-fwarn-incomplete-patterns</option>,
997 <option>-fwarn-dodgy-exports</option>,
998 <option>-fwarn-dodgy-imports</option>,
999 <option>-fwarn-unused-matches</option>,
1000 <option>-fwarn-unused-imports</option>, and
1001 <option>-fwarn-unused-binds</option>.</para>
1002 </listitem>
1003 </varlistentry>
1004
1005 <varlistentry>
1006 <term><option>-Wall</option>:</term>
1007 <listitem>
1008 <indexterm><primary><option>-Wall</option></primary></indexterm>
1009 <para>Turns on all warning options that indicate potentially
1010 suspicious code. The warnings that are
1011 <emphasis>not</emphasis> enabled by <option>-Wall</option>
1012 are
1013 <option>-fwarn-tabs</option>,
1014 <option>-fwarn-incomplete-uni-patterns</option>,
1015 <option>-fwarn-incomplete-record-updates</option>,
1016 <option>-fwarn-monomorphism-restriction</option>,
1017 <option>-fwarn-auto-orphans</option>,
1018 <option>-fwarn-implicit-prelude</option>,
1019 <option>-fwarn-missing-local-sigs</option>,
1020 <option>-fwarn-missing-import-lists</option>.</para>
1021 </listitem>
1022 </varlistentry>
1023
1024 <varlistentry>
1025 <term><option>-w</option>:</term>
1026 <listitem>
1027 <indexterm><primary><option>-w</option></primary></indexterm>
1028 <para>Turns off all warnings, including the standard ones and
1029 those that <literal>-Wall</literal> doesn't enable.</para>
1030 </listitem>
1031 </varlistentry>
1032
1033 <varlistentry>
1034 <term><option>-Werror</option>:</term>
1035 <listitem>
1036 <indexterm><primary><option>-Werror</option></primary></indexterm>
1037 <para>Makes any warning into a fatal error. Useful so that you don't
1038 miss warnings when doing batch compilation. </para>
1039 </listitem>
1040 </varlistentry>
1041
1042 <varlistentry>
1043 <term><option>-Wwarn</option>:</term>
1044 <listitem>
1045 <indexterm><primary><option>-Wwarn</option></primary></indexterm>
1046 <para>Warnings are treated only as warnings, not as errors. This is
1047 the default, but can be useful to negate a
1048 <option>-Werror</option> flag.</para>
1049 </listitem>
1050 </varlistentry>
1051
1052 </variablelist>
1053
1054 <para>The full set of warning options is described below. To turn
1055 off any warning, simply give the corresponding
1056 <option>-fno-warn-...</option> option on the command line.</para>
1057
1058 <variablelist>
1059
1060 <varlistentry>
1061 <term><option>-fdefer-type-errors</option>:</term>
1062 <listitem>
1063 <indexterm><primary><option>-fdefer-type-errors</option></primary>
1064 </indexterm>
1065 <indexterm><primary>warnings</primary></indexterm>
1066 <para>Defer as many type errors as possible until runtime.
1067 At compile time you get a warning (instead of an error). At
1068 runtime, if you use a value that depends on a type error, you
1069 get a runtime error; but you can run any type-correct parts of your code
1070 just fine. See <xref linkend="defer-type-errors"/></para>
1071 </listitem>
1072 </varlistentry>
1073
1074 <varlistentry>
1075 <term><option>-fhelpful-errors</option>:</term>
1076 <listitem>
1077 <indexterm><primary><option>-fhelpful-errors</option></primary>
1078 </indexterm>
1079 <indexterm><primary>warnings</primary></indexterm>
1080 <para>When a name or package is not found in scope, make
1081 suggestions for the name or package you might have meant instead.</para>
1082 <para>This option is on by default.</para>
1083 </listitem>
1084 </varlistentry>
1085
1086 <varlistentry>
1087 <term><option>-fwarn-unrecognised-pragmas</option>:</term>
1088 <listitem>
1089 <indexterm><primary><option>-fwarn-unrecognised-pragmas</option></primary>
1090 </indexterm>
1091 <indexterm><primary>warnings</primary></indexterm>
1092 <indexterm><primary>pragmas</primary></indexterm>
1093 <para>Causes a warning to be emitted when a
1094 pragma that GHC doesn't recognise is used. As well as pragmas
1095 that GHC itself uses, GHC also recognises pragmas known to be used
1096 by other tools, e.g. <literal>OPTIONS_HUGS</literal> and
1097 <literal>DERIVE</literal>.</para>
1098
1099 <para>This option is on by default.</para>
1100 </listitem>
1101 </varlistentry>
1102
1103 <varlistentry>
1104 <term><option>-fwarn-pointless-pragmas</option>:</term>
1105 <listitem>
1106 <indexterm><primary><option>-fwarn-pointless-pragmas</option></primary>
1107 </indexterm>
1108 <indexterm><primary>warnings</primary></indexterm>
1109 <indexterm><primary>pragmas</primary></indexterm>
1110 <para>Causes a warning to be emitted when GHC detects that a
1111 module contains a pragma that has no effect.</para>
1112
1113 <para>This option is on by default.</para>
1114 </listitem>
1115 </varlistentry>
1116
1117 <varlistentry>
1118 <term><option>-fwarn-warnings-deprecations</option>:</term>
1119 <listitem>
1120 <indexterm><primary><option>-fwarn-warnings-deprecations</option></primary>
1121 </indexterm>
1122 <indexterm><primary>warnings</primary></indexterm>
1123 <indexterm><primary>deprecations</primary></indexterm>
1124 <para>Causes a warning to be emitted when a
1125 module, function or type with a WARNING or DEPRECATED pragma
1126 is used. See <xref linkend="warning-deprecated-pragma"/> for more
1127 details on the pragmas.</para>
1128
1129 <para>This option is on by default.</para>
1130 </listitem>
1131 </varlistentry>
1132
1133 <varlistentry>
1134 <term><option>-fwarn-amp</option>:</term>
1135 <listitem>
1136 <indexterm><primary><option>-fwarn-amp</option></primary>
1137 </indexterm>
1138 <indexterm><primary>amp</primary></indexterm>
1139 <indexterm><primary>applicative-monad proposal</primary></indexterm>
1140 <para>Causes a warning to be emitted when a definition
1141 is in conflict with the AMP (Applicative-Monad proosal),
1142 namely:
1143 1. Instance of Monad without Applicative;
1144 2. Instance of MonadPlus without Alternative;
1145 3. Custom definitions of join/pure/&lt;*&gt;</para>
1146
1147 <para>This option is on by default.</para>
1148 </listitem>
1149 </varlistentry>
1150
1151 <varlistentry>
1152 <term><option>-fwarn-deprecated-flags</option>:</term>
1153 <listitem>
1154 <indexterm><primary><option>-fwarn-deprecated-flags</option></primary>
1155 </indexterm>
1156 <indexterm><primary>deprecated-flags</primary></indexterm>
1157 <para>Causes a warning to be emitted when a deprecated
1158 commandline flag is used.</para>
1159
1160 <para>This option is on by default.</para>
1161 </listitem>
1162 </varlistentry>
1163
1164 <varlistentry>
1165 <term><option>-fwarn-unsupported-calling-conventions</option>:</term>
1166 <listitem>
1167 <indexterm><primary><option>-fwarn-unsupported-calling-conventions</option></primary>
1168 </indexterm>
1169 <para>Causes a warning to be emitted for foreign declarations
1170 that use unsupported calling conventions. In particular,
1171 if the <literal>stdcall</literal> calling convention is used
1172 on an architecture other than i386 then it will be treated
1173 as <literal>ccall</literal>.</para>
1174 </listitem>
1175 </varlistentry>
1176
1177 <varlistentry>
1178 <term><option>-fwarn-dodgy-foreign-imports</option>:</term>
1179 <listitem>
1180 <indexterm><primary><option>-fwarn-dodgy-foreign-imports</option></primary>
1181 </indexterm>
1182 <para>Causes a warning to be emitted for foreign imports of
1183 the following form:</para>
1184
1185 <programlisting>
1186 foreign import "f" f :: FunPtr t
1187 </programlisting>
1188
1189 <para>on the grounds that it probably should be</para>
1190
1191 <programlisting>
1192 foreign import "&amp;f" f :: FunPtr t
1193 </programlisting>
1194
1195 <para>The first form declares that `f` is a (pure) C
1196 function that takes no arguments and returns a pointer to a
1197 C function with type `t`, whereas the second form declares
1198 that `f` itself is a C function with type `t`. The first
1199 declaration is usually a mistake, and one that is hard to
1200 debug because it results in a crash, hence this
1201 warning.</para>
1202 </listitem>
1203 </varlistentry>
1204
1205 <varlistentry>
1206 <term><option>-fwarn-dodgy-exports</option>:</term>
1207 <listitem>
1208 <indexterm><primary><option>-fwarn-dodgy-exports</option></primary>
1209 </indexterm>
1210 <para>Causes a warning to be emitted when a datatype
1211 <literal>T</literal> is exported
1212 with all constructors, i.e. <literal>T(..)</literal>, but is it
1213 just a type synonym.</para>
1214 <para>Also causes a warning to be emitted when a module is
1215 re-exported, but that module exports nothing.</para>
1216 </listitem>
1217 </varlistentry>
1218
1219 <varlistentry>
1220 <term><option>-fwarn-dodgy-imports</option>:</term>
1221 <listitem>
1222 <indexterm><primary><option>-fwarn-dodgy-imports</option></primary>
1223 </indexterm>
1224 <para>Causes a warning to be emitted in the following cases:</para>
1225 <itemizedlist>
1226 <listitem>
1227 <para>When a datatype <literal>T</literal> is imported with all
1228 constructors, i.e. <literal>T(..)</literal>, but has been
1229 exported abstractly, i.e. <literal>T</literal>.
