2 % (c) The GRASP/AQUA Project, Glasgow University, 1993-1998
4 \section[Specialise]{Stamping out overloading, and (optionally) polymorphism}
7 module Specialise ( specProgram ) where
9 #include "HsVersions.h"
11 import CmdLineOpts ( opt_D_verbose_core2core, opt_D_dump_spec, opt_D_dump_rules )
12 import Id ( Id, idName, idType, mkTemplateLocals, mkUserLocal,
13 idSpecialisation, setIdNoDiscard, isExportedId,
14 modifyIdInfo, idUnfolding
16 import IdInfo ( zapSpecPragInfo )
20 import Type ( Type, mkTyVarTy, splitSigmaTy, splitFunTysN,
21 tyVarsOfType, tyVarsOfTypes, tyVarsOfTheta, applyTys,
22 mkForAllTys, boxedTypeKind
24 import PprType ( {- instance Outputable Type -} )
25 import Subst ( Subst, mkSubst, substTy, emptySubst, substBndrs, extendSubstList,
26 substId, substAndCloneId, substAndCloneIds, lookupIdSubst
28 import Var ( TyVar, mkSysTyVar, setVarUnique )
32 import CoreUtils ( applyTypeToArgs )
33 import CoreUnfold ( certainlyWillInline )
34 import CoreFVs ( exprFreeVars, exprsFreeVars )
35 import CoreLint ( beginPass, endPass )
36 import PprCore ( pprCoreRules )
37 import Rules ( addIdSpecialisations )
39 import UniqSupply ( UniqSupply,
40 UniqSM, initUs_, thenUs, thenUs_, returnUs, getUniqueUs,
41 getUs, setUs, uniqFromSupply, splitUniqSupply, mapUs
43 import Name ( nameOccName, mkSpecOcc, getSrcLoc )
45 import Maybes ( MaybeErr(..), catMaybes )
46 import ErrUtils ( dumpIfSet )
48 import List ( partition )
49 import Util ( zipEqual, zipWithEqual, mapAccumL )
56 %************************************************************************
58 \subsection[notes-Specialise]{Implementation notes [SLPJ, Aug 18 1993]}
60 %************************************************************************
62 These notes describe how we implement specialisation to eliminate
65 The specialisation pass works on Core
66 syntax, complete with all the explicit dictionary application,
67 abstraction and construction as added by the type checker. The
68 existing type checker remains largely as it is.
70 One important thought: the {\em types} passed to an overloaded
71 function, and the {\em dictionaries} passed are mutually redundant.
72 If the same function is applied to the same type(s) then it is sure to
73 be applied to the same dictionary(s)---or rather to the same {\em
74 values}. (The arguments might look different but they will evaluate
77 Second important thought: we know that we can make progress by
78 treating dictionary arguments as static and worth specialising on. So
79 we can do without binding-time analysis, and instead specialise on
80 dictionary arguments and no others.
89 and suppose f is overloaded.
91 STEP 1: CALL-INSTANCE COLLECTION
93 We traverse <body>, accumulating all applications of f to types and
96 (Might there be partial applications, to just some of its types and
97 dictionaries? In principle yes, but in practice the type checker only
98 builds applications of f to all its types and dictionaries, so partial
99 applications could only arise as a result of transformation, and even
100 then I think it's unlikely. In any case, we simply don't accumulate such
101 partial applications.)
106 So now we have a collection of calls to f:
110 Notice that f may take several type arguments. To avoid ambiguity, we
111 say that f is called at type t1/t2 and t3/t4.
113 We take equivalence classes using equality of the *types* (ignoring
114 the dictionary args, which as mentioned previously are redundant).
116 STEP 3: SPECIALISATION
118 For each equivalence class, choose a representative (f t1 t2 d1 d2),
119 and create a local instance of f, defined thus:
121 f@t1/t2 = <f_rhs> t1 t2 d1 d2
123 f_rhs presumably has some big lambdas and dictionary lambdas, so lots
124 of simplification will now result. However we don't actually *do* that
125 simplification. Rather, we leave it for the simplifier to do. If we
126 *did* do it, though, we'd get more call instances from the specialised
127 RHS. We can work out what they are by instantiating the call-instance
128 set from f's RHS with the types t1, t2.
130 Add this new id to f's IdInfo, to record that f has a specialised version.
132 Before doing any of this, check that f's IdInfo doesn't already
133 tell us about an existing instance of f at the required type/s.
134 (This might happen if specialisation was applied more than once, or
135 it might arise from user SPECIALIZE pragmas.)
139 Wait a minute! What if f is recursive? Then we can't just plug in
140 its right-hand side, can we?
142 But it's ok. The type checker *always* creates non-recursive definitions
143 for overloaded recursive functions. For example:
145 f x = f (x+x) -- Yes I know its silly
149 f a (d::Num a) = let p = +.sel a d
151 letrec fl (y::a) = fl (p y y)
155 We still have recusion for non-overloaded functions which we
156 speciailise, but the recursive call should get specialised to the
157 same recursive version.
