2 % (c) The GRASP/AQUA Project, Glasgow University, 1992-1998
4 \section[SpecConstr]{Specialise over constructors}
11 #include "HsVersions.h"
14 import CoreLint ( showPass, endPass )
15 import CoreUtils ( exprType, mkPiTypes )
16 import CoreFVs ( exprsFreeVars )
17 import CoreSubst ( Subst, mkSubst, substExpr )
18 import CoreTidy ( tidyRules )
19 import PprCore ( pprRules )
20 import WwLib ( mkWorkerArgs )
21 import DataCon ( dataConRepArity, isVanillaDataCon, dataConTyVars )
22 import Type ( tyConAppArgs, tyVarsOfTypes )
23 import Rules ( matchN )
24 import Unify ( coreRefineTys )
25 import Id ( Id, idName, idType, isDataConWorkId_maybe,
26 mkUserLocal, mkSysLocal, idUnfolding, isLocalId )
30 import Name ( nameOccName, nameSrcLoc )
31 import Rules ( addIdSpecialisations, mkLocalRule, rulesOfBinds )
32 import OccName ( mkSpecOcc )
33 import ErrUtils ( dumpIfSet_dyn )
34 import DynFlags ( DynFlags, DynFlag(..) )
35 import BasicTypes ( Activation(..) )
36 import Maybes ( orElse, catMaybes, isJust )
37 import Util ( zipWithEqual, lengthAtLeast, notNull )
38 import List ( nubBy, partition )
45 -----------------------------------------------------
47 -----------------------------------------------------
52 drop n (x:xs) = drop (n-1) xs
54 After the first time round, we could pass n unboxed. This happens in
55 numerical code too. Here's what it looks like in Core:
57 drop n xs = case xs of
62 _ -> drop (I# (n# -# 1#)) xs
64 Notice that the recursive call has an explicit constructor as argument.
65 Noticing this, we can make a specialised version of drop
67 RULE: drop (I# n#) xs ==> drop' n# xs
69 drop' n# xs = let n = I# n# in ...orig RHS...
71 Now the simplifier will apply the specialisation in the rhs of drop', giving
73 drop' n# xs = case xs of
77 _ -> drop (n# -# 1#) xs
81 We'd also like to catch cases where a parameter is carried along unchanged,
82 but evaluated each time round the loop:
84 f i n = if i>0 || i>n then i else f (i*2) n
86 Here f isn't strict in n, but we'd like to avoid evaluating it each iteration.
87 In Core, by the time we've w/wd (f is strict in i) we get
89 f i# n = case i# ># 0 of
91 True -> case n of n' { I# n# ->
94 True -> f (i# *# 2#) n'
96 At the call to f, we see that the argument, n is know to be (I# n#),
97 and n is evaluated elsewhere in the body of f, so we can play the same
103 We must be careful not to allocate the same constructor twice. Consider
104 f p = (...(case p of (a,b) -> e)...p...,
105 ...let t = (r,s) in ...t...(f t)...)
106 At the recursive call to f, we can see that t is a pair. But we do NOT want
107 to make a specialised copy:
108 f' a b = let p = (a,b) in (..., ...)
109 because now t is allocated by the caller, then r and s are passed to the
110 recursive call, which allocates the (r,s) pair again.
113 (a) the argument p is used in other than a case-scrutinsation way.
114 (b) the argument to the call is not a 'fresh' tuple; you have to
115 look into its unfolding to see that it's a tuple
117 Hence the "OR" part of Note [Good arguments] below.
119 ALTERNATIVE: pass both boxed and unboxed versions. This no longer saves
120 allocation, but does perhaps save evals. In the RULE we'd have
123 f (I# x#) = f' (I# x#) x#
125 If at the call site the (I# x) was an unfolding, then we'd have to
126 rely on CSE to eliminate the duplicate allocation.... This alternative
127 doesn't look attractive enough to pursue.
130 Note [Good arguments]
131 ~~~~~~~~~~~~~~~~~~~~~
134 * A self-recursive function. Ignore mutual recursion for now,
135 because it's less common, and the code is simpler for self-recursion.
139 a) At a recursive call, one or more parameters is an explicit
140 constructor application
142 That same parameter is scrutinised by a case somewhere in
143 the RHS of the function
147 b) At a recursive call, one or more parameters has an unfolding
148 that is an explicit constructor application
150 That same parameter is scrutinised by a case somewhere in
151 the RHS of the function
153 Those are the only uses of the parameter (see Note [Reboxing])
156 What to abstract over
157 ~~~~~~~~~~~~~~~~~~~~~
158 There's a bit of a complication with type arguments. If the call
161 f p = ...f ((:) [a] x xs)...
