1 <!DOCTYPE Article PUBLIC "-//OASIS//DTD DocBook V3.1//EN">
3 <Article id="building-guide">
7 <Title>Building the Glasgow Functional Programming Tools Suite</Title>
8 <Author><OtherName>The GHC Team</OtherName></Author>
9 <Address><Email>glasgow-haskell-{users,bugs}@haskell.org</Email></Address>
10 <PubDate>November 2001</PubDate>
13 <para>The Glasgow fptools suite is a collection of Functional
14 Programming related tools, including the Glasgow Haskell
15 Compiler (GHC). The source code for the whole suite is kept in
16 a single CVS repository and shares a common build and
17 installation system.</para>
19 <para>This guide is intended for people who want to build or
20 modify programs from the Glasgow <Literal>fptools</Literal>
21 suite (as distinct from those who merely want to
22 <Emphasis>run</Emphasis> them). Installation instructions are
23 now provided in the user guide.</para>
25 <para>The bulk of this guide applies to building on Unix
26 systems; see <XRef LinkEnd="winbuild"> for Windows notes.</para>
32 <sect1 id="sec-getting">
33 <title>Getting the sources</title>
35 <para>You can get your hands on the <literal>fptools</literal>
41 <term><indexterm><primary>Source
42 distributions</primary></indexterm>Source distributions</term>
44 <para>You have a supported platform, but (a) you like
45 the warm fuzzy feeling of compiling things yourself;
46 (b) you want to build something ``extra”—e.g., a
47 set of libraries with strictness-analysis turned off; or
48 (c) you want to hack on GHC yourself.</para>
50 <para>A source distribution contains complete sources for
51 one or more projects in the <literal>fptools</literal>
52 suite. Not only that, but the more awkward
53 machine-independent steps are done for you. For example, if
55 <command>happy</command><indexterm><primary>happy</primary></indexterm>
56 you'll find it convenient that the source distribution
57 contains the result of running <command>happy</command> on
58 the parser specifications. If you don't want to alter the
59 parser then this saves you having to find and install
60 <command>happy</command>. You will still need a working
61 version of GHC (preferably version 4.08+) on your machine in
62 order to compile (most of) the sources, however.</para>
67 <term>The CVS repository.</term>
68 <indexterm><primary>CVS repository</primary>
71 <para>We make releases infrequently. If you want more
72 up-to-the minute (but less tested) source code then you need
73 to get access to our CVS repository.</para>
75 <para>All the <literal>fptools</literal> source code is held
76 in a CVS repository. CVS is a pretty good source-code
77 control system, and best of all it works over the
80 <para>The repository holds source code only. It holds no
81 mechanically generated files at all. So if you check out a
82 source tree from CVS you will need to install every utility
83 so that you can build all the derived files from
86 <para>More information about our CVS repository can be found
87 in <xref linkend="sec-cvs">.</para>
92 <para>If you are going to do any building from sources (either
93 from a source distribution or the CVS repository) then you need to
94 read all of this manual in detail.</para>
98 <title>Using the CVS repository</title>
100 <para>We use <ulink url="http://www.cvshome.org/">CVS</ulink> (Concurrent Version System) to keep track of our
101 sources for various software projects. CVS lets several people
102 work on the same software at the same time, allowing changes to be
103 checked in incrementally. </para>
105 <para>This section is a set of guidelines for how to use our CVS
106 repository, and will probably evolve in time. The main thing to
107 remember is that most mistakes can be undone, but if there's
108 anything you're not sure about feel free to bug the local CVS
109 meister (namely Jeff Lewis
110 <email>jlewis@galconn.com</email>). </para>
112 <sect2 id="cvs-access">
113 <title>Getting access to the CVS Repository</title>
115 <para>You can access the repository in one of two ways:
116 read-only (<xref linkend="cvs-read-only">), or read-write (<xref
117 linkend="cvs-read-write">).</para>
119 <sect3 id="cvs-read-only">
120 <title>Remote Read-only CVS Access</title>
122 <para>Read-only access is available to anyone - there's no
123 need to ask us first. With read-only CVS access you can do
124 anything except commit changes to the repository. You can
125 make changes to your local tree, and still use CVS's merge
126 facility to keep your tree up to date, and you can generate
127 patches using 'cvs diff' in order to send to us for
130 <para>To get read-only access to the repository:</para>
134 <para>Make sure that <application>cvs</application> is
135 installed on your machine.</para>
138 <para>Set your <literal>$CVSROOT</literal> environment variable to
139 <literal>:pserver:anoncvs@glass.cse.ogi.edu:/cvs</literal></para>
142 <para>Run the command</para>
146 <para>The password is simply <literal>cvs</literal>. This
147 sets up a file in your home directory called
148 <literal>.cvspass</literal>, which squirrels away the
149 dummy password, so you only need to do this step once.</para>
153 <para>Now go to <xref linkend="cvs-first">.</para>
158 <sect3 id="cvs-read-write">
159 <title>Remote Read-Write CVS Access</title>
161 <para>We generally supply read-write access to folk doing
162 serious development on some part of the source tree, when
163 going through us would be a pain. If you're developing some
164 feature, or think you have the time and inclination to fix
165 bugs in our sources, feel free to ask for read-write
166 access. There is a certain amount of responsibility that goes
167 with commit privileges; we are more likely to grant you access
168 if you've demonstrated your competence by sending us patches
169 via mail in the past.</para>
171 <para>To get remote read-write CVS access, you need to do the
172 following steps.</para>
176 <para>Make sure that <literal>cvs</literal> and
177 <literal>ssh</literal> are both installed on your
182 <para>Generate a DSA private-key/public-key pair, thus:</para>
186 <para>(<literal>ssh-keygen</literal> comes with
187 <literal>ssh</literal>.) Running <literal>ssh-keygen
188 -d</literal> creates the private and public keys in
189 <literal>$HOME/.ssh/id_dsa</literal> and
190 <literal>$HOME/.ssh/id_dsa.pub</literal> respectively
191 (assuming you accept the standard defaults).</para>
193 <para><literal>ssh-keygen -d</literal> will only work if
194 you have Version 2 <literal>ssh</literal> installed; it
195 will fail harmlessly otherwise. If you only have Version
196 1 you can instead generate an RSA key pair using plain</para>
201 <para>Doing so creates the private and public RSA keys in
202 <literal>$HOME/.ssh/identity</literal> and
203 <literal>$HOME/.ssh/identity.pub</literal>
206 <para>[Deprecated.] Incidentally, you can force a Version
207 2 <literal>ssh</literal> to use the Version 1 protocol by
208 creating <literal>$HOME/config</literal> with the
209 following in it:</para>
217 <para>In both cases, <literal>ssh-keygen</literal> will
218 ask for a <firstterm>passphrase</firstterm>. The
219 passphrase is a password that protects your private key.
220 In response to the 'Enter passphrase' question, you can
224 <para>[Recommended.] Enter a passphrase, which you
225 will quote each time you use CVS.
226 <literal>ssh-agent</literal> makes this entirely
230 <para>[Deprecated.] Just hit return (i.e. use an empty
231 passphrase); then you won't need to quote the
232 passphrase when using CVS. The downside is that
233 anyone who can see into your <literal>.ssh</literal>
234 directory, and thereby get your private key, can mess
235 up the repository. So you must keep the
236 <literal>.ssh</literal> directory with draconian
237 no-access permissions.</para>
243 [Windows users.] The programs <command>ssh-keygen1</command>, <command>ssh1</command>, and <command>cvs</command>,
244 seem to lock up <command>bash</command> entirely if they try to get user input (e.g. if
245 they ask for a password). To solve this, start up <filename>cmd.exe</filename>
246 and run it as follows:
248 c:\tmp> set CYGWIN32=tty
249 c:\tmp> c:/user/local/bin/ssh-keygen1
252 <para>[Windows users.] To protect your
253 <literal>.ssh</literal> from access by anyone else,
254 right-click your <literal>.ssh</literal> directory, and
255 select <literal>Properties</literal>. If you are not on
256 the access control list, add yourself, and give yourself
257 full permissions (the second panel). Remove everyone else
258 from the access control list. Don't leave them there but
259 deny them access, because 'they' may be a list that
264 <para>Send a message to to the CVS repository
265 administrator (currently Jeff Lewis
266 <email>jeff@galconn.com</email>), containing:</para>
269 <para>Your desired user-name.</para>
272 <para>Your <literal>.ssh/id_dsa.pub</literal> (or
273 <literal>.ssh/identity.pub</literal>).</para>
276 <para>He will set up your account.</para>
280 <para>Set the following environment variables:</para>
284 <constant>$HOME</constant>: points to your home directory. This is where CVS
285 will look for its <filename>.cvsrc</filename> file.
291 <constant>$CVS_RSH</constant> to <filename>ssh</filename>
293 <para>[Windows users.] Setting your <literal>CVS_RSH</literal> to
294 <literal>ssh</literal> assumes that your CVS client
295 understands how to execute shell script
296 ("#!"s,really), which is what
297 <literal>ssh</literal> is. This may not be the case on
298 Win32 platforms, so in that case set <literal>CVS_RSH</literal> to
299 <literal>ssh1</literal>.</para>
303 <para><literal>$CVSROOT</literal> to
304 <literal>:ext:</literal><replaceable>your-username</replaceable>
305 <literal>@cvs.haskell.org:/home/cvs/root</literal>
306 where <replaceable>your-username</replaceable> is your user name on
307 <literal>cvs.haskell.org</literal>.
309 <para>The <literal>CVSROOT</literal> environment variable will
310 be recorded in the checked-out tree, so you don't need to set
311 this every time. </para>
317 <constant>$CVSEDITOR</constant>: <filename>bin/gnuclient.exe</filename>
318 if you want to use an Emacs buffer for typing in those long commit messages.
324 <constant>$SHELL</constant>: To use bash as the shell in Emacs, you need to
325 set this to point to <filename>bash.exe</filename>.
336 Put the following in <filename>$HOME/.cvsrc</filename>:
347 These are the default options for the specified CVS commands,
348 and represent better defaults than the usual ones. (Feel
349 free to change them.)
353 [Windows users.] Filenames starting with <filename>.</filename> were illegal in
354 the 8.3 DOS filesystem, but that restriction should have
355 been lifted by now (i.e., you're using VFAT or later filesystems.) If
356 you're still having problems creating it, don't worry; <filename>.cvsrc</filename> is entirely
364 <para>[Experts.] Once your account is set up, you can get
365 access from other machines without bothering Jeff, thus:</para>
368 <para>Generate a public/private key pair on the new
372 <para>Use ssh to log in to
373 <literal>cvs.haskell.org</literal>, from your old
377 <para>Add the public key for the new machine to the file
378 <literal>$HOME/ssh/authorized_keys</literal> on
379 <literal>cvs.haskell.org</literal>.
380 (<literal>authorized_keys2</literal>, I think, for Version
384 <para>Make sure that the new version of
385 <literal>authorized_keys</literal> still has 600 file
394 <sect2 id="cvs-first">
395 <title>Checking Out a Source Tree</title>
399 <para>Make sure you set your <literal>CVSROOT</literal>
400 environment variable according to either of the remote
401 methods above. The Approved Way to check out a source tree
402 is as follows:</para>
405 $ cvs checkout fpconfig
408 <para>At this point you have a new directory called
409 <literal>fptools</literal> which contains the basic stuff
410 for the fptools suite, including the configuration files and
411 some other junk. </para>
413 <para>[Windows users.] The following messages appear to be harmless:
415 setsockopt IPTOS_LOWDELAY: Invalid argument
416 setsockopt IPTOS_THROUGHPUT: Invalid argument
421 <para>You can call the fptools directory whatever you like,
422 CVS won't mind: </para>
425 $ mv fptools <replaceable>directory</replaceable>
428 <para> NB: after you've read the CVS manual you might be
429 tempted to try</para>
431 $ cvs checkout -d <replaceable>directory</replaceable> fpconfig
434 <para>instead of checking out <literal>fpconfig</literal>
435 and then renaming it. But this doesn't work, and will
436 result in checking out the entire repository instead of just
437 the <literal>fpconfig</literal> bit.</para>
439 $ cd <replaceable>directory</replaceable>
440 $ cvs checkout ghc hslibs libraries
443 <para>The second command here checks out the relevant
444 modules you want to work on. For a GHC build, for instance,
445 you need at least the <literal>ghc</literal>,
446 <literal>hslibs</literal> and <literal>libraries</literal>
447 modules (for a full list of the projects available, see
448 <xref linkend="projects">).</para>
453 <sect2 id="cvs-committing">
454 <title>Committing Changes</title>
456 <para>This is only if you have read-write access to the
457 repository. For anoncvs users, CVS will issue a "read-only
458 repository" error if you try to commit changes.</para>
462 <para>Build the software, if necessary. Unless you're just
463 working on documentation, you'll probably want to build the
464 software in order to test any changes you make.</para>
468 <para>Make changes. Preferably small ones first.</para>
472 <para>Test them. You can see exactly what changes you've
473 made by using the <literal>cvs diff</literal> command:</para>
477 <para>lists all the changes (using the
478 <literal>diff</literal> command) in and below the current
479 directory. In emacs, <literal>C-c C-v =</literal> runs
480 <literal>cvs diff</literal> on the current buffer and shows
481 you the results.</para>
485 <para>Before checking in a change, you need to update your
492 <para>This pulls in any changes that other people have made,
493 and merges them with yours. If there are any conflicts, CVS
494 will tell you, and you'll have to resolve them before you
495 can check your changes in. The documentation describes what
496 to do in the event of a conflict.</para>
498 <para>It's not always necessary to do a full cvs update
499 before checking in a change, since CVS will always tell you
500 if you try to check in a file that someone else has changed.
501 However, you should still update at regular intervals to
502 avoid making changes that don't work in conjuction with
503 changes that someone else made. Keeping an eye on what goes
504 by on the mailing list can help here.</para>
508 <para>When you're happy that your change isn't going to
509 break anything, check it in. For a one-file change:</para>
512 $ cvs commit <replaceable>filename</replaceable>
515 <para>CVS will then pop up an editor for you to enter a
516 "commit message", this is just a short description
517 of what your change does, and will be kept in the history of
520 <para>If you're using emacs, simply load up the file into a
521 buffer and type <literal>C-x C-q</literal>, and emacs will
522 prompt for a commit message and then check in the file for
525 <para>For a multiple-file change, things are a bit
526 trickier. There are several ways to do this, but this is the
527 way I find easiest. First type the commit message into a
528 temporary file. Then either</para>
531 $ cvs commit -F <replaceable>commit-message</replaceable> <replaceable>file_1</replaceable> .... <replaceable>file_n</replaceable>
534 <para>or, if nothing else has changed in this part of the
538 $ cvs commit -F <replaceable>commit-message</replaceable> <replaceable>directory</replaceable>
541 <para>where <replaceable>directory</replaceable> is a common
542 parent directory for all your changes, and
543 <replaceable>commit-message</replaceable> is the name of the
544 file containing the commit message.</para>
546 <para>Shortly afterwards, you'll get some mail from the
547 relevant mailing list saying which files changed, and giving
548 the commit message. For a multiple-file change, you should
549 still get only <emphasis>one</emphasis> message.</para>
554 <sect2 id="cvs-update">
555 <title>Updating Your Source Tree</title>
557 <para>It can be tempting to cvs update just part of a source
558 tree to bring in some changes that someone else has made, or
559 before committing your own changes. This is NOT RECOMMENDED!