163 All this is crystal clear when the function is applied to *constant
164 types*; that is, types which have no type variables inside. But what if
165 it is applied to non-constant types? Suppose we find a call of f at type
166 t1/t2. There are two possibilities:
168 (a) The free type variables of t1, t2 are in scope at the definition point
169 of f. In this case there's no problem, we proceed just as before. A common
170 example is as follows. Here's the Haskell:
175 After typechecking we have
177 g a (d::Num a) (y::a) = let f b (d'::Num b) (x::b) = +.sel b d' x x
178 in +.sel a d (f a d y) (f a d y)
180 Notice that the call to f is at type type "a"; a non-constant type.
181 Both calls to f are at the same type, so we can specialise to give:
183 g a (d::Num a) (y::a) = let f@a (x::a) = +.sel a d x x
184 in +.sel a d (f@a y) (f@a y)
187 (b) The other case is when the type variables in the instance types
188 are *not* in scope at the definition point of f. The example we are
189 working with above is a good case. There are two instances of (+.sel a d),
190 but "a" is not in scope at the definition of +.sel. Can we do anything?
191 Yes, we can "common them up", a sort of limited common sub-expression deal.
194 g a (d::Num a) (y::a) = let +.sel@a = +.sel a d
195 f@a (x::a) = +.sel@a x x
196 in +.sel@a (f@a y) (f@a y)
198 This can save work, and can't be spotted by the type checker, because
199 the two instances of +.sel weren't originally at the same type.
203 * There are quite a few variations here. For example, the defn of
204 +.sel could be floated ouside the \y, to attempt to gain laziness.
205 It certainly mustn't be floated outside the \d because the d has to
208 * We don't want to inline f_rhs in this case, because
209 that will duplicate code. Just commoning up the call is the point.
211 * Nothing gets added to +.sel's IdInfo.
213 * Don't bother unless the equivalence class has more than one item!
215 Not clear whether this is all worth it. It is of course OK to
216 simply discard call-instances when passing a big lambda.
218 Polymorphism 2 -- Overloading
220 Consider a function whose most general type is
222 f :: forall a b. Ord a => [a] -> b -> b
224 There is really no point in making a version of g at Int/Int and another
225 at Int/Bool, because it's only instancing the type variable "a" which
226 buys us any efficiency. Since g is completely polymorphic in b there
227 ain't much point in making separate versions of g for the different
230 That suggests that we should identify which of g's type variables
231 are constrained (like "a") and which are unconstrained (like "b").
232 Then when taking equivalence classes in STEP 2, we ignore the type args
233 corresponding to unconstrained type variable. In STEP 3 we make
234 polymorphic versions. Thus:
236 f@t1/ = /\b -> <f_rhs> t1 b d1 d2
245 f a (d::Num a) = let g = ...
247 ...(let d1::Ord a = Num.Ord.sel a d in g a d1)...
249 Here, g is only called at one type, but the dictionary isn't in scope at the
250 definition point for g. Usually the type checker would build a
251 definition for d1 which enclosed g, but the transformation system
252 might have moved d1's defn inward. Solution: float dictionary bindings
253 outwards along with call instances.
257 f x = let g p q = p==q
263 Before specialisation, leaving out type abstractions we have
265 f df x = let g :: Eq a => a -> a -> Bool
267 h :: Num a => a -> a -> (a, Bool)
268 h dh r s = let deq = eqFromNum dh
269 in (+ dh r s, g deq r s)
273 After specialising h we get a specialised version of h, like this:
275 h' r s = let deq = eqFromNum df
276 in (+ df r s, g deq r s)
278 But we can't naively make an instance for g from this, because deq is not in scope
279 at the defn of g. Instead, we have to float out the (new) defn of deq
280 to widen its scope. Notice that this floating can't be done in advance -- it only
281 shows up when specialisation is done.
283 User SPECIALIZE pragmas
284 ~~~~~~~~~~~~~~~~~~~~~~~
285 Specialisation pragmas can be digested by the type checker, and implemented
286 by adding extra definitions along with that of f, in the same way as before
288 f@t1/t2 = <f_rhs> t1 t2 d1 d2
290 Indeed the pragmas *have* to be dealt with by the type checker, because
291 only it knows how to build the dictionaries d1 and d2! For example
293 g :: Ord a => [a] -> [a]
294 {-# SPECIALIZE f :: [Tree Int] -> [Tree Int] #-}
296 Here, the specialised version of g is an application of g's rhs to the
297 Ord dictionary for (Tree Int), which only the type checker can conjure
298 up. There might not even *be* one, if (Tree Int) is not an instance of
299 Ord! (All the other specialision has suitable dictionaries to hand
302 Problem. The type checker doesn't have to hand a convenient <f_rhs>, because
303 it is buried in a complex (as-yet-un-desugared) binding group.
306 f@t1/t2 = f* t1 t2 d1 d2
308 where f* is the Id f with an IdInfo which says "inline me regardless!".
309 Indeed all the specialisation could be done in this way.
310 That in turn means that the simplifier has to be prepared to inline absolutely
311 any in-scope let-bound thing.
314 Again, the pragma should permit polymorphism in unconstrained variables:
316 h :: Ord a => [a] -> b -> b
317 {-# SPECIALIZE h :: [Int] -> b -> b #-}
319 We *insist* that all overloaded type variables are specialised to ground types,
320 (and hence there can be no context inside a SPECIALIZE pragma).