163 then our specialised function look like
165 f_spec x xs = let p = (:) [a] x xs in ....as before....
167 This only makes sense if either
168 a) the type variable 'a' is in scope at the top of f, or
169 b) the type variable 'a' is an argument to f (and hence fs)
171 Actually, (a) may hold for value arguments too, in which case
172 we may not want to pass them. Supose 'x' is in scope at f's
173 defn, but xs is not. Then we'd like
175 f_spec xs = let p = (:) [a] x xs in ....as before....
177 Similarly (b) may hold too. If x is already an argument at the
178 call, no need to pass it again.
180 Finally, if 'a' is not in scope at the call site, we could abstract
181 it as we do the term variables:
183 f_spec a x xs = let p = (:) [a] x xs in ...as before...
185 So the grand plan is:
187 * abstract the call site to a constructor-only pattern
188 e.g. C x (D (f p) (g q)) ==> C s1 (D s2 s3)
190 * Find the free variables of the abstracted pattern
192 * Pass these variables, less any that are in scope at
193 the fn defn. But see Note [Shadowing] below.
196 NOTICE that we only abstract over variables that are not in scope,
197 so we're in no danger of shadowing variables used in "higher up"
203 In this pass we gather up usage information that may mention variables
204 that are bound between the usage site and the definition site; or (more
205 seriously) may be bound to something different at the definition site.
208 f x = letrec g y v = let x = ...
211 Since 'x' is in scope at the call site, we may make a rewrite rule that
213 RULE forall a,b. g (a,b) x = ...
214 But this rule will never match, because it's really a different 'x' at
215 the call site -- and that difference will be manifest by the time the
216 simplifier gets to it. [A worry: the simplifier doesn't *guarantee*
217 no-shadowing, so perhaps it may not be distinct?]
219 Anyway, the rule isn't actually wrong, it's just not useful. One possibility
220 is to run deShadowBinds before running SpecConstr, but instead we run the
221 simplifier. That gives the simplest possible program for SpecConstr to
222 chew on; and it virtually guarantees no shadowing.
224 Note [Specialising for constant parameters]
225 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
226 This one is about specialising on a *constant* (but not necessarily
227 constructor) argument
229 foo :: Int -> (Int -> Int) -> Int
231 foo m f = foo (f m) (+1)
235 lvl_rmV :: GHC.Base.Int -> GHC.Base.Int
237 \ (ds_dlk :: GHC.Base.Int) ->
238 case ds_dlk of wild_alH { GHC.Base.I# x_alG ->
239 GHC.Base.I# (GHC.Prim.+# x_alG 1)
241 T.$wfoo :: GHC.Prim.Int# -> (GHC.Base.Int -> GHC.Base.Int) ->
244 \ (ww_sme :: GHC.Prim.Int#) (w_smg :: GHC.Base.Int -> GHC.Base.Int) ->
245 case ww_sme of ds_Xlw {
247 case w_smg (GHC.Base.I# ds_Xlw) of w1_Xmo { GHC.Base.I# ww1_Xmz ->
248 T.$wfoo ww1_Xmz lvl_rmV
253 The recursive call has lvl_rmV as its argument, so we could create a specialised copy
254 with that argument baked in; that is, not passed at all. Now it can perhaps be inlined.
256 When is this worth it? Call the constant 'lvl'
257 - If 'lvl' has an unfolding that is a constructor, see if the corresponding
258 parameter is scrutinised anywhere in the body.
260 - If 'lvl' has an unfolding that is a inlinable function, see if the corresponding
261 parameter is applied (...to enough arguments...?)
263 Also do this is if the function has RULES?
267 Note [Specialising for lambda parameters]
268 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
269 foo :: Int -> (Int -> Int) -> Int
271 foo m f = foo (f m) (\n -> n-m)
273 This is subtly different from the previous one in that we get an
274 explicit lambda as the argument:
276 T.$wfoo :: GHC.Prim.Int# -> (GHC.Base.Int -> GHC.Base.Int) ->
279 \ (ww_sm8 :: GHC.Prim.Int#) (w_sma :: GHC.Base.Int -> GHC.Base.Int) ->
280 case ww_sm8 of ds_Xlr {
282 case w_sma (GHC.Base.I# ds_Xlr) of w1_Xmf { GHC.Base.I# ww1_Xmq ->
285 (\ (n_ad3 :: GHC.Base.Int) ->
286 case n_ad3 of wild_alB { GHC.Base.I# x_alA ->
287 GHC.Base.I# (GHC.Prim.-# x_alA ds_Xlr)
293 I wonder if SpecConstr couldn't be extended to handle this? After all,
294 lambda is a sort of constructor for functions and perhaps it already
295 has most of the necessary machinery?
