import Type hiding ( substTy, extendTvSubst )
import SimplEnv
import SimplUtils
+import FamInstEnv ( FamInstEnv )
import Id
+import MkId ( mkImpossibleExpr, seqId )
import Var
import IdInfo
import Coercion
import FamInstEnv ( topNormaliseType )
import DataCon ( dataConRepStrictness, dataConUnivTyVars )
import CoreSyn
-import NewDemand ( isStrictDmd )
+import NewDemand ( isStrictDmd, splitStrictSig )
import PprCore ( pprParendExpr, pprCoreExpr )
import CoreUnfold ( mkUnfolding, callSiteInline, CallCtxt(..) )
import CoreUtils
-import Rules ( lookupRule )
-import BasicTypes ( isMarkedStrict )
-import CostCentre ( currentCCS )
+import CoreArity ( exprArity )
+import Rules ( lookupRule, getRules )
+import BasicTypes ( isMarkedStrict, Arity )
+import CostCentre ( currentCCS, pushCCisNop )
import TysPrim ( realWorldStatePrimTy )
import PrelInfo ( realWorldPrimId )
import BasicTypes ( TopLevelFlag(..), isTopLevel,
; env1 <- go (zapFloats env_with_info) triples
; return (env0 `addRecFloats` env1) }
-- addFloats adds the floats from env1,
- -- *and* updates env0 with the in-scope set from env1
+ -- _and_ updates env0 with the in-scope set from env1
where
add_rules :: SimplEnv -> (InBndr,InExpr) -> (SimplEnv, (InBndr, OutBndr, InExpr))
-- Add the (substituted) rules to the binder
simplLazyBind env top_lvl is_rec bndr bndr1 rhs rhs_se
= do { let rhs_env = rhs_se `setInScope` env
- (tvs, body) = collectTyBinders rhs
+ (tvs, body) = case collectTyBinders rhs of
+ (tvs, body) | not_lam body -> (tvs,body)
+ | otherwise -> ([], rhs)
+ not_lam (Lam _ _) = False
+ not_lam _ = True
+ -- Do not do the "abstract tyyvar" thing if there's
+ -- a lambda inside, becuase it defeats eta-reduction
+ -- f = /\a. \x. g a x
+ -- should eta-reduce
+
; (body_env, tvs') <- simplBinders rhs_env tvs
- -- See Note [Floating and type abstraction]
- -- in SimplUtils
+ -- See Note [Floating and type abstraction] in SimplUtils
- -- Simplify the RHS; note the mkRhsStop, which tells
- -- the simplifier that this is the RHS of a let.
- ; let rhs_cont = mkRhsStop (applyTys (idType bndr1) (mkTyVarTys tvs'))
- ; (body_env1, body1) <- simplExprF body_env body rhs_cont
+ -- Simplify the RHS
+ ; (body_env1, body1) <- simplExprF body_env body mkBoringStop
-- ANF-ise a constructor or PAP rhs
; (body_env2, body2) <- prepareRhs body_env1 body1
; (env', rhs')
<- if not (doFloatFromRhs top_lvl is_rec False body2 body_env2)
then -- No floating, just wrap up!
- do { rhs' <- mkLam tvs' (wrapFloats body_env2 body2)
+ do { rhs' <- mkLam env tvs' (wrapFloats body_env2 body2)
; return (env, rhs') }
else if null tvs then -- Simple floating
else -- Do type-abstraction first
do { tick LetFloatFromLet
; (poly_binds, body3) <- abstractFloats tvs' body_env2 body2
- ; rhs' <- mkLam tvs' body3
- ; return (extendFloats env poly_binds, rhs') }
+ ; rhs' <- mkLam env tvs' body3
+ ; let env' = foldl (addPolyBind top_lvl) env poly_binds
+ ; return (env', rhs') }
; completeBind env' top_lvl bndr bndr1 rhs' }
\end{code}
-> SimplM SimplEnv
simplNonRecX env bndr new_rhs
+ | isDeadBinder bndr -- Not uncommon; e.g. case (a,b) of b { (p,q) -> p }
+ = return env -- Here b is dead, and we avoid creating
+ | otherwise -- the binding b = (a,b)
= do { (env', bndr') <- simplBinder env bndr
- ; completeNonRecX env' NotTopLevel NonRecursive
- (isStrictId bndr) bndr bndr' new_rhs }
+ ; completeNonRecX env' (isStrictId bndr) bndr bndr' new_rhs }
completeNonRecX :: SimplEnv
- -> TopLevelFlag -> RecFlag -> Bool
+ -> Bool
-> InId -- Old binder
-> OutId -- New binder
-> OutExpr -- Simplified RHS
-> SimplM SimplEnv
-completeNonRecX env top_lvl is_rec is_strict old_bndr new_bndr new_rhs
+completeNonRecX env is_strict old_bndr new_bndr new_rhs
= do { (env1, rhs1) <- prepareRhs (zapFloats env) new_rhs
; (env2, rhs2) <-
- if doFloatFromRhs top_lvl is_rec is_strict rhs1 env1
+ if doFloatFromRhs NotTopLevel NonRecursive is_strict rhs1 env1
then do { tick LetFloatFromLet
; return (addFloats env env1, rhs1) } -- Add the floats to the main env
else return (env, wrapFloats env1 rhs1) -- Wrap the floats around the RHS
where
is_val = n_val_args > 0 -- There is at least one arg
-- ...and the fun a constructor or PAP
- && (isDataConWorkId fun || n_val_args < idArity fun)
+ && (isConLikeId fun || n_val_args < idArity fun)
go _ env other
= return (False, env, other)
\end{code}
| exprIsTrivial expr
= return (env, expr)
| otherwise -- See Note [Take care] below
- = do { var <- newId FSLIT("a") (exprType expr)
- ; env' <- completeNonRecX env NotTopLevel NonRecursive
- False var var expr
- ; return (env', substExpr env' (Var var)) }
+ = do { var <- newId (fsLit "a") (exprType expr)
+ ; env' <- completeNonRecX env False var var expr
+-- pprTrace "makeTrivial" (vcat [ppr var <+> ppr (exprArity (substExpr env' (Var var)))
+-- , ppr expr
+-- , ppr (substExpr env' (Var var))
+-- , ppr (idArity (fromJust (lookupInScope (seInScope env') var))) ]) $
+ ; return (env', substExpr env' (Var var)) }
+ -- The substitution is needed becase we're constructing a new binding
+ -- a = rhs
+ -- And if rhs is of form (rhs1 |> co), then we might get
+ -- a1 = rhs1
+ -- a = a1 |> co
+ -- and now a's RHS is trivial and can be substituted out, and that
+ -- is what completeNonRecX will do
\end{code}
-- Use the substitution to make quite, quite sure that the
-- substitution will happen, since we are going to discard the binding
- | otherwise
- = let
+ | otherwise
+ = return (addNonRecWithUnf env new_bndr new_rhs unfolding wkr)
+ where
+ unfolding | omit_unfolding = NoUnfolding
+ | otherwise = mkUnfolding (isTopLevel top_lvl) new_rhs
+ old_info = idInfo old_bndr
+ occ_info = occInfo old_info
+ wkr = substWorker env (workerInfo old_info)
+ omit_unfolding = isNonRuleLoopBreaker occ_info
+ -- or not (activeInline env old_bndr)
+ -- Do *not* trim the unfolding in SimplGently, else
+ -- the specialiser can't see it!
+
+-----------------
+addPolyBind :: TopLevelFlag -> SimplEnv -> OutBind -> SimplEnv
+-- Add a new binding to the environment, complete with its unfolding
+-- but *do not* do postInlineUnconditionally, because we have already
+-- processed some of the scope of the binding
+-- We still want the unfolding though. Consider
+-- let
+-- x = /\a. let y = ... in Just y
+-- in body
+-- Then we float the y-binding out (via abstractFloats and addPolyBind)
+-- but 'x' may well then be inlined in 'body' in which case we'd like the
+-- opportunity to inline 'y' too.