321 We *permit* unconstrained type variables to be specialised to
323 - or left as a polymorphic type variable
324 but nothing in between. So
326 {-# SPECIALIZE h :: [Int] -> [c] -> [c] #-}
328 is *illegal*. (It can be handled, but it adds complication, and gains the
332 SPECIALISING INSTANCE DECLARATIONS
333 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
336 instance Foo a => Foo [a] where
338 {-# SPECIALIZE instance Foo [Int] #-}
340 The original instance decl creates a dictionary-function
343 dfun.Foo.List :: forall a. Foo a -> Foo [a]
345 The SPECIALIZE pragma just makes a specialised copy, just as for
346 ordinary function definitions:
348 dfun.Foo.List@Int :: Foo [Int]
349 dfun.Foo.List@Int = dfun.Foo.List Int dFooInt
351 The information about what instance of the dfun exist gets added to
352 the dfun's IdInfo in the same way as a user-defined function too.
355 Automatic instance decl specialisation?
356 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
357 Can instance decls be specialised automatically? It's tricky.
358 We could collect call-instance information for each dfun, but
359 then when we specialised their bodies we'd get new call-instances
360 for ordinary functions; and when we specialised their bodies, we might get
361 new call-instances of the dfuns, and so on. This all arises because of
362 the unrestricted mutual recursion between instance decls and value decls.
364 Still, there's no actual problem; it just means that we may not do all
365 the specialisation we could theoretically do.
367 Furthermore, instance decls are usually exported and used non-locally,
368 so we'll want to compile enough to get those specialisations done.
370 Lastly, there's no such thing as a local instance decl, so we can
371 survive solely by spitting out *usage* information, and then reading that
372 back in as a pragma when next compiling the file. So for now,
373 we only specialise instance decls in response to pragmas.
376 SPITTING OUT USAGE INFORMATION
377 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
379 To spit out usage information we need to traverse the code collecting
380 call-instance information for all imported (non-prelude?) functions
381 and data types. Then we equivalence-class it and spit it out.
383 This is done at the top-level when all the call instances which escape
384 must be for imported functions and data types.
386 *** Not currently done ***
389 Partial specialisation by pragmas
390 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
391 What about partial specialisation:
393 k :: (Ord a, Eq b) => [a] -> b -> b -> [a]
394 {-# SPECIALIZE k :: Eq b => [Int] -> b -> b -> [a] #-}
398 {-# SPECIALIZE k :: Eq b => [Int] -> [b] -> [b] -> [a] #-}
400 Seems quite reasonable. Similar things could be done with instance decls:
402 instance (Foo a, Foo b) => Foo (a,b) where
404 {-# SPECIALIZE instance Foo a => Foo (a,Int) #-}
405 {-# SPECIALIZE instance Foo b => Foo (Int,b) #-}
407 Ho hum. Things are complex enough without this. I pass.
410 Requirements for the simplifer
411 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
412 The simplifier has to be able to take advantage of the specialisation.
414 * When the simplifier finds an application of a polymorphic f, it looks in
415 f's IdInfo in case there is a suitable instance to call instead. This converts
417 f t1 t2 d1 d2 ===> f_t1_t2
419 Note that the dictionaries get eaten up too!
421 * Dictionary selection operations on constant dictionaries must be
424 +.sel Int d ===> +Int
426 The obvious way to do this is in the same way as other specialised
427 calls: +.sel has inside it some IdInfo which tells that if it's applied
428 to the type Int then it should eat a dictionary and transform to +Int.
430 In short, dictionary selectors need IdInfo inside them for constant
433 * Exactly the same applies if a superclass dictionary is being
436 Eq.sel Int d ===> dEqInt
438 * Something similar applies to dictionary construction too. Suppose
439 dfun.Eq.List is the function taking a dictionary for (Eq a) to
440 one for (Eq [a]). Then we want
442 dfun.Eq.List Int d ===> dEq.List_Int
444 Where does the Eq [Int] dictionary come from? It is built in
445 response to a SPECIALIZE pragma on the Eq [a] instance decl.
447 In short, dfun Ids need IdInfo with a specialisation for each
448 constant instance of their instance declaration.
450 All this uses a single mechanism: the SpecEnv inside an Id
453 What does the specialisation IdInfo look like?
454 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
456 The SpecEnv of an Id maps a list of types (the template) to an expression
460 For example, if f has this SpecInfo:
462 [Int, a] -> \d:Ord Int. f' a
464 it means that we can replace the call
466 f Int t ===> (\d. f' t)
468 This chucks one dictionary away and proceeds with the
469 specialised version of f, namely f'.
472 What can't be done this way?
473 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
474 There is no way, post-typechecker, to get a dictionary for (say)
475 Eq a from a dictionary for Eq [a]. So if we find
479 we can't transform to
484 eqList :: (a->a->Bool) -> [a] -> [a] -> Bool
486 Of course, we currently have no way to automatically derive
487 eqList, nor to connect it to the Eq [a] instance decl, but you
488 can imagine that it might somehow be possible. Taking advantage
489 of this is permanently ruled out.
491 Still, this is no great hardship, because we intend to eliminate
492 overloading altogether anyway!
496 A note about non-tyvar dictionaries
497 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
498 Some Ids have types like
500 forall a,b,c. Eq a -> Ord [a] -> tau
502 This seems curious at first, because we usually only have dictionary
503 args whose types are of the form (C a) where a is a type variable.