297 Furthermore, there's an immediate win, because you don't need to allocate the lamda
298 at the call site; and if perchance it's called in the recursive call, then you
299 may avoid allocating it altogether. Just like for constructors.
301 Looks cool, but probably rare...but it might be easy to implement.
303 -----------------------------------------------------
304 Stuff not yet handled
305 -----------------------------------------------------
307 Here are notes arising from Roman's work that I don't want to lose.
313 foo :: Int -> T Int -> Int
315 foo x t | even x = case t of { T n -> foo (x-n) t }
316 | otherwise = foo (x-1) t
318 SpecConstr does no specialisation, because the second recursive call
319 looks like a boxed use of the argument. A pity.
321 $wfoo_sFw :: GHC.Prim.Int# -> T.T GHC.Base.Int -> GHC.Prim.Int#
323 \ (ww_sFo [Just L] :: GHC.Prim.Int#) (w_sFq [Just L] :: T.T GHC.Base.Int) ->
324 case ww_sFo of ds_Xw6 [Just L] {
326 case GHC.Prim.remInt# ds_Xw6 2 of wild1_aEF [Dead Just A] {
327 __DEFAULT -> $wfoo_sFw (GHC.Prim.-# ds_Xw6 1) w_sFq;
329 case w_sFq of wild_Xy [Just L] { T.T n_ad5 [Just U(L)] ->
330 case n_ad5 of wild1_aET [Just A] { GHC.Base.I# y_aES [Just L] ->
331 $wfoo_sFw (GHC.Prim.-# ds_Xw6 y_aES) wild_Xy
337 data a :*: b = !a :*: !b
340 foo :: (Int :*: T Int) -> Int
342 foo (x :*: t) | even x = case t of { T n -> foo ((x-n) :*: t) }
343 | otherwise = foo ((x-1) :*: t)
345 Very similar to the previous one, except that the parameters are now in
346 a strict tuple. Before SpecConstr, we have
348 $wfoo_sG3 :: GHC.Prim.Int# -> T.T GHC.Base.Int -> GHC.Prim.Int#
350 \ (ww_sFU [Just L] :: GHC.Prim.Int#) (ww_sFW [Just L] :: T.T
352 case ww_sFU of ds_Xws [Just L] {
354 case GHC.Prim.remInt# ds_Xws 2 of wild1_aEZ [Dead Just A] {
356 case ww_sFW of tpl_B2 [Just L] { T.T a_sFo [Just A] ->
357 $wfoo_sG3 (GHC.Prim.-# ds_Xws 1) tpl_B2 -- $wfoo1
360 case ww_sFW of wild_XB [Just A] { T.T n_ad7 [Just S(L)] ->
361 case n_ad7 of wild1_aFd [Just L] { GHC.Base.I# y_aFc [Just L] ->
362 $wfoo_sG3 (GHC.Prim.-# ds_Xws y_aFc) wild_XB -- $wfoo2
366 We get two specialisations:
367 "SC:$wfoo1" [0] __forall {a_sFB :: GHC.Base.Int sc_sGC :: GHC.Prim.Int#}
368 Foo.$wfoo sc_sGC (Foo.T @ GHC.Base.Int a_sFB)
369 = Foo.$s$wfoo1 a_sFB sc_sGC ;
370 "SC:$wfoo2" [0] __forall {y_aFp :: GHC.Prim.Int# sc_sGC :: GHC.Prim.Int#}
371 Foo.$wfoo sc_sGC (Foo.T @ GHC.Base.Int (GHC.Base.I# y_aFp))
372 = Foo.$s$wfoo y_aFp sc_sGC ;
374 But perhaps the first one isn't good. After all, we know that tpl_B2 is
375 a T (I# x) really, because T is strict and Int has one constructor. (We can't
376 unbox the strict fields, becuase T is polymorphic!)
