1 <chapter id="introduction-GHC">
2 <title>Introduction to GHC</title>
4 <para>This is a guide to using the Glasgow Haskell Compiler (GHC):
5 an interactive and batch compilation system for the <ulink
6 url="http://www.haskell.org/">Haskell 98</ulink>
9 <para>GHC has two main components: an interactive Haskell
10 interpreter (also known as GHCi), described in <xref
11 linkend="ghci">, and a batch compiler, described throughout <xref
12 linkend="using-ghc">. In fact, GHC consists of a single program
13 which is just run with different options to provide either the
14 interactive or the batch system.</para>
16 <para>The batch compiler can be used alongside GHCi: compiled
17 modules can be loaded into an interactive session and used in the
18 same way as interpreted code, and in fact when using GHCi most of
19 the library code will be pre-compiled. This means you get the best
20 of both worlds: fast pre-compiled library code, and fast compile
21 turnaround for the parts of your program being actively
24 <para>GHC supports numerous language extensions, including
25 concurrency, a foreign function interface, exceptions, type system
26 extensions such as multi-parameter type classes, local universal and
27 existential quantification, functional dependencies, scoped type
28 variables and explicit unboxed types. These are all described in
29 <xref linkend="ghc-language-features">.</para>
31 <para>GHC has a comprehensive optimiser, so when you want to Really
32 Go For It (and you've got time to spare) GHC can produce pretty fast
33 code. Alternatively, the default option is to compile as fast as
34 possible while not making too much effort to optimise the generated
35 code (although GHC probably isn't what you'd describe as a fast
38 <para>GHC's profiling system supports “cost centre
39 stacks”: a way of seeing the profile of a Haskell program in a
40 call-graph like structure. See <xref linkend="profiling"> for more
43 <para>GHC comes with a large collection of libraries, with
44 everything from parser combinators to networking. The libraries are
45 described in separate documentation.</para>
47 <Sect1 id="mailing-lists-GHC">
48 <Title>Meta-information: Web sites, mailing lists, etc.</Title>
50 <IndexTerm><Primary>mailing lists, Glasgow Haskell</Primary></IndexTerm>
51 <IndexTerm><Primary>Glasgow Haskell mailing lists</Primary></IndexTerm>
53 <para>On the World-Wide Web, there are several URLs of likely
58 <para><ulink URL="http://www.haskell.org/" >Haskell home
63 <para><ulink URL="http://www.haskell.org/ghc/">GHC home
69 URL="http://www.cs.nott.ac.uk/~gmh/faq.html">comp.lang.functional
75 <para>We run the following mailing lists about Glasgow Haskell.
76 We encourage you to join, as you feel is appropriate.</para>
80 <term>glasgow-haskell-users:</term>
82 <para>This list is for GHC users to chat among themselves.