504 But this doesn't hold for the functions arising from instance decls,
505 which sometimes get arguements with types of form (C (T a)) for some
508 Should we specialise wrt this compound-type dictionary? We used to say
510 "This is a heuristic judgement, as indeed is the fact that we
511 specialise wrt only dictionaries. We choose *not* to specialise
512 wrt compound dictionaries because at the moment the only place
513 they show up is in instance decls, where they are simply plugged
514 into a returned dictionary. So nothing is gained by specialising
517 But it is simpler and more uniform to specialise wrt these dicts too;
518 and in future GHC is likely to support full fledged type signatures
520 f ;: Eq [(a,b)] => ...
523 %************************************************************************
525 \subsubsection{The new specialiser}
527 %************************************************************************
529 Our basic game plan is this. For let(rec) bound function
530 f :: (C a, D c) => (a,b,c,d) -> Bool
532 * Find any specialised calls of f, (f ts ds), where
533 ts are the type arguments t1 .. t4, and
534 ds are the dictionary arguments d1 .. d2.
536 * Add a new definition for f1 (say):
538 f1 = /\ b d -> (..body of f..) t1 b t3 d d1 d2
540 Note that we abstract over the unconstrained type arguments.
544 [t1,b,t3,d] |-> \d1 d2 -> f1 b d
546 to the specialisations of f. This will be used by the
547 simplifier to replace calls
548 (f t1 t2 t3 t4) da db
550 (\d1 d1 -> f1 t2 t4) da db
552 All the stuff about how many dictionaries to discard, and what types
553 to apply the specialised function to, are handled by the fact that the
554 SpecEnv contains a template for the result of the specialisation.
556 We don't build *partial* specialisations for f. For example:
558 f :: Eq a => a -> a -> Bool
559 {-# SPECIALISE f :: (Eq b, Eq c) => (b,c) -> (b,c) -> Bool #-}
561 Here, little is gained by making a specialised copy of f.
562 There's a distinct danger that the specialised version would
563 first build a dictionary for (Eq b, Eq c), and then select the (==)
564 method from it! Even if it didn't, not a great deal is saved.
566 We do, however, generate polymorphic, but not overloaded, specialisations:
568 f :: Eq a => [a] -> b -> b -> b
569 {#- SPECIALISE f :: [Int] -> b -> b -> b #-}
571 Hence, the invariant is this:
573 *** no specialised version is overloaded ***
576 %************************************************************************
578 \subsubsection{The exported function}
580 %************************************************************************
583 specProgram :: UniqSupply -> [CoreBind] -> IO [CoreBind]
586 beginPass "Specialise"
588 let binds' = initSM us (go binds `thenSM` \ (binds', uds') ->
589 returnSM (dumpAllDictBinds uds' binds'))
591 endPass "Specialise" (opt_D_dump_spec || opt_D_verbose_core2core) binds'
593 dumpIfSet opt_D_dump_rules "Top-level specialisations"
594 (vcat (map dump_specs (concat (map bindersOf binds'))))
598 go [] = returnSM ([], emptyUDs)
599 go (bind:binds) = go binds `thenSM` \ (binds', uds) ->
600 specBind emptySubst bind uds `thenSM` \ (bind', uds') ->
601 returnSM (bind' ++ binds', uds')
603 dump_specs var = pprCoreRules var (idSpecialisation var)
606 %************************************************************************
608 \subsubsection{@specExpr@: the main function}
610 %************************************************************************
613 specVar :: Subst -> Id -> CoreExpr
614 specVar subst v = case lookupIdSubst subst v of
618 specExpr :: Subst -> CoreExpr -> SpecM (CoreExpr, UsageDetails)
619 -- We carry a substitution down:
620 -- a) we must clone any binding that might flaot outwards,
621 -- to avoid name clashes
622 -- b) we carry a type substitution to use when analysing
623 -- the RHS of specialised bindings (no type-let!)
625 ---------------- First the easy cases --------------------
626 specExpr subst (Type ty) = returnSM (Type (substTy subst ty), emptyUDs)
627 specExpr subst (Var v) = returnSM (specVar subst v, emptyUDs)
628 specExpr subst (Lit lit) = returnSM (Lit lit, emptyUDs)
630 specExpr subst (Note note body)
631 = specExpr subst body `thenSM` \ (body', uds) ->
632 returnSM (Note (specNote subst note) body', uds)
635 ---------------- Applications might generate a call instance --------------------
636 specExpr subst expr@(App fun arg)
639 go (App fun arg) args = specExpr subst arg `thenSM` \ (arg', uds_arg) ->
640 go fun (arg':args) `thenSM` \ (fun', uds_app) ->
641 returnSM (App fun' arg', uds_arg `plusUDs` uds_app)
643 go (Var f) args = case specVar subst f of
644 Var f' -> returnSM (Var f', mkCallUDs f' args)
645 e' -> returnSM (e', emptyUDs) -- I don't expect this!