+
+addPolyBind top_lvl env (NonRec poly_id rhs)
+ = addNonRecWithUnf env poly_id rhs unfolding NoWorker
+ where
+ unfolding | not (activeInline env poly_id) = NoUnfolding
+ | otherwise = mkUnfolding (isTopLevel top_lvl) rhs
+ -- addNonRecWithInfo adds the new binding in the
+ -- proper way (ie complete with unfolding etc),
+ -- and extends the in-scope set
+
+addPolyBind _ env bind@(Rec _) = extendFloats env bind
+ -- Hack: letrecs are more awkward, so we extend "by steam"
+ -- without adding unfoldings etc. At worst this leads to
+ -- more simplifier iterations
+
+-----------------
+addNonRecWithUnf :: SimplEnv
+ -> OutId -> OutExpr -- New binder and RHS
+ -> Unfolding -> WorkerInfo -- and unfolding
+ -> SimplEnv
+-- Add suitable IdInfo to the Id, add the binding to the floats, and extend the in-scope set
+addNonRecWithUnf env new_bndr rhs unfolding wkr
+ = ASSERT( isId new_bndr )
+ WARN( new_arity < old_arity || new_arity < dmd_arity,
+ (ptext (sLit "Arity decrease:") <+> ppr final_id <+> ppr old_arity
+ <+> ppr new_arity <+> ppr dmd_arity) $$ ppr rhs )
+ -- Note [Arity decrease]
+ final_id `seq` -- This seq forces the Id, and hence its IdInfo,
+ -- and hence any inner substitutions
+ addNonRec env final_id rhs
+ -- The addNonRec adds it to the in-scope set too
+ where
+ dmd_arity = length $ fst $ splitStrictSig $ idNewStrictness new_bndr
+ old_arity = idArity new_bndr
+
-- Arity info
- new_bndr_info = idInfo new_bndr `setArityInfo` exprArity new_rhs
+ new_arity = exprArity rhs
+ new_bndr_info = idInfo new_bndr `setArityInfo` new_arity
-- Unfolding info
-- Add the unfolding *only* for non-loop-breakers
-- (for example) be no longer strictly demanded.
-- The solution here is a bit ad hoc...
info_w_unf = new_bndr_info `setUnfoldingInfo` unfolding
- `setWorkerInfo` worker_info
+ `setWorkerInfo` wkr
- final_info | loop_breaker = new_bndr_info
- | isEvaldUnfolding unfolding = zapDemandInfo info_w_unf `orElse` info_w_unf
+ final_info | isEvaldUnfolding unfolding = zapDemandInfo info_w_unf `orElse` info_w_unf
| otherwise = info_w_unf
-
+
final_id = new_bndr `setIdInfo` final_info
- in
- -- These seqs forces the Id, and hence its IdInfo,
- -- and hence any inner substitutions
- final_id `seq`
- -- pprTrace "Binding" (ppr final_id <+> ppr unfolding) $
- return (addNonRec env final_id new_rhs)
- where
- unfolding = mkUnfolding (isTopLevel top_lvl) new_rhs
- worker_info = substWorker env (workerInfo old_info)
- loop_breaker = isNonRuleLoopBreaker occ_info
- old_info = idInfo old_bndr
- occ_info = occInfo old_info
\end{code}
+Note [Arity decrease]
+~~~~~~~~~~~~~~~~~~~~~
+Generally speaking the arity of a binding should not decrease. But it *can*
+legitimately happen becuase of RULES. Eg
+ f = g Int
+where g has arity 2, will have arity 2. But if there's a rewrite rule
+ g Int --> h
+where h has arity 1, then f's arity will decrease. Here's a real-life example,
+which is in the output of Specialise:
+
+ Rec {
+ $dm {Arity 2} = \d.\x. op d
+ {-# RULES forall d. $dm Int d = $s$dm #-}
+
+ dInt = MkD .... opInt ...
+ opInt {Arity 1} = $dm dInt
+
+ $s$dm {Arity 0} = \x. op dInt }
+
+Here opInt has arity 1; but when we apply the rule its arity drops to 0.
+That's why Specialise goes to a little trouble to pin the right arity
+on specialised functions too.
%************************************************************************
\begin{code}
simplExpr :: SimplEnv -> CoreExpr -> SimplM CoreExpr
-simplExpr env expr = simplExprC env expr (mkBoringStop expr_ty')
- where
- expr_ty' = substTy env (exprType expr)
- -- The type in the Stop continuation, expr_ty', is usually not used
- -- It's only needed when discarding continuations after finding
- -- a function that returns bottom.
- -- Hence the lazy substitution
-
+simplExpr env expr = simplExprC env expr mkBoringStop
simplExprC :: SimplEnv -> CoreExpr -> SimplCont -> SimplM CoreExpr
-- Simplify an expression, given a continuation
do { ty' <- simplType env ty
; rebuild env (Type ty') cont }
-simplExprF' env (Case scrut bndr case_ty alts) cont
+simplExprF' env (Case scrut bndr _ alts) cont
| not (switchIsOn (getSwitchChecker env) NoCaseOfCase)
= -- Simplify the scrutinee with a Select continuation
simplExprF env scrut (Select NoDup bndr alts env cont)
do { case_expr' <- simplExprC env scrut case_cont
; rebuild env case_expr' cont }
where
- case_cont = Select NoDup bndr alts env (mkBoringStop case_ty')
- case_ty' = substTy env case_ty -- c.f. defn of simplExpr
+ case_cont = Select NoDup bndr alts env mkBoringStop
simplExprF' env (Let (Rec pairs) body) cont
= do { env' <- simplRecBndrs env (map fst pairs)
seqType new_ty `seq` return new_ty
where
new_ty = substTy env ty
+
+---------------------------------
+simplCoercion :: SimplEnv -> InType -> SimplM OutType
+simplCoercion env co
+ = do { co' <- simplType env co
+ ; return (optCoercion co') }
\end{code}
Stop {} -> return (env, expr)
CoerceIt co cont -> rebuild env (mkCoerce co expr) cont
Select _ bndr alts se cont -> rebuildCase (se `setFloats` env) expr bndr alts cont
- StrictArg fun ty _ info cont -> rebuildCall env (fun `App` expr) (funResultTy ty) info cont
+ StrictArg fun _ info cont -> rebuildCall env (fun `App` expr) info cont
StrictBind b bs body se cont -> do { env' <- simplNonRecX (se `setFloats` env) b expr
; simplLam env' bs body cont }
ApplyTo _ arg se cont -> do { arg' <- simplExpr (se `setInScope` env) arg
simplCast :: SimplEnv -> InExpr -> Coercion -> SimplCont
-> SimplM (SimplEnv, OutExpr)
simplCast env body co0 cont0
- = do { co1 <- simplType env co0
+ = do { co1 <- simplCoercion env co0
; simplExprF env body (addCoerce co1 cont0) }
where
addCoerce co cont = add_coerce co (coercionKind co) cont
add_coerce co1 (s1, _k2) (CoerceIt co2 cont)
| (_l1, t1) <- coercionKind co2
- -- coerce T1 S1 (coerce S1 K1 e)
+ -- e |> (g1 :: S1~L) |> (g2 :: L~T1)
-- ==>
- -- e, if T1=K1
- -- coerce T1 K1 e, otherwise
+ -- e, if S1=T1
+ -- e |> (g1 . g2 :: S1~T1) otherwise
--
-- For example, in the initial form of a worker
-- we may find (coerce T (coerce S (\x.e))) y
| otherwise = CoerceIt (mkTransCoercion co1 co2) cont
add_coerce co (s1s2, _t1t2) (ApplyTo dup (Type arg_ty) arg_se cont)
- -- (f `cast` g) ty ---> (f ty) `cast` (g @ ty)
+ -- (f |> g) ty ---> (f ty) |> (g @ ty)
-- This implements the PushT rule from the paper
