1 %************************************************************************
3 \section[glasgow-exts]{Glasgow extensions to Haskell}
4 \index{Haskell, Glasgow extensions}
5 \index{extensions, Glasgow Haskell}
7 %************************************************************************
9 As with all known Haskell systems, GHC implements some extensions to
11 To use them, you'll need to give
12 a \tr{-fglasgow-exts}%
13 \index{-fglasgow-exts option} option.
15 Virtually all of the Glasgow extensions serve to give you access to the
16 underlying facilities with which we implement Haskell. Thus, you can
17 get at the Raw Iron, if you are willing to write some non-standard
18 code at a more primitive level. You need not be ``stuck'' on
19 performance because of the implementation costs of Haskell's
20 ``high-level'' features---you can always code ``under'' them. In an
21 extreme case, you can write all your time-critical code in C, and then
22 just glue it together with Haskell!
24 Executive summary of our extensions:
26 \item[Unboxed types and primitive operations:] You can get right down
27 to the raw machine types and operations; included in this are
28 ``primitive arrays'' (direct access to Big Wads of Bytes).
29 Please see \Sectionref{glasgow-unboxed} and following.
31 \item[Calling out to C:] Just what it sounds like. We provide {\em
32 lots} of rope that you can dangle around your neck.
33 Please see \Sectionref{glasgow-ccalls}.
35 \item[Low-level monadic I/O:] Monadic I/O is now standard with Haskell~1.3;
36 you can still get access to the system at a lower level (the ``PrimIO'' level).
38 \item[``HBC-ish'' extensions:] Extensions implemented because people said,
39 ``HBC does Y. Could you teach GHC to do the same?'' Please see
40 \Sectionref{glasgow-hbc-exts} for a quick list.
43 Before you get too carried away working at the lowest level (e.g.,
44 sloshing \tr{MutableByteArray#}s around your program), you may wish to
45 check if there are system libraries that provide a ``Haskellised
46 veneer'' over the features you want. See \Sectionref{syslibs}.
48 \Sectionref{ghc-prelude} is the definitive guide for many of the low-level facilities in GHC.
50 %Pieter Hartel led an interesting comparison-of-many-compilers (and
51 %many languages) in which GHC got to show off its extensions. We did
52 %very well! For the full details, check out
53 %\tr{pub/computer-systems/functional/packages/pseudoknot.tar.Z} on \tr{ftp.fwi.uva.nl}.
56 %************************************************************************
58 \subsection[glasgow-unboxed]{Unboxed types}
59 \index{Unboxed types (Glasgow extension)}
61 %************************************************************************
63 These types correspond to the ``raw machine'' types you would use in
64 C: \tr{Int#} (long int), \tr{Double#} (double),
65 \tr{Addr#} (void *), etc. The {\em primitive
66 operations} (PrimOps) on these types are what you might expect; e.g.,
67 \tr{(+#)} is addition on \tr{Int#}s, and is the machine-addition that
68 we all know and love---usually one instruction.
70 A numerically-intensive program using unboxed types can go a {\em lot}
71 faster than its ``standard'' counterpart---we saw a threefold speedup
74 Please see \Sectionref{ghc-libs-ghc} for the details of
75 unboxed types and the operations on them.
77 %************************************************************************
79 \subsection[glasgow-ST-monad]{Primitive state-transformer monad}
80 \index{state transformers (Glasgow extensions)}
82 %************************************************************************
84 This monad underlies our implementation of arrays, mutable and immutable,
85 and our implementation of I/O, including ``C calls''.
87 You probably won't use the monad directly, but you might use all those
90 More information on the state-related types can be found in the
91 elsewhere, \Sectionref{ghc-prelude}.
93 %************************************************************************
95 \subsection[glasgow-prim-arrays]{Primitive arrays, mutable and otherwise}
96 \index{primitive arrays (Glasgow extension)}
97 \index{arrays, primitive (Glasgow extension)}
99 %************************************************************************
101 GHC knows about quite a few flavours of Large Swathes of Bytes.
103 First, GHC distinguishes between primitive arrays of (boxed) Haskell
104 objects (type \tr{Array# obj}) and primitive arrays of bytes (type
107 Second, it distinguishes between...
110 Arrays that do not change (as with ``standard'' Haskell arrays); you
111 can only read from them. Obviously, they do not need the care and
112 attention of the state-transformer monad.