560 Quite often changes in one part of the tree are dependent on
561 changes in another part of the tree (the
562 <literal>mk/*.mk</literal> files are a good example where
563 problems crop up quite often). Having an inconsistent tree is a
564 major cause of headaches. </para>
566 <para>So, to avoid a lot of hassle, follow this recipe for
567 updating your tree: </para>
571 $ cvs update -Pd 2>&1 | tee log</screen>
573 <para>Look at the log file, and fix any conflicts (denoted by a
574 <quote>C</quote> in the first column). If you're using multiple
575 build trees, then for every build tree you have pointing at this
576 source tree, you need to update the links in case any new files
577 have appeared: </para>
580 $ cd <replaceable>build-tree</replaceable>
581 $ lndir <replaceable>source-tree</replaceable>
584 <para>Some files might have been removed, so you need to remove
585 the links pointing to these non-existent files:</para>
588 $ find . -xtype l -exec rm '{}' \;
591 <para>To be <emphasis>really</emphasis> safe, you should do
594 <screen>$ gmake all</screen>
596 <para>from the top-level, to update the dependencies and build
597 any changed files. </para>
600 <sect2 id="cvs-tags">
601 <title>GHC Tag Policy</title>
603 <para>If you want to check out a particular version of GHC,
604 you'll need to know how we tag versions in the repository. The
605 policy (as of 4.04) is:</para>
609 <para>The tree is branched before every major release. The
610 branch tag is <literal>ghc-x-xx-branch</literal>, where
611 <literal>x-xx</literal> is the version number of the release
612 with the <literal>'.'</literal> replaced by a
613 <literal>'-'</literal>. For example, the 4.04 release lives
614 on <literal>ghc-4-04-branch</literal>.</para>
618 <para>The release itself is tagged with
619 <literal>ghc-x-xx</literal> (on the branch). eg. 4.06 is
620 called <literal>ghc-4-06</literal>.</para>
624 <para>We didn't always follow these guidelines, so to see
625 what tags there are for previous versions, do <literal>cvs
626 log</literal> on a file that's been around for a while (like
627 <literal>fptools/ghc/README</literal>).</para>
631 <para>So, to check out a fresh GHC 4.06 tree you would
635 $ cvs co -r ghc-4-06 fpconfig
637 $ cvs co -r ghc-4-06 ghc hslibs
641 <sect2 id="cvs-hints">
642 <title>General Hints</title>
646 <para>As a general rule: commit changes in small units,
647 preferably addressing one issue or implementing a single
648 feature. Provide a descriptive log message so that the
649 repository records exactly which changes were required to
650 implement a given feature/fix a bug. I've found this
651 <emphasis>very</emphasis> useful in the past for finding out
652 when a particular bug was introduced: you can just wind back
653 the CVS tree until the bug disappears.</para>
657 <para>Keep the sources at least *buildable* at any given
658 time. No doubt bugs will creep in, but it's quite easy to
659 ensure that any change made at least leaves the tree in a
660 buildable state. We do nightly builds of GHC to keep an eye
661 on what things work/don't work each day and how we're doing
662 in relation to previous verions. This idea is truely wrecked
663 if the compiler won't build in the first place!</para>
667 <para>To check out extra bits into an already-checked-out
668 tree, use the following procedure. Suppose you have a
669 checked-out fptools tree containing just ghc, and you want
670 to add nofib to it:</para>
681 $ cvs update -d nofib
684 <para>(the -d flag tells update to create a new
685 directory). If you just want part of the nofib suite, you
690 $ cvs checkout nofib/spectral
693 <para>This works because <literal>nofib</literal> is a
694 module in its own right, and spectral is a subdirectory of
695 the nofib module. The path argument to checkout must always
696 start with a module name. There's no equivalent form of this
697 command using <literal>update</literal>.</para>
703 <sect1 id="projects">
704 <title>What projects are there?</title>
706 <para>The <literal>fptools</literal> suite consists of several
707 <firstterm>projects</firstterm>, most of which can be downloaded,
708 built and installed individually. Each project corresponds to a
709 subdirectory in the source tree, and if checking out from CVS then
710 each project can be checked out individually by sitting in the top
711 level of your source tree and typing <command>cvs checkout
712 <replaceable>project</replaceable></command>.</para>
714 <para>Here is a list of the projects currently available:</para>
718 <term><literal>ghc</literal></term>
719 <indexterm><primary><literal>ghc</literal></primary>
720 <secondary>project</secondary></indexterm>
722 <para>The <ulink url="http://www.haskell.org/ghc/">Glasgow
723 Haskell Compiler</ulink> (minus libraries). Absolutely
724 required for building GHC.</para>
729 <term><literal>glafp-utils</literal></term>
730 <indexterm><primary><literal>glafp-utils</literal></primary><secondary>project</secondary></indexterm>
732 <para>Utility programs, some of which are used by the
733 build/installation system. Required for pretty much
739 <term><literal>green-card</literal></term>
740 <indexterm><primary><literal>green-card</literal></primary><secondary>project</secondary></indexterm>
743 url="http://www.haskell.org/greencard/">Green Card</ulink>
744 system for generating Haskell foreign function
750 <term><literal>haggis</literal></term>
751 <indexterm><primary><literal>haggis</literal></primary><secondary>project</secondary></indexterm>
754 url="http://www.dcs.gla.ac.uk/fp/software/haggis/">Haggis</ulink>
755 Haskell GUI framework.</para>
760 <term><literal>haddock</literal></term>
761 <indexterm><primary><literal>haddock</literal></primary><secondary>project</secondary></indexterm>
764 url="http://www.haskell.org/haddock/">Haddock</ulink>
765 documentation tool.</para>
770 <term><literal>happy</literal></term>
771 <indexterm><primary><literal>happy</literal></primary><secondary>project</secondary></indexterm>
774 url="http://www.haskell.org/happy/">Happy</ulink> Parser
780 <term><literal>hdirect</literal></term>
781 <indexterm><primary><literal>hdirect</literal></primary><secondary>project</secondary></indexterm>
784 url="http://www.haskell.org/hdirect/">H/Direct</ulink>
785 Haskell interoperability tool.</para>
790 <term><literal>hood</literal></term>
791 <indexterm><primary><literal>hood</literal></primary><secondary>project</secondary></indexterm>
793 <para>The <ulink url="http://www.haskell.org/hood/">Haskell
794 Object Observation Debugger</ulink>.</para>
799 <term><literal>hslibs</literal></term>
800 <indexterm><primary><literal>hslibs</literal></primary><secondary>project</secondary></indexterm>
802 <para>Supplemental libraries for GHC
803 (<emphasis>required</emphasis> for building GHC).</para>
808 <term><literal>libraries</literal></term>
809 <indexterm><primary><literal></literal></primary><secondary>project</secondary></indexterm>
811 <para>Hierarchical Haskell library suite
812 (<emphasis>required</emphasis> for building GHC).</para>
817 <term><literal>mhms</literal></term>
818 <indexterm><primary><literal></literal></primary><secondary>project</secondary></indexterm>
820 <para>The Modular Haskell Metric System.</para>
825 <term><literal>nofib</literal></term>
826 <indexterm><primary><literal>nofib</literal></primary><secondary>project</secondary></indexterm>
828 <para>The NoFib suite: A collection of Haskell programs used
829 primarily for benchmarking.</para>
834 <term><literal>testsuite</literal></term>
835 <indexterm><primary><literal>testsuite</literal></primary><secondary>project</secondary></indexterm>
837 <para>A testing framework, including GHC's regression test
843 <para>So, to build GHC you need at least the
844 <literal>ghc</literal>, <literal>libraries</literal> and
845 <literal>hslibs</literal> projects (a GHC source distribution will
846 already include the bits you need).</para>
849 <sect1 id="sec-build-checks">
850 <title>Things to check before you start</title>
852 <para>Here's a list of things to check before you get
858 <indexterm><primary>Disk space needed</primary></indexterm>
859 <para>Disk space needed: from about 100Mb for a basic GHC
860 build, up to probably 500Mb for a GHC build with everything
861 included (libraries built several different ways,
866 <para>Use an appropriate machine / operating system. <xref
867 linkend="sec-port-info"> lists the supported platforms; if
868 yours isn't amongst these then you can try porting GHC (see
869 <xref linkend="sec-porting-ghc">).</para>
873 <para>Be sure that the “pre-supposed” utilities are
874 installed. <Xref LinkEnd="sec-pre-supposed">
879 <para>If you have any problem when building or installing the
880 Glasgow tools, please check the “known pitfalls” (<Xref
881 LinkEnd="sec-build-pitfalls">). Also check the FAQ for the
882 version you're building, which is part of the User's Guide and
883 available on the <ulink URL="http://www.haskell.org/ghc/" >GHC web
886 <indexterm><primary>bugs</primary><secondary>known</secondary></indexterm>
888 <para>If you feel there is still some shortcoming in our
889 procedure or instructions, please report it.</para>
891 <para>For GHC, please see the <ulink
892 url="http://www.haskell.org/ghc/docs/latest/set/bug-reporting.html">bug-reporting
893 section of the GHC Users' Guide</ulink>, to maximise the
894 usefulness of your report.</para>
896 <indexterm><primary>bugs</primary><secondary>seporting</secondary></indexterm>
897 <para>If in doubt, please send a message to
898 <email>glasgow-haskell-bugs@haskell.org</email>.
899 <indexterm><primary>bugs</primary><secondary>mailing
900 list</secondary></indexterm></para>
905 <sect1 id="sec-port-info">
906 <title>What machines the Glasgow tools run on</title>
908 <indexterm><primary>ports</primary><secondary>GHC</secondary></indexterm>
909 <indexterm><primary>GHC</primary><secondary>ports</secondary></indexterm>
910 <indexterm><primary>platforms</primary><secondary>supported</secondary></indexterm>
912 <para>The main question is whether or not the Haskell compiler
913 (GHC) runs on your platform.</para>
915 <para>A “platform” is a
916 architecture/manufacturer/operating-system combination, such as
917 <literal>sparc-sun-solaris2</literal>. Other common ones are
918 <literal>alpha-dec-osf2</literal>,
919 <literal>hppa1.1-hp-hpux9</literal>,
920 <literal>i386-unknown-linux</literal>,
921 <literal>i386-unknown-solaris2</literal>,
922 <literal>i386-unknown-freebsd</literal>,
923 <literal>i386-unknown-cygwin32</literal>,
924 <literal>m68k-sun-sunos4</literal>,
925 <literal>mips-sgi-irix5</literal>,
926 <literal>sparc-sun-sunos4</literal>,
927 <literal>sparc-sun-solaris2</literal>,
928 <literal>powerpc-ibm-aix</literal>.</para>
930 <para>Some libraries may only work on a limited number of
931 platforms; for example, a sockets library is of no use unless the
932 operating system supports the underlying BSDisms.</para>
935 <title>What platforms the Haskell compiler (GHC) runs on</title>
937 <indexterm><primary>fully-supported platforms</primary></indexterm>
938 <indexterm><primary>native-code generator</primary></indexterm>
939 <indexterm><primary>registerised ports</primary></indexterm>
940 <indexterm><primary>unregisterised ports</primary></indexterm>
942 <para>The GHC hierarchy of Porting Goodness: (a) Best is a
943 native-code generator; (b) next best is a
944 “registerised” port; (c) the bare minimum is an
945 “unregisterised” port.
946 (“Unregisterised” is so terrible that we won't say
947 more about it).</para>
949 <para>We use Sparcs running Solaris 2.7 and x86 boxes running
950 FreeBSD and Linux, so those are the best supported platforms,
951 unsurprisingly.</para>
953 <para>Here's everything that's known about GHC ports. We
954 identify platforms by their “canonical”
955 CPU/Manufacturer/OS triple.</para>
959 <term>alpha-dec-{osf,linux,freebsd,openbsd,netbsd}:</term>
960 <indexterm><primary>alpha-dec-osf</primary></indexterm>
961 <indexterm><primary>alpha-dec-linux</primary></indexterm>
962 <indexterm><primary>alpha-dec-freebsd</primary></indexterm>
963 <indexterm><primary>alpha-dec-openbsd</primary></indexterm>
964 <indexterm><primary>alpha-dec-netbsd</primary></indexterm>
967 <para>The OSF port is currently working (as of GHC version
968 5.02.1) and well supported. The native code generator is
969 currently non-working. Other operating systems will
970 require some minor porting.</para>
975 <term>sparc-sun-sunos4</term>
976 <indexterm><primary>sparc-sun-sunos4</primary></indexterm>
978 <para>Probably works with minor tweaks, hasn't been tested
984 <term>sparc-sun-solaris2</term>
985 <indexterm><primary>sparc-sun-solaris2</primary></indexterm>
987 <para>Fully supported (at least for Solaris 2.7),
988 including native-code generator.</para>
993 <term>hppa1.1-hp-hpux (HP-PA boxes running HPUX 9.x)</term>
994 <indexterm><primary>hppa1.1-hp-hpux</primary></indexterm>
996 <para>A registerised port is available for version 4.08,
997 but GHC hasn't been built on that platform since (as far
998 as we know). No native-code generator.</para>
1003 <term>i386-unknown-linux (PCs running Linux, ELF binary format)</term>
1004 <indexterm><primary>i386-*-linux</primary></indexterm>
1006 <para>GHC works registerised and has a native code
1007 generator. You <Emphasis>must</Emphasis> have GCC 2.7.x
1008 or later. NOTE about <literal>glibc</literal> versions:
1009 GHC binaries built on a system running <literal>glibc
1010 2.0</literal> won't work on a system running
1011 <literal>glibc 2.1</literal>, and vice versa. In general,
1012 don't expect compatibility between
1013 <literal>glibc</literal> versions, even if the shared
1014 library version hasn't changed.</para>
1019 <term>i386-unknown-freebsd (PCs running FreeBSD 2.2 or
1021 <indexterm><primary>i386-unknown-freebsd</primary></indexterm>
1023 <para>GHC works registerised. Pre-built packages are
1024 available in the native package format, so if you just
1025 need binaries you're better off just installing the
1026 package (it might even be on your installation
1032 <term>i386-unknown-openbsd (PCs running OpenBSD)</term>
1033 <indexterm><primary>i386-unknown-openbsd</primary></indexterm>
1035 <para>Supported, with native code generator. Packages are
1036 available through the ports system in the native package
1042 <term>i386-unknown-netbsd (PCs running NetBSD and
1044 <indexterm><primary>i386-unknown-netbsd</primary></indexterm>
1046 <para>Will require some minor porting effort, but should
1047 work registerised.</para>
1052 <term>i386-unknown-mingw32 (PCs running Windows)</term>
1053 <indexterm><primary>i386-unknown-mingw32</primary></indexterm>
1055 <para>Fully supported under Win9x, WinNT, Win2k, and
1056 WinXP. Includes a native code generator. Building from
1057 source requires a recent <ulink
1058 url="http://www.cygwin.com/">Cygwin</ulink> distribution
1059 to be installed.</para>
1064 <term>ia64-unknown-linux</term>
1065 <indexterm><primary>ia64-unknown-linux</primary></indexterm>
1067 <para>GHC currently works unregisterised. A registerised
1068 port is in progress.</para>
1073 <term>mips-sgi-irix5</term>
1074 <indexterm><primary>mips-sgi-irix[5-6]</primary></indexterm>
1076 <para>Port has worked in the past, but hasn't been tested
1077 for some time (and will certainly have rotted in various
1078 ways). As usual, we don't have access to machines and
1079 there hasn't been an overwhelming demand for this port,
1080 but feel free to get in touch.</para>
1085 <term>powerpc-ibm-aix</term>
1086 <indexterm><primary>powerpc-ibm-aix</primary></indexterm>
1088 <para>Port currently doesn't work, needs some minimal
1089 porting effort. As usual, we don't have access to
1090 machines and there hasn't been an overwhelming demand for
1091 this port, but feel free to get in touch.</para>
1096 <term>powerpc-apple-darwin</term>
1097 <indexterm><primary>powerpc-apple-darwin</primary></indexterm>
1099 <para>Supported registerised. No native code
1105 <term>powerpc-apple-linux</term>
1106 <indexterm><primary>powerpc-apple-linux</primary></indexterm>
1108 <para>Not supported (yet).</para>
1113 <para>Various other systems have had GHC ported to them in the
1114 distant past, including various Motorola 68k boxes. The 68k
1115 support still remains, but porting to one of these systems will
1116 certainly be a non-trivial task.</para>
1120 <title>What machines the other tools run on</title>
1122 <para>Unless you hear otherwise, the other tools work if GHC
1128 <sect1 id="sec-pre-supposed">
1129 <title>Installing pre-supposed utilities</title>
1131 <indexterm><primary>pre-supposed utilities</primary></indexterm>
1132 <indexterm><primary>utilities, pre-supposed</primary></indexterm>
1134 <para>Here are the gory details about some utility programs you
1135 may need; <command>perl</command>, <command>gcc</command> and
1136 <command>happy</command> are the only important
1137 ones. (PVM<indexterm><primary>PVM</primary></indexterm> is
1138 important if you're going for Parallel Haskell.) The
1139 <command>configure</command><indexterm><primary>configure</primary></indexterm>
1140 script will tell you if you are missing something.</para>
1146 <indexterm><primary>pre-supposed: Perl</primary></indexterm>
1147 <indexterm><primary>Perl, pre-supposed</primary></indexterm>
1149 <para><emphasis>You have to have Perl to proceed!</emphasis>
1150 Perl version 5 at least is required. GHC has been known to
1151 tickle bugs in Perl, so if you find that Perl crashes when
1152 running GHC try updating (or downgrading) your Perl
1153 installation. Versions of Perl that we use and are known to
1154 be fairly stable are 5.005 and 5.6.1.</para>
1156 <para>For Win32 platforms, you should use the binary
1157 supplied in the InstallShield (copy it to
1158 <filename>/bin</filename>). The Cygwin-supplied Perl seems
1161 <para>Perl should be put somewhere so that it can be invoked
1162 by the <literal>#!</literal> script-invoking
1163 mechanism. The full pathname may need to be less than 32
1164 characters long on some systems.</para>
1169 <term>GNU C (<command>gcc</command>)</term>
1170 <indexterm><primary>pre-supposed: GCC (GNU C
1171 compiler)</primary></indexterm> <indexterm><primary>GCC (GNU C
1172 compiler), pre-supposed</primary></indexterm>
1174 <para>We recommend using GCC version 2.95.2 on all
1175 platforms. Failing that, version 2.7.2 is stable on most
1176 platforms. Earlier versions of GCC can be assumed not to
1177 work, and versions in between 2.7.2 and 2.95.2 (including
1178 <command>egcs</command>) have varying degrees of stability
1179 depending on the platform.</para>
1181 <para>If your GCC dies with “internal error” on
1182 some GHC source file, please let us know, so we can report
1183 it and get things improved. (Exception: on iX86
1184 boxes—you may need to fiddle with GHC's
1185 <option>-monly-N-regs</option> option; see the User's
1191 <term>GNU Make</term>
1192 <indexterm><primary>make</primary><secondary>GNU</secondary>
1195 <para>The fptools build system makes heavy use of features
1196 specific to GNU <command>make</command>, so you must have
1197 this installed in order to build any of the fptools
1204 <indexterm><primary>Happy</primary></indexterm>
1206 <para>Happy is a parser generator tool for Haskell, and is
1207 used to generate GHC's parsers. Happy is written in
1208 Haskell, and is a project in the CVS repository
1209 (<literal>fptools/happy</literal>). It can be built from
1210 source, but bear in mind that you'll need GHC installed in
1211 order to build it. To avoid the chicken/egg problem,
1212 install a binary distribtion of either Happy or GHC to get
1213 started. Happy distributions are available from <ulink
1214 url="http://www.haskell.org/happy/">Happy's Web
1215 Page</ulink>.</para>
1220 <term>Autoconf</term>
1221 <indexterm><primary>pre-supposed: Autoconf</primary></indexterm>
1222 <indexterm><primary>Autoconf, pre-supposed</primary></indexterm>
1224 <para>GNU Autoconf is needed if you intend to build from the
1225 CVS sources, it is <emphasis>not</emphasis> needed if you
1226 just intend to build a standard source distribution.</para>
1228 <para>Autoconf builds the <command>configure</command>
1229 script from <filename>configure.in</filename> and
1230 <filename>aclocal.m4</filename>. If you modify either of
1231 these files, you'll need <command>autoconf</command> to
1232 rebuild <filename>configure</filename>.</para>
1237 <term><command>sed</command></term>
1238 <indexterm><primary>pre-supposed: sed</primary></indexterm>
1239 <indexterm><primary>sed, pre-supposed</primary></indexterm>
1241 <para>You need a working <command>sed</command> if you are
1242 going to build from sources. The build-configuration stuff
1243 needs it. GNU sed version 2.0.4 is no good! It has a bug
1244 in it that is tickled by the build-configuration. 2.0.5 is
1245 OK. Others are probably OK too (assuming we don't create too
1246 elaborate configure scripts.)</para>
1251 <para>One <literal>fptools</literal> project is worth a quick note
1252 at this point, because it is useful for all the others:
1253 <literal>glafp-utils</literal> contains several utilities which
1254 aren't particularly Glasgow-ish, but Occasionally Indispensable.