| Just (tyvar,_) <- splitForAllTy_maybe s1s2
, not (isCoVar tyvar)
= ApplyTo dup (Type ty') (zapSubstEnv env) (addCoerce (mkInstCoercion co ty') cont)
where
- ty' = substTy arg_se arg_ty
+ ty' = substTy (arg_se `setInScope` env) arg_ty
-- ToDo: the PushC rule is not implemented at all
add_coerce co (s1s2, _t1t2) (ApplyTo dup arg arg_se cont)
| not (isTypeArg arg) -- This implements the Push rule from the paper
, isFunTy s1s2 -- t1t2 must be a function type, becuase it's applied
- -- co : s1s2 :=: t1t2
- -- (coerce (T1->T2) (S1->S2) F) E
+ -- (e |> (g :: s1s2 ~ t1->t2)) f
-- ===>
- -- coerce T2 S2 (F (coerce S1 T1 E))
+ -- (e (f |> (arg g :: t1~s1))
+ -- |> (res g :: s2->t2)
--
- -- t1t2 must be a function type, T1->T2, because it's applied
+ -- t1t2 must be a function type, t1->t2, because it's applied
-- to something but s1s2 might conceivably not be
--
-- When we build the ApplyTo we can't mix the out-types
-- Example of use: Trac #995
= ApplyTo dup new_arg (zapSubstEnv env) (addCoerce co2 cont)
where
- -- we split coercion t1->t2 :=: s1->s2 into t1 :=: s1 and
- -- t2 :=: s2 with left and right on the curried form:
- -- (->) t1 t2 :=: (->) s1 s2
+ -- we split coercion t1->t2 ~ s1->s2 into t1 ~ s1 and
+ -- t2 ~ s2 with left and right on the curried form:
+ -- (->) t1 t2 ~ (->) s1 s2
[co1, co2] = decomposeCo 2 co
new_arg = mkCoerce (mkSymCoercion co1) arg'
- arg' = substExpr arg_se arg
+ arg' = substExpr (arg_se `setInScope` env) arg
add_coerce co _ cont = CoerceIt co cont
\end{code}
simplLam env [] body cont = simplExprF env body cont
- -- Type-beta reduction
-simplLam env (bndr:bndrs) body (ApplyTo _ (Type ty_arg) arg_se cont)
- = ASSERT( isTyVar bndr )
- do { tick (BetaReduction bndr)
- ; ty_arg' <- simplType (arg_se `setInScope` env) ty_arg
- ; simplLam (extendTvSubst env bndr ty_arg') bndrs body cont }
-
- -- Ordinary beta reduction
+ -- Beta reduction
simplLam env (bndr:bndrs) body (ApplyTo _ arg arg_se cont)
= do { tick (BetaReduction bndr)
; simplNonRecE env bndr (arg, arg_se) (bndrs, body) cont }
simplLam env bndrs body cont
= do { (env', bndrs') <- simplLamBndrs env bndrs
; body' <- simplExpr env' body
- ; new_lam <- mkLam bndrs' body'
+ ; new_lam <- mkLam env' bndrs' body'
; rebuild env' new_lam cont }
------------------
simplNonRecE :: SimplEnv
-> InId -- The binder
-> (InExpr, SimplEnv) -- Rhs of binding (or arg of lambda)
- -> ([InId], InExpr) -- Body of the let/lambda
+ -> ([InBndr], InExpr) -- Body of the let/lambda
-- \xs.e
-> SimplCont
-> SimplM (SimplEnv, OutExpr)
-- Why? Because of the binder-occ-info-zapping done before
-- the call to simplLam in simplExprF (Lam ...)
+ -- First deal with type applications and type lets
+ -- (/\a. e) (Type ty) and (let a = Type ty in e)
+simplNonRecE env bndr (Type ty_arg, rhs_se) (bndrs, body) cont
+ = ASSERT( isTyVar bndr )
+ do { ty_arg' <- simplType (rhs_se `setInScope` env) ty_arg
+ ; simplLam (extendTvSubst env bndr ty_arg') bndrs body cont }
+
simplNonRecE env bndr (rhs, rhs_se) (bndrs, body) cont
| preInlineUnconditionally env NotTopLevel bndr rhs
= do { tick (PreInlineUnconditionally bndr)
(StrictBind bndr bndrs body env cont) }
| otherwise
- = do { (env1, bndr1) <- simplNonRecBndr env bndr
+ = ASSERT( not (isTyVar bndr) )
+ do { (env1, bndr1) <- simplNonRecBndr env bndr
; let (env2, bndr2) = addBndrRules env1 bndr bndr1
; env3 <- simplLazyBind env2 NotTopLevel NonRecursive bndr bndr2 rhs rhs_se
; simplLam env3 bndrs body cont }
simplNote :: SimplEnv -> Note -> CoreExpr -> SimplCont
-> SimplM (SimplEnv, OutExpr)
simplNote env (SCC cc) e cont
+ | pushCCisNop cc (getEnclosingCC env) -- scc "f" (...(scc "f" e)...)
+ = simplExprF env e cont -- ==> scc "f" (...e...)
+ | otherwise
= do { e' <- simplExpr (setEnclosingCC env currentCCS) e
; rebuild env (mkSCC cc e') cont }
completeCall :: SimplEnv -> Id -> SimplCont -> SimplM (SimplEnv, OutExpr)
completeCall env var cont
- = do { dflags <- getDOptsSmpl
- ; let (args,call_cont) = contArgs cont
+ = do { let (args,call_cont) = contArgs cont
-- The args are OutExprs, obtained by *lazily* substituting
-- in the args found in cont. These args are only examined
-- to limited depth (unless a rule fires). But we must do
-- We used to use the black-listing mechanism to ensure that inlining of
-- the wrapper didn't occur for things that have specialisations till a
-- later phase, so but now we just try RULES first
- --
- -- Note [Rules for recursive functions]
- -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- -- You might think that we shouldn't apply rules for a loop breaker:
- -- doing so might give rise to an infinite loop, because a RULE is
- -- rather like an extra equation for the function:
- -- RULE: f (g x) y = x+y
- -- Eqn: f a y = a-y
- --
- -- But it's too drastic to disable rules for loop breakers.
- -- Even the foldr/build rule would be disabled, because foldr
- -- is recursive, and hence a loop breaker:
- -- foldr k z (build g) = g k z
- -- So it's up to the programmer: rules can cause divergence
- ; rules <- getRules
- ; let in_scope = getInScope env
- maybe_rule = case activeRule dflags env of
- Nothing -> Nothing -- No rules apply
- Just act_fn -> lookupRule act_fn in_scope
- rules var args
- ; case maybe_rule of {
- Just (rule, rule_rhs) -> do
- tick (RuleFired (ru_name rule))
- (if dopt Opt_D_dump_rule_firings dflags then
- pprTrace "Rule fired" (vcat [
- text "Rule:" <+> ftext (ru_name rule),
- text "Before:" <+> ppr var <+> sep (map pprParendExpr args),
- text "After: " <+> pprCoreExpr rule_rhs,
- text "Cont: " <+> ppr call_cont])
- else
- id) $
- simplExprF env rule_rhs (dropArgs (ruleArity rule) cont)
+ --
+ -- See also Note [Rules for recursive functions]
+ ; mb_rule <- tryRules env var args call_cont
+ ; case mb_rule of {
+ Just (n_args, rule_rhs) -> simplExprF env rule_rhs (dropArgs n_args cont) ;
-- The ruleArity says how many args the rule consumed
+ ; Nothing -> do -- No rules
- ; Nothing -> do -- No rules
------------- Next try inlining ----------------
- { let arg_infos = [interestingArg arg | arg <- args, isValArg arg]
+ { dflags <- getDOptsSmpl
+ ; let arg_infos = [interestingArg arg | arg <- args, isValArg arg]
n_val_args = length arg_infos
interesting_cont = interestingCallContext call_cont
active_inline = activeInline env var
Just unfolding -- There is an inlining!