115 Arrays that may be changed or ``mutated.'' All the operations on them
116 live within the state-transformer monad and the updates happen {\em
119 \item[``Static'' (in C land):]
120 A C~routine may pass an \tr{Addr#} pointer back into Haskell land.
121 There are then primitive operations with which you may merrily grab
122 values over in C land, by indexing off the ``static'' pointer.
124 \item[``Stable'' pointers:]
125 If, for some reason, you wish to hand a Haskell pointer (i.e., {\em
126 not} an unboxed value) to a C~routine, you first make the pointer
127 ``stable,'' so that the garbage collector won't forget that it exists.
128 That is, GHC provides a safe way to pass Haskell pointers to C.
130 Please see \Sectionref{glasgow-stablePtrs} for more details.
132 \item[``Foreign objects'':]
133 A ``foreign object'' is a safe way to pass an external object (a
134 C~allocated pointer, say) to Haskell and have Haskell do the Right
135 Thing when it no longer references the object. So, for example, C
136 could pass a large bitmap over to Haskell and say ``please free this
137 memory when you're done with it.''
139 Please see \Sectionref{glasgow-foreignObjs} for more details.
142 The libraries section give more details on all these
143 ``primitive array'' types and the operations on them,
144 \Sectionref{ghc-prelude}.
147 %************************************************************************
149 \subsection[own-mainPrimIO]{Using your own @mainPrimIO@}
150 \index{mainPrimIO, rolling your own}
152 %************************************************************************
154 Normally, the GHC runtime system begins things by called an internal
155 function @mainPrimIO :: PrimIO ()@ which, in turn, fires up
158 To subvert the above process, you need only provide a
159 @mainPrimIO :: PrimIO ()@ of your own (in a module named \tr{GHCmain}).
161 Here's a little example, stolen from Alastair Reid:
163 module GHCmain ( mainPrimIO ) where
167 mainPrimIO :: PrimIO ()
170 _ccall_ printf "%d\n" (14::Int)
172 sleep :: Int -> PrimIO ()
173 sleep t = _ccall_ sleep t
176 %************************************************************************
178 \subsection[glasgow-ccalls]{Calling~C directly from Haskell}
179 \index{C calls (Glasgow extension)}
180 \index{_ccall_ (Glasgow extension)}
181 \index{_casm_ (Glasgow extension)}
183 %************************************************************************
185 %Besides using a \tr{-fglasgow-exts} flag, your modules need to include...
187 %import PreludePrimIO
190 GOOD ADVICE: Because this stuff is not Entirely Stable as far as names
191 and things go, you would be well-advised to keep your C-callery
192 corraled in a few modules, rather than sprinkled all over your code.
193 It will then be quite easy to update later on.
195 WARNING AS OF 2.03: Yes, the \tr{_ccall_} stuff probably {\em will
196 change}, to something better, of course! One step in that direction
197 is Green Card, a foreign function interface pre-processor for Haskell
198 (``Glasgow'' Haskell in particular) --- check out
200 ftp://ftp.dcs.gla.ac.uk/pub/haskell/glasgow/green-card.ANNOUNCE
201 ftp://ftp.dcs.gla.ac.uk/pub/haskell/glasgow/green-card-src.tar.gz
204 %************************************************************************
206 \subsubsection[ccall-intro]{\tr{_ccall_} and \tr{_casm_}: an introduction}
208 %************************************************************************
210 The simplest way to use a simple C function
212 double fooC( FILE *in, char c, int i, double d, unsigned int u )
214 is to provide a Haskell wrapper:
216 fooH :: Char -> Int -> Double -> Word -> PrimIO Double
217 fooH c i d w = _ccall_ fooC (``stdin''::Addr) c i d w
219 The function @fooH@ will unbox all of its arguments, call the C
220 function \tr{fooC} and box the corresponding arguments.
