ffi-chap.xml 20.5 KB
Newer Older
<?xml version="1.0" encoding="iso-8859-1"?>
rrt's avatar
rrt committed
2 3
<!-- FFI docs as a chapter -->

<chapter id="ffi">
5 6 7
Foreign function interface (FFI)

chak's avatar
chak committed
  <para>GHC (mostly) conforms to the Haskell 98 Foreign Function Interface
10 11 12
  Addendum 1.0, whose definition is available from <ulink url="http://haskell.org/"><literal>http://haskell.org/</literal></ulink>.</para>

  <para>To enable FFI support in GHC, give the <option>-fffi</option><indexterm><primary><option>-fffi</option></primary>
Simon Marlow's avatar
Simon Marlow committed
    </indexterm> flag, or
14 15 16 17
the <option>-fglasgow-exts</option><indexterm><primary><option>-fglasgow-exts</option></primary>
    </indexterm> flag which implies <option>-fffi</option>

Simon Marlow's avatar
Simon Marlow committed
18 19 20 21
  <para>GHC implements a number of GHC-specific extensions to the FFI
    Addendum.  These extensions are described in <xref linkend="ffi-ghcexts" />, but please note that programs using
    these features are not portable.  Hence, these features should be
    avoided where possible.</para>

  <para>The FFI libraries are documented in the accompanying library
Simon Marlow's avatar
Simon Marlow committed
24 25
  documentation; see for example the
    <ulink url="../libraries/base/Control-Concurrent.html"><literal>Foreign</literal></ulink> module.</para>
chak's avatar
chak committed

  <sect1 id="ffi-ghcexts">
chak's avatar
chak committed
28 29 30
    <title>GHC extensions to the FFI Addendum</title>

    <para>The FFI features that are described in this section are specific to
Simon Marlow's avatar
Simon Marlow committed
    GHC.  Your code will not be portable to other compilers if you use them.</para>
chak's avatar
chak committed
32 33 34 35 36 37 38 39 40 41 42 43 44

      <title>Unboxed types</title>

      <para>The following unboxed types may be used as basic foreign types
      (see FFI Addendum, Section 3.2): <literal>Int#</literal>,
      <literal>Word#</literal>, <literal>Char#</literal>,
      <literal>Float#</literal>, <literal>Double#</literal>,
      <literal>Addr#</literal>, <literal>StablePtr# a</literal>,
      <literal>MutableByteArray#</literal>, <literal>ForeignObj#</literal>,
      and <literal>ByteArray#</literal>.</para>

    <sect2 id="ffi-newtype-io">
46 47 48 49 50 51 52 53 54 55 56 57
      <title>Newtype wrapping of the IO monad</title>
      <para>The FFI spec requires the IO monad to appear in various  places,
      but it can sometimes be convenient to wrap the IO monad in a
      <literal>newtype</literal>, thus:
  newtype MyIO a = MIO (IO a)
     (A reason for doing so might be to prevent the programmer from
	calling arbitrary IO procedures in some part of the program.)
<para>The Haskell FFI already specifies that arguments and results of
foreign imports and exports will be automatically unwrapped if they are 
newtypes (Section 3.2 of the FFI addendum).  GHC extends the FFI by automatically unwrapping any newtypes that
59 60 61 62 63 64 65 66 67 68
wrap the IO monad itself.
More precisely, wherever the FFI specification requires an IO type, GHC will
accept any newtype-wrapping of an IO type.  For example, these declarations are
   foreign import foo :: Int -> MyIO Int
   foreign import "dynamic" baz :: (Int -> MyIO Int) -> CInt -> MyIO Int
chak's avatar
chak committed

  <sect1 id="ffi-ghc">
72 73 74 75 76
    <title>Using the FFI with GHC</title>

    <para>The following sections also give some hints and tips on the
    use of the foreign function interface in GHC.</para>

    <sect2 id="foreign-export-ghc">
      <title>Using <literal>foreign export</literal> and <literal>foreign import ccall "wrapper"</literal> with GHC</title>
79 80