-> do { tick (UnfoldingDone var)
; (if dopt Opt_D_dump_inlinings dflags then
- pprTrace ("Inlining done" ++ showSDoc (ppr var)) (vcat [
+ pprTrace ("Inlining done: " ++ showSDoc (ppr var)) (vcat [
text "Before:" <+> ppr var <+> sep (map pprParendExpr args),
text "Inlined fn: " <+> nest 2 (ppr unfolding),
text "Cont: " <+> ppr call_cont])
------------- No inlining! ----------------
-- Next, look for rules or specialisations that match
--
- rebuildCall env (Var var) (idType var)
+ rebuildCall env (Var var)
(mkArgInfo var n_val_args call_cont) cont
}}}}
rebuildCall :: SimplEnv
- -> OutExpr -> OutType -- Function and its type
+ -> OutExpr -- Function
-> ArgInfo
-> SimplCont
-> SimplM (SimplEnv, OutExpr)
-rebuildCall env fun fun_ty (ArgInfo { ai_strs = [] }) cont
+rebuildCall env fun (ArgInfo { ai_strs = [] }) cont
-- When we run out of strictness args, it means
-- that the call is definitely bottom; see SimplUtils.mkArgInfo
-- Then we want to discard the entire strict continuation. E.g.
| not (contIsTrivial cont) -- Only do this if there is a non-trivial
= return (env, mk_coerce fun) -- contination to discard, else we do it
where -- again and again!
- cont_ty = contResultType cont
+ fun_ty = exprType fun
+ cont_ty = contResultType env fun_ty cont
co = mkUnsafeCoercion fun_ty cont_ty
mk_coerce expr | cont_ty `coreEqType` fun_ty = expr
| otherwise = mkCoerce co expr
-rebuildCall env fun fun_ty info (ApplyTo _ (Type arg_ty) se cont)
+rebuildCall env fun info (ApplyTo _ (Type arg_ty) se cont)
= do { ty' <- simplType (se `setInScope` env) arg_ty
- ; rebuildCall env (fun `App` Type ty') (applyTy fun_ty ty') info cont }
+ ; rebuildCall env (fun `App` Type ty') info cont }
-rebuildCall env fun fun_ty
+rebuildCall env fun
(ArgInfo { ai_rules = has_rules, ai_strs = str:strs, ai_discs = disc:discs })
(ApplyTo _ arg arg_se cont)
- | str || isStrictType arg_ty -- Strict argument
+ | str -- Strict argument
= -- pprTrace "Strict Arg" (ppr arg $$ ppr (seIdSubst env) $$ ppr (seInScope env)) $
simplExprF (arg_se `setFloats` env) arg
- (StrictArg fun fun_ty cci arg_info' cont)
+ (StrictArg fun cci arg_info' cont)
-- Note [Shadowing]
| otherwise -- Lazy argument
-- have to be very careful about bogus strictness through
-- floating a demanded let.
= do { arg' <- simplExprC (arg_se `setInScope` env) arg
- (mkLazyArgStop arg_ty cci)
- ; rebuildCall env (fun `App` arg') res_ty arg_info' cont }
+ (mkLazyArgStop cci)
+ ; rebuildCall env (fun `App` arg') arg_info' cont }
where
- (arg_ty, res_ty) = splitFunTy fun_ty
arg_info' = ArgInfo { ai_rules = has_rules, ai_strs = strs, ai_discs = discs }
cci | has_rules || disc > 0 = ArgCtxt has_rules disc -- Be keener here
| otherwise = BoringCtxt -- Nothing interesting
-rebuildCall env fun _ _ cont
+rebuildCall env fun _ cont
= rebuild env fun cont
\end{code}
discard the entire application and replace it with (error "foo"). Getting
all this at once is TOO HARD!
+
+%************************************************************************
+%* *
+ Rewrite rules
+%* *
+%************************************************************************
+
+\begin{code}
+tryRules :: SimplEnv -> Id -> [OutExpr] -> SimplCont
+ -> SimplM (Maybe (Arity, CoreExpr)) -- The arity is the number of
+ -- args consumed by the rule
+tryRules env fn args call_cont
+ = do { dflags <- getDOptsSmpl
+ ; rule_base <- getSimplRules
+ ; let in_scope = getInScope env
+ rules = getRules rule_base fn
+ maybe_rule = case activeRule dflags env of
+ Nothing -> Nothing -- No rules apply
+ Just act_fn -> lookupRule act_fn in_scope
+ fn args rules
+ ; case (rules, maybe_rule) of {
+ ([], _) -> return Nothing ;
+ (_, Nothing) -> return Nothing ;
+ (_, Just (rule, rule_rhs)) -> do
+
+ { tick (RuleFired (ru_name rule))
+ ; (if dopt Opt_D_dump_rule_firings dflags then
+ pprTrace "Rule fired" (vcat [
+ text "Rule:" <+> ftext (ru_name rule),
+ text "Before:" <+> ppr fn <+> sep (map pprParendExpr args),
+ text "After: " <+> pprCoreExpr rule_rhs,
+ text "Cont: " <+> ppr call_cont])
+ else
+ id) $
+ return (Just (ruleArity rule, rule_rhs)) }}}
+\end{code}
+
+Note [Rules for recursive functions]
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+You might think that we shouldn't apply rules for a loop breaker:
+doing so might give rise to an infinite loop, because a RULE is
+rather like an extra equation for the function:
+ RULE: f (g x) y = x+y
+ Eqn: f a y = a-y
+
+But it's too drastic to disable rules for loop breakers.
+Even the foldr/build rule would be disabled, because foldr
+is recursive, and hence a loop breaker:
+ foldr k z (build g) = g k z
+So it's up to the programmer: rules can cause divergence
+
+
%************************************************************************
%* *
Rebuilding a cse expression
%* *
%************************************************************************
-Blob of helper functions for the "case-of-something-else" situation.
+Note [Case elimination]
+~~~~~~~~~~~~~~~~~~~~~~~
+The case-elimination transformation discards redundant case expressions.
+Start with a simple situation:
+
+ case x# of ===> e[x#/y#]
+ y# -> e
+
+(when x#, y# are of primitive type, of course). We can't (in general)
+do this for algebraic cases, because we might turn bottom into
+non-bottom!
+
+The code in SimplUtils.prepareAlts has the effect of generalise this
+idea to look for a case where we're scrutinising a variable, and we
+know that only the default case can match. For example:
+
+ case x of
+ 0# -> ...
+ DEFAULT -> ...(case x of
+ 0# -> ...
+ DEFAULT -> ...) ...
+
+Here the inner case is first trimmed to have only one alternative, the
+DEFAULT, after which it's an instance of the previous case. This
+really only shows up in eliminating error-checking code.
+
+We also make sure that we deal with this very common case:
+
+ case e of
+ x -> ...x...
+
+Here we are using the case as a strict let; if x is used only once
+then we want to inline it. We have to be careful that this doesn't
+make the program terminate when it would have diverged before, so we
+check that
+ - e is already evaluated (it may so if e is a variable)
+ - x is used strictly, or
+
+Lastly, the code in SimplUtils.mkCase combines identical RHSs. So
+
+ case e of ===> case e of DEFAULT -> r
+ True -> r
+ False -> r
+
+Now again the case may be elminated by the CaseElim transformation.
