% % (c) The GRASP/AQUA Project, Glasgow University, 1993-1996 % \section[SaAbsInt]{Abstract interpreter for strictness analysis} \begin{code} #include "HsVersions.h" module SaAbsInt ( findStrictness, findDemand, absEval, widen, fixpoint, isBot ) where IMP_Ubiq(){-uitous-} import CoreSyn import CoreUnfold ( Unfolding(..), UfExpr, RdrName, SimpleUnfolding(..), FormSummary ) import CoreUtils ( unTagBinders ) import Id ( idType, getIdStrictness, getIdUnfolding, dataConTyCon, dataConArgTys ) import IdInfo ( StrictnessInfo(..), wwPrim, wwStrict, wwEnum, wwUnpack ) import Demand ( Demand(..) ) import MagicUFs ( MagicUnfoldingFun ) import Maybes ( maybeToBool ) import Outputable ( Outputable(..){-instance * []-} ) import PprStyle ( PprStyle(..) ) import Pretty ( ppStr ) import PrimOp ( PrimOp(..) ) import SaLib import TyCon ( maybeTyConSingleCon, isEnumerationTyCon, TyCon{-instance Eq-} ) import Type ( maybeAppDataTyConExpandingDicts, isPrimType ) import TysWiredIn ( intTyCon, integerTyCon, doubleTyCon, floatTyCon, wordTyCon, addrTyCon ) import Util ( isIn, isn'tIn, nOfThem, zipWithEqual, pprTrace, panic, pprPanic, assertPanic ) returnsRealWorld x = False -- ToDo: panic "SaAbsInt.returnsRealWorld (ToDo)" \end{code} %************************************************************************ %* * \subsection[AbsVal-ops]{Operations on @AbsVals@} %* * %************************************************************************ Least upper bound, greatest lower bound. \begin{code} lub, glb :: AbsVal -> AbsVal -> AbsVal lub val1 val2 | isBot val1 = val2 -- The isBot test includes the case where lub val1 val2 | isBot val2 = val1 -- one of the val's is a function which -- always returns bottom, such as \y.x, -- when x is bound to bottom. lub (AbsProd xs) (AbsProd ys) = AbsProd (zipWithEqual "lub" lub xs ys) lub _ _ = AbsTop -- Crude, but conservative -- The crudity only shows up if there -- are functions involved -- Slightly funny glb; for absence analysis only; -- AbsBot is the safe answer. -- -- Using anyBot rather than just testing for AbsBot is important. -- Consider: -- -- f = \a b -> ... -- -- g = \x y z -> case x of -- [] -> f x -- (p:ps) -> f p -- -- Now, the abstract value of the branches of the case will be an -- AbsFun, but when testing for z's absence we want to spot that it's -- an AbsFun which can't possibly return AbsBot. So when glb'ing we -- mustn't be too keen to bale out and return AbsBot; the anyBot test -- spots that (f x) can't possibly return AbsBot. -- We have also tripped over the following interesting case: -- case x of -- [] -> \y -> 1 -- (p:ps) -> f -- -- Now, suppose f is bound to AbsTop. Does this expression mention z? -- Obviously not. But the case will take the glb of AbsTop (for f) and -- an AbsFun (for \y->1). We should not bale out and give AbsBot, because -- that would say that it *does* mention z (or anything else for that matter). -- Nor can we always return AbsTop, because the AbsFun might be something -- like (\y->z), which obviously does mention z. The point is that we're -- glbing two functions, and AbsTop is not actually the top of the function -- lattice. It is more like (\xyz -> x|y|z); that is, AbsTop returns -- poison iff any of its arguments do. -- Deal with functions specially, because AbsTop isn't the -- top of their domain. glb v1 v2 | is_fun v1 || is_fun v2 = if not (anyBot v1) && not (anyBot v2) then AbsTop else AbsBot where is_fun (AbsFun _ _ _) = True is_fun (AbsApproxFun _ _) = True -- Not used, but the glb works ok is_fun other = False -- The non-functional cases are quite straightforward glb (AbsProd xs) (AbsProd ys) = AbsProd (zipWithEqual "glb" glb xs ys) glb AbsTop v2 = v2 glb v1 AbsTop = v1 glb _ _ = AbsBot -- Be pessimistic combineCaseValues :: AnalysisKind -> AbsVal -- Value of scrutinee -> [AbsVal] -- Value of branches (at least one) -> AbsVal -- Result -- For strictness analysis, see if the scrutinee is bottom; if so -- return bottom; otherwise, the lub of the branches. combineCaseValues StrAnal AbsBot branches = AbsBot combineCaseValues