222 So, if you want to do C-calling, you have to confront the underlying
223 I/O system (at the ``PrimIO'' level).
225 %The code in \tr{ghc/lib/glaExts/*.lhs} is not too obtuse.
226 %That code, plus \tr{lib/prelude/Builtin.hs}, give examples
227 %of its use. The latter includes the implementations of \tr{error} and
230 One of the annoyances about \tr{_ccall_}s is when the C types don't quite
231 match the Haskell compiler's ideas. For this, the \tr{_casm_} variant
232 may be just the ticket (NB: {\em no chance} of such code going through
233 a native-code generator):
236 = _casm_ ``%r = getenv((char *) %0);'' name >>= \ litstring@(A# str#) ->
238 if (litstring == ``NULL'') then
239 Left ("Fail:oldGetEnv:"++name)
241 Right (unpackCString# str#)
245 The first literal-literal argument to a \tr{_casm_} is like a
246 \tr{printf} format: \tr{%r} is replaced with the ``result,''
247 \tr{%0}--\tr{%n-1} are replaced with the 1st--nth arguments. As you
248 can see above, it is an easy way to do simple C~casting. Everything
249 said about \tr{_ccall_} goes for \tr{_casm_} as well.
251 %************************************************************************
253 \subsubsection[glasgow-foreign-headers]{Using function headers}
254 \index{C calls---function headers}
256 %************************************************************************
258 When generating C (using the \tr{-fvia-C} directive), one can assist
259 the C compiler in detecting type errors by using the \tr{-#include}
260 directive to provide \tr{.h} files containing function headers.
264 typedef unsigned long *StgForeignObj;
267 void initialiseEFS (StgInt size);
268 StgInt terminateEFS (void);
269 StgForeignObj emptyEFS(void);
270 StgForeignObj updateEFS (StgForeignObj a, StgInt i, StgInt x);
271 StgInt lookupEFS (StgForeignObj a, StgInt i);
274 You can find appropriate definitions for \tr{StgInt},
275 \tr{StgForeignObj}, etc using \tr{gcc} on your architecture by
276 consulting \tr{ghc/includes/StgTypes.lh}. The following table
277 summarises the relationship between Haskell types and C types.
280 C type name & Haskell Type \\ \hline
281 %----- & ---------------
282 \tr{StgChar} & \tr{Char#}\\
283 \tr{StgInt} & \tr{Int#}\\
284 \tr{StgWord} & \tr{Word#}\\
285 \tr{StgAddr} & \tr{Addr#}\\
286 \tr{StgFloat} & \tr{Float#}\\
287 \tr{StgDouble} & \tr{Double#}\\
289 \tr{StgArray} & \tr{Array#}\\
290 \tr{StgByteArray} & \tr{ByteArray#}\\
291 \tr{StgArray} & \tr{MutableArray#}\\
292 \tr{StgByteArray} & \tr{MutableByteArray#}\\
294 \tr{StgStablePtr} & \tr{StablePtr#}\\
295 \tr{StgForeignObj} & \tr{ForeignObj#}
298 Note that this approach is only {\em essential\/} for returning
299 \tr{float}s (or if \tr{sizeof(int) != sizeof(int *)} on your
300 architecture) but is a Good Thing for anyone who cares about writing
301 solid code. You're crazy not to do it.
303 %************************************************************************
305 \subsubsection[glasgow-stablePtrs]{Subverting automatic unboxing with ``stable pointers''}
306 \index{stable pointers (Glasgow extension)}
308 %************************************************************************
310 The arguments of a \tr{_ccall_} are automatically unboxed before the
311 call. There are two reasons why this is usually the Right Thing to do:
314 C is a strict language: it would be excessively tedious to pass
315 unevaluated arguments and require the C programmer to force their
316 evaluation before using them.
318 \item Boxed values are stored on the Haskell heap and may be moved
319 within the heap if a garbage collection occurs---that is, pointers
320 to boxed objects are not {\em stable\/}.