+
+
+Further notes about case elimination
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+Consider: test :: Integer -> IO ()
+ test = print
+
+Turns out that this compiles to:
+ Print.test
+ = \ eta :: Integer
+ eta1 :: State# RealWorld ->
+ case PrelNum.< eta PrelNum.zeroInteger of wild { __DEFAULT ->
+ case hPutStr stdout
+ (PrelNum.jtos eta ($w[] @ Char))
+ eta1
+ of wild1 { (# new_s, a4 #) -> PrelIO.lvl23 new_s }}
+
+Notice the strange '<' which has no effect at all. This is a funny one.
+It started like this:
+
+f x y = if x < 0 then jtos x
+ else if y==0 then "" else jtos x
+
+At a particular call site we have (f v 1). So we inline to get
+
+ if v < 0 then jtos x
+ else if 1==0 then "" else jtos x
+
+Now simplify the 1==0 conditional:
+
+ if v<0 then jtos v else jtos v
+
+Now common-up the two branches of the case:
+
+ case (v<0) of DEFAULT -> jtos v
+
+Why don't we drop the case? Because it's strict in v. It's technically
+wrong to drop even unnecessary evaluations, and in practice they
+may be a result of 'seq' so we *definitely* don't want to drop those.
+I don't really know how to improve this situation.
\begin{code}
---------------------------------------------------------
-- Eliminate the case if possible
-rebuildCase :: SimplEnv
- -> OutExpr -- Scrutinee
- -> InId -- Case binder
- -> [InAlt] -- Alternatives (inceasing order)
- -> SimplCont
- -> SimplM (SimplEnv, OutExpr)
+rebuildCase, reallyRebuildCase
+ :: SimplEnv
+ -> OutExpr -- Scrutinee
+ -> InId -- Case binder
+ -> [InAlt] -- Alternatives (inceasing order)
+ -> SimplCont
+ -> SimplM (SimplEnv, OutExpr)
--------------------------------------------------
-- 1. Eliminate the case if there's a known constructor
rebuildCase env scrut case_bndr [(_, bndrs, rhs)] cont
-- See if we can get rid of the case altogether
- -- See the extensive notes on case-elimination above
+ -- See Note [Case eliminiation]
-- mkCase made sure that if all the alternatives are equal,
-- then there is now only one (DEFAULT) rhs
| all isDeadBinder bndrs -- bndrs are [InId]
-- exprOkForSpeculation was intended for.
var_demanded_later _ = False
+rebuildCase env scrut case_bndr alts@[(_, bndrs, rhs)] cont
+ | all isDeadBinder (case_bndr : bndrs) -- So this is just 'seq'
+ = -- For this case, see Note [Rules for seq] in MkId
+ do { let rhs' = substExpr env rhs
+ out_args = [Type (substTy env (idType case_bndr)),
+ Type (exprType rhs'), scrut, rhs']
+ -- Lazily evaluated, so we don't do most of this
+ ; mb_rule <- tryRules env seqId out_args cont
+ ; case mb_rule of
+ Just (n_args, res) -> simplExprF (zapSubstEnv env)
+ (mkApps res (drop n_args out_args))
+ cont
+ Nothing -> reallyRebuildCase env scrut case_bndr alts cont }
+
+rebuildCase env scrut case_bndr alts cont
+ = reallyRebuildCase env scrut case_bndr alts cont
--------------------------------------------------
-- 3. Catch-all case
--------------------------------------------------
-rebuildCase env scrut case_bndr alts cont
+reallyRebuildCase env scrut case_bndr alts cont
= do { -- Prepare the continuation;
-- The new subst_env is in place
(env', dup_cont, nodup_cont) <- prepareCaseCont env alts cont
-- Simplify the alternatives
; (scrut', case_bndr', alts') <- simplAlts env' scrut case_bndr alts dup_cont
- ; let res_ty' = contResultType dup_cont
- ; case_expr <- mkCase scrut' case_bndr' res_ty' alts'
- -- Notice that rebuildDone returns the in-scope set from env', not alt_env
- -- The case binder *not* scope over the whole returned case-expression
- ; rebuild env' case_expr nodup_cont }
+ -- Check for empty alternatives
+ ; if null alts' then missingAlt env case_bndr alts cont
+ else do
+ { case_expr <- mkCase scrut' case_bndr' alts'
+
+ -- Notice that rebuild gets the in-scope set from env, not alt_env
+ -- The case binder *not* scope over the whole returned case-expression
+ ; rebuild env' case_expr nodup_cont } }
\end{code}
simplCaseBinder checks whether the scrutinee is a variable, v. If so,
way, there's a chance that v will now only be used once, and hence
inlined.
-Note [no-case-of-case]
-~~~~~~~~~~~~~~~~~~~~~~
-There is a time we *don't* want to do that, namely when
--fno-case-of-case is on. This happens in the first simplifier pass,
-and enhances full laziness. Here's the bad case:
- f = \ y -> ...(case x of I# v -> ...(case x of ...) ... )
-If we eliminate the inner case, we trap it inside the I# v -> arm,
-which might prevent some full laziness happening. I've seen this
-in action in spectral/cichelli/Prog.hs:
- [(m,n) | m <- [1..max], n <- [1..max]]
-Hence the check for NoCaseOfCase.
-
-Note [Suppressing the case binder-swap]
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-There is another situation when it might make sense to suppress the
-case-expression binde-swap. If we have
-
- case x of w1 { DEFAULT -> case x of w2 { A -> e1; B -> e2 }
- ...other cases .... }
-
-We'll perform the binder-swap for the outer case, giving
-
- case x of w1 { DEFAULT -> case w1 of w2 { A -> e1; B -> e2 }
- ...other cases .... }
-
-But there is no point in doing it for the inner case, because w1 can't
-be inlined anyway. Furthermore, doing the case-swapping involves
-zapping w2's occurrence info (see paragraphs that follow), and that
-forces us to bind w2 when doing case merging. So we get
-
- case x of w1 { A -> let w2 = w1 in e1
- B -> let w2 = w1 in e2
- ...other cases .... }
-
-This is plain silly in the common case where w2 is dead.
-
-Even so, I can't see a good way to implement this idea. I tried
-not doing the binder-swap if the scrutinee was already evaluated
-but that failed big-time:
-
- data T = MkT !Int
-
- case v of w { MkT x ->
- case x of x1 { I# y1 ->
- case x of x2 { I# y2 -> ...
-
-Notice that because MkT is strict, x is marked "evaluated". But to
-eliminate the last case, we must either make sure that x (as well as
-x1) has unfolding MkT y1. THe straightforward thing to do is to do
-the binder-swap. So this whole note is a no-op.
+Historical note: we use to do the "case binder swap" in the Simplifier
+so there were additional complications if the scrutinee was a variable.
+Now the binder-swap stuff is done in the occurrence analyer; see
+OccurAnal Note [Binder swap].
Note [zapOccInfo]
~~~~~~~~~~~~~~~~~
-If we replace the scrutinee, v, by tbe case binder, then we have to nuke
-any occurrence info (eg IAmDead) in the case binder, because the
-case-binder now effectively occurs whenever v does. AND we have to do
-the same for the pattern-bound variables! Example:
-
- (case x of { (a,b) -> a }) (case x of { (p,q) -> q })
-
-Here, b and p are dead. But when we move the argment inside the first
-case RHS, and eliminate the second case, we get
-
- case x of { (a,b) -> a b }
-
-Urk! b is alive! Reason: the scrutinee was a variable, and case elimination
-happened.
-
-Indeed, this can happen anytime the case binder isn't dead:
+If the case binder is not dead, then neither are the pattern bound
+variables:
case <any> of x { (a,b) ->
case x of { (p,q) -> p } }
Here (a,b) both look dead, but come alive after the inner case is eliminated.
after the outer case, and that makes (a,b) alive. At least we do unless
the case binder is guaranteed dead.
-Note [Case of cast]
-~~~~~~~~~~~~~~~~~~~
-Consider case (v `cast` co) of x { I# ->
- ... (case (v `cast` co) of {...}) ...