1230 </para>
1231 </listitem>
1232 <listitem>
1233 <para>When an <literal>import</literal> statement hides an
1234 entity that is not exported.</para>
1235 </listitem>
1236 </itemizedlist>
1237 </listitem>
1238 </varlistentry>
1239
1240 <varlistentry>
1241 <term><option>-fwarn-overflowed-literals</option>:</term>
1242 <listitem>
1243 <indexterm><primary><option>-fwarn-overflowed-literals</option></primary>
1244 </indexterm>
1245 <para>
1246 Causes a warning to be emitted if a literal will overflow,
1247 e.g. <literal>300 :: Word8</literal>.
1248 </para>
1249 </listitem>
1250 </varlistentry>
1251
1252 <varlistentry>
1253 <term><option>-fwarn-empty-enumerations</option>:</term>
1254 <listitem>
1255 <indexterm><primary><option>-fwarn-empty-enumerations</option></primary>
1256 </indexterm>
1257 <para>
1258 Causes a warning to be emitted if an enumeration is
1259 empty, e.g. <literal>[5 .. 3]</literal>.
1260 </para>
1261 </listitem>
1262 </varlistentry>
1263
1264 <varlistentry>
1265 <term><option>-fwarn-lazy-unlifted-bindings</option>:</term>
1266 <listitem>
1267 <indexterm><primary><option>-fwarn-lazy-unlifted-bindings</option></primary>
1268 </indexterm>
1269 <para>Causes a warning to be emitted when an unlifted type
1270 is bound in a way that looks lazy, e.g.
1271 <literal>where (I# x) = ...</literal>. Use
1272 <literal>where !(I# x) = ...</literal> instead. This will be an
1273 error, rather than a warning, in GHC 7.2.
1274 </para>
1275 </listitem>
1276 </varlistentry>
1277
1278 <varlistentry>
1279 <term><option>-fwarn-duplicate-constraints</option>:</term>
1280 <listitem>
1281 <indexterm><primary><option>-fwarn-duplicate-constraints</option></primary></indexterm>
1282 <indexterm><primary>duplicate constraints, warning</primary></indexterm>
1283
1284 <para>Have the compiler warn about duplicate constraints in a type signature. For
1285 example
1286 <programlisting>
1287 f :: (Eq a, Show a, Eq a) => a -> a
1288 </programlisting>
1289 The warning will indicate the duplicated <literal>Eq a</literal> constraint.
1290 </para>
1291
1292 <para>This option is on by default.</para>
1293 </listitem>
1294 </varlistentry>
1295
1296 <varlistentry>
1297 <term><option>-fwarn-duplicate-exports</option>:</term>
1298 <listitem>
1299 <indexterm><primary><option>-fwarn-duplicate-exports</option></primary></indexterm>
1300 <indexterm><primary>duplicate exports, warning</primary></indexterm>
1301 <indexterm><primary>export lists, duplicates</primary></indexterm>
1302
1303 <para>Have the compiler warn about duplicate entries in
1304 export lists. This is useful information if you maintain
1305 large export lists, and want to avoid the continued export
1306 of a definition after you've deleted (one) mention of it in
1307 the export list.</para>
1308
1309 <para>This option is on by default.</para>
1310 </listitem>
1311 </varlistentry>
1312
1313 <varlistentry>
1314 <term><option>-fwarn-hi-shadowing</option>:</term>
1315 <listitem>
1316 <indexterm><primary><option>-fwarn-hi-shadowing</option></primary></indexterm>
1317 <indexterm><primary>shadowing</primary>
1318 <secondary>interface files</secondary></indexterm>
1319
1320 <para>Causes the compiler to emit a warning when a module or
1321 interface file in the current directory is shadowing one
1322 with the same module name in a library or other
1323 directory.</para>
1324 </listitem>
1325 </varlistentry>
1326
1327 <varlistentry>
1328 <term><option>-fwarn-identities</option>:</term>
1329 <listitem>
1330 <indexterm><primary><option>-fwarn-identities</option></primary></indexterm>
1331 <para>Causes the compiler to emit a warning when a Prelude numeric
1332 conversion converts a type T to the same type T; such calls
1333 are probably no-ops and can be omitted. The functions checked for
1334 are: <literal>toInteger</literal>,
1335 <literal>toRational</literal>,
1336 <literal>fromIntegral</literal>,
1337 and <literal>realToFrac</literal>.
1338 </para>
1339 </listitem>
1340 </varlistentry>
1341
1342 <varlistentry>
1343 <term><option>-fwarn-implicit-prelude</option>:</term>
1344 <listitem>
1345 <indexterm><primary><option>-fwarn-implicit-prelude</option></primary></indexterm>
1346 <indexterm><primary>implicit prelude, warning</primary></indexterm>
1347 <para>Have the compiler warn if the Prelude is implicitly
1348 imported. This happens unless either the Prelude module is
1349 explicitly imported with an <literal>import ... Prelude ...</literal>
1350 line, or this implicit import is disabled (either by
1351 <option>-XNoImplicitPrelude</option> or a
1352 <literal>LANGUAGE NoImplicitPrelude</literal> pragma).</para>
1353
1354 <para>Note that no warning is given for syntax that implicitly
1355 refers to the Prelude, even if <option>-XNoImplicitPrelude</option>
1356 would change whether it refers to the Prelude.
1357 For example, no warning is given when
1358 <literal>368</literal> means
1359 <literal>Prelude.fromInteger (368::Prelude.Integer)</literal>
1360 (where <literal>Prelude</literal> refers to the actual Prelude module,
1361 regardless of the imports of the module being compiled).</para>
1362
1363 <para>This warning is off by default.</para>
1364 </listitem>
1365 </varlistentry>
1366
1367 <varlistentry>
1368 <term><option>-fwarn-incomplete-patterns</option>,
1369 <option>-fwarn-incomplete-uni-patterns</option>:
1370 </term>
1371 <listitem>
1372 <indexterm><primary><option>-fwarn-incomplete-patterns</option></primary></indexterm>
1373 <indexterm><primary><option>-fwarn-incomplete-uni-patterns</option></primary></indexterm>
1374 <indexterm><primary>incomplete patterns, warning</primary></indexterm>
1375 <indexterm><primary>patterns, incomplete</primary></indexterm>
1376
1377 <para>The option <option>-fwarn-incomplete-patterns</option> warns
1378 about places where
1379 a pattern-match might fail at runtime.
1380 The function
1381 <function>g</function> below will fail when applied to
1382 non-empty lists, so the compiler will emit a warning about
1383 this when <option>-fwarn-incomplete-patterns</option> is
1384 enabled.
1385
1386 <programlisting>
1387 g [] = 2
1388 </programlisting>
1389
1390 This option isn't enabled by default because it can be
1391 a bit noisy, and it doesn't always indicate a bug in the
1392 program. However, it's generally considered good practice
1393 to cover all the cases in your functions, and it is switched
1394 on by <option>-W</option>.</para>
1395
1396 <para>The flag <option>-fwarn-incomplete-uni-patterns</option> is
1397 similar, except that it
1398 applies only to lambda-expressions and pattern bindings, constructs
1399 that only allow a single pattern:
1400
1401 <programlisting>
1402 h = \[] -> 2
1403 Just k = f y
1404 </programlisting>
1405
1406 </para>
1407 </listitem>
1408 </varlistentry>
1409
1410 <varlistentry>
1411 <term><option>-fwarn-incomplete-record-updates</option>:</term>
1412 <listitem>
1413 <indexterm><primary><option>-fwarn-incomplete-record-updates</option></primary></indexterm>
1414 <indexterm><primary>incomplete record updates, warning</primary></indexterm>
1415 <indexterm><primary>record updates, incomplete</primary></indexterm>
1416
1417 <para>The function
1418 <function>f</function> below will fail when applied to
1419 <literal>Bar</literal>, so the compiler will emit a warning about
1420 this when <option>-fwarn-incomplete-record-updates</option> is
1421 enabled.</para>
1422
1423 <programlisting>
1424 data Foo = Foo { x :: Int }
1425 | Bar
1426
1427 f :: Foo -> Foo
1428 f foo = foo { x = 6 }
1429 </programlisting>
1430
1431 <para>This option isn't enabled by default because it can be
1432 very noisy, and it often doesn't indicate a bug in the
1433 program.</para>
1434 </listitem>
1435 </varlistentry>
1436
1437 <varlistentry>
1438 <term>
1439 <option>-fwarn-missing-fields</option>:
1440 <indexterm><primary><option>-fwarn-missing-fields</option></primary></indexterm>
1441 <indexterm><primary>missing fields, warning</primary></indexterm>
1442 <indexterm><primary>fields, missing</primary></indexterm>
1443 </term>
1444 <listitem>
1445
1446 <para>This option is on by default, and warns you whenever
1447 the construction of a labelled field constructor isn't
1448 complete, missing initializers for one or more fields. While
1449 not an error (the missing fields are initialised with
1450 bottoms), it is often an indication of a programmer error.</para>
1451 </listitem>
1452 </varlistentry>
1453
1454 <varlistentry>
1455 <term>
1456 <option>-fwarn-missing-import-lists</option>:
1457 <indexterm><primary><option>-fwarn-import-lists</option></primary></indexterm>
1458 <indexterm><primary>missing import lists, warning</primary></indexterm>
1459 <indexterm><primary>import lists, missing</primary></indexterm>
1460 </term>
1461 <listitem>
1462
1463 <para>This flag warns if you use an unqualified
1464 <literal>import</literal> declaration
1465 that does not explicitly list the entities brought into scope. For
1466 example
1467 </para>
1468
1469 <programlisting>
1470 module M where
1471 import X( f )
1472 import Y
1473 import qualified Z
1474 p x = f x x
1475 </programlisting>
1476
1477 <para>
1478 The <option>-fwarn-import-lists</option> flag will warn about the import
1479 of <literal>Y</literal> but not <literal>X</literal>
1480 If module <literal>Y</literal> is later changed to export (say) <literal>f</literal>,
1481 then the reference to <literal>f</literal> in <literal>M</literal> will become
1482 ambiguous. No warning is produced for the import of <literal>Z</literal>
1483 because extending <literal>Z</literal>'s exports would be unlikely to produce
1484 ambiguity in <literal>M</literal>.