1255 Like <command>lndir</command> for creating symbolic link
1258 <sect2 id="pre-supposed-gph-tools">
1259 <title>Tools for building parallel GHC (GPH)</title>
1263 <term>PVM version 3:</term>
1264 <indexterm><primary>pre-supposed: PVM3 (Parallel Virtual Machine)</primary></indexterm>
1265 <indexterm><primary>PVM3 (Parallel Virtual Machine), pre-supposed</primary></indexterm>
1267 <para>PVM is the Parallel Virtual Machine on which
1268 Parallel Haskell programs run. (You only need this if you
1269 plan to run Parallel Haskell. Concurent Haskell, which
1270 runs concurrent threads on a uniprocessor doesn't need
1271 it.) Underneath PVM, you can have (for example) a network
1272 of workstations (slow) or a multiprocessor box
1275 <para>The current version of PVM is 3.3.11; we use 3.3.7.
1276 It is readily available on the net; I think I got it from
1277 <literal>research.att.com</literal>, in
1278 <filename>netlib</filename>.</para>
1280 <para>A PVM installation is slightly quirky, but easy to
1281 do. Just follow the <filename>Readme</filename>
1282 instructions.</para>
1287 <term><command>bash</command>:</term>
1288 <indexterm><primary>bash, presupposed (Parallel Haskell only)</primary></indexterm>
1290 <para>Sadly, the <command>gr2ps</command> script, used to
1291 convert “parallelism profiles” to PostScript,
1292 is written in Bash (GNU's Bourne Again shell). This bug
1293 will be fixed (someday).</para>
1299 <sect2 id="pre-supposed-doc-tools">
1300 <title>Tools for building the Documentation</title>
1302 <para>The following additional tools are required if you want to
1303 format the documentation that comes with the
1304 <literal>fptools</literal> projects:</para>
1308 <term>DocBook</term>
1309 <indexterm><primary>pre-supposed: DocBook</primary></indexterm>
1310 <indexterm><primary>DocBook, pre-supposed</primary></indexterm>
1312 <para>All our documentation is written in SGML, using the
1313 DocBook DTD. Instructions on installing and configuring
1314 the DocBook tools are in the installation guide (in the
1315 GHC user guide).</para>
1321 <indexterm><primary>pre-supposed: TeX</primary></indexterm>
1322 <indexterm><primary>TeX, pre-supposed</primary></indexterm>
1324 <para>A decent TeX distribution is required if you want to
1325 produce printable documentation. We recomment teTeX,
1326 which includes just about everything you need.</para>
1331 <para> In order to actually build any documentation, you need to
1332 set <constant>SGMLDocWays</constant> in your
1333 <filename>build.mk</filename>. Valid values to add to this list
1334 are: <literal>dvi</literal>, <literal>ps</literal>,
1335 <literal>pdf</literal>, <literal>html</literal>, and
1336 <literal>rtf</literal>.</para>
1339 <sect2 id="pre-supposed-other-tools">
1340 <title>Other useful tools</title>
1345 <indexterm><primary>pre-supposed: flex</primary></indexterm>
1346 <indexterm><primary>flex, pre-supposed</primary></indexterm>
1348 <para>This is a quite-a-bit-better-than-Lex lexer. Used
1349 to build a couple of utilities in
1350 <literal>glafp-utils</literal>. Depending on your
1351 operating system, the supplied <command>lex</command> may
1352 or may not work; you should get the GNU version.</para>
1359 <sect1 id="sec-building-from-source">
1360 <title>Building from source</title>
1362 <indexterm><primary>Building from source</primary></indexterm>
1363 <indexterm><primary>Source, building from</primary></indexterm>
1365 <para>You've been rash enough to want to build some of the Glasgow
1366 Functional Programming tools (GHC, Happy, nofib, etc.) from
1367 source. You've slurped the source, from the CVS repository or
1368 from a source distribution, and now you're sitting looking at a
1369 huge mound of bits, wondering what to do next.</para>
1371 <para>Gingerly, you type <command>make</command>. Wrong
1374 <para>This rest of this guide is intended for duffers like me, who
1375 aren't really interested in Makefiles and systems configurations,
1376 but who need a mental model of the interlocking pieces so that
1377 they can make them work, extend them consistently when adding new
1378 software, and lay hands on them gently when they don't
1381 <sect2 id="sec-source-tree">
1382 <title>Your source tree</title>
1384 <para>The source code is held in your <emphasis>source
1385 tree</emphasis>. The root directory of your source tree
1386 <emphasis>must</emphasis> contain the following directories and
1391 <para><filename>Makefile</filename>: the root
1396 <para><filename>mk/</filename>: the directory that contains
1397 the main Makefile code, shared by all the
1398 <literal>fptools</literal> software.</para>
1402 <para><filename>configure.in</filename>,
1403 <filename>config.sub</filename>,
1404 <filename>config.guess</filename>: these files support the
1405 configuration process.</para>
1409 <para><filename>install-sh</filename>.</para>
1413 <para>All the other directories are individual
1414 <emphasis>projects</emphasis> of the <literal>fptools</literal>
1415 system—for example, the Glasgow Haskell Compiler
1416 (<literal>ghc</literal>), the Happy parser generator
1417 (<literal>happy</literal>), the <literal>nofib</literal>
1418 benchmark suite, and so on. You can have zero or more of these.
1419 Needless to say, some of them are needed to build others.</para>
1421 <para>The important thing to remember is that even if you want
1422 only one project (<literal>happy</literal>, say), you must have
1423 a source tree whose root directory contains
1424 <filename>Makefile</filename>, <filename>mk/</filename>,
1425 <filename>configure.in</filename>, and the project(s) you want
1426 (<filename>happy/</filename> in this case). You cannot get by
1427 with just the <filename>happy/</filename> directory.</para>
1431 <title>Build trees</title>
1432 <indexterm><primary>build trees</primary></indexterm>
1433 <indexterm><primary>link trees, for building</primary></indexterm>
1435 <para>If you just want to build the software once on a single
1436 platform, then your source tree can also be your build tree, and
1437 you can skip the rest of this section.</para>
1439 <para>We often want to build multiple versions of our software
1440 for different architectures, or with different options
1441 (e.g. profiling). It's very desirable to share a single copy of
1442 the source code among all these builds.</para>
1444 <para>So for every source tree we have zero or more
1445 <emphasis>build trees</emphasis>. Each build tree is initially
1446 an exact copy of the source tree, except that each file is a
1447 symbolic link to the source file, rather than being a copy of
1448 the source file. There are “standard” Unix
1449 utilities that make such copies, so standard that they go by
1451 <command>lndir</command><indexterm><primary>lndir</primary></indexterm>,
1452 <command>mkshadowdir</command><indexterm><primary>mkshadowdir</primary></indexterm>
1453 are two (If you don't have either, the source distribution
1454 includes sources for the X11
1455 <command>lndir</command>—check out
1456 <filename>fptools/glafp-utils/lndir</filename>). See <Xref
1457 LinkEnd="sec-storysofar"> for a typical invocation.</para>
1459 <para>The build tree does not need to be anywhere near the
1460 source tree in the file system. Indeed, one advantage of
1461 separating the build tree from the source is that the build tree
1462 can be placed in a non-backed-up partition, saving your systems
1463 support people from backing up untold megabytes of
1464 easily-regenerated, and rapidly-changing, gubbins. The golden
1465 rule is that (with a single exception—<XRef
1466 LinkEnd="sec-build-config">) <emphasis>absolutely everything in
1467 the build tree is either a symbolic link to the source tree, or
1468 else is mechanically generated</emphasis>. It should be
1469 perfectly OK for your build tree to vanish overnight; an hour or
1470 two compiling and you're on the road again.</para>
1472 <para>You need to be a bit careful, though, that any new files
1473 you create (if you do any development work) are in the source
1474 tree, not a build tree!</para>
1476 <para>Remember, that the source files in the build tree are
1477 <emphasis>symbolic links</emphasis> to the files in the source
1478 tree. (The build tree soon accumulates lots of built files like
1479 <filename>Foo.o</filename>, as well.) You can
1480 <emphasis>delete</emphasis> a source file from the build tree
1481 without affecting the source tree (though it's an odd thing to
1482 do). On the other hand, if you <emphasis>edit</emphasis> a
1483 source file from the build tree, you'll edit the source-tree
1484 file directly. (You can set up Emacs so that if you edit a
1485 source file from the build tree, Emacs will silently create an
1486 edited copy of the source file in the build tree, leaving the
1487 source file unchanged; but the danger is that you think you've
1488 edited the source file whereas actually all you've done is edit
1489 the build-tree copy. More commonly you do want to edit the
1490 source file.)</para>
1492 <para>Like the source tree, the top level of your build tree
1493 must be (a linked copy of) the root directory of the
1494 <literal>fptools</literal> suite. Inside Makefiles, the root of
1495 your build tree is called
1496 <constant>$(FPTOOLS_TOP)</constant><indexterm><primary>FPTOOLS_TOP</primary></indexterm>.
1497 In the rest of this document path names are relative to
1498 <constant>$(FPTOOLS_TOP)</constant> unless
1499 otherwise stated. For example, the file
1500 <filename>ghc/mk/target.mk</filename> is actually
1501 <filename><constant>$(FPTOOLS_TOP)</constant>/ghc/mk/target.mk</filename>.</para>
1504 <sect2 id="sec-build-config">
1505 <title>Getting the build you want</title>
1507 <para>When you build <literal>fptools</literal> you will be
1508 compiling code on a particular <emphasis>host
1509 platform</emphasis>, to run on a particular <emphasis>target
1510 platform</emphasis> (usually the same as the host
1511 platform)<indexterm><primary>platform</primary></indexterm>.
1512 The difficulty is that there are minor differences between
1513 different platforms; minor, but enough that the code needs to be
1514 a bit different for each. There are some big differences too:
1515 for a different architecture we need to build GHC with a
1516 different native-code generator.</para>
1518 <para>There are also knobs you can turn to control how the
1519 <literal>fptools</literal> software is built. For example, you
1520 might want to build GHC optimised (so that it runs fast) or
1521 unoptimised (so that you can compile it fast after you've
1522 modified it. Or, you might want to compile it with debugging on
1523 (so that extra consistency-checking code gets included) or off.
1526 <para>All of this stuff is called the
1527 <emphasis>configuration</emphasis> of your build. You set the
1528 configuration using a three-step process.</para>
1532 <term>Step 1: get ready for configuration.</term>
1534 <para>Change directory to
1535 <constant>$(FPTOOLS_TOP)</constant> and
1537 <command>autoconf</command><indexterm><primary>autoconf</primary></indexterm>
1538 (with no arguments). This GNU program converts
1539 <filename><constant>$(FPTOOLS_TOP)</constant>/configure.in</filename>
1540 to a shell script called
1541 <filename><constant>$(FPTOOLS_TOP)</constant>/configure</filename>.