646 go other args = specExpr subst other
648 ---------------- Lambda/case require dumping of usage details --------------------
649 specExpr subst e@(Lam _ _)
650 = specExpr subst' body `thenSM` \ (body', uds) ->
652 (filtered_uds, body'') = dumpUDs bndrs' uds body'
654 returnSM (mkLams bndrs' body'', filtered_uds)
656 (bndrs, body) = collectBinders e
657 (subst', bndrs') = substBndrs subst bndrs
658 -- More efficient to collect a group of binders together all at once
659 -- and we don't want to split a lambda group with dumped bindings
661 specExpr subst (Case scrut case_bndr alts)
662 = specExpr subst scrut `thenSM` \ (scrut', uds_scrut) ->
663 mapAndCombineSM spec_alt alts `thenSM` \ (alts', uds_alts) ->
664 returnSM (Case scrut' case_bndr' alts', uds_scrut `plusUDs` uds_alts)
666 (subst_alt, case_bndr') = substId subst case_bndr
668 spec_alt (con, args, rhs)
669 = specExpr subst_rhs rhs `thenSM` \ (rhs', uds) ->
671 (uds', rhs'') = dumpUDs args uds rhs'
673 returnSM ((con, args', rhs''), uds')
675 (subst_rhs, args') = substBndrs subst_alt args
677 ---------------- Finally, let is the interesting case --------------------
678 specExpr subst (Let bind body)
680 cloneBindSM subst bind `thenSM` \ (rhs_subst, body_subst, bind') ->
682 -- Deal with the body
683 specExpr body_subst body `thenSM` \ (body', body_uds) ->
685 -- Deal with the bindings
686 specBind rhs_subst bind' body_uds `thenSM` \ (binds', uds) ->
689 returnSM (foldr Let body' binds', uds)
691 -- Must apply the type substitution to coerceions
692 specNote subst (Coerce t1 t2) = Coerce (substTy subst t1) (substTy subst t2)
693 specNote subst note = note
696 %************************************************************************
698 \subsubsection{Dealing with a binding}
700 %************************************************************************
703 specBind :: Subst -- Use this for RHSs
705 -> UsageDetails -- Info on how the scope of the binding
706 -> SpecM ([CoreBind], -- New bindings
707 UsageDetails) -- And info to pass upstream
709 specBind rhs_subst bind body_uds
710 = specBindItself rhs_subst bind (calls body_uds) `thenSM` \ (bind', bind_uds) ->
712 bndrs = bindersOf bind
713 all_uds = zapCalls bndrs (body_uds `plusUDs` bind_uds)
714 -- It's important that the `plusUDs` is this way round,
715 -- because body_uds may bind dictionaries that are
716 -- used in the calls passed to specDefn. So the
717 -- dictionary bindings in bind_uds may mention
718 -- dictionaries bound in body_uds.
720 case splitUDs bndrs all_uds of
722 (_, ([],[])) -- This binding doesn't bind anything needed
723 -- in the UDs, so put the binding here
724 -- This is the case for most non-dict bindings, except
725 -- for the few that are mentioned in a dict binding
726 -- that is floating upwards in body_uds
727 -> returnSM ([bind'], all_uds)
729 (float_uds, (dict_binds, calls)) -- This binding is needed in the UDs, so float it out
730 -> returnSM ([], float_uds `plusUDs` mkBigUD bind' dict_binds calls)
733 -- A truly gruesome function
734 mkBigUD bind@(NonRec _ _) dbs calls
735 = -- Common case: non-recursive and no specialisations
736 -- (if there were any specialistions it would have been made recursive)
737 MkUD { dict_binds = listToBag (mkDB bind : dbs),
738 calls = listToCallDetails calls }
740 mkBigUD bind dbs calls
742 MkUD { dict_binds = unitBag (mkDB (Rec (bind_prs bind ++ dbsToPairs dbs))),
744 calls = listToCallDetails calls }
746 bind_prs (NonRec b r) = [(b,r)]
747 bind_prs (Rec prs) = prs
750 dbsToPairs ((bind,_):dbs) = bind_prs bind ++ dbsToPairs dbs
752 -- specBindItself deals with the RHS, specialising it according
753 -- to the calls found in the body (if any)
754 specBindItself rhs_subst (NonRec bndr rhs) call_info
755 = specDefn rhs_subst call_info (bndr,rhs) `thenSM` \ ((bndr',rhs'), spec_defns, spec_uds) ->
757 new_bind | null spec_defns = NonRec bndr' rhs'
758 | otherwise = Rec ((bndr',rhs'):spec_defns)
759 -- bndr' mentions the spec_defns in its SpecEnv
760 -- Not sure why we couln't just put the spec_defns first
762 returnSM (new_bind, spec_uds)
764 specBindItself rhs_subst (Rec pairs) call_info
765 = mapSM (specDefn rhs_subst call_info) pairs `thenSM` \ stuff ->
767 (pairs', spec_defns_s, spec_uds_s) = unzip3 stuff
768 spec_defns = concat spec_defns_s
769 spec_uds = plusUDList spec_uds_s
770 new_bind = Rec (spec_defns ++ pairs')
772 returnSM (new_bind, spec_uds)
775 specDefn :: Subst -- Subst to use for RHS
776 -> CallDetails -- Info on how it is used in its scope
777 -> (Id, CoreExpr) -- The thing being bound and its un-processed RHS
778 -> SpecM ((Id, CoreExpr), -- The thing and its processed RHS
779 -- the Id may now have specialisations attached
780 [(Id,CoreExpr)], -- Extra, specialised bindings
781 UsageDetails -- Stuff to fling upwards from the RHS and its
782 ) -- specialised versions
784 specDefn subst calls (fn, rhs)
785 -- The first case is the interesting one
786 | n_tyvars == length rhs_tyvars -- Rhs of fn's defn has right number of big lambdas
787 && n_dicts <= length rhs_bndrs -- and enough dict args
788 && not (null calls_for_me) -- And there are some calls to specialise
789 && not (certainlyWillInline fn) -- And it's not small
790 -- If it's small, it's better just to inline
791 -- it than to construct lots of specialisations
792 = -- Specialise the body of the function
793 specExpr subst rhs `thenSM` \ (rhs', rhs_uds) ->
795 -- Make a specialised version for each call in calls_for_me
796 mapSM spec_call calls_for_me `thenSM` \ stuff ->
798 (spec_defns, spec_uds, spec_env_stuff) = unzip3 stuff
800 fn' = addIdSpecialisations zapped_fn spec_env_stuff
802 returnSM ((fn',rhs'),
804 rhs_uds `plusUDs` plusUDList spec_uds)
806 | otherwise -- No calls or RHS doesn't fit our preconceptions
807 = specExpr subst rhs `thenSM` \ (rhs', rhs_uds) ->
808 returnSM ((zapped_fn, rhs'), [], rhs_uds)
811 zapped_fn = modifyIdInfo zapSpecPragInfo fn
812 -- If the fn is a SpecPragmaId, make it discardable
813 -- It's role as a holder for a call instance is o'er
814 -- But it might be alive for some other reason by now.