1485 </para>
1486 </listitem>
1487 </varlistentry>
1488
1489 <varlistentry>
1490 <term><option>-fwarn-missing-methods</option>:</term>
1491 <listitem>
1492 <indexterm><primary><option>-fwarn-missing-methods</option></primary></indexterm>
1493 <indexterm><primary>missing methods, warning</primary></indexterm>
1494 <indexterm><primary>methods, missing</primary></indexterm>
1495
1496 <para>This option is on by default, and warns you whenever
1497 an instance declaration is missing one or more methods, and
1498 the corresponding class declaration has no default
1499 declaration for them.</para>
1500 <para>The warning is suppressed if the method name
1501 begins with an underscore. Here's an example where this is useful:
1502 <programlisting>
1503 class C a where
1504 _simpleFn :: a -> String
1505 complexFn :: a -> a -> String
1506 complexFn x y = ... _simpleFn ...
1507 </programlisting>
1508 The idea is that: (a) users of the class will only call <literal>complexFn</literal>;
1509 never <literal>_simpleFn</literal>; and (b)
1510 instance declarations can define either <literal>complexFn</literal> or <literal>_simpleFn</literal>.
1511 </para>
1512 <para>The MINIMAL pragma can be used to change which combination of methods will be required for instances of a particular class. See <xref linkend="minimal-pragma"/>.</para>
1513 </listitem>
1514 </varlistentry>
1515
1516 <varlistentry>
1517 <term><option>-fwarn-missing-signatures</option>:</term>
1518 <listitem>
1519 <indexterm><primary><option>-fwarn-missing-signatures</option></primary></indexterm>
1520 <indexterm><primary>type signatures, missing</primary></indexterm>
1521
1522 <para>If you would like GHC to check that every top-level
1523 function/value has a type signature, use the
1524 <option>-fwarn-missing-signatures</option> option. As part of
1525 the warning GHC also reports the inferred type. The
1526 option is off by default.</para>
1527 </listitem>
1528 </varlistentry>
1529
1530 <varlistentry>
1531 <term><option>-fwarn-missing-local-sigs</option>:</term>
1532 <listitem>
1533 <indexterm><primary><option>-fwarn-missing-local-sigs</option></primary></indexterm>
1534 <indexterm><primary>type signatures, missing</primary></indexterm>
1535
1536 <para>If you use the
1537 <option>-fwarn-missing-local-sigs</option> flag GHC will warn
1538 you about any polymorphic local bindings. As part of
1539 the warning GHC also reports the inferred type. The
1540 option is off by default.</para>
1541 </listitem>
1542 </varlistentry>
1543
1544 <varlistentry>
1545 <term><option>-fwarn-name-shadowing</option>:</term>
1546 <listitem>
1547 <indexterm><primary><option>-fwarn-name-shadowing</option></primary></indexterm>
1548 <indexterm><primary>shadowing, warning</primary></indexterm>
1549
1550 <para>This option causes a warning to be emitted whenever an
1551 inner-scope value has the same name as an outer-scope value,
1552 i.e. the inner value shadows the outer one. This can catch
1553 typographical errors that turn into hard-to-find bugs, e.g.,
1554 in the inadvertent capture of what would be a recursive call in
1555 <literal>f = ... let f = id in ... f ...</literal>.</para>
1556 <para>The warning is suppressed for names beginning with an underscore. For example
1557 <programlisting>
1558 f x = do { _ignore &lt;- this; _ignore &lt;- that; return (the other) }
1559 </programlisting>
1560 </para>
1561 </listitem>
1562 </varlistentry>
1563
1564 <varlistentry>
1565 <term><option>-fwarn-orphans, -fwarn-auto-orphans</option>:</term>
1566 <listitem>
1567 <indexterm><primary><option>-fwarn-orphans</option></primary></indexterm>
1568 <indexterm><primary><option>-fwarn-auto-orphans</option></primary></indexterm>
1569 <indexterm><primary>orphan instances, warning</primary></indexterm>
1570 <indexterm><primary>orphan rules, warning</primary></indexterm>
1571
1572 <para>These flags cause a warning to be emitted whenever the
1573 module contains an "orphan" instance declaration or rewrite rule.
1574 An instance declaration is an orphan if it appears in a module in
1575 which neither the class nor the type being instanced are declared
1576 in the same module. A rule is an orphan if it is a rule for a
1577 function declared in another module. A module containing any
1578 orphans is called an orphan module.</para>
1579 <para>The trouble with orphans is that GHC must pro-actively read the interface
1580 files for all orphan modules, just in case their instances or rules
1581 play a role, whether or not the module's interface would otherwise
1582 be of any use. See <xref linkend="orphan-modules"/> for details.
1583 </para>
1584 <para>The flag <option>-fwarn-orphans</option> warns about user-written
1585 orphan rules or instances. The flag <option>-fwarn-auto-orphans</option>
1586 warns about automatically-generated orphan rules, notably as a result of
1587 specialising functions, for type classes (<literal>Specialise</literal>)
1588 or argument values (<literal>SpecConstr</literal>).</para>
1589 </listitem>
1590 </varlistentry>
1591
1592 <varlistentry>
1593 <term>
1594 <option>-fwarn-overlapping-patterns</option>:
1595 <indexterm><primary><option>-fwarn-overlapping-patterns</option></primary></indexterm>
1596 <indexterm><primary>overlapping patterns, warning</primary></indexterm>
1597 <indexterm><primary>patterns, overlapping</primary></indexterm>
1598 </term>
1599 <listitem>
1600 <para>By default, the compiler will warn you if a set of
1601 patterns are overlapping, e.g.,</para>
1602
1603 <programlisting>
1604 f :: String -&#62; Int
1605 f [] = 0
1606 f (_:xs) = 1
1607 f "2" = 2
1608 </programlisting>
1609
1610 <para>where the last pattern match in <function>f</function>
1611 won't ever be reached, as the second pattern overlaps
1612 it. More often than not, redundant patterns is a programmer
1613 mistake/error, so this option is enabled by default.</para>
1614 </listitem>
1615 </varlistentry>
1616
1617 <varlistentry>
1618 <term><option>-fwarn-tabs</option>:</term>
1619 <listitem>
1620 <indexterm><primary><option>-fwarn-tabs</option></primary></indexterm>
1621 <indexterm><primary>tabs, warning</primary></indexterm>
1622 <para>Have the compiler warn if there are tabs in your source
1623 file.</para>
1624
1625 <para>This warning is off by default.</para>
1626 </listitem>
1627 </varlistentry>
1628
1629 <varlistentry>
1630 <term><option>-fwarn-type-defaults</option>:</term>
1631 <listitem>
1632 <indexterm><primary><option>-fwarn-type-defaults</option></primary></indexterm>
1633 <indexterm><primary>defaulting mechanism, warning</primary></indexterm>
1634 <para>Have the compiler warn/inform you where in your source
1635 the Haskell defaulting mechanism for numeric types kicks
1636 in. This is useful information when converting code from a
1637 context that assumed one default into one with another,
1638 e.g., the &lsquo;default default&rsquo; for Haskell 1.4 caused the
1639 otherwise unconstrained value <constant>1</constant> to be
1640 given the type <literal>Int</literal>, whereas Haskell 98
1641 and later
1642 defaults it to <literal>Integer</literal>. This may lead to
1643 differences in performance and behaviour, hence the
1644 usefulness of being non-silent about this.</para>
1645
1646 <para>This warning is off by default.</para>
1647 </listitem>
1648 </varlistentry>
1649
1650 <varlistentry>
1651 <term><option>-fwarn-monomorphism-restriction</option>:</term>
1652 <listitem>
1653 <indexterm><primary><option>-fwarn-monomorphism-restriction</option></primary></indexterm>
1654 <indexterm><primary>monomorphism restriction, warning</primary></indexterm>
1655 <para>Have the compiler warn/inform you where in your source
1656 the Haskell Monomorphism Restriction is applied. If applied silently
1657 the MR can give rise to unexpected behaviour, so it can be helpful
1658 to have an explicit warning that it is being applied.</para>
1659
1660 <para>This warning is off by default.</para>
1661 </listitem>
1662 </varlistentry>
1663
1664 <varlistentry>
1665 <term><option>-fwarn-unused-binds</option>:</term>
1666 <listitem>
1667 <indexterm><primary><option>-fwarn-unused-binds</option></primary></indexterm>
1668 <indexterm><primary>unused binds, warning</primary></indexterm>
1669 <indexterm><primary>binds, unused</primary></indexterm>
1670 <para>Report any function definitions (and local bindings)
1671 which are unused. For top-level functions, the warning is
1672 only given if the binding is not exported.</para>
1673 <para>A definition is regarded as "used" if (a) it is exported, or (b) it is
1674 mentioned in the right hand side of another definition that is used, or (c) the
1675 function it defines begins with an underscore. The last case provides a
1676 way to suppress unused-binding warnings selectively. </para>
1677 <para> Notice that a variable
1678 is reported as unused even if it appears in the right-hand side of another
1679 unused binding. </para>
1680 </listitem>
1681 </varlistentry>
1682
1683 <varlistentry>
1684 <term><option>-fwarn-unused-imports</option>:</term>
1685 <listitem>
1686 <indexterm><primary><option>-fwarn-unused-imports</option></primary></indexterm>
1687 <indexterm><primary>unused imports, warning</primary></indexterm>
1688 <indexterm><primary>imports, unused</primary></indexterm>
1689
1690 <para>Report any modules that are explicitly imported but
1691 never used. However, the form <literal>import M()</literal> is
1692 never reported as an unused import, because it is a useful idiom
1693 for importing instance declarations, which are anonymous in Haskell.</para>
1694 </listitem>
1695 </varlistentry>
1696
1697 <varlistentry>
1698 <term><option>-fwarn-unused-matches</option>:</term>
1699 <listitem>
1700 <indexterm><primary><option>-fwarn-unused-matches</option></primary></indexterm>
1701 <indexterm><primary>unused matches, warning</primary></indexterm>
1702 <indexterm><primary>matches, unused</primary></indexterm>
1703
1704 <para>Report all unused variables which arise from pattern
1705 matches, including patterns consisting of a single variable.