83 If you have a specific question about GHC, please check the
84 FAQ first (<xref linkend="faq">).</para>
88 <term>list email address:</term>
90 <para><email>glasgow-haskell-users@haskell.org</email></para>
95 <term>subscribe at:</term>
98 url="http://www.haskell.org/mailman/listinfo/glasgow-haskell-users"><literal>http://www.haskell.org/mailman/listinfo/glasgow-haskell-users</literal></ulink>.</para>
103 <term>admin email address:</term>
105 <para><email>glasgow-haskell-users-admin@haskell.org</email></para>
110 <term>list archives:</term>
113 url="http://www.haskell.org/pipermail/glasgow-haskell-users/"><literal>http://www.haskell.org/pipermail/glasgow-haskell-users/</literal></ulink></para>
121 <term>glasgow-haskell-bugs:</term>
123 <para>Send bug reports for GHC to this address! The sad and
124 lonely people who subscribe to this list will muse upon
125 what's wrong and what you might do about it.</para>
129 <term>list email address:</term>
131 <para><email>glasgow-haskell-bugs@haskell.org</email></para>
136 <term>subscribe at:</term>
139 url="http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs"><literal>http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs</literal></ulink>.</para>
144 <term>admin email address:</term>
146 <para><email>glasgow-haskell-bugs-admin@haskell.org</email></para>
151 <term>list archives:</term>
154 url="http://www.haskell.org/pipermail/glasgow-haskell-bugs/"><literal>http://www.haskell.org/pipermail/glasgow-haskell-bugs/</literal></ulink></para>
162 <term>cvs-ghc:</term>
164 <para>The hardcore GHC developers hang out here. This list
165 also gets commit message from the CVS repository. There are
166 several other similar lists for other parts of the CVS
167 repository (eg. <literal>cvs-hslibs</literal>,
168 <literal>cvs-happy</literal>, <literal>cvs-hdirect</literal>
173 <term>list email address:</term>
175 <para><email>cvs-ghc@haskell.org</email></para>
180 <term>subscribe at:</term>
183 url="http://www.haskell.org/mailman/listinfo/cvs-ghc"><literal>http://www.haskell.org/mailman/listinfo/cvs-ghc</literal></ulink>.</para>
188 <term>admin email address:</term>
190 <para><email>cvs-ghc-admin@haskell.org</email></para>
195 <term>list archives:</term>
198 url="http://www.haskell.org/pipermail/cvs-ghc/"><literal>http://www.haskell.org/pipermail/cvs-ghc/</literal></ulink></para>
206 <para>There are several other haskell and GHC-related mailing
207 lists served by <literal>www.haskell.org</literal>. Go to <ulink
208 url="http://www.haskell.org/mailman/listinfo/"><literal>http://www.haskell.org/mailman/listinfo/</literal></ulink>
209 for the full list.</para>
211 <para>Some Haskell-related discussion also takes place in the
212 Usenet newsgroup <literal>comp.lang.functional</literal>.</para>
216 <sect1 id="bug-reporting">
217 <title>Reporting bugs in GHC</title>
218 <indexterm><primary>bugs</primary><secondary>reporting</secondary>
220 <indexterm><primary>reporting bugs</primary>
223 <para>Glasgow Haskell is a changing system so there are sure to be
226 <para>To report a bug, either:</para>
230 <para>Go to the <ulink
231 url="http://sourceforge.net/projects/ghc/">SoureForge GHC
232 page</ulink>, go to the <quote>bugs</quote> section, click on
233 <quote>submit</quote>, and enter your bug report. You can also
234 check the outstanding bugs here and search the archives to make
235 sure it hasn't already been reported. Or:</para>
238 <para>Email your bug report to
239 <Email>glasgow-haskell-bugs@haskell.org</Email>. </para>
244 <title>How do I tell if I should report my bug?</title>
246 <para>Take a look at the FAQ (<xref linkend="faq">) and <xref
247 linkend="wrong">, which will give you some guidance as to
248 whether the behaviour you're seeing is really a bug or
251 <para>If it is a bug, then it might have been reported before:
252 try searching the mailing list archives. The archives don't
253 have a built-in search facility, but we find that <ulink
254 url="http://www.google.com/">Google</ulink>'s site search works
256 <quote><literal>site:www.haskell.org</literal></quote> followed
257 by your search term into Google.</para>
259 <para>If in doubt, just report it.</para>
263 <title>What to put in a bug report</title>
264 <indexterm><primary>bug reports</primary><secondary>contents</secondary></indexterm>
266 <para>The name of the bug-reporting game is: facts, facts,
267 facts. Don't omit them because “Oh, they won't be
268 interested…”</para>
272 <para>What kind of machine are you running on, and exactly
273 what version of the operating system are you using?