817 (tyvars, theta, tau) = splitSigmaTy fn_type
818 n_tyvars = length tyvars
819 n_dicts = length theta
821 (rhs_tyvars, rhs_ids, rhs_body) = collectTyAndValBinders rhs
822 rhs_dicts = take n_dicts rhs_ids
823 rhs_bndrs = rhs_tyvars ++ rhs_dicts
824 body = mkLams (drop n_dicts rhs_ids) rhs_body
825 -- Glue back on the non-dict lambdas
827 calls_for_me = case lookupFM calls fn of
829 Just cs -> fmToList cs
831 ----------------------------------------------------------
832 -- Specialise to one particular call pattern
833 spec_call :: ([Maybe Type], ([DictExpr], VarSet)) -- Call instance
834 -> SpecM ((Id,CoreExpr), -- Specialised definition
835 UsageDetails, -- Usage details from specialised body
836 ([CoreBndr], [CoreExpr], CoreExpr)) -- Info for the Id's SpecEnv
837 spec_call (call_ts, (call_ds, call_fvs))
838 = ASSERT( length call_ts == n_tyvars && length call_ds == n_dicts )
839 -- Calls are only recorded for properly-saturated applications
841 -- Suppose f's defn is f = /\ a b c d -> \ d1 d2 -> rhs
842 -- Supppose the call is for f [Just t1, Nothing, Just t3, Nothing] [dx1, dx2]
844 -- Construct the new binding
845 -- f1 = SUBST[a->t1,c->t3, d1->d1', d2->d2'] (/\ b d -> rhs)
846 -- PLUS the usage-details
847 -- { d1' = dx1; d2' = dx2 }
848 -- where d1', d2' are cloned versions of d1,d2, with the type substitution applied.
850 -- Note that the substitution is applied to the whole thing.
851 -- This is convenient, but just slightly fragile. Notably:
852 -- * There had better be no name clashes in a/b/c/d
855 -- poly_tyvars = [b,d] in the example above
856 -- spec_tyvars = [a,c]
857 -- ty_args = [t1,b,t3,d]
858 poly_tyvars = [tv | (tv, Nothing) <- rhs_tyvars `zip` call_ts]
859 spec_tyvars = [tv | (tv, Just _) <- rhs_tyvars `zip` call_ts]
860 ty_args = zipWithEqual "spec_call" mk_ty_arg rhs_tyvars call_ts
862 mk_ty_arg rhs_tyvar Nothing = Type (mkTyVarTy rhs_tyvar)
863 mk_ty_arg rhs_tyvar (Just ty) = Type ty
864 rhs_subst = extendSubstList subst spec_tyvars [DoneTy ty | Just ty <- call_ts]
866 cloneBinders rhs_subst rhs_dicts `thenSM` \ (rhs_subst', rhs_dicts') ->
868 inst_args = ty_args ++ map Var rhs_dicts'
870 -- Figure out the type of the specialised function
871 spec_id_ty = mkForAllTys poly_tyvars (applyTypeToArgs rhs fn_type inst_args)
873 newIdSM fn spec_id_ty `thenSM` \ spec_f ->
874 specExpr rhs_subst' (mkLams poly_tyvars body) `thenSM` \ (spec_rhs, rhs_uds) ->
876 -- The rule to put in the function's specialisation is:
877 -- forall b,d, d1',d2'. f t1 b t3 d d1' d2' = f1 b d
878 spec_env_rule = (poly_tyvars ++ rhs_dicts',
880 mkTyApps (Var spec_f) (map mkTyVarTy poly_tyvars))
882 -- Add the { d1' = dx1; d2' = dx2 } usage stuff
883 final_uds = foldr addDictBind rhs_uds (my_zipEqual "spec_call" rhs_dicts' call_ds)
885 returnSM ((spec_f, spec_rhs),
890 my_zipEqual doc xs ys
891 | length xs /= length ys = pprPanic "my_zipEqual" (ppr xs $$ ppr ys $$ (ppr fn <+> ppr call_ts) $$ ppr rhs)
892 | otherwise = zipEqual doc xs ys
895 %************************************************************************
897 \subsubsection{UsageDetails and suchlike}
899 %************************************************************************
904 dict_binds :: !(Bag DictBind),
905 -- Floated dictionary bindings
906 -- The order is important;
907 -- in ds1 `union` ds2, bindings in ds2 can depend on those in ds1
908 -- (Remember, Bags preserve order in GHC.)