1544 <para>Some projects, including GHC, have their own
1545 configure script. If there's an
1546 <constant>$(FPTOOLS_TOP)/<project>/configure.in</constant>,
1547 then you need to run <command>autoconf</command> in that
1548 directory too.</para>
1550 <para>Both these steps are completely
1551 platform-independent; they just mean that the
1552 human-written file (<filename>configure.in</filename>) can
1553 be short, although the resulting shell script,
1554 <command>configure</command>, and
1555 <filename>mk/config.h.in</filename>, are long.</para>
1557 <para>In case you don't have <command>autoconf</command>
1558 we distribute the results, <command>configure</command>,
1559 and <filename>mk/config.h.in</filename>, with the source
1560 distribution. They aren't kept in the repository,
1566 <term>Step 2: system configuration.</term>
1568 <para>Runs the newly-created <command>configure</command>
1569 script, thus:</para>
1572 ./configure <optional><parameter>args</parameter></optional>
1575 <para><command>configure</command>'s mission is to scurry
1576 round your computer working out what architecture it has,
1577 what operating system, whether it has the
1578 <Function>vfork</Function> system call, where
1579 <command>yacc</command> is kept, whether
1580 <command>gcc</command> is available, where various obscure
1581 <literal>#include</literal> files are, whether it's a
1582 leap year, and what the systems manager had for lunch. It
1583 communicates these snippets of information in two
1590 <filename>mk/config.mk.in</filename><indexterm><primary>config.mk.in</primary></indexterm>
1592 <filename>mk/config.mk</filename><indexterm><primary>config.mk</primary></indexterm>,
1593 substituting for things between
1594 “<literal>@</literal>” brackets. So,
1595 “<literal>@HaveGcc@</literal>” will be
1596 replaced by “<literal>YES</literal>” or
1597 “<literal>NO</literal>” depending on what
1598 <command>configure</command> finds.
1599 <filename>mk/config.mk</filename> is included by every
1600 Makefile (directly or indirectly), so the
1601 configuration information is thereby communicated to
1602 all Makefiles.</para>
1606 <para> It translates
1607 <filename>mk/config.h.in</filename><indexterm><primary>config.h.in</primary></indexterm>
1609 <filename>mk/config.h</filename><indexterm><primary>config.h</primary></indexterm>.
1610 The latter is <literal>#include</literal>d by
1611 various C programs, which can thereby make use of
1612 configuration information.</para>
1616 <para><command>configure</command> takes some optional
1617 arguments. Use <literal>./configure --help</literal> to
1618 get a list of the available arguments. Here are some of
1619 the ones you might need:</para>
1623 <term><literal>--with-ghc=<parameter>path</parameter></literal></term>
1624 <indexterm><primary><literal>--with-ghc</literal></primary>
1627 <para>Specifies the path to an installed GHC which
1628 you would like to use. This compiler will be used
1629 for compiling GHC-specific code (eg. GHC itself).
1630 This option <emphasis>cannot</emphasis> be specified
1631 using <filename>build.mk</filename> (see later),
1632 because <command>configure</command> needs to
1633 auto-detect the version of GHC you're using. The
1634 default is to look for a compiler named
1635 <literal>ghc</literal> in your path.</para>
1640 <term><literal>--with-hc=<parameter>path</parameter></literal></term>
1641 <indexterm><primary><literal>--with-hc</literal></primary>
1644 <para>Specifies the path to any installed Haskell
1645 compiler. This compiler will be used for compiling
1646 generic Haskell code. The default is to use
1647 <literal>ghc</literal>.</para>
1652 <term><literal>--with-gcc=<parameter>path</parameter></literal></term>
1653 <indexterm><primary><literal>--with-gcc</literal></primary>
1656 <para>Specifies the path to the installed GCC. This
1657 compiler will be used to compile all C files,
1658 <emphasis>except</emphasis> any generated by the
1659 installed Haskell compiler, which will have its own
1660 idea of which C compiler (if any) to use. The
1661 default is to use <literal>gcc</literal>.</para>
1666 <para><command>configure</command> caches the results of
1667 its run in <filename>config.cache</filename>. Quite often
1668 you don't want that; you're running
1669 <command>configure</command> a second time because
1670 something has changed. In that case, simply delete
1671 <filename>config.cache</filename>.</para>
1676 <term>Step 3: build configuration.</term>
1678 <para>Next, you say how this build of
1679 <literal>fptools</literal> is to differ from the standard
1680 defaults by creating a new file
1681 <filename>mk/build.mk</filename><indexterm><primary>build.mk</primary></indexterm>
1682 <emphasis>in the build tree</emphasis>. This file is the
1683 one and only file you edit in the build tree, precisely
1684 because it says how this build differs from the source.
1685 (Just in case your build tree does die, you might want to
1686 keep a private directory of <filename>build.mk</filename>
1687 files, and use a symbolic link in each build tree to point
1688 to the appropriate one.) So
1689 <filename>mk/build.mk</filename> never exists in the
1690 source tree—you create one in each build tree from
1691 the template. We'll discuss what to put in it
1697 <para>And that's it for configuration. Simple, eh?</para>
1699 <para>What do you put in your build-specific configuration file
1700 <filename>mk/build.mk</filename>? <emphasis>For almost all
1701 purposes all you will do is put make variable definitions that
1702 override those in</emphasis>
1703 <filename>mk/config.mk.in</filename>. The whole point of
1704 <filename>mk/config.mk.in</filename>—and its derived
1705 counterpart <filename>mk/config.mk</filename>—is to define
1706 the build configuration. It is heavily commented, as you will
1707 see if you look at it. So generally, what you do is look at
1708 <filename>mk/config.mk.in</filename>, and add definitions in
1709 <filename>mk/build.mk</filename> that override any of the
1710 <filename>config.mk</filename> definitions that you want to
1711 change. (The override occurs because the main boilerplate file,
1712 <filename>mk/boilerplate.mk</filename><indexterm><primary>boilerplate.mk</primary></indexterm>,
1713 includes <filename>build.mk</filename> after
1714 <filename>config.mk</filename>.)</para>
1716 <para>For example, <filename>config.mk.in</filename> contains
1717 the definition:</para>
1720 GhcHcOpts=-O -Rghc-timing
1723 <para>The accompanying comment explains that this is the list of
1724 flags passed to GHC when building GHC itself. For doing
1725 development, it is wise to add <literal>-DDEBUG</literal>, to
1726 enable debugging code. So you would add the following to
1727 <filename>build.mk</filename>:</para>
1729 <para>or, if you prefer,</para>
1732 GhcHcOpts += -DDEBUG
1735 <para>GNU <command>make</command> allows existing definitions to
1736 have new text appended using the “<literal>+=</literal>”
1737 operator, which is quite a convenient feature.)</para>
1739 <para>If you want to remove the <literal>-O</literal> as well (a
1740 good idea when developing, because the turn-around cycle gets a
1741 lot quicker), you can just override
1742 <literal>GhcLibHcOpts</literal> altogether:</para>
1745 GhcHcOpts=-DDEBUG -Rghc-timing
1748 <para>When reading <filename>config.mk.in</filename>, remember
1749 that anything between “@...@” signs is going to be substituted
1750 by <command>configure</command> later. You
1751 <emphasis>can</emphasis> override the resulting definition if
1752 you want, but you need to be a bit surer what you are doing.
1753 For example, there's a line that says:</para>
1759 <para>This defines the Make variables <constant>YACC</constant>
1760 to the pathname for a <command>yacc</command> that
1761 <command>configure</command> finds somewhere. If you have your
1762 own pet <command>yacc</command> you want to use instead, that's
1763 fine. Just add this line to <filename>mk/build.mk</filename>:</para>
1769 <para>You do not <emphasis>have</emphasis> to have a
1770 <filename>mk/build.mk</filename> file at all; if you don't,
1771 you'll get all the default settings from
1772 <filename>mk/config.mk.in</filename>.</para>
1774 <para>You can also use <filename>build.mk</filename> to override
1775 anything that <command>configure</command> got wrong. One place
1776 where this happens often is with the definition of
1777 <constant>FPTOOLS_TOP_ABS</constant>: this
1778 variable is supposed to be the canonical path to the top of your
1779 source tree, but if your system uses an automounter then the
1780 correct directory is hard to find automatically. If you find
1781 that <command>configure</command> has got it wrong, just put the
1782 correct definition in <filename>build.mk</filename>.</para>
1786 <sect2 id="sec-storysofar">
1787 <title>The story so far</title>
1789 <para>Let's summarise the steps you need to carry to get
1790 yourself a fully-configured build tree from scratch.</para>
1794 <para> Get your source tree from somewhere (CVS repository
1795 or source distribution). Say you call the root directory
1796 <filename>myfptools</filename> (it does not have to be
1797 called <filename>fptools</filename>). Make sure that you
1798 have the essential files (see <XRef
1799 LinkEnd="sec-source-tree">).</para>
1804 <para>(Optional) Use <command>lndir</command> or
1805 <command>mkshadowdir</command> to create a build tree.</para>
1809 $ mkshadowdir . /scratch/joe-bloggs/myfptools-sun4
1812 <para>(N.B. <command>mkshadowdir</command>'s first argument
1813 is taken relative to its second.) You probably want to give
1814 the build tree a name that suggests its main defining
1815 characteristic (in your mind at least), in case you later
1820 <para>Change directory to the build tree. Everything is
1821 going to happen there now.</para>
1824 $ cd /scratch/joe-bloggs/myfptools-sun4
1830 <para>Prepare for system configuration:</para>
1836 <para>(You can skip this step if you are starting from a
1837 source distribution, and you already have
1838 <filename>configure</filename> and
1839 <filename>mk/config.h.in</filename>.)</para>
1841 <para>Some projects, including GHC itself, have their own
1842 configure scripts, so it is necessary to run autoconf again
1843 in the appropriate subdirectories. eg:</para>
1846 $ (cd ghc; autoconf)
1851 <para>Do system configuration:</para>
1857 <para>Don't forget to check whether you need to add any
1858 arguments to <literal>configure</literal>; for example, a
1859 common requirement is to specify which GHC to use with
1860 <option>--with-ghc=<replaceable>ghc</replaceable></option>.</para>
1864 <para>Create the file <filename>mk/build.mk</filename>,
1865 adding definitions for your desired configuration
1874 <para>You can make subsequent changes to
1875 <filename>mk/build.mk</filename> as often as you like. You do
1876 not have to run any further configuration programs to make these
1877 changes take effect. In theory you should, however, say
1878 <command>gmake clean</command>, <command>gmake all</command>,
1879 because configuration option changes could affect
1880 anything—but in practice you are likely to know what's
1885 <title>Making things</title>
1887 <para>At this point you have made yourself a fully-configured
1888 build tree, so you are ready to start building real
1891 <para>The first thing you need to know is that <emphasis>you
1892 must use GNU <command>make</command>, usually called
1893 <command>gmake</command>, not standard Unix
1894 <command>make</command></emphasis>. If you use standard Unix
1895 <command>make</command> you will get all sorts of error messages
1896 (but no damage) because the <literal>fptools</literal>
1897 <command>Makefiles</command> use GNU <command>make</command>'s
1898 facilities extensively.</para>
1900 <para>To just build the whole thing, <command>cd</command> to
1901 the top of your <literal>fptools</literal> tree and type
1902 <command>gmake</command>. This will prepare the tree and build
1903 the various projects in the correct order.</para>
1907 <sect2 id="sec-standard-targets">
1908 <title>Standard Targets</title>
1909 <indexterm><primary>targets, standard makefile</primary></indexterm>
1910 <indexterm><primary>makefile targets</primary></indexterm>
1912 <para>In any directory you should be able to make the following:</para>
1916 <term><literal>boot</literal></term>
1918 <para>does the one-off preparation required to get ready
1919 for the real work. Notably, it does <command>gmake
1920 depend</command> in all directories that contain programs.
1921 It also builds the necessary tools for compilation to
1924 <para>Invoking the <literal>boot</literal> target
1925 explicitly is not normally necessary. From the top-level
1926 <literal>fptools</literal> directory, invoking
1927 <literal>gmake</literal> causes <literal>gmake boot
1928 all</literal> to be invoked in each of the project
1929 subdirectories, in the order specified by
1930 <literal>$(AllTargets)</literal> in
1931 <literal>config.mk</literal>.</para>
1933 <para>If you're working in a subdirectory somewhere and
1934 need to update the dependencies, <literal>gmake
1935 boot</literal> is a good way to do it.</para>
1940 <term><literal>all</literal></term>
1942 <para>makes all the final target(s) for this Makefile.
1943 Depending on which directory you are in a “final
1944 target” may be an executable program, a library
1945 archive, a shell script, or a Postscript file. Typing
1946 <command>gmake</command> alone is generally the same as
1947 typing <command>gmake all</command>.</para>
1952 <term><literal>install</literal></term>
1954 <para>installs the things built by <literal>all</literal>
1955 (except for the documentation). Where does it install
1956 them? That is specified by
1957 <filename>mk/config.mk.in</filename>; you can override it
1958 in <filename>mk/build.mk</filename>, or by running
1959 <command>configure</command> with command-line arguments
1960 like <literal>--bindir=/home/simonpj/bin</literal>; see
1961 <literal>./configure --help</literal> for the full
1967 <term><literal>install-docs</literal></term>
1969 <para>installs the documentation. Otherwise behaves just
1970 like <literal>install</literal>.</para>
1975 <term><literal>uninstall</literal></term>
1977 <para>reverses the effect of
1978 <literal>install</literal>.</para>
1983 <term><literal>clean</literal></term>
1985 <para>Delete all files from the current directory that are
1986 normally created by building the program. Don't delete
1987 the files that record the configuration, or files
1988 generated by <command>gmake boot</command>. Also preserve
1989 files that could be made by building, but normally aren't
1990 because the distribution comes with them.</para>
1995 <term><literal>distclean</literal></term>
1997 <para>Delete all files from the current directory that are
1998 created by configuring or building the program. If you
1999 have unpacked the source and built the program without
2000 creating any other files, <literal>make
2001 distclean</literal> should leave only the files that were
2002 in the distribution.</para>
2007 <term><literal>mostlyclean</literal></term>
2009 <para>Like <literal>clean</literal>, but may refrain from
2010 deleting a few files that people normally don't want to
2016 <term><literal>maintainer-clean</literal></term>
2018 <para>Delete everything from the current directory that
2019 can be reconstructed with this Makefile. This typically
2020 includes everything deleted by
2021 <literal>distclean</literal>, plus more: C source files
2022 produced by Bison, tags tables, Info files, and so
2025 <para>One exception, however: <literal>make
2026 maintainer-clean</literal> should not delete
2027 <filename>configure</filename> even if
2028 <filename>configure</filename> can be remade using a rule
2029 in the <filename>Makefile</filename>. More generally,
2030 <literal>make maintainer-clean</literal> should not delete
2031 anything that needs to exist in order to run
2032 <filename>configure</filename> and then begin to build the
2038 <term><literal>check</literal></term>
2040 <para>run the test suite.</para>
2045 <para>All of these standard targets automatically recurse into
2046 sub-directories. Certain other standard targets do not:</para>
2050 <term><literal>configure</literal></term>
2052 <para>is only available in the root directory
2053 <constant>$(FPTOOLS_TOP)</constant>; it has
2054 been discussed in <XRef
2055 LinkEnd="sec-build-config">.</para>
2060 <term><literal>depend</literal></term>
2062 <para>make a <filename>.depend</filename> file in each
2063 directory that needs it. This <filename>.depend</filename>
2064 file contains mechanically-generated dependency
2065 information; for example, suppose a directory contains a
2066 Haskell source module <filename>Foo.lhs</filename> which
2067 imports another module <literal>Baz</literal>. Then the
2068 generated <filename>.depend</filename> file will contain
2069 the dependency:</para>
2075 <para>which says that the object file
2076 <filename>Foo.o</filename> depends on the interface file
2077 <filename>Baz.hi</filename> generated by compiling module
2078 <literal>Baz</literal>. The <filename>.depend</filename>
2079 file is automatically included by every Makefile.</para>
2084 <term><literal>binary-dist</literal></term>
2086 <para>make a binary distribution. This is the target we
2087 use to build the binary distributions of GHC and
2093 <term><literal>dist</literal></term>
2095 <para>make a source distribution. Note that this target
2096 does “make distclean” as part of its work;
2097 don't use it if you want to keep what you've built.</para>
2102 <para>Most <filename>Makefile</filename>s have targets other
2103 than these. You can discover them by looking in the
2104 <filename>Makefile</filename> itself.</para>
2108 <title>Using a project from the build tree</title>
2110 <para>If you want to build GHC (say) and just use it direct from
2111 the build tree without doing <literal>make install</literal>
2112 first, you can run the in-place driver script:
2113 <filename>ghc/compiler/ghc-inplace</filename>.</para>
2115 <para> Do <emphasis>NOT</emphasis> use
2116 <filename>ghc/compiler/ghc</filename>, or
2117 <filename>ghc/compiler/ghc-5.xx</filename>, as these are the
2118 scripts intended for installation, and contain hard-wired paths
2119 to the installed libraries, rather than the libraries in the
2122 <para>Happy can similarly be run from the build tree, using
2123 <filename>happy/src/happy-inplace</filename>.</para>
2127 <title>Fast Making</title>
2129 <indexterm><primary>fastmake</primary></indexterm>
2130 <indexterm><primary>dependencies, omitting</primary></indexterm>
2131 <indexterm><primary>FAST, makefile variable</primary></indexterm>
2133 <para>Sometimes the dependencies get in the way: if you've made
2134 a small change to one file, and you're absolutely sure that it
2135 won't affect anything else, but you know that
2136 <command>make</command> is going to rebuild everything anyway,
2137 the following hack may be useful:</para>
2143 <para>This tells the make system to ignore dependencies and just
2144 build what you tell it to. In other words, it's equivalent to
2145 temporarily removing the <filename>.depend</filename> file in
2146 the current directory (where <command>mkdependHS</command> and
2147 friends store their dependency information).</para>
2149 <para>A bit of history: GHC used to come with a
2150 <command>fastmake</command> script that did the above job, but
2151 GNU make provides the features we need to do it without
2152 resorting to a script. Also, we've found that fastmaking is
2153 less useful since the advent of GHC's recompilation checker (see
2154 the User's Guide section on "Separate Compilation").</para>
2158 <sect1 id="sec-makefile-arch">
2159 <title>The <filename>Makefile</filename> architecture</title>
2160 <indexterm><primary>makefile architecture</primary></indexterm>
2162 <para><command>make</command> is great if everything
2163 works—you type <command>gmake install</command> and lo! the
2164 right things get compiled and installed in the right places. Our
2165 goal is to make this happen often, but somehow it often doesn't;
2166 instead some weird error message eventually emerges from the
2167 bowels of a directory you didn't know existed.</para>
2169 <para>The purpose of this section is to give you a road-map to
2170 help you figure out what is going right and what is going
2174 <title>Debugging</title>
2176 <para>Debugging <filename>Makefile</filename>s is something of a
2177 black art, but here's a couple of tricks that we find
2178 particularly useful. The following command allows you to see
2179 the contents of any make variable in the context of the current
2180 <filename>Makefile</filename>:</para>
2182 <screen>$ make show VALUE=HS_SRCS</screen>
2184 <para>where you can replace <literal>HS_SRCS</literal> with the
2185 name of any variable you wish to see the value of.</para>
2187 <para>GNU make has a <option>-d</option> option which generates
2188 a dump of the decision procedure used to arrive at a conclusion
2189 about which files should be recompiled. Sometimes useful for
2190 tracking down problems with superfluous or missing
2191 recompilations.</para>
2195 <title>A small project</title>
2197 <para>To get started, let us look at the
2198 <filename>Makefile</filename> for an imaginary small
2199 <literal>fptools</literal> project, <literal>small</literal>.