StrAnal other_scrutinee branches -- Scrutinee can only be AbsBot, AbsProd or AbsTop = ASSERT(ok_scrutinee) foldr1 lub branches where ok_scrutinee = case other_scrutinee of { AbsTop -> True; -- i.e., cool AbsProd _ -> True; -- ditto _ -> False -- party over } -- For absence analysis, check if the scrutinee is all poison (isBot) -- If so, return poison (AbsBot); otherwise, any nested poison will come -- out from looking at the branches, so just glb together the branches -- to get the worst one. combineCaseValues AbsAnal AbsBot branches = AbsBot combineCaseValues AbsAnal other_scrutinee branches -- Scrutinee can only be AbsBot, AbsProd or AbsTop = ASSERT(ok_scrutinee) let result = foldr1 glb branches tracer = if at_least_one_AbsFun && at_least_one_AbsTop && no_AbsBots then pprTrace "combineCase:" (ppr PprDebug branches) else id in -- tracer ( result -- ) where ok_scrutinee = case other_scrutinee of { AbsTop -> True; -- i.e., cool AbsProd _ -> True; -- ditto _ -> False -- party over } at_least_one_AbsFun = foldr ((||) . is_AbsFun) False branches at_least_one_AbsTop = foldr ((||) . is_AbsTop) False branches no_AbsBots = foldr ((&&) . is_not_AbsBot) True branches is_AbsFun x = case x of { AbsFun _ _ _ -> True; _ -> False } is_AbsTop x = case x of { AbsTop -> True; _ -> False } is_not_AbsBot x = case x of { AbsBot -> False; _ -> True } \end{code} @isBot@ returns True if its argument is (a representation of) bottom. The ``representation'' part is because we need to detect the bottom {\em function} too. To detect the bottom function, bind its args to top, and see if it returns bottom. Used only in strictness analysis: \begin{code} isBot :: AbsVal -> Bool isBot AbsBot = True isBot (AbsFun arg body env) = isBot (absEval StrAnal body env) -- Don't bother to extend the envt because -- unbound variables default to AbsTop anyway isBot other = False \end{code} Used only in absence analysis: \begin{code} anyBot :: AbsVal -> Bool anyBot AbsBot = True -- poisoned! anyBot AbsTop = False anyBot (AbsProd vals) = any anyBot vals anyBot (AbsFun arg body env) = anyBot (absEval AbsAnal body env) anyBot (AbsApproxFun _ _) = False -- AbsApproxFun can only arise in absence analysis from the Demand -- info of an imported value; whatever it is we're looking for is -- certainly not present over in the imported value. \end{code} @widen@ takes an @AbsVal@, $val$, and returns and @AbsVal@ which is approximated by $val$. Furthermore, the result has no @AbsFun@s in it, so it can be compared for equality by @sameVal@. \begin{code} widen :: AnalysisKind -> AbsVal -> AbsVal widen StrAnal (AbsFun arg body env) = AbsApproxFun (findDemandStrOnly env body arg) (widen StrAnal abs_body) where abs_body = absEval StrAnal body env {- OLD comment... This stuff is now instead handled neatly by the fact that AbsApproxFun contains an AbsVal inside it. SLPJ Jan 97 | isBot abs_body = AbsBot -- It's worth checking for a function which is unconditionally -- bottom. Consider -- -- f x y = let g y = case x of ... -- in (g ..) + (g ..) -- -- Here, when we are considering strictness of f in x, we'll -- evaluate the body of f with x bound to bottom. The current -- strategy is to bind g to its *widened* value; without the isBot -- (...) test above, we'd bind g to an AbsApproxFun, and deliver -- Top, not Bot as the value of f's rhs. The test spots the -- unconditional bottom-ness of g when x is bottom. (Another -- alternative here would be to bind g to its exact abstract -- value, but that entails lots of potential re-computation, at -- every application of g.) -} widen StrAnal (AbsProd vals) = AbsProd (map (widen StrAnal) vals) widen StrAnal other_val = other_val widen AbsAnal (AbsFun arg body env) | anyBot abs_body = AbsBot -- In the absence-analysis case it's *essential* to check -- that the function has no poison in its body. If it does, -- anywhere, then the whole function is poisonous. | otherwise = AbsApproxFun (findDemandAbsOnly env body arg) (widen AbsAnal abs_body) where abs_body = absEval AbsAnal body env widen AbsAnal (AbsProd vals) = AbsProd (map (widen AbsAnal) vals) -- It's desirable to do a good job of widening for product -- values. Consider -- -- let p = (x,y) -- in ...