910 calls :: !CallDetails
913 type DictBind = (CoreBind, VarSet)
914 -- The set is the free vars of the binding
915 -- both tyvars and dicts
917 type DictExpr = CoreExpr
919 emptyUDs = MkUD { dict_binds = emptyBag, calls = emptyFM }
921 type ProtoUsageDetails = ([DictBind],
922 [(Id, [Maybe Type], ([DictExpr], VarSet))]
925 ------------------------------------------------------------
926 type CallDetails = FiniteMap Id CallInfo
927 type CallInfo = FiniteMap [Maybe Type] -- Nothing => unconstrained type argument
928 ([DictExpr], VarSet) -- Dict args and the vars of the whole
929 -- call (including tyvars)
930 -- [*not* include the main id itself, of course]
931 -- The finite maps eliminate duplicates
932 -- The list of types and dictionaries is guaranteed to
933 -- match the type of f
935 unionCalls :: CallDetails -> CallDetails -> CallDetails
936 unionCalls c1 c2 = plusFM_C plusFM c1 c2
938 singleCall :: (Id, [Maybe Type], [DictExpr]) -> CallDetails
939 singleCall (id, tys, dicts)
940 = unitFM id (unitFM tys (dicts, call_fvs))
942 call_fvs = exprsFreeVars dicts `unionVarSet` tys_fvs
943 tys_fvs = tyVarsOfTypes (catMaybes tys)
944 -- The type args (tys) are guaranteed to be part of the dictionary
945 -- types, because they are just the constrained types,
946 -- and the dictionary is therefore sure to be bound
947 -- inside the binding for any type variables free in the type;
948 -- hence it's safe to neglect tyvars free in tys when making
949 -- the free-var set for this call
950 -- BUT I don't trust this reasoning; play safe and include tys_fvs
952 -- We don't include the 'id' itself.
954 listToCallDetails calls
955 = foldr (unionCalls . mk_call) emptyFM calls
957 mk_call (id, tys, dicts_w_fvs) = unitFM id (unitFM tys dicts_w_fvs)
958 -- NB: the free vars of the call are provided
960 callDetailsToList calls = [ (id,tys,dicts)
961 | (id,fm) <- fmToList calls,
962 (tys,dicts) <- fmToList fm
967 || length spec_tys /= n_tyvars
968 || length dicts /= n_dicts
969 = emptyUDs -- Not overloaded
972 = MkUD {dict_binds = emptyBag,
973 calls = singleCall (f, spec_tys, dicts)
976 (tyvars, theta, tau) = splitSigmaTy (idType f)
977 constrained_tyvars = tyVarsOfTheta theta
978 n_tyvars = length tyvars
979 n_dicts = length theta
981 spec_tys = [mk_spec_ty tv ty | (tv, Type ty) <- tyvars `zip` args]
982 dicts = [dict_expr | (_, dict_expr) <- theta `zip` (drop n_tyvars args)]
984 mk_spec_ty tyvar ty | tyvar `elemVarSet` constrained_tyvars
989 ------------------------------------------------------------
990 plusUDs :: UsageDetails -> UsageDetails -> UsageDetails
991 plusUDs (MkUD {dict_binds = db1, calls = calls1})
992 (MkUD {dict_binds = db2, calls = calls2})
993 = MkUD {dict_binds = d, calls = c}
995 d = db1 `unionBags` db2
996 c = calls1 `unionCalls` calls2
998 plusUDList = foldr plusUDs emptyUDs
1000 -- zapCalls deletes calls to ids from uds
1001 zapCalls ids uds = uds {calls = delListFromFM (calls uds) ids}
1003 mkDB bind = (bind, bind_fvs bind)
1005 bind_fvs (NonRec bndr rhs) = exprFreeVars rhs
1006 bind_fvs (Rec prs) = foldl delVarSet rhs_fvs bndrs
1009 rhs_fvs = unionVarSets [exprFreeVars rhs | (bndr,rhs) <- prs]
1011 addDictBind (dict,rhs) uds = uds { dict_binds = mkDB (NonRec dict rhs) `consBag` dict_binds uds }
1013 dumpAllDictBinds (MkUD {dict_binds = dbs}) binds
1014 = foldrBag add binds dbs
1016 add (bind,_) binds = bind : binds
1018 dumpUDs :: [CoreBndr]
1019 -> UsageDetails -> CoreExpr
1020 -> (UsageDetails, CoreExpr)
1021 dumpUDs bndrs uds body
1022 = (free_uds, foldr add_let body dict_binds)
1024 (free_uds, (dict_binds, _)) = splitUDs bndrs uds
1025 add_let (bind,_) body = Let bind body
1027 splitUDs :: [CoreBndr]
1029 -> (UsageDetails, -- These don't mention the binders
1030 ProtoUsageDetails) -- These do
1032 splitUDs bndrs uds@(MkUD {dict_binds = orig_dbs,
1033 calls = orig_calls})
1035 = if isEmptyBag dump_dbs && null dump_calls then
1036 -- Common case: binder doesn't affect floats
1040 -- Binders bind some of the fvs of the floats
1041 (MkUD {dict_binds = free_dbs,
1042 calls = listToCallDetails free_calls},
1043 (bagToList dump_dbs, dump_calls)
1047 bndr_set = mkVarSet bndrs
1049 (free_dbs, dump_dbs, dump_idset)
1050 = foldlBag dump_db (emptyBag, emptyBag, bndr_set) orig_dbs
1051 -- Important that it's foldl not foldr;
1052 -- we're accumulating the set of dumped ids in dump_set
1054 -- Filter out any calls that mention things that are being dumped