      <indexterm><primary><literal>foreign export
      </literal></primary><secondary>with GHC</secondary>
82 83 84

      <para>When GHC compiles a module (say <filename>M.hs</filename>)
85 86
      which uses <literal>foreign export</literal> or 
      <literal>foreign import "wrapper"</literal>, it generates two
87 88 89
      additional files, <filename>M_stub.c</filename> and
      <filename>M_stub.h</filename>.  GHC will automatically compile
      <filename>M_stub.c</filename> to generate
90 91
      <filename>M_stub.o</filename> at the same time.</para>

92 93 94 95 96 97 98 99 100
      <para>For a plain <literal>foreign export</literal>, the file
      <filename>M_stub.h</filename> contains a C prototype for the
      foreign exported function, and <filename>M_stub.c</filename>
      contains its definition.  For example, if we compile the
      following module:</para>

module Foo where

chak's avatar
chak committed
foreign export ccall foo :: Int -> IO Int
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121

foo :: Int -> IO Int
foo n = return (length (f n))

f :: Int -> [Int]
f 0 = []
f n = n:(f (n-1))</programlisting>

      <para>Then <filename>Foo_stub.h</filename> will contain
      something like this:</para>

#include "HsFFI.h"
extern HsInt foo(HsInt a0);</programlisting>

      <para>and <filename>Foo_stub.c</filename> contains the
      compiler-generated definition of <literal>foo()</literal>.  To
      invoke <literal>foo()</literal> from C, just <literal>#include
      "Foo_stub.h"</literal> and call <literal>foo()</literal>.</para>

122 123 124 125 126
      <para>The <filename>foo_stub.c</filename> and
	<filename>foo_stub.h</filename> files can be redirected using the
	<option>-stubdir</option> option; see <xref linkend="options-output"

Simon Marlow's avatar
Simon Marlow committed
127 128 129 130 131 132 133
      <para>When linking the program, remember to include
        <filename>M_stub.o</filename> in the final link command line, or
        you'll get link errors for the missing function(s) (this isn't
        necessary when building your program with <literal>ghc
        &ndash;&ndash;make</literal>, as GHC will automatically link in the
        correct bits).</para>

      <sect3 id="using-own-main"> 
135 136 137 138 139 140
	<title>Using your own <literal>main()</literal></title>

	<para>Normally, GHC's runtime system provides a
	<literal>main()</literal>, which arranges to invoke
	<literal>Main.main</literal> in the Haskell program.  However,
	you might want to link some Haskell code into a program which
ross's avatar
ross committed
	has a main function written in another language, say C.  In
142 143 144 145 146 147 148 149
	order to do this, you have to initialize the Haskell runtime
	system explicitly.</para>

	<para>Let's take the example from above, and invoke it from a
	standalone C program.  Here's the C code:</para>

#include &lt;stdio.h&gt;
#include "HsFFI.h"

152 153 154
#include "foo_stub.h"

extern void __stginit_Foo ( void );
159 160 161 162 163

int main(int argc, char *argv[])
  int i;

164 165 166 167
  hs_init(&amp;argc, &amp;argv);

  for (i = 0; i &lt; 5; i++) {
170 171 172
    printf("%d\n", foo(2500));

174 175 176
  return 0;

177 178 179 180 181 182
	<para>We've surrounded the GHC-specific bits with
	<literal>#ifdef __GLASGOW_HASKELL__</literal>; the rest of the
	code should be portable across Haskell implementations that
	support the FFI standard.</para>

	<para>The call to <literal>hs_init()</literal>
183 184
	initializes GHC's runtime system.  Do NOT try to invoke any
	Haskell functions before calling
Simon Marlow's avatar
Simon Marlow committed
	<literal>hs_init()</literal>: bad things will
186 187
	undoubtedly happen.</para>