380 %************************************************************************
382 \subsection{Top level wrapper stuff}
384 %************************************************************************
387 specConstrProgram :: DynFlags -> UniqSupply -> [CoreBind] -> IO [CoreBind]
388 specConstrProgram dflags us binds
390 showPass dflags "SpecConstr"
392 let (binds', _) = initUs us (go emptyScEnv binds)
394 endPass dflags "SpecConstr" Opt_D_dump_spec binds'
396 dumpIfSet_dyn dflags Opt_D_dump_rules "Top-level specialisations"
397 (pprRules (tidyRules emptyTidyEnv (rulesOfBinds binds')))
401 go env [] = returnUs []
402 go env (bind:binds) = scBind env bind `thenUs` \ (env', _, bind') ->
403 go env' binds `thenUs` \ binds' ->
404 returnUs (bind' : binds')
408 %************************************************************************
410 \subsection{Environment: goes downwards}
412 %************************************************************************
415 data ScEnv = SCE { scope :: InScopeEnv,
416 -- Binds all non-top-level variables in scope
421 type InScopeEnv = VarEnv HowBound
423 type ConstrEnv = IdEnv ConValue
424 data ConValue = CV AltCon [CoreArg]
425 -- Variables known to be bound to a constructor
426 -- in a particular case alternative
429 instance Outputable ConValue where
430 ppr (CV con args) = ppr con <+> interpp'SP args
432 refineConstrEnv :: Subst -> ConstrEnv -> ConstrEnv
433 -- The substitution is a type substitution only
434 refineConstrEnv subst env = mapVarEnv refine_con_value env
436 refine_con_value (CV con args) = CV con (map (substExpr subst) args)
438 emptyScEnv = SCE { scope = emptyVarEnv, cons = emptyVarEnv }
440 data HowBound = RecFun -- These are the recursive functions for which
441 -- we seek interesting call patterns
443 | RecArg -- These are those functions' arguments; we are
444 -- interested to see if those arguments are scrutinised
446 | Other -- We track all others so we know what's in scope
447 -- This is used in spec_one to check what needs to be
448 -- passed as a parameter and what is in scope at the
449 -- function definition site
451 instance Outputable HowBound where
452 ppr RecFun = text "RecFun"
453 ppr RecArg = text "RecArg"
454 ppr Other = text "Other"
456 lookupScopeEnv env v = lookupVarEnv (scope env) v
458 extendBndrs env bndrs = env { scope = extendVarEnvList (scope env) [(b,Other) | b <- bndrs] }
459 extendBndr env bndr = env { scope = extendVarEnv (scope env) bndr Other }
464 -- we want to bind b, and perhaps scrut too, to (C x y)
465 extendCaseBndrs :: ScEnv -> Id -> CoreExpr -> AltCon -> [Var] -> ScEnv
466 extendCaseBndrs env case_bndr scrut DEFAULT alt_bndrs
467 = extendBndrs env (case_bndr : alt_bndrs)
469 extendCaseBndrs env case_bndr scrut con@(LitAlt lit) alt_bndrs
470 = ASSERT( null alt_bndrs ) extendAlt env case_bndr scrut (CV con []) []
472 extendCaseBndrs env case_bndr scrut con@(DataAlt data_con) alt_bndrs
473 | isVanillaDataCon data_con
474 = extendAlt env case_bndr scrut (CV con vanilla_args) alt_bndrs
477 = extendAlt env1 case_bndr scrut (CV con gadt_args) alt_bndrs
479 vanilla_args = map Type (tyConAppArgs (idType case_bndr)) ++
480 map varToCoreExpr alt_bndrs
482 gadt_args = map (substExpr subst . varToCoreExpr) alt_bndrs
483 -- This call generates some bogus warnings from substExpr,
484 -- because it's inconvenient to put all the Ids in scope
485 -- Will be fixed when we move to FC
487 (alt_tvs, _) = span isTyVar alt_bndrs
488 Just (tv_subst, is_local) = coreRefineTys data_con alt_tvs (idType case_bndr)
489 subst = mkSubst in_scope tv_subst emptyVarEnv -- No Id substitition
490 in_scope = mkInScopeSet (tyVarsOfTypes (varEnvElts tv_subst))
492 env1 | is_local = env
493 | otherwise = env { cons = refineConstrEnv subst (cons env) }
496 extendAlt :: ScEnv -> Id -> CoreExpr -> ConValue -> [Var] -> ScEnv
497 extendAlt env case_bndr scrut val alt_bndrs
499 env1 = SCE { scope = extendVarEnvList (scope env) [(b,Other) | b <- case_bndr : alt_bndrs],
500 cons = extendVarEnv (cons env) case_bndr val }