1706 For instance <literal>f x y = []</literal> would report
1707 <varname>x</varname> and <varname>y</varname> as unused. The
1708 warning is suppressed if the variable name begins with an underscore, thus:
1709 <programlisting>
1710 f _x = True
1711 </programlisting>
1712 </para>
1713 </listitem>
1714 </varlistentry>
1715
1716 <varlistentry>
1717 <term><option>-fwarn-unused-do-bind</option>:</term>
1718 <listitem>
1719 <indexterm><primary><option>-fwarn-unused-do-bind</option></primary></indexterm>
1720 <indexterm><primary>unused do binding, warning</primary></indexterm>
1721 <indexterm><primary>do binding, unused</primary></indexterm>
1722
1723 <para>Report expressions occurring in <literal>do</literal> and <literal>mdo</literal> blocks
1724 that appear to silently throw information away.
1725 For instance <literal>do { mapM popInt xs ; return 10 }</literal> would report
1726 the first statement in the <literal>do</literal> block as suspicious,
1727 as it has the type <literal>StackM [Int]</literal> and not <literal>StackM ()</literal>, but that
1728 <literal>[Int]</literal> value is not bound to anything. The warning is suppressed by
1729 explicitly mentioning in the source code that your program is throwing something away:
1730 <programlisting>
1731 do { _ &lt;- mapM popInt xs ; return 10 }
1732 </programlisting>
1733 Of course, in this particular situation you can do even better:
1734 <programlisting>
1735 do { mapM_ popInt xs ; return 10 }
1736 </programlisting>
1737 </para>
1738 </listitem>
1739 </varlistentry>
1740
1741 <varlistentry>
1742 <term><option>-fwarn-wrong-do-bind</option>:</term>
1743 <listitem>
1744 <indexterm><primary><option>-fwarn-wrong-do-bind</option></primary></indexterm>
1745 <indexterm><primary>apparently erroneous do binding, warning</primary></indexterm>
1746 <indexterm><primary>do binding, apparently erroneous</primary></indexterm>
1747
1748 <para>Report expressions occurring in <literal>do</literal> and <literal>mdo</literal> blocks
1749 that appear to lack a binding.
1750 For instance <literal>do { return (popInt 10) ; return 10 }</literal> would report
1751 the first statement in the <literal>do</literal> block as suspicious,
1752 as it has the type <literal>StackM (StackM Int)</literal> (which consists of two nested applications
1753 of the same monad constructor), but which is not then &quot;unpacked&quot; by binding the result.
1754 The warning is suppressed by explicitly mentioning in the source code that your program is throwing something away:
1755 <programlisting>
1756 do { _ &lt;- return (popInt 10) ; return 10 }
1757 </programlisting>
1758 For almost all sensible programs this will indicate a bug, and you probably intended to write:
1759 <programlisting>
1760 do { popInt 10 ; return 10 }
1761 </programlisting>
1762 </para>
1763 </listitem>
1764 </varlistentry>
1765
1766 </variablelist>
1767
1768 <para>If you're feeling really paranoid, the
1769 <option>-dcore-lint</option>
1770 option<indexterm><primary><option>-dcore-lint</option></primary></indexterm>
1771 is a good choice. It turns on heavyweight intra-pass
1772 sanity-checking within GHC. (It checks GHC's sanity, not
1773 yours.)</para>
1774
1775 </sect1>
1776
1777 &packages;
1778
1779 <sect1 id="options-optimise">
1780 <title>Optimisation (code improvement)</title>
1781
1782 <indexterm><primary>optimisation</primary></indexterm>
1783 <indexterm><primary>improvement, code</primary></indexterm>
1784
1785 <para>The <option>-O*</option> options specify convenient
1786 &ldquo;packages&rdquo; of optimisation flags; the
1787 <option>-f*</option> options described later on specify
1788 <emphasis>individual</emphasis> optimisations to be turned on/off;
1789 the <option>-m*</option> options specify
1790 <emphasis>machine-specific</emphasis> optimisations to be turned
1791 on/off.</para>
1792
1793 <sect2 id="optimise-pkgs">
1794 <title><option>-O*</option>: convenient &ldquo;packages&rdquo; of optimisation flags.</title>
1795
1796 <para>There are <emphasis>many</emphasis> options that affect
1797 the quality of code produced by GHC. Most people only have a
1798 general goal, something like &ldquo;Compile quickly&rdquo; or
1799 &ldquo;Make my program run like greased lightning.&rdquo; The
1800 following &ldquo;packages&rdquo; of optimisations (or lack
1801 thereof) should suffice.</para>
1802
1803 <para>Note that higher optimisation levels cause more
1804 cross-module optimisation to be performed, which can have an
1805 impact on how much of your program needs to be recompiled when
1806 you change something. This is one reason to stick to
1807 no-optimisation when developing code.</para>
1808
1809 <variablelist>
1810
1811 <varlistentry>
1812 <term>
1813 No <option>-O*</option>-type option specified:
1814 <indexterm><primary>-O* not specified</primary></indexterm>
1815 </term>
1816 <listitem>
1817 <para>This is taken to mean: &ldquo;Please compile
1818 quickly; I'm not over-bothered about compiled-code
1819 quality.&rdquo; So, for example: <command>ghc -c
1820 Foo.hs</command></para>
1821 </listitem>
1822 </varlistentry>
1823
1824 <varlistentry>
1825 <term>
1826 <option>-O0</option>:
1827 <indexterm><primary><option>-O0</option></primary></indexterm>
1828 </term>
1829 <listitem>
1830 <para>Means &ldquo;turn off all optimisation&rdquo;,
1831 reverting to the same settings as if no
1832 <option>-O</option> options had been specified. Saying
1833 <option>-O0</option> can be useful if
1834 eg. <command>make</command> has inserted a
1835 <option>-O</option> on the command line already.</para>
1836 </listitem>
1837 </varlistentry>
1838
1839 <varlistentry>
1840 <term>
1841 <option>-O</option> or <option>-O1</option>:
1842 <indexterm><primary>-O option</primary></indexterm>
1843 <indexterm><primary>-O1 option</primary></indexterm>
1844 <indexterm><primary>optimise</primary><secondary>normally</secondary></indexterm>
1845 </term>
1846 <listitem>
1847 <para>Means: &ldquo;Generate good-quality code without
1848 taking too long about it.&rdquo; Thus, for example:
1849 <command>ghc -c -O Main.lhs</command></para>
1850 </listitem>
1851 </varlistentry>
1852
1853 <varlistentry>
1854 <term>
1855 <option>-O2</option>:
1856 <indexterm><primary>-O2 option</primary></indexterm>
1857 <indexterm><primary>optimise</primary><secondary>aggressively</secondary></indexterm>
1858 </term>
1859 <listitem>
1860 <para>Means: &ldquo;Apply every non-dangerous
1861 optimisation, even if it means significantly longer
1862 compile times.&rdquo;</para>
1863
1864 <para>The avoided &ldquo;dangerous&rdquo; optimisations
1865 are those that can make runtime or space
1866 <emphasis>worse</emphasis> if you're unlucky. They are
1867 normally turned on or off individually.</para>
1868
1869 <para>At the moment, <option>-O2</option> is
1870 <emphasis>unlikely</emphasis> to produce better code than
1871 <option>-O</option>.</para>
1872 </listitem>
1873 </varlistentry>
1874 </variablelist>
1875
1876 <para>We don't use a <option>-O*</option> flag for day-to-day
1877 work. We use <option>-O</option> to get respectable speed;
1878 e.g., when we want to measure something. When we want to go for
1879 broke, we tend to use <option>-O2</option> (and we go for
1880 lots of coffee breaks).</para>
1881
1882 <para>The easiest way to see what <option>-O</option> (etc.)
1883 &ldquo;really mean&rdquo; is to run with <option>-v</option>,
1884 then stand back in amazement.</para>
1885 </sect2>
1886
1887 <sect2 id="options-f">
1888 <title><option>-f*</option>: platform-independent flags</title>
1889
1890 <indexterm><primary>-f* options (GHC)</primary></indexterm>
1891 <indexterm><primary>-fno-* options (GHC)</primary></indexterm>
1892
1893 <para>These flags turn on and off individual optimisations.
1894 They are normally set via the <option>-O</option> options
1895 described above, and as such, you shouldn't need to set any of
1896 them explicitly (indeed, doing so could lead to unexpected
1897 results). A flag <option>-fwombat</option> can be negated by
1898 saying <option>-fno-wombat</option>. The flags below are off
1899 by default, except where noted below. See <xref linkend="options-f-compact"/>
1900 for a compact list.
1901 </para>
1902
1903 <variablelist>
1904 <varlistentry>
1905 <term>
1906 <option>-favoid-vect</option>
1907 <indexterm><primary><option></option></primary></indexterm>
1908 </term>
1909 <listitem>
1910 <para>Part of <link linkend="dph">Data Parallel Haskell
1911 (DPH)</link>.</para>
1912
1913 <para><emphasis>Off by default.</emphasis> Enable the
1914 <emphasis>vectorisation</emphasis> avoidance optimisation. This
1915 optimisation only works when used in combination with the
1916 <option>-fvectorise</option> transformation.</para>
1917
1918 <para>While vectorisation of code using DPH is often a big win, it
1919 can also produce worse results for some kinds of code. This
1920 optimisation modifies the vectorisation transformation to try to
1921 determine if a function would be better of unvectorised and if
1922 so, do just that.</para>
1923 </listitem>
1924 </varlistentry>
1925
1926 <varlistentry>
1927 <term>
1928 <option>-fcase-merge</option>
1929 <indexterm><primary><option></option></primary></indexterm>
1930 </term>
1931 <listitem>
1932 <para><emphasis>On by default.</emphasis>
1933 Merge immediately-nested case expressions that scrutinse the same variable. Example
1934 <programlisting>
1935 case x of
1936 Red -> e1
1937 _ -> case x of
1938 Blue -> e2
1939 Green -> e3
1940 ==>
1941 case x of
1942 Red -> e1
1943 Blue -> e2
1944 Green -> e2
1945 </programlisting>
1946 </para>
1947 </listitem>
1948 </varlistentry>
1949
1950 <varlistentry>
1951 <term>
1952 <option>-fcse</option>
1953 <indexterm><primary><option>-fcse</option></primary></indexterm>
1954 </term>
1955 <listitem>
1956 <para><emphasis>On by default.</emphasis>. Enables the common-sub-expression
1957 elimination optimisation.