Simon Marlow's avatar
Simon Marlow committed
	<para>We pass references to <literal>argc</literal> and
	<literal>argv</literal> to <literal>hs_init()</literal>
190 191 192 193
	so that it can separate out any arguments for the RTS
	(i.e. those arguments between

194 195 196 197 198 199 200 201 202 203 204 205 206 207
	<para>Next, we call
	</indexterm>, a GHC-specific interface which is required to
	initialise the Haskell modules in the program.  The argument
	to <function>hs_add_root</function> should be the name of the
	initialization function for the "root" module in your program
	- in other words, the module which directly or indirectly
	imports all the other Haskell modules in the program.  In a
	standalone Haskell program the root module is normally
	<literal>Main</literal>, but when you are using Haskell code
	from a library it may not be.  If your program has multiple
	root modules, then you can call
	<function>hs_add_root</function> multiple times, one for each
	root.  The name of the initialization function for module
208 209 210
	<replaceable>M</replaceable> is
	<literal>__stginit_<replaceable>M</replaceable></literal>, and
	it may be declared as an external function symbol as in the
211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245
	code above.  Note that the symbol name should be transformed
	according to the Z-encoding:</para>

	<tgroup cols="2" align="left" colsep="1" rowsep="1">
246 247

	<para>After we've finished invoking our Haskell functions, we
Simon Marlow's avatar
Simon Marlow committed
248 249
	can call <literal>hs_exit()</literal>, which terminates the

251 252 253 254 255 256 257 258
	<para>There can be multiple calls to
	<literal>hs_init()</literal>, but each one should be matched
	by one (and only one) call to
	<literal>hs_exit()</literal><footnote><para>The outermost
	<literal>hs_exit()</literal> will actually de-initialise the
	system.  NOTE that currently GHC's runtime cannot reliably
	re-initialise after this has happened.</para>
259 260 261 262 263 264 265

	<para>NOTE: when linking the final program, it is normally
	easiest to do the link using GHC, although this isn't
	essential.  If you do use GHC, then don't forget the flag
	  </indexterm>, otherwise GHC will try to link
	to the <literal>Main</literal> Haskell module.</para>
266 267

Simon Marlow's avatar
Simon Marlow committed
268 269 270
      <sect3 id="ffi-library">
        <title>Making a Haskell library that can be called from foreign

Simon Marlow's avatar
Simon Marlow committed
272 273 274 275
        <para>The scenario here is much like in <xref linkend="using-own-main"
            />, except that the aim is not to link a complete program, but to
          make a library from Haskell code that can be deployed in the same
          way that you would deploy a library of C code.</para>

Simon Marlow's avatar
Simon Marlow committed
277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302
        <para>The main requirement here is that the runtime needs to be
          initialized before any Haskell code can be called, so your library
          should provide initialisation and deinitialisation entry points,
          implemented in C or C++.  For example:</para>

 HsBool mylib_init(void){
   int argc = ...
   char *argv[] = ...

   // Initialize Haskell runtime
   hs_init(&amp;argc, &amp;argv);

   // Tell Haskell about all root modules

   // do any other initialization here and
   // return false if there was a problem
   return HS_BOOL_TRUE;

 void mylib_end(void){

        <para>The initialisation routine, <literal>mylib_init</literal>, calls
Simon Marlow's avatar
Simon Marlow committed
304 305
          <literal>hs_init()</literal> and <literal>hs_add_root()</literal> as
          normal to initialise the Haskell runtime, and the corresponding
          deinitialisation function <literal>mylib_end()</literal> calls
Simon Marlow's avatar
Simon Marlow committed
307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348
          <literal>hs_exit()</literal> to shut down the runtime.</para>

      <sect3 id="hs-exit">
        <title>On the use of <literal>hs_exit()</literal></title>

        <para><literal>hs_exit()</literal> normally causes the termination of
          any running Haskell threads in the system, and when
          <literal>hs_exit()</literal> returns, there will be no more Haskell
          threads running.  The runtime will then shut down the system in an
          orderly way, generating profiling
          output and statistics if necessary, and freeing all the memory it