(case p of (x,y) -> x)... -- -- Now, is y absent in this expression? Currently the -- analyser widens p before looking at p's scope, to avoid -- lots of recomputation in the case where p is a function. -- So if widening doesn't have a case for products, we'll -- widen p to AbsBot (since when searching for absence in y we -- bind y to poison ie AbsBot), and now we are lost. widen AbsAnal other_val = other_val -- WAS: if anyBot val then AbsBot else AbsTop -- Nowadays widen is doing a better job on functions for absence analysis. \end{code} @crudeAbsWiden@ is used just for absence analysis, and always returns AbsTop or AbsBot, so it widens to a two-point domain \begin{code} crudeAbsWiden :: AbsVal -> AbsVal crudeAbsWiden val = if anyBot val then AbsBot else AbsTop \end{code} @sameVal@ compares two abstract values for equality. It can't deal with @AbsFun@, but that should have been removed earlier in the day by @widen@. \begin{code} sameVal :: AbsVal -> AbsVal -> Bool -- Can't handle AbsFun! #ifdef DEBUG sameVal (AbsFun _ _ _) _ = panic "sameVal: AbsFun: arg1" sameVal _ (AbsFun _ _ _) = panic "sameVal: AbsFun: arg2" #endif sameVal AbsBot AbsBot = True sameVal AbsBot other = False -- widen has reduced AbsFun bots to AbsBot sameVal AbsTop AbsTop = True sameVal AbsTop other = False -- Right? sameVal (AbsProd vals1) (AbsProd vals2) = and (zipWithEqual "sameVal" sameVal vals1 vals2) sameVal (AbsProd _) AbsTop = False sameVal (AbsProd _) AbsBot = False sameVal (AbsApproxFun str1 v1) (AbsApproxFun str2 v2) = str1 == str2 && sameVal v1 v1 sameVal (AbsApproxFun _ _) AbsTop = False sameVal (AbsApproxFun _ _) AbsBot = False sameVal val1 val2 = panic "sameVal: type mismatch or AbsFun encountered" \end{code} @evalStrictness@ compares a @Demand@ with an abstract value, returning @True@ iff the abstract value is {\em less defined} than the demand. (@True@ is the exciting answer; @False@ is always safe.) \begin{code} evalStrictness :: Demand -> AbsVal -> Bool -- True iff the value is sure -- to be less defined than the Demand evalStrictness (WwLazy _) _ = False evalStrictness WwStrict val = isBot val evalStrictness WwEnum val = isBot val evalStrictness (WwUnpack demand_info) val = case val of AbsTop -> False AbsBot -> True AbsProd vals -> or (zipWithEqual "evalStrictness" evalStrictness demand_info vals) _ -> trace "evalStrictness?" False evalStrictness WwPrim val = case val of AbsTop -> False other -> -- A primitive value should be defined, never bottom; -- hence this paranoia check pprPanic "evalStrictness: WwPrim:" (ppr PprDebug other) \end{code} For absence analysis, we're interested in whether "poison" in the argument (ie a bottom therein) can propagate to the result of the function call; that is, whether the specified demand can {\em possibly} hit poison. \begin{code} evalAbsence (WwLazy True) _ = False -- Can't possibly hit poison -- with Absent demand evalAbsence (WwUnpack demand_info) val = case val of AbsTop -> False -- No poison in here AbsBot -> True -- Pure poison AbsProd vals -> or (zipWithEqual "evalAbsence" evalAbsence demand_info vals) _ -> panic "evalAbsence: other" evalAbsence other val = anyBot val -- The demand is conservative; even "Lazy" *might* evaluate the -- argument arbitrarily so we have to look everywhere for poison \end{code} %************************************************************************ %* * \subsection[absEval]{Evaluate an expression in the abstract domain} %* * %************************************************************************ \begin{code} -- The isBottomingId stuf is now dealt with via the Id's strictness info -- absId anal var env | isBottomingId var -- = case anal of -- StrAnal -> AbsBot -- See discussion below -- AbsAnal -> AbsTop -- Just want to see if there's any poison in -- error's arg absId anal var env = let result = case (lookupAbsValEnv env var, getIdStrictness var, getIdUnfolding