2200 Each project in <literal>fptools</literal> has its own directory
2201 in <constant>FPTOOLS_TOP</constant>, so the
2202 <literal>small</literal> project will have its own directory
2203 <constant>FPOOLS_TOP/small/</constant>. Inside the
2204 <filename>small/</filename> directory there will be a
2205 <filename>Makefile</filename>, looking something like
2208 <indexterm><primary>Makefile, minimal</primary></indexterm>
2211 # Makefile for fptools project "small"
2214 include $(TOP)/mk/boilerplate.mk
2216 SRCS = $(wildcard *.lhs) $(wildcard *.c)
2219 include $(TOP)/target.mk
2222 <para>this <filename>Makefile</filename> has three
2227 <para>The first section includes
2230 One of the most important
2231 features of GNU <command>make</command> that we use is the ability for a <filename>Makefile</filename> to
2232 include another named file, very like <command>cpp</command>'s <literal>#include</literal>
2237 a file of “boilerplate” code from the level
2238 above (which in this case will be
2239 <filename><constant>FPTOOLS_TOP</constant>/mk/boilerplate.mk</filename><indexterm><primary>boilerplate.mk</primary></indexterm>).
2240 As its name suggests, <filename>boilerplate.mk</filename>
2241 consists of a large quantity of standard
2242 <filename>Makefile</filename> code. We discuss this
2243 boilerplate in more detail in <XRef LinkEnd="sec-boiler">.
2244 <indexterm><primary>include, directive in
2245 Makefiles</primary></indexterm> <indexterm><primary>Makefile
2246 inclusion</primary></indexterm></para>
2248 <para>Before the <literal>include</literal> statement, you
2249 must define the <command>make</command> variable
2250 <constant>TOP</constant><indexterm><primary>TOP</primary></indexterm>
2251 to be the directory containing the <filename>mk</filename>
2252 directory in which the <filename>boilerplate.mk</filename>
2253 file is. It is <emphasis>not</emphasis> OK to simply say</para>
2256 include ../mk/boilerplate.mk # NO NO NO
2260 <para>Why? Because the <filename>boilerplate.mk</filename>
2261 file needs to know where it is, so that it can, in turn,
2262 <literal>include</literal> other files. (Unfortunately,
2263 when an <literal>include</literal>d file does an
2264 <literal>include</literal>, the filename is treated relative
2265 to the directory in which <command>gmake</command> is being
2266 run, not the directory in which the
2267 <literal>include</literal>d sits.) In general,
2268 <emphasis>every file <filename>foo.mk</filename> assumes
2270 <filename><constant>$(TOP)</constant>/mk/foo.mk</filename>
2271 refers to itself.</emphasis> It is up to the
2272 <filename>Makefile</filename> doing the
2273 <literal>include</literal> to ensure this is the case.</para>
2275 <para>Files intended for inclusion in other
2276 <filename>Makefile</filename>s are written to have the
2277 following property: <emphasis>after
2278 <filename>foo.mk</filename> is <literal>include</literal>d,
2279 it leaves <constant>TOP</constant> containing the same value
2280 as it had just before the <literal>include</literal>
2281 statement</emphasis>. In our example, this invariant
2282 guarantees that the <literal>include</literal> for
2283 <filename>target.mk</filename> will look in the same
2284 directory as that for <filename>boilerplate.mk</filename>.</para>
2288 <para> The second section defines the following standard
2289 <command>make</command> variables:
2290 <constant>SRCS</constant><indexterm><primary>SRCS</primary></indexterm>
2291 (the source files from which is to be built), and
2292 <constant>HS_PROG</constant><indexterm><primary>HS_PROG</primary></indexterm>
2293 (the executable binary to be built). We will discuss in
2294 more detail what the “standard variables” are,
2295 and how they affect what happens, in <XRef
2296 LinkEnd="sec-targets">.</para>
2298 <para>The definition for <constant>SRCS</constant> uses the
2299 useful GNU <command>make</command> construct
2300 <literal>$(wildcard $pat$)</literal><indexterm><primary>wildcard</primary></indexterm>,
2301 which expands to a list of all the files matching the
2302 pattern <literal>pat</literal> in the current directory. In
2303 this example, <constant>SRCS</constant> is set to the list
2304 of all the <filename>.lhs</filename> and
2305 <filename>.c</filename> files in the directory. (Let's
2306 suppose there is one of each, <filename>Foo.lhs</filename>
2307 and <filename>Baz.c</filename>.)</para>
2311 <para>The last section includes a second file of standard
2313 <filename>target.mk</filename><indexterm><primary>target.mk</primary></indexterm>.
2314 It contains the rules that tell <command>gmake</command> how
2315 to make the standard targets (<Xref
2316 LinkEnd="sec-standard-targets">). Why, you ask, can't this
2317 standard code be part of
2318 <filename>boilerplate.mk</filename>? Good question. We
2319 discuss the reason later, in <Xref
2320 LinkEnd="sec-boiler-arch">.</para>
2322 <para>You do not <emphasis>have</emphasis> to
2323 <literal>include</literal> the
2324 <filename>target.mk</filename> file. Instead, you can write
2325 rules of your own for all the standard targets. Usually,
2326 though, you will find quite a big payoff from using the
2327 canned rules in <filename>target.mk</filename>; the price
2328 tag is that you have to understand what canned rules get
2329 enabled, and what they do (<Xref
2330 LinkEnd="sec-targets">).</para>
2334 <para>In our example <filename>Makefile</filename>, most of the
2335 work is done by the two <literal>include</literal>d files. When
2336 you say <command>gmake all</command>, the following things
2341 <para><command>gmake</command> figures out that the object
2342 files are <filename>Foo.o</filename> and
2343 <filename>Baz.o</filename>.</para>
2347 <para>It uses a boilerplate pattern rule to compile
2348 <filename>Foo.lhs</filename> to <filename>Foo.o</filename>
2349 using a Haskell compiler. (Which one? That is set in the
2350 build configuration.)</para>
2354 <para>It uses another standard pattern rule to compile
2355 <filename>Baz.c</filename> to <filename>Baz.o</filename>,
2356 using a C compiler. (Ditto.)</para>
2360 <para>It links the resulting <filename>.o</filename> files
2361 together to make <literal>small</literal>, using the Haskell
2362 compiler to do the link step. (Why not use
2363 <command>ld</command>? Because the Haskell compiler knows
2364 what standard libraries to link in. How did
2365 <command>gmake</command> know to use the Haskell compiler to
2366 do the link, rather than the C compiler? Because we set the
2367 variable <constant>HS_PROG</constant> rather than
2368 <constant>C_PROG</constant>.)</para>
2372 <para>All <filename>Makefile</filename>s should follow the above
2373 three-section format.</para>
2377 <title>A larger project</title>
2379 <para>Larger projects are usually structured into a number of
2380 sub-directories, each of which has its own
2381 <filename>Makefile</filename>. (In very large projects, this
2382 sub-structure might be iterated recursively, though that is
2383 rare.) To give you the idea, here's part of the directory
2384 structure for the (rather large) GHC project:</para>
2394 ...source files for documentation...
2397 ...source files for driver...
2400 parser/...source files for parser...
2401 renamer/...source files for renamer...
2405 <para>The sub-directories <filename>docs</filename>,
2406 <filename>driver</filename>, <filename>compiler</filename>, and
2407 so on, each contains a sub-component of GHC, and each has its
2408 own <filename>Makefile</filename>. There must also be a
2409 <filename>Makefile</filename> in
2410 <filename><constant>$(FPTOOLS_TOP)</constant>/ghc</filename>.
2411 It does most of its work by recursively invoking
2412 <command>gmake</command> on the <filename>Makefile</filename>s
2413 in the sub-directories. We say that
2414 <filename>ghc/Makefile</filename> is a <emphasis>non-leaf
2415 <filename>Makefile</filename></emphasis>, because it does little
2416 except organise its children, while the
2417 <filename>Makefile</filename>s in the sub-directories are all
2418 <emphasis>leaf <filename>Makefile</filename>s</emphasis>. (In
2419 principle the sub-directories might themselves contain a
2420 non-leaf <filename>Makefile</filename> and several
2421 sub-sub-directories, but that does not happen in GHC.)</para>
2423 <para>The <filename>Makefile</filename> in
2424 <filename>ghc/compiler</filename> is considered a leaf
2425 <filename>Makefile</filename> even though the
2426 <filename>ghc/compiler</filename> has sub-directories, because
2427 these sub-directories do not themselves have
2428 <filename>Makefile</filename>s in them. They are just used to
2429 structure the collection of modules that make up GHC, but all
2430 are managed by the single <filename>Makefile</filename> in
2431 <filename>ghc/compiler</filename>.</para>
2433 <para>You will notice that <filename>ghc/</filename> also
2434 contains a directory <filename>ghc/mk/</filename>. It contains
2435 GHC-specific <filename>Makefile</filename> boilerplate code.
2436 More precisely:</para>
2440 <para><filename>ghc/mk/boilerplate.mk</filename> is included
2441 at the top of <filename>ghc/Makefile</filename>, and of all
2442 the leaf <filename>Makefile</filename>s in the
2443 sub-directories. It in turn <literal>include</literal>s the
2444 main boilerplate file
2445 <filename>mk/boilerplate.mk</filename>.</para>
2449 <para><filename>ghc/mk/target.mk</filename> is
2450 <literal>include</literal>d at the bottom of
2451 <filename>ghc/Makefile</filename>, and of all the leaf
2452 <filename>Makefile</filename>s in the sub-directories. It
2453 in turn <literal>include</literal>s the file
2454 <filename>mk/target.mk</filename>.</para>
2458 <para>So these two files are the place to look for GHC-wide
2459 customisation of the standard boilerplate.</para>
2462 <sect2 id="sec-boiler-arch">
2463 <title>Boilerplate architecture</title>
2464 <indexterm><primary>boilerplate architecture</primary></indexterm>
2466 <para>Every <filename>Makefile</filename> includes a
2467 <filename>boilerplate.mk</filename><indexterm><primary>boilerplate.mk</primary></indexterm>
2468 file at the top, and
2469 <filename>target.mk</filename><indexterm><primary>target.mk</primary></indexterm>
2470 file at the bottom. In this section we discuss what is in these
2471 files, and why there have to be two of them. In general:</para>
2475 <para><filename>boilerplate.mk</filename> consists of:</para>
2479 <para><emphasis>Definitions of millions of
2480 <command>make</command> variables</emphasis> that
2481 collectively specify the build configuration. Examples:
2482 <constant>HC_OPTS</constant><indexterm><primary>HC_OPTS</primary></indexterm>,
2483 the options to feed to the Haskell compiler;
2484 <constant>NoFibSubDirs</constant><indexterm><primary>NoFibSubDirs</primary></indexterm>,
2485 the sub-directories to enable within the
2486 <literal>nofib</literal> project;
2487 <constant>GhcWithHc</constant><indexterm><primary>GhcWithHc</primary></indexterm>,
2488 the name of the Haskell compiler to use when compiling
2489 GHC in the <literal>ghc</literal> project.</para>
2493 <para><emphasis>Standard pattern rules</emphasis> that
2494 tell <command>gmake</command> how to construct one file
2495 from another.</para>
2499 <para><filename>boilerplate.mk</filename> needs to be
2500 <literal>include</literal>d at the <emphasis>top</emphasis>
2501 of each <filename>Makefile</filename>, so that the user can
2502 replace the boilerplate definitions or pattern rules by
2503 simply giving a new definition or pattern rule in the
2504 <filename>Makefile</filename>. <command>gmake</command>
2505 simply takes the last definition as the definitive one.</para>
2507 <para>Instead of <emphasis>replacing</emphasis> boilerplate
2508 definitions, it is also quite common to
2509 <emphasis>augment</emphasis> them. For example, a
2510 <filename>Makefile</filename> might say:</para>
2516 <para>thereby adding “<option>-O</option>” to
2518 <constant>SRC_HC_OPTS</constant><indexterm><primary>SRC_HC_OPTS</primary></indexterm>.</para>
2522 <para><filename>target.mk</filename> contains
2523 <command>make</command> rules for the standard targets
2524 described in <Xref LinkEnd="sec-standard-targets">. These
2525 rules are selectively included, depending on the setting of
2526 certain <command>make</command> variables. These variables
2527 are usually set in the middle section of the
2528 <filename>Makefile</filename> between the two
2529 <literal>include</literal>s.</para>
2531 <para><filename>target.mk</filename> must be included at the
2532 end (rather than being part of
2533 <filename>boilerplate.mk</filename>) for several tiresome
2539 <para><command>gmake</command> commits target and
2540 dependency lists earlier than it should. For example,
2541 <FIlename>target.mk</FIlename> has a rule that looks
2545 $(HS_PROG) : $(OBJS)
2546 $(HC) $(LD_OPTS) $< -o $@
2549 <para>If this rule was in
2550 <filename>boilerplate.mk</filename> then
2551 <constant>$(HS_PROG)</constant><indexterm><primary>HS_PROG</primary></indexterm>
2553 <constant>$(OBJS)</constant><indexterm><primary>OBJS</primary></indexterm>
2554 would not have their final values at the moment
2555 <command>gmake</command> encountered the rule. Alas,
2556 <command>gmake</command> takes a snapshot of their
2557 current values, and wires that snapshot into the rule.