503 Var v -> -- Bind the scrutinee in the ConstrEnv if it's a variable
504 -- Also forget if the scrutinee is a RecArg, because we're
505 -- now in the branch of a case, and we don't want to
506 -- record a non-scrutinee use of v if we have
507 -- case v of { (a,b) -> ...(f v)... }
508 SCE { scope = extendVarEnv (scope env1) v Other,
509 cons = extendVarEnv (cons env1) v val }
512 -- When we encounter a recursive function binding
514 -- we want to extend the scope env with bindings
515 -- that record that f is a RecFn and x,y are RecArgs
516 extendRecBndr env fn bndrs
517 = env { scope = scope env `extendVarEnvList`
518 ((fn,RecFun): [(bndr,RecArg) | bndr <- bndrs]) }
522 %************************************************************************
524 \subsection{Usage information: flows upwards}
526 %************************************************************************
531 calls :: !(IdEnv ([Call])), -- Calls
532 -- The functions are a subset of the
533 -- RecFuns in the ScEnv
535 occs :: !(IdEnv ArgOcc) -- Information on argument occurrences
536 } -- The variables are a subset of the
537 -- RecArg in the ScEnv
539 type Call = (ConstrEnv, [CoreArg])
540 -- The arguments of the call, together with the
541 -- env giving the constructor bindings at the call site
543 nullUsage = SCU { calls = emptyVarEnv, occs = emptyVarEnv }
545 combineUsage u1 u2 = SCU { calls = plusVarEnv_C (++) (calls u1) (calls u2),
546 occs = plusVarEnv_C combineOcc (occs u1) (occs u2) }
548 combineUsages [] = nullUsage
549 combineUsages us = foldr1 combineUsage us
551 lookupOcc :: ScUsage -> Var -> (ScUsage, ArgOcc)
552 lookupOcc (SCU { calls = sc_calls, occs = sc_occs }) bndr
553 = (SCU {calls = sc_calls, occs = delVarEnv sc_occs bndr},
554 lookupVarEnv sc_occs bndr `orElse` NoOcc)
556 lookupOccs :: ScUsage -> [Var] -> (ScUsage, [ArgOcc])
557 lookupOccs (SCU { calls = sc_calls, occs = sc_occs }) bndrs
558 = (SCU {calls = sc_calls, occs = delVarEnvList sc_occs bndrs},
559 [lookupVarEnv sc_occs b `orElse` NoOcc | b <- bndrs])
561 data ArgOcc = NoOcc -- Doesn't occur at all; or a type argument
562 | UnkOcc -- Used in some unknown way
564 | ScrutOcc (UniqFM [ArgOcc]) -- See Note [ScrutOcc]
566 | BothOcc -- Definitely taken apart, *and* perhaps used in some other way
570 An occurrence of ScrutOcc indicates that the thing is *only* taken apart or applied.
572 Functions, litersl: ScrutOcc emptyUFM
573 Data constructors: ScrutOcc subs,
575 where (subs :: UniqFM [ArgOcc]) gives usage of the *pattern-bound* components,
576 The domain of the UniqFM is the Unique of the data constructor
578 The [ArgOcc] is the occurrences of the *pattern-bound* components
579 of the data structure. E.g.
580 data T a = forall b. MkT a b (b->a)
581 A pattern binds b, x::a, y::b, z::b->a, but not 'a'!
585 instance Outputable ArgOcc where
586 ppr (ScrutOcc xs) = ptext SLIT("scrut-occ") <+> ppr xs
587 ppr UnkOcc = ptext SLIT("unk-occ")
588 ppr BothOcc = ptext SLIT("both-occ")
589 ppr NoOcc = ptext SLIT("no-occ")
591 combineOcc NoOcc occ = occ
592 combineOcc occ NoOcc = occ
593 combineOcc (ScrutOcc xs) (ScrutOcc ys) = ScrutOcc (plusUFM_C combineOccs xs ys)
594 combineOcc UnkOcc UnkOcc = UnkOcc
595 combineOcc _ _ = BothOcc
597 combineOccs :: [ArgOcc] -> [ArgOcc] -> [ArgOcc]
598 combineOccs xs ys = zipWithEqual "combineOccs" combineOcc xs ys
600 conArgOccs :: ArgOcc -> AltCon -> [ArgOcc]
601 -- Find usage of components of data con; returns [UnkOcc...] if unknown
602 -- See Note [ScrutOcc] for the extra UnkOccs in the vanilla datacon case
604 conArgOccs (ScrutOcc fm) (DataAlt dc)
605 | Just pat_arg_occs <- lookupUFM fm dc
606 = tyvar_unks ++ pat_arg_occs
608 tyvar_unks | isVanillaDataCon dc = [UnkOcc | tv <- dataConTyVars dc]
611 conArgOccs other con = repeat UnkOcc
615 %************************************************************************
617 \subsection{The main recursive function}
619 %************************************************************************
621 The main recursive function gathers up usage information, and
622 creates specialised versions of functions.