var) of (Just abs_val, _, _) -> abs_val -- Bound in the environment (Nothing, NoStrictnessInfo, CoreUnfolding (SimpleUnfolding _ _ unfolding)) -> -- We have an unfolding for the expr -- Assume the unfolding has no free variables since it -- came from inside the Id absEval anal (unTagBinders unfolding) env -- Notice here that we only look in the unfolding if we don't -- have strictness info (an unusual situation). -- We could have chosen to look in the unfolding if it exists, -- and only try the strictness info if it doesn't, and that would -- give more accurate results, at the cost of re-abstract-interpreting -- the unfolding every time. -- We found only one place where the look-at-unfolding-first -- method gave better results, which is in the definition of -- showInt in the Prelude. In its defintion, fromIntegral is -- not inlined (it's big) but ab-interp-ing its unfolding gave -- a better result than looking at its strictness only. -- showInt :: Integral a => a -> [Char] -> [Char] -- ! {-# GHC_PRAGMA _A_ 1 _U_ 122 _S_ -- "U(U(U(U(SA)AAAAAAAAL)AA)AAAAASAAASA)" {...} _N_ _N_ #-} -- --- 42,44 ---- -- showInt :: Integral a => a -> [Char] -> [Char] -- ! {-# GHC_PRAGMA _A_ 1 _U_ 122 _S_ -- "U(U(U(U(SL)LLLLLLLLL)LL)LLLLLSLLLLL)" _N_ _N_ #-} (Nothing, strictness_info, _) -> -- Includes MagicUnfolding, NoUnfolding -- Try the strictness info absValFromStrictness anal strictness_info in -- pprTrace "absId:" (ppBesides [ppr PprDebug var, ppStr "=:", pp_anal anal, ppStr ":=",ppr PprDebug result]) $ result where pp_anal StrAnal = ppStr "STR" pp_anal AbsAnal = ppStr "ABS" absEvalAtom anal (VarArg v) env = absId anal v env absEvalAtom anal (LitArg _) env = AbsTop \end{code} \begin{code} absEval :: AnalysisKind -> CoreExpr -> AbsValEnv -> AbsVal absEval anal (Var var) env = absId anal var env absEval anal (Lit _) env = AbsTop -- What if an unboxed literal? That's OK: it terminates, so its -- abstract value is AbsTop. -- For absence analysis, a literal certainly isn't the "poison" variable \end{code} Discussion about \tr{error} (following/quoting Lennart): Any expression \tr{error e} is regarded as bottom (with HBC, with the \tr{-ffail-strict} flag, on with \tr{-O}). Regarding it as bottom gives much better strictness properties for some functions. E.g. \begin{verbatim} f [x] y = x+y f (x:xs) y = f xs (x+y) i.e. f [] _ = error "no match" f [x] y = x+y f (x:xs) y = f xs (x+y) \end{verbatim} is strict in \tr{y}, which you really want. But, it may lead to transformations that turn a call to \tr{error} into non-termination. (The odds of this happening aren't good.) Things are a little different for absence analysis, because we want to make sure that any poison (?????) \begin{code} absEval StrAnal (Prim SeqOp [TyArg _, e]) env = ASSERT(isValArg e) if isBot (absEvalAtom StrAnal e env) then AbsBot else AbsTop -- This is a special case to ensure that seq# is strict in its argument. -- The comments below (for most normal PrimOps) do not apply. absEval StrAnal (Prim op es) env = AbsTop -- The arguments are all of unboxed type, so they will already -- have been eval'd. If the boxed version was bottom, we'll -- already have returned bottom. -- Actually, I believe we are saying that either (1) the -- primOp uses unboxed args and they've been eval'ed, so -- there's no need to force strictness here, _or_ the primOp -- uses boxed args and we don't know whether or not it's -- strict, so we assume laziness. (JSM) absEval AbsAnal (Prim op as) env = if any anyBot [absEvalAtom AbsAnal a env | a <- as, isValArg a] then AbsBot else AbsTop -- For absence analysis, we want to see if the poison shows up... absEval anal (Con con as) env | has_single_con = AbsProd [absEvalAtom anal a env | a <- as, isValArg a] | otherwise -- Not single-constructor = case anal of StrAnal -> -- Strictness case: it's easy: it certainly terminates AbsTop AbsAnal -> -- In the absence case we need to be more -- careful: look to see if there's any -- poison in the components if any anyBot [absEvalAtom AbsAnal a env | a <- as, isValArg a] then