-We'd like to eliminate the inner case. We can get this neatly by
-arranging that inside the outer case we add the unfolding
- v |-> x `cast` (sym co)
-to v. Then we should inline v at the inner case, cancel the casts, and away we go
-
Note [Improving seq]
~~~~~~~~~~~~~~~~~~~
Consider
(Otherwise, there's a danger that we'll simply drop the 'seq' altogether, before
LiberateCase gets to see it.)
-Note [Case elimination]
-~~~~~~~~~~~~~~~~~~~~~~~
-The case-elimination transformation discards redundant case expressions.
-Start with a simple situation:
-
- case x# of ===> e[x#/y#]
- y# -> e
-
-(when x#, y# are of primitive type, of course). We can't (in general)
-do this for algebraic cases, because we might turn bottom into
-non-bottom!
-
-The code in SimplUtils.prepareAlts has the effect of generalise this
-idea to look for a case where we're scrutinising a variable, and we
-know that only the default case can match. For example:
-
- case x of
- 0# -> ...
- DEFAULT -> ...(case x of
- 0# -> ...
- DEFAULT -> ...) ...
-
-Here the inner case is first trimmed to have only one alternative, the
-DEFAULT, after which it's an instance of the previous case. This
-really only shows up in eliminating error-checking code.
-
-We also make sure that we deal with this very common case:
-
- case e of
- x -> ...x...
-
-Here we are using the case as a strict let; if x is used only once
-then we want to inline it. We have to be careful that this doesn't
-make the program terminate when it would have diverged before, so we
-check that
- - e is already evaluated (it may so if e is a variable)
- - x is used strictly, or
-
-Lastly, the code in SimplUtils.mkCase combines identical RHSs. So
-
- case e of ===> case e of DEFAULT -> r
- True -> r
- False -> r
-
-Now again the case may be elminated by the CaseElim transformation.
-
-
-Further notes about case elimination
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-Consider: test :: Integer -> IO ()
- test = print
-
-Turns out that this compiles to:
- Print.test
- = \ eta :: Integer
- eta1 :: State# RealWorld ->
- case PrelNum.< eta PrelNum.zeroInteger of wild { __DEFAULT ->
- case hPutStr stdout
- (PrelNum.jtos eta ($w[] @ Char))
- eta1
- of wild1 { (# new_s, a4 #) -> PrelIO.lvl23 new_s }}
-
-Notice the strange '<' which has no effect at all. This is a funny one.
-It started like this:
-f x y = if x < 0 then jtos x
- else if y==0 then "" else jtos x
-
-At a particular call site we have (f v 1). So we inline to get
-
- if v < 0 then jtos x
- else if 1==0 then "" else jtos x
-
-Now simplify the 1==0 conditional:
-
- if v<0 then jtos v else jtos v
-
-Now common-up the two branches of the case:
-
- case (v<0) of DEFAULT -> jtos v
-
-Why don't we drop the case? Because it's strict in v. It's technically
-wrong to drop even unnecessary evaluations, and in practice they
-may be a result of 'seq' so we *definitely* don't want to drop those.
-I don't really know how to improve this situation.
\begin{code}
-simplCaseBinder :: SimplEnv -> OutExpr -> OutId -> [InAlt]
- -> SimplM (SimplEnv, OutExpr, OutId)
-simplCaseBinder env0 scrut0 case_bndr0 alts
- = do { (env1, case_bndr1) <- simplBinder env0 case_bndr0
-
- ; fam_envs <- getFamEnvs
- ; (env2, scrut2, case_bndr2) <- improve_seq fam_envs env1 scrut0
- case_bndr0 case_bndr1 alts
- -- Note [Improving seq]
-
- ; let (env3, case_bndr3) = improve_case_bndr env2 scrut2 case_bndr2
- -- Note [Case of cast]
-
- ; return (env3, scrut2, case_bndr3) }
- where
-
- improve_seq fam_envs env scrut case_bndr case_bndr1 [(DEFAULT,_,_)]
- | Just (co, ty2) <- topNormaliseType fam_envs (idType case_bndr1)
- = do { case_bndr2 <- newId FSLIT("nt") ty2
- ; let rhs = DoneEx (Var case_bndr2 `Cast` mkSymCoercion co)
- env2 = extendIdSubst env case_bndr rhs
- ; return (env2, scrut `Cast` co, case_bndr2) }
-
- improve_seq _ env scrut _ case_bndr1 _
- = return (env, scrut, case_bndr1)
-
-
+improveSeq :: (FamInstEnv, FamInstEnv) -> SimplEnv
+ -> OutExpr -> InId -> OutId -> [InAlt]
+ -> SimplM (SimplEnv, OutExpr, OutId)
+-- Note [Improving seq]
+improveSeq fam_envs env scrut case_bndr case_bndr1 [(DEFAULT,_,_)]
+ | Just (co, ty2) <- topNormaliseType fam_envs (idType case_bndr1)
+ = do { case_bndr2 <- newId (fsLit "nt") ty2
+ ; let rhs = DoneEx (Var case_bndr2 `Cast` mkSymCoercion co)
+ env2 = extendIdSubst env case_bndr rhs
+ ; return (env2, scrut `Cast` co, case_bndr2) }
+
+improveSeq _ env scrut _ case_bndr1 _
+ = return (env, scrut, case_bndr1)
+
+{-
improve_case_bndr env scrut case_bndr
- | switchIsOn (getSwitchChecker env) NoCaseOfCase
- -- See Note [no-case-of-case]
- = (env, case_bndr)
+ -- See Note [no-case-of-case]
+ -- | switchIsOn (getSwitchChecker env) NoCaseOfCase
+ -- = (env, case_bndr)
| otherwise -- Failed try; see Note [Suppressing the case binder-swap]
-- not (isEvaldUnfolding (idUnfolding v))
_ -> (env, case_bndr)
where
- case_bndr' = zapOccInfo case_bndr
+ case_bndr' = zapIdOccInfo case_bndr
env1 = modifyInScope env case_bndr case_bndr'
-
-
-zapOccInfo :: InId -> InId -- See Note [zapOccInfo]
-zapOccInfo b = b `setIdOccInfo` NoOccInfo
+-}
\end{code}
simplAlts :: SimplEnv
-> OutExpr
-> InId -- Case binder
- -> [InAlt] -> SimplCont
+ -> [InAlt] -- Non-empty
+ -> SimplCont
-> SimplM (OutExpr, OutId, [OutAlt]) -- Includes the continuation
-- Like simplExpr, this just returns the simplified alternatives;
-- it not return an environment
simplAlts env scrut case_bndr alts cont'
= -- pprTrace "simplAlts" (ppr alts $$ ppr (seIdSubst env)) $
- do { let alt_env = zapFloats env
- ; (alt_env', scrut', case_bndr') <- simplCaseBinder alt_env scrut case_bndr alts
+ do { let env0 = zapFloats env
- ; (imposs_deflt_cons, in_alts) <- prepareAlts alt_env' scrut case_bndr' alts
+ ; (env1, case_bndr1) <- simplBinder env0 case_bndr
+
+ ; fam_envs <- getFamEnvs
+ ; (alt_env', scrut', case_bndr') <- improveSeq fam_envs env1 scrut
+ case_bndr case_bndr1 alts
+
+ ; (imposs_deflt_cons, in_alts) <- prepareAlts alt_env' scrut' case_bndr' alts
; alts' <- mapM (simplAlt alt_env' imposs_deflt_cons case_bndr' cont') in_alts
; return (scrut', case_bndr', alts') }
evald_v = zapped_v `setIdUnfolding` evaldUnfolding
go _ _ = pprPanic "cat_evals" (ppr con $$ ppr vs $$ ppr the_strs)
+ -- See Note [zapOccInfo]
-- zap_occ_info: if the case binder is alive, then we add the unfolding
-- case_bndr = C vs
-- to the envt; so vs are now very much alive
-- case e of t { (a,b) -> ...(case t of (p,q) -> p)... }
-- ==> case e of t { (a,b) -> ...(a)... }
-- Look, Ma, a is alive now.