        <para>It isn't always possible to terminate a Haskell thread forcibly:
          for example, the thread might be currently executing a foreign call,
          and we have no way to force the foreign call to complete.  What's
          more, the runtime must
          assume that in the worst case the Haskell code and runtime are about
          to be removed from memory (e.g. if this is a <link linkend="win32-dlls">Windows DLL</link>,
          <literal>hs_exit()</literal> is normally called before unloading the
          DLL).  So <literal>hs_exit()</literal> <emphasis>must</emphasis> wait
          until all outstanding foreign calls return before it can return

        <para>The upshot of this is that if you have Haskell threads that are
          blocked in foreign calls, then <literal>hs_exit()</literal> may hang
          (or possibly busy-wait) until the calls return.  Therefore it's a
          good idea to make sure you don't have any such threads in the system
          when calling <literal>hs_exit()</literal>.  This includes any threads
          doing I/O, because I/O may (or may not, depending on the
          type of I/O and the platform) be implemented using blocking foreign

        <para>The GHC runtime treats program exit as a special case, to avoid
          the need to wait for blocked threads when a standalone
          executable exits.  Since the program and all its threads are about to
          terminate at the same time that the code is removed from memory, it
          isn't necessary to ensure that the threads have exited first.
          (Unofficially, if you want to use this fast and loose version of
          <literal>hs_exit()</literal>, then call
          <literal>shutdownHaskellAndExit()</literal> instead).</para> 
352 353 354 355 356 357
    <sect2 id="glasgow-foreign-headers">
      <title>Using function headers</title>

      <indexterm><primary>C calls, function headers</primary></indexterm>

      <para>When generating C (using the <option>-fvia-C</option>
Simon Marlow's avatar
Simon Marlow committed
      flag), one can assist the C compiler in detecting type
      errors by using the <option>-&num;include</option> directive
      (<xref linkend="options-C-compiler"/>) to provide
361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376
      <filename>.h</filename> files containing function

      <para>For example,</para>

#include "HsFFI.h"

void         initialiseEFS (HsInt size);
HsInt        terminateEFS (void);
HsForeignObj emptyEFS(void);
HsForeignObj updateEFS (HsForeignObj a, HsInt i, HsInt x);
HsInt        lookupEFS (HsForeignObj a, HsInt i);

      <para>The types <literal>HsInt</literal>,
chak's avatar
chak committed
377 378
      <literal>HsForeignObj</literal> etc. are described in the H98 FFI
379 380 381 382 383 384 385 386

      <para>Note that this approach is only
      <emphasis>essential</emphasis> for returning
      <literal>float</literal>s (or if <literal>sizeof(int) !=
      sizeof(int *)</literal> on your architecture) but is a Good
      Thing for anyone who cares about writing solid code.  You're
      crazy not to do it.</para>

387 388 389 390 391 392 393 394
What if you are importing a module from another package, and
a cross-module inlining exposes a foreign call that needs a supporting
<option>-&num;include</option>?  If the imported module is from the same package as
the module being compiled, you should supply all the <option>-&num;include</option>
that you supplied when compiling the imported module.  If the imported module comes
from another package, you won't necessarily know what the appropriate 
<option>-&num;include</option> options are; but they should be in the package 
Simon Marlow's avatar
Simon Marlow committed
395 396 397 398
configuration, which GHC knows about.  So if you are building a package using
        Cabal, remember to put all those include files in the package
        description (see the <literal>includes</literal> field in the Cabal
399 400 401 402 403 404

It is also possible, according the FFI specification, to put the 
<option>-&num;include</option> option in the foreign import 
declaration itself:
  foreign import "foo.h f" f :: Int -> IO Int
406 407 408
When compiling this module, GHC will generate a C file that includes
the specified <option>-&num;include</option>.  However, GHC
ross's avatar
ross committed
<emphasis>disables</emphasis> cross-module inlining for such foreign
410 411 412 413 414 415 416 417
calls, because it doesn't transport the <option>-&num;include</option>
information across module boundaries.  (There is no fundamental reason for this;
it was just tiresome to implement.  The wrapper, which unboxes the arguments
etc, is still inlined across modules.)  So if you want the foreign call itself
to be inlined across modules, use the command-line and package-configuration
<option>-&num;include</option> mechanism.