AbsBot else AbsTop where has_single_con = maybeToBool (maybeTyConSingleCon (dataConTyCon con)) \end{code} \begin{code} absEval anal (Lam (ValBinder binder) body) env = AbsFun binder body env absEval anal (Lam other_binder expr) env = absEval anal expr env absEval anal (App f a) env | isValArg a = absApply anal (absEval anal f env) (absEvalAtom anal a env) absEval anal (App expr _) env = absEval anal expr env \end{code} For primitive cases, just GLB the branches, then LUB with the expr part. \begin{code} absEval anal (Case expr (PrimAlts alts deflt)) env = let expr_val = absEval anal expr env abs_alts = [ absEval anal rhs env | (_, rhs) <- alts ] -- Don't bother to extend envt, because unbound vars -- default to the conservative AbsTop abs_deflt = absEvalDefault anal expr_val deflt env in combineCaseValues anal expr_val (abs_deflt ++ abs_alts) absEval anal (Case expr (AlgAlts alts deflt)) env = let expr_val = absEval anal expr env abs_alts = [ absEvalAlgAlt anal expr_val alt env | alt <- alts ] abs_deflt = absEvalDefault anal expr_val deflt env in let result = combineCaseValues anal expr_val (abs_deflt ++ abs_alts) in {- (case anal of StrAnal -> id _ -> pprTrace "absCase:ABS:" (ppAbove (ppCat [ppr PprDebug expr, ppr PprDebug result, ppr PprDebug expr_val, ppr PprDebug abs_deflt, ppr PprDebug abs_alts]) (ppr PprDebug (keysFM env `zip` eltsFM env))) ) -} result \end{code} For @Lets@ we widen the value we get. This is nothing to do with fixpointing. The reason is so that we don't get an explosion in the amount of computation. For example, consider: \begin{verbatim} let g a = case a of q1 -> ... q2 -> ... f x = case x of p1 -> ...g r... p2 -> ...g s... in f e \end{verbatim} If we bind @f@ and @g@ to their exact abstract value, then we'll ``execute'' one call to @f@ and {\em two} calls to @g@. This can blow up exponentially. Widening cuts it off by making a fixed approximation to @f@ and @g@, so that the bodies of @f@ and @g@ are not evaluated again at all when they are called. Of course, this can lose useful joint strictness, which is sad. An alternative approach would be to try with a certain amount of ``fuel'' and be prepared to bale out. \begin{code} absEval anal (Let (NonRec binder e1) e2) env = let new_env = addOneToAbsValEnv env binder (widen anal (absEval anal e1 env)) in -- The binder of a NonRec should *not* be of unboxed type, -- hence no need to strictly evaluate the Rhs. absEval anal e2 new_env absEval anal (Let (Rec pairs) body) env = let (binders,rhss) = unzip pairs rhs_vals = cheapFixpoint anal binders rhss env -- Returns widened values new_env = growAbsValEnvList env (binders `zip` rhs_vals) in absEval anal body new_env absEval anal (SCC cc expr) env = absEval anal expr env absEval anal (Coerce c ty expr) env = absEval anal expr env \end{code} \begin{code} absEvalAlgAlt :: AnalysisKind -> AbsVal -> (Id,[Id],CoreExpr) -> AbsValEnv -> AbsVal absEvalAlgAlt anal (AbsProd arg_vals) (con, args, rhs) env = -- The scrutinee is a product value, so it must be of a single-constr -- type; so the constructor in this alternative must be the right one -- so we can go ahead and bind the constructor args to the components -- of the product value. ASSERT(length arg_vals == length args) let new_env = growAbsValEnvList env (args `zip` arg_vals) in absEval anal rhs new_env absEvalAlgAlt anal other_scrutinee (con, args, rhs) env = -- Scrutinised value is Top or Bot (it can't be a function!) -- So just evaluate the rhs with all constr args bound to Top. -- (If the scrutinee is Top we'll never evaluated this function -- call anyway!) ASSERT(ok_scrutinee) absEval anal rhs env where ok_scrutinee = case other_scrutinee of { AbsTop -> True; -- i.e., OK AbsBot -> True; -- ditto _ -> False -- party over } absEvalDefault :: AnalysisKind -> AbsVal -- Value of scrutinee -> CoreCaseDefault -> AbsValEnv -> [AbsVal] -- Empty or singleton absEvalDefault anal scrut_val NoDefault env = [] absEvalDefault anal scrut_val (BindDefault binder expr) env = [absEval anal expr (addOneToAbsValEnv env binder