323 It is possible to subvert the unboxing process by creating a ``stable
324 pointer'' to a value and passing the stable pointer instead. For example, to
325 pass/return an integer lazily to C functions \tr{storeC} and
326 \tr{fetchC}, one might write:
328 storeH :: Int -> PrimIO ()
329 storeH x = makeStablePtr x >>= \ stable_x ->
330 _ccall_ storeC stable_x
333 fetchH x = _ccall_ fetchC >>= \ stable_x ->
334 deRefStablePtr stable_x >>= \ x ->
335 freeStablePtr stable_x >>
339 The garbage collector will refrain from throwing a stable pointer away
340 until you explicitly call one of the following from C or Haskell.
342 void freeStablePointer( StgStablePtr stablePtrToToss )
343 freeStablePtr :: StablePtr a -> PrimIO ()
346 As with the use of \tr{free} in C programs, GREAT CARE SHOULD BE
347 EXERCISED to ensure these functions are called at the right time: too
348 early and you get dangling references (and, if you're lucky, an error
349 message from the runtime system); too late and you get space leaks.
351 %Doesn't work in ghc-0.23 - best to just keep quiet about them.
353 %And to force evaluation of the argument within \tr{fooC}, one would
354 %call one of the following C functions (according to type of argument).
357 %void performIO ( StgStablePtr stableIndex /* StablePtr s (PrimIO ()) */ );
358 %StgInt enterInt ( StgStablePtr stableIndex /* StablePtr s Int */ );
359 %StgFloat enterFloat ( StgStablePtr stableIndex /* StablePtr s Float */ );
362 %ToDo ADR: test these functions!
364 %Note Bene: \tr{_ccall_GC_} must be used if any of these functions are used.
367 %************************************************************************
369 \subsubsection[glasgow-foreignObjs]{Pointing outside the Haskell heap}
370 \index{foreign objects (Glasgow extension)}
372 %************************************************************************
374 There are two types that \tr{ghc} programs can use to reference
375 (heap-allocated) objects outside the Haskell world: \tr{Addr} and
378 If you use \tr{Addr}, it is up to you to the programmer to arrange
379 allocation and deallocation of the objects.
381 If you use \tr{ForeignObj}, \tr{ghc}'s garbage collector will
382 call upon the user-supplied {\em finaliser} function to free
383 the object when the Haskell world no longer can access the object.
384 (An object is associated with a finaliser function when the abstract
385 Haskell type @ForeignObj@ is created). The finaliser function is
386 expressed in C, and is passed as argument the object:
389 void foreignFinaliser ( StgForeignObj fo )
391 when the Haskell world can no longer access the object. Since
392 \tr{ForeignObj}s only get released when a garbage collection occurs,
393 we provide ways of triggering a garbage collection from within C and
396 void StgPerformGarbageCollection()
397 performGC :: PrimIO ()
400 %************************************************************************
402 \subsubsection[glasgow-avoiding-monads]{Avoiding monads}
403 \index{C calls to `pure C'}
404 \index{unsafePerformPrimIO (GlaExts)}
406 %************************************************************************
408 The \tr{_ccall_} construct is part of the \tr{PrimIO} monad because 9
409 out of 10 uses will be to call imperative functions with side effects
410 such as \tr{printf}. Use of the monad ensures that these operations
411 happen in a predictable order in spite of laziness and compiler
414 There are three situations where one might like to use
415 @unsafePerformPrimIO@ to avoid the monad:
418 Calling a function with no side-effects:
420 atan2d :: Double -> Double -> Double
421 atan2d y x = unsafePerformPrimIO (_ccall_ atan2d y x)
423 sincosd :: Double -> (Double, Double)
424 sincosd x = unsafePerformPrimIO $
425 newDoubleArray (0, 1) >>= \ da ->
426 _casm_ ``sincosd( %0, &((double *)%1[0]), &((double *)%1[1]) );'' x da
428 readDoubleArray da 0 >>= \ s ->
429 readDoubleArray da 1 >>= \ c ->
433 \item Calling a set of functions which have side-effects but which can
434 be used in a purely functional manner.
436 For example, an imperative implementation of a purely functional
437 lookup-table might be accessed using the following functions.