418 419 420 421 422 423 424 425
      <sect3 id="finding-header-files">
	<title>Finding Header files</title>

	<para>Header files named by the <option>-&num;include</option>
	option or in a <literal>foreign import</literal> declaration
	are searched for using the C compiler's usual search path.
	You can add directories to this search path using the
	<option>-I</option> option (see <xref
427 428 429 430 431 432 433 434 435 436 437

	<para>Note: header files are ignored unless compiling via C.
	If you had been compiling your code using the native code
	generator (the default) and suddenly switch to compiling via
	C, then you can get unexpected errors about missing include
	files.  Compiling via C is enabled automatically when certain
	options are given (eg. <option>-O</option> and
	<option>-prof</option> both enable

439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455

      <title>Memory Allocation</title>

      <para>The FFI libraries provide several ways to allocate memory
      for use with the FFI, and it isn't always clear which way is the
      best.  This decision may be affected by how efficient a
      particular kind of allocation is on a given compiler/platform,
      so this section aims to shed some light on how the different
      kinds of allocation perform with GHC.</para>

	  <term><literal>alloca</literal> and friends</term>
	    <para>Useful for short-term allocation when the allocation
	    is intended to scope over a given <literal>IO</literal>
ross's avatar
ross committed
	    computation.  This kind of allocation is commonly used
457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494
	    when marshalling data to and from FFI functions.</para>

	    <para>In GHC, <literal>alloca</literal> is implemented
	    using <literal>MutableByteArray#</literal>, so allocation
	    and deallocation are fast: much faster than C's
	    <literal>malloc/free</literal>, but not quite as fast as
	    stack allocation in C.  Use <literal>alloca</literal>
	    whenever you can.</para>

	    <para>Useful for longer-term allocation which requires
	    garbage collection.  If you intend to store the pointer to
	    the memory in a foreign data structure, then
	    <literal>mallocForeignPtr</literal> is
	    <emphasis>not</emphasis> a good choice, however.</para>

	    <para>In GHC, <literal>mallocForeignPtr</literal> is also
	    implemented using <literal>MutableByteArray#</literal>.
	    Although the memory is pointed to by a
	    <literal>ForeignPtr</literal>, there are no actual
	    finalizers involved (unless you add one with
	    <literal>addForeignPtrFinalizer</literal>), and the
	    deallocation is done using GC, so
	    <literal>mallocForeignPtr</literal> is normally very

	    <para>If all else fails, then you need to resort to
	    <literal>Foreign.malloc</literal> and
	    <literal>Foreign.free</literal>.  These are just wrappers
ross's avatar
ross committed
	    around the C functions of the same name, and their
496 497 498 499 500 501 502 503 504
	    efficiency will depend ultimately on the implementations
	    of these functions in your platform's C library.  We
	    usually find <literal>malloc</literal> and
	    <literal>free</literal> to be significantly slower than
	    the other forms of allocation above.</para>

ross's avatar
ross committed
506 507 508 509 510 511
	    <para>Pools are currently implemented using
	    <literal>malloc/free</literal>, so while they might be a
	    more convenient way to structure your memory allocation
	    than using one of the other forms of allocation, they
	    won't be any more efficient.  We do plan to provide an
ross's avatar
ross committed
	    improved-performance implementation of Pools in the
513 514 515 516 517
	    future, however.</para>
520 521 522

<!-- Emacs stuff:
     ;;; Local Variables: ***
523 524
     ;;; mode: xml ***
     ;;; sgml-parent-document: ("users_guide.xml" "book" "chapter") ***
525 526
     ;;; End: ***