scrut_val)] \end{code} %************************************************************************ %* * \subsection[absApply]{Apply an abstract function to an abstract argument} %* * %************************************************************************ Easy ones first: \begin{code} absApply :: AnalysisKind -> AbsVal -> AbsVal -> AbsVal absApply anal AbsBot arg = AbsBot -- AbsBot represents the abstract bottom *function* too absApply StrAnal AbsTop arg = AbsTop absApply AbsAnal AbsTop arg = if anyBot arg then AbsBot else AbsTop -- To be conservative, we have to assume that a function about -- which we know nothing (AbsTop) might look at some part of -- its argument \end{code} An @AbsFun@ with only one more argument needed---bind it and eval the result. A @Lam@ with two or more args: return another @AbsFun@ with an augmented environment. \begin{code} absApply anal (AbsFun binder body env) arg = absEval anal body (addOneToAbsValEnv env binder arg) \end{code} \begin{code} absApply StrAnal (AbsApproxFun demand val) arg = if evalStrictness demand arg then AbsBot else val absApply AbsAnal (AbsApproxFun demand val) arg = if evalAbsence demand arg then AbsBot else val #ifdef DEBUG absApply anal (AbsProd _) arg = panic ("absApply: Duff function: AbsProd." ++ show anal) #endif \end{code} %************************************************************************ %* * \subsection[findStrictness]{Determine some binders' strictness} %* * %************************************************************************ @findStrictness@ applies the function \tr{\ ids -> expr} to \tr{[bot,top,top,...]}, \tr{[top,bot,top,top,...]}, etc., (i.e., once with @AbsBot@ in each argument position), and evaluates the resulting abstract value; it returns a vector of @Demand@s saying whether the result of doing this is guaranteed to be bottom. This tells the strictness of the function in each of the arguments. If an argument is of unboxed type, then we declare that function to be strict in that argument. We don't really have to make up all those lists of mostly-@AbsTops@; unbound variables in an @AbsValEnv@ are implicitly mapped to that. See notes on @addStrictnessInfoToId@. \begin{code} findStrictness :: StrAnalFlags -> [Type] -- Types of args in which strictness is wanted -> AbsVal -- Abstract strictness value of function -> AbsVal -- Abstract absence value of function -> [Demand] -- Resulting strictness annotation findStrictness strflags [] str_val abs_val = [] findStrictness strflags (ty:tys) str_val abs_val = let demand = findRecDemand strflags [] str_fn abs_fn ty str_fn val = absApply StrAnal str_val val abs_fn val = absApply AbsAnal abs_val val demands = findStrictness strflags tys (absApply StrAnal str_val AbsTop) (absApply AbsAnal abs_val AbsTop) in demand : demands \end{code} \begin{code} findDemandStrOnly str_env expr binder -- Only strictness environment available = findRecDemand strflags [] str_fn abs_fn (idType binder) where str_fn val = absEval StrAnal expr (addOneToAbsValEnv str_env binder val) abs_fn val = AbsBot -- Always says poison; so it looks as if -- nothing is absent; safe strflags = getStrAnalFlags str_env findDemandAbsOnly abs_env expr binder -- Only absence environment available = findRecDemand strflags [] str_fn abs_fn (idType binder) where str_fn val = AbsBot -- Always says non-termination; -- that'll make findRecDemand peer into the -- structure of the value. abs_fn val = absEval AbsAnal expr (addOneToAbsValEnv abs_env binder val) strflags = getStrAnalFlags abs_env findDemand str_env abs_env expr binder = findRecDemand strflags [] str_fn abs_fn (idType binder) where str_fn val = absEval StrAnal expr (addOneToAbsValEnv str_env binder val) abs_fn val = absEval AbsAnal expr (addOneToAbsValEnv abs_env binder val) strflags = getStrAnalFlags str_env \end{code} @findRecDemand@ is where we finally convert strictness/absence info into ``Demands'' which we can pin on Ids (etc.). NOTE: What do we do if something is {\em both} strict and absent? Should \tr{f x y z = error "foo"} says that \tr{f}'s arguments