1958 Switching this off can be useful if you have some <literal>unsafePerformIO</literal>
1959 expressions that you don't want commoned-up.</para>
1960 </listitem>
1961 </varlistentry>
1962
1963 <varlistentry>
1964 <term>
1965 <option>-fdicts-cheap</option>
1966 <indexterm><primary><option></option></primary></indexterm>
1967 </term>
1968 <listitem>
1969 <para>A very experimental flag that makes dictionary-valued
1970 expressions seem cheap to the optimiser.
1971 </para>
1972 </listitem>
1973 </varlistentry>
1974
1975 <varlistentry>
1976 <term>
1977 <option>-fdo-lambda-eta-expansion</option>
1978 <indexterm><primary><option></option></primary></indexterm>
1979 </term>
1980 <listitem>
1981 <para><emphasis>On by default.</emphasis>
1982 Eta-expand let-bindings to increase their arity.
1983 </para>
1984 </listitem>
1985 </varlistentry>
1986
1987 <varlistentry>
1988 <term>
1989 <option>-fdo-eta-reduction</option>
1990 <indexterm><primary><option></option></primary></indexterm>
1991 </term>
1992 <listitem>
1993 <para><emphasis>On by default.</emphasis>
1994 Eta-reduce lambda expressions, if doing so gets rid of a whole
1995 group of lambdas.
1996 </para>
1997 </listitem>
1998 </varlistentry>
1999
2000 <varlistentry>
2001 <term>
2002 <option>-feager-blackholing</option>
2003 <indexterm><primary><option></option></primary></indexterm>
2004 </term>
2005 <listitem>
2006 <para>Usually GHC black-holes a thunk only when it switches
2007 threads. This flag makes it do so as soon as the thunk is
2008 entered. See <ulink url="http://research.microsoft.com/en-us/um/people/simonpj/papers/parallel/">
2009 Haskell on a shared-memory multiprocessor</ulink>.
2010 </para>
2011 </listitem>
2012 </varlistentry>
2013
2014 <varlistentry>
2015 <term>
2016 <option>-fexcess-precision</option>
2017 <indexterm><primary><option>-fexcess-precision</option></primary></indexterm>
2018 </term>
2019 <listitem>
2020 <para>When this option is given, intermediate floating
2021 point values can have a <emphasis>greater</emphasis>
2022 precision/range than the final type. Generally this is a
2023 good thing, but some programs may rely on the exact
2024 precision/range of
2025 <literal>Float</literal>/<literal>Double</literal> values
2026 and should not use this option for their compilation.</para>
2027
2028 <para>
2029 Note that the 32-bit x86 native code generator only
2030 supports excess-precision mode, so neither
2031 <option>-fexcess-precision</option> nor
2032 <option>-fno-excess-precision</option> has any effect.
2033 This is a known bug, see <xref linkend="bugs-ghc" />.
2034 </para>
2035 </listitem>
2036 </varlistentry>
2037
2038 <varlistentry>
2039 <term>
2040 <option>-fexpose-all-unfoldings</option>
2041 <indexterm><primary><option></option></primary></indexterm>
2042 </term>
2043 <listitem>
2044 <para>An experimental flag to expose all unfoldings, even for very
2045 large or recursive functions. This allows for all functions to be
2046 inlined while usually GHC would avoid inlining larger functions.
2047 </para>
2048 </listitem>
2049 </varlistentry>
2050
2051 <varlistentry>
2052 <term>
2053 <option>-ffloat-in</option>
2054 <indexterm><primary><option></option></primary></indexterm>
2055 </term>
2056 <listitem>
2057 <para><emphasis>On by default.</emphasis>
2058 Float let-bindings inwards, nearer their binding site. See
2059 <ulink url="http://research.microsoft.com/en-us/um/people/simonpj/papers/float.ps.gz">
2060 Let-floating: moving bindings to give faster programs (ICFP'96)</ulink>.
2061 </para>
2062
2063 <para>This optimisation moves let bindings closer to their use
2064 site. The benefit here is that this may avoid unnecessary
2065 allocation if the branch the let is now on is never executed. It
2066 also enables other optimisation passes to work more effectively
2067 as they have more information locally.
2068 </para>
2069
2070 <para>This optimisation isn't always beneficial though (so GHC
2071 applies some heuristics to decide when to apply it). The details
2072 get complicated but a simple example is that it is often beneficial
2073 to move let bindings outwards so that multiple let bindings can be
2074 grouped into a larger single let binding, effectively batching
2075 their allocation and helping the garbage collector and allocator.
2076 </para>
2077 </listitem>
2078 </varlistentry>
2079
2080 <varlistentry>
2081 <term>
2082 <option>-ffull-laziness</option>
2083 <indexterm><primary><option>-ffull-laziness</option></primary></indexterm>
2084 </term>
2085 <listitem>
2086 <para><emphasis>On by default.</emphasis>
2087 Run the full laziness optimisation (also known as let-floating),
2088 which floats let-bindings outside enclosing lambdas, in the hope
2089 they will be thereby be computed less often. See
2090 <ulink url="http://research.microsoft.com/en-us/um/people/simonpj/papers/float.ps.gz">Let-floating:
2091 moving bindings to give faster programs (ICFP'96)</ulink>.
2092 Full laziness increases sharing, which can lead to increased memory
2093 residency.
2094 </para>
2095
2096 <para>NOTE: GHC doesn't implement complete full-laziness.
2097 When optimisation in on, and <option>-fno-full-laziness</option>
2098 is not given, some transformations that increase sharing are
2099 performed, such as extracting repeated computations from a loop.
2100 These are the same transformations that a fully lazy
2101 implementation would do, the difference is that GHC doesn't
2102 consistently apply full-laziness, so don't rely on it.
2103 </para>
2104 </listitem>
2105 </varlistentry>
2106
2107 <varlistentry>
2108 <term>
2109 <option>-ffun-to-thunk</option>
2110 <indexterm><primary><option>-fignore-asserts</option></primary></indexterm>
2111 </term>
2112 <listitem>
2113 <para>Worker-wrapper removes unused arguments, but usually we do
2114 not remove them all, lest it turn a function closure into a thunk,
2115 thereby perhaps creating a space leak and/or disrupting inlining.
2116 This flag allows worker/wrapper to remove <emphasis>all</emphasis>
2117 value lambdas. Off by default.
2118 </para>
2119 </listitem>
2120 </varlistentry>
2121
2122 <varlistentry>
2123 <term>
2124 <option>-fignore-asserts</option>
2125 <indexterm><primary><option>-fignore-asserts</option></primary></indexterm>
2126 </term>
2127 <listitem>
2128 <para>Causes GHC to ignore uses of the function
2129 <literal>Exception.assert</literal> in source code (in
2130 other words, rewriting <literal>Exception.assert p
2131 e</literal> to <literal>e</literal> (see <xref
2132 linkend="assertions"/>). This flag is turned on by
2133 <option>-O</option>.
2134 </para>
2135 </listitem>
2136 </varlistentry>
2137
2138 <varlistentry>
2139 <term>
2140 <option>-fignore-interface-pragmas</option>
2141 <indexterm><primary><option>-fignore-interface-pragmas</option></primary></indexterm>
2142 </term>
2143 <listitem>
2144 <para>Tells GHC to ignore all inessential information when reading interface files.
2145 That is, even if <filename>M.hi</filename> contains unfolding or strictness information
2146 for a function, GHC will ignore that information.</para>
2147 </listitem>
2148 </varlistentry>
2149
2150 <varlistentry>
2151 <term>
2152 <option>-flate-dmd-anal</option>
2153 <indexterm><primary><option>-flate-dmd-anal</option></primary></indexterm>
2154 </term>
2155 <listitem>
2156 <para><emphasis>Off by default.</emphasis>Run demand analysis
2157 again, at the end of the simplification pipeline. We found some opportunities
2158 for discovering strictness that were not visible earlier; and optimisations like
2159 <literal>SpecConstr</literal> can create functions with unused arguments which
2160 are eliminated by late demand analysis. Improvements are modest, but so is the
2161 cost. See notes on the <ulink href="http://ghc.haskell.org/trac/ghc/wiki/LateDmd">Trac wiki page</ulink>.
2162 </para>
2163 </listitem>
2164 </varlistentry>
2165
2166 <varlistentry>
2167 <term>
2168 <option>-fliberate-case</option>
2169 <indexterm><primary><option>-fliberate-case</option></primary></indexterm>
2170 </term>
2171 <listitem>
2172 <para><emphasis>Off by default, but enabled by -O2.</emphasis>
2173 Turn on the liberate-case transformation. This unrolls recursive
2174 function once in its own RHS, to avoid repeated case analysis of
2175 free variables. It's a bit like the call-pattern specialiser
2176 (<option>-fspec-constr</option>) but for free variables rather than
2177 arguments.
2178 </para>
2179 </listitem>
2180 </varlistentry>
2181
2182 <varlistentry>
2183 <term>
2184 <option>-fliberate-case-threshold=N</option>
2185 <indexterm><primary><option>-fliberate-case-threshold</option></primary></indexterm>
2186 </term>
2187 <listitem>
2188 <para>Set the size threshold for the liberate-case transformation.
2189 </para>
2190 </listitem>
2191 </varlistentry>
2192
2193 <varlistentry>
2194 <term>
2195 <option>-fmax-relevant-bindings=N</option>
2196 <indexterm><primary><option>-fmax-relevant-bindings</option></primary></indexterm>
2197 </term>
2198 <listitem>
2199 <para>The type checker sometimes displays a fragment of the type environment
2200 in error messages, but only up to some maximum number, set by this flag.