441 update :: EFS x -> Int -> x -> EFS x
442 lookup :: EFS a -> Int -> a
444 empty = unsafePerformPrimIO (_ccall_ emptyEFS)
446 update a i x = unsafePerformPrimIO $
447 makeStablePtr x >>= \ stable_x ->
448 _ccall_ updateEFS a i stable_x
450 lookup a i = unsafePerformPrimIO $
451 _ccall_ lookupEFS a i >>= \ stable_x ->
452 deRefStablePtr stable_x
455 You will almost always want to use \tr{ForeignObj}s with this.
457 \item Calling a side-effecting function even though the results will
458 be unpredictable. For example the \tr{trace} function is defined by:
461 trace :: String -> a -> a
463 = unsafePerformPrimIO (
464 ((_ccall_ PreTraceHook sTDERR{-msg-}):: PrimIO ()) >>
465 fputs sTDERR string >>
466 ((_ccall_ PostTraceHook sTDERR{-msg-}):: PrimIO ()) >>
469 sTDERR = (``stderr'' :: Addr)
472 (This kind of use is not highly recommended --- it is only really
473 useful in debugging code.)
476 %************************************************************************
478 \subsubsection[ccall-gotchas]{C-calling ``gotchas'' checklist}
479 \index{C call dangers}
481 %************************************************************************
483 And some advice, too.
487 \tr{_ccall_} is part of the \tr{PrimIO} monad --- not the 1.3 \tr{IO} Monad.
490 primIOToIO :: PrimIO a -> IO a
492 to promote a \tr{_ccall_} to the \tr{IO} monad.
495 For modules that use \tr{_ccall_}s, etc., compile with \tr{-fvia-C}.\index{-fvia-C option}
496 You don't have to, but you should.
498 Also, use the \tr{-#include "prototypes.h"} flag (hack) to inform the
499 C compiler of the fully-prototyped types of all the C functions you
500 call. (\Sectionref{glasgow-foreign-headers} says more about this...)
502 This scheme is the {\em only} way that you will get {\em any}
503 typechecking of your \tr{_ccall_}s. (It shouldn't be that way,
507 Try to avoid \tr{_ccall_}s to C~functions that take \tr{float}
508 arguments or return \tr{float} results. Reason: if you do, you will
509 become entangled in (ANSI?) C's rules for when arguments/results are
510 promoted to \tr{doubles}. It's a nightmare and just not worth it.
511 Use \tr{doubles} if possible.
513 If you do use \tr{floats}, check and re-check that the right thing is
514 happening. Perhaps compile with \tr{-keep-hc-file-too} and look at
515 the intermediate C (\tr{.hc} file).
518 The compiler uses two non-standard type-classes when
519 type-checking the arguments and results of \tr{_ccall_}: the arguments
520 (respectively result) of \tr{_ccall_} must be instances of the class
521 \tr{CCallable} (respectively \tr{CReturnable}). (Neither class
522 defines any methods --- their only function is to keep the
525 The type checker must be able to figure out just which of the
526 C-callable/returnable types is being used. If it can't, you have to
527 add type signatures. For example,
531 is not good enough, because the compiler can't work out what type @x@ is, nor
532 what type the @_ccall_@ returns. You have to write, say:
534 f :: Int -> PrimIO Double
538 This table summarises the standard instances of these classes.
540 % ToDo: check this table against implementation!
542 \begin{tabular}{llll}
543 Type &CCallable&CReturnable & Which is probably... \\ \hline
544 %------ ---------- ------------ -------------
545 \tr{Char} & Yes & Yes & \tr{unsigned char} \\
546 \tr{Int} & Yes & Yes & \tr{long int} \\
547 \tr{Word} & Yes & Yes & \tr{unsigned long int} \\
548 \tr{Addr} & Yes & Yes & \tr{char *} \\
549 \tr{Float} & Yes & Yes & \tr{float} \\
550 \tr{Double} & Yes & Yes & \tr{double} \\
551 \tr{()} & No & Yes & \tr{void} \\
552 \tr{[Char]} & Yes & No & \tr{char *} (null-terminated) \\
554 \tr{Array} & Yes & No & \tr{unsigned long *}\\
555 \tr{ByteArray} & Yes & No & \tr{unsigned long *}\\
556 \tr{MutableArray} & Yes & No & \tr{unsigned long *}\\
557 \tr{MutableByteArray} & Yes & No & \tr{unsigned long *}\\
559 \tr{State} & Yes & Yes & nothing!\\
561 \tr{StablePtr} & Yes & Yes & \tr{unsigned long *}\\
562 \tr{ForeignObjs} & Yes & Yes & see later\\
565 The brave and careful programmer can add their own instances of these
566 classes for the following types:
569 A {\em boxed-primitive} type may be made an instance of both
570 \tr{CCallable} and \tr{CReturnable}.