2558 (In contrast, the commands executed when the rule
2559 “fires” are only substituted at the moment
2560 of firing.) So, the rule must follow the definitions
2561 given in the <filename>Makefile</filename> itself.</para>
2565 <para>Unlike pattern rules, ordinary rules cannot be
2566 overriden or replaced by subsequent rules for the same
2567 target (at least, not without an error message).
2568 Including ordinary rules in
2569 <filename>boilerplate.mk</filename> would prevent the
2570 user from writing rules for specific targets in specific
2575 <para>There are a couple of other reasons I've
2576 forgotten, but it doesn't matter too much.</para>
2583 <sect2 id="sec-boiler">
2584 <title>The main <filename>mk/boilerplate.mk</filename> file</title>
2585 <indexterm><primary>boilerplate.mk</primary></indexterm>
2587 <para>If you look at
2588 <filename><constant>$(FPTOOLS_TOP)</constant>/mk/boilerplate.mk</filename>
2589 you will find that it consists of the following sections, each
2590 held in a separate file:</para>
2594 <term><filename>config.mk</filename></term>
2595 <indexterm><primary>config.mk</primary></indexterm>
2597 <para>is the build configuration file we discussed at
2598 length in <Xref LinkEnd="sec-build-config">.</para>
2603 <term><filename>paths.mk</filename></term>
2604 <indexterm><primary>paths.mk</primary></indexterm>
2606 <para>defines <command>make</command> variables for
2607 pathnames and file lists. This file contains code for
2608 automatically compiling lists of source files and deriving
2609 lists of object files from those. The results can be
2610 overriden in the <filename>Makefile</filename>, but in
2611 most cases the automatic setup should do the right
2614 <para>The following variables may be set in the
2615 <filename>Makefile</filename> to affect how the automatic
2616 source file search is done:</para>
2620 <term><literal>ALL_DIRS</literal></term>
2621 <indexterm><primary><literal>ALL_DIRS</literal></primary>
2624 <para>Set to a list of directories to search in
2625 addition to the current directory for source
2631 <term><literal>EXCLUDE_SRCS</literal></term>
2632 <indexterm><primary><literal>EXCLUDE_SRCS</literal></primary>
2635 <para>Set to a list of source files (relative to the
2636 current directory) to omit from the automatic
2637 search. The source searching machinery is clever
2638 enough to know that if you exclude a source file
2639 from which other sources are derived, then the
2640 derived sources should also be excluded. For
2641 example, if you set <literal>EXCLUDED_SRCS</literal>
2642 to include <filename>Foo.y</filename>, then
2643 <filename>Foo.hs</filename> will also be
2649 <term><literal>EXTRA_SRCS</literal></term>
2650 <indexterm><primary><literal>EXCLUDE_SRCS</literal></primary>
2653 <para>Set to a list of extra source files (perhaps
2654 in directories not listed in
2655 <literal>ALL_DIRS</literal>) that should be
2661 <para>The results of the automatic source file search are
2662 placed in the following make variables:</para>
2666 <term><literal>SRCS</literal></term>
2667 <indexterm><primary><literal>SRCS</literal></primary></indexterm>
2669 <para>All source files found, sorted and without
2670 duplicates, including those which might not exist
2671 yet but will be derived from other existing sources.
2672 <literal>SRCS</literal> <emphasis>can</emphasis> be
2673 overriden if necessary, in which case the variables
2674 below will follow suit.</para>
2679 <term><literal>HS_SRCS</literal></term>
2680 <indexterm><primary><literal>HS_SRCS</literal></primary></indexterm>
2682 <para>all Haskell source files in the current
2683 directory, including those derived from other source
2684 files (eg. Happy sources also give rise to Haskell
2690 <term><literal>HS_OBJS</literal></term>
2691 <indexterm><primary><literal>HS_OBJS</literal></primary></indexterm>
2693 <para>Object files derived from
2694 <literal>HS_SRCS</literal>.</para>
2699 <term><literal>HS_IFACES</literal></term>
2700 <indexterm><primary><literal>HS_IFACES</literal></primary></indexterm>
2702 <para>Interface files (<literal>.hi</literal> files)
2703 derived from <literal>HS_SRCS</literal>.</para>
2708 <term><literal>C_SRCS</literal></term>
2709 <indexterm><primary><literal>C_SRCS</literal></primary></indexterm>
2711 <para>All C source files found.</para>
2716 <term><literal>C_OBJS</literal></term>
2717 <indexterm><primary><literal>C_OBJS</literal></primary></indexterm>
2719 <para>Object files derived from
2720 <literal>C_SRCS</literal>.</para>
2725 <term><literal>SCRIPT_SRCS</literal></term>
2726 <indexterm><primary><literal>SCRIPT_SRCS</literal></primary></indexterm>
2728 <para>All script source files found
2729 (<literal>.lprl</literal> files).</para>
2734 <term><literal>SCRIPT_OBJS</literal></term>
2735 <indexterm><primary><literal>SCRIPT_OBJS</literal></primary></indexterm>
2737 <para><quote>object</quote> files derived from
2738 <literal>SCRIPT_SRCS</literal>
2739 (<literal>.prl</literal> files).</para>
2744 <term><literal>HSC_SRCS</literal></term>
2745 <indexterm><primary><literal>HSC_SRCS</literal></primary></indexterm>
2747 <para>All <literal>hsc2hs</literal> source files
2748 (<literal>.hsc</literal> files).</para>
2753 <term><literal>HAPPY_SRCS</literal></term>
2754 <indexterm><primary><literal>HAPPY_SRCS</literal></primary></indexterm>
2756 <para>All <literal>happy</literal> source files
2757 (<literal>.y</literal> or <literal>.hy</literal> files).</para>
2762 <term><literal>OBJS</literal></term>
2763 <indexterm><primary>OBJS</primary></indexterm>
2765 <para>the concatenation of
2766 <literal>$(HS_OBJS)</literal>,
2767 <literal>$(C_OBJS)</literal>, and
2768 <literal>$(SCRIPT_OBJS)</literal>.</para>
2773 <para>Any or all of these definitions can easily be
2774 overriden by giving new definitions in your
2775 <filename>Makefile</filename>.</para>
2777 <para>What, exactly, does <filename>paths.mk</filename>
2778 consider a <quote>source file</quote> to be? It's based
2779 on the file's suffix (e.g. <filename>.hs</filename>,
2780 <filename>.lhs</filename>, <filename>.c</filename>,
2781 <filename>.hy</filename>, etc), but this is the kind of
2782 detail that changes, so rather than enumerate the source
2783 suffices here the best thing to do is to look in
2784 <filename>paths.mk</filename>.</para>
2789 <term><filename>opts.mk</filename></term>
2790 <indexterm><primary>opts.mk</primary></indexterm>
2792 <para>defines <command>make</command> variables for option
2793 strings to pass to each program. For example, it defines
2794 <constant>HC_OPTS</constant><indexterm><primary>HC_OPTS</primary></indexterm>,
2795 the option strings to pass to the Haskell compiler. See
2796 <Xref LinkEnd="sec-suffix">.</para>
2801 <term><filename>suffix.mk</filename></term>
2802 <indexterm><primary>suffix.mk</primary></indexterm>
2804 <para>defines standard pattern rules—see <Xref
2805 LinkEnd="sec-suffix">.</para>
2810 <para>Any of the variables and pattern rules defined by the
2811 boilerplate file can easily be overridden in any particular
2812 <filename>Makefile</filename>, because the boilerplate
2813 <literal>include</literal> comes first. Definitions after this
2814 <literal>include</literal> directive simply override the default
2815 ones in <filename>boilerplate.mk</filename>.</para>
2818 <sect2 id="sec-suffix">
2819 <title>Pattern rules and options</title>
2820 <indexterm><primary>Pattern rules</primary></indexterm>
2823 <filename>suffix.mk</filename><indexterm><primary>suffix.mk</primary></indexterm>
2824 defines standard <emphasis>pattern rules</emphasis> that say how
2825 to build one kind of file from another, for example, how to
2826 build a <filename>.o</filename> file from a
2827 <filename>.c</filename> file. (GNU <command>make</command>'s
2828 <emphasis>pattern rules</emphasis> are more powerful and easier
2829 to use than Unix <command>make</command>'s <emphasis>suffix
2830 rules</emphasis>.)</para>
2832 <para>Almost all the rules look something like this:</para>
2837 $(CC) $(CC_OPTS) -c $< -o $@
2840 <para>Here's how to understand the rule. It says that
2841 <emphasis>something</emphasis><filename>.o</filename> (say
2842 <filename>Foo.o</filename>) can be built from
2843 <emphasis>something</emphasis><filename>.c</filename>
2844 (<filename>Foo.c</filename>), by invoking the C compiler (path
2845 name held in <constant>$(CC)</constant>), passing to it
2846 the options <constant>$(CC_OPTS)</constant> and
2847 the rule's dependent file of the rule
2848 <literal>$<</literal> (<filename>Foo.c</filename> in
2849 this case), and putting the result in the rule's target
2850 <literal>$@</literal> (<filename>Foo.o</filename> in this
2853 <para>Every program is held in a <command>make</command>
2854 variable defined in <filename>mk/config.mk</filename>—look
2855 in <filename>mk/config.mk</filename> for the complete list. One
2856 important one is the Haskell compiler, which is called
2857 <constant>$(HC)</constant>.</para>
2859 <para>Every program's options are are held in a
2860 <command>make</command> variables called
2861 <constant><prog>_OPTS</constant>. the
2862 <constant><prog>_OPTS</constant> variables are
2863 defined in <filename>mk/opts.mk</filename>. Almost all of them
2864 are defined like this:</para>
2867 CC_OPTS = $(SRC_CC_OPTS) $(WAY$(_way)_CC_OPTS) $($*_CC_OPTS) $(EXTRA_CC_OPTS)
2870 <para>The four variables from which
2871 <constant>CC_OPTS</constant> is built have the following
2876 <term><constant>SRC_CC_OPTS</constant><indexterm><primary>SRC_CC_OPTS</primary></indexterm>:</term>
2878 <para>options passed to all C compilations.</para>
2883 <term><constant>WAY_<way>_CC_OPTS</constant>:</term>
2885 <para>options passed to C compilations for way
2886 <literal><way></literal>. For example,
2887 <constant>WAY_mp_CC_OPTS</constant>
2888 gives options to pass to the C compiler when compiling way
2889 <literal>mp</literal>. The variable
2890 <constant>WAY_CC_OPTS</constant> holds
2891 options to pass to the C compiler when compiling the
2892 standard way. (<Xref LinkEnd="sec-ways"> dicusses
2893 multi-way compilation.)</para>
2898 <term><constant><module>_CC_OPTS</constant>:</term>
2900 <para>options to pass to the C compiler that are specific
2901 to module <literal><module></literal>. For example,
2902 <constant>SMap_CC_OPTS</constant> gives the
2903 specific options to pass to the C compiler when compiling
2904 <filename>SMap.c</filename>.</para>
2909 <term><constant>EXTRA_CC_OPTS</constant><indexterm><primary>EXTRA_CC_OPTS</primary></indexterm>:</term>
2911 <para>extra options to pass to all C compilations. This
2912 is intended for command line use, thus:</para>
2915 gmake libHS.a EXTRA_CC_OPTS="-v"
2922 <sect2 id="sec-targets">
2923 <title>The main <filename>mk/target.mk</filename> file</title>
2924 <indexterm><primary>target.mk</primary></indexterm>
2926 <para><filename>target.mk</filename> contains canned rules for
2927 all the standard targets described in <Xref
2928 LinkEnd="sec-standard-targets">. It is complicated by the fact
2929 that you don't want all of these rules to be active in every
2930 <filename>Makefile</filename>. Rather than have a plethora of
2931 tiny files which you can include selectively, there is a single
2932 file, <filename>target.mk</filename>, which selectively includes
2933 rules based on whether you have defined certain variables in
2934 your <filename>Makefile</filename>. This section explains what
2935 rules you get, what variables control them, and what the rules
2936 do. Hopefully, you will also get enough of an idea of what is
2937 supposed to happen that you can read and understand any weird
2938 special cases yourself.</para>
2942 <term><constant>HS_PROG</constant><indexterm><primary>HS_PROG</primary></indexterm>.</term>
2944 <para>If <constant>HS_PROG</constant> is defined,
2945 you get rules with the following targets:</para>
2949 <term><filename>HS_PROG</filename><indexterm><primary>HS_PROG</primary></indexterm></term>
2951 <para>itself. This rule links
2952 <constant>$(OBJS)</constant> with the Haskell
2953 runtime system to get an executable called
2954 <constant>$(HS_PROG)</constant>.</para>
2959 <term><literal>install</literal><indexterm><primary>install</primary></indexterm></term>
2962 <constant>$(HS_PROG)</constant> in
2963 <constant>$(bindir)</constant>.</para>
2972 <term><constant>C_PROG</constant><indexterm><primary>C_PROG</primary></indexterm></term>
2974 <para>is similar to <constant>HS_PROG</constant>,
2975 except that the link step links
2976 <constant>$(C_OBJS)</constant> with the C
2977 runtime system.</para>
2982 <term><constant>LIBRARY</constant><indexterm><primary>LIBRARY</primary></indexterm></term>
2984 <para>is similar to <constant>HS_PROG</constant>,
2985 except that it links
2986 <constant>$(LIB_OBJS)</constant> to make the
2987 library archive <constant>$(LIBRARY)</constant>,
2988 and <literal>install</literal> installs it in
2989 <constant>$(libdir)</constant>.</para>
2994 <term><constant>LIB_DATA</constant><indexterm><primary>LIB_DATA</primary></indexterm></term>
2996 <para>…</para>
3001 <term><constant>LIB_EXEC</constant><indexterm><primary>LIB_EXEC</primary></indexterm></term>
3003 <para>…</para>
3008 <term><constant>HS_SRCS</constant><indexterm><primary>HS_SRCS</primary></indexterm>, <constant>C_SRCS</constant><indexterm><primary>C_SRCS</primary></indexterm>.</term>
3010 <para>If <constant>HS_SRCS</constant> is defined
3011 and non-empty, a rule for the target
3012 <literal>depend</literal> is included, which generates
3013 dependency information for Haskell programs. Similarly
3014 for <constant>C_SRCS</constant>.</para>
3019 <para>All of these rules are “double-colon” rules,
3023 install :: $(HS_PROG)
3024 ...how to install it...
3027 <para>GNU <command>make</command> treats double-colon rules as
3028 separate entities. If there are several double-colon rules for
3029 the same target it takes each in turn and fires it if its
3030 dependencies say to do so. This means that you can, for
3031 example, define both <constant>HS_PROG</constant> and
3032 <constant>LIBRARY</constant>, which will generate two rules for
3033 <literal>install</literal>. When you type <command>gmake
3034 install</command> both rules will be fired, and both the program
3035 and the library will be installed, just as you wanted.</para>
3038 <sect2 id="sec-subdirs">
3039 <title>Recursion</title>
3040 <indexterm><primary>recursion, in makefiles</primary></indexterm>
3041 <indexterm><primary>Makefile, recursing into subdirectories</primary></indexterm>
3043 <para>In leaf <filename>Makefile</filename>s the variable
3044 <constant>SUBDIRS</constant><indexterm><primary>SUBDIRS</primary></indexterm>
3045 is undefined. In non-leaf <filename>Makefile</filename>s,
3046 <constant>SUBDIRS</constant> is set to the list of
3047 sub-directories that contain subordinate
3048 <filename>Makefile</filename>s. <emphasis>It is up to you to
3049 set <constant>SUBDIRS</constant> in the
3050 <filename>Makefile</filename>.</emphasis> There is no automation
3051 here—<constant>SUBDIRS</constant> is too important to
3054 <para>When <constant>SUBDIRS</constant> is defined,
3055 <filename>target.mk</filename> includes a rather neat rule for
3056 the standard targets (<Xref LinkEnd="sec-standard-targets"> that
3057 simply invokes <command>make</command> recursively in each of
3058 the sub-directories.</para>
3060 <para><emphasis>These recursive invocations are guaranteed to
3061 occur in the order in which the list of directories is specified
3062 in <constant>SUBDIRS</constant>. </emphasis>This guarantee can
3063 be important. For example, when you say <command>gmake
3064 boot</command> it can be important that the recursive invocation
3065 of <command>make boot</command> is done in one sub-directory
3066 (the include files, say) before another (the source files).