2201 The default is 6. Turning it off with <option>-fno-max-relevant-bindings</option>
2202 gives an unlimited number. Syntactically top-level bindings are also
2203 usually excluded (since they may be numerous), but
2204 <option>-fno-max-relevant-bindings</option> includes them too.
2205 </para>
2206 </listitem>
2207 </varlistentry>
2208
2209 <varlistentry>
2210 <term>
2211 <option>-fno-state-hack</option>
2212 <indexterm><primary><option>-fno-state-hack</option></primary></indexterm>
2213 </term>
2214 <listitem>
2215 <para>Turn off the "state hack" whereby any lambda with a
2216 <literal>State#</literal> token as argument is considered to be
2217 single-entry, hence it is considered OK to inline things inside
2218 it. This can improve performance of IO and ST monad code, but it
2219 runs the risk of reducing sharing.
2220 </para>
2221 </listitem>
2222 </varlistentry>
2223
2224 <varlistentry>
2225 <term>
2226 <option>-fomit-interface-pragmas</option>
2227 <indexterm><primary><option>-fomit-interface-pragmas</option></primary></indexterm>
2228 </term>
2229 <listitem>
2230 <para>Tells GHC to omit all inessential information from the
2231 interface file generated for the module being compiled (say M).
2232 This means that a module importing M will see only the
2233 <emphasis>types</emphasis> of the functions that M exports, but
2234 not their unfoldings, strictness info, etc. Hence, for example,
2235 no function exported by M will be inlined into an importing module.
2236 The benefit is that modules that import M will need to be
2237 recompiled less often (only when M's exports change their type, not
2238 when they change their implementation).</para>
2239 </listitem>
2240 </varlistentry>
2241
2242 <varlistentry>
2243 <term>
2244 <option>-fomit-yields</option>
2245 <indexterm><primary><option>-fomit-yields</option></primary></indexterm>
2246 </term>
2247 <listitem>
2248 <para><emphasis>On by default.</emphasis> Tells GHC to omit
2249 heap checks when no allocation is being performed. While this improves
2250 binary sizes by about 5%, it also means that threads run in
2251 tight non-allocating loops will not get preempted in a timely
2252 fashion. If it is important to always be able to interrupt such
2253 threads, you should turn this optimization off. Consider also
2254 recompiling all libraries with this optimization turned off, if you
2255 need to guarantee interruptibility.
2256 </para>
2257 </listitem>
2258 </varlistentry>
2259
2260 <varlistentry>
2261 <term>
2262 <option>-fpedantic-bottoms</option>
2263 <indexterm><primary><option>-fpedantic-bottoms</option></primary></indexterm>
2264 </term>
2265 <listitem>
2266 <para>Make GHC be more precise about its treatment of bottom (but see also
2267 <option>-fno-state-hack</option>). In particular, stop GHC
2268 eta-expanding through a case expression, which is good for
2269 performance, but bad if you are using <literal>seq</literal> on
2270 partial applications.
2271 </para>
2272 </listitem>
2273 </varlistentry>
2274
2275 <varlistentry>
2276 <term>
2277 <option>-fregs-graph</option>
2278 <indexterm><primary><option></option></primary></indexterm>
2279 </term>
2280 <listitem>
2281 <para><emphasis>Off by default, but enabled by -O2. Only applies in
2282 combination with the native code generator.</emphasis>
2283 Use the graph colouring register allocator for register allocation
2284 in the native code generator. By default, GHC uses a simpler,
2285 faster linear register allocator. The downside being that the
2286 linear register allocator usually generates worse code.
2287 </para>
2288 </listitem>
2289 </varlistentry>
2290
2291 <varlistentry>
2292 <term>
2293 <option>-fregs-iterative</option>
2294 <indexterm><primary><option></option></primary></indexterm>
2295 </term>
2296 <listitem>
2297 <para><emphasis>Off by default, only applies in combination with
2298 the native code generator.</emphasis>
2299 Use the iterative coalescing graph colouring register allocator for
2300 register allocation in the native code generator. This is the same
2301 register allocator as the <option>-freg-graph</option> one but also
2302 enables iterative coalescing during register allocation.
2303 </para>
2304 </listitem>
2305 </varlistentry>
2306
2307 <varlistentry>
2308 <term>
2309 <option>-fsimpl-tick-factor=<replaceable>n</replaceable></option>
2310 <indexterm><primary><option>-fsimpl-tick-factor</option></primary></indexterm>
2311 </term>
2312 <listitem>
2313 <para>GHC's optimiser can diverge if you write rewrite rules (
2314 <xref linkend="rewrite-rules"/>) that don't terminate, or (less
2315 satisfactorily) if you code up recursion through data types
2316 (<xref linkend="bugs-ghc"/>). To avoid making the compiler fall
2317 into an infinite loop, the optimiser carries a "tick count" and
2318 stops inlining and applying rewrite rules when this count is
2319 exceeded. The limit is set as a multiple of the program size, so
2320 bigger programs get more ticks. The
2321 <option>-fsimpl-tick-factor</option> flag lets you change the
2322 multiplier. The default is 100; numbers larger than 100 give more
2323 ticks, and numbers smaller than 100 give fewer.
2324 </para>
2325
2326 <para>If the tick-count expires, GHC summarises what simplifier
2327 steps it has done; you can use
2328 <option>-fddump-simpl-stats</option> to generate a much more
2329 detailed list. Usually that identifies the loop quite
2330 accurately, because some numbers are very large.
2331 </para>
2332 </listitem>
2333 </varlistentry>
2334
2335 <varlistentry>
2336 <term>
2337 <option>-funfolding-creation-threshold=<replaceable>n</replaceable></option>:
2338 <indexterm><primary><option>-funfolding-creation-threshold</option></primary></indexterm>
2339 <indexterm><primary>inlining, controlling</primary></indexterm>
2340 <indexterm><primary>unfolding, controlling</primary></indexterm>
2341 </term>
2342 <listitem>
2343 <para>(Default: 45) Governs the maximum size that GHC will allow a
2344 function unfolding to be. (An unfolding has a &ldquo;size&rdquo;
2345 that reflects the cost in terms of &ldquo;code bloat&rdquo; of
2346 expanding (aka inlining) that unfolding at a call site. A bigger
2347 function would be assigned a bigger cost.)
2348 </para>
2349
2350 <para>Consequences: (a) nothing larger than this will be inlined
2351 (unless it has an INLINE pragma); (b) nothing larger than this
2352 will be spewed into an interface file.
2353 </para>
2354
2355 <para>Increasing this figure is more likely to result in longer
2356 compile times than faster code. The
2357 <option>-funfolding-use-threshold</option> is more useful.
2358 </para>
2359 </listitem>
2360 </varlistentry>
2361
2362 <varlistentry>
2363 <term>
2364 <option>-funfolding-use-threshold=<replaceable>n</replaceable></option>
2365 <indexterm><primary><option>-funfolding-use-threshold</option></primary></indexterm>
2366 <indexterm><primary>inlining, controlling</primary></indexterm>
2367 <indexterm><primary>unfolding, controlling</primary></indexterm>
2368 </term>
2369 <listitem>
2370 <para>(Default: 8) This is the magic cut-off figure for unfolding
2371 (aka inlining): below this size, a function definition will be
2372 unfolded at the call-site, any bigger and it won't. The size
2373 computed for a function depends on two things: the actual size of
2374 the expression minus any discounts that
2375 apply (see <option>-funfolding-con-discount</option>).
2376 </para>
2377
2378 <para>The difference between this and
2379 <option>-funfolding-creation-threshold</option> is that this one
2380 determines if a function definition will be inlined <emphasis>at
2381 a call site</emphasis>. The other option determines if a
2382 function definition will be kept around at all for potential
2383 inlining.
2384 </para>
2385 </listitem>
2386 </varlistentry>
2387
2388 <varlistentry>
2389 <term>
2390 <option>-fvectorise</option>
2391 <indexterm><primary><option></option></primary></indexterm>
2392 </term>
2393 <listitem>
2394 <para>Part of <link linkend="dph">Data Parallel Haskell
2395 (DPH)</link>.</para>
2396
2397 <para><emphasis>Off by default.</emphasis> Enable the
2398 <emphasis>vectorisation</emphasis> optimisation transformation. This
2399 optimisation transforms the nested data parallelism code of programs
2400 using DPH into flat data parallelism. Flat data parallel programs
2401 should have better load balancing, enable SIMD parallelism and
2402 friendlier cache behaviour.</para>
2403 </listitem>
2404 </varlistentry>
2405
2406 <varlistentry>
2407 <term>
2408 <option>-fspec-constr</option>
2409 <indexterm><primary><option>-fspec-constr</option></primary></indexterm>
2410 </term>
2411 <listitem>
2412 <para><emphasis>Off by default, but enabled by -O2.</emphasis>
2413 Turn on call-pattern specialisation; see
2414 <ulink url="http://research.microsoft.com/en-us/um/people/simonpj/papers/spec-constr/index.htm">
2415 Call-pattern specialisation for Haskell programs</ulink>.
2416 </para>
2417
2418 <para>This optimisation specializes recursive functions according to
2419 their argument "shapes". This is best explained by example so
2420 consider:
2421 <programlisting>
2422 last :: [a] -> a
2423 last [] = error "last"
2424 last (x : []) = x
2425 last (x : xs) = last xs
2426 </programlisting>
2427 In this code, once we pass the initial check for an empty list we
2428 know that in the recursive case this pattern match is redundant. As
2429 such <option>-fspec-constr</option> will transform the above code
2430 to:
2431 <programlisting>
2432 last :: [a] -> a
2433 last [] = error "last"
2434 last (x : xs) = last' x xs
2435 where
2436 last' x [] = x
2437 last' x (y : ys) = last' y ys
2438 </programlisting>
2439 </para>
2440
2441 <para>As well avoid unnecessary pattern matching it also helps avoid
2442 unnecessary allocation. This applies when a argument is strict in
2443 the recursive call to itself but not on the initial entry. As
2444 strict recursive branch of the function is created similar to the
2445 above example.