572 A boxed primitive type is any data type with a
573 single unary constructor with a single primitive argument. For
574 example, the following are all boxed primitive types:
579 data XDisplay = XDisplay Addr#
580 data EFS a = EFS# ForeignObj#
584 instance CCallable (EFS a)
585 instance CReturnable (EFS a)
588 \item Any datatype with a single nullary constructor may be made an
589 instance of \tr{CReturnable}. For example:
593 instance CReturnable MyVoid
596 \item As at version 2.02, \tr{String} (i.e., \tr{[Char]}) is still
597 not a \tr{CReturnable} type.
599 Also, the now-builtin type \tr{PackedString} is neither
600 \tr{CCallable} nor \tr{CReturnable}. (But there are functions in
601 the PackedString interface to let you get at the necessary bits...)
605 The code-generator will complain if you attempt to use \tr{%r}
606 in a \tr{_casm_} whose result type is \tr{PrimIO ()}; or if you don't
607 use \tr{%r} {\em precisely\/} once for any other result type. These
608 messages are supposed to be helpful and catch bugs---please tell us
609 if they wreck your life.
612 If you call out to C code which may trigger the Haskell garbage
613 collector (examples of this later...), then you must use the
614 \tr{_ccall_GC_} or \tr{_casm_GC_} variant of C-calls. (This does not
615 work with the native code generator - use \tr{\fvia-C}.) This stuff is
616 hairy with a capital H!
619 %************************************************************************
621 %\subsubsection[ccall-good-practice]{C-calling ``good practice'' checklist}
623 %************************************************************************
625 %************************************************************************
627 \subsubsection[glasgow-prim-interface]{Access to the \tr{PrimIO} monad}
628 \index{PrimIO monad (Glasgow extension)}
629 \index{I/O, primitive (Glasgow extension)}
631 %************************************************************************
633 The \tr{IO} monad (new in Haskell~1.3) catches errors and passes them
634 along. It is built on top of the \tr{ST} state-transformer monad.
636 A related (and inter-operable-with) monad is the \tr{PrimIO} monad
637 (NB: the level at which @_ccall_@s work...), where you handle errors
640 Should you wish to use the \tr{PrimIO} monad directly, you can import
641 \tr{GlaExts}. It makes available the usual monadic stuff (@>>=@,
642 @>>@, @return@, etc.), as well as these functions:
644 -- for backward compatibility:
645 returnPrimIO :: a -> PrimIO a
646 thenPrimIO :: PrimIO a -> (a -> PrimIO b) -> PrimIO b
647 seqPrimIO :: PrimIO a -> PrimIO b -> PrimIO b
650 fixPrimIO :: (a -> PrimIO a) -> PrimIO a
651 forkPrimIO :: PrimIO a -> PrimIO a
652 listPrimIO :: [PrimIO a] -> PrimIO [a]
653 mapAndUnzipPrimIO :: (a -> PrimIO (b,c)) -> [a] -> PrimIO ([b],[c])
654 mapPrimIO :: (a -> PrimIO b) -> [a] -> PrimIO [b]
656 unsafePerformPrimIO :: PrimIO a -> a
657 unsafeInterleavePrimIO :: PrimIO a -> PrimIO a
658 -- and they are not called "unsafe" for nothing!