3067 Generally, put the most independent sub-directory first, and the
3068 most dependent last.</para>
3071 <sect2 id="sec-ways">
3072 <title>Way management</title>
3073 <indexterm><primary>way management</primary></indexterm>
3075 <para>We sometimes want to build essentially the same system in
3076 several different “ways”. For example, we want to build GHC's
3077 <literal>Prelude</literal> libraries with and without profiling,
3078 so that there is an appropriately-built library archive to link
3079 with when the user compiles his program. It would be possible
3080 to have a completely separate build tree for each such “way”,
3081 but it would be horribly bureaucratic, especially since often
3082 only parts of the build tree need to be constructed in multiple
3086 <filename>target.mk</filename><indexterm><primary>target.mk</primary></indexterm>
3087 contains some clever magic to allow you to build several
3088 versions of a system; and to control locally how many versions
3089 are built and how they differ. This section explains the
3092 <para>The files for a particular way are distinguished by
3093 munging the suffix. The <quote>normal way</quote> is always
3094 built, and its files have the standard suffices
3095 <filename>.o</filename>, <filename>.hi</filename>, and so on.
3096 In addition, you can build one or more extra ways, each
3097 distinguished by a <emphasis>way tag</emphasis>. The object
3098 files and interface files for one of these extra ways are
3099 distinguished by their suffix. For example, way
3100 <literal>mp</literal> has files
3101 <filename>.mp_o</filename> and
3102 <filename>.mp_hi</filename>. Library archives have their
3103 way tag the other side of the dot, for boring reasons; thus,
3104 <filename>libHS_mp.a</filename>.</para>
3106 <para>A <command>make</command> variable called
3107 <constant>way</constant> holds the current way tag.
3108 <emphasis><constant>way</constant> is only ever set on the
3109 command line of <command>gmake</command></emphasis> (usually in
3110 a recursive invocation of <command>gmake</command> by the
3111 system). It is never set inside a
3112 <filename>Makefile</filename>. So it is a global constant for
3113 any one invocation of <command>gmake</command>. Two other
3114 <command>make</command> variables,
3115 <constant>way_</constant> and
3116 <constant>_way</constant> are immediately derived from
3117 <constant>$(way)</constant> and never altered. If
3118 <constant>way</constant> is not set, then neither are
3119 <constant>way_</constant> and
3120 <constant>_way</constant>, and the invocation of
3121 <command>make</command> will build the <quote>normal
3122 way</quote>. If <constant>way</constant> is set, then the other
3123 two variables are set in sympathy. For example, if
3124 <constant>$(way)</constant> is “<literal>mp</literal>”,
3125 then <constant>way_</constant> is set to
3126 “<literal>mp_</literal>” and
3127 <constant>_way</constant> is set to
3128 “<literal>_mp</literal>”. These three variables are
3129 then used when constructing file names.</para>
3131 <para>So how does <command>make</command> ever get recursively
3132 invoked with <constant>way</constant> set? There are two ways
3133 in which this happens:</para>
3137 <para>For some (but not all) of the standard targets, when
3138 in a leaf sub-directory, <command>make</command> is
3139 recursively invoked for each way tag in
3140 <constant>$(WAYS)</constant>. You set
3141 <constant>WAYS</constant> in the
3142 <filename>Makefile</filename> to the list of way tags you
3143 want these targets built for. The mechanism here is very
3144 much like the recursive invocation of
3145 <command>make</command> in sub-directories (<Xref
3146 LinkEnd="sec-subdirs">). It is up to you to set
3147 <constant>WAYS</constant> in your
3148 <filename>Makefile</filename>; this is how you control what
3149 ways will get built.</para>
3153 <para>For a useful collection of targets (such as
3154 <filename>libHS_mp.a</filename>,
3155 <filename>Foo.mp_o</filename>) there is a rule which
3156 recursively invokes <command>make</command> to make the
3157 specified target, setting the <constant>way</constant>
3158 variable. So if you say <command>gmake
3159 Foo.mp_o</command> you should see a recursive
3160 invocation <command>gmake Foo.mp_o way=mp</command>,
3161 and <emphasis>in this recursive invocation the pattern rule
3162 for compiling a Haskell file into a <filename>.o</filename>
3163 file will match</emphasis>. The key pattern rules (in
3164 <filename>suffix.mk</filename>) look like this:
3168 $(HC) $(HC_OPTS) $< -o $@
3175 <para>You can invoke <command>make</command> with a
3176 particular <literal>way</literal> setting yourself, in order
3177 to build files related to a particular
3178 <literal>way</literal> in the current directory. eg.
3184 will build files for the profiling way only in the current
3191 <title>When the canned rule isn't right</title>
3193 <para>Sometimes the canned rule just doesn't do the right thing.
3194 For example, in the <literal>nofib</literal> suite we want the
3195 link step to print out timing information. The thing to do here
3196 is <emphasis>not</emphasis> to define
3197 <constant>HS_PROG</constant> or
3198 <constant>C_PROG</constant>, and instead define a special
3199 purpose rule in your own <filename>Makefile</filename>. By
3200 using different variable names you will avoid the canned rules
3201 being included, and conflicting with yours.</para>
3205 <sect1 id="sec-porting-ghc">
3206 <title>Porting GHC</title>
3208 <para>This section describes how to port GHC to a currenly
3209 unsupported platform. There are two distinct
3210 possibilities:</para>
3214 <para>The hardware architecture for your system is already
3215 supported by GHC, but you're running an OS that isn't
3216 supported (or perhaps has been supported in the past, but
3217 currently isn't). This is the easiest type of porting job,
3218 but it still requires some careful bootstrapping. Proceed to
3219 <xref linkend="sec-booting-from-hc">.</para>
3223 <para>Your system's hardware architecture isn't supported by
3224 GHC. This will be a more difficult port (though by comparison
3225 perhaps not as difficult as porting gcc). Proceed to <xref
3226 linkend="unregisterised-porting">.</para>
3230 <sect2 id="sec-booting-from-hc">
3231 <title>Booting/porting from C (<filename>.hc</filename>) files</title>
3233 <indexterm><primary>building GHC from .hc files</primary></indexterm>
3234 <indexterm><primary>booting GHC from .hc files</primary></indexterm>
3235 <indexterm><primary>porting GHC</primary></indexterm>
3237 <para>Bootstrapping GHC on a system without GHC already
3238 installed is achieved by taking the intermediate C files (known
3239 as HC files) from a GHC compilation on a supported system to the
3240 target machine, and compiling them using gcc to get a working
3243 <para><emphasis>NOTE: GHC version 5.xx is significantly harder
3244 to bootstrap from C than previous versions. We recommend
3245 starting from version 4.08.2 if you need to bootstrap in this
3246 way.</emphasis></para>
3248 <para>HC files are architecture-dependent (but not
3249 OS-dependent), so you have to get a set that were generated on
3250 similar hardware. There may be some supplied on the GHC
3251 download page, otherwise you'll have to compile some up
3252 yourself, or start from <emphasis>unregisterised</emphasis> HC
3253 files - see <xref linkend="unregisterised-porting">.</para>
3255 <para>The following steps should result in a working GHC build
3256 with full libraries:</para>
3260 <para>Unpack the HC files on top of a fresh source tree
3261 (make sure the source tree version matches the version of
3262 the HC files <emphasis>exactly</emphasis>!). This will
3263 place matching <filename>.hc</filename> files next to the
3264 corresponding Haskell source (<filename>.hs</filename> or
3265 <filename>.lhs</filename>) in the compiler subdirectory
3266 <filename>ghc/compiler</filename> and in the libraries
3267 (subdirectories of <filename>hslibs</filename> and
3268 <literal>libraries</literal>).</para>
3272 <para>The actual build process is fully automated by the
3273 <filename>hc-build</filename> script located in the
3274 <filename>distrib</filename> directory. If you eventually
3275 want to install GHC into the directory
3276 <replaceable>dir</replaceable>, the following
3277 command will execute the whole build process (it won't
3278 install yet):</para>
3281 foo% distrib/hc-build --prefix=<replaceable>dir</replaceable>
3283 <indexterm><primary>--hc-build</primary></indexterm>
3285 <para>By default, the installation directory is
3286 <filename>/usr/local</filename>. If that is what you want,
3287 you may omit the argument to <filename>hc-build</filename>.
3288 Generally, any option given to <filename>hc-build</filename>
3289 is passed through to the configuration script
3290 <filename>configure</filename>. If
3291 <filename>hc-build</filename> successfully completes the
3292 build process, you can install the resulting system, as
3302 <sect2 id="unregisterised-porting">
3303 <title>Porting GHC to a new architecture</title>
3305 <para>The first step in porting to a new architecture is to get
3306 an <firstterm>unregisterised</firstterm> build working. An
3307 unregisterised build is one that compiles via vanilla C only.
3308 By contrast, a registerised build uses the following
3309 architecture-specific hacks for speed:</para>
3313 <para>Global register variables: certain abstract machine
3314 <quote>registers</quote> are mapped to real machine
3315 registers, depending on how many machine registers are
3317 <filename>ghc/includes/MachRegs.h</filename>).</para>
3321 <para>Assembly-mangling: when compiling via C, we feed the
3322 assembly generated by gcc though a Perl script known as the
3323 <firstterm>mangler</firstterm> (see
3324 <filename>ghc/driver/mangler/ghc-asm.lprl</filename>). The
3325 mangler rearranges the assembly to support tail-calls and
3326 various other optimisations.</para>
3330 <para>In an unregisterised build, neither of these hacks are
3331 used — the idea is that the C code generated by the
3332 compiler should compile using gcc only. The lack of these
3333 optimisations costs about a factor of two in performance, but
3334 since unregisterised compilation is usually just a step on the
3335 way to a full registerised port, we don't mind too much.</para>
3338 <title>Building an unregisterised port</title>
3340 <para>The first step is to get some unregisterised HC files.
3341 Either (a) download them from the GHC site (if there are
3342 some available for the right version of GHC), or
3343 (b) build them yourself on any machine with a working
3344 GHC. If at all possible this should be a machine with the
3345 same word size as the target.</para>
3347 <para>There is a script available which should automate the
3348 process of doing the 2-stage bootstrap necessary to get the
3349 unregisterised HC files - it's available in <ulink
3350 url="http://cvs.haskell.org/cgi-bin/cvsweb.cgi/fptools/distrib/cross-port"><filename>fptools/distrib/cross-port</filename></ulink>
3353 <para>Now take these unregisterised HC files to the target
3354 platform and bootstrap a compiler from them as per the
3355 instructions in <xref linkend="sec-booting-from-hc">. In
3356 <filename>build.mk</filename>, you need to tell the build
3357 system that the compiler you're building is
3358 (a) unregisterised itself, and (b) builds
3359 unregisterised binaries. This varies depending on the GHC
3360 version you're bootstraping:</para>
3363 # build.mk for GHC 4.08.x
3364 GhcWithRegisterised=NO
3368 # build.mk for GHC 5.xx
3369 GhcUnregisterised=YES
3372 <para>Version 5.xx only: use the option
3373 <option>--enable-hc-boot-unregisterised</option> instead of
3374 <option>--enable-hc-boot</option> when running
3375 <filename>./configure</filename>.</para>
3377 <para>The build may not go through cleanly. We've tried to
3378 stick to writing portable code in most parts of the compiler,
3379 so it should compile on any POSIXish system with gcc, but in
3380 our experience most systems differ from the standards in one
3381 way or another. Deal with any problems as they arise - if you
3382 get stuck, ask the experts on
3383 <email>glasgow-haskell-users@haskell.org</email>.</para>
3385 <para>Once you have the unregisterised compiler up and
3386 running, you can use it to start a registerised port. The
3387 following sections describe the various parts of the system
3388 that will need architecture-specific tweaks in order to get a
3389 registerised build going.</para>
3391 <para>Lots of useful information about the innards of GHC is
3392 available in the <ulink
3393 url="http://www.cse.unsw.edu.au/~chak/haskell/ghc/comm/">GHC
3394 Commentary</ulink>, which might be helpful if you run into
3395 some code which needs tweaking for your system.</para>
3399 <title>Porting the RTS</title>
3401 <para>The following files need architecture-specific code for a
3402 registerised build:</para>
3406 <term><filename>ghc/includes/MachRegs.h</filename></term>
3407 <indexterm><primary><filename>MachRegs.h</filename></primary>
3410 <para>Defines the STG-register to machine-register
3411 mapping. You need to know your platform's C calling
3412 convention, and which registers are generally available
3413 for mapping to global register variables. There are
3414 plenty of useful comments in this file.</para>
3418 <term><filename>ghc/includes/TailCalls.h</filename></term>
3419 <indexterm><primary><filename>TailCalls.h</filename></primary>
3422 <para>Macros that cooperate with the mangler (see <xref
3423 linkend="sec-mangler">) to make proper tail-calls
3428 <term><filename>ghc/rts/Adjustor.c</filename></term>
3429 <indexterm><primary><filename>Adjustor.c</filename></primary>
3433 <literal>foreign import "wrapper"</literal>
3435 <literal>foreign export dynamic</literal>).
3436 Not essential for getting GHC bootstrapped, so this file
3437 can be deferred until later if necessary.</para>
3441 <term><filename>ghc/rts/StgCRun.c</filename></term>
3442 <indexterm><primary><filename>StgCRun.c</filename></primary>
3445 <para>The little assembly layer between the C world and
3446 the Haskell world. See the comments and code for the
3447 other architectures in this file for pointers.</para>
3451 <term><filename>ghc/rts/MBlock.h</filename></term>
3452 <term><filename>ghc/rts/MBlock.c</filename></term>
3453 <indexterm><primary><filename>MBlock.h</filename></primary>
3455 <indexterm><primary><filename>MBlock.c</filename></primary>
3458 <para>These files are really OS-specific rather than
3459 architecture-specific. In <filename>MBlock.h</filename>
3460 is specified the absolute location at which the RTS
3461 should try to allocate memory on your platform (try to
3462 find an area which doesn't conflict with code or dynamic
3463 libraries). In <filename>Mblock.c</filename> you might
3464 need to tweak the call to <literal>mmap()</literal> for
3471 <sect3 id="sec-mangler">
3472 <title>The mangler</title>
3474 <para>The mangler is an evil Perl-script that rearranges the
3475 assembly code output from gcc to do two main things:</para>
3479 <para>Remove function prologues and epilogues, and all
3480 movement of the C stack pointer. This is to support
3481 tail-calls: every code block in Haskell code ends in an
3482 explicit jump, so we don't want the C-stack overflowing
3483 while we're jumping around between code blocks.</para>
3486 <para>Move the <firstterm>info table</firstterm> for a
3487 closure next to the entry code for that closure. In
3488 unregisterised code, info tables contain a pointer to the
3489 entry code, but in registerised compilation we arrange
3490 that the info table is shoved right up against the entry
3491 code, and addressed backwards from the entry code pointer
3492 (this saves a word in the info table and an extra
3493 indirection when jumping to the closure entry
3498 <para>The mangler is abstracted to a certain extent over some
3499 architecture-specific things such as the particular assembler
3500 directives used to herald symbols. Take a look at the
3501 definitions for other architectures and use these as a
3502 starting point.</para>
3506 <title>The native code generator</title>
3508 <para>The native code generator isn't essential to getting a
3509 registerised build going, but it's a desirable thing to have
3510 because it can cut compilation times in half. The native code
3511 generator is described in some detail in the <ulink
3512 url="http://www.cse.unsw.edu.au/~chak/haskell/ghc/comm/">GHC
3513 commentary</ulink>.</para>
3519 <para>To support GHCi, you need to port the dynamic linker
3520 (<filename>fptools/ghc/rts/Linker.c</filename>). The linker
3521 currently supports the ELF and PEi386 object file formats - if
3522 your platform uses one of these then you probably don't have
3523 to do anything except fiddle with the
3524 <literal>#ifdef</literal>s at the top of
3525 <filename>Linker.c</filename> to tell it about your OS.</para>
3527 <para>If your system uses a different object file format, then
3528 you have to write a linker — good luck!</para>
3534 <sect1 id="sec-build-pitfalls">
3535 <title>Known pitfalls in building Glasgow Haskell
3537 <indexterm><primary>problems, building</primary></indexterm>
3538 <indexterm><primary>pitfalls, in building</primary></indexterm>
3539 <indexterm><primary>building pitfalls</primary></indexterm></title>
3542 WARNINGS about pitfalls and known “problems”:
3551 One difficulty that comes up from time to time is running out of space
3552 in <literal>TMPDIR</literal>. (It is impossible for the configuration stuff to
3553 compensate for the vagaries of different sysadmin approaches to temp
3555 <indexterm><primary>tmp, running out of space in</primary></indexterm>
3557 The quickest way around it is <command>setenv TMPDIR /usr/tmp</command><indexterm><primary>TMPDIR</primary></indexterm> or
3558 even <command>setenv TMPDIR .</command> (or the equivalent incantation with your shell
3561 The best way around it is to say
3564 export TMPDIR=<dir>
3567 in your <filename>build.mk</filename> file.