2446 </para>
2447 </listitem>
2448 </varlistentry>
2449
2450 <varlistentry>
2451 <term>
2452 <option>-fspecialise</option>
2453 <indexterm><primary><option>-fspecialise</option></primary></indexterm>
2454 </term>
2455 <listitem>
2456 <para><emphasis>On by default.</emphasis>
2457 Specialise each type-class-overloaded function defined in this
2458 module for the types at which it is called in this module. Also
2459 specialise imported functions that have an INLINABLE pragma
2460 (<xref linkend="inlinable-pragma"/>) for the types at which they
2461 are called in this module.
2462 </para>
2463 </listitem>
2464 </varlistentry>
2465
2466 <varlistentry>
2467 <term>
2468 <option>-fstatic-argument-transformation</option>
2469 <indexterm><primary><option>-fstatic-argument-transformation</option></primary></indexterm>
2470 </term>
2471 <listitem>
2472 <para>Turn on the static argument transformation, which turns a
2473 recursive function into a non-recursive one with a local
2474 recursive loop. See Chapter 7 of
2475 <ulink url="http://research.microsoft.com/en-us/um/people/simonpj/papers/santos-thesis.ps.gz">
2476 Andre Santos's PhD thesis</ulink>
2477 </para>
2478 </listitem>
2479 </varlistentry>
2480
2481 <varlistentry>
2482 <term>
2483 <option>-fstrictness</option>
2484 <indexterm><primary><option></option></primary></indexterm>
2485 </term>
2486 <listitem>
2487 <para> <emphasis>On by default.</emphasis>.
2488 Switch on the strictness analyser. There is a very old paper about GHC's
2489 strictness analyser, <ulink url="http://research.microsoft.com/en-us/um/people/simonpj/papers/simple-strictnes-analyser.ps.gz">
2490 Measuring the effectiveness of a simple strictness analyser</ulink>,
2491 but the current one is quite a bit different.
2492 </para>
2493
2494 <para>The strictness analyser figures out when arguments and
2495 variables in a function can be treated 'strictly' (that is they
2496 are always evaluated in the function at some point). This allow
2497 GHC to apply certain optimisations such as unboxing that
2498 otherwise don't apply as they change the semantics of the program
2499 when applied to lazy arguments.
2500 </para>
2501 </listitem>
2502 </varlistentry>
2503
2504 <varlistentry>
2505 <term>
2506 <option>-funbox-strict-fields</option>:
2507 <indexterm><primary><option>-funbox-strict-fields</option></primary></indexterm>
2508 <indexterm><primary>strict constructor fields</primary></indexterm>
2509 <indexterm><primary>constructor fields, strict</primary></indexterm>
2510 </term>
2511 <listitem>
2512 <para>This option causes all constructor fields which are marked
2513 strict (i.e. &ldquo;!&rdquo;) to be unpacked if possible. It is
2514 equivalent to adding an <literal>UNPACK</literal> pragma to every
2515 strict constructor field (see <xref linkend="unpack-pragma"/>).
2516 </para>
2517
2518 <para>This option is a bit of a sledgehammer: it might sometimes
2519 make things worse. Selectively unboxing fields by using
2520 <literal>UNPACK</literal> pragmas might be better. An alternative
2521 is to use <option>-funbox-strict-fields</option> to turn on
2522 unboxing by default but disable it for certain constructor
2523 fields using the <literal>NOUNPACK</literal> pragma (see
2524 <xref linkend="nounpack-pragma"/>).</para>
2525 </listitem>
2526 </varlistentry>
2527
2528 <varlistentry>
2529 <term>
2530 <option>-funbox-small-strict-fields</option>:
2531 <indexterm><primary><option>-funbox-small-strict-fields</option></primary></indexterm>
2532 <indexterm><primary>strict constructor fields</primary></indexterm>
2533 <indexterm><primary>constructor fields, strict</primary></indexterm>
2534 </term>
2535 <listitem>
2536 <para><emphasis>On by default.</emphasis>. This option
2537 causes all constructor fields which are marked strict
2538 (i.e. &ldquo;!&rdquo;) and which representation is smaller
2539 or equal to the size of a pointer to be unpacked, if
2540 possible. It is equivalent to adding an
2541 <literal>UNPACK</literal> pragma (see <xref
2542 linkend="unpack-pragma"/>) to every strict constructor
2543 field that fulfils the size restriction.
2544 </para>
2545
2546 <para>For example, the constructor fields in the following
2547 data types
2548 <programlisting>
2549 data A = A !Int
2550 data B = B !A
2551 newtype C = C B
2552 data D = D !C
2553 </programlisting>
2554 would all be represented by a single
2555 <literal>Int#</literal> (see <xref linkend="primitives"/>)
2556 value with
2557 <option>-funbox-small-strict-fields</option> enabled.
2558 </para>
2559
2560 <para>This option is less of a sledgehammer than
2561 <option>-funbox-strict-fields</option>: it should rarely make things
2562 worse. If you use <option>-funbox-small-strict-fields</option>
2563 to turn on unboxing by default you can disable it for certain
2564 constructor fields using the <literal>NOUNPACK</literal> pragma (see
2565 <xref linkend="nounpack-pragma"/>).</para>
2566
2567 <para>
2568 Note that for consistency <literal>Double</literal>,
2569 <literal>Word64</literal>, and <literal>Int64</literal> constructor
2570 fields are unpacked on 32-bit platforms, even though they are
2571 technically larger than a pointer on those platforms.
2572 </para>
2573 </listitem>
2574 </varlistentry>
2575
2576 </variablelist>
2577
2578 </sect2>
2579
2580 </sect1>
2581
2582 &code-gens;
2583
2584 &phases;
2585
2586 &shared_libs;
2587
2588 <sect1 id="using-concurrent">
2589 <title>Using Concurrent Haskell</title>
2590 <indexterm><primary>Concurrent Haskell</primary><secondary>using</secondary></indexterm>
2591
2592 <para>GHC supports Concurrent Haskell by default, without requiring a
2593 special option or libraries compiled in a certain way. To get access to
2594 the support libraries for Concurrent Haskell, just import
2595 <ulink
2596 url="&libraryBaseLocation;/Control-Concurrent.html"><literal>Control.Concurrent</literal></ulink>. More information on Concurrent Haskell is provided in the documentation for that module.</para>
2597
2598 <para>
2599 Optionally, the program may be linked with
2600 the <option>-threaded</option> option (see
2601 <xref linkend="options-linker" />. This provides two benefits:
2602
2603 <itemizedlist>
2604 <listitem>
2605 <para>It enables the <option>-N</option><indexterm><primary><option>-N<replaceable>x</replaceable></option></primary><secondary>RTS option</secondary></indexterm> RTS option to be
2606 used, which allows threads to run in
2607 parallel<indexterm><primary>parallelism</primary></indexterm>
2608 on a
2609 multiprocessor<indexterm><primary>multiprocessor</primary></indexterm><indexterm><primary>SMP</primary></indexterm>
2610 or
2611 multicore<indexterm><primary>multicore</primary></indexterm>
2612 machine. See <xref linkend="using-smp" />.</para>
2613 </listitem>
2614 <listitem>
2615 <para>If a thread makes a foreign call (and the call is
2616 not marked <literal>unsafe</literal>), then other
2617 Haskell threads in the program will continue to run
2618 while the foreign call is in progress.
2619 Additionally, <literal>foreign export</literal>ed
2620 Haskell functions may be called from multiple OS
2621 threads simultaneously. See
2622 <xref linkend="ffi-threads" />.</para>
2623 </listitem>
2624 </itemizedlist>
2625 </para>
2626
2627 <para>The following RTS option(s) affect the behaviour of Concurrent
2628 Haskell programs:<indexterm><primary>RTS options, concurrent</primary></indexterm></para>
2629
2630 <variablelist>
2631 <varlistentry>
2632 <term><option>-C<replaceable>s</replaceable></option></term>
2633 <listitem>
2634 <para><indexterm><primary><option>-C<replaceable>s</replaceable></option></primary><secondary>RTS option</secondary></indexterm>
2635 Sets the context switch interval to <replaceable>s</replaceable>
2636 seconds. A context switch will occur at the next heap block
2637 allocation after the timer expires (a heap block allocation occurs
2638 every 4k of allocation). With <option>-C0</option> or
2639 <option>-C</option>, context switches will occur as often as
2640 possible (at every heap block allocation). By default, context
2641 switches occur every 20ms.</para>
2642 </listitem>
2643 </varlistentry>
2644 </variablelist>
2645 </sect1>
2646
2647 <sect1 id="using-smp">
2648 <title>Using SMP parallelism</title>
2649 <indexterm><primary>parallelism</primary>
2650 </indexterm>
2651 <indexterm><primary>SMP</primary>
2652 </indexterm>
2653
2654 <para>GHC supports running Haskell programs in parallel on an SMP
2655 (symmetric multiprocessor).</para>
2656
2657 <para>There's a fine distinction between
2658 <emphasis>concurrency</emphasis> and <emphasis>parallelism</emphasis>:
2659 parallelism is all about making your program run
2660 <emphasis>faster</emphasis> by making use of multiple processors
2661 simultaneously. Concurrency, on the other hand, is a means of
2662 abstraction: it is a convenient way to structure a program that must
2663 respond to multiple asynchronous events.</para>
2664
2665 <para>However, the two terms are certainly related. By making use of
2666 multiple CPUs it is possible to run concurrent threads in parallel,
2667 and this is exactly what GHC's SMP parallelism support does. But it
2668 is also possible to obtain performance improvements with parallelism
2669 on programs that do not use concurrency. This section describes how to
2670 use GHC to compile and run parallel programs, in <xref
2671 linkend="lang-parallel" /> we describe the language features that affect
2672 parallelism.</para>
2673
2674 <sect2 id="parallel-compile-options">
2675 <title>Compile-time options for SMP parallelism</title>
2676
2677 <para>In order to make use of multiple CPUs, your program must be
2678 linked with the <option>-threaded</option> option (see <xref
2679 linkend="options-linker" />). Additionally, the following
2680 compiler options affect parallelism:</para>
2681
2682 <variablelist>
2683 <varlistentry>
2684 <term><option>-feager-blackholing</option></term>
2685 <indexterm><primary><option>-feager-blackholing</option></primary></indexterm>
2686 <listitem>
2687 <para>
2688 Blackholing is the act of marking a thunk (lazy
2689 computuation) as being under evaluation. It is useful for
2690 three reasons: firstly it lets us detect certain kinds of
2691 infinite loop (the <literal>NonTermination</literal>
2692 exception), secondly it avoids certain kinds of space
2693 leak, and thirdly it avoids repeating a computation in a
2694 parallel program, because we can tell when a computation
2695 is already in progress.</para>
2696
2697 <para>
2698 The option <option>-feager-blackholing</option> causes
2699 each thunk to be blackholed as soon as evaluation begins.