are all strict (because of bottoming effect of \tr{error}) or all absent (because they're not used)? Well, for practical reasons, we prefer absence over strictness. In particular, it makes the ``default defaults'' for class methods (the ones that say \tr{defm.foo dict = error "I don't exist"}) come out nicely [saying ``the dict isn't used''], rather than saying it is strict in every component of the dictionary [massive gratuitious casing to take the dict apart]. But you could have examples where going for strictness would be better than absence. Consider: \begin{verbatim} let x = something big in f x y z + g x \end{verbatim} If \tr{x} is marked absent in \tr{f}, but not strict, and \tr{g} is lazy, then the thunk for \tr{x} will be built. If \tr{f} was strict, then we'd let-to-case it: \begin{verbatim} case something big of x -> f x y z + g x \end{verbatim} Ho hum. \begin{code} findRecDemand :: StrAnalFlags -> [TyCon] -- TyCons already seen; used to avoid -- zooming into recursive types -> (AbsVal -> AbsVal) -- The strictness function -> (AbsVal -> AbsVal) -- The absence function -> Type -- The type of the argument -> Demand findRecDemand strflags seen str_fn abs_fn ty = if isPrimType ty then -- It's a primitive type! wwPrim else if not (anyBot (abs_fn AbsBot)) then -- It's absent -- We prefer absence over strictness: see NOTE above. WwLazy True else if not (all_strict || (num_strict && is_numeric_type ty) || (isBot (str_fn AbsBot))) then WwLazy False -- It's not strict and we're not pretending else -- It's strict (or we're pretending it is)! case (maybeAppDataTyConExpandingDicts ty) of Nothing -> wwStrict Just (tycon,tycon_arg_tys,[data_con]) | tycon `not_elem` seen -> -- Single constructor case, tycon not already seen higher up let cmpnt_tys = dataConArgTys data_con tycon_arg_tys prod_len = length cmpnt_tys compt_strict_infos = [ findRecDemand strflags (tycon:seen) (\ cmpnt_val -> str_fn (mkMainlyTopProd prod_len i cmpnt_val) ) (\ cmpnt_val -> abs_fn (mkMainlyTopProd prod_len i cmpnt_val) ) cmpnt_ty | (cmpnt_ty, i) <- cmpnt_tys `zip` [1..] ] in if null compt_strict_infos then if isEnumerationTyCon tycon then wwEnum else wwStrict else wwUnpack compt_strict_infos where not_elem = isn'tIn "findRecDemand" Just (tycon,_,_) -> -- Multi-constr data types, *or* an abstract data -- types, *or* things we don't have a way of conveying -- the info over module boundaries (class ops, -- superdict sels, dfns). if isEnumerationTyCon tycon then wwEnum else wwStrict where (all_strict, num_strict) = strflags is_numeric_type ty = case (maybeAppDataTyConExpandingDicts ty) of -- NB: duplicates stuff done above Nothing -> False Just (tycon, _, _) | tycon `is_elem` [intTyCon, integerTyCon, doubleTyCon, floatTyCon, wordTyCon, addrTyCon] -> True _{-something else-} -> False where is_elem = isIn "is_numeric_type" -- mkMainlyTopProd: make an AbsProd that is all AbsTops ("n"-1 of -- them) except for a given value in the "i"th position. mkMainlyTopProd :: Int -> Int -> AbsVal -> AbsVal mkMainlyTopProd n i val = let befores = nOfThem (i-1) AbsTop afters = nOfThem (n-i) AbsTop in AbsProd (befores ++ (val : afters)) \end{code} %************************************************************************ %* * \subsection[fixpoint]{Fixpointer for the strictness analyser} %* * %************************************************************************ The @fixpoint@ functions take a list of \tr{(binder, expr)} pairs, an environment, and returns the abstract value of each binder. The @cheapFixpoint@ function makes a conservative approximation, by binding each of the variables to Top in their own right hand sides. That allows us to make rapid progress, at the cost of a less-than-wonderful approximation. \begin{code} cheapFixpoint :: AnalysisKind -> [Id] -> [CoreExpr] -> AbsValEnv -> [AbsVal] cheapFixpoint AbsAnal [id] [rhs] env = [crudeAbsWiden (absEval AbsAnal rhs new_env)] where new_env = addOneToAbsValEnv env id AbsTop -- Unsafe starting point! -- In the just-one-binding case, we guarantee to -- find a fixed point