625 scExpr :: ScEnv -> CoreExpr -> UniqSM (ScUsage, CoreExpr)
626 -- The unique supply is needed when we invent
627 -- a new name for the specialised function and its args
629 scExpr env e@(Type t) = returnUs (nullUsage, e)
630 scExpr env e@(Lit l) = returnUs (nullUsage, e)
631 scExpr env e@(Var v) = returnUs (varUsage env v UnkOcc, e)
632 scExpr env (Note n e) = scExpr env e `thenUs` \ (usg,e') ->
633 returnUs (usg, Note n e')
634 scExpr env (Lam b e) = scExpr (extendBndr env b) e `thenUs` \ (usg,e') ->
635 returnUs (usg, Lam b e')
637 scExpr env (Case scrut b ty alts)
638 = do { (alt_usgs, alt_occs, alts') <- mapAndUnzip3Us sc_alt alts
639 ; let (alt_usg, b_occ) = lookupOcc (combineUsages alt_usgs) b
640 scrut_occ = foldr combineOcc b_occ alt_occs
641 -- The combined usage of the scrutinee is given
642 -- by scrut_occ, which is passed to scScrut, which
643 -- in turn treats a bare-variable scrutinee specially
644 ; (scrut_usg, scrut') <- scScrut env scrut scrut_occ
645 ; return (alt_usg `combineUsage` scrut_usg,
646 Case scrut' b ty alts') }
649 = do { let env1 = extendCaseBndrs env b scrut con bs
650 ; (usg,rhs') <- scExpr env1 rhs
651 ; let (usg', arg_occs) = lookupOccs usg bs
652 scrut_occ = case con of
653 DataAlt dc -> ScrutOcc (unitUFM dc arg_occs)
654 other -> ScrutOcc emptyUFM
655 ; return (usg', scrut_occ, (con,bs,rhs')) }
657 scExpr env (Let bind body)
658 = scBind env bind `thenUs` \ (env', bind_usg, bind') ->
659 scExpr env' body `thenUs` \ (body_usg, body') ->
660 returnUs (bind_usg `combineUsage` body_usg, Let bind' body')
662 scExpr env e@(App _ _)
663 = do { let (fn, args) = collectArgs e
664 ; (fn_usg, fn') <- scScrut env fn (ScrutOcc emptyUFM)
665 -- Process the function too. It's almost always a variable,
666 -- but not always. In particular, if this pass follows float-in,
667 -- which it may, we can get
668 -- (let f = ...f... in f) arg1 arg2
669 -- We use scScrut to record the fact that the function is called
670 -- Perhpas we should check that it has at least one value arg,
671 -- but currently we don't bother
673 ; (arg_usgs, args') <- mapAndUnzipUs (scExpr env) args
674 ; let call_usg = case fn of
675 Var f | Just RecFun <- lookupScopeEnv env f
676 -> SCU { calls = unitVarEnv f [(cons env, args)],
679 ; return (combineUsages arg_usgs `combineUsage` fn_usg
680 `combineUsage` call_usg,
684 ----------------------
685 scScrut :: ScEnv -> CoreExpr -> ArgOcc -> UniqSM (ScUsage, CoreExpr)
686 -- Used for the scrutinee of a case,
687 -- or the function of an application
688 scScrut env e@(Var v) occ = returnUs (varUsage env v occ, e)
689 scScrut env e occ = scExpr env e
692 ----------------------
693 scBind :: ScEnv -> CoreBind -> UniqSM (ScEnv, ScUsage, CoreBind)
694 scBind env (Rec [(fn,rhs)])
696 = scExpr env_fn_body body `thenUs` \ (usg, body') ->
697 specialise env fn bndrs body' usg `thenUs` \ (rules, spec_prs) ->
698 -- Note body': the specialised copies should be based on the
699 -- optimised version of the body, in case there were
700 -- nested functions inside.
702 SCU { calls = calls, occs = occs } = usg
704 returnUs (extendBndr env fn, -- For the body of the letrec, just
705 -- extend the env with Other to record
706 -- that it's in scope; no funny RecFun business
707 SCU { calls = calls `delVarEnv` fn, occs = occs `delVarEnvList` val_bndrs},
708 Rec ((fn `addIdSpecialisations` rules, mkLams bndrs body') : spec_prs))
710 (bndrs,body) = collectBinders rhs
711 val_bndrs = filter isId bndrs
712 env_fn_body = extendRecBndr env fn bndrs
715 = mapAndUnzipUs do_one prs `thenUs` \ (usgs, prs') ->
716 returnUs (extendBndrs env (map fst prs), combineUsages usgs, Rec prs')
718 do_one (bndr,rhs) = scExpr env rhs `thenUs` \ (usg, rhs') ->
719 returnUs (usg, (bndr,rhs'))
721 scBind env (NonRec bndr rhs)
722 = scExpr env rhs `thenUs` \ (usg, rhs') ->
723 returnUs (extendBndr env bndr, usg, NonRec bndr rhs')
725 ----------------------
727 | Just RecArg <- lookupScopeEnv env v = SCU { calls = emptyVarEnv,
728 occs = unitVarEnv v use }
729 | otherwise = nullUsage
733 %************************************************************************
735 \subsection{The specialiser}