2700 The default is "lazy blackholing", whereby thunks are only
2701 marked as being under evaluation when a thread is paused
2702 for some reason. Lazy blackholing is typically more
2703 efficient (by 1-2&percnt; or so), because most thunks don't
2704 need to be blackholed. However, eager blackholing can
2705 avoid more repeated computation in a parallel program, and
2706 this often turns out to be important for parallelism.
2707 </para>
2708
2709 <para>
2710 We recommend compiling any code that is intended to be run
2711 in parallel with the <option>-feager-blackholing</option>
2712 flag.
2713 </para>
2714 </listitem>
2715 </varlistentry>
2716 </variablelist>
2717 </sect2>
2718
2719 <sect2 id="parallel-options">
2720 <title>RTS options for SMP parallelism</title>
2721
2722 <para>There are two ways to run a program on multiple
2723 processors:
2724 call <literal>Control.Concurrent.setNumCapabilities</literal> from your
2725 program, or use the RTS <option>-N</option> option.</para>
2726
2727 <variablelist>
2728 <varlistentry>
2729 <term><option>-N<optional><replaceable>x</replaceable></optional></option></term>
2730 <listitem>
2731 <para><indexterm><primary><option>-N<replaceable>x</replaceable></option></primary><secondary>RTS option</secondary></indexterm>
2732 Use <replaceable>x</replaceable> simultaneous threads when
2733 running the program. Normally <replaceable>x</replaceable>
2734 should be chosen to match the number of CPU cores on the
2735 machine<footnote><para>Whether hyperthreading cores should be counted or not is an
2736 open question; please feel free to experiment and let us know what
2737 results you find.</para></footnote>. For example,
2738 on a dual-core machine we would probably use
2739 <literal>+RTS -N2 -RTS</literal>.</para>
2740
2741 <para>Omitting <replaceable>x</replaceable>,
2742 i.e. <literal>+RTS -N -RTS</literal>, lets the runtime
2743 choose the value of <replaceable>x</replaceable> itself
2744 based on how many processors are in your machine.</para>
2745
2746 <para>Be careful when using all the processors in your
2747 machine: if some of your processors are in use by other
2748 programs, this can actually harm performance rather than
2749 improve it.</para>
2750
2751 <para>Setting <option>-N</option> also has the effect of
2752 enabling the parallel garbage collector (see
2753 <xref linkend="rts-options-gc" />).</para>
2754
2755 <para>The current value of the <option>-N</option> option
2756 is available to the Haskell program
2757 via <literal>Control.Concurrent.getNumCapabilities</literal>, and
2758 it may be changed while the program is running by
2759 calling <literal>Control.Concurrent.setNumCapabilities</literal>.</para>
2760 </listitem>
2761 </varlistentry>
2762 </variablelist>
2763
2764 <para>The following options affect the way the runtime schedules
2765 threads on CPUs:</para>
2766
2767 <variablelist>
2768 <varlistentry>
2769 <term><option>-qa</option></term>
2770 <indexterm><primary><option>-qa</option></primary><secondary>RTS
2771 option</secondary></indexterm>
2772 <listitem>
2773 <para>Use the OS's affinity facilities to try to pin OS
2774 threads to CPU cores. This is an experimental feature,
2775 and may or may not be useful. Please let us know
2776 whether it helps for you!</para>
2777 </listitem>
2778 </varlistentry>
2779 <varlistentry>
2780 <term><option>-qm</option></term>
2781 <indexterm><primary><option>-qm</option></primary><secondary>RTS
2782 option</secondary></indexterm>
2783 <listitem>
2784 <para>Disable automatic migration for load balancing.
2785 Normally the runtime will automatically try to schedule
2786 threads across the available CPUs to make use of idle
2787 CPUs; this option disables that behaviour. Note that
2788 migration only applies to threads; sparks created
2789 by <literal>par</literal> are load-balanced separately
2790 by work-stealing.</para>
2791
2792 <para>
2793 This option is probably only of use for concurrent
2794 programs that explicitly schedule threads onto CPUs
2795 with <literal>Control.Concurrent.forkOn</literal>.
2796 </para>
2797 </listitem>
2798 </varlistentry>
2799 </variablelist>
2800 </sect2>
2801
2802 <sect2>
2803 <title>Hints for using SMP parallelism</title>
2804
2805 <para>Add the <literal>-s</literal> RTS option when
2806 running the program to see timing stats, which will help to tell you
2807 whether your program got faster by using more CPUs or not. If the user
2808 time is greater than
2809 the elapsed time, then the program used more than one CPU. You should
2810 also run the program without <literal>-N</literal> for
2811 comparison.</para>
2812
2813 <para>The output of <literal>+RTS -s</literal> tells you how
2814 many &ldquo;sparks&rdquo; were created and executed during the
2815 run of the program (see <xref linkend="rts-options-gc" />), which
2816 will give you an idea how well your <literal>par</literal>
2817 annotations are working.</para>
2818
2819 <para>GHC's parallelism support has improved in 6.12.1 as a
2820 result of much experimentation and tuning in the runtime
2821 system. We'd still be interested to hear how well it works
2822 for you, and we're also interested in collecting parallel
2823 programs to add to our benchmarking suite.</para>
2824 </sect2>
2825 </sect1>
2826
2827 <sect1 id="options-platform">
2828 <title>Platform-specific Flags</title>
2829
2830 <indexterm><primary>-m* options</primary></indexterm>
2831 <indexterm><primary>platform-specific options</primary></indexterm>
2832 <indexterm><primary>machine-specific options</primary></indexterm>
2833
2834 <para>Some flags only make sense for particular target
2835 platforms.</para>
2836
2837 <variablelist>
2838
2839 <varlistentry>
2840 <term><option>-msse2</option>:</term>
2841 <listitem>
2842 <para>
2843 (x86 only, added in GHC 7.0.1) Use the SSE2 registers and
2844 instruction set to implement floating point operations when using
2845 the <link linkend="native-code-gen">native code generator</link>.
2846 This gives a substantial performance improvement for floating
2847 point, but the resulting compiled code
2848 will only run on processors that support SSE2 (Intel Pentium 4 and
2849 later, or AMD Athlon 64 and later). The
2850 <link linkend="llvm-code-gen">LLVM backend</link> will also use SSE2
2851 if your processor supports it but detects this automatically so no
2852 flag is required.
2853 </para>
2854 <para>
2855 SSE2 is unconditionally used on x86-64 platforms.
2856 </para>
2857 </listitem>
2858 </varlistentry>
2859
2860 <varlistentry>
2861 <term><option>-msse4.2</option>:</term>
2862 <listitem>
2863 <para>
2864 (x86 only, added in GHC 7.4.1) Use the SSE4.2 instruction set to
2865 implement some floating point and bit operations when using the
2866 <link linkend="native-code-gen">native code generator</link>. The
2867 resulting compiled code will only run on processors that
2868 support SSE4.2 (Intel Core i7 and later). The
2869 <link linkend="llvm-code-gen">LLVM backend</link> will also use
2870 SSE4.2 if your processor supports it but detects this automatically
2871 so no flag is required.
2872 </para>
2873 </listitem>
2874 </varlistentry>
2875
2876 </variablelist>
2877
2878 </sect1>
2879
2880 &runtime;
2881
2882 <sect1 id="ext-core">
2883 <title>Generating and compiling External Core Files</title>
2884
2885 <indexterm><primary>intermediate code generation</primary></indexterm>
2886
2887 <para>GHC can dump its optimized intermediate code (said to be in &ldquo;Core&rdquo; format)
2888 to a file as a side-effect of compilation. Non-GHC back-end tools can read and process Core files; these files have the suffix
2889 <filename>.hcr</filename>. The Core format is described in <ulink url="../../core.pdf">
2890 <citetitle>An External Representation for the GHC Core Language</citetitle></ulink>,
2891 and sample tools
2892 for manipulating Core files (in Haskell) are available in the
2893 <ulink url="http://hackage.haskell.org/package/extcore">extcore package on Hackage</ulink>. Note that the format of <literal>.hcr</literal>
2894 files is <emphasis>different</emphasis> from the Core output format that GHC generates
2895 for debugging purposes (<xref linkend="options-debugging"/>), though the two formats appear somewhat similar.</para>
2896
2897 <para>The Core format natively supports notes which you can add to
2898 your source code using the <literal>CORE</literal> pragma (see <xref
2899 linkend="pragmas"/>).</para>
2900
2901 <variablelist>
2902
2903 <varlistentry>
2904 <term>
2905 <option>-fext-core</option>
2906 <indexterm><primary><option>-fext-core</option></primary></indexterm>
2907 </term>
2908 <listitem>
2909 <para>Generate <literal>.hcr</literal> files.</para>
2910 </listitem>
2911 </varlistentry>
2912
2913 </variablelist>
2914
2915 <para>Currently (as of version 6.8.2), GHC does not have the ability to read in External Core files as source. If you would like GHC to have this ability, please <ulink url="http://hackage.haskell.org/trac/ghc/wiki/MailingListsAndIRC">make your wishes known to the GHC Team</ulink>.</para>
2916
2917 </sect1>
2918
2919 &debug;
2920 &flags;
2921
2922 </chapter>
2923
2924 <!-- Emacs stuff:
2925 ;;; Local Variables: ***
2926 ;;; sgml-parent-document: ("users_guide.xml" "book" "chapter") ***
2927 ;;; End: ***
2928 -->