1055 orig_call_list = callDetailsToList orig_calls
1056 (dump_calls, free_calls) = partition captured orig_call_list
1057 captured (id,tys,(dicts, fvs)) = fvs `intersectsVarSet` dump_idset
1058 || id `elemVarSet` dump_idset
1060 dump_db (free_dbs, dump_dbs, dump_idset) db@(bind, fvs)
1061 | dump_idset `intersectsVarSet` fvs -- Dump it
1062 = (free_dbs, dump_dbs `snocBag` db,
1063 dump_idset `unionVarSet` mkVarSet (bindersOf bind))
1065 | otherwise -- Don't dump it
1066 = (free_dbs `snocBag` db, dump_dbs, dump_idset)
1070 %************************************************************************
1072 \subsubsection{Boring helper functions}
1074 %************************************************************************
1077 lookupId:: IdEnv Id -> Id -> Id
1078 lookupId env id = case lookupVarEnv env id of
1082 ----------------------------------------
1083 type SpecM a = UniqSM a
1088 getUniqSM = getUniqueUs
1089 getUniqSupplySM = getUs
1090 setUniqSupplySM = setUs
1094 mapAndCombineSM f [] = returnSM ([], emptyUDs)
1095 mapAndCombineSM f (x:xs) = f x `thenSM` \ (y, uds1) ->
1096 mapAndCombineSM f xs `thenSM` \ (ys, uds2) ->
1097 returnSM (y:ys, uds1 `plusUDs` uds2)
1099 cloneBindSM :: Subst -> CoreBind -> SpecM (Subst, Subst, CoreBind)
1100 -- Clone the binders of the bind; return new bind with the cloned binders
1101 -- Return the substitution to use for RHSs, and the one to use for the body
1102 cloneBindSM subst (NonRec bndr rhs)
1103 = getUs `thenUs` \ us ->
1105 (subst', us', bndr') = substAndCloneId subst us bndr
1108 returnUs (subst, subst', NonRec bndr' rhs)
1110 cloneBindSM subst (Rec pairs)
1111 = getUs `thenUs` \ us ->
1113 (subst', us', bndrs') = substAndCloneIds subst us (map fst pairs)
1116 returnUs (subst', subst', Rec (bndrs' `zip` map snd pairs))
1118 cloneBinders subst bndrs
1119 = getUs `thenUs` \ us ->
1121 (subst', us', bndrs') = substAndCloneIds subst us bndrs
1124 returnUs (subst', bndrs')
1127 newIdSM old_id new_ty
1128 = getUniqSM `thenSM` \ uniq ->
1130 -- Give the new Id a similar occurrence name to the old one
1131 name = idName old_id
1132 new_id = mkUserLocal (mkSpecOcc (nameOccName name)) uniq new_ty (getSrcLoc name)
1134 -- If the old Id was exported, make the new one non-discardable,
1135 -- else we will discard it since it doesn't seem to be called.
1136 new_id' | isExportedId old_id = setIdNoDiscard new_id
1137 | otherwise = new_id
1142 = getUniqSM `thenSM` \ uniq ->
1143 returnSM (mkSysTyVar uniq boxedTypeKind)
1147 Old (but interesting) stuff about unboxed bindings
1148 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1150 What should we do when a value is specialised to a *strict* unboxed value?
1152 map_*_* f (x:xs) = let h = f x
1156 Could convert let to case:
1158 map_*_Int# f (x:xs) = case f x of h# ->
1162 This may be undesirable since it forces evaluation here, but the value
1163 may not be used in all branches of the body. In the general case this
1164 transformation is impossible since the mutual recursion in a letrec
1165 cannot be expressed as a case.
1167 There is also a problem with top-level unboxed values, since our
1168 implementation cannot handle unboxed values at the top level.
1170 Solution: Lift the binding of the unboxed value and extract it when it
1173 map_*_Int# f (x:xs) = let h = case (f x) of h# -> _Lift h#
1178 Now give it to the simplifier and the _Lifting will be optimised away.
1180 The benfit is that we have given the specialised "unboxed" values a
1181 very simplep lifted semantics and then leave it up to the simplifier to
1182 optimise it --- knowing that the overheads will be removed in nearly
1185 In particular, the value will only be evaluted in the branches of the
1186 program which use it, rather than being forced at the point where the
1187 value is bound. For example:
1189 filtermap_*_* p f (x:xs)
1196 filtermap_*_Int# p f (x:xs)
1197 = let h = case (f x) of h# -> _Lift h#
1200 True -> case h of _Lift h#
1204 The binding for h can still be inlined in the one branch and the
1205 _Lifting eliminated.
1208 Question: When won't the _Lifting be eliminated?
1210 Answer: When they at the top-level (where it is necessary) or when
1211 inlining would duplicate work (or possibly code depending on
1212 options). However, the _Lifting will still be eliminated if the
1213 strictness analyser deems the lifted binding strict.