737 %************************************************************************
742 -> [CoreBndr] -> CoreExpr -- Its RHS
743 -> ScUsage -- Info on usage
744 -> UniqSM ([CoreRule], -- Rules
745 [(Id,CoreExpr)]) -- Bindings
747 specialise env fn bndrs body body_usg
748 = do { let (_, bndr_occs) = lookupOccs body_usg bndrs
750 ; mb_calls <- mapM (callToPats (scope env) bndr_occs)
751 (lookupVarEnv (calls body_usg) fn `orElse` [])
753 ; let good_calls :: [([Var], [CoreArg])]
754 good_calls = catMaybes mb_calls
755 in_scope = mkInScopeSet $ unionVarSets $
756 [ exprsFreeVars pats `delVarSetList` vs
757 | (vs,pats) <- good_calls ]
758 uniq_calls = nubBy (same_call in_scope) good_calls
760 mapAndUnzipUs (spec_one env fn (mkLams bndrs body))
761 (uniq_calls `zip` [1..]) }
763 -- Two calls are the same if they match both ways
764 same_call in_scope (vs1,as1)(vs2,as2)
765 = isJust (matchN in_scope vs1 as1 as2)
766 && isJust (matchN in_scope vs2 as2 as1)
768 callToPats :: InScopeEnv -> [ArgOcc] -> Call
769 -> UniqSM (Maybe ([Var], [CoreExpr]))
770 -- The VarSet is the variables to quantify over in the rule
771 -- The [CoreExpr] are the argument patterns for the rule
772 callToPats in_scope bndr_occs (con_env, args)
773 | length args < length bndr_occs -- Check saturated
776 = do { prs <- argsToPats in_scope con_env (args `zip` bndr_occs)
777 ; let (good_pats, pats) = unzip prs
778 pat_fvs = varSetElems (exprsFreeVars pats)
779 qvars = filter (not . (`elemVarEnv` in_scope)) pat_fvs
780 -- Quantify over variables that are not in sccpe
781 -- See Note [Shadowing] at the top
784 then return (Just (qvars, pats))
785 else return Nothing }
787 ---------------------
790 -> CoreExpr -- Rhs of the original function
791 -> (([Var], [CoreArg]), Int)
792 -> UniqSM (CoreRule, (Id,CoreExpr)) -- Rule and binding
794 -- spec_one creates a specialised copy of the function, together
795 -- with a rule for using it. I'm very proud of how short this
796 -- function is, considering what it does :-).
802 f = /\b \y::[(a,b)] -> ....f (b,c) ((:) (a,(b,c)) (x,v) (h w))...
803 [c::*, v::(b,c) are presumably bound by the (...) part]
805 f_spec = /\ b c \ v::(b,c) hw::[(a,(b,c))] ->
806 (...entire RHS of f...) (b,c) ((:) (a,(b,c)) (x,v) hw)
808 RULE: forall b::* c::*, -- Note, *not* forall a, x
812 f (b,c) ((:) (a,(b,c)) (x,v) hw) = f_spec b c v hw
815 spec_one env fn rhs ((vars_to_bind, pats), rule_number)
816 = getUniqueUs `thenUs` \ spec_uniq ->
819 fn_loc = nameSrcLoc fn_name
820 spec_occ = mkSpecOcc (nameOccName fn_name)
822 -- Put the type variables first; the type of a term
823 -- variable may mention a type variable
824 (tvs, ids) = partition isTyVar vars_to_bind
826 spec_body = mkApps rhs pats
827 body_ty = exprType spec_body
829 (spec_lam_args, spec_call_args) = mkWorkerArgs bndrs body_ty
830 -- Usual w/w hack to avoid generating
831 -- a spec_rhs of unlifted type and no args
833 rule_name = mkFastString ("SC:" ++ showSDoc (ppr fn <> int rule_number))
834 spec_rhs = mkLams spec_lam_args spec_body
835 spec_id = mkUserLocal spec_occ spec_uniq (mkPiTypes spec_lam_args body_ty) fn_loc
836 rule_rhs = mkVarApps (Var spec_id) spec_call_args
837 rule = mkLocalRule rule_name specConstrActivation fn_name bndrs pats rule_rhs
839 returnUs (rule, (spec_id, spec_rhs))
841 -- In which phase should the specialise-constructor rules be active?
842 -- Originally I made them always-active, but Manuel found that
843 -- this defeated some clever user-written rules. So Plan B
844 -- is to make them active only in Phase 0; after all, currently,
845 -- the specConstr transformation is only run after the simplifier
846 -- has reached Phase 0. In general one would want it to be
847 -- flag-controllable, but for now I'm leaving it baked in
849 specConstrActivation :: Activation
850 specConstrActivation = ActiveAfter 0 -- Baked in; see comments above
853 %************************************************************************
855 \subsection{Argument analysis}
857 %************************************************************************
859 This code deals with analysing call-site arguments to see whether
860 they are constructor applications.