660 -- to convert back and forth between IO and PrimIO
661 ioToPrimIO :: IO a -> PrimIO a
662 primIOToIO :: PrimIO a -> IO a
666 %************************************************************************
668 \subsection[glasgow-hbc-exts]{``HBC-ish'' extensions implemented by GHC}
669 \index{HBC-like Glasgow extensions}
670 \index{extensions, HBC-like}
672 %************************************************************************
675 %-------------------------------------------------------------------
676 \item[@fromInt@ method in class @Num@:]
677 It's there. Converts from an \tr{Int} to the type.
679 %-------------------------------------------------------------------
680 \item[@toInt@ method in class @Integral@:]
681 Converts from type type to an \tr{Int}.
683 %-------------------------------------------------------------------
684 \item[Overlapping instance declarations:]
685 \index{overlapping instances}
686 \index{instances, overlapping}
688 In \tr{instance <context> => Class (T x1 ... xn)}, the \tr{xi}s can be
689 {\em types}, rather than just {\em type variables}.
691 Thus, you can have an instance \tr{instance Foo [Char]}, as well as
692 the more general \tr{instance Foo [a]}; the former will be used in
693 preference to the latter, where applicable.
695 As Lennart says, ``This is a dubious feature and should not be used
698 See also: \tr{SPECIALIZE instance} pragmas, in \Sectionref{faster}.
700 %-------------------------------------------------------------------
701 % \item[Signal-handling I/O request:]
702 % \index{signal handling (extension)}
703 % \index{SigAction I/O request}
704 % The Haskell-1.2 I/O request \tr{SigAction n act} installs a signal handler for signal
705 % \tr{n :: Int}. The number is the usual UNIX signal number. The action
714 % The corresponding continuation-style I/O function is the unsurprising:
716 % sigAction :: Int -> SigAct -> FailCont -> SuccCont -> Dialogue
719 % When a signal handler is installed with \tr{SACatch}, receipt of the
720 % signal causes the current top-level computation to be abandoned, and
721 % the specified dialogue to be executed instead. The abandoned
722 % computation may leave some partially evaluated expressions in a
723 % non-resumable state. If you believe that your top-level computation
724 % and your signal handling dialogue may share subexpressions, you should
725 % execute your program with the \tr{-N} RTS option, to prevent
728 % The \tr{-N} option is not available with concurrent/parallel programs,
729 % so great care should be taken to avoid shared subexpressions between
730 % the top-level computation and any signal handlers when using threads.
732 %-------------------------------------------------------------------
733 %\item[Simple time-out mechanism, in ``monadic I/O'':]
734 %\index{time-outs (extension)}
736 %This function is available:
738 %timeoutIO :: Int -> IO Void -> IO (IO Void)
741 %Wait that many seconds, then abandon the current computation and
742 %perform the given I/O operation (second argument). Uses the
743 %signal-handling, so it returns the previous signal-handler (in case
744 %you want to re-install it). As above, you may need to execute your
745 %program with the RTS flag \tr{-N}, to prevent black-holing.
749 %************************************************************************
751 %\subsection[glasgow-compiler-namespace]{Fiddlings the compiler's built-in namespaces}
753 %************************************************************************
755 %This is really only used for compiling the prelude. It's turgid and
756 %will probably change.
758 % \begin{description}
759 % \item[\tr{-no-implicit-prelude}:]
760 % \index{-no-implicit-prelude option}
762 % ???? (Tells the parser not to read \tr{Prelude.hi}).
764 % \item[\tr{-fhide-builtin-names}:]
765 % \index{-fhide-builtin-names option}
766 % This hides {\em all} Prelude names built-in to the compiler.
768 % \item[\tr{-fmin-builtin-names}:]
769 % \index{-fmin-builtin-names option}
770 % This hides all but a few of the Prelude names that are built-in to the
771 % compiler. @:@ (cons) is an example of one that would remain visible.
773 % \item[\tr{-fhide-builtin-instances}:]
774 % \index{-fhide-builtin-instances option}
775 % This suppresses the compiler's own ideas about what instances already
776 % exist (e.g., \tr{instance Eq Int}).
778 % This flag is used when actually compiling the various instance
779 % declarations in the Prelude.