in just one iteration, -- because we are using only a two-point domain. -- This improves matters in cases like: -- -- f x y = letrec g = ...g... -- in g x -- -- Here, y isn't used at all, but if g is bound to -- AbsBot we simply get AbsBot as the next -- iteration too. cheapFixpoint anal ids rhss env = [widen anal (absEval anal rhs new_env) | rhs <- rhss] -- We do just one iteration, starting from a safe -- approximation. This won't do a good job in situations -- like: -- \x -> letrec f = ...g... -- g = ...f...x... -- in -- ...f... -- Here, f will end up bound to Top after one iteration, -- and hence we won't spot the strictness in x. -- (A second iteration would solve this. ToDo: try the effect of -- really searching for a fixed point.) where new_env = growAbsValEnvList env [(id,safe_val) | id <- ids] safe_val = case anal of -- The safe starting point StrAnal -> AbsTop AbsAnal -> AbsBot \end{code} \begin{verbatim} mkLookupFun :: (key -> key -> Bool) -- Equality predicate -> (key -> key -> Bool) -- Less-than predicate -> [(key,val)] -- The assoc list -> key -- The key -> Maybe val -- The corresponding value mkLookupFun eq lt alist s = case [a | (s',a) <- alist, s' `eq` s] of [] -> Nothing (a:_) -> Just a \end{verbatim} \begin{code} fixpoint :: AnalysisKind -> [Id] -> [CoreExpr] -> AbsValEnv -> [AbsVal] fixpoint anal [] _ env = [] fixpoint anal ids rhss env = fix_loop initial_vals where initial_val id = case anal of -- The (unsafe) starting point StrAnal -> if (returnsRealWorld (idType id)) then AbsTop -- this is a massively horrible hack (SLPJ 95/05) else AbsBot AbsAnal -> AbsTop initial_vals = [ initial_val id | id <- ids ] fix_loop :: [AbsVal] -> [AbsVal] fix_loop current_widened_vals = let new_env = growAbsValEnvList env (ids `zip` current_widened_vals) new_vals = [ absEval anal rhs new_env | rhs <- rhss ] new_widened_vals = map (widen anal) new_vals in if (and (zipWith sameVal current_widened_vals new_widened_vals)) then current_widened_vals -- NB: I was too chicken to make that a zipWithEqual, -- lest I jump into a black hole. WDP 96/02 -- Return the widened values. We might get a slightly -- better value by returning new_vals (which we used to -- do, see below), but alas that means that whenever the -- function is called we have to re-execute it, which is -- expensive. -- OLD VERSION -- new_vals -- Return the un-widened values which may be a bit better -- than the widened ones, and are guaranteed safe, since -- they are one iteration beyond current_widened_vals, -- which itself is a fixed point. else fix_loop new_widened_vals \end{code} For absence analysis, we make do with a very very simple approach: look for convergence in a two-point domain. We used to use just one iteration, starting with the variables bound to @AbsBot@, which is safe. Prior to that, we used one iteration starting from @AbsTop@ (which isn't safe). Why isn't @AbsTop@ safe? Consider: \begin{verbatim} letrec x = ...p..d... d = (x,y) in ... \end{verbatim} Here, if p is @AbsBot@, then we'd better {\em not} end up with a ``fixed point'' of @d@ being @(AbsTop, AbsTop)@! An @AbsBot@ initial value is safe because it gives poison more often than really necessary, and thus may miss some absence, but will never claim absence when it ain't so. Anyway, one iteration starting with everything bound to @AbsBot@ give bad results for f = \ x -> ...f... Here, f would always end up bound to @AbsBot@, which ain't very clever, because then it would introduce poison whenever it was applied. Much better to start with f bound to @AbsTop@, and widen it to @AbsBot@ if any poison shows up. In effect we look for convergence in the two-point @AbsTop@/@AbsBot@ domain. What we miss (compared with the cleverer strictness analysis) is spotting that in this case f = \ x y -> ...y...(f x y')... \tr{x} is actually absent, since it is only passed round the loop, never used. But who cares about missing that? NB: despite only having a two-point domain, we may still have many iterations, because there are several variables involved at once.