3568 Then GHC and the other <literal>fptools</literal> programs will use the appropriate directory
3577 In compiling some support-code bits, e.g., in <filename>ghc/rts/gmp</filename> and even
3578 in <filename>ghc/lib</filename>, you may get a few C-compiler warnings. We think these
3586 When compiling via C, you'll sometimes get “warning: assignment from
3587 incompatible pointer type” out of GCC. Harmless.
3594 Similarly, <command>ar</command>chiving warning messages like the following are not
3598 ar: filename GlaIOMonad__1_2s.o truncated to GlaIOMonad_
3599 ar: filename GlaIOMonad__2_2s.o truncated to GlaIOMonad_
3609 In compiling the compiler proper (in <filename>compiler/</filename>), you <emphasis>may</emphasis>
3610 get an “Out of heap space” error message. These can vary with the
3611 vagaries of different systems, it seems. The solution is simple:
3618 If you're compiling with GHC 4.00 or later, then the
3619 <emphasis>maximum</emphasis> heap size must have been reached. This
3620 is somewhat unlikely, since the maximum is set to 64M by default.
3621 Anyway, you can raise it with the
3622 <option>-optCrts-M<size></option> flag (add this flag to
3623 <constant><module>_HC_OPTS</constant>
3624 <command>make</command> variable in the appropriate
3625 <filename>Makefile</filename>).
3632 For GHC < 4.00, add a suitable <option>-H</option> flag to the <filename>Makefile</filename>, as
3641 and try again: <command>gmake</command>. (see <Xref LinkEnd="sec-suffix"> for information about
3642 <constant><module>_HC_OPTS</constant>.)
3644 Alternatively, just cut to the chase:
3648 % make EXTRA_HC_OPTS=-optCrts-M128M
3657 If you try to compile some Haskell, and you get errors from GCC about
3658 lots of things from <filename>/usr/include/math.h</filename>, then your GCC was
3659 mis-installed. <command>fixincludes</command> wasn't run when it should've been.
3661 As <command>fixincludes</command> is now automagically run as part of GCC installation,
3662 this bug also suggests that you have an old GCC.
3670 You <emphasis>may</emphasis> need to re-<command>ranlib</command><indexterm><primary>ranlib</primary></indexterm> your libraries (on Sun4s).
3674 % cd $(libdir)/ghc-x.xx/sparc-sun-sunos4
3675 % foreach i ( `find . -name '*.a' -print` ) # or other-shell equiv...
3677 ? # or, on some machines: ar s $i
3682 We'd be interested to know if this is still necessary.
3690 GHC's sources go through <command>cpp</command> before being compiled, and <command>cpp</command> varies
3691 a bit from one Unix to another. One particular gotcha is macro calls
3696 SLIT("Hello, world")
3700 Some <command>cpp</command>s treat the comma inside the string as separating two macro
3701 arguments, so you get
3705 :731: macro `SLIT' used with too many (2) args
3709 Alas, <command>cpp</command> doesn't tell you the offending file!
3711 Workaround: don't put weird things in string args to <command>cpp</command> macros.
3722 <Sect1 id="winbuild"><Title>Notes for building under Windows</Title>
3725 This section summarises how to get the utilities you need on your
3726 Win95/98/NT/2000 machine to use CVS and build GHC. Similar notes for
3727 installing and running GHC may be found in the user guide. In general,
3728 Win95/Win98 behave the same, and WinNT/Win2k behave the same.
3729 You should read the GHC installation guide sections on Windows (in the user
3730 guide) before continuing to read these notes.
3734 <sect2><Title>Cygwin and MinGW</Title>
3736 <para> The Windows situation for building GHC is rather confusing. This section
3737 tries to clarify, and to establish terminology.</para>
3739 <sect3 id="ghc-mingw"><title>GHC-mingw</title>
3741 <para> <ulink url="http://www.mingw.org">MinGW (Minimalist GNU for Windows)</ulink>
3742 is a collection of header
3743 files and import libraries that allow one to use <command>gcc</command> and produce
3744 native Win32 programs that do not rely on any third-party DLLs. The
3745 current set of tools include GNU Compiler Collection (<command>gcc</command>), GNU Binary
3746 Utilities (Binutils), GNU debugger (Gdb), GNU make, and a assorted
3749 <para>The GHC that we distribute includes, inside the distribution itself, the MinGW <command>gcc</command>,
3750 <command>as</command>, <command>ld</command>, and a bunch of input/output libraries.
3751 GHC compiles Haskell to C (or to
3752 assembly code), and then invokes these MinGW tools to generate an executable binary.
3753 The resulting binaries can run on any Win32 system.
3755 <para> We will call a GHC that targets MinGW in this way <emphasis>GHC-mingw</emphasis>.</para>
3757 <para> The down-side of GHC-mingw is that the MinGW libraries do not support anything like the full
3758 Posix interface. So programs compiled with GHC-mingw cannot import the (Haskell) Posix
3759 library; they have to do
3760 their input output using standard Haskell I/O libraries, or native Win32 bindings.
3764 <sect3 id="ghc-cygwin"><title>GHC-cygwin</title>
3766 <para>There <emphasis>is</emphasis> a way to get the full Posix interface, which is to use Cygwin.
3767 <ulink url="http://www.cygwin.com">Cygwin</ulink> is a complete Unix simulation that runs on Win32.
3768 Cygwin comes with a shell, and all the usual Unix commands: <command>mv</command>, <command>rm</command>,
3769 <command>ls</command>, plus of course <command>gcc</command>, <command>ld</command> and so on.
3770 A C program compiled with the Cygwin <command>gcc</command> certainly can use all of Posix.
3772 <para>So why doesn't GHC use the Cygwin <command>gcc</command> and libraries? Because
3773 Cygwin comes with a DLL <emphasis>that must be linked with every runnable Cygwin-compiled program</emphasis>.
3774 A program compiled by the Cygwin tools cannot run at all unless Cygwin is installed.
3775 If GHC targeted Cygwin, users would have to install Cygwin just to run the Haskell programs
3776 that GHC compiled; and the Cygwin DLL would have to be in the DLL load path.
3777 Worse, Cygwin is a moving target. The name of the main DLL, <literal>cygwin1.dll</literal>
3778 does not change, but the implementation certainly does. Even the interfaces to functions
3779 it exports seem to change occasionally. So programs compiled by GHC might only run with
3780 particular versions of Cygwin. All of this seems very undesirable.
3783 Nevertheless, it is certainly possible to build a version of GHC that targets Cygwin;
3784 we will call that <emphasis>GHC-cygwin</emphasis>. The up-side of GHC-cygwin is
3785 that Haskell programs compiled by GHC-cygwin can import the (Haskell) Posix library.
3789 <sect3><title>Summary</title>
3791 <para>Notice that "GHC-mingw" means "GHC that <emphasis>targets</emphasis> MinGW". It says nothing about
3792 how that GHC was <emphasis>built</emphasis>. It is entirely possible to have a GHC-mingw that was built
3793 by compiling GHC's Haskell sources with a GHC-cygwin, or vice versa.</para>
3795 <para>We distribute only a GHC-mingw built by a GHC-mingw; supporting
3796 GHC-cygwin too is beyond our resources. The GHC we distribute
3797 therefore does not require Cygwin to run, nor do the programs it
3798 compiles require Cygwin.</para>
3800 <para>The instructions that follow describe how to build GHC-mingw. It is
3801 possible to build GHC-cygwin, but it's not a supported route, and the build system might
3806 <Sect2><Title>Installing and configuring Cygwin</Title>
3808 <para>You don't need Cygwin to <emphasis>use</emphasis> GHC,
3809 but you do need it to <emphasis>build</emphasis> GHC.</para>
3811 <para> Install Cygwin from <ulink url="http://www.cygwin.com/">http://www.cygwin.com/</ulink>.
3812 The installation process is straightforward; we install it in <Filename>c:/cygwin</Filename>.
3813 Both <command>cvs</command> and <command>ssh</command>
3814 come with Cygwin, but you'll need them, so make sure you select them when running
3815 the Cygwin installer.
3818 <para> Now set the following user environment variables:
3821 <listitem><para> Add <filename>c:/cygwin/bin</filename> and <filename>c:/cygwin/usr/bin</filename> to your
3822 <constant>PATH</constant></para></listitem>
3826 Set <constant>MAKE_MODE</constant> to <Literal>UNIX</Literal>. If you
3827 don't do this you get very weird messages when you type
3828 <Command>make</Command>, such as:
3830 /c: /c: No such file or directory
3835 <listitem><para> Set <constant>SHELL</constant> to
3836 <Filename>c:/cygwin/bin/sh</Filename>. When you invoke a shell in Emacs, this
3837 <constant>SHELL</constant> is what you get.
3840 <listitem><para> Set <constant>HOME</constant> to point to your
3841 home directory. This is where, for example,
3842 <command>bash</command> will look for your <filename>.bashrc</filename>
3843 file. Ditto <command>emacs</command> looking for <filename>.emacsrc</filename>
3849 There are a few other things to do:
3853 Some script files used in the make system start with "<Command>#!/bin/perl</Command>",
3854 (and similarly for <Command>bash</Command>). Notice the hardwired path!
3855 So you need to ensure that your <Filename>/bin</Filename> directory has the following
3858 <listitem> <para><Command>sh</Command></para></listitem>
3859 <listitem> <para><Command>perl</Command></para></listitem>
3860 <listitem> <para><Command>cat</Command></para></listitem>
3862 All these come in Cygwin's <Filename>bin</Filename> directory, which you probably have
3863 installed as <Filename>c:/cygwin/bin</Filename>. By default Cygwin mounts "<Filename>/</Filename>" as
3864 <Filename>c:/cygwin</Filename>, so if you just take the defaults it'll all work ok.
3865 (You can discover where your Cygwin
3866 root directory <Filename>/</Filename> is by typing <Command>mount</Command>).
3867 Provided <Filename>/bin</Filename> points to the Cygwin <Filename>bin</Filename>
3868 directory, there's no need to copy anything.
3874 By default, cygwin provides the command shell <filename>ash</filename>
3875 as <filename>sh.exe</filename>. It has a couple of 'issues', so
3876 in your <filename>/bin</filename> directory, make sure that <filename>
3877 bash.exe</filename> is also provided as <filename>sh.exe</filename>
3878 (i.e. overwrite the old <filename>sh.exe</filename> with a copy of
3879 <filename>bash.exe</filename>).
3885 <para>Finally, here are some things to be aware of when using Cygwin:
3887 <listitem> <para>Cygwin doesn't deal well with filenames that include
3888 spaces. "<filename>Program Files</filename>" and "<filename>Local files</filename>" are
3892 <listitem> <para> Cygwin implements a symbolic link as a text file with some
3893 magical text in it. So other programs that don't use Cygwin's
3894 I/O libraries won't recognise such files as symlinks.
3895 In particular, programs compiled by GHC are meant to be runnable
3896 without having Cygwin, so they don't use the Cygwin library, so
3897 they don't recognise symlinks.
3901 Win32 has a <command>find</command> command which is not the same as Cygwin's find.
3902 You will probably discover that the Win32 <command>find</command> appears in your <constant>PATH</constant>
3903 before the Cygwin one, because it's in the <emphasis>system</emphasis> <constant>PATH</constant>
3904 environment variable, whereas you have probably modified the <emphasis>user</emphasis> <constant>PATH</constant>
3905 variable. You can always invoke <command>find</command> with an absolute path, or rename it.
3912 <Sect2><Title>Other things you need to install</Title>
3914 <para>You have to install the following other things to build GHC:
3918 Install an executable GHC, from <ulink url="http://www.haskell.org/ghc">http://www.haskell.org/ghc</ulink>.
3919 This is what you will use to compile GHC. Add it in your
3920 <constant>PATH</constant>: the installer tells you the path element
3921 you need to add upon completion.
3927 Install an executable Happy, from <ulink url="http://www.haskell.org/happy">http://www.haskell.org/happy</ulink>.
3928 Happy is a parser generator used to compile the Haskell grammar. Add it in your
3929 <constant>PATH</constant>.
3935 <para>GHC uses the <emphasis>mingw</emphasis> C compiler to
3936 generate code, so you have to install that. Just pick up a mingw bundle at
3937 <ulink url="http://www.mingw.org/">http://www.mingw.org/</ulink>.
3938 We install it in <filename>c:/mingw</filename>.
3944 <para> Finally, check out a copy of GHC sources from
3945 the CVS repository, following the instructions above (<xref linkend="cvs-access">).
3952 <Sect2><Title>Building GHC</Title>
3955 Now go read the documentation above on building from source (<xref linkend="sec-building-from-source">);
3956 the bullets below only tell
3957 you about Windows-specific wrinkles.</para>
3961 Run <Command>autoconf</Command> both in <filename>fptools</filename>
3962 and in <filename>fptools/ghc</filename>. If you omit the latter step you'll
3963 get an error when you run <filename>./configure</filename>:
3966 creating mk/config.h
3967 mk/config.h is unchanged
3969 running /bin/sh ./configure --cache-file=.././config.cache --srcdir=.
3970 ./configure: ./configure: No such file or directory
3971 configure: error: ./configure failed for ghc
3976 <listitem> <para><command>autoconf</command> seems to create the file <filename>configure</filename>
3977 read-only. So if you need to run autoconf again (which I sometimes do for safety's sake),
3980 /usr/bin/autoconf: cannot create configure: permission denied
3982 Solution: delete <filename>configure</filename> first.
3987 You either need to add <filename>ghc</filename> to your
3988 <constant>PATH</constant> before you invoke
3989 <Command>configure</Command>, or use the <Command>configure</Command>
3990 option <option>--with-ghc=c:/ghc/ghc-some-version/bin/ghc</option>.
3996 After <command>autoconf</command> run <command>./configure</command> in
3997 <filename>fptools/</filename> thus:
4000 ./configure --host=i386-unknown-mingw32 --with-gcc=/mingw/bin/gcc
4002 This is the point at which you specify that you are building GHC-mingw
4003 (see <xref linkend="ghc-mingw">).
4005 Both these options are important! It's possible to get into
4006 trouble using the wrong C compiler!</para>
4009 If you want to build GHC-cygwin (<xref linkend="ghc-cygwin">)
4010 you'll have to do something more like:
4012 ./configure --with-gcc=...the Cygwin gcc...
4017 <listitem><para> Do not attempt to build the documentation.
4018 It needs all kinds of wierd Jade stuff that we haven't worked out for
4019 Win32.</para></listitem>