- zap_occ_info | isDeadBinder case_bndr' = \ident -> ident
- | otherwise = zapOccInfo
+ zap_occ_info = zapCasePatIdOcc case_bndr'
addBinderUnfolding :: SimplEnv -> Id -> CoreExpr -> SimplEnv
addBinderUnfolding env bndr rhs
- = modifyInScope env bndr (bndr `setIdUnfolding` mkUnfolding False rhs)
+ = modifyInScope env (bndr `setIdUnfolding` mkUnfolding False rhs)
addBinderOtherCon :: SimplEnv -> Id -> [AltCon] -> SimplEnv
addBinderOtherCon env bndr cons
- = modifyInScope env bndr (bndr `setIdUnfolding` mkOtherCon cons)
+ = modifyInScope env (bndr `setIdUnfolding` mkOtherCon cons)
+
+zapCasePatIdOcc :: Id -> Id -> Id
+-- Consider case e of b { (a,b) -> ... }
+-- Then if we bind b to (a,b) in "...", and b is not dead,
+-- then we must zap the deadness info on a,b
+zapCasePatIdOcc case_bndr
+ | isDeadBinder case_bndr = \ pat_id -> pat_id
+ | otherwise = \ pat_id -> zapIdOccInfo pat_id
\end{code}
All this should happen in one sweep.
\begin{code}
-knownCon :: SimplEnv -> OutExpr -> AltCon -> [OutExpr]
+knownCon :: SimplEnv -> OutExpr -> AltCon
+ -> [OutExpr] -- Args *including* the universal args
-> InId -> [InAlt] -> SimplCont
-> SimplM (SimplEnv, OutExpr)
knownCon env scrut con args bndr alts cont
= do { tick (KnownBranch bndr)
- ; knownAlt env scrut args bndr (findAlt con alts) cont }
+ ; case findAlt con alts of
+ Nothing -> missingAlt env bndr alts cont
+ Just alt -> knownAlt env scrut args bndr alt cont
+ }
+-------------------
knownAlt :: SimplEnv -> OutExpr -> [OutExpr]
- -> InId -> (AltCon, [CoreBndr], InExpr) -> SimplCont
+ -> InId -> InAlt -> SimplCont
-> SimplM (SimplEnv, OutExpr)
-knownAlt env scrut _ bndr (DEFAULT, bs, rhs) cont
- = ASSERT( null bs )
- do { env' <- simplNonRecX env bndr scrut
- -- This might give rise to a binding with non-atomic args
- -- like x = Node (f x) (g x)
- -- but simplNonRecX will atomic-ify it
- ; simplExprF env' rhs cont }
-
-knownAlt env scrut _ bndr (LitAlt _, bs, rhs) cont
- = ASSERT( null bs )
- do { env' <- simplNonRecX env bndr scrut
- ; simplExprF env' rhs cont }
knownAlt env scrut the_args bndr (DataAlt dc, bs, rhs) cont
- = do { let dead_bndr = isDeadBinder bndr -- bndr is an InId
- n_drop_tys = length (dataConUnivTyVars dc)
- ; env' <- bind_args env dead_bndr bs (drop n_drop_tys the_args)
+ = do { let n_drop_tys = length (dataConUnivTyVars dc)
+ ; env' <- bind_args env bs (drop n_drop_tys the_args)
; let
-- It's useful to bind bndr to scrut, rather than to a fresh
-- binding x = Con arg1 .. argn
-- args are aready OutExprs, but bs are InIds
; env'' <- simplNonRecX env' bndr bndr_rhs
- ; -- pprTrace "knownCon2" (ppr bs $$ ppr rhs $$ ppr (seIdSubst env'')) $
- simplExprF env'' rhs cont }
+ ; simplExprF env'' rhs cont }
where
- -- Ugh!
- bind_args env' _ [] _ = return env'
+ zap_occ = zapCasePatIdOcc bndr -- bndr is an InId
+
+ -- Ugh!
+ bind_args env' [] _ = return env'
- bind_args env' dead_bndr (b:bs') (Type ty : args)
+ bind_args env' (b:bs') (Type ty : args)
= ASSERT( isTyVar b )
- bind_args (extendTvSubst env' b ty) dead_bndr bs' args
+ bind_args (extendTvSubst env' b ty) bs' args
- bind_args env' dead_bndr (b:bs') (arg : args)
+ bind_args env' (b:bs') (arg : args)
= ASSERT( isId b )
- do { let b' = if dead_bndr then b else zapOccInfo b
+ do { let b' = zap_occ b
-- Note that the binder might be "dead", because it doesn't
-- occur in the RHS; and simplNonRecX may therefore discard
-- it via postInlineUnconditionally.
-- Nevertheless we must keep it if the case-binder is alive,
-- because it may be used in the con_app. See Note [zapOccInfo]
; env'' <- simplNonRecX env' b' arg
- ; bind_args env'' dead_bndr bs' args }
+ ; bind_args env'' bs' args }
- bind_args _ _ _ _ =
+ bind_args _ _ _ =
pprPanic "bind_args" $ ppr dc $$ ppr bs $$ ppr the_args $$
text "scrut:" <+> ppr scrut
+
+knownAlt env scrut _ bndr (_, bs, rhs) cont
+ = ASSERT( null bs ) -- Works for LitAlt and DEFAULT
+ do { env' <- simplNonRecX env bndr scrut
+ ; simplExprF env' rhs cont }
+
+
+-------------------
+missingAlt :: SimplEnv -> Id -> [InAlt] -> SimplCont -> SimplM (SimplEnv, OutExpr)
+ -- This isn't strictly an error, although it is unusual.
+ -- It's possible that the simplifer might "see" that
+ -- an inner case has no accessible alternatives before
+ -- it "sees" that the entire branch of an outer case is
+ -- inaccessible. So we simply put an error case here instead.
+missingAlt env case_bndr alts cont
+ = WARN( True, ptext (sLit "missingAlt") <+> ppr case_bndr )
+ return (env, mkImpossibleExpr res_ty)
+ where
+ res_ty = contResultType env (substTy env (coreAltsType alts)) cont
\end{code}
-- continunation)
-- No need to make it duplicatable if there's only one alternative
-prepareCaseCont env [_] cont = return (env, cont, mkBoringStop (contResultType cont))
+prepareCaseCont env [_] cont = return (env, cont, mkBoringStop)
prepareCaseCont env _ cont = mkDupableCont env cont
\end{code}
mkDupableCont env cont
| contIsDupable cont
- = return (env, cont, mkBoringStop (contResultType cont))
+ = return (env, cont, mkBoringStop)
mkDupableCont _ (Stop {}) = panic "mkDupableCont" -- Handled by previous eqn
= do { (env', dup, nodup) <- mkDupableCont env cont
; return (env', CoerceIt ty dup, nodup) }
-mkDupableCont env cont@(StrictBind bndr _ _ se _)
- = return (env, mkBoringStop (substTy se (idType bndr)), cont)
- -- See Note [Duplicating strict continuations]
+mkDupableCont env cont@(StrictBind {})
+ = return (env, mkBoringStop, cont)
+ -- See Note [Duplicating StrictBind]
-mkDupableCont env cont@(StrictArg _ fun_ty _ _ _)
- = return (env, mkBoringStop (funArgTy fun_ty), cont)
- -- See Note [Duplicating strict continuations]
+mkDupableCont env (StrictArg fun cci ai cont)
+ -- See Note [Duplicating StrictArg]
+ = do { (env', dup, nodup) <- mkDupableCont env cont
+ ; (env'', fun') <- mk_dupable_call env' fun
+ ; return (env'', StrictArg fun' cci ai dup, nodup) }
+ where
+ mk_dupable_call env (Var v) = return (env, Var v)
+ mk_dupable_call env (App fun arg) = do { (env', fun') <- mk_dupable_call env fun
+ ; (env'', arg') <- makeTrivial env' arg
+ ; return (env'', fun' `App` arg') }
+ mk_dupable_call _ other = pprPanic "mk_dupable_call" (ppr other)
+ -- The invariant of StrictArg is that the first arg is always an App chain
mkDupableCont env (ApplyTo _ arg se cont)
= -- e.g. [...hole...] (...arg...)