274 (on a Unix system, <Command>uname -a</Command> or <Command>cat
275 /etc/motd</Command> will show the desired information.)</para>
279 <para> What version of GCC are you using? <Command>gcc
280 -v</Command> will tell you.</para>
284 <para> Run the sequence of compiles/runs that caused the
285 offending behaviour, capturing all the input/output in a
286 “script” (a UNIX command) or in an Emacs shell
287 window. We'd prefer to see the whole thing.</para>
291 <para> Be sure any Haskell compilations are run with a
292 <Option>-v</Option> (verbose) flag, so we can see exactly
293 what was run, what versions of things you have, etc.</para>
297 <para> What is the program behaviour that is wrong, in your
302 <para> If practical, please send enough source files for us
303 to duplicate the problem.</para>
307 <para> If you are a Hero and track down the problem in the
308 compilation-system sources, please send us patches relative
309 to a known released version of GHC, or whole files if you
316 <sect1 id="version-numbering">
317 <title>GHC version numbering policy</title>
318 <indexterm><primary>version, of ghc</primary></indexterm>
320 <para>As of GHC version 6.0, we have adopted the following policy
321 for numbering GHC versions:</para>
325 <term>Stable Releases</term>
327 <para>These are numbered <literal><replaceable>x</replaceable>.<replaceable>y</replaceable>.<replaceable>z</replaceable></literal>, where
328 <replaceable>y</replaceable> is <emphasis>even</emphasis>, and
329 <replaceable>z</replaceable> is the patchlevel number (the trailing
330 <literal>.<replaceable>z</replaceable></literal> can be omitted if <replaceable>z</replaceable>
331 is zero). Patchlevels are bug-fix releases only, and never
332 change the programmer interface to any system-supplied code.
333 However, if you install a new patchlevel over an old one you
334 will need to recompile any code that was compiled against the
335 old libraries.</para>
337 <para>The value of <literal>__GLASGOW_HASKELL__</literal>
338 (see <xref linkend="c-pre-processor">) for a major release
339 <literal><replaceable>x</replaceable>.<replaceable>y</replaceable>.<replaceable>z</replaceable></literal>
340 is the integer <replaceable>xyy</replaceable> (if
341 <replaceable>y</replaceable> is a single digit, then a leading zero
342 is added, so for example in version 6.2 of GHC,
343 <literal>__GLASGOW_HASKELL__==602</literal>).</para>
345 <primary><literal>__GLASGOW_HASKELL__</literal></primary>
351 <term>Snapshots/unstable releases</term>
353 <para>We may make snapshot releases of the current
354 development sources from time to time, and the current
355 sources are always available via the CVS repository (see the
356 <ulink url="http://www.haskell.org/ghc/">GHC web
357 site</ulink> for details).</para>
359 <para>Snapshot releases are named
360 <literal><replaceable>x</replaceable>.<replaceable>y</replaceable>.YYYYMMDD</literal>
361 where <literal>y</literal> is <emphasis>odd</emphasis>, and
362 <literal>YYYYMMDD</literal> is the date of the sources from
363 which the snapshot was built. In theory, you can check out
364 the exact same sources from the CVS repository using this
367 <para>The value of <literal>__GLASGOW_HASKELL__</literal>
368 for a snapshot release is the integer
369 <replaceable>xyy</replaceable>. You should never write any
370 conditional code which tests for this value, however: since
371 interfaces change on a day-to-day basis, and we don't have
372 finer granularity in the values of
373 <literal>__GLASGOW_HASKELL__</literal>, you should only
374 conditionally compile using predicates which test whether
375 <literal>__GLASGOW_HASKELL__</literal> is equal to, later
376 than, or earlier than a given major release.</para>
378 <primary><literal>__GLASGOW_HASKELL__</literal></primary>
384 <para>The version number of your copy of GHC can be found by
385 invoking <literal>ghc</literal> with the
386 <literal>––version</literal> flag (see <xref
387 linkend="options-help">).</para>
396 ;;; Local Variables: ***
398 ;;; sgml-parent-document: ("users_guide.sgml" "book" "chapter") ***