862 ---------------------
863 good_arg :: ConstrEnv -> IdEnv ArgOcc -> (CoreBndr, CoreArg) -> Bool
864 -- See Note [Good arguments] above
865 good_arg con_env arg_occs (bndr, arg)
866 = case is_con_app_maybe con_env arg of
867 Just _ -> bndr_usg_ok arg_occs bndr arg
870 bndr_usg_ok :: IdEnv ArgOcc -> Var -> CoreArg -> Bool
871 bndr_usg_ok arg_occs bndr arg
872 = case lookupVarEnv arg_occs bndr of
873 Just ScrutOcc -> True -- Used only by case scrutiny
874 Just Both -> case arg of -- Used by case and elsewhere
875 App _ _ -> True -- so the arg should be an explicit con app
877 other -> False -- Not used, or used wonkily
881 -- argToPat takes an actual argument, and returns an abstracted
882 -- version, consisting of just the "constructor skeleton" of the
883 -- argument, with non-constructor sub-expression replaced by new
884 -- placeholder variables. For example:
885 -- C a (D (f x) (g y)) ==> C p1 (D p2 p3)
887 argToPat :: InScopeEnv -- What's in scope at the fn defn site
888 -> ConstrEnv -- ConstrEnv at the call site
889 -> CoreArg -- A call arg (or component thereof)
891 -> UniqSM (Bool, CoreArg)
892 -- Returns (interesting, pat),
893 -- where pat is the pattern derived from the argument
894 -- intersting=True if the pattern is non-trivial (not a variable or type)
895 -- E.g. x:xs --> (True, x:xs)
896 -- f xs --> (False, w) where w is a fresh wildcard
897 -- (f xs, 'c') --> (True, (w, 'c')) where w is a fresh wildcard
898 -- \x. x+y --> (True, \x. x+y)
899 -- lvl7 --> (True, lvl7) if lvl7 is bound
900 -- somewhere further out
902 argToPat in_scope con_env arg@(Type ty) arg_occ
903 = return (False, arg)
905 argToPat in_scope con_env (Var v) arg_occ -- Don't uniqify existing vars,
906 = return (interesting, Var v) -- so that we can spot when we pass them twice
908 interesting = not (isLocalId v) || v `elemVarEnv` in_scope
910 argToPat in_scope con_env arg arg_occ
914 is_value_lam (Lam v e) -- Spot a value lambda, even if
915 | isId v = True -- it is inside a type lambda
916 | otherwise = is_value_lam e
917 is_value_lam other = False
919 argToPat in_scope con_env arg arg_occ
920 | Just (CV dc args) <- is_con_app_maybe con_env arg
922 ScrutOcc _ -> True -- Used only by case scrutinee
923 BothOcc -> case arg of -- Used by case scrut
924 App {} -> True -- ...and elsewhere...
926 other -> False -- No point; the arg is not decomposed
927 = do { args' <- argsToPats in_scope con_env (args `zip` conArgOccs arg_occ dc)
928 ; return (True, mk_con_app dc (map snd args')) }
930 argToPat in_scope con_env arg arg_occ
931 = do { uniq <- getUniqueUs
932 ; let id = mkSysLocal FSLIT("sc") uniq (exprType arg)
933 ; return (False, Var id) }
935 argsToPats :: InScopeEnv -> ConstrEnv
936 -> [(CoreArg, ArgOcc)]
937 -> UniqSM [(Bool, CoreArg)]
938 argsToPats in_scope con_env args
941 do_one (arg,occ) = argToPat in_scope con_env arg occ
946 is_con_app_maybe :: ConstrEnv -> CoreExpr -> Maybe ConValue
947 is_con_app_maybe env (Var v)
948 = case lookupVarEnv env v of
949 Just stuff -> Just stuff
950 -- You might think we could look in the idUnfolding here
951 -- but that doesn't take account of which branch of a
952 -- case we are in, which is the whole point
954 Nothing | isCheapUnfolding unf
955 -> is_con_app_maybe env (unfoldingTemplate unf)
958 -- However we do want to consult the unfolding
959 -- as well, for let-bound constructors!
963 is_con_app_maybe env (Lit lit)
964 = Just (CV (LitAlt lit) [])
966 is_con_app_maybe env expr
967 = case collectArgs expr of
968 (Var fun, args) | Just con <- isDataConWorkId_maybe fun,
969 args `lengthAtLeast` dataConRepArity con
970 -- Might be > because the arity excludes type args
971 -> Just (CV (DataAlt con) args)
975 mk_con_app :: AltCon -> [CoreArg] -> CoreExpr
976 mk_con_app (LitAlt lit) [] = Lit lit
977 mk_con_app (DataAlt con) args = mkConApp con args