do { (env', dup_cont, nodup_cont) <- mkDupableCont env cont
; arg' <- simplExpr (se `setInScope` env') arg
; (env'', arg'') <- makeTrivial env' arg'
- ; let app_cont = ApplyTo OkToDup arg'' (zapSubstEnv env') dup_cont
+ ; let app_cont = ApplyTo OkToDup arg'' (zapSubstEnv env'') dup_cont
; return (env'', app_cont, nodup_cont) }
-mkDupableCont env cont@(Select _ case_bndr [(_, bs, _rhs)] se _case_cont)
+mkDupableCont env cont@(Select _ case_bndr [(_, bs, _rhs)] _ _)
-- See Note [Single-alternative case]
-- | not (exprIsDupable rhs && contIsDupable case_cont)
-- | not (isDeadBinder case_bndr)
- | all isDeadBinder bs -- InIds
- = return (env, mkBoringStop scrut_ty, cont)
- where
- scrut_ty = substTy se (idType case_bndr)
+ | all isDeadBinder bs -- InIds
+ && not (isUnLiftedType (idType case_bndr))
+ -- Note [Single-alternative-unlifted]
+ = return (env, mkBoringStop, cont)
mkDupableCont env (Select _ case_bndr alts se cont)
= -- e.g. (case [...hole...] of { pi -> ei })
; (env'', alts'') <- mkDupableAlts env' case_bndr' alts'
; return (env'', -- Note [Duplicated env]
- Select OkToDup case_bndr' alts'' (zapSubstEnv env'')
- (mkBoringStop (contResultType dup_cont)),
+ Select OkToDup case_bndr' alts'' (zapSubstEnv env'') mkBoringStop,
nodup_cont) }
; (final_bndrs', final_args) -- Note [Join point abstraction]
<- if (any isId used_bndrs')
then return (used_bndrs', varsToCoreExprs used_bndrs')
- else do { rw_id <- newId FSLIT("w") realWorldStatePrimTy
+ else do { rw_id <- newId (fsLit "w") realWorldStatePrimTy
; return ([rw_id], [Var realWorldPrimId]) }
- ; join_bndr <- newId FSLIT("$j") (mkPiTypes final_bndrs' rhs_ty')
+ ; join_bndr <- newId (fsLit "$j") (mkPiTypes final_bndrs' rhs_ty')
-- Note [Funky mkPiTypes]
; let -- We make the lambdas into one-shot-lambdas. The
join_rhs = mkLams really_final_bndrs rhs'
join_call = mkApps (Var join_bndr) final_args
- ; return (addNonRec env join_bndr join_rhs, (con, bndrs', join_call)) }
+ ; return (addPolyBind NotTopLevel env (NonRec join_bndr join_rhs), (con, bndrs', join_call)) }
-- See Note [Duplicated env]
\end{code}
but zapping it (as we do in mkDupableCont, the Select case) is safe, and
at worst delays the join-point inlining.
-Note [Small alterantive rhs]
+Note [Small alternative rhs]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
It is worth checking for a small RHS because otherwise we
get extra let bindings that may cause an extra iteration of the simplifier to
True -> $j s
(the \v alone is enough to make CPR happy) but I think it's rare
-Note [Duplicating strict continuations]
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-Do *not* duplicate StrictBind and StritArg continuations. We gain
-nothing by propagating them into the expressions, and we do lose a
-lot. Here's an example:
- && (case x of { T -> F; F -> T }) E
+Note [Duplicating StrictArg]
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+The original plan had (where E is a big argument)
+e.g. f E [..hole..]
+ ==> let $j = \a -> f E a
+ in $j [..hole..]
+
+But this is terrible! Here's an example:
+ && E (case x of { T -> F; F -> T })
Now, && is strict so we end up simplifying the case with
an ArgOf continuation. If we let-bind it, we get
-
- let $j = \v -> && v E
+ let $j = \v -> && E v
in simplExpr (case x of { T -> F; F -> T })
(ArgOf (\r -> $j r)
And after simplifying more we get
-
- let $j = \v -> && v E
+ let $j = \v -> && E v
in case x of { T -> $j F; F -> $j T }
Which is a Very Bad Thing
+What we do now is this
+ f E [..hole..]
+ ==> let a = E
+ in f a [..hole..]
+Now if the thing in the hole is a case expression (which is when
+we'll call mkDupableCont), we'll push the function call into the
+branches, which is what we want. Now RULES for f may fire, and
+call-pattern specialisation. Here's an example from Trac #3116
+ go (n+1) (case l of
+ 1 -> bs'
+ _ -> Chunk p fpc (o+1) (l-1) bs')
+If we can push the call for 'go' inside the case, we get
+call-pattern specialisation for 'go', which is *crucial* for
+this program.
+
+Here is the (&&) example:
+ && E (case x of { T -> F; F -> T })
+ ==> let a = E in
+ case x of { T -> && a F; F -> && a T }
+Much better!
+
+Notice that
+ * Arguments to f *after* the strict one are handled by
+ the ApplyTo case of mkDupableCont. Eg
+ f [..hole..] E
+
+ * We can only do the let-binding of E because the function
+ part of a StrictArg continuation is an explicit syntax
+ tree. In earlier versions we represented it as a function
+ (CoreExpr -> CoreEpxr) which we couldn't take apart.
+
+Do *not* duplicate StrictBind and StritArg continuations. We gain
+nothing by propagating them into the expressions, and we do lose a
+lot.
+
+The desire not to duplicate is the entire reason that
+mkDupableCont returns a pair of continuations.
+
+Note [Duplicating StrictBind]
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+Unlike StrictArg, there doesn't seem anything to gain from
+duplicating a StrictBind continuation, so we don't.
+
The desire not to duplicate is the entire reason that
mkDupableCont returns a pair of continuations.
-The original plan had:
-e.g. (...strict-fn...) [...hole...]
- ==>
- let $j = \a -> ...strict-fn...
- in $j [...hole...]
Note [Single-alternative cases]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
When x is inlined into its full context, we find that it was a bad
idea to have pushed the outer case inside the (...) case.
+Note [Single-alternative-unlifted]
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+Here's another single-alternative where we really want to do case-of-case:
+
+data Mk1 = Mk1 Int#
+data Mk1 = Mk2 Int#
+
+M1.f =
+ \r [x_s74 y_s6X]
+ case
+ case y_s6X of tpl_s7m {
+ M1.Mk1 ipv_s70 -> ipv_s70;
+ M1.Mk2 ipv_s72 -> ipv_s72;
+ }
+ of
+ wild_s7c
+ { __DEFAULT ->
+ case
+ case x_s74 of tpl_s7n {
+ M1.Mk1 ipv_s77 -> ipv_s77;
+ M1.Mk2 ipv_s79 -> ipv_s79;
+ }
+ of
+ wild1_s7b
+ { __DEFAULT -> ==# [wild1_s7b wild_s7c];
+ };
+ };
+
+So the outer case is doing *nothing at all*, other than serving as a
+join-point. In this case we really want to do case-of-case and decide
+whether to use a real join point or just duplicate the continuation.
+
+Hence: check whether the case binder's type is unlifted, because